Journal of Neuroendocrinology, 2003, Vol.

15, 390–397

Clock Genes and the Long-Term Regulation of Prolactin

Secretion: Evidence for a Photoperiod/Circannual
Timer in the Pars Tuberalis

G. A. Lincoln,
H. Andersson
and D. Hazleriggy

MRC Human Reproductive Sciences Unit, Centre for Reproductive Biology, Chancellor’s Building, Edinburgh, UK.
yACERO, Department of Agriculture and Forestry, University of Aberdeen, Aberdeen, UK.

Key words: clock genes, melatonin, pituitary gland, season.

Abstract
Prolactin secretion is regulated by photoperiod through changes in the 24-h melatonin profile and displays
circannual rhythmicity under constant photoperiod. These two processes appear to occur principally within
the pituitary gland, controlled by the pars tuberalis. This is evident because: (i) hypothalamic-pituitary
disconnected (HPD) sheep show marked changes in prolactin secretion in response to switches in
photoperiod and manipulations of melatonin, similar to brain-intact controls; (ii) HPD sheep also show
photoperiod-specific, long-term cycles in prolactin secretion under constant long or short days, with the
timing maintained even when prolactin secretion is blocked for 2–3 months; and (iii) pars tuberalis cells, but
not lactotrophs, express high concentrations of melatonin (MT1) receptor, and exhibit a duration-sensitive,
cAMP-dependant, inhibitory response to physiological concentrations of melatonin. This suggests the
existence of an intrinsic, reversible photoperiod-circannual timer in pars tuberalis cells. A full complement of
clock genes (Bmal1, Clock, Per1, Per2, Cry1 and Cry2) are expressed in the ovine pars tuberalis, and undergo
24-h cyclical expression as observed in a cell autonomous, circadian clock. Activation of Per genes occurs in
the early day (melatonin off-set), while activation of Cry genes occurs in the early night (melatonin on-set).
This temporal association is evident under both long and short days, thus the Per–Cry interval varies directly
with photoperiod. Because, PER : CRY, protein : protein interactions affect stability, nuclear entry and gene
transcription based on rodent data, the change in phasing of Per/Cry expression provides a potential
mechanism for decoding the long day/short day melatonin signal. A speculative, but testable, extension of
this hypothesis is that intrinsically regulated changes in the phase of Per/Cry rhythms, regulates both
photorefractoriness and the generation of circannual rhythms in prolactin secretion.

constant environmental conditions, and the annual change in day-

Seasonal prolactin secretion
Prolactin secretion is increased in summer and decreased in winter
in a wide range of mammals adapted to temperate and cold cli-
mates. This includes domesticated and wild ungulates, mustelids,
canids, rodents, primates and marsupials where the high prolactin
secretion in summer is implicated in the seasonal regulation of
growth, food intake, gonadal activity, paternal behaviour, lactation
and/or the timing of embryonic diapause (1, 2). One of the best
characterized roles of seasonal prolactin is in the control of the
pelage growth/moult cycle where prolactin acts through prolactin
receptors expressed in dermal papilla of hair follicles to reactivate
fibre growth in spring and to determine the phenotype of the hair
produced in summer compared to winter (3, 4). In long-lived
species, such as sheep and deer, the long-term prolactin cycle is
generated endogenously and displays circannual rhythmicity under
length (photoperiod) is utilized as the predictive cue to time the cycle
to the natural environment (5). In short-lived species, such as small
rodents, photoperiod and temperature more acutely modulate pro-
lactin secretion. Across all mammalian groups, long photoperiods
stimulate, and short photoperiods inhibit prolactin secretion, sug-
gesting that the physiological control mechanisms that govern sea-
sonal prolactin secretion have been conserved during evolution (6).
The suprachiasmatic nuclei (SCN)–melatonin–pars
tuberalis relay
Prolactin secretion is regulated by photoperiod through changes in
the duration of nocturnal melatonin secretion by the pineal gland.
In this sensory/neuroendocrine relay, light acts through specia-
lized, nonvisual photoreceptors in the retina to dictate melatonin

Correspondence to: Dr Gerald A. Lincoln, MRC Human Reproductive Sciences Unit, Centre for Reproductive Biology, Chancellor’s Building, 49 Little France
Crescent, Edinburgh EH16 4SB, UK (e-mail: g.lincoln@hrsu.mrc.ac.uk).

# 2003 Blackwell Publishing Ltd


release through two distinct mechanisms (7, 8). First, periodic
exposure to light every 24-h entrains the circadian rhythms gene-
rated by the central pacemaker neurones located in the SCN of the
anterior hypothalamus. The SCN governs the timing of diurnal
rhythms in locomotor activity, feeding behaviour, body tempera-
ture and pituitary hormone secretion, as well as the nightly release
of melatonin. Second, light acutely inhibits melatonin production
by the pineal gland via the retinal-hypothalamic-sympathetic
innervation and the control of the rate-limiting enzyme N-acet-
yletransferase. These control mechanisms dictate that melatonin is
secreted only at night, and the duration of release varies quantita-
tively with nightlength, therefore daylength. The 24-h melatonin
rhythm thus provides an endocrine index of time of year.
Decoding the melatonin signal occurs in the brain and pituitary
gland through cells that express melatonin receptors, and different
target tissues are thought to control the different seasonal rhythms
(6, 9). The pars tuberalis of the stalk of the pituitary gland is spe-
cifically implicated in photoperiodic regulation of prolactin secre-
tion (10) and the premammillary region of the hypothalamus in the
control of gonadotrophin secretion and gonadal activity (11). The
pars tuberalis expresses melatonin (MT1) receptors at high density
compared with sites in the brain, and the tissue is strategically
placed at the interface between median eminence and pars distalis,
to regulate neuroendocrine functions. Pars tuberalis cells secrete
prolactin-releasing factors (tuberalins), and the currently accepted
view is that melatonin dictates the secretory function of the pars
tuberalis, and the paracrine products relay the effects to the
lactotroph to produce the photoperiodic response (12, 13).
The strongest evidence for a pivotal role of the pars tuberalis is
provided by studies in hypothalamic-pituitary disconnected Soay
rams (14–16). In this model, hypothalamic-pituitary disconnected
(HPD) surgery destroys the arcuate nucleus and median eminence
of the hypothalamus (17), but leaves a viable isolated pituitary
gland with pars tuberalis tissue still expressing melatonin receptors
(18). HPD rams show little or no prolactin-response to treatment
with dopamine receptor and noradrenaline receptor antagonists,
confirming the loss of the hypothalamic inhibitory pathways that
normally provide the homeostatic control of prolactin in the intact
animal (19). Despite the hypothalamic lesion, HPD rams continue
to express normal long-term patterns of prolactin secretion in
response to changes in photoperiod and the chronic administration
of melatonin (14). The timing and amplitude of the photoperidic
responses, are very similar to those expressed by sham-operated
controls, implying that the melatonin signal acts largely, or exclu-
sively, at the level of the pituitary gland to control the seasonal
prolatin cycle. Moreover, observed photoperiod-related changes
in content and turnover of dopamine in the ovine hypothalamus/
median eminence in intact animals (20, 21) may be the consequ-
ence rather than the cause of the changes in prolactin secretion due
to the homeostatic, positive-feedback response to prolactin (19,
22). With only pars tuberalis cells in the ovine pituitary expressing
melatonin receptors, the inference is that all long-term timed
changes in prolactin release observed in the HPD sheep, reflect
the function of the melatonin-target cells in the pars tuberalis.
Photorefractoriness and circannual rhythm generation
in HPD sheep
Figure 1 presents the long-term changes in prolactin secretion
recorded in a group of HPD Soay rams to illustrate the complex
# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397
Photoperiod and circannual timer 391
control that still exists in these animals under alternating photo-
period (Fig. 1A), or a prolonged constant photoperiod (Fig. 1B). In
this study, 12 rams received the standardized HPD operation under
short days when blood prolactin concentrations were low. The
surgery induced a short period of increased prolactin secretion, but
the suppressive effect of the short-day photoperiod persisted
(Fig. 1A) (weeks 0–16). The HPD rams were then exposed to
alternating 16-weekly periods of long days (16 : 8 h light/dark) and
short days (8 : 16 h light/dark) for three cycles, which effectively
entrained the seasonal prolactin cycle to the artificial 32-week
periodicity, as seen previously (14, 16), but with one notable
exception (Fig. 1A, S1). This HPD ram failed to adjust to the
changes in photoperiod and continued to express a long-term
prolactin cycle with a periodicity close to 48 weeks. This is the
first observed case of an HPD ram expressing a circannual rhythm
in prolactin release. The ram was overtly similar to the others in
the group, and showed all the clinical signs of complete pituitary
disconnection including polyurea, increasing obesity and perma-
nent regression of the testes.
At 8–10 weeks into long and short days, blood samples were
collected from the HPD sheep to characterize the diurnal pat-
terns of melatonin secretion (Fig. 2A). As predicted, the blood
melatonin concentrations were increased during the dark phase
under both photoperiods, with the duration of the nocturnal
melatonin peak markedly extended under short days. However,
the ram that failed to entrain its prolactin cycle (S1), showed a
delayed onset of melatonin secretion under short days, and the
duration of the melatonin peak was unaffected by photo-
period. For all rams, there was an inverse relationship between
melatonin duration and the 24-h mean blood prolactin concen-
tration, with the outlier ram exhibiting a long-day response
irrespective of photoperiod (Fig. 2B). These data confirm that
HPD sheep can generate a normal photoperiod-controlled 24-h
blood melatonin profile, and that it is a change in melatonin
duration that potentially determines the level of prolactin secre-
tion.
The HPD sheep used in this illustration were also exposed to a
change from short days to prolonged constant long days for
48 weeks (Fig. 1B). This induced a synchronized sequence of
activation, decline and reactivation in prolactin secretion in all
animals, with the exception of the unresponsive ram that still
expressed an unentrained circannual rhythm. For the majority, the
occurrence of an intrinsic prolactin cycle with consistent timing is
similar to that described previously in HPD rams (23). The mela-
tonin rhythm continues to reflect the photoperiod under prolonged
long days, thus the relationship between melatonin duration and
the level of prolactin secretion must change spontaneously, as the
animals become photorefractory, and produce long-term cycles in
prolactin secretion. The notable animal, in which the melatonin
profile did not change with photoperiod, was essentially respond-
ing as if on constant long days, and this permitted the expression of
the circannual pattern.
To investigate whether that the intrinsic cyclical changes in
prolactin secretion might be controlled at the level of the mela-
tonin target cell in the pars tuberalis, or at the level of the lacto-
troph itself, photorefractory HPD sheep were treated chronically
with the dopamine agonist, bromocriptine (24). This treatment
blocked the release of prolactin, but once terminated, the timing of
the long-term prolactin cycle was parallel to that of control HPD
animals living under constant long days. Thus, it seems more
392 Photoperiod and circannual timer

FIG. 1. Individual long-term prolactin profiles in three selected hypothalamic-pituitary disconnected (HPD) Soay rams exposed to: (A) alternating 16-weekly
periods of short days (SD, 8 : 16 h light/dark) and long days (LD, 16 : 8 h light/dark), and (B) a switch from short days (SD) to prolonged constant long days (LD)
for 48 weeks, and then back to short days (SD). The HPD operations were performed at 4 weeks into short days (arrow). This experiment included 12 HPD rams,
S2 (top) is representative of 9/12 animals (rapidly entrained), S4 (middle) is representative of 2/12 animals (more slowly entrained) and S1 (bottom) was the only
animal that remained unentrained and continued to express a circannual rhythm in blood prolactin concentrations.

likely that the pars tuberalis cell, rather than the lactotroph, is the
generator of the long rhythmicity.
Overall, the HPD sheep model provides compelling evidence
that the photoperiodic control of long-term cycles in prolactin
secretion does not depend on the direct hypothalamic control of
the lactotroph through the classical dopaminergic pathway, but is
mediated through the melatonin signalling via the pars tuberalis.
The ability of HPD animals to express an acute (inductive)
response over weeks following a switch in photoperiod, to express
a chronic (refractory) response over months under constant photo-
period, and even to express a circannual cycle in prolactin secre-
tion with a period >40 months, suggests that the pars tuberalis
functions as a photoperiod and circannual timer.
Decoding the melatonin signal in the pars tuberalis
Since pars tuberalis cells can be studied in vivo, and are readily
manipulated in vitro, they provide a model system to investigate
the molecular mechanisms whereby the melatonin signal is
decoded to produce a biological response (25). To date, the best
characterized cellular response to melatonin in pars tuberalis cells
is the suppression of intracellular cAMP production due to inhibi-
tion of adenylyl cyclase, and the associated block of a range of
cAMP-dependant processes including activation of protein kinase
A, phosphorylation of the cAMP-response element binding pro-
tein and the induction of the transcription factor, c-fos (26, 27).
The suppressive response occurs at picomolar concentrations,
which is considered to be physiological, and appears to be medi-
ated primarily through the melatonin 1a (MT1) receptor, coupled
to the Gi class of G-proteins (10, 28). These intracellular responses
occur within minutes; however, prolonged exposure to melatonin
over hours causes a duration-dependant change in the sensitivity
of the signalling mechanisms. This is seen as a marked increase in
the accumulation of cAMP in ovine pars tuberalis cells stimulated
with forskolin, and a partial loss of the inhibitory response to a
second exposure to melatonin (29). Because this time-dependent
sensitivity could result in long-term up-regulation, or down-reg-
ulation of gene transription in the target cell according to the

# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397

 Photoperiod and circannual timer 393

FIG. 2. (A) 24-h melatonin profiles in hypothalamic-pituitary disconnected (HPD) Soay rams sampled every 15 min for 24 h under short days (8 : 16 h light/dark;
week 72, Fig. 1) and long days (16 : 8 h light/dark; week 88, Fig. 1) The values are mean  SEM for group (*, n ¼ 11), and individual values for the unentrained
outlier animal (*, S1, Fig. 1). Zeitgeber time (ZT) 0 was time of ‘lights on’ and the closed bar depicts the period of darkness. (B) Prolactin response summarizing
the inverse correlation between the duration of the nocturnal melatonin peak and the 24-h mean blood plasma concentrations of prolactin in the HPD rams
sampled under short days (SD) and long days (LD). The correlation coefficient was calculated based on the individual values for the main group (*, n ¼ 11). The
values for the outlier animal (&, S1) are also shown.

melatonin duration in vivo, the cAMP sensitization/desensitization
response is thought to be part of the decoding mechanism (25).
More specific information is provided by in situ hybridization
studies in the sheep, Syrian and Siberian hamsters where the 24-h
rhythmic expression of the acutely inducible genes, Period 1
(Per1) and inducible cAMP early repressor (ICER) was measured
in the pars tuberalis (30, 31). For animals maintained on a 24-h
light/dark cycle, peak expression for both Per1 and ICER occurred
early in the light phase [zeitgeber time (ZT) 3–7; ZT 0 ¼ lights
on]. Moreover, the level of maximal expression was affected by
photoperiod, with a higher amplitude peak under long days
compared to short days (32). Other studies involving timed
injection of melatonin and pinealectomy have shown that the
cycle in Per1 and ICER mRNA expression in the pars tuberalis is
intimately regulated by the melatonin profile, and support the view
that the daytime peak in gene expression results from disinhibition
by melatonin, acting through cAMP signalling (32, 33). Besides
the inhibitory effect of the nocturnal signal, melatonin has been
shown to sensitize the pars tuberalis cell to the stimulatory effect
of adenosine, which induces cAMP-associated responses during
the light phase (34). Taken together, the results support a hypoth-
esis that melatonin duration is decoded into a change in amplitude
of expression of a range of acutely inducible genes, that in turn
regulate the expression of a cascade of other genes, to control the
secretory function of the pars tuberalis cell (31). A short daily
melatonin signal (long days), up-regulates transcription to pro-
duce an active summer pars tuberalis cell and a summer pheno-
type, while a long melatonin signal (short days) produces the
inactive winter phenotype.
Clock gene expression in the ovine pars tuberalis
The mammalian clock genes that generate circadian rhythms in
the SCN are now well characterized (35–37). Because these genes
are essential for time keeping in the central pacemaker, it might
be predicted that they also provide the clockwork in the pars
tuberalis, forming part of the mechanism for decoding the photo-
period-regulated melatonion signal. Most recently, the 24-h pat-
tern of expression of the major canonical clock genes has been
measured in the pars tuberalis in sheep entrained to short and long
days (38). The results for Per1 and Cry1 are briefly summarized in
Fig. 3.
All the selected clock genes (Clock, Bmal1, Per1, Per2, Cry 1
and Cry2) were rhythmically expressed in the ovine pars tuberalis.

# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397

394 Photoperiod and circannual timer

FIG. 3. Double-plotted, 24-h profiles of Per1 (*) and Cry1 (*) mRNA expression in the pars tuberalis (PT) and blood plasma melatonin concentrations in Soay
rams sampled every 4 h for 24 h (n ¼ 4/time point; mean  SEM) under (A) short days (8 : 16 h light/dark) and (B) long days (16 : 8 h light/dark). Zeitgeber time
(ZT) 0 was time of ‘lights on’ and the closed bar depicts the period of darkness. The interval (c) between the peak in Per and Cry expression is shown (horizontal
line) for the two photoperiods. (C) Representative autoradiograms of Per1 and Cry1 expression in the PT at ZT 3 and ZT 15 under short days to illustrate the
contrast in time of expression for these two clock genes. Per1 is maximally expressed in the PT at ZT 3, and also rhythmically expressed in periventricular
(PeVN) hypothalamus adjacent to the third cerebral ventricle (3V) (38); Cry 1 is maximally expressed in the PT at ZT 15 with no specific expression in the PeVN.

The notable feature was that, unlike the situation in the SCN, half of the light phase (ZT 3–7; ZT 0 ¼ lights on), and the peak
photoperiod differentially affected the timing of expression of the in expression of Cry 1 and Cry2 occurred in the early dark phase
Period and Cryptochrome genes in the pars tuberalis. Specifically, (ZT 11–15 under short days and ZT 19 under long days). This
the 24-h peak in expression of Per1 and Per2 occurred in the first produced a change in the relative phasing of expression for the two

# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397

 Photoperiod and circannual timer 395
sets of genes, with the interval between the Per and Cry maxima
(termed psi, c) consistently shorter under short days compared to
long days (Fig. 3A,B). These patterns of gene expression were very
closely related to the diurnal rhythm in blood concentrations of
melatonin, with the activation of Per1 closely linked to the off-set
of melatonin secretion. The earlier and more gradual decline in
melatonin under short days was associated with an earlier and
lower amplitude increase in Per1 expression, compared with the
more precipitous changes under long days. This is consistent with
a disinhibition mechanism by which melatonin controls Per1
expression in the ovine PT (30). In contrast, the activation of
Cry1 and Cry 2 was closely linked to the on-set of melatonin
secretion, suggesting a positive effect of melatonin on the expres-
sion of the cryptochrome genes.
These data therefore support an alternative, internal coinci-
dence model for the molecular decoding of the melatonin signal.
The photoperiod-induced change in melatonin duration is trans-
lated into a corresponding change in phasing of the 24-h rhythm in
expression of the Period and Cryptochrome genes. Changes in
phasing may be critically important, because the stability of newly
synthesized PER and CRY protein in the cytoplasm is likely to
depend on the formation of heterodymic PER/CRY protein : pro-
tein complexes based on the situation in the rodent SCN (39, 40).
This determines the entry of the clock proteins into the cell
nucleus, where they acts to modulate transcription of target genes.
In the pars tuberalis, the presumption is that the altered coin-
cidence between Per and Cry expression, affects the maximum
concentration of the PER/CRY heterodimers and the duration
when they are present during the 24-h, which affects transcription
of the genes that control the timing function of the pars tuberalis. FIG. 4. Model for the decoding of the melatonin signal in a melatonin target
This is likely if there is a fixed time relationship between mRNA cell (pars tuberalis). This proposes that the decoding mechanism depends on
the differential regulation of the circadian rhythm in expression of two sets of
expression for a specific clock gene as revealed by in situ clock genes by the melatonin signal. The ‘dawn’ component of the signal
hybridization, and the translation into protein. Future studies in (melatonin decrease) controls the 24-h Per1/Per2 rhythm, phasing the peak of
sheep will need to measure the 24-h rhythm in clock proteins in the expression to the early day (A), while the ‘dusk’ component (melatonin
pars tuberalis by immunocytochemistry, and investigate the pre- increase) controls, via a second transcriptional relay, the timing of the 24-h
Cry1/Cry2 rhythm, phasing the peak of expression to the early night (B). The
dicted formation of heterodimers and the change from cytoplas- Per–Cry interval (c) thus varies directly with the length of the photoperiod,
mic to nuclear localization. and determines the period when the protein products of these genes (CRYs
and PERs) are coincident in the cytoplasm. The level of coincidence in turn
determines the formation of PER/CRY protein complexes that potentially
Generalized model for decoding the melatonin signal control, the transcription of other clock genes as part of the cellular
clockwork (C), and the transcription of downstream genes (D) that dictate the
Our preferred model for a melatonin decoding mechanism is secretory activity of the melatonin target cell. For the pars tuberalis cell, the
presented in Fig. 4. This proposes that: (i) The decoding mechan- global transcriptional control is up-regulated under long days, producing
ism depends on the differential regulation of the circadian rhythm increased synthesis of a prolactin-releasing factor that drives the lactotrophs
(summer phenotype), and down-regulated under short days (winter
in expression of two sets of clock genes by the melatonin signal. phenotype).
(ii) The ‘dawn’ component of the signal (melatonin decrease)
controls the 24-h Per1/Per2 rhythm, phasing the peak of expres-
sion to the early day (Fig. 4A), while the ‘dusk’ component
(melatonin increase) controls, via a different transcriptional relay, cell, we propose that the global transcriptional control is up-
the timing of the 24-h Cry1/Cry2 rhythm, phasing the peak of regulated under long days, producing increased synthesis of a
expression to the early night (Fig. 4B). This is consistent with the prolactin-releasing factor that drives the lactotrophs, and down-
observed daily patterns of expression of these four genes in the regulated under short days.
ovine pars tuberalis under long and short days (38). (iii) The Per– A more speculative, but equally testable, extension of this
Cry interval (c) thus varies directly with the length of the internal coincidence concept is the possibility that photorefrac-
photoperiod, and determines the period when the protein products toriness is the result of uncoupling of the melatonin-control of the
of these genes (CRYs and PERs) are coincident in the cytoplasm. clock genes such that the Per–Cry interval (c) no longer reflects
(iv) The level of coincidence in turn determines the formation of the ambient photoperiod. This is unlikely to involve the timing of
PER/CRY protein complexes that potentially control the tran- Per1 because the expression of this clock gene has been shown to
scription of other clock genes as part of the cellular clockwork (C), stay locked to the onset of the light phase in both photosensitive
and the transcription of downstream genes (D) that dictate the and refractory animals (A.S. Loudon, personal communication).
secretory activity of the melatonin-target cell. For the prolactin We predict that the melatonin control of Cry expression may form

# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397

396 Photoperiod and circannual timer
the basis of photorefractoriness. There is also the possibility that
circannual rhythms are generated by the intrinsic control of the
Per–Cry interval that oscillates between stages of decreased
coincident (long-day type) and increased coincident (short-day
type) under a constant photoperiod/melatonin signal, producing a
cycle with a periodicity of 10–12 months. The data from the HPD
sheep studies support the view that the processes of decoding the
melatonin signal and the generation of a circannual rhythm are
tissue-specific and cell autonomous, and that they may share a
common molecular mechanism.
Acknowledgements
We are grateful to Dr Sophie Messager for supplying the molecular probes for the
ovine clock genes, to Professor Iain Clarke for carrying out the HPD surgery, to
Norah Anderson, Marjorie Thompson and the staff at the Marshall Building for
dedicated work in sampling and care of the Soay rams, to Kirsten Brown, Ian
Swanston and Irene Cooper for expert help with the in situ hybridization and
radioimmunoassay methodology, and to Ted Pinner for the artwork.
References
1 Lincoln GA. Significance of seasonal cycles in prolactin secretion in male
mammals. In: Serio M, ed. Perspectives in Andrology. New York: Raven
Press, 1989: 299–306.
2 Hindes L, Loudon ASI. Mechanisms of seasonality in marsupials: a
comparative review. In: Saunders N, Hindes L, eds. Marsupial Biology.
Sydney: UNSW Press, 1995: 41–70.
3 Loudon ASI, Jabbour HN. Seasonal rhythms in the physiology of deer:
endogenous clocks and target tissues. In: Milne JA. Recent Developments
in Deer Biology. Edinburgh: Moredun Research Institute Publications,
1994: 197–207.
4 Choy VJ, Nixon AJ, Pearson AJ. Distribution of prolactin receptor
immunoreactivity in ovine skin and changes during the wool follicle
cycle. J Endocrinol 1997; 155: 265–275.
5 Jansen HT, Jackson GL. Circannual rhythms in the ewe: patterns of
ovarian cycles and prolactin under two different constant photoperiods.
Biol Reprod 1993; 49: 627–634.
6 Lincoln GA. Melatonin modulation of prolactin and gonadotrophin
secretion. Systems ancient and modern. In: Olcese J, ed. Melatonin After
Four Decades. New York: Kluwer Academic/Plenum Publications, 1999:
137–153.
7 Tamarkin L, Baird CJ, Almeida OFX. Melatonin: a coordinating signal
for mammalian reproduction. Science 1985; 227: 714–720.
8 Stehle JH, von Gall C, Schomerus C, Korf HW. Of rodents and ungulates
and melatonin: creating a uniform code for darkness by different signaling
mechanisms. J Biol Rhythms 2001; 16: 312–325.
9 Hastings MH, Walker AP, Powers JB. Differential effects of photoper-
iodic history on the responses of gonadotrophin and prolactin in inter-
mediate daylengths in the male Syrian hamster. J Biol Rhythms 1989; 4:
335–350.
10 Morgan PJ. The pars tuberalis: the missing link in the photoperio-
dic regulation of prolactin secretion. J Neuroendocrinol 2000; 12:
287–295.
11 Malpaux B, Daveau A, Maurice-Mandon F, Duarte G, Chemineau P.
Evidence that melatonin acts in the premammillary hypothalamic area to
control reproduction in the ewe: presence of binding sites and stimulation
of luteinizing hormone secretion by in situ microimplant delivery.
Endocrinology 1998; 139: 1508–1516.
12 Hazlerigg DG, Hastings MH, Morgan PJ. Production of a prolactin
releasing factor by the ovine pars tuberalis. J Neuroendocrinol 1996;
8: 489–492.
13 Stirland JA, Johnston JD, Cagampang FRA, Morgan PJ, Castro MG,
White MRH, Davis JRE, Loudon ASI. Photoperiodic regulation of
prolactin gene expression in the Syrian hamster by a pars tuberalis-
derived factor. J Neuroendocrinol 2001; 13: 147–157.
14 Lincoln GA, Clarke IJ. Photoperiodically-induced cycles in the secretion
of prolactin in hypothalamo-pituitary disconnected rams: evidence for
translation of the melatonin signal in the pituitary gland. J Neuroendo-
crinol 1994; 6: 251–260.
# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397
15 Lincoln GA, Clarke IJ. Evidence that melatonin acts in the pituitary gland
through a dopamine-independent mechanism to mediate effects of day-
length on the secretion of prolactin in the ram. J Neuroendocrinol 1995; 7:
637–643.
16 Lincoln GA, Clarke IJ, Sweeney T. ‘Hamster-like’ cycles in testicular size
in the absence of gonadotrophin secretion in HPD rams exposed to long-
term changes in photoperiod and treatment with melatonin. J Neuroen-
docrinol 1996; 8: 855–866.
17 Lincoln GA, Rhind SM, Pompolo S, Clarke IJ. Hypothalamic control of
photoperiod-induced cycles in food intake, body weight, and metabolic
hormones in rams. Am J Physiol Reg Int Comp Physiol 2001; 281: R76–
R90.
18 Williams L, Lincoln GA, Mercer JG, Barrett P, Morgan PJ, Clarke IJ.
Melatonin receptors in the brain and pituitary gland of hypothalamo-
pituitary disconnected Soay rams. J Neuroendocrinol 1997; 9: 639–643.
19 Lincoln GA, Clarke IJ. Noradrenaline and dopamine regulation of
prolactin secretion in sheep: role in prolactin homeostasis but not
photoperiodism. J Neuroendocrinol 2001; 14: 36–44.
20 Thiery JC. Monoamine content of the stalk-median eminence and
hypothalamus in female sheep as affected by day length. J Neuroendo-
crinol 1991; 3: 407–411.
21 Le Corre S, Chemineau P. Control of photoperiodic inhibition of luteiniz-
ing hormone secretion by dopaminergic and serotinergic systems in
ovariectomized Ile-de-France ewes supplemented with estradiol. J
Reprod Fertil 1993; 97: 367–373.
22 Curlewis JD, McNeilly AS. Prolactin short-loop feedback and prolactin
inhibition of luteinizing hormone secretion during the breeding season
and seasonal anoestrous in the ewe. Neuroendocrinology 1991; 54: 279–
285.
23 Lincoln GA, Clarke IJ. Role of the pituitary gland in the development of
photorefractoriness and generation of long-term changes in prolactin
secretion in rams. Biol Reprod 2000, 2002; 62: 432–438.
24 Lincoln GA, Andersson H, Clarke IJ. Calendar timing in the pituitary
gland is independent of the output signal. The Fifth International Con-
gress in Neuroendocrinology. Bristol: Abstract P8.
25 Hazlerigg DG, Morgan PJ, Messager S. The cellular basis to the decoding
of photoperiodic time in mammals  what can we learn from the pars
tuberalis? J Biol Rhythm 2001; 16: 326–335.
26 McNulty S, Ross AW, Shiu KY, Morgan PJ, Hastings MH. Phosphoryla-
tion of CREB in ovine pars tuberalis is regulated both by cyclic AMP-
dependent and cyclic AMP-independent mechanisms. J Neuroendocrinol
1996; 8: 635–645.
27 Ross AW, Barrett P, Mercer JG, Morgan PJ. Melatonin suppresses the
induction of AP-1 transcription factor components in the pars tuberalis of
the pituitary. Mol Cell Endocrinol 1996; 123: 71–80.
28 Reppert SM, Weaver DR, Ebisawa T. Cloning and characterisation of a
mammalian melatonin receptor that mediates reproductive and circadian
responses. Neuron 1994; 13: 1–9.
29 Hazlerigg DG, Gonzalez-Brito A, Lawson W, Hastings MH, Morgan PJ.
Prolonged exposure to melatonin leads to time-dependent sensitization of
adenylate cyclase and down-regulates melatonin receptors in pars tuber-
alis cells from ovine pituitary. Endocrinology 1993; 132: 285–292.
30 Morgan PJ, Ross AW, Scott Graham E, Adam C, Messager S, Barrett P.
oPer1 is an early response gene under photoperiodic regulation in the
ovine pars tuberalis. J Neuroendocrinol 1998; 10: 319–323.
31 Messager S, Ross AW, Barrett P, Morgan PJ. Decoding photoperiodic
time through Per1 and ICER gene amplitude. Proc Natl Acad Sci USA
1999; 96: 9938–9943.
32 Messager S, Hazlerigg DG, Mercer JG, Morgan PJ. Photoperiod differ-
entially regulates the expression of Per1 and ICER in the pars tuberalis
and the suprachiasmatic nucleus of the Siberian hamster. Eur J Neurosci
2000; 12: 2865–2870.
33 Messager S, Garabette ML, Hastings MH, Hazlerigg DG. Tissue-specific
abolition of Per1 expression in the pars tuberalis by pinealectomy in the
Syrian hamster. Neuroreport 2001; 12: 579–582.
34 Von Gall C, Garabette ML, Kell CA, Frenzel S, Dehghani F, Schumm-
Draeger P-M, Weaver DR, Korf H-W, Hastings MH, Stehle JH. Rhythmic
gene expression in pituitary depends on heterologous sensitisation by the
neurohormone melatonin. Nature Neurosci 2002; 5: 234–238.
35 Reppert SM, Weaver DR. Molecular analysis of mammalian circadian
rhythms. Ann Rev Physiol 2001; 63: 647–676.
36 Young MW, Kay SA. Time zones: a comparative genetics of circadian
clocks. Nature Rev Genet 2001; 2: 702–715.
37 Reppert SM, Weaver DR. Coordination of circadian timing in mammals.
Nature 2002; 418: 935–941.

38 Lincoln GA, Messager S, Andersson H, Hazlerigg D. Temporal expres-
sion of seven clock genes in the suprachiasmatic nucleus and the pars
tuberalis of the sheep: evidence for an internal coincidence timer. Proc
Natl Acad Sci USA 2002; 99: 13890–13895.
39 Shearman LP, Sriram S, Weaver DR, Maywood ES, Chaves I, Zheng B,
Kume K, Lee CC, van der Horst GT, Hastings MH, Reppert SM.
# 2003 Blackwell Publishing Ltd, Journal of Neuroendocrinology, 15, 390–397
Photoperiod and circannual timer 397
Interacting molecular loops in the mammalian circadian clock. Science
2000; 288: 1013–1019.
40 Lowrey PL, Shimomura K, Antoch MP, Yamazaki S, Zemenides PD,
Ralph MR, Menaker M, Takahashi JS. Positional syntenic cloning and
functional characterization of the mammalian circadian mutation tau.
Science 2000; 288: 483–492.