J. Pineal Res.

2002; 32:3440 Copyright Munksgaard, 2002

Journal of Pineal Research
ISSN 0742-3098

Inuence of light intensity, spectrum and orientation on

sea bass plasma and ocular melatonin

Bayarri MJ, Madrid JA, Sanchez-Vazquez FJ. Inuence of light intensity, M.J. Bayarri, J.A. Madrid and
spectrum and orientation on sea bass plasma and ocular melatonin. J. Pineal F.J. Sanchez-Vazquez
Res. 2002; 32:3440. Munksgaard, 2002 Department of Physiology, Faculty of Biology,
University of Murcia, Murcia, Spain
Abstract: Melatonin is involved in the transduction of light information and
the photoperiodic control of many important physiological functions in sh.
Although articial photoperiods have been used to improve sh growth
and manipulate reproduction, there is little information about the
characteristics of light `quality'. In this paper we describe the eects of a light
pulse in the middle of the dark phase on plasma and ocular melatonin in
European sea bass. We rst determined the light intensity necessary to elicit a
melatonin response using white light of varying intensities (0.6600 lW/cm2,
experiment 1). Secondly, we tested the effect of the light spectrum on
melatonin production using three differently coloured lights (half-peak
bandwidth 434477, 498575 and 610687 nm for the blue, green and red
lamp, respectively, experiment 2) and, nally, we determined the effect of
light orientation (downwards directed versus upwards directed, experiment 3).
The results show that the minimum light intensity needed to inhibit or
stimulate melatonin levels in both plasma and the eye was 6.0 lW/cm2.
A linear correlation was found between the logarithm of light intensity and
the relative inhibition. In addition, the blue wavelength was more effective in Key words: Dicentrarchus labrax, light pulses,
decreasing melatonin levels in the former and increasing the levels in the melatonin, upwards light, wavelength
latter. Nevertheless, red light at sufcient intensity proved effective at Address reprint requests to
signicantly suppressing circulating melatonin. Downwards light had a F.J. Sanchez-Vazquez, Department of
Physiology, Faculty of Biology,
greater effect than upward-directed illumination in suppressing plasma
University of Murcia, 30100-Murcia, Spain.
melatonin. In conclusion, the results point to the importance of giving proper E-mail: javisan@um.es
consideration to the characteristics of light, to adequately control melatonin Received February 28, 2001;
production and its related physiological processes. accepted June 13, 2001.

In sh, melatonin rhythms are thought to regulate the

Introduction
The central role of melatonin in the synchronization of
the vertebrate circadian system is widely acknowledged
(Cassone, 1998). In sh, as in other vertebrates, circulating
melatonin is derived mainly from the pineal gland, which
transduces light information and secretes melatonin into
the blood (Falcon et al., 1992). Although the pineal was
initially considered the only source of melatonin, Gern and
Ralph (1979) demonstrated that the trout retina synthesizes
melatonin rhythmically. Unlike in higher vertebrates,
melatonin production in sh has been sparingly investi-
gated. In sea bass, melatonin rhythms in the plasma and eye
were rst reported by Sanchez-Vazquez et al. (1997). These
authors pointed out the special interest of this sh species
for melatonin studies because sea bass exhibit inverse
proles in the plasma and eye, with plasma melatonin rising
at night while ocular melatonin increases during daytime.
Further investigations revealed that acute exposure of sea
bass to light at night provoked a drop in plasma melatonin
but a rise in ocular melatonin (Iigo et al., 1997).
temporal co-ordination of many physiological processes,
including smoltication and reproduction (for a review, see
Ekstrom and Meissl, 1997). Indeed, exposing sh to
articial photoperiods has been used extensively to control
the timing of smoltication completion and gonadal devel-
opment in salmonids (Hansen et al., 1992; Sigholt et al.,
1995) and sea bass (Prat et al., 1999). Melatonin adminis-
tration has even been successfully tested to mimic the short
photoperiodic eects in masu salmon (Amano et al., 2000)
and melatonin implants and/or pinealectomy can modify
the timing of sea water adaptation of Atlantic salmon
(Porter et al., 1998). However, the light characteristics
(intensity and spectrum) required by sh to respond to a
given light treatment remain little explored. Although some
electrophysiological studies have described the luminance
and chromatic response of the pineal (Ekstrom and Meissl,
1997), better knowledge of such light requirements for
melatonin regulation is needed, particularly regarding the
`quality' of light, as sh inhabit a highly dynamic photo-
environment, where the water column acts as a potent lter

34

which modies both the light spectrum and intensity (Nicol,
1989).
The wavelength limits causing a photobiological response
can be determined using polychromatic action spectra,
which may be complex to interpret but can be more directly
related to ambient responses than precise analytical action
spectra which require sophisticated equipment and a range
of at least six wavelengths (Coohill, 1991).
The objective of the present paper was to investigate the
inhibitory/stimulatory eect of a light pulse on circulating
and ocular melatonin in sea bass. To this end, we took
samples in the middle of the dark phase to measure
melatonin in dierent groups of sea bass after a 1 hr expo-
sure to white light of varying intensities (0.6600 lW/cm2,
experiment 1). Once the light intensity threshold necessary
to elicit a melatonin response had been determined, in a
second experiment we tested the effect of the light spectrum
on melatonin production using three different coloured
lamps with half-peak bandwidth (434477, 498575 and
610687 nm for the blue, green and red lamp, respectively).
Finally, the effect of light orientation (downwards versus
upwards) on melatonin levels was investigated in a third
experiment to test the effectiveness of such angular distri-
bution of light.
Materials and methods
Animals and housing
For the study, which consisted of three experiments running
from April to June, three dierent lots of animals were used:
for the rst experiment, 57 sea bass of 48.9 6.6 g in body
weight were used; for the second experiment, 63 sea bass of
57.9 14.0 g, from the same stock, and for the third 29 sea
bass of 56.5 15.3 g (mean S.D.) from a dierent
stock. Fish came from the IEO (Spanish Oceanographic
Institute) at Mazarron (Murcia), except from the third lot,
which came from CULMAREX, S.A., at Aguilas (Murcia).
At both sites the sh were reared under natural photoperiod,
temperature and salinity conditions. They were transferred
to the laboratory after a short acclimatization period.
Three to eight weeks prior to the experiments, the sh were
maintained in tanks in isolated chambers, with individual
water circuits, under controlled environmental conditions
(water temperature and salinity were maintained constant at
23C and 2.5%, respectively). The tanks were lled with
articial seawater (Premium, Sera, Toronto, Canada), well
aerated and continuously recycled through biological and
gravel lters. Illumination (on average 118 lW/cm2 at the
water surface) was supplied by an individual light source
in each chamber, consisting of a white uorescent tube
(GRO-LUX, 40 W, Germany) located at the top of the
tanks. The light schedule in each chamber was controlled by
an electronic timer, set at 12 hr light:12 hr dark (LD 12:12).
Animals were fed ad libitum twice a day at scheduled
times with a commercial diet (Proaqua, Palencia, Spain).
Experimental design
The three experiments were designed to investigate the
eect of 1 hr long light pulses of dierent light intensities
Light characteristics and sea bass melatonin
(experiment 1), spectrum (experiment 2), and orienta-
tion (experiment 3). Light measurements were taken in a
dark room with a spectroradiometer (Analytical Spectral
Devices FieldSpec Handheld, Boulder, CO, USA), placing
the light source 28 cm (the same distance between lights
and the water surface) from the light receptor.
Experiment 1: inuence of light intensity
Here we tested the eect of ve dierent pulses of white light
with decreasing intensities (600, 60, 6, 2.4 and 0.6 lW/cm2)
on melatonin production. To this end, ve groups of eight
sea bass were housed individually in ve isolated chambers,
each one with a light phase of 12 hr. However, lights
switched on differed by 30 min in each chamber (07.00,
07.30, 08.00, 08.30 and 09.00 hr), to enable the light pulse
to be given and three groups of sh to be sampled in one
night. The experiment was performed on two consecutive
nights. The light pulses were given in the middle of the dark
phase using a fan-cooled light source (Leica CLS 100,
innitely adjustable brightness with constant colour tem-
perature, Heerburgg, Switzerland), using an optic bre ring
diffuser (Leica near-vertical illuminator). A blue lter
(Medilan, Lan Optics International, Burlington, MA,
USA) was attached to the light source for colour compen-
sation to produce a white, day-light spectrum (see Fig. 1).
Immediately after the light exposure, sh were anaesthet-
ized with phenoxy-ethanol (0.3 ppm) and sampled under
the same light intensity that they had been exposed to
during the light pulse. As controls, two other groups of
eight sh were sampled in the middle of the light phase
(ML) and the dark phase (MD). The latter was sampled
using a dim red light and covering the sh head with
aluminium foil. For the sampling, body weight and length
were measured and blood and eye samples were taken.
Fig. 1. Spectral composition, expressed as the percentage of irra-
diance between 390 and 750 nm wavelengths, of the white daylight
(Leica CLS 100, blue ltered and innitely adjustable brightness
with constant colour temperature) used in experiments 1 and 3, and
the three coloured lights used in experiment 2. The `blue-appearing'
light (half-peak bandwidth 434477 nm with a peak of 454 nm)
was provided by an actinic uorescent lamp (Marine-Glo, Japan),
while the green (half-peak bandwidth 498575 nm with a peak of
544 nm) and the red (half-peak bandwidth 610687 nm with a
peak of 640 nm) lights were obtained by ltering the Leica CLS 100
light source. Irradiance (lW/cm2) was measured by spectroradio-
meter (Analytical Spectral Devices FieldSpec Handheld), with the
light receptor placed at 28 cm from the light source.
35
Bayarri et al.
Blood was removed by caudal punction with heparinized
syringes and collected into heparinized Eppendorf tubes.
The sh were then decapitated and the crystalline lens and
cornea were removed, while the rest of the eye cup was
frozen immediately at )70C. After centrifugation, plasma
was frozen at )20C.
Experiment 2: inuence of light spectrum
In this experiment we investigated the eect of three
coloured light sources with three dierent spectra: blue
(half-peak bandwidth 434477 with a peak of 454 nm),
green (half-peak bandwidth 498575 with a peak of
544 nm) and red (half-peak bandwidth 610687 with a
peak of 640 nm). The blue light was obtained using an
actinic uorescent lamp (Marine Glo, Hagen, Japan) with a
blue lter for colour correction (Fig. 1). For the green and
red light we used the same light source as in experiment 1,
but ltered through a green (Medilan) or red lter (Cokin
3RED, Paris, France), respectively (see Fig. 1 for spectra).
Two light intensities were tested: threshold (6.0 lW/cm2)
and subthreshold (2.4 lW/cm2), as previously determined in
experiment 1 with a daylight source. Light intensity was
adjusted by partially covering the lamps with aluminium
foil. Fish were housed in six groups of 78 sea bass, and, two
additional groups of sh were used as ML and MD controls.
As in experiment 1, the sh were anaesthetized with
phenoxy-ethanol and sampled under the same light spec-
trum and intensity that they had been exposed to during the
light pulse.
Experiment 3: inuence of light orientation
In this study we compared the eects of exposing sh with a
light pulse given at the water surface or at the bottom of a
glass aquarium. To this end, we used a group of eight sea
bass in a 60-L glass aquarium, the walls and surface of
which were covered with black card to avoid reection.
We used the same white light source as used in experiment 1.
The light diuser was placed 28 cm from the bottom of the
aquarium, and the light intensity was adjusted to
60 lW/cm2 (well above the light threshold for plasma
melatonin), measuring the illumination level inside the tank
with an underwater light receiver. For the group receiving
the downwards light, we used seven sea bass exposed to a
light pulse of 60 lW/cm2 at the water surface and supplied
downwards. We also sampled ML and MD groups as
controls. Fish were transferred to the glass aquarium just
before the light pulse began and, as soon as the pulse
nished, they were anaesthetized and sampled.
Melatonin analysis
To determine melatonin concentration in plasma and retina,
the immunoenzymoassay method was used, with a com-
mercial enzyme-linked immunoabsorbent assay (ELISA)
kit (IBL, Hamburg, Germany). Plasma samples were puri-
ed immediately after defrosting, with C18 phase extraction
columns in the centrifuge. Eye cup samples were individu-
ally homogenized at 0C in 2 mL phosphate-buffered saline
(PBS) 10 mM phosphate buer containing 140 mM NaCl
36
and 0.1% sodium azide, pH 7.5 with 1% albumin (Fraction V;
Sigma, St Louis, MO, USA). They were then puried with
the extraction columns. After that, 50 lL of each sample
were added to dierent wells of an ELISA plate precoated
with capture antibody. They were incubated with the
melatonin-biotin and antiserum solutions for 15 hr at 4C.
The wells were then washed with the assay buer (phos-
phate buer with tween and stabilizer) and the plate was
incubated with the enzyme-labelled solution (antibiotin-
alkaline phosphatase in TRIS buer with stabilizers) for
2 hr at room temperature and constant shaking. After
washing the plate again, it was incubated with the p-nitro-
phenyl phosphate solution for 30 min, before adding 50 lL
of the stop solution [1 N NaOH with 0.25 M ethylenedi-
amine tetracetic acid (EDTA)]. Absorbance was read at
405 nm.
Data analysis
Data are expressed as mean S.E.M. values. Excel and
SPSS were used for data analysis. To determine the
relationship between light intensity and the relative inhi-
bition or stimulation of melatonin production, a regression
analysis was performed. The percentage of plasma mela-
tonin inhibition was calculated using the following for-
mula: 100)(100 x/y), where x is the concentration of
circulating melatonin after a light pulse and y is the
melatonin level of the MD control. Similarly, the relative
stimulation of ocular melatonin was dened as 100 x/z,
where z is the melatonin level in the ML control. In the
three experiments, the statistical dierences between mean
melatonin levels were analysed by ANOVA followed by
Duncan's test, with P < 0.05 taken as the statistically
signicant threshold.
Results
Plasma and ocular melatonin showed opposing levels in
ML and MD, the highest melatonin levels occurring in the
dark for plasma and in the light for the eye. One hour long
light pulses of sucient intensity also had opposing eects
on plasma and ocular melatonin, with increasing light
intensity generally inducing an increase in ocular melatonin
and a decrease in plasma melatonin (Fig. 2).
Ocular melatonin peaked signicantly after 1 hr of 6 and
600 lW/cm2 light exposure although unexpectedly it did
not increase after the 60 lW/cm2 pulse (Fig. 2A). On the
other hand, the threshold light pulse value which induced a
signicant decrease in plasma melatonin compared with the
MD control, was the light intensity of 6.0 lW/cm2. Above
this light intensity, plasma melatonin levels remained low
and did not differ from those of the ML control (Fig. 2B).
Of note is the fact that the light intensity threshold required
to signicantly stimulate/inhibit melatonin was similar for
the pineal and the eye, respectively.
The regression analysis of the inhibiting or stimulating
eect of the light pulses on melatonin production revealed a
linear correlation between melatonin and the logarithm of
light intensity (Fig. 3). In the plasma, light pulses of grea-
ter light intensity induced stronger relative inhibitions
(Fig. 3B). In contrast, in the eye, the relative stimulation

Fig. 2. Impact of 1 hr light pulses of increasing intensities (0.6, 2.4,
6, 60 and 600 lW/cm2) on melatonin levels in the sea bass eye (A)
and plasma (B). Mid-light (ML, white bar) and mid-night (MD,
black bar) samples were also taken as controls. The values are the
mean S.E.M. The number of replicates in each group is indi-
cated at the bottom of the bars. Dierent letters indicate statisti-
cally signicant dierences from each other (ANOVA, Duncan's
test, P < 0.05).
Fig. 3. Linear regression analysis between the logarithm of light
intensity and the relative inhibition/stimulation of ocular (A) and
plasma (B) melatonin, respectively. The percentage of inhibition of
plasma melatonin was dened as 100)(100 x/y), where x is the
concentration of circulating melatonin after a light pulse and y is
the melatonin level of the MD control, while the relative stimula-
tion of ocular melatonin was dened as 100 x/z, where z is the
melatonin level in the ML control.
of ocular melatonin tended to increase with increasing light
intensities, although the response was limited, and so not
signicant in the statistical analysis (Fig. 3A).
As with white light, coloured light pulses inuenced
ocular and plasma melatonin dierently although in this
case the response also varied depending on the type of light
Light characteristics and sea bass melatonin
Fig. 4. Impact of 1 hr light pulses of three coloured lights (blue,
green and red) of two intensities (2.4 and 6 lW/cm2). The group
receiving the red light pulse of 2.4 lW/cm2 is labelled R2.4, the red
light pulse of 6 lW/cm2, R6, the green light pulse of 2.4 lW/cm2,
G2.4, and so on. The values are the mean S.E.M. The number
of replicates in each group is indicated at the bottom of the bars.
Different letters indicate statistically signicant differences from
each other (ANOVA, Duncan's test, P < 0.05).
and the intensity used (Fig. 4). In the case of ocular
melatonin, maximum sensitivity was shifted towards shor-
ter wavelengths (blue) as a red light pulse (6.0 lW/cm2)
failed to signicantly increase melatonin concentrations.
However, the green light pulses (2.4 and 6.0 lW/cm2)
increased ocular melatonin concentrations compared with
the MD control, while the blue light proved effective at
2.4 lW/cm2 and at a higher intensity further increased
melatonin concentration up to the levels recorded in ML
(Fig. 4A).
Plasma melatonin was also modied by the type and
intensity of light, the lower intensity red light pulses not
decreasing melatonin levels signicantly, although green
and blue did. Indeed, the blue (and somewhat the red) light
pulse of 6.0 lW/cm2 suppressed plasma melatonin to the
levels of the ML control group. Curiously, the green light
pulses also decreased melatonin concentration, although
not to the ML control levels (Fig. 4B).
As seen in the preceding experiments, melatonin showed
contrary ML and MD levels in the plasma and the eye.
When the sh were exposed to a 1 hr white light pulse of
60 lW/cm2 supplied at the water surface (downwards) or
at the glass bottom (upwards), the response differed
between the tissues (Fig. 5). In the eye, neither light pulse
increased the melatonin content, although the downwards-
oriented light provoked a decrease (Fig. 5A). In plasma, the
light pulse appeared more eective at suppressing melato-
nin when given at the water surface than at the bottom of
the aquarium. As a result, the upward-light pulse group did
not dier signicantly from the MD group, while the
downward-light pulse group showed a melatonin concen-
tration which fell almost to the ML control level (Fig. 5B).
37
Bayarri et al.
Fig. 5. Eect of the light orientation on ocular (A) and plasma
(B) melatonin levels. Fish were exposed to a 1 hr light pulse of
60 lW/cm2 given at the water surface (downwards) or at the bot-
tom (upwards) of a glass aquarium. The values are the mean
S.E.M. The number of replicates in each group is indicated at the
bottom of the bars. Different letters indicate statistically signicant
differences from each other (ANOVA, Duncan's test, P < 0.05).
Discussion
The underwater photic environment diers greatly from the
terrestrial, as sun rays are partially reected at the water
surface, refracted on entering the water and scattered and
dierentially absorbed during passage through the water.
As a result, with increasing depth, sunlight is altered in its
spectral composition and attenuated in intensity (Nicol,
1989). Indeed, our results revealed that light of decreasing
intensity (simulating greater depth), dierent spectral
quality (blue, green and red) and orientation (downwards-
versus upwards-directed), strongly inuenced melatonin
production in sea bass. The light threshold required to
modify melatonin contents was 6 lW/cm2 for both the eye
and plasma, indicating a similar sensitivity to white light (at
least as regards melatonin production, exp. 1). In addition,
light of shorter wavelengths (blue) proved more effective
than those of the green and red spectrum (exp. 2), and down-
wards light had a greater effect than upwards illumination
in suppressing plasma melatonin (exp. 3).
Light intensity and the growth rate of sh have received
considerable attention (Brett, 1979). Indeed, there is little
information about the inuence on growth of light quality
but more about light intensity and much more about
photoperiod (Bouef and Le Bail, 1999). For instance, to
develop normally and grow, a minimal threshold of light is
required by sh larvae, whose response may be wavelength-
dependent (Gehrke, 1994). In the case of sea bass, both
growth and feed eciency appeared linearly correlated with
light irradiance in the 3.569.5 lx range (Hidalgo et al.,
1993). As regards melatonin, light pulses of dierent
38
intensities have previously been used to investigate melato-
nin production in a teleost sh, brook trout (Salvelinus
fontinalis) (Zachmann et al., 1992). Despite methodological
dierences (particularly as regards the light source used),
serum melatonin in brook trout decreased in an intensity-
dependent manner when subjected to light pulses of 0.2
2000 lx, with signicant changes being found at intensities
of 2 lx and higher. Although comparisons between these
values and the present results should be made with care
(dierent light sources and intensity conversions), brook
trout as well as sea bass melatonin appeared to exhibit great
sensitivity to light, with a threshold of about 6 lW/cm2
(around 10 lx). Ocular melatonin in brook trout, as in sea
bass, increased with increasing light intensities, although
with a slightly higher threshold (20 lx).
To date, the role played by melatonin in the eye is not
fully understood, although the fact that ocular melatonin
rhythms are common among vertebrates suggests an
important role in retinal function (Iuvone, 1986; Cahill
et al., 1991). Melatonin eects are mediated by specic high-
anity receptors coupled to quanosine 5-triphosphate
(GTP)-binding proteins (Reppert, 1997). In mammals,
ocular melatonin is under the control of a circadian oscil-
lator, persists in vitro and is temperature-compensated and
light-entrained, so that production is high at night both in
the eye and in the pineal (Tosini, 2000). This nocturnal
increase in ocular melatonin strikingly contrasts with our
ndings in sea bass, which not only showed high melatonin
levels at ML, but also responded to light pulses at MD by
increasing their melatonin content in the eye. This photo-
response of sea bass ocular melatonin to light pulses reects
previous ndings, which also showed higher ocular mela-
tonin levels during the daytime and at night after exposure
to light for 1 hr (Iigo et al., 1997). The role of this diurnal
melatonin rhythm in the eye, however, is unclear. In
mammals, ocular melatonin has specic receptors in the
retina and does not contribute to circulating levels, indica-
ting that melatonin may act locally as a neuromodulator,
regulating, together with dopamine, the retinal physiology
and light and dark adaptation (Tosini, 2000). In sh, there
is evidence to support the existence of two types of
serotonin N-acetyltransferase, the key enzyme controlling
melatonin biosynthesis, with dierent kinetic characteristics
in the trout retina and pineal, and suggesting a dierent
endocrine and paracrine role for plasma and ocular
melatonin, respectively (Benyassi et al., 2000). For instance,
in the sea bass eye, melatonin may be involved in the
control of retinomotor responses to light (myoid contrac-
tion and pigment dispersion) in a phasic manner (morning
melatonin peaks), supplying a single signal, rather than
sustaining high melatonin levels. In addition, ocular mela-
tonin rhythms in sea bass are not consistent throughout the
year and may disappear at certain seasons for yet unknown
reasons (Garc a-Allegue et al., 2001). This complex beha-
viour of the sea bass eye may explain why ocular melatonin
levels varied considerably between the three experiments
and did not respond consistently to the light pulses.
Clear oceanic waters are most transparent to blue light of
460475 nm and strongly absorb photons of longer and
shorter wavelengths, which are further absorbed and
scattered by dissolved pigments and suspended particles

in coastal waters, where maximal light transmission is
shifted towards longer wavelengths (Nicol, 1989). Curi-
ously, the photopigments of sh inhabiting dierent photo-
environments (deep sea, coastal and freshwater) show
adaptations, so that maximum sensitivity occurs at the
available wavelengths of the corresponding environment,
although the response is displaced towards shorter wave-
lenghs for coastal and freshwater species (Lythgoe, 1980).
Our results in sea bass, a coastal species, are consistent with
the above observations, as the maximal stimulation and
suppression of ocular and plasma melatonin, respectively,
were induced by the `blue-appearing' light with half-peak
bandwidth 434477 nm. In mammals, on the other hand,
melatonin synthesis (N-acetyltransferase activity) in the rat
pineal is quickly suppressed, specially by `yellow-green'
light (kmax 548 nm), but less affected by blue or red light
(Jarmak et al., 1998). Nevertheless, in the Syrian hamster a
blue uorescent light (half-peak bandwidth 435500 nm)
was the most effective in suppressing pineal melatonin
(Brainard et al., 1984). The spectral sensitivity of phase
shifting circadian locomotor rhythms have also been
investigated, revealing that the sensitivity curve has a
maximum nearing 500 nm in hamsters (Takahashi et al.,
1984), while in mice discrete pulses of both kmax 359 nm
and kmax 511 nm produced large phase shifts (Provencio
and Foster, 1995). In the sh pineal, electrophysiological
techniques have shown that photoreceptor cells display
variable spectral sensitivity, which suggests the existence of
two photopigments (rodopsin and porphiropsin) with peak
sensitivities of about 460500 and 520560 nm (Ekstrom
and Meissl, 1997). In the sh retina, four cone types with
absorption spectra ranging from 360 to 620 nm have been
described (Kraaij et al., 1998). Further research, however,
is required to look for possible differences in the spectrum
for retinal and pineal responses, and its role in melatonin
production and spectral sensitivity.
Although the structure of the sea bass retina has been
well studied (Mani-Ponset et al., 1993), the sensitivity of the
photoreceptors to dierent wavelength are unknown. In
striped bass (Morone saxatilis), maximum absorbance for
rods is 528 nm, for single cone, 542 nm and for twin cone,
605 nm (Miller and Korenbrot, 1993). Sea bass are most
likely to have colour vision because the anatomy of the eye
and behavioural responses (colour preferences), point to its
capacity to discriminate colour dierences (Paspatis et al.,
1994).
The minimum light intensity to suppress melatonin may
be time-dependent, because the minimum light intensity
decreases as the duration of exposure increases (Aoki et al.,
1998). Nevertheless, the pineal appears to be very sensitive
to light and NAT inhibition in rats can be observed after
15 s exposure to light (Jarmak et al., 1998). In sh, in vitro
studies suggest that melatonin production is most sensitive
to light stimulus exceeding 30 min and with wavelengths
peaking near 500 nm (Max and Menaker, 1992). Melatonin
release in superfusion cultures of the trout pineal is
dependent on light irradiance and increases with decreasing
light intensity (Yanez and Meissl, 1996). A similar relation
between light intensity and plasma melatonin was also
found in sea bass, melatonin increasing (relative inhibition
decreased) as light intensity decreases (Fig. 3B).
Light characteristics and sea bass melatonin
Daylight penetration in the sea is greatest in the direction
of the sun (zenith light), at least in shallow waters, where
downwards light is not equally distributed in all azimuths
because of scattering (Nicol, 1989). Submersible lights may
be used in net cages to apply articial photoperiods, but
little is known about the eectiveness of such angular
distribution of light. Most sun rays reach the sh head
dorsally (except scattering light which may reach it later-
ally), as the pineal is located at the top of the brain and thus
receives downward-directed light. As we observed in experi-
ment 3, downwards light of sucient light intensity proved
signicantly more eective than upwards light in suppres-
sing plasma melatonin. In the case of ocular melatonin,
however, there were no signicant dierences, which may
be explained by the fact that the lateral eyes can receive
light from many directions.
In conclusion, our results provide the rst evidence for
an eect of light irradiance, spectrum and orientation, on
sea bass ocular and plasma melatonin. Our data suggested
that both the pineal and the retina, where the plasma and
ocular melatonin are produced, are equally sensitive to
light and showed a similar light threshold (6.0 lW/cm2).
Photons of shorter wavelengths (blue) also appeared more
effective than those of longer wavelengths (red), although
the latter may signicantly affect plasma melatonin at
threshold intensities so that extreme care should be taken
when using supposedly `safe' dim red lights to sample sea
bass during nigh-time. Finally, as plasma melatonin must
be suppressed below a certain threshold before articial
photoperiods are capable of altering melatonin rhythms
(Porter et al., 1999), not only the number of light hours
(photoperiod) but also the power, colour and orientation
of lamps (underwater illumination) should be considered
when designing an effective articial lighting system for
sh.
Acknowledgements
This research was funded by the CICYT project no.
1FD97-1699 to Dr F.J. Sanchez-Vazquez. Authors thank
CULMAREX, S.A and the IEO (Spanish Oceanographic
Institute) at Mazarron (Murcia), for kindly providing the
sh, and Rosa Garc a-Allegue and Jose Luis Patino for
their help during sampling.
References
AMANO, M., M. IIGO, K. IKUTA, S. KITAMURA, H. YAMADA,
K. YAMAMORI (2000) Roles of melatonin in gonadal matur-
ation of underlying precocious male masu salmon. General
Comp. Endocrinol. 120:190197.
AOKI, H., N. YAMADA, Y. OZEKI, H. YAMANE, N. KATO (1998)
Minimum light intensity required to suppress nocturnal mela-
tonin concentration in human saliva. Neurosci. Lett. 252:
9194.
BENYASSI, A., C. SCHWARTZ, S.L. COON, D.C. KLEIN, J. FALCON
(2000) Melatonin synthesis: arylalkalylamine N-acetyltrans-
ferase in trout retina and pineal organ are dierent. Neuro-
report 11:255258.
BOUEF, G., P.Y. LE BAIL (1999) Does light have an inuence on sh
growth? Aquaculture 177:129152.
39
Bayarri et al.
BRAINARD, G.C., B.A. RICHARDSON, T.S. KING, R.J. REITER (1984)
The inuence of dierent light spectra on the suppression of
pineal melatonin content in the Syrian hamster. Brain Res.
294:333339.
BRETT, J.R. (1979) Environmental factors and growth. In: Fish
Physiology, Vol. VIII. W.S. Hoar, D.J. Randall, J.R. Brett, eds.
Academic Press, New York, pp. 599675.
CAHILL, G.H., M.S. GRACE, J.C. BERSHARSE (1991) Rhythmic
regulation of retinal melatonin: metabolic pathways, neuro-
chemical mechanisms, and the ocular circadian clock. Cell.
Mol. Neurobiol. 11:529560.
CASSONE, V.M. (1998) Melatonin role in vertebrate circadian
rhythms. Chronobiol. Int. 15:457473.
COOHILL, T.P. (1991) Action spectra again? Photochem. Photobiol.
54:859870.
EKSTROM, P., H. MEISSL (1997) The pineal organ of teleost shes.
Rev. Fish Biol. Fish. 7:199284.
FALCON, J., C. THIABAULT, V. BEGAY, A. ZACHMANN, J.P. COLLIN
(1992) Regulation of the rhythmic melatonin secretion by sh
pineal photoreceptor cells. In: Rhythms in Fishes M.A. Ali, ed.
Plenum, New York, pp. 167199.
GARCIA-ALLEGUE, R., J.A. MADRID, F.J. SANCHEZ-VAZQUEZ
(2001) Melatonin rhythms in European sea bass plasma and
eye: inuence of seasonal photoperiod and water temperature.
J. Pineal Res. 31:6875.
GEHRKE, P.C. (1994) Inuence of light intensity and wavelength on
phototactic behaviour of larval silver perch Bidyamus bidyanus
and golden perch Macquaria ambigua and the eectiveness of
light traps. J. Fish Biol. 44:741751.
GERN, W.A., C.L. RALPH (1979) Melatonin synthesis by the retina.
Science 204:183184.
HANSEN, T., S.O. STEFANSSON, G.L. TARANGER (1992) Growth and
sexual maturation in Atlantic salmon, Salmo salar L., reared
in sea cages at two dierent light regimes. Aquacult. Fish.
Manag. 23:275280.
HIDALGO, F., J.J. NOGALES, A. ORTEGA (1993) Preliminary study
on the eect of light intensity on growth and food eciency in
sea bass (Dicentrachus labrax). Actas IV Congreso Nac.
Acuicult. 1:109113.
IIGO, M., F.J. SANCHEZ-VAZQUEZ, J.A. MADRID, S. ZAMORA,
M. TABATA (1997) Unusual responses to light and darkness
of ocular melatonin in European sea bass. Neuroreport 8:
16311635.
IUVONE, P.M. (1986) Rhythms of melatonin biosynthesis in retina:
involvement of calcium, cyclic AMP and dopamine in the
regulation of serotonin N-acetyltransferase. In: Pineal and
Retinal Relationships. P.J. O'Brien, D.C. Klein, eds. Academic
Press, New York, pp. 197218.
JARMAK, A., J.B. ZAWILSKA, J.Z. NOWAK (1998) Eect of white and
monochromatic lights of various wavelengths on the nocturnal
serotonin N-acetyltransferase activity suppression in the pineal
gland of rat. Klinika Oczna 100:7780.
KRAAIJ, D.A., M. KAMERMANS, H. SPEKRIJSE (1998) Spectral
sensitivity of the feedback signal from horizontal cells to cones
in goldsh retina. Vis. Neurosci. 15:799808.
LYTHGOE, J.N. (1980) Vision in sh: ecological adaptations.
In: Environmental Physiology of Fishes M.A. Ali, ed. Plenum
Press, New York, pp. 15391550.
MANI-PONSET, L., J.P. DIAZ, P. DIVANACH, M. KENTOURI, R.
CONNES (1993) Structure de la retine et potentialite visuelles
40
susceptibles d'inuer sur le comportement trophique du loup
(Dicentrarchus Labrax) adulte et en cours du developpement.
Eur. Aquacult. Soc. Spec. Publishers 18:359372.
MAX, M., M. MENAKER (1992) Regulation of melatonin produc-
tion by light, darkness, and temperature in the trout pineal.
J. Comp. Physiol. A170:479489.
MILLER, J.L., J.I. KORENBROT (1993) Phototransduction and
adaptation in rods, single cones, and twin cones of the striped
bass retina: a comparative study. Vis. Neurosci. 10:653667.
NICOL, J.A.C. (1989) The photoenvironment. In: The Eyes of
Fishes. Clarendon Press, Oxford.
PASPATIS, M., M. KENTOURI, M. ANTHOUARD, P. DIVANACH
(1994) 2-choice discrimination and preference of sea bass
Dicentrarchus labrax among 9 colours. In: Measures for
Success. P. Kestemont, J. Muir, F. Sevila, P. Williot, eds.
EAS Spec Publ No 21, Oostende, pp 188190.
PORTER, M.J.R., N.J. DUNCAN, D. MITCHELL, N.R. BROMAGE
(1999) The use of cage lighting to reduce plasma melatonin in
Atlantic salmon (Salmo salar) and its eects on the inhibition
of grilsing. Aquaculture 176:237244.
PORTER, M.J.R., C.F. RANDALL, N.R. BROMAGE, J.E. THORPE
(1998) The role of melatonin and the pineal gland on devel-
opment and smoltication of Atlantic salmon (Salmo salar)
parr. Aquaculture 168:139155.
PRAT, F., S. ZANUY, N. BROMAGE, M. CARRILLO (1999) Eects of
constant short and long photoperiod regimes on the spawning
performance and sex steroid levels of female and male sea bass.
J. Fish Biol. 54:125137.
PROVENCIO, I., R.G. FOSTER (1995) Circadian rhythms in mice can
be regulated by photoreceptors with cone-like characteristics.
Brain Res. 694:183190.
REPPERT, S.M. (1997) Melatonin receptors: molecular biology of a
new family of G protein-coupled receptors. J. Biol. Rhythms
12:528531.
SANCHEZ-VAZQUEZ, F.J., M. IIGO, J.A. MADRID, S. ZAMORA,
M. TABATA (1997) Daily cycles in plasma and ocular melatonin
in demand-fed sea bass, Dicentrarchus labrax L. J. Comp.
Physiol. 167B:409415.
SIGHOLT, T., M. STAURNES, H.J. JAKOBSEN, T. ASGARD (1995)
Eects of continuous light and short-day photoperiod on
smolting, seawater survival and growth in Atlantic salmon
(Salmo salar). Aquaculture 130:373388.
TAKAHASHI, J.S., P.J. DECOURSEY, L. BAUMAN, M. MENAKER
(1984) Spectral sensitivity of a novel photoreceptive system
mediating entrainment of mammalian circadian system.
Nature 308:186188.
TOSINI, G. (2000) Melatonin circadian rhythm in the retina of
mammals. Chronobiol. Int. 17:599612.
YANEZ, J., H. MEISSL (1996) Secretion of the metoxyindolamines
melatonin, 5-metoxytriptophol, 5-metoxyindoleacetic acid, and
5-metoxytriptamine from trout pineal organs in superfusion
culture: eect of light intensity. General Comp. Endocrinol.
101:165172.
ZACHMANN, A., S.C.M. KNIJFF, M.A. ALI, M. ANCTIL (1992)
Eects of photoperiod and dierent intensities of light expo-
sure on melatonin levels in the blood, pineal organ, and retina
of the brook trout (Salvelinus fontinalis Mitchill). Can. J. Zool.
70:2529.