NMR Practical

Pulse calibration, basic 2D and 3D heteronuclear NMR experiments

EMBO Practical course on structure determination by NMR

EMBL Heidelberg, Sept. 2001

Remco Sprangers, Michael Sattler (EMBL Heidelberg)

Table of contents:
1. Introduction 2
2. The Bruker language and our nomenclature. 3
Powerlevels 3
Pulses 3
Phases 3
Shaped pulses 4
Delays 4
Gradients 4
Example 5
Processing 6
3. Calibration 7
offset 7
proton hardpulse 7
nitrogen hardpulse 7
carbon hardpulse 8
carbon shaped pulses 9
Bloch Siegert phase 12
Waterflipback pulse 13
Gradients 13
WURST pulse 13
4. 1D and 2D experiments 15
1d presat 15
1d WATERGATE 15
15N HSQC 15
15N Trosy 16
Half Trosy 17
13C HMQS 17
13C HSQC 18
Filters 18
5. 3D experiments 19
wrong HNCA 19

1
1. Introduction

The practical will be done during two sessions, 2.5 h each. The first part concerns basic
steps for heteronuclear triple resonance experiments, e.g. determination of pulses
(rectangular & shapes), phases, gradients, etc. During the second part some basic 2D
experiments will be set up and compared (hqsc, hmqc, trosy). Furthermore, two basis
filter experiments will be analyzed. Finally, a wrong HNCA pulse sequence, where some
"mistakes" / programming errors have been included intentionally, has to be debugged.

The different NMR experiments and pulse programs are stored in the data sets indicated
below. For each data set there it is briefly described what the pulse program does and
what the important parameters are. Printouts of some of the pulseprograms are included.
Figures for many of the pulse programs can be found in the lecture notes.

Some name conventions and a short manual for the spectrometer are available at the
spectrometers / on the local web pages. On the Sattler homepage more pulseprograms
can be found. A very brief overview of the Bruker language is included here and should
be read before the practical start.

2
2. The Bruker language and our nomenclature.
There are 4 channels on the spectrometers (f1,f2,f3 and f4). Here, we use f1 for protons,
f2 for carbon and f3 for nitrogen (can be checked with the command edasp in
XWINNMR).

Basically, the pulseprogram is executed line by line. When a line is finished the program
continues on the next line. When two commands are written on one line they start both at
the beginning. The line lasts as long as the longest command on that line. A more visual
representation of the pulseprogram is generated using the commands ppsim, ppg or
pulsdisp in XWINNMR.

Powerlevels

Powerlevels are called pl and are set on a certain channel (in db)
e.g. pl1:f1 sets the powerlevel on the value pl1 on channel 1 (protons)

note: the lower the db, the higher the power

CHECK THE POWER LIMITS AT THE SPECTROMETER,
DON’T BURN THE PROBE!!

Pulses

Rectangular pulses are indicated by p (there is p1 to p31)

We use the following nomenclature:

p1 Proton 90deg
p2 Proton 180deg
p3 Carbon 90deg
p4 Carbon 180deg
p5 Nitrogen 90deg
p6 Nitrogen 180deg
p16 and p19 Gradient

e.g: p1:f1 (means a 90 deg pulse on protons)

Phases

Pulses can be given a certain phase (ph) (if nothing is specified, x is used)
e.g: p2 ph7:f1 (means a 180 deg hardpulse on protons with phase 7)

the phases are defined at the bottom of the pulseprogram.

e.g: ph7 = 0 2 (means ph7 is cycled between 0 and 2)
Where: 0=x, 1=y, 2=-x, 3=-y

Phases can be incremented during the pulseprogram with ip

3
 e.g: 10m ip3 increases phase3 with 90deg at the start of a 10ms delay

Shaped pulses

Shaped pulses are indicated with sp (there is sp1 to sp31) and come with some
variables including:

spnam = the name of the shaped pulse

spoffs = is the offset of the pulse

We use the following nomenclature:

p13:sp1 90deg on resonance

p13:sp2 90deg time reverse on resonance
p14:sp3 180deg on resonance
p14:sp4 180deg off resonance
p13:sp5 90deg off resonance
p13:sp6 90deg time reverse off resonance
(note: normally on resonance is on Cα and off resonance is on C’)

Where the pulselength is defined by p and the powerlevel is defined by sp

Furthermore, a phase needs to be defined
e.g (p14:sp4 ph0):f2 is a shaped pulse with name spnam4, offset spoffs4,
powerlevel sp4 and length p14, having a phase of ph0 (a 180 deg on the C' when
o2p is at 40ppm and spoffs is set accordingly)

Delays

Delays are indicated by d (there is d1 to d31) and can either be given in the
dataset (in XWINNMR) or calculated on top of the pulseprogram. (e.g.
"d6=9.0m" or "d30=d6+6u- d7+p13*2")
a delay can also be written directly in the pulseprogram (e.g 20u or 10m)
and a delay can be incremented or decremented during the pulseprogram with id
and dd e.g. id10 means the d10 is incremented with value in10

Gradients

Gradients are called gp and come with gpnam (the name of the gradient) and a p
with gives the length of the gradient. The strength of the gradient is given by gpx,
gpy and gpz for the three dimensions (gp goes from 0 to 100%)
e.g. p16:gp3 (is a gradient of length p16, with name gpnam3 and strength
gpx3,gpy3 and gpz3)
Gradients can be inverted using the following trick:
p16:gp1*EA

4
 igrad EA
where EA is a list containing a 1 and a –1. The command igrad gives EA the next
value in the list. (Used for echo/antiecho type of experiments)

Example

The simplest case: a 1d on protons with presat.

Pulseprogram Explanation

;mszgpr |
;avance-version | Some commands
;1D sequence with f1 presaturation |

#include <Avance.incl> Import some parameters

1 ze Start
2 20u pl7:f1 20us delay, the powerlevel for protons is set to pl7
(p29 ph0):f1 ; presaturation at pl7 55dB a pulse on protons of length p29 and phase ph1 (see below)
3u 3us delay
20u pl1:f1 20us delay, powerlevel for protons to pl1
p1 ph1 ; 90 degree 1H 0dB pulse p1 (90deg protons) with phase1 (see below)
go=2 ph31 receiver open with phase 31
d11 wr #0 delay of length d11 and write fid to disk
exit end

ph0=0 phase0 is x
ph1=0 2 2 0 1 3 3 1 phase1 goes as x,-x,-x,x,y,-y,-y,y (incremented every scan)
ph31=0 2 2 0 1 3 3 1 receiver phase

;pl1 : f1 channel - power level for pulse (default) |

;pl9 : f1 channel - power level for presaturation | Some remarks
;p1 : f1 channel - 90 degree high power pulse ||

What also has to be set in the dataset (in XWINNMR) in this case:

offset for protons: o1 (in Hz) or o1p (in ppm)

number of scans: ns (multiple of the phase cycle)
number of dummy scans: ds
the receiver gain: rg (the higher the number the more sensitive)

An overview of most parameters used are given in ASED and EDA in XWINNMR.

5
Processing
An overview of the processing parameters can be found under EDP
1d spectra can be processed using the commands:
ft = fourier transform
fp = ft + use predefined phases (see EDP)
efp = fp + exponential window function (line broadening is defined by lb)
gfp = fp + gaussian window function (defined by lb and gb)
2d spectra are processed using:
xfb = process both dimensions
xf2 = xf2, process 1 dimension
xf1 = process the other dimension
abs1 = baseline correction in f1
abs2 = baseline correction in f2
rser # = extract 1d number # from a 2d, can be further processed as a 1d

6
3. Calibration of frequency offset, pulses, gradients etc.

During this session the calibration of proton, nitrogen and carbon pulses will be
explained.
Change to directory "cal 1 1 embo nmr" by typing "emcal" in XWINNMR

cal/1 Emzgpr (pulseprogram above)

calibration of the offset. (make sure water is on resonance)
presat (p29, pl7) , 1H pulse (p1, pl1), detection

cal/2 Emzgproff
calibration of the 1H pulse
off resonance presat (p18*10, sp0, spoffs0), 1H pulse (p1, pl1) , detection
Optimize p1 and pl1 for zero signal (180deg pulse)

;EMzgproff
;avance-version
;1D sequence with f1 presaturation
; off resonance presat

#include <Avance.incl>

"d12=20u" Set some delays…..

"p18=100m"

1 ze
2 d1 Delay of length d1
d12 pl7:f1 Check pl7
3u
18 (p18:sp0 ph0):f1 ; squ.1000, 100m, not < 55 db Check the spman0, spoffs0
2u
lo to 18 times 10 Loop (presat is 10*p18)
3u
10u pl1:f1
p1 ph1
go=2 ph31
50m wr #0
exit

ph0=0
ph1=0 2 2 0 1 3 3 1
ph10=0 2 2 0 1 3 3 1
ph27=0 2 2 0 1 3 3 1
ph31=0 2 2 0 1 3 3 1

cal/3 EMinvp90f3
calibration of 15N hard pulse
off resonance presat, 90° 1H, delay of 1/(2JHN) , 90° 15N , 90° or 180° on 15N, 180° 1H,
delay of 1/(2JHN), detection.

7
 Optimize p4 and pl4 for pulse calibration. (Unclear?: make a little drawing of
the pulseprogram and use the product operators).

; invp90f3
; for calibration of rectangular pulse on F3 (15N)
; p3 pl3: ~ 90deg
; p4 pl4: 1) 90deg: phase spectrum
; 2) 180deg: optimize for zero signal

; F1 1H 4.7ppm (H2O)
; F3 15N ~118ppm (or select amide with unique 1H shift in 1D)

"d2=5.4m" ; 1/2J(N,H)
"d22=d2-p3-p4-16u"
"p18=100m" ; for off resonance presaturation set SPOFFS0

1 ze
2 d1 do:f3
d12 pl7:f1
18 (p18:sp0 ph0):f1 ; sp0=squ.1000, 100ms, sp0=pl7
2u
lo to 18 times 10 ; off resonance presaturation loop
3u
20u pl1:f1
(p1 ph1):f1 90deg proton
d2 pl3:f3 d2
(p2 ph3):f1 180deg proton
(p3 ph2):f3 ; 90 deg 90deg nitrogen
3u
10u pl4:f3 change powerlevel
(p4 ph4):f3 ; 90/180 deg 90 (signal) or 180 (no signal)
3u
d22 pl13:f3 set decoupling powerlevel
go=2 ph31 cpd3:f3 decoupling on
d11 wr #0 do:f3
exit

ph0=0
ph1=0
ph2=0 0 2 2
ph3=0
ph4=0 2
ph31=0 2 2 0

Why use d22 (which is less than d2)

cal/4 EMinvp90f2
calibration of 13C hard pulse
off resonance presat, 90° 1H, delay of 1/(2JHC) , 90° 13C, 90/180 on 13C, 180° 1H, delay
of 1/(2JHC), detection
Optimize p4 and pl4 for pulse calibration, same principle as for the 15N
calibration

8
 ; invp90f2

; for calibration of rectangular pulse on F2 (13C)

; p3 pl3: ~ 90deg
; p4 pl4: 1) 90deg: phase spectrum
; 2) 180deg: optimize for zero signal

; F1 1H 4.7ppm (H2O)
; F2 13C ~20ppm (select methyl group with unique 1H shift in 1D)

"d2=3.8m" ; 1/2J(C,H)
"d22=d2-p3-p4-16u"
"p18=100m"

1 ze
2 d1 do:f2
d12 pl7:f1
18 (p18:sp0 ph0):f1
2u
lo to 18 times
3u
20u pl1:f1
(p1 ph1):f1
d2 pl3:f2
(p2 ph3):f1
(p3 ph2):f2 ; 90 deg
3u
10u pl4:f2
(p4 ph4):f2 ; 90/180 deg
3u
d22 pl12:f2
go=2 ph31 cpd2:f2
d11 wr #0 do:f2
exit

ph0=0
ph1=0
ph2=0 0 2 2
ph3=0
ph4=0 2
ph31=0 2 2 0

cal/5 EMinvg4
Use the shape tool in XWINNMR to check the difference in excitation profile
between the 90 and 180 deg hard pulses and the shaped pulses (g3,g4).

Calibration of 90° (g4) 13C shaped pulses:

; invg4
; for calibration of 90deg shaped pulse on f2 (i.e. G4)
; 1) acquire spectrum with program as is, phase to absorption

9
; 2) delete the ";;", acquire another spectrum, and optimize sp1 &
sp2 for zero signal

; F1 1H 4.7ppm (H2O)
; F2 13C ~20ppm (select methyl group with unique 1H shift in 1D)

"d20=d2+p13"
Check o2p
1 ze
2 d1 do:f2
3u
20u pl7:f1
(p29 ph0):f1 ; presaturation
d13
20u pl1:f1
20u pl2:f2
(p1 ph1):f1
d2;;0 ; d20 or d2 Compensate for p13
3u
(p13:sp1 ph3):f2 ; 90 G4 (z->xy) spnam1 spoffs1
(p2 ph0):f1
(p13:sp2 ph4):f2 ; 90 G4tr(xy->z) spnam2 spoffs2
3u
;; (p13:sp1 ph4):f2 ; 90 G4
d2 pl12:f2
go=2 ph31 cpd2:f2
d11 wr #0 do:f2
exit

ph0=0
ph1= 0
ph3= 0 2
ph4= 0 0 2 2
;ph4= 1 1 3 3
ph2= 0
ph8=1
ph29=0
ph31=0 2 2 0

1. off resonance presat, 90° 1H, delay of 1/(2J HC) , 90° shaped pulse (g4) (p13, sp1,
spoffs1), 180° 1H, 90° shaped pulse (g4 time reversed), delay of 1/(2J HC), detection
=> get a spectrum that has to be phased and stored for reference.

2. off resonance presat, 90° 1H, delay of 1/(2J HC) + delay of p13, 90° shaped pulse (g4)
(p13, sp1, spoffs1), 180° 1H, 90° shaped pulse (g4 time reverse), 90° shaped pulse (g4),
delay of 1/(2J HC), detection.

why change a delay here, and not in the 13C and 15N hard pulse calibration?
=> optimize sp1 and sp2 for zero signal

10
Eminvg3
Calibration of 180° 13C shaped pulses (change the pulseprogram in this
dataset to Eminvg3 (use pu in XWINNMR)

off resonance presat, 90° 1H, delay of 1/(2J HC) , 90° shaped pulse, 180° 1H, 180° shaped
pulse on 13C (g3 = p14 sp3 spoffs3, or an other 180° shaped pulse).

; invg3

; for calibration of 180deg shaped pulse (i.e. G3)

; 1) acquire and phase spectrum with pulse program invg4
; 2) then use this pulse program, optimize sp3 for zero signal

; F1 1H 4.7ppm (H2O)
; F2 13C ~20ppm (select methyl group with unique 1H shift in 1D)

"d22=d2-p14+p13-d13"

1 ze
2 d11 do:f2
20u pl1:f1
20u pl2:f2
20u pl7:f1
(p29 ph0):f1 ; presaturation
3u
30u pl1:f1
(p1 ph1):f1 90 protons
d2 ;d2=1/2J(HC)
(p13:sp1 ph3):f2 ; 90 G4 90 carbon, g4
(p2 ph0):f1 180 protons
(p14:sp3 ph4):f2 ; 180 G3 180 carbon, g3
d13
d22 pl12:f2
go=2 ph31 cpd2:f2
d11 wr #0 do:f2
d11
exit

ph0=0
ph1= 0
ph2= 0
ph3= 0 2
ph4= 0
;ph4 = 1
ph31=0 2

=> check the effect the shaped pulses have, compared to the normal hard
pulses, by modifying the pulseprogram EMinvg3 and using hard pulses on
carbon

11
cal/6 EMinvg3g4bs
Determination / compensation of Bloch Siegert phases

; EMinvg3g4bs
; Bloch Siegert Phase (BSP) determination

"d2=3.57m"
"d4=(p14-p2)/2"
"d23=100u"

1 ze
2 d1 do:f2
20u pl7:f1
3 (p18:sp0 ph0):f1 ; presat
lo to 3 times 10
3u
20u pl1:f1
20u pl2:f2
(p1 ph0):f1 ; 90 proton
d2
3u
;-------------- start mq coherence HxCx ---
(p13:sp1 ph3):f2 ; 90 13Ca
3u
d23
3u
(p14:sp3 ph12:r):f2 (d4 p2 ph0):f1 ; 180 proton 180deg ca
3u
d23
3u
(p13:sp2 ph4:r):f2 ; 90 on Ca
;------------- end mq coherence HxCx ---
3u
d2 pl12:f2
go=2 ph31 cpd:f2
d11 wr #0 do:f2
d11
exit

ph0=0
ph3= 0 2
ph4= 0 0 2 2
ph12=0
ph31=0 2 2 0

; p1 proton 90
; p2 proton 180
; p13 sp1 carbon 90 (G4)
; p13 sp2 carbon 90 (G4tr)
; pl1 power for protons
; ph12 phase to correct for BSP (zero order)

1. (Change phcor12 and phcor4 to 0) Run pulseprogram. Spectrum should be

very similar to the one in cal/3.
=> phase and save spectrum in procno2 (wrp 2 y)

12
 2. Change the phase of the last 90° pulse on 13C (ph4) by 90°(from "x" to
"y")
=> no signal should be observed (why?), save spectrum in procno3

3. Include a 180° off-resonance pulse (on C’) in the delay after the 180 deg
on the Cα. This will introduces a Bloch Siegert phase on the Cα. Change
delays accordingly!
=> residual signal is as a result of BSP, save spectrum in procno4

4. change the phase of the 180° on-resonance 13C pulse (phcor12)

=> compensate for the zero order BSP, compare spectrum with the one in
procno3

5. change delays around the 180deg on Cα

=> compensate first order BSP, compare spectrum with procno3

Other option for BSP compensation:

6. add a second 180° off-resonance 13C pulse, change delays accordingly and
confirm that the BSP is compensated for.

cal/2 Calibration of water flip back pulse

Write a WFB pulse in the pulseprogram in cal/2 and calibrate it.
(use a 2.5ms gauss pulse, 270deg; (p27:sp11 ph27:r):f1 )
Calibrate both the powerlevel and the phase correction.

Now change the parameters so that the WFB is on resonance.

Using gs, test whether the watersignal can be minimized further by changing
the power and phase of the wfb pulse.

cal/7 EMgradcal calibration of gradients for echo-antiecho

1. 1D SE HSQC with gradient coherence selection using gradients p16:gp2

and p16:gp3. If ratio gp2:gp3 is γH:γN signal should be maximum. However,
the gradient could be non-linear.
=> use PAROPT on gpz3 and look for maximum signal

2. Also check whether the WFB pulse can be optimized further

cal/8 Emwurstcal calibration of an adiabatic WURST pulse

Define a shape of a wurst pulse in the shapetool (use 2359 us pulselength,

1000 points)
Save the pulse under Emwurst

13
Calibrate this pulse (powerlevel) and see that after a certain powerlevel the
pulse stays 180deg.

14
 4. 1D & basic 2D pulse sequences (heteronuclear correlation)

During this session some “real” experiments will be set up and analyzed.

Type "re 1h 1 1 v embo" to access the 1d 1H data sets.

Type "re 15n 1 1 v embo" to access the 15N data sets.
Type "re 13c 1 1 v embo" to access the 13C data sets.

embo/1 EMzgpr presat on resonance and 90° 1H.

embo/2 EM1dwg 1D spectrum using WATERGATE 3-9-19 for water
suppression
=> compare the water suppression.

2D pulse sequences
Set up and compare a normal SE HN hsqc and a 15N SE trosy:

embo/3 EMhnhsqcse H,N HSQC using sensitivity enhancement with gradients

; Ehnhsqc_dec d4
; X-nucleus on f3 (p5 ph8 3u p6 ph0 3u p5 ph8):f3 (d27 p2 ph0):f1
d4
#include <Avance.incl> (p1 ph0):f1
#include <Grad.incl> d29
(p2 ph0):f1
"d29=p16+d16+3u+100u" 3u
"d27=(p6*2+6u-p2)/2" p16:gp3
"d4=2.7m" d16 pl13:f3
;"d28=p2+p16+d16+6u" ;-----------------------------------------------------------
"d28=p2+p16+d16+p14*2+6u" 100u BLKGRAD
"d0=3u+in0/2" ; 90/-180 phase in F1 go=2 ph31 cpd3:f3
;"d0=3u" d1 do:f3 wr #0 if #0 zd
"d8=p1" 5m ip6 igrad EA
"d9=p5" 5m ip6
lo to 3 times 2 ; E/AE
1 ze 10m id0
2 d1 do:f3 5m ip3
3 15m 5m ip3 ; States-TPPI
4 6m 5m ip31
5 3m 5m ip31
6 1m lo to 4 times l0
30u pl1:f1 exit
30u pl3:f3
;----------------- INEPT (Hz -> HzNy) -----------------------
(p1 ph0):f1 ph0=0
d4 ph1=1
(p5 ph8 3u p6 ph0 3u p5 ph8):f3 (d27 p2 ph0):f1 ph2=0
d4 ph3=0 2
(p1 ph1):f1 ph4=0
3u ph5=0
20u pl11:f1 ph6=1
(p27:sp11 ph27:r):f1 ; H2O sel. gauss_1.256, 2.5m, 270deg ph8=1
3u ph27=0 ; ph27=0
100u UNBLKGRAD ph31=0 2
p16:gp1
d16 pl1:f1
(p5 ph3):f3
;--------------------- 15N RT evol. (t1) --------------------- ;gpz1 70,strength z-gradient 1

15
 d0 ; td1/2 ;gpz2 90,strength z-gradient 2 (E/A)
(p14:sp3 ph0):f2 ; dec. Ca (on res. Caliph. G3 268u) ;gpz3 9.1,strength z-gradient 3 (E/A)
(p2 ph0):f1 ; dec. 1H ;gpnam1 SINE.100
(p14:sp4 ph0):f2 ; dec. CO (G3 268u offset: +16500Hz) ;gpnam2 SINE.100
d0 ; td1/2 ;gpnam3 SINE.100
p16:gp2*EA ;p14 180deg G3 pulselength (250u)
d16 ;sp3 180deg G3 powerlevel (onres)
(p6 ph0):f3 ;sp4 180deg G3 powerlevel (offres)
3u ;d1 relaxation delay (1sec)
d28 ;d16 gradient recovery delay
;------------------ SE (HzNy -> Hx) -------------------------- ;p1 proton 90deg hardpulse
(p5 ph4):f3 ;p2 proton 180deg hardpulse
d8 ;p5 nitrogen 90deg hardpulse
(p1 ph0):f1 ;p6 nitrogen 180deg hardpulse
d4 ;p16 gradient length
(p6 ph0):f3 ;pl1 powerlevel for proton hardpulse
(p2 ph0):f1 ;pl2 powerlevel for carbon hardpulse
d4 ;pl3 powerlevel for nitrogen hardpulse
(p5 ph6):f3 ;p27 waterflipback pulse (2.5 ms, 270deg)
d9 ;pl11 powerlevel for proton decoupling
(p1 ph8):f1 ;pl13 powerlevel for nitogen decoupling
;in0 nitrogen indirect
;l0 number of nitrogen increments
;sp11 powerlevel for waterflipback pulse
;pcpd3 pulselenght for nitrogen decoupling

Change the pulseprogram so thatthere is no decoupling during t1 and t2

=> 4 multiplet-lines should appear due to the 1J(H,N) coupling

embo/4 EMtrosy H,N TROSY with E/AE gradient selection

=> 1 peak (corresponding to the slowest relaxing multiplett-line from 15n/5

16
; mstrosy2 d16 exit
; trosyetf3gp (p1 ph3:r)
; avance-version ;
; 2D H-1/X correlation via TROSY (p5 ph4):f3 ph0=0
; using sensitivity improvement d0 gron0 ph2=1
; phase sensitive using Echo/Antiecho gradient selection d0 gron0*-1 ph3=1
; using f3 - channel 10u groff ph4= 0 2
; M. Czisch & R. Boelens, J. Magn. Reson. 134, 158-160 (1998) p19:gp2*-1*EA ph6=1 ; = 1 alpha/beta
; K. Pervushin, G. Wider & K. Wuethrich, J. Biomol. NMR 12, d16 ph7=0 ; = 0 alpha/beta
; 345-348 (1998) (p6 ph0):f3 ph31=0 2
; A. Meissner, T. Schulte-Herbrueggen, J. Briand & O.W. DELTA
Sorensen, Mol. Phys. 96, p19:gp2*EA
; 1137-1142 (1998) d16 ;pl1 : f1 channel - power level for pulse (default)
; J. Weigelt, J. Am. Chem. Soc. 120, 10778-10779 (1998) ; ;pl3 : f3 channel - power level for pulse (default)
; M. Rance, J.P. Loria & A.G. Palmer III, J. Magn. Reson. 136, (p1 ph6:r) ;p1 : f1 channel - 90 degree high power pulse
91-101 (1999) 4u ;p2 : f1 channel - 180 degree high power pulse
; G. Zhu, X.M. Kong & K.H. Sze, J. Biomol. NMR 13, 77-81 p16:gp3 ;p16: homospoil/gradient pulse [1 msec]
(1999) d16 ;p19: gradient pulse 2 [500 usec]
DELTA1 pl1:f1 ;;;p21: f3 channel - 90 degree high power pulse
#include <Avance.incl> (CEN_HN2 p2 ph0) (p6 ph0):f3 ;p5: f3 channel - 90 degree high power pulse
#include <Grad.incl> 4u ;;;;p22: f3 channel - 180 degree high power pulse
#include <Delay.incl> DELTA1 ;p6: f3 channel - 180 degree high power pulse
p16:gp3 ;d0 : incremented delay (2D) [3 usec]
"p2=p1*2" d16 ;d1 : relaxation delay; 1-5 * T1
"p6=p5*2" (p1 ph0) ;30m: delay for disk I/O [30 msec]
(p5 ph2):f3 ;d16: delay for homospoil/gradient recovery
"d4=2.7m" 4u ;d4 : 1/(4J)YH
;;"d0=3u" p16:gp4 ;cnst4: = J(YH)
d16 ;l3: loop for phase sensitive 2D using E/A method: l3 =
;"DELTA=d0*2+p5*4/3.1416+10u" DELTA1 td1/2
"DELTA=d0*2+p5*4/3.1416+10u+in0/2" (p6 ph0):f3 ;in0: 1/(2 * SW(X)) = DW(X)
"DELTA1=d4-p16-d16-4u" d22 ;nd0: 2
"d22=(p19+d16+8u)/2" ; tauG/2 (p2 ph0) ;NS: 4 * n
"d23=DELTA1-d22" d23 ;DS: 16
"d17=d16+p6" p16:gp4 ;td1: number of experiments
d16 ;MC2: echo-antiecho
"CEN_HN1=(p5-p1)/2" 4u
"CEN_HN2=(p6-p2)/2" (p5 ph7:r):f3 ;for z-only gradients:
4u ;gpz0: 3%
p19:gp5 ;gpz1: 30%
1 ze d17 ;gpz2: 80%
2 42m 4u BLKGRAD ;gpz3: 45%
3 42m go=2 ph31 ;gpz4: 50%
4 d1 pl1:f1 30m wr #0 if #0 zd ;gpz5: 16.2%
50u UNBLKGRAD 6m ip6*2 igrad EA
(p1 ph0) 6m ip7*2 ;use gradient files:
4u lo to 3 times 2 ;gpnam1: SINE.100
p16:gp1 30m id0 ;gpnam2: SINE.50
d16 6m ip4*2 ;gpnam3: SINE.100
DELTA1 6m ip31*2 ;gpnam4: SINE.100
(CEN_HN2 p2 ph0) (p6 ph0):f3 lo to 4 times l0 ;gpnam5: SINE.50
4u
DELTA1
p16:gp1

For the TROSY experiment:

1. confirm that the 15N Zeeman magnetization is added to and not
subtracted from you signal
2. change phases/gradients in order to select the other (undesired) multiplett
lines.

embo/5 Emhalftrosy The Trosy principle is only used in one dimension

1. check signal to noise compared to HSQC and Trosy

embo/6 EMhchmqc H,C HMQC using presat for water suppression

17
 embo/7 EMhchsqc H,C HSQC using presat for water suppression
1. compare the 13C signals of aromatic and methyl signals in the HMQC vs.
HSQC implementation. Make sure the decoupling power is sufficient
2. modify the pulse sequence (rename it!) in order to use an adiabatic 180°
pulse (WURST, remember?) during the H,C INEPT transfers and compare
the effect for the 13C signals far off-resonance

Filters
2 types of filters will be demonstrated. Both showing the unlabelled components in a
mixture of protein (doubly labeled) and peptide (unlabelled) and purge all magnetization
of protons bound to 15N and 13C. Those experiments can be easily used as building
blocks for filtered 2D and 3D experiments.

embo/8 Emfilter 1d filter, using normal pulses.

; mszgxfcng (p29 ph29):f1 (p13:sp2 ph4):f2 ;g4tr.256 aliph.

; 1D, presat 3u 3u
; 12C aliph. + ar. double filter (G4) 100u UNBLKGRAD 20u pl13:f3 pl12:f2 BLKGRAD
; 14N filter 20u pl1:f1 (p10 ph8):f1
; with Gradients (p1 ph1):f1 go=2 ph31 ;;cpd2:f2 ;;cpd3:f3
23u d11 do:f2 do:f3 wr #0
#include <Avance.incl> p16:gp1 exit
#include <Grad.incl> d16
d14 ;3.1m ph0= 0
"d14=3.1m-p16-d16" (p13:sp5 ph5):f2 ;g3.256 arom. ph1= 0
"d15=0.5m" ; 0.5m d15 ;0.5m ph2= 0 1 2 3
"d6=1.1m" ; 0.9m -> 1.1m (p13:sp1 ph3):f2 ;g4.256 aliph. ph3= 0 0 0 0 2 2 2 2
"d7=d14+p13+d15-p5*4-6u-p14-d6-p5" (p2 ph2):f1 ph4= 1 1 1 1 1 1 1 1 3 3 3 3 3 3 3 3
"p10=0.5m" ; trim pulse 1H (p5 ph14 3u p6 ph4 3u p5 ph14):f3 ph14=0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2
(p14:sp3 ph0):f2 ;g3.256 aliph. ph5 =0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 ze d6 ;0.5m 2222222222222222
2 d11 do:f3 do:f2 (p5 ph3):f3 ;90 (15N) ph8=3
20u pl2:f2 d7 ;~ 3.1m ph29=0
20u pl3:f3 p16:gp1 ph31=0 2
20u pl7:f1 d16

embo/9 Emfilterwurst 1d filter, using a WURST pulses for carbon inversion

18
; rsxfcn_wurst p19:gp1 ph31=0 2
; 1D, using wurst pulse (kay 97) d19 fq2:f2 ;50 ppm (wurst optimum)
; 12C aliph. + ar. double filter (G4) d14
; 14N filter ;p1 proton 90 deg
; with Gradients (d29 p2 ph0):f1 (p15:sp7 ph0):f2 (d28 p6 ph0):f3 ;p2 proton 180 deg
;p5 nitrogen 90 deg
#include <Avance.incl> 3u ;p6 nitrogen 180 deg
#include <Grad.incl> d15 fq2:f2 ;13C to 67ppm for purge ;p10 proton trim pulse
(p3 ph15 3u p3 ph9):f2 ;p13 180 deg g3
"p2=p1*2" ;p14 90 deg g4
;-------------------end first 13C filter, beginning 2nd ;p15 wurst pulselength 2359us
"p6=p5*2" ;p19 gradient pulse
3u ;p27 wfb 270deg 2.5m
;"d26=p5-(p3+p3+3u)" p19:gp1 ;p28 wfb 90deg 1.0m
"d26=(p3+p3+3u)-p5" d19 ;p29 presat pulse length
"d28=(p15-p6)/2" ;center wurst and p6 d18
"d29=(p15-p2)/2" ;center wurst and p2 p19:gp2 ;pl1 proton powerlevel
"d27=d28-p5" d19 fq2:f2 ;13C at 50ppm ;pl2 carbon powerlevel
;pl3 nitrogen powerlevel
"d4=2.2m" (d29 p2 ph0):f1 (p15:sp7 ph0):f2 (p5 ph17 d27 p6 ;pl7 presat powerlevel
ph0):f3 ;pl12 carbon decoupling power
"d14=d4-p19-d19-3u-d29" ;Ta=2.2m ;pl13 nitrogen decoupling power
"d15=d4-d29-p3-p3-3u-3u" ;Ta=2.2m 3u
;"d17=d4-p19-d19-3u-d29-p5-20u" p19:gp2 ;sp1 90 deg g4 power
"d17=d4-p19-d19-3u-d29-p5-d26-20u" d19 fq2:f2 ;13C at 67 ;sp2 90 deg g4 power
"d18=d4-p19-d19-p19-d19-d29-3u" d17 BLKGRAD ;sp3 180 deg g3 power
;(d26 p3 ph16 3u p3 ph9):f2 (p5 ph0):f3 ;sp4 180 deg g3 power
(p3 ph16 3u p3 ph9):f2 (d26 p5 ph0):f3 ;sp5 90 deg g4 power
;sp6 90 deg g4 power
1 ze ;----------------------------------------end 2nd filter ;sp7 wurst power (+/- 15dB)
1m pl1:f1 pl2:f2 pl3:f3 3u ;sp11 wfb
2 d11 do:f3 do:f2 ;sp12 wfb
20u pl2:f2 go=2 ph31
20u pl3:f3 d11 do:f2 do:f3 wr #0 ;d11 recycle delay
20u pl7:f1 exit ;d19 gradient recovery delay
(p29 ph29):f1
3u ph0= 0 ;gpnam1 SINE.50
100u UNBLKGRAD ph1= 0 2 ;gpz1 gradient
20u pl1:f1 ph9= 1 ; carbon after purge ;gpnam2 SINE.50
3u ph15=0 0 2 2 ;first carbon purge ;gpz2 gradient
(p1 ph1):f1 ph16=0 0 0 0 2 2 2 2 ;second carbon purge
;----------------------beginning first 13c filter ph17=0 0 2 2 ;nitrogen purge ;spnam7 wurst40.2000
3u ph29=0 ;presat

Run both pulse programs and compare their performance.

5. 3D pulse sequences

Attached is an HNCA pulse sequence with an error. Change the pulseprogram, to get a
proper 3D HNCA. Check the 2D planes of the 3D experiment.

embo/10

19
; EMhnca_wrong (p5 ph3):f3 d5
; semi-constant-time evolution (15N) d21 (p1 ph0)
; N->H: COS-INEPT & gradient coherence selection 20u pl11:f1 d29
; minimizing water saturation: p27 2.5m gauss_1.256 (p30 ph8):f1 (p2 ph0)
3u 3u
#include <Avance.incl> d22 cpds1:f1 GRADIENT(cnst23)
#include <Grad.incl> (p14:sp3 ph0):f2 (d27 p6 ph0):f3 d16 pl13:f3
d23 50u BLKGRAD
;; GRDPROG: 3sineea_2 40 90 9.1 (p5 ph0):f3 go=2 ph31 cpd3:f3
"d4=2.6m" ; 1/4J(H,N) (p13:sp1 ph4):f2 d1 do:f3 wr #0 if #0 zd
"d21=5.4m" ; 1/2J(H,N) d10 3m ip16
"d23=11m" ; 1/8J(N,CA) (p14:sp4 ph0):f2 (d27 p6 ph0):f3 3m ip16
;;"d0=3u+in0/2" d10 lo to 3 times 2
;;"d14=3u+in0/2" (p14:sp3 ph13):f2 10m ip6*2
"d20=d23" 3u 10m ip16*2
;;d10=3u+in10/2 (p14:sp4 ph0):f2 ; BSP 10m ip26*2
"d22=d23-d21-23u-p30" 3u 5m ip31
"d24=d23-d21-p14" (p13:sp2 ph0):f2 5m ip31
"d25=d23-3u-p14-d24-29u-p30-p14*2-p16-d16" (p5 ph5):f3 10m id0
"d29=3u+p16+d16+50u" d20 ; dd20 IN20=d20/l0 10m id14
"d7=(p6*2+6u-p2)/2" (p6 ph0):f3 10m dd20
"d27=(p14-p6)/2" d14 ; id14 IN14=IN0-IN20 lo to 4 times l0 ; 15N
"d8=p1" (p14:sp3 ph0):f2 5m rd0
"d9=p5" d24 5m rd14
(p14:sp4 ph0):f2 5m rd20
1 d11 ze d0 ; in0 IN0=1/2SWH 5m ip4
d11 pl13:f3 3u do:f1 lo to 5 times 2
20u pl2:f2 (p30 ph18):f1 10m id10
2 d1 do:f2 do:f3 (p14:sp4 ph0):f2 lo to 6 times l10 ; 13c
6m 3u exit
3 20m 20u
4 15m GRADIENT(cnst22) ph0= 0
5 10m d16 ph2= 1
6 d11 pl1:f1 d25 ;;pl1:f1 ph27=2 ; 2 --> H2O +z
p1 ph0 (p5 ph6):f3 ph3= 0 0 2 2
d4 pl3:f3 d8 ph4= 1 3
(p5 ph0 3u p6 ph0 3u p5 ph0):f3 (d7 p2 ph0):f1 (p1 ph0):f1 ph5= 0
d4 d4 ph6= 0
(p1 ph2):f1 (p6 ph6):f3 ph13=0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3
3u (p2 ph0):f1 ph16= 3
(p27:sp11 ph27:r):f1 ; H2O selective d4 ph26= 1
3u (p5 ph16):f3 ph12= 1
20u pl1:f1 d9 ph8= 1
100u UNBLKGRAD (p1 ph12):f1 ph18= 3
GRADIENT(cnst21) d4 ph31=0 2 2 0
d16 (p5 ph6 3u p6 ph26 3u p5 ph6):f3 (d7 p2 ph0):f1

20