Download Characterization of the day-night variation of retinal melatonin

Characterization of the Day-Night Variation of
Retinal Melatonin Content in the Chick
Steven M. Repperr and Stephen M. Sagar
By monitoring two time points (one at mid-light and the other at mid-dark), the day-night variation
of melatonin content in the retina of 19-day-old chicks was characterized. Melatonin was detected
in the retina plus attached pigment epithelium by a specific radioimmunoassay, and its identity was
verified by high performance liquid chromatography with electrochemical detection. Melatonin content
in the posterior pole of the eye showed a fivefold day-night variation, with high levels during the
dark period of diurnal lighting. Light exposure during the dark period lowered the normally high
nighttime value; maintenance of darkness during the normal light period did not alter the low melatonin
values typical of mid-light. Pineal melatonin content responded similarly to the above lighting manipulations. Neither pinealectomy nor optic nerve transection had an effect on retina-pigment epithelium melatonin or its light-dark rhythm. We next examined the relative contributions of retinal
and pineal melatonin to blood levels. Pinealectomy reduced the normally high mid-dark plasma
melatonin value by 80%. The addition of bilateral enucleation reduced the mid-dark value by another
9% of control values. The day-night variation of retina-pigment epithelium melatonin was first evident
in the embryo 2 days prior to hatching and persisted through adulthood. It was concluded that the
chick retina from the latter stages of embryonic development is capable of rhythmically synthesizing
melatonin; that retinal melatonin content displays a photically controlled circadian rhythm in phase
with, but independent of, the pineal gland; and that the retinal rhythm is not regulated by afferent
optic nerve fibers. The pineal gland is the major source of plasma melatonin in the intact chick, with
at most a small contribution from the retina. Invest Ophthalmol Vis Sci 24:294-300, 1983
Our interests in melatonin physiology and the potential endocrine functions of the retina led us to
characterize the retinal rhythm of melatonin content
in the chick by examining its response to alterations
in environmental lighting, by delineating the influence of the pineal gland and of retinal optic nerve
afferents in the generation of the eye rhythm, and by
studying the ontogeny (embryo through adult) of the
day-night variation in retinal melatonin content. An
important aim of our studies was to assess the relative
contributions of retinal and pineal melatonin to circulating levels of the hormone.
It is well established that the major source of melatonin for homoiotherms is the pineal gland.19 This
organ synthesizes melatonin from serotonin by use
of two enzymes, N-acetyltransferase (NAT) and hydroxyindole-O-methyl transferase (HIOMT), with
NAT controlling the well-described daily oscillations
of pineal melatonin production in rats10 and chickens.1 The pineal melatonin rhythm, in turn, is reflected accurately in blood, urine, and cerebrospinal
fluid, with high levels occurring at night.19 Interestingly, the retinae of a number of vertebrate species
have been known for some time to contain the enzymatic machinery necessary for melatonin biosynthesis.715 A resurgence of interest in this retinal capability has led to the recent findings of large daily
rhythms in NAT activity4 and melatonin content8 in
the chick retina.
Materials and Methods
Animals
Zero day fertile eggs and 12-day-old White Leghorn
Chicks {Gallus domesticus) were purchased from Spafas, Inc. (Norwich, CT). Eggs were kept in a warmed
(37 C), humidified incubator equipped with an automated lighting system; intensity of illumination at
the mid-portion of the incubator was 600 lux. Chicks
were communally housed in a temperature-controlled brooder that was quartered in a light-controlled room; intensity of illumination was 400 to 600
lux at the food and water bins. Blind chicks (see
From the Neuroendocrine Research Laboratory, Children's and
Neurology Services, Massachusetts General Hospital and Harvard
Medical School, Boston, Massachusetts.
Supported by PHS grant HD 14427 and a postdoctoral fellowship from the National Eye Institute. Steven M. Reppert is a research fellow of the Charles A. King Trust, Boston, Massachusetts.
Submitted for publication July 27, 1982.
Reprint requests: Steven M. Reppert, Children's Service, Massachusetts General Hospital, Boston, MA 02114.
0146-0404/83/0300/294/$ 1.15 © Association for Research in Vision and Ophthalmology
294
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
No. 3
RETINAL MELATONIN IN THE CHICK / Repperr ond Sogor
• Mid-light
• Mid-dark
10
Light at night
Dark during day
Pineal
1,
§
§ 10
1°
^ .3
s
Eye
reotaxic frame. All surgical procedures were performed at 14 to 15 days of age.
Pinealectomy was performed by the technique of
Binkley et al.5 For sham operations, the skin over the
head was incised, the skull etched with a trephine,
and the incision closed. Completeness of pinealectomy was corrfirmed on postmortem examination.
Blinding was accomplished by bilateral orbital enucleation. Following enucleation, the vacant orbit was
packed with gel foam and eyelids were sutured closed.
For optic nerve transection, each optic nerve was
visualized directly by medial retraction of the globe
through a ventrolateral incision of the lower eyelid.
The optic nerve was transected with an iris scissors.
The opposite eye, in which the optic nerve was visualized but not cut, served as a control. Optic nerve
transection was confirmed on postmortem examination.
Sample Collection
Plasma
.15
0
295
fa
m
Fig. 1. Photic regulation of pineal, ocular, and plasma melatonin.
At 19 days of age animals were killed at mid-light or mid-dark.
Animals were also killed at the subjective mid-dark time with lights
left on throughout the experimental dark period or at the subjective
mid-light time with lights left off throughout the experimental light
period. Eye (middle panel) refers to the posterior hemisphere of
one eye. Data are presented as mean ± SE of six to eight chicks
in each group.
"Surgical Procedures") were housed within a fenced
off portion of the communal brooder in order to protect them from sighted animals. Purina Chick Starteena and water were available at all times. Unless
otherwise specified, the daily light-dark cycle consisted of 12 hrs of light per day (LD 12:12) with lights
on from 0600 hrs EST. For experiments involving
hens, animals were reared by the animal husbandry
department of the University of Connecticut at Storrs
and housed there in the diurnal lighting cycle described above.
Surgical Procedures
Chicks were anesthetized with pentobarbital (40
mg/kg, IM), and the heads were secured in a rat ste-
At the end of an experiment, animals were killed
by decapitation; killing in darkness was accomplished
with the aid of a low intensity red light (Kodak safe
light filter no. 2). Trunk blood was collected in tubes
containing 100 units sodium heparin and centrifuged;
the resulting plasma was decanted and stored at
-20 C until analysis. The pineal gland was removed
rapidly, frozen on dry ice, and stored at -20 C. Eyes
were removed from the skull and placed on ice. The
globes were hemisected at the equator, and the vitreous was removed. For the experiments depicted in
Figures 1 and 2, the posterior hemisphere was frozen
on dry ice and stored at —20 C. For all subsequent
experiments, the retina plus attached pigment epithelium was dissected out and stored as above. We
found that the eye could remain at 0 C for up to 3
hrs prior to dissection without any significant change
in melatonin concentration.
Melatonin Radioimmunoassay (RIA)
Melatonin was measured using the assay procedure
of Rollag and Niswender22 (antiserum 1055) as modified by Reppert et al.20 At the time of analysis, tissues
were sonicated for 5 sec in the following volumes of
PBS (0.01 M phosphate buffer with 0.9% NaCl, pH
7.0): pineals, 120 /A; retina pigment epithelium, 1 ml;
and posterior pole, 2 ml. The pineal sonicates were
then diluted 1:40 in PBS. Melatonin from either a
200- or 400-jil portion of each sonicate was extracted
into 5 ml chloroform. The chloroform was washed,
and melatonin quantified as previously described.20
For measurement of melatonin in chick plasma, an
additional extraction step using petroleum ether was
used.20 The assay procedure for plasma melatonin
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
296
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / March 1983
Vol. 24
was validated by demonstrating (1) parallel inhibition
curves between serial dilutions of extracted nighttime
plasma samples and the melatonin standard, and (2)
quantitative recovery of 125 pg, 250 pg, and 500 pg
of melatonin from 1 -ml samples of pooled daytime
plasma. The extraction efficiency for all samples was
greater than 95%; values reported are not corrected
for the small loss during extraction. The limit of assay
sensitivity was 0.5 pg/tube. The within and between
assay coefficients of variation were 10 and 15%, respectively.
• Sham
M Pnx
Chemical Validation of Retinal Melatonin by HighPerformance Liquid Chromatography (HPLC)
0
Mid-light
Mid-dark
Fig. 2. Effects of pinealectomy (Pnx) on the day-night variation
of ocular melatonin. At 14 days of age, animals were divided into
two groups; one group was pinealectomized and the other received
sham operations. Five days later animals from each group were
killed at mid-light or mid-dark. At each time point, eight pinealectomized animals and four sham animals were used. Data are
presented as mean ± SE.
4r
Radioimmunoassay analysis of retina-pigment epithelium melatonin was corroborated by HPLC with
electrochemical detection using the method of Mefford and Barchas1' with modifications (Fig. 3). Pooled
chloroform extracts of ocular tissue were dried under
vacuum in darkness and redissolved in 0.1 ml HC1.
After centrifugation, aliquots of the supernatant were
injected into a 0.4 X 30 cm micron Bondapak C|8
reversed phase column (Waters Assoc, Milford, MA)
eluted with a mobile phase consisting of 0.1 M sodium acetate buffer (pH 4.7), 0.1 mM EDTA, and
25% methanol at a flow rate of 1.5 ml/min. A LC-4
electrochemical detector (Bioanalytical Systems, West
Lafayette, IN) with a glassy carbon electrode was used
at a potential of +0.90 V. Peaks were identified by
retention time and melatonin quantified by peak
height.
Experimental Protocols
Two time points (one at mid-light and the other
at mid-dark) were used to monitor daily melatonin
rhythms in the eye, pineal, and plasma. These time
points correspond to the nadir and peak, respectively,
of the daily melatonin rhythms in all three tissues in
the chicken, as studied in diurnal lighting.2'814 The
term ocular melatonin is used in the text to signify
melatonin content in the posterior hemisphere of the
eye. Other experimental details appear in the "Results" section and the figure legends.
1
MEL
Statistical Methods
0
L
20
iO
20
tO
0
20
10
Statistical analyses were performed using the MannWhitney U/Wilcoxon nonparametric test.
TIME (mm)
Fig. 3. High performance liquid chromatograms of (a) 10 pmoles
each of 5-hydroxytryptophan (5-HT), N-acetyl serotonin (NAS)
and melatonin (MEL); b, pooled extracts of retina-pigment epithelium of chicks killed at mid-dark; and c, pooled extracts of
retina-pigment epithelium of chicks killed at mid-light. Melatonin
concentrations calculated from the chromatograms were 1.3 ng/
retina at mid-light and 3.9 ng/retina at mid-dark. The same extracts
analyzed by RIA gave values of 1.2 and 5.0 ng/retina, respectively.
Results
Photic Regulation of Pineal,
Ocular, and Plasma Melatonin
Day-night variations of melatonin concentrations
in the pineal gland, eye, and plasma were examined
in chicks exposed to diurnal lighting and killed at
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
No. 3
297
RETINAL MELATONIN IN THE CHICK / Repperr and Sogor
either mid-light or mid-dark. In this lighting cycle,
there are prominent, synchronous day-night oscillations in pineal, ocular, and plasma melatonin, with
high values at night (Fig. 1); the magnitudes of the
nocturnal excursions are 17-fold in the pineal, fivefold in the eye, and tenfold in plasma.
Light suppression of nocturnal melatonin production was examined in the chick by killing animals at
the mid-dark time with lights left on throughout the
experimental dark period. This manipulation suppresses plasma and ocular melatonin levels such that
they become similar to normal mid-light values (Fig.
1). The treatment also lowers pineal melatonin levels,
but not to normal midday values.
To determine whether melatonin levels can decrease without a light cue, animals were killed at the
mid-light time with lights left off throughout the experimental light period. This results in low ocular and
plasma melatonin levels like those normally found
during the light portion of the day (Fig. 1). Pineal
melatonin content is also reduced, compared to middark levels, in this environment, but the resulting
value is higher than that usually found at mid-light.
•
Sham H Pnx
7/\ Pnx and Blinded
I
i
s
.2
Effects of Pinealectomy on Ocular
and Plasma Melatonin Levels
Pinealectomy does not alter the day-night variation
in ocular melatonin content (Fig. 2). In contrast, pineal removal has a marked effect on the normally
high mid-dark plasma melatonin concentration, lowering it by 80% (Fig. 4). Pinealectomy does not, however, alter the normally low mid-light plasma melatonin level. Interestingly, a small but significant (P
< 0.01) day-night variation in plasma melatonin levels, with the high value at night, is still evident in
pinealectomized chicks.
Effects of Pinealectomy plus
Blinding on Plasma Melatonin
To examine the contribution of the retina to circulating melatonin levels, blind-pinealectomized animals were killed at either mid-light or mid-dark, and
their circulating melatonin concentrations compared
to those of pinealectomized animals. Beyond the reduction incurred by pinealectomy alone (see preceding section), blinding causes a further 9% reduction
of the mid-dark plasma melatonin value (Fig. 4). This
procedure does not alter the mean mid-light plasma
melatonin concentration. Blind-pinealectomized
chicks do not appear to exhibit a day-night variation
in plasma melatonin levels.
Effects of Optic Nerve Transection on RetinaPigment Epithelium Melatonin
The role of optic nerve afferents to the retina in
the generation of the day-night variation in retina-
0
Mid-light
Mid-dark
Fig. 4. Effects of pinealectomy (Pnx) and blinding plus pinealectomy on the day-night variation in plasma melatonin. At 14
days of age, groups of animals received one of three operations:
pinealixtomy, pinealectomy plus blinding, or sham pinealectomy.
Five days later, animals from each group were killed at mid-light
or mid-dark. The data depicted represent a computation of results
from three separate experiments. Data are presented as mean ± SE,
and the number of animals is depicted at the base of each bar.
*P < 0.01; tP < 0.001.
pigment epithelium was also examined. As shown in
Figure 5, optic nerve transection does not affect the
eye rhythm, as studied under diurnal lighting conditions.
Ontogeny of the Day-Night Variation in RetinaPigment Epithelium Melatonin
Developmentally, a significant (P < 0.01) day-night
variation in retina-pigment epithelium melatonin
content, with high nighttime levels, is first detected
in the late embryo, 2 days prior to hatching (Fig. 6).
The increase in the magnitude of the variation during
maturation results primarily from an increase in the
mid-dark value. A similar developmental pattern was
found for pineal melatonin content (data not shown).
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
298
Vol. 24
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1983
Optic Nerve Transection
pineal gland of each animal, we have shown that the
day-night variations in both organs are synchronous
and regulated by environmental lighting in a similar
manner; high nighttime levels are suppressed by light
whereas the rhythms persist in darkness. We also dissected retina from pigment epithelium and found, as
others have,15'8 that in the chick ocular melatonin and
HIOMT are localized to the retina (Sagar and Reppert, unpublished data). The results of our pinealectomy experiment show that the pineal gland is not
the source of retinal melatonin since ocular melatonin levels continue to oscillate in animals lacking
pineals. Taken together, the findings to date indicate
that the chick retina rhythmically synthesizes melatonin and that retinal melatonin content displays a
photically controlled circadian rhythm in phase with,
but independent of, the pineal rhythm.
The results of our developmental study suggest that
melatonin is produced rhythmically by the retina as
early as the latter stages of embryonic development.
This coincides with the time that day-night variations
^S
<b
L
Mid-light
Mid-dark
Fig. 5. Effects of optic nerve transection on the day-night variation of retina-pigment epithelium melatonin. Optic nerve transection (right eye) and sham operation (left eye) were performed
in each animal at 14 days of age. Five days later animals were
killed at mid-light or mid-dark. Data are the mean ± SE of six to
eight animals in each group.
Examination of retina-pigment epithelium melatonin
content in hens showed that the day-night variation
is prominently manifested in the adult animal.
Discussion
4
After Binkley et al discovered a prominent daily
rhythm in NAT activity in the chick retina, Hamm
and Menaker8 showed that both NAT activity and
melatonin content exhibit synchronous daily retinal
rhythms in this species and respond similarly to manipulations of the daily light-dark cycle. They also
showed that the retinal NAT rhythm persists after
pinealectomy. The results presented here confirm and
extend those observations.
By monitoring melatonin concentrations at two
time points in the posterior pole of the eye and in the
I
"7 (
0
0
•
•
i
i
Mid-light
Mid-dark
i
i
-
I l i
y
I
/
2 -
/
/
/
A
s
I
/
f
/
/
/
•o
0
-5
Birth
5
15
20 Adult
Age (d)
Fig. 6. Developmental pattern of the day-night variation in retina-pigment epithelium melatonin. At 16 and 19 days of incubation, embryos were removed from their eggs and killed at mid-light
or mid-dark. On the day of birth and at 2 and 19 days of age chicks
were killed at mid-light or mid-dark. The adult animals were hens
housed in diurnal lighting and killed at mid-light or mid-dark. Data
are the mean ± SE of five to six animals in each group. *P < 0.01.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
No. 3
RETINAL MELATONIN IN THE CHICK / Repperr and Sagar
in pineal NAT activity3 and in pineal melatonin content are first detectable in the chick embryo. Thus,
it seems that the rhythmic production of melatonin
by the retina and pineal gland exhibit similar developmental patterns.
An interesting question arises as to the site of the
endogenous oscillator that generates the retinal
rhythm. It does not reside in the pineal gland since
the retinal rhythm persists after pinealectomy. Another possibility is that the rhythm might be generated
by neural input from the brain. One way that neural
information could reach the retina is via optic nerve
afferents, which are prevalent in avian species.21 This
does not appear to be the case, however, since optic
nerve transection does not affect the eye rhythm in
diurnal lighting. Other possible neural inputs to the
retina that could be considered in this regard, such
as sympathetic fibers, require further evaluation. It
is worth noting that Binkley and coworkers6 have
shown by using patch experiments and monitoring
retinal NAT activity that within an individual chick
the two retinae can respond to light independently
of each other, suggesting that each eye contains an
endogenous mechanism for rhythm generation.
A major focus of our study was to examine the
contribution of retinal melatonin to circulating concentrations of the substance. Such a contribution
seemed possible since melatonin readily passes
through cell membranes and many biologic barriers,
including the blood-brain barrier and the placenta18;
this difFusibility is due to the nonpolar, lipophilic
nature of this small molecule (mol. wt. 232). Also,
based on studies with the rat pineal gland, melatonin
is not stored in secretory vesicles, but instead diffuses
rapidly into blood upon production.1019
We examined the contribution of retinal melatonin
to blood levels by first showing that our RIA system
detects a large day-night variation in plasma melatonin concentrations in the intact chick (Fig. 1). The
variation that we found (mean mid-day value of 31
pg/ml; mean mid-dark value of 322 pg/ml) is in good
agreement with those reported by three different laboratories (range of mid-light values 10 to 50 pg/ml;
range of mid-dark values 210 to 350 pg/ml) where
bioassay17 and two other melatonin RIA systems were
used.912 We next showed that the pineal gland is the
major source of circulating melatonin concentrations
at night, since the increase at mid-dark is nearly
abolished in pinealectomized animals; this again is
consistent with an earlier report.14 We also found that
retinal melatonin contributes at best a small, albeit
statistically significant, amount of melatonin to the
mid-dark circulating levels. Furthermore, animals
lacking pineals exhibit a detectable day-night variation in blood melatonin, with the higher levels at
night. Whether this small variation actually repre-
299
sents a daily rhythm of retinal origin or is of physiologic significance is not possible to determine from
the two time points examined in our experiment.
Based on the biophysical properties of melatonin
discussed above, the apparently small retinal contribution to circulating levels of the hormone was unexpected. If one estimates from Figure 1 the amount
of melatonin produced by each eye, assuming that
melatonin levels reflect production rates in the retina,
as suggested by the parallel rises and falls of NAT
activity and melatonin content in the retina,8 then
we would have expected both retinae together to be
responsible for roughly 50% of the high nighttime
circulating melatonin concentrations. That this is not
the case is difficult to explain. A possible explanation
is that retinal melatonin might be metabolized locally
within the eye.
Since our results suggest that the contribution of
retinal melatonin to blood is quite small, the most
likely function of retinally derived melatonin is in the
regulation of rhythmic phenomena within the eye
itself. There are three processes in the retina and pigment epithelium of various species known to display
circadian rhythms: pigment migration, the elongation
and contraction of the inner segment of photoreceptors (retinomotor movements), and rod outer segment disk shedding.2124 There is some evidence in
nonavian species that exogenously administered melatonin can influence each of these processes. Pigment
migration in mammals13 and retinomotor movements in amphibians16 can be influenced by melatonin. Moreover, rod outer segment shedding in rats
can be modulated by systemically administered melatonin.23 The role of retinal melatonin in the regulation of these mechanisms in the chick deserves systematic study. Furthermore, the hypothesis that in
other species including mammals, the retina may be
a target organ for melatonin should be tested.
Key words: retina, pineal gland, melatonin, circadian
rhythms, chicks
References
1. Binkley S: Pineal gland biorhythms: N-acetyltransferase in
chickens and rats. Fed Proc 35:2347, 1976.
2. Binkley S, MacBride SE, Klein DC, and Ralph CL: Pineal
enzymes: Regulation of avian melatonin synthesis. Science
181:273, 1973.
3. Binkley S and Geller EB: Pineal enzymes in chickens: Development of daily rhythmicity. Gen Comp Endocrinol 27:424,
1975.
4. Binkley S, Hryshchyshyn M, and Reilly K: N-acetyltransferase
activity responds to environmental lighting in the eye as well
as in the pineal gland. Nature 281:479, 1979.
5. Binkley S, Kluth E, and Menaker M: Pineal function in sparrows: circadian rhythms and body temperature. Science
174:311, 1971.
6. Binkley S, Reilly K, and Hernandez T: N-acetyltransferase in
the chick retina. II. Interactions of the eyes and pineal gland
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017
300
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1983
in response to light. J Comp Physiol 140:181, 1980.
7. Cardinali DP and Rosner JM: Retinal localization of the hydroxyindole-O-methyl transferase (HIOMT) in the rat. Endocrinology 89:301, 1971.
8. Hamm HE and Menaker M: Retinal rhythms in chicks: Circadian variation in melatonin and serotonin N-acetyltransferase activity. Proc Natl Acad Sci USA 77:4948, 1980.
9. Kennaway DJ, Frith RG, Phillipou G, Matthews CD, and
Seamark RF: A specific radioimmunoassay for melatonin in
biological tissue and fluids and its validation by gas chromatography-mass spectrometry. Endocrinology 101:119, 1977.
10. Klein DC: The pineal gland: A model of neuroendocrine regulation. Res Publ Assoc Res Nerv Ment Dis 56:303, 1978.
11. Mefford IN and Barchas JD: Determination of tryptophan and
metabolites in rat brain and pineal tissue by reversed-phase
high-performance liquid chromatography with electrochemical detection. J Chromatogr 181:187, 1980.
12. Pang SF, Brown GM, Grota LJ, Chambers JW, and Rodman
RL: Determination of N-acetylserotonin and melatonin activities in the pineal gland, retina, harderian gland, brain and
serum of rats and chickens. Neuroendocrinology 23:1, 1977.
13. Pang SF and Yew DT: Pigment aggregation by melatonin in
the retinal pigment epithelium and choroid of guinea-pigs,
Cavia porcellus. Experientia 35:231, 1979.
14. Pelham RW: A serum melatonin rhythm in chickens and its
abolition by pinealectomy. Endocrinology 96:543, 1975.
Vol. 24
15. Quay WB: Retinal and pineal hydroxyindole-O-methyl transferase activity in vertebrates. Life Sci 4:983, 1965.
16. Quay WB and McLoed RW: Melatonin and photic stimulation
of cone contraction in the retina of larval Xenopus laevis. Anat
Rec 160:491, 1968.
17. Ralph CL, Pelham RW, MacBride SE, and Reilly DP: Persistent rhythms of pineal and serum melatonin in cockerels in
continuous darkness. J Endocrinol 63:319, 1974.
18. Reppert SM, Chez RA, Anderson A, and Klein DC: Maternalfetal transfer of melatonin in the non-human primate. Pediatr
Res 13:788, 1979.
19. Reppert SM and Klein DC: Mammalian pineal gland: basic
and clinical aspects. In The Endocrine Functions of the Brain.
M Motta, editor. New York, Raven Press, 1980, p. 327.
20. Reppert SM, Perlow MJ, Tamarkin L, and Klein DC: A diurnal melatonin rhythm in primate cerebrospinal fluid. Endocrinology 104:295, 1979.
21. Rodieck RW: The Vertebrate Retina; Principles of Structure
and Function. San Francisco, W. H. Freeman and Co, 1973.
22. Rollag MD and Niswender GD: Radioimmunoassay of serum
concentrations of melatonin in sheep exposed to different lighting regimens. Endocrinology 98:482, 1976.
23. White MP and Fisher LJ: Melatonin effects on circadian rod
outer segment shedding. Neuroscience Abstracts 6:344, 1980.
24. Young RW: The daily rhythm of shedding and degradation
of rod and cone outer segment membranes in the chick retina.
Invest Ophthalmol Vis Sci 17:105, 1978.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933340/ on 05/04/2017