Sie sind auf Seite 1von 170

Westflische Wilhelms-Universitt Mnster

Institut fr Physiologie I
Geschftsfhrender Direktor: Univ.-Prof. Dr. rer. nat. H.-C. Pape

Zellulre Mechanismen der physiologischen


und pathophysiologischen Rhythmogenese im
thalamokortikalen System

Kumulative Habilitationsschrift
zur Erlangung der Venia legendi fr das Fach Physiologie
an der Medizinischen Fakultt
der Westflischen Wilhelms-Universitt Mnster

vorgelegt von
Dr. rer. nat. Philippe Coulon
Mnster
2012

Formale Grundlagen zum Habilitationsantrag

Dieser Antrag basiert auf 17 wissenschaftlichen Publikationen (16 Originalarbeiten


und ein bersichtsartikel), sowie 19 Tagungsbeitrgen (siehe Anhang). Von den
Originalarbeiten und bersichtsartikeln sind 15 bereits publiziert und unterlagen
dem peer-review-Verfahren [gefordert: 12], eine Originalarbeit ist zur Begutachtung
eingereicht (submitted) und eine Arbeit befindet sich im Revisionsverfahren (under
revision). Der Verfasser dieser Habilitationsschrift ist Erstautor, Letztautor oder
Korrespondenzautor in sieben bereits verffentlichten Publikationen (Erstautor: 5;
Letztautor: 1; korrespondierender Autor: 1) [gefordert: 6]. Gem "Journal Citation
Report" des "Science Citation Index" und unter Beachtung der Vorgaben des
Fachbereichsrats vom 15.05.2012, wurde jeder der 15 Verffentlichungen ein
"impact factor" zugewiesen. Zusammengenommen addiert sich der kumulative
"impact factor" zu 68,5 Punkten. Gem dem Punktesystem der Habilitationskriterien der Medizinischen Fakultt der Westflischen Wilhelms-Universitt Mnster
addieren sich die 15 Publikationen zu 92 Punkten (Klasse 1: 10; Klasse 2: 4; Klasse
3: 1).

Im folgenden Text wurden 8 thematisch zusammenhngende Publikationen fr eine


kumulative Habilitationsschrift gem der Habilitationsordnung zu einer zusammenhngenden Arbeit zusammengefasst [gefordert: 5]. Von diesen Publikationen sind
fnf der Klasse 1 zuzuordnen, drei sind der Klasse 2 zuzuordnen. Der Antragsteller
ist dabei Erstautor auf vier Publikationen der Klasse 1 [gefordert: 3].

Einleitung

Inhaltsverzeichnis
Formale Grundlagen zum Habilitationsantrag ................................................................................. 2
Inhaltsverzeichnis ................................................................................................................................. 3
1.

2.

Einleitung ..................................................................................................................................... 4
1.1.

Der Thalamus und das thalamokortikale System ......................................................... 7

1.2.

Der Nucleus reticularis thalami (NRT) ......................................................................... 11

1.3.

Thalamische Erregungsmuster bei Schlaf und Epilepsie .......................................... 13

1.4.

Hypothesen ...................................................................................................................... 15

Eigene Arbeiten ......................................................................................................................... 18


2.1.

Die Rolle von HCN-Kanlen in der Rhythmogenese ................................................. 18

2.2.
Intrazellulres Calcium beeinflusst elektrophysiologische Aktivittsmuster in
thalamischen Neuronen ............................................................................................................... 22

3.

2.3.

Aktivierung thalamischer Afferenzen: bertragung an den Kortex .......................... 31

2.4.

Schlaf, Narkose und Epilepsie: Gemeinsame Mechanismen................................... 35

2.5.

Die Rolle von G-Proteinen bei der Signalverarbeitung im Thalamus ..................... 38

Zusammenfassung ................................................................................................................... 40

Literaturverzeichnis ............................................................................................................................ 43
Anhang ................................................................................................................................................ 52
Der kumulativen Habilitationsschrift zugrundeliegende Publikationen ........................ 52
Lehrerfahrung ....................................................................................................................... 53
Eingeworbene Frdermittel und Preise ............................................................................ 56
Gesamtpublikationsliste ...................................................................................................... 56
Lebenslauf Philippe Coulon ............................................................................................. 63
Danksagungen ................................................................................................................................... 64

Einleitung

1.

Einleitung

Im Zentralnervensystem (ZNS) generieren Nervenzellen elektrische Signale,


die in binrer Weise Informationen weiterleiten knnen (Aktionspotentiale,
neuronale Erregungen). Dabei ist nicht die Amplitude der Signale von
Bedeutung, sondern deren Frequenz. Erregungsmuster knnen z. B. Sinnesinformationen kodieren, motorische Ablufe steuern oder auch komplexe
Empfindungen abspeichern. Werden diese Signale jedoch ber weite Teile
des ZNS in rhythmischem Einklang generiert, knnen diese Funktionen nicht
mehr erfllt werden. Dieser Zustand tritt unter physiologischen Bedingungen
whrend des Schlafs auf, kann aber auch pathophysiologischen Ursprungs
sein, wie bei epileptischen Anfllen.

Die Einstellung der Rhythmogenese neuronaler Netzwerke ist also entscheidend fr die Funktion des ZNS. Eines der wichtigsten Netzwerke zur
Generierung von rhythmischer Aktivitt ist das thalamokortikale System,
welches in einer vereinfachten Betrachtung aus dem Thalamus, dem
Nucleus reticularis thalami (NRT) und dem Kortex gebildet wird. Der
Thalamus ist die Schnittstelle zwischen Empfindung und Wahrnehmung.
Abhngig davon, ob der Organismus wach ist oder schlft, bestimmen dabei
die Nervenzellverbnde des Thalamus in erster Instanz, welche Informationen zur Grohirnrinde (Kortex) weitergeleitet werden. Integrative
Prozesse des Kortex werden so vom Thalamus mageblich mitbestimmt
4

Einleitung

(Pape et al., 2005) und z. B. whrend des Schlafs von sensorischen


Eingngen entkoppelt.

Es erscheint verwunderlich, dass die Funktion des Schlafs nach wie vor
unklar ist. Es wird angenommen, dass sie u. a. in der Regeneration der
neuronalen Energiereserven liegt. Auerdem findet offenbar whrend des
Schlafs eine Konsolidierung von Gedchtnisinhalten statt. Obwohl Schlaf fr
Sugetiere lebensnotwendig ist und Schlafentzug von Menschen als Qual
empfunden wird, ist Schlafmangel nur sehr selten lebensbedrohend. Es
scheint, dass Schlafentzug zu einem groen Teil und ber lngere Zeitrume
kompensiert werden kann, wobei mglicherweise lokale neuronale Verbnde
in einen schlafhnlichen Zustand bergehen (Krueger et al., 2008; Coulon et
al., 2012). Eine Voraussetzung fr den Schlaf bei Sugetieren ist die
synchrone Aktivitt elektrischer Entladungen im thalamokortikalen System
(Coulon et al., 2012).

Neben der physiologischen synchronen Aktivitt whrend der Schlafphasen


gibt es jedoch auch pathophysiologische Formen synchroner Aktivitt, wie
epileptische Anflle. Die Anflle der Absence-Epilepsie haben ihren Ursprung
im thalamokortikalen System (Crunelli & Leresche, 2002a) und nutzen somit
wahrscheinlich dieselben Mechanismen thalamokortikaler Synchronizitt, die
auch fr die Regulation der Schlaf-Wach-Aktivitt mageblich sind. Die
Absence-Epilepsie ist eine nicht-konvulsive, d.h. ohne Muskelkrmpfe
5

Einleitung

einhergehende Form der Epilepsie. Die Anflle werden als generalisiert


bezeichnet, weil sich die typischen elektrischen Aktivittsmuster fast ber
den gesamten Kortex erstrecken. Die bei Kindern und Jugendlichen
auftretende Absence-Epilepsie verschwindet in 70% der Flle mit der
Pubertt wieder, geht jedoch manchmal in andere Formen der Epilepsie
ber.

Auerdem

sind

hufig

Einschrnkungen

der

kognitiven

Leistungsfhigkeit die Folge (Wirrell, 2003). Es gibt mehrere Formen dieses


Epilepsie-Typs, zu denen die Absence-Epilepsie im Kindesalter (engl.
childhood absence epilepsy, CAE) und jene im Jugendalter (engl. juvenile
absence epilepsy, JAE) gehren. Die ersten Anflle der CAE treten zumeist
im Alter von 5-7 Jahren auf (Crunelli & Leresche, 2002a; Wirrell, 2003) und
zeichnen sich durch milde motorische Symptome aus, wie ein Zittern der
Augenlider oder eine Dorsalflektion der Zehen (Crunelli & Leresche, 2002a).
Die Erkrankung bleibt daher oft zunchst unentdeckt und wird im Umfeld des
Kindes, z. B. in der Schule, fr Tagtrumerei gehalten. Die Betroffenen sind
whrend eines Anfalls nicht ansprechbar, reagieren nicht auf Umweltreize
und behalten keine Erinnerung an Geschehnisse, die whrend des Anfalls
stattfinden. Die psychosoziale Prognose fr die Kinder ist oft schlecht,
sowohl was schulische Leistungen betrifft, als auch fr soziale und
zwischenmenschliche Beziehungen (Crunelli & Leresche, 2002a; Wirrell,
2003).

Einleitung

Ziel der Arbeiten, die hier zusammengefasst wurden ist es, gemeinsame
Mechanismen der Rhythmogenese bei Schlaf-Wach-Regulation und der
Epilepsie-Entstehung zu untersuchen und hieraus ein tieferes Verstndnis
der Steuerung rhythmischer Aktivitt im ZNS unter physiologischen und
pathophysiologischen Bedingungen zu erhalten.

1.1. Der Thalamus und das thalamokortikale System

In

einer

vereinfachten

Betrachtung

der

sensorischen

Informations-

verarbeitung bilden Teile des Thalamus (die sog. Relaiskerne), des NRT und
der sensorischen Areale des Kortex ein neuronales Netzwerk, das sog.
thalamokortikale System. Das Schema in Abb. 1 zeigt die Verschaltungen
sensorischer Eingnge in diesem Netzwerk an einem Beispiel. Visuelle
Signale, die ber die Sehbahn von der Netzhaut (Retina, RET) an das
Corpus geniculatum laterale pars dorsalis (CGLd) des Thalamus gesendet
werden, werden bei Wachheit an die korrespondierenden und topographisch
geordneten sensorischen Zentren des Kortex weitergeleitet. Das CGLd ist
ebenfalls topographisch geordnet und besteht aus mindestens zwei Typen
von Nervenzellen: Thalamokortikale Schaltneuronen (TC Neuronen, rote
Ellipse in Abb. 1) und lokale Interneuronen (grne Ellipse). TC Neuronen
projizieren mit ihren Axonen zum Kortex und terminieren dort mit ihren
Synapsen auf kortikalen Neuronen der Schicht 4. Gleichzeitig besitzen sie

Einleitung

Kollateralen, die zum NRT abzweigen und dort auf hemmenden Neuronen
terminieren, die ebenfalls topographisch geordnet sind (Lam et al., 2006).

Schicht 1 und 2

Schicht 6
Erregung
Hemmung

NR
T

NOT
NPB

RET

Thalamus
Cortex

BV
NPB
NOT

NPB

Abb. 1: Das thalamokortikale System. Schema der Verschaltungen der drei Hauptzentren
des thalamokortikalen Systems: Relaiskerne des Thalamus (Thalamus), Nucleus reticularis
thalami (NRT), Kortex. Nach Coulon et al. (2012), modifiziert nach Sherman & Guillery
(1996).

Einleitung

Diese hemmenden NRT-Neuronen besitzen eine intrathalamische rekurrente


Verschaltung; sie projizieren nicht zum Kortex (Pinault, 2004; Fuentealba &
Steriade, 2005; Sherman & Guillery, 2006). Kortikale Pyramidenzellen der
Schicht 6 projizieren zurck zum Thalamus, wobei von diesen Projektionen
ebenfalls Kollateralen zum NRT abzweigen. Die Neuronen des NRT erhalten
somit

Eingnge

sowohl

aus

dem

Kortex

(von

kortikothalamischen

Projektionen), als auch aus dem Thalamus (von

thalamokortikalen

Projektionen), die berwiegend erregend sind. Auerdem erhalten sie


weitere inhibitorische Eingnge, z. B. aus dem basalen Vorderhirn (BV). Die
Neuronen des Hirnstamms aktivieren ber ihre Projektionen durch
verschiedene Neurotransmitter das Vorderhirn. Dies stellt einen wichtigen
Mechanismus dar, um den Verhaltenszustand des gesamten Organismus zu
beeinflussen. Dabei werden die Neurotransmitter Acetylcholin (ACh),
Noradrenalin (NA) und Serotonin (5-hydroxytryptamin, 5-HT) aus den
Synapsen der Neuronen des Hirnstamms auf thalamische Neuronen
ausgeschttet. Diese Ausschttung bewirkt dabei mehrheitlich die Aktivierung
von G-Protein-gekoppelten Rezeptoren. Die absteigenden Eingnge von
kortikalen Neuronen wirken ber den erregenden Neurotransmitter Glutamat,
vor allem ber metabotrope Glutamatrezeptoren (Salt, 2002).

Die genannten drei Typen von Neuronen im thalamokortikalen System (TC


Neuronen,

NRT-Neuronen,

kortikale

Pyramidenzellen)

spielen

eine

magebliche Rolle bei der Erzeugung rhythmisch-oszillatorischer Aktivitt im


9

Einleitung

thalamokortikalen System. Es lassen sich grundstzlich zwei elektrogene


Aktivittsmuster unterscheiden, die abhngig vom Verhaltenszustand sind
und

ein

bistabiles Membranruhepotential

in

thalamischen

Neuronen

bedingen:

1. Whrend Wachheit sind TC Neuronen depolarisiert und knnen


einzelne schnelle Aktionspotentiale (APs) generieren (Steriade et al.,
1997). Vermutlich ist dies die Basis fr die Weiterleitung sensorischer
Informationen der Sinnesorgane.

2. Whrend ruhiger Wachheit oder whrend des Tiefschlafs ist das


Membranpotential von TC Neuronen hyperpolarisiert. Die Neuronen
zeigen rhythmische, stereotype Entladungsmuster: Eine von Ca 2+
getragene,

verhltnismig

langsame,

niederschwellige

und

vorbergehende Depolarisation (engl. low threshold Ca2+-spike, LTS)


fhrt zur salvenartigen Generierung einiger schneller Aktionspotentiale
(engl. burst, Salvenaktivitt). Diese sich wiederholende Aktivitt , die
als Schlafspindeln und -Wellen im Elektroenzephalogramm (EEG)
sichtbar ist, unterdrckt vermutlich die eingehenden sensorischen
Signale und bildet die Basis rhythmisch-oszillatorischer Aktivitt im
gesamten Vorderhirn.

10

Einleitung

Diese Aktivittsmuster knnen unter anderem durch eine Vernderung der


elektrophysiologischen Eigenschaften der beteiligten Neuronen beeinflusst
werden, insbesondere ber die Art und Anzahl der Ionenkanle in der
Plasmamembran. Diese Ionenkanle knnen auch selbst in ihren Eigenschaften moduliert werden und so zu unterschiedlichen Aktivittsmustern
fhren. So knnen z. B. Neuronen des Hirnstamms ber ihre Synapsen auf
thalamischen Neuronen eine Reihe von Neurotransmittern ausschtten, die
Ionenkanle derart beeinflussen, dass die thalamischen Neuronen von
Salvenaktivitt auf tonische Aktivitt wechseln (Broicher et al., 2008b; Coulon
et al., 2010).

1.2. Der Nucleus reticularis thalami (NRT)

Der NRT gilt als Zentrum fr die Modulation der sensorischen Schaltneuronen im Thalamus und scheint fr die Generierung rhythmischer Aktivitt
im thalamokortikalen System von besonderer Bedeutung zu sein. Er gilt als
Schrittmacher (engl. pacemaker) langsamer Oszillationen im Thalamus
(Bal & McCormick, 1993; Llinas & Ribary, 1993; Steriade et al., 1997;
Fuentealba & Steriade, 2005). Auerdem spielt dieser Nucleus eine
Schlsselrolle bei der Kontrolle selektiver Aufmerksamkeit (Crick, 1984;
Guillery et al., 1998).

11

Einleitung

Die Neuronen des NRT bilden eine dnne Schicht aus Zellen um den
dorsalen Thalamus. Einzelne Zellverbnde innerhalb dieser Schicht sind
topographisch nach sensorischen Eingngen geordnet (Lam et al., 2006).
Alle Neuronen im NRT nutzen fr die chemisch-synaptische bertragung den
inhibitorischen Neurotransmitter -Aminobuttersure (engl. -aminobutyric
acid, GABA) (Pinault, 2004). Wie andere Neuronen des Thalamus sind auch
diese dazu in der Lage, zwei Aktivittszustnde einzunehmen. Allerdings
scheint dem NRT dabei eher die Funktion der Generierung von
Salvenaktivitt zuzukommen. Diese kann mit einer hheren Frequenz
erzeugt werden, als einzelne Aktionspotentiale (Steriade et al., 1986; Coulon
et al., 2009).

Neuronen des NRT sind untereinander komplex verschaltet. Einige der


GABAergen Synapsen terminieren direkt innerhalb des Nucleus. Auerdem
sind

einige

NRT-Neuronen

ber

elektrische

Synapsen

miteinander

verbunden (Landisman et al., 2002; Connors & Long, 2004). Charakteristisch


ist dabei, dass elektrische Synapsen meist zwischen Zellen in unmittelbarer
Nachbarschaft auftreten. Zellen, die ber elektrische Synapsen untereinander in Verbindung stehen, weisen zudem offenbar keine GABAerge
Verbindung untereinander auf (Landisman et al., 2002).

Eingnge von Neuronen des basalen Vorderhirns blockieren die Schlafspindeln und haben somit einen desynchronisierenden Effekt (Steriade et al.,
12

Einleitung

1987). Es ist allerdings nach wie vor unklar, ob der NRT diese Schlafspindeln
selbst generiert (De Gennaro & Ferrara, 2003) oder ob erst eine Interaktion
zwischen NRT und TC Neuronen dies bewerkstelligt (Bal & McCormick,
1993; Huguenard & McCormick, 2007).

Der NRT war Gegenstand zahlreicher Untersuchungen, die unter verschiedenen experimentellen Bedingungen, wie Stimulation, Lsion oder Isolation
durchgefhrt wurden, und zum Ziel hatten, die Netzwerkeigenschaften dieses
Kerns aufzuklren. Auerdem wurden einzelne Neuronen elektrophysiologisch untersucht oder deren elektrische Aktivitt wurde anhand eines
Computermodells simuliert, um die Eigenschaften der NRT-Neuronen auf
Einzelzellebene zu erfassen (siehe Pinault, 2004). Dennoch ist die genaue
Funktion dieses Kerngebiets und seiner Neuronen bislang unklar.

1.3. Thalamische Erregungsmuster bei Schlaf und Epilepsie

Die rhythmischen Vernderungen der elektrischen Aktivitt des menschlichen


Gehirns knnen mit der Elektroenzephalographie gemessen werden. Die
Aktivittsmuster whrend des Schlafs unterscheiden sich gravierend von
denen bei Wachheit (Berger, 1929). Nach seinen Frequenzmustern im EEG
wird der menschliche Schlaf in fnf Stadien eingeteilt (Rechtschaffen &
Kales, 1968), die jedoch fr experimentelle Studien zur Vereinfachung auf
zwei Stadien reduziert werden: REM-Schlaf oder paradoxer Schlaf (engl.
13

Einleitung

rapid eye movement sleep, REM-sleep, paradoxical sleep) und Tiefschlaf


(engl. slow-wave sleep oder non-REM-sleep) (Reinoso-Surez et al.,
2011). Das EEG whrend der Tiefschlafphasen wird von langsamen Wellen
hoher Amplitude dominiert. Dabei zeigen Neuronen des thalamokortikalen
Systems Salvenaktivitt in einem Frequenzbereich von 0,5 bis 4 Hz (Deltawellen, -Wellen) oder Spindelaktivitt mit einer Frequenz von 7-14 Hz
(McCarley et al., 1983; Roy et al., 1984; Amzica & Steriade, 1995).

Synchrone Aktivitt rekrutiert individuelle TC Neuronen durch ein konzertiertes Wechselspiel von GABAergen inhibitorischen postsynaptischen
Potentialen, die aus dem NRT eintreffen, und zwei Typen spannungsgesteuerter Ionenkanle: T-Typ Ca2+-Kanle, durch die der Strom IT fliet
und hyperpolarisationsaktivierte Kationenkanle (HCN-Kanle), durch die der
Strom Ih fliet (McCormick & Pape, 1990; Steriade et al., 1993; McCormick &
Bal, 1994; Gerard et al., 2012).

Den -Wellen beim Schlaf hneln die bilateralen Spitze-Welle-Entladungen


(engl. spike-and-wave-discharge, SWD), die bei der Absence-Epilepsie
auftreten. Diese Anflle dauern nur ca. 5 bis 20 s, knnen jedoch bis zu
mehrere hundert Mal pro Tag auftreten (Crunelli & Leresche, 2002a). Da
typische Antikonvulsiva, wie Carbamazepin, Gabapentin, Phenytoin oder
Tiagabin die Symptomatik verstrken, wurden Medikamente erprobt, die
gezielt in die Mechanismen der Rhythmogenese des thalamokortikalen
14

Einleitung

Systems eingreifen (Crunelli & Leresche, 2002b). Ethosuximid, Levetiracetam, Trimethadion und Valproat konnten die Anfallshufigkeit und -dauer
verringern. Fr Ethosuximid konnte gezeigt werden, dass es u. a. niedrigschwellige Ca2+-Kanle des T-Typs hemmt (Broicher et al., 2007). Bei bis zu
40% aller Patienten, die unter Absence-Epilepsie leiden, liegt eine genetische
Ursache der Erkrankung vor, wobei Mutationen in T-Typ Ca2+-Kanlen und
GABAA-Rezeptoren nachgewiesen wurden (Wallace et al., 2001; Chen et al.,
2003; Scheffer, 2003; Khosravani et al., 2004; Khosravani & Zamponi, 2006;
Vitko

et

al.,

2007).

Auerdem

wurde

gezeigt,

dass

rhythmische,

physiologische Schlafaktivitt in der Form von Schlafspindeln in SWD


permutieren knnen (Gloor, 1978).

1.4. Hypothesen

Die Rhythmogenese im thalamokortikalen System basiert auf drei wichtigen


Mechanismen: a) Intrinsische elektrophysiologische Eigenschaften der
beteiligten

thalamischen

Neuronen;

b)

Eigenschaften

des

lokalen

synaptischen Netzwerkes und c) Aktivittszustnde extrathalamischer


Eingangssysteme (Steriade, 1991). Zu diesen wichtigen Grundmechanismen
sollten im Rahmen der hier beschriebenen Arbeiten wichtige Ergnzungen
vorgenommen werden. Die dazugehrigen Hypothesen lauten:
zu a):

15

Einleitung

1. Die intrinsischen elektrophysiologischen Eigenschaften werden durch


reziproke Interaktionen zwischen unterschiedlichen Ionenkanaltypen
verndert (Budde et al., 2008; Coulon et al., 2009; Gerard et al.,
2012).
2. Die intrinsischen elektrophysiologischen Eigenschaften werden durch
intrazellulre Ca2+-Signalwege beeinflusst (Coulon et al., 2009;
Rankovic et al., 2010), wobei die nderungen der intrazellulren Ca2+Konzentration ([Ca2+]i) wiederum Einfluss auf Ionenkanle nimmt, die
mageblich die Rhythmogenese modulieren.
zu b):
3. Die Eigenschaften des lokalen synaptischen Netzwerkes werden
durch intrazellulre Ca2+-abhngige Mechanismen beeinflusst, wobei
die distinkten Ca2+-Signale im NRT und im CGLd auf der
unterschiedlichen Auslsung grundstzlich hnlicher Mechanismen
beruhen (Budde et al., 2000; Cueni et al., 2008; Coulon et al., 2009).
4. Die Eigenschaften des lokalen synaptischen Netzwerkes werden
durch Inhalationsnarkotika in hnlicher Weise beeinflusst wie bei
natrlichem Schlaf, wobei Effekte zum Tragen kommen, die denjenigen von Hirnstammtransmittern entgegengesetzt sind (Budde et al.,
2008).
5. Durch spannungssensitive Farbstoffe kann die rumliche und zeitliche
Ausbreitung der durch thalamokortikale Projektionen im Kortex
hervorgerufenen Erregung untersucht werden. In zuknftigen Arbeiten
16

Einleitung

kann diese Methode Verwendung finden, um die Einflsse vernderter


elektrophysiologischer Eigenschaften einzelner thalamischer Neuronen auf das thalamokortikale System im Epilepsiemodell zu untersuchen (Broicher et al., 2010; Broicher & Speckmann, 2012).
zu c):
6. Die extrathalamischen Einflsse, insbesondere durch Hirnstammtransmitter, werden in der Zielzelle ber G-Proteine integriert, die
somit ein wichtiges Modulationsziel darstellen (Broicher et al., 2008b;
Coulon et al., 2010).

Im Rahmen der Beantwortung dieser Hypothesen konnte darber hinaus das


neue Prinzip der Koexistenz lokaler und globaler Schlafmechanismen
entwickelt werden, welches in einem bersichtsartikel beschrieben wird
(Coulon et al., 2012).

Auerdem kristallisierte sich im Verlaufe der Arbeiten eine zentrale


Bedeutung elektrischer Kopplung im NRT fr die thalamische Rhythmogenese heraus. Die noch nicht publizierten Vorarbeiten dazu fhrten zur
erfolgreichen

Einwerbung

von

Frdermitteln:

Die

Rolle

elektrischer

Synapsen in einem Epilepsiemodell: Ca2+-vermittelte Wechselwirkungen


zwischen

Neuronen

des

Nucleus

Reticularis

Thalami.

Innovative

Medizinische Forschung (IMF) Mnster. 138.000 Euro (CO 1 2 10 08).

17

Eigene Arbeiten

2.

Eigene Arbeiten

2.1. Die Rolle von HCN-Kanlen in der Rhythmogenese

[GERARD E, HOCHSTRATE P, DIERKES PW, COULON P (2012): Functional


properties and cell type specific distribution of Ih channels in leech
neurons. J Exp Biol., 215(2), 227-238.]

Schlaf und schlafhnliche Zustnde finden sich nicht nur bei Vertebraten
sondern z. B. auch bei Nematoden (Caenorhabditis elegans, (Raizen et al.,
2008)), Fruchtfliegen (Drosophila melanogaster (Cirelli & Bushey, 2008)) und
selbst im Pflanzenreich (McClung, 2006). Beim medizinischen Blutegel kann
erhhte oder verringerte motorische Aktivitt durch Mechanismen ausgelst
werden, die mit der Schlaf-/Wachregulation bei Sugetieren vergleichbar sind
(Hashemzadeh-Gargari & Friesen, 1989). Ein Schlafzustand wurde beim
medizinischen Blutegel zwar nicht beschrieben, jedoch treten z. B. vor und
nach der Nahrungsaufnahme stille Ruhezustnde auf (quiescence resting,
(Dickinson & Lent, 1984)) die zumindest schlafhnlich sind.

Dem hyperpolarisationsaktivierten Kationenstrom Ih kommt bei Oszillationen


des Membranpotentials eine wichtige Bedeutung zu (Abb. 2). Dieser Strom
kommt sowohl bei Vertebraten, als auch bei Evertebraten vor (Pape, 1996)
und fliet durch Ionenkanlen, die durch eine Membranhyperpolarisation
18

Eigene Arbeiten

aktiviert und von zyklischen Nukleotiden moduliert werden (HCN-Kanle,


engl. hyperpolarisation-activated cyclic nucleotide-gated cation conductance).

2+

Abb. 2: HCN-Kanle knnen im Wechselspiel mit dem T-Typ Ca -Kanal Oszillationen des
Membranpotentials verursachen. Die Rckkopplung kann dabei ber das eingestrmte Ca

2+

2+

und die Ca -abhngige Adenylylcyclase (AC) stattfinden, die wiederum HCN-Kanal


modulierendes cAMP produziert. Modifiziert nach Coulon et al. (2012).

In Neuronen erfllt dieser Strom mindestens zwei Aufgaben: 1. Rhythmogenese und 2. Stabilisierung des Membranpotentials. Diese beiden Funktionen bedingen einander, da durch eine dauerhafte Aktivierung des Ih um das
19

Eigene Arbeiten

Ruhemembranpotential herum eine Schwankung des Membranpotentials


verhindert wird (Gerard et al., 2012), jedoch gengend MembranpotentialVernderung eine Aktivierung (bei Hyperpolarisation) oder Deaktivierung (bei
Depolarisation)

des

Stroms

zur

Folge

hat.

Ist

der Auslser

der

Membranpotential-Vernderung z. B. ein Strom durch Ca2+-Kanle des TTyps (IT), welcher selbst spannungsaktiviert ist, so ergeben sich aus dem
Wechselspiel zwischen Aktivierung und Deaktivierung Oszillationen des
Membranpotentials: Die Aktivierung des IT depolarisiert die Membran,
wodurch der Ih deaktiviert wird. Diese Deaktivierung bewirkt aber zusammen
mit der zeitabhngigen Inaktivierung des IT eine Hyperpolarisation, die
wiederum Ih aktiviert (Pape, 1996). Dies hat eine Depolarisation zur Folge,
die wieder IT aktiviert, do dass der Zyklus von neuem beginnt.

Der Nematode Caenorhabditis elegans wurde bereits als Modellsystem zur


Analyse der Schlaffunktion etabliert (Raizen et al., 2008) und es konnte
gezeigt werden, dass das intrazellulre Signalmolekl cAMP in C. elegans,
Drosophila und in Sugetieren schlafrelevante Erregungsmuster unterdrckt.
Um universelle Konzepte der Schlaffunktion erkennen zu knnen, wurden
Untersuchungen an orthologen Ionenkanlen, die in Sugetieren wichtig fr
den Schlaf sind, im medizinischen Blutegel, Hirudo verbana, durchgefhrt.
Dabei wurde eine hauptschlich stabilisierende Wirkung des Ih auf das
Membranpotential sensorischer Neuronen gezeigt, wo ein IT-hnlicher Strom
nicht nachzuweisen war. Ein IT-hnlicher Strom wurde beim medizinischen
20

Eigene Arbeiten

Blutegel nur in rhythmisch aktiven Herz-Interneuronen gezeigt. Alle anderen


untersuchten Neuronen zeigten einen Ca2+-Strom, der eher dem L-Typ bei
Vertebraten entsprach (Dierkes et al., 1997, 2004). Die Eigenschaften des Ih
bei sensorischen Neuronen des medizinischen Blutegels entsprachen weitgehend denen in Vertebraten-Neuronen (Gerard et al., 2012), so dass davon
ausgegangen werden kann, dass die fehlende Rhythmogenese in den
meisten Blutegel-Neuronen nicht auf die Eigenschaften dieses Ionenkanals
zurckzufhren sind. Mglicherweise ist jedoch die Abwesenheit von IT dafr
verantwortlich. Aus diesen Ergebnissen lsst sich schlussfolgern, dass IT das
Ausma der Erzeugung rhythmisch-oszillatorischer elektrischer Aktivitt in
Neuronen mageblich mitbestimmt.

Zusammenfassend kann festgehalten werden, dass fr die Rhythmogenese


offenbar ein Zusammenspiel zwischen mindestens zwei Ionenkanaltypen
erforderlich ist. Im vorliegenden Fall fhrt die Expression der HCN-Kanle
allein zu einer Stabilisierung des Membranpotentials. Erst beim gemeinsamen auftreten mit T-Typ Ca2+-Kanlen wird rhythmische Aktivitt beobachtet. Ein ganz hnliches Phnomen lsst sich bei NRT-Neuronen beobachten,
wo in Dendriten ein Zusammenspiel zwischen IT und einem Strom durch
Ca2+-aktivierte K+-Kanle fr die Rhythmogenese verantwortlich ist (Cueni et
al., 2008) (siehe folgendes Kapitel).

21

Eigene Arbeiten

2.2. Intrazellulres Calcium beeinflusst elektrophysiologische


Aktivittsmuster in thalamischen Neuronen

[COULON P, HERR D, KANYSHKOVA T, MEUTH P, BUDDE T, PAPE HC (2009):


Burst discharges in neurons of the thalamic reticular nucleus are shaped
by calcium-induced calcium release. Cell Calcium, 46, 333-346.]

[RANKOVIC V, EHLING P, LANDGRAF P, COULON P, KREUTZ MR, MUNSCH T,


AND

BUDDE T (2010): Intracellular Ca2+ release-dependent inactivation of

Ca2+ currents in thalamocortical relay neurons. Eur J Neurosci., 31(3),


439 - 449.]

Ca2+ und die Vernderung seiner intrazellulren Konzentration ist ein


wichtiger Signalgeber fr zahlreiche intrazellulre Prozesse in Nervenzellen
(Berridge, 1998; Berridge et al., 2000; Pape et al., 2004). Whrend
schlafrelevanter rhythmisch-oszillatorischer Salvenaktivitt in TC und NRTNeuronen spielt Ca2+ eine wichtige Rolle (Budde et al., 2000; Cueni et al.,
2008; Coulon et al., 2009). Dies ist wenig verwunderlich, da Ca2+ als
intrazellulrer Botenstoff zahlreiche Funktionen erfllt, unter anderem die
Regulation der Freisetzung von Neurotransmittern in der chemisch-synaptischen bertragung, neuronaler Erregbarkeit und synaptischer Plastizitt
(Berridge, 1998).

22

Eigene Arbeiten

Zu den wichtigsten Quellen fr Zunahmen der [Ca2+]i gehren spannungsgesteuerte Ca2+-Kanle, die ber die Plasmamembran einen Einstrom von
Ca2+ aus dem Extrazellulrraum in das Cytosol ermglichen (Cain & Snutch,
2010). Dabei unterscheidet man grob zwischen hochschwelligen Ca2+Kanlen (engl. high voltage activated, HVA) und niederschwelligen Ca2+Kanlen (engl. low voltage activated, LVA).

HVA-Ca2+-Kanle, zu denen neben den L-Typ-Kanlen auch solche des Typs


P/Q, N und R gehren, werden bei Potentialen positiv von ca. -45 mV
aktiviert und inaktivieren nur partiell. LVA-Ca2+-Kanle, zu denen bislang nur
die T-Typ Ca2+-Kanle gezhlt werden, aktivieren bei Potentialen positiv von
ca. -100 mV, jedoch inaktivieren sie schnell und nahezu vollstndig (Hille,
2001). Der T-Typ Ca2+-Kanal ist fr das bistabile Membranpotential (siehe
Kap. 1.1.) thalamischer Neuronen mitverantwortlich.

Eine

weitere

wichtige

Quelle

fr

Erhhungen

der

[Ca2+]i

ist

das

endoplasmatische Retikulum (ER), das ber Ca2+-ATPasen, wie z. B. die


SERCA (engl. sarco-/endoplasmic reticulum Ca2+ ATPase) mit Ca2+ aus
dem Cytosol gefllt wird. Als Botenstoff, sog. second messenger, kann Ca2+
selbst ber die Aktivierung von Ryanodin-Rezeptoren (RyR) und Rezeptoren
fr Inositol-1,4,5-trisphosphat (IP3-Rezeptoren, IP3R) seine Freisetzung aus
diesen intrazellulren Speichern bewirken (Kostyuk & Verkhratsky, 1994).

23

Eigene Arbeiten

Dieser Prozess wird als Ca2+-induzierte Ca2+-Freisetzung bezeichnet (engl.


calcium-induced calcium-release, CICR).

Fr TC Neuronen des CGLd konnte bereits gezeigt werden, dass ihr


Aktivittsmuster von CICR beeinflusst wird (siehe Abb. 3) (Budde et al.,
2000).

2+

2+

Abb. 3: Ca -induzierte Ca -Freisetzung stabilisiert tonische Aktivitt in TC Neuronen. Als


mgliches Zielprotein fr das freigesetzte Ca

2+

2+

wurden Ca -abhngige K -Kanle des BK-

Typs vermutet (Budde et al., 2000). Modifiziert nach Coulon et al. (2012).

24

Eigene Arbeiten

ber Ryanodin-Rezeptoren fhrt der Ca2+-Einstrom durch HVA-Ca2+-Kanle


zu einer Freisetzung von Ca2+ aus intrazellulren Speichern. Dies fhrt zu
einer Rckkopplung, die die Neuronen fr tonische Aktionspotentialgenerierung stabilisiert und so schlafrelevante Aktivitt unterdrckt. Als
zugrundeliegender

Mechanismus

wurde

die

Aktivierung

von

Ca 2+-

abhngigen K+-Kanlen des BK-Typs vermutet (Budde et al., 2000).

Der T-Typ Ca2+-Strom ist in NRT-Neuronen besonders stark ausgeprgt und


lang anhaltend (Broicher et al., 2007, 2008a). Dahingegen ist der Ih
verhltnismig schwach ausgeprgt (Santoro et al., 2000). In NRTNeuronen findet das oszillatorische Wechselspiel daher zwischen I T und
einem durch den Ca2+-Einstrom aktivierten K+-Kanal statt (Cueni et al.,
2008). NRT-Neuronen bertragen whrend Schlafspindel-Aktivitt rhythmisch
GABAerge inhibitorische postsynaptische Potentiale (IPSPs) auf TC
Neuronen. Die dadurch ausgelsten Hyperpolarisationen de-inaktivieren IT,
whrend parallel der hyperpolarisationsaktivierte Ih aktiviert wird. Die
nachfolgende, durch den Ih ausgelste Depolarisation aktiviert dann den
zuvor de-inaktivierten IT. Dies verursacht einen sog. rebound-burst in Form
von Salvenaktivitt. Rhythmisch wiederkehrende IPSPs aktivieren ebenfalls
den Ih, wodurch wieder IT aktiviert werden kann. Alternativ kann die Ihbedingte Depolarisation des Membranpotentials so stark sein, dass alle
Ionenkanle de- oder inaktiviert werden, so dass die Spindelaktivitt zum
Erliegen kommt (Bal & McCormick, 1996).
25

Eigene Arbeiten

Bislang konnte ein CICR im NRT nicht gezeigt werden (Richter et al., 2005;
Cueni et al., 2008). Mit Hilfe der whole-cell patch-clamp Technik und der 2Photonen-Laserscanning-Fluoreszenzmikroskopie mit Ca2+-Indikatorfarbstoffen konnte gezeigt werden, dass eine vergleichbare Freisetzung im NRT
durch rhythmisch-oszillatorische Aktivitt erzielt wird. Der dafr erforderliche
Ca2+-Einstrom findet ber T-Typ Ca2+-Kanle statt. Die Freisetzung von Ca2+
findet, hnlich wie im dorsalen Thalamus, ber Ryanodin-Rezeptoren statt
(Abb. 4). Dieser CICR ber Ryanodin-Rezeptoren erfolgt im NRT hauptschlich im Bereich des Zellsomas (Coulon et al., 2009). Das dabei
freigesetzte Ca2+ konnte durch einen hier ebenfalls noch ungeklrten Mechanismus oszillatorisch ausgelste LTS frequenzabhngig unterdrcken und so
die Fhigkeit der NRT-Neuronen, einer vorgegebenen Frequenz zu folgen,
behindern (Coulon et al., 2009). In der Folge bedeutet dies einen Verlust der
Fhigkeit zur

Generierung von

Salvenaktivitt

und

eine

Dmpfung

stereotyper Entladungsmuster. Eine pharmakologische Blockierung der


SERCAs fhrt auf lange Sicht durch die verbleibende Hintergrundaktivitt der
Ryanodin- und IP3-Rezeptoren zur Entleerung der intrazellulren Ca2+Speicher und erlaubt so eine Verbesserung der Fhigkeit von NRTNeuronen, einer vorgegebenen Frequenz zu folgen (Coulon et al., 2009).

Anhand dieser Erkenntnisse wird klar, dass sowohl L-Typ-, als auch T-Typ
Ca2+-Kanle zelltypspezifisch einen CICR auslsen knnen. Obwohl dies auf
gnzlich unterschiedliche Weise geschieht und auf gegenstzliche zellulre
26

Eigene Arbeiten

Aktivittsmuster Einfluss nimmt, hat es im Netzwerk des thalamokortikalen


Systems einen gemeinsamen Effekt: L-Typ-vermittelter CICR in TCNeuronen stabilisiert den tonischen Modus, T-Typ-vermittelter CICR in NRTNeuronen vermindert gleichzeitig Salvenaktivitt. Beides fhrt letztlich zu
einer Unterdrckung schlaf- und epilepsierelevanter Salvenaktivitt.

Abb. 4: Calcium-induzierte Calcium Freisetzung hemmt Salvenaktivitt in NRT Neuronen.


2+

Dabei wird eine Aktivierung von Ca -abhngigen K -Kanlen vermutet, die zu den
Membranpotential-Oszillationen phasenverschoben ist (Coulon et al., 2009). Modifiziert nach
Coulon et al. (2012).

27

Eigene Arbeiten

Ein Grund fr diese unterschiedlichen Reaktionen in NRT und TC Neuronen


mag in der unterschiedlichen subzellulren Lokalisation der entsprechenden
Ionenkanle zu finden sein. So konnte gezeigt werden, dass in Dendriten
und im Soma von NRT-Neuronen unterschiedliche T-Typ Ca2+-Kanle
exprimiert werden (Joksovic et al., 2005a), dass Ryanodin-Rezeptoren
jedoch hauptschlich in der Nhe des Zellkerns vorkommen (Coulon et al.,
2009). Auerdem unterscheidet sich die Kontrolle der [Ca2+]i in NRTNeuronen von der in TC-Neuronen: In NRT-Neuronen fhrt die Blockierung
der SERCAs zu einer Vernderung der [Ca2+]i, whrend dies in TC-Neuronen
nicht der Fall ist (Budde et al., 2000; Coulon et al., 2009). Dies lsst den
Schluss zu, dass NRT-Neuronen fr die zellulre Ca2+-Homostase auf das
ER und SERCAs angewiesen sind, whrend TC-Neuronen ber andere
Ca2+-Puffersysteme verfgen. Dies wurde als eine Mglichkeit zur differenzierten Kontrolle der Kopplung der Ca2+-Freisetzung an einen Effektor
betrachtet (engl. Ca2+-release-to-effector-coupling, (Coulon et al., 2012)),
wobei die Effektoren noch unbekannt sind. Im Falle der TC-Neuronen ist es
mglicherweise ein Ca2+-aktivierter K+-Kanal des BK-Typs (Budde et al.,
2000). In NRT-Neuronen wird eine Beteiligung eines Ca2+-aktivierten K+Kanals des SK2-Typs ausgeschlossen (Coulon et al., 2009). Ein mglicher
Effektor wre hier neben BK-Kanlen ein Ca2+-aktivierter Kationenkanal, der
in diesen Zellen bereits beschrieben wurde (Bal & McCormick, 1993;
Destexhe et al., 1994) und einen nicht-selektiven Kationenstrom (engl. nonselective cation conductance, ICAN) trgt. In NRT-Neuronen knnten die
28

Eigene Arbeiten

beschriebenen Ryanodin-Rezeptoren dazu dienen, stereotype Entladungsmuster zu detektieren und ber die dauerhafte Aktivierung Ca2+-aktivierter
Ionenkanle zu unterbinden (Coulon et al., 2012).

ber die intrazellulre Ca2+-Konzentration knnte auerdem eine Reihe


weiterer Prozesse in Gang gebracht werden, die Aktivittswechsel zwischen
Schlaf- und Wachzustand regulieren und pathophysiologische Hypersynchronizitt (wie z. B. Epilepsie) unterbinden knnten. So werden die im NRT
nachgewiesenen elektrischen Synapsen von hohen [Ca 2+]i inhibiert (Decrock
et al., 2011).

Zusammenfassend lsst sich festhalten, dass der NRT aufgrund seiner


Eigenschaften dafr geeignet zu sein scheint, synchrone Erregung im
thalamokortikalen System einzudmmen. Es ist daher naheliegend, dass
dieser Nucleus Mechanismen besitzt, die dazu dienen knnen, oszillatorische Aktivitt

bei

Bedarf

zu

unterbinden.

Im

Rahmen

der

hier

beschriebenen Arbeiten konnte erstmals demonstriert werden, dass dies


ber intrazellulres Ca2+ erfolgen kann.

Einige spannungsgesteuerte Ca2+-Kanle werden durch hohe intrazellulre


Ca2+-Konzentrationen inhibiert, ein Mechanismus der als Ca2+-abhngige
Inaktivierung bezeichnet wurde (engl. Ca2+-dependent inactivation, CDI)
(Budde et al., 2002). Es wurde zunchst angenommen, dass durch die
29

Eigene Arbeiten

Kanle einstrmendes Ca2+ seinen eigenen Einstrom hemmt. Es konnte


jedoch spter gezeigt werden, dass auch Ca2+ aus intrazellulren Speichern
eine solche CDI bewirken bzw. untersttzen kann (Rankovic et al., 2010).
Dabei wurde zunchst CICR durch die Applikation von Koffein ausgelst,
wodurch bereits HVA-Ca2+-Kanle gehemmt wurden. Diese CDI war
reduziert, wenn Ryanodin-Rezeptoren oder die SERCA inhibiert waren. Eine
Hemmung der IP3-Rezeptoren hatte jedoch nicht diesen Effekt. Wurde Ca2+
durch

Ba2+

ersetzt

oder

durch

den

Ca2+-Chelator

1,2-Bis(2-

aminophenoxy)ethan-N,N,N,N-tetraacetat (BAPTA) komplexiert, so wurde


CDI ebenfalls unterdrckt. Offenbar ermglicht dieses CICR-vermittelte
Rckkopplungssystem die Begrenzung eines Ca2+-Einstroms ber HVACa2+-Kanle. Whrend neuronaler Aktivitt findet somit eine Rckkopplung
zwischen Ca2+-Einstrom und den intrazellulren Speichern statt.

30

Eigene Arbeiten

2.3. Aktivierung thalamischer Afferenzen: bertragung an den Kortex

[BROICHER T, BIDMON HJ, KAMUF B, COULON P*, GORJI A, PAPE HC,


SPECKMANN EJ, BUDDE T (2010): Thalamic afferent activation of
supragranular layers in auditory cortex in vitro: a voltage sensitive dye
study. Neuroscience, 165, 371-385. *(corresponding author)]

Fr die Entstehung der rhythmischen EEG-Aktivitt, die bei Schlaf und


Epilepsie registriert werden kann, ist die Kommunikation zwischen Thalamus,
NRT und Kortex von entscheidender Bedeutung. Um Vernderungen in
dieser Kommunikation untersuchen zu knnen, aber auch um Auswirkungen
vernderter elektrophysiologischer Eigenschaften der beteiligten Neuronen
auf das gesamte Netzwerk untersuchen zu knnen, ist idealerweise ein
Prparat zu nutzen, bei dem die thalamischen Afferenzen intakt sind. Hierfr
wurden bereits einige Anstze etabliert, sowohl fr somatosensorische
Bahnen (Agmon & Connors, 1991), als auch fr auditorische (Cruikshank et
al., 2002) und visuelle (MacLean et al., 2006). In einem solchen Prparat
fhrt eine elektrische Stimulation thalamischer Neuronen zu einer Aktivittsnderung in den neuronalen Netzwerken des Kortex. Zahlreiche Arbeiten
hatten zum Ziel, ein solches Prparat zu etablieren und zu nutzen (Rose &
Metherate, 2001, 2005; Cruikshank et al., 2002; Kaur et al., 2005; Kotak et
al., 2008; Lee & Sherman, 2008). Auf der Basis dieser Anstze wurden
auditorische thalamokortikale Signale hinsichtlich der rumlichen und
31

Eigene Arbeiten

zeitlichen Ausbreitung in Abhngigkeit des Alters der Versuchstiere


untersucht (Broicher et al., 2010). Dafr wurden Hirnschnitte von Musen
angefertigt, die um 25 von der Horizontalebene geneigt waren. Dadurch
blieben die auditorischen Afferenzen zwischen Thalamus und Kortex intakt.
Die elektrische Stimulation im Nucleus geniculatum mediale fhrte zur
Auslsung unterschiedlicher Erregungsmuster in den untersuchten auditorischen Arealen des Kortex (primrer auditorischer Kortex, A1; sekundrer
auditorischer Kortex, AuV; Assoziative Areale des Temporallappens, TeA).
Auch zwischen den kortikalen Schichten lieen sich komplexe Unterschiede
feststellen (infragranulr, I; granulr, G; supragranulr, S). Diese Muster
wurden mit einem spannungssensitiven Farbstoff registriert, der in die
Zellmembran der Neuronen gebracht wurde (Khling et al., 2000, 2002;
Straub et al., 2003; Broicher & Speckmann, 2012). Es konnte gezeigt
werden, dass bei ausgewachsenen Musen die thalamischen Afferenzen
zunchst die supragranulren und granulren Schichten erregen, wobei
diese Erregung hauptschlich ber non-NMDA-Rezeptoren erfolgte. GABAARezeptor-vermittelte Inhibition begrenzte dabei sowohl die Amplitude der
Antworten, als auch deren horizontale und vertikale Ausbreitung. Die
bevorzugte Ausbreitung der Signale war in horizontaler Richtung. Die grte
Amplitude

der Antworten

wurde

in

den

supragranulren

Schichten

gemessen, von wo aus die Amplitude mit zunehmender Entfernung von der
Pia mater encephali abnahm. Entwicklungsabhngig nahm die Erregung der
infragranulren Schichten zunehmend ab, so dass diese in jngeren Tieren
32

Eigene Arbeiten

strker aktiviert waren als bei ausgewachsenen Tieren. Zwischen einem Alter
von 31 und 64 Tagen postnatal (p31 bis p64) nahm die Aktivitt in den
infragranulren Schichten kontinuierlich ab und war ab einem Alter von ca.
65 Tagen (p65) kaum mehr messbar. Eine Stimulation mit hoher Frequenz
hatte vergleichsweise geringe Effekte im auditorischen Kortex zur Folge, im
Gegensatz zum somatosensorischen Kortex, wo durch Stimulation des VB
starke Aktivierung erzielt werden konnte (Laaris et al., 2000; Beierlein et al.,
2002; Llinas et al., 2002). Auch die Latenzen waren in supragranulren und
granulren Schichten krzer als in den infragranulren. Die Ergebnisse zur
Amplitude und Latenz stimmen gut mit den anatomisch-morphologischen
Beschreibungen der Terminalien thalamokortikaler Synapsen aus dem
ventralen Teil des Nucleus geniculatum mediale berein (Linden & Schreiner,
2003; Winer & Lee, 2007). Whrend die geringe Amplitude ein Artefakt der
geringeren Neuronendichte sein knnte, deuten die langen Latenzen in den
infragranulren Schichten auf eine intrakortikale Aktivierung hin. Die laterale
Ausbreitung in den supragranulren und granulren Schichten war zudem
nicht von der infragranulren Aktivitt abhngig, so dass davon ausgegangen
werden kann, dass diese Ausbreitung ber die hheren Schichten luft.
Infragranulre Schichten allein zeigten keine laterale Ausbreitung. Wenn die
ausgelste Aktivitt sich ber die Grenzen des A1 ausbreitete, dann waren
die krzesten Latenzen ausschlielich in den supragranulren Schichten des
A1 zu finden. Im AuV und TeA waren die krzesten Latenzen in den
granulren Schichten zu finden, whrend die grten Amplituden ebenfalls in
33

Eigene Arbeiten

den supragranulren vorkamen. Dieses spezifische rumlich-zeitliche Muster


der Erregungsentstehung deutet auf eine Fokussierung der thalamokortikalen Projektionen auf die granulren und supragranulren Schichten
des primren auditorischen Kortex hin, mit einer nachfolgend starken horizontalen Ausbreitung.

Durch spannungssensitive Farbstoffe kann erfolgreich die rumliche und


zeitliche Erregungsausbreitung thalamokortikaler Projektionen im Kortex
untersucht werden. In zuknftigen Arbeiten kann diese Methode, mit der
neuronale Membranpotentiale und nicht mehrdeutige, sekundr induzierte
Feldpotentiale gemessen werden, Verwendung finden, um die Einflsse
vernderter elektrophysiologischer Eigenschaften einzelner thalamischer
Neuronen auf die thalamokortikalen Projektionen im Epilepsiemodell zu
untersuchen.

34

Eigene Arbeiten

2.4. Schlaf, Narkose und Epilepsie: Gemeinsame Mechanismen

[BUDDE T, COULON P*, PAWLOWSKI M, MEUTH P, MEUTH SG, PAPE HC


(2008): Reciprocal modulation of Ih and ITASK in thalamocortical relay
neurons by halothane. Pflgers Arch. 456(6), 1061-73. *(T.B. and P.C. are
equally contributing first authors)]

Schlaf, Narkose und Bewusstlosigkeit werden hufig intuitiv miteinander


verglichen, obwohl das EEG whrend der klinischen Narkose weder Stadien,
noch die charakteristischen Zyklen zeigt, die whrend des natrlichen
Schlafs auftreten. Umgekehrt lsst sich whrend des natrlichen Schlafs kein
isoelektrisches EEG registrieren, welches jedoch bei tiefer Narkose auftreten
kann. Trotzdem gibt es deutliche Parallelen zwischen der Narkose und
natrlichem Schlaf. Die Aktivittsnderungen, die Ansthetika an ihren
spezifischen molekularen Wirkorten auslsen, hneln denen, die bei
natrlichem Schlaf auftreten (Nelson et al., 2002). Die berwiegende
Mehrheit dieser molekularen Wirkorte sind Ionenkanle, die durch klinisch
relevante

Konzentrationen

volatiler

Ansthetika

moduliert

werden

(Campagna et al., 2003). Diese volatilen Ansthetika, wie z. B. Isofluran,


Sevofluran oder das heute klinisch nicht mehr verwendete Halothan,
verstrken einerseits die durch GABAA- und Glycin-Rezeptoren vermittelten
IPSPs (Mihic et al., 1997), andererseits hemmen sie exzitatorische

35

Eigene Arbeiten

Ionenkanle, wie z. B. Rezeptoren fr die Neurotransmitter Acetylcholin,


Serotonin und Glutamat (Franks & Lieb, 1994, 1998).

In Neuronen des ventrobasalen Komplexes (VB) konnte solch ein


Mechanismus nach Gabe des intravensen Ansthetikums Pentobarbital
identifiziert werden (Wan et al., 2003). Dabei wurde Ih inhibiert und K+-Kanle
(IKir

und

IKL)

aktiviert.

Gleichzeitig

wurde

dort

gezeigt,

dass

die

Offenwahrscheinlichkeit von GABAA Cl--Kanlen erhht wurde. hnliche


Effekte wurden durch die Applikation des Ansthetikums Propofol erzielt
(Ying & Goldstein, 2005). Auch der T-Typ Ca2+-Kanal wird durch eine Reihe
von Ansthetika moduliert (Soltesz et al., 1991; Todorovic et al., 1998;
Joksovic et al., 2005a, 2005b; Joksovic & Todorovic, 2010). In TC Neuronen
fhrte die Applikation des ansthetischen Alkohols 1-Octanol zu einer
reversiblen Blockade der Salvenaktivitt und reduzierte den Eingangswiderstand von Neuronen, was letztlich durch Kurzschluss der Membranstrme eine Abschwchung synaptischer Eingnge verursacht (KurzschlussHemmung, engl. shunting inhibition) (Soltesz et al., 1991). Eine solche
Kurzschluss-Hemmung war die Hauptursache fr die Unterdrckung der
tonischen Aktivitt in TC Neuronen nach Applikation von Isofluran (Ries &
Puil, 1999a, 1999b). Isofluran und 1-Octanol verursachten hnliche Effekte in
NRT-Neuronen: Isofluran reduzierte die Erregbarkeit (Joksovic & Todorovic,
2010) und 1-Octanol hemmte T-Typ Ca2+-Kanle (Joksovic et al., 2010).

36

Eigene Arbeiten

In TC Neuronen wurde eine reziproke Wirkung klinisch relevanter


Konzentrationen volatiler Ansthetika auf zwei Ionenkanaltypen nachgewiesen (Budde et al., 2008). Einerseits wurde Ih inhibiert, wobei auch die
spannungsgesteuerte Aktivierung zu negativeren Potentialen hin verschoben
wurde, andererseits wurden K+-Kanle des TASK-Typs aktiviert. Beide
Effekte fhren in einer sich gegenseitig verstrkenden Weise zu einer
Hyperpolarisation des Ruhemembranpotentials. Diese Hyperpolarisation
verhindert wiederum tonische Aktivitt und knnte eine mgliche Erklrung
fr das Fehlen bewusster sensorischer Wahrnehmung whrend der
Ansthesie darstellen. Da diese Ionenkanle im Thalamus in groer Zahl zu
finden sind, ist deren Modulation wahrscheinlich ein essentieller Teil der
klinischen Wirkung volatiler Ansthetika (Budde et al., 2008). Ob ein
hnlicher Mechanismus whrend des natrlichen Schlafs eine Rolle spielt, ist
bislang unklar.

Obwohl Schlaf und Ansthesie weiterhin als unterschiedliche Zustnde des


Gehirns betrachtet werden mssen, sind die zugrundeliegenden Mechanismen mglicherweise hnlich und knnten schlussendlich hnliche nderungen auf molekularer Ebene auslsen. Daher wurde die Applikation klinisch
relevanter Konzentrationen volatiler Ansthetika im Hirnschnitt als Modellsystem fr die Untersuchung der Schlafinduktion vorgeschlagen (Coulon et
al., 2012).

37

Eigene Arbeiten

2.5. Die Rolle von G-Proteinen bei der Signalverarbeitung im


Thalamus

[BROICHER T, WETTSCHURECK N, MUNSCH T, COULON P, MEUTH SG,


KANYSHKOVA T, SEIDENBECHER T, OFFERMANNS S, PAPE HC, BUDDE T
(2008): Muscarinic ACh receptor-mediated control of thalamic activity via
Gq/G11-family G-proteins. Pflgers Arch., 456(6), 1049-60.]

[COULON P, KANYSHKOVA T, BROICHER T, MUNSCH T, W ETTSCHURECK N,


SEIDENBECHER T, MEUTH SG, OFFERMANNS S, PAPE HC, BUDDE T (2010):
Activity modes in thalamocortical relay neurons are modulated by G q/G11
family G-proteins. Front Cell Neurosci., 4:132.]

Zahlreiche Neurotransmitter lsen einen Aktivittswechsel von TC Neuronen


aus. Diese Neurotransmitter werden von Synapsen ausgeschttet, deren
Neuronen in cholinergen Kernen des Hirnstamms (Acetylcholin), im Locus
coeruleus (Noradrenalin) oder den Raphe-Kernen (Serotonin) liegen. In
hnlicher Weise wirkt Glutamat, das von corticothalamischen Axonterminalien ausgeschttet wird und schlielich mehrheitlich an metabotrope
Glutamatrezeptoren bindet (Salt, 2002). Diese Neurotransmitter haben
allesamt eine depolarisierende Wirkung auf das Membranpotential von TC
Neuronen, wodurch Burst-Aktivitt unterdrckt und tonische Aktivitt
gefrdert wird. Diese Neurotransmitter-Systeme scheinen allesamt auf G38

Eigene Arbeiten

Proteine der Gq/G11-Familie zu konvergieren, was sich aus der Tatsache


schlieen lsst, dass in genetischen knock-out Musen, in denen dieses GProtein fehlt, die Wirkung dieser Neurotransmitter stark reduziert ist.
Auerdem konnte gezeigt werden, dass die Applikation eines beliebigen
Transmitters aus dieser Gruppe die Wirkung eines nachfolgend zustzlich
applizierten beliebigen anderen Transmitters begrenzt (Coulon et al., 2010).
Hieraus entstand die Idee, dass G-Proteine der Gq/G11-Familie als
Flaschenhals in der Erregung des Thalamus fungieren (Broicher et al.,
2008b; Coulon et al., 2012). Obwohl also unterschiedliche Neurotransmitter
unterschiedliche Rezeptortypen aktivieren, begrenzen die gemeinsam
genutzten G-Proteine die mgliche gleichzeitige Gesamtwirkung aller
getesteten Neurotransmitter. Das bedeutet, dass diese G-Proteine ein
wichtiges modulatorisches Zentrum darstellen, dass in Zukunft fr die
Entwicklung therapeutischer Anstze grere Beachtung finden sollte.

39

Zusammenfassung

3.

Zusammenfassung

Die in der vorliegenden Arbeit referierten Publikationen liefern neue Erkenntnisse bezglich der Rhythmogenese des thalamokortikalen Systems. Dabei
wurden

sowohl

elektrophysiologische

Ableitungen

von

Ionenstrmen

untersucht, als auch Vernderungen intrazellulrer Ca2+-Ionenkonzentrationen, die durch diese Ionenstrme bedingt waren.

Dem hyperpolarisationsaktivierten Kationenstrom Ih kommt bei Oszillationen


des Membranpotentials eine wichtige Bedeutung zu. Die Eigenschaften des
Ih bei sensorischen Neuronen des medizinischen Blutegels entsprachen
weitgehend denen in Vertebraten-Neuronen. Die fehlende Rhythmogenese in
den meisten Blutegel-Neuronen knnte folglich auf die Abwesenheit von IT
zurckzufhren sein, der mglicherweise das Ausma der Rhythmogenese
in Neuronen mageblich mitbestimmt. Dieser IT konnte eine Freisetzung
intrazellulrer Ca2+-Ionen in Neuronen des Nucleus reticularis thalami
auslsen. Darber hinaus konnte eine Auswirkung auf die mgliche
Frequenz der Salvenaktivitt gezeigt werden. Eine weitere wichtige Rolle der
Ca2+-Freisetzung aus intrazellulren Speichern findet sich in der Hemmung
spannungsgesteuerter Ca2+-Kanle. Offenbar ermglicht dieses Rckkopplungssystem die Begrenzung eines Ca2+-Einstroms ber hochschwellige
Ca2+-Kanle u. a. durch eine Ca2+-Freisetzung aus intrazellulren Speichern
whrend neuronaler Aktivitt. Fr die Entstehung rhythmischer Aktivitt ist die
40

Zusammenfassung

Kommunikation zwischen thalamischen Relaiskernen, NRT und Kortex von


entscheidender Bedeutung. Durch spannungssensitive Farbstoffe kann diese
Kommunikation

erfolgreich

bezglich

der

rumlichen

und

zeitlichen

Erregungsausbreitung thalamokortikaler Projektionen im Kortex untersucht


werden. In zuknftigen Arbeiten kann diese Methode Verwendung finden, um
die Einflsse vernderter elektrophysiologischer Eigenschaften einzelner
thalamischer Neuronen auf die thalamokortikalen Projektionen im Epilepsiemodell zu untersuchen. Auditorische thalamokortikale Signale zeigen
entwicklungsabhngige, spezifische rumlich-zeitliche Muster der Erregungsentstehung, die auf eine Fokussierung der thalamokortikalen Projektionen
auf die granulren und supragranulren Schichten des primren auditorischen Kortex hindeuten, mit einer nachfolgend starken horizontalen
Ausbreitung. Diese aufsteigenden Signale sind bei Schlaf aber auch whrend
der Narkose offensichtlich unterbunden. In TC Neuronen wurde nachgewiesen, dass volatile Ansthetika in klinisch relevanten Konzentrationen zwei
Ionenkanaltypen reziprok modulieren. Im Zusammenspiel fhrt dies zu einer
Unterdrckung tonischer Aktivitt und somit zur Unterdrckung der
Weiterleitung sensorischer Informationen an den Kortex. Da diese Ionenkanle im Thalamus in groer Zahl zu finden sind, ist deren Modulation
wahrscheinlich ein essentieller Teil der klinischen Wirkung volatiler
Ansthetika. Volatile Ansthetika verhalten sich demzufolge antagonistisch
zu den Neurotransmittern, die von Hirnstamm-Neuronen auf TC Neuronen
ausgeschttet werden. Die Neurotransmitter-Rezeptoren der TC Neuronen
41

Zusammenfassung

konvergieren offenbar gemeinsam auf G-Proteine der Gq/G11-Familie, so


dass die Ausschttung eines Transmitters die Wirkung eines nachfolgend
ausgeschtteten Transmitters begrenzt. G-Proteine der Gq/G11-Familie
scheinen somit als Flaschenhals in der Physiologie des Thalamus zu
fungieren.

Die hier beschriebenen Arbeiten zeigen Mechanismen auf, die zu Vernderungen der Aktivittsmuster von Neuronen im thalamokortikalen System
beitragen. Hierdurch wurde es ermglicht, in einem umfangreichen bersichtsartikel das neue Prinzip der Koexistenz zentraler und dezentraler
Schlafmechanismen zu beschreiben.

[COULON P, BUDDE T, PAPE HC (2012): The sleep relay - the role of the
thalamus in central and de-central sleep regulation. Pflgers Arch. 463(1),
53-71.]

42

Literaturverzeichnis

Literaturverzeichnis
Agmon A & Connors BW (1991). Thalamocortical responses of mouse
somatosensory (barrel) cortex in vitro. Neuroscience 41, 365379.
Amzica F & Steriade M (1995). Disconnection of intracortical synaptic
linkages disrupts synchronization of a slow oscillation. J Neurosci 15,
46584677.
Bal T & McCormick DA (1993). Mechanisms of oscillatory activity in guineapig nucleus reticularis thalami in vitro: A mammalian pacemaker. J
Physiol (Lond) 468, 669691.
Bal T & McCormick DA (1996). What stops synchronized thalamocortical
oscillations? Neuron 17, 297308.
Beierlein M, Fall CP, Rinzel J & Yuste R (2002). Thalamocortical bursts
trigger recurrent activity in neocortical networks: layer 4 as a frequencydependent gate. J Neurosci 22, 98859894.
Berger H (1929). ber das Elektrenkephalogramm des Menschen. Arch
Psychiat Nervenkr 87, 527570.
Berridge MJ (1998). Neuronal calcium signaling. Neuron 21, 1326.
Berridge MJ, Lipp P & Bootman MD (2000). The versatility and universality of
calcium signalling. Nat Rev Mol Cell Biol 1, 1121.
Broicher T, Bidmon HJ, Kamuf B, Coulon P, Gorji A, Pape HC, Speckmann
EJ & Budde T (2010). Thalamic afferent activation of supragranular
layers in auditory cortex in vitro: a voltage sensitive dye study.
Neuroscience 165, 371385.
Broicher T, Kanyshkova T, Meuth P, Pape HC & Budde T (2008a).
Correlation of T-channel coding gene expression, IT, and the low
threshold Ca2+ spike in the thalamus of a rat model of absence
epilepsy. Mol Cell Neurosci 39, 384399.
Broicher T, Seidenbecher T, Meuth P, Munsch T, Meuth SG, Kanyshkova T,
Pape HC & Budde T (2007). T-current related effects of antiepileptic
drugs and a Ca2+ channel antagonist on thalamic relay and local circuit
interneurons in a rat model of absence epilepsy. Neuropharmacology 53,
431446.
43

Literaturverzeichnis

Broicher T & Speckmann EJ (2012). Living human brain slices: network


analysis using voltage sensitive dyes. In Neuromethods, Isolated neural
circuits, ed. Ballanyi K. Springer.
Broicher T, Wettschureck N, Munsch T, Coulon P, Meuth SG, Kanyshkova T,
Seidenbecher T, Offermanns S, Pape HC & Budde T (2008b).
Muscarinic ACh receptor-mediated control of thalamic activity via
G(q)/G(11)-family G-proteins. Pflgers Arch 456, 10491060.
Budde T, Coulon P, Pawlowski M, Japes A, Meuth P, Meuth SG & Pape HC
(2008). Reciprocal modulation of Ih and ITASK in thalamocortical relay
neurons by halothane. Pflgers Arch 456, 10611073.
Budde T, Meuth S & Pape H (2002). Calcium-dependent inactivation of
neuronal calcium channels. Nat Rev Neurosci 3, 873883.
Budde T, Sieg F, Braunewell K, Gundelfinger E & Pape H (2000). Ca2+induced Ca2+ release supports the relay mode of activity in
thalamocortical cells. Neuron 26, 483492.
Cain SM & Snutch TP (2010). Contributions of T-type calcium channel
isoforms to neuronal firing. Channels 4, 475482.
Campagna JA, Miller KW & Forman SA (2003). Mechanisms of actions of
inhaled anesthetics. N Engl J Med 348, 21102124.
Chen Y, Lu J, Pan H, Zhang Y, Wu H, Xu K, Liu X, Jiang Y, Bao X, Yao Z,
Ding K, Lo WH, Qiang B, Chan P, Shen Y & Wu X (2003). Association
between genetic variation of CACNA1H and childhood absence
epilepsy. Ann Neurol 54, 239243.
Cirelli C & Bushey D (2008). Sleep and wakefulness in drosophila
melanogaster. Ann N Y Acad Sci 1129, 323329.
Connors BW & Long MA (2004). Electrical synapses in the mammalian brain.
Annu Rev Neurosci 27, 393418.
Coulon P, Budde T & Pape HC (2012). The sleep relay: the role of the
thalamus in central and decentral sleep regulation. Pflgers Arch 463,
5371.
Coulon P, Herr D, Kanyshkova T, Meuth P, Budde T & Pape HC (2009).
Burst discharges in neurons of the thalamic reticular nucleus are shaped
by calcium-induced calcium release. Cell Calcium 46, 333346.
44

Literaturverzeichnis

Coulon P, Kanyshkova T, Broicher T, Munsch T, Wettschureck N,


Seidenbecher T, Meuth SG, Offermanns S, Pape HC & Budde T (2010).
Activity modes in thalamocortical relay neurons are modulated by
Gq/G11 family G-proteins - serotonergic and glutamatergic signalling.
Front Cell Neurosci 4, 110.
Crick F (1984). Function of the thalamic reticular complex: the searchlight
hypothesis. PNAS 81, 45864590.
Cruikshank SJ, Rose HJ & Metherate R (2002). Auditory thalamocortical
synaptic transmission in vitro. J Neurophysiol 87, 361384.
Crunelli V & Leresche N (2002a). Childhood absence epilepsy: Genes,
channels, neurons and networks. Nat Rev Neurosci 3, 371382.
Crunelli V & Leresche N (2002b). Block of Thalamic T-Type Ca2+ Channels
by Ethosuximide Is Not the Whole Story. Epilepsy Curr 2, 5356.
Cueni L, Canepari M, Lujan R, Emmenegger Y, Watanabe M, Bond CT,
Franken P, Adelman JP & Luthi A (2008). T-type Ca2+ channels, SK2
channels and SERCAs gate sleep-related oscillations in thalamic
dendrites. Nat Neurosci 11, 683692.
Decrock E, Vinken M, Bol M, DHerde K, Rogiers V, Vandenabeele P, Krysko
DV, Bultynck G & Leybaert L (2011). Calcium and connexin-based
intercellular communication, a deadly catch? Cell Calcium 50, 310321.
Destexhe A, Contreras D, Sejnowski TJ & Steriade M (1994). A model of
spindle rhythmicity in the isolated thalamic reticular nucleus. J
Neurophysiol 72, 803818.
Dickinson MH & Lent CM (1984). Feeding behavior of the medicinal leech,
Hirudo medicinalis L. J Comp Physiol [A] 154, 449455.
Dierkes PW, Hochstrate P & Schlue WR (1997). Voltage-dependent Ca2+
influx into identified leech neurones. Brain Res 746, 285293.
Dierkes PW, Wende V, Hochstrate P & Schlue WR (2004). L-type Ca2+
channel antagonists block voltage-dependent Ca2+ channels in
identified leech neurons. Brain Res 1013, 159167.
Franks NP & Lieb WR (1994). Molecular and cellular mechanisms of general
anaethesia. Nature 367, 607614.

45

Literaturverzeichnis

Franks NP & Lieb WR (1998). Which molecular targets are most relevant to
general anaesthesia? Toxicol Lett 100-101, 18.
Fuentealba P & Steriade M (2005). The reticular nucleus revisited: intrinsic
and network properties of a thalamic pacemaker. Progr Neurobiol 75,
125141.
De Gennaro L & Ferrara M (2003). Sleep spindles: an overview. Sleep Med
Rev 7, 423440.
Gerard E, Hochstrate P, Dierkes PW & Coulon P (2012). Functional
properties and cell type specific distribution of I(h) channels in leech
neurons. J Exp Biol 215, 227238.
Gloor P (1978). Generalized epilepsy with bilateral synchronous spike and
wave discharge. New findings concerning its physiological mechanisms.
Electroencephalogr Clin Neurophysiol Suppl 34, 245249.
Guillery RW, Feig SL & Lozsadi DA (1998). Paying attention to the thalamic
reticular nucleus. Trends Neurosci 21, 2832.
Hashemzadeh-Gargari H & Friesen W (1989). Modulation of swimming
activity in the medicinal leech by serotonin and octopamine. Comp
Biochem Physiol C 94, 295302.
Hille B (2001). Ion channels of excitable membranes, 3rd edn. Sinauer
Associates, Sunderland.
Huguenard JR & McCormick DA (2007). Thalamic synchrony and dynamic
regulation of global forebrain oscillations. Trends Neurosci 30, 350356.
Joksovic PM, Bayliss DA & Todorovic SM (2005a). Different kinetic
properties of two T-type Ca2+ currents of rat reticular thalamic neurones
and their modulation by enflurane. J Physiol (Lond) 566, 125142.
Joksovic PM, Brimelow BC, Murbartin J, Perez-Reyes E & Todorovic SM
(2005b). Contrasting anesthetic sensitivities of T-type Ca2+ channels of
reticular thalamic neurons and recombinant Cav3.3 channels. Br J
Pharmacol 144, 5970.
Joksovic PM, Choe WJ, Nelson MT, Orestes P, Brimelow BC & Todorovic
SM (2010). Mechanisms of inhibition of t-type calcium current in the
reticular thalamic neurons by 1-octanol: implication of the protein kinase
C pathway. Mol Pharmacol 77, 8794.
46

Literaturverzeichnis

Joksovic PM & Todorovic SM (2010). Isoflurane modulates neuronal


excitability of the nucleus reticularis thalami in vitro. Ann N Y Acad Sci
1199, 3642.
Kaur S, Rose HJ, Lazar R, Liang K & Metherate R (2005). Spectral
integration in primary auditory cortex: laminar processing of afferent
input, in vivo and in vitro. Neuroscience 134, 10331045.
Khosravani H, Altier C, Simms B, Hamming KS, Snutch TP, Mezeyova J,
McRory JE & Zamponi GW (2004). Gating effects of mutations in the
Cav3.2 T-type calcium channel associated with childhood absence
epilepsy. J Biol Chem 279, 96819684.
Khosravani H & Zamponi G (2006). Voltage-gated calcium channels and
idiopathic generalized epilepsies. Physiol Rev 86, 941966.
Kostyuk P & Verkhratsky A (1994). Calcium stores in neurons and glia.
Neuroscience 63, 381404.
Kotak VC, Takesian AE & Sanes DH (2008). Hearing loss prevents the
maturation of GABAergic transmission in the auditory cortex. Cereb
Cortex 18, 20982108.
Krueger JM, Rector DM, Roy S, Van Dongen HPA, Belenky G & Panksepp J
(2008). Sleep as a fundamental property of neuronal assemblies. Nat
Rev Neurosci 9, 910919.
Khling R, Hohling JM, Straub H, Kuhlmann D, Kuhnt U, Tuxhorn I, Ebner A,
Wolf P, Pannek HW, Gorji A & Speckmann EJ (2000). Optical monitoring
of neuronal activity during spontaneous sharp waves in chronically
epileptic human neocortical tissue. J Neurophysiol 84, 21612165.
Khling R, Reinel J, Vahrenhold J, Hinrichs K & Speckmann EJ (2002).
Spatio-temporal patterns of neuronal activity: analysis of optical imaging
data using geometric shape matching. J Neurosci Methods 114, 1723.
Laaris N, Carlson GC & Keller A (2000). Thalamic-evoked synaptic
interactions in barrel cortex revealed by optical imaging. J Neurosci 20,
15291537.
Lam Y-W, Nelson CS & Sherman SM (2006). Mapping of the functional
interconnections between thalamic reticular neurons using
photostimulation. J Neurophysiol 96, 25932600.

47

Literaturverzeichnis

Landisman CE, Long MA, Beierlein M, Deans MR, Paul DL & Connors BW
(2002). Electrical synapses in the thalamic reticular nucleus. J Neurosci
22, 10021009.
Lee CC & Sherman SM (2008). Synaptic properties of thalamic and
intracortical inputs to layer 4 of the first- and higher-order cortical areas
in the auditory and somatosensory systems. J Neurophysiol 100, 317
326.
Linden JF & Schreiner CE (2003). Columnar transformations in auditory
cortex? A comparison to visual and somatosensory cortices. Cereb
Cortex 13, 8389.
Llinas R & Ribary U (1993). Coherent 40-Hz oscillation characterizes dream
state in humans. PNAS 90, 20782081.
Llinas RR, Leznik E & Urbano FJ (2002). Temporal binding via cortical
coincidence detection of specific and nonspecific thalamocortical inputs:
a voltage-dependent dye-imaging study in mouse brain slices. PNAS 99,
449454.
MacLean JN, Fenstermaker V, Watson BO & Yuste R (2006). A visual
thalamocortical slice. Nat Methods 3, 129134.
McCarley RW, Benoit O & Barrionuevo G (1983). Lateral geniculate nucleus
unitary discharge in sleep and waking: state- and rate-specific aspects. J
Neurophysiol 50, 798818.
McClung CR (2006). Plant circadian rhythms. Plant Cell 18, 792803.
McCormick DA & Bal T (1994). Sensory gating mechanisms of the thalamus.
Curr Opin Neurobiol 4, 550556.
McCormick DA & Pape HC (1990). Properties of a hyperpolarizationactivated cation current and its role in rhythmic oscillation in thalamic
relay neurones. J Physiol (Lond) 431, 291318.
Mihic SJ, Ye Q, Wick MJ, Koltchine VV, Krasowski MD, Finn SE, Mascia MP,
Valenzuela CF, Hanson KK, Greenblatt EP, Harris RA & Harrison NL
(1997). Sites of alcohol and volatile anaesthetic action on GABAA and
glycine receptors. Nature 389, 385389.
Nelson LE, Guo TZ, Lu J, Saper CB, Franks NP & Maze M (2002). The
sedative component of anesthesia is mediated by GABAA receptors in
an endogenous sleep pathway. Nat Neurosci 5, 979984.
48

Literaturverzeichnis

Pape HC (1996). Queer current and pacemaker: the hyperpolarizationactivated cation current in neurons. Annu Rev Physiol 58, 299327.
Pape HC, Meuth SG, Seidenbecher T, Munsch T & Budde T (2005). The
thalamus: the gate to consciousness and rhythm generator in the brain.
Neuroforum 2/05, 4454.
Pape HC, Munsch T & Budde T (2004). Novel vistas of calcium-mediated
signalling in the thalamus. Pflgers Arch 448, 131138.
Pinault D (2004). The thalamic reticular nucleus: structure, function and
concept. Brain Res Brain Res Rev 46, 131.
Raizen DM, Zimmerman JE, Maycock MH, Ta UD, You Y, Sundaram MV &
Pack AI (2008). Lethargus is a Caenorhabditis elegans sleep-like state.
Nature 451, 569572.
Rankovic V, Ehling P, Coulon P, Landgraf P, Kreutz MR, Munsch T & Budde
T (2010). Intracellular Ca2+ release-dependent inactivation of Ca2+
currents in thalamocortical relay neurons. Eur J Neurosci 31, 439449.
Rechtschaffen A & Kales A (1968). A manual of standardized terminology,
techniques and scoring system for sleep stages of human subject. U. S.
National Institute of Neurological Diseases and Blindness, Neurological
Information Network, Bethesda, MD.
Reinoso-Surez F, De Andrs I & Garzn M (2011). Functional anatomy of
the sleep-wakefulness cycle: wakefulness. Adv Anat Embryol Cell Biol
208, 1128.
Richter TA, Kolaj M & Renaud LP (2005). Low voltage-activated Ca2+
channels are coupled to Ca2+-induced Ca2+ release in rat thalamic
midline neurons. J Neurosci 25, 82678271.
Ries CR & Puil E (1999a). Mechanism of anesthesia revealed by shunting
actions of isoflurane on thalamocortical neurons. J Neurophysiol 81,
17951801.
Ries CR & Puil E (1999b). Ionic mechanism of isofluranes actions on
thalamocortical neurons. J Neurophysiol 81, 18021809.
Rose HJ & Metherate R (2001). Thalamic stimulation largely elicits
orthodromic, rather than antidromic, cortical activation in an auditory
thalamocortical slice. Neuroscience 106, 331340.
49

Literaturverzeichnis

Rose HJ & Metherate R (2005). Auditory thalamocortical transmission is


reliable and temporally precise. J Neurophysiol 94, 20192030.
Roy JP, Clercq M, Steriade M & Deschenes M (1984). Electrophysiology of
neurons of lateral thalamic nuclei in cat: mechanisms of long-lasting
hyperpolarization. J Neurophysiol 51, 12201235.
Salt TE (2002). Glutamate receptor functions in sensory relay in the
thalamus. Philos Trans R Soc Lond B Biol Sci 357, 17591766.
Santoro B, Chen S, Luthi A, Pavlidis P, Shumyatsky GP, Tibbs GR &
Siegelbaum SA (2000). Molecular and functional heterogeneity of
hyperpolarization-activated pacemaker channels in the mouse CNS. J
Neurosci 20, 52645275.
Scheffer IE (2003). Severe infantile epilepsies: molecular genetics challenge
clinical classification. Brain 126, 513514.
Sherman SM & Guillery RW (1996). Functional organization of
thalamocortical relays. J Neurophysiol 76, 13671395.
Sherman SM & Guillery RW (2006). Exploring the thalamus and its role in
cortical function. MIT Press, Cambridge, Massachusetts.
Soltesz I, Lightowler S, Leresche N, Jassik-Gerschenfeld D, Pollard CE &
Crunelli V (1991). Two inward currents and the transformation of lowfrequency oscillations of rat and cat thalamocortical cells. J Physiol
(Lond) 441, 175197.
Steriade M (1991). Alertness, quiet sleep, dreaming. Cereb Cortex 9, 279
357.
Steriade M, Domich L & Oakson G (1986). Reticularis thalami neurons
revisited: activity changes during shifts in states of vigilance. J Neurosci
6, 6881.
Steriade M, Domich L, Oakson G & Deschenes M (1987). The deafferented
reticular thalamic nucleus generates spindle rhythmicity. J Neurophysiol
57, 260273.
Steriade M, Jones EG & McCormick DA (1997). Thalamus, 1. edn. Elsevier,
Amsterdam.
Steriade M, McCormick DA & Sejnowski TJ (1993). Thalamocortical
oscillations in the sleeping and aroused brain. Science 262, 679685.
50

Literaturverzeichnis

Straub H, Kuhnt U, Hohling JM, Kohling R, Gorji A, Kuhlmann D, Tuxhorn I,


Ebner A, Wolf P, Pannek HW, Lahl R & Speckmann EJ (2003).
Stimulus-induced patterns of bioelectric activity in human neocortical
tissue recorded by a voltage sensitive dye. Neuroscience 121, 587604.
Todorovic SM, Prakriya M, Nakashima YM, Nilsson KR, Han M, Zorumski
CF, Covey DF & Lingle CJ (1998). Enantioselective blockade of t-type
Ca2+ current in adult rat sensory neurons by a steroid that lacks
gamma-aminobutyric acid-modulatory activity. Mol Pharmacol 54, 918
927.
Vitko I, Bidaud I, Arias JM, Mezghrani A, Lory P & Perez-Reyes E (2007).
The I-II loop controls plasma membrane expression and gating of
Ca(v)3.2 T-type Ca2+ channels: a paradigm for childhood absence
epilepsy mutations. J Neurosci 27, 322330.
Wallace RH, Marini C, Petrou S, Harkin LA, Bowser DN, Panchal RG,
Williams DA, Sutherland GR, Mulley JC, Scheffer IE & Berkovic SF
(2001). Mutant GABAA receptor g2-subunit in childhood absence
epilepsy and febrile seizures. Nat Genet 28, 4952.
Wan X, Mathers DA & Puil E (2003). Pentobarbital modulates intrinsic and
GABA-receptor conductances in thalamocortical inhibition. Neuroscience
121, 947958.
Winer JA & Lee CC (2007). The distributed auditory cortex. Hear Res 229, 3
13.
Wirrell EC (2003). Natural history of absence epilepsy in children. Can J
Neurol Sci 30, 184188.
Ying S-W & Goldstein PA (2005). Propofol-block of SK channels in reticular
thalamic neurons enhances GABAergic inhibition in relay neurons. J
Neurophysiol 93, 19351948.

51

Anhang

Anhang
Der kumulativen Habilitationsschrift zugrundeliegende Publikationen

1. BUDDE T, COULON P*, PAWLOWSKI M, MEUTH P, MEUTH SG, PAPE HC


(2008): Reciprocal modulation of Ih and ITASK in thalamocortical relay
neurons by halothane. Pflgers Arch, 456(6), 1061-73. *(T.B. and P.C.
are equally contributing first authors). IF 2011: 4.463
2. BROICHER T, WETTSCHURECK N, MUNSCH T, COULON P, MEUTH SG,
KANYSHKOVA T, SEIDENBECHER T, OFFERMANNS S, PAPE HC, BUDDE T
(2008): Muscarinic ACh receptor-mediated control of thalamic activity
via G(q)/G(11)-family G-proteins. Pflgers Arch, 456(6), 1049-60. IF
2011: 4.463
3. COULON P, HERR D, KANYSHKOVA T, MEUTH P, BUDDE T, PAPE HC (2009):
Burst discharges in neurons of the thalamic reticular nucleus are
shaped by calcium-induced calcium release. Cell Calcium, 46, 333346. IF 2009: 4.288
4. BROICHER T, BIDMON HJ, KAMUF B, COULON P*, GORJI A, PAPE HC,
SPECKMANN EJ, BUDDE T (2010): Thalamic afferent activation of
supragranular layers in auditory cortex in vitro: a voltage sensitive dye
study. Neuroscience, 165, 371-385. *(corresponding author). IF
2011: 3.380
5. RANKOVIC V, EHLING P, LANDGRAF P, COULON P, KREUTZ MR, MUNSCH T,
2+
AND BUDDE T (2010): Intracellular Ca release-dependent inactivation
2+
of Ca currents in thalamocortical relay neurons. Eur J Neurosci,
31(3), 439 - 449. IF 2011: 3.631
6. COULON P, KANYSHKOVA T, BROICHER T, MUNSCH T, W ETTSCHURECK N,
SEIDENBECHER T, MEUTH SG, OFFERMANNS S, PAPE HC, BUDDE T
(2010): Activity modes in thalamocortical relay neurons are modulated
by Gq/G11 family G-proteins. Front Cell Neurosci, 4:132. IF 2011:
4.171
7. COULON P, BUDDE T, PAPE HC (2012): The sleep relay - the role of the
thalamus in central and de-central sleep regulation. Pflgers Arch,
463(1), 53-71. Review. IF 2011: 4.463
8. GERARD E, HOCHSTRATE P, DIERKES PW, COULON P (2012): Functional
properties and cell type specific distribution of Ih channels in leech
neurons. J Exp Biol, 215(2), 227-238. IF 2011: 2.996
52

Anhang

Lehrerfahrung
Der Antragsteller war vom 09.01.2001 bis 31.03.2005 in der curricularen
Lehre des Fachbereichs Biologie an der Mathematisch-Naturwissenschaftlichen Fakultt der Heinrich-Heine-Universitt Dsseldorf ttig. Im
Rahmen dieser Ttigkeit wurden folgende, auf einem Arbeitszeugnis
besttigte und dokumentierte Lehrveranstaltungen inhaltlich, fachlich und
didaktisch gestaltet und durchgefhrt:
bungen in Neurophysiologie und Stoffwechselphysiologie fr Studierende
im Grundstudium: Computersimulationen. Mit dem Programm Neurosimul
(H. Kettenmann) wurden komplexe neurophysiologische Zusammenhnge
erklrt und demonstriert. Die Ttigkeit umfasste eine umfangreiche
Vorbesprechung ber 1,5 h. Die Veranstaltung wurde durchschnittlich von ca.
180 Studenten besucht, der Umfang betrug 4 SWS und die Veranstaltung
fand jeweils im Wintersemester statt. Der Antragsteller war zusammen mit
AOR Dr. Eberhard von Berg alleinig fr die inhaltliche Gestaltung verantwortlich.
bungen

in

Neurophysiologie

fr

Studierende

im

Hauptstudium:

Extrazellulre Ableitungen am Regenwurm Lumbricus spec.. Am Frischprparat des Riesennervenfasersystems von Lumbricus spec. wurde das
Prinzip der extrazellulren Ableitung demonstriert und vermittelt. Das
Praktikumsmodul wurde dabei vom Antragsteller vollstndig selbstttig
gestaltet, weiterentwickelt und alleinig durchgefhrt. Diese bung hatte einen
Umfang von 4 SWS und fand im Sommersemester statt.
bungen in Neurobiologie fr Studierende im Hauptstudium. In diesem
sechswchigen Blockpraktikum fr fortgeschrittene Studierende der Biologie,
das als Einzelkurs auch fr Examenskandidaten durchgefhrt wurde, wurden
die Studierenden vom Antragsteller in verschiedenen neurobiologischen
53

Anhang

Messtechniken ausgebildet. Es wurden u. a. intrazellulre Ableitungen an


einem Frischprparat von Ganglien des medizinischen Blutegels Hirudo
medicinalis bzw. H. verbana durchgefhrt, um das Prinzip intrazellulrer
Ableitungen zu vermitteln. Dabei wurden aktuelle Themen aus der im Institut
durchgefhrten wissenschaftlichen Arbeiten behandelt, u. a. wurde die
Funktion von AMPA/Kainat Rezeptoren untersucht, die Volumenregulation in
Retzius Neuronen und die potentiometrische Volumenbestimmung unter
Verwendung der Voltage-Clamp-Technik entwickelt. Die bungen hatten
einen

Umfang

von

SWS.

Der

Ablauf

des

Einzelkurses

fr

Examenskandidaten war hiermit vergleichbar, jedoch fand die Veranstaltung


nach Vereinbarung statt. Der Umfang betrug auch hier 4 SWS.
Literaturseminar Neurobiologie. Hierbei wurden von den Studenten unter
Anleitung aktuelle Themen der Neurobiologie recherchiert und vorgetragen.
Die Veranstaltung fand parallel zum Blockpraktikum statt und hatte einen
Umfang von 2 SWS.

Der Antragsteller ist seit 2006 an der Westflischen Wilhelms-Universitt


Mnster in der curricularen Lehre im Fachbereich Medizin ttig. In der
Veranstaltung Seminar in Physiologie fr Medizinstudenten wurden
folgende

Themen

eigenverantwortlich

behandelt:

Atmung,

Niere

glomerulre Filtration, Herz, Kreislauf, Blutgefsystem und Blutgerinnung


und Leistung im Wintersemester sowie im Sommersemester die Themen
Erregungsentstehung, -leitung und -bertragung im Nervensystem, motorisches System, somatoviszerale Sensibilitt, auditorisches System, visuelles
System, Wachen und Schlafen sowie Lernen und Gedchtnis.

Auerdem ist der Antragsteller seit dem Sommersemester 2008 Prfer fr


die Abschlusstestate im Praktikum Physiologie fr Mediziner und staatlich
bestellter Prfer fr die mndlich-praktischen Prfungen im Ersten Abschnitt
54

Anhang

der rztlichen Prfung (Physikum) nach der Approbationsordnung fr rzte


(AppO 2002).
Im Rahmen der Veranstaltung Hochschultag der Westflischen WilhelmsUniversitt Mnster hat der Antragsteller von 2006 bis 2008 zusammen mit
Frau Prof. Dr. Heidrun Straub und Dr. Jrg Lesting die Teilveranstaltung
Forschungsreisen ins Gehirn - Aktuelle Einblicke in die Neurobiologie
durchgefhrt. Dabei bernahm der Antragsteller sowohl die Vorlesung (1 h),
als auch einen Teil der Laborfhrungen. Die Veranstaltung wurde von bis zu
180 Teilnehmern besucht.

Fr

den

Sonderforschungsbereich

TRR

58

Furcht,

Angst,

Angst-

erkrankungen wurden seit 2009 ca. vier Mal pro Jahr (nach Vereinbarung)
Laborfhrungen und Kurzvorlesungen fr Oberstufenschler durchgefhrt.
Die Veranstaltung wurde von durchschnittlich ca. 20 Teilnehmern besucht.

Innerhalb des Instituts fr Physiologie wurden fr Doktoranden die Vorlesungen Basic principles of electrophysiology und Calcium Imaging
(insgesamt 1 SWS) abgehalten.
Fr das Otto Creutzfeld Center for Cognitive and Behavioral Neuroscience
wird am 08.11.2012 die Vorlesung Sleep and Epilepsy: From Spindles to
Spikes im Rahmen der Ringvorlesung OCC Lecture Series WS 2012/13
(Nr.: 132906) gehalten.

55

Anhang

Eingeworbene Frdermittel und Preise


1. Ca2+ vermittelte Ca2+-Freisetzung in Neuronen des Nucleus
Reticularis Thalami, Innovative Medizinische Forschung (IMF)
Mnster. 15.000 Euro (CO 2 1 08 03).
2. Die Rolle elektrischer Synapsen in einem Epilepsiemodell: Ca2+vermittelte Wechselwirkungen zwischen Neuronen des Nucleus
Reticularis Thalami. Innovative Medizinische Forschung (IMF)
Mnster. 138.000 Euro (CO 1 2 10 08).
3. Award for an outstanding poster, 87th meeting of the German
Physiological Society, Cologne 2008 [COULON P, PAWLOWSKI M, MEUTH
P, MEUTH SG, PAPE HC, BUDDE T: Reciprocal modulation of Ih and ITASK
in thalamocortical relay neurons by halothane. Acta Physiologica, 192,
663 (suppl.), PW 04-07].

Gesamtpublikationsliste
Artikel in peer-review-Journalen
1. DIERKES PW, COULON P, NEUMANN S, SCHLUE WR (2002):
Potentiometric measurement of cell volume changes and intracellular
ion concentrations under voltage-clamp conditions in invertebrate
nerve cells. Anal Bioanal Chem, 373, 762-766. IF 2011: 3.778 Klasse 1
2. COULON P, W STEN HJ, HOCHSTRATE P, DIERKES PW (2008): Swellingactivated chloride channels in leech Retzius neurons. J Exp Biol, 211,
630-641. IF 2011: 2.996 - Klasse 2
3. BUDDE T, COULON P*, PAWLOWSKI M, MEUTH P, MEUTH SG, PAPE HC
(2008): Reciprocal modulation of Ih and ITASK in thalamocortical relay
neurons by halothane. Pflgers Arch, 456(6), 1061-73. *(T.B. and P.C.
are equally contributing first authors). IF 2011: 4.463 - Klasse 1
4. BROICHER T, WETTSCHURECK N, MUNSCH T, COULON P, MEUTH SG,
KANYSHKOVA T, SEIDENBECHER T, OFFERMANNS S, PAPE HC, BUDDE T
(2008): Muscarinic ACh receptor-mediated control of thalamic activity
via Gq/G11-family G-proteins. Pflgers Arch, 456(6), 1049-60. IF 2011:
4.463 - Klasse 1
5. DNGI M, HIRNET D, COULON P, PAPE HC, DEITMER JW, LOHR C (2009):
GABA uptake-dependent Ca2+ signaling in olfactory bulb astrocytes: a
56

Anhang

new mechanism of GABAergic signaling. PNAS, 106, 17570-17575. IF


2011: 9.681 - Klasse 1
6. COULON P, HERR D, KANYSHKOVA T, MEUTH P, BUDDE T, PAPE HC (2009):
Burst discharges in neurons of the thalamic reticular nucleus are
shaped by calcium-induced calcium release. Cell Calcium, 46, 333346. IF 2009: 4.288 - Klasse 1
7. BROICHER T, BIDMON HJ, KAMUF B, COULON P*, GORJI A, PAPE HC,
SPECKMANN EJ, BUDDE T (2010): Thalamic afferent activation of
supragranular layers in auditory cortex in vitro: a voltage sensitive dye
study. Neuroscience, 165, 371-385. *(corresponding author). IF
2011: 3.380 - Klasse 2
8. RANKOVIC V, EHLING P, LANDGRAF P, COULON P, KREUTZ MR, MUNSCH T,
2+
AND BUDDE T (2010): Intracellular Ca release-dependent inactivation
2+
of Ca currents in thalamocortical relay neurons. Eur J Neurosci.,
31(3), 439-449. IF 2010: 3.658; IF 2011: 3.631 - Klasse 2
9. COULON P, KANYSHKOVA T, BROICHER T, MUNSCH T, W ETTSCHURECK N,
SEIDENBECHER T, MEUTH SG, OFFERMANNS S, PAPE HC, BUDDE T
(2010): Activity modes in thalamocortical relay neurons are modulated
by Gq/G11 family G-proteins. Front Cell Neurosci, 4:132. IF 2011: 4.171
- Klasse 1
10. GREBER B, COULON P, ZHANG M, MORITZ S, FRANK S, MLLER MOLINA
AJ, ARAZO-BRAVO MJ, HAN DW, PAPE HC, SCHLER HR (2011): FGF
signaling inhibits neural induction in human embryonic stem cells.
EMBO J, 30(24), 4874-4884. IF 2011: 9.205 - Klasse 1
11. COULON P, BUDDE T, PAPE HC (2012): The sleep relay - the role of the
thalamus in central and de-central sleep regulation. Pflgers Arch,
463(1), 53-71. Review. IF 2011: 4.463 - Klasse 1
12. GERARD E, HOCHSTRATE P, DIERKES PW, COULON P (2012): Functional
properties and cell type specific distribution of Ih channels in leech
neurons. J Exp Biol, 215(2), 227-238. IF 2011: 2.996 - Klasse 2
13. SADEGHIAN H, JAFARIAN M, KARIMZADEH F, KAFAMI L, KAZEMI H, COULON
P, GHABAEE M, GORJI A (2012): Neuronal death by repetitive cortical
spreading depression in juvenile rat brain. Exp Neurol, 233, 438-446.
IF 2011: 4.699 - Klasse 1
14. JNGLING K, LESTING J, COULON P, LIU X, REINSCHEID R, AND PAPE HC
(2012): Cellular properties of neuropeptide S expressing neurons at
57

Anhang

the locus coeruleus in the mouse brain stem. J Physiol (Lond),


590(16), 3701-3717. IF 2011: 4.718 - Klasse 1
15. KARIMZADEH F, JAFARYAN M, GHARAKHANI M, RAZEGHI S, MOHAMADZADEH
E, KHALLAGHI B, KAZEMI H, COULON P, GORJI A (2012): Behavioural and
histopathological assessment of the effects of periodic fasting on
pentylenetetrazol-induced seizures in rats. Nutr Neurosci, in press. IF
2011: 1.563 - Klasse 3
16. MEUTH SG, GBEL K, KANYSHKOVA T, RITTER M, SCHWIND W,
BIELASZEWSKA M, EHLING P, LEBNITZ P, COULON P, PAVENSTDT H,
KARCH H, PETERS G, BUDDE T, W IENDL H, PAPE HC: Thalamic
involvement in patients with neurologic impairment due to shiga toxinproducing E. coli. Ann Neurol, revision submitted.
17. EHLING P, CERINA M, MEUTH P, BISTA P, COULON P, MEUTH SG, PAPE HC,
2+
+
AND BUDDE T: Ca -dependent large conductance K currents in
thalamocortical relay neurons of epileptic WAG/Rij and non-epileptic
ACI rats. Submitted to Pflgers Arch.
Erst- oder Letztautor (inkl. geteilten Erstautorenschaften und korrespondierender Autor): 7
Erst- oder Letztautor (inkl. geteilten Erstautorenschaften): 6
bersichtsartikel (Reviews): 1
Kumulativer impact: 68.495 (ohne submitted und under revision)
Durchschnittlicher impact factor (von 15): 4.566
Punkte: 92
Buchrezensionen
1. Bear, Connors, Paradiso: Neurowissenschaften. BIOspektrum 15(03):
347.
Abstracts und Tagungsbeitrge
1. COULON P, DIERKES PW, NEUMANN S, SCHLUE WR (2002):
Potentiometric measurement of cell volume changes under voltageclamp conditions in invertebrate nerve cells. 81st Annual Meeting of
the DPG, Tbingen 2002 Pflgers Arch., 443 (suppl.): S236.
http://www.springerlink.com/content/w5endfa8yx4fqw01/fulltext.pdf
2. COULON P, KLEES G, DIERKES PW, SCHLUE WR (2003): Effect of
hyposmotic conditions on cell volume and electrophysiological
properties of leech Retzius neurones. In: Elsner, N., Zimmermann, H.
(Eds.): The Neurosciences from Basic Research to Therapy.
58

Anhang

Proceedings of the 29th Goettingen Neurobiology Conference and the


5th Meeting of the German Neuroscience Society 2003, Georg
Thieme Verlag Stuttgart, S795.
3. COULON P, KLEES G, LANGER J, DIERKES PW, SCHLUE WR (2003):
Effect of osmotic alterations on cell volume and electrophysiological
properties of leech Retzius neurones. Poujeol P, Petersen O (Eds.):
Proceedings of the 3rd FEPS Congress, Nice, France 2003, OC12-3.
http://www.feps.org/abstracts/FEPS2003Proceeding.pdf
4. COULON P, DIERKES PW, HOCHSTRATE P, SCHLUE WR (2005): Swellingactivated chloride current in leech Retzius neurones. Proceedings of
the 30th Goettingen Neurobiology Conference and the 6th Meeting of
the German Neuroscience Society 2005, Neuroforum 1/2005 (suppl.),
315B.
5. COULON P, KANYSHKOVA T, MEUTH P, MEUTH SG, PAWLOWSKI M, PAPE
HC, BUDDE T (2008): Reciprocal modulation of Ih and ITASK in
thalamocortical relay neurons by halothane. Acta Physiologica, 192,
663 (suppl.), PW 04-07, 87th DPG meeting, Cologne 2008 (DPG
award for outstanding poster).
6. KAMUF B, BUDDE T, SEIDENBECHER T, SPECKMANN EJ, PAPE HC, BIDMON
HJ, COULON P, GORJI A, BROICHER T (2008): Thalamcortical projections
in a horizontal brain slice preparation. Acta Physiologica, 192, 663
(suppl.), PM 10-02, 87th DPG meeting, Cologne 2008.
7. COULON P, HERR D, BUDDE T, PAPE HC (2009): Low voltage activated
calcium channels are coupled to ryanodine receptors in neurons of the
thalamic reticular nucleus. Neuroforum, 2009: 1, Suppl.: T6-3A.
8. DOENGI M, COULON P, PAPE HC, DEITMER JW, LOHR C (2009): GABA
transport-mediated calcium signaling in olfactory bulb astrocytes.
Neuroforum, 2009: 1, Suppl.: T9-7A.
9. DOENGI M, HIRNET D, COULON P, PAPE HC, DEITMER JW, LOHR C (2009):
GABA transport-mediated calcium signaling in olfactory bulb
astrocytes. NeuroVisionen 5, Sept 3rd 2009, Bochum, Germany. P. 51.
Winner of a poster prize.
http://www.rd.ruhr-uni-bochum.de/neuro/events/neurovisionen5/index.html
10. MEUTH P, KANYSHKOVA T, PAWLOWSKI M, DUB C, BENDER RA,
BREWSTER AL, BAUMANN A, BARAM TZ, EHLING P, COULON P, MEUTH SG,
PAPE HC, & BUDDE T (2009): HCN channel expression in thalamic
59

Anhang

neurons impacts maturation of thalamocortical oscillations.


NeuroVisionen 5, Sept 3rd 2009, Bochum, Germany. P. 104.
11. COULON P, HERR D, KANYSHKOVA T, MEUTH P, BUDDE T, PAPE HC (2009):
Calcium induced calcium release modulates burst discharges in
neurons of the thalamic reticular nucleus. Program No. 818.3, Poster
No.: B121. 2009 Neuroscience Meeting, Chicago, IL: Society for
Neuroscience, 2009. Online.
http://www.sfn.org/skins/main/pdf/abstracts/am2009/poster_presentatio
ns/PosterPresentation_WedPM.pdf
12. MEUTH P, KANYSHKOVA T, PAWLOWSKI M, DUBE C, BENDER RA,
BREWSTER AL, BAUMANN A, BARAM TZ, EHLING P, COULON P, MEUTH SG,
PAPE HC, BUDDE T (2010): HCN channel expression in thalamic
neurons impacts maturation of thalamocortical oscillations. Acta
Physiologica, 198, 677 (suppl.), O-TUE-2-5.
http://www.blackwellpublishing.com/aphmeeting/abstract.asp?MeetingI
D=769&id=86014&meeting=APSABS2010677
13. COULON P (2010): Burst Discharges in the Thalamic Reticular Nucleus:
Ion channel properties and their role in Absence Epilepsy. Dec 10th 13th. First International Epilepsy Congress, Razavi Hospital, Mashad,
Iran.
14. CERINA M, COULON P, PAPE HC, BUDDE T (2011): The role of KCNQ
channels in the thalamus. Proceedings of the 9th Gttingen Meeting of
the German Neuroscience Society, 33rd Gttingen Neurobiology
Conference. Poster No.: T6-11C.
https://www.nwg-goettingen.de/2011/upload/file/Proceedings_2011.pdf
15. GREBER B, COULON P, ZHANG M, MORITZ S, FRANK S, ARAZO-BRAVO
MJ, HAN DW, PAPE HC, SCHLER HR (2011): FGF signaling inhibits
neural induction in human embryonic stem cells. Kompetenznetzwerk
Stammzellforschung NRW, 6th International Meeting: 5.-6. April 2011 in
Essen, Germany. Poster No. 28.
16. CERINA M, SZKUDLAREK H, PAPE HC, BUDDE T, COULON P (2011): KCNQ
channels in the thalamus: role of a neuronal brake in sensory
perception. NeuroVisionen 7, Oct 7th 2011, Essen, P. 63.
http://www.uk-essen.de/neurovisionen7/downloads/abstraktbandneurovisionen7.pdf
17. GREBER B, COULON P, ZHANG M, MORITZ S, FRANK S, MLLER MOLINA
AJ, ARAZO-BRAVO MJ, HAN DW, PAPE HC, SCHLER HR (2011): FGF
signaling specifically inhibits neural induction in human embryonic
60

Anhang

stem cells. Stem Cell Programming & Reprogramming, December 8


10, 2011, Lisbon, Portugal. P1.58
18. CERINA M, SZKUDLAREK H, KANYSHKOVA T, MEUTH SG, PAPE HC, BUDDE
T, COULON P (2012): KCNQ Channels in the thalamus: role of a
neuronal brake in somatosensory perception. Acta Physiologica, 204,
689 (suppl.), P128
http://www.blackwellpublishing.com/aphmeeting/abstract.asp?MeetingI
D=785&id=98588
19. COULON P (2012): From sleep and spindles to spikes and seizures:
Neurophysiological data on absence epilepsy. Young Researchers in
Pediatric Epileptology: From Gene to Disease. Kiel, August 23rd - 25th
http://www.epilepsiegenetik.uni-kiel.de/Englisch/Meeting/Flyer.pdf
http://www.epilepsiegenetik.uni-kiel.de/Englisch/Meeting/Program.pdf
Erst- oder Letztautor: 11
Acknowledged contributions
1. JNGLING, K., SEIDENBECHER, T., SOSULINA, L., LESTING, J., SANGHA, S.,
CLARK, S., OKAMURA, N., DUANGDAO, D., XU, Y.-L., REINSCHEID, R. AND
PAPE, H.-C. (2008): Neuropeptide S-mediated control of fear
expression and extinction: role of intercalated GABAergic neurons in
the amygdala. Neuron, 59, 293-310.
2. BROCKHAUS J, PAPE HC (2010): Abnormalities in GABAergic synaptic
transmission of intralaminar thalamic neurons in a genetic rat model of
absence epilepsy. Mol Cell Neurosci, 46, 444-451
3. GORJI A, MITTAG C, SHAHABI P, SEIDENBECHER T, PAPE HC (2011):
Seizure-related activity of intralaminar thalamic neurons in a genetic
model of absence epilepsy. Neurobiol Dis, 43, 266-274
4. GHADIRI MK, KOZIAN M, GHAFFARIAN N, STUMMER W, KAZEMI H,
SPECKMANN EJ AND GORJI A (2012): Sequential changes in neuronal
activity in single neocortical neurons after spreading depression.
Cephalalgia, 32(2), 116-124.
5. LANGE MD, DOENGI M, LESTING J, PAPE HC AND JNGLING K (2011):
Heterosynaptic long-term potentiation at interneuronprincipal neuron
61

Anhang

synapses in the amygdala requires nitric oxide signalling. J Physiol,


590(1), 131-143.
Abschlussarbeiten
1. COULON P (2005): Elektrophysiologie der Volumenregulation von
Retzius-Neuronen im Zentralnervensystem des medizinischen
Blutegels.
Dissertation,
Heinrich-Heine-Universitt
Dsseldorf,
Mathematisch-Naturwissenschaftliche Fakultt.
http://diss.ub.uni-duesseldorf.de/ebib/diss/diss_files/1022.pdf
2. COULON P (2001): Einfluss extrazellulrer Ionen auf das Volumen und
die elektrophysiologischen Eigenschaften identifizierter BlutegelNeuronen. Diplomarbeit, Heinrich-Heine-Universitt Dsseldorf
(Universitts- und Landesbibliothek Dsseldorf: bioa780.c855).
Wissenschaftliche Vortrge, Invited Talks
1. Effect of osmotic alterations on cell volume and electrophysiological
properties of leech Retzius neurones. 3rd FEPS Congress, Nice,
France 2003, OC12-3.
2. Swelling-activated chloride channels in leech Retzius neurons.
Department of Anesthesia (Stuart Forman and Keith Miller) of the
Massachusetts General Hospital, Harvard Medical School, Boston (03.
Oktober 2005).
3. Swelling-activated chloride channels in leech Retzius neurons. Institut
fr Physiologie (Andreas Lckhoff and Frank Khn), RWTH Aachen
(09. September 2005).
4. Channel surfing in the thalamus. From selective attention to
anaesthesia: ion channels in the thalamocortical system and their
physiological and pathophysiological relevance, Evotec AG, Hamburg
(07. Dezember 2010)
5. From sleep and spindles to spikes and seizures: Neurophysiological
data on absence epilepsy. Young Researchers in Pediatric
Epileptology: From Gene to Disease. Kiel, August 23rd - 25th (24.
August 2012)

62

Anhang

Lebenslauf Philippe Coulon


Persnliche Informationen
Geburtstag:
Geburtsort:
Nationalitt:
Adresse:

22. April 1975


Hamburg
deutsch
Rigaweg 8, 48159 Mnster

Besetzte Positionen
09/2011 heute
01/2006 09/2011
07/2005 01/2006
12/2001 03/2005
09/2000 11/2001

Senior Post-Doc, Physiologie I, Mnster


Wissenschaftlicher Angestellter (Post-Doc), Physiologie I,
Mnster
Coordinator of the International Graduate School of
Neuroscience (IGSN), Bochum
Wissenschaftlicher Angestellter (Doktorand) und Promotionsstudent, Institut fr Neurobiologie, Dsseldorf
Studentische Hilfskraft, Institut fr Neurobiologie,
Dsseldorf

Ausbildung
01/2005 heute
12/2001 01/2005
10/94 12/2001
11/87 6/94
07/81 11/87

Post-Doc
Promotion zum Dr. rer. nat. (magna cum laude)
Studium
der
Biologie
(Heinrich-Heine-Universitt
Dsseldorf), Abschluss: Dipl.-Biol. (sehr gut)
Erzbischfliches St. Suitbertus Gymnasium in DsseldorfKaiserswerth, Abitur (1,8)
Schulen in Nrnberg, Hamburg und London

Gutachter-, Prfer- und Beratungsttigkeiten


- seit 03.07.2008 staatlich bestellter Prfer fr das Fach Physiologie fr die
mndlich-praktischen Prfungen im Ersten Abschnitt der rztlichen Prfung
(Physikum)
- Gutachter der Heinrich-Hertz-Stiftung
- Gutachter des IMF, "Innovative Medizinische Forschung an der Medizinischen
Fakultt Mnster"
- Reviewer bei Brain Research, Regulatory Peptides, European Journal of
Physiology - Pflgers Archiv, Journal of Experimental Biology, PLoS ONE
- 2003-2005: Gewhltes Mitglied der Senatskommission fr Forschung,
Forschungstransfer und wissenschaftlichen Nachwuchs
- 2009: Berater fr das Patent-Mining an der Universitt Mnster
Sonstige Interessen
Sportklettern, Bouldern (Trainer C), Naturfotografie, Musik (Trompete)
Mitgliedschaften
- Neurowissenschaftliche Gesellschaft, NWG, seit 2003)
- Frderverein Epilepsieforschung an der Universitt Mnster e.V., seit 2007
- Society for Neuroscience (2009-2010)

63

Danksagungen

Danksagungen
Prof. Dr. Hans-Christian Pape gilt besonderer Dank fr seine Untersttzung und die
hervorragende Infrastruktur des Instituts. Prof. Dr. Thomas Budde bin ich fr seine
Ideen und die stetige, tatkrftige und freundschaftliche Untersttzung zu groem
Dank verpflichtet. Prof. em. Dr. Erwin-Josef Speckmann gilt besonderer Dank fr
seine Ermunterungen und seine guten Ratschlge. David Herr, Christina Neyer,
Manuela Cerina und Denise Kohmann gilt mein besonderer Dank dafr, dass sie
sich dafr entschieden haben, mit mir zusammen zu arbeiten. Dr. Susan Sangha,
Dr. Ludmila Sosulina, Dr. Tilman Broicher, Dr. Christian Kluge, Dr. Jrg Lesting, Dr.
Kay Jngling, PD Dr. Christian Stock und Dr. Michael Dngi danke ich sehr fr die
zahllosen

fachlichen

und

freundschaftlichen

Gesprche.

Birgit

Herrenpoth,

Bernadette Reinecke, Petra Berenbrock, Hubert Bumer, Elisabeth Bning,


Svetlana Kiesling, Ingrid Winkelhues und Elke Na danke ich fr die beraus gute
Zusammenarbeit. Auerdem danke ich allen anderen Mitarbeitern des Instituts fr
Physiologie I fr die gute Arbeitsatmosphre. Christian Schelp, Dr. Georg Pohland,
Dr. Ednan Gerard, Dr. Peter Hochstrate, Dr. Peter Mullen und Anne Brockerhoff
danke ich fr viele erhellende Gesprche, die richtigen Ratschlge und die
langjhrige Freundschaft. David Gregori, Hendrik Fritze, Dr. Dajana Klein, Dr. Anna
Martinsohn, Anna Vssing, Iris Lindemann, Janina Strunk und Marlies Hepting
danke ich dafr, dass sie Mnster zu einer Heimat fr mich gemacht haben. Und
schlielich danke ich meiner Familie fr die moralische und tatkrftige Untersttzung
in allen Phasen des Lebens. Danke!

64

227
The Journal of Experimental Biology 215, 227-238
2012. Published by The Company of Biologists Ltd
doi:10.1242/jeb.062836

RESEARCH ARTICLE
Functional properties and cell type specific distribution of Ih channels in leech
neurons
Ednan Gerard1, Peter Hochstrate1, Paul-Wilhelm Dierkes2 and Philippe Coulon3,*
1

Institut fr Neurobiologie, Heinrich-Heine-Universitt, 40225 Dsseldorf, Germany, 2Abteilung fr Didaktik der Biowissenschaften,
Johann Wolfgang Goethe-Universitt, Sophienstrae 1-3, 60487 Frankfurt/Main, Germany and 3Institut fr Physiologie I,
Westflische Wilhelms-Universitt, Robert-Koch-Str. 27a, 48149 Mnster, Germany
*Author for correspondence (coulon@uni-muenster.de)

Accepted 9 October 2011

SUMMARY
The hyperpolarisation-activated cation current (Ih) has been described in many vertebrate and invertebrate species and cell types.
In neurons, Ih is involved in rhythmogenesis, membrane potential stabilisation and many other functions. In this work, we
investigate the distribution and functional properties of Ih in identified leech neurons of intact segmental ganglia. We found Ih in
the mechanosensory touch (T), pressure (P) and noxious (N) neurons, as well as in Retzius neurons. The current displayed its
largest amplitude in P neurons and we investigated its biophysical and pharmacological properties in these cells. Ih was halfmaximally activated at 65mV and fully activated at 100mV. The current mutually depended on both Na+ and K+ with a
permeability ratio pNa/pK of ~0.21. The reversal potential was approximately 35mV. The time course of activation could be
approximated by a single time constant of ~370ms at 60mV, but required two time constants at 80mV of ~80 and ~560ms. The
current was half-maximally blocked by 0.3mmoll1Cs+ but was insensitive to the bradycardic agent ZD7288. The physiological
function of this channel could be a subtle alteration of the firing behaviour of mechanosensory neurons as well as a stabilisation
of the resting membrane potential.
Key words: leech, P neuron, Ih channel, caesium pharmacology, ZD7288.

INTRODUCTION

The hyperpolarisation-activated cation current (Ih) is involved in a


range of functions of cells across species (Biel et al., 2009; Pape,
1996). In neurons, the current allows the control of rhythmic activity
[e.g. in the thalamocortical system (McCormick and Pape, 1990b)],
plays a key role in determining and stabilising the resting membrane
potential (Em) (Ludwig et al., 2003; Meuth et al., 2006; Pape, 1996)
and has a wide range of other functions in processes such as dendritic
integration, synaptic plasticity, synaptic transmission and the
temporal processing of visual signals in the retina (for a review, see
Biel et al., 2009). The current has several characteristic properties:
(1) it is carried by both Na+ and K+ and has a reversal potential
(Erev) of approximately 20mV under physiological conditions,
hence it is inwardly directed at rest, causing a membrane
depolarisation upon activation; (2) it is activated by hyperpolarising
voltage deflections negative to approximately 55mV; (3) its
activation is facilitated by cyclic adenosine monophosphate (cAMP);
and (4) it is blocked by low concentrations of Cs+ and by the
bradycardic agent ZD7288.
The first systematic description of a hyperpolarisation-activated
cation conductance in leech neurons by Arbas and Calabrese
showed that the electrical properties of heart interneurons (HN cells)
include a Cs+-sensitive component that is vital for these cells
oscillatory behaviour (Arbas and Calabrese, 1987a; Arbas and
Calabrese, 1987b). This component appeared as a slow partial
repolarisation after hyperpolarising current injections (voltage
sag, see Fig.1). Later, the underlying current was shown to have
the typical properties of neuronal Ih found in mammals, in that it

was fully activated near 70 to 80mV, its activation kinetics were


voltage dependent between 100 and 60mV (with time constants
of 700ms and 2s, respectively) and it was carried by both Na+ and
K+ (Angstadt and Calabrese, 1989).
In leech HN cells, Ih helps to control heartbeat: a series of
inhibitory postsynaptic potentials hyperpolarises the membrane and
causes a slow activation of Ih, which, in turn, mediates the
depolarising phase (Angstadt and Calabrese, 1989). This function
is similar to that in the thalamic neurons of mammals. Here, the
current can provide pacemaker depolarisations after the preceding
activity of hyperpolarising currents, which then trigger
depolarisation-activated conductances such as low-threshold Ca2+
channels (Akasu et al., 1993; McCormick and Pape, 1990a;
McCormick and Pape, 1990b; Pape and McCormick, 1989). Another
function is to stabilise neurons near their Em by lowering the input
resistance upon hyperpolarising stimuli, causing both reduced
amplitude of postsynaptic potentials (shunting inhibition) and a
return to more positive potentials. Ih was reported to play a role in
stabilising the resting membrane potential in thalamic relay neurons,
nodose sensory neurons and hippocampal interneurons (Doan and
Kunze, 1999; Lupica et al., 2001; Meuth et al., 2006). Such a role
has yet to be described for leech neurons, and a detailed analysis
of Ih across leech neurons is not yet available.
Besides HN cells, a significant Ih that activates near 70mV was
also found in cultured leech Retzius neurons (Angstadt, 1999). In
Retzius neurons of intact ganglia, voltage sags appeared to be
substantially smaller, but Ih was nevertheless thought to be involved
in the control of spontaneous action potential activity (Angstadt,

THE JOURNAL OF EXPERIMENTAL BIOLOGY

228

E. Gerard and others

1999; Coulon et al., 2008). Marked voltage sags have been described
in pressure (P) neurons in situ (Jansen and Nicholls, 1973; Klees
et al., 2005), but the underlying mechanism has not been
investigated. Ih was also found in salivary glands of the giant
Amazon leech, Haementeria ghilianii (Wuttke and Berry, 1992).
The aims of this study were to: (1) determine which neurons show
Ih or the characteristic voltage sag; (2) characterise key properties
such as activation kinetics, voltage dependence, effects of
impermeable cations, pharmacological profile and ion fluxes that
constitute the current; and (3) elucidate the possible functional role
this current plays in leech P neurons by assessing the excitability
and the effects on the time course of AP firing.
We found that Ih preferentially appears in leech mechanosensory
neurons and is largest in P neurons. We concentrated on P neurons
for the biophysical and pharmacological characterisation of the
current properties and for a discussion of the currents functional
role.

lateral packet there are two pairs of P neurons (P1, P2), which can
be distinguished by their electrophysiological characteristics.
Depolarising current injections evoke large-amplitude action
potentials and hyperpolarising current injections evoke characteristic
voltage sags (see Fig.1). The two anterior-lateral packets contain
N, T and AP neurons. N and AP neurons are similar in size to P
neurons, but N neurons generate very large action potentials with
a prominent afterhyperpolarisation immediately after microelectrode
impalement. AP neurons show spontaneous action potentials with
low amplitude (~10mV) at a frequency of ~5Hz and a characteristic
pagoda shape (Nicholls, 1987). T neurons are smaller in size and
do not show spontaneous action potentials. AE neurons are situated
adjacent to neuron 251 and generate spontaneous low-amplitude
(~7mV) action potentials at ~10Hz. Leydig neurons are adjacent
to P2 neurons, smaller in size and fire spontaneously at less than
4Hz (Arbas and Calabrese, 1990).

MATERIALS AND METHODS


Preparation

The standard leech saline (SLS) used for bath perfusion had the
following composition (in mmoll1): 85 NaCl, 4 KCl, 2 CaCl2, 1
MgCl2 and 10 HEPES. The pH was adjusted to 7.40 with 1moll1
NaOH, which increased the Na+ concentration by 4mmoll1. The
osmolality of the SLS was 190mosmolkg1 H2O (Osmomat 030,
Gonotec, Berlin, Germany). Na+-free solution was prepared by
substitution of Na+ with N-methyl-D-glucammonium (NMDG+,
prepared from N-methyl-D-glucamine and HCl; Sigma-Aldrich,
Munich, Germany). Because of differences in osmotic activity,
105mmoll1 NMDG-Cl was used to substitute 89mmoll1 NaCl.
In solutions containing 89mmoll1 Li+, Na+ was omitted and the
pH was adjusted using LiOH. Finally, in solutions with varying K+
concentrations, K+ was either removed, or added to the solution
without osmotic compensation. Monovalent and polyvalent cations
were added as chloride salts without osmotic compensation.
In most preparations, the activation curve of Ih can be shifted
towards more positive potentials by elevations in the intracellular
cAMP concentration (Pape, 1996), thus increasing active Ih under
resting conditions. In order to determine whether leech Ih channels
are also sensitive to cAMP, we used the membrane-permeable cAMP
analogue dibutyryl-cAMP (dbcAMP). Similarly, we applied 3isobutyl-1-methylxanthine (IBMX, Sigma-Aldrich) to inhibit
phosphodiesterases and, thus, increase the intracellular cAMP
concentration (Bobker and Williams, 1989). To directly block
Ih, we used Cs+ (CsCl, Sigma-Aldrich) or ZD7288
(4-ethylphenylamino-1,2-dimethyl-6-methylaminopyrimidinium
chloride; Tocris, Bristol, UK). For experiments in the presence of
ZD7288, the bath perfusion was stopped after adding ZD7288 to
the SLS.

Bath solutions and drugs

Preparation was done as described previously (Coulon et al., 2008).


Leeches were purchased from a commercial supplier (Zaug GmbH,
Biebertal, Germany), or taken from the laboratorys own breeding
stock, and identified as Hirudo verbana Carena 1820 (Annelida,
Hirudinea) (see Siddall et al., 2007). Intact segmental ganglia were
continuously superfused at a rate of ~5mlmin1, exchanging the
chamber volume (0.05ml) approximately 100 times per minute.
Electrophysiological recordings

Neurons were impaled by two sharp electrolyte-filled


microelectrodes for recording Em and for current injection. The
electrodes were pulled on a vertical puller (Narishige PE-2, London,
UK) from borosilicate capillaries (outer/inner diameter:
1.5mm/0.86mm, 0.15mm filament; GC150F-15, Clark
Electromedical Instruments, Pangbourne, UK) and filled with
0.5moll1 K2SO4 and 5mmoll1 KCl. Electrode resistance was
~90M. Electrodes were connected to a custom-made two-electrode
voltage-clamp amplifier (modified TEC-05L, NPI Instruments,
Tamm, Germany). The bath was grounded via an agar bridge. The
output signals were digitised by an A/D converter (custom built,
Zentralwerkstatt Biologie, Heinrich-Heine-Universitt Dsseldorf,
Germany; or Digidata 1322A, Molecular Devices, Sunnyvale, CA,
USA) and were saved and analysed on a computer running in-house
acquisition software (Eberhard von Berg, Institut fr Neurobiologie,
Heinrich-Heine-Universitt Dsseldorf) or pClamp (Molecular
Devices). Sampling frequency was 1kHz for slow signals or 10kHz
for fast signals. Stimulus parameters were programmed on a MAX
21 pulse generator (Zeitz Instruments, Augsburg, Germany). The
experimental setup was shielded by a Faraday cage and dampened
by shock absorbers. Recordings were performed at room temperature
(21C). Fast current oscillations lasting roughly 10ms after holding
potential deflections in the voltage-clamp mode are not shown in
the current traces to improve clarity.
Identification of neurons

The 21 segmental ganglia contain ~400 neurons arranged in six


packets. The neurons investigated in this work can be divided into
neurosecretory (Retzius and Leydig), motor [annulus erector (AE)
and anterior pagoda (AP)] and mechanosensory [touch (T), pressure
(P) and noxious (N)]. When ganglia are fixed ventral side up, Retzius
neurons are easily identifiable based on their large size (~80m
diameter of the soma) and central position. In the adjacent, posterior-

Calculation of ion selectivity and analysis of experimental


data

Ih is carried by cation channels that are selective for Na+ and K+


(Ho et al., 1994). Determining Erev for Ih allows the calculation of
the permeability ratio (pNa/pK) by rearranging Goldmans equation
(see Aidley, 1989):
pNa
=
pK

K + e
o

zFErev
RT

zFErev
RT

K +
i

Na+ Na+
i
o

(1)

where [K+]o and [Na+]o are the extracellular and [K+]i and [Na+]i
the intracellular concentrations of K+ and Na+, respectively, z is the
number of elementary charges transferred, R is the universal gas
constant, T is the absolute temperature and Erev is the reversal

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Ih channels in leech neurons


potential. The conductance ratio (gNa/gK) can be calculated using
the formalism from Hodgkin and Huxley (Hodgkin and Huxley,
1952) (see Aidley, 1989):
gNa ( EK Erev )
=
,
gK ( Erev ENa )

(2)

where EK and ENa are the equilibrium potentials for K+ and Na+,
respectively. At a given Em, Ih can be calculated by (Aidley, 1989):
Ih gK(Em EK) + gNa(Em ENa).

(3)

By combining Eqns 2 and 3, gNa and gK, as well as the currents


carried by Na+ and K+, can be calculated. For the intracellular and
extracellular concentrations of Na+ and K+, data obtained with ionsensitive microelectrodes were used (see below); these data were
also used to calculate the equilibrium potentials of Na+ and K+.
Experimental data was analysed and fitted using Origin Software
(Microcal, Northampton, MA, USA). Boltzmanns equation was
used to calculate the voltage sensitivity of Ih activation, Hills
equation was used to calculate the doseresponse curve, and single
or sums of e-functions were used to describe the activation and
deactivation kinetics of the current.
RESULTS
Voltage sag in identified leech neurons

The injection of a negative current into Retzius neurons or into the


mechanosensory T, P and N neurons induced a membrane
hyperpolarisation that was maximal after approximately 50ms, but
subsequently became smaller. The Em reached a less negative plateau
a few hundred milliseconds later (Fig.1). This voltage sag was
virtually abolished after the addition of 15mmoll1 Cs+ to the bath
(see Fig.11A), indicating that it was mediated by Ih channels. The
amplitude of the voltage sag was determined as the depolarisation
between the peak of the negative voltage deflection and the potential
at the end of a given current injection (Table1). A voltage sag was
absent in Leydig, AE and AP neurons (see Fig.1). The voltage sag
was largest by far in P neurons (see Fig.1, Table1), prominent in
N neurons and small in Retzius neurons, where a voltage sag was

Table1. Size of the voltage sag in touch (T), pressure (P), noxious
(N) and Retzius neurons
Cell type

Sag amplitude (mV)

Em@ICC (mV)

ICC (nA)

T
P
N
Retzius

1.81.0
24.74.7
9.04.2
2.11.9

10021
1056
10717
10111

2
3
2
6

4
8
8
9

The sag amplitude was measured using injection currents (ICC) that
hyperpolarised the cells to approximately 100mV (Em@ICC, where Em is
membrane potential). Voltage sags were absent in Leydig, annulus
erector (AE) and anterior pagoda (AP) neurons (see Fig.1).

discernible in the majority of cells (five out of nine). The smallest


voltage sag was observed in three out of four tested T neurons.
The functional properties of the Ih channels were investigated in
more detail in P neurons, because, in these cells, the voltage sag
was largest. Furthermore, P neurons are sufficiently large and robust
to routinely and reliably allow the application of the two-electrode
voltage-clamp technique.
Ih in leech P neurons

A typical voltage-clamp experiment showing the activation of Ih in


a P neuron is presented in Fig.2. Initially, the Em of the cell was
clamped to a reference potential of 50mV, which was slightly more
negative than the mean Em of leech P neurons (44.34.9mV, N67)
(see Schlue and Deitmer, 1984). This required a holding current of
0.40.3nA (N40). Clamping the neuron to 80mV induced an
inward current (1 in Fig.2B), reflecting the net change in
membrane currents through ion channels that were already active
at 50mV (instantaneous current, Ii). Subsequently, an additional
current mediated by Ih channels developed, which was evident from
its biophysical and pharmacological properties (see below). Ih was
maximal after approximately 0.5s and persisted throughout the
hyperpolarising pulse, indicating that the channels did not inactivate.
Provided that the Ii remained constant during the hyperpolarising
pulse, the Ih that was activated when stepping from 50 to 80mV
was obtained by subtracting Ii from the current at the end of the

20
Retzius
40

6 nA

3 nA

4.5 nA

P
4 nA

60
80

Em (mV)

100
120
500 ms
20
Leydig
2 nA

40
60

AE
5 nA

5 nA

AP

229

Fig.1. Voltage sags in identified leech neurons.


Upon injection of a hyperpolarising current of the
given amplitude, Retzius neurons as well as the
mechanosensory touch (T), pressure (P) and
noxious (N) neurons showed voltage sags; i.e.
after reaching a maximum hyperpolarisation, the
membrane potential (Em) repolarised partially.
After cessation of the injection current, a transient
depolarisation occurred, which in P neurons
sometimes caused the generation of one or more
action potentials. In Leydig, annulus erector (AE)
and anterior pagoda (AP) neurons, voltage sags
were not observed. The injection current (2 to
6nA for 500 or 800ms) was adjusted such that
the cells initially hyperpolarised to approximately
100mV. Note the occurrence of spontaneous
action potentials in Leydig, AE and AP neurons.
The action potentials of P and Leydig neurons are
truncated.

80
100
120

THE JOURNAL OF EXPERIMENTAL BIOLOGY

230

E. Gerard and others

50 mV

80 mV

4
3

Tail current

Instantaneous current, Ii

2
Steady-state current, ISS

Hyperpolarisation-activated current, Ih

Fig.2. Hyperpolarisation-activated cation (Ih)


channels in a leech P neuron. Shifting Em from the
reference potential of 50 to 80mV (A) induced
the slow activation of a persistent inward current
(B), which is mediated by Ih channels, as evident
from the biophysical and pharmacological
properties. The amplitude of Ih is given by the
difference between the instantaneous current at
the beginning of the voltage step (Ii; 1) and the
membrane current at its end (steady-state or slow
inward current, ISS; 2). After return to the reference
potential, a slowly falling tail current remained (3),
which reflects the deactivation of the Ih channels.
After complete deactivation, the current returned to
resting values (4). The holding current at 50mV
was 0.40.3nA (N40).

1 nA
200 ms

pulse (steady-state or slow inward current, ISS; 2 in Fig.2B). After


returning to 50mV, an inwardly directed tail current (3 in Fig.2B)
remained, which slowly declined as the Ih channels deactivated (4
in Fig.2B).
The activation of Ih was evident at Em values slightly more
negative than 50mV (Fig.3). With increasing hyperpolarisation,
Ih became more and more prominent. At 80mV, it was
approximately as large as Ii (Fig.3B, Fig.4B). The tail current
saturated with increasing, preceding hyperpolarisation, indicating
that Ih channel activation was all but complete at approximately
80mV (see Fig.5).
The dependence of the Ii on the Em was linear, the slope
corresponding to an average input resistance of 23M (Fig.3B, red
line). In contrast, the membrane current at the end of the
hyperpolarising pulse increased over-proportionally because of the

increasing activation of Ih channels. At Em values more negative


than 70mV, the slope of the currentvoltage relationship appeared
to approach a constant value, corresponding to an input resistance
of 9M. At still more negative potentials the slope became smaller
again, corresponding to an input resistance of 14M (see Fig.8B,
black squares).
Kinetics of activation and deactivation

The speed of Ih activation increased with growing membrane


hyperpolarisation, as was previously described for other preparations
(e.g. Budde et al., 2008) (see Fig.3A). We approximated the
activation kinetics by exponential functions as illustrated in Fig.4,
which shows the membrane currents induced by hyperpolarising
the cell membrane to 60 or 80mV. When the time course of Ih
activation was fitted by a single exponential function, time constants

B
50 mV
Em (mV)
85

80

75

70

65

60

55

50

45
0.0

80 mV

1.0
Ii
1.5
2.0

1 nA
ISS
200 ms

Holding current (nA)

0.5

2.5
3.0

Fig.3. Voltage dependence of Ii and ISS. (A)Membrane currents upon shifting Em from 50 to 80mV (in 2mV steps) for 2s, recorded at intervals of 30s.
With increasing hyperpolarisation, the activation of Ih became progressively faster, while the decay of the tail current appeared to follow an invariant time
course. (B)Relationship between membrane current and Em, either at the beginning of the hyperpolarising step (Ii, open circles) or at its end, after activation
of the Ih channels (ISS, closed circles). Data points are means s.d. of 23 experiments. The voltage dependence of Ii was well fitted by a straight line (red
line), the slope of which corresponded to an input resistance of 23M. In contrast, ISS increased over-proportionally because of the increasing activation of
the Ih channels. The difference between the two currentvoltage relationships (ISSIi) gives the amplitude of Ih activated by the respective voltage step.

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Ih channels in leech neurons

231

Em (mV)

60 mV

130 120 110 100

=368 ms

90

80

70

60

50

40

30
0
0.1

0.3

0.4

80 mV
0.5

Tail current (nA)

0.2

0.6

=158 ms
1=78 ms
2=557 ms

0.7
0.8

1 nA
0.9

200 ms
0
0.2 nA
0
Fig.4. Kinetics of Ih activation and deactivation. (A)At moderate
hyperpolarisations (e.g. 60mV), the activation of Ih was well fitted by a
single exponential function. The upper trace shows the membrane current
upon a hyperpolarisation from 50 to 60mV (black trace), together with
the approximation of Ih by a single exponential function with t368ms and
an amplitude of 0.38nA (red curve). The lower trace shows the difference
between recorded trace and calculated curve at an increased
magnification, allowing an assessment of fit quality. See B for scale bars.
(B)At stronger hyperpolarisations, the approximation of Ih by a single
exponential function became progressively worse, but we obtained a
satisfying fit using a sum of two exponential functions. The black trace
shows the membrane current upon a hyperpolarisation to 80mV, together
with the approximation by a single exponential function (t158ms; red
curve) or the sum of two exponential functions (t178ms, t2557ms; green
curve). The fast component contributed 0.93nA to the total current and
the slow component contributed 0.33nA. The lower two traces show the
difference between the recorded trace and the curves calculated with one
exponential function (red) or the sum of two exponential functions (green).
The decline of the tail currents in A and B could be reasonably fitted by a
single exponential function with a time constant of approximately 200ms.

of 36874ms (for 60mV) and 15842ms (for 80mV) were


obtained (N12). At an Em of 60mV, the recorded trace and
calculated curve corresponded very well to each other, but deviations
were evident at 80mV (see Fig.4B, red trace). At this holding
potential, a good fit was obtained by the sum of two exponential
functions with time constants of 7822 and 55784ms, whereby
the fast component contributed approximately 75% to the total Ih.
The decay of the tail current could be approximated by a single
exponential function, with time constants ranging from 150 to
230ms, and it was independent of the strength of the preceding
hyperpolarisation (data not shown). In some recordings, the
activation of Ih appeared to follow a slightly sigmoidal time course
(see traces in Fig.2B, Fig.3A, Fig.4B; see Discussion).
Voltage dependence of Ih channel activity

The tail current was recorded at constant electromotive forces, so


that its amplitude should be proportional to the activity of the Ih
channels at the end of the hyperpolarising pulse. The dependence
of the tail current on the preceding Em is shown in Fig.5. An
approximation of the experimental data by a Boltzmann function
gave a half-maximal activation of the Ih channels at 65mV.

Fig.5. Dependence of Ih channel activity on Em. Relationship between the


tail current recorded at 50mV and Em during the preceding
hyperpolarisation. Data were obtained from 21 P neurons. In seven cells,
Em was shifted in 5mV steps up to 120mV and, in the remaining 14 cells,
in 2mV steps up to 80mV (pulse duration 2s, pulse interval 30s). The
continuous line represents an approximation of the data by a Boltzmann
function, according to which the Ih channels are half-maximally activated at
Em65mV.

According to this approximation, the activation of Ih channels started


at 30mV, and at 100mV the channels were fully activated. The
data in Fig.5 also illustrate the large variability of the tail current
in individual cells. Following hyperpolarisation to 80mV, the tail
current amplitudes ranged between 0.1 and 0.8nA.
Erev and ion permeability

The Erev of Ih was determined by analysing the current flow through


fully activated Ih channels at different Em (Fig.6A). For this
purpose, Em was shifted very briefly to a defined test potential and
Ii at this potential was determined. Then, after a short pause, the
cells were hyperpolarised to 100mV for 3s to achieve complete
Ih channel activation. Finally, Em was clamped again to the test
potential. The membrane current measured at this time point is the
sum of the Ii and the current mediated by the fully activated Ih
channels. Its amplitude ideally depends on electromotive driving
forces only.
The dependence of the maximal Ih on Em is shown in Fig.6B.
Linear fits of the data gave a mean Erev of 3517mV (N7). Using
this value and data on the intracellular and extracellular
concentrations of Na+ and K+ previously obtained in our laboratory
with ion-sensitive microelectrodes, pNa/pK was calculated to be 0.21.
The calculation is based on the following values: [K+]o5.8mmoll1
(at a bath concentration of 4mmoll1 K+) (see Schlue and Deitmer,
1980), [K+]i97mmoll1 (Schlue, 1991) (see also Klees et al., 2005;
Schlue and Deitmer, 1984), [Na+]o89mmoll1 (assumed to
correspond to the Na+ concentration of the bath solution) and
[Na+]i7.5mmoll1 (Hintz, 1999; Lucht, 1998) (see also Schlue,
1991). The equilibrium potentials were calculated to ENa+62mV
and EK72mV, and from the electromotive forces at Erev we
obtained a gNa/gK of 0.38. It could be argued that [K+]o should be
identical to the bath. In this case, pNa/pK would be 0.25 and gNa/gK
would be 0.47. However, when the ganglion capsule was intact, the
extracellular potassium concentration in the nerve cell body region
was determined to be 5.80.6mmoll1 (mean s.d., N27) (Schlue
and Deitmer, 1980), whereas it was identical to the bath only after

THE JOURNAL OF EXPERIMENTAL BIOLOGY

232

E. Gerard and others

50 mV

90

2s

80

70

Em (mV)
60
50

40

30
0.5

80 mV

80 mV

100 mV

1.0
1.5

2s

Ih (nA)

0.5

2.0
2.5
3.0
Ih

3.5

1 nA
50 ms

1s

Fig.6. Reversal potential of Ih. (A)Experimental protocol. To determine the instantaneous membrane current (Ii) at a given test potential (here, 80mV), Em
was shifted correspondingly for 50ms (upper trace). During this period, the membrane current increased slightly, probably because of the onset of Ih
channel activation (lower trace). Ii was estimated by visual extrapolation of the recorded trace to the beginning of the voltage jump (dark grey circle). After a
2s pause, Em was shifted to 100mV for 3s to achieve full activation of the Ih channels. Subsequently, Em was shifted back to the test potential, in order to
determine the maximum Ih at this potential. For this, the previously recorded Ii was subtracted from the current recorded immediately after the potential jump
from 100mV to the test potential (light grey circle). This protocol was applied every 10s at different test potentials to determine the membrane currents
mediated by the fully activated Ih channels at 30 to 90mV in 10mV increments. (B)Currentvoltage relationships. The data points represent the
membrane currents mediated by the fully activated Ih channels at different Em, as recorded from seven P neurons. In two cells, a reversal of Ih was not
observed; in one cell Ih could only be measured between 90 and 60mV. An error-weighted linear fit of the averaged data gave an Erev of Ih of 35mV
(N7, red line).

opening the ganglion capsule. Because the ganglion capsule was


always intact in our experiments, we assumed [K+]o was equal to
5.8mmoll1 for all subsequent calculations and discussions.
Block by external Cs+ and insensitivity to ZD7288

A characteristic feature of Ih channels is the blockade by


submillimolar concentrations of extracellular Cs+ (DiFrancesco,
1982; Kaupp and Seifert, 2001; Pape, 1996). In P neurons, Ih was
half-maximally blocked at a Cs+ concentration of 0.3mmoll1. At

2 mmol l1 Cs+
3 min

5mmoll1 Cs+, the block was virtually complete (Fig.7). The onset
of the block was fast and reversible within a few minutes. The time
course of Ih activation was unaffected. The extracellular application
of Ba2+, Co2+ or Ni2+ (5mmoll1 each), or of the K+ channel blocker
tetraethylammonium (TEA, 20mmoll1), for up to 10min had no
effect on Ih (data not shown).
The bradycardic agent ZD7288 selectively blocks Ih channels in
a variety of vertebrate and invertebrate preparations. The halfmaximal effect is mostly exerted at concentrations of 10 to

1.0
3 min

Relative Ih

0.8

0.5 nA

0.6

0.4

0.2

500 ms
0
0

1
Cs+ (mmol l1)

10

Fig.7. Ih is blocked by Cs+. (A)The presence of 2mmoll1 Cs+ reversibly suppressed Ih. Traces show the membrane current upon shifting Em from 50 to
80mV for 1s. Wash-in and wash-out time of Cs+ was 3min, as indicated. (B)Ih in solutions containing various Cs+ concentrations were normalised to the
mean current recorded before and after Cs+ application. The approximation of the data by using the Hill formalism (red line) gave a half-maximal block of Ih
at a Cs+ concentration of 0.3mmoll1. The Hill coefficient was 0.98. Data points are means s.d. of four to six experiments.

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Ih channels in leech neurons

233

B
Em(mV)
110 100

90

80

70

60

50

0
1

110 mV

Ii

Iss, Na+-free

4
Iss

Holding current (nA)

50 mV

5
0.5 nA

500 ms
Fig.8. Ih in Na+-free solution. (A)Membrane currents upon shifting Em from 50 to 110mV (in 5mV steps) for 2s, recorded at intervals of 30s. The
recordings were started 3min after replacing extracellular Na+ by NMDG+. In Na+-free solution, Ih was virtually abolished. The holding current for clamping
the cells to the reference potential was slightly reduced (0.20.16nA, N7, cf. Fig.2). (B)Relationship between the membrane current and Em, either at the
beginning of the hyperpolarising step (Ii; open circles) or at its end, after activation of the Ih channels (ISS, Na+-free; closed circles). Data points are means
s.d. of five experiments. The voltage dependence of Ii was fitted by a straight line, the slope of which indicated an input resistance of 27M (see Fig.3). For
comparison, the ISS recorded in SLS is also given (black squares; means s.d., N7).

50moll1 (see Gasparini and DiFrancesco, 1997; Pirtle et al., 2010;


Shin et al., 2001). In P neurons, however, even concentrations of
110mmoll1, applied for up to 10min, had no effect on Ih.
Ih in Na+-free solution

After replacing extracellular Na+ by NMDG+, Ih was virtually


abolished (Fig.8). At 100mV, the mean Ih amplitude was 0.2nA,
as compared with 2.2nA in SLS (Fig.8B). The tail current was
inverted and strongly reduced, with maximum amplitudes of
approximately ~0.1nA (cf. Fig.5); the time constant of deactivation
varied between 250 and 400ms. The voltage dependence of
activation appeared to be unchanged: a Boltzmann approximation
gave a half maximal activation at approximately 70mV (data not
shown). The effects of replacing extracellular Na+ by NMDG+ were
fully reversible within a few minutes.

Replacing extracellular Na+ by Li+ also suppressed Ih and


inverted the tail current (data not shown). The current for clamping
the cells to 50mV was increased (+1.30.6nA, N6). These effects
occurred within 1min after exchanging Na+ for Li+. Recovery was
poor and, even 20min after return to SLS, only a weak Ih was
detectable. The tail current remained inverted and the current for
clamping the cells to 50mV often stayed increased.
Effect of extracellular K+

After reducing the K+ concentration of the bath solution, Ih and


the tail current were strongly diminished (Fig.9). Both currents
were markedly increased when the extracellular K+ concentration
was raised. At low K+ concentrations, Erev of Ih was shifted towards
more negative values, corresponding to the shift in the K+
equilibrium potential. Under nominally K+-free conditions, it was

B
[K+]o (mmol l1)

50 mV
K+-free

48

16

80 mV

12

1
Ih (nA)

K+-free
1
2
4

7
40
8

1 nA

5
3

8
200 ms
16 mmol l1 K+

Fig.9. Effect of extracellular K+ on Ih. (A)Superimposed membrane currents upon shifting Em from 50 to 80mV for 2s, recorded in the same P neuron
5min after adjusting the K+ concentration in the bath as indicated. For better comparison, the traces were slightly shifted vertically to bring the instantaneous
currents recorded immediately after shifting Em to congruence. (B)Amplitudes of Ih at 80mV at different K+ concentrations in the bath solution. Data are
means s.d. of five to 40 experiments (N-values are shown to the right of the error bars).

THE JOURNAL OF EXPERIMENTAL BIOLOGY

E. Gerard and others

A
80

75

70

Em (mV)
65 60

B
55

50
0

80

75

70

Em (mV)
65
60

55

0.5

ISS

0.2

1.5

0.3

2.0

0.4

2.5
3.0

80

75

70

65

60

55

0.1

1.0

50
0
0.5

Holding current (nA)

Ii

0.5

D
80

75

70

65

1.0
Ii

50

60

55

50
0

Tail current (nA)

234

0.5

1.5
2.0

1.0

2.5
ISS

3.0
3.5

1.5

Fig.10. Effect of cyclic nucleotides on Ih. (A,C)Ii (open symbols) and ISS (closed symbols) in the presence (circles) and absence (squares) of dibutyryl-cAMP
(dbcAMP, 1mmoll1; A) or the phosphodiesterase blocker IBMX (1mmoll1; C). For clarity, error bars are only shown exemplarily. Although Ih was
decreased by ~50% in the presence of dbcAMP, the instantaneous current was unchanged. In IBMX, the amplitude of Ih remained unchanged but an overall
decrease in the holding current was observed. (B,D)Relationship between tail currents at 50mV and the preceding membrane hyperpolarisation, measured
in the presence of dbcAMP (B) or IBMX (D). The recordings were started 10min after adding dbcAMP or IBMX to the bath solution; single data points were
measured at 10s intervals. The continuous lines indicate approximations of the data by the Boltzmann equation, according to which the Ih channels were
half-maximally activated at 66mV (dbcAMP) or 67mV (IBMX). Note that in the presence of IBMX, the maximum tail current was similar to that in SLS,
whereas it was reduced by more than 50% in the presence of dbcAMP (cf. Fig.5).

measured to be 6323mV (N12). Surprisingly, the Erev at


increased bath K+ concentrations were also found to be more
negative than in SLS (8mmoll1 K+: 405mV, N8; 16mmoll1:
3913mV, N5).
Effects of cAMP and IBMX

In most cells, cAMP shifts the voltage range of Ih channel activation


to more positive values (see Kaupp and Seifert, 2001). In the
presence of the membrane-permeable cAMP analogue, dbcAMP,
or the phosphodiesterase inhibitor, IBMX (each 1mmoll1), the Ih
channels of leech P neurons were half-maximally activated at
virtually the same Em as in SLS (Fig.10). Amplitudes of Ih and tail
current were reduced by approximately 50% upon dbcAMP
application, whereas they were unchanged in the presence of
IBMX. The instantaneous current was almost unaffected by both
dbcAMP and IBMX.
Effect of Ih on membrane excitability

The activation of Ih channels reduced the depolarising current


necessary to elicit an action potential (Fig.11A), suggesting an
increased excitability. Furthermore, approximately 30% of the cells
generated action potentials without the injection of a depolarising
current upon repolarisation from hyperpolarised potentials (see
Fig.1). These effects cannot be attributed to accommodation,
because a membrane hyperpolarisation in the presence of millimolar
concentrations of Cs+ never facilitated action-potential generation
(Fig.11A).

After activation of the Ih channels, the afterhyperpolarisation


following an action potential shortened and subsequently reached
a transient maximum (Fig.11B, arrow). We never observed this
when there was no preceding activation of the Ih channels. Ih channel
activation did not affect amplitude and time course of the action
potential. In the presence of Cs+, amplitude, afterhyperpolarisation
and latency (stimulus to peak) of the action potentials were reduced,
whereas the time course remained unchanged (Fig.11C).
Ih in other leech neurons

When using the voltage-clamp technique to assess Ih in AP, Leydig,


N, T and Retzius neurons, Ih could be excluded in Leydig but not
in AP neurons. The maximum Ih current varied from cell type to
cell type and occurred at voltages between 80 and 120mV
(Table2). As Ii may already contain Ih, we additionally determined
the difference between the steady-state currents under control
conditions (i.e. with activated Ih) and in the presence of 2mmoll1
Cs+ (i.e. with Ih blocked; Cs+-sensitive current, ICs; Table2). We
found that the Cs+-sensitive current was consistently larger than the
Ih current as determined from the difference between ISS and Ii. This
corroborates that Ih is to some degree active at 50mV, and/or that
there is a rapidly activating Ih component, which is included in Ii,
as has been reported previously (Budde et al., 2008; Mistrk et al.,
2006; Proenza et al., 2002). When ICs is set in relation to the total
holding current at a given holding potential, thus giving an estimate
of the contribution of Ih to the overall membrane conductance, it
becomes apparent that the mechanosensory N and P neurons have

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Ih channels in leech neurons


40

235

20
0

Em (mV)

20

3 nA

40
60
+5 nA

80
100

Cs+

120
140

100 ms

10 ms

100 ms

10 ms

10 ms

Fig.11. Effect of Ih channel activation on membrane excitability and action potential waveform. (A)Membrane excitability. In this P neuron, action potentials
could be reliably elicited by injecting a depolarising current of +5nA for 5ms, provided that the Ih channels had been activated by a preceding
hyperpolarisation (3nA for 0.3s; black trace). However, if the Ih channels were blocked by 1mmoll1 Cs+, action potentials were never generated in
response to this stimulus (red trace). Each trace represents the superimposition of five single recordings to demonstrate the reproducibility of the waveforms.
The effect of Cs+ was fully reversible within a few minutes. Identical results were obtained from five cells in total. (B)Action potential waveform. After
activation of the Ih channels by a hyperpolarising current (3nA for 0.3s), the afterhyperpolarisation following an action potential was shortened, and
subsequently reached a transient maximum (red trace, arrow). This was never observed without preceding Ih channel activation (black trace). Amplitude and
time course of the action potentials were not affected by Ih channel activation. As in A, each trace represents the superimposition of five single recordings.
Traces were horizontally shifted to bring action potentials into congruence. In this P neuron, the injection of +2nA for 5ms was sufficient to reliably elicit an
action potential, even without Ih channel activation. (C)Effect of Cs+ on action potentials. Superimposed traces of five consecutively elicited action potentials
under control conditions (black) and in the presence of 1mmoll1 Cs+ (red). Traces were horizontally shifted to bring action potentials into congruence. Cs+
reduced the peak amplitudes and the afterhyperpolarisation. Moreover, the latency from the beginning of the current injection (arrows mark the stimulus
artefact, +2nA for 10ms) to the peak of the action potential was also reduced. The effect of Cs+ was fully reversible. Similar results were obtained for five
cells in total.

the largest Ih, followed by T, AP and Retzius neurons, whereas in


Leydig neurons Ih appears to be virtually absent. We note that the
values in Table2 can only be a rough estimate, as Cs+ may inhibit
conductances other than Ih.
DISCUSSION

For leech HN cells, and briefly for Retzius and P neurons, Ih has
been described before (Angstadt, 1999; Angstadt and Calabrese,
1989; Arbas and Calabrese, 1987b; Coulon et al., 2008; Klees et
al., 2005). However, a detailed description of the distribution and
biophysical properties of leech Ih across cell types was lacking. The
findings presented here can be summarised as follows. First, voltage
sags and the underlying Ih were observed in T, P, N and Retzius
neurons. Second, the current was largest in P neurons. Third, the
biophysical properties of the current largely matched those described
in other preparations: (a) half-maximum activation occurred at
65mV, complete activation at 100mV; (b) the activation could
be fitted by a single time constant of ~370ms at 60mV but required
two time constants of ~80 and ~560ms at 80mV; (c) Erev was
determined to be 3517mV; and (d) the current mutually depended
on both extracellular Na+ and K+, with a pNa/pK of 0.21. Fourth,
half-maximal block of Ih was achieved at a Cs+ concentration of
0.3mmoll1. The current was insensitive to ZD7288. Fifth, after
elevation of the intracellular cAMP concentration, Ih in leech P
neurons showed a blockade, rather than a voltage shift of the
activation curve. Sixth, the channels caused bursts of APs when a
neuron was relieved from hyperpolarisation and modulated the form
of afterhyperpolarisations. And finally, the function of Ih seems to
include a subtle alteration of firing behaviour in leech
mechanosensory neurons as well as a stabilising effect on Em.

Activation kinetics

In some experiments, it was evident that Ih activation followed a


sigmoidal time course (see Fig.3A), as was found previously in leech
heart interneurons [see fig.2 in Angstadt and Calabrese (Angstadt
and Calabrese, 1989)] and in most other systems (see Hestrin, 1987;
Kaupp and Seifert, 2001). This phenomenon was interpreted as a
delay-to-onset of the time-dependent component of Ih (Angstadt and
Calabrese, 1989), i.e. the delay between the instantaneous
component and the voltage-activated component of the membrane
current. Unfortunately, in our experiments the early lag phase was
Table2. Hyperpolarisation-activated cation currents (Ih) were
observed in T, P, N, AP and Retzius neurons
Cell type

Ih (nA)

ICs (nA)

%ICs of ISS

AP
Leydig
N
P
T
Retzius

0.10.2
+0.40.2
0.60.4
1.90.4
0.20.3
0.40.4

0.80.6
0.20.8
0.91.0
2.81.4
0.40.3
1.41.7

159
0.227
3929
4724
169
1019

5
5
8
7
4
8

Ih was measured 500ms after a voltage step from 50mV to the holding
potential that gave the largest Ih (see Fig.2). When the current required for
the voltage step in the presence of Cs+ was deducted from the current
measured under control conditions with active Ih (ICs), the resulting
values were always larger. The contribution of the current blocked by Cs+
to the overall current required for the voltage step (%ICs of ISS) reveals
that N and P neurons have the largest relative Ih and will thus be
influenced most by Ih activation. Because AE neurons are rather small and
difficult to access using the two electrode current- and voltage-clamp
techniques, AE neurons were not analysed (but see Fig.1).

THE JOURNAL OF EXPERIMENTAL BIOLOGY

236

E. Gerard and others

short and mostly hidden by the initial oscillations in the holding


current. When the early lag was ignored, the activation of Ih could
be well approximated by a single exponential function, if Em was
not too negative (Fig.4A). At potentials that were more negative,
a reasonable fit was only possible by applying a sum of two
exponential functions (Fig.4B). This biphasic activation has been
reported previously (Budde et al., 1994; Pape, 1996) and was
attributed to the existence of two distinct populations of Ih channels.

In some systems that selectively express HCN1, cAMP shifted the


activation curve by only a few millivolts (Santoro et al., 1998) or
not at all (Shin et al., 2001). The insensitivity of the voltage
dependence of activation to cAMP points to the expression of the
HCN1 and HCN3 subtypes in leech neurons. In support of HCN1
expression is the comparably fast activation kinetics (Shin et al.,
2001). In support of HCN3 expression is the inhibition by cAMP.

Ion selectivity

The dependence of Ih on the presence of extracellular Na+ and K+


suggests that the fluxes of Na+ and K+ through the Ih channels are
dependent on one another. Thus, after replacing extracellular Na+,
Ih was all but absent. Under these conditions, a small Na+ efflux
can be expected. However, this was not taken into account for the
calculations, as the intracellular Na+ concentration drops in the
absence of extracellular Na+ to values of 1mmoll1 and lower (Hintz,
1999), so that a Na+ efflux should be negligibly small. If, under
these conditions, the K+ component of the current persisted, an
inward current of 0.59nA would be expected at 100mV
(gK21nS). However, the measured current was on average only
0.21nA, indicating that gK was reduced to roughly one-third of its
original value in the absence of extracellular Na+.
Ih was suppressed in the absence of K+ in the bath solution (Fig.9),
arguing in favour of a mutual dependency of the two ions on one
another within the channel pore. In a K+-free bath solution, the
extracellular K+ concentration in the vicinity of the neurons somata
was measured to be 1.60.6mmoll1 (Schlue and Deitmer, 1980).
Under K+-free conditions, the lowest measured intracellular K+ was
78mmoll1 (Schlue, 1991; Schlue and Deitmer, 1984) and
intracellular Na+ increased to ~20mmoll1 (Lucht, 1998). With these
ion concentrations and assuming that extracellular Na+ remains at
89mmoll1, Ih can be calculated with the given gK and gNa to
0.56nA at 80mV. However, the measured current was only
0.330.16nA (N12; Fig.9B), suggesting that gNa was reduced by
at least 40% in the K+-free bath solution. This mutual dependency
of ion conductances within a channel pore is a characteristic feature
of multi-ion pores (Hestrin, 1987; Hille, 2001).

Multi-ion pores

From the determined Erev of Ih (3517mV, N7; Fig.6), pNa/pK


was calculated to be 0.21 by using the Goldman equation. This
calculation was based on intracellular and extracellular
concentrations of Na+ and K+ measured previously by ion-sensitive
microelectrodes. gNa/gK was determined to be 0.38. Comparison with
permeability and conductance ratios obtained from other
preparations reveals that the values calculated here are within the
same range [pNa/pK between 0.2 and 0.4 (for a review, see Pape,
1996); gNa/gK between 0.25 and 0.67 (e.g. Bolvar et al., 2008; Macri
et al., 2002)].
At 80mV, Ih was on average approximately 1.3nA (see Figs2,
3). With this value and the above conductance ratio, gK is 21nS
and gNa is 8nS. These values, in turn, allow us to calculate the tail
current when returning from 80mV to 50mV to 0.44nA, which
corresponds nicely with the experimental data (see Fig.5).
The suppression of Ih after replacing Na+ by Li+ shows that the
channels are virtually impermeable to Li+, which is in accordance
with previous findings (Biel et al., 2009; Pape, 1996; Wollmuth and
Hille, 1992).
Pharmacological profile of Ih in leech neurons

One of the characteristics of Ih is its sensitivity to relatively low


concentrations of Cs+ [0.15mmoll1 (Pape, 1996)]. Cs+ also blocks
other voltage-dependent ion channels, such as delayed rectifier and
inward rectifier K+ channels, complicating its use as a sole criterion
for the identification of Ih. In leech P neurons, Ih was insensitive to
Ni2+, Co2+ or Ba2+(5mmoll1 each), as well as to TEA+ (20mmoll1;
data not shown).
Another characteristic of Ih is its sensitivity to ZD7288. Leech P
neurons were, however, completely insensitive to ZD7288 at
concentrations as high as 10mmoll1. Nevertheless, in our view,
most other experimental data shown here strongly argue in favour
of the assumption that leech Ih channels are very similar, albeit not
identical, to classical, ZD7288-sensitive Ih channels found in other
preparations. Although it is beyond the scope of this work to
elucidate why ZD7288 does not block the channels, it is not
uncommon that known and established selective blockers are
ineffective in the leech preparation in known concentration ranges
(e.g. Johansen and Kleinhaus, 1986). Moreover, an ineffective
blockade of leech Ih by ZD7288 was reported previously for HN
cells (Tobin and Calabrese, 2005).
Leech Ih channels also show unusual behaviour concerning their
reactions to cAMP. Ih channels are often regulated by cAMP such
that they shift their voltage dependency of activation towards more
positive potentials, whereas kinetics and amplitudes remain largely
unaltered (Kaupp and Seifert, 2001; Pape, 1996; Pape and
McCormick, 1989). In leech P neurons, the membrane-permeable
analogue dbcAMP did not change the voltage dependency of
activation, but instead caused a 50% reduction of the current. Lack
of regulation by cAMP has been reported before and has been
attributed to both HCN1 and HCN3 subunits of the channel, with
HCN3 even being slightly inhibited by cAMP (Biel et al., 2009).

Ih or the lack thereof in other neurons of the leech

Leydig neurons fire spontaneously at low rates [<4Hz (Arbas and


Calabrese, 1990)] and were found to regulate heart rate in the leech.
We show here that Ih is all but absent in Leydig neurons, so that
the intrinsic rhythm observed in these neurons cannot be mediated
by Ih. The only non-mechanosensory neurons showing considerable
Ih were Retzius neurons. In these cells, the current may affect
spontaneous action-potential generation (Angstadt, 1999) and does
not seem to be involved in volume regulation (Coulon, 2005) (see
Coulon et al., 2008). The voltage sag and Ih were smallest in T
neurons, and it is questionable whether such a small current can
have a functional relevance.
Physiological relevance

The nervous system of the medicinal leech is organised in ganglia


that contain various neurons with well-known functions. This offers
the opportunity to correlate physiological and functional aspects. The
hyperpolarisation-activated inward current of leech HN cells was
slightly different in its biophysical properties when compared with
the Ih described here in leech P neurons. It began to activate near
50mV, became fully activated between 70 and 80mV, and had
slow and voltage-dependent activation kinetics with single time
constants ranging from 2s at 60mV to 700ms at 100mV. It had
an Erev of 215mV and was blocked by 15mmoll1 extracellular

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Ih channels in leech neurons


Cs+ (Angstadt and Calabrese, 1989). HN cells interconnect by
inhibitory chemical synapses, and their normal electrical activity
consists of bursts of action potentials separated by periods of
inhibition. The function of Ih in HN cells seems obvious: the cells
hyperpolarise during the silent state so that Ih activates. This activation
depolarises the membrane and helps the cells escape from the
hyperpolarisation caused by inhibitory inputs. Accordingly, Cs+
slows (Masino and Calabrese, 2002; Tobin and Calabrese, 2005) or
even abolishes oscillatory behaviour (Angstadt and Calabrese, 1989).
The neurons described in this work displaying Ih were mostly
mechanosensory and not spontaneously active. Thus, a pacemaker
function of Ih like that described for HN cells can be excluded.
Instead, two further functions may exist in leech P neurons: (1) a
stabilising effect on Em and (2) a subtle alteration of the firing
behaviour. These neurons respond to mechanical stimuli of the skin
with barrages of action potentials, proportional to the stimulus
intensity, followed by a prominent hyperpolarisation that lasts for
seconds or minutes and is caused by the activation of Ca2+-activated
K+ channels and the activity of the Na+/K+-ATPase (Baylor and
Nicholls, 1969; Jansen and Nicholls, 1973; Scuri et al., 2002). This
hyperpolarisation will lead to the activation of Ih, which then helps
to determine both the strength and duration of this hyperpolarisation.
Moreover, a marked increase of the afterhyperpolarisation following
each action potential is observed after prolonged sensory stimulation
(Scuri et al., 2002). The afterhyperpolarisation is considered to have
two major functions: it limits the firing frequency of the neuron and
is responsible for generating spike-frequency adaptation. An
activation of Ih should alleviate both effects, and indeed, we
observed a shorter afterhyperpolarisation after Ih activation
(Fig.11B) as well as the occasional spontaneous generation of one
or several action potentials upon repolarisation. Hyperpolarisations
activate Ih, allowing counteracting depolarisations, so that the Em
is shifted back to its original value. Vice versa, depolarisations will
cause deactivation of Ih, which would result in a counteracting
hyperpolarisation. Additionally, after prolonged mechanosensory
stimulation, and the subsequent strong hyperpolarisation, the
activation of Ih could help to preserve the effect of mechanosensory
inputs by increasing the excitability. Although this could be a
common trait of mechanosensory neurons in the leech, it is unclear
whether this is a necessity for somatosensory signalling (i.e. a
mechanism to reduce spike-frequency adaptation) or an amplifying
mechanism for post-stimulus intervals (providing increased
sensitivity to stimuli). The idea of shunting inhibition by reducing
the input resistance of P neurons is refuted somewhat by the fact
that deactivation of Ih is slow: before it is complete, e.g. during
depolarisation from a hyperpolarised Em, Ih increases the neurons
excitability by a further depolarising tail potential. Additionally, it
could be argued that increased excitability and shortened
afterhyperpolarisation increase the neurons firing rate, thereby
possibly providing a more reliable sensory relay.
Although Ih is probably active near rest, the blockade with Cs+
did not hyperpolarise Em. This could be explained by an unspecific
action of Cs+. K+ channels that may be active in leech P neurons
could be blocked, e.g. the delayed rectifier and inward rectifier K+
channels (Pape, 1996). This would mask the hyperpolarisation
caused by blocking Ih with a depolarisation caused by blocking K+
conductances. To what extent this occurs in leech P neurons is,
however, unclear.
LIST OF SYMBOLS AND ABBREVIATIONS
AE
AP

annulus erector (neuron)


anterior pagoda (neuron)

cAMP
EK
Em
ENa
Erev
gNa/gK
HN
IBMX
Ih
Ii
ISS
N
P
pNa/pK
R
SLS
T
T
z
ZD7288

237

cyclic adenosine monophosphate


equilibrium potential for K+
resting membrane potential
equilibrium potential for Na+
reversal potential
conductance ratio
heart interneuron
3-isobutyl-1-methylxanthine
hyperpolarisation-activated cation current
instantaneous current
steady-state or slow inward current
noxious (neuron)
pressure (neuron)
permeability ratio
universal gas constant
standard leech saline
touch (neuron)
absolute temperature
number of elementary charges transferred
4-ethylphenylamino-1,2-dimethyl-6-methylaminopyrimidinium
chloride

ACKNOWLEDGEMENTS
The authors thank Claudia Roderigo and Simone Durry for excellent technical
assistance and Prof. Dr Thomas Budde and Dr Susan Sangha for helpful
comments on the manuscript.

FUNDING
This research received no specific grant from any funding agency in the public,
commercial, or not-for-profit sectors.

REFERENCES
Aidley, D. J. (1989). The Physiology of Excitable Cells. Cambridge: Cambridge
University Press.
Akasu, T., Shoji, S. and Hasuo, H. (1993). Inward rectifier and low-threshold calcium
currents contribute to the spontaneous firing mechanism in neurons of the rat
suprachiasmatic nucleus. Pflgers Arch. 425, 109-116.
Angstadt, J. D. (1999). Persistent inward currents in cultured Retzius cells of the
medicinal leech. J. Comp. Physiol. A 184, 49-61.
Angstadt, J. D. and Calabrese, R. L. (1989). A hyperpolarization-activated inward
current in heart interneurons of the medicinal leech. J. Neurosci. 9, 2846-2857.
Arbas, E. A. and Calabrese, R. L. (1987a). Ionic conductances underlying the activity
of interneurons that control heartbeat in the medicinal leech. J. Neurosci. 7, 39453952.
Arbas, E. A. and Calabrese, R. L. (1987b). Slow oscillations of membrane potential in
interneurons that control heartbeat in the medicinal leech. J. Neurosci. 7, 3953-3960.
Arbas, E. A. and Calabrese, R. L. (1990). Leydig neuron activity modulates heartbeat
in the medicinal leech. J. Comp. Physiol. A 167, 665-671.
Baylor, D. A. and Nicholls, J. G. (1969). Chemical and electrical synaptic connexions
between cutaneous mechanoreceptor neurones in the central nervous system of the
leech. J. Physiol. 203, 591-609.
Biel, M., Wahl-Schott, C., Michalakis, S. and Zong, X. (2009). Hyperpolarizationactivated cation channels: from genes to function. Physiol. Rev. 89, 847-885.
Bobker, D. H. and Williams, J. T. (1989). Serotonin augments the cationic current Ih
in central neurons. Neuron 2, 1535-1540.
Bolvar, J. J., Tapia, D., Arenas, G., Castanon-Arreola, M., Torres, H. and
Galarraga, E. (2008). A hyperpolarization-activated, cyclic nucleotide-gated (Ih-like)
cationic current and HCN gene expression in renal inner medullary collecting duct
cells. Am. J. Physiol. 294, C893-C906.
Budde, T., White, J. A. and Kay, A. R. (1994). Hyperpolarization-activated Na+-K+
current (Ih) in neocortical neurons is blocked by external proteolysis and internal
TEA. J. Neurophysiol. 72, 2737-2742.
Budde, T., Coulon, P., Pawlowski, M., Japes, A., Meuth, P., Meuth, S. G. and
Pape, H. C. (2008). Reciprocal modulation of Ih and ITASK in thalamocortical relay
neurons by halothane. Pflgers Arch. 456, 1061-1073.
Coulon, P. (2005). Elektrophysiologie der Volumenregulation von Retzius-Neuronen im
Zentralnervensystem des medizinischen Blutegels. PhD thesis, Heinrich-HeineUniversitt, Dsseldorf, Germany.
Coulon, P., Wsten, H. J., Hochstrate, P. and Dierkes, P. W. (2008). Swellingactivated chloride channels in leech Retzius neurons. J. Exp. Biol. 211, 630-641.
DiFrancesco, D. (1982). Block and activation of the pace-maker channel in calf
Purkinje fibres: effects of potassium, caesium and rubidium. J. Physiol. 329, 485507.
Doan, T. N. and Kunze, D. L. (1999). Contribution of the hyperpolarization-activated
current to the resting membrane potential of rat nodose sensory neurons. J. Physiol.
514, 125-138.
Gasparini, S. and DiFrancesco, D. (1997). Action of the hyperpolarization-activated
current (Ih) blocker ZD 7288 in hippocampal CA1 neurons. Pflgers Arch. 435, 99106.
Hestrin, S. (1987). The properties and function of inward rectification in rod
photoreceptors of the tiger salamander. J. Physiol. 390, 319-333.

THE JOURNAL OF EXPERIMENTAL BIOLOGY

238

E. Gerard and others

Hille, B. (2001). Ion Channels of Excitable Membranes. Sunderland, MA: Sinauer


Associates.
Hintz, K. (1999). Elektrophysiologische charakterisierung der Mg2+-regulation bei
identifizierten neuronen und neuropil-gliazellen im zentralnervensystem des
medizinischen blutegels. PhD thesis, Heinrich-Heine-Universitt, Dsseldorf,
Germany.
Ho, W. K., Brown, H. F. and Noble, D. (1994). High selectivity of the i(f) channel to
Na+ and K+ in rabbit isolated sinoatrial node cells. Pflgers Arch. 426, 68-74.
Hodgkin, A. L. and Huxley, A. F. (1952). A quantitative description of membrane
current and its application to conduction and excitation of nerve. J. Physiol. 117,
500-544.
Jansen, J. K. and Nicholls, J. G. (1973). Conductance changes, an electrogenic
pump and the hyperpolarization of leech neurones following impulses. J. Physiol.
229, 635-655.
Johansen, J. and Kleinhaus, A. L. (1986). Differential sensitivity of tetrodotoxin of
nociceptive neurons in 4 species of leeches. J. Neurosci. 6, 3499-3504.
Kaupp, U. B. and Seifert, R. (2001). Molecular diversity of pacemaker ion channels.
Annu. Rev. Physiol. 63, 235-257.
Klees, G., Hochstrate, P. and Dierkes, P. W. (2005). Sodium-dependent potassium
channels in leech P neurons. J. Membr. Biol. 208, 27-38.
Lucht, M. (1998). Elektrophysiologische und pharmakologische Charakterisierung von
5-Hydroxytryptamin-Rezeptoren und Second-messenger-Kaskaden bei identifizierten
Neuronen des Blutegel-Nervensystems. PhD thesis, Heinrich-Heine-Universtitt,
Dsseldorf, Germany.
Ludwig, A., Budde, T., Stieber, J., Moosmang, S., Wahl, C., Holthoff, K.,
Langebartels, A., Wotjak, C., Munsch, T., Zong, X. et al. (2003). Absence
epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2.
EMBO J. 22, 216-224.
Lupica, C. R., Bell, J. A., Hoffman, A. F. and Watson, P. L. (2001). Contribution of
the hyperpolarization-activated current (Ih) to membrane potential and GABA release
in hippocampal interneurons. J. Neurophysiol. 86, 261-268.
Macri, V., Proenza, C., Agranovich, E., Angoli, D. and Accili, E. A. (2002).
Separable gating mechanisms in a mammalian pacemaker channel. J. Biol. Chem.
277, 35939-35946.
Masino, M. A. and Calabrese, R. L. (2002). Period differences between segmental
oscillators produce intersegmental phase differences in the leech heartbeat timing
network. J. Neurophysiol. 87, 1603-1615.
McCormick, D. A. and Pape, H. C. (1990a). Properties of a hyperpolarizationactivated cation current and its role in rhythmic oscillation in thalamic relay neurones.
J. Physiol. 431, 291-318.
McCormick, D. A. and Pape, H. C. (1990b). Noradrenergic and serotonergic
modulation of a hyperpolarization-activated cation current in thalamic relay neurones.
J. Physiol. 431, 319-342.
Meuth, S. G., Kanyshkova, T., Meuth, P., Landgraf, P., Munsch, T., Ludwig, A.,
Hofmann, F., Pape, H. C. and Budde, T. (2006). Membrane resting potential of

thalamocortical relay neurons is shaped by the interaction among TASK3 and HCN2
channels. J. Neurophysiol. 96, 1517-1529.
Mistrk, P., Pfeifer, A. and Biel, M. (2006). The enhancement of HCN channel
instantaneous current facilitated by slow deactivation is regulated by intracellular
chloride concentration. Pflgers Arch. 452, 718-727.
Nicholls, J. (1987). The search for connections. In Studies of Regeneration in the
Nervous System of a Leech, Magnes Lecture Series, Vol. 2, p. 25. Sunderland, MA:
Sinauer Associates.
Pape, H. C. (1996). Queer current and pacemaker: the hyperpolarization-activated
cation current in neurons. Annu. Rev. Physiol. 58, 299-327.
Pape, H. C. and McCormick, D. A. (1989). Noradrenaline and serotonin selectively
modulate thalamic burst firing by enhancing a hyperpolarization-activated cation
current. Nature 340, 715-718.
Pirtle, T. J., Willingham, K. and Satterlie, R. A. (2010). A hyperpolarization-activated
inward current alters swim frequency of the pteropod mollusk Clione limacina. Comp.
Biochem. Physiol. A Mol. Integr. Physiol. 157, 319-327.
Proenza, C., Angoli, D., Agranovich, E., Macri, V. and Accili, E. A. (2002).
Pacemaker channels produce an instantaneous current. J. Biol. Chem. 277, 51015109.
Santoro, B., Liu, D. T., Yao, H., Bartsch, D., Kandel, E. R., Siegelbaum, S. A. and
Tibbs, G. R. (1998). Identification of a gene encoding a hyperpolarization-activated
pacemaker channel of brain. Cell 93, 717-729.
Schlue, W. R. (1991). Effects of ouabain on intracellular ion activities of sensory
neurons of the leech central nervous system. J. Neurophysiol. 65, 736-746.
Schlue, W. R. and Deitmer, J. W. (1980). Extracellular potassium in neuropile and
nerve cell body region of the leech central nervous system. J. Exp Biol. 87, 23-43.
Schlue, W. R. and Deitmer, J. W. (1984). Potassium distribution and membrane
potential of sensory neurons in the leech nervous system. J. Neurophysiol. 51, 689704.
Scuri, R., Mozzachiodi, R. and Brunelli, M. (2002). Activity-dependent increase of
the AHP amplitude in T sensory neurons of the leech. J. Neurophysiol. 88, 24902500.
Shin, K. S., Rothberg, B. S. and Yellen, G. (2001). Blocker state dependence and
trapping in hyperpolarization-activated cation channels: evidence for an intracellular
activation gate. J. Gen. Physiol. 117, 91-101.
Siddall, M. E., Trontelj, P., Utevsky, S. Y., Nkamany, M. and Macdonald, K. S.
(2007). Diverse molecular data demonstrate that commercially available medicinal
leeches are not Hirudo medicinalis. Proc. R. Soc. Lond B 274, 1481-1487.
Tobin, A. E. and Calabrese, R. L. (2005). Myomodulin increases Ih and inhibits the
Na/K pump to modulate bursting in leech heart interneurons. J. Neurophysiol. 94,
3938-3950.
Wollmuth, L. P. and Hille, B. (1992). Ionic selectivity of Ih channels of rod
photoreceptors in tiger salamanders. J. Gen. Physiol. 100, 749-765.
Wuttke, W. A. and Berry, M. S. (1992). Modulation of inwardly rectifying Na+-K+
channels by serotonin and cyclic nucleotides in salivary gland cells of the leech,
Haementeria. J. Membr. Biol. 127, 57-68.

THE JOURNAL OF EXPERIMENTAL BIOLOGY

Cell Calcium 46 (2009) 333346

Contents lists available at ScienceDirect

Cell Calcium
journal homepage: www.elsevier.com/locate/ceca

Burst discharges in neurons of the thalamic reticular nucleus are


shaped by calcium-induced calcium release
Philippe Coulon a,b, , David Herr a , Tatyana Kanyshkova a , Patrick Meuth a ,
Thomas Budde a,b , Hans-Christian Pape a,b
a
b

Institut fr Physiologie I, Westflische Wilhelms-Universitt Mnster, Robert-Koch-Str. 27a, 48149 Mnster, Germany
Institut fr Experimentelle Epilepsieforschung, Westflische-Wilhelms-Universitt Mnster, Robert-Koch-Str. 27a, 48149 Mnster, Germany

a r t i c l e

i n f o

Article history:
Received 15 June 2009
Received in revised form 1 September 2009
Accepted 27 September 2009
Available online 14 November 2009
Keywords:
Thalamic reticular nucleus
Ryanodine receptors
SERCA
CICR
Calcium
Two-photon microscopy

a b s t r a c t
The nucleus reticularis thalami (NRT) is a layer of inhibitory neurons that surrounds the dorsal thalamus. It
appears to be the pacemaker of certain forms of slow oscillations in the thalamus and was proposed to be
a key determinant of the internal attentional searchlight as well as the origin of hypersynchronous activity
during absence seizures. Neurons of the NRT exhibit a transient depolarization termed low threshold spike
(LTS) following sustained hyperpolarization. This is caused by the activation of low-voltage-activated
Ca2+ channels (LVACC). Although the role of these channels in thalamocortical oscillations was studied
in great detail, little is known about the downstream intracellular Ca2+ signalling pathways and their
feedback onto the oscillations. A signalling triad consisting of the sarco(endo)plasmic reticulum calcium
ATPase (SERCA), Ca2+ activated K+ channels (SK2), and LVACC is active in dendrites of NRT neurons and
shapes rhythmic oscillations. The aim of our study was to nd out (i) if and how Ca2+ -induced Ca2+
release (CICR) via ryanodine receptors (RyR) can be evoked in NRT neurons and (ii) how the released
Ca2+ affects burst activity. Combining electrophysiological, immunohistochemical, and two-photon Ca2+
imaging techniques, we show that CICR in NRT neurons takes place by a cell-type specic coupling of
LVACC and RyR. CICR could be evoked by the application of caffeine, by activation of LVACC, or by repetitive
LTS generation. During the latter, CICR contributed 30% to the resulting build-up of [Ca2+ ]i . CICR was
abolished by cyclopiazonic acid, a specic blocker for SERCA, or by high concentrations of ryanodine
(50 M). Unlike other thalamic nuclei, in the NRT the activation of high-voltage-activated Ca2+ channels
failed to evoke CICR. While action potentials contributed little to the build-up of [Ca2+ ]i upon repetitive LTS
generation, the Ca2+ released via RyR signicantly reduced the number of action potentials during an LTS
and reduced the neurons low threshold activity, thus potentially reducing hypersynchronicity. This effect
persisted in the presence of the SK2 channel blocker apamin. We conclude that the activation of LVACC
specically causes CICR via RyR in neurons of the NRT, thereby adding a Ca2+ -dependent intracellular
route to the mechanisms determining rhythmic oscillatory bursting in this nucleus.
2009 Elsevier Ltd. All rights reserved.

1. Introduction
The nucleus reticularis thalami (NRT) was proposed to be a key
determinant of the internal attentional searchlight as well as the
origin of hypersynchronous activity during absence seizures [1,2].
Endogenous NRT oscillations were shown to be regulated by Ca2+
entry through LVACC and the subsequent interplay between SERCA
activity and Ca2+ activated K+ channels (SK2) [3]. This intracellular
Ca2+ signalling complex was proposed to play a role in the amplication of low-frequency thalamic oscillations into large scale EEG
waves and it was suggested that physiologically controlled SERCA
activity could represent a method of regulating intrinsic oscillatory

Corresponding author. Tel.: +49 0251 83 58112; fax: +49 0251 83 55551.
E-mail address: coulon@uni-muenster.de (P. Coulon).
0143-4160/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ceca.2009.09.005

strength. Our specic hypothesis was that Ca2+ released from intracellular stores directly modulates such oscillatory activity. Intrinsic
oscillations can facilitate CICR, since during such activity considerable amounts of Ca2+ enter the cell. However, whether CICR occurs
in the NRT under these conditions was not known.
The NRT is the central modulatory site for the thalamocortical (TC) relay, and it is involved in generating the major brain
rhythms observed during sleep, wakefulness, and dreaming [46].
Oscillations in this nucleus arise as cyclic interactions between
the GABAergic NRT neurons and TC neurons of other thalamic
nuclei [7]. Burst ring in NRT neurons evokes rhythmic inhibitory
post-synaptic potentials (IPSPs) in TC neurons and an associated deinactivation of the low-voltage-activated (LVA) Ca2+ current that is
then activated upon repolarization and triggers a burst of fast action
potentials. This burst activity is transferred via excitatory synaptic
connections back to NRT neurons.

334

P. Coulon et al. / Cell Calcium 46 (2009) 333346

Similarly to neurons of other thalamic nuclei, NRT neurons


express LVACC (or T-type Ca2+ channels), enabling them to generate a low threshold Ca2+ potential [8] or low threshold spike (LTS)
that is crowned by a burst of action potentials (APs) after sustained
hyperpolarization. De-inactivation of the LVA Ca2+ current in NRT
neurons is achieved by repolarization to the relatively hyperpolarized resting state or by intra-NRT GABAergic connections. The
LTS triggers another GABA release that hyperpolarizes TC neurons
and NRT neurons alike, and the cycle starts again. Rhythmic burst
activity spreads as a propagating wave through the TC network,
resulting in spindle waves in the EEG [710].
By allowing Ca2+ inux into the cell, voltage-gated Ca2+ channels provide the main source of intracellular Ca2+ . Another source is
the endoplasmic reticulum (ER) which accumulates Ca2+ via SERCAs. Acting as a second messenger, Ca2+ itself mediates its own
release from these internal stores by activating ryanodine receptors (RyR) and IP3 receptors [11]. Beyond being a charge carrier,
Ca2+ is known to play a critical role in many thalamic functions [12].
It has been shown that thalamocortical relay neurons of the lateral
geniculate nucleus are modulated in their state of activity by CICR
[13]. Here, HVA Ca2+ channels were coupled to RyR, and CICR supported the relay mode of activity. CICR also occurred upon LVACC
activation in thalamic midline neurons [14]. Our study combines
electrophysiological, immunohistochemical, and two-photon Ca2+
imaging techniques to investigate whether Ca2+ release from intracellular stores occurs in NRT neurons, which mechanisms lead to
its activation, and what its functional consequences are concerning
burst and oscillatory activity.
2. Materials and methods
2.1. Preparation and electrophysiology
All animal preparations were done according to the European
Communities Council Directive of 24 November 1986 (86/609/EEC).
SpragueDawley rats (postnatal days 1027) were anaesthetized
with isourane and decapitated after loss of tail-withdrawal reex.
A block of brain tissue containing the NRT was removed and
submerged in ice-cold oxygenated saline containing (in mM):
sucrose, 200; PIPES, 20; KCl, 2.5; NaH2 PO4 , 1.25; MgSO4 , 10;
CaCl2 , 0.5; dextrose, 10; pH 7.35 with NaOH. NRT slices were
prepared as coronal sections on a vibratome. Before recording,
slices were kept in articial cerebrospinal uid (ACSF) containing (in mM): NaCl, 125; KCl, 2.5; NaH2 PO4 , 1.25; NaHCO3 , 24;
MgSO4 , 2; CaCl2 , 2; glucose, 10; pH adjusted to 7.35 with carbogen (95% O2 , 5% CO2 ). After transferring a slice to the recording
chamber, the NRT was identied by its position and structure
(Fig. 1A) and, after gaining whole cell access, NRT neurons were
identied by their electrophysiological characteristics (Fig. 1C and
D). During current-clamp recordings, slices were superfused with
ACSF containing (in mM): NaCl, 120; KCl, 2.5; NaH2 PO4 , 1.25;
NaHCO3 , 22; MgSO4 , 2; CaCl2 , 3; glucose, 20; pH 7.35 with carbogen. For voltage-clamp recordings the following alterations
were made to this solution (in mM): CsCl, 3; tetraethylammonium chloride (TEA-Cl), 5; 4-aminopyridine (4-AP), 2; tetrodotoxin,
0.0005; glucose, 10. Electrodes for whole cell recordings with a
resistance of 23 M were pulled from borosilicate glass capillaries (GC150T-10(F); Clark Electromedical Instruments, Pangbourne,
UK). For current-clamp recordings, electrodes were lled with a
saline solution (internal solution) containing (in mM): NaCl, 10;
K-gluconate, 105; K3 -citrate, 20; HEPES, 10; EGTA, 0.25; Mg-ATP,
3; Na2 -GTP, 0.5; MgCl2 , 0.5. For voltage-clamp recordings, electrodes were lled with an internal solution containing (in mM):
NaCl, 10; CsMeSO4 , 100; KCl, 1; HEPES, 10; EGTA, 0.25; Mg-ATP,
3; Na2 -GTP, 0.5; MgCl2 , 0.5; creatine phosphate, 15; QX314-Cl,
3.35 ((dimethylphenylcarbamoylmethyl)triethylammonium chlo-

ride, a blocker of voltage-gated Na+ channels); TEA-Cl, 15; 4-AP,


5. The pH was set to 7.25 with KOH or CsOH. Oregon Green
488 BAPTA-1 (OGB-1; 200 M, hexapotassium salt, Ca2+ indicator, green uorescence; KD 206 nM) [15,16], and Alexa-594
hydrazide (A594; 50 M, red uorescence) were added to internal solutions. In some experiments we used the lower afnity
indicator OGB-6F (KD 3 M). Dyes were from Invitrogen (Karlsruhe, Germany). Electrodes were connected to an EPC-10 amplier
(HEKA Elektronik, Lamprecht, Germany). After gaining access to
the cell, dyes were allowed to equilibrate by diffusion for at least
5 min. Access resistances were in the range of 520 M. Series
resistance compensation of 30% was routinely applied. Voltageand current-clamp experiments were controlled by the software
PatchMaster (HEKA Elektronik) operating on an IBM-compatible
personal computer. Liquid junction potential was not corrected
for (Vm = Vp 14 mV for current-clamp recordings and 2.3 mV for
voltage-clamp recordings). Caffeine (Sigma-Aldrich, Munich, Germany) was water soluble; apamin (Ascent Scientic, Bristol, UK),
ryanodine (Tocris, Bristol, UK; Alomone Labs, Jerusalem, Israel;
Ascent Scientic), thapsigargin (Alomone Labs), and CPA (Alomone
Labs) were dissolved in DMSO (nal DMSO concentration 0.2%).
Pressure application of caffeine was achieved with a pneumatic
drug ejection system (PDES 02D-LA-2, NPI electronics GmbH,
Tamm, Germany).
2.2. Calcium imaging
A custom-built two-photon laser-scanning microscope was
used, built around a confocal scan head (FV 300, Olympus, Hamburg, Germany) that was mounted on an upright microscope
(BX51WI, Olympus). We used a 60 objective lens (LumFl 60W/IR,
NA 1.1 or LumPlanFL 60W/IR, NA 0.9, both from Olympus). A
pulsed femtosecond Ti:Sapphire laser (Mai-Tai HP, Spectra-Physics,
Darmstadt, Germany) tuned to a  of 790 nm was used for
excitation. Epiuorescence was detected by two photomultiplier
tubes (Olympus). A dichroic mirror (SDM570, Olympus), a 530 nm
short-pass lter, and a 585 nm long-pass lter (BA530RIF, BA585IF,
Olympus) were used to separate the red from the green uorescence. Reected excitation light was blocked by IR blocking lters
(green channel: HC 680/SP, OD > 8 for excitation light, Semrock,
AHF Analysentechnik AG, Germany; red channel: 625SP, Chroma,
Frstenfeldbruck, Germany). Frames were captured at 210 Hz
with 12 bits per pixel. Fluorescence signals were analyzed with
Fluoview software (Olympus) or with custom Matlab routines
(R2008a, The MathWorks Inc., Ismaning, Germany). The uorescence ratio green/red (FG/R , OGB-1/A594) was used to avoid errors
due to volume changes or movement artefacts [17]. Unless stated
otherwise, we included large parts of the soma and a part of one
or several proximal dendrites (515 m) in a region of interest. After the experiments, confocal sections of the neurons were
recorded and stacked in three-dimensional reconstructions (see
Fig. 1B).
In order to estimate intracellular Ca2+ concentrations
([Ca2+ ]i ) we used the equation given by Tsien [18]: [Ca]2+ /KD =
(f fmin )/(fmax f). Minimal FG/R (fmin ) was estimated for a subset of cells by incubating slices for 1 h in A23187 (Calcimycin,
20100 M) [19] prior to patch clamping. After dye equilibration,
cells were incubated in 5 mM BAPTA. The uorescence ratio measured under these conditions was 0.42 0.02 (n = 8). In another
subset of cells maximal FG/R (fmax ) was estimated by disrupting the cell membrane with the patch electrode [20], thereby
allowing the rapid entry of extracellular Ca2+ into the cytosol.
This caused a rapid Ca2+ increase to an FG/R of 2.3 0.2 (n = 8).
The resulting dynamic range (Rf ) of 5.5 corresponds well with
values reported elsewhere [15]. Together with the KD reported
for OGB-1 of 206 nM [15,16], the resting [Ca2+ ]i for a subset of

P. Coulon et al. / Cell Calcium 46 (2009) 333346

335

Fig. 1. Basic properties of NRT neurons. (A) The NRT was visible as a thin band between the internal capsule and the external medullary lamina or ventral posterolateral
thalamic nucleus. The arrows indicate two patch pipettes, one for the application of drugs (right) and another for whole cell patch clamp recordings. (B) NRT neuron in a
three-dimensional reconstruction after establishing the whole cell patch clamp conguration and equilibration of the pipette solution with the cytosol. The patch electrode
was removed. (C) Superposition of three current-clamp traces from an NRT neuron (upper traces) obtained by varying current injections (lower traces). By applying a DC
offset (not shown), cells were set to 60 mV. During the injection of depolarizing currents the cells showed a series of APs (tonic ring), and after cessation of hyperpolarizing
currents, the neurons produced one or several LTS (burst ring). The LTS is shown to the right at an expanded timescale. It consisted of a long-lasting depolarization crowned
by several APs. APs are truncated. (D) The frequency (fAP ), the number of action potentials (nAP ), and the width of the LTS (widthLTS ) of 180 NRT neurons were plotted in a
three-dimensional graph. The black squares represent the three-dimensional value set. The grey symbols show the projections onto the plane spanning between each pair of
axes: fAP nAP (circles), fAP widthLTS (triangles), widthLTS nAP (squares). Neurons of the NRT showed variations in the values shown but could not be subdivided into different
cell types.

10 cells was estimated to be 39 11 nM, which was comparable


to the resting [Ca2+ ]i determined in other thalamic cells [21].
Furthermore, we estimated the intracellular Ca2+ for a transient
evoked by a typical voltage step from 50 to +5 mV for 2 s (suppl.
Fig. 1A). The spatially averaged maximal [Ca2+ ]i was well below
indicator-saturating levels. Finally, to further verify that the Ca2+
indicator was not saturated under our experimental conditions,
we compared the high afnity indicator OGB-1 with the lower

afnity indicator OGB-6F. Transients of different amplitudes were


evoked by stepping the holding potential from 60 to 30, 10,
0, and +5 mV (suppl. Fig. 1B). The normalized Ca2+ responses were
linear for both dyes. For all experiments that needed activation
of HVA Ca2+ channels we chose to step to a holding potential of
5 mV in order to evoke transients well below any saturating
effects. All other measured Ca2+ transients had similar or lower
amplitudes.

336

P. Coulon et al. / Cell Calcium 46 (2009) 333346

2.3. Immunohistochemistry
SpragueDawley rats (postnatal days 1827) were deeply
anaesthetized using pentobarbital (50 mg/kg body weight) and
transcardially perfused with PBS, followed by ice-cold 4% PFA/PBS
for 3540 min. Brains were removed, postxed for 4 h in 4%
PFA/PBS, and cryoprotected with 25% sucrose. Coronal sections
(40 m) were cut at the level of the NRT, washed several times
with TBS, and blocked with 10% normal horse serum (NHS), 2%
BSA, 0.3% Triton X-100 in TBS for 2 h to minimize nonspecic
binding before incubation of slices with primary antibodies in
2% NHS, 2% BSA, 0.3% Triton X-100 in TBS at 4 C for 1618 h.
The following antibodies were used: mouse anti-RyR2 (1:500,
Dianova, Hamburg, Germany, MA3-925), rabbit anti-RyR3 (1:1000,
Chemicon/Millipore, Schwalbach/Ts., Germany), rabbit anti-CaV 3.3
(1:1000, Alomone Labs), and rabbit anti-CaV 3.2 (1:1000, Alomone
Labs). Antibodies against SK2 (guinea pig anti-SK2) were a kind
gift from Dr. Rafael Lujan (Universidad de Castilla-La Mancha,

Albacete, Spain). After washing (3 10 min with TBS), sections were


exposed to Cy2- or Cy3-conjugated donkey-IgG (1:300, Dianova,
Germany) for 1.5 h, washed again and cover slipped with Immumount (Shandon, Pittsburgh, PA, USA). Negative controls were
routinely performed by occluding the primary antibody from the
staining procedure. No positive immunological signal was detected
(not shown).
2.4. Data analysis
Results are presented as mean S.E.M. Unless stated otherwise, control conditions were directly compared to altered
experimental conditions (e.g. in the presence of a substance or
at a different stimulus frequency) using data from the same
cell. Signicance was tested using a paired two-sided Students t-test. Differences were considered statistically signicant if
p < 0.05. Asterisks mark signicant (*p < 0.05) or highly signicant
(**p < 0.01) changes.

Fig. 2. Ryanodine receptors, Ca2+ activated K+ channels, and LVA Ca2+ channels are co-expressed in NRT neurons. Fluorescence images of NRT slice preparations with
immunohistochemical labelling of proteins involved in NRT Ca2+ signalling. Photographs in (A) each show a part of the NRT anked medially (left) by the ventrobasal
thalamic complex (VB) and laterally (right) by the internal capsule (ic, see rst photograph for illustration). Photographs in (B) each show a set of superimposed confocal
images, showing individual neurons at a larger magnication (compare scale bars). RyR2 and 3 were co-expressed in the same neurons (top left) with RyR3 delivering a
much stronger signal and showing a somatic expression with particularly strong signals from outside the area where the nucleus is expected (top right). CaV 3.3 and RyR2
were also co-expressed (bottom left), as were RyR3 and SK2 (top right) and CaV 3.2 and SK2 (bottom right).

P. Coulon et al. / Cell Calcium 46 (2009) 333346

337

3. Results
3.1. Basic properties of NRT neurons
Neurons of the NRT showed one to several LTS after a hyperpolarizing current injection under current-clamp conditions (Fig. 1C,
right trace, see also Fig. 6A, upper traces, Fig. 7A, 8C, and 9A) or
a comparably long-lasting LVA Ca2+ current under voltage-clamp
conditions (Fig. 5A), as described previously [22,23]. The following
resting properties were found: membrane capacitance: 62 1.6 pF
(n = 331), input resistance: 291 7 M (n = 234), resting membrane potential: 67 0.6 mV (n = 222), width of LTS: 286 11 ms
(n = 180), number of APs crowning a rebound LTS (after recovery
from 85.8 0.4 mV to the resting membrane potential): 4.9 0.2
(n = 180), frequency of AP ring during LTS (measured from the
rst two APs): 151 3.2 Hz (n = 172). Three of these parameters
(width of LTS, number of APs, and AP frequency) are plotted in
Fig. 1D. All parameters were comparable to previously published
values [3,2325]. Some cells (38 out of 204 tested under currentclamp conditions) showed several LTS upon returning to 60 mV or
bursted spontaneously (not shown). Another subset of cells (24 out
of 204) did not show any rebound LTS under these conditions, with
6 cells not showing an LTS at all during current-clamp recordings.
These cells were excluded from further analysis.
Although there was some variety among neurons, we found
the electrophysiological data of the 180 cells investigated under
current-clamp conditions and plotted in Fig. 1D to be insufcient
to subdivide the data set into distinctive cell types, as has been done
previously [25,26].
3.2. Ca2+ signalling proteins in NRT neurons

Fig. 3. Caffeine-induced Ca2+ transients in NRT neurons. (A) Somatic Ca2+ signals
after local application of caffeine (arrows). After the addition of the SERCA blocker
CPA (30 M) basal Ca2+ levels increased and, after 1015 min, caffeine applications
ceased to have an effect. (B) Dendritic (50 m from soma) Ca2+ signals after application of caffeine. Note the reduced scale. (C) Multiple applications of 20 mM caffeine
caused comparable Ca2+ transients.

In order to verify that RyR, LVA Ca2+ channels, and SK2 channels are all co-expressed in neurons of the NRT, we performed
immunohistochemical stainings. We found that RyR2 and 3 are coexpressed alongside with CaV 3.2 and 3.3 LVA Ca2+ channels and
SK2 (see Fig. 2). Type 3 RyR gave a stronger signal than type 2
(Fig. 2A) and were typically located in or near the soma, especially
around the area where the nucleus can be expected (Fig. 2B, top
right photograph). Fast LVA Ca2+ channels (CaV 3.2, Fig. 2B, bottom
right photograph) were also found to be expressed near the soma.
SK2 channels seemed to be expressed strongly in the proximal dendrites (Fig. 2B, bottom right, red), conrming previous ndings [3].

sients reached 96 9% of the initial amplitude; p = 0.65, n = 6). Bath


application of 30 M CPA caused a rise of [Ca2+ ]i (19.0 3.2% FG/R ,
n = 6), from which cells recovered largely (see Fig. 3A and B, and
compare resting Ca2+ levels in Fig. 6A, lower traces). Eventually (i.e.
after 10 min), CPA abolished the caffeine-evoked Ca2+ transients
in four out of six cells (see Fig. 3A and B) and reduced them in the
remaining two cells. On average, the caffeine-induced [Ca2+ ]i transient amplitude in the presence of CPA was 17 11% (p = 6 104 ,
n = 6) of control values. Local application of a HEPES-buffered ACSF
alone did not cause a Ca2+ transient.

3.3. Intracellular Ca2+ stores in NRT neurons

3.4. Ca2+ entry through HVA Ca2+ channels fails to trigger CICR

In the NRT, fast LVA Ca2+ channels of the CaV 3.1 and 3.2 type
appear to be expressed on somatic regions of the plasma membrane while slowly inactivating LVA Ca2+ channels (CaV 3.3) are
preferentially expressed on proximal dendrites [27]. In a previous study, Ca2+ signals of the NRT were investigated in dendritic
regions [3] and it was found that intracellular Ca2+ stores play a
major role in a signalling triad consisting of Ca2+ activated K+ channels of the SK2 type, T-type Ca2+ channels, and SERCAs. In this work
we focused on Ca2+ signals from somatic and proximal dendritic
regions of interest and CICR through RyR. Unless stated otherwise,
we included large parts of the soma and a part of one or several
proximal dendrites (extending no more than 15 m from the soma)
in a region of interest, averaging the cellular Ca2+ signals. Bath application of 10 or 20 mM caffeine resulted in a transient rise in [Ca2+ ]i
(not shown). Local application of 20 mM caffeine in HEPES-buffered
ACSF resulted in Ca2+ transients that were reected in uorescence
changes of 20.5 1.8% FG/R (soma, n = 6, Fig. 3A) and 34.5 5.7% FG/R
(dendrite, 1050 m from soma, n = 4, Fig. 3B). When repeatedly
elicited, these transients reduced their amplitude only slightly (two
out of six cells) or not at all (Fig. 3C; on average, subsequent tran-

To test for CICR upon activation of voltage-gated Ca2+ channels,


HVA (Fig. 4A) and LVA (Fig. 5A) Ca2+ currents were activated by
depolarizing the membrane potential under voltage-clamp conditions. This allowed us to simultaneously record the current induced
by Ca2+ inux and changes in [Ca2+ ]i . To account for any run-down
effect of HVA Ca2+ currents during the course of our experiments,
we used an approach described in previous studies [13,28,29]: To
distinguish between Ca2+ entering the cell through the plasma
membrane and Ca2+ released from intracellular stores, we estimated the net transmembrane charge movement by calculating the
integral of the Ca2+ current. We then used these values to normalize
the simultaneously recorded uorescence transients. For recording
HVA Ca2+ currents, cells were rst held at 45 mV in order to inactivate LVA Ca2+ currents and were subsequently stepped to 5 mV
for increasing amounts of time (pulse width, 50, 200, 500, 1000,
1500, and 2000 ms, see Fig. 4A for example traces), thus allowing increasing amounts of Ca2+ to enter the cell. The amplitude
of the uorescence signal (Fig. 4B) was divided by the net transmembrane charge movement to render the unit calcium transient
(norm. (FG/R )/Q) [29], which has been shown to faithfully detect a

338

P. Coulon et al. / Cell Calcium 46 (2009) 333346

Fig. 4. Ca2+ transients evoked by HVA Ca2+ channels. (A) Voltage steps from 45 to 5 mV caused HVA Ca2+ currents. (B) Increasing step duration (pulse width) caused
increasing changes in [Ca2+ ]i . (C) Plotting the unit calcium transient (rel. (FG/R )/Q) against the pulse width rendered more or less constant values, suggesting no contribution
of CICR (black boxes, n = 9). When 50100 M ryanodine was present in the pipette solution, a separate set of neurons behaved similarly (dashed line, open boxes, n = 15).
Values were normalized to the longest pulse width. The value at 1000 ms was signicantly reduced in ryanodine (p = 0.03). (D) To verify that CICR did not contribute to the
changes in [Ca2+ ]i under these conditions, the unit calcium transient was recorded both under control conditions and in the presence of 30 M CPA in the same set of neurons
(n = 4). Again, no indication of CICR was observed. Note that the slightly increased values at the rst pulse were caused by noise in the very small uorescence signals at a
pulse width of 50 ms.

contribution of CICR to the changes in [Ca2+ ]i [29,13]. In our experiments, the normalized unit calcium transient was independent
of the (increasing) pulse width of the HVA Ca2+ current activation (Fig. 4C), indicating that all Ca2+ detected by uorescence had
entered via the plasma membrane and was thus also detected as a
charge movement. It could be argued, however, that increasing activation of the Na+ /Ca2+ -exchanger, or other Ca2+ extrusion systems,
act to mask or even reverse an effect resulting from CICR. Since CICR
was reported to occur via RyR elsewhere in the thalamus [13], we
blocked this release in a separate set of experiments by using high
concentrations of ryanodine (50100 M) in the pipette solution.
The normalized unit calcium transient was still independent of the
pulse width under these conditions (Fig. 4C, dashed line). In order
to rule out a masked contribution of IP3 receptors and to eliminate
Ca2+ sequestration by SERCAs, we measured the unit calcium transient before and after the application of CPA (10 min, Fig. 4D). Again,
the normalized unit calcium transient remained independent of the
pulse width. We conclude that intracellular Ca2+ sources do not
contribute to an increase in [Ca2+ ]i induced by HVA Ca2+ channels.

3.5. CICR in NRT neurons is coupled to LVA Ca2+ channels


It was shown previously that LVA Ca2+ channels can trigger
CICR in midline thalamic neurons [14]. Because of the fast inactivation of LVA Ca2+ channels we used a protocol different from
that used for HVA Ca2+ channels to test for CICR. To allow sufcient amounts of Ca2+ to enter the neurons, we repeatedly activated
and de-inactivated the channels by stepping the holding potential from 100 to 45 mV for activation and back to 100 mV for
de-inactivation (Fig. 5A). While the [Ca2+ ]i will eventually reach
similar levels as observed after HVA Ca2+ channel activation (see
Section 4), this technique prevents the use of the unit calcium transient: With an increasing number of consecutive activations the
Ca2+ extrusion systems are also allowed an increasing amount of
time to do their work. Ca2+ that was detected as a charge movement as it entered the cell will not be detected as a uorescence
signal if it has been transported back out again. This will lead to
an increasing underestimation of the uorescence signal as the
number of consecutive LVA Ca2+ channel activations increases,

P. Coulon et al. / Cell Calcium 46 (2009) 333346

339

Fig. 5. Ca2+ transients evoked by LVA Ca2+ channels (A) Voltage steps from 100 to 45 mV caused LVA Ca2+ currents that fully inactivated within 200 ms. When activated
repeatedly at 2 Hz (with intermediate de-inactivation), LVA currents did not show any signs of run-down, as can be seen in the overlay of the rst and the last in a set of
20 traces (left and middle panel, middle panel recorded after 10 min in 30 M CPA). The overlay of currents under control conditions and in CPA shows that no direct effect
on LVA Ca2+ currents occurred (right panel). (B) Five, ten, and twenty consecutive LVA Ca2+ channel activations at 2 Hz caused Ca2+ transients of increasing amplitude. (C)
In CPA, the transients were reduced in the same neuron (left panel, dashed line/open boxes, n = 6). When ryanodine was added to the bath (50 M), the transients were
reduced in a similar way (right panel, n = 7). In another set of cells, when ryanodine (100 M) was present in the pipette solution, CPA did not have an effect (not shown,
n = 6). Fluorescence values were normalized to the response at 20 LVA activations under control conditions.

nally rendering negative slopes for a plot of the unit calcium transient against the number of activations. However, we found that
the simultaneously measured Ca2+ currents were constant over
time: The rst and last pulse in a series of 20 consecutive LVA
activations were very similar (Fig. 5A, left traces) and this was
also true after 10 min in CPA (middle traces). Thus, an approach
using the unit calcium transient is not mandatory. Instead, we
chose to directly plot the amplitude of the uorescence Ca2+ signal
(see Fig. 5B) against the number of consecutive LVA Ca2+ channel
activations (Fig. 5C; 5, 10, and 20 consecutive LVA Ca2+ current activations). Fluorescence amplitudes were normalized for each cell to
the value obtained with 20 activations under control conditions.
In order to detect a possible inuence of CICR on this relation,
we once again blocked SERCAs with 30 M CPA (10 min) or 2 M
thapsigargin (not shown). In the presence of either blocker, the
resulting Ca2+ uorescence transients were smaller in all cases,

being signicant at 10 and 20 activations (p = 0.014 and p = 0.004,


respectively; n = 6; Fig. 5C, left panel). To verify that this effect
was due to CICR and to characterize the mechanism by which
this presumed release takes place, the experiment was repeated
with ryanodine present in the internal solution at RyR-blocking
concentrations (100 M; not shown). Under these conditions the
effect of CPA was abolished (n = 6). In another set of experiments,
to directly show an effect of ryanodine and in order to verify that
the release mechanism depends on RyR, we bath applied ryanodine
at high concentrations (50 M, 10 min). The resulting reduction of
the normalized uorescence amplitude was similar to that caused
by CPA (Fig. 5C, right panel) with signicance already occurring
after 5 consecutive LVA Ca2+ channel activations (p = 0.02; 10:
p = 0.002; 20: p = 5 106 ; n = 7). We conclude that a Ca2+ inux
via LVA Ca2+ channels induces CICR through RyR in NRT neurons.

340

P. Coulon et al. / Cell Calcium 46 (2009) 333346

It could be argued that the slight increase of basal [Ca2+ ]i that


persisted after CPA incubation can render the indicator closer to
saturation. Under such conditions the uorescence signal expected
for a given Ca2+ rise would be smaller. However, the Ca2+ transients elicited by LVA Ca2+ channel activation were smaller than
those during HVA Ca2+ channel activation, thus any unspecic effect
would have occurred during both LVA and HVA Ca2+ channel activation (in the presence of CPA: 0.23 0.05 FG/R by 20 consecutive
LVA Ca2+ channel activations, n = 6; 0.45 0.05 FG/R by a 2 s HVA
Ca2+ channel activation, n = 4). However, the HVA Ca2+ channelinduced [Ca2+ ]i increase was unchanged in the presence of CPA.
Moreover, the following ndings argue against this scenario: (1) we
estimated the [Ca2+ ]i changes and found that they were typically
well below saturation levels (see suppl. Fig. 1A). (2) The high afnity indicator OGB-1 behaved in much the same way as the lower
afnity indicator OGB-6F when similar Ca2+ transients were evoked
(suppl. Fig. 1B).
3.6. Ca2+ transients evoked by low threshold Ca2+ potentials
Repeatedly evoked rebound bursts consisting of an LTS and
crowning APs (Fig. 6A, upper traces) caused a build-up of [Ca2+ ]i ,

indicative of CICR [14]. This build-up was reduced in the presence of CPA (Fig. 6A, lower traces). Even with just a single LTS,
a signicant effect of CPA was discernible and the effect continuously increased with an increasing number of LTS (Fig. 6B).
The application of TTX (0.51 M) caused the abolition of APs
(Fig. 6D) but left the LTS-evoked increase in [Ca2+ ]i almost unaffected. This revealed that APs contributed little to the build-up
of [Ca2+ ]i (7.8 3.8% at 0.3 Hz, n = 9, p = 0.07; 18.0 7.3% at 3 Hz,
n = 8, p = 0.04; Fig. 6C and E). These results conrm previous ndings [3] that an increase of [Ca2+ ]i by a rhythmic generation of
LTS mainly depends on LVA Ca2+ channels and not on APs. Additionally, we show that CICR can be elicited by this rhythmic burst
activity.
3.7. CICR controls repetitive low threshold activity
For the experiments shown in Fig. 6B we elicited up to 32
LTS in succession. As shown in Fig. 7A (left superimposed traces)
the delay between return from step hyperpolarization and the
rst AP of the subsequent LTS considerably decreased between
the rst and the 32nd LTS in a series. This was similar in the
presence of CPA (right traces). However, the number of APs crown-

Fig. 6. Ca2+ transients evoked by LTS and the contribution of action potentials to LTS-evoked increases in [Ca2+ ]i . (A) Repeated LTS (30 at 1 Hz, upper traces, APs truncated)
caused a build-up of [Ca2+ ]i (lower traces). In the presence of 30 M CPA this build-up was signicantly reduced (lower right trace, 73 3%, p = 5 105 , n = 8). Note the
increased number of APs evoked by a rebound burst in the presence of the blocker (see Figs. 79). (B) One or several LTS (2, 4, 8, 16, and 32) at a frequency of 1 Hz caused Ca2+
transients that were signicantly reduced in the presence of CPA (n = 5). Values were normalized to [Ca2+ ]i changes resulting from 32 LTS under control conditions. (C) In the
presence of TTX (0.51 M) Ca2+ transients evoked by 40 LTS at 0.3 Hz or 100 LTS at 3 Hz were largely unchanged when compared to control conditions. (D) LTS under control
conditions and in the presence of TTX. (E) At 0.3 Hz Ca2+ transients were not signicantly changed in the presence of TTX (n = 9), while at 3 Hz the change was signicant,
albeit small (n = 8).

P. Coulon et al. / Cell Calcium 46 (2009) 333346

341

Fig. 7. Ca2+ released from internal stores shapes the form of the LTS. (A) By release from hyperpolarizing current injections, 32 LTS were successively evoked. The delay to
the rst AP shortened between the rst and the 32nd LTS. In parallel, the number of crowning APs was reduced (left superimposed traces). The differences between the rst
and the 32nd LTS were very similar in CPA (right superimposed traces), but the number of APs was increased in both cases when compared to controls. APs are truncated.
(B) Statistical analysis revealed a signicantly shorter delay (p = 0.03) in ve out of six cells under control conditions (left panel, black bars) which was probably due to an
accompanying depolarization by 5.5 1.0 mV (n = 5). Although similar, the difference was not signicant in CPA (grey bars). The depolarization between the rst and last LTS
was only slightly reduced here (5.1 2.2 mV). The number of APs crowning the 32nd LTS was signicantly increased in CPA (right panel, p = 0.025).

ing each LTS was increased. The rst LTS in such a series of
32 showed an average delay between the onset of the LTS and
the rst AP of 170 38 ms (n = 6; 1 Hz; Fig. 7B, left panel). In
ve out of six cells, the last LTS showed a signicantly shorter
delay (78 15 ms, n = 5, p = 0.03). The delays between release from
hyperpolarization and the rst AP were similar in the presence
of CPA (1st LTS: 164 62 ms, last LTS: 82 25 ms, n = 5, p = 0.10,
Fig. 7B, left panel). The number of APs elicited during the rst
LTS was only slightly increased in CPA, whereas the 32nd LTS
generated signicantly more APs (3.3 0.8 under control conditions, 4.0 0.9 in the presence of CPA; n = 6, p = 0.025; Fig. 7B, right
panel).
In order to simulate oscillatory activity, and to evoke a usedependent cumulative activation/inactivation of the LVA Ca2+
channels, we repeatedly hyperpolarized NRT neurons at varying frequencies (1, 2, and 3 Hz) after an initial depolarizing step
(Fig. 8A). We chose parameters that would just sufce to evoke a
rebound burst at 1 Hz (i.e. 100 or 150 pA, see lower trace) in
order to assess the frequency-following capability and low threshold activity. We simultaneously recorded the [Ca2+ ]i , showing a
build-up as described in Fig. 6. After each set of hyperpolarizations,
Ca2+ was allowed to return to resting values before continuing the
experiments. Some hyperpolarizations failed to evoke a rebound
burst even at 1 Hz (see Fig. 8B, left trace), and the number of LTS per
hyperpolarization decreased as the frequency increased. This was
true both for the rst ve LTS in a series, that were usually most
reliably evoked, as well as for the last ve LTS, that were evoked

at comparably high [Ca2+ ]i (Fig. 8D, black bars). The number of


failures to evoke a rebound LTS with a current-induced hyperpolarization was somewhat decreased in ryanodine when viewed across
all frequencies (1, 2, and 3 Hz, total maximum number of hyperpolarizations was 90: 19.6 7.4 failures under control conditions,
12.1 4.6 failures in ryanodine, p = 0.13). Under control conditions,
the number of LTS per hyperpolarization at 3 Hz (n = 7) was signicantly reduced from the number at 1 Hz (n = 11, p = 0.026, unpaired
t-test). In a subset of cells (n = 6), both stimulation frequencies were
applied, also showing signicant (p = 0.025) differences (paired ttest). In ryanodine, the number of LTS per hyperpolarization for
the rst 5 events increased at all frequencies (Fig. 8B, right trace,
Fig. 8D, grey bars) rendering the remaining frequency-dependent
decrease of the number of LTS per hyperpolarization not signicant
(n.s.). Furthermore, ryanodine altered the shape of the rebound
LTS, producing more APs per LTS (Fig. 8C, compare left and right
trace). Combined with the increased low threshold activity (LTS per
hyperpolarization), this produced more APs per trace at all frequencies, with 1 and 3 Hz showing a signicant effect (1 Hz: p = 0.025,
n = 10, 3 Hz: p = 0.046, n = 7; Fig. 8E). In the presence of ryanodine
the neurons input resistance was not signicantly decreased (control: 327 38 M, in ryanodine: 295 40 M, n = 6, p = 0.08). We
conclude that Ca2+ released from ryanodine receptors contributes
to the characteristic frequency and time dependence of burst oscillations in NRT neurons.
In order to assess a possible contribution of SK2 channels
to the ryanodine-evoked increase in low threshold activity, we

342

P. Coulon et al. / Cell Calcium 46 (2009) 333346

Fig. 8. Released Ca2+ modulates burst discharges. (A) Current-clamp recording of an NRT neuron (Em : upper trace, injected current: lower trace). Neurons were set to an Em of
60 mV by DC current injection. The dashed line marks 90 mV. To evoke rebound LTS, currents of 100 or 150 pA were injected after positive current injection, mimicking
a shift from tonic to burst activity. The highlighted areas mark sections shown in (B) and (C) at an expanded timescale. (B) Release from the rst ve hyperpolarizations most
reliably evoked a rebound LTS. In this case, three out of ve releases from hyperpolarization caused an LTS under control conditions (left trace), whereas all ve releases from
hyperpolarization were successful in the presence of ryanodine (right trace). Voltage scale as in A. (C) The form of the LTS in the presence of ryanodine differed from that
under control conditions, most notably the number of APs. (D) Statistical analysis of the rst and last ve events. Under control conditions (black bars) at 3 Hz, the number of
LTS per hyperpolarization was signicantly reduced when compared to 1 Hz for the rst ve events (p = 0.026, unpaired t-test). In the presence of ryanodine this was lessened
and no longer signicant (n = 7 and 10). (E) The number of APs per trace was compared between control conditions (black bars) and in the presence of ryanodine (grey bars).
It was increased in the presence of ryanodine, which was most obvious at 1 Hz and at 3 Hz (1 Hz: p = 0.025, n = 10, 3 Hz: p = 0.046, n = 7). APs are truncated.

blocked SK2 using apamin (100 nM). In the presence of apamin,


the SK2-dependent afterhyperpolarization was completely abolished (Fig. 9A, compare grey and black trace). In the presence
of apamin the neurons input resistance signicantly decreased
(control: 294 22 M, in apamin: 249 29 M, n = 7, p = 0.01) and
at 3 Hz the low threshold activity was generally lower (compare
Fig. 9C with Fig. 8D). Under these conditions the application of
ryanodine left the form of the LTS as well as the input resistance

unaltered (226 26 M in apamin, 224 21 M in apamin and


ryanodine, n = 7, p = 0.91). However, the increase in low threshold
activity induced by the application of ryanodine remained unaffected by the presence of apamin (Fig. 9B and C), suggesting that
the mechanism by which CICR modulates the low threshold activity does not rely on SK2 channels. The number of failures to evoke a
rebound LTS was signicantly decreased in ryanodine when viewed
across all frequencies (1, 2, and 3 Hz, total maximum number of

P. Coulon et al. / Cell Calcium 46 (2009) 333346

343

Fig. 9. The ryanodine-evoked modulation of burst discharges does not rely on SK2 channel function. (A) The application of apamin (100 nM) abolished the afterhyperpolarization following an LTS and suppressed spontaneous repetition of LTS generation as was reported previously [3]. (B) Representative trace taken under similar conditions
as shown in Fig. 8B with apamin present in the bath solution. In the presence of ryanodine a similar modulation of burst discharges occurred. The example shows the last
ve traces of a neuron that was repeatedly hyperpolarized at 3 Hz with a current of 150 pA. (C) Both the rst and the last ve events showed signicantly fewer LTS per
hyperpolarization at 3 Hz when compared to 1 Hz (in the presence of 100 nM apamin; rst ve events: p = 0.003, n = 7; last ve events: p = 0.043, n = 7). After ryanodine was
added to the bath solution this effect was reduced in much the same way as shown in Fig. 8D. Here too, the remaining decrease was not signicant (n.s.; p = 0.12). (D) The
number of action potentials elicited in a sweep (see Fig. 8A) were similarly increased in the presence of apamin. APs are truncated.

hyperpolarizations was 90: 29.7 9.6 failures under control conditions with apamin, 11.6 7.1 failures in ryanodine with apamin,
p = 0.009). In ryanodine and apamin, the number of LTS per hyperpolarization for the rst and last 5 events was increased at all
frequencies (with the exception of the rst 5 traces at 1 Hz, Fig. 9B;
Fig. 9C, grey bars; 2 Hz not shown). Again, ryanodine rendered the
remaining frequency-dependent decrease of the number of LTS per
hyperpolarization not signicant. Although ryanodine did not seem
to alter the shape of the rebound LTS in the presence of apamin by
as much as in its absence, here too, more APs were observed per
LTS (Fig. 9D). Combined with the increased low threshold activity
(LTS per hyperpolarization), more APs per trace were generated
at all frequencies, with 1 and 2 Hz showing a signicant effect
(Fig. 9D).

4. Discussion
While the inputoutput connections as well as cellular electrogenic mechanisms of the NRT have been studied extensively
[3,6,25], the dynamics of intracellular Ca2+ signalling are far from
being completely understood. To our knowledge, the present study
is the rst to provide evidence for Ca2+ release mechanisms and
their functional signicance in NRT neurons. The results can be
summarized as follows: (1) We demonstrated Ca2+ release in NRT
neurons. (2) CICR via RyR was coupled to LVA Ca2+ channels, could
be evoked by repetitive LTS generation, and contributed 30% to
stimulus evoked transients and Ca2+ build-up during repetitive
stimulation. It was abolished in the presence of CPA or ryanodine.
(3) HVA Ca2+ currents failed to evoke CICR and APs contributed little

344

P. Coulon et al. / Cell Calcium 46 (2009) 333346

to the build-up of [Ca2+ ]i upon repetitive LTS generation. (4) Ca2+


released from intracellular stores modulated the shape of the LTS
and contributed to the time- and frequency-dependence of repetitive burst discharges by a mechanism not involving SK2 channels.
In conclusion, our results add RyR to the Ca2+ signalling tool kit that
is active in NRT neurons and shapes thalamocortical oscillations.
4.1. NRT Ca2+ signalling pathways
The NRT neurons unique ability to generate oscillatory bursting has been suggested to depend on a Ca2+ signalling triad that,
among LVA Ca2+ channels includes SK2 channels and SERCAs [3].
It has been proposed previously that CICR plays a major role in
mammalian Ca2+ signalling triads formed by voltage-gated Ca2+
channels, RyR, and SK channels [30]. Moreover, it was shown that
blocking sequestration by SERCAs facilitates oscillatory activity [3],
thereby already bringing an ER-dependent mechanism into focus.
Our results add to this by demonstrating, for the rst time, that the
intracellular Ca2+ stores in NRT neurons show CICR, and by identifying the release mechanism: active RyR comprise an additional
pathway in the Ca2+ signalling scenario, yielding an NRT-specic
signalling quad.
4.2. Ca2+ signalling proteins expressed in the NRT
The presence of mRNA encoding for skeletal muscle, cardiac, and
brain type RyR (sRyR, cRyR, and bRyR or RyR1, 2, and 3, respectively)
was shown for neurons of the rabbit NRT [31], with RyR2 giving the
strongest signal. In the mouse brain, RyR1 seems to be primarily
expressed in cerebellar Purkinje cells, with some expression also
shown for hippocampal and olfactory bulb cells, but little in the
thalamus [3133]. RyR1 is less Ca2+ sensitive than RyR 2 or 3 and
exhibits a lower gain of CICR activity [34,35]. Moreover, RyR1 is
known to be involved in voltage-induced Ca2+ release rather than
CICR [36]. Therefore, in this study we concentrated on RyR2 and 3.
It was known from previous studies that mRNA for SK1 and SK2
channels is present in neurons of the NRT [37]. Recently, Cueni and
colleagues showed that SK2 channels and SERCAs in dendrites of
the NRT interact with LVA Ca2+ channels to produce the typical
dampened burst discharge patterns [3]. While RyR2 and 3 were
both present in NRT neurons (this study), type 3 RyR gave a stronger
signal than type 2 (Fig. 2A), which may distinguish neurons of the
rat NRT from neurons of the rabbit NRT [31]. RyR3 in NRT neurons
were typically located in or near the soma, especially around the
area where the nucleus can be expected (Fig. 2B).
In brain slice preparations of the NRT, classical fast LVA Ca2+
channels (CaV 3.1 and 3.2) appear to be expressed on somatic
regions of the plasma membrane while slowly inactivating LVA
Ca2+ channels (CaV 3.3) are preferentially expressed on proximal
dendrites [27]. These channels may be coupled to different RyR
(skeletal muscle, cardiac, or brain type; RyR type 1, 2, or 3, respectively) which in turn possibly release Ca2+ from functionally and
locally distinct stores, as previously proposed [31]. While the previous studies already indicated the presence of all channels required
for the interaction proposed in this work, we have veried this by
immunohistochemical stainings (Fig. 2). The channel proteins necessary to explain our experimental ndings are all co-expressed in
NRT neurons. Moreover, our results show that RyR of type 3 are primarily localized in the soma, maybe preferentially activated by Ca2+
entering through classical fast LVA Ca2+ channels (CaV 3.2, Fig. 2B)
that were found to be expressed near the soma as well. Whether
the more diffuse expression of RyR type 2 plays a role in dendritic
Ca2+ signalling remains to be claried.
Interestingly, CICR already occurred after a single LTS. This
high sensitivity may arise from a close local arrangement in
microdomains. This is further backed by the nding that HVA

Ca2+ channels failed to elicit CICR, although the overall HVA Ca2+
channel-induced Ca2+ increase was generally larger. The spatial
coexpression of LVA Ca2+ channels and SERCAs was shown previously [3], suggesting a close proximity of these channels to the
respective Ca2+ release sites.
4.3. CICR and LVA Ca2+ channels in NRT neurons
It could be argued that an amplication mechanism such as CICR
would result in a supra-linear dependence of [Ca2+ ]i on the number of LTS- or LVA Ca2+ channel activations, comparable to what
has been reported for HVA Ca2+ channel-induced Ca2+ transients
[29,13]. However, HVA Ca2+ channel stimulation is continuous,
since these channels do not completely inactivate, while LVA Ca2+
channel activation has to occur repetitively to de-inactivate the
channels in between stimuli [14,23]. This mimics the putative physiological scenario of oscillatory bursting but limits the amount
of Ca2+ that can be accumulated within a given amount of time,
because of the counteracting effects of Ca2+ sequestration and
extrusion and the time consuming de-inactivation process. As a
result, to induce Ca2+ accumulations of comparable amplitude, LVA
Ca2+ currents must be activated over a period of time that is 10 to
20 times longer than for HVA Ca2+ currents (HVA Ca2+ channel activation of 500 ms: 0.31 0.04 FG/R , n = 9; 20 LVA Ca2+ channel
activation @ 2 Hz: 0.32 0.06 FG/R , n = 6). During this increased
time, larger amounts of Ca2+ can be expected to be passively or
actively extruded from the neuron or transported back into the
stores. Thus, the effect of CICR will not lead to a supra-linear Ca2+
accumulation but may instead be a prerequisite for an increase of
[Ca2+ ]i to occur, as has been proposed previously [14].
4.4. Physiological and pathophysiological relevance of CICR in the
NRT
The interplay between NRT neurons and TC relay neurons during the early phases of sleep has been studied extensively. Here, the
physiological function of the NRT is believed to lie in the generation of sleep spindles and the structuring of slow oscillatory activity
[6]. During attentive wakefulness the GABAergic NRT neurons are
tonically active at low frequencies [38]. By suppressing background
activity, the projections into the dorsal thalamus enhance the transfer of afferent information via relay neurons. However, since NRT
neurons reach intraburst frequencies during slow-wave sleep that
are higher than the fastest tonic discharges during steady states
of wakefulness [38], the nucleus primary function seems to be
reected in its ability to produce burst discharges. This idea is further strengthened by the ndings of this study, since CICR seems
to play a role only during burst activity and modulates it in return.
Burst activity in the NRT has been implicated in the generation and/or maintenance of generalized nonconvulsive seizures in
rodent models [39,40]. Moreover, an increase of LVA Ca2+ current in
the NRT, relating to an increase of the expression of CaV 3.2, has been
observed in a rat model of absence epilepsy [41,42]. Under both
physiological and pathophysiological conditions, burst ring in the
NRT and associated thalamic synaptic rebounds are characterized
by a typical temporal pattern, indicated by the waxing and waning of the LTS burst oscillation and the acceleration/deceleration
sequence of APs within each burst [2]. Important differences of
pathophysiological activity, such as absence seizures, are the high
degree of synchronization, more stereotyped oscillations at low
frequencies, and the more sustained nature of the synchronized
oscillatory activity [43]. While the exact contribution of CICR to
these phenomena is not clear, the present results demonstrate that
CICR signicantly contributes to the frequency-dependent characteristics of the LTS activity pattern and the APs generated by

P. Coulon et al. / Cell Calcium 46 (2009) 333346

each LTS. The dampening effect of released Ca2+ may present a


physiological means of preventing stereotyped oscillations at low
frequencies as well as a possible source of malfunction under pathophysiological conditions. Future studies will have to determine
whether this is true.
The number of APs elicited during late LTS generation was signicantly increased in the presence of CPA and ryanodine. It was
shown previously, that CPA increases the number of oscillatory
repetitions of the rebound burst. This was attributed to SERCAs
functioning as a Ca2+ sink, competing with SK2 for the Ca2+ entering through LVA Ca2+ channels. The loss of this function was shown
to augment activation of SK2 channels. However, this explanation
cannot account for the increased number of APs during an LTS, and
thus, the mechanism causing this increase remains elusive. Instead,
we showed that it does not require SK2 channels (Fig. 9D). Limiting
AP ring during an LTS may be an important role of CICR in the NRT
and would render a loss of function of RyR and SERCAs a putative
source for hypersynchronicity in the TC system.
The specic ability of the NRT to generate oscillatory bursting is mirrored in our nding that LVA Ca2+ channels selectively
activate RyR, and by the impact the released Ca2+ has onto the
shape of the LTS, the number of concomitant APs, and the low
threshold activity. By reducing the low threshold activity of NRT
neurons, ryanodine receptors may comprise an important mechanism for the dampening of oscillatory activity and, thus, for the
regulation of synchronization. In fact, it could be argued that intact
intracellular Ca2+ stores could allow a delayed action of released
Ca2+ on Ca2+ activated ion channels, such as SK2 or CAN (a Ca2+
activated non-selective cation conductance), providing an activation that is out of phase with the oscillatory activity and the direct
action of Ca2+ entering via LVA Ca2+ channels. We showed that the
effects of ryanodine persisted in the presence of the SK2 channel
blocker apamin, suggesting that the attenuation could occur by the
CAN current that was shown to be present in NRT neurons [44,45]
and should be activated at high [Ca2+ ]i . This cation conductance
has been shown to play an important role in NRT neurons and
particularly in the attenuation of oscillations. Future studies will
have to determine, by which mechanism RyRs exert their inuence.
Conict of interest
The authors declare that they have no conict of interest.
Acknowledgements
The authors wish to thank Elke Na and Birgit Herrenpoth for
excellent technical assistance, Dr. Rafael Lujan for kindly providing anti-SK2 antibodies, Dr. Tilman Broicher and Dr. Christian Lohr
for helpful comments on an earlier version of the manuscript, as
well as Dr. Hartmut Schmidt for sharing his technical expertise.
This work was supported by a grant from Innovative Medizinische Forschung at the University of Mnster (IMF, CO 210803, to
P.C.), the Deutsche Forschungsgemeinschaft (Pa 336/17-1 to H.C.P.),
and the Max-Planck-Research Award (to H.C.P.).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ceca.2009.09.005.
References
[1] F. Crick, Function of the thalamic reticular complex: the searchlight hypothesis,
PNAS 81 (1984) 45864590.

345

[2] P. Fuentealba, M. Steriade, The reticular nucleus revisited: intrinsic and


network properties of a thalamic pacemaker, Progr. Neurobiol. 75 (2005)
125141.
[3] L. Cueni, M. Canepari, R. Lujan, Y. Emmenegger, M. Watanabe, C.T. Bond, P.
Franken, J.P. Adelman, A. Luthi, T-type Ca2+ channels, SK2 channels and SERCAs
gate sleep-related oscillations in thalamic dendrites, Nat. Neurosci. 11 (2008)
683692.
[4] R. Llinas, U. Ribary, Coherent 40-Hz oscillation characterizes dream state in
humans, PNAS 90 (1993) 20782081.
[5] M. Steriade, D.A. McCormick, T.J. Sejnowski, Thalamocortical oscillations in the
sleeping and aroused brain, Science 262 (1993) 679685.
[6] M. Steriade, E.G. Jones, D.A. McCormick, Thalamus, 1st ed., Elsevier, Amsterdam,
1997.
[7] D.A. McCormick, T. Bal, Sleep and arousal: thalamocortical mechanisms, Annu.
Rev. Neurosci. 20 (1997) 185215.
[8] V. Crunelli, D.W. Cope, S.W. Hughes, Thalamic T-type Ca2+ channels and NREM
sleep, Cell Calcium 40 (2006) 175190.
[9] D. Contreras, M. Steriade, Spindle oscillation in cats: the role of corticothalamic
feedback in a thalamically generated rhythm, J. Physiol. (Lond.) 490.1 (1996)
159179.
[10] S.M. Sherman, R.W. Guillery, Functional organization of thalamocortical relays,
J. Neurophysiol. 76 (1996) 13671395.
[11] P. Kostyuk, A. Verkhratsky, Calcium stores in neurons and glia, Neuroscience
63 (1994) 381404.
[12] H.-C. Pape, T. Munsch, T. Budde, Novel vistas of calcium-mediated signalling in
the thalamus, Pgers Arch. 448 (2004) 131138.
[13] T. Budde, F. Sieg, K.H. Braunewell, E.D. Gundelnger, H.-C. Pape, Ca2+ -induced
Ca2+ release supports the relay mode of activity in thalamocortical cells, Neuron
26 (2000) 483492.
[14] T.A. Richter, M. Kolaj, L.P. Renaud, Low voltage-activated Ca2+ channels are coupled to Ca2+ -induced Ca2+ release in rat thalamic midline neurons, J. Neurosci.
25 (2005) 82678271.
[15] M. Maravall, Z.F. Mainen, B.L. Sabatini, K. Svoboda, Estimating intracellular calcium concentrations and buffering without wavelength ratioing, Biophys. J. 78
(2000) 26552667.
[16] C.D. Wilms, H. Schmidt, J. Eilers, Quantitative two-photon Ca2+ imaging via
uorescence lifetime analysis, Cell Calcium 40 (2006) 7379.
[17] R. Yasuda, B.L. Sabatini, K. Svoboda, Plasticity of calcium channels in dendritic
spines, Nat. Neurosci. 6 (2003) 948955.
[18] R.Y. Tsien, Fluorescent probes of cell signaling, Annu. Rev. Neurosci. 12 (1989)
227253.
[19] D. Thomas, S.C. Tovey, T.J. Collins, M.D. Bootman, M.J. Berridge, P. Lipp, A comparison of uorescent Ca2+ indicator properties and their use in measuring
elementary and global Ca2+ signals, Cell Calcium 28 (2000) 213223.
[20] J. Garcia, K.G. Beam, Calcium transients associated with the T type calcium
current in myotubes, J. Gen. Physiol. 104 (1994) 11131128.
[21] S.G. Meuth, H.-C. Pape, T. Budde, Modulation of Ca2+ currents in rat thalamocortical relay neurons by activity and phosphorylation, Eur. J. Neurosci. 15 (2002)
16031614.
[22] J.R. Huguenard, D.A. Prince, A novel T-type current underlies prolonged Ca(2+)dependent burst ring in GABAergic neurons of rat thalamic reticular nucleus,
J. Neurosci. 12 (1992) 38043817.
[23] T. Broicher, T. Kanyshkova, P. Meuth, H.C. Pape, T. Budde, Correlation of Tchannel coding gene expression, IT, and the low threshold Ca2+ spike in
the thalamus of a rat model of absence epilepsy, Mol. Cell Neurosci. (2008),
doi:10.1016/j.mcn.2008.07.012.
[24] J.R. Huguenard, Low-threshold calcium currents in central nervous system neurons, Annu. Rev. Physiol. 58 (1996) 3293248.
[25] S.-H. Lee, G. Govindaiah, C.L. Cox, Heterogeneity of ring properties among rat
thalamic reticular nucleus neurons, J. Physiol. (Lond.) 582 (2007) 195208.
[26] R. Spreaco, M. de Curtis, C. Frassoni, G. Avanzini, Electrophysiological characteristics of morphologically identied reticular thalamic neurons from rat
slices, Neuroscience 27 (1988) 629638.
[27] P.M. Joksovic, D.A. Bayliss, S.M. Todorovic, Different kinetic properties of two
T-type Ca2+ currents of rat reticular thalamic neurones and their modulation
by enurane, J. Physiol. (Lond.) 566 (2005) 125142.
[28] A. Shmigol, D. Eisber, A. Verkhratsky, Cyclic ADP-ribose enhances Ca2+ -induced
Ca2+ release in isolated mouse sensory neurons, J. Physiol. (Lond.) 483 (1995)
63P64P.
[29] A. Shmigol, A. Verkhratsky, G. Isenberg, Calcium-induced calcium release in rat
sensory neurons, J. Physiol. (Lond.) 489.3 (1995) 627636.
[30] M. Stocker, Ca2+ -activated K+ channels: molecular determinants and function
of the SK family, Nat. Rev. Neurosci. 5 (2004) 758770.
[31] T. Furuichi, D. Furutama, Y. Hakamata, J. Nakai, H. Takeshima, K. Mikoshiba,
Multiple types of ryanodine receptor/Ca2+ release channels are differentially
expressed in rabbit brain, J. Neurosci. 14 (1994) 47944805.
[32] G. Kuwajima, A. Futatsugi, M. Niinobe, S. Nakanishi, K. Mikoshiba, Two types
of ryanodine receptors in mouse brain: skeletal muscle type exclusively in
purkinje cells and cardiac muscle type in various neurons, Neuron 9 (1992)
11331142.
[33] G. Giannini, A. Conti, S. Mammarella, M. Scrobogna, Sorrentino V, The
ryanodine receptor/calcium channel genes are widely and differentially
expressed in murine brain and peripheral tissues, J. Cell Biol. 128 (1995)
893904.
[34] A.R. Marks, Intracellular calcium-release channels: regulators of cell life and
death, Am. J. Physiol. 272 (1997) H597605.

346

P. Coulon et al. / Cell Calcium 46 (2009) 333346

[35] T. Murayama, Y. Ogawa, RyR1 exhibits lower gain of CICR activity than RyR3 in
the SR: evidence for selective stabilization of RyR1 channel, Am. J. Physiol. Cell
Physiol. 287 (2004) C3645.
[36] C. Franzini-Armstrong, F. Protasi, Ryanodine receptors of striated muscles:
a complex channel capable of multiple interactions, Physiol. Rev. 77 (1997)
699729.
[37] M. Stocker, P. Pedarzani, Differential distribution of three Ca2+ -activated K+
channel subunits, SK1, SK2, and SK3, in the adult rat central nervous system,
Mol. Cell. Neurosci. 15 (2000) 476493.
[38] M. Steriade, L. Domich, G. Oakson, Reticularis thalami neurons revisited: activity changes during shifts in states of vigilance, J. Neurosci. 6 (1986) 6881.
[39] G. Avanzini, M. Vergnes, R. Spreaco, C Marescaux, Calcium-dependent regulation of genetically determined spike and waves by the reticular thalamic
nucleus of rats, Epilepsia 34 (1993) 17.
[40] V. Crunelli, N. Leresche, Childhood absence epilepsy: genes, channels, neurons
and networks, Nat. Rev. Neurosci. 3 (2002) 371382.

[41] E. Tsakiridou, L. Bertollini, M. De Curtis, G. Avanzini, H.C. Pape, Selective increase


in T-type calcium conductance of reticular thalamic neurons in a rat model of
absence epilepsy, J. Neurosci. 15 (1995) 31103117.
[42] E. Talley, G. Solrzano, A. Depaulis, E. Perez-Reyes, D. Bayliss, Lowvoltage-activated calcium channel subunit expression in a genetic model
of absence epilepsy in the rat, Brain Res. Mol. Brain Res. 75 (2000)
159165.
[43] M. Steriade, D. Contreras, Relations between cortical and thalamic cellular
events during transition from sleep patterns to paroxysmal activity, J. Neurosci.
15 (1995) 623642.
[44] T. Bal, D.A. McCormick, Mechanisms of oscillatory activity in guinea-pig nucleus
reticularis thalami in vitro: a mammalian pacemaker, J. Physiol. (Lond.) 468
(1993) 669691.
[45] A. Destexhe, D. Contreras, T.J. Sejnowski, M. Steriade, A model of spindle rhythmicity in the isolated thalamic reticular nucleus, J. Neurophysiol. 72 (1994)
803818.

European Journal of Neuroscience

European Journal of Neuroscience, Vol. 31, pp. 439449, 2010

doi:10.1111/j.1460-9568.2010.07081.x

SYNAPTIC MECHANISMS

Intracellular Ca2+ release-dependent inactivation of Ca2+


currents in thalamocortical relay neurons
Vladan Rankovic,1 Petra Ehling,2 Philippe Coulon,2 Peter Landgraf,3 Michael R. Kreutz,3 Thomas Munsch1,*
and Thomas Budde2,4,*
1

Institut fur Physiologie, Otto-von-Guericke-Universitat, Magdeburg, Germany


Institut fur Physiologie I, Westfalische Wilhelms-Universitat Munster, Robert-Koch-Strasse 27a, D-48149 Munster, Germany
3
PG Neuroplastizitat, Leibniz Institut fur Neurobiologie, Magdeburg, Germany
4
Institut fur Experimentelle Epilepsieforschung, Westfalische Wilhelms-Universitat Munster, Munster, Germany
2

Keywords: Ca2+-induced Ca2+ release, Ca2+ signalling, high-voltage-activated Ca2+ channels, rat, ryanodine receptors,
thalamic function

Abstract
Neuronal Ca2+ channels are rapidly inactivated by a mechanism that is termed Ca2+-dependent inactivation (CDI). In this study we
investigated the influence of intracellular Ca2+ release on CDI of high-voltage-activated Ca2+ channels in rat thalamocortical relay
neurons by combining voltage-clamp, Ca2+ imaging and immunological techniques. Double-pulse protocols revealed CDI, which
depended on the length of the conditioning pulses. Caffeine caused a concentration-dependent increase in CDI that was
accompanied by an increase in the duration of Ca2+ transients. Inhibition of ryanodine receptors and endoplasmic Ca2+ pumps (by
thapsigargin or cyclopiazonic acid) resulted in a reduction of CDI. In contrast, inhibition of inositol 1,4,5-tris-phosphate receptors by
intracellular application of 2-aminoethoxy diphenyl borate or heparin did not influence CDI. The block of transient receptor potential
channels by extracellular application of 2-aminoethoxy diphenyl borate, however, resulted in a significant reduction of CDI. The
central role of L-type Ca2+ channels was emphasized by the near-complete block of CDI by nifedipine, an effect only surpassed when
Ca2+ was replaced by Ba2+ and chelated by 1,2-bis(o-aminophenoxy)ethane-N,N,N,N,-tetraacetic acid (BAPTA). Trains of action
potential-like stimuli induced a strong reduction in high-voltage-activated Ca2+ current amplitude, which was significantly reduced
when intracellular Ca2+ stores were made inoperative by thapsigargin or Ba2+ BAPTA. Western blotting revealed expression of Ltype Ca2+ channels in thalamic and hippocampal tissue but not liver tissue. In summary, these results suggest a cross-signalling
between L-type Ca2+ channels and ryanodine receptors that controls the amount of Ca2+ influx during neuronal activity.

Introduction
Voltage-gated Ca2+ channels of the plasma membrane have three
subfamilies (CaV1, CaV2 and CaV3) (Lacinova, 2005). They are
composed of 10 pore-forming a1 channel subunits and are important
components of a universal cellular Ca2+ signalling tool kit (Berridge
et al., 2000). Depending on the specic voltage-gated Ca2+ channel
subtype, Ca2+-induced Ca2+ release (CICR) from intracellular stores is
a mechanism that amplies the Ca2+ entry through plasma membrane
Ca2+ channels, whereas Ca2+-dependent inactivation (CDI) of these
plasma membrane channels represents an important negative feedback
mechanism (Budde et al., 2002; Bardo et al., 2006). With respect to
the thalamus, only recently has a role for Ca2+ ions beyond being
charge carriers started to evolve, and contributions of low-voltageactivated (CaV3) as well as high-voltage-activated (HVA) (CaV1 and
CaV2) Ca2+ channels to intracellular Ca2+ signals have been found

(Pape et al., 2004; Cueni et al., 2009). In thalamocortical relay (TC)


neurons of the dorsal part of the lateral geniculate nucleus (dLGN),
CICR contributes to intracellular Ca2+ transients, leads to the
activation of Ca2+-dependent K+ channels and thereby supports
regular tonic ring (Budde et al., 2000). Furthermore, CDI, which is
under multiple biochemical and activity-dependent control, has been
shown to limit Ca2+ entry into TC neurons (Armstrong, 1989; Meuth
et al., 2001, 2002, 2005). Although CICR-dependent CDI has been
described in ventricular myocytes (Takamatsu et al., 2003), this type
of cross-signalling between Ca2+ channels of the plasma membrane
and the endoplasmic reticulum (ER) membrane remains to be
elucidated in neurons. As both mechanisms are active in TC neurons,
this cell type is an ideal candidate for investigating possible
interactions between CICR and CDI in central neurons.

Correspondence: Professor Dr Thomas Budde, 2Institut fur Physiologie I, as above.


E-mail: tbudde@uni-muenster.de

Materials and methods

*T.M. and T.B. contributed equally to this work.

Thalamic slices were prepared from juvenile LongEvans rats at


postnatal day 1433. After anaesthesia with isourane, animals were

Received 17 November 2008, revised 27 November 2009, accepted 1 December 2009

Tissue preparation

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd

440 V. Rankovic et al.


decapitated and a block of tissue containing the dLGN was rapidly
removed and placed in chilled (24C), oxygenated slicing solution
(pH 7.35, with NaOH) containing (in mm): 195 sucrose, 11 glucose,
20 piperazine-N,N-bis(2-ethanesulfonic acid) (PIPES), 2.5 KCl, 10
MgSO4 and 0.5 CaCl2. Coronal slices of the thalamus were cut at
300 lm on a vibratome and kept submerged in articial cerebrospinal
uid (pH 7.35, with 95% O2 5% CO2) containing (in mm): 125 NaCl,
2.5 KCl, 1.25 NaH2PO4, 2226 NaHCO3, 2 MgSO4, 2 CaCl2 and 10
glucose. All animal procedures were approved by local authorities
(Bezicksregierung Munster).

All values were presented as mean SEM. As we were able to


demonstrate a Gauss distribution for the three main parameters
(current amplitude, Dinact and Rinact) analysed in the present study
under control conditions, statistical signicance was evaluated by
Students t-test. Where applicable, data under control conditions and in
the presence of drugs were compared for the same cells. Values of
P = 0.05 were considered statistically signicant. WaveMetrics Igor
Pro 4.0 and Microcal Origin 6.0 software were used for data analysis
and gure plotting.

Ca 2+ imaging
Patch-clamp recordings
Whole-cell recordings were performed on visually identied TC
neurons of the dLGN at room temperature (2123C), using glass
microelectrodes pulled from borosilicate glass capillaries (GC150TF10; Clark Electromedical Instruments, Pangbourne, UK) that were
connected to an EPC-9 2 amplier (double patch clamp; HEKA,
Lambrecht, Germany). The typical electrode resistance was 24 MX,
whereas access resistance was 515 MX. Series resistance compensation was routinely used (> 30%). Voltage-clamp experiments were
governed by Pulse software (HEKA Electronics). For standard recordings, the following solutions were used: (i) extracellular solution
(in mm): 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 2226 NaHCO3, 2
MgSO4, 2 CaCl2, 10 glucose, 0.001 tetrodotoxin and 4 4-aminopyridine, pH 7.35 with NaOH; and (ii) intracellular solution (in mm): 85
Cs-gluconate, 10 Cs3-citrate, 10 NaCl, 1 KCl, 1.1 EGTA, 0.1 CaCl2,
0.25 MgCl2, 10 HEPES, 15 tetraethylammonium (TEA)-Cl, 3 Mg-ATP
and 0.5 Na-GTP, pH 7.25 with CsOH. For measurements with trains of
action potential-like stimuli, the following intracellular solution was
used (in mm): 95 CsMeSO4, 10 NaCl, 1 KCl, 1.1 EGTA, 0.1 CaCl2, 0.25
MgCl, 10 HEPES, 15 TEA-Cl, 3 Mg-ATP, 0.5 Na-GTP, 5
4-aminopyridine, 3.35 QX-314-Cl and 15 phosphocreatine, pH 7.25
with CsOH.
The HVA Ca2+ currents were evoked from a holding potential of
)40 mV. For conditioning pulses, the voltage was stepped to varying
potentials ()40 to +60 mV, 200 ms duration; in some experiments the
conditioning pulse length was varied between 50 and 1000 ms),
followed by a brief gap ()40 mV, 50 ms) and a subsequent analysing
test pulse to a xed potential of +10 mV (200 ms). Double pulses were
applied every 6.4 s. In one set of experiments, double pulses were
repeated every 60 s. For standard recordings, Ca2+ was used as the charge
carrier and 1.1 mm EGTA was included in the intracellular solution. In
one set of experiments Ba2+ was used as the charge carrier and 11 mm
BAPTA was included in the intracellular solution. The degree of
inactivation (Dinact) (see bracket in Fig. 1B) was determined as follows
Dinact 1I min =I max   100%
with Imin representing the minimal test pulse amplitude after a preceding conditioning pulse, as measured at the beginning of the voltage step
to +10 mV (upper arrowhead in Fig. 1B), and Imax representing the
maximal test pulse amplitude (lower arrowhead), as measured without a
preceding conditioning pulse. Time-dependent inactivation of the
conditioning pulse to +10 mV was assessed by calculating the
inactivation ratio (Rinact), which was determined as follows
Rinact I 200 =I peak
with I200 and Ipeak representing the current amplitude at the end of
the 200 ms pulse and the peak amplitude, respectively. Rinact values
can be expected to decrease as time-dependent inactivation becomes
faster.

Thalamic slices were placed under an upright microscope (Axioskop,


Zeiss) and whole-cell recordings were performed using a
Cs-gluconate-based internal solution containing neither EGTA nor
CaCl2. Fluorometric Ca2+ measurements were performed with BisFura-2 (Molecular Probes; Invitrogen) added to the pipette solution at
a concentration of 100 lm. Double pulses were elicited every 10 s.
The dye was alternately excited at 350 and 380 nm using a
monochromator, and images were captured at 1 Hz with 12 bit
resolution on a CCD camera (PentaMAX; Princeton Inc., NJ, USA).
The background-corrected 350 : 380 ratio was calculated ofine.
For the determination of changes in basal Ca2+ levels, the highafnity Ca2+ indicator Oregon Green 488 BAPTA-1 (200 lm, hexapotassium salt) and a reference dye (Alexa-594 hydrazide, 50 lm, red
uorescence) were added to internal solutions (Coulon et al., 2009).
A custom-built two-photon laser-scanning microscope was used.
Briey, we used a 60 objective lens (LumFl 60W IR, NA 1.1;
Olympus) and a pulsed femtosecond Ti:Sapphire laser (Mai-Tai HP;
Spectra-Physics, Darmstadt, Germany) tuned to a k of 790 nm for
excitation, and the uorescence ratio green : red (FG R, Oregon Green
488 BAPTA-1 Alexa-594 hydrazide) was used to avoid errors due to
volume changes or movement artefacts (Yasuda et al., 2003).

Results
Ca 2+-dependent inactivation is active in thalamocortical relay
neurons of acute brain slice preparations
To conrm the presence of CDI in dLGN TC neurons of thalamic
slices, HVA Ca2+ currents were recorded from more than 150 cells.
A double-pulse voltage protocol (Fig. 1A; see Materials and methods)
was used that effectively discloses CDI (Budde et al., 2002). If CDI is
operative, the current evoked by the test pulse should exhibit an
inverted U-shaped dependence (due to the normalization procedure
used in the present study) on the conditioning pulse potential, with
maximal inactivation occurring at the peak of the conditioning pulse
currentvoltage (IV ) relationship. Under standard conditions, the IV
relationship of the conditioning pulse (Fig. 1C, open squares)
demonstrated HVA Ca2+ currents (Fig. 1B) with an activation
threshold negative to )30 mV, a maximal inward current at around
+10 mV and an apparent reversal potential at around +40 mV. The test
pulse IV (peak amplitude of the test pulse current plotted vs.
conditioning pulse voltage) showed an inverted U-shape with the
minimal current occurring at +10 mV (Fig. 1C, closed squares), as
expected for a CDI mechanism. In the following the Dinact was used as
a simple measure for the strength of CDI. This parameter was
determined by dividing the minimal test pulse amplitude (following
the conditioning pulse to +10 mV; upper arrowhead in Fig. 1B) by the
maximal test pulse amplitude (occurring when no preceding conditioning pulse was applied; lower arrowhead in Fig. 1B) and calculating the percent change in current amplitude. With respect to the

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

CICR-dependent CDI in the thalamus 441

conditioning pulse

HVA Ca 2+ currents

+60 mV

1 nA

200 ms

test pulse
+10 mV

Dinact

-40 mV
200ms

200ms

D
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

conditioning pulse I/V

***

60

***
50

*** (120)

40
30
20

-40 -20
0
20
40
60
Conditioning pulse potential (mV)

(5)

(5)

test pulse I/V

Dinact (%)

I/Imax

(5)

(5)

100
1000
Conditioning pulse length (ms)

Fig. 1. Identication of CDI in TC neurons in the slice preparation. (A) Scheme of the double-pulse protocol used to elicit HVA Ca2+ currents. TC neurons were
held at -40 mV and stepped to varying potentials (conditioning pulse, )40 to +60 mV, 200 ms duration). This was followed by a brief pause ()40 mV, 50 ms) and a
subsequent analysing test pulse to a xed potential of +10 mV (200 ms). (B) Representative current traces elicited by the pulse protocol shown in A. The Dinact,
which was taken as a measure of the strength of CDI, is indicated by the two arrowheads and bracket. (C) Mean IV relationship evoked by the conditioning pulse
(open squares; data from 10 randomly chosen cells were averaged) and the test pulse depending on conditioning pulses of increasing depolarization (closed squares;
data from 120 cells were averaged) are shown. Normalized current amplitudes are plotted vs. the conditioning pulse potential. (D) Dependency of Dinact on the length
of the conditioning pulse. Signicance was tested against values obtained by the shortest conditioning pulse duration of 50 ms (the number of observations is
indicated in brackets) ***P < 0.001.

maximal test pulse current amplitude of )1502 44 pA (n = 120),


the Dinact under experimental control conditions was 39.1 0.5%
(Fig. 1C). By default, the test pulse IV was obtained with the
conditioning pulses altered in 10 mV increments from )40 mV
(Fig. 1B and C). From the presented data it became clear that there
was no full recovery of the test pulse current amplitude to baseline
levels at the most depolarized potentials. To exclude an inuence of
this experimental paradigm on the results obtained, the stimulus
protocol was repeated in a reversed manner by altering the
conditioning pulse from +60 in )10 mV decrements. Under these
conditions, the Dinact was unchanged (41.9 1.8%, n = 14), whereas
the lack of full recovery was shifted to the most hyperpolarized
potentials. The hysteresis found in both types of experiments probably
resulted from the incapability of the cells to fully restore baseline
intracellular Ca2+ levels between stimulations (see below). Therefore,
a double-pulse protocol was used in one set of experiments, consisting
of only three conditioning pulses ()40, +10 and +60 mV) applied at
60 s time intervals. Under these conditions, a full recovery of the test
pulse amplitude (96.6 1.1%, n = 22; data not shown) was achieved
for the most depolarized conditioning pulse, which was signicantly
(P < 0.001) different from the incomplete recovery seen during faster
stimulation (81.7 0.6%, n = 107). The time dependency of HVA
Ca2+ current inactivation was assessed by determining the Rinact. This
parameter was calculated by dividing the current amplitude at the end
of the 200 ms pulse (conditioning pulse to +10 mV) by the peak

amplitude. Under control conditions, current amplitudes strongly


decreased with Rinact = 0.54 0.01 (n = 130), thereby indicating a
fast inactivation process.
After the initial equilibration period of 5 min following the
establishment of whole-cell conditions, the peak current amplitude of
the conditioning pulse to +10 mV was )1516 42 pA (n = 130). As
has been noted before (Budde et al., 1998), HVA Ca2+ currents showed a
7 1% (n = 10) run-down over the course (20 min) of a typical
experiment (data not shown). To exclude the possibility that run-down
had inuenced the degree of CDI, the extracellular Ca2+ concentration
was reduced from 2 to 1 mm and from 3 to 1.5 mm, resulting in a
signicant 40.2 1.8% (n = 6; P < 0.01) and 42.9 5.0% (n = 7;
P < 0.01) reduction in peak current amplitude, respectively (data not
shown). Despite signicantly smaller overall current amplitudes, the
Dinact was not signicantly changed (P > 0.05) when the driving force
decreased (3 mm: 38.9 3.1%; 2 mm: 42.2 2.2%; 1.5 mm:
37.2 2.0%; 1 mm: 40.2 1.8%). Taken together, these ndings
demonstrate a CDI mechanism in TC neurons of the acute slice
preparation that is very similar to cells after acute isolation.
Next, the dependency of Dinact on the conditioning pulse length was
investigated. With a pulse length exceeding 100 ms, a signicant
(P < 0.0001) increase in inactivation was observed (Fig. 1D). Dinact
averaged 22.2 2.0 (n = 5), 23.5 2.3 (n = 5), 39.1 0.5 (n = 120),
51.1 1.1 (n = 5) and 56.8 2.0% (n = 5) for conditioning pulse
durations of 50, 100, 200, 500 and 1000 ms, respectively.

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

442 V. Rankovic et al.


Modulators of ryanodine receptors and endoplasmic Ca 2+
pumps alter Ca2+-dependent inactivation in thalamocortical
relay neurons
The CICR is a prominent feature of TC neurons (Budde et al., 2000)
that is based on close spatial interaction between L-type Ca2+ channels
and ryanodine receptors (RyR) (Chavis et al., 1996). Therefore, we
probed the possibility that the negative feedback of intracellular Ca2+
on the HVA Ca2+ current amplitude is based not only on Ca2+ entering
via the plasma membrane channels but also on Ca2+ that is released
from intracellular stores. Therefore, CICR was stimulated with
caffeine using concentrations ranging from 0.5 to 20 mm. With
concentrations exceeding 0.5 mm, the extracellular application of
caffeine signicantly (P < 0.01) reduced peak current amplitudes
(Budde et al., 2000) of the conditioning pulse to +10 mV by
)5.9 9.9 (0.5 mm; n = 5), )27.3 6.9 (1 mm; n = 6),
)37.1 4.7 (5 mm; n = 6), )29.4 4.2 (10 mm; n = 16) and
)32.1 4.9% (20 mm; n = 5; Fig. 2A). This effect was accompanied
by an increase in Dinact (0.5 mm: 44.2 2.1%, P < 0.05; 1 mm:
45.2 2.4%, P < 0.05; 5 mm: 46.7 2%, P < 0.01; 10 mm:
50.9 1.1%, P < 0.001; 20 mm: 54.6 0.9%, P < 0.001) and a
decrease (P < 0.05 for 1 and 10 mm caffeine; P < 0.001 for 20 mm
caffeine) in Rinact (0.5 mm: 0.51 0.01; 1 mm: 0.50 0.03; 5 mm:
0.51 0.06; 10 mm: 0.48 0.03; 20 mm: 0.27 0.02) at 200 ms
conditioning pulse length (Fig. 2B and C). When conditioning pulses
of different lengths were applied, the addition of 10 mm caffeine
induced a signicant increase in Dinact in comparison to control
conditions for pulse durations 100 ms (50 ms: 19.4 1.5%, n = 10;
100 ms: 29.3 1.2%, n = 10, P < 0.01; 200 ms: 48.5 1.4%,
n = 15, P < 0.001; 500 ms: 61.6 1.2%, n = 14, P < 0.001;
1000 ms: 67.2 1.2%, n = 14, P < 0.001; Fig. 2D).
We performed Ca2+ imaging experiments to assess alterations of
intracellular Ca2+ levels during caffeine application and to further
establish a specic contribution of CICR to CDI. TC neurons were
loaded with the Ca2+ indicator dye bis-Fura-2 (100 lm) via the
recording pipette (Fig. 3A). Calcium currents were evoked by using
double pulses with conditioning voltage steps to )40, +10 and
+60 mV (500 ms prepulse duration; see pulse protocol in Fig. 3B).
Under control conditions, the uorescence ratio averaged
0.50 0.015 (n = 3). When double pulses were applied, amplitudes
of the Ca2+ transients under control conditions with respect to the
basal Ca2+ level were 0.077 0.015 (rst double pulse; 1, ),
0.14 0.018 (second double pulse; 2, ) and 0.142 0.025 (third
double pulse; 3, ) (Fig. 3C). To achieve a rough estimation of the
duration of each transient, we determined the uorescence ratios
reached at the end of each double-pulse sweep (i.e. at 10, 20 and
30 s after starting the electrophysiological stimulation marked by
the arrowheads in Fig. 3B) and calculated the degree of recovery to
baseline with respect to the maximal amplitude of each transient.
Under control conditions, Ca2+ transients recovered by 85.8 2.2%
(rst double pulse), 66.9 2.3% (second double pulse) and
52.1 5.7% (third double pulse) (Fig. 3D). After the application
of 10 mm caffeine, basal Ca2+ levels increased by less than 10% to
ratio values of 0.552 0.012% (Fig. 3B). Under these conditions,
amplitudes of the Ca2+ transients with respect to the basal Ca2+
level were 0.095 0.024% (rst double pulse), 0.158 0.032%
(second double pulse) and 0.143 0.034% (third double pulse)
(Fig. 3C), and recovered by 58.7 2.0% (rst double pulse),
37.1 5.9% (second double pulse) and 26.9 6.1% (third double
pulse) during a time period of 10 s (Fig. 3D). These data indicate
that the increase in the duration of Ca2+ transients is based on the
occurrence of CICR.

To demonstrate the contribution of the ER to CDI in TC neurons,


we functionally disabled intracellular Ca2+ stores by application of
inhibitors of the sarcoplasmic endoplasmic reticulum Ca2+ ATPase
(SERCA). In the presence of thapsigargin or cyclopiazonic acid
(CPA), Dinact was signicantly (P < 0.0001) decreased to 24.8 1.0%
(n = 4) and 24.3 1.0% (n = 4), respectively (Fig. 4B and C; 200 ms
conditioning pulse duration). This effect was accompanied by a
signicant (P < 0.0001) increase in Rinact (thapsigargin: 0.68 0.01,
n = 4; CPA: 0.69 0.02, n = 4) but peak current amplitudes of the
conditioning pulse to +10 mV were nearly unchanged (thapsigargin:
)1505 123 pA, n = 4; CPA: )1580 170 pA, n = 4). To further
prove the contribution of RyR, we applied ryanodine (10 lm) and
found a signicant reduction in Dinact with prepulse durations between
100 and 500 ms in comparison to control conditions (50 ms:
18.1 1.2%, n = 14; 100 ms: 19.1 1.0%, n = 14, P < 0.05;
200 ms: 28.9 0.8%, n = 22, P < 0.001; 500 ms: 44.6 1.3%,
n = 14, P < 0.05; 1000 ms: 57.4 1.1%, n = 14; Figs 2D and 4D).
This effect was characterized by a concordant signicant (P < 0.0001)
increase in Rinact (0.69 0.01; 200 ms pulse length) and an
unchanged peak current amplitude (1686 176 pA; 200 ms pulse
length) for the conditioning pulse to +10 mV. In order to assess the
effect of intracellular Ca2+ store modulators on basal Ca2+ levels, we
performed Ca2+ imaging experiments using Oregon Green 488
BAPTA-1. Although the SERCA pump inhibitor CPA induced a
signicant (P < 0.04) increase (16.1 4.4%, n = 4) in basal Ca2+
levels, ryanodine had only a small effect on basal Ca2+ levels
(9.8 9.6%, n = 3; data not shown).
Next, inhibitors of inositol 1,4,5-tris-phosphate-induced Ca2+
release were tested. Intracellular application of 2-aminoethoxy diphenyl borate (2APB) (100 lm) or heparin (2 mg mL) had no effect on
Dinact (2APB: 39.5 2.1%; n = 6; heparin: 38.4 2.4%; n = 4;
Fig. 4E and F; 200 ms conditioning pulse length), peak current
amplitude (2APB: -1641 114 pA; heparin: )1351 91 pA) and
Rinact (2APB: 0.61 0.04; heparin: 0.58 0.02; conditioning pulse to
+10 mV). The possible contribution of store-operated Ca2+ entry was
probed by extracellular application of 2APB, which is also a blocker
of multiple types of transient receptor potential channels (Xu et al.,
2005). As is shown in Fig. 5, Dinact was signicantly (P < 0.05)
reduced to 26.5 1.8% (n = 6; Fig. 5B and C; 200 ms conditioning
pulse length) when TC neurons were challenged with extracellular
2APB (100 lm; Fig. 5A, lower left panel). This effect was accompanied by a signicant (P < 0.0001) increase in Rinact (0.74 0.01,
n = 6) but peak current amplitudes of the conditioning pulse to
+10 mV remained unchanged ()1453 94 pA, n = 6). Taken
together, these ndings indicate a contribution of ryanodine-sensitive
intracellular Ca2+ stores and CICR to Dinact, which depends on store
replenishment by extracellular Ca2+.

Ca 2+-dependent inactivation mainly affects L-type


Ca 2+ channels
It has been shown before that CICR as well as CDI in TC neurons
depend on L-type Ca2+ channels. A western blot analysis using a
CaV1.2-specic antibody revealed the expression of L-type Ca2+
channels in the membrane fractions of rat thalamic and hippocampal
tissue but not liver tissue (see Supplementary material, Fig. S1). In the
presence of nifedipine (1 lm), HVA Ca2+ current parameters were
signicantly (P < 0.01) altered with peak amplitudes and Rinact
averaging to )797 33 pA and 0.81 0.01 (n = 8), respectively.
Under these conditions, Dinact was signicantly (P < 0.0001) reduced
to 17.6 0.7% (n = 8; Fig. 6B; 200 ms conditioning pulse length).

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

CICR-dependent CDI in the thalamus 443

control
1

-0.4

10 mM
caffeine

-0.6
2

+10 mV

I/Imax

caffeine
1
2

-40 mV

2
1

-0.8
control

-1.0

1nA

-30
0
30
60
Conditioning pulse potential (mV)

200ms

***
*

**

*
(5)
(6)

40

(16)

(5)

**

40
30
(10)

10

15

20

Caffeine concentration (mM)

(14)

***

(14)

(10)

10 (14)

(120)

***
(15)

50

20

***
(14)

***
(14)

10 mM caffeine

60

(5)

Dinact (%)

50

35

70

***

55

45

control

80

60

Dinact (%)

***
***
******
**
******

***
(22)

(14)

**

100

10 M ryanodine

1000

Conditioning pulse length (ms)

Fig. 2. Effect of caffeine on CDI. (A) Scheme of the double-pulse protocol (left) and representative current traces recorded under control conditions (upper right
panel) and during extracellular application of caffeine (10 mm; lower right panel) with no conditioning pulse (1) and a conditioning pulse to +10 mV (2) are shown
(traces were taken from the same cell). (B) Mean IV relationship of normalized currents evoked by the test pulse under control conditions (closed squares) and in the
presence of 10 mm caffeine (closed circles). (C) Concentration-dependent effect of caffeine on Dinact. (D) Dependency of Dinact on the length of the conditioning pulse
in the presence of caffeine (10 mm; closed squares) and ryanodine (10 lm; closed triangles). Signicance was tested against values obtained with the shortest prepulse
duration (50 ms) for each substance. The dashed line indicates the t through the control data points shown in Fig. 1D. *P < 0.05, **P < 0.01, ***P < 0.001.

With L-type channels blocked, CPA (Dinact = 16.8 1.5%;


Rinact = 0.84 0.01; peak current amplitude = )888 41 pA; n =
5; Fig. 6C) as well as ryanodine (Dinact = 17.9 0.5%; Rinact =
0.77 0.02; peak current amplitude = )734 83 pA; n = 6; Fig. 6D)
were not able to induce an additional effect on HVA Ca2+ current
inactivation parameters and peak current amplitudes (conditioning
pulse to +10 mV). In order to assess the relative contribution of CDI
vs. voltage-dependent inactivation (VDI), Ba2+ (3 mm) was used as
charge carrier, whereas 11 mm BAPTA was included in the intracellular solution. Under these conditions, the mean peak current
amplitude was )792 75 pA, with Dinact and Rinact averaging
13.2 1.7% and 0.86 0.01 (n = 7), respectively (Fig. 6E; 200 ms
conditioning pulses; conditioning pulse to +10 mV).
Control of high-voltage-activated Ca2+ current amplitude by
Ca2+-induced Ca2+ release during trains of simulated action
potentials
In order to test functional aspects of the inuence of CICR on
CDI, trains of 200 action potential-like stimuli (see Fig. 7F for

pulse protocol) were applied at 30 Hz, which is a typical


frequency of tonic ring observed in a slice recording, from
a holding potential of )50 mV. Control and test pulses (200 ms,
from )40 to 0 mV) were applied 60 s before and immediately
after the train, respectively (Fig. 7A). To assure minimal interference with exogenous Ca2+ buffering, recordings were performed in the absence of the low-afnity Ca2+ buffers citrate and
gluconate. Under these conditions, the inward current amplitude
during train stimulation gradually declined and the peak current
amplitude of the test pulse with respect to the control pulse was
decreased by 36.5 0.7% (n = 9; Fig. 7B and E). When thapsigargin (10 lm) was present in the intracellular solution, the effect
of train stimulation was signicantly (P < 0.01) reduced with
the decrease in test pulse amplitude averaging 22.6 1.2% (n = 12;
Fig. 7C and E). With 3 mm Ba2+ in the extracellular solution
and 11 mm BAPTA in the intracellular solution, the degree of
current amplitude reduction was 12.5 1.7% (n = 6; Fig. 7D and
E). Taken together, these ndings indicate that CICR triggered by
trains of simulated action potentials contributes to CDI in TC
neurons.

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

444 V. Rankovic et al.

0.75

+60 mV

2, 2

Ratio 350/380

3, 3
2
1
3
20 m

0.60

control

0.50

20

100

0.05

60

80

100

0.10

1, 1

40

Time (s)

0.15

0.00

-40 mV

caffeine

Recovery to baseline (%)

Ratio 350/380

0.65

control
caffeine

0.20

+10 mV

1, 1

0.55

(1) soma
(2) somato-dendritic junction
(3) proximal dendrite
(4) background

0.70

2, 2

3, 3

control
caffeine

**

80

**
*

60
40
20
0

1, 1

2, 2

3, 3

Fig. 3. Ca2+ imaging in TC neurons. (A) Typical TC neuron loaded with 100 lm bis-Fura-2. Regions of interest in different cellular compartments are indicated.
(B) Recording of somatic uorescence ratio changes in TC neurons under control conditions (black trace) and during application of 10 mm caffeine (grey trace).
Ca2+ transients were elicited by three double pulses (see scheme) causing three Ca2+ peaks. In this set of experiments, each double-pulse application consisted of a
variable conditioning pulse ()40, +10 and +60 mV, 500 ms; 1, 2 and 3), a pause of 50 ms and a test pulse xed to +10 mV (200 ms; , and ). The resulting
three Ca2+ transients are labelled accordingly (rst double pulse: 1, ; second double pulse: 2, ; third double pulse: 3, ). For clarity, ratio traces were shifted by
2.5 s in time. The beginning of electrical stimulation is indicated by the arrowheads. Note that the recovery of intracellular Ca2+ concentration back to baseline levels
is delayed in caffeine after termination of electrical stimulation. (C) Bar graph representation of Ca2+ transient peak amplitudes (measured from the basal Ca2+ level
for each recording condition) under control conditions (black bars) and in the presence of caffeine (10 mm; white bars). (D) Relative recovery of uorescence ratio
back to baseline levels with respect to the transient peak amplitude at 10, 20 and 30 s after starting stimulation of the cell (control, black bars; 10 mm caffeine, open
bars). Signicance was calculated for the comparison of the two recording conditions. *P < 0.05, **P < 0.01.

Discussion
2+

In this study, evidence for a negative feedback of intracellular Ca


release on CDI of HVA Ca2+ channels in central neurons has been
demonstrated for the rst time. Our results can be summarized as
follows. (i) Classical double-pulse protocols revealed the occurrence
of CDI in TC neurons in acute brain slice preparations. CDI was
dependent on the length of the conditioning pulse. (ii) Caffeine, a
known activator of RyR, induced a concentration-dependent
increase in the degree of CDI, which was accompanied by an increase
in basal Ca2+ levels and a prolongation of HVA Ca2+ channel-evoked
Ca2+ transients. (iii) Ryanodine, a modulator of RyR, at a concentration known to block the receptors, decreased the degree of CDI for
conditioning pulses exceeding 100 ms in duration. (iv) Functional
inactivation of intracellular Ca2+ stores by application of the SERCA
pump inhibitors thapsigargin or CPA reduced the degree of
CDI despite an accompanying increased in basal intracellular Ca2+
levels. (v) Inhibition of store-operated Ca2+ inux reduces the degree

of CDI. (vi) Trains of action potential-like stimuli induced an ERdependent reduction in HVA Ca2+ current amplitude. (vii) The degree
of CDI in the presence of the specic L-type Ca2+ channel blocker
nifedipine was strongly reduced and close to the value that was
reached when Ca2+-dependent signalling was completely disabled by
using Ba2+ BAPTA. (viii) CaV1.2 channels were expressed in dLGN
tissue.

Ca 2+-dependent inactivation in thalamocortical relay neurons


The existence of CDI of HVA Ca2+ currents in TC neurons has been
shown before in acutely isolated cell preparations. Based on these
previous studies, the basic features of CDI in TC neurons are
as follows: (i) the degree of CDI is 3540% under control conditions;
(ii) L-type and Q-type channels are governed by CDI; and (iii) CDI is
inuenced by a number of cellular processes, including repetitive
neuronal activity, phosphorylation, Ca2+-binding proteins and the

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

CICR-dependent CDI in the thalamus 445

B
control

thapsigargin

CPA

ryanodine

heparin

2APB

1 nA

+10 mV

200 ms

60
-40 mV

Dinact (%)

50
40

(120)

30

***

***

(4)

(4)

(6)

***

(4)

(22)

20
10
He p a ir n

ar
in

in
e
ry
an
od

CP
A

n
th
ap
si

ga
rg
i

l
nt
ro
co

APB

Ry
CPA

he
p

haps i
T
c o n VC

2A
PB

Fig. 4. Inuence of the ER on CDI. Representative current traces (upper panels; see pulse protocol in the inset) and bar graph representation of Dinact (lower panels)
obtained under control conditions (A) and in the presence of thapsigargin (10 lm; extracellular application) (B), CPA (30 lm; extracellular application) (C),
ryanodine (10 lm; extracellular application) (D), 2APB (100 lm, intracellular application) (E) and heparin (2 mg mL; intracellular application) (F). The scale bars
in F apply to all current traces. ***P < 0.001.

-0.4

60

extracellular 2APB application

-0.6
I/Imax

control

* * * *

-0.8

2APB

* *

2APB

Dinact (%)

50
control

40

(120)

(6)

30
20

c o n VC

AP Be xt

2A
PB

-40 -20
0
20
40
60
Conditioning pulse potential (mV)

nt
ro

200 ms

-1.0

co

1 nA

10

Fig. 5. Inuence of store-operated Ca2+ inux on CDI. (A) Representative current traces recorded under control conditions (middle panel) and during extracellular
application of 2APB (100 lm; lower panel) elicited by the indicated pulse protocol (upper panel; all traces were taken from the same cell). (B) Mean IV relationship
of normalized currents evoked by the test pulse under control conditions (closed squares) and in the presence of 2APB (closed circles). (C) Bar graph representation
of Dinact under different recording conditions (as indicated). *P < <0.05.

cytoskeleton (Meuth et al., 2001, 2002, 2005). The present study adds
to these ndings by demonstrating an inuence of intracellular Ca2+
release on the degree of CDI. Although acutely isolated cell
preparations have the advantage of a low run-down of HVA Ca2+
currents (Budde et al., 1998), the cells are devoid of most of their
dendritic tree and synaptic connections. Here, we conrmed the
occurrence of CDI in TC neurons in the slice preparation by using the
well-established double-pulse protocol (Eckert & Tillotson, 1981;
Armstrong, 1989) and found a degree of CDI of 39% under
comparable control conditions. Although this is very similar to our
previous ndings, the data interpretation may be complicated by the
occurrence of HVA Ca2+ current run-down and a partial U-shape of

the test pulse IV. As a reduction of the driving force of Ca2+ currents
has only very little effect on the degree of CDI in TC neurons in slices
(this study) and after acute isolation (Meuth et al., 2001), an inuence
of run-down on the inactivation process seems unlikely. The partial
U-shape of the inactivation curve may be explained as follows. (i) The
Ca2+ imaging data presented here indicate that removal of Ca2+ from
the cytoplasm between voltage steps is incomplete with stimulation
frequencies around 0.1 Hz. Indeed, when the stimulation frequency
was reduced to 0.0167 Hz, the inactivation curve returned to baseline
with the most depolarized conditioning pulses. This view is in
agreement with the slow Ca2+ transients found under low-afnity but
high-capacity Ca2+-binding conditions (EGTA, citrate and gluconate)

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

446 V. Rankovic et al.

B
nifedipine

nifedipine
+ CPA

E
nifedipine
+ ryanodine

Ba2+/BAPTA

1 nA

control

200 ms

60

+10 mV

Dinact (%)

50
40

(120)

-40 mV

30
20

***

***

***

(8)

(5)

(6)

***
(7)

10
Ba 2 + BAPTA

2+

/B

AP
TA

Ni fRy

Ba

ne
ni
fe

di
pi

l
nt
ro
co

Ni fCP A

n
+ ifed
ry ip
an i n
od e
in
e

Nfi

c o n VC

n
+ ifed
CP ip
A ine

Fig. 6. Inuence of L-type Ca2+ channels on CDI. Representative current traces (upper panel; see inset for the pulse protocol) and bar graph representation of Dinact
(lower panel) obtained under control conditions (A), in the presence of nifedipine (1 lm; extracellular application) (B) and in the presence of nifedipine and CPA (C),
nifedipine and ryanodine (D), and Ba2+ (3 mm, as charge carrier) with BAPTA (11 mm, as Ca2+ chelator) (E). Scale bars in E apply to all current traces.
***P < 0.001.

(Meuth et al., 2002), which were also used in the present study.
(ii) The partially U-shaped curve is consistent with the dominance of
VDI over CDI at strongly depolarized potentials (Hadley & Hume,
1987; Hadley & Lederer, 1991). However, the shift in hysteresis of the
inactivation curve when the order of the conditioning pulses was
reversed (from the most depolarized potentials to the most hyperpolarized potentials) indicates only a small contribution of VDI with
200 ms pulses. The rather strong Dinact with long conditioning pulses
(5001000 ms) indicates that a signicant part of the inactivation can
be attributed to the slow VDI process during prolonged depolarization
(Lacinova, 2005; Lacinova & Hofmann, 2005).
Interaction between Ca 2+-induced Ca 2+ release and
Ca 2+-dependent inactivation
A prototypical feature of Ca2+ signalling in cardiac cells and neurons
is the release of Ca2+ from intracellular stores via the activation of
RyR by Ca2+ inux through closely connected L-type Ca2+ channels
(Berridge et al., 2000). In ventricular myocytes, a signicant fraction
of CDI depends on CICR and is therefore termed CICR-dependent
CDI (Grantham & Cannell, 1996; Takamatsu et al., 2003). Thus, the
negative feedback of intracellular Ca2+ release on CDI in TC neurons
is in good agreement with the detection of RyR2, RyR3 and CaV1.2 in
TC neurons and dLGN tissue by specic uorescent blockers (Budde
et al., 1998) and antibodies (Budde et al., 2000; Meuth et al., 2001).
In a previous study on isolated TC neurons it was assumed that 10 mm
of the low-afnity Ca2+ buffer citrate interfered with intracellular
signalling such that Ca2+ entry via the plasma membrane and the
subsequent release via the ER were separated in time (Meuth et al.,
2002). Although this may indeed be true for uorescent measurements
of bulk intracellular Ca2+ levels (Budde et al., 2000), the following

ndings of the present study argue in favour of a contribution of CICR


to CDI due to closely conned local mechanisms.
1. We have demonstrated previously that a supra-linear relation exists
between the amplitude of intracellular Ca2+ transients and the net
Ca2+ inux via HVA Ca2+ channels (the latter varied through
variations in pulse length), thereby demonstrating the occurrence of
CICR in TC neurons (Budde et al., 2000). In a similar way, the
degree of CDI revealed a steep increase with conditioning pulses
exceeding 100 ms duration, thereby displaying the classical
behaviour of a CICR-involving process (Hua et al., 1993; Llano
et al., 1994; Shmigol et al., 1995).
2. The stimulation by caffeine is a second characteristic that is
consistent between CICR and CDI. Caffeine interacts with RyR in
the ER and makes them more sensitive to Ca2+ so that Ca2+ release
mechanisms can be activated by smaller Ca2+ increases (Kostyuk
& Verkhratsky, 1995). As a typical result of caffeine stimulation,
the amplitude and duration of depolarization-induced Ca2+
transients are increased (Friel & Tsien, 1992; Kano et al., 1995;
Usachev & Thayer, 1997). Indeed, the Ca2+ imaging experiments
presented here revealed an increase in duration of Ca2+ transients
that was accompanied by an increase in the degree of CDI. The
lack of a signicant increase in Ca2+ transient amplitudes probably
based on the accompanying reduction in HVA Ca2+ current peak
amplitude.
3. The neuronal ER is equipped with Ca2+ pumps and Ca2+ release
channels that help to accumulate and liberate Ca2+ into and from
this intracellular compartment, respectively (Kostyuk & Verkhratsky, 1995). As indicated by the effects of the specic SERCA
pump blockers thapsigargin and CPA, SERCA-driven uptake of
Ca2+ into the ER participates in CDI. Furthermore, whereas

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

CICR-dependent CDI in the thalamus 447

test pulse

control pulse
0 mV
99 - 101

1-3

-40 mV

198 - 200

B
control

C
thapsigargin

Test current reduction (%)

Ba2+/
BAPTA

40

(5)

30

(8)

-45 to +50 mV
3ms

***
20

(6)

***

-50 to -45 mV
1 ms

10
0

+50 mV, 1 ms

c o n rt o
l

control

Ba B
/ AP TA

t a
hps i

thapsi

+50 to -50 mV
3 ms

2+

Ba /BAPTA

Fig. 7. Interaction between CDI and intracellular Ca2+ stores during trains of action potential-like stimuli. (A) Scheme of the stimulation protocol. A voltage
step (200 ms, from )40 to 0 mV) was applied 1 min before (control pulse) and immediately after (test pulse) a train (200 pulses of 30 Hz) of short
depolarizations. Current traces evoked by pulses before (left column), during (three middle columns; responses to the rst, second, third, 99th, 100th, 101st,
198th, 199th and 200th pulses in a train are shown) and following (right column) train stimulation under control conditions (B), during extracellular
application of thapsigargin (10 lm) (C) and using Ba2+ (3 mm, as charge carrier) with BAPTA (11 mm, as Ca2+ chelator) (D). (E) Mean bar graph
representation of the reduction of the test current peak amplitude in relation to the control current. (F) Scheme of the action potential-like voltage protocol.
***P < 0.001.

specic block of RyR signicantly reduces CDI, block of inositol


1,4,5-tris-phosphate receptors had no effect. The degree of CDI
for short conditioning pulses (50100 ms) under control conditions (23%) was in the same range as for longer (200 ms)
conditioning pulses in the presence of ER inhibitors (2530%),

thereby supporting the recruitment of CICR with longer conditioning pulse durations. It is interesting to note that the block of
transient receptor potential channels indicates that relling of
intracellular Ca2+ stores depends on Ca2+ inow from the
extracellular space.

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

448 V. Rankovic et al.


4. Although a correlation between HVA Ca2+ current and Ca2+
transient amplitudes has been described in TC neurons previously
(Meuth et al., 2002), the results of the present study indicate that an
increase in basal intracellular Ca2+ levels is not unambiguously
correlated with the degree of CDI. This is evident from the nding
that caffeine increased both the basal Ca2+ concentration and CDI,
whereas CPA increased the basal Ca2+ concentration but decreased
CDI. Also, the peak current amplitudes of HVA Ca2+ currents were
not unambiguously correlated to CDI. The 40% reduction of the
HVA Ca2+ current peak amplitude by reducing the driving force (all
types of HVA Ca2+ channels are activated by voltage steps) had no
inuence on CDI, whereas a very similar reduction of the HVA
Ca2+ current peak amplitude by nifedipine (only non-L-type Ca2+
channels are activated by voltage steps) was accompanied by a
strong decrease in CDI. These data indicate that a modulation of
CDI only occurs if a component of the CDI machinery is
specically targeted.
5. Although the reduction of CDI by nifedipine was signicantly
(P < 0.001) stronger compared with ryanodine, thapsigargin or
CPA, modulators of ER function (CPA and ryanodine) had no
additional effect on cells pretreated with the L-type Ca2+ channel
blocker. Therefore, the present study emphasizes the central role of
L-type Ca2+ channels in CDI as these channels are both governed
by CDI (Meuth et al., 2001, 2002) and trigger CICR (Budde et al.,
2000) in TC neurons.
6. It is concluded that the coupling between L-type Ca2+ channels and
RyR in TC neurons occurs in a closely conned area and persists in
the presence of moderate exogenous Ca2+ buffering. Increased pulse
duration (allowing increased lateral diffusion of Ca2+) or the
application of caffeine (increasing the sensitivity of RyR for Ca2+)
is necessary to switch this interaction from a local to a more global
event (resulting in increased CDI). This conclusion is corroborated
by previous ndings in TC neurons and observations in heart cells.
In TC neurons, the clustering of L-type Ca2+ channels on the plasma
membrane (Budde et al., 1998), and spatial spread of CICR across
the ER, may ll up a shell of Ca2+ ions beneath the plasma
membrane and allow Ca2+ cooperativity (Sherman et al., 1990;
Kay, 1991; Bertram et al., 1999; Meuth et al., 2001). In heart cells,
experimental evidence strongly supports local control of Ca2+
release where a cluster of RyR is juxtaposed with a group of L-type
Ca2+ channels, forming a couplon or Ca2+ release unit that
functions autonomously, and a progressive recruitment of couplons
is the basis of graded Ca2+ release (Zahradnikova et al., 2007;
Polakova et al., 2008; Fowler, 2009). Although high concentrations
of intracellular EGTA are necessary to prevent spatial overlap
between Ca2+ released from neighbouring couplons, interaction
between couplons is enhanced by the b-adrenoceptor-stimulated
recruitment of L-type Ca2+ channels and Ca2+ release units (Song
et al., 1998, 2001). In addition, the ability of L-type Ca2+ channelbound calmodulin to sense strong local and weaker global Ca2+
signals in a lobe-specic manner may contribute to this scenario
(Calin-Jageman & Lee, 2008).

oscillations (Pedroarena & Llinas, 1997; Lo et al., 2002; Pape et al.,


2004). Indeed, tonic action potential ring (Budde et al., 2000), as
well as repetitive activation of HVA Ca2+ currents (Meuth et al.,
2002), induces an increase in the intracellular Ca2+ concentration that
follows a saturating hyperbolic function, thereby pointing to the Ca2+limiting inuence of CDI (in combination with VDI). The present
ndings add to this scenario by demonstrating that tonic stimulation
with action potential-like voltage steps induces a decrease in HVA
Ca2+ current amplitude that partially depends on CICR. Therefore, the
close functional (and spatial) coupling between Ca2+ entry via HVA
Ca2+ channels, CICR, CDI and the relling of intracellular Ca2+ stores
tightly controls intracellular Ca2+ levels and, thus, Ca2+-dependent
processes. It is interesting to note that caffeine-induced oscillations of
intracellular Ca2+ levels in sympathetic ganglia cells are based on a
similar combination of Ca2+ signalling tools, namely CICR, CDI of
plasma membrane Ca2+ inux and two types of Ca2+ pumps
(Cseresnyes et al., 1999).
In summary, our ndings indicate that CDI in TC neurons is
partially under the control of CICR, thereby pointing to a crosssignalling between L-type Ca2+ channels and RyR which controls the
amount of Ca2+ inux during neuronal activity.

Supporting Information
Additional supporting information may be found in the online version
of this article:
Fig. S1. Western blot analysis of the membrane fraction of rat
hippocampus, thalamus, and liver.
Please note: As a service to our authors and readers, this journal
provides supporting information supplied by the authors. Such
materials are peer-reviewed and may be re-organized for online
delivery, but are not copy-edited or typeset by Wiley-Blackwell.
Technical support issues arising from supporting information (other
than missing les) should be addressed to the authors.

Acknowledgements
This work was supported by DFG (BU 1019 8-1 and GRK 1167-P1). P.E. was
a fellow of the Otto Creutzfeldt Center for Cognitive and Behavioral
Neuroscience Munster. Thanks are due to A. Jahn, M. Marunde, E. Na and
R. Ziegler for excellent technical assistance. This work was performed in partial
fulllment of the PhD theses of P.E. and V.R.

Abbreviations
2APB, 2-aminoethoxy diphenyl borate; CDI, Ca2+-dependent inactivation;
CICR, Ca2+-induced Ca2+ release; CPA, cyclopiazonic acid; Dinact, degree of
inactivation; dLGN, dorsal part of the lateral geniculate nucleus; ER,
endoplasmic reticulum; HVA, high-voltage-activated; IV, current-voltage;
Rinact, inactivation ratio; RyR, ryanodine receptors; SERCA, sarcoplasmic endoplasmic Ca2+ ATPase; TC, thalamocortical relay; VDI, voltagedependent inactivation.

References
2+

Possible functional relevance of high-voltage-activated Ca


channels and Ca 2+-induced Ca 2+ release-dependent Ca 2+dependent inactivation

It has been suggested previously that HVA Ca2+ channels and their
inherent CDI process are most relevant for neuronal activity of dLGN
TC neurons seen during depolarized membrane states, including tonic
ring of action potentials, plateau-like depolarizations evoked by optic
tract stimulation during early development and fast high-threshold

Armstrong, D.L. (1989) Calcium channel regulation by calcineurin, a Ca2+activated phosphatase in mammalian brain. Trends Neurosci., 12, 117122.
Bardo, S., Cavazzini, M.G. & Emptage, N. (2006) The role of the endoplasmic
reticulum Ca2+ store in the plasticity of central neurons. Trends Pharmacol.
Sci., 27, 7884.
Berridge, M.J., Lipp, P. & Bootman, M.D. (2000) The versatility and
universality of calcium signalling. Nat. Rev. Mol. Cell Biol., 1, 1121.
Bertram, R., Smith, G.D. & Sherman, A. (1999) Modeling study of the effects
of overlapping Ca2+ microdomains on neurotransmitter release. Biophys. J.,
76, 735750.

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

CICR-dependent CDI in the thalamus 449


Budde, T., Munsch, T. & Pape, H.-C. (1998) Distribution of L-type calcium
channels in rat thalamic neurons. Eur. J. Neurosci., 10, 586597.
Budde, T., Sieg, F., Braunewell, K.H., Gundelnger, E.D. & Pape, H.-C. (2000)
Ca2+-induced Ca2+ release supports the relay mode of activity in thalamocortical cells. Neuron, 26, 483492.
Budde, T., Meuth, S. & Pape, H.-C. (2002) Calcium-dependent inactivation of
neuronal calcium channels. Nat. Rev. Neurosci., 3, 873883.
Calin-Jageman, I. & Lee, A. (2008) Ca(v)1 L-type Ca2+ channel signaling
complexes in neurons. J. Neurochem., 105, 573583.
Chavis, P., Fagni, L., Lansman, J.B. & Bockaert, J. (1996) Functional coupling
between ryanodine receptors and L-type calcium channels in neurons.
Nature, 382, 719722.
Coulon, P., Herr, D., Kanyshkova, T., Meuth, P., Budde, T. & Pape, H.-C.
(2009) Burst discharges in neurons of the thalamic reticular nucleus are
shaped by calcium-induced calcium release. Cell Calcium, 46, 333346.
Cseresnyes, Z., Bustamante, A.I. & Schneider, M.F. (1999) Caffeine-induced
[Ca2+] oscillations in neurones of frog sympathetic ganglia. J. Physiol.
(Lond.), 514(Pt 1), 8399.
Cueni, L., Canepari, M., Adelman, J.P. & Luthi, A. (2009) Ca(2+) signaling by
T-type Ca(2+) channels in neurons. Pugers Arch., 457, 11611172.
Eckert, R. & Tillotson, D.L. (1981) Calcium-mediated inactivation of the
calcium conductance in caesium-loaded giant neurones of Aplysia californica. J. Physiol. (Lond.), 314, 265280.
Fowler, M.R. (2009) Local is as local does: the unitary nature of SR Ca2+
release in cardiac ventricular myocytes. J. Physiol. (Lond.), 587, 301302.
Friel, D.D. & Tsien, R.W. (1992) A caffeine- and ryanodine-sensitive Ca2+
store in bullfrog sympathetic neurones modulates effects of Ca2+ entry on
[Ca2+]i. J. Physiol. (Lond.), 450, 217246.
Grantham, C.J. & Cannell, M.B. (1996) Ca2+ inux during the cardiac action
potential in guinea pig ventricular myocytes. Circ. Res., 79, 194200.
Hadley, R.W. & Hume, J.R. (1987) An intrinsic potential-dependent inactivation mechanism associated with calcium channels in guinea-pig myocytes.
J. Physiol. (Lond.), 389, 205222.
Hadley, R.W. & Lederer, W.J. (1991) Ca2+ and voltage inactivate Ca2+ channels
in guinea-pig ventricular myocytes through independent mechanisms.
J. Physiol. (Lond.), 444, 257268.
Hua, S.Y., Nohmi, M. & Kuba, K. (1993) Characteristics of Ca2+ release
induced by Ca2+ inux in cultured bullfrog sympathetic neurones. J. Physiol.
(Lond.), 464, 245272.
Kano, M., Garaschuk, O., Verkhratsky, A. & Konnerth, A. (1995) Ryanodine
receptor-mediated intracellular calcium release in rat cerebellar Purkinje
neurones. J. Physiol. (Lond.), 487.1, 116.
Kay, A.R. (1991) Inactivation kinetics of calcium currents of acutely
dissociated CA1 pyramidal cells of the mature guinea-pig hippocampus.
J. Physiol. (Lond.), 437, 2748.
Kostyuk, P.G. & Verkhratsky, A.N. (1995) Calcium Signaling in the Nervous
System. Wiley, Chichester, New York.
Lacinova, L. (2005) Voltage-dependent calcium channels. Gen. Physiol.
Biophys., 24(Suppl. 1), 178.
Lacinova, L. & Hofmann, F. (2005) Ca2+- and voltage-dependent inactivation
of the expressed L-type Ca(v)1.2 calcium channel. Arch. Biochem. Biophys.,
437, 4250.

Llano, I., DiPolo, R. & Marty, A. (1994) Calcium-induced calcium release in


cerebellar Purkinje cells. Neuron, 12, 663673.
Lo, F.S., Ziburkus, J. & Guido, W. (2002) Synaptic mechanisms regulating the
activation of a Ca2+-mediated plateau potential in developing relay cells of
the LGN. J. Neurophysiol., 87, 11751185.
Meuth, S.G., Budde, T. & Pape, H.-C. (2001) Differential control of highvoltage activated Ca2+ current components by a Ca2+-dependent inactivation
mechanism in thalamic relay neurons. Thalamus Relat. Syst., 1, 3138.
Meuth, S.G., Pape, H.-C. & Budde, T. (2002) Modulation of Ca2+ currents in
rat thalamocortical relay neurons by activity and phosphorylation. Eur.
J. Neurosci., 15, 16031614.
Meuth, S.G., Kanyshkova, T., Landgraf, P., Pape, H.C. & Budde, T. (2005)
Inuence of Ca 2+-binding proteins and the cytoskeleton on Ca2+-dependent
inactivation of high-voltage activated Ca2+ currents in thalamocortical relay
neurons. Pugers Arch., 450, 111122.
Pape, H.-C., Munsch, T. & Budde, T. (2004) Novel vistas of calcium-mediated
signalling in the thalamus. Pugers Arch., 448, 131138.
Pedroarena, C. & Llinas, R. (1997) Dendritic calcium conductances generate
high-frequency oscillation in thalamocortical neurons. PNAS, 94, 724728.
Polakova, E., Zahradnikova, A. Jr, Pavelkova, J., Zahradnik, I. & Zahradnikova, A. (2008) Local calcium release activation by DHPR calcium channel
openings in rat cardiac myocytes. J. Physiol., 586, 38393854.
Sherman, A., Keizer, J. & Rinzel, J. (1990) Domain model for Ca2+inactivation of Ca2+ channels at low channel density. Biophys. J., 58, 985
995.
Shmigol, A., Verkhratsky, A. & Isenberg, G. (1995) Calcium-induced calcium
release in rat sensory neurons. J. Physiol. (Lond.), 489.3, 627636.
Song, L.S., Sham, J.S., Stern, M.D., Lakatta, E.G. & Cheng, H. (1998) Direct
measurement of SR release ux by tracking Ca2+ spikes in rat cardiac
myocytes. J. Physiol. (Lond.), 512(Pt 3), 677691.
Song, L.S., Wang, S.Q., Xiao, R.P., Spurgeon, H., Lakatta, E.G. & Cheng, H.
(2001) beta-Adrenergic stimulation synchronizes intracellular Ca(2+) release
during excitation-contraction coupling in cardiac myocytes. Circ. Res., 88,
794801.
Takamatsu, H., Nagao, T., Ichijo, H. & Adachi-Akahane, S. (2003) L-type
Ca2+ channels serve as a sensor of the SR Ca2+ for tuning the efcacy of
Ca2+-induced Ca2+ release in rat ventricular myocytes. J. Physiol. (Lond.),
552, 415424.
Usachev, Y.M. & Thayer, S.A. (1997) All-or-none Ca2+ release from
intracellular stores triggered by Ca2+ inux through voltage-gated Ca2+
channels in rat sensory neurons. J. Neurosci., 17, 74047414.
Xu, S.Z., Zeng, F., Boulay, G., Grimm, C., Harteneck, C. & Beech, D.J.
(2005) Block of TRPC5 channels by 2-aminoethoxydiphenyl borate: a
differential, extracellular and voltage-dependent effect. Br. J. Pharmacol.,
145, 405414.
Yasuda, H., Higashi, H., Kudo, Y., Inoue, T., Hata, Y., Mikoshiba, K. &
Tsumoto, T. (2003) Imaging of calcineurin activated by long-term depression-inducing synaptic inputs in living neurons of rat visual cortex. Eur. J.
Neurosci., 17, 287297.
Zahradnikova, A. Jr, Polakova, E., Zahradnik, I. & Zahradnikova, A. (2007)
Kinetics of calcium spikes in rat cardiac myocytes. J. Physiol. (Lond.), 578,
677691.

The Authors (2010). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 31, 439449

Neuroscience 165 (2010) 371385

THALAMIC AFFERENT ACTIVATION OF SUPRAGRANULAR LAYERS


IN AUDITORY CORTEX IN VITRO: A VOLTAGE SENSITIVE DYE
STUDY
T. BROICHER,a1 H.-J. BIDMON,b,c B. KAMUF,a
P. COULON,a,d* A. GORJI,a H.-C. PAPE,a,d
E.-J. SPECKMANNa AND T. BUDDEa,d

pragranular and granular layers in the auditory cortex of adult


mice. This activation is predominantly mediated by nonNMDA receptors, while GABAA receptor-mediated inhibition
limits the horizontal and vertical spread of activity. 2010
IBRO. Published by Elsevier Ltd. All rights reserved.

a
Westflische Wilhelms-Universitt Mnster, Institut fr Physiologie I,
Robert-Koch-Strasse 27a, D-48149 Mnster, Germany
b
C. u. O. Vogt Institut fr Hirnforschung, Universittsstr. 1, D-40225
Dsseldorf, Germany

Key words: thalamocortical, voltage sensitive dye, thalamus,


cortex, auditory.

Biologisch-Medizinisches Forschungszentrum, Universittsstr. 1,


D-40225 Dsseldorf, Germany

The initial step in cortical processing of sensory information is the arrival of afferent information via the thalamus.
Except for olfactory signals, all sensory information is relayed to primary sensory cortical areas through specific
thalamic nuclei. From the primary sensory areas the information is thought to be projected to second and higher
order cortical areas, initializing a complex activity pattern,
including parallel processing and feedback projections.
Thalamocortical brain slice preparations have proven
to be valuable tools in the study of thalamocortical interactions. Previous in vitro studies relied on stimulation on
the border between layer 6 and the white matter, or of layer
4, to mimic thalamic input to the cortex. To date, three
different thalamocortical brain slice preparations have
been described: one somatosensory (Agmon and Connors, 1991), one auditory (Cruikshank et al., 2002), and
one visual (MacLean et al., 2006). The most thoroughly
studied is the somatosensory thalamocortical slice, in
which optical imaging has been successfully used to visualize cortical responses to thalamic input. Using these
methods, it has been shown that thalamic input to the
barrel cortex evokes cortical excitation initially in layers 4
and 5, from where the activity spreads in vertical and
horizontal directions (Laaris et al., 2000; Llinas et al.,
2002).
The primary auditory cortex (A1) is believed to differ
from other well studied primary sensory areas (visual, V1,
and somatosensory, S1) in terms of localization of
thalamocortical input and intracortical circuitry (Atzori et
al., 2001; Linden and Schreiner, 2003; Winer and Lee,
2007; Barbour and Callaway, 2008). In contrast to the
concentrated thalamocortical innervation of spiny stellate
neurons of layer 4 in S1 and V1, the auditory cortex seems
to receive afferents in a more graded fashion, with the
majority of terminations being located in layers 3 and 4.
Furthermore, the thalamorecipient cell types differ between the cortices, due to the absence of spiny stellate
cells in A1 (Smith and Populin, 2001; Linden and Schreiner, 2003; Winer et al., 2005; Barbour and Callaway,
2008).

Institut fr Experimentelle Epilepsieforschung, Robert-KochStrasse 27a, 48149 Mnster, Germany

AbstractWe studied auditory thalamocortical interactions


in vitro, using an auditory thalamocortical brain slice preparation. Cortical activity evoked by electrical stimulation of the
medial geniculate nucleus (MGN) was investigated through
field potential recordings and voltage sensitive dyes. Experiments were performed in slices obtained from adult mice
(9 14 weeks). Stimulus evoked activity was detected in the
granular and supragranular layers after a short latency (5 6
ms). In 9 14 weeks old mice infragranular activity was detected in 10 of 24 preparations and was found to be increased
in younger mice (p 31 64). In 14 of 24 slices a prominent
horizontal spread was observed, which extended into cortical
areas lateral to A1. In these experiments, the shortest onset
latencies and largest signal amplitudes were located in the
supragranular layers of A1. In areas lateral to A1, shortest
onset latencies were located in the granular layer, while largest signal amplitudes were found in the supragranular layers.
Evoked cortical activity was sensitive to removal of extracellular Ca2 or application of 6-cyano-7-nitroquinoxaline-2,3dione (CNQX, 10 M). Short repetitive stimulation, resembling thalamic burst activity (three pulses at 100 Hz), resulted
in an increase of signal amplitude and excited area by 25%,
without changing the overall spatiotemporal activity profile.
Blockade of N-methyl-D-aspartate receptors by 2-amino-5phosphonopentanoate (AP5, 50 M) reduced amplitudes and
excited area by 1530%, irrespective of stimulation frequency. Application of bicuculline (10 M) greatly increased
cortical responses to thalamic stimulation. Under these conditions, evoked activity displayed a pronounced horizontal
spread in combination with a 23-fold increase in amplitude.
In conclusion, afferent thalamic inputs primarily activate su1
Present address: Department of Bioengineering, University of Utah,
20 South 2030 East, B.P. RB, Room 108, Salt Lake City, UT 84112,
USA
*Corresponding author. Tel: 49-251-83-58112; fax: 49-251-83-55551.
E-mail address: coulon@uni-muenster.de (P. Coulon) or tilman.
broicher@utah.edu (T. Broicher).
Abbreviations: ACSF, artificial cerebrospinal fluid; AuV, ventral secondary auditory cortex; A1, primary auditory cortex; AP5, amino-5phosphonopentanoate; HF, high frequency; MGN, medial geniculate
nucleus; NMDA, N-methyl-D-aspartate; PBS, phosphate buffered saline; S1, primary somatosensory cortex; TeA, association area of the
temporal lobe; WFA, Wisteria floribunda agglutinin.

0306-4522/10 $ - see front matter 2010 IBRO. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuroscience.2009.10.025

371

372

T. Broicher et al. / Neuroscience 165 (2010) 371385

It was shown recently, that thalamic neuronal signaling


mechanisms undergo significant developmental alterations within the first 60 postnatal days (Kanyshkova et al.,
2009). To the best of our knowledge, the spatiotemporal
profile of activity evoked by stimulation of the MGN in
auditory thalamocortical slices has not been described in
adult animals. Previous studies using this preparation employed juvenile mice (Rose and Metherate, 2001; Cruikshank et al., 2002; Kaur et al., 2005), while investigations in
rats (Kubota et al., 1997, 1999) relied on stimulation of
the border between layer 6 and the white matter or of the
cortex to mimic thalamic input to the auditory cortex. In the
present study, we adapted the auditory thalamocortical
brain slice preparation developed by Cruikshank et al.,
(2002) to study cortical activity evoked by stimulation of the
MGN in adult mice. We used a combination of field potential and voltage sensitive dye recordings to investigate
whether cortices of adult mice show a different activity
pattern than described for younger animals (Cruikshank et
al., 2002). We found that supragranular excitation was
markedly more pronounced in adult mice (9 14 weeks)
than was described previously for younger animals (p 13
19) (Cruikshank et al., 2002). Moreover, we found that
infragranular activity underwent developmental changes.
Another focus of this approach lay on obtaining a brain
slice preparation which would allow us to measure onset
latencies, spatial propagation, and area of evoked cortical
activity in the thalamocortical system of adult mice. We
intend to use this preparation in future studies to investigate the influence of pathological changes in conduction
velocities by demyelination on cortical activity evoked by
thalamic stimulation.
Here, we describe the responses of auditory cortical
areas to thalamic input, as assessed by voltage sensitive
dye and field potential recordings, and compare these
results to known responses of other sensory areas and to
previous descriptions of the auditory cortex (Albowitz and
Kuhnt, 1993a,b; Nelson and Katz, 1995; Yuste et al., 1997;
Laaris et al., 2000; Contreras and Llinas, 2001; Rose and
Metherate, 2001, 2005; Beierlein et al., 2002; Cruikshank
et al., 2002; Llinas et al., 2002; Kaur et al., 2005; Lee and
Sherman, 2008; Sato et al., 2008).

EXPERIMENTAL PROCEDURES
Unless stated otherwise, the experimental procedures were performed on adult mice (C57BL/6, 9 14 weeks) of both genders and
all chemicals were obtained from Sigma (Sigma-Aldrich, Munich,
Germany). All animal procedures were approved by the local
authorities.

Preparation of brain slices


To obtain slices containing intact auditory thalamocortical projections, we adapted a method developed by Cruikshank et al. (2002)
(see also Kawai et al., 2007). Mice were deeply anesthetized with
isoflurane (Abbott, Wiesbaden, Germany) and decapitated. Brains
were removed and glued with the dorsal side onto a 25 agarramp (Fig. 1). Subsequently, the brains were placed into a vibratome and superfused with 4 C artificial cerebrospinal fluid
(ACSF, containing in mM: NaCl, 124; KCl, 4; NaH2PO4, 1.24;
MgSO4, 1.3; NaHCO3, 26; CaCl2, 2; Glucose, 10) and cut into 500

Fig. 1. Agar ramp used for preparation of angled horizontal slices. (A)
Teflon form used to produce agar ramps. The depicted form yields 20
ramps. (B) Posterior view of mouse brain glued on a 25 agar ramp.
The ramp is glued onto the vibratome mounting platform. (C) Birds eye
view of a mouse brain glued on a 25 agar ramp in the vibratome.
Blade used to cut slices is visible in the upper right corner. The
cerebellum and the bulbus olfactorius have been removed. Scale Bars
indicate 1 cm.

m sections. Slices were stained with 12.5 g/ml of the voltage


sensitive dye RH 795 (Invitrogen, Karlsruhe, Germany, dissolved
in ACSF at 31 C, 1 h incubation), before being transferred into a
holding chamber filled with ACSF at 31 C. All solutions used
during preparation and staining were gassed with carbogen (95%
O2 and 5% CO2), stabilizing the pH at 7.4. Slices were allowed to
rest for 60 min before recording commenced. In about twothirds of the preparations, one slice per animal contained intact
thalamocortical projections.

Patch-Clamp
Patch-Clamp recordings were performed as described previously
(Broicher et al., 2008). Briefly, whole cell recordings in brain slice
preparations were done using glass microelectrodes pulled from

T. Broicher et al. / Neuroscience 165 (2010) 371385


borosilicate glass capillaries (GC150T-10, Clark Electromedical
Instruments, Pangbourne, UK) and filled with (in mM): K-gluconate, 95; K3-citrate, 20; NaCl, 10; CaCl2, 0.5; MgCl2, 1; HEPES,
10; K-BAPTA, 5; phosphocreatine, 15; Na2-GTP, 0.5; MgATP, 3.
The pH was set to 7.24 with KOH, and the osmolality was measured (290 mOsm/kg). Electrodes were connected to an
EPC-10 amplifier (HEKA Elektronik, Friedland, Germany) and had
a resistance ranging from 1.8 to 2.5 M. Pulse software (HEKA)
was used for stimulus application and recording of data. For
current-clamp experiments in brain slice preparations, the following bath solution was used (in mM): NaCl, 120; KCl, 2.5; CaCl2, 2;
MgSO4, 2; NaH2PO4, 1.25; HEPES, 30; dextrose, 10. The pH was
adjusted to 7.24 with NaOH and the osmolality was measured to
be 300 mOsm/kg. Series resistance ranged from 5 to 15 M. All
recordings were performed at room temperature.

Recording and analysis of voltage sensitive dye


signals and field potentials
Optical and electrical signals were recorded in a submerged-type
recording chamber mounted on an inverted microscope (Zeiss,
Gttingen, Germany). Optical recordings were performed as described previously (Khling et al., 2000, 2002; Straub et al., 2003).
Briefly, fluorescence signals were acquired by a 464 element
hexagonal photodiode array, through a 20 or 10 objective,
corresponding to a total imaged area of 0.416 or 1.856 mm2
per array position, respectively. Sampling intervals were 1.274 ms
and Xenon lamp illumination was applied for 1300 ms. Detection
periods consisted of three trials (30 s interval) and were separated
by at least 3 min. Excitation wavelength was bandpass filtered at
54620 nm, emitted light was low pass filtered at 590 nm, allowing the transmission of the excitation and emission maxima of 530
and 712 nm, respectively. A dichroic mirror was used to reflect the
excitation light and to allow transmission of the emitted light (cutoff wavelength 580 nm, FT580, Zeiss). Recording of optical signals was governed by Neuroplex software (RedShirtImaging, Decatur, GA, USA). Stimuli were applied via a custom-made concentric bipolar electrode placed in the MGN close to the exit of the
superior thalamic radiation (Fig. 5A, B). Pulse width of individual
stimuli and stimulation intensity were set to obtain optimal responses and were kept constant for the rest of the experiment
(150 450 s; 1 mA). Optical signals were expressed as fractional changes of resting light intensity (IrestIrecording/Irest). Unfiltered averages of seven adjacent diodes (0.006 mm2 @ 20 and
0.028 mm2 @ 10 objective) and three trials were used for
analysis of signal amplitudes, onset latencies, duration, and normalized amplitude. Analysis of optical signals was done using
Neuroplex, MATLAB (MathWorks, Ismaning, Germany), and Origin (OriginLab, Additive GmbH, Friedrichsdorf, Germany) software. All optical signals were inverted. For the analysis of amplitudes and latency, one (20 objective) optical signal or two averaged optical signals (10 objective) per cortical layer group
(infragranular, granular, and supragranular) and array position
were used. Amplitudes are given as peak amplitude minus baseline amplitude (mean of the 76 ms preceding the stimulus). Amplitude maxima were taken from the first 25 ms after stimulus.
Onset latency was determined manually. For the analysis of the
area excited by the stimulus, maximal amplitudes were compared
to the baseline. If the amplitude exceeded the mean baseline
amplitude plus four times the baseline standard deviation, a diode
was considered activated. To determine the duration of optical
signals, the baseline was fitted with a linear function and the time
needed to cross the linear fit of the baseline, after the maximal
signal amplitude was reached, was taken as a measure for the
duration of the optical signal. For construction of cortical input
maps, data were converted to pseudocolor images and scaled to
the maximal signal of the recording session. Activity of the three
cortical layers was analyzed and compared between four age groups
(see Fig. 4B). To this end, normalized averaged amplitudes were

373

measured 40 ms after the stimulus and age dependent infragranular


activity is given as the averaged amplitude of optical signals relative
to the averaged amplitude measured in the granular and supragranular layers (norm. amplitude, Fig. 4B). Cortical field potentials were
recorded using ACSF filled borosilicate glass pipettes with 0.51.5
M resistance, connected to an AgAgClKCl bridge. Data were
acquired using a custom-made amplifier and AxoScope software
(Molecular Devices, Sunnyvale, CA, USA).
Data are presented as meanstandard error, unless indicated otherwise. Statistical significances were determined using
unpaired and paired two sample t-tests.

Histology
For biocytin staining, slices were cut as described above and
placed into an interface chamber at 30 C in standard ACSF.
Biocytin crystals were applied to the primary auditory cortex using
the tip of a needle. After biocytin application, slices were incubated
for 6 h in the interface chamber. Then, slices were fixed in 4%
paraformaldehyde (PFA) in phosphate buffered saline (PBS) overnight. For immunohistochemistry of slices used in physiological
experiments, a 4% formamide (FA) fixation was used. After fixation, the brain slices (500 m) were soaked in PBS (0.1 M, pH 7.4)
containing 25% sucrose (w/v) and 10% glycerol (v/v) for 24 h at
4 C. The slices were oriented flat onto the tissue holder of the
Frigomobil (Leica, Bensheim, Germany) combined with a microtome (Model SM 2000 R; Leica) and frozen. The slices were
resectioned to a thickness of 50 m, rinsed in PBS three times for
10 min and endogenous peroxidase activity was blocked by incubating all slices in PBS containing 1% H2O2 (Sigma) for 20 min.
Slices were rinsed three times in PBS and alternate slices were
either used for immunohistochemistry or for biocytin tracing.

Biocytin tracing
For biocytin labeling we followed a previously described protocol
(Staiger et al., 1999) with minor modifications. The 50 m thick
slices were incubated in PBS containing 10% normal goat serum
(NGS, Vector Laboratories, Burlingame) and 0.3% saponin
(Sigma) for 30 min followed by an incubation in AB-Complex (ABC
elite kit, Vector Laboratories, Peterborough, UK) at a final dilution
of 1:300 with PBS for 90 min, according to manufacturers instructions. Slices were consequently rinsed in PBS for 5 min in PBS
1, TrisHCl (0.05 M; pH 7.6) 2, and TrisHCl (pH 8.1) 2
followed by a pre-incubation in Ni-DAB solution (50 ml TrisHCl,
pH 8.1, containing 7.5 mg 3=-3-diaminobenzidine tetrahydrochloride (DAB, Sigma) and 0.2 g ammonium nickel sulfate (Ni, Fluka,
Buchs, Switzerland)) for 10 min. Afterward the Ni-DAB solution
was replaced by a freshly prepared solution containing in addition
0.01% H2O2 to induce the peroxidase reaction. The reaction was
terminated according to visual inspection after approximately 10
min. Sections were rinsed in TrisHCl (0.05 M, pH 7.6) mounted
on glass slides, air-dried, cleared in xylene, and coverslipped with
Entellan (Merck, Darmstadt, Germany).

Cortical parcellation
Sections collected for the identification of functional cortical
subdivisions were immunohistochemically and histochemically
stained according to previous protocols (Bidmon et al., 1997)
using antibodies SMI 31 (Sternberger monoclonal Inc., Baltimore,
MD, USA) for the labeling of axons, SMI 311 for the labeling of
non-phosphorylated neurofilaments, or Wisteria floribunda agglutinin (WFA, Sigma) histochemistry for the region-specific presence
of the so-called perineuronal nets (Celio and Blumcke, 1994). In
brief, for immunohistochemistry, sections were incubated in 2%
NGS (Vector Laboratories) containing 10% fetal calf serum, dissolved in PBS containing 0.3% saponin for 1 h, followed by an
incubation either in antibody SMI 31 (final dilution 1:500), or in

374

T. Broicher et al. / Neuroscience 165 (2010) 371385


antibody SMI 311 (final dilution 1:300) for 48 h at 4 C under gentle
shaking. Afterward, the sections were rinsed in PBS (415 min
each) and incubated in biotinylated secondary antibody (goat
anti-mouse, final dilution 1:500, Dianova, Hamburg, Germany)
containing 1% NGS for 24 h at 4 C. After rinsing the section 4
times in PBS (15 min each), they were incubated in AB-Complex
(ABC elite kit, Vector Laboratories) at a final dilution of 1:300 with
PBS for 90 min. Sections were subsequently rinsed for 5 min in
PBS 1, TrisHCl (0.05 M; pH 7.6) 2, and TrisHCl (pH 8.1) 2,
followed by a pre-incubation in DAB (final concentration 0.75
mg/ml) for 10 min, which was then replaced by freshly prepared
DAB-solution containing 0.01% H2O2 to start the reaction. The
reaction was terminated by several rinses in TrisHCl (pH 7.6).
Sections were mounted on slides, air-dried, cleared, and coverslipped as described above. For the cortical parcellation by WFA
(Sigma), we used a concentration of 1 g WFA/ml PBS following
a standard staining procedure (Bruckner et al., 1994).
Sections were microscopically evaluated, and single high resolution images were produced using a microscope (Axioplan 2,
Zeiss) equipped with a CCD camera (AxioCam HRc) using AxioVision Rel. 4.5 software (Zeiss). For the imaging of whole sections
at higher resolution, we used a computer linked scanning microscope (4 or 10 apo-plan objectives; Zeiss) using
AxioVision module MOSAIX (Zeiss). The final resolution was 1
m/pixel (4 objective) or 0.52 m/pixel (10 objective, Fig. 2A).
Afterward, the obtained images were converted into TIF-format for
further processing.

RESULTS
We recorded stimulus evoked cortical activity in auditory
thalamocortical preparations (n24 slices with intact connections, slices derived from 23 animals aged p 6597). To
verify connectivity in our preparations, we injected biocytin
into the primary auditory cortex (A1) of thalamocortical
slices (n11 animals), which led to a retrograde staining of
thalamocortical fibers and cell bodies within the MGN (Fig.
2AC). To further identify cortical areas excited by stimulation of the MGN, we performed immunohistochemical
stainings of the pan-neuronal neurofilament (SMI-311; Fig.
2D). In the plane of the slices, the primary auditory cortex
was flanked by the ventral secondary auditory cortex (AuV)
and the association area of the temporal lobe (TeA, Fig.
2D). To control for possible unspecific effects of the voltage
sensitive dye, we performed whole cell patch-clamp recordings in thalamocortical projection neurons of the MGN and
neurons in A1. The recorded neurons (MGN, n5; A1, n7)
displayed resting parameters (resting membrane potential:

Fig. 2. Biocytin and immunohistochemical staining of angled horizontal brain slices. (A) Overview of an angled horizontal brain slice after
peroxidase staining of biocytin. The inset on the upper left shows the
entire slice, which was trimmed frontally and along the midline. The
medial geniculate nucleus (MGN), the lateral geniculate nucleus
(LGN), the nucleus reticularis thalami (NRT), the ventrobasal nuclear
complex (VB), and the primary auditory cortex (A1) are indicated.
Application of biocytin into the primary auditory cortex led to a retrograde staining of projection neurons in the MGN. Labeled neurons are

more clearly visible at a higher magnification, as shown in B (arrow


heads in A and B for orientation). Note the lower intensity labeling of
many MGN neurons, which is absent in the surrounding areas. Boxes
1 and 2, as well as fiber tracts transversing the white matter (asterisk)
below cortical layer 6 are shown at high magnification in C, as indicated. The right panel in C shows two MGN neurons at higher magnification. (D) Immunohistochemical staining of the pan-neuronal
marker for non-phosphorylated neurofilament (SMI311) in another
slice. SMI-311 staining allowed the identification of cortical areas. The
perirhinal cortex (Pir), the association area of the temporal lobe (TeA),
the primary auditory cortex (A1), the ventral secondary auditory cortex
(AuV), and the medial geniculate nucleus are indicated. The dotted
lines mark the borders of A1. Scale bars in A represent 2 mm (inset)
and 0.5 mm. In B scale bar indicates 100 m, while in C it denotes 50
m for the left three panels and 20 m for the right panel. Scale bar in
D indicates 1 mm.

T. Broicher et al. / Neuroscience 165 (2010) 371385

MGN: 65.34.7 mV; A1: 78.31.9 mV; capacitance:


MGN: 39.912.4 pF; A1: 56.08.5 pF; input resistance:
MGN: 16131 M; A1: 20454 M) and generated action
potentials (peak of action potential: MGN: 38.23.2 mV; A1:
39.61.7 mV; half-width: MGN: 2.960.45 ms, A1:
3.260.62 ms) as described previously (Tennigkeit et al.,
1998; Contreras, 2004).
Evoked activity in the auditory thalamocortical slice
Cortical activity evoked by stimulation of the auditory thalamus was recorded using field potential electrodes and
voltage sensitive dyes. The stimulation electrode was
placed in the rostromedial end of the MGN (Fig. 5A, B).
Cortical field potentials recorded in the granular layer
(layer 4) consisted of two negative early onset components, which were followed by later components (Fig. 3A,
1st component black arrow, 2nd component white arrow).
The latency of the first negative field potential deflection
was 1.80.1 ms, while the latency of the second component was 4.80.3 ms with an average amplitude of
0.30.05 mV (n23). On average, optical signals had a
slightly longer onset latency, with a minimal onset latency
per slice of 5.60.3 ms (n24). Optical signals consisted
of a single component or of an initial peak followed by a
plateau (Fig. 3B, left panels, 8B) and displayed a slow rise
time (onset to maximum 12.50.5 ms, n24).
Pharmacological approaches were used to identify the
origin of the recorded signals and to correlate optical signals with field potential recordings. Nominal Ca2 free
ACSF abolished all optical signals and field potential deflections, except for the initial negative field potential component (Fig. 3A, third panel from the left; 3D, area excited
by stimulus; n4). Addition of 0.5 M tetrodotoxin (TTX)
eliminated the remaining field potential deflection (Fig. 3A
right panel; n4). In a separate series of experiments,
addition of 10 M 6-cyano-7-nitroquinoxaline-2,3-dione
(CNQX), an antagonist of nonN-methyl-D-aspartate (NMDA)
glutamate receptors, to the ACSF abolished all optical signals
and all field potential deflections, except for the initial negative
field potential component (Fig. 3B, third panel from the left;
3E, area excited by stimulus; n4). This indicates that the
initial negative field potential deflection represents the afferent thalamocortical fiber volley or the antidromic activation of
corticothalamic fibers. Consistent with this conclusion, the
initial field potential component was not detected in our voltage sensitive dye recordings. The later field potential deflection is likely to correspond to a population spike embedded in
a field excitatory post-synaptic potential (fEPSP). Furthermore, the pharmacological data indicate that the optical signals are caused by postsynaptic activity.
For analyses of cortical activity the cortex was divided
into supragranular, granular, and infragranular layers (Fig.
4A), with supragranular layers showing significantly more
activity. High frequency stimulation evoked higher amplitudes and longer signal durations (see Fig. 5E). When
analyzing the averaged amplitudes of optical signals we
found that the magnitude of infragranular activity was age
dependent. When normalized to the averaged amplitudes
observed in the supragranular and granular layers, infra-

375

granular amplitudes decreased with increasing age of the


animals (Fig. 4B, normalized averaged amplitude: p 31
46: 0.560.08, n6; p 47 64: 0.420.03, n8; p 65 82:
0.310.03, n12; p 83100: 0.310.06, n6; P0.05 for
p 65 82 and p 83100 when compared to p 31 46; oneway ANOVA), while no difference was observed in the
granular or supragranular layers. In order to rule out any
errors due to developmental differences we here present
data from experiments using animals p 65 or older, unless
stated otherwise.
In contrast to observations made in somatosensory
thalamocortical slices (Laaris et al., 2000; Llinas et al.,
2002), where thalamic stimulation evoked maximal responses in the granular and infragranular layers (layers 5
and 6), the magnitude of evoked activity was graded in
auditory thalamocortical slices, as indicated by the distribution of optical signal amplitudes. Maximal amplitudes of
optical signals were recorded in the supragranular layers
(layers 1, 2, and 3) from where signal amplitudes progressively decreased with increasing distance from the pial
surface (Fig. 5D, photodiode array positions indicated in
Fig. 5A). Maximal amplitudes in the supragranular layers
were 1.180.12103 I/I (n24), maximal signals in the
granular layer had an amplitude of 0.70.08103 I/I
(n24), while only 10 of 24 slices displayed activity in
the infragranular layers. The largest signal amplitudes
in the infragranular layers were comparatively small
(0.30.07103 I/I, no response slices included with amplitude of zero; mean amplitude of responding slices was
0.60.07103 I/I; n10, Fig. 4A). The differences in signal amplitudes between cortical layers were significant (supra
vs. granular P0.01; supra vs. infra P0.001; granular vs.
infra P0.001). On average, the maximal amplitude recorded
in the infragranular layers had 567% of the maximal signal
amplitude of the entire slice and in no case has the largest
signal been recorded in the infragranular layers.
Spread of optical signals
To assess the spatial extent of optical signals, we constructed overlays of pseudocolor images of the fluorescence signals from successive recordings and bright field
images of the slice (Fig. 5A shows positions of photodiode
arrays; 5D-G shows color coded activity maps). In terms of
layer distribution, two major patterns of cortical activity
were observed: (i) in 10 of 24 slices activity was evoked in
all cortical layers, while (ii) in the remaining 14 slices
cortical activity was restricted to the granular and supragranular layers. Investigating the lateral spread of stimulus
evoked activity revealed that 10 of 24 slices generated
activity only in A1, while the remaining 14 slices displayed
stimulus evoked activity in cortical areas lateral to A1,
namely the ventral secondary auditory cortex (AuV, four
slices), the association area of the temporal lobe (TeA, one
slice) or both (nine slices). There was no obvious correlation between the presence of activity in the infragranular
layers and the lateral spread of activity.
In the following, slices were grouped according to the
lateral spread of evoked activity, into local response slices,
in which activity was detected only in A1, and extended

376

T. Broicher et al. / Neuroscience 165 (2010) 371385

Fig. 3. Effects of high frequency stimulation, removal of extracellular Ca2, CNQX, and TTX on evoked activity. (A) Voltage sensitive dye signals
(VSD, upper traces) and field potential recordings (Field, lower traces) under control conditions (single pulse stimulation, Ctrl), upon stimulation with
three pulses at 100 Hz (Ctrl HF), in nominal Ca2 free ACSF (0 Ca2), and in 0 Ca2 with 0.5 M TTX added to the bath solution (0 Ca2 TTX). In
field potential traces, black arrows denote the presumed afferent thalamocortical fiber volley, white arrows denote the first presumed postsynaptic
population spike. Note the absence of compound action potentials only in the presence of TTX. All recordings derive from the same slice. Arrows below
VSD signals indicate time of stimulus in A and B. (B) Same conventions as in A. Next to control responses to single and high frequency stimulation, the
responses in the presence of 10 M CNQX are shown for low and high frequency stimulation (HF). All recordings derive from the same slice. All VSD signals
shown were recorded in the supragranular layers of A1. Changes in the size of the excited area due to high frequency stimulation (HF) are shown in (C),
changes due to the removal of extracellular Ca2 are shown in (D), and the effects of 10 M CNQX on the size of the excited area are shown in (E). The
depicted points in C, D, and E correspond to time bins: left points 0 26 ms after stimulus, middle points 2751 ms after stimulus, right points 52102 ms
after stimulus. Filled symbols represent control conditions while open symbols represent alterations in stimulus or pharmacological manipulation. * P0.05.

T. Broicher et al. / Neuroscience 165 (2010) 371385

Fig. 4. (A) Maximal amplitudes per cortical layer and slice under control
conditions (Ctrl) and with high frequency stimulation (Ctrl HF). No responses in a given layer and slice were included with an amplitude of zero
(see text). (B) Age dependence of normalized averaged optical signal
amplitudes in the infragranular layers. Infragranular activity was measured as the averaged amplitude of the optical signal in the infragranular
layers normalized to the averaged amplitudes in granular and supragranular layers and compared between four age groups (see methods).
Differences between p 65 82 and p 83100 compared to the youngest
group were significant (one-way ANOVA, * P0.05).

response slices, in which the activity was detected in A1


and adjacent areas. In both local and extended response
slices, the earliest response to thalamic stimulation was
detected in A1. Minimal onset latencies were recorded in
the supragranular and/or granular layers, but not in the
infragranular layers. The minimal onset latency per slice
was identical between the two spatial response types
(5.60.2 ms in local response slices, 5.70.4 ms in extended response slices).
For subsequent analysis, the cortex was subdivided
horizontally into layers (infragranular, granular, supragranular) and vertically into areas (AuV, A1, TeA; Fig. 6). In
local response slices, the average minimal onset latency
was located in the granular layer (6.50.6 ms), supragranular activity commenced about a millisecond later
(7.30.8 ms), while signal onset in the infragranular layers
was delayed by 3 4 ms (10.22 ms, n10; Fig. 6B).
The latency difference was significant for the granular
versus the infragranular layers (P0.05). The largest
signals were recorded in the supragranular layers
(1.10.4103 I/I), followed by signal amplitudes observed in the granular layer (0.70.08103 I/I), and in

377

the infragranular layers (0.50.09103 I/I, responding


slices only, n10; Fig. 6D). The differences in amplitude
were significant for the supragranular versus the infragranular layers (P0.05).
In contrast to the above findings, the lowest average
minimal onset latency in extended response slices was
located in the supragranular layers of A1 (6.20.5 ms; Fig.
6A), followed by signal onset in the granular layer of A1
and AuV (7.80.8 and 7.80.9 ms, respectively) and after
these by the supragranular layers of AuV, the infragranular
layers of A1, and the granular layer of the TeA (8.30.8
ms, 8.71 ms, 8.80.8 ms, respectively). The slowest
signal onsets were observed in the infragranular layers of
AuV, and the supra- and infragranular layers of the TeA
(9.91.6 ms, 9.61.4 ms, 11.21.9 ms, respectively;
n14; Fig. 6A). Onset latency differences between different areas were significant for the supragranular layers of
AuV versus A1 (P0.05) and TeA versus A1 (P0.05),
while within area differences were significant for the infragranular versus the supragranular layers of A1 (P0.05).
The largest signal amplitudes in extended response slices
were recorded in the supragranular layers of A1 and the
TeA (1.340.15103 I/I, 1.310.16103 I/I, respectively; Fig. 6C). Signals observed in the granular layer of A1
and the supragranular layer of the AuV were of intermediate
amplitudes (0.830.09103 I/I, 0.820.08103 I/I, respectively), while the smallest signal amplitudes were found
in the granular layers of AuV and the TeA, and the infragranular layers of all areas (range, 0.78 0.49 103 I/I; n14; Fig.
6C). Amplitude differences were significant for the supragranular layers of the AuV versus A1 and the TeA (both
P0.01), while differences between the granular layer of the
same areas were close to significant (both P0.08). Differences in signal amplitude between layers within an area were
significant for the infragranular layers versus the supragranular layers in all areas (P0.05P0.01) and the granular
versus the supragranular layers (P0.05P0.01), except
for the AuV, where differences just failed to reach significance
(P0.06).
Effect of high frequency stimulation
To investigate whether a temporal input pattern resembling
thalamic burst mode transmission (Sherman and Guillery,
2006; Broicher et al., 2008) alters the cortical response
profile, we used a stimulation protocol composed of three
stimuli repeated at a frequency of 100 Hz (Fig. 3A, B, high
frequency stimulation is indicated by HF; 5E, G). Compared to single stimuli, this high frequency stimulation
increased the area, as well as the amplitude of optical
signals recorded in the cortex. On average, maximal amplitudes in the supragranular and granular layers were
increased by 1525% by changing the stimulation from a
single shock to three pulses at 100 Hz (n14; supragranular layers P0.05; Fig. 4A). In the infragranular layers
amplitude changes were minor, but high stimulation frequency increased the number of slices which generated
activity in the infragranular layers from 5 to 7 of a total of 14
slices stimulated with three pulses at 100 Hz. Similarly, the
area excited by thalamic stimulation increased with high

T. Broicher et al. / Neuroscience 165 (2010) 371385

im
St

Stim

Scaling

Ctrl

Ctrl HF

AP5

5 10-4
I/I

378

64 ms

AP5 HF

3.8 ms

3.8 ms

3.8 ms

3.8 ms

7.6

7.6

7.6

7.6

10.2

10.2

10.2

10.2

12.7

12.7

12.7

12.7

15.3

15.3

15.3

15.3

17.8

17.8

17.8

17.8

20.4

20.4

20.4

20.4

31.9

31.9

31.9

31.9

42

42

42

42

52.2

52.2

52.2

52.2

62.4

62.4

62.4

62.4

72.6

72.6

72.6

72.6

Fig. 5. Pseudocolor maps of evoked activity. (A) Overview of the slice from the top. Slices were trimmed frontally and along the midline. The
stimulation electrode is indicated (Stim) and was placed in the medial geniculate nucleus. The electrode to the left was used for fixation. The white
hexagons indicate the positions of the photodiode array (10 objective). (B) Composite overview picture of the slice from below. The position of the
stimulation electrode is visible as a black dot and is indicated (Stim). (C) Voltage sensitive dye trace used for scaling. Period used for scaling of
the pseudocolor images is indicated by the horizontal line below the VSD trace. Trace used for scaling was recorded in the supragranular layers of
the primary auditory cortex under high frequency stimulation. (D) Enlarged sections containing the three photodiode array positions, displaying the
evoked activity under control conditions using single pulse stimulation. Time after stimulus is indicated on the lower right of each frame. (E) Evoked
activity using high frequency stimulation (three pulses at 100 Hz). (F) Evoked activity using single pulse stimulation in the presence of 50 M AP5.
(G) Evoked activity using high frequency stimulation under the influence of 50 M AP5. Scale bars in A and B, as well as in D through G represent
1 mm.

T. Broicher et al. / Neuroscience 165 (2010) 371385

379

Latency
A

AuV

A1

TeA

I
G
S

6-7 ms
7-8 ms
8-9 ms

AuV

A1

TeA

I
G
S

9-10 ms
>10 ms
No signal

Amplitude
C

AuV

A1

TeA

I
G
S

> 12 10-4 I/I


10-12 10-4 I/I

AuV

A1

I
G
S

TeA

8-10 10-4 I/I


6-8 10-4 I/I
4-6 10-4 I/I
No signal

Fig. 6. Summary of minimal onset latencies (A and B) and maximal amplitudes (C and D) per slice using single pulse stimulation. Slices were grouped
according to lateral spread of activity into extended response slices (A and C) and local response (B and D) slices. The cortex was subdivided into areas
(ventral secondary auditory cortex, AuV; primary auditory cortex, A1; association area of the temporal lobe, TeA) and layers (infragranular, I; granular, G,
shaded line; supragranular, S). Onset latencies and amplitudes were binned, color coded (scale on the right), and plotted on the corresponding regions of
the slice.

stimulation frequency. For the analysis of excited area,


three temporal bins were used (time after stimulus: 0 25.5
ms, bin 1; 26.8 51 ms, bin 2; 52.2102 ms, bin 3). The
area excited by a single stimulus to the thalamus was
0.710.16 mm2, 0.190.07 mm2, and 0.220.07 mm2 for
bins 13, respectively (n14). Using three stimuli at a
frequency of 100 Hz increased the excited area to
0.990.2 mm2, 0.240.05 mm2, and 0.430.12 mm2 for
bins 13, respectively (bin 1 and 3, P0.05; n14; Fig.
3C, HF indicates high frequency stimulation). Furthermore,
the duration of cortical activity was increased by high frequency input. The average maximal duration of optical
signals per slice using a single stimulus was 16933 ms,
15534 ms, and 10460 ms in the supragranular, granular, and infragranular layers, respectively (n14, data not
shown). Changing the stimulus to three pulses at 100 Hz

increased the average duration of optical signals to


462103 ms, 31387 ms, and to 14951 ms in the
supragranular, granular, and infragranular layers, respectively (supragranular P0.01; n14). Importantly, and in
contrast to observations in the somatosensory barrel cortex (Laaris et al., 2000; Beierlein et al., 2002), the overall
spatiotemporal pattern of evoked activity did not change
markedly as a result of high frequency stimulation (Fig. 5D,
E, high frequency stimulation is indicated by HF). The most
notable alteration in this respect was the elongated response duration.
Effect of NMDA receptor blockade
To assess the involvement of NMDA receptors in evoked
cortical activity in the auditory thalamocortical preparation,

380

T. Broicher et al. / Neuroscience 165 (2010) 371385

we used the NMDA receptor antagonist AP5. Using single


pulse stimulation, bath applied AP5 (50 M) slightly reduced the maximal signal amplitudes, and the area excited
by thalamic stimulation (Figs. 5F and 7A, D, E). In the
supragranular layers signal amplitudes were reduced by
16%, while we observed only a marginal decrease in signal
amplitude in the granular layer (5%), and a decrease of
11% in the infragranular layers (n6, Fig. 7D). Application
of AP5 reduced the area excited by a single stimulation of
the thalamus by 32%, 25%, and 40% for time bins 13,
respectively (n6, Fig. 7E). Likewise, the maximal duration of optical signals was slightly reduced in the supragranular and infragranular layers after application of AP5.
Here, signal duration was reduced from 28065 to
19155 ms and from 11662 to 538 ms, respectively
(n6, data not shown). In the granular layer we observed
a marginal increase in signal duration (from 16852 to
17680 ms, n6). In summary, application of 50 M AP5
slightly dampened evoked cortical activity without changing the overall spatiotemporal activity pattern (Fig. 5F).
As high frequency input is known to activate NMDA
receptors, we investigated the effects of 50 M AP5 on the
cortical responses to high frequency stimulation of the
thalamus (three pulses at 100 Hz). Under these conditions,
AP5 had effects similar to those observed upon single
pulse stimulation of the thalamus (Figs. 5G and 7B, D, G;
HF indicates high frequency stimulation). Maximal amplitudes were reduced by 25% in the supragranular layers
and by 18% in the granular layer, while there was a slight
increase in amplitude in the infragranular layers (10%,
n3, Fig. 7D). Furthermore, the excited area was reduced
by 31%, 52%, and 56% for time bins 13, respectively (Fig.
7G). The duration of optical responses was reduced by
78% in the supragranular layers and by 26% in the granular layer, while we observed a slight increase of signal
duration in the infragranular layers of 9% (n3, data not
shown). These data indicate that there is no major recruitment of NMDA receptors by high frequency input from the
thalamus in the auditory cortex. In three slices, adding 10
M CNQX to the recording solution already containing 50
M AP5 led to the disappearance of all optical responses
(n3, Fig. 7A, B, F).
Effect of GABAA receptor blockade
To visualize the involvement of inhibitory, GABAA mediated neurotransmission, we investigated the effects of 10
M bicuculline (a GABAA receptor antagonist) on evoked
cortical activity. Bicuculline application greatly increased
the amplitude (P0.05P0.001; Fig. 7C, H), the duration, and the area of cortical responses (P0.05
P0.001; Fig. 7I; n3). The evoked activity spread vertically, encompassing all cortical layers, and horizontally in
both rostral and caudal directions (Fig. 8A, C). Interestingly, infragranular signals were still of lower amplitude
than granular or supragranular responses, but the interlaminar difference was reduced (Fig. 7H). Furthermore,
supragranular activity seemed to lead the excitation
spread. These experiments indicate that GABAA receptor-

mediated inhibition limits the spatiotemporal spread of activity in auditory thalamocortical slices.

DISCUSSION
The present study employed the auditory thalamocortical
brain slice preparation in adult mice (Rose and Metherate,
2001, 2005; Cruikshank et al., 2002; Kaur et al., 2005;
Kotak et al., 2008; Lee and Sherman, 2008). We used this
preparation to study cortical activity evoked by activation of
thalamic afferents via stimulation of the MGN. Optical imaging using voltage sensitive dyes in combination with
conventional field potential recordings allowed us to study
the spatiotemporal pattern of evoked activity at high temporal and spatial resolution. To the best of our knowledge,
this is the first study to report the spatiotemporal characteristics of thalamically evoked activity in the auditory cortex of adult mice.
The major findings of the present study are (1) a prominent supragranular and granular excitation with pronounced horizontal spread, (2) a dramatic developmental
decrease of activity in the infragranular layers, (3) a predominance of non-NMDA receptor mediated activity, (4)
comparatively small effects of high frequency stimulation
and NMDA receptor blockade, (5) a pronounced influence
of GABAA receptor-mediated inhibition on the spatiotemporal spread of activity.
The basic spatiotemporal profile of evoked activity in
slices displaying activity only in A1 (local response slices)
matches previous descriptions (Kubota et al., 1997, 1999;
Cruikshank et al., 2002; Kaur et al., 2005), while the activity in slices with activity spreading beyond the boundaries of A1 into AuV and TeA (extended response slices)
differed from previous reports in several aspects. In this
subset of samples, minimal onset latencies were located in
the supragranular layers of A1. The activity in areas lateral
to A1 (AuV and TeA) displayed shortest onset latencies in
the granular layer, while the largest signal amplitudes were
found in the supragranular layers. Notably, the evoked
activity in the supragranular layers of the TeA was of
comparable amplitude to the activity in the supragranular
layers of A1, while displaying significantly longer onset
latencies. The layer distribution of evoked activity, as well
as the influence of NMDA receptors and high frequency
stimulation, differed from observations made in somatosensory thalamocortical slices after stimulation of the ventrobasal nuclear complex (Laaris et al., 2000; Beierlein et
al., 2002; Llinas et al., 2002), indicating differences in
cortical processing of thalamic inputs in rodent A1 as compared to S1.
Technical considerations
When interpreting the results of the present study, some
technical issues need to be addressed. The optical signals
described here reflect average membrane potential
changes in columns of tissue extending 100 m into the
slice (Salzberg et al., 1977; Cohen and Salzberg, 1978;
Cohen et al., 1978; Wu and Cohen, 1993; Grinvald and
Hildesheim, 2004; Baker et al., 2005). Furthermore, optical

T. Broicher et al. / Neuroscience 165 (2010) 371385

381

Fig. 7. Effects of AP5, CNQX, and bicuculline on evoked activity. (A) Voltage sensitive dye signals (VSD, upper traces) and field potential recordings
(Field, lower traces) under control conditions (Ctrl), in the presence of 50 M AP5 (AP5), and in the presence of 50 M AP5 in combination with 10
M CNQX (AP5 CNQX). All recordings derive from the same slice. (B) VSD and field potential recording from the same slice shown in A using high
frequency stimulation (three pulses at 100 Hz, HF), under control conditions (Ctrl HF), in the presence of 50 M AP5 (AP5 HF), as well as of 50 M
AP5 in combination with 10 M CNQX (AP5 CNQX HF). Scale bars apply to A and B. Arrows below VSD signal denote the time of stimulation. (C)
VSD and field potential recordings under control conditions (Ctrl, upper panels) and in the presence of 10 M bicuculline (Bicu, lower panels). VSD
signals in A-C derived from the supragranular layers of A1. (D) Maximal amplitudes under control conditions (Ctrl) and after application of 50 M AP5
(AP5) under single pulse and high frequency stimulation (HF) in the infragranular layers (I), the granular layer (G), and the supragranular layers (S).
(E, F, and G) show changes in the excited area due to application of 50 M AP5 (E), 50 M AP5 in combination with 10 M CNQX (F), and 50 M
AP5 under high frequency stimulation (G). (H) Maximal amplitudes before (Ctrl) and after application of 10 M bicuculline (Bicu) for the infragranular
layers (I), the granular layer (G), and the supragranular layers (S). (I) Changes in excited area induced by 10 M bicuculline (Bicu). The depicted points
in E, F, G, and I correspond to time bins: left point 0 26 ms after stimulus, middle point 2751 ms after stimulus, right point 52102 ms after stimulus.
Closed symbols represent control conditions, while open symbols represent alterations in stimulation or pharmacological manipulation.

T. Broicher et al. / Neuroscience 165 (2010) 371385

im
t
S

50 ms
scaling interval

7.6 ms

17.8 ms

10.2 ms

22.9 ms

12.7 ms

26.8 ms

5 10-4 I/I

382

Fig. 8. Pseudocolor maps of evoked activity in the presence of 10 M bicuculline. (A) Top view of the slice. The stimulation electrode located in the
medial geniculate nucleus is indicated (Stim) and the positions of the photodiode array are displayed as shaded hexagons (20 objective). (B) Voltage
sensitive dye trace used to scale pseudocolor images. Period used for scaling is indicated by the horizontal line below the VSD trace. (C) Enlarged
sections of the slice containing all photodiode array positions, showing the evoked activity at the indicated time points after stimulation. Scale bar in
A and C indicates 1 mm.

signal amplitudes are directly proportional to the magnitude of voltage deflections as well as the amount of membrane experiencing a potential change, resulting in a
strong influence of dendritic potential changes (Cinelli and
Salzberg, 1990, 1992). This implies that the strongest
voltage sensitive dye signals may be located in a different
layer than the somata of the excited neurons. Furthermore,
the density of neuronal membrane (somata and dendrites)
will differ between different layers of cortex, which will
affect the amplitudes of voltage sensitive dye signals. In
particular, neuronal density is likely to be higher in the
supragranular layers in comparison to the infragranular
layers (Weedman and Ryugo, 1996), which could lead to
an overestimation of signal amplitudes in the supragranular layers compared to the infragranular layers. In addition,
dye concentrations may vary between different layers of
cortex, and non-neuronal signals may add to neuronal
signals, distorting the results. Differences in dye concentrations are unlikely to be of influence to our data, as all
signals are given as fluorescence changes relative to the
resting light intensity. Non-neuronal voltage changes have

been described to display much slower time courses (on


the order of seconds vs. ms) than neuronal signals (Konnerth et al., 1987; Garaschuk et al., 2006), which leads us
to conclude that the signals described here originate from
neuronal rather than non-neuronal activity, although a contribution of fast signals originating from non-neuronal
sources cannot be completely ruled out (Konnerth et al.,
1987).
Our voltage sensitive dye signals most likely correspond to EPSPs within the sampled volume of tissue.
This notion is supported by the finding that putative
thalamocortical compound action potentials are reflected in our field potential recordings, but not in our
voltage sensitive dye signals. This may be related to the
brief duration of single action potentials and to the
strong myelination of thalamocortical fibers (Salzberg et
al., 1977; Cohen and Salzberg, 1978; Cohen et al.,
1978; Grinvald and Hildesheim, 2004; Baker et al.,
2005). Furthermore, our experiments in nominal Ca2
free solution and TTX indicate that the evoked activity in
the auditory thalamocortical preparation is caused pri-

T. Broicher et al. / Neuroscience 165 (2010) 371385

marily by orthodromic activation of thalamocortical fibers, as was previously demonstrated by Rose and
Metherate (2001).
The pseudocolor maps constructed to study the spatiotemporal distribution of evoked cortical activity were assembled from successive recordings, in order to cover
larger areas of the slice by repositioning the photodiode
array. In most cases evoked activity recorded at different
points in time could be combined into a single map without
abrupt changes at the photodiode array borders. Additionally, all optical signals analyzed in this study are averages
from three successive recordings separated by 30 s. This
indicates that the activity described here represents stereotypical behavior of our preparations, which can be
evoked repetitively.
Spatiotemporal distribution of evoked activity in the
auditory cortex
The most robust finding of the present study was a graded
distribution of optical signal amplitudes. The largest amplitudes were located in the supragranular layers, from where
amplitudes progressively decreased with increasing distance from the pial surface. Furthermore, we detected a
developmental decrease of activity in the infragranular
layers. At age p 65 and above, very little or no activity in the
infragranular layers was found in most of our preparations.
Taking the results of previous studies into account (Rose
and Metherate, 2001; Cruikshank et al., 2002; Kaur et al.,
2005), infragranular activity due to thalamic stimulation
seems to decrease during development. While the previous studies used younger animals (p 1319) and found
moderate levels of infragranular activity, our study shows a
very low level of activity from p 65 onwards, and a graded
decrease between p 31 and p 64 (Fig. 4B).
The earliest signal onsets were located in the supragranular layers, in the granular layer, or both, while infragranular activity had longer onset latencies. The amplitude
and latency results match the anatomical descriptions of
the terminations of thalamocortical synapses coming from
the ventral division of the medial geniculate body (Linden
and Schreiner, 2003; Winer and Lee, 2007).
While the low amplitude activity in the infragranular
layers was a novel and very robust finding in our experiments it should be interpreted with caution, as a lower
neuronal density in the infragranular layers might confound
this result. In any case, signal onsets are unlikely to be
influenced by this. We consistently observed the longest
onset latencies in the infragranular layers. This hints at an
intracortical as opposed to a thalamocortical activation
mechanism and points to a more efficient interconnection
between the granular and supragranular layers in comparison to the interconnection between the infragranular and
the upper layers in A1, AuV, and TeA. Furthermore, the
spread of activity was independent of the presence or
absence of infragranular excitation, which indicates that, in
the auditory cortex, lateral spread of activity can be
achieved by the supragranular and granular layers alone.
An isolated spread of activity in the infragranular layers
was not observed in our experiments. In slices in which the

383

evoked activity extended beyond the borders of A1, average minimal onset latencies of the entire slice were located
in the supragranular layers of A1. In the AuV and the TeA,
average minimal onset latencies were located in the granular layer, while largest signal amplitudes were found in the
supragranular layers. To the best of our knowledge, this
peculiar spatiotemporal activity pattern has not been observed before. It is consistent with a highly focal localization of thalamocortical projections in the granular and/or
supragranular layers of A1 as well as with a more distributed localization of thalamic input in the supragranular
layers of A1 and the granular layers of A1, AuV, and
possibly the TeA. The former scenario implies a stronger
horizontal spread of activity, while the latter implies a
larger area of thalamic input with a predominant upward spread of activity to the supragranular layers.
Interestingly, the average overall shortest onset latency
was located in different layers depending on the extent
of the lateral spread of activity (extended response
slices vs. local response slices). Possibly, the differences in lateral spread of activity were caused by differences in the localization of thalamic input and the integrity of intracortical circuitry, which are likely to vary with
small differences in the individual slicing procedures and
between different animals.
Effects of glutamate and GABAA receptor blockade
and high frequency stimulation
Our pharmacological experiments indicated a strong influence of non-NMDA glutamate receptors along with a modest contribution of NMDA receptors. These findings are in
agreement with a previous study in rats (Kubota et al.,
1997). Interestingly, an increase of stimulation frequency
neither markedly changed the spatiotemporal activity pattern, nor did it increase the relative effects of AP5, indicating that under our recording conditions there was no
marked increase in NMDA receptor recruitment by high
frequency input. Notably, this result differs from NMDA
receptor dependent high frequency induced long-lasting
activity observed in A1 of slices of juvenile mice (Rose and
Metherate, 2005). This discrepancy might be explained by
a difference in the number of pulses and the stimulation
frequency (mostly 10 pulses at 40 Hz by Rose and Metherate (2005), vs. three pulses at 100 Hz in the present
study) and/or the age of the animals used in the experiments. Blockade of GABAA mediated inhibition substantially restructured evoked activity. In the presence of bicuculline, stimulation of the thalamus generated high amplitude activity, which spread horizontally in both directions.
Although signals in the infragranular layers still displayed
the lowest amplitude, the inter-laminar differences were
smaller in the absence of GABAA receptor-mediated inhibition. This might be explained by an unopposed vertical
spread of activity. Furthermore, supragranular activity led
the spreading excitation, pointing at a more efficient horizontal interconnection within the supragranular layers
compared to the deeper layers.

384

T. Broicher et al. / Neuroscience 165 (2010) 371385

Comparison to previous studies and other cortical


areas
Our data are in good agreement with two previous studies
on evoked activity in slices of the auditory cortex using
voltage sensitive dyes (Kubota et al., 1997, 1999). Although Kubota and colleagues used slices from rats and
usually stimulated at the border between the white matter
and layer 6, the basic spatiotemporal activity pattern described by them was very similar to what we found here.
Interestingly, Kubota and colleagues observed supragranular activity irrespective of electrode position (white
matter/layer 6 or supragranular) but strong infragranular
activity only when the infragranular layers were stimulated
directly.
It should be noted that in vivo recordings in rats using
non-characteristic frequency stimulation resulted in short
latency activation of the infragranular layers (Kaur et al.,
2005), while stimulation using the characteristic frequency
evoked activity with short latency current sinks in layers 3
and 4 in the same study as well as in studies in awake
monkeys (Fishman et al., 2000a,b; Kaur et al., 2005). A
possible explanation for this difference is a non-lemniscal
projection to the infragranular layers in vivo, which has
been severed in our brain slice preparations. This interpretation is supported by in vitro experiments by Kaur and
colleagues (2005) using the auditory thalamocortical preparation in juvenile mice, where activity in areas lateral to
the initially activated region (off focus) was found to be
predominantly located in the supragranular rather than the
infragranular layers.
Furthermore, the weak activation of the infragranular
layers in A1 is in contrast to infragranular activity observed using voltage sensitive dyes in the somatosensory barrel cortex after stimulation of the ventrobasal
nucleus of the thalamus (Laaris et al., 2000; Llinas et al.,
2002), in the visual cortex after stimulation of the border
between the white matter and layer 6 (Albowitz and
Kuhnt, 1993b; Nelson and Katz, 1995; Contreras and
Llinas, 2001), and in the insular cortex after layer 4
stimulation (Sato et al., 2008).
A pronounced NMDA receptor dependent facilitation of
evoked cortical activity upon high frequency stimulation
has been described in the somatosensory cortex (Laaris et
al., 2000; Beierlein et al., 2002). Even though we did
observe a facilitation of evoked cortical activity after stimulation with three pulses at 100 Hz in the auditory cortex as
opposed to four pulses at 100 Hz by Laaris and colleagues
(2000), the magnitude of the facilitation was weaker than
that described in the somatosensory system. In line with
this, we did not observe a relative increase of AP5 effects
on activity evoked by high frequency stimulation as compared to activity evoked by single pulse stimulation, pointing at a different functional role of NMDA receptors in A1
as compared to S1.
Future directions
We adapted the auditory thalamocortical slice to obtain a
preparation which would allow us to study the functional

impact of conductance velocity changes induced by demyelination on the thalamocortical system. The goal of the
present study was to provide solid ground for future investigations by describing the basic patterns of evoked cortical
activity in adult mice. A combination of voltage sensitive
dye imaging and field potential recordings, as employed
here, will allow investigating changes in evoked activity in
both temporal and spatial dimensions. We believe that the
auditory thalamocortical slice is well suited for such investigations, as the fiber tract from MGN to A1 is sufficiently
long to detect changes in onset latency and it allows the
effects of intracortical and subcortical demyelination to be
addressed separately.
AcknowledgmentsThe Authors would like to thank Birgit Herrenpoth for excellent technical assistance. This work was supported by Interdisziplinres Zentrum fr Klinische Forschung (Bud
3/005/07), Deutsche Forschungsgesellschaft (BU1014/8-1/9-1;
PA336/17-1), and the Max-Planck-Gesellschaft (Max-PlanckResearch-Award to HCP).

REFERENCES
Agmon A, Connors BW (1991) Thalamocortical responses of mouse
somatosensory (barrel) cortex in vitro. Neuroscience 41:365379.
Albowitz B, Kuhnt U (1993a) Evoked changes of membrane potential
in guinea pig sensory neocortical slices: an analysis with voltagesensitive dyes and a fast optical recording method. Exp Brain Res
93:213225.
Albowitz B, Kuhnt U (1993b) The contribution of intracortical connections to horizontal spread of activity in the neocortex as revealed by
voltage sensitive dyes and a fast optical recording method. Eur
J Neurosci 5:1349 1359.
Atzori M, Lei S, Evans DI, Kanold PO, Phillips-Tansey E, McIntyre O,
McBain CJ (2001) Differential synaptic processing separates stationary from transient inputs to the auditory cortex. Nat Neurosci
4:1230 1237.
Baker BJ, Kosmidis EK, Vucinic D, Falk CX, Cohen LB, Djurisic M,
Zecevic D (2005) Imaging brain activity with voltage- and calciumsensitive dyes. Cell Mol Neurobiol 25:245282.
Barbour DL, Callaway EM (2008) Excitatory local connections of superficial neurons in rat auditory cortex. J Neurosci 28:11174
11185.
Beierlein M, Fall CP, Rinzel J, Yuste R (2002) Thalamocortical bursts
trigger recurrent activity in neocortical networks: layer 4 as a
frequency-dependent gate. J Neurosci 22:98859894.
Bidmon HJ, Wu J, Godecke A, Schleicher A, Mayer B, Zilles K (1997)
Nitric oxide synthase-expressing neurons are area-specifically distributed within the cerebral cortex of the rat. Neuroscience 81:
321330.
Broicher T, Kanyshkova T, Meuth P, Pape HC, Budde T (2008) Correlation of T-channel coding gene expression, IT, and the low
threshold Ca2 spike in the thalamus of a rat model of absence
epilepsy. Mol Cell Neurosci 39:384 399.
Bruckner G, Seeger G, Brauer K, Hartig W, Kacza J, Bigl V (1994)
Cortical areas are revealed by distribution patterns of proteoglycan
components and parvalbumin in the Mongolian gerbil and rat. Brain
Res 658:67 86.
Celio MR, Blumcke I (1994) Perineuronal netsa specialized form of
extracellular matrix in the adult nervous system. Brain Res Brain
Res Rev 19:128 145.
Cinelli AR, Salzberg BM (1990) Multiple site optical recording of transmembrane voltage (MSORTV), single-unit recordings, and evoked
field potentials from the olfactory bulb of skate (Raja erinacea).
J Neurophysiol 64:17671790.

T. Broicher et al. / Neuroscience 165 (2010) 371385


Cinelli AR, Salzberg BM (1992) Dendritic origin of late events in optical
recordings from salamander olfactory bulb. J Neurophysiol 68:
786 806.
Cohen LB, Salzberg BM (1978) Optical measurement of membrane
potential. Rev Physiol Biochem Pharmacol 83:35 88.
Cohen LB, Salzberg BM, Grinvald A (1978) Optical methods for monitoring neuron activity. Annu Rev Neurosci 1:171182.
Contreras D (2004) Electrophysiological classes of neocortical neurons. Neural Netw 17:633 646.
Contreras D, Llinas R (2001) Voltage-sensitive dye imaging of neocortical spatiotemporal dynamics to afferent activation frequency.
J Neurosci 21:94039413.
Cruikshank SJ, Rose HJ, Metherate R (2002) Auditory thalamocortical
synaptic transmission in vitro. J Neurophysiol 87:361384.
Fishman YI, Reser DH, Arezzo JC, Steinschneider M (2000a) Complex tone processing in primary auditory cortex of the awake
monkey. II. Pitch versus critical band representation. J Acoust Soc
Am 108:247262.
Fishman YI, Reser DH, Arezzo JC, Steinschneider M (2000b) Complex tone processing in primary auditory cortex of the awake
monkey. I. Neural ensemble correlates of roughness. J Acoust Soc
Am 108:235246.
Garaschuk O, Milos RI, Grienberger C, Marandi N, Adelsberger H,
Konnerth A (2006) Optical monitoring of brain function in vivo: from
neurons to networks. Pflugers Arch 453:385396.
Grinvald A, Hildesheim R (2004) VSDI: a new era in functional imaging
of cortical dynamics. Nat Rev Neurosci 5:874 885.
Kanyshkova T, Pawlowski M, Meuth P, Dub C, Bender RA, Brewster
AL, Baumann A, Baram TZ, Pape H-C, Budde T (2009) Postnatal
expression pattern of HCN channel isoforms in thalamic neurons:
relationship to maturation of thalamocortical oscillations. J Neurosci 29(27):8847 8857.
Kaur S, Rose HJ, Lazar R, Liang K, Metherate R (2005) Spectral
integration in primary auditory cortex: laminar processing of afferent input, in vivo and in vitro. Neuroscience 134:10331045.
Kawai H, Lazar R, Metherate R (2007) Nicotinic control of axon excitability regulates thalamocortical transmission. Nat Neurosci 10:
1168 1175.
Khling R, Hohling JM, Straub H, Kuhlmann D, Kuhnt U, Tuxhorn I,
Ebner A, Wolf P, Pannek HW, Gorji A, Speckmann EJ (2000)
Optical monitoring of neuronal activity during spontaneous sharp
waves in chronically epileptic human neocortical tissue. J Neurophysiol 84:21612165.
Khling R, Reinel J, Vahrenhold J, Hinrichs K, Speckmann EJ (2002)
Spatio-temporal patterns of neuronal activity: analysis of optical
imaging data using geometric shape matching. J Neurosci Methods 114:1723.
Konnerth A, Obaid AL, Salzberg BM (1987) Optical recording of electrical activity from parallel fibres and other cell types in skate
cerebellar slices in vitro. J Physiol 393:681702.
Kotak VC, Takesian AE, Sanes DH (2008) Hearing loss prevents the
maturation of GABAergic transmission in the auditory cortex.
Cereb Cortex 18:2098 2108.
Kubota M, Nasu M, Taniguchi I (1999) Layer-specific horizontal propagation of excitation in the auditory cortex. Neuroreport 10:
28652867.
Kubota M, Sugimoto S, Horikawa J, Nasu M, Taniguchi I (1997)
Optical imaging of dynamic horizontal spread of excitation in rat
auditory cortex slices. Neurosci Lett 237:77 80.

385

Laaris N, Carlson GC, Keller A (2000) Thalamic-evoked synaptic


interactions in barrel cortex revealed by optical imaging. J Neurosci
20:1529 1537.
Lee CC, Sherman SM (2008) Synaptic properties of thalamic and
intracortical inputs to layer 4 of the first- and higher-order cortical
areas in the auditory and somatosensory systems. J Neurophysiol
100:317326.
Linden JF, Schreiner CE (2003) Columnar transformations in auditory
cortex? A comparison to visual and somatosensory cortices. Cereb
Cortex 13:83 89.
Llinas RR, Leznik E, Urbano FJ (2002) Temporal binding via cortical
coincidence detection of specific and nonspecific thalamocortical
inputs: a voltage-dependent dye-imaging study in mouse brain
slices. Proc Natl Acad Sci U S A 99:449 454.
MacLean JN, Fenstermaker V, Watson BO, Yuste R (2006) A visual
thalamocortical slice. Nat Methods 3:129 134.
Nelson DA, Katz LC (1995) Emergence of functional circuits in ferret
visual cortex visualized by optical imaging. Neuron 15:2334.
Rose HJ, Metherate R (2001) Thalamic stimulation largely elicits
orthodromic, rather than antidromic, cortical activation in an auditory thalamocortical slice. Neuroscience 106:331340.
Rose HJ, Metherate R (2005) Auditory thalamocortical transmission is
reliable and temporally precise. J Neurophysiol 94:2019 2030.
Salzberg BM, Grinvald A, Cohen LB, Davila HV, Ross WN (1977)
Optical recording of neuronal activity in an invertebrate central
nervous system: simultaneous monitoring of several neurons.
J Neurophysiol 40:12811291.
Sato H, Shimanuki Y, Saito M, Toyoda H, Nokubi T, Maeda Y,
Yamamoto T, Kang Y (2008) Differential columnar processing in
local circuits of barrel and insular cortices. J Neurosci 28:3076
3089.
Sherman SM, Guillery RW (2006) Exploring the thalamus and its role
in cortical function. Cambridge, MA: MIT Press.
Smith PH Populin LC (2001) Fundamental differences between the
thalamocortical recipient layers of the cat auditory and visual cortices. J Comp Neurol 436:508 519.
Staiger JF, Kotter R, Zilles K, Luhmann HJ (1999) Connectivity in the
somatosensory cortex of the adolescent rat: an in vitro biocytin
study. Anat Embryol (Berl) 199:357365.
Straub H, Kuhnt U, Hohling JM, Kohling R, Gorji A, Kuhlmann D,
Tuxhorn I, Ebner A, Wolf P, Pannek HW, Lahl R, Speckmann EJ
(2003) Stimulus-induced patterns of bioelectric activity in human
neocortical tissue recorded by a voltage sensitive dye. Neuroscience 121:587 604.
Tennigkeit F, Schwarz DW, Puil E (1998) Modulation of bursts and
high-threshold calcium spikes in neurons of rat auditory thalamus.
Neuroscience 83:10631073.
Weedman DL, Ryugo DK (1996) Pyramidal cells in primary auditory
cortex project to cochlear nucleus in rat. Brain Res 706:97102.
Winer JA, Lee CC (2007) The distributed auditory cortex. Hear Res
229:313.
Winer JA, Miller LM, Lee CC, Schreiner CE (2005) Auditory thalamocortical transformation: structure and function. Trends Neurosci
28:255263.
Wu JY, Cohen LB (1993) Fast multisite optical measurement of membrane potential. In: Fluorescent and luminescent probes for biological activity (Mason WT, ed), pp 389 404. London: Academic.
Yuste R, Tank DW, Kleinfeld D (1997) Functional study of the rat
cortical microcircuitry with voltage-sensitive dye imaging of neocortical slices. Cereb Cortex 7:546 558.

(Accepted 13 October 2009)


(Available online 17 October 2009)

Pflugers Arch - Eur J Physiol (2008) 456:10611073


DOI 10.1007/s00424-008-0482-9

CELLULAR NEUROPHYSIOLOGY

Reciprocal modulation of Ih and ITASK in thalamocortical


relay neurons by halothane
Thomas Budde & Philippe Coulon &
Matthias Pawlowski & Patrick Meuth &
Tatyana Kanyshkova & Ansgar Japes & Sven G. Meuth &
Hans-Christian Pape

Received: 7 November 2007 / Revised: 12 February 2008 / Accepted: 21 February 2008 / Published online: 14 May 2008
# Springer-Verlag 2008

Abstract By combining electrophysiological, immunohistochemical, and computer modeling techniques, we examined the effects of halothane on the standing outward current
(ISO) and the hyperpolarization-activated current (Ih) in rat
thalamocortical relay (TC) neurons of the dorsal lateral
geniculate nucleus (dLGN). Hyperpolarizing voltage steps
elicited an instantaneous current component (Ii) followed
by a slower time-dependent current that represented Ih.
Halothane reduced Ih by shifting the voltage dependency of
activation toward more negative potentials and by reducing
the maximal conductance. Moreover, halothane augmented
Ii and ISO. During the blockade of Ih through Cs+, the
currentvoltage relationship of the halothane-sensitive
current closely resembled the properties of a current
through members of the TWIK-related acid-sensitive K+
(TASK) channel family (ITASK). Computer simulations in a

single-compartment TC neuron model demonstrated that


the modulation of Ih and ITASK is sufficient to explain the
halothane-induced hyperpolarization of the membrane
potential observed in current clamp recordings. Immunohistochemical staining revealed protein expression of the
hyperpolarization-activated cyclic nucleotide-gated (HCN)
channel proteins HCN1, HCN2, and HCN4. Together with
the dual effect of halothane on Ih properties, these results
suggest that Ih in TC neurons critically depends on HCN1/
HCN2 heterodimers. It is concluded that the reciprocal
modulation of Ih and ITASK is an important mechanism of
halothane action in the thalamus.
Keywords TC neurons . Anesthetic . Halothane . HCN .
TASK

Introduction
Thomas Budde and Philippe Coulon are equally contributing first
authors.
T. Budde (*) : P. Coulon : M. Pawlowski : P. Meuth :
T. Kanyshkova : H.-C. Pape
Institut fr Physiologie I,
Westflische Wilhelms-Universitt Mnster,
Robert-Koch-Str. 27a,
48149 Mnster, Germany
e-mail: tbudde@uni-muenster.de
A. Japes
Institut fr Anorganische und Analytische Chemie,
Westflische Wilhelms-Universitt Mnster,
Corrensstr. 30/36,
48149 Mnster, Germany
S. G. Meuth
Klinik fr Neurologie, Julius-Maximilians-Universitt Wrzburg,
Josef-Schneider-Strae 11,
97080 Wrzburg, Germany

Our understanding of the reversible action of inhaled


anesthetics on central nervous function is far from
complete. Today, ion channels are considered to be their
most likely molecular and functional targets [5]. Inhaled
anesthetics enhance inhibitory postsynaptic channel activity
(GABAA and glycine receptors; [29]) and inhibit excitatory
channel activity (acetylcholine, serotonin, and glutamate
receptors; [11, 12]). Volatile anesthetics cause neurons to
hyperpolarize, which was attributed to the activation of K+
leak currents (IKL) and suggested to be a possible
mechanism of action [11, 38, 42, 43, 52]. This view is
corroborated by the finding that several two-pore-domain
K+ (K2P) channels are activated (TASK1, TASK3; TWIKrelated K+ channel 1 [TREK1]) by inhalational anesthetics.
In contrast, one type (TWIK-related halothane-inhibited K+
channel [THIK]) is inhibited [38, 41]. The involvement of

1062

TREK1 in general anesthesia was demonstrated by showing


that TREK1-deficient mice revealed reduced sensitivity to
halothane, sevoflurane, desflurane, and isoflurane [13]. In a
similar way, TASK1- and TASK3-deficient mice were less
sensitive to halothane and isoflurane [18, 19]. Furthermore,
hyperpolarization-activated cyclic nucleotide-gated (HCN)
channels are modulated by the binding of halothane,
making them potentially relevant targets for the clinical
action of inhaled anesthetics [6]. HCN channels expressed
in HEK 293 cells and in higher motoneurons have been
shown to be affected by halothane in two ways: a negative
shift of the voltage dependency of activation mediated by
HCN1 and a reduction in current amplitude mediated by
HCN2.
Anesthesia is often intuitively compared to sleep, although
the electroencephalogram (EEG) pattern during anesthesia
does not show distinct stages or cycling between stages that is
characteristic of naturally occurring sleep. Moreover, in
anesthesia, an isoelectric EEG may occur that is never seen
during sleep. The neurophysiological basis of anesthesia is,
therefore, commonly assumed to be different from sleep, but
there are, nevertheless, some parallels [56]. Anesthetic effects
at specific molecular targets in specific brain regions alter
brain activity in a way that is very similar to the alterations
observed during sleep [33]. Together with behavioral
similarities, these results raise the question, whether neuronal
networks that are normally involved in the generation or
maintenance of sleep, are also influenced by anesthetics.
The thalamocortical network is the major cellular
substrate of sleep-related activity. During periods of slowwave sleep or drowsiness, a marked reduction of the
response to sensory stimuli is accompanied by rhythmic
burst firing of thalamocortical relay (TC) neurons which are
hyperpolarized during these states [22]. On the other hand,
wakefulness is associated with TC neurons being depolarized and showing tonic single-spike activity [50]. In this
state, the neurons faithfully relay the sensory information
from the sensory organs through the thalamus to the
primary sensory areas of the cortex, which is commonly
assumed to be a prerequisite for the conscious perception of
our environment. A strong arousing function is exerted by
the ascending brainstem system (for review, see [22]) which
governs the switch from burst to tonic firing by the release
of neurotransmitters that predominantly act on two types of
ion channels: TASK channels, which give rise to ITASK [27]
and are affected by ACh, and HCN channels, giving rise to
Ih [20, 24], which are affected by noradrenalin and
serotonin. Modulation by brainstem transmitters shifts the
voltage dependency of activation of HCN channels to more
depolarized potentials and inhibits ITASK. Consequently, the
membrane depolarizes and the input resistance increases,
thereby favoring the tonic firing mode. This arousing
mechanism is obviously inactive under anesthesia.

Pflugers Arch - Eur J Physiol (2008) 456:10611073

In the dorsal part of the lateral geniculate nucleus


(dLGN), which is the primary gateway for the processing
and relaying of visual information in the thalamus, ITASK
and Ih contribute to the standing outward current (ISO) of
TC neurons. The interplay between these currents critically
determines the resting membrane potential (Vrest), and their
state of modulation controls the activity mode of these
cells. On the molecular level, these functions are based on
the mRNA expression of TASK1, TASK3, and all HCN
isoforms (HCN14) and the protein coexpression of HCN2
and TASK3 in identified TC neurons [28]. In these cells,
halothane increases a K+ conductance [43], and we have
shown previously that this is at least in part due to ion
channels of the TASK family [27]. However, the reversal
potential of the halothane-sensitive component was more
positive than expected of a TASK current alone, suggesting
the contribution of at least one other ion channel.
In the present study, electrophysiological techniques,
immunohistochemical staining, and computer simulations
were used to investigate whether the interplay of Ih and
ITASK is the basis for the clinical effect of anesthetics that
are antagonistic to the arousing brainstem effects in the
thalamus.

Materials and methods


Preparation
All animal preparations were done according to the
European Communities Council Directive of 24 November
1986 (86/609/EEC). Rats (postnatal days 1230) were
anesthetized with isoflurane and decapitated. In rapid
sequence, after surgically removing a scull cap caudal to
the bregma, a block of brain tissue containing the thalamus
was removed from the cranial vault and submerged in icecold aerated (O2) saline solution containing (in mM):
sucrose, 200; PIPES, 20; KCl, 2.5; NaH2PO4, 1.25;
MgSO4, 10; CaCl2, 0.5; dextrose, 10; pH 7.35, adjusted
with NaOH. Thalamic slices were prepared as coronal
sections on a vibratome. Before recording, slices were kept
submerged in artificial cerebrospinal fluid (ACSF) containing (in mM): NaCl, 125; KCl, 2.5; NaH2PO4, 1.25;
NaHCO3, 24; MgSO4, 2; CaCl2, 2; dextrose, 10; pH
adjusted to 7.35 by bubbling with carbogen (95% O2 and
5% CO2).
Whole-cell patch clamp
Recordings were performed on thalamocortical relay
neurons of the dLGN at room temperature in a solution
containing (in mM): NaCl, 120; KCl, 2.5; NaH2PO4, 1.25;
HEPES, 30; MgSO4, 2; CaCl2, 2; dextrose, 10; pH 7.35

Pflugers Arch - Eur J Physiol (2008) 456:10611073

was adjusted with HCl. Individual cells were visually


identified by infrared differential interference contrast
videomicroscopy. Membrane currents were measured with
glass microelectrodes pulled from borosilicate glass capillaries (GC150T-10; Clark Electromedical Instruments,
Pangbourne, UK) and filled with (in mM): K-gluconate,
95; K3-citrate, 20; NaCl, 10; HEPES, 10; MgCl2, 1; CaCl2,
0.5; BAPTA, 3; Mg-ATP, 3; and Na-GTP, 0.5. The internal
solution was set to a pH of 7.25 with KOH and an
osmolality of 295 mOsm/kg. The electrodes were
connected to an EPC-10 amplifier (HEKA Elektronik,
Lamprecht, Germany) with a chlorinated silver wire.
Electrode resistances were in the range of 23 M with
access resistances in the range of 520 M. Series
resistance compensation of 30% was routinely applied.
Voltage clamp experiments were controlled by the software
Pulse or PatchMaster (HEKA Elektronik) operating on an
IBM-compatible personal computer. Measurements were
corrected for liquid junction potential. All results are
presented as the meanSEM. Substance effects were tested
for statistical significance using the nonparametric Mann
Whitney test (Graph Pad Prism software; Graph Pad, San
Diego, CA, USA). Where applicable, the parametric t test
was used (Origin software). Differences were considered
statistically significant if p<0.05.
To increase the stability of whole-cell recordings and
account for increasingly fast activation kinetics of Ih, the
pulse length was shortened by 500 ms with increasing
hyperpolarization (1.5 s pulse length at 133 mV). This
protocol was used for experiments in which cells had to
tolerate long recording times with extracellular halothane
application. In experiments where halothane was applied
directly with the pipette solution, protocols eliciting long
pulses ranging from 15.5 s at 53 mV to 3.5 s at 133 mV
(1.5 s decrement) were used. Slight differences in the
values of half-maximal activation could be observed
between protocols (see also [44]). We found, while shorter
steps are better tolerated by the cells, halothane effects were
slightly underestimated. Steady-state activation of Ih, p(V),
was estimated by normalizing the tail current amplitudes (I)
50 ms after stepping to a constant potential (indicated by
the circles in Fig. 1a) from a variable amplitude step using
the following equation:
pV I  Imin =Imax  Imin
with Imax being the tail current amplitude for the voltage
step from 133 to 103 mV and Imin for the voltage step
from 43 to 103 mV. Ih activation was well accounted for
by a Boltzmann distribution of the following form:
pV 1=1 expV  Vh =k
where Vh is the voltage of half-maximal activation and k is
the slope factor. The fully activated current vs. voltage

1063

relationship was obtained from instantaneous tail currents.


Ih was evoked by a hyperpolarizing step to 123 mV from a
holding potential of 43 mV. The post step test potential
was varied between 123 and 33 mV in 10 mV increments. The tail current amplitude was plotted against the
test pulse potential, fitted using a linear regression, and
fully activated conductance and reversal potential were
calculated from slope and zero-point, respectively.
The amplitude of Ih (indicated by the bracket in Fig. 1a,
left current traces) was calculated by subtracting the
instantaneous current amplitude (Ii) from the steady-state
current (ISS). The density of Ih was calculated by dividing
the Ih current amplitude at 133 mV by the membrane
capacitance obtained during whole-cell recordings. The
density of Ii was calculated accordingly.
During current clamp recordings, the instantaneous
frequency (fi) of action potential generation was determined
by analyzing the first two action potentials elicited by a
depolarizing current pulse.
Drugs
Halothane, isoflurane, sevoflurane, BaCl2, and CsCl were
obtained from Sigma (Deisenhofen, Germany). CsCl and
BaCl2 were prepared as stock solution in distilled water and
added to the perfusion medium. In one set of experiments,
volatile anesthetics were added in liquid form to the
extracellular solution and administered by perfusion (3
5 min) of the thalamic slices. In another set of experiments,
halothane was delivered to an individual cell via the patch
electrode solution. In both cases, after adding the anesthetic
to the respective aqueous solution at room temperature,
solutions were sonicated for 5 min. If not stated otherwise,
halothane was used at a concentration of 0.1% (v/v). At 25C,
saturation with halothane in aqueous solution occurs at
0.345% and is slightly increased at room temperature.
Therefore, solutions to which 1% halothane were added
can be considered saturated.
Concentration of halothane in the recording chamber
Several factors influence the final halothane concentration
in the recording chamber: outgassing during the preparation
of the solution, absorption by tubing, evaporation along the
line of application and the recording chamber, and delayed
uptake into the slice preparation during the application.
Even the decline in the concentration of a volatile
anesthetic along the length of a typical bath chamber has
been shown to be as high as 33% [39]. Based on the dose
response characteristic of cloned TASK channels expressed
in cell cultures [38], we concluded that only about 10% of
the nominal halothane concentration reaches the target
cells.

1064

Pflugers Arch - Eur J Physiol (2008) 456:10611073

intracellular application

extracellular application

Ih
halo 0.1 %

control

-43 mV

1.0

1.0

0.8

0.8
control

control

0.6
p(V)

p(V)

0.6
0.4

halo 0.1 %

control

halo 0.1 %

0.2

-133mV

0.0

0.4

halo 0.1 %

0.2
0.0

-140 -120 -100

-80

-60

-40

-140 -120 -100

V (mV)

4
2
Ihc on

Ihhal o

Ih

halo

I Lc on

con

ILhalo

Ii

halo

current density (pA/pF)

current density (pA/pF)

con

-40

**

12

10

-60

V (mV)

12

-80

10
8

**

6
4
2
0

con

Ih

halo

con

Ii

halo

Fig. 1 Halothane effect on hyperpolarization-activated inward currents in TC neurons. Representative current traces and analyses
obtained from experiments with intracellular (a) and extracellular (b)
application of halothane are shown. Upper panels current traces
obtained by stepping from a holding potential of 43 to 53, 73,
93, 113, and 133 mV (inset, middle panel) are shown (left
superimposed traces control conditions, right superimposed traces
presence of 0.1% halothane). The tail current voltage was 103 mV.
The duration of each hyperpolarizing step was shortened as command
potentials became more negative. This accounts for increasingly fast
activation kinetics of Ih and improves cell viability. The total trace

length is 18 and 7 s in a and b, respectively. Scale bars represent 3 s/


400 pA (a) and 1 s/300 pA (b, see text for details). The bracket
indicates the amplitude of Ih. The circle indicates the tail current
amplitude measured to calculate the steady-state activation curve.
Middle panels mean steady-state activation curves were obtained by
plotting normalized tail current amplitudes against the step potential
and fitting with a Boltzmann function. Lower panels mean bar graph
representations of current density. Ih amplitude was calculated as Iss
Ii. Current densities were calculated by dividing amplitudes at
133 mV by the membrane capacitance obtained during whole-cell
recordings. *p<0.05, **p<0.01

After adding 0.1% of halothane, the recording solution


was expected to contain 9.4 mM of the anesthetic. To assess
the actual concentration of halothane in the recording
chamber, we measured it using gas chromatography (GC).
GC analyses were carried out on an Agilent HP5890 Series
II (Agilent Technologies, Waldbronn, Germany) equipped
with a flame ionization detector and a Gerstel MPS2
autosampler (Gerstel, Mhlheim/Ruhr, Germany). GC conditions were as follows: 30 m0.25 mm i.d. Agilent DB17ms fused silica capillary column with 0.25 m film
thickness; carrier gas hydrogen at 1.4 ml/min; isothermal

temperature program at 40C; splitless injection, injector


temperature 250C; detector temperature 250C.
Probes were taken from both the original solution, i.e.,
before passing the tubing leading to the bath chamber, and
from the recording chamber, and conveyed into gas-tight
containers. For calibration, a stock solution of halothane
was prepared in cold dimethyl sulfoxide. The solution was
mixed and kept cold and sealed to avoid evaporation of
the analyte. A calibration curve was recorded using final
concentrations from 0 to 12 mM in the same solution used
for the electrophysiological recordings. The necessity of

Pflugers Arch - Eur J Physiol (2008) 456:10611073

an internal standard was tested. No additional volatile


compound was found. Calibration solutions or sample
solutions (200 l) were prepared in a glass vial (1.5 ml)
with septum, vigorously shaken, and equilibrated at room
temperature. Thirty microliters of the headspace was
injected using a gas-tight 100-l syringe. Five measurements of independent solutions were done, showing a high
reproducibility.
From the 9.4 mM (0.1% v/v) halothane added to the
recording solution, only 2.960.02 mM remained in solution
after mixing. After passing the tubing and the bath chamber,
only 1.240.04 mM (n=5) remained in solution. In addition,
we tested a 0.2% halothane solution and found that, of the
18.8 mM halothane added, only 3.940.08 mM were
dissolved in the recording solution and only 2.18
0.06 mM remained in solution after passing the tubing and
recording chamber (all n=5).
The minimum alveolar concentration (MAC) of halothane, the concentration that prevents movement in
response to a noxious stimulus in 50% of the animals, is
around 1% for rats, corresponding to a concentration of
around 0.3 mM in an artificial cerebrospinal fluid (ACSF),
while it is 2.2% for isoflurane, corresponding to 0.71 mM
in ACSF [39]. For young rats, the MAC value is slightly
increased, and around 1.9% in juvenile rats [42]. This
corresponds to a halothane concentration of about 0.7 mM
in ACSF at 22C [30]. Therefore, the halothane concentration used in most of our experiments (nominal 0.1%, which
relates to 1.2 mM in the recording chamber) is about 1.8
times higher and, thus, not necessarily in the clinical range.
However, measurable effects on ISO were already found
with a nominal concentration of 0.02% (see Fig. 3b, d). It is
noted that short application times (<5 min) were necessary
to achieve reversible effects. For a detailed discussion of
halothane concentrations used in situ compared with those
used for anesthesia in vivo, see [42].

1065

Immunohistochemistry
LongEvans rats (postnatal days 2527) were deeply
anesthetized using pentobarbital (50 mg/kg body weight)
and transcardially perfused with PBS, followed by an icecold 4% PFA/PBS for 3540 min. Brains were removed,
postfixed for 4 h in 4% PFA/PBS, and cryoprotected with
30% sucrose. Coronal sections (20 m) were cut at the
level of the dLGN, mounted onto Polysine slide glass
(Menzel, Germany), and air dried. For detection of HCN2,
fresh-frozen sections were used. In this case, brains from
isoflurane-anesthetized rats were removed and frozen in
50C isopentane. Cryostat coronal sections of 20 m
thickness were cut at the level of the dLGN, thaw-mounted
onto Polysine slide glass, air dried, and fixed in 4% PFA/
PBS for 10 min.
After permeabilization with 0.1% Triton X-100 in PBS
for 10 min and several washings with PBS, sections were
blocked with 10% normal horse serum (NHS), 2% BSA in
PBS for 3 h to minimize nonspecific binding before
incubation of slices with primary antibodies: rabbit antiHCN1 (1:500, Alomone Labs, Israel), goat anti-HCN2
(1:250, Santa Cruz Biotechnology, USA) and rabbit antiHCN4 (1:100, Alomone Labs, Israel) in 2% NHS, 2% BSA
in PBS at 4C for 1618 h. After washing (310 min with
PBS), sections were exposed to Cy3-conjugated donkey
antirabbit IgG or Cy2 donkey antigoat IgG (1:400 in 2%
NHS, 2% BSA in PBS, Dianova, Germany) for 1.5 h,
washed again, and coverslipped with Immumount. For the
negative controls, occlusion of the primary antibody from
the staining procedure was routinely performed with no
positive immunological signal detected (see Fig. 6, insets).

Results
Intracellular and extracellular modulation of Ih

Computer simulations with NEURON


A previously described single-compartment TC neuron
model [16, 23] was adapted and used within the NEURON
Simulation Environment [14, 25, 26]. The model is based
on the mathematical description of Ipassive, IHH2, ITASK,
IAslow, IT, and Ih and displays the two typical modes of
action potential generation in thalamic cells: burst firing
with two to six action potentials riding on a low-threshold
Ca2+ spike and single-spike activity with tonic trains of
action potentials. All simulations were calculated at 25C.
The magnitude of halothane-dependent effects on ITASK
(30% increase) and Ih (10 mV hyperpolarizing shift in
the activation curve) used for computer modeling were
estimated from the experiments achieved with 0.1%
halothane.

We examined effects of halothane on the hyperpolarizationactivated current (Ih) in rat TC neurons in dLGN. In Fig. 1,
we compare the controls (upper left panels) to the results
obtained in the presence of halothane (upper right panels).
First, halothane was added to the pipette solution, thus
allowing diffusion into the cytosol (Fig. 1a). Ih was
activated from a holding potential of 43 mV by using
hyperpolarizing voltage steps of increasing (V=10 mV)
amplitude and decreasing (t=1500 ms) duration (3.5 s at
133 mV, see the Materials and methods section)
followed by a constant step to 93 mV (Fig. 1a, upper
panel; the inset shows a scheme of the voltage protocol).
An analysis of the deactivating currents revealed a halfmaximal value of Ih activation (Vh) at a membrane potential
of 861 mV (n=16) under control conditions (Fig. 1a

1066

Pflugers Arch - Eur J Physiol (2008) 456:10611073

intracellular application in Ba2+

extracellular application in Ba2+

-93 mV

halo 0.1 %

control

-113 mV

1.0

1.0

0.8

0.8
control

0.4

0.6
p(V)

p(V)

0.6
halo 0.1 %

0.4

0.2

0.2

0.0

0.0

-140 -120 -100

-80

-60

-40

control
halo 0.1 %

-140 -120 -100

V (mV)
1.2

**

con
C onR e v

normalized Gh

1.0

I hco n

Ih hal o

Ih

halo

-40

I Lc on

con

I Lha lo

Ii

halo

0.6

H a lo R ev

-20

0.4

0.0

halo

-10

0.8

-30

0.2
con

-60

Erev

current density (pA/pF)

-80

V (mV)

C on S l

con

H a lo S l

-40

halo

Fig. 2 Halothane effect on isolated Ih in TC neurons. a Representative current traces and analyses obtained from experiments with
intracellular application of halothane in the presence of Ba2+ (in the
bathing solution) are shown. Current traces (upper panel; long
protocol; scale bar represent 3 s and 100 pA), mean steady-state
activation curves (middle panel), and mean bar graph representations
of current density (lower panel) were obtained as described in Fig. 1.
b Superimposed current traces (upper panel; short protocol) recorded
in Ba2+-containing extracellular solution in the presence (gray traces)

and absence (black traces) of halothane (currents in response to


voltage commands from 43 to 93, 103, and 113 mV are shown;
scale bars represent 1 s and 200 pA). Mean steady-state activation
curves (middle panel) were obtained by plotting the absolute tail
current amplitudes against the step potential and fitting with a
Boltzmann function. The lower panel shows the mean bar graph
representations of the normalized fully activated conductance (left)
and the reversal potential (right) obtained from the fully activated
currentvoltage relationship of Ih (not shown). *p<0.05, **p<0.01

middle panel, filled circles) and a significantly (p<0.001)


more hyperpolarized value (992 mV, n=7) for cells
recorded with halothane added to the pipette solution
(Fig. 1a, middle panel, open circles). A comparison of the
current densities of Ii and Ih at 133 mV revealed a
halothane-induced increase (p<0.05) in Ii (control=6.5
0.9 pA/pF, n=6; halothane=9.71.0 pA/pF, n=7) and a
decrease (p<0.02) in Ih (control=4.30.4 pA/pF, n=6;
halothane=3.10.1 pA/pF, n=7; Fig. 1a, lower panel).
To more directly demonstrate the effect of halothane on Ih
in TC neurons, control currents and currents in the presence
of halothane (35 min after wash in) were recorded from the

same cells (Fig. 1b, upper panel). In this set of experiments,


shorter activation protocols were used (see the Materials
and methods section, holding potential=43 mV, V=
10 mV, t=500 ms, 1.5 s duration at 133 mV).
Extracellular application resulted in a shift of the activation
curve of Ih toward more negative holding potentials (control:
Vh =901 mV; halothane: Vh =971 mV; n=7; p<0.002;
Fig. 1b, middle panel) and the simultaneous increase (p<
0.003) in Ii (control=8.71.0 pA/pF, n=6; halothane=10.7
1.3 pA/pF, n=7) and decrease (p<0.008) in Ih (control=3.5
0.3 pA/pF; halothane=2.50.2 pA/pF; n=7; Fig. 1b, lower
panel).

Pflugers Arch - Eur J Physiol (2008) 456:10611073

1067

Effect on Ih during blockade of IKL

To test whether modulation of Ih is a more general


feature of inhalational anesthetics, the widely used substances isoflurane and sevoflurane were tested (short
activation protocol; 12 mM Ba2+; data not shown). At a
concentration of 0.1% (control: Vh =931 mV; isoflurane:
Vh =1041 mV; n=3/control: Vh =941 mV; sevoflurane:
Vh =1011 mV; n=3) and 0.2% (control: Vh =941 mV;
isoflurane: Vh =1022 mV; n=3), both substances induced
a significant shift of the voltage dependency of Ih (p<0.04).
This hyperpolarizing shift was associated with a significant
(p<0.04) decrease in Ih density (0.1% isoflurane=4.3
0.4 pA/pF under control conditions, 2.60.3 pA/pF in the
presence of isoflurane, n = 3; 0.2% isoflurane = 4.7
0.5 pA/pF under control conditions, 2.20.3 pA/pF in
the presence of isoflurane, n=3; 0.1% sevoflurane=3.7
0.6 pA/pF under control conditions, 2.70.5 pA/pF in the
presence of sevoflurane, n=3). Because both substances
induced an increase in leak currents even in the presence
of 2 mM Ba2+, an analysis of Ih conductance and reversal
potential was not possible by using the tail current
protocol as above.

To assess the effect of halothane on Ih alone, experiments


were performed in the presence of 1 mM Ba2+, which, at this
concentration, is known to block TASK and inward rectifier
channels in TC neurons [27]. Under these recording
conditions, Ii density in the presence (0.50.2 pA/pF, n=7)
and absence (0.40.1 pA/pF, n=6) of intracellular halothane
was indistinguishable (p>0.7). In contrast, Ih density
(control=3.10.2 pA/pF, n=6; halothane=2.20.1 pA/pF,
n=7; p<0.004; Fig. 2a, lower panel) and voltage dependency
(control: Vh =861 mV, n=7; halothane: Vh =921 mV;
n=7; p<0.002; Fig. 2a, middle panel) were significantly
different between the two recording conditions. To more
directly demonstrate the effect of halothane on Ih in a given
cell, extracellular application was used while stimulating the
cell with short activation protocols. While Ii was unchanged,
Ih was clearly reduced (Fig. 2b, upper panel). This effect was
accompanied by a hyperpolarizing shift in the activation
curve (control: Vh =931 mV; halothane: Vh =992 mV;
n=6; p<0.003; Fig. 2b, middle panel) and a decrease of the
fully activated Ih conductance (Gh; control: Gh =4.50.7 nS;
halothane: Gh =4.00.7 nS; n=9; p<0.02), but no significant
changes of the reversal potential were found (Erev; control:
Erev =402 mV; halothane: Erev =373 mV; n=9; Fig. 2b,
lower panel).

Next, the effect of halothane on the ISO (arrowhead in


Fig. 3c) of the TC neurons was analyzed. We have shown

halothane

1%

1.6

increase in ISO (%)

0.2%

1.4
Inor

0.1%

1.2
1.0
0.8

0.01%

1000

60
40
20
0
0.01

2000

0.1

halothane concentration (%)

time (s)

d
-28 mV
800 ms

300

-138 mV

0.2% in Cs+

250
halo 0.2% in

200

Cs+

I (pA)

Fig. 3 Halothane effect on the


standing outward current (ISO)
of TC neurons. a Normalized
mean ISO amplitude vs. time
plots for the indicated halothane
concentrations. Horizontal bar
marks the duration of halothane
application. b Doseresponse
curve of the halothane effect on
ISO. The increase in ISO amplitude is plotted against the nominal halothane concentration. c
Experiments were performed in
the continuous presence of Cs+
to block Ih. Current response to
ramp stimulation (see inset) in
the presence (gray trace) and
absence (black trace) of halothane are shown. Arrowhead
indicates ISO amplitude. Scale
bars represent 200 ms and
50 pA, respectively. d Current
voltage relationship of the halothane-sensitive current with
0.02% (open squares) and 0.2%
(in the presence of Cs+, closed
circles) halothane added to the
external solution

Effect of halothane on ISO

control

150
100

0.02%

50
0

-140 -120 -100 -80 -60 -40 -20


V (mV)

1068

Effect of halothane on thalamic activity modes


The functional consequences of halothane application on TC
neuron activity were probed under current clamp conditions.
To ensure tonic responses, recordings were obtained at
depolarized membrane potentials using DC injection (VM =
59.70.2 mV, n=6). Under these conditions, depolarizing
current steps elicited tonic firing (fi =20.02.6 Hz, n=6;
Fig. 4b, left panel). The application of halothane resulted in a
significant (p<0.004) hyperpolarization of the membrane
potential to VM =65.51.2 mV (Fig. 4a) and a change from
tonic to burst firing (fi =152.617.8 Hz; n=5; Fig. 4b, right
panel). It is noted that most low-threshold Ca2+ spikes under
these recording conditions were crowned by only a single
action potential. In one out of six cells, no burst was
observed. This was possibly due to halothane reducing the
input resistance and T-type Ca2+ current [42].

-55
halo 0.1%

-60
V (mV)

earlier that current through TASK channels significantly


contributes to this steady-state current component [26, 27].
Application of halothane (35 min) at different concentrations
resulted in a fully reversible (Fig. 3a) and dose-dependent
(Fig. 3b) increase in ISO amplitude at 28 mV (0.01%
halothane=0.53.3% normalized current increase, n=4;
0.02% halothane=5.90.2%, n=3; 0.1% halothane=28.7
2.9%, n=3; 0.2% halothane=49.23.2%, n=5; 1% halothane=59.09.2%, n=7). At pH 6.0, the increasing effect of
1% halothane on ISO was reduced to 48.06.7% (n=3; data
not shown). Furthermore, 0.1% halothane decreased ISO
amplitudes by 12.82.1% (n=4) under acidic recording
conditions (data not shown). To further characterize the
current component sensitive to halothane, the membrane
potential was ramped from 28 mV in 800 ms to 138 mV
once every 20 s (Fig. 3c, inset). The rate of hyperpolarization
(0.14 mV/ms) is sufficiently slow to allow the membrane
current to reach steady-state at each potential and it is
expected that only constitutively open channels can follow
the ramp. The currentvoltage relationship of the halothanesensitive current was obtained by subtracting the control
currents from those recorded in the presence of halothane
(i.e., halothanecontrol) and was characterized by an
outward rectification (mean current at 28 mV=104.9
4.9 pA) and a reversal potential (Erev) of 96.97.0 mV
(n=3; data not shown), i.e., approximately 7 mV positive to
the expected K+ equilibrium potential (EK =104 mV). As
this slight deviation from EK could be attributed to the contribution of Ih, experiments were performed in the presence
of the Ih blocker Cs+ [35]. As before, the application of
halothane (0.2%) resulted in a strong increase in ISO
amplitude (Fig. 3c). The currentvoltage relationship of the
halothane-sensitive current revealed strong outward rectification and a reversal (Erev =105.52.2 mV; n=4) very close
to EK (Fig. 3d, closed circles).

Pflugers Arch - Eur J Physiol (2008) 456:10611073

-65

-70

1000

2000

time (s)

control

-60 mV

halo 0.1%

-66 mV

control

-10 mV shift in Ih
30% increase in ITASK
model cell

-60 mV

-66 mV

Fig. 4 Halothane effect on the membrane potential of TC neurons. a


Normalized mean voltage vs. time plot. Horizontal bar indicates the
duration of halothane application. b Voltage response to a depolarizing
current step under control conditions (left trace) and in the presence of
halothane (right trace). The slightly depolarized holding potential of
60 mV was obtained by DC current injection (Vrest =71 mV). This
current was not changed during the experiment. c Voltage response of
a single-compartment TC neuron computer model to a depolarizing
current step under control conditions (left trace) and after shifting Ih
and increasing ITASK (right trace). The resting membrane potential of
the computer model was 70 mV. The control value of 60 mV was
achieved by DC current injection which was not changed in the course
of the simulation. The low-threshold Ca2+ spike with the crowning
burst of action potentials is displayed at an expanded time scale. Scale
bars represent 100 ms and 20 mV in the main panel and 50 ms and
20 mV in the inset

To further characterize the effect of halothane on TC


neurons, continuous current clamp recordings including
manual voltage clamp were performed (Fig. 5). Again,
recordings were obtained at depolarized membrane potentials using DC injection (VM =57.00.9 mV; DC=79.3
10.8 pA, n=7). Bath application of halothane (0.2%)
resulted in a significant (p<0.001) membrane hyperpolarization of 6.51.0 mV (n=7). The injection of hyperpolarizing (150 pA) and depolarizing current pulses
(+200 pA, 800 ms, 0.4 Hz) resulted in deflections of the
membrane potential which indicated an apparent input

Pflugers Arch - Eur J Physiol (2008) 456:10611073


1

1069
4

Expression of HCN channel proteins in dLGN

-60 mV

10
mV
0.2% halothane

30 s

-60 mV

2s

Fig. 5 Typical response of a TC neuron to halothane. Application of


halothane (0.2%) resulted in a membrane hyperpolarization and
decrease in apparent input resistance. Responses of the cell to the
constant current pulses before (1), during (2 and 3; the latter with
manual voltage clamp), and after 10 min recovery (4; note that the
timeline is noncontinuous) from halothane are expanded for comparison. The low-threshold Ca2+ spike is displayed at an expanded time
scale. The black bar represents halothane application as indicated.
Dashed lines indicate membrane potential before the application of
halothane. Fast Na+/K+-mediated action potentials are truncated

resistance of 28430 M (n=7). Compensation of the


halothane-induced hyperpolarization with intracellular injection of DC current (manual voltage clamp) revealed a
significant (p<0.02) decrease in input resistance associated
with the response (21822 M, n=7). Moreover, while
under control conditions, every depolarizing current step
elicited tonic firing (Fig. 5, insets 1 and 4); this was
abolished in the presence of halothane (Fig. 5, inset 2) even
after compensating for the halothane-induced hyperpolarization (Fig. 5, inset 3).
To determine whether the modulation of Ih and ITASK
are sufficient to account for the halothane effects seen
under current clamp conditions, we used computer
modeling techniques [28]. The model cell was held at a
potential of 60 mV using DC injection. From this
potential, a sufficiently strong step depolarization evoked
tonic firing (fi =18 Hz; Fig. 4c, left panel). Incorporating
the effect of 0.1% halothane on Ih (10 mV shift in Vh)
and ITASK (30% increase in ISO) into the computer model
resulted in a hyperpolarization of the membrane potential
to 66 mV which was accompanied by an induction of
typical burst firing (fi =98 Hz; Fig. 4c, right panel) upon
step depolarization.

The HCN family of ion channels represents the molecular


substrate for Ih in native cells. Halothane seems to exert a
dual effect on the electrophysiological properties of Ih: a
negative shift of the activation potential and a reduction of
the current amplitude. The negative shift of the activation
potential can be attributed to HCN1, and the reduction of
the current amplitude can be attributed to HCN2 [6]. To
verify the presence of the channel isoform proteins in TC
neurons and thus, that the electrophysiological effects can
indeed be caused by an action of halothane on these two
isoforms, we applied immunohistochemical staining. The
results indicate that HCN1, HCN2, and HCN4 are
expressed in dLGN (Fig. 6). HCN1 (Fig. 6, left panel),
HCN2 (Fig. 6, middle panel), and HCN4 (Fig. 6, right
panel) show a clear somatic localization with some
dendritic staining for HCN1.

Discussion
Effect of halothane on HCN channels in TC neurons
We have shown previously that the halothane-sensitive
current in TC neurons had a reversal potential positive to
the K+ equilibrium potential and wasin addition to the
TASK channelscarried by as yet uncharacterized Na+
and/or Ca2+ channels [27]. This conclusion is corroborated
by the finding that Na+-permeable HCN channels are
modulated by halothane. Inhibition of Ih by halothane in
native TC neurons is characterized by two effects: a
hyperpolarizing shift in the activation curve and a
decrease in the maximal available current. Recent studies
on cloned channels in expression systems indicate subunit-specific effects of clinically relevant concentrations of
halothane with the shift in Ih voltage dependency and the
reduction in the maximal available current being mediated
by HCN1 and HCN2, respectively [6], and both effects
can be observed concurrently in HCN1HCN2 heteromeric channels. Therefore, our results suggest that Ih in
TC neurons is critically based on functional HCN1HCN2
heterodimers, which is corroborated by the findings that
HCN1 and HCN2 mRNA expression was found in TC
neurons [4, 28] and the respective proteins are present in
the plasma membrane (this study). Although HCN2 was
found to be the dominant isoform in TC neurons [20, 28],
only a small reduction in the maximal available current
was observed (Fig. 2b, lower panel). This discrepancy
may be explained by the finding that the predominant
action of halothane on HCN1HCN2 heterodimers
depends on the relief of the C-terminal-dependent basal
inhibition by cAMP [6].

1070

Pflugers Arch - Eur J Physiol (2008) 456:10611073

Fig. 6 Immunohistochemical labeling of HCN channels in dLGN.


Specific antibodies directed against HCN1 (left images), HCN2
(middle images), and HCN4 (right images) were detected by
fluorescent microscopy and detection of fluorochrome-coupled appropriate secondary antibodies (Cy2, HCN2; Cy3, HCN1, and HCN4;

upper row low resolution; lower row high resolution). The insets
(upper row) show negative controls (occlusion of the primary
antibody). Images were deconvoluted using the Zeiss AxioVision
(4.6.3) software package

However, the conclusion that Ih in TC neurons is based on


HCN1HCN2 heterodimers is complicated by several
circumstances: (1) TC neurons express HCN4 ([36]; this
study). HCN channel subunits assemble to a variety of
different heteromers, including HCN1/HCN4 and HCN2/
HCN4 [1, 32]. The modulation of HCN4 and heteromeric
channels containing HCN4 has not yet been determined. (2)
The manifestation of the two halothane effects on Ih depends
on the degree of HCN channel relief from basal inhibition by
cAMP [6]. Therefore, the more prominent effect of halothane
on Ih voltage dependency in TC neurons may be attributed to
significant basal adenylate cyclase activity, resulting in basal
cAMP levels of around 10 nM and influencing Ih voltage
dependency [4]. (3) Halothane has been shown to directly
affect adenylate cyclase in rat cerebellum and cerebral cortex
[54], thereby increasing the rate of cAMP production. This
would lead to a depolarizing shift of the voltage dependence
of Ih with Gh remaining unaltered [21, 37].

dominant isoform [2628]. Although the results presented


in this study demonstrate that TASK channels are a major
target of halothane in TC cells, the interpretation of the
halothane effect is complicated by the finding that TREK1
and THIK2 genes are expressed in dLGN [28]. Because
numerous current components underlie ISO in TC neurons
[28], the specific contribution of ion channel subtypes is
difficult to assess and has not been performed for K2P
channels other than TASK. Nevertheless, the presence of an
ISO component that is inhibited by cAMP [3] and activated
by arachidonic acid [28] is in agreement with the functional
expression of TREK1 [15]. Therefore, TREK1 cannot be
excluded to contribute to the halothane-sensitive outward
current in TC neurons. Furthermore, it is likely that the
reduction of ISO by halothane seen during extracellular
acidification is caused by the inhibition of THIK2 [41],
which points to a complex activation and inhibition of K2P
channels in TC neurons by halothane.

Correlation of K2P channel expression and halothane effects


in TC neurons

Halothane suspends sensory system relay

TC neurons carry a prominent pH-sensitive outwardly


rectifying K+ current that is mediated by TASK1 and
TASK3 channels. The TASK3 channel represents the

The thalamocortical neuronal network exerts two distinct


states of activity: (1) low-frequency oscillatory activity
(delta activity and spindling) during natural sleep and deep
anesthesia and (2) high-frequency oscillatory activity in the

Pflugers Arch - Eur J Physiol (2008) 456:10611073

gamma range and tonic activity during waking and REM


sleep [45, 49]. TC neurons accordingly possess two modes
of activity: (1) hyperpolarized and rhythmically bursting
during states of slow-wave sleep and anesthesia, (2)
depolarized and tonic firing of single action potentials that
relay incoming sensory signals through the dorsal thalamus
toward the cortex [46]. The thalamus is, therefore, regarded
as being one of the central structures in the control of the
sleepwake cycle and in the induction and maintenance of
anesthesia.
Adenosine is directly linked to the energy metabolism of
cells and is regarded as a sleep-inducing factor [2, 40]. In
the central nervous system, an increase in neuronal activity
enhances energy consumption and extracellular adenosine
concentrations. In most brain areas, high extracellular
adenosine concentrations via A1 adenosine receptors decrease neuronal activity and, thus, energy consumption. In
the thalamus, adenosine hyperpolarizes TC neurons due to
an increase in membrane potassium conductance and a
decrease in Ih, which possibly results in an inhibition of
tonic firing and promotes burst discharges [34]. Similarly,
halothane will strongly disfacilitate conditions of faithful
signal transfer in the thalamocortical system in that the
hyperpolarization in relay neurons shifts the membrane
potential out of the range of tonic action potential firing.
These findings indicate that the intuitive analogy between
natural sleep and anesthesia is not only a matter of common
sense but indeed has a cellular basis because halothane and
adenosine modulate very similar ionic mechanisms to
counteract the ascending brainstem system.
Both TASK and HCN channels have been identified as
major molecular determinants of the resting membrane
potential, and the reciprocal modulation of the corresponding
currents has been shown to be the most effective way to
toggle between the two activity modes of TC neurons [28].
Thus, in addition to NA, serotonin [22], and adenosine [34],
halothane represents another example for this type of
modulatory mechanism. Moreover, reciprocal modulation
of a TASK-like current and Ih has been observed before in
brainstem motoneurons and has been attributed to the
decreased motoneuronal excitability that accompanies
anesthetic-induced immobilization [48]. The present study
adds to this scenario by showing that very similar
mechanisms lead to sleep-like activity of TC neurons,
thereby pointing to a more general principle.
It could be argued that the action of halothane on TC
neurons is a specific blockade of ion channels rather than
the basis for its anesthetic effect. Indeed, halothane seems
to specifically bind to a number of ion channels (for review,
see [55]). However, halothane acts on target brain regions
that are in perfect strategic positions to affect essential
parameters of anesthesia, including consciousness (thalamus), pain perception (thalamus), and voluntary movement

1071

(brainstem motoneurons): (1) Halothane action on TC


neurons may lead to the breakdown of sensory transmission
and the loss of conscious awareness via the induction of a
membrane hyperpolarization and, hence, the switch from
tonic to burst firing (this study). (2) The membrane
hyperpolarization and decrease in excitability of hypoglossal motoneurons may lead to the dampening of muscle
activity [48]. (3) The bradycardic action of halothane may
have its cause in its effect on cardiac HCN channels. (4)
The counteracting balance of HCN and TASK is essential
for maintaining the resting membrane potential in central
neurons [7, 28, 48]. Given the widespread distribution of
TASK and HCN channels in the brain, it may be concluded
that the modulation of these channels is an essential part of
the clinical action of halothane [31, 53]. This view is
further corroborated by the finding that isoflurane and
sevoflurane also act on Ih (this study) and ITASK [38, 47].
Is the reciprocal modulation of TASK and HCN channels
sufficient to explain the halothane effect?
Although the reciprocal modulation of TASK and HCN
channels seems to be a more general mechanism of
halothane action in central areas for motor control and
awareness, and computer modeling indicates that this effect
is sufficient to explain the halothane-induced hyperpolarization seen in TC neurons, additional effects can be
expected. The shunting effect of the increased K+ conductance on postsynaptic potentials, Na+/K+ action potentials,
and low-threshold Ca2+ spikes [42, 51] will block the
transfer of all sensory and motor activity coded as tonic or
burst firing [10, 58], contributing in analgesia, loss of
awareness, and the suppression of motor activity and
leading to the observed isoelectric EEG during anesthesia.
Furthermore, anesthetic sites on GABA receptors cause
lower GABA concentrations to work but also increase the
amplitude of GABAergic signals [17], while the amplitudes
of excitatory postsynaptic cholinergic and/or glutamatergic
currents are decreased [9] (for review, see [5]), although the
ratio between these effects may vary between cell types.
For example, in somatosensory TC neurons in Wistar rats
in vivo, the volatile anesthetic isoflurane exerts a concentration-dependent effect on GABAergic inhibition leading
to a block of thalamocortical information transfer and, to a
lesser extent, a suppression of glutamatergic signals and
synapses [8, 57].
Acknowledgements The authors wish to thank E. Na and A. Jahn
for the excellent technical assistance. We also wish to thank Prof. J.
Andersson for his kind help with the gas chromatography and Prof. E.
Pogatzki-Zahn for kindly providing the sevoflurane. This study was
supported by DFG (BU 1019/7-1; Pa 336/17-1), Innovative Medizinische Forschung (IMF; BU 120501), and Interdisziplinres Zentrum fr
Klinische Forschung (IZKF; Bud/005/07 to TB; A-54 to SGM).

1072

References
1. Altomare C, Terragni B, Brioschi C, Milanesi R, Pagliuca C,
Viscomi C, Moroni A, Baruscotti M, DiFrancesco D (2003)
Heteromeric HCN1HCN4 channels: a comparison with native
pacemaker channels from the rabbit sinoatrial node. J Physiol
(Lond) 549:347359
2. Basheer R, Strecker RE, Thakkar MM, McCarley RW (2004)
Adenosine and sleepwake regulation. Prog Neurobiol 73:379396
3. Budde T, Biella G, Munsch T, Pape H-C (1997) Lack of regulation
by intracellular Ca2+ of the hyperpolarization-activated cation
current in rat thalamic neurons. J Physiol (Lond) 503.1:7985
4. Budde T, Caputi L, Kanyshkova T, Staak R, Abrahamczik C,
Munsch T, Pape HC (2005) Impaired regulation of thalamic
pacemaker channels through an imbalance of subunit expression
in absence epilepsy. J Neurosci 25:98719882
5. Campagna JA, Miller KW, Forman SA (2003) Mechanisms of
actions of inhaled anesthetics. N Engl J Med 348:21102124
6. Chen X, Sirois JE, Lei Q, Talley EM, Lynch C III, Bayliss DA (2005)
HCN subunit-specific and cAMP-modulated effects of anesthetics on
neuronal pacemaker currents. J Neurosci 25:58035814
7. Day M, Carr DB, Ulrich S, Ilijic E, Tkatch T, Surmeier DJ (2005)
Dendritic excitability of mouse frontal cortex pyramidal neurons
is shaped by the interaction among HCN, Kir2, and Kleak
channels. J Neurosci 25:87768787
8. Detsch O, Kochs E, Siemers M, Bromm B, Vahle-Hinz C (2002)
Differential effects of isoflurane on excitatory and inhibitory
synaptic inputs to thalamic neurones in vivo. Br J Anaesth
89:294300
9. Dilger JP, Vidal AM, Mody AI, Liu Y (1994) Evidence for direct
actions of general anesthetics on an ion channel protein. A new look
at a unified mechanism of action. Anesthesiology 81:431442
10. Fanselow EE, Sameshima K, Baccala LA, Nicolelis MA (2001)
Thalamic bursting in rats during different awake behavioral states.
Proc Natl Acad Sci U S A 98:1533015335
11. Franks NP, Lieb WR (1994) Molecular and cellular mechanisms
of general anaethesia. Nature 367:607614
12. Franks NP, Lieb WR (1998) Which molecular targets are most
relevant to general anaesthesia? Toxicol Lett 100101:18
13. Heurteaux C, Guy N, Laigle C, Blondeau N, Duprat F, Mazzuca
M, Lang-Lazdunski L, Widmann C, Zanzouri M, Romey G,
Lazdunski M (2004) TREK-1, a K(+) channel involved in
neuroprotection and general anesthesia. EMBO J 23:26842695
14. Hines ML, Carnevale NT (2001) NEURON: a tool for neuroscientists. Neuroscientist 7:123135
15. Honore E (2007) The neuronal background K2P channels: focus
on TREK1. Nat Rev Neurosci 8:251261
16. Huguenard JR, McCormick DA (1992) Simulation of the currents
involved in rhythmic oscillations in thalamic relay neurons. J
Neurophysiol 68:13731383
17. Jones MV, Harrison NL (1993) Effects of volatile anesthetics on
the kinetics of inhibitory postsynaptic currents in cultured rat
hippocampal neurons. J Neurophysiol 70:13391349
18. Linden A-M, Aller MI, Leppa E, Vekovischeva O, Aitta-aho T,
Veale EL, Mathie A, Rosenberg P, Wisden W, Korpi ER (2006)
The in vivo contributions of TASK-1-containing channels to the
actions of inhalation anesthetics, the {alpha}2 adrenergic sedative
dexmedetomidine, and cannabinoid agonists. J Pharmacol Exp
Ther 317:615626
19. Linden A-M, Sandu C, Aller MI, Vekovischeva OY, Rosenberg
PH, Wisden W, Korpi ER (2007) TASK-3 knockout mice exhibit
exaggerated nocturnal activity, impairments in cognitive functions, and reduced sensitivity to inhalation anesthetics. J Pharmacol Exp Ther 323:924934

Pflugers Arch - Eur J Physiol (2008) 456:10611073


20. Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff K,
Langebartels A, Wotjak C, Munsch T, Zong X, Feil S, Feil R,
Lancel M, Chien KR, Konnerth A, Pape HC, Biel M, Hofmann F
(2003) Absence epilepsy and sinus dysrhythmia in mice lacking
the pacemaker channel HCN2. EMBO J 22:216224
21. Maingret F, Patel AJ, Lesage F, Lazdunski M, Honore E (1999)
Mechano- or acid stimulation, two interactive modes of activation
of the TREK-1 potassium channel. J Biol Chem 274:26691
26696
22. McCormick DA (1992) Neurotransmitter actions in the thalamus
and cerebral cortex and their role in neuromodulation of
thalamocortical activity. Prog Neurobiol 39:337388
23. McCormick DA, Huguenard JR (1992) A model of the electrophysiological properties of thalamocortical relay neurons. J
Neurophysiol 68:13841400
24. McCormick DA, Pape H-C (1990) Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in
thalamic relay neurones. J Physiol (Lond) 431:291318
25. Meuth P, Meuth SG, Jacobi D, Broicher T, Pape HC, Budde T
(2005) Get the rhythm: modeling of neuronal activity. Journal of
Undergraduate Neuroscience Education 4(1):A1A11
26. Meuth SG, Aller MI, Munsch T, Schuhmacher T, Seidenbecher T,
Kleinschnitz C, Pape HC, Wiendl H, Wisden W, Budde T (2006a)
The contribution of TASK-1-containing channels to the function
of dorsal lateral geniculate thalamocortical relay neurons. Mol
Pharmacol 69:14681476
27. Meuth SG, Budde T, Kanyshkova T, Broicher T, Munsch T, Pape
H-C (2003) Contribution of TWIK-related acid-sensitive K+
channel 1 (TASK1) and TASK3 channels to the control of activity
modes in thalamocortical neurons. J Neurosci 23:64606469
28. Meuth SG, Kanyshkova T, Meuth P, Landgraf P, Munsch T,
Ludwig A, Hofmann F, Pape HC, Budde T (2006b) The
membrane resting potential of thalamocortical relay neurons is
shaped by the interaction among TASK3 and HCN2 channels. J
Neurophysiol 96:15171529
29. Mihic SJ, Ye Q, Wick MJ, Koltchine VV, Krasowski MD, Finn
SE, Mascia MP, Valenzuela CF, Hanson KK, Greenblatt EP,
Harris RA, Harrison NL (1997) Sites of alcohol and volatile
anaesthetic action on GABAA and glycine receptors. Nature
389:385389
30. Mikulec AA, Pittson S, Amagasu SM, Monroe FA, MacIver MB
(1998) Halothane depresses action potential conduction in
hippocampal axons. Brain Res 796:231238
31. Monteggia LM, Eisch AJ, Tang MD, Kaczmarek LK, Nestler EJ
(2000) Cloning and localization of the hyperpolarization-activated
cyclic nucleotide-gated channel family in rat brain. Brain Res Mol
Brain Res 81:129139
32. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L,
Moosmang S, Ludwig A, Biel M (2003) Role of subunit
heteromerization and N-linked glycosylation in the formation of
functional hyperpolarization-activated cyclic nucleotide-gated
channels. J Biol Chem 278:4378143786
33. Nelson LE, Guo TZ, Lu J, Saper CB, Franks NP, Maze M (2002)
The sedative component of anesthesia is mediated by GABAA
receptors in an endogenous sleep pathway. Nat Neurosci 5:979984
34. Pape H-C (1992) Adenosine promotes burst activity in guinea-pig
geniculocortical neurones through two different ionic mechanisms. J Physiol (Lond) 447:729753
35. Pape H-C (1996) Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annu Rev Physiol
58:299327
36. Pape HC, Kanyshkova T, Broicher T, Budde T (2007) Developmental and functional profile of the thalamic hyperpolarizationactivated cation current, Ih, in absence epilepsy. Thalamus &
Related Systems DOI 10.1017/S1472928807000180

Pflugers Arch - Eur J Physiol (2008) 456:10611073


37. Patel A, Lazdunski M (2004) The 2P-domain K+ channels: role in
apoptosis and tumorigenesis. Pflugers Arch 448:261273
38. Patel AJ, Honore E, Lesage F, Fink M, Romey G, Lazdunski M
(1999) Inhalational anesthetics activate two-pore-domain background K+ channels. Nat Neurosci 2:422426
39. Pearce RA (1996) Volatile anaesthetic enhancement of pairedpulse depression investigated in the rat hippocampus in vitro. J
Physiol (Lond) 492:823840
40. Porkka-Heiskanen T, Alanko L, Kalinchuk A, Stenberg D (2002)
Adenosine and sleep. Sleep Med Rev 6:321332
41. Rajan S, Wischmeyer E, Karschin C, Preisig-Muller R, Grzeschik
K-H, Daut J, Karschin A, Derst C (2001) THIK-1 and THIK-2, a
novel subfamily of tandem pore domain K+ channels. J Biol
Chem 276:73027311
42. Ries CR, Puil E (1999a) Mechanism of anesthesia revealed by
shunting actions of isoflurane on thalamocortical neurons. J
Neurophysiol 81:17951801
43. Ries CR, Puil E (1999b) Ionic mechanism of isofluranes actions
on thalamocortical neurons. J Neurophysiol 81:18021809
44. Seifert R, Scholten A, Gauss R, Mincheva A, Lichter P, Kaupp UB
(1999) Molecular characterization of a slowly gating human
hyperpolarization-activated channel predominantly expressed in
thalamus, heart, and testis. Proc Natl Acad Sci U S A 96:93919396
45. Sherman SM, Guillery RW (1996) Functional organization of
thalamocortical relays. J Neurophysiol 76:13671395
46. Sherman SM, Guillery RW (2001) Exploring the thalamus.
Academic Press, San Diego
47. Sirois JE, Lei Q, Talley EM, Lynch C 3rd, Bayliss DA (2000) The
TASK-1 two-pore domain K+ channel is a molecular substrate for
neuronal effects of inhalation anesthetics. J Neurosci 20:63476354
48. Sirois JE, Lynch C 3rd, Bayliss DA (2002) Convergent and
reciprocal modulation of a leak K+ current and Ih by an inhalational

1073

49.

50.
51.

52.

53.

54.

55.

56.
57.

58.

anaesthetic and neurotransmitters in rat brainstem motoneurones. J


Physiol (Lond) 541:717729
Steriade M (1997) Synchronized activities of coupled oscillators
in the cerebral cortex and thalamus at different levels of vigilance.
Cereb Cortex 7:583604
Steriade M, Jones EG, McCormick DA (1997) Thalamus.
Elsevier, Amsterdam
Sugiyama K, Muteki T, Shimoji K (1992) Halothane-induced
hyperpolarization and depression of postsynaptic potentials of
guinea pig thalamic neurons in vitro. Brain Res 576:97103
Talley EM, Bayliss DA (2002) Modulation of TASK-1 (Kcnk3)
and TASK-3 (Kcnk9) potassium channels. Volatile anesthetics and
neurotransmitters share a molecular site of action. J Biol Chem
277:1773317742
Talley EM, Solorzano G, Lei Q, Kim D, Bayliss DA (2001) CNS
distribution of members of the two-pore-domain (KCNK) potassium channel family. J Neurosci 21:74917505
Triner L, Vulliemoz Y, Woo S-Y, Verosky M (1980) Halothane
effect on cAMP generation and hydrolysis in rat brain. Eur J
Pharmacol 66:7380
Trudell JR, Bertaccini E (2002) Molecular modelling of
specific and non-specific anaesthetic interactions. Br J Anaesth
89:3240
Tung A, Mendelson WB (2004) Anesthesia and sleep. Sleep Med
Rev 8:213225
Vahle-Hinz C, Detsch O, Siemers M, Kochs E (2007) Contributions of GABAergic and glutamatergic mechanisms to isofluraneinduced suppression of thalamic somatosensory information
transfer. Exp Brain Res 176:159172
Weyand TG, Boudreaux M, Guido W (2001) Burst and tonic
response modes in thalamic neurons during sleep and wakefulness. J Neurophysiol 85:11071118

Pflugers Arch - Eur J Physiol (2008) 456:10491060


DOI 10.1007/s00424-008-0473-x

CELLULAR NEUROPHYSIOLOGY

Muscarinic ACh receptor-mediated control of thalamic


activity via Gq/G11-family G-proteins
Tilman Broicher & Nina Wettschureck &
Thomas Munsch & Philippe Coulon & Sven G. Meuth &
Tatyana Kanyshkova & Thomas Seidenbecher &
Stefan Offermanns & Hans-Christian Pape &
Thomas Budde

Received: 18 October 2007 / Revised: 16 January 2008 / Accepted: 7 February 2008 / Published online: 19 March 2008
# Springer-Verlag 2008

Abstract A genetic knock out was used to determine the


specific contribution of Gq/G11-family G-proteins to the
function of thalamocortical relay (TC) neurons. Disruption
of Gq function in a conditional forebrain-specific Gq/

G11-double-deficient mouse line Gq G=
11 had no
effects on the resting membrane potential (Vrest) and the
amplitude of the standing outward current (ISO). Stimulation of muscarinic acetylcholine (ACh) receptors (mAChR;
muscarine, 50 M) induced a decrease in ISO amplitude in
wild-type mice (364%, n=5), a constitutive G11-deficient
=
mouse line (G11 ; 363%, n=8), and Gq =G=
(11
11
2%, n=16). Current-clamp recordings revealed a muscarineinduced positive shift in Vrest of 232 mV (n=6), 185 mV
Tilman Broicher, Nina Wettschureck and Thomas Munsch are equally
contributing first authors.
T. Broicher : P. Coulon : T. Kanyshkova : T. Seidenbecher :
H.-C. Pape : T. Budde (*)
Institut fr Physiologie I,
Westflische Wilhelms-Universitt Mnster,
Robert-Koch-Str. 27a,
48149, Mnster, Germany
e-mail: tbudde@uni-muenster.de
N. Wettschureck : S. Offermanns
Institut fr Pharmakologie, Universitt Heidelberg,
Im Neuenheimer Feld 366,
69120, Heidelberg, Germany
T. Munsch
Institut fr Physiologie, Otto-von-Guericke-Universitt,
Leipziger Str. 44,
39120, Magdeburg, Germany
S. G. Meuth
Neurologische Klinik, Bayerische Julius-Maximilians-Universitt,
Josef-Schneider-Str. 11,
97080, Wrzburg, Germany

=

(n=5),
. and 21 mV (n=9) in wild type, G11 , and
=
Gq G11 , respectively. This depolarization was associated
with a change in TC neuron activity from burst to tonic firing
=
in wild type and G11 , but not in Gq =G=
11 . The use of
specific antibodies and of pharmacological agents with
preferred affinity points to the contribution of m1AChR and
m3AChR. In conclusion, we present two novel aspects of the
physiology of the thalamocortical system by demonstrating
that the depolarization of TC neurons, which is induced by
the action of transmitters of ascending brainstem fibers, is
governed roughly equally by both m1AChR and m3AChR
and is transduced by Gq but not by G11.
Keywords Thalamic function . Gene knock out .
TASK channels . Thalamocortical relay neurons .
G-proteins . Muscarinic ACh receptors

Introduction
The thalamocortical system is characterized by highly
synchronized oscillatory burst activity (<15 Hz) during
states of slow wave sleep and tonic generation of action
potentials and high-frequency oscillations (~40 Hz) during
wakefulness and rapid eye movement (REM) sleep [36].
Neurons located in brainstem cholinergic nuclei, the locus
coeruleus, and the raphe nuclei release acetylcholine (ACh),
noradrenaline (NA), and serotonin (5-HT) into the thalamus
and mediate the switch between activity modes [24]. A
common action of these transmitters is a depolarizing shift
of the membrane potential of thalamocortical relay (TC)
neurons, leading to cessation of rhythmic bursts and
occurrence of tonic activity. One crucial step of membrane
depolarization is the decrease in a leak K+ conductance

1050

(IKL), the molecular nature of which has recently been


attributed to members of the two pore-domain K+ (K2P)
channels, namely TASK-1 and TASK-3 [26, 27]. Similar to
the modulation of IKL upon, for instance, activation of
muscarinic ACh receptors (mAChR) [24], TASK-1 and
TASK-3 are inhibited by the activation of m1AChR coupled
to Gq/G11 [4, 5, 18]. Whether TASK channels are directly
modulated by Gq or indirectly via phospholipase C
activation and depletion of phosphatidyl-4,5-inositolbisphosphate (PIP2) is currently a matter of discussion.
Since a wide variety of modulators in the brain exert
their biological effects by the activation of receptors
coupled to members of the Gq/G11 family of heterotrimeric
G-proteins, genetic mouse models have been shown to be
useful tools to investigate the physiology and pathophysiology of G-protein-mediated signaling [41]. In this study,
we used conditional.forebrain-specific Gq/G11-doubledeficient mice Gq G=
11 [40] to analyze the role of
Gq/G11-mediated signaling in TC neurons.

Materials and methods


Mice
Generation of forebrain-specific Gq/G11-deficient mice
and genotyping of the gnaqflox allele, of gna11-wildtype
and -knockout alleles, and of the Cre-transgene have been
described previously [40, 43]. The genetic background was
predominantly C57BL6/N (fourth-generation backcross),
and non-littermate C57BL6/N mice or littermates with the
genotype Camkcre4-/-; gnaqfl/fl; gna11-/- were used as
controls. In accordance with former observations on cardiac
growth and development [32], no significant differences
were observed between the two control groups.
Immunohistochemistry and LacZ staining
Mice were deeply anesthetized with pentobarbital (100 mg/kg
i.p.) and perfused with 4% paraformaldehyde (PFA) via the left
ventricle. Brains were post-fixed overnight and stored in 0.5%
PFA at 4C. Vibratome sections (50 m) were cut and
incubated overnight at 4C with mouse anti-GAD67 antibody
(1:2.500; Chemicon, Hofheim/Ts, Germany), mouse anti NeuN
antibody (1:2.000; Chemicon), mouse anti parvalbumin (1:500;
Chemicon), and rabbit anti--galactosidase antibody (1:8.000;
ICN, Irvine, CA, USA). After washing, slices were incubated
with fluorescein-isothiocyanate- and tetramethylrhodamine
isothiocyanate-labeled secondary antibodies (1:200; Jackson
ImmunoResearch, PA, USA), washed twice in phosphatebuffered saline (PBS) and mounted in Mowiol (Sigma,
Taufkirchen, Germany). In some cases, nuclei were stained
with 4,6-diamidino-2-phenylindoledihydrochloride (DAPI).

Pflugers Arch - Eur J Physiol (2008) 456:10491060

LacZ staining was performed according to standard protocols


[35].
Muscarinic ACh receptor staining was performed on
20 m coronal cryosections of mouse brain tissue that had
been cut at the level of dorsal lateral geniculate nucleus
(dLGN), thaw-mounted onto polysine slide glasses, air dried,
and fixed in 4% PFA/PBS for 10 min. The following
antibodies were used: monoclonal mouse anti-NeuN
(1:1,000; Abcam); rabbit anti-MR1 (1:150; Biogenesis,
UK); rabbit anti-MR3 (1:150; Biogenesis); and secondary
antibodies conjugated with Cy2 or Cy3 (Jackson ImmunoResearch Laboratories). For negative controls, occlusion of
the primary antibody from the staining procedure was routinely performed with no positive immunological signal detected.
Western blotting
Cholate extracts from cerebral cortex, thalamus, and
brainstem of 10-week old mice were prepared. An amount
of 5 g of protein per lane were electrophoresed on 10%
sodium dodecyl sulfate polyacrylamide electrophoresis gels
and blotted onto nitrocellulose membranes. Membranes
were incubated overnight at 4C with rabbit anti-Gq/G11
antibodies (1:500; Santa Cruz Biotechnology, Santa Cruz,
CA, USA), followed by incubation with horseradish
peroxidase-conjugated secondary antibodies (1:2,000;
Sigma-Aldrich, St. Louis, MO, USA). Chemiluminescence
was developed using the ECLplus kit (Amersham Biosciences,
Freiburg, Germany).
Preparation of brain slices for electrophysiology
At postnatal days 1824, mice were deeply anesthetized
using halothane and decapitated as described earlier [26].
Briefly, thalamic slices were prepared as coronal sections
on a vibratome (Series 1000 Classic, St. Louis, USA) in an
ice-chilled solution containing (in mM): sucrose, 200;
PIPES, 20; KCl, 2.5; NaH2PO4, 1.25; MgSO4, 10; CaCl2,
0.5; dextrose, 10; pH 7.35 adjusted with NaOH. Prior to
recording, slices were kept submerged in artificial cerebrospinal fluid (in mM): NaCl, 125; KCl, 2.5; NaH2PO4, 1.25;
NaHCO3, 24; MgSO4, 2; CaCl2, 2; dextrose, 10; pH was
adjusted to 7.35 by bubbling with a mixture of 95% O2 and
5% CO2.
Patch-clamp recordings
=

Wild type and G11 mice were investigated in comparison to Gq =G=


mice. Where applicable, results of the
11
two control groups are stated separately. Whole-cell
recording pipettes (23 M) were prepared from borosilicate glass (GT150T-10, Clark Electromedical Instruments,
Pangbourne, UK) and filled with an intracellular solution

Pflugers Arch - Eur J Physiol (2008) 456:10491060

containing (in mM): K-gluconate, 95; K3-citrate, 20; NaCl,


10; 4-2-hydroxyethyl-1-piperazineethanesulfonic acid
(HEPES), 10; MgCl2, 1; CaCl2, 0.5; 1,2-bis(2-aminophenoxy)ethane-N,N,N,N-tetraacetic acid (BAPTA), 3; Mgadenosine triphosphate, 3; Na2-guanosine triphosphate,
0.5; pH 7.25 with KOH; osmolality 295 mOsm/kg;
~80 nM free Ca2+. In the recording chamber, slices were
continuously superfused with an oxygenated solution
containing (in mM): NaCl, 125; KCl, 2.5; NaH2PO4,
1.25; HEPES, 30; MgSO4, 2; CaCl2, 2; dextrose, 10;
pH 7.35 with HCl; osmolality 305 mOsmol/kg. In one set
of experiments, KCl was increased to 15 mM, while NaCl
was decreased accordingly. Whole-cell patch-clamp recordings were obtained using an EPC-10 amplifier (HEKA
Elektronik, Lamprecht, Germany). Analog signals were
digitized and acquired onto a computer using Pulse
software (HEKA). During current-clamp recordings, the
instantaneous frequency (fi) of action potential generation
was determined by analyzing the first two action potentials
elicited by a depolarizing current pulse.
Recordings were only performed on TC neurons. The
distinction to GABAergic interneurons was based on
previously established physiological and morphological
criteria [30, 33, 34]. Interneurons generally had a bipolar
appearance, with two dendrites arising from a relatively
small soma. By contrast, TC cells generally had a
multipolar appearance with three or more dendrites radiating out from the soma. TC cells always showed robust burst
firing on release from hyperpolarization which was clearly
different from interneurons [2]. Low-threshold Ca2+ spikes
were absent or rather variable in geniculate interneurons.
All cells included in the analysis had a resting membrane
potential more hyperpolarized than 60 mV and revealed
TC neuron-like burst firing from hyperpolarized potentials
(<70 mV) and tonic firing from depolarized potentials
(>60 mV; with potentials in the range of about 60 to
50 mV delayed onset of firing could be observed, which is
another typical TC neuron property). Access resistance was
in the range of 515 M, and series resistance compensation of more than 30% was routinely used. A liquidjunction potential of 8 mV was taken into account.
All results are presented as meanSEM, and differences
stated as significant were tested with p<0.05. Substance
effects were tested for statistical significance using a nonparametric MannWhitney test.
Quantitative real-time polymerase chain reaction
Messenger RNA from control and Gq/G11-deficient
thalami was prepared with the RNeasy Mini Kit (Qiagen,
Hilden, Germany). Following reverse transcription, quantitative real-time polymerase chain reaction (PCR) was
performed using the Platinum SYBR Green qPCR Super-

1051

Mix UGD (Invitrogen, Karlsruhe, Germany) and the ABI


PRISM 7700 Sequence detection system (Applied Biosystems, Lincoln, CA, USA). Primers were as follows: Kir2.3:
5-gccaacctgagcaacaagtc-3/5-agaagaggaggccgaagaag-3;
HCN1: 5- agagatatgcctgctgaccaa-3/5-cattgaaattgtccacc
gaaa-3; HCN2: 5-ctgcgtgaggagattgtgaa-3/5-aagttggggtctg
cattgg-3; HCN3: 5-cctgtgtggacgcactacc-3/5-ccttagctt
taggggcaaca-3; HCN4: 5-ggtggcagtgggagtagtg-3/5atgtcttccgaggcaaagtg-3; TASK-1: 5- tgggtgaacgcatca-aca3/5-acgacacgaaaccgatgag-3; TASK-3: 5-aattcgccggttccttc
tac-3/5-cttgccagcatcagttc-ca-3; m1 AChR: 5-agcagctcaga
gaggtcacagcca-3/5-gggcctcttgactgtatttgggga-3; m2 AChR:
5-tctgcgggacggtgggactg-3/5-cccgggatggaggaggcttct-3; m3
AChR: 5-acagaagcggaggcagaaaacttt-3/5-cttgaaggacagagg
tagagtagc-3; m4 AChR: 5-cagcccgccgca-ctactaagatg-3/5gaagcgccccctggtggag-3; m5 AChR: 5-gaccgccattgccgctttc
tata-3/5-ggtggccttggcttcgtcctct-3. The resulting bands were
normalized against -actin (5-tgacgttgacatccgtaaagac-3/ 5tgctaggagccagagcagtaa-3).
Drugs
Pirenzepine, 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP), and oxotremorine M (OxoM) were obtained
from Tocris/Biotrend Chemikalien GmbH (Cologne, Germany). Muscarine was obtained from Sigma (Deisenhofen,
Germany). Drugs were prepared as stock solutions in
distilled water or directly added to the perfusion medium.
4-DAMP was dissolved in dimethyl sulfoxide (final solvent
concentration 0.2%). Antagonists were applied to the cells
via the extracellular perfusion solution about 10 min before
the muscarinic agonist was added.

Results
Cre recombinase-mediated inactivation of Gq and G11
in the thalamus
To determine the degree of Cre-mediated recombination in
the thalamus of forebrain-specific Gq/G11-double-deficient
mice, we analyzed the expression of Gq/G11 by Western
blotting (Fig. 1a). Thalami from mutant mice showed a
massive reduction of Gq/G11 immunoreactivity, comparable to that observed in the cerebral cortex of these mice.
The residual Gq/G11 immunoreactivity observed in
mutant thalami is probably due to non-recombined interneurons, glial cells, and thalamic blood vessels. In contrast
to the thalamus, no recombination was observed in the
brainstem (Fig. 1a, lower panel). In hippocampus and
cortex, Camkcre4-mediated recombination has been shown
to be restricted to principal neurons [22]. To test whether
this was also true in the thalamus, we performed -

1052

Pflugers Arch - Eur J Physiol (2008) 456:10491060


-NeuN

b
a

-gal

cortex

G11
Gq

40 kD
nRet

nRet

40 kD
brainstem

G11
Gq
Fb- Gq/G11-/-

G11-/-

Gq-/-

Wild-type

40 kD

nRet

dLGN

-gal-FITC /
Parv-TRITC / DAPI

G11
Gq

dLGN

GAD67 /
gal

thalamus

Fig. 1 Cre-mediated recombination in thalamic nuclei. a Western blot


with Gq/G11-specific antibodies performed on cholate extracts from
cerebral cortex, thalamus, and brainstem from wild-type mice,
constitutive Gq-deficient mice Gq= , constitutive G11-deficient
, orforebrain-specific Gq/G11-double-deficient mice
mice G=

q
=
Fb  Gq =G11 . b -galactosidase staining (-gal, right) in mice
carrying the Camkcre4 transgene and a Rosa26LacZ reporter

construct. An adjacent section was stained with an antibody directed


against NeuN. nRet is outlined in black (5 objective). c Double
immunofluorescent staining with antibodies directed against GAD67
(green) and -gal (red) in the dorsal lateral geniculate nucleus (dLGN,
left) or the nRet (right). d Double immunofluorescent staining with
antibodies directed against -gal (green) and parvalbumin (Parv, red)
in the dLGN (nuclear staining with DAPI in blue; 63 objective)

galactosidase staining of brain slices from mice which


carried both the Camkcre4 transgene [21] and a Rosa26LacZ reporter construct [35]. We found strong expression
of -galactosidase in all thalamic nuclei except the reticular
thalamic nucleus (nRet; Fig. 1b). To exclude a recombination of local thalamic interneurons, we performed a double
staining of thalamic sections with antibodies directed
against GAD67, a marker for GABAergic neurons, and galactosidase (Fig. 1c). We did not detect any expression of
-galactosidase in GAD67-positive cells in the dLGN or
the localization of the nucleus reticularis thalami (nRet),
suggesting that thalamic TC cells, but neither local
interneurons nor nucleus reticularis neurons, are recombined in forebrain-specific Gq/G11-double-deficient mice
(Fig. 1c, two visual fields per slice were analyzed, two
slices each, with a total number of 33 and 31 GAD-67positive neurons for dLGN and nRet, respectively).
Furthermore, the ratio of -galactosidase-positive/GAD67negative cells to -galactosidase-negative/GAD67-positive
cells in dLGN was 3.5:1, thereby matching the wellestablished TC neuron-to-interneuron ratio in rodents [2,
11, 33]. Next, the -galactosidase-positive cell population

was further characterized by co-staining with antibodies


directed against -galactosidase and parvalbumin, a marker
for TC neurons [1, 7, 15, 16, 20]. The analysis of 118 cells
(three independent slices were analyzed, each with four
visual fields) revealed that 94.5% of -galactosidasepositive cells were also parvalbumin positive, i.e., they
were relay neurons (Fig. 1d).

Muscarinic
 AChR-mediated signaling
in Gq G11 = mice
The standing outward current (ISO) in TC neurons is carried
in part by TASK-1, TASK-3, and inward rectifier K+
channels which have recently been shown to be inhibited
by activation of mAChR [2628]. To assess the contribution of Gq/G11-mediated signaling, the effect of muscarine
=
was compared in wild type, G11 , and Gq/G11deficient mice. TC neurons were kept at a potential of
28 mV by using the whole-cell configuration of the patchclamp technique. Under steady-state conditions, ISO was
45479 pA (n=10) in wild-type mice and 40625 pA in

Pflugers Arch - Eur J Physiol (2008) 456:10491060

1053

b
1500

-28 mV

1500

800 ms

I (pA)

1000
-138 mV

500
0

-500

Gq/G11-/-

Inor

G11-/-

Vhold = -28 mV

1000

2000

time (s)

3000

decrease in apmlitude (%)

0.5

1.0

q /G

1.5

2.0

300
200
G11-/-

-10
n = 15

-20

100
0

-30

Gq/G11-/-

-100

-40
n=8

-50

Fig. 2 Muscarine effect on ISO. a, b Current responses to hyperpolarizing ramp protocols (the holding potential was ramped from
28 mV to 138 mV in 800 ms; see inset) under control conditions
(black traces) and during application of muscarine (gray traces) in
=
=
G11 (a) and Gq =G11 (b). Note that the outward current
deflection following the hyperpolarizing ramp depends on the
+
activation of voltage-dependent K currents with the early phase
mainly carried by IA [3]. c Normalized mean ISO amplitude vs. time

=

muscarine (50 M)

0.0

11 -/-

11 -/-

1.2

0.8

muscarine

time (s)

time (s)

0.6

500

0.0 0.2 0.4 0.6 0.8 1.0 1.2

1.0

ISO

1000

muscarine

-500

Gq/G11-/-

I (pA)

2000

G11-/-

I (pA)

2000

G11 (n=13; Fig. 2a). Application of muscarine (50 M)


induced a reversible reduction in ISO amplitude in wild-type
=
mice (364%, n=5) and G11 (363%, n=8; Fig. 2a,c,d).
In order to determine the currentvoltage (IV) relationship of
the muscarine-sensitive current, the membrane potential was
ramped in 800 ms from the holding level to 138 mV
(Fig. 2a, inset), and currents recorded during the application
of muscarine and under control conditions were subtracted.
The resulting IV relationship shows the contribution of
inwardly and outwardly rectifying components and reversed
=
at 1065 mV (n=7; G11 ), i.e., close to the expected K+
equilibrium potential (EK =104 mV; Fig. 2e). When the
extracellular K+ concentration was increased to 15 mM, the
muscarine-sensitive current reversed at 552 mV (n=3;
EK =57.mV), which points to a current carried by K+ ions.
In Gq G=
11 , the average ISO amplitude (40924 pA,
n=26) was indistinguishable from control animals. The effect
of muscarine, however, was significantly (p<0.0001) reduced
to 112% (n=16; Fig. 2b,c,d) with the corresponding IV
relationship being weakly inwardly rectifying (Fig. 2e).
The functional consequence of reduced muscarinedependent signaling was investigated under current clamp
conditions. Vrest of TC neurons was not different between

n=5

-120

-80

-40

V (mV)

.
=
=
plot (black squares, G11 ; grey circles, Gq G11 ). Horizontal
bar indicates the period of substance application. d Mean bar graph
representation of the reduction in ISO amplitude induced by muscarine
=
application
(black bar, G11 ; white bar, wild type; grey bar,
.
=
Gq G11 ). e The mean IV relationship of the muscarine-sensitive
current was calculated by graphical subtraction of currents during drug
action from those under control conditions (i.e., control minus
muscarine)

=

wild type (Vrest =711


. mV, n=18), G11 - (Vrest =69
=
1 mV, n=27), and Gq G11 -deficient mice (Vrest =70
1 mV, n=32). Cells were held at a slightly hyperpolarized
membrane potential of Vh =72 mV by DC current injection
in order to sufficiently de-inactivate the T-type Ca2+
conductance. Under these conditions, depolarizing current
steps (100200 pA, 800 ms; Fig. 3a, inset) elicited a burst
of 26 action potentials crowning a low-threshold Ca2+
spike (LTS; Fig. 3a,b). Firing frequencies were very similar
=
in wild type (fi =15611
Hz, n=6), G11 (fi =1355 Hz,
.
=
n=5), and Gq G11 (fi =1635 Hz, n=10) mice. In
=
wild type (232 mV, n=6) and G11 mice (185 mV,
n=5), muscarine induced a strong depolarizing shift of the
membrane potential (Fig. 3c,d). This shift was accompanied
by a switch from burst to tonic firing in all cells tested (wild
=
type: fi = 26 4 Hz, n = 6; G11 : fi = 34 5 Hz, n = 5;
Fig. 3a). In double-deficient mice, the muscarine-dependent
depolarizing shift (Fig. 3c) was significantly (p<0.002)
smaller (21 mV, n=9; Fig. 3d) and was still associated
with burst firing (fi =1527 Hz, n=8; Fig. 3b; two cells
revealed only a single LTS-induced action potential). In
summary, these data indicate that mAChR-signaling in TC
neurons critically depends on the presence of Gq.

1054

Pflugers Arch - Eur J Physiol (2008) 456:10491060

muscarine

control

40

control

muscarine

20

20
0

G
11-/-

V (mV)

V (mV)

40

-20
-40

G
q/G
11-/-

-20
-40
-60

-60

-80

-80

500 ms

500 ms

d
-20

muscarine (50 M)

depolarization (mV)

-30

V (mV)

-40

G
11-/-

-50
-60
-70

G
q/G
11-/-

-80

1000

2000

3000

time (s)
Fig. 3 Muscarine effect on firing properties. Functional consequences
of muscarine administration were tested under whole-cell current=
=
clamp conditions in G11 (a) and Gq =G11 (b). Cells of both
genotypes were held at a potential of approximately 72 mV by DC
current injection. This holding current was not changed during the
course of the experiment. Cells were challenged using 100200 pA
depolarizing current pulses (800 ms duration; see inset in a). Under
control conditions, depolarizing current pulses starting from hyperpolarized membrane potentials evoked typical burst firing (left traces).


Muscarinic receptor subtypes involved in Gq G11
signaling
In an effort to distinguish between receptor subtypes, we
tested the muscarinic agonist OxoM with some preference
=
to m1AChR [12, 37]. In G11 , application of OxoM
(10 M) induced a 292% (n=9; Fig. 4a,c) decrease in ISO
amplitudes under voltage-clamp conditions and a 122 mV
depolarization (from Vh) of the membrane potential under
current clamp conditions (n=7,
. Fig. 4d). Both effects were
significantly reduced in Gq G=
(ISO reduction: 13
11
3%, n=8, p<0.001; Fig. 4b,c; depolarization: 41 mV, n=
5; Fig. 4d, p<0.03). In all. control mice (Fig. 4e, upper
=
panel) but none of the Gq G11 (Fig. 4e, lower panel),
the OxoM-induced depolarization was accompanied by a
switch to tonic firing.
To further identify mAChR subtypes, we used antagonists with preferred affinity to one of the known Gqcoupled receptors. Application of the m1AChR-affinitive
antagonist pirenzepine (10 M) [14] induced a non=
significant increase in ISO amplitude in G11 (32%,

30

n=6
n=5

20

10

n=9
0

G
11-/-

WT

G
q/G
11-/-

Addition of muscarine (right traces) led to a marked depolarization of


the membrane potential with a consecutive shift of the activity mode
=
from burst to tonic firing in control mice (a) but not in Gq =G11
(b). c Voltage vs. time plot of two individual cells (black squares,
=
control animal; grey circles, Gq =G11 ). Horizontal bar indicates
the period of substance application. d Mean bar graph representation
of the depolarization induced by muscarine application (black bar,
=
=
G11 ; white bar, wild type; grey bar, Gq =G11 )

.
=
n =7) and Gq G11 (5 3%, n =5) mice, possibly
indicating the activity of ambient ACh or basal receptor
activity (Fig. 5a,d). Subsequent application of muscarine in
=
G11 mice induced a reversible reduction in ISO amplitude which was significantly (p<0.0002) smaller (183%,
n=7) compared to the control effect (34%, see above).
When muscarine
was applied in the presence of pirenzepine
.
=
in Gq G11 mice, the reduction in ISO amplitude was
62% (n=5; Fig. 5a,b) which was significantly (p<0.02)
=
smaller compared to G11 . The IV relationship of the
muscarine-sensitive current in the presence of pirenzepine
=
in G11 revealed inwardly and outwardly rectifying
components
and reversed at 1041 mV (n=7, Fig. 5c).
.
=
In Gq G11 mice, the small rather inwardly rectifying
muscarine-sensitive current in the presence of pirenzepine
reversed at 1032 mV (n=5; Fig. 5c).
Next, we used the m3AChR-affinitive antagonist 4-DAMP
[29]. Application of 4-DAMP (1 M) induced an increase
in
.
=
=
ISO amplitude in G11 (5 4%, n = 6) and Gq G11
(71%, n=3) mice, which was significant in the latter case
=
(Fig. 5b,d). Subsequent application of muscarine in G11

Pflugers Arch - Eur J Physiol (2008) 456:10491060

-500

I (pA)

Vhold = -28 mV

500

0.2

0.4 0.6 0.8


time (s)

1.0

-500

1.2

d
G
q/G
11-/-

1.0

0.8
11-/G

0.6

Vhold = -28 mV

1000
2000
time (s)

induced a reversible reduction in ISO amplitude (242%,


n=6) which was significantly (p<0.002) smaller compared
to the control effect (34%, see above). When .
muscarine was
applied in the presence of 4-DAMP in Gq G=
mice,
11
the reduction in ISO amplitude was 72% (n=3; Fig. 5b,d),
which was significantly (p<0.0003) smaller compared to
=
G11 . The IV relationship of the muscarine-sensitive
=
current in the presence of 4-DAMP in G11 mice revealed
inward and outward rectification
and reversed at 1052 mV
.
=
(n=6, Fig. 5e). In Gq G11 mice, the small rather inwardly rectifying muscarine-sensitive current in the presence
of 4-DAMP reversed at 1032 mV (n=3; Fig. 5e). When
=
pirenzepine and 4-DAMP were co-applied to cells from G11
mice, ISO increased by 53%, and subsequent application of
muscarine induced reduction in ISO of 63% which
. was
undistinguishable from the muscarine effect in Gq G=
11
mice (n=5; Fig. 4d).
To provide further evidence that m1AChR and m3AChR
are functionally expressed in TC neurons, we performed
antibody labeling. Application of m1AChR- (Fig. 6a, left
images) and m3AChR-specific (Fig. 6b, left images) antibodies revealed clear soma staining in the .dLGN of wild
=
type (Fig. 6a,b, upper rows) and Gq G11 mice
(Fig. 6a,b, lower rows) which was co-localized (Fig. 6a,b,
right merged images) with the neuron-specific cell marker
NeuN in all inspected cells (Fig. 6a,b, middle images; a
total of 100 cells were inspected; three independent slices
were analyzed, each with three visual fields).

e
15

OxoM (10 M)

0.0

n=7

V (mV)

c
1.2

OxoM

OxoM

0.0

G
q/G
11-/-

1000

depolarization (mV)

I (pA)

500

G
11-/-

1000

10

0.2

co n

KO

G
q/
G
11-/-

0.6 0.8
time (s)

1.0

1.2

G
11-/-

-40

40

G
11-/-

0.4

40

-80

n=5

V (mV)

I (pA)

Fig. 4 OxoM effect on ISO and


firing properties. a, b Current
response to ramp protocols under control conditions (black
traces) and during application of
=
OxoM (grey traces) in G11
=
(a) and Gq =G11 (b). The
hyperpolarizing ramp protocol
was used as described above. c
Mean ISO amplitude vs. time
=
plot (black squares, G11 ;
=
gray circles, Gq =G11 ). The
horizontal bar indicates the period of substance application. d
Mean bar graph representation
of the depolarization induced by
OxoM application. e OxoM ef=
fect on firing pattern in.G11
(upper panel) and Gq G=
11
(lower panel; scale bar represents 500 ms)

1055

control

OxoM

G
q/G
11-/-

-40
-80

These data indicate that cholinergic signaling via


mAChR in TC neurons is mediated by m1AChR and
m3AChR, with both subtypes roughly equally contributing
to the overall response.

Possible compensatory mechanisms for Gq G11 loss
To explore the possibility of compensatory changes in
response to Gq/G11 deficiency, we performed a quantitative PCR analysis on the transcripts of the known
mAChR subtypes (m1AChRm5AChR) and a number of
putative effector channels (TASK-1, TASK-3, HCN14,
and KIR2.3; Fig. 7). We did not find differences in the
relative expression of these transcripts between the two
genotypes (Fig. 7).
To provide evidence that modulation of ISO by alternative mechanisms is intact in Gq/G11-deficient mice, we
used TASK channel blockers in voltage- and current-clamp
experiments. The general TASK channel blocker bupivacaine (20 M) and the preferential TASK-1 channel blocker
=
anandamide (10 M) reduced ISO at 28 mV in G11
mice (bupivacaine:
435%, n=8; anandamide: 233, n=5)
.
=
and in Gq G11 mice (bupivacaine: 375%, n=3;
anandamide: 195, n=3; data not shown) in an undistinguishable manner. Under current-clamp conditions, bupivacaine induced an 183 mV depolarization of the membrane
potential from holding potential (72 mV), which was
associated with a switch from burst (fi =12117 Hz) to

1056

Pflugers Arch - Eur J Physiol (2008) 456:10491060

pirenzepine ( 10 M)

1.2

1.2

muscarine (50 M)

1.1
Gq/G11-/-

0.9
0.8

G11-/-

0.7

Gq/G11-/-

G11-/-

Vhold = -28 mV

1000

2000

3000

Vhold = -28 mV

0.6

1000

2000

pir

10

P ir

musc /
pir

P i rMu sc

-50
-150

3000

4-DAMP

DA M P

musc /
4-DAMP

D A MP Mu sc

pir /
4-DAMP

-100

-50

V (mV)

time (s)
musc /
pir /
4-DAMP

P i rDA MP

P i rDA MP Mu sc

-10

in 4-DAMP

100

I (pA)

d
change in amplitude (%)

50

0.7

time (s)

G11-/-

I (pA)

Inor

Inor

0.8

100

Gq/G11-/-

1.0

0.9

in pirenzepine

muscarine (50 M)

1.1

1.0

0.6

4-DAMP (1 M)

G11-/-

50

Gq/G11-/-

0
Gq/G11-/-

-20

G11-/-

-50
-150

-30

-100

-50

V (mV)

Fig. 5 m1AChR- and m3AChR-dependent signaling in TC neurons.


a, b Normalized ISO mean amplitude vs. time plots (black squares,
=
=
G11 ; gray circles, Gq =G11 ). The horizontal bar indicates the
period of substance application. c, e Mean IV relationships of
muscarine-sensitive currents (blocker as indicated; black squares,

.
=
tonic (fi =4922 Hz) firing in Gq G11 mice (n=3;
data not shown).
These data argue against any substantial compensatory
mechanisms in TC neurons of Gq =G=
mice at the level
11
of effector channels and mAChR and demonstrate that the
Gq-independent modulation of TASK channels is intact.

=

=

G11 ; gray circles, Gq =G11 ). d Mean bar graph representation

of the change in ISO current amplitude under experimental conditions


=
as indicated (musc
muscarine; pir pirenzepine; black bar, G11 ;
.
=
gray bar, Gq G11 )

to two important novel aspects of the physiology of the


thalamocortical system by demonstrating that the depolarization of TC neurons induced by the release of ACh from
ascending brainstem fibers is governed roughly equally by
both m1AChR and m3AChR and is transduced by Gq but
not by G11.
mAChR-dependent signaling in the thalamus

Discussion
The results of the present study can be summarized as
=
follows: (1) Forebrain-specific Gq =G11 -deficient mice
show a strong reduction of Gq/G11 protein levels in the
thalamus. Cre-mediated recombination is observed only in
TC neurons not in local GABAergic interneurons or neurons
of the nRET. (2) The actions of ACh in TC neurons depend
on intracellular signaling cascades mainly activated by
m1AChR and m3AChR coupled to Gq-proteins. The effects
of muscarine and OxoM under current clamp conditions
=
were significantly reduced in Gq =G11 . As a consequence, stimulation of mAChR could not induce a shift from
burst to tonic firing of action potentials. (3) The steady-state
parameters
Vrest and ISO amplitude were unchanged in
.
=
Gq G11 with no evidence for compensatory mechanisms on the level of effector channels. (4) These data point

The flow of sensory information from the sensory organs to


the cerebral cortex via the thalamus is highly regulated by
inputs from the ascending activating brainstem system,
releasing ACh, 5-HT, and noradrenalin [36]. These neuromodulators influence the state of the thalamus by altering
specific ionic channels through activation of G-proteincoupled membrane receptors [24]. During wakefulness and
REM sleep, the release of ACh from mesopontine cholinergic neurons is permanently increased, resulting in
sustained depolarization of TC neurons and the appearance
of tonic activity and high-frequency (2060 Hz) oscillations
[36]. Cholinergic inputs to dLGN TC neurons mostly contact
proximal dendritic shafts and retinothalamic synapses and
are thus in close spatial relationship to visual inputs [38].
dLGN TC neurons from several species in vitro reveal an
ACh-induced change in firing mode due to a depolarization

Pflugers Arch - Eur J Physiol (2008) 456:10491060


Fig. 6 Immunological detection
of m1AChR and m3AChR in
dLGN. Specific antibodies
against m1AChR (a, left images)
or m3AChR (b, left images)
were co-incubated with an antibody directed against NeuN
(middle images) in wild type
(upper rows in a and b) and
=
Gq =G11 (lower rows in a
and b). Merged images (right
images) indicate co-localization.
Scale bars 20 m

1057

m1AChR

NeuN

merge

m3AChR

NeuN

merge

WT

Gq/G11-/-

WT

Gq/G11-/-

application of ACh to extracellularly recorded cat dLGN TC


neurons results in marked excitation of both physiologically
identified X and Y cells [10].
The
. strong reduction of muscarinic depolarization in
=
Gq G11 is in agreement with the presence of
m1AChRm5AChR transcripts in the thalamus [39] and

Gq/G11-/G11-/-

3
2
1

5
m

4
m

3
m

2
m

1
m

SK

1
TA

N4

SK
TA

HC

N3
HC

N2
HC

N1
HC

r2

.3

0
Ki

Fig. 7 Relative expression of


different effector channel
subtypes as well as muscarinic
receptors in G11- (black) and
Gq/G11-deficient (gray) thalami as determined by quantitative real-time PCR. Data were
normalized to -actin expression. (m15 muscarinic
acetylcholine receptor
subtypes 15)

relative expression (normalized to -actin)

of the neuron through activation of muscarinic and, in


addition, nicotinic receptors [23, 25]. The nicotinic response
typically results in rapid depolarization mediated by an
increase in membrane cation conductance, while the muscarinic response is slower in onset and longer lasting in duration
and results from a block of a resting K+ conductance. In vivo

1058

the dLGN (this study) although a detailed study of mAChR


expression in thalamic cell types is still missing. In TC
neurons, the inhibition of TASK-1 and TASK-3 channels is
the molecular correlate of IKL modulation by ACh [24, 26,
27]. Based on the effects of the antagonists pirenzepine [14]
and 4-DAMP [29], it is concluded that m1AChR and
m3AChR largely mediate the effect of muscarine on ISO and
Vrest. It should be noted that despite preferential actions on
one receptor subtype, OxoM (m1AChR), pirenzepine
(m1AChR), and 4-DAMP (m3AChR) are not ideally specific
and may bind to other mAChRs as well (http://www.iuphardb.org/index.jsp). Therefore, taking the results of expression
analyses and the pharmacological experiments into account,
it may be concluded that mainly m1AChR and m3AChR
mediate the effect of ACh in TC neurons.
Furthermore, the findings presented here are in agreement with both a model of direct channel modulation by
Gq [5] as well as a Gq/G11-mediated depletion of PIP2 and
subsequent closure of TASK channels in native neurons [4,
18]. To which extent the coupling of Gq-coupled receptors
to TASK channels represents a general scheme in native
neurons remains to be established, since the suppression of
leak currents appears to be independent from Gq and
G11 in hippocampal neurons [17].
Specificity and plasticity of G-proteins and TC neurons
The high structural homology, overlapping expression
patterns, and identical coupling properties to downstream
effectors of Gq and G11 have raised the question of their
functional specificity [9]. The muscarine-sensitive compo=
nent in G11 TC neurons is very similar to the analogous
current in rats [27], wild-type BL6 mice (this study), and
other control mice [26]. Therefore, our findings suggest that
the Gq subunit is responsible for TASK and Kir channel
regulation in TC neurons. A similar functional specificity of
Gq and G11 has been found in hippocampal neurons
where Gq, but not G11, mediates specifically the action of
cholinergic and glutamatergic agonists on small-conductance
Ca2+-activated K+ channels [17].
The contribution of inward rectifier channels to the
muscarine-sensitive component and to ISO was estimated to
be around 30 pA at membrane potentials positive to EK
[28]. In addition, TASK channels carry some inward
current at potentials negative to EK [26, 27]. Therefore, it
is feasible to conclude that most of the loss of outward
current and some of the loss of inward current can be
attributed to TASK channels. Since no changes in Kir2.3,
TASK-1,
and TASK-3 expression levels were found in
.
=
Gq G11 , the ratio between currents through TASK and
inward rectifier channels can be expected to be stable in
both genotypes. As Vrest of TC neurons critically depends
on the functional antagonism between TASK and HCN

Pflugers Arch - Eur J Physiol (2008) 456:10491060

channels [6, 28], we quantitatively analyzed HCN channel


=
transcripts. No differences were observed between G11
=
and Gq =G11 , a finding which is in good agreement with
unchanged steady-state (Vrest, ISO) properties in both
genotypes. In this respect, it should be noted that there is
no clear linear and stable relationship between mRNA
levels and protein expression. Modulation of translation,
posttranslational changes, and distribution between the
different cellular pools may lead to functional differences
despite unchanged mRNA levels.
While Gq and G11 have very similar effector-coupling
properties and may substitute each other [31], the remaining
effect of receptor agonists in Gq =G=
raise the
11
questions of its molecular basis and possible compensatory
involvement of other G-protein families. Since muscarine is
an unspecific agonist of mAChR, all subtypes expressed in
TC neurons will be activated. While m1AChR, m3AChR,
and m5AChR are coupled to Gq-proteins, the typical
coupling of m2AChR and m4AChR is to Gi/o-proteins
[42]. The mean amplitude of the muscarine-sensitive
current at 28 mV in control mice (210 pA) in the present
study was matched by the summed muscarine-sensitive
currents in the presence of m1AChR (100 pA) and
m3AChR (110 pA) blockers. The
. residual muscarine=
sensitive current at 28 mV in Gq G11 in the presence
or absence of mAChR antagonists is roughly 30 pA and
reveals inward rectification. In rat TC neurons, the effect of
muscarine on ISO is nearly completely blocked in the
presence of Ba2+ (150 M; Budde, unpublished results). It
is therefore concluded that this component is carried by
inward rectifier channels governed by m2AChR and
m4AChR coupled to Gi/o-proteins, thereby representing a
typical coupling scheme [42]. Due to the lack of specific
antagonists, the contribution of m5AChR could not be
determined, allowing no further clarification of its elusive
functional role [8].
Findings from hippocampal neurons of Go-deficient
mice suggest that in the absence of this abundant G-protein,
ion channels may be readily regulated by G-proteins with
different properties [13]. However, this does not seem to be
the case with TC neurons for the following reasons: (1) The
severe epileptic phenotype [44], the reduced life span, and
the unchanged expression level of mAChR and effector
channels argue against large-scale compensatory mechanisms in TC neurons. (2) A lack of plastic compensation in
TC neurons has been noted before in mice deficient for
HCN2 [19] and TASK-1 [26]. Future studies have to show
the coupling of mAChR receptors to parallel pathways,
activated by -subunits, by G-proteins not belonging to
the Gq/G11-family and/or by G-protein-independent mechanisms in TC neurons as well as G-protein-dependent
signaling in general in interneurons to further gain understanding of receptor signaling in the thalamus.

Pflugers Arch - Eur J Physiol (2008) 456:10491060

Together with earlier observations [26, 27], the current


findings unravel a highly specific pathway in rodent TC
neurons where TASK-1 and TASK-3 channels are modulated
by activation of m1AChR and m3AChR via Gq-proteins. The
activity of this pathway is necessary to provide a sufficient
depolarization for switching the activity mode of TC neurons
from sleep-associated burst firing to tonic firing seen during
wakefulness and REM sleep.
Acknowledgements This work was supported by DFG (BU 1019/71; PA 336/17-1), Innovative Medical Research Fund (IMF) of the
University of Muenster Medical School, and IZKF Bud3/005/07. T.
Broicher was a fellow of the Boehringer Ingelheim Foundation.
Thanks are due to A. Jahn, E. Nass, A. Markovic, and R. Ziegler for
excellent technical assistance.

References
1. Alcantara S, Ferrer I (1994) Postnatal development of parvalbumin
immunoreactivity in the cerebral cortex of the cat. J Comp Neurol
348:133149
2. Broicher T, Kanyshkova T, Landgraf P, Rankovic V, Meuth P,
Meuth SG, Pape HC, Budde T (2007) Specific expression of lowvoltage-activated calcium channel isoforms and splice variants in
thalamic local circuit interneurons. Mol Cell Neurosci 36:132145
3. Budde T, Mager R, Pape H-C (1992) Different types of potassium
outward current in relay neurons acutely isolated from the rat
lateral geniculate nucleus. Eur J Neurosci 4:708722
4. Chemin J, Girard C, Duprat F, Lesage F, Romey G, Lazdunski M
(2003) Mechanisms underlying excitatory effects of group I
metabotropic glutamate receptors via inhibition of 2P domain K+
channels. Embo J 22:54035411
5. Chen X, Talley EM, Patel N, Gomis A, McIntire WE, Dong B,
Viana F, Garrison JC, Bayliss DA (2006) Inhibition of a
background potassium channel by Gq protein {alpha}-subunits.
Proc Natl Acad Sci U S A 103:34223427
6. Day M, Carr DB, Ulrich S, Ilijic E, Tkatch T, Surmeier DJ (2005)
Dendritic excitability of mouse frontal cortex pyramidal neurons
is shaped by the interaction among HCN, Kir2, and Kleak
channels. J Neurosci 25:87768787
7. Del Rio MR, DeFelipe J (1994) A study of SMI 32-stained
pyramidal cells, parvalbumin-immunoreactive chandelier cells,
and presumptive thalamocortical axons in the human temporal
neocortex. J Comp Neurol 342:389408
8. Eglen RM, Nahorski SR (2000) The muscarinic M(5) receptor: a
silent or emerging subtype? Br J Pharmacol 130:1321
9. Exton JH (1996) Regulation of phosphoinositide phospholipases
by hormones, neurotransmitters, and other agonists linked to G
proteins. Annu Rev Pharmacol Toxicol 36:481509
10. Eysel UT, Pape H-C, Van Schayck R (1986) Exitatory and
differential disinhibitory actions of acetylcholine in the lateral
geniculate nucleus of the cat. J Physiol (Lond) 370:233254
11. Gabbott PL, Bacon SJ (1994) Two types of interneuron in the
dorsal lateral geniculate nucleus of the rat: a combined NADPH
diaphorase histochemical and GABA immunocytochemical study.
J Comp Neurol 350:281301
12. Ge ZD, Zhang DX, Chen YF, Yi FX, Zou AP, Campbell WB, Li
PL (2003) Cyclic ADP-ribose contributes to contraction and Ca2+
release by M1 muscarinic receptor activation in coronary arterial
smooth muscle. J Vasc Res 40:2836

1059
13. Greif GJ, Sodickson DL, Bean BP, Neer EJ, Mende U (2000)
Altered regulation of potassium and calcium channels by GABA
(B) and adenosine receptors in hippocampal neurons from mice
lacking Galpha(o). J Neurophysiol 83:10101018
14. Hammer R, Berrie CP, Birdsall NJ, Burgen AS, Hulme EC (1980)
Pirenzepine distinguishes between different subclasses of muscarinic receptors. Nature 283:9092
15. Hashikawa T, Rausell E, Molinari M, Jones EG (1991) Parvalbumin- and calbindin-containing neurons in the monkey medial
geniculate complex: differential distribution and cortical layer
specific projections. Brain Res 544:335341
16. Jones EG, Hendry SH (1989) Differential calcium binding protein
immunoreactivity distinguishes classes of relay neurons in
monkey thalamic nuclei. Eur J Neurosci 1:222246
17. Krause M, Offermanns S, Stocker M, Pedarzani P (2002)
Functional specificity of G alpha q and G alpha 11 in the
cholinergic and glutamatergic modulation of potassium currents
and excitability in hippocampal neurons. J Neurosci 22:666673
18. Lopes CM, Rohacs T, Czirjak G, Balla T, Enyedi P, Logothetis
DE (2005) PIP2 hydrolysis underlies agonist-induced inhibition
and regulates voltage gating of two-pore domain K+ channels. J
Physiol 564:117129
19. Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff K,
Langebartels A, Wotjak C, Munsch T, Zong X, Feil S, Feil R,
Lancel M, Chien KR, Konnerth A, Pape HC, Biel M, Hofmann F
(2003) Absence epilepsy and sinus dysrhythmia in mice lacking
the pacemaker channel HCN2. Embo J 22:216224
20. Magnusson A, Dahlfors G, Blomqvist A (1996) Differential
distribution of calcium-binding proteins in the dorsal column
nuclei of rats: a combined immunohistochemical and retrograde
tract tracing study. Neuroscience 73:497508
21. Mantamadiotis T, Lemberger T, Bleckmann SC, Kern H, Kretz O,
Martin Villalba A, Tronche F, Kellendonk C, Gau D, Kapfhammer
J, Otto C, Schmid W, Schutz G (2002) Disruption of CREB
function in brain leads to neurodegeneration. Nat Genet 31:4754
22. Marsicano G, Goodenough S, Monory K, Hermann H, Eder M,
Cannich A, Azad SC, Cascio MG, Gutierrez SO, van der Stelt M,
Lopez-Rodriguez ML, Casanova E, Schutz G, Zieglgansberger W,
Di Marzo V, Behl C, Lutz B (2003) CB1 cannabinoid receptors
and on-demand defense against excitotoxicity. Science 302:8488
23. McCormick DA (1992) Cellular mechanisms underlying cholinergic
and noradrenergic modulation of neuronal firing mode in the cat and
guinea pig dorsal lateral geniculate nucleus. J Neurosci 12:278289
24. McCormick DA (1992) Neurotransmitter actions in the thalamus
and cerebral cortex and their role in neuromodulation of
thalamocortical activity. Prog Neurobiol 39:337388
25. McCormick DA, Prince DA (1987) Actions of acetylcholine in
the guinea-pig and cat medial and lateral geniculate nuclei, in
vitro. J Physiol 392:147165
26. Meuth SG, Aller MI, Munsch T, Schuhmacher T, Seidenbecher T,
Kleinschnitz C, Pape HC, Wiendl H, Wisden W, Budde T (2006)
The contribution of TASK-1-containing channels to the function
of dorsal lateral geniculate thalamocortical relay neurons. Mol
Pharmacol 69:14681476
27. Meuth SG, Budde T, Kanyshkova T, Broicher T, Munsch T, Pape
H-C (2003) Contribution of TWIK-related acid-sensitive K+
channel 1 (TASK1) and TASK3 channels to the control of activity
modes in thalamocortical neurons. J Neurosci 23:64606469
28. Meuth SG, Kanyshkova T, Meuth P, Landgraf P, Munsch T,
Ludwig A, Hofmann F, Pape HC, Budde T (2006) The membrane
resting potential of thalamocortical relay neurons is shaped by the
interaction among TASK3 and HCN2 channels. J Neurophysiol
96:15171529
29. Michel AD, Stefanich E, Whiting RL (1989) Direct labeling of rat
M3-muscarinic receptors by [3H]4DAMP. Eur J Pharmacol
166:459466

1060
30. Munsch T, Budde T, Pape H-C (1997) Voltage-activated intracellular calcium transients in thalamic relay cells and interneurons.
Neuroreport 8:24112418
31. Offermanns S (1999) New insights into the in vivo function of
heterotrimeric G-proteins through gene deletion studies. Naunyn
Schmiedebergs Arch Pharmacol 360:513
32. Offermanns S, Zhao LP, Gohla A, Sarosi I, Simon MI, Wilkie TM
(1998) Embryonic cardiomyocyte hypoplasia and craniofacial
defects in G alpha q/G alpha 11-mutant mice. Embo J 17:4304
4312
33. Pape H-C, Budde T, Mager R, Kisvarday Z (1994) Prevention of
Ca2+-mediated action potentials in GABAergic local circuit
neurons of the thalamus by a transient K+ current. J Physiol
(Lond) 478:403422
34. Pape HC, McCormick DA (1995) Electrophysiological and
pharmacological properties of interneurons in the cat dorsal lateral
geniculate nucleus. Neuroscience 68:11051125
35. Soriano P (1999) Generalized lacZ expression with the ROSA26
Cre reporter strain. Nat Genet 21:7071
36. Steriade M, Jones EG, McCormick DA (1997) Thalamus.
Elsevier, Amsterdam
37. Suh BC, Horowitz LF, Hirdes W, Mackie K, Hille B (2004)
Regulation of KCNQ2/KCNQ3 current by G protein cycling: the
kinetics of receptor-mediated signaling by Gq. J Gen Physiol
123:663683

Pflugers Arch - Eur J Physiol (2008) 456:10491060


38. Uhlrich DJ, Tamamaki N, Murphy PC, Sherman SM (1995)
Effects of brain stem parabrachial activation on receptive field
properties of cells in the cat's lateral geniculate nucleus. J
Neurophysiol 73:24282447
39. Wei J, Walton EA, Milici A, Buccafusco JJ (1994) m1-m5
muscarinic receptor distribution in rat CNS by RT-PCR and
HPLC. J Neurochem 63:815821
40. Wettschureck N, Moers A, Hamalainen T, Lemberger T, Schutz G,
Offermanns S (2004) Heterotrimeric G proteins of the Gq/11
family are crucial for the induction of maternal behavior in mice.
Mol Cell Biol 24:80488054
41. Wettschureck N, Moers A, Offermanns S (2004) Mouse models to
study G-protein-mediated signaling. Pharmacol Ther 101:7589
42. Wettschureck N, Offermanns S (2005) Mammalian G proteins and
their cell type specific functions. Physiol Rev 85:11591204
43. Wettschureck N, Rutten H, Zywietz A, Gehring D, Wilkie TM,
Chen J, Chien KR, Offermanns S (2001) Absence of pressure
overload induced myocardial hypertrophy after conditional inactivation of Galphaq/Galpha11 in cardiomyocytes. Nat Med
7:12361240
44. Wettschureck N, van der Stelt M, Tsubokawa H, Krestel H, Moers A,
Petrosino S, Schutz G, Di Marzo V, Offermanns S (2006) Forebrainspecific inactivation of Gq/G11 family G proteins results in agedependent epilepsy and impaired endocannabinoid formation. Mol
Cell Biol 26:58885894

Original Research Article

published: 25 October 2010


doi: 10.3389/fncel.2010.00132

CELLULAR NEUROSCIENCE

Activity modes in thalamocortical relay neurons are


modulated by Gq/G11 family G-proteins serotonergic and
glutamatergic signaling
Philippe Coulon1, Tatyana Kanyshkova1, Tilman Broicher 1, Thomas Munsch 2, Nina Wettschureck 3,
Thomas Seidenbecher1, Sven G. Meuth1,4, Stefan Offermanns 3, Hans-Christian Pape1 and Thomas Budde1*
Institut fr Physiologie I, Westflische Wilhelms-Universitt Mnster, Mnster, Germany
Institut fr Physiologie, Otto-von-Guericke-Universitt, Magdeburg, Germany
3
Max-Planck-Institut fr Herz- und Lungenforschung, Abteilung Pharmakologie, Bad Nauheim, Germany
4
Department of Neurology, Inflammatory Disorders of the Nervous System and Neurooncology, Westflische Wilhelms-Universitt Mnster, Mnster, Germany
1
2

Edited by:
Barry W. Connors, Brown University,
USA
Reviewed by:
Charles Cox, University of Illinois at
Urbana-Champiagn, USA
Barry W. Connors, Brown University,
USA
*Correspondence:
Thomas Budde, Institut fr Physiologie
I, Westflische Wilhelms-Universitt
Mnster, Robert-Koch-Street 27a,
48149 Mnster, Germany.
e-mail: tbudde@uni-muenster.de
Present address:
Tilman Broicher, Department of
Bioengineering, University of Utah, 20
South 2030 East, Salt Lake City, UT
84112, USA

The authors Philippe Coulon, Tatyana


Kanyshkova, and Tilman Broicher
contributed equally to this work.

In thalamocortical relay (TC) neurons, G-protein-coupled receptors play an important part in


the control of activity modes. A conditional Gq knockout on the background of a constitutive
G11 knockout (Gq/G11/) was used to determine the contribution of Gq/G11 family G-proteins
to metabotropic serotonin (5-HT) and glutamate (Glu) function in the dorsal part of the
lateral geniculate nucleus (dLGN). In control mice, current clamp recordings showed that
-m-5-HT induced a depolarization of Vrest which was sufficient to suppress burst firing. This
depolarization was concentration-dependent (100M: +61mV, n=10; 200M: +101mV,
n=7) and had a conditioning effect on the activation of other Gq-mediated pathways. The
depolarization was significantly reduced in Gq/G11/ (100M: 31mV, n=11; 200M:
51mV, n=6) and was apparently insufficient to suppress burst firing. Activating Gq-coupled
muscarinic receptors affected the magnitude of -m-5-HT-induced effects in a reciprocal
manner. Furthermore, the depolarizing effect of mGluR1 agonists was significantly reduced in
Gq/G11/ mice. Immunohistochemical stainings revealed binding of 5-HT2CR- and mGluR1-,
but not of 5-HT2AR-specific antibodies in the dLGN of Gq/G11/ mice. In conclusion, these
findings demonstrate that transmitters of ascending brainstem fibers and corticofugal fibers
both signal via a central element in the form of Gq/G11-mediated pathways to control activity
modes in the TC system.
Keywords: thalamic function, muscarinic receptor, metabotropic glutamate receptor, serotonin receptor, gene knockout,
G-protein

Introduction
The thalamocortical (TC) network takes up two states of activity:
slow and highly synchronized oscillatory burst activity during slow
wave sleep, and tonic generation of action potentials alongside fast
oscillations during mental alertness and REM sleep. Slow oscillatory activity has a frequency of <15Hz, while fast oscillations
occur at 40 Hz (Steriade et al., 1997). Chemically-coded projections from the brainstem activate the forebrain by releasing
acetylcholine (ACh), noradrenalin (NA), and serotonin (5-HT).
These neurotransmitters mainly act on G-protein-coupled membrane receptors (McCormick, 1992a). In a similar way, the arousing
action of glutamate, released from corticothalamic axons, is mediated by metabotropic glutamate receptors (mGluR) (Salt, 2002).
A common action of these neurotransmitters is a depolarizing
shift of the membrane potential of TC neurons, causing rhythmic bursting to cease and tonic activity to commence. Membrane
depolarization is caused to a large part by the inhibition of a leak
K+ conductance (IKL), the molecular correlate of which are the
two pore-domain K+ (K2P) channels TASK-1 and TASK-3 (Meuth
etal., 2003, 2006). The activation of muscarinic ACh receptors
(mAChR) and mGluR1 coupled to Gq/G11 inhibits TASK-1 and

Frontiers in Cellular Neuroscience

TASK-3 (Exton, 1996; Chemin et al., 2003; Lopes et al., 2005;


Chen etal., 2006) in a similar fashion as IKL (McCormick, 1992a;
Salt,2002).
Recently, it has been shown that the reduction of the standing outward current (ISO), and the depolarization induced by the
activation of m1AChR and m3AChR, depends on the action of Gq
in TC neurons (Broicher et al., 2008b). Acetylcholine is one of
several neurotransmitters that play a major role in the modulation
of thalamic states of activity. Thus, we were interested in other
neurotransmitter systems depending on the presence of Gq and
involved in Gq-mediated signaling in TC neurons. We used the
same genetic strategy to investigate the role of mGluR and metabotropic 5-HT receptors (5-HTR) in conditional forebrain-specific
Gq/G11-double-deficient mice (Gq/G11/) (Wettschureck
etal., 2004b).

Materials and Methods


Mice

All animal handling and procedures were approved by the local


authorities. Generation of forebrain-specific Gq/G11 deficient
mice and genotyping of the gnaqflox allele, of gna11-wildtype

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 1

Coulon etal.

Gq/11 signaling in the thalamus

and -knockout alleles, and of the Cre-transgene has been


described previously (Wettschureck et al., 2001, 2004b). The
genetic background was predominantly C57BL6/N (4th generation backcross). As controls, littermates with the genotype
Camkcre4/; gnaqfl/fl; gna11/ (named G11/ in the following)
were used. Former studies have shown that no significant differences exist between Camkcre4/; gnaqfl/fl; gna11/ and nonlittermate C57BL6/N mice (Offermanns etal., 1998; Broicher
et al., 2008b). For this reason, we have also included control
experiments with C57BL6/Nmice.

Drugs

()-1-Aminocyclopentane-trans-1,3-dicarboxylic acid (t-ACPD),


(RS)-3,5-Dihydroxyphenylglycine (DHPG) and -methyl-5hydroxytryptamine (-m-5-HT) were obtained from Tocris (Biozol,
Eching, Germany) or from Biotrend (Biotrend Chemikalien GmbH,
Cologne, Germany). Muscarine was obtained from Sigma-Aldrich
(Sigma, Deisenhofen, Germany) and CP 809101 was purchased from
Biozol (Eching, Germany). Drugs were prepared as stock solutions
in distilled water or directly added to the perfusion medium.
Immunofluorescence

Preparation of brain slices for electrophysiological


experiments

At postnatal days 1824 (G11/ and Gq/G11/) or 1128


(C57BL6/N) mice were deeply anesthetized using isoflurane and
decapitated as described earlier (Meuth etal., 2006). Briefly, thalamic slices were prepared as coronal sections on a vibratome (Series
1000 Classic, St. Louis, USA) in an ice-chilled solution containing
(in mM): Sucrose, 200; PIPES, 20; KCl, 2.5; NaH2PO4, 1.25; MgSO4,
10; CaCl2, 0.5; dextrose, 10. The pH was adjusted to 7.35 with
NaOH. Prior to recording, slices were kept at room temperature
submerged in artificial cerebrospinal fluid (ACSF) that contained
(in mM): NaCl, 125; KCl, 2.5; NaH2PO4, 1.25; NaHCO3, 24; MgSO4,
2; CaCl2, 2; dextrose, 10. The pH was adjusted to 7.35 by gassing
with carbogen (95% O2, 5% CO2).
Patch-clamp recordings

Whole-cell recording pipettes (23 M) were prepared from


borosilicate glass capillaries [GT150T-10(F), Clark Electromedical
Instruments, Pangbourne, UK] and filled with an intracellular solution containing (in mM): K-gluconate, 95; K3-citrate, 20; NaCl, 10;
HEPES, 10; MgCl2, 1; CaCl2, 0.5; BAPTA, 3; Mg-ATP, 3; Na-GTP,
0.5. The pH was adjusted to 7.25 with KOH, the osmolality was
295 mOsm/kg. In the recording chamber, slices were continuously superfused with a solution containing (in mM): NaCl, 120;
KCl, 2.5; NaH2PO4, 1.25; HEPES, 30; MgSO4, 2; CaCl2, 2; dextrose,
10. The pH was adjusted to 7.25 with HCl and osmolality was
305mOsm/kg. Recordings were performed at room temperature.
Whole-cell patch-clamp electrodes were attached to an EPC-10
amplifier (HEKA Elektronik, Lamprecht, Germany), and digitized
signals were saved to a computer using Pulse software (HEKA).
During current clamp recordings, the instantaneous frequency (fi)
of action potential generation was determined by analyzing the
first two action potentials that were generated upon a depolarizing
current pulse.
Recordings were only performed on recombined TC neurons. We distinguished these from non-recombined GABAergic
interneurons based on our previously established physiological and
morphological criteria (Broicher etal., 2008a). All cells had a resting membrane potential negative of 60mV, the access resistance
was in the range of 515M and series resistance compensation
of 30% or more was routinely applied. A liquid junction potential
of 82mV (n=10) was measured and taken into account.
All results are presented as meanSEM and differences were
considered significant when p<0.05. Substance effects were tested
for statistical significance using a modified Students t-test for
smallsamples.

Frontiers in Cellular Neuroscience

Gq/G11/ mice (postnatal days 1827) were deeply anesthetized


using pentobarbital (50mg/kg body weight) and transcardially perfused with PBS, followed by an ice-cold 4% PFA/PBS for 3540min.
Brains were removed, postfixed for 4h in 4% PFA/PBS and cryoprotected with 25% sucrose. Coronal sections (40m) were cut at the
level of the dLGN, washed several times with TBS, and blocked with
10% normal horse serum (NHS), 2% BSA, and 0.3% Triton X-100
in TBS for 2h to minimize non-specific binding before incubation of
slices with primary antibodies in 2% NHS, 2% BSA, and 0.3% Triton
X-100 in TBS at 4C for 1618h. The following antibodies were used:
rabbit anti-5-HT2A (1:500, ImmunoStar, Hudson, WI, USA), rabbit
anti-5-HT2C (1:1000, ImmunoStar), rabbit anti-mGluR1a (1:1000,
Novus Biologicals Inc., Littleton, CO, USA), mouse anti-NeuN (1:150,
Millipore, Schwalbach, Germany), and mouse anti-MAP2 (1:200,
Sigma). After washing (310min with TBS), sections were exposed
to Cy2- or Cy3-conjugated donkey-IgG (1:300, Dianova, Germany)
for 1.5h, washed again, and cover slipped with Immumount. For
negative controls, occlusion of the primary antibody from the staining
procedure was routinely performed with no positive immunological
signal detected. Densitometric analysis of immunofluorescence was
performed by using the fluorescence measuring function of ImageJ
software (public domain, National Institutes of Health). One hundred
square areas (30m30m) were placed on different positions of
dLGN images and the mean fluorescence intensity was determined
after background subtraction. This analysis was repeated for three
slices taken from different animals.

Results
5-HT receptor signaling and competition between different
receptor classes in Gq/G11/ mice

In order to mimic an arousal by brainstem and cortical inputs to


the thalamus, we have chosen the paradigm described in the following. A hyperpolarized membrane potential negative to 70mV
is necessary to achieve burst firing of TC neurons in response to a
depolarizing current step in vivo (Steriade, 1991), and thus, cells
were held at about 71mV by DC current injection. Rather small
sleep-related variations of the membrane potential (10mV) found
in TC neurons are sufficient to mediate a switch between burst and
tonic firing in vivo (Hirsch etal., 1983). The experiments described
below tested to what extent the knockout of Gq has affected the
ability of TC neurons to perform this switch.
Acetylcholine plays a major role in the modulation of thalamic states
of activity and the function of this transmitter depends on Gq-coupled
muscarinic receptors. However, it is unknown which G-proteins
are targeted by the other brainstem neurotransmitters. Thus, we
tested neurotransmitter candidates expected to be connected to

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 2

Coulon etal.

Gq/11 signaling in the thalamus

Gq/G11-mediated signaling pathways. Because of the known coupling


of 5-HT2 receptors to Gq/G11 family G-proteins (Roth etal., 1998),
we applied the 5-HT2 receptor agonist -m-5-HT and monitored
changes in TC neurons under current clamp conditions. A shift
from burst to tonic firing relies on the initial membrane potential.
In order to provide comparable conditions between neurons with a
different resting membrane potential and to ensure robust bursting
with two or more action potentials riding on top of a low-threshold
Ca2+ spike (LTS), all TC neurons investigated here were set to a membrane potential of 70.70.2mV (n=70) by DC offset. The resting
membrane potential was 69.40.9mV and the applied DC offset
current was 102.4pA (n=70). Of 70 neurons investigated under
these conditions, only 1 had a resting membrane potential positive
to 65mV (63mV) and almost all (68 of 70 TC neurons) were
able to generate a LTS. Cells with a membrane potential positive to
65mV and cells that could not generate a LTS were excluded from
further analysis. Step depolarization resulted in high-frequency
burst firing in both mice genotypes (G11/: fi=1517Hz, n=17;
Gq/G11/: fi=1598Hz, n=17; Figure1A). In G11/, the
application of 100 and 200M -m-5-HT depolarized the resting

Figure1 | Membrane potential and muscarinic signaling are influenced


by -m-5-HT. (A) Firing pattern in a control mouse (upper panel) and in Gq/
G11/ (lower panel) under control conditions (left traces) and in the presence
of -m-5-HT (right traces). The insets show the low-threshold Ca2+ potentials
at a ten times expanded timescale. (B) Mean bar graph representation of the
depolarization induced by two different substance concentrations as
indicated. (C) Voltage vs. time plot of two individual cells from control animals
(black line: application of 100M -m-5-HT followed by application of 50M

Frontiers in Cellular Neuroscience

membrane potential of TC neurons by 6 1 mV (n = 10) and


101mV (n=7), respectively (Figure1B). We also tested the
effect of a more specific agonist for 5-HT2C receptors (CP 809101)
on the membrane potential in current clamp recordings (Siuciak
etal., 2007). This agonist produced comparable depolarizations
(10M: 8.32.0mV, n=5; 100M: 17.51.4mV, n=4). This
suggests that 5-HT2C receptors in the dLGN could be responsible
for a large portion of the observed effect. The response to -m5-HT was significantly reduced in Gq/G11/ (100M: 31mV,
n=11, p<0.001; 200M: 51mV, n=6, p<0.01). In the presence
of 200 M -m-5-HT, depolarizing current steps elicited tonic
firing in G11/ (fi=6914Hz, n=7; data not shown). During
application of 100M -m-5-HT in Gq/G11/ burst firing was
either preserved (fi=1266Hz, n=4) or the LTS was crowned by
a single action potential (n=7, Figure1A, lower right trace). For
the other two recording conditions (100M in controls, 200M
in Gq/G11/) depolarizing current steps either evoked an intermediate response with slow bursting (fi100Hz) followed by 14
tonic action potentials (not shown) or passive membrane responses
(Figure1A, upper right trace). See also Sherman (1996).

muscarine; gray line: application of 200M -m-5-HT followed by application


of 50M muscarine). Arrows illustrate plateau of the substance effects.
Muscarine was applied in the continuous presence of -m-5-HT. (D) Plot of
the depolarization induced by -m-5-HT vs. the depolarization induced by
muscarine in individual cells from control animals (filled circles). Open circles
represent the numerical sum of the two depolarizations. Straight black
line=linear regression of the data points. Dashed line=mean value of the
summed effects.

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 3

Coulon etal.

Gq/11 signaling in the thalamus

To further characterize Gq-dependent signaling in TC neurons


we applied muscarine (50M) on top of different concentrations
of -m-5-HT in G11/. At the concentration we used, muscarine is
known to depolarize G11/ TC neurons by about 18mV (Broicher
etal., 2008b). 100M -m-5-HT induced a depolarizing shift of the
membrane potential of 71mV (n=4; Figure1C, black trace).
The following muscarine-induced depolarization was only 71mV
(n=4; Figure1C black trace). The combined depolarization was
14mV. An increased concentration of -m-5-HT (200M) induced
a stronger depolarization of 101mV which was accompanied by a
reduced muscarine effect averaging 31mV (n=7; Figure1C gray
trace), so that the combined depolarization was similar to the previous experiment. Plotting the amplitude of the muscarine effect as a
function of the amplitude of the -m-5-HT effect revealed a clear
linear dependency of the two parameters (Figure1D; closed circles), i.e., the larger the effect of -m-5-HT, the smaller the effect of
muscarine. The sum of the combined substance effects (Figure1D,
open circles) was independent of the recording conditions and averaged 13.50.4mV (Figure1D, dashed line). Obviously, this was
an upper limit for a combined effect of muscarine and -m-5-HT
on VM. To verify this, we reversed the order of agonist application
using C57BL6/N mice. Now, the initial muscarine effect (50M) was
153mV followed by a strongly reduced -m-5-HT effect (100M)
of 32mV, summing up to a combined effect of 184mV (n=6).
Agonist-induced depolarization was significantly reduced in Gq/
G11/ (100M -m-5-HT/50M muscarine: 41mV/21mV,
p<0.01/p<0.001, n=6; 200M -m-5-HT/50M muscarine:
51mV/21mV, p<0.01/p=0.27, n=5).

As shown in Figure1D, the variability of a combined effect of


the two agonists, namely -m-5-HT and muscarine is rather small
(open circles). Moreover, the variation in the effect of -m-5-HT is
comparable to the variation in the effect of muscarine. Still, when
applied together, the combined effect does not exceed a given value.
Or, in other words, both agonist effects vary, depending on how
occupied the system is by the presence of the respective other agonist. This suggests that both mechanisms of activation involve a
common and limiting mechanism.
In summary, these data demonstrate that serotonergic signaling
depends on Gq-type G-proteins, and suggest that the same pool of
G-proteins can be accessed by different transmitter pathways.
Metabotropic GluR-mediated signaling in Gq/G11/ mice

Since glutamate and mGluR1 are also known to depolarize TC


neurons by their action on IKL (McCormick and von Krosigk,
1992; Salt, 2002) and, moreover, are known to be coupled to Gq/G11
(Wettschureck etal., 2004a), we investigated the effect of the mGluR
agonist t-ACPD on the TC neurons firing patterns. TC neurons set
to a membrane potential of 71.20.3mV (n=28) by DC offset
showed burst responses to depolarizing current steps as described
above (G11/: fi=1395Hz, n=16; Gq/G11/: fi=1464Hz,
n=12; Figures2A,B, left panels). Application of t-ACPD resulted in a
depolarizing shift of the membrane potential, the amplitude of which
was significantly (p<0.001) larger in G11/ (50M: 231mV, n=7;
100M: 172mV, n=8) than in Gq/G11/ (50M: 112mV,
n=7; 100M: 82mV, n=6; Figures2C,D). The effects of 50
and 100M t-ACPD were not significantly different, indicating a

Figure2 | t-ACPD effect on firing properties. (A,B) Functional consequences of t-ACPD administration during current clamp recordings in a control mouse (A) and
in Gq/G11/ (B). (C) Mean voltage vs. time plot (black squares, control animal; gray circles, Gq/G11/). The horizontal bar indicates substance application. (D) Mean
bar graph representation of the depolarization induced by two different substance concentrations as indicated.

Frontiers in Cellular Neuroscience

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 4

Coulon etal.

Gq/11 signaling in the thalamus

saturating response. In both mouse strains, the t-ACPD-dependent


depolarization was strong enough to induce a shift from burst to
tonic firing of action potentials, which is also reflected in the firing
frequency (G11/, 50M: fi=322Hz, n=7; G11/, 100M:
fi=4411Hz, n=8; Gq/G11/, 50M: fi=383Hz, n=7;
Figures2A,B). In Gq/G11/ at 100M t-ACPD four out of six cells
revealed an intermediate response with a slow burst (fi=969Hz,
n=4) followed by 28 tonic action potentials (data not shown).
Only one cell showed tonic firing (27Hz), and the remaining cell
was lost in the course of the experiment.
To further elucidate the role of mGluR we used the selective group
I agonist (RS)-3,5-DHPG (DHPG) (Ito etal., 1992). Application of
DHPG (50 or 100M) induced a reversible depolarization of the
membrane potential of 132mV (50M, n=7) or 241mV
(100 M, n = 5) under current clamp conditions in G11/ or
C57BL6/N (Figures3C,D). This effect was significantly (p<0.01)
reduced in Gq/G11/ (50M, 51mV, n=8; Figures3C,D). In
control mice (Figure3A), but not in Gq/G11/ (Figure3B), the
DHPG-induced depolarization was accompanied by a switch to
tonic firing of action potentials.
It could be argued that the application of a neuromodulator
could lead to changes in interneuron firing (Sherman, 2004; Munsch
etal., 2005), which in turn could lead to changes in RMP unrelated
to the activation of Gq-coupled receptors of the recorded cell. To
test for this, we applied 50M DHPG in the presence of 0.5M
TTX, eliminating all action potential firing and excluding presynaptic effects. Under these conditions, the observed depolarization
was very similar to control conditions (123mV, n=3).

In summary, these data indicate that the action of glutamate in


the thalamus critically depends on intact Gq-mediated intracellular
signaling pathways.
Expression of 5-HT and mGlu receptor subtypes in dLGN

To provide further evidence that Gq-coupled receptors are functionally expressed in TC neurons we performed immunohistochemical stainings by antibody labeling.
Application of 5-HT2AR-specific antibodies did not lead to
staining by binding of the secondary antibody (Figure4B, middle
image). Strong staining was observed upon application of 5-HT2CR(Figure 4C, middle image) and mGluR1-specific antibodies
(Figure4D, middle image). We co-stained slices with the neuronspecific nucleus marker NeuN (Figures4BD left images; merged
images to the right). This revealed that 5-HT2CR and mGluR1
were not somatically expressed in neurons of the LGN. We verified
this by examining slices that were co-stained with a marker for
microtubule associated protein 2 (MAP2) at a higher magnification. The images revealed that neither the staining for 5-HT2C (not
shown) nor for mGluR1 clearly marked the outline of the soma
(Figure S1 in Supplementary Material). We could not observe any
differences between stainings in the dLGN of control and Gq/
G11/ mice (Figure S2 in Supplementary Material).
The densitometric analysis of 5-HT2CR-specific fluorescence
revealed significant differences between control (mean fluorescence intensity=321a.u.; n=3 independent slices) and Gq/
G11/ mice (mean fluorescence intensity=411a.u.; n=3; data
not shown). No differences were found for mGluR1-specific

Figure3 | DHPG effect on firing properties. (A,B) Functional consequences of DHPG administration during current clamp recordings in a control mouse (A) and in
Gq/G11/ (B). (C) Mean voltage vs. time plot (black squares, control animal; gray circles, Gq/G11/). The horizontal bar indicates substance application. (D) Mean
bar graph representation of the DHPG-induced depolarization.

Frontiers in Cellular Neuroscience

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 5

Coulon etal.

Gq/11 signaling in the thalamus

Figure4 | Immunohistochemical detection of mGluR and 5-HTR subtypes


in dLGN of Gq/G11/. (A) Schematic drawing of a frontal brain slice showing
the position of the dLGN. In B-D the dLGN is marked by a dashed line. Specific
antibodies for NeuN, 5-HT2AR, 5-HT2CR, and for mGluR1 were used and labeled

with secondary antibodies conjugated to Cy2 or Cy3. 5-HT2AR-specific antibodies


did not cause staining (B). 5-HT2CR- (C) and mGluR1-specific antibodies (D)
caused strong staining. 5-HT2CR and mGluR1 did not appear to be somatically
expressed. Scale bars indicate 100 m.

uorescence (control: mean fluorescence intensity=441a.u.;


fl
Gq/G11/: mean fluorescence intensity=421a.u.; n=3 independent slices each; data not shown). These findings point to an
attempt to compensate the loss of 5-HT2C receptor function by an
increased number of receptors in Gq/G11/ mice.

G-protein-coupled membrane receptors (McCormick, 1992a).


In addition, the excitatory transmitter glutamate exerts powerful
activation of thalamic neurons, thereby driving the sensory relay
in the thalamus (Salt, 2002). Corticofugal inputs, in particular, are
connected to mGluR (Godwin etal., 1996b).

Discussion

5-HT receptor-dependent signaling

The results of the present study can be summarized as follows:


(1) The action of 5-HT and glutamate in TC neurons depends on
Gq-coupled intracellular signaling cascades. The effects of -m5-HT, t-ACPD, and DHPG under current clamp conditions were significantly reduced in Gq/G11/. As a consequence, the stimulation
of metabotropic 5-HTR and GluR could not induce a full shift from
burst to tonic firing of action potentials in Gq/G11/ mice. (2)
The amplitude of membrane potential depolarization by muscarine
depends on the degree of prior utilization of the Gq-dependent pathway. (3) It is concluded that Gq/G11 family G-proteins play a central
role in the state-dependent control of thalamic activity modes.

Waking, but not REM sleep, is accompanied by increased serotonergic activity in the thalamus (Steriade etal., 1997). The role of 5-HT
in the thalamus seems complex and is not yet fully understood:
Cellular effects may be direct or indirect and show regional differences. In vitro studies demonstrated that 5-HT causes a small
depolarization and a shift in voltage dependency of the hyperpolarization activated cation current, Ih. The latter is achieved via
GS-proteins and cAMP production (McCormick and Pape, 1990;
Lee and McCormick, 1996). Moreover, inhibition of an ISO component occurred, resembling the current mediated by TASK channels
(S. G. Meuth and T. Budde, unpublished results). In consequence,
oscillatory burst activity is suppressed. An indirect modulation of Ih
via Gq-proteins may include IP3-induced Ca2+ release from intracellular stores and subsequent activation of a Ca2+-dependent adenylyl
cyclase (Lthi and McCormick, 1998). The findings of the present
study indicate that a significant part of the 5-HT-induced depolarization is mediated by Gq/G11-coupled receptors. All receptors of
the 5-HT2 subclass are coupled to Gq/G11-proteins. These in turn,
activate PLC-. 5-HT2A and 5-HT2C are widely distributed in the

Gq /G11-dependent signaling in the thalamus

The flow of sensory information from the sensory organs to the


cerebral cortex via the thalamus is highly regulated by inputs from
the ascending activating brainstem system, releasing ACh, 5-HT,
and noradrenalin (Steriade etal., 1997). These neuromodulators
exert their influence onto the state of the thalamus by altering
specific ion channels. This is achieved through the activation of

Frontiers in Cellular Neuroscience

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 6

Coulon etal.

Gq/11 signaling in the thalamus

brain and are present in the rodent dLGN (Li etal., 2004). Thus,
the strong reduction of the effect of -m-5-HT in Gq/G11/ is in
good agreement with a 5-HT2 expression in dLGN and is possibly
connected to the modulation of TASK channels.
In the course of this study, we made the interesting observation that muscarinic and serotonergic receptors seem to compete
for the same Gq-protein-coupled signaling pathway. The effect of
activation of one receptor class is strongly and negatively correlated to the strength of prior activation of the other receptor class,
thereby suggesting convergence onto the same limited pool of
Gq-protein-coupled signaling pathways. This view is supported by
results obtained from cat and guinea pig TC neurons that suggested
acetylcholine- and noradrenalin-induced slow depolarizations to
occur through the activation of the same second-messenger system
(McCormick, 1992b).
mGluR-dependent signaling

Group I mGluRs consist of mGluR1 and mGluR5 that are positively coupled to PLC-. Several types of mGluRs are expressed in
the dLGN of different species, with retinal (mGluR5) and cortical (mGluR1) inputs accessing specific subtypes (Godwin etal.,
1996b; Lourenco Neto etal., 2000). Application of t-ACPD and
selective mGluR1 agonists depolarizes TC neurons and switches
their activity mode from burst to tonic firing, thereby mediating
TC transmission (McCormick and von Krosigk, 1992; Godwin
etal., 1996a; Salt, 2002). The results of the present study are in line
with these findings. However, the remaining t-ACPD effect in Gq/
G11/ indicates that mGluRs not coupled to Gq/G11 contribute to
the response. This conclusion is corroborated by the finding that
mGluR3, mGluR4, and mGluR7 are expressed in rodent dLGN
(Lourenco Neto et al., 2000). During postnatal development,
specific changes in the subcellular location of mGluRs have been
observed (Liu etal., 1998) which are the basis for the topographical association to different input systems (Godwin etal., 1996b;
Turner and Salt, 2000).
Residual effects of neurotransmitters in Gq/G11/ mice

While Gq and G11 have very similar effector-coupling properties


and may substitute each other (Offermanns, 1999), the remaining
effect of receptor agonists in Gq/G11/ may result from different
mechanisms:
(1) Other G-protein families may compensate for the lack of
Gq in deficient mice. Findings from hippocampal neurons of Go-deficient mice have shown that in the absence
of Go, ion channels will be regulated by other G-proteins
with different properties (Greif et al., 2000). In particular
G15 and G16 can link a variety of predominantly Gqcoupled receptors to the PLC pathway (Offermanns and
Simon,1995).
(2) Receptors, intracellular signaling proteins, and effectors of the
Gq-dependent pathways may be up-regulated. However, the
Gq/G11/ mice develop a severe epileptic phenotype with
a reduced life span, which argues against an effective compensation mechanism (Wettschureck et al., 2006; Broicher
etal., 2008b). Indeed, mRNA expression of effector channels
(TASK channels, HCN channels) and all mAChR subtypes

Frontiers in Cellular Neuroscience

were unchanged in dLGN tissue (Wettschureck etal., 2006;


Broicher etal., 2008b) and a lack of strong plastic compensation has been noted before in TC neurons of mice deficient
for HCN2 (Ludwig etal., 2003) and TASK-1 (Meuth etal.,
2006). Nevertheless, the densitometric analysis of immunohistochemical staining in the present study indicated an
increased expression of 5-HT2CR in Gq/G11/ mice, thereby possibly showing a futile attempt to compensate the loss
of receptor function by an increased number of membrane
receptors.
(3) G-protein signaling outside the canonical seven transmembrane domain receptors and G-protein independent
pathways of these receptors may exist. Recent evidence
indicates that G-proteins play important roles in receptor
tyrosine kinase signaling and may be activated by accessory proteins (Marty and Ye, 2010; Sato and Ishikawa,
2010). Furthermore, increasing evidence indicates that
ERK, JAK/STATs, Src-family tyrosine kinases, -arrestins,
and PDZ domain-containing proteins directly relay signals
from seven transmembrane domain receptors, independent of G-proteins (Sun et al., 2007). Future studies have
to show the coupling of metabotropic ACh, 5-HT, and Glu
receptors to parallel pathways, activated by -subunits, by
G-proteins not belonging to the Gq/G11-family, and/or by
G-proteinindependent mechanisms.
(4) Gq may be incompletely eliminated. The CaMKII-Cre mouse
line has been shown to express Cre in forebrain principal
neurons, but not in forebrain interneurons (Mantamadiotis
etal., 2002; Marsicano etal., 2003). With respect to recombination efficiency within the population of forebrain principal
neurons it is generally assumed that recombination is complete (Marsicano etal., 2003). Within a particular forebrain
principal neuron, Cre expression always results in a complete
inactivation of Gq, since the CaMKII promoter chosen to
drive Cre expression is very strong. We are therefore positive
that we can rule out any partial Gq inactivation (i.e., only
one of the two Gq alleles is recombined) in Cre-expressing
principal neurons.

Expression of 5-HT and mGlu receptors

Immunohistochemical stainings provided evidence that Gqcoupled receptors are functionally expressed in TC neurons.
However, the application of 5-HT2AR-specific antibodies failed to
stain the tissue. Why this is the case remains unclear. The strong
staining that was observed upon application of 5-HT2CR- and
mGluR1-specific antibodies was not reduced in Gq/G11/ mice,
suggesting that the deletion of Gq does not lead to a down regulation of its upstream receptors (compare Figure4 and Figure S2
in Supplementary Material). Co-staining slices with the neuronspecific nucleus marker NeuN revealed that 5-HT2CR and mGluR1
were not somatically expressed in neurons of the LGN. This was
confirmed in a co-staining with MAP2.
The finding that a specific agonist for 5-HT2C receptors had the
same effect on the membrane potential in current clamp recordings
than -m-5-HT in control mice suggests that 5-HT2C receptors in
the dLGN could be the functionally dominant isoform.

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 7

Coulon etal.

Gq/11 signaling in the thalamus

Convergence of transmitter pathways

The strong convergence of several transmitter/receptor systems on


Gq/G11 family G-proteins in TC neurons leads to the question of
the degree of redundancy in their signaling. The finding (i) of differential subcellular locations of specific membrane receptors, (ii)
of the topographical organization of different input systems, (iii)
of different state-dependent releases of neurotransmitters, and (iv)
of the formation of tight receptor/effector protein complexes in
the forebrain indicate a separation of function between the different Gq/G11-coupled receptor classes in the same neuronal population. Thus, spatial separation and state-dependent activation allow

References
Broicher, T., Kanyshkova, T., Meuth, P.,
Pape, H. C., and Budde, T. (2008a).
Correlation of T-channel coding gene
expression, IT, and the low threshold
Ca2+ spike in the thalamus of a rat
model of absence epilepsy. Mol. Cell.
Neurosci. 39, 384399.
Broicher, T., Wettschureck, N., Munsch, T.,
Coulon, P., Meuth, S. G., Kanyshkova,
T., Seidenbecher, T., Offermanns, S.,
Pape, H. C., and Budde, T. (2008b).
Muscarinic ACh receptor-mediated
control of thalamic activity via G(q)/
G(11)-family G-proteins. Pflugers
Arch. 456, 10491060.
Chemin, J., Girard, C., Duprat, F., Lesage,
F., Romey, G., and Lazdunski, M.
(2003). Mechanisms underlying excitatory effects of group I metabotropic
glutamate receptors via inhibition of
2P domain K+ channels. EMBO J. 22,
54035411.
Chen, X., Talley, E. M., Patel, N., Gomis,
A., McIntire, W. E., Dong, B., Viana,
F., Garrison, J. C., and Bayliss, D. A.
(2006). Inhibition of a background
potassium channel by Gq protein
{alpha}-subunits. Proc. Natl. Acad.
Sci. U. S.A.103, 34223427.
Exton, J. H. (1996). Regulation of phosphoinositide phospholipases by hormones, neurotransmitters, and other
agonists linked to G proteins. Annu.
Rev. Pharmacol. Toxicol. 36, 481509.
Godwin, D. W., Vaughan, J. W., and
Sherman, S. M. (1996a). Metabotropic
glutamate receptors switch visual
response mode of lateral geniculate
nucleus cells from burst to tonic. J.
Neurophysiol. 76, 18001816.
Godwin, D. W., Van Horn, S. C., Eriir, A.,
Sesma, M., Romano, C., and Sherman,
S. M. (1996b). Ultrastructural localization suggests that retinal and cortical
inputs access different metabotropic
glutamate receptors in the lateral
geniculate nucleus. J. Neurosci. 16,
81818192.
Greif, G. J., Sodickson, D. L., Bean, B.
P., Neer, E. J., and Mende, U. (2000).
Altered regulation of potassium and
calcium channels by GABAB and

Frontiers in Cellular Neuroscience

adenosine receptors in hippocampal


neurons from mice lacking galpha o.
J. Neurophysiol. 83, 10101018.
Hirsch, J. C., Fourment, A., and Marc,
M. E. (1983). Sleep-related variations
of membrane potential in the lateral
geniculate body relay neurons of the
cat. Brain Res. 259, 308312.
Ito, I., Kohda, A., Tanabe, S., Hirose,
E., Hayashi, M., Mitsunaga, S., and
Sugiyama, H. (1992). 3,5-Dihydroxyphenyl-glycine: a potent agonist of
metabotropic glutamate receptors.
Neuroreport 3, 10131016.
Lee, K. H., and McCormick, D. A. (1996).
Abolition of spindle oscillations by
serotonin and norepinephrine in
the ferret lateral geniculate and perigeniculate nuclei in vitro. Neuron 17,
309321.
Li, Q. H., Nakadate, K., Tanaka-Nakadate,
S., Nakatsuka, D., Cui, Y., and
Watanabe, Y. (2004). Unique expression patterns of 5-HT2A and 5-HT2C
receptors in the rat brain during postnatal development: western blot and
immunohistochemical analyses. J.
Comp. Neurol. 469, 128140.
Liu, X. B., Munoz, A., and Jones, E. G.
(1998). Changes in subcellular localization of metabotropic glutamate
receptor subtypes during postnatal
development of mouse thalamus. J.
Comp. Neurol. 395, 450465.
Lopes, C. M., Rohacs, T., Czirjak, G.,
Balla, T., Enyedi, P., and Logothetis,
D. E. (2005). PIP2 hydrolysis underlies agonist-induced inhibition and
regulates voltage gating of two-pore
domain K+ channels. J. Physiol. (Lond.)
564, 117129.
Lourenco Neto, F., Schadrack, J., Berthele,
A., Zieglgansberger, W., Tolle, T.
R., and Castro-Lopes, J. M. (2000).
Differential distribution of metabotropic glutamate receptor subtype
mRNAs in the thalamus of the rat.
Brain Res. 854, 93105.
Ludwig, A., Budde, T., Stieber, J.,
Moosmang, S., Wahl, C., Holthoff, K.,
Langebartels, A., Wotjak, C., Munsch,
T., Zong, X., Feil, S., Feil, R., Lancel,
M., Chien, K. R., Konnerth, A., Pape,

ivergent action of several metabotropic receptor classes, which


d
seem to utilize the same pool of Gq/G11-proteins. These findings,
therefore, suggest a scenario in which Gq/G11-proteins could function as a central element in thalamic physiology.

Acknowledgments
Supported by the German Research Foundation (DFG): FOR1086,
TP2 to Thomas Budde and Sven G. Meuth. Supported by the
Interdisziplinres Zentrum fr Klinische Forschung Mnster,
Bud3/010/10 to Thomas Budde. The authors wish to thank A. Jahn, E.
Nass, A. Markovic, and R. Ziegler for excellent technical assistance.

H. C., Biel, M., and Hofmann, F.


(2003). Absence epilepsy and sinus
dysrhythmia in mice lacking the
pacemaker channel HCN2. EMBO J.
22, 216224.
Lthi, A., and McCormick, D. A. (1998).
Periodicity of thalamic synchronized oscillations: the role of Ca2+mediated upregulation of Ih. Neuron
20, 553563.
Mantamadiotis, T., Lemberger, T.,
Bleckmann, S. C., Kern, H., Kretz,
O., Villalba, A. M., Tronche, F.,
Kellendonk, C., Gau, D., Kapfhammer,
J., Otto, C., Schmid, W., and Schutz, G.
(2002). Disruption of CREB function
in brain leads to neurodegeneration.
Nat. Genet. 31, 4754.
Marsicano, G., Goodenough, S.,
Monory, K., Hermann, H., Eder, M.,
Cannich, A., Azad, S. C., Cascio, M.
G., Gutierrez, S. O., van der Stelt, M.,
Lopez-Rodriguez, M. L., Casanova,
E., Schutz, G., Zieglgansberger, W., Di
Marzo, V., Behl, C., and Lutz, B. (2003).
CB1 cannabinoid receptors and ondemand defense against excitotoxicity.
Science 302, 8488.
Marty, C., and Ye, R. D. (2010).
Heterotrimeric G protein signaling
outside the realm of seven transmembrane domain receptors. Mol.
Pharmacol. 78, 1218.
M c C o r m i c k , D. A . ( 1 9 9 2 a ) .
Neurotransmitter actions in the thalamus and cerebral cortex and their
role in neuromodulation of thalamocortical activity. Prog. Neurobiol. 39,
337388.
McCormick, D. A. (1992b). Cellular
mechanisms underlying cholinergic
and noradrenergic modulation of
neuronal firing mode in the cat and
guinea pig dorsal lateral geniculate
nucleus. J. Neurosci. 12, 278289.
McCormick, D. A., and Pape, H.-C. (1990).
Noradrenergic and serotonergic
modulation of a hyperpolarizationactivated cation current in thalamic
relay neurones. J. Physiol. (Lond.) 431,
319342.
McCormick, D. A., and von Krosigk, M.
(1992). Corticothalamic activation

www.frontiersin.org

modulates thalamic firing through


glutamate metabotropic receptors. Proc. Natl. Acad. Sci. U.S.A. 89,
27742778.
Meuth, S. G., Aller, M. I., Munsch, T.,
Schuhmacher, T., Seidenbecher, T.,
Kleinschnitz, C., Pape, H. C., Wiendl,
H., Wisden, W., and Budde, T.
(2006). The contribution of TASK-1containing channels to the function of
dorsal lateral geniculate thalamocortical relay neurons. Mol. Pharmacol. 69,
14681476.
Meuth, S. G., Budde, T., Kanyshkova,
T., Broicher, T., Munsch, T., and
Pape, H.-C. (2003). Contribution of
TWIK-related acid-sensitive K+ channel 1 (TASK1) and TASK3 channels
to the control of activity modes in
thalamocortical neurons. J. Neurosci.
23, 64606469.
Munsch, T., Yanagawa, Y., Obata, K., and
Pape, H. C. (2005). Dopaminergic
control of local interneuron activity
in the thalamus. Eur. J. Neurosci. 21,
290294.
Offermanns, S. (1999). New insights into
the in vivo function of heterotrimeric
G-proteins through gene deletion
studies. Naunyn Schmiedebergs Arch.
Pharmacol. 360, 513.
Offermanns, S., and Simon, M. I. (1995).
G and G couple a wide variety of
receptors to phospholipase C. J. Biol.
Chem. 270, 1517515180.
Offermanns, S., Zhao, L. P., Gohla, A.,
Sarosi, I., Simon, M. I., and Wilkie, T.
M. (1998). Embryonic cardiomyocyte
hypoplasia and craniofacial defects in
G alpha q/G alpha 11-mutant mice.
EMBO J. 17, 43044312.
Roth, B. L., Willins, D. L., Kristiansen,
K., and Kroeze, W. K. (1998). 5-Hydroxytryptamine2-family receptors (5-hydroxytryptamine2A,
5-hydroxytryptamine2B, 5-hydroxytryptamine2C): where structure
meets function. Pharmacol. Ther. 79,
231257.
Salt, T. E. (2002). Glutamate receptor
functions in sensory relay in the thalamus. Philos. Trans. R. Soc. Lond., B,
Biol. Sci. 357, 17591766.

October 2010 | Volume 4 | Article 132 | 8

Coulon etal.

Sato, M., and Ishikawa, Y. (2010). Accessory


proteins for heterotrimeric G-protein:
implication in the cardiovascular system. Pathophysiology 17, 8999.
Sherman, S. M. (1996). Dual reponse
modes in lateral geniculate neurons:
mechanisms and functions. Vis.
Neurosci. 13, 205213.
Sherman, S. M. (2004). Interneurons
and triadic circuitry of the thalamus.
Trends Neurosci. 27, 670675.
Siuciak, J. A., McCarthy, S. A., Chapin,
D. S., Reed, T. M., Vorhees, C. V., and
Repaske, D. R. (2007). Behavioral
and neurochemical characterization of mice deficient in the phosphodiesterase-1B (PDE1B) enzyme.
Neuropharmacology 53, 113124.
Steriade, M. (1991). Alertness, quiet sleep,
dreaming. Cereb. Cortex 9, 279357.
Steriade, M., Jones, E. G., and McCormick,
D. A. (1997). Thalamus, 1st Edn.
Amsterdam: Elsevier.

Frontiers in Cellular Neuroscience

Gq/11 signaling in the thalamus

Sun, Y., McGarrigle, D., and Huang, X.-Y.


(2007). When a G protein-coupled
receptor does not couple to a G protein. Mol. Biosyst. 3, 849854.
Turner, J. P., and Salt, T. E. (2000). Synaptic
activation of the group I metabotropic
glutamate receptor mGlu1 on the thalamocortical neurons of the rat dorsal lateral geniculate nucleus in vitro.
Neuroscience 100, 493505.
Wettschureck, N., Moers, A., and
Offermanns, S. (2004a). Mouse models to study G-protein-mediated signaling. Pharmacol. Ther. 101, 7589.
Wettschureck, N., Moers, A., Hamalainen,
T., Lemberger, T., Schutz, G., and
Offermanns,S.(2004b).Heterotrimeric
G proteins of the Gq/11 family are
crucial for the induction of maternal
behavior in mice. Mol. Cell. Biol. 24,
80488054.
Wettschureck, N., Rutten, H., Zywietz,
A., Gehring, D., Wilkie, T. M., Chen,

J., Chien, K. R., and Offermanns, S.


(2001). Absence of pressure overload
induced myocardial hypertrophy after
conditional inactivation of Galphaq/
Galpha11 in cardiomyocytes. Nat.
Med. 7, 12361240.
Wettschureck, N., van der Stelt, M.,
Tsubokawa, H., Krestel, H., Moers,
A., Petrosino, S., Schutz, G., Di
Marzo, V., and Offermanns, S.
(2006). Forebrain-specific inactivation of Gq/G11 family G proteins
results in age-dependent epilepsy
and impaired endocannabinoid
formation. Mol. Cell. Biol. 26,
58885894.
Conflict of Interest Statement: The
authors declare that the research was
conducted in the absence of any commercial or financial relationships that
could be construed as a potential conflict
of interest.

www.frontiersin.org

Received: 06 July 2009; accepted: 28


September 2010; published online: 25
October 2010.
Citation: Coulon P, Kanyshkova T,
Broicher T, Munsch T, Wettschureck N,
Seidenbecher T, Meuth SG, Offermanns
S, Pape H-C and Budde T (2010) Activity
modes in thalamocortical relay neurons are
modulated by Gq/G11 family G-proteins
serotonergic and glutamatergic signaling.
Front. Cell. Neurosci. 4:132. doi: 10.3389/
fncel.2010.00132
Copyright 2010 Coulon, Kanyshkova,
Broicher, Munsch, Wettschureck,
Seidenbecher, Meuth, Offermanns,
Pape and Budde. This is an open-access
article subject to an exclusive license
agreement between the authors and the
Frontiers Research Foundation, which
permits unrestricted use, distribution,
and reproduction in any medium, provided the original authors and source
are credited.

October 2010 | Volume 4 | Article 132 | 9

Coulon etal.

Gq/11 signaling in the thalamus

Supplementary material

Figure S1 | Immunohistochemical staining of mGluR1 and MAP2. Neurons of the dLGN were stained using specific antibodies against microtubule
associated protein 2 (MAP2) and against mGluR1 conjugated to Cy2 and Cy3, respectively. While MAP2 was clearly present in the soma, mGluR1 appears to be
absent there. Scale bars indicate 10 m.

Figure S2 | Immunohistochemical detection of mGluR and 5-HTR subtypes in dLGN of G11/ mice. Specific antibodies and secondary antibodies were the
same as in Figure4. 5-HT2AR-specific antibodies again did not cause staining (A). 5-HT2CR- (B) and mGluR1-specific antibodies (C) caused strong staining as in the
dLGN of Gq/G11/ mice. Scale bars indicate 100 m.

Frontiers in Cellular Neuroscience

www.frontiersin.org

October 2010 | Volume 4 | Article 132 | 10

Pflugers Arch - Eur J Physiol (2012) 463:5371


DOI 10.1007/s00424-011-1014-6

INVITED REVIEW

The sleep relaythe role of the thalamus in central


and decentral sleep regulation
Philippe Coulon & Thomas Budde & Hans-Christian Pape

Received: 26 May 2011 / Revised: 8 August 2011 / Accepted: 11 August 2011 / Published online: 13 September 2011
# Springer-Verlag 2011

Abstract Surprisingly, the concept of sleep, its necessity


and function, the mechanisms of action, and its elicitors are
far from being completely understood. A key to sleep
function is to determine how and when sleep is induced.
The aim of this review is to merge the classical concepts of
central sleep regulation by the brainstem and hypothalamus
with the recent findings on decentral sleep regulation in
local neuronal assemblies and sleep regulatory substances
that create a scenario in which sleep is both local and use
dependent. The interface between these concepts is provided by thalamic cellular and network mechanisms that
support rhythmogenesis of sleep-related activity. The
brainstem and the hypothalamus centrally set the pace for
sleep-related activity throughout the brain. Decentral
regulation of the sleepwake cycle was shown in the
cortex, and the homeostat of non-rapid-eye-movement sleep
is made up by molecular networks of sleep regulatory
substances, allowing individual neurons or small neuronal
assemblies to enter sleep-like states. Thalamic neurons
provide state-dependent gating of sensory information via
their ability to produce different patterns of electrogenic
activity during wakefulness and sleep. Many mechanisms
of sleep homeostasis or sleep-like states of neuronal
assemblies, e.g. by the action of adenosine, can also be
found in thalamic neurons, and we summarize cellular and
network mechanisms of the thalamus that may elicit nonREM sleep. It is argued that both central and decentral

This article is published as part of the Special Issue on Sleep.


P. Coulon (*) : T. Budde : H.-C. Pape
Institut fr Physiologie I,
Westflische Wilhelms-Universitt Mnster,
Robert Koch Str. 27a,
48149 Mnster, Germany
e-mail: coulon@uni-muenster.de

regulators ultimately target the thalamus to induce global


sleep-related oscillatory activity. We propose that future
studies should integrate ideas of central, decentral, and
thalamic sleep generation.
Keywords Sleep . Thalamus . Cortex . Brainstem . Sleep
wake cycle . NREMS . Adenosine . Nitric oxide .
Hypothalamus . Homeostasis . Hypothalamus . Burst firing .
Neural network . Rhythmogenesis

Introduction
Sleep occupies roughly one third of our lifetime, and yet,
sleep research is unable to fully identify, let alone explain,
the nature, mechanisms, and functions of sleep. Reducing
sleep time is a common route in a notoriously busy society,
but prolonged sleep loss will cause (among other defects)
impaired metabolism, impaired immune function, impaired
cognitive function, and will, ultimately, lead to death [149].
Many pathophysiological states, such as lethargic encephalitis [190], insomnia [47], narcolepsy [114], apnoea [186],
cataplexy [126], or epilepsy [104], are related to disturbances in the mechanisms of sleep induction or maintenance.
Despite these valuable findings, the fundamental question of sleep research remains largely unanswered: Why do
we need to sleep? One of the main functions of sleep was
proposed to be memory consolidation [46, 102], but as
fundamental as this is, loss of memory consolidation cannot
explain the grave effects of sleep loss or disturbed sleep,
nor can the reduced immune and endocrine functions [120,
121], or impaired thermal regulation [180] that are hallmark
symptoms of sleep deprivation.
A key to sleep function could be to determine how and
when sleep is induced. The proposed mechanisms of sleep

54

induction have led to three main lines of evidence: (a) the


classical concept of central sleep regulation by thalamocortical afferents, the brainstem, and the hypothalamus
[108, 131, 187]; (b) regulation by thalamic neurons, which
provide a state-dependent gating of sensory information
based on their ability to produce different patterns of
electrogenic activity during wakefulness and sleep [92]; and
(c) more recently decentral regulation of the sleepwake
cycle, including cortical contributions [86, 131] and
individual cortical neurons being in the ON or OFF state
[188].
The aim of this review is to merge the classical concepts
of central sleep regulation with the recent findings on local
neuronal assemblies and sleep regulatory substances that
create a scenario in which sleep is both local and use
dependent. The interface between these concepts is provided by the thalamic cellular and network mechanisms that
support the rhythmogenesis of sleep-related activity. The
role of the thalamus in sleep regulation seems to be
consistently underestimated, although it possibly underlies
the regulation of the sleepwake cycle: both central and
decentral regulators must ultimately target the thalamus to
utilize the cellular mechanisms that induce stable and global
sleep-related oscillatory activity and allow the statedependent gating of sensory information.

Central regulators of sleep


Brainstem
The rhythmic changes in electrical activity of the human
brain, as studied by the electroencephalogram (EEG), differ
between wakefulness and sleep [9]. Although five different
sleep stages have since been described in humans [150],
this scenario is simplified for experimental purposes into
two sleeping stages: slow-wave sleep, also called non-REM
sleep, and rapid-eye-movement sleep (REM sleep or
paradoxical sleep) [151].
From the 1930s onward, the encphale isol and cerveau
isol preparations from Bremer, Moruzzi, and Magoun gave
rise to the idea that activity in the brainstem reticular
formation was the key to the cortical activity patterns
observed in the EEG during arousal [119]. This led to the
concept of passive sleep, which is based on the idea that
absence of all arousal activity from the brainstem will cause
the thalamocortical system to revert to a stable oscillatory
activity by default. Experimentally, this was based on
numerous findings showing that local stimulation in the
brainstem causes arousal and EEG desynchronisation [119],
while lesions of the ascending fibres were followed by
lethargy or coma. Two concepts were thought to act in
concert: (a) The brainstem actively maintains wakefulness

Pflugers Arch - Eur J Physiol (2012) 463:5371

by an intrinsic mechanism (the brainstem reticular activating concept) and (b) sensory input from the periphery is
needed for the maintenance of wakefulness (maintenance of
cerebral tone by sensory pathway activity, see [176]).
The main argument against theories of passive sleep has
been that sleep or sleep-like states can be induced by
stimulation either of sensory receptors or of central
structures [176]. The theories of active sleep have evolved
around the fact that peripheral or central low-frequency
stimulation causes sleep or EEG synchronization (see
[118]). Moreover, afferent volleys from the vagus nerve
are relevant for producing sleep with a synchronized EEG
pattern [147] and earlier findings showed that repetitive
visual stimulation evoked sleep-like behavioural signs with
matching EEG patterns in cats and humans [38, 58, 98].
However, the sites that cause EEG synchrony after
stimulation are numerous and applying synchronous stimuli
was considered nonsensical for the study of normal brain
activity, since they disturb natural patterns of neuronal
discharges and thus cannot be used to induce natural states
of sleep (see [176]). Lesion and pharmacological studies
have provided evidence that two major cerebral sites are
implicated in active sleep generation: The nucleus of the
solitary tract of the medulla [8, 96] and the preoptic area of
the basal forebrain ([112, 124, 156], see [176]). However,
the idea of active sleep induction and an active hypnogenic
focus also requires data at the cellular level. Neurons
promoting sleep were thought to at least require projections
toward the cholinergic and noradrenergic neurons in the
upper brainstem core [176]. This could provide a disfacilitation of brainstem to thalamus excitatory pathways,
supporting thalamic synchronization [97].
Hypothalamus
The ventro-lateral preoptic nucleus was suggested to initiate
sleep onset by several mechanisms of reciprocal inhibition
of the following [131]: (a) the cholinergic, noradrenergic,
and serotonergic arousal systems in the brainstem; (b)
histaminergic systems of the tuberomammillary nucleus in
the posterior hypothalamus; and (c) cholinergic systems of
the basal forebrain. All of these are in addition modulated
by the orexinergic arousal system of the lateral hypothalamus [64]. When active, these systems promote wakefulness, and accordingly, their inhibition promotes sleep. In
this scenario, the ventro-lateral preoptic nucleus is a switch
to initiate sleep: With increasing time of neuronal activity
during wakefulness, sleep regulatory substances accumulate
and eventually trigger the ventro-lateral preoptic nucleus
[131] in conjunction with circadian input from the suprachiasmatic nucleus [2] in the anterior hypothalamus, the
circadian rhythm of which is controlled by a series of genes
[87]. Sleep regulatory substances include adenosine in the

Pflugers Arch - Eur J Physiol (2012) 463:5371

pons and hypothalamus, which will be separately dealt with


later in this review.

Local network regulation of sleep


Any attempt to explain the neurophysiological models
evolving around mechanisms of sleep-related activity and
sleep-inducing circuits has to involve the biochemical
mechanisms that also help to keep track of past sleep
wake activity [85], accounting for a sleep-regulating
homeostatic component [144]. This homeostat of nonrapid-eye-movement (NREM) sleep is made up by molecular networks of several sleep regulatory substances [86],
including tumour necrosis factor- (TNF-), interleukin-1
(IL-1), growth hormone-releasing hormone (GHRH),
prostaglandin D2 (PG), and adenosine (for review, see
[86]). Moreover, brain-derived neurotrophic factor (BDNF)
has been suggested to play a key role in sleep homeostasis
[67], and it was suggested that regulation of cortistatinexpressing interneurons by activity-dependent BDNF
expression may contribute to the regulation of sleep
behaviour [103].
While adenosine acts on thalamic and basal forebrain
neurons to promote sleep (see also [7]), GHRH, TNF-,
and IL-1 act directly on hypothalamic preoptic neurons to
promote NREM sleep [127]. TNF- and IL-1 also act on
the locus coeruleus [40] and IL-1 additionally acts on
raphe serotonergic neurons [99] and on GHRH-receptive aminobutyric acid (GABA)-containing neurons in the
hypothalamus [41].
Sleep regulatory substances are produced in response to
cellular activity and metabolism, and sleep loss increases
IL-1, TNF-, and adenosine levels, while antagonists of
those substances given before sleep deprivation reduced
rebound sleep. Different time courses in the mechanisms of
action of sleep regulatory factors were suggested to allow
the brain to (a) induce activity-dependent oscillations that
are related to local sleep-like states of neuronal assemblies
(e.g. cortical columns) and (b) monitor past neuronal
network activity to react with increased sleep or wakefulness [86]. Transcription and translation events that lead to
the production of circadian clock proteins such as Cry1, 2,
PER1, 2, BMAL, or CLOCK require hours or days, and
would allow tracking sleepwake history on a longer
timescale [87, 2]. Labile substances like adenosine or the
gaseous nitric oxide (NO) are sleep regulatory substances
which will have more direct effects in this scenario, altering
blood flow, second messenger systems, or metabolic rates.
Finally, neurotransmitters act within milliseconds and can
alter synaptic transmission, and thus, the state of a neuronal
assembly. It was proposed that such a biochemical network
operates in cortical columns and other neuronal assemblies

55

to affect the state [86]. The homeostatic drive to sleep


regulation by accumulation of sleep regulatory substances
has certainly given insight into how the brain is organized
to produce sleep.
Taken together, many findings support the hypothesis
that (a) sleep is initiated within the cortex after local
production of several sleep regulatory substances and that
(b) whole organism sleep is coordinated via cortical
connections with the subcortical sites. Sleep regulatory
circuits may integrate information about the state of cortical
columns with information about the time of day, sensory
input, mental activity, or other determinants of a whole
animal sleepwake state.

The thalamus: an interface to sleep regulation


The thalamocortical network: centres and distributed
systems
It is obvious that neither wakefulness nor any sleep state
can be solely explained by the action of a single centre (see
[176]) but can rather be expected to be the result of an
interplay of many centres and distributed systems. In fact,
sleep is characterized by synchronized events in a multitude
of synaptically coupled neurons in thalamocortical systems
[177], and research has unravelled a number of cellular and
network mechanisms that cause this activity during sleep,
or blocks it during wakefulness. It was already mentioned
that whole organism sleep appears to be coordinated via
cortical connections with the subcortical sites. It is unclear
whether decentral sleep regulation, as found in cortical
neuronal assemblies, can also be found in thalamic
neuronal assemblies. Investigations that clarify whether
such functional subunits exist in the thalamus will most
likely help us understand the role of the thalamus and the
concept of decentral sleep regulation in whole organism sleep.
The thalamocortical system is comprised of three major
brain areas: The cortex, the nucleus reticularis thalami
(NRT), and the dorsal thalamus (Fig. 1). Oscillations,
typical of sleep states, are generated as the result of
synaptic interactions between three major cell types [177]:
inhibitory GABA-containing neurons of the NRT, thalamocortical relay neurons (TC neurons), and cortical pyramidal
neurons. In a simplified schema, TC neurons project to
corresponding (e.g. sensory) cortical areas and terminate on
neurons of layer four. These projections also form collaterals that terminate on NRT neurons. The cortical pyramidal
neurons of layer six project back to the dorsal thalamus and
the NRT in a topographical manner. Consequentially, NRT
neurons receive excitatory inputs from both the thalamus
and the cortex via thalamocortical and corticothalamic
projections. NRT neurons project back to the thalamus

56

Pflugers Arch - Eur J Physiol (2012) 463:5371

LTS) crowned by several fast Na+/K+ APs, which probably


interrupt the relay of sensory information [71, 105, 177].
This rhythmic oscillatory bursting provides an important
mechanism in the generation of synchronized oscillatory
activity in the forebrain, which is represented as sleep
spindle oscillations and slow/delta waves in the EEG during
quiet sleep.
The nucleus reticularis thalami

Fig. 1 Schematic view of the thalamocortical system. Modified from


Sherman and Guillery ([163], J. Neurophysiol., Am. Physiol. Soc.,
modified with permission). The system is comprised out of three
major brain areas: the cortex, the nucleus reticularis thalami (NRT),
and the dorsal thalamus. The lateral geniculate nucleus of the thalamus
contains at least two types of neurons: thalamocortical relay (TC)
neurons (red ellipse) and local interneurons (green ellipse). The TC
neurons project to the cortex (layer IV) and collaterals terminate on
GABAergic NRT neurons. Cortical neurons (layer VI) project to the
thalamus and also branch collaterals that terminate on NRT neurons.
The GABAergic NRT neurons project back to the thalamus but not to
the cortex. Thalamic neurons receive inputs from the basal forebrain
(BF), the parabrachial nucleus (PBR), the nucleus of the optic tract
(NOT), and the sensory organs (e.g. the retina, RET)

and also innervate other neurons of the NRT, but they do


not project back to the cortex [56, 142]. Thalamic neurons
receive inputs from the basal forebrain, the parabrachial
nucleus, the nucleus of the optic tract, and the sensory
organs. For a more comprehensive description of the
organisation of the thalamus, including first and higher
order relay nuclei as well as driver and follower neurons,
the reader is referred to a series of excellent review articles
and books [63, 77, 92, 161164, 191].
Thalamic neurons show two distinct states of electrogenic activity that allow the state-dependent gating of
sensory information during sleep and wakefulness (Fig. 2).
During wakefulness TC neurons are depolarised and show
tonic activity of fast Na+/K+-mediated action potentials
(APs) [175], which is probably the basis for a faithful relay
of sensory information from the sensory organs through the
thalamus to the primary sensory areas of the cortex. During
periods of slow-wave sleep or drowsiness TC neurons are
hyperpolarised and show rhythmic, rather stereotyped,
bursts: low-threshold Ca2+ potentials (low-threshold spikes,

The NRT is a central modulatory site for the thalamocortical


relay and has been considered to be a pacemaker of slow
oscillations in the thalamus. It is a key determinant of the
internal attentional searchlight and is thought to be involved
in the generation of major brain rhythms, including those
observed during sleep, dreaming, and wakefulness [91, 175,
177]. It has widespread topographical and functionally
structured synaptic projections into the thalamus [62, 143]
and receives strong inputs from collaterals of thalamocortical, and in particular, from corticothalamic projections,
while projections from the NRT never target the cortex [56,
142]. The basal forebrain projections to the thalamic
reticular nucleus have a desynchronising effect by blocking
synchronous spindle oscillations [174]. However, it is still
unclear whether the NRT by itself is capable of generating
sleep spindles [39] or whether an interaction between NRT
and TC neurons is essential for the genesis of these
oscillations [5, 69]. The NRT was investigated in stimulation, lesion, isolation, computer modelling, and cellular
studies [5, 25, 27, 35, 45, 56, 173]; for review see [142]. It
meets the definition of a centre since it is an anatomically
defined and localized set of neurons that appear to serve a
single (albeit complex) function. Moreover, it was suggested that it consists of a single type of GABAergic
neurons. In contrast, several subtypes of neurons have been
identified by some groups [90, 168, 169] in rodents, but
others have shown uniform cellular properties and found no
basis for a subdivision into more than one cell class [15, 27,
128]. The neurons of the NRT are electrically coupled by
gap junctions (electrical synapses) of the connexin 36 type.
It was shown that these gap junctions can mediate a
modulation of burst firing states, a synchronization of
electrical activity in general, and could even convey a
switch between bursting and tonically spiking mode of
activity from one neuron to another [88, 89]. Gap junctions
allow a low pass filtering by conveying slower signals more
efficiently than fast signals. This is of particular interest in
thalamic neurons that display rebound LTS: the fast APs
have little influence on the postsynaptic neuron, despite
their large amplitude, while the much smaller but slower
Ca2+ potential is passed on more efficiently. This could
allow slow rhythms to spread across the nucleus without
being influenced by fast AP firing [89].

Pflugers Arch - Eur J Physiol (2012) 463:5371

57

Fig. 2 Thalamic neurons exhibit two states of activity. During sleep


TC neurons are hyperpolarised and show stereotyped, Ca2+-mediated
bursts (left). During wakefulness TC neurons are depolarised and
show tonic activity of fast Na+/K+-mediated action potentials (right)
which is probably the basis for a faithful relay of sensory information
(see text for details). Thalamocortical synchrony required for sleep
arises out of an interplay between several mechanisms that ultimately
suspend the sensory relay: suppression of mechanisms that support
tonic firing or activation of mechanisms which uphold oscillatory
activity. The ascending brainstem system governs the switch from
burst to tonic firing by the release of the neurotransmitters
acetylcholine (ACh), noradrenalin (NA), and serotonin (5-hydroxytrytamine, 5-HT) that predominantly act on two types of ion currents:

ITASK and Ih [93, 110]. Both reciprocally interact as major determinants of the resting membrane potential [113]. The brainstem transmitters inactivate ITASK and shift the voltage dependency of activation
of Ih to more depolarised potentials. Consequently, the membrane
depolarises and the input resistance increases, thereby favouring the
tonic firing mode. Orexins, histamine, and nitric oxide (NO) exert a
similar influence, while adenosine (but also volatile anaesthetics such
as halothane) suppresses tonic firing by a local modulation of these
ion channels. GABAergic input from the NRT or local interneurons
provide a hyperpolarisation that allows a de-inactivation of T-type Ca2
+
channels and activation of Ih, which then allows a cyclic interaction
that generates rhythmic burst activity (see Fig. 3 for details)

Gap junctions are known to be regulated by numerous


parameters such as intracellular pH, transjunctional voltage,
and intracellular Ca2+ [24]. It seems likely that electrical
synapses of the NRT play a major role in the generation
and/or maintenance of sleep-related activity in the thalamocortical system. Further studies are required to elucidate
the role of this both highly complex and remarkably simple
nucleus and its unique characteristics.

neurons, whose activity generalizes by progressive recruitment of neurons in the thalamocortical network. This
recruitment is thought to underlie the waxing phase of a
sleep spindle wave [26, 83]. The rhythm itself is thought to
base upon the reciprocal interactions between NRT neurons
and interconnected TC neurons, orchestrated in their
spatiotemporal pattern through corticofugal influences (see
Fig. 1) [25]: Individual TC neurons follow network
oscillations due to a fine tuned interplay between
GABAergic inhibitory postsynaptic potentials (IPSPs)
that are received from NRT neurons, and two types of
voltage-gated membrane conductances, which are as follows: the T-type Ca2+ current (IT) and the hyperpolarisationactivated cation current (Ih; see Figs. 2 and 3) [107, 110,
177].
Additionally, a long-lasting potassium-dependent IPSP
mediated by GABAB receptors primes TC neurons for burst
firing by activating low-threshold calcium potentials [32].
During spindling, NRT neurons rhythmically convey
GABAergic IPSPs onto TC neurons at 714 Hz. The
membrane hyperpolarisation provided by the IPSPs causes
a de-inactivation of IT, paralleled by an activation of Ih. The
repolarisation then activates IT again, triggering a so-called
rebound burst, representing the already mentioned LTS
which is crowned by several fast Na+/K+ APs. The

Slow-wave sleep oscillations


Slow synchronous oscillations of electrical activity, such as
delta waves and spindle waves, are commonly found in
thalamocortical circuits and are thought to underlie synchronized EEG states, such as slow-wave sleep [31, 39,
177]. During slow-wave sleep, sleep spindle oscillations
occur frequently and are a well-studied example of a
synchronized activity between neuronal ensembles [107,
177]. The synchronized waves of activity appear at 7
14 Hz with gradually increasing (waxing) and decreasing
(waning) amplitudes during early stages of slow-wave
sleep. Sleep spindle waves are grouped as sequences of
12 s that recur periodically every 320 s [107, 177, 195].
There is a common agreement that these oscillations are
initiated by a small number of reciprocally interconnected

58

Pflugers Arch - Eur J Physiol (2012) 463:5371

Fig. 3 Thalamocortical relay neurons: intrinsic properties determine


sleep-related activity. Hyperpolarised membrane potentials activate the
hyperpolarisation-activated cyclic nucleotide-gated cation (HCN)
channels and de-inactivate T-type Ca2+ channels. HCN channels then
provide a depolarisation that, in turn, triggers a transient current
through T-type Ca2+ channels. The hyperpolarisation-activated cation

conductance Ih has been shown to undergo a positive shift of the


activation curve when the intracellular concentration of cAMP rises
[17, 133]. This occurs when an increase of intracellular Ca2+ increases
the adenylyl cyclase activity, causing, in turn, the intracellular cAMP
concentration to increase [95]

rhythmically occurring IPSPs also activate Ih, providing at


first a depolarisation that can activate IT but also causing a
stronger depolarisation that leads to inactivation or prevents
de-inactivation of IT, ultimately terminating the spindling
oscillation in TC neurons [6].

issue [33, 34]. Many more studies show that besides being
a charge carrier in rhythmic burst discharges during slowwave sleep, Ca 2+ is also crucially involved in the
generation and maintenance of sleep-related activity (for
review see [138, 175]):

The role of Ca2+ in sleep-related activity of thalamic


neurons

Ca2+ and the hyperpolarisation-activated cation channels


in TC neurons

Intracellular Ca2+ ions play an important role during


rhythmic bursting of TC and NRT neurons in sleep-related
activity [20, 27, 35, 138]. This is not surprising, since Ca2+
controls numerous neuronal functions, including transmitter
release, neuronal excitability, and synaptic plasticity [10].
For thalamic neurons, one of the main sources of Ca2+ is its
influx via T-type Ca2+ channels (for review, see [21]). These
channels play a role in membrane potential bistability
observed in thalamic neurons (i.e. the existence of two
stable resting membrane potentials). They give rise to the
window current, a key component in the generation of the
slow (<1 Hz) sleep rhythm also reviewed in this special

The hyperpolarisation-activated cyclic nucleotide-gated cation (HCN) channels give rise to Ih. A regulation of Ih by
intracellular Ca2+ has been shown to be particularly
important during sleep spindling and slow-wave sleep [43,
44, 181]. In an attempt to provide a basis for the waxing and
waning behaviour of sleep spindles, a mathematical model of
synchronized oscillations in thalamic networks has implemented an activity-dependent up-regulation of Ih by intracellular Ca2+ (entering via T-type Ca2+ channels) in TC
neurons [43]. Here, a Ca2+ influx through the T-Type Ca2+
channels and the resulting positive shift of the activation
curve of Ih will increase the available Ih-mediated conduc-

Pflugers Arch - Eur J Physiol (2012) 463:5371

tance at membrane potentials near rest. This would facilitate


Ih recruitment by IPSPs, favouring the termination of sleep
spindle waves. Removal and/or sequestration of Ca2+ by
plasmalemmal or endoplasmic Ca2+ pumps, or by binding to
intracellular Ca2+ buffers during the quiescent period then
allows the build-up of a new sleep spindle, possibly
contributing to the typical waxing and waning behaviour. A
direct influence of intracellular Ca2+ was absent when using
the whole cell patch clamp technique [17], consistent with a
lack of intracellular regulatory Ca2+ binding sites on the
channel protein [94]. However, an indirect influence of Ca2+
on Ih exists: (a) an increase of intracellular Ca2+ increases the
adenylyl cyclase activity, causing (b) the intracellular cAMP
concentration to increase [95], and in turn, (c) causes a shift
of the activation curve of Ih towards more positive potentials
[17, 133] (see Fig. 3).
Ca2+-induced Ca2+ release in the NRT
Intracellular Ca2+ ions were also suggested to both support
and dampen repetitive burst firing in NRT neurons [27, 35].
In dendrites of NRT neurons, a signalling triad was
described that involves (a) the T-type Ca2+ current, (b) a
Ca2+-activated K+ current of the SK2 type, and (c) Ca2+
pumps of the SERCA type (sarco/endoplasmic reticulum
Ca2+ ATPase; see Fig. 4). It was shown that a selective
interaction between T-type Ca2+ channels and SK2 K+
channels is based on the spatial coexpression of both
channel types and that this forms the basis for the robust
oscillatory properties of NRT neurons [35]: an LTS and the
accompanying Ca2+ influx through T-type Ca2+ channels,
the largest inlet for Ca2+ in NRT neurons, cause a
depolarisation and an increase of the intracellular Ca2+
concentration, which in turn triggers the opening of SK2
channels, K+ efflux, and a subsequent hyperpolarisation as
IT inactivates [35]. This will also cause de-inactivation of Ttype channels, so that repolarisation to the resting membrane potential then triggers another LTS by once more
activating T-type Ca2+ channels. SERCAs influence this
oscillatory interplay by competing with the SK2 channels
for incoming Ca2+. When the sequestration of Ca2+ into the
ER was prevented by selective SERCA blockers, more
rebound bursts were observed. Mice lacking the SK2
channels showed a marked decrement of EEG power in
the SWS and spindle bands. Moreover, the mice displayed a
disturbed and fragmented NREM sleep [35].
Because of its unique set of features, the NRT is ideally
suited to prevent hyperexcitatory tendencies in the thalamocortical system. Thus, it is feasible to assume that a system to
dampen oscillatory burst discharges exists. During rhythmic
oscillatory bursting in NRT neurons, Ca2+-induced Ca2+
release (CICR) via ryanodine receptors occurs in somatic
regions (Fig. 4) [27]. This released Ca2+ was shown to exert

59

a dampening effect on oscillatory activity by reducing the


capability of NRT neurons to follow a given frequency of
oscillations. This was proposed to be a physiological means
to prevent stereotypic oscillations at low frequencies, but it
may also be a source of regulation for sleep-related activity.
Depletion of intracellular Ca2+ stores increases the ability of
NRT neurons to repeatedly generate rebound bursts in an
SK2 independent manner [27]. It has recently been shown
that somatic burst discharge cause large Ca2+ influx into
distal dendrites of NRT neurons, providing a physiological
quantification of the distribution of T-type Ca2+ channels in
NRT neurons [29]. Because of the dendritic localisation of
the majority of Ca2+ channels activated during an LTS, the
physiological relevance of somatically expressed subtype 3
ryanodine receptors (RyR3) in NRT neurons is questionable
[48]. However, the classical fast CaV3.1 and 3.2 appear to be
somatically expressed [72] and RyR2, that also seem to be
present in NRT neurons [27] are more diffusely expressed.
The differential contribution of these channels to Ca2+
signalling in NRT neurons is unclear. Moreover, if dampening repeated burst discharges is a physiological function of
somatic RyR3, a separation from the site of the largest Ca2+
entry could be necessary to enable a specific (and delayed)
response to longer lasting series of bursts, as opposed to
immediate responses to a single burst.
Ca2+-induced Ca2+ release in the dorsal thalamus
In TC neurons, a CICR via ryanodine receptors was shown to
support the tonic firing mode and thus suppress sleep-related
activity [20]. Unlike neurons of the NRT, which show CICR
after repeated rebound bursting, and thus show a CICR
largely based on Ca2+ influx via T-type Ca2+ channels (see
Fig. 4), TC neurons showed CICR after tonic firing, and thus
depended largely on Ca2+ influx through high-voltageactivated Ca2+ channels (see Fig. 5), probably of the L-type.
While NRT neurons lose their ability to create repeated burst
discharges when CICR is active, TC neurons are supported in
tonic firing by CICR in a positive feedback manner.
Finally, in thalamic midline neurons CICR also occurred
upon activation of T-type Ca2+ channels [152]. Similar to
NRT neurons, the transient nature of the T-type current
required repeated bursts to cause a build-up of [Ca2+]i,
indicative of CICR. This much resembles the reaction to
oscillatory activity in the NRT.
Intracellular Ca2+ signalling across thalamic nuclei
regulates sleep-related activity
Both low-voltage-activated Ca2+ channels of the T-type and
high-voltage-activated Ca2+ channels of the L-type thus
represent a mechanism to trigger CICR, however under
different circumstances. The reason for the different

60

Pflugers Arch - Eur J Physiol (2012) 463:5371

Fig. 4 Ca2+ in the nucleus reticularis thalami. Ca2+ influx through Ttype Ca2+ channels in dendrites of NRT neurons was shown to activate
Ca2+-activated K+ channels of the SK2 type to regulate the strength of
the endogenous NRT oscillation [35]: depolarisations by Ca2+ influx
are followed by hyperpolarisations due to K+ efflux through SK2
channels. The hyperpolarisation de-inactivates T-type Ca2+ channels
and the cycle starts again. Sarco-(endo-)plasmic reticulum Ca2+
ATPase (SERCA) competes with SK2 channels for cytosolic Ca2+,
which interrupts the cyclic activity. Ca2+-induced Ca2+ release (CICR)

in NRT neurons probably takes place in somatic regions by a cell-type


specific coupling of T-type Ca2+ channels and ryanodine receptors.
Ca2+ released via ryanodine receptors (RyR) shapes the bursting
activity of NRT neurons by an as yet unknown mechanism that does
not involve the SK2 channels but could involve the Ca2+-activated
unselective cation (CAN) channels. RyR could provide an out-of-phase
activation of Ca2+-activated conductances outside the cyclic
interaction between T-type Ca2+ channels and SK2 channels

triggering of CICR (L-type or T-type Ca2+ channels in TC


or NRT/Midline neurons, respectively) might be found in a
close spatial coexpression of plasmalemmal and endoplasmic Ca2+ channels, but more research in this area is needed,
e.g. using high resolution Ca2+ imaging in subcellular
compartments, in order to clarify this.
The control of intracellular Ca2+ varies between the cell
types: Blocking the SERCAs did not alter the basal cytosolic
Ca2+ levels in TC neurons, indicating that TC neurons rely on
endogenous Ca2+ buffers [165]. In contrast, NRT neurons rely
on the endoplasmic reticulum, showing Ca2+ changes when
this mechanism is compromised by pharmacological inhibition of SERCAs [27]. This could allow a fine tuning of the
Ca2+-release-to-effector-coupling and could aid to explain
the reciprocal effects of CICR in TC and NRT neurons.
The mechanisms that couple CICR and the released Ca2+
to the effector systems are not yet known. The observed
changes in spike width and afterhyperpolarisation after
blocking ryanodine receptors in TC neurons [20] point to a
Ca2+-activated K+ current termed IC or IK(Ca) and the
corresponding BK channels. Ca2+-activated K+ channels
are known to exist in TC and NRT neurons [19, 70, 179]. In

particular, IK(Ca) has been described in thalamic neurons


[141] and was found to be involved in the generation of
intermittent tonic series of APs in a model of TC neurons
[183]. Firing responses to threshold current injections were
inhibited by a BK Ca2+-activated current, and the AP firing
rate was limited by the activity of a SK Ca2+-activated
current in TC neurons [81]. Ca2+-activated currents have,
however, only negligible effects on LTS high frequency
firing in TC neurons [184]. This was different in local
interneurons [197] and NRT neurons [5]. The all but absent
adaptation in TC neurons (see [20]), and the stable tonic
firing with intact CICR, points to a very transiently active
K+ conductance that is tightly coupled to changes in
intracellular Ca2+.
In NRT neurons, ryanodine receptors may prevent stereotypic oscillatory behaviour by detecting repeated Ca2+ influx,
the released Ca2+ subsequently leading to a permanent
activation of Ca2+-activated K+ channels different from the
abovementioned SK2 channels. These channels could then
counteract oscillations by a hyperpolarisation that is out of
phase with the oscillatory activity, as well as a shunting
inhibition. A possible candidate for this is ICAN, a Ca2+-

Pflugers Arch - Eur J Physiol (2012) 463:5371

61

Fig. 5 Ca2+ induced Ca2+


release modulates activity mode
of relay neurons. Thalamocortical relay neurons of the dorsal
part of the lateral geniculate
nucleus are modulated in their
state of activity by CICR [20].
High-voltage-activated Ca2+
channels, probably of the
L-type, were coupled to RyR
and CICR supported the relay
mode of activity. The
mechanism of action is as yet
unknown but probably involves
Ca2+-activated K+ channels of
the BKCa type. Ca2+ ions
provide a negative feedback
mechanism termed Ca2+
dependent inhibition (CDI) of
High-voltage-activated Ca2+
channels. TASK channels
contribute to this
scenario by largely determining
the resting membrane potential

activated non-selective cation conductance, which has been


shown to exist in NRT neurons [5, 45] and could play an
important role in the attenuation of rhythmic burst firing.
Intracellular Ca2+ can modulate a range of other potential
targets that could help to control sleep-related activity. The
inhibition of gap junction mediated electrical coupling (see
above) by an increase of the intracellular Ca2+ concentration (for review see [42]) could constitute a mechanism to
prevent the spread of oscillatory activity across the nucleus.
Vice versa, a depletion of intracellular Ca2+ stores could
lead to an increased generation of oscillatory burst activity
in the NRT, supporting its role as a pacemaker of
thalamocortical synchrony [56], as well as a suppression
of tonic firing in TC neurons. This leads to the question of
how intracellular Ca2+ stores in thalamic neurons are
replenished. Future studies will have to determine, whether
store-operated ion channels such as ICRAC can support tonic
activity or prevent oscillatory bursting in thalamic neurons.
Ca2+-induced Ca2+ release in TC neurons was also shown
to be facilitated by cyclic ADP ribose [20]. The production of
cyclic ADP ribose is catalyzed by ADP ribosyl cyclase, which
in turn is stimulated by cyclic GMP [60]. Increases of
intracellular cyclic GMP levels can be induced by the
activation of guanylyl cyclase via NO [84]. There are multiple
potential sources of NO in the thalamus: cholinergic terminals
originating from brainstem nuclei, GABAergic NRT neurons,
and local GABAergic interneurons [193]. NO probably
modulates TC neurons by activation of the cyclic GMP
system, which then causes a positive shift of the activation
curve of Ih [135], thereby controlling oscillatory activity.

NO causes a robust, long-lasting depolarisation in NRT


neurons via an ACh and cGMP-dependent pathway, inhibiting
burst activity [192, 193]. A Ca2+-dependent activation of Ih
in NRT neurons could occur, providing a depolarisation that
lifts the membrane potential out of the range of the bursting
mode. Additionally, a shunting inhibition would take place
and could reduce the efficacy of both intra-NRT GABAergic
signalling and cortical glutamatergic inputs.
Sensory responses and responses to excitatory amino
acids are facilitated after NO-induced increases of the
intracellular cGMP concentration in TC neurons [160],
interrupting oscillatory burst activity and enhancing the
relay mode of activity in TC neurons. An increase of cGMP
levels can be assumed to also result in an increased
production of cyclic ADP ribose, facilitating CICR [20],
and thus further strengthening the relay mode of activity. A
similar mechanism has not yet been shown in NRT neurons
but would support the relay mode of TC neurons by
dampening oscillatory activity in NRT neurons. In this way,
the neurotransmitters of the ascending brainstem system
may be functionally coupled to the intracellular Ca2+
release mechanisms described in NRT and TC neurons.
Anaesthesia and sleep cellular mechanisms in the thalamus
Anaesthesia is often intuitively compared to sleep, although
the EEG patterns during anaesthesia neither show distinct
stages (or cycling between stages) observed during natural
sleep, nor does natural sleep involve an isoelectric EEG, as
sometimes observed under deep anaesthesia. There are,

62

nevertheless, some parallels between sleep and anaesthesia.


Apart from behavioural similarities (or rather the absence of
behaviour), anaesthetic effects at specific molecular targets
in specific brain regions cause activity changes that much
resemble sleep-related activity [125]. The majority of
molecular targets are ion channels modulated by clinically
effective concentrations of volatile anaesthetics (for review
see [22]). These anaesthetics enhance inhibitory postsynaptic channel activity (GABAA and glycine receptors [115])
and inhibit excitatory channel activity (acetylcholine,
serotonin, and glutamate receptors [54, 55]).
Many anaesthetics have been shown to modulate ion
channels in the thalamus, in a way that could explain
anaesthetic action. In TC neurons, in a mutually cooperative
way, two ion channel systems are altered in the presence of
clinically relevant concentrations of a range of volatile
anaesthetics [18]: (a) The activation curve of Ih is shifted
towards more negative potentials and the maximally available current is reduced, thus rendering a lower portion of
available Ih at membrane potentials near rest. This removes a
depolarising component (Na+ influx through HCN channels)
and thus causes hyperpolarisation, hindering tonic firing. (b)
In parallel, a K+ current, most likely carried by TASK
channels, is increased in the presence of volatile anaesthetics.
Both mechanisms may contribute to the breakdown of
sensory transmission and a loss of conscious awareness.
The widespread distribution of these ion channels led to the
suggestion that this is an essential part of the clinical action
of volatile anaesthetics [18]. Whether a similar modulation
occurs during natural sleep remains to be determined.
In neurons of the ventrobasal complex (VB), the intravenous anaesthetic pentobarbital reduced the input resistance at
potentials near the AP threshold [189]. While it inhibited Ih, it
increased IKir and the K+ leak current. At the same time it
increased the open probability of GABAA chloride channels,
modulating the time course of inhibitory postsynaptic
currents. All of these modulations will ultimately disable
the sensory relay by preventing tonic firing of APs.
The intravenous general anaesthetic propofol inhibits the
thalamic sensory relay by a shunting inhibition at clinically
relevant concentrations. This is primarily mediated by a
potentiation of the GABAA receptor chloride channelmediated conductance [194]. As a result, inhibitory postsynaptic currents evoked in VB neurons by electrical stimulation
of the NRT were markedly increased in amplitude, decay
time, and charge transfer.
The T-type Ca2+ channel has also been shown to be
modulated by a whole range of anaesthetics [7274, 167,
182]. In TC neurons the application of the anaesthetic
alcohol 1-octanol reversibly blocked burst activity and
decreased the cell input resistance [167] enabling a shunting
inhibition. Such a shunting inhibition (together with a
hyperpolarisation) was the main cause for the inhibition of

Pflugers Arch - Eur J Physiol (2012) 463:5371

tonic firing of APs in TC neurons after the application of the


volatile anaesthetic isoflurane [153, 154].
The effects of both isoflurane and 1-octanol have also
been studied in NRT neurons, where excitability was found
to be reduced in the presence of isoflurane [75] and native
T-channels were inhibited by 1-octanol [74]. The mechanism of T-channel inhibition was proposed to be a
stabilization of the inactive states of the channel as well
as an inhibition of the Ca2+-dependent PKC signalling
pathway, which was proposed to be the molecular substrate
for a modulation of T-channels.
Although sleep and anaesthesia must be regarded as
different brain states, the underlying mechanisms could be
similar and ultimately have similar effects. Anaesthetic action
as a model for investigating mechanisms of sleep induction
can deliver valuable insights: The effects at overlapping
specific molecular targets in specific brain regions seem to
cause similar changes in anaesthesia and sleep-related activity.
G-Proteins as a central element in thalamic physiology
The switch between activity modes of thalamocortical relay
neurons is mediated via neurotransmitters released from
brainstem cholinergic nuclei (acetylcholine), the locus coeruleus (noradrenalin), and the raphe nuclei (serotonin). In a
similar way, the arousing action of glutamate, released from
corticothalamic axons, is mediated by metabotropic glutamate
receptors [157]. A common action of these neurotransmitter
systems is a depolarising shift of the membrane potential,
causing rhythmic bursting to cease and tonic activity to
commence. These transmitter systems all seem to converge
onto Gq/G11 family G- proteins. Yet, spatial separation and
state-dependent differences in activation seem to allow
divergent action of several metabotropic receptor classes that
seem to utilize the same pool of Gq/G11 family G-proteins.
This is strongly supported by the finding that these transmitters mutually saturate their responses, pointing to a
scenario in which Gq/G11 family G-proteins could function
as a bottleneck in thalamic physiology [16, 28].
Developmental regulation of HCN channels
and sleep-related activity
In the mammalian ontogeny, REM and NREM sleep are
thought to develop either from two different immature sleep
states or one common form of early sleep, termed presleep
[12, 13, 53]. When the maturation of presleep into REM
and NREM sleep is complete, state-dependent differentiated
activity can be observed in a neocortical EEG. In young
rodents (<~p12), slow-wave activity is not observed, and
rhythmic thalamocortical activity appears no earlier than at
postnatal day 12 [78, 100, 158]. A recent study points to a
scenario in which Ih, as an essential molecular prerequisite

Pflugers Arch - Eur J Physiol (2012) 463:5371

for intrinsic oscillatory activity in TC neurons, gradually


matures, and as a threshold of current density is reached,
enables the appearance of delta activity in single TC
neurons, and thus in the thalamocortical network [80]. This
is corroborated by computer modelling, which suggests that
this threshold is reached between p12 and p14, and by
dynamic clamp studies in cat TC neurons, which show that
the occurrence of spontaneous delta activity depends on the
presence of an artificially introduced Ih of sufficient
amplitude [68]. A maturation of Ih contributes at least
partially to the development of slow-wave sleep rhythms
and the establishment of adult sleepwake patterns in rats.
Local thalamic interneurons and sleep-related activity
Intrinsic properties of thalamic interneurons
Thalamic local circuit interneurons utilize GABA as a
major neurotransmitter and possess a unique combination
of electrophysiological and morphological properties that
distinguishes them from TC neurons. These properties
include relatively short duration APs, a relative lack of
low-threshold Ca2+-mediated burst activity, a high input
resistance and small, oblong dimensions [14, 109, 122,
137]. They make up as much as 25% of neurons in the
dorsal thalamus [50, 57, 61, 76, 116, 129]. The ascending
fibres from the brainstem and the corticofugal axons are
capable of controlling the strength and pattern of intrathalamic inhibitory mechanisms [106, 176, 177]. GABA
release in these interneurons occurs in both axons and
dendrites and while Na+ APs evoke widespread Ca2+
signals throughout dendritic and axonal arbours, Ca2+
spikes mediated by L-type Ca2+ channels specifically
trigger dendritic GABA release [1]. These Na+ and Ca2+
spikes regulate the rapid and prolonged release of GABA
from interneurons, determining spatial and temporal features of the feedforward inhibition [11]. Additionally,
cholinergic input modulates the disynaptic feedforward
inhibition in a context-dependent manner [4].
Although evidence is accumulating that interneurons play a
vital role in the shaping of thalamic activity, very little is
known about the role these neurons play during sleep. For
cortical interneurons, cortistatin, a neuropeptide expressed
primarily in a subset of cortical GABAergic interneurons was
shown to be implicated in sleep homeostasis [103]. Moreover,
a subpopulation of cortical GABAergic interneurons, which
express the enzyme neuronal nitric oxide synthase (NOS),
has recently been found to be activated during SWS and
these neurons were suggested to be the anatomical link for
understanding homeostatic sleep regulation [59, 82]. Investigating interneuron function in producing sleep-related
activity in the thalamus could possibly help to identify
similar mechanisms.

63

Modulation of thalamic interneurons in the thalamocortical


system
Activation of the ascending cholinergic pathways seems to
exert an inhibitory influence upon interneurons of the
dorsal part of the lateral geniculate nucleus (dLGN) in cats:
Stimulation of the region of the brainstem containing these
cholinergic inputs results in a reduction of IPSPs in TC
neurons [37, 52, 166] and application of ACh to interneurons hyperpolarises the membrane potential, presumably
through a K+ conductance increase [109]. The activity of
interneurons in the dLGN may be strongly influenced by
activity in the retina and the cortex (by ionotropic glutamate
receptors), other interneurons (through GABAA receptors,
and to a lesser degree, GABAB receptors), and the
ascending cholinergic system, while activity in the noncholinergic brainstem or in hypothalamic systems seem to
exert a more subtle influence [137].
The brainstem modifies intrathalamic inhibitory influences
The intrinsic capabilities of these interneurons to generate
tonic series of APs at high frequencies can be assumed to
support the linear transformation of synaptic signals [137],
supporting wakefulness. The release of ACh from brainstem cholinergic neurons should result in a reduction of
neuronal excitability and decreased AP discharge in
interneurons and NRT neurons, thus reducing overall
intrathalamic inhibitory influences on TC neurons.
Thalamic interneurons support network oscillations
GABAergic neurons in the thalamus were proposed to
allow reinforcing corticothalamic input by providing a
phase-locked hyperpolarising influence on rhythmically
active relay neurons, sustaining and synchronizing individual neurons delta activity into a large scale network
oscillation [172]. Interestingly, although interneurons in
the dLGN do not show an LTS, they do possess T-type Ca2+
channels. Burst firing is prevented by a functional balance
between IT and the fast transient K+ current IA [79, 134]. It
was speculated that regulatory influences are capable of
shifting the balance between these two currents. This could
cause the selective appearance of burst or burst-like activity
in interneurons and would thus support their function in
synchronizing slow oscillatory burst firing of the synaptic
network [134].
Thalamic interneurons and dopamine
Local interneurons are also essential players when it comes
to the action of the neurotransmitter dopamine. The role of
dopamine in the regulation of the sleepwake cycle as well

64

as wake-active dopaminergic neurons has recently been


established (for review see [117]). However, the role of
dopamine in the regulation of sleep-related activity in the
thalamus is unclear. It was suggested, that dopamine acts by
indirectly inhibiting relay cell activity through the excitation of local interneurons [3, 196]. Indeed, dopamine
modulates the inhibitory synaptic transmission to dLGN
TC neurons by activation of D2-like receptors on local
GABAergic interneurons [123]. As a functional consequence, increased activity of the dopaminergic system
would facilitate GABAergic interactions through a depolarisation of GABAergic interneurons and the increase in
frequency of GABAergic synaptic events.

Possible decentral mechanisms in the thalamus link


to the brainstem and cortex
Many key players and potential candidates to orchestrate
local neuronal assemblies into sleep-like states in a
decentral manner seem present in the thalamus. Prominent
examples of such key players that act in concert in the
brainstem, the thalamus, and the cortex are the ATP
metabolite adenosine, NO, and glia cells. The energy
metabolism is among the most essential functions of life
and could represent the core of sleep need [146]. Adenosine
provides a direct link between cellular processes and the
energy metabolism of neurons.
Adenosine
Injection of adenosine into the ventricle of a cat brain
caused a state that closely resembled natural sleep of
30 min duration [49]. Pharmacological studies revealed that
adenosine agonists promoted, but antagonists such as
caffeine, inhibited both NREM and REM sleep [66, 170].
Increased neuronal activity enhances energy consumption and, in turn, increases extracellular adenosine concentrations throughout the brain (for review, see [86]). This
decreases neuronal activity in several brain areas, among
them the basal forebrain, through A1 adenosine receptors.
Reduced neuronal activity also reduces the need for energy,
allowing intracellular ATP stores to be replenished, extraand intracellular adenosine concentrations to decrease, and
electrical activity and blood flow to return back to a level
that can support vigilance [86].
Cortex
The increase of adenosine concentration during sleep
deprivation in cortical areas suggests that it may have a
role in the decentral regulation of sleep homeostasis [146]
and delivers a link between decentral and central sleep

Pflugers Arch - Eur J Physiol (2012) 463:5371

regulation. As such, adenosine can be regarded a sleepinducing factor [7, 145]. For instance, A1 adenosine
receptors decreased excitatory postsynaptic currents elicited
by stimulation of the ventrobasal thalamus in both
inhibitory and excitatory neurons in the barrel cortex [51].
Agonists for adenosine receptors A1 and A2A applied to the
prefrontal cortex modulate cortical ACh release, behavioural arousal, EEG delta power, and sleep. Additionally,
cortical adenosine agonist or antagonist application significantly altered ACh release in the pontine reticular
formation [185]. Obviously, the prefrontal cortex is able to
modulate behavioural arousal via descending inputs to the
pontine brainstem. The A1 adenosine receptors within the
prefrontal cortex may comprise part of a descending system
that can inhibit wakefulness.
Brainstem and hypothalamus
Adenosine can regulate sleep-related activity in the pons and
the hypothalamus (see [146]). In the hypothalamus, adenosine promotes NREM sleep by inhibiting the histaminergic
neurons in the tuberomammillary nucleus [130]. In the pons,
an adenosine A1 receptor agonist or an inhibitor of the
adenylyl cyclase increase REM sleep in rats, while an A2A
agonist increased ACh release in the pontine reticular
formation and both REM and NREM sleep [23, 101], the
latter probably via GABA-mediated inhibition of arousal
promoting neurons in the pontine reticular formation.
In a simplified view, the duration spent in the state of
wakefulness is inversely proportional to arousal activity.
This is reflected in diminishing discharge activity of
mesopontine cholinergic neurons [148] that play a major
role during arousal [177]. During wakefulness, when the
activity of these neurons is high, increased metabolic
activity may cause an increase in both intracellular and
extracellular adenosine. Whole cell patch clamp and
extracellular recordings in acute brainstem slice preparations
have shown that mesopontine cholinergic neurons are under
tonic inhibitory control of such endogenous adenosine. This is
mediated by increasing an inwardly rectifying K+ current and
a reduction of Ih [148]. Both alterations may act in concert in
mesopontine cholinergic neurons to reduce excitability and
increase the tendency of thalamic neurons to burst.
Thalamus
In the thalamus, adenosine hyperpolarises TC neurons and
reduces the apparent input resistance due to an A1-receptormediated increase in membrane K+ conductance [132].
Additionally, Ih is reduced in TC neurons in the presence of
an A1 receptor agonist, possibly due to an inhibition of
adenylyl cyclase activity. As described in detail above, Ih
has been proposed to be particularly important during

Pflugers Arch - Eur J Physiol (2012) 463:5371

rhythmic burst activity of TC neurons [110], which occurs


in vivo predominantly during certain periods of EEG
synchronization, such as slow-wave sleep or drowsiness
[36, 178]. The increase in Ih that occurs after brainstem
activation of -adrenergic or serotonergic receptors in TC
neurons substantially reduces amplitude and duration of
strong membrane hyperpolarisations and decreases the
input resistance of the membrane, thereby probably
providing an effective mechanism to dampen spindle
oscillations [111, 132, 136]. Vice versa, the decrease in Ih
by adenosine may be capable of inhibiting the effects of
activation of -adrenergic or serotonergic receptors, thus
facilitating rebound burst firing in TC neurons. This
provides a link between central and decentral regulators of
sleep homeostasis: adenosine affects both the oscillator and
resonator of sleep-related electrical activity.

65

intracellular cGMP concentration in TC neurons [160],


interrupting oscillatory burst activity and enhancing the relay
mode of activity in TC neurons, as described in detail above.
Additionally, NO causes a robust, long-lasting depolarisation
in NRT neurons via an ACh and cGMP-dependent pathway,
this time inhibiting burst activity [192, 193]. NO is
considered to have direct sleep-promoting actions in the
cortex [86], although the role of NO in cortical NREM sleep
regulation is not yet fully understood and has led to some
controversy: Both decreases and increases in NREMS were
reported after systemic administration of NOS inhibitors (for
review, see [127]). NO effects on sleep may vary with brain
structures since many sources for NO exist. But because of
the volatile nature of the substance, its actions are local.
Studies of such local NO actions may help to unravel the
complex role of this molecule in both decentral and central
sleep regulation.

Mechanisms of nitric oxide


Glial mechanisms
Another crucial link between central and decentral sleep
regulation is given by the gaseous NO, not only by the finding
that NO possibly stimulates the accumulation of adenosine
[155]. Sensory responses and responses to excitatory amino
acids are facilitated after NO-induced increases of the

Fig. 6 From molecules to networks: thalamic mediators of sleeprelated activity. The hypothetical homeostat of NREM sleep is made
up by molecular networks of several sleep regulatory substances,
including tumour necrosis factor- (TNF-), interleukin-1 (IL-1),
growth hormone-releasing hormone (GHRH), prostaglandin D2 (PG),
brain-derived neurotrophic factor (BDNF), and adenosine. On a
cellular level, many studies show that differential regulation of
intrinsic cellular properties support or suspend thalamic sleep-related
activity. Finally, neuronal networks, e.g. the functional syncytium of
electrically coupled NRT neurons or the intra- and intercellular
networks between NRT neurons, TC neurons, and thalamic local
interneurons impose, support, or maintain sleep-related activity during
whole organism sleep. It is unknown whether individual neuronal
sleep-like states as described for cortical neurons also exist in thalamic

Thalamocortical sleep dynamics and their modulation by


the ascending arousal system and locally released neurochemicals may depend on glia cells, in particular on
purinergic signalling from astrocytes [65], providing yet

neurons. Locally released neurochemicals may depend on purinergic


signalling from astrocytes which can modulate cortical slow oscillations, sleep behaviour, and sleep-dependent cognitive functions. In
particular, astrocytes were found to drive TC neuron excitation via
spontaneous intracellular Ca2+ waves that propagate along thalamic
astrocytes and induce N-methyl-D-aspartate (NMDA)-receptor-mediated currents in TC neurons. These spontaneous transients were
probably produced by Ca2+ release from internal stores mediated by
IP3 receptors. Astrocytes process and transmit information on a longer
timescale, while their interaction with neurons triggers faster processes. Network, neuronal, and molecular pacemakers provide the basis
for phase relations between excitatory and inhibitory signals that can
render different functional states

66

another link between the thalamus and cortex. Astrocytes


can modulate cortical slow oscillations, sleep behaviour,
and sleep-dependent cognitive function. In particular, they
were found to drive TC neuron excitation via spontaneous
intracellular Ca2+ waves that propagate along thalamic
astrocytes and induce N-methyl-D-aspartate-receptor-mediated currents in TC neurons (Fig. 6) [30, 139]. These
spontaneous transients were probably produced by Ca2+
release from internal stores mediated by IP3 receptors [140]
and depended largely on L-type Ca2+ channels probably
needed for store replenishment. This has been a first
demonstration of a complex astrocyteneuron network in
the thalamus. While astrocytes process and transmit
information on a longer timescale, their interaction with
neurons triggers faster processes (for review see [30]). Such
inter-network connectivity between network, neuronal, and
molecular pacemakers is ideally suited as a basis for phase
relations between excitatory and inhibitory signals that can
render different functional states. Future research must take
the role of glia in the regulation of sleep-related activity and
the sleepwake cycle into account.
Integration of central and decentral mechanisms in future
studies
Although the properties of individual neurons may define
the mode of operation of their neuronal networks, a full
understanding of the characteristics of neuronal networks
requires much more: The intra-network connectivity, driving forces, cellular pacemakers, and internal as well as
external phase relations [171] between excitatory and
inhibitory signals can render different functional states,
despite the fact that their individual components have
similar intrinsic properties (see [176]). Behavioural state
control may result from a modulation of excitability in
neuronal networks. Thus, increased knowledge of changing
excitatory and inhibitory modes of operation of these
networks can help to elucidate the mechanisms that cause
the resulting behavioural state (i.e. sleep or wakefulness).
Expanding experimental evidence and network modelling is
required in order to achieve this, and many studies have
extended our understanding in this regard.
In this review, we have summarized mechanisms of sleep
induction with the aim of merging the classical concepts of
central sleep regulation with the recent findings on local
neuronal assemblies and sleep regulatory substances.
Although cellular and network mechanisms of the thalamus
that are thought to elicit non-REM sleep have been studied
extensively, the role of the thalamus in sleep research seems
to be consistently underestimated. Recent findings on
decentral sleep regulation create a scenario in which sleep
is both local and use dependent. The interface between this
newer concept and the classical views of central sleep

Pflugers Arch - Eur J Physiol (2012) 463:5371

regulation is provided by the thalamus and there is no doubt


that thalamic neurons play an important role in sleep
induction, given their ability to produce different patterns of
electrogenic activity during wakefulness and sleep. Many
mechanisms of sleep homeostasis or sleep-like states of
neuronal assemblies, e.g. by the action of adenosine, can
also be found in thalamic neurons and both central and
decentral regulators must ultimately target the thalamus to
induce whole organism sleep. While many mechanisms of
sleep homeostasis or sleep-like states of neuronal assemblies can also be found in thalamic neurons, several
mediators found to be active in cortical neuronal assemblies, such as tumour necrosis factor-, interleukin-1,
growth hormone-releasing hormone, or prostaglandin D2,
have simply not yet been tested on thalamic networks.
Much is known about circadian and homeostatic factors
that prime the brain for sleep, and these mechanisms are
reviewed in other articles of this special issue. The
disruption or disorganization of mechanisms involved in
sleep generation may lead to insomnia, headaches, or
epilepsy. But although sleep is vital, death by sleep
deprivation is very rare. It is obvious that, as important as
sleep is, it can be compensated to a large part and for
extended periods. Our everyday experience supports this
notion. Possibly, if global sleep is absent, this causes
individual neurons to enter a local, sleep-like state for
recovery. Cognitive deficits observed during prolonged
sleep deprivation on the one side, and intermediary states,
such as parasomnias or sleepwalking in humans or
unihemispheric sleep in marine mammals and migratory
birds on the other, support the idea of local sleep. Still,
natural sleep is a global phenomenon, and thalamocortical
synchrony is a prerequisite. Further studies should, therefore, integrate ideas of local and global sleep generation.
Acknowledgements The authors thank Ingrid Winkelhues for the
assistance in the preparation of Fig. 1 and the reviewers for their
helpful comments. The work reported of herein was enabled by grants
to HCP (Max-Planck-Research Award 2007), TB (Interdisciplinary
Centre for Clinical Research Mnster, IZKF, Bud3/010/10; German
Research Foundation, DFG, BU1019/8-1/9-2), and PC (Innovative
Medical Research of the Medical Faculty Mnster, CO 2 1 08 03 and
CO 1 2 10 08).

References
1. Acuna-Goycolea C, Brenowitz SD, Regehr WG (2008) Active
dendritic conductances dynamically regulate GABA release from
thalamic interneurons. Neuron 57:420431
2. Albrecht U (2011) Circadian rhythms and sleep - the metabolic
connection. Pflugers Arch. doi: 10.1007/s00424-011-0986-6
3. Albrecht D, Quschling U, Zippel U, Davidowa H (1996) Effects
of dopamine on neurons of the lateral geniculate nucleus: an
iontophoretic study. Synapse 23:7078

Pflugers Arch - Eur J Physiol (2012) 463:5371


4. Antal M, Acuna-Goycolea C, Pressler RT, Blitz DM, Regehr WG
(2010) Cholinergic activation of M2 receptors leads to contextdependent modulation of feedforward inhibition in the visual
thalamus. PLoS Biol 8:e1000348
5. Bal T, McCormick DA (1993) Mechanisms of oscillatory activity
in guinea-pig nucleus reticularis thalami in vitro: a mammalian
pacemaker. J Physiol (Lond) 468:669691
6. Bal T, McCormick DA (1996) What stops synchronized
thalamocortical oscillations? Neuron 17:297308
7. Basheer R, Strecker RE, Thakkar MM, McCarley RW (2004)
Adenosine and sleepwake regulation. Progr Neurobiol 73:379396
8. Batini C, Moruzzi G, Palestini M, Rossi G, Zanchetti A (1958)
Presistent patterns of wakefulness in the pretrigeminal midpontine
preparation. Science 128:3032
9. Berger H (1929) ber das Elektrenkephalogramm des Menschen.
Arch Psychiat Nervenkr 87:527570
10. Berridge MJ (1998) Neuronal calcium signaling. Neuron 21:1326
11. Blitz DM, Regehr WG (2005) Timing and specificity of feedforward inhibition within the LGN. Neuron 45:917928
12. Blumberg M, Karlsson K, Seelke A, Mohns E (2005) The
ontogeny of mammalian sleep: a response to Frank and Heller
(2003). J Sleep Res 14:9198
13. Blumberg M, Seelke A, Lowen S, Karlsson K (2005) Dynamics
of sleepwake cyclicity in developing rats. PNAS 102:14860
14864
14. Broicher T, Kanyshkova T, Landgraf P, Rankovic V, Meuth P,
Meuth SG, Pape HC, Budde T (2007) Specific expression of
low-voltage-activated calcium channel isoforms and splice
variants in thalamic local circuit interneurons. Mol Cell Neurosci
36:132145
15. Broicher T, Kanyshkova T, Meuth P, Pape HC, Budde T (2008)
Correlation of T-channel coding gene expression, IT, and the low
threshold Ca2+ spike in the thalamus of a rat model of absence
epilepsy. Mol Cell Neurosci 39:384399
16. Broicher T, Wettschureck N, Munsch T, Coulon P, Meuth SG,
Kanyshkova T, Seidenbecher T, Offermanns S, Pape HC, Budde
T (2008) Muscarinic ACh receptor-mediated control of thalamic
activity via G(q)/G(11)-family G-proteins. Pflugers Arch
456:10491060
17. Budde T, Biella G, Munsch T, Pape H-C (1997) Lack of
regulation by intracellular Ca2+ of the hyperpolarizationactivated cation current in rat thalamic neurons. J Physiol (Lond)
503(1):7985
18. Budde T, Coulon P, Pawlowski M, Japes A, Meuth P, Meuth SG,
Pape HC (2008) Reciprocal modulation of Ih and ITASK in
thalamocortical relay neurons by halothane. Pflugers Arch
456:10611073
19. Budde T, Mager R, Pape H-C (1992) Different types of
potassium outward current in relay neurons acutely isolated
from the rat lateral geniculate nucleus. Eur J Neurosci 4:708722
20. Budde T, Sieg F, Braunewell KH, Gundelfinger ED, Pape H-C
(2000) Ca2+-induced Ca2+ release supports the relay mode of
activity in thalamocortical cells. Neuron 26:483492
21. Cain SM, Snutch TP (2010) Contributions of T-type calcium
channel isoforms to neuronal firing. Channels 4:475482
22. Campagna JA, Miller KW, Forman SA (2003) Mechanisms of
actions of inhaled anesthetics. N Engl J Med 348:21102124
23. Coleman C, Baghdoyan H, Lydic R (2006) Dialysis delivery of
an adenosine A2A agonist into the pontine reticular formation of
C57BL/6J mouse increases pontine acetylcholine release and
sleep. J Neurochem 96:17501759
24. Connors BW, Long MA (2004) Electrical synapses in the
mammalian brain. Annu Rev Neurosci 27:393418
25. Contreras D, Destexhe A, Sejnowski TJ, Steriade M (1996)
Control of spatiotemporal coherence of a thalamic oscillation by
corticothalamic feedback. Science 274:771774

67
26. Contreras D, Steriade M (1996) Spindle oscillation in cats: the
role of corticothalamic feedback in a thalamically generated
rhythm. J Physiol (Lond) 490(1):159179
27. Coulon P, Herr D, Kanyshkova T, Meuth P, Budde T, Pape H-C
(2009) Burst discharges in neurons of the thalamic reticular
nucleus are shaped by calcium-induced calcium release. Cell
Calcium 46:333346
28. Coulon P, Kanyshkova T, Broicher T, Munsch T, Wettschureck N,
Seidenbecher T, Meuth SG, Offermanns S, Pape H-C, Budde T
(2010) Activity modes in thalamocortical relay neurons are
modulated by Gq/G11 family G-proteins-serotonergic and
glutamatergic signalling. Front Cell Neurosci 4:110
29. Crandall SR, Govindaiah G, Cox CL (2010) Low-threshold Ca2+
current amplifies distal dendritic signaling in thalamic reticular
neurons. J Neurosci 30:1541915429
30. Crunelli V, Blethyn KL, Cope DW, Hughes SW, Parri HR, Turner
JP, Tth TI, Williams SR (2002) Novel neuronal and astrocytic
mechanisms in thalamocortical loop dynamics. Philos Trans R
Soc Lond B Biol Sci 357:16751693
31. Crunelli V, Cope DW, Hughes SW (2006) Thalamic T-type Ca2+
channels and NREM sleep. Cell Calcium 40:175190
32. Crunelli V, Leresche N (1991) A role for GABAB receptors in
excitation and inhibition of thalamocortical cells. TINS 14:1621
33. Crunelli V, Lrincz ML, Errington AC, Hughes SW, (2011)
Activity of cortical and thalamic neurons during the slow (<1
Hz) rhythm in the mouse in vivo. Pflugers Arch. doi: 10.1007/
s00424-011-1011-9
34. Crunelli V, Tth TI, Cope DW, Blethyn K, Hughes SW (2005)
The window T-type calcium current in brain dynamics of
different behavioural states. J Physiol (Lond) 562:121129
35. Cueni L, Canepari M, Lujan R, Emmenegger Y, Watanabe
M, Bond CT, Franken P, Adelman JP, Luthi A (2008) T-type
Ca2+ channels, SK2 channels and SERCAs gate sleeprelated oscillations in thalamic dendrites. Nat Neurosci
11:683692
36. Curr Dossi RC, Nunez A, Steriade M (1992) Electrophysiology
of a slow (0.54Hz) intrinsic oscillation of cat thalamocortical
neurones in vivo. J Physiol (Lond) 447:215234
37. Curr Dossi R, Par D, Steriade M (1992) Various types of
inhibitory postsynaptic potentials in anterior thalamic cells are
differentially altered by stimulation of laterodorsal tegmental
cholinergic nucleus. Neuroscience 47:279289
38. Daum I, Leonard J, Hehl F (1988) Development of sleep during
monotonous stimulation as related to individual differences.
Integr Psychol Behav Sci 23:118124
39. De Gennaro L, Ferrara M (2003) Sleep spindles: an overview.
Sleep Med Rev 7:423440
40. De Sarro G, Gareri P, Sinopoli VA, David E, Rotiroti D (1997)
Comparative, behavioural and electrocortical effects of tumor
necrosis factor-[alpha] and interleukin-1 microinjected into the
locus coeruleus of rat. Life Sci 60:555564
41. De A, Churchill L, Obal F, Simasko SM, Krueger JM (2002)
GHRH and IL1[beta] increase cytoplasmic Ca2+ levels in
cultured hypothalamic GABAergic neurons. Brain Res
949:209212
42. Decrock E, Vinken M, Bol M, DHerde K, Rogiers V,
Vandenabeele P, Krysko DV, Bultynck G, Leybaert L (2011)
Calcium and connexin-based intercellular communication, a
deadly catch? Cell Calcium. doi:10.1016/j.ceca.2011.05.007
43. Destexhe A, Babloyantz A, Sejnowski TJ (1993) Ionic mechanisms for intrinsic slow oscillations in thalamic relay neurons.
Biophys J 65:15381552
44. Destexhe A, Bal T, Mccormick DA, Sejnowski TJ (1996) Ionic
mechanisms underlying synchronized oscillations and propagating
waves in a model of ferret thalamic slices. J Neurophysiol
76:20492070

68
45. Destexhe A, Contreras D, Sejnowski TJ, Steriade M (1994) A
model of spindle rhythmicity in the isolated thalamic reticular
nucleus. J Neurophysiol 72:803818
46. Diekelmann S, Born J (2010) The memory function of sleep. Nat
Rev Neurosci 11:114126
47. Ebert B, Wafford KA, Deacon S (2006) Treating insomnia:
current and investigational pharmacological approaches. Pharmacol
Ther 112:612629
48. Errington AC, Connelly WM (2011) Dendritic T-type Ca2+
channels: giving a boost to thalamic reticular neurons. J Neurosci
31:55515553
49. Feldberg W, Sherwood SL (1954) Injections of drugs into the
lateral ventricle of the cat. J Physiol (Lond) 123:148167
50. Fitzpatrick D, Penny GR, Schmechel DE (1984) Glutamic acid
decarboxylase-immunoreactive neurons and terminals in the
lateral geniculate nucleus of the cat. J Neurosci 4:18091829
51. Fontanez D, Porter J (2006) Adenosine A1 receptors decrease
thalamic excitation of inhibitory and excitatory neurons in the
barrel cortex. Neuroscience 137:11771184
52. Francesconi W, Muller CM, Singer W (1988) Cholinergic
mechanisms in the reticular control of transmission in the cat
lateral geniculate nucleus. J Neurophysiol 59:16901718
53. Frank M, Heller H (2003) The ontogeny of mammalian sleep: a
reappraisal of alternative hypotheses. J Sleep Res 12:2534
54. Franks NP, Lieb WR (1994) Molecular and cellular mechanisms
of general anaethesia. Nature 367:607614
55. Franks NP, Lieb WR (1998) Which molecular targets are most
relevant to general anaesthesia? Toxicol Lett 100101:18
56. Fuentealba P, Steriade M (2005) The reticular nucleus revisited:
intrinsic and network properties of a thalamic pacemaker. Progr
Neurobiol 75:125141
57. Gabbott PL, Somogyi J, Stewart MG, Hamori J (1986) A quantitative
investigation of the neuronal composition of the rat dorsal lateral
geniculate nucleus using GABA-immunocytochemistry. Neuroscience 19:101111
58. Gastaut H, Bert J (1961) Electroencephalographic detection of sleep
induced by repetitive sensory stimuli. In: Wolstenholme G, OConnor
C (eds) The Nature of Sleep. Churchill, London, pp 260283
59. Gerashchenko D, Wisor JP, Burns D, Reh RK, Shiromani PJ,
Sakurai T, de la Iglesia HO, Kilduff TS (2008) Identification of a
population of sleep-active cerebral cortex neurons. PNAS
105:1022710232
60. Graeff RM, Franco L, De Flora A, Lee HC (1998) Cyclic GMPdependent and -independent effects on the synthesis of the
calcium messengers cyclic ADP-ribose and nicotinic acid
adenine dinucleotide phosphate. J Biol Chem 273:118125
61. Guillery R (1966) A study of Golgi preparations from the dorsal
lateral geniculate nucleus of the adult cat. J Comp Neurol
128:2150
62. Guillery RW, Feig SL, Lozsadi DA (1998) Paying attention to the
thalamic reticular nucleus. Trends Neurosci 21:2832
63. Guillery RW, Sherman SM (2002) Thalamic relay functions and
their role in corticocortical communication: generalizations from
the visual system. Neuron 33:163175
64. Haas HL, Lin, JS (2011) Waking with the hypothalamus.
Pflugers Arch. doi: 10.1007/s00424-011-0996-4
65. Halassa MM (2011) Thalamocortical dynamics of sleep: roles of
purinergic neuromodulation. Semin Cell Dev Biol 22(2):245251
66. Huang Z-L, Urade Y, Hayaishi O (2007) Prostaglandins and
adenosine in the regulation of sleep and wakefulness. Curr Opin
Pharmacol 7:3338
67. Huber R, Tononi G, Cirelli C (2007) Exploratory behavior, cortical
BDNF expression, and sleep homeostasis. Sleep 30:129139
68. Hughes SW, Cope DW, Crunelli V (1998) Dynamic clamp study
of Ih modulation of burst firing and [delta] oscillations in
thalamocortical neurons in vitro. Neuroscience 87:541550

Pflugers Arch - Eur J Physiol (2012) 463:5371


69. Huguenard JR, McCormick DA (2007) Thalamic synchrony and
dynamic regulation of global forebrain oscillations. Trends
Neurosci 30:350356
70. Huguenard JR, Prince DA (1991) Slow inactivation of a TEAsensitive K current in acutely isolated rat thalamic relay neurons.
J Neurophysiol 66:13161328
71. Jahnsen H, Llinas R (1984) Electrophysiological properties of
guinea-pig thalamic neurones: an in vitro study. J Physiol
349:205226
72. Joksovic PM, Bayliss DA, Todorovic SM (2005) Different
kinetic properties of two T-type Ca2+ currents of rat reticular
thalamic neurones and their modulation by enflurane. J Physiol
(Lond) 566:125142
73. Joksovic PM, Brimelow BC, Murbartin J, Perez-Reyes E,
Todorovic SM (2005) Contrasting anesthetic sensitivities of Ttype Ca2+ channels of reticular thalamic neurons and recombinant
Cav3.3 channels. Br J Pharmacol 144:5970
74. Joksovic PM, Choe WJ, Nelson MT, Orestes P, Brimelow BC,
Todorovic SM (2010) Mechanisms of inhibition of T-type calcium
current in the reticular thalamic neurons by 1-octanol: implication of
the protein kinase C pathway. Mol Pharmacol 77:8794
75. Joksovic PM, Todorovic SM (2010) Isoflurane modulates
neuronal excitability of the nucleus reticularis thalami in vitro.
Ann NY Acad Sci 1199:3642
76. Jones EG (1985) The thalamus. Plenum, New York
77. Jones EG (2002) Thalamic circuitry and thalamocortical synchrony. Philos Trans R Soc Lond B Biol Sci 357:16591673
78. Jouvet-Mounier D, Astic L, Lacote D (1970) Ontogenesis of the
states of sleep in rat, cat, and guinea pig during the first postnatal
month. Dev Psychobiol 2:216239
79. Kanyshkova T, Broicher T, Meuth S, Pape H-C, Budde T (2011)
A-type K+ currents in intralaminar thalamocortical relay neurons. Pflugers Arch 461:545556
80. Kanyshkova T, Pawlowski M, Meuth P, Dub C, Bender RA,
Brewster AL, Baumann A, Baram TZ, Pape H-C, Budde T
(2009) Postnatal expression pattern of HCN channel isoforms in
thalamic neurons: relationship to maturation of thalamocortical
oscillations. J Neurosci 29:88478857
81. Kasten MR, Rudy B, Anderson MP (2007) Differential regulation of action potential firing in adult murine thalamocortical
neurons by Kv3.2, Kv1, and SK potassium and N-type calcium
channels. J Physiol (Lond) 584:565582
82. Kilduff TS, Cauli B, Gerashchenko D (2011) Activation of
cortical interneurons during sleep: an anatomical link to
homeostatic sleep regulation? Trends Neurosci 34:1019
83. Kim U, Bal T, McCormick DA (1995) Spindle waves are
propagating sychronized oscillations in the ferret LGNd in vitro.
J Nuerophysiol 74:13011323
84. Koesling D, Humbert P, Schultz G (1995) The NO receptor:
characterization and regulation of soluble guanylyl cyclase. In:
Vincent SR (ed) Nitric oxide in the nervous system. Academic,
London, pp 4350
85. Krueger J, Churchill L, Rector D (2009) Cytokines and other
neuromodulators. In: Stickgold R, Walker M (eds) The neuroscience of sleep. Elsevier, Oxford
86. Krueger JM, Rector DM, Roy S, Van Dongen HPA, Belenky G,
Panksepp J (2008) Sleep as a fundamental property of neuronal
assemblies. Nat Rev Neurosci 9:910919
87. Landgraf D, Shostak A, Oster H (2011) Clock genes and sleep.
Pflugers Arch. doi: 10.1007/s00424-011-1003-9
88. Landisman CE, Connors BW (2005) Long-term modulation of
electrical synapses in the mammalian thalamus. Science
310:18091813
89. Landisman CE, Long MA, Beierlein M, Deans MR, Paul DL,
Connors BW (2002) Electrical synapses in the thalamic reticular
nucleus. J Neurosci 22:10021009

Pflugers Arch - Eur J Physiol (2012) 463:5371


90. Lee SH, Govindaiah G, Cox CL (2007) Heterogeneity of firing
properties among rat thalamic reticular nucleus neurons. J
Physiol 582:195208
91. Llinas R, Ribary U (1993) Coherent 40-Hz oscillation characterizes
dream state in humans. PNAS 90:20782081
92. Llinas RR, Steriade M (2006) Bursting of thalamic neurons and
states of vigilance. J Neurophysiol 95:32973308
93. Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff
K, Langebartels A, Wotjak C, Munsch T, Zong X, Feil S, Feil R,
Lancel M, Chien KR, Konnerth A, Pape HC, Biel M, Hofmann F
(2003) Absence epilepsy and sinus dysrhythmia in mice lacking
the pacemaker channel HCN2. EMBO J 22:216224
94. Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M (1998) A
family of hyperpolarization-activated mammalian cation channels.
Nature 393:587591
95. Lthi A, McCormick DA (1999) Modulation of a pacemaker
current through Ca2+-induced stimulation of cAMP production.
Nat Neurosci 2:634641
96. Magni F, Moruzzi G, Rossi G, Zanchetti A (1959) EEG arousal
following inactivation of the lower brain stem by selective
injection of barbiturate into the vertebral circulation. Arch Ital
Biol 97:3346
97. Mancia M, Margnelli M, Mariotti M, Spreafico R, Broggi G
(1974) Brain stem-thalamus reciprocal influences in the cat.
Brain Res 69:297314
98. Mancia M, Meulders M, Santibanez H (1959) Synchronisation
de llectroencphalogramme provoque par la stimulation
visuelle rptitive chez le chat mdiopontin prtrigminal.
Arch Int Physiol Biochem 67:661670
99. Manfridi A, Brambilla D, Bianchi S, Mariotti M, Opp MR, Imeri
L (2003) Interleukin-1 enhances non-rapid eye movement sleep
when microinjected into the dorsal raphe nucleus and inhibits
serotonergic neurons in vitro. Eur J Neurosci 18:10411049
100. Mares P, Maresov D, Trojan S, Fischer J (1982) Ontogenetic
development of rhythmic thalamo-cortical phenomena in the rat.
Brain Res Bull 8:765769
101. Marks GA, Birabil CG (1998) Enhancement of rapid eye
movement sleep in the rat by cholinergic and adenosinergic
agonists infused into the pontine reticular formation. Neuroscience
86:2937
102. Marshall L, Born J (2007) The contribution of sleep to
hippocampus-dependent memory consolidation. Trends Cogn
Sci 11:442450
103. Martinowich K, Schloesser R, Jimenez D, Weinberger D, Lu B
(2011) Activity-dependent brain-derived neurotrophic factor
expression regulates cortistatin-interneurons and sleep behavior.
Mol Brain 4:11
104. Matos G, Andersen ML, do Valle AC, Tufik S (2010) The
relationship between sleep and epilepsy: evidence from clinical
trials and animal models. J Neurol Sci 295:17
105. McCormick DA (1992) Neurotransmitter actions in the thalamus
and cerebral cortex and their role in neuromodulation of
thalamocortical activity. Prog Neurobiol 39:337388
106. McCormick DA (1992) Cellular mechanisms underlying cholinergic and noradrenergic modulation of neuronal firing mode in
the cat and guinea pig dorsal lateral geniculate nucleus. J
Neurosci 12:278289
107. McCormick DA, Bal T (1994) Sensory gating mechanisms of the
thalamus. Curr Opin Neurobiol 4:550556
108. McCormick DA, Bal T (1997) Sleep and arousal: thalamocortical
mechanisms. Annu Rev Neurosci 20:185215
109. McCormick DA, Pape H-C (1988) Acetylcholine inhibits
identified interneurons in the cat lateral geniculate nucleus.
Nature 334:246248
110. McCormick DA, Pape HC (1990) Properties of a
hyperpolarization-activated cation current and its role in

69

111.

112.
113.

114.
115.

116.

117.
118.
119.

120.

121.

122.

123.

124.
125.

126.
127.
128.

129.

130.

131.

132.

rhythmic oscillation in thalamic relay neurones. J Physiol


(Lond) 431:291318
McCormick DA, Pape HC (1990) Noradrenergic and serotonergic modulation of a hyperpolarization-activated cation current in
thalamic relay neurones. J Physiol (Lond) 431:319342
McGinty D, Sterman M (1968) Sleep suppression after basal
forebrain lesions in the cat. Science 160:12531255
Meuth SG, Kanyshkova T, Meuth P, Landgraf P, Munsch T,
Ludwig A, Hofmann F, Pape HC, Budde T (2006) Membrane
resting potential of thalamocortical relay neurons is shaped by
the interaction among TASK3 and HCN2 channels. J Neurophysiol 96:15171529
Mignot E, Lin L (2009) Narcolepsy. In: Stickgold R, Walker M
(eds) The Neuroscience of Sleep. Elsevier, pp 270277
Mihic SJ, Ye Q, Wick MJ, Koltchine VV, Krasowski MD, Finn SE,
Mascia MP, Valenzuela CF, Hanson KK, Greenblatt EP, Harris RA,
Harrison NL (1997) Sites of alcohol and volatile anaesthetic action
on GABAA and glycine receptors. Nature 389:385389
Montero V, Singer W (1985) Ultrastructural identification of
somata and neural processes immunoreactive to antibodies
against glutamic acid decarboxylase (GAD) in the dorsal lateral
geniculate nucleus of the cat. Exp Brain Res 59:151165
Monti JM, Monti D (2007) The involvement of dopamine in the
modulation of sleep and waking. Sleep Med Rev 11:113133
Moruzzi G (1972) The sleepwaking cycle. Ergeb Physiol 64:1165
Moruzzi G, Magoun H (1949) Brain stem reticular formation and
activation of the EEG. Electroencephalogr Clin Neurophysiol
1:455473
Mullington J (2009) Endocrine function during sleep and sleep
deprivation. In: Stickgold R, Walker M (eds) The Neuroscience
of Sleep. Elsevier, pp 209212
Mullington J (2009) Immune function during sleep and sleep
deprivation. In: Stickgold R, Walker M (eds) The Neuroscience
of Sleep. Elsevier, pp 213217
Munsch T, Budde T, Pape HC (1997) Voltage-activated intracellular calcium transients in thalamic relay cells and interneurons.
Neuroreport 8:24112418
Munsch T, Yanagawa Y, Obata K, Pape HC (2005) Dopaminergic
control of local interneuron activity in the thalamus. Eur J
Neurosci 21:290294
Nauta W (1946) Hypothalamic regulation of sleep in rats; an
experimental study. J Neurophysiol 9:285316
Nelson LE, Guo TZ, Lu J, Saper CB, Franks NP, Maze M (2002)
The sedative component of anesthesia is mediated by GABAA
receptors in an endogenous sleep pathway. Nat Neurosci 5:979984
Nishino S (2009) Cataplexy. In: Stickgold R, Walker M (eds) The
Neuroscience of Sleep. Elsevier, pp 278284
Obal FJ, Krueger J (2003) Biochemical regulation of non-rapideye-movement sleep. Front Biosci 8:d520d550
Ohara PT, Lieberman AR (1985) The thalamic reticular nucleus
of the adult rat: experimental anatomical studies. J Neurocytol
14:365411
Ohara PT, Lieberman AR, Hunt SP, Wu JY (1983) Neural elements
containing glutamic acid decarboxylase (GAD) in the dorsal lateral
geniculate nucleus of the rat; Immunohistochemical studies by light
and electron microscopy. Neuroscience 8:189211
Oishi Y, Huang Z-L, Fredholm BB, Urade Y, Hayaishi O (2008)
Adenosine in the tuberomammillary nucleus inhibits the histaminergic system via A1 receptors and promotes non-rapid eye
movement sleep. PNAS 105:1999219997
Pace-Schott EF, Hobson JA (2002) The neurobiology of sleep:
genetics, cellular physiology and subcortical networks. Nat Rev
Neurosci 3:591605
Pape H-C (1992) Adenosine promotes burst activity in guineapig geniculocortical neurones through two different ionic
mechanisms. J Physiol (Lond) 447:729753

70
133. Pape HC (1996) Queer current and pacemaker: the
hyperpolarization-activated cation current in neurons. Annu
Rev Physiol 58:299327
134. Pape H-C, Budde T, Mager R, Kisvarday Z (1994) Prevention of
Ca2+-mediated action potentials in GABAergic local circuit
neurons of the thalamus by a transient K+ current. J Physiol
(Lond) 478(3):403422
135. Pape H-C, Mager R (1992) Nitric oxide controls oscillatory
activity in thalamocortical neurons. Neuron 9:441448
136. Pape HC, McCormick DA (1989) Noradrenaline and serotonin
selectively modulate thalamic burst firing by enhancing a
hyperpolarization-activated cation current. Nature 340:715718
137. Pape HC, McCormick DA (1995) Electrophysiological and
pharmacological properties of interneurons in the cat dorsal
lateral geniculate nucleus. Neuroscience 68:11051125
138. Pape HC, Munsch T, Budde T (2004) Novel vistas of calciummediated signalling in the thalamus. Eur J Physiol (Pflgers
Arch) 448:131138
139. Parri HR, Crunelli V (2001) Pacemaker calcium oscillations in
thalamic astrocytes in situ. Neuroreport 12:38973900
140. Parri HR, Gould TM, Crunelli V (2001) Spontaneous astrocytic
Ca2+ oscillations in situ drive NMDAR-mediated neuronal
excitation. Nat Neurosci 4:803812
141. Perez Velazquez JL, Carlen PL (1996) Development of firing
patterns and electrical properties in neurons of the rat ventrobasal
thalamus. Brain Res 91:164170
142. Pinault D (2004) The thalamic reticular nucleus: structure,
function and concept. Brain Res Brain Res Rev 46:131
143. Pinault D, Deschnes M (1998) Projection and innervation
patterns of individual thalamic reticular axons in the thalamus
of the adult rat: a three-dimensional, graphic, and morphometric
analysis. J Comp Neurol 391:180203
144. Porkka-Heiskanen T (2011) Methylxanthines and sleep. In:
Fredholm BB (ed) Methylxanthines, handbook of experimental
pharmacology. Springer, Berlin, Heidelberg, pp 331348
145. Porkka-Heiskanen T, Alanko L, Kalinchuk A, Stenberg D (2002)
Adenosine and sleep. Sleep Med Rev 6:321332
146. Porkka-Heiskanen T, Kalinchuk A (2011) Adenosine, energy
metabolism and sleep homeostasis. Sleep Med Rev 15:123135
147. Puizillout J, Ternaux J, Foutz A, Dell P (1973) Slow wave sleep
with phasic discharges. Triggering by vago-aortic stimulation.
Rev Electroencephalogr Neurophysiol Clin 3(2137):143
148. Rainnie DG, Grunze HCR, McCarley RW, Greene RW (1994)
Adenosine inhibition of mesopontine cholinergic neurons:
implications for EEG arousal. Science 263(689692):144
149. Rechtschaffen A, Bergmann B (2002) Sleep deprivation in the
rat: an update of the 1989 paper. Sleep 25:1824
150. Rechtschaffen A, Kales A (1968) A manual of standardized
terminology, techniques and scoring system for sleep stages of
human subject. U.S. National Institute of Neurological Diseases and
Blindness, Neurological Information Network, Bethesda, MD
151. Reinoso-Surez F, De Andrs I, Garzn M (2011) Functional
anatomy of the sleepwakefulness cycle: wakefulness. Adv Anat
Embryol Cell Biol 208(1128):147
152. Richter TA, Kolaj M, Renaud LP (2005) Low voltage-activated
Ca2+ channels are coupled to Ca2+-induced Ca2+ release in rat
thalamic midline neurons. J Neurosci 25(82678271):148
153. Ries CR, Puil E (1999) Ionic mechanism of isofluranes actions
on thalamocortical neurons. J Neurophysiol 81:18021809
154. Ries CR, Puil E (1999) Mechanism of anesthesia revealed by
shunting actions of isoflurane on thalamocortical neurons. J
Neurophysiol 81:17951801
155. Rosenberg PA, Li Y, Le M, Zhang Y (2000) Nitric oxidestimulated increase in extracellular adenosine accumulation in rat
forebrain neurons in culture is associated with atp hydrolysis and
inhibition of adenosine kinase activity. J Neurosci 20:62946301

Pflugers Arch - Eur J Physiol (2012) 463:5371


156. Sallanon M, Denoyer M, Kitahama K, Aubert C, Gay N, Jouvet
M (1989) Long-lasting insomnia induced by preoptic neuron
lesions and its transient reversal by muscimol injection into the
posterior hypothalamus in the cat. Neuroscience 32:669683
157. Salt TE (2002) Glutamate receptor functions in sensory relay in the
thalamus. Philos Trans R Soc Lond B Biol Sci 357:17591766
158. Seelke A, Blumberg M (2008) The microstructure of active and
quiet sleep as cortical delta activity emerges in infant rats. Sleep
31:691699
159. Selbach O, Haas HL (2006) Hypocretins: the timing of sleep and
waking. Chronobiol Int 23(12):6370
160. Shaw PJ, Salt TE (1997) Modulation of sensory and excitatory
amino acid responses by nitric oxide donors and glutathione in
the ventrobasal thalamus of the rat. Eur J Neurosci 9:15071513
161. Sherman SM (2001) Tonic and burst firing: dual modes of
thalamocortical relay. Trends Neurosci 24:122126
162. Sherman SM (2007) The thalamus is more than just a relay. Curr
Opin Neurobiol 17:417422
163. Sherman SM, Guillery RW (1996) Functional organization of
thalamocortical relays. J Neurophysiol 76(3):13671395
164. Sherman SM, Guillery RW (2006) Exploring the thalamus and
its role in cortical function. MIT Press, Cambridge
165. Sieg F, Obst K, Gorba T, Riederer B, Pape HC, Wahle P (1998)
Postnatal expression pattern of calcium-binding proteins in
organotypic thalamic cultures and in the dorsal thalamus in vivo.
Brain Res 110:8395
166. Singer W (1977) Control of thalamic transmission by corticofugal and ascending reticular pathways in the visual system.
Physiol Rev 57:386420
167. Soltesz I, Lightowler S, Leresche N, Jassik-Gerschenfeld D,
Pollard CE, Crunelli V (1991) Two inward currents and the
transformation of low-frequency oscillations of rat and cat
thalamocortical cells. J Physiol (Lond) 441:175197
168. Spreafico R, Battaglia G, Frassoni C (1991) The reticular
thalamic nucleus (RTN) of the rat: cytoarchitectural, Golgi,
immunocytochemical, and horseradish peroxidase study. J Comp
Neurol 304:478490
169. Spreafico R, de Curtis M, Frassoni C, Avanzini G (1988)
Electrophysiological characteristics of morphologically identified
reticular thalamic neurons from rat slices. Neuroscience 27:629638
170. Stenberg D (2007) Cell Mol Life Sci 64:11871204
171. Steriade M (2006) Grouping of brain rhythms in corticothalamic
systems. Neuroscience 137:10871106
172. Steriade M, Curro Dossi RC, Nunez A (1991) Network modulation
of a slow intrinsic oscillation of cat thalamocortical neurons
implicated in sleep delta waves: cortically induced synchronization
and brainstem cholinergic suppression. J Neurosci 11:32003217
173. Steriade M, Domich L, Oakson G (1986) Reticularis thalami
neurons revisited: activity changes during shifts in states of
vigilance. J Neurosci 6:6881
174. Steriade M, Domich L, Oakson G, Deschenes M (1987) The
deafferented reticular thalamic nucleus generates spindle rhythmicity. J Neurophysiol 57:260273
175. Steriade M, Jones EG, McCormick DA (1997) Thalamus.
Elsevier, Amsterdam
176. Steriade M, McCarley R (1990) Brainstem control of wakefulness and sleep. Plenum, New York
177. Steriade M, McCormick DA, Sejnowski TJ (1993) Thalamocortical oscillations in the sleeping and aroused brain. Science
262:679685
178. Steriade M, Pare D, Hu B, Deschenes M (1990) The visual
thalamocortical system and its modulation by the brain stem
core. Prog Sens Physiol 10:1124
179. Stocker M, Pedarzani P (2000) Differential distribution of three
Ca2+-activated K+ channel subunits, SK1, SK2, and SK3, in the
adult rat central nervous system. Mol Cell Neurosci 15:476493

Pflugers Arch - Eur J Physiol (2012) 463:5371


180. Szymusiak R (2009) Thermoregulation during sleep and sleep
deprivation. In: Stickgold R, Walker M (eds) The Neuroscience
of Sleep. Elsevier, pp 218222
181. Terman D, Bose A, Kopell N (1996) Functional reorganization in
thalamocortical networks: transition between spindling and delta
sleep rhythms. PNAS 93:1541715422
182. Todorovic SM, Lingle CJ (1998) Pharmacological properties of Ttype Ca2+ current in adult rat sensory neurons: effects of
anticonvulsant and anesthetic agents. J Neurophysiol 79:240252
183. Tth TI, Crunelli V (1997) Simulation of intermittent action
potential firing in thalamocortical neurons. Neuroreport 8:2889
2892
184. Tscherter A, David F, Ivanova T, Deleuze C, Renger JJ, Uebele
VN, Shin HS, Bal T, Leresche N, Lambert RC (2011) Minimal
alterations in T-type calcium channel gating markedly modify
physiological firing dynamics. J Physiol (Lond) 589:17071724
185. Van Dort CJ, Baghdoyan HA, Lydic R (2009) Adenosine A1 and
A2A receptors in mouse prefrontal cortex modulate acetylcholine release and behavioral arousal. J Neurosci 29:871881
186. Veasey S (2009) Sleep apnea. In: Stickgold R, Walker M (eds)
The Neuroscience of Sleep. Elsevier, pp 263269
187. Villablanca J (2004) Counterpointing the functional role of the
forebrain and of the brainstem in the control of the sleep-waking
system. J Sleep Res 13:179208
188. Vyazovskiy VV, Olcese U, Hanlon EC, Nir Y, Cirelli C, Tononi G
(2011) Local sleep in awake rats. Nature 472:443447

71
189. Wan X, Mathers DA, Puil E (2003) Pentobarbital modulates
intrinsic and GABA-receptor conductances in thalamocortical
inhibition. Neuroscience 121:947958
190. Ward C (2011) On doing nothing: descriptions of sleep, fatigue,
and motivation in encephalitis lethargica. Mov Disord. doi:10.1002/
mds.23545, Epub ahead of print
191. Wurtz RH, Sommer MA, Cavanaugh J (2005) Drivers from the
deep: the contribution of collicular input to thalamocortical
processing. Prog Brain Res 149:207225
192. Yang S, Cox CL (2007) Modulation of inhibitory activity by
nitric oxide in the thalamus. J Neurophysiol 97:33863395
193. Yang S, Cox CL (2008) Excitatory and anti-oscillatory actions of
nitric oxide in thalamus. J Physiol (Lond) 586:36173628
194. Ying SW, Goldstein PA (2005) Propofol suppresses synaptic
responsiveness of somatosensory relay neurons to excitatory
input by potentiating GABA(A) receptor chloride channels. Mol
Pain 1:2
195. Zeitlhofer J, Gruber G, Anderer P, Asenbaum S, Schimicek P,
Saletu B (1997) Topographic distribution of sleep spindles in
young healthy subjects. J Sleep Res 6:149155
196. Zhao Y, Kerscher N, Eysel U, Funke K (2002) D1 and D2
receptor-mediated dopaminergic modulation of visual responses
in cat dorsal lateral geniculate nucleus. J Physiol (Lond)
539:223238
197. Zhu JJ, Uhlrich DJ, Lytton WW (1999) Burst firing in identified
rat geniculate interneurons. Neuroscience 91:14451460