Implication of the visual system in the
regulation of activity cycles in the absence of
solar light: 2-[125I]iodomelatonin binding sites
and melatonin receptor gene expression in the
brains of demersal deep-sea gadiform ®sh
Imants G. Priede1*, Lynda M. Williams2, Hans-Joachim Wagner3, Amanda Thom2,
Ian Brierley2, Martin A. Collins1, Shaun P. Collin4 , Nigel R. Merrett5
and Cynthia Yau1
1
Department of Zoology, University of Aberdeen,Tillydrone Avenue, Aberdeen AB24 2TZ, UK
The Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB, UK
3
Anatomisches Institut, UniversitÌt TÏbingen, O«sterbergstrasse 3, D-72074 TÏbingen, Germany
4
Department of Zoology, University of Western Australia, Nedlands 6907, Western Australia, Australia
5
Department of Zoology, Natural History Museum, Cromwell Road, London SW7 5BD, UK
2
Relative eye size, gross brain morphology and central localization of 2-[125I]iodomelatonin binding sites
and melatonin receptor gene expression were compared in six gadiform ¢sh living at di¡erent depths in
the north-east Atlantic Ocean: Phycis blennoides (capture depth range 265^1260 m), Nezumia aequalis (445^
1512 m), Coryphaenoides rupestris (706^1932 m), Trachyrincus murrayi (1010^1884 m), Coryphaenoides guentheri
(1030 m) and Coryphaenoides (Nematonurus) armatus (2172^4787 m). Amongst these, the eye size range was
0.15^0.35 of head length with a value of 0.19 for C. (N.) armatus, the deepest species. Brain morphology
re£ected behavioural di¡erences with well-developed olfactory regions in P. blennoides, T. murrayi and
C. (N.) armatus and evidence of olfactory de¢cit in N. aequalis, C. rupestris and C. guentheri. All species had a
clearly de¢ned optic tectum with 2-[125I]iodomelatonin binding and melatonin receptor gene expression
localized to speci¢c brain regions in a similar pattern to that found in shallow-water ¢sh. Melatonin
receptors were found throughout the visual structures of the brains of all species. Despite living beyond
the depth of penetration of solar light these ¢sh have retained central features associated with the
coupling of cycles of growth, behaviour and reproduction to the diel light^dark cycle. How this functions
in the deep sea remains enigmatic.
Keywords: melatonin; deep sea; eye; brain; ¢sh; gadiform
1. INTRODUCTION
Teleost ¢sh have colonized even the deepest parts of the
world's oceans through the acquisition of special adaptations for surviving under conditions of high pressure, low
temperature, darkness and sparse food supply (Childress
1995; Merrett & Haedrich 1997). A less obvious problem
in this extreme environment is temporal synchronization
of physiological and behavioural events. In vertebrate
animals, the hormone melatonin is secreted from the
pineal gland during the hours of darkness. This melatonin
signal plays a key role in the timing of the rhythms of
growth, behaviour and reproduction (Arendt 1995).
Annual periodicity has been discovered in the growth
and reproduction of several deep-sea ¢sh species (Gordon
1979; Merrett & Haedrich 1997) together with evidence
for annual migrations (Jones et al. 1998) and seasonal
*
Author for correspondence ([email protected]).
Proc. R. Soc. Lond. B (1999) 266, 2295^2302
Received 29 July 1999 Accepted 25 August 1999
changes in swimming speeds (Priede et al. 1994a). The
question arises as to how these seasonal rhythms are
regulated in the absence of solar light.
In this study, we investigated whether the hormone
melatonin could be implicated in the control of rhythms
in deep-sea ¢sh by investigating the occurrence of
melatonin receptors in the brains and pituitaries of six
species from di¡erent depths. Detection was by means of
in vitro autoradiography for speci¢c 2-[125I]iodomelatonin
binding sites and melatonin receptor gene expression
using in situ hybridization. In mammals, with the possible
exception of humans, the highest concentration of
speci¢c, high-a¤nity, 2-[125I]iodomelatonin binding sites
and melatonin receptor gene expression is localized to the
pars tuberalis of the pituitary (Morgan et al. 1994). The
suprachiasmatic nucleus (SCN), the biological clock of
the brain, is also a target for the melatonin signal in
many mammalian species. In ¢sh, including gold¢sh,
salmon and rainbow trout, 2-[125I]iodomelatonin binding
2295
& 1999 The Royal Society
I. G. Priede and others The visual system in regulation of activity cycles
sites and melatonin receptor gene expression have been
localized to regions of the brain concerned with the
processing of visual signals, particularly the optic tectum
(Martinoli et al. 1991; EkstrÎm & Vanecek 1992; Davies et
al. 1994). Using in situ hybridization, Brierley et al. (1999)
localized period gene (Per1) expression to the optic tectum
of both trout and cod indicating that, in ¢sh, clock genes
are associated with the visual centres of the brain.
Smith et al. (1996) ¢rst identi¢ed 2-[125I]iodomelatonin
binding sites in the brain of the deep-sea ¢sh Coryphaenoides (Nematonurus) armatus. Binding sites were absent
from the optic tectum, although positive elsewhere in the
brain, suggesting a reduced or highly modi¢ed role for
the melatonin signal in this abyssal species in the absence
of solar light. We have now undertaken a more extensive
study localizing gene expression as well as 2-[125I]iodomelatonin binding in a series of species, of the order
Gadiformes, which occur at depths from 300 m, where
solar light may have an in£uence, down to 5000 m,
beyond the reach of any visually important solar radiation: Phycis blennoides (BrÏnich, 1768) (family Gadidae),
Nezumia aequalis (GÏnther, 1878) (family Macrouridae),
Coryphaenoides rupestris Gunnerus, 1765 (family Macrouridae), Trachyrincus murrayi GÏnther, 1887 (family Macrouridae), Coryphaenoides guentheri (Vaillant, 1888) (family
Macrouridae) and Coryphaenoides (Nematonurus) armatus
(Hector, 1875) (family Macrouridae) (¢gure 1).
0
Phycis blennoides
1000
Nezumia aequalis
2000
depth (m)
2296
Coryphaenoides rupestris
3000
Trachyrincus murrayi
4000
Coryphaenoides guentheri
5000
Coryphaenoides (N.) armatus
2. MATERIAL AND METHODS
(a) Collection of specimens
Fish were captured during August 1997 on Cruise 134 of the
RRS Challenger in the Porcupine Seabight area of the north-east
Atlantic (508 N, 138 W) using an Otter Trawl Semi-balloon 14
net towed along the bottom on a single warp. The depths of
occurrence of each species and depths of capture of specimens
are indicated in ¢gure 1. The brains and pituitaries were
removed as quickly as possible, frozen in iso-pentane chilled
over dry ice (CO2), wrapped in foil and stored at 70 8C for
transport. The nomenclature of the brain nuclei and tectal
layering follows that of Northcutt (1983).
(b) Eye size
Measurements were taken of the eye diameters (deye) and
head lengths (lhead) (from the snout to the posterior margin of
the operculum) from illustrations in Whitehead et al.(1986) of all
demersal species occurring in the Porcupine Seabight, northeast Atlantic Ocean, as listed by Priede et al. (1994). The relative
eye diameter was then expressed as a ratio (deye/lhead) for interspeci¢c comparison.
(c) Localization of 2-[125I ]iodomelatonin binding sites
Cryostat sections (20 mm) were cut and thaw mounted onto
gelatin-coated slides. 2-[125I]iodomelatonin binding studies were
carried out as described previously (Helliwell & Williams 1992)
and the incubation time optimized for sections of ¢sh brain and
pituitary (Davies et al. 1994). Brie£y, slides were incubated with
100 pM 2-[125I]iodomelatonin (NEN Dupont Ltd, UK) in
25 mM Tris^HCl bu¡er containing 4 mM CaCl2 for 2 h at
room temperature in the presence and absence of either 10 6 M
melatonin or 10 6 M guanosine 5-O-3-thiotriphosphate
(GTP gS). Slides were then washed in ice-cold bu¡er, air dried
and apposed to Kodak X-OMAT AR ¢lm along with [125I]
Proc. R. Soc. Lond. B (1999)
Figure 1. Diagrams of the gross morphology of the brains of
the six species of deep-sea demersal ¢sh investigated. The
vertical bars show the depth ranges of each species and the
depths of capture of the particular specimens examined.
Lateral (upper) and dorsal (lower) brain outlines show the
telencephalon, optic tectum, cerebellum and medulla. The
hypothalamus is located ventral to the optic tectum.
Super¢cial and deep regions in which 2-[125I]iodomelatonin
binding was detected are shaded. Vertical bars depict
relatively intense binding over whole layers of neural tissue.
Dots depict binding in localized nuclei or at low intensity.
The drawings are not to scale. The maximum total lengths of
these ¢sh are P. blennoides 80 cm, N. aequalis 27+ cm, C. rupestris
150 cm, T. murrayi 40+ cm, C. guentheri 47 cm and
C. (N.) armatus 100 cm.
polymer standards (Amersham International, UK). Films were
exposed for two to six weeks and the sections were stained with
toluidine blue.
(d) PCR and DNA sequencing
PCR ampli¢cation was performed on rainbow trout genomic
DNA using degenerate primers designed to span the most highly
conserved regions of the melatonin receptor corresponding to
transmembrane domains 3 and 7. Each reaction cycle (32 loops)
consisted of incubations at 94 8C for 30 s, 59 8C for 1min and
72 8C for 1min with Ampli Taq DNA polymerase (Perkin Elmer
Cetus). The PCR products were separated by agarose (1.5%) gel
electrophoresis and subcloned in pGEM-T.
DNA sequencing was carried out using the ABI Prism Dye
Terminator Cycle Sequencing Kit (Perkin Elmer) and the
samples were run and analysed using an Applied Biosystems
DNA Sequencer (Model 373).
The visual system in regulation of activity cycles
I. G. Priede and others
2297
(e) Riboprobe synthesis
For riboprobe synthesis, a PCR fragment corresponding to
the melatonin RT1.4 receptor was inserted into the plasmid
pGEM-T. The plasmid was linearized with either Apa I or Sac I
to give rise to the anti-sense and sense probes, respectively. In
vitro transcription with either T7 or SP6 RNA polymerase
(Promega, Southampton, UK) was carried out. Brie£y, a 1 mg
sample of the linearized DNA template was incubated with
1 transcription bu¡er, 0.4 mM rCTP, 0.4 mM rGTP, 0.4 mM
rATP, 10 mM dithiothreitol (DTT), 1U ml 1 RNAse inhibitor,
1000 Ci mmol 1 8 mCi 35S-a-thio-UTP (NEN) and the appropriate RNA polymerase (T7 for sense and SP6 for anti-sense)
for 1^1.5 h at 37 8C. The DNA template was digested with 10 U
RQ-1 DNAse for 30 min at 37 8C. The riboprobe is separated
from the other components using a Chromaspin 30 column and
checked for incorporation by counting on a liquid scintillation
analyser (Packard 1900TR). The riboprobe was then diluted in
a hybridization bu¡er of 50% formamide, 0.3 M NaCl,
1 Denhardts, 10 mM Tris, 1mM EDTA, 10% dextran sulphate,
10 mM DTT and 0.485 mg ml 1 transfer RNA at a ¢nal concentration of 5^7 103 cpm ml 1 and stored at 70 8C until use in
the in situ hybridization studies.
(f) In situ hybridization
Adjacent sections were ¢xed in 4% 0.1M paraformaldehyde^
phosphate bu¡er (PB) solution (pH 7.0). The slides were rinsed
with 0.1M PB (2 5 min) and then acetylated by immersion in
0.1M triethanolamine (2 min) and 0.0025% acetic anhydride
(10 min). After rinsing (0.1M PB, 2 2 min) the sections were
dehydrated in a graded series of ethanol and dried under
vacuum for 60 min. The sections were then covered with 70 ml
hybridization bu¡er containing 5^7 103 cpm ml 1 of probe and
a cover slip sealed over the sections with DPX mounting
medium and incubated for 16^18 h at 56^59 8C. The slides were
then cooled, the DPX removed and the slides immersed in
4 standard saline citrate (1 trisodium citrate 15 mM, NaCl
150 mM, pH 7.0 (SSC)) for 30 min. After removal of the cover
slips the slides were again washed in 4 SSC (4 5 min) and
immersed in 0.1 SSC containing RNase A (10 mg ml 1 RNase
A, 0.5 M NaCl, 10 mM Tris (pH 8.0) and 1mM EDTA) for
30 min at 37 8C. The slides were then washed to a ¢nal stringency of 0.1 SSC at 60 8C and dehydrated with graded concentrations of ethanol. The sections were then air dried and
exposed to ¢lm (Hyper¢lm-b Max, Amersham) for two weeks.
All chemicals were obtained from Sigma Chemical Co. (Poole,
Dorset, UK) unless otherwise stated.
3. RESULTS
(a) Eye size
Taking 68 species of demersal ¢sh into consideration,
the relative eye size tends to decrease with increasing
depth (¢gure 2). However, at shallower depths, eye size
ranges from 0.07 in the tripod ¢sh Bathypterois dubius to
0.47 in Epigonus telescopus, whereas deeper than 3000 m
eye size converges on an apparent optimum size for
abyssal life of approximately 0.20. The relative eye size in
the six species analysed in the present study lies in the
middle of the general trend for this ¢sh assemblage.
(b) Brain morphology
The brains examined showed considerable variation in
the relative size of the major subdivisions (¢gure 1).
Proc. R. Soc. Lond. B (1999)
Figure 2. Relative eye size in demersal ¢sh species from the
Porcupine Seabight, north-east Atlantic Ocean, in relation to
the mean depth of occurrence. Large closed dots, species in
this study; open circles, other species living in this area.
Although a quantitative analysis of the relative size of
each subdivision will not be attempted here, some qualitative observations regarding the size of the sensory and
motor nuclei may be related to behaviour. In P. blennoides,
the telencephalon, optic tectum and cerebellum are comparable in size. Easily indenti¢able vagal lobes are
situated dorsally. In N. aequalis, the brain is compressed
antero-posteriorly, where the telencephalon is reduced but
the optic tectum, hypothalamus, cerebellum and medulla
(with vagal lobes) are relatively enlarged. In the brain of
C. rupestris, the telencephalon is not reduced but the hindbrain is dominated by hypertrophy of the cerebellum and
vagal lobes. This enlargement of the hindbrain is
illustrated in transverse sections, where the valvula cerebelli and corpus cerebelli protrude anteriorly between the
two lobes of the optic tectum (¢gure 3c). In contrast, the
hindbrain in T. murrayi is reduced and the forebrain is
highly developed, where the large telecephalon is divided
into four anterior lobes accounting for almost half the
total brain mass. The brain of C. guentheri tapers anteroposteriorly with a small telencephalon, medium-sized
tectum and a large cerebellum. The brain of
C. (N.) armatus possesses a small optic tectum, a large
telencephalon and a well-developed cerebellum which
extends dorsally into the spacious cranial cavity.
(c) 2-[125I ]iodomelatonin binding
All species examined showed speci¢c 2-[125I]iodomelatonin binding discretely localized to identi¢able regions of
the brain. This was completely displaced by 1 mM melatonin and greatly reduced by the addition of 10^6 M
GTP gS, indicating the presence of a G-protein-coupled
melatonin receptor. The highest concentration of binding
was found over the inner layers of the optic tectum: the
periventricular grey zone, the deep white zone and the
super¢cial white and grey zone. No binding was observed
in the outer layers of the super¢cial white and grey zone,
except for weak binding in C. rupestris and T. murrayi.
Binding was never observed in the specialized midline
roof of the midbrain, the torus longitudinalis (¢gure 2),
but was always found in the torus semicircularis, which
was located ventral to the optic tectum and is derived
embryologically from the posterior midbrain roof. In
N. aequalis, C. rupestris, T. murrayi and C. guentheri binding
was observed in localized nuclei of the tegmentum.
2298
I. G. Priede and others The visual system in regulation of activity cycles
125I-mel
control
ot
(a)
Mel 1b
control
tl
ts
pgz
ot
(b)
pgz
tl
(c)
c
ts
ot
c
pgz
c
c
n
(d)
(e)
ot
pgz
ot
pgz
ts
(f )
ot
tl
pgz
ts
Figure 3. Autoradiographs of coronal sections of the brain and underlying pituitary of the ¢sh studied showing the distribution
of 2-[125I]iodomelatonin binding (125I-mel) and in situ hybridization (Mel 1b) and corresponding controls. (a) Phycis blennoides.
(b) Nezumia aequalis. (c) Coryphaenoides rupestris. (d) Trachyrincus murrayi. (e) Coryphaenoides guentheri. ( f ) Coryphaenoides (Nematonurus)
armatus. c, cerebellum; n, tegmental nucleus; ot, optic tectum; pgz, periventricular grey zone; tl, torus longitudinalis; ts, torus
semicircularis.
However, the location of these nuclei varied between
species and exact homologies could not be established. No
binding was observed in the hypothalamus except for
clear bilateral labelling in C. rupestris (¢gure 3c).
Speci¢c 2-[125I]iodomelatonin binding was observed in
the pre-tectal area in all species except N. aequalis, where
this area is reduced in size. In all species, both the optic
Proc. R. Soc. Lond. B (1999)
tract and pre-optic region showed comparatively weak
binding. The exact localization and intensity of binding
was very variable in these regions. No binding was found
in the telencephalon except in P. blennoides.
Speci¢c 2-[125I]iodomelatonin binding was found in the
granular layer of the valvula and corpus cerebelli in all
¢sh except in Coryphaenoides spp. where binding was either
The visual system in regulation of activity cycles
I. G. Priede and others
2299
Table 1. Summary of the distribution of ligand binding and gene expression for the melatonin receptor in the brains of deep demersal
¢sh
(Iod, 2-[125I]iodomelatonin binding; Mel1A, gene expression. High +++, medium ++, low + and not detectable 0. Depth of
capture: P. blennoides 747 m, N. aequalis 984 and 1188 m, C. rupestris 1148 m, T. murrayi 1541 m, C. guentheri 2441 and 2565 m, and
C.(N.) armatus 4143 m.)
P. blennoides
Iod
telencephalon
+
pre-optic area
++
optic tract
+
pre-tectum
+
hypothalamus
0
tegmental nuclei
++
torus semicircularis
+
optic tectum
periventricular grey zone ++
deep white and super¢cial +++
white and grey zones
torus longitudinalis
0
valvula cerebelli
++
corpus cerebelli
++
vagal lobe
0
medulla
+
N. aequalis
C. guentheri
C. (N.) armatus
Iod
Mel1A
Iod
Mel1A
Iod
Mel1A
Iod
Mel1A
Iod
Mel1A
0
+
0
0
0
0
0
0
0
0
0
0
+++
+
0
+
0
0
+
+
+
0
+
+
+
++
+
++
0
0
0
0
+
++
0
+
++
+
+
0
+
++
0
+
0
0
+
0
+
0
0
+
+
0
+
+
0
0
0
+
0
0
0
0
0
++
++
0
0
++
0
0
0
0
0
+
0
+++
0
0
+++
+++
0
+
+++
++
0
+
++
++
0
+
+
+
0
+
+
++
0
+++
++
++
0
0
0
++
+
0
+
++
++
++
++
+
0
0
0
0
+
++
+
+
+
+
0
+
++
0
+
+++
++
++
0
0
0
+
+
+
+
+
+
+
0
0
0
0
0
+
++
++
0
0
0
0
The pattern of gene expression detected with the
Mel1b riboprobe matched that of speci¢c 2-[125I]iodomelatonin binding in the tectal region. Melatonin
receptor gene expression was consistently found in all
species in the optic tectum with the signal clearly localized to the periventricular grey zone but absent in the
more peripheral layers where 2-[125I]iodomelatonin
binding was present. The torus longitudinalis always
showed strong gene expression, whereas 2-[125I]iodomelatonin binding was always absent (¢gure 3 and table 1).
In the valvula and corpus cerebelli, melatonin receptor
gene expression was detected in all species except
C. (N.) armatus. In other brain regions, localization of
melatonin receptor gene expression was highly variable
from species to species.
In the midbrain, melatonin receptor gene expression
was only present in the torus semicircularis in N. aequalis
and T. murrayi, whereas in the tegmentum the melatonin
gene was expressed in discrete nuclei in N. aequalis, C. rupestris and C. (N.) armatus but no speci¢c 2-[125I]iodomelatonin
binding could be identi¢ed in these areas. In the hypothalamus, melatonin receptor gene expression was found in
N. aequalis, C. rupestris and T. murrayi with discretely localized bilateral labelling inT. murrayi (¢gure 3).
Posteriorly, vagal lobe melatonin receptor gene expression was generally absent. However, in N. aequalis and
C. rupestris, a weak, di¡use signal was observed
throughout the lobe. In the medulla, melatonin receptor
Proc. R. Soc. Lond. B (1999)
T. murrayi
Mel1A
absent (C. rupestris and C. armatus) or weak (C. guentheri)
(¢gure 3). Binding was also found in the medulla, but
only in species-speci¢c locations, such as the motor
region in P. blennoides, the reticular formation in
N. aequalis, the ventral tegmentum in T. murrayi and in
discrete unidenti¢ed structures in C. (N.) armatus.
(d) In situ hybridization
C. rupestris
gene expression was only found in N. aequalis and
C. rupestris.
Anteriorly, melatonin receptor gene expression was
generally absent in the pre-tectal area and was never
found in the optic tract. In the pre-optic area gene
expression was evident in P. blennoides, T. murrayi and
C. (N.) armatus. There was no melatonin receptor gene
expression in the telencephalon.
4. DISCUSSION
At depths greater than 1000 m, the only visually relevant light present is of biological origin and it has been
suggested that vision becomes less important, based on a
decline in relative eye size with depth (Montgomery &
Pankhurst 1997). Analysis of relative eye size amongst
demersal ¢sh of the north-east Atlantic (¢gure 2) shows
that, despite a trend for a decrease in relative eye size
with an increase in depth, the values for even the deepestdwelling ¢sh are comparable with many shallow-water
species. The sensitivity of eyes to point sources of light
characteristic of the deep sea is a function of the absolute
area of the pupil. Relative eye size normalized to head
length can give only a crude index of the importance of
the visual system. Comparison of ¢sh of di¡erent sizes
could also give rise to misleading results. Despite the
limitations of this index, contrary to Montgomery &
Pankhurst (1997), we conclude that there is no evidence
that the visual systems of deep-dwelling ¢sh have
regressed; indeed, very high sensitivities have evolved to
detect the low, prevailing, light levels (Douglas et al. 1998;
Wagner et al. 1998).
Although the species of ¢sh used in this study were
selected for their close taxonomic relationship and their
relative eye sizes are very similar, their brain morphologies were found to be remarkably diverse, possibly
2300
I. G. Priede and others The visual system in regulation of activity cycles
re£ecting speci¢c adaptations. Nezumia aequalis, C. rupestris
and C. guentheri are never attracted to bait placed on the
sea £oor (Priede et al. 1994b). These are species with relatively small olfactory bulbs and telencephali suggesting
that their olfactory capability is restricted. This olfactory
de¢cit may explain their inability to locate non-motile
dead food items. They feed on free-swimming and
bottom-living prey (Mauchline & Gordon 1986; Gordon
& Mauchline 1990), for which presumably lateral line,
tactile and visual senses are important.
Phycis blennoides has elongate pelvic ¢ns, which we
suggest may have a tactile gustatory function. Trachyrincus
murrayi has a characteristic elongation of its snout, a
greatly enlarged telencephalon and an elaborate olfactory
apparatus, which we suggest may aid in intraspeci¢c
chemical communication and feeding. This species prefers
active, pelagic prey (N. R. Merrett, unpublished data).
Coryphaenoides (Nematonurus) armatus is a well-known
scavenger in the deep sea, rapidly locating carrion by
virtue of its olfactory system (Wilson & Smith 1984;
Armstrong et al. 1992). The importance of olfaction is
further realized in this species by comparing the ratio of
olfactory axons to optic axons, which peaks at 5.9:1 in
larger individuals (S. P. Collin and H.-J. Wagner, unpublished data). It is likely that solar light is visually irrelevant to all the species examined but P. blennoides,
N. aequalis and C. rupestris may experience circadian
changes in bioluminescent light levels associated with
downward migration of organisms of the deep scattering
layer at night. Below 2000 m, the only light is dim bioluminescent events occurring at intervals of 10 min or more
(I. G. Priede, P. Bagley and S. J. Way, unpublished observations). However, tidal changes in the bottom current
direction and speed (Lampitt et al. 1983; Wilson & Smith
1984; Armstrong et al. 1992) may entrain bioluminescent
events providing a periodic optical signal to the ¢sh.
All species examined possess functional melatonin
receptors, notably in the tectal region of the brain.
Despite considerable variation in brain morphology, a
constant feature in all the specimens was the presence of
melatonin receptor gene expression in the periventricular
grey zone of the optic tectum but not in the more peripheral layers. However, 2-[125I]iodomelatonin binding
occurred to varying degrees in all the layers of the optic
tectum indicating that, as in the trout (Mazurais et al.
1999), the location of functional receptors and gene
expression is spatially di¡erentiated in accordance with
the laminar organization of the optic tectum (Butler
1992). This implies that the receptor gene is expressed in
piriform neurons of the periventricular grey zone and the
receptor protein is transported radially along dendrites in
the deep white and super¢cial white and grey zones.
A similar dichotomy occurs in the torus longitudinalis,
which always shows receptor gene expression but no 2[125I]iodomelatonin binding. This paired ridge of densely
packed, small, granular cells is unique to actinopterygian
¢sh. Physiological recordings show strong visual responses
and visuomechanoreceptive and visuomotor coordination
(Meek & Nieuwenhuys 1997). These cells send parallel
¢bres to the outer layer of the super¢cial white and grey
zone of the optic tectum. Therefore, these neurons probably account for some of the iodomelatonin binding found
in the optic tectum.
Proc. R. Soc. Lond. B (1999)
In the midbrain, 2-[125I]iodomelatonin binding and
receptor gene expression were detected to varying degrees
in the torus semicircularis and in nuclei of the ventral
tegmentum. This is similar to results in gold¢sh (Martinoli et al. 1991), salmon (EkstrÎm & Vanecek 1992) and
trout (Davies et al. 1994).
Within the valvula and corpus cerebelli, 2-[125I]iodomelatonin binding and melatonin receptor gene expression were found in all species except the deepest-living
species C. (N.) armatus. In trout, melatonin gene expression
is associated with the ganglionic layer of the cerebellum
which contains Purkinje cells and it is suggested that the
functional receptor is transported along these dendrites to
the molecular layer (Mazurais et al. 1999). Such detail
could not be resolved in the present study but it should be
noted that the cerebellum probably receives inputs from
all the sensory systems and is implicated in visual
processing in trout (Reid & Westerman 1975) and gold¢sh (Pastor et al. 1997). Mazurais et al. (1999) detected
2-[125I]iodomelatonin binding in the medulla and vagal
lobes in trout, but no receptor gene expression. The
results in the present study are quite variable with some
evidence of gene expression in both the medulla and the
vagal lobes in N. aequalis and C. rupestris, where the hindbrain is well developed.
In the forebrain, the results are generally comparable
with those found in the trout, where no receptor gene
expression was found in the telencephalon but occasionally some 2-[125I]iodomelatonin binding did occur
(Mazurais et al. 1999). The pre-optic area, optic tract and
pre-tectum also showed melatonin receptor gene expression in some instances and 2-[125I]iodomelatonin binding
was more common.
The present investigation is restricted by a lack of
detailed knowledge of brain morphology in deep-sea ¢sh.
However, it is apparent that all the species examined here
possess melatonin receptors in those brain regions implicated in the processing of visual signals. It is interesting
to note that clearer di¡erentiation and a more intense
signal of receptor binding sites and gene expression may
be found in shallow-dwelling species (Mazurais et al.
1999) and such a trend is apparent in ¢gure 3 with the
clearest di¡erentiation in P. blennoides, the shallowest
species in this study. However, this may be an artefact of
the collection technique with inevitable delays in recovering ¢sh from the greatest depths.
Evolution of the present deep-sea ¢sh fauna, including
the Macrouridae and Gadidae, has occurred in the last
65 million years and is associated with the development
of the modern deep ocean circulation of oxygenated
water (White 1987; Howes 1991). The extensive central
melatonin receptor system is found in widely divergent
orders of shallow-water ¢sh, the Salmoniformes
(Mazurais et al. 1999), Cypriniformes (Martinoli et al.
1991) and the Gadiformes (present study) suggesting it
was present in ancestral forms at the time of the origin of
the teleost fauna in the upper Jurassic, some 200 million
years ago. Thus, the ¢sh in the present study have almost
certainly retained a central melatonin receptor system
that was present in shallower-dwelling ancestors.
The reproductive, behavioural and growth cycles of
¢sh are highly seasonal and it is probable that melatonin
acting on central receptors mediates these cycles. In a
The visual system in regulation of activity cycles
wide taxonomic range of teleost ¢sh such as the pike
(Falcon et al. 1989), gold¢sh (Kezuka et al. 1989) and
white sucker (Zachman et al. 1992) the pineal releases
melatonin rhythmically under the control of a circadian
oscillator. However, in salmon and trout there is no endogenous rhythm and secretion is directly under the control
of the light^dark cycle (Gern & Greenhouse 1988). Even
in ¢sh with endogenous circadian cycles, the rhythm is
in£uenced by light and temperature.
Until very recently, the deep sea was perceived as a
dark, stable environment with a constantly low temperature and the fauna dependent largely on a constant rain
of organic matter from primary production in the photic
zone above (Wells et al. 1931). Time-lapse cameras (Billett
et al. 1983; Rice et al. 1986) and sediment traps (Honjo &
Docherty 1988; Smith et al. 1994) have revealed that sedimentation of particulate organic matter to the abyssal sea
£oor occurs mostly in the months following the surface
plankton bloom observed at temperate latitudes in spring.
This provides an annually pulsed input of food imposing
a seasonal cycle on microbial and sediment infaunal
metabolism (Smith et al. 1992; Pfannkuche 1993; Pfannkuche & Lochte 1993) and macro- and megafauna
activity above the sediment (Smith et al. 1993). In the
Paci¢c Ocean, Priede et al. (1994) discovered a seasonal
doubling of swimming speeds of C. (N.) armatus. The
otoliths of C. rupestris also have distinct growth rings
(Bergstad 1995), as do the other species considered, indicating a seasonal pattern of growth whilst seasonal reproduction has been shown to occur in C. rupestris, T. murrayi
(Gordon 1979), C. guentheri and N. aequalis (Merrett &
Haedrich 1997). It is reasonable to assume that all the ¢sh
in this present study show some degree of seasonality,
which may be mediated by melatonin.
Smith et al. (1996) detected a restricted distribution of
melatonin receptors in the brain of C. (N.) armatus;
however, in the present study melatonin binding and gene
expression was evident in the optic tectum of all species
including the deepest occurring species C.(N.) armatus.
The ¢sh sampled by Smith et al. (1996) were small ¢sh
captured in April, whereas the ¢sh in the present study
were larger ¢sh captured in August during a di¡erent
phase of the putative deep ocean migration cycle of this
species. Thus, the di¡erence in the spatial distribution of
the central receptors could be a seasonal or developmental e¡ect.
We conclude that, despite living without visually detectable solar light, deep demersal gadiform ¢sh have, in addition to a sensitive visual system, functional central
melatonin receptors which may play a role in the regulation
of cycles of growth, activity and reproduction. If the melatonin is pineal derived it seems unlikely that there would
ever be su¤cient di¡use light in the deep sea to penetrate
the cranium and act directly on a pineal photoreceptor.
The visual system may be implicated in the transduction of
deep-sea bioluminescent signals to the melatonin secretion
system or some other internal or external stimuli may in£uence the secretion of melatonin in the absence of solar light.
The possibility of seasonal variation in bioluminescence
coupled to seasonal changes in water £ow and/or biological
activity cannot be excluded.
It would appear that the function of the melatonin
system, including its receptor molecules and their localiProc. R. Soc. Lond. B (1999)
I. G. Priede and others
2301
zation and expression in parts of the brain associated with
light perception and/or vision, is a very conservative
feature, even though the mediation of or entrainment to
rhythmic temporal cues in the abyss most probably relies
on di¡erent sensory systems. In this way, the original
function of melatonin as a neurochemical signal for
seasonality, synchronizing reproduction or other behaviour, could still be conserved in deep-sea ¢sh.
We thank Bob Duthie (RRI Graphics) for work on the illustrations and the ship's company of the RRS Challenger (Cruise 134)
for capture of ¢sh specimens. The ship was chartered by the
University of Aberdeen and the research was partially supported
by the NERC.
REFERENCES
Arendt, J. 1995 Melatonin and the mammalian pineal gland. London:
Chapman & Hall.
Armstrong, J. D., Bagley, P. M. & Priede, I. G. 1992
Photographic and acoustic tracking observations of the behaviour of the grenadier, Coryphaenoides (Nematonurus) armatus,
the eel, Synaphobranchus bathybius, and other abyssal demersal
¢sh in the North Atlantic Ocean. Mar. Biol. 112, 535^544.
Bergstad, O. 1995 Age determination of deep-water ¢shes:
experiences, status and challenges for the future. In Deepwater ¢sheries of the North Atlantic Ocean slope (ed. A. G. Hopper),
pp. 267^283. Dordrecht, The Netherlands: Kluwer.
Billett, D. S. M., Lampitt, R. S., Rice, A. L. & Mantoura,
R. F. C. 1983 Seasonal sedimentation of phytoplankton to the
deep-sea benthos. Nature 302, 520^522.
Brierley, I., Mazurais, D., Drew, J., Morgan, P., Ross, A.,
Anglade, I., Randall, C., Bromage, N., Kah, O. & Williams,
L. M. 1999 Melatonin receptors and Per1 gene expression in
the teleost ¢sh brain. In Proceedings of the Sixth International.
Symposium of Reproductive Physiology Fish (ed. B. Norberg,
S. Kjesbu, E. Andersson, S. O. Stefansson, G. L. Taranger &
Th. B. Wather), 4^9 July, Bergen, Norway. (In the press.)
Butler, A. B. 1992 Variation of tectal morphology in teleost
¢shes. Brain Behav. Evol. 40, 256^272.
Childress, J. J. 1995 Are there physiological and biochemical
adaptations of metabolism in deep-sea animals? Trends Ecol.
Evol. 10, 30^36.
Davies, B., Hannah, L. T., Randall, C. F., Bromage, N. &
Williams, L. M. 1994 Central melatonin binding sites in
rainbow trout (Onchorhynchus mykiss). Gen. Comp. Endocrinol. 96,
19^26.
Douglas, R. H., Partridge, J. C. & Marshall, N. J. 1998 The
eyes of deep-sea ¢sh. I. Lens pigmenation, tapeta and visual
pigments. Prog. Retinal Eye Res. 17, 597^636.
EkstrÎm, P. & Vanecek, J. 1992 Localization of 2[125I]iodomelatonin binding sites in the brain of the Atlantic
salmon, Salmo salar L. Neuroendocrinology 55, 529^537.
Falcon, J., Brun Marmillon, J., Claustrat, B. & Collin, J. P. 1989
Regulation of melatonin secretion in a photoreceptive pineal
organ: an in vitro study in the pike. J. Neurosci. 9, 1943^1950.
Gern, W. A. & Greenhouse, S. S. 1988 Examination of in vitro
melatonin secretion from superfused trout (Salmogairdneri)
pineal organs maintained under die1 illumination or constant
darkness. Gen. Comp. Endocrinol. 71, 163^174.
Gordon, J. D. M. 1979 Lifestyle and phenology in deep sea
anacanthine teleosts. Symp. Zool. Soc. Lond. 44, 327^359.
Gordon, J. D. M. & Mauchline, J. 1990 Depth-related trends in
diet of a deep sea bottom-living ¢sh assemblage of the
Rockall Trough. In Trophic relations in the marine environment (ed.
M. Barnes & R. N. Gibson), pp. 439^452. Oxford, UK:
Aberdeen University Press.
2302
I. G. Priede and others The visual system in regulation of activity cycles
Helliwell, R. J. A. & Williams, L. M. 1992 Melatonin binding
sites in the ovine brain and pituitary: characterisation during
the oestrous cycle. J. Neuroendocrinol. 4, 287^294.
Honjo, S. & Docherty, K. W. 1988 Large aperture time-series
sediment traps; design objectives, construction and application. Deep-sea Res. 35, 133^149.
Howes, G. J. 1991 Biogeography of gadoid ¢shes. J. Biogeogr. 18,
596^622.
Jones, E. G., Collins, M. A., Bagley, P. M., Addison, S. &
Priede, I. G. 1998 The fate of cetacean carcasses in the deep
sea: observations on consumption rates and succession of
scavenging species in the abyssal north-east Atlantic Ocean.
Proc. R. Soc. Lond. B 265, 1119^1127.
Kezuka, H., Aida, K. & Hanyu, I. 1989 Melatonin secretion
from gold¢sh pineal gland in organ culture. Gen. Comp.
Endocrinol. 75, 217^221.
Lampitt, R. S., Merrett, N. R. & Thurston, M. H. 1983 Interrelations of necrophagous amphipods, a ¢sh predator and
tidal currents in the deep sea. Mar. Biol. 74, 73^78.
Martinoli, M. G., Williams, L. M., Kah, O., Titchener, L. T. &
Pelletier, G. 1991 Distribution of central melatonin binding sites
in the gold¢sh (Carassius auratus). Mol. Cell. Neurosci. 2,78^85.
Mauchline, J. & Gordon, J. D. M. 1986 Foraging strategies of
deep sea ¢sh. Mar. Ecol. Prog. Ser. 27, 227^238.
Mazurais, D., Brierley, I., Anglade, I., Drew, J., Randall, C.,
Bromage, N., Michel, D., Kah, O. & Williams, L. M. 1999
Central melatonin receptors in the rainbow trout: comparative distribution of ligand binding and gene expression. J.
Comp. Neurol. 409, 313^324.
Meek, J. & Nieuwenhuys, R. 1997 Holosteans and teleosts. InThe
central nervous system of vertebrates (ed. R. Nieuwenhuys, H. J. T.
Donkelaar & C. Nicholson), pp.759^937. Berlin: Springer.
Merrett, N. R. & Haedrich, R. L. 1997 Deep-sea demersal ¢sh and
¢sheries. London: Chapman & Hall.
Montgomery, J. & Pankhurst, N. 1997 Sensory physiology. In
Deep-sea ¢shes (ed. D. J. Randall & A. P. Farrell), pp. 325^349.
London: Academic Press.
Morgan, P. J., Barrett, P., Howell, H. E. & Helliwell, R. 1994
Melatonin receptors: localization, molecular pharmacology
and physiological signi¢cance. Neurochem. Int. 24, 101^146.
Northcutt, R. G. 1983 Evolution of the optic tectum in ray¢nned ¢shes. In Fish neurobiology. 2. Higher brain areas and functions (ed. R. E. Davis & R. G. Northcutt), pp. 1^42. Ann
Arbor, MI: University of Michigan Press.
Pastor, A. M., De la Cruz, R. R. & Baker, R. 1997
Characterisation of Purkinje cells in the gold¢sh cerebellum
during eye movement and adaptive modi¢cation of the
vestibulo-ocular re£ex. Prog. Brain Res. 114, 359^381.
Pfannkuche, O. 1993 Benthic response to the sedimentation of
particulate organic matter at the BIOTRANS station 478 N
208 W. Deep-sea Res. II 40, 135^149.
Pfannkuche, O. & Lochte, K. 1993 Open ocean pelago-benthic
coupling: cyanobacteria as tracers of sediment salp faeces.
Deep-sea Res. I 40, 727^737.
Proc. R. Soc. Lond. B (1999)
Priede, I. G., Bagley, P. M. & Smith Jr, K. L. 1994a Seasonal
change in activity of abyssal demersal scavenging grenadiers
Coryphaenoides (Nematonurus) armatus in the eastern North
Paci¢c Ocean. Limnol. Oceanogr. 39, 279^285.
Priede, I. G., Bagley, P. M., Smith, A., Creasey, S. & Merrett,
N. R. 1994b Scavenging deep demersal ¢shes of the Porcupine
Seabight (NE Atlantic Ocean); observations by baited
camera, trap and trawl. J. Mar. Biol. Assoc. UK 74, 481^498.
Reid, M. D. & Westerman, R. A. 1975 Visual responses in the
cerebellum of the trout, Salmo gairdneri. Comp. Biochem. Physiol.
50, 259^262.
Rice, A. L., Billett, D. S. M., Fry, J., John, A. W. G., Lampitt,
R. S., Mantoura, R. F. C. & Morris, R. J. 1986 Seasonal
deposition of phytodetritus to the deep-sea £oor. Proc. R. Soc.
Edinb. B 88, 265^279.
Smith, A., Trudeau, V. L., Williams, L. M., Martinoli, M. G. &
Priede, I. G. 1996 Melatonin receptors are present in nonoptic regions of the brain of a deep-sea ¢sh living in the
absence of solar light. J. Neuroendocrinol. 8, 655^658.
Smith Jr, K. L., Baldwin, R. J. & Williams, P. M. 1992
Reconciling particulate organic carbon £ux and sediment
community oxygen consumption in the deep North Paci¢c
Ocean. Nature 359, 313^316.
Smith Jr, K. L., Kaufman, R. S. & Wake¢eld, W. W. 1993
Mobile megafaunal activity monitored with a time-lapse
camera in the abyssal North Paci¢c. Deep-sea Res. I 40,
2307^2324.
Smith Jr, K. L., Kaufman, R. S. & Baldwin, R. J. 1994
Coupling of near-bottom pelagic and benthic processes at
abyssal depths in the eastern North Paci¢c Ocean. Limnol.
Oceanogr. 39, 1101^1118.
Wagner, H.-J., FrÎhlich, E., Negishi, K. & Collin, S. P. 1998
The eyes of deep-sea ¢sh. II. Functional morphology of the
retina. Prog. Retinal Eye Res. 17, 637^685.
Wells, H. G., Huxley, J. & Wells, G. P. 1931 The science of life.
London: Cassell.
White, B. N. 1987 Oceanic anoxic events and allopatric speciation in the deep sea. Biol. Oceanogr. 5, 243^259.
Whitehead, P. J. P., Bauchot, M.-L., Hureau, J.-C., Neilsen, J.
& Tortonese, E. 1986 Fishes of the North-eastern Atlantic and the
Mediterranean, vol. II. Paris: UNESCO.
Wilson Jr, R. R. & Smith Jr, K. L. 1984 E¡ect of near-bottom
currents on detection of bait by the abyssal grenadier ¢shes
Coryphaenoides spp., recorded in situ with a video camera freevehicle. Mar. Biol. 84, 83^91.
Zachmann, A., Falcon, A., Knij¡, S. C. M., Bolliet, V. & Ali,
M. A. 1992 E¡ects of photoperiod and temperature on
rhythmic melatonin secretion from the pineal organ of the
white sucker (Catostomus commersoni) in vitro. Gen. Comp.
Endocrinol. 86, 26^33.
As this paper exceeds the maximum length normally permitted,
the authors have agreed to contribute to production costs.