Journal of Neuroendocrinology, 2002, Vol. 14, 318–329
Photorefractoriness of Immune Function in Male Siberian Hamsters
(Phodopus sungorus)
B. J. Prendergast,* K. E. Wynne-Edwards,† S. M. Yellon‡ and R. J. Nelson*
*Departments of Psychology and Neuroscience, The Ohio State University, Columbus, OH, USA.
†Department of Biology, Queens University, Kingston, Ontario, Canada.
‡Center for Perinatal Biology, Department of Physiology, Loma Linda University School of Medicine, Loma Linda, CA, USA.
Key words: reproduction, lymphocyte, photoperiodism, seasonality, melatonin.
Abstract
Short days induce multiple changes in reproductive and immune function in Siberian hamsters.
Short-day reproductive inhibition in this species is regulated by an endogenous timing mechanism;
after approximately 20 weeks in short days, neuroendocrine refractoriness to short-day patterns of
melatonin develops, triggering spontaneous recrudescence of the reproductive system. It is unknown
whether analogous mechanisms control immune function, or if photoperiodic changes in immune
function are masked by prevailing photoperiod. In Experiment 1, 3 weeks of exposure to long days
was not sufficient to induce long-day-like enhancement of in vitro lymphocyte proliferation in
short-day adapted male Siberian hamsters. Experiment 2 tested the hypothesis that immunological
photorefractoriness is induced by prolonged exposure to short days. Adult male hamsters were
gonadectomized or sham-gonadectomized and housed in long (14 h light/day) or short (10 h
light/day) photoperiods for 12, 32 or 40 weeks. Somatic and reproductive regression occurred after
12 weeks in short days, and spontaneous recrudescence was complete after 32–40 weeks in short
days, indicative of somatic and reproductive photorefractoriness. In gonad-intact hamsters, 12 weeks
of exposure to short days decreased the number of circulating granulocytes and increased the
number of B-like lymphocytes. After 32 weeks in short days, these measures were restored to
long-day values, indicative of photorefractoriness; castration eliminated these effects of photoperiod.
In both intact and castrated hamsters, in vitro proliferation of splenic lymphocytes was inhibited by
12 weeks of exposure to short days. After 40 weeks in short days lymphocyte proliferation was
restored to long-day values in intact hamsters, but remained suppressed in castrated hamsters.
These results suggest that short-day-induced inhibition of lymphocyte function does not depend on
gonadal regression, but that spontaneous recrudescence of this measure is dependent on gonadal
recrudescence. In Experiment 3, in vitro treatment with melatonin enhanced basal proliferation of
lymphocytes from male hamsters exposed to short days for 12 weeks, but had no effect on
lymphocytes of photorefractory hamsters or long-day control hamsters. Lymphocytes of castrated
hamsters were unresponsive to in vitro melatonin, suggesting that photoperiodic changes in gonadal
hormone secretion may be required to activate mechanisms which permit differential responsiveness
to melatonin depending on phase in the annual reproductive cycle. Together, these data indicate
that, similar to the reproductive system, the immune system of male Siberian hamsters exhibits
refractoriness to short days.
Seasonal changes in temperature and food availability
are common at nonequatorial latitudes; in the face of these
recurring environmental challenges, animals must maintain
energy balance, survive and reproduce. Species inhabiting
these regions year-round often possess adaptations which
permit seasonal adjustments in physiology (e.g. seasonal
reproduction, hibernation) that ultimately restrict energetically expensive activities (e.g. breeding, lactation) to energetically favourable times of year (1). Seasonal changes in day
length (photoperiod) provide relatively accurate indicators
Correspondence to: Brian J. Prendergast, Department of Psychology, The Ohio State University, Townshend Hall, Columbus, OH 43210, USA
(e-mail: [email protected]).
#
2002 Blackwell Science Ltd
Recrudescence of immune function 319
of phase in the annual geophysical cycle and can function
as proximate cues for triggering such physiological adaptations
(1, 2). The neuroendocrine system encodes seasonal changes
in photoperiod by varying the duration of pineal melatonin
secretion in proportion to the length of the night (3). Relatively longer nights during late summer, autumn and winter
are associated with extended nocturnal melatonin signals,
relative to the shorter nights of spring and early summer when
the duration of nocturnal melatonin secretion is compressed
(4). Changes in the duration of nocturnal melatonin signals
are decoded by melatonin target tissues in the brain to alter
reproductive physiology over the course of the year (5, 6).
With few exceptions, photoperiodic changes in physiology
and behaviour are pineal-dependent and are mediated by
changes in the duration of nocturnal melatonin secretion (7, 8).
Seasonal changes in immune function are ubiquitous in
mammals (9) and likely contribute to annual patterns of
illness and death (10). In common with reproductive cycles,
seasonal changes in immunocompetence are driven primarily
by changes in day length (11). Siberian hamsters (Phodopus
sungorus) are seasonally breeding, photoperiodic rodents
that undergo a constellation of immunological changes when
exposed to short, winter-like, photoperiods. Short-day lengths
sufficient to inhibit reproductive function are associated with
reduced phagocytosis and oxidative burst activities, blunted
T-cell dependent humoral immunity, suppressed in vitro
basal lymphocyte proliferation, and enhanced natural killer
cell cytotoxicity (12–15). The mechanisms by which short
days alter these components of immune function remain
largely unknown; in vivo (13, 16) and in vitro (15) data
indicate immunomodulatory roles for both melatonin and
seasonal changes in gonadal steroid hormone secretion in the
photoperiodic control of immune function in this species.
In common with other seasonal breeders, Siberian hamsters
must not only inhibit reproductive function in autumn,
but also reactivate the reproductive system months later to
coincide with the onset of spring. Rather than waiting for
longer spring day lengths to appear, hamsters use an endogenous seasonal timing mechanism to accomplish this vernal
reproductive reactivation. Briefly, hamsters undergo gonadal
regression when exposed to short photoperiods (=12 h
light/day) (17). Reproductive inhibition in short days is not
sustained indefinitely, however. After approximately 20 weeks
in short days, hamsters spontaneously revert to a long-day,
reproductively competent, phenotype (18). Termed ‘spontaneous recrudescence’, this recovery of reproductive function
occurs despite continued exposure to photoperiods that were
previously inadequate to sustain reproductive function (19).
After spontaneous recrudescence has occurred, hamsters do
not exhibit another transition to the short-day reproductive
phenotype unless first exposed to an indeterminate episode of
long days, which restores responsiveness to short days and
melatonin (20). An animal is thus considered photorefractory
to short days for a given trait when a photoperiod previously
sufficient to sustain the short-day phenotype in that trait (e.g.
reproductive function) no longer does so (19). Photorefractoriness is a fundamental feature of photoperiodic seasonal
timekeeping mechanisms in mammals and, in all photoperiodic traits described to date, invariably follows prolonged
exposure to short days. Reproductive photorefractoriness
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
is mediated by a spontaneous loss of responsiveness to
melatonin by neural tissues after 5–6 months of exposure to
short days (6). The physiological basis of this spontaneous
change in responsiveness to melatonin is unknown; however,
its functional significance is clear: reproductive hormone
secretion and gametogenesis require many weeks to be fully
reactivated from a state of winter quiescence (21). Photorefractoriness in mid-winter initiates the recovery of reproductive function several weeks in advance of the appearance
of long-day lengths (18) and thereby permits early spring
breeding (22).
The issue of photorefractoriness has received little attention
with regard to immune function (23). It is unknown, for
example, whether photoperiod-responsive components of the
immune system become refractory to short days. It may be the
case that, unlike in the reproductive system, the transition
from winter-like to summer-like immunocompetence can be
accomplished in a matter of hours or days. If this were
so, there would be no obvious advantage associated with
photorefractoriness mechanisms governing photoresponsive
immune traits: hamsters could simply wait for the appearance
of longer day lengths in the spring to initiate rapid seasonal
transitions in immune function. Alternatively, photorefractoriness may be an inevitable process common to all physiological systems regulated by day length and melatonin; in this
case, refractoriness of a trait to short days should ensue after
an extended interval of exposure to short days, independent of
the trait’s capacity for rapid phenotypic transition. A related
issue is the immunomodulatory effect of gonadal hormones
on immune function in Siberian hamsters (16). Spontaneous
recrudescence of reproductive hormone secretion may be
sufficient to drive changes in the immune system. In addition,
given the direct effect of melatonin on lymphocytes of some
photoperiodic rodents (14, 15), Siberian hamsters lymphocytes provide a model in which to test the hypothesis that
a spontaneous change in responsiveness (i.e. refractoriness)
to melatonin after chronic exposure to short days is a characteristic common among melatonin-responsive tissues, rather
than a phenomenon limited to neural melatonin targets that
govern reproductive function (6).
The objectives of the following experiments were to test
hypotheses related to the phenomenology and mechanisms of
photorefractoriness of the immune system. First, to estimate
the lability of photoresponsive immune traits, we assessed
whether exposure of immunologically photoinhibited male
hamsters to photostimulatory long days would trigger a rapid
or a more gradual restoration of the long-day immune phenotype. Next, to test the hypothesis that initial components of
cell-mediated immune function become refractory to short
days, age-matched groups of male Siberian hamsters were
exposed to short days for 12, 32 or 40 weeks to induce regression, photorefractoriness and spontaneous recrudescence of
the reproductive system. Several measures of immune function
were then assessed. Because immune function is influenced by
gonadal hormones in this species (16), we conducted parallel
experiments in castrated hamsters, to test the hypothesis that
refractoriness of immune function is dependent upon spontaneous recrudescence of the reproductive system. Finally, we
assessed whether isolated lymphocytes of photorefractory
hamsters were refractory to melatonin in vitro.
320 Recrudescence of immune function
Nonresponder exclusions
In most Siberian hamster populations, a subset of individuals fails to exhibit
the modal reproductive and somatic responses to short days (i.e. gonadal
regression, decrease in body weight, pelage moult). Short-day changes in some
measures of immune function do not occur in these ‘nonresponders’ (14); these
individuals were excluded from analyses. Nonresponders were defined as individuals that failed to exhibit either a moult to winter pelage (in castrated
hamsters) or a decrease of at least 40% in testis volume (in intact hamsters)
after 8 weeks (Experiment 1) or 12 weeks (Experiment 2) in short days. These
criteria resulted in the removal of four hamsters from Experiment 1 and 24
hamsters from Experiment 2.
Methods
Animals
One hundred and 64 male Siberian hamsters (Phodopus sungorus) used in the
present experiment were derived from a breeding colony maintained at Johns
Hopkins University, established with hamsters provided by Dr Katherine
Wynne-Edwards at Queens University (Kingston, Ontario, Canada). Hamsters were housed in polypropylene cages from birth in a room illuminated for
14 h per day (14 L) with fluorescent light (LD; lights on 01.00 h, Eastern
Standard Time). Food (LabDiet 5001; PMI Nutrition, Brentwood, MO, USA)
and tap water were provided ad libitum; ambient temperature was maintained
at 21t2uC and relative humidity was held constant at 50t5%. Hamsters were
weaned at 18–25 days of age, segregated by sex and housed in LD (1–4 animals
per cage) until experimental treatments began.
Lymphocyte proliferation assay
In vitro splenocyte proliferation was assessed in the presence of the mitogen
concanavalin A (ConA; Sigma, St Louis, MO, USA). Measurement of
mitogen-stimulated lymphocyte proliferation is perhaps the most widespread
functional in vitro assessment of the cellular arm of the immune system (25);
implicit in the performance of this assay is the assumption that elevated
in-vitro proliferation reflects elevated in vivo immune response. A polyclonal
mitogen such as ConA triggers essentially the same growth response
(proliferation) as does an antigen (26). At autopsy, spleens were harvested
under aseptic conditions and immediately suspended in culture medium
(RPMI-1640, Mediatech; Herndon, VA, USA). Spleens were weighed, and
splenic lymphocytes were extracted from whole tissue under a sterile laminarflow hood. Lymphocytes were first dispersed by flushing spleens with 2 ml
RPMI from a syringe. Spleen fragments and ejected lymphocytes were gently
ground against and then passed through a 70-mm Falcon cell strainer.
Lymphocytes were resuspended in 4 ml RPMI and separated by Ficoll
(NycoPrep 1Step-1.077/265 Animal Ficoll; Accurate Chemical and Scientific;
Westbury, NY, USA) density-gradient centrifugation (600 g for 20 min). The
resulting mononuclear cell suspensions were retained, washed twice (600 g
for 10 min, each) with RPMI, and resuspended in 1 ml of supplemented
culture medium [RPMI-1640/Hepes supplemented with 1% penicillin
(5000 U/ml)/streptomycin (5000 ml/ml), 1% L-glutamine (2 mM/ml), 0.1%
2-mercaptoethanol (5r10x2 M/ml) and 10% heat-inactivated fetal bovine
serum]. Lymphocytes were counted and assessed for viability on a haemacytometer by Trypan blue exclusion. Viable cells (which exceeded 95%) were
adjusted to 2r106 cells/ml with supplemented culture medium.
Lymphocyte proliferation (blastogenesis) and responsiveness to the T-cell
mitogen ConA were assessed. Basal proliferation was determined in a portion
of a sterile flat-bottomed 96-well culture plate filled with 50 ml of mitogen-free
culture media and a 50-ml aliquot of lymphocyte cell suspensions (1r105 cells).
In Experiment 1, other wells contained ConA, serially diluted with 50 ml of
culture media at concentrations of 40, 20, 10, 5, 2.5, 1.25 and 0.625 mg/ml,
to yield a final volume of 100 ml in each well. In Experiment 2, unstimulated
proliferation of splenic lymphocytes was assessed in the absence of ConA
(basal proliferation). All samples were assayed in duplicate.
Culture plates were incubated at 37 uC with 5% CO2 for 48 h. This interval
is sufficient for induction of splenocyte function and substantial proliferation
in this and other rodent species (27, 28). Splenocyte proliferation was assessed
using a colourimetric assay based on a tetrazolium salt (3-(4,5-demethylthiazol2-yl)-5-(3-carboxy-methoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; MTS)
with 0.92 mg/ml of phenazine methosulphate (PMS) in sterile Dulbecco’s
PBS (Promega; Madison, WI, USA). Tetrazolium salts are incorporated into
active mitochondria; the salts are bioreduced into a coloured formazan
product in metabolically active cells. The quantity of formazan product as
measured by the amount of absorbance of 490 nm light is directly proportional
to the number of living cells in culture (29); the accumulation of coloured
formazan product (i.e. the product of tetrazolium salt based cell-counting
techniques) thus provides a measure of cell number (not mitosis per se). In the
absence of direct measures of apoptosis and cell death, absorbance readings
reflect the net sum of proliferation and cell death. However, under conditions of
mitogen stimulation, the accumulation of coloured formazan products is
strongly and positively correlated with tritiated thymidine uptake (a direct
measure of blastogenesis (30, 31). After the 48 h incubation, the MTS/PMS
reagent was added to each incubation well (20 ml per 100 ml incubation
volume). Plates were then incubated at 37uC with 5% CO2 for an additional
4 h. The optical density (OD) of each well was determined with a microplate
reader (Bio-Rad Benchmark, Bioxab; Richmond, CA, USA) equipped with a
490-nm filter. Mean OD values for each set of duplicates were used in statistical
analyses. Dose–response curves to mitogens and melatonin were constructed
based on group mean values within each photoperiod treatment.
Experiment 1: Latency to immunoenhancement by long days
Photoperiod manipulations
After weaning, 25 male hamsters were transferred to a long-day photoperiod
(16 L). At 5–7 months of age, singly housed hamsters were either transferred to
a short-day photoperiod (8 L; lights on 07.00 h; n=18) or remained in long
days (16 L to 16 L; n=7). Eight weeks later (week 0), gonadal regression
(>50% decrease in testis volume) was verified in 14 of 18 hamsters housed in
8 L. Seven gonadally regressed hamsters were then returned to 16 L (8 L to
16 L; n=7) and seven hamsters remained in 8 L (8 L to 8 L; n=7). Hamsters
were killed 3 weeks later, at which time somatic, reproductive and immune
measures were determined.
Experiment 2: Spontaneous recrudescence of immune function in short days
Surgical procedures
At 2–3 months of age (week 0), 139 male hamsters from our long-day breeding
colony were either surgically castrated (n=68) or subjected to a sham
castration surgery (externalization of each testicle without ligation or incision)
under sodium pentobarbital anaesthesia (n=71). Because of the substantial
number of animals involved, a group of 8–18 animals underwent week 0
surgical treatments every several weeks over the course of 5 months. This
staggered experimental design controls, to some degree, for any possible timeof-year variables. Immediately after surgery, intact and castrated (x) hamsters
were randomly assigned to a photoperiod treatment group (see below). A total
of 13 hamsters died over the 40-week experiment (eight intact, five castrated).
Groups described below reflect final sample sizes after 24 nonresponder
hamsters (12 intact, 12 castrated) were excluded from analyses (see
Nonresponder Exclusions).
Photoperiod manipulations
After surgery (week 0), a randomly selected group of castrated and gonadally
intact hamsters was transferred to a short-day photoperiod (SD; 10 L; lights
on 05.00 h) where they remained for the next 32 weeks (SD32, n=10; xSD32,
n=11). Another group remained in LD for the entire experiment (LD32,
n=10; xLD32, n=10). A third group was housed in LD for 20 weeks, at which
time (week 20) they were transferred to SD for the next 12 weeks (SD12, n=16;
xSD12, n=16). To verify that spontaneous recrudescence was complete, a
subset of SD32 and xSD32 hamsters remained in SD for an additional 8 weeks
[designated SD40 (n=10) and xSD40 (n=8), respectively] along with appropriate long-day controls which remained in LD until week 40 (LD40, n=5;
xLD40, n=10). SD32 and SD40 hamsters were combined and treated as a
single group for all body weight, pelage and testis volume analyses up to and
including week 32. LD32 and LD40 hamsters were likewise combined for
analyses, as were xSD32/xSD40 and xLD32/xLD40 hamsters.
Somatic and reproductive measures
At predetermined intervals throughout the experiments, hamsters were
weighed (t0.1 g), and the length and width of the left testis was measured
(t0.1 mm) under light anaesthesia induced by methoxyflurane vapours. The
product of testis width squared times testis length provided a measure of
estimated testis volume (ETV) that is highly correlated (r>0.9) with testis
weight (18). Stage in the pelage colour cycle was also assessed in each hamster
using an integer scale of 1–4 (1=dark ‘summer’ fur, 4=white ‘winter’ fur)
without knowledge of the animal’s treatment condition (24). Hamsters were
killed by cervical dislocation, and trunk blood was collected; testis weights
(t0.1 mg) and body weights were determined at autopsy.
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Recrudescence of immune function 321
Assessment of immune cell phenotypes (Flow Cytometry)
At predetermined intervals between weeks 12 and 32, 1 ml of blood was taken
from a subset of hamsters via the retro-orbital sinus and collected into sterile
heparinized collection vials. Samples were shipped to Loma Linda University
(Loma Linda, CA, USA), and 24–48 h later formed elements in erythrocytelysed whole blood were gated and analysed by FACSort flow cytometry
following previously described methods (32) which were empirically validated
for the Siberian hamster. Forward and side-scatter characteristics identified
morphologies that were consistent with granulocytes (primarily neutrophils
and eosinophils), monocytes and a population of B-like lymphocytes. This
latter immunophenotype was further found to cross-react with an antisera
specific for immunoglobulin M (S. M. Yellon and O. R. Fagoaga, unpublished
observations). Samples were analysed by flow cytometry in a total of six separate assays. Because of substantial interassay variability, data were expressed
as a percentage of long-day control values. At least one LD32 hamster was
included in every assay of castrated hamsters, and at least one xLD32 hamster
was included in each assay of intact hamsters. These samples served as internal
controls. The relative proportions of monocytes, granulocytes and B-like
lymphocytes were calculated as a percentage of internal controls.
Experiment 3: Lymphocyte responses to melatonin
In a separate assay, splenic lymphocytes from hamsters in Experiment 2
were incubated in vitro in the presence or absence of melatonin (N-acetyl5-methoxytryptamine) and lymphocyte proliferation was assessed. Ficollseparated lymphocytes were obtained as described above (see Lymphocyte
Proliferation Assay). Melatonin (Sigma) was dissolved in 95% ethanol and
diluted to a concentration of 50 pg/ml, 500 pg/ml or 1000 pg/ml with culture
medium. Then, 100 ml of each melatonin concentration was added to individual culture wells, along with 50 ml of culture medium and 50 ml of the
lymphocyte suspensions (to yield a final volume of 200 ml per well). Culture
medium without added melatonin served as a control solution (i.e. 0 pg/ml).
After 48 h of incubation, assay plates were treated with the MTS/PMS reagent
and, 4 h later, OD measurements were determined at 490 nm with a microplate
reader. All samples were assayed in duplicate, and the mean OD values for
each set of duplicates were used in statistical analyses (see Lymphocyte
Proliferation Assay).
Statistical analysis
Between-group means for all dependent variables except pelage colour were
compared by between-subjects ANOVA using Statview 5 (SAS Institute Inc.,
Cary, NC, USA). Fisher’s least significant difference tests were used to evaluate
pairwise comparisons where permitted by a significant F-value. Within-group
means were compared using paired t-tests (two-tailed) to determine whether
a trait exhibited significant changes between two consecutive sampling points.
Distributions of pelage scores were compared by the Kruskall–Wallis nonparametric test for multiple treatment groups and, where permitted, by
a significant H-statistic, followed by Mann–Whitney U-tests for pairwise
comparisons. P<0.05 was considered statistically significant for observed
mean differences.
Results
Experiment 1: Latency to immunoenhancement by long days
Hamsters exhibited gonadal (F=12.2, d.f.=6,54, P<0.0001)
and somatic (F=8.43, d.f.=6,54, P<0.0001) regression
over the first 8 weeks of exposure to 8 L, characteristic of
responsiveness to short days. Both groups of 8 L hamsters
(8 L to 16 L and 8 L to 8 L) had comparably regressed testes
that were significantly smaller than those of 16 L to 16 L
hamsters at week 8 (P<0.0001, both comparisons). Transfer
of photoregressed hamsters to long days at week 8 resulted in
slight, but nonsignificant increases in ETV and body weight
over the next 3 weeks (Fig. 1A,B). After 3 weeks of exposure to
long days, testes of 8 L to 16 L hamsters were heavier than
those of hamsters that had remained in short days (P<0.05),
though still substantially smaller in weight than those of 16 L
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
to 16 L hamsters (P<0.0001; Fig. 1C). Epididymal white
adipose tissue and seminal vesicle weights were smaller in
8 L to 8 L hamsters relative to 16 L to 16 L hamsters at
week 11 (P<0.0001, both comparisons) and were not significantly increased by transfer to long days (P>0.80, both
comparisons; data not shown).
Photoperiod manipulations significantly influenced in vitro
proliferation of splenic lymphocytes. Basal (spontaneous)
proliferation of lymphocytes was significantly lower in
hamsters housed in 8 L for 11 weeks relative to hamsters
kept in long days over the same time interval (P<0.05;
Fig. 1D). Transfer of photoregressed hamsters to long days
for 3 weeks did not result in a significant elevation in basal
lymphocyte proliferation above short-day values: proliferation of lymphocytes from 8 L to 16 L hamsters was comparable to that of 8 L to 8 L hamsters (P>0.70) and tended to
remain lower than that of 16 L to 16 L hamsters (P=0.06;
Fig. 1D). In vitro treatment of splenic lymphocytes with the
mitogen ConA induced significant increases in proliferation
(F=23.9, d.f.=7,14; P<0.0001; Fig. 2B); however, neither
the magnitude nor the pattern of responsiveness to ConA was
affected by photoperiod treatments (P>0.20, all comparisons;
data not shown).
Experiment 2: Spontaneous recrudescence of
immune function in short days
Body weight
After 6 weeks of exposure to short days (i.e. week 6), gonadintact SD32 and SD40 hamsters weighed significantly less
than hamsters housed in long days (P<0.001); body weights
of short-day hamsters remained significantly reduced through
week 20 (week 12, P<0.001; week 20, P<0.001), after which
they no longer differed from long-day control values (week 32,
P>0.05; week 40, P>0.35; Fig. 2A). In castrated males, body
weights of xSD32 and xSD40 hamsters followed a similar
pattern of transient reduction during weeks 0–20, followed
by spontaneous regrowth (Fig. 2B). Both intact and castrated hamsters transferred to short days at week 20 (SD12
and xSD12, respectively) underwent significant reductions in
body weight as measured 12 weeks later (intact hamsters,
P<0.0001; castrated hamsters, P<0.0005). Thus, body weight
of hamsters exposed to short days for i32 weeks exhibited
regression followed by spontaneous recrudescence, indicative
of photorefractoriness to short days. No significant increases
in body weight occurred in short-day hamsters between weeks
32 and 40 (intact hamsters: t=0.53, d.f.=9, P>0.60, Fig. 3A;
castrated hamsters: t=0.59, d.f.=7, P>0.55, Fig. 3B).
Pelage
Similar to body weight, pelage colour of hamsters exposed
to short days for i32 weeks exhibited a complete cycle of
regression, recrudescence, and refractoriness. Intact and castrated (data not shown) hamsters exhibited comparable patterns of fur moult. Fur colour was significantly lighter after
6 weeks in short days (intact hamsters: U=105.0, P<0.05;
castrated hamsters: U=102.0, P<0.01), and remained significantly lighter than that of long-day control animals until
approximately week 32. Pelage colour of photorefractory
hamsters was restored to the long-day phenotype by week 32
322 Recrudescence of immune function
(A)
(B)
1000
55
900
50
Estimated testis volume (mg)
800
700
Body weight (g)
45
600
500
400
40
35
300
200
30
16 L to 16 L
8 L to 16 L
8 L to 8 L
100
0
16 L to 16 L
8 L to 16 L
8 L to 8 L
25
0
4
8
11
0
Week
4
8
11
Week
(C)
(D)
1200
0.8
1000
800
Absorbance units (OD)
Paired testis weight (mg)
0.7
600
400
0.6
0.5
0.4
200
0
0.3
16 L to 16 L
8 L to 16 L
8 L to 8 L
16 L to 16 L
8 L to 16 L
8 L to 8 L
FIG. 1. (A) Mean (tSEM) estimated testis volumes and (B) body weights of hamsters in Experiment 1. Hamsters were housed in long day lengths
(16 L to 16 L) or short day lengths (8 L to 8 L) for 11 weeks, or in short day lengths beginning on week 0 and transferred to long day lengths on week 8
(8 L to 16 L). (C) Mean (tSEM) paired testis weights on week 11. (D) Mean (tSEM) basal (0 mg/ml ConA) in vitro proliferation of splenic lymphocytes
from hamsters in Experiment 1 (expressed as absorbance units at 490 nm) as determined on week 11. *P=0.05 versus 16 L to 16 L group. †Pf0.05 versus
all other groups.
in intact hamsters (U=120.0, P>0.05) and by week 40 in
castrated hamsters (U=0, P>0.95). Pelage of SD12 and
xSD12 hamsters exhibited significant moult after exposure to
short days beginning at week 20 (intact, H=25.5, P<0.0001;
castrated, H=32.5, P<0.0001).
Fig. 3A). Testis size of SD32 and SD40 hamsters remained
significantly lower than those of long-day control hamsters for
the next 26–34 weeks. Gonadal recrudescence in SD32/SD40
hamsters commenced after 12–20 weeks of exposure to short
days, as testis size did not change between weeks 6 and 12
(t=1.34, d.f.=19, P=0.20), but increased significantly
between weeks 12 and 20 (t=5.42, d.f.=19, P<0.0001;
Fig. 3A). Estimated testis sizes were restored to approximately
90% of initial (week 0) values after 32 weeks in short days,
Testis size
Intact hamsters underwent gonadal regression within the first
6 weeks of exposure to short days (t=3.00, d.f.=19, P<0.01;
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Recrudescence of immune function 323
(A)
(A)
140
55
LD32/LD40
SD12
SD32/SD40
120
% Week 0 testis volume
Body weight (g)
50
45
40
35
30
25
6
12
0
45
40
35
Castrated
25
6
12
6
20
12
(B )
xLD32/xLD40
xSD12
xSD32/xSD40
0
LD32/LD40
SD12
SD32/SD40
1000
30
32
40
Week
FIG. 2. Mean (tSEM) body weights of hamsters in Experiment 2.
Hamsters were either sham-operated (A) or surgically castrated (B) on
week 0. Hamsters were then housed in either long (LD32/LD40; xLD32/
xLD40) or short (SD32/SD40; xSD32/xSD40) days for 32–40 weeks, or in
long days for 20 weeks and transferred to short days on week 20 (SD12;
xSD12). *Pf0.05 versus all other groups. †P=0.05 versus LD32/LD40
group.
and did not increase significantly thereafter (t=0.06, d.f.=9,
P>0.95; Fig. 3A). In contrast, hamsters transferred to short
days at week 20 underwent complete gonadal collapse over
the next 12 weeks (t=12.59, d.f.=15, P<0.0001; Fig. 3A).
Paired testis weights obtained at autopsy indicated that
hamsters housed in long days for 32 weeks retained fully
developed (>600 mg) testes (Fig. 3B); testis weights of photorefractory hamsters exposed to short days for 32 weeks did not
differ from those of long-day control values (P>0.10), indicating testicular recrudescence was complete (Fig. 3B). Moreover,
testis weights were comparable between hamsters housed in
short days for 32 and 40 weeks (P>0.65; Fig. 3B). Hamsters
housed in long days from weeks 0–20 and in short days
from weeks 20–32 (SD12 treatment) exhibited complete
gonadal regression at week 32; testis weights of SD12
hamsters were significantly lower than those of both long-day
control hamsters and photorefractory hamsters (P<0.0001,
both comparisons; Fig. 3B) characteristic of reproductive
responsiveness to short days.
#
40
0
40
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Paired testis weights (mg)
Body weight (g)
50
32
60
Week
(B)
55
20
80
20
Intact
0
100
20
Week
Week 32
32
40
Week 40
800
600
400
200
25
LD32
SD12
SD32
LD40
SD40
FIG. 3. (A) Mean (tSEM) testis volumes (expressed as a percentage of
week 0 testis volume) and (B) paired testis weights of hamsters in
Experiment 2. Hamsters were housed in either long or short days for
32–40 weeks and autopsied on week 32 (LD32, SD32) or week 40 (LD40,
SD40), and hamsters were housed in long days for 20 weeks, transferred to
short days on week 20, and autopsied on week 32 (SD12). *Pf0.05 versus
all other groups.
Immune cell phenotypes
Blood samples obtained at week 32 revealed significant
effects of castration on circulating lymphocyte phenotypes.
Irrespective of photoperiod treatments, castration significantly decreased the proportion of both monocytes (F=6.29,
d.f.=1,58, P<0.05) and granulocytes (F=31.2, d.f.=1,59,
P<0.0001) relative to intact hamsters, and significantly
increased the number of B-like lymphocytes (F=40.4,
d.f.=1,58, P<0.0001).
In intact hamsters, photoperiod treatments had no effect on
the proportions of circulating monocytes (F=0.69, d.f.=2,28,
P>0.50; Fig. 4A) but did significantly affect the proportions
of circulating granulocytes (F=11.3, d.f.=2,28, P<0.0005;
Fig. 4B), and B-like lymphocytes (F=11.1, d.f.=2,28,
P<0.0005; Fig. 4C). Twelve weeks of exposure to short days
was associated with a significant decrease in the proportion of
granulocytes relative to long-day values (P<0.01). Moreover,
after 32 weeks in short days, the proportion of granulocytes
no longer resembled acute-short-day values (P<0.0001)
and did not differ significantly from that of long-day
324 Recrudescence of immune function
Castrates
500
500
400
400
300
300
200
200
100
100
% Relative to control (xLD32)
180
LD32 SD12 SD32
xLD32 xSD12 xSDR32
Intacts
Castrates
% Relative to control (xLD32)
180
160
160
140
140
120
120
100
100
80
80
60
60
40
40
20
20
0
0
(C)
160
Lymphocyte proliferation
Intact hamsters: Spontaneous proliferation of splenic lymphocytes in vitro depended upon photoperiod treatments
(F=5.17, d.f.=2,33, P=0.01; Fig. 5A). Basal proliferation
was higher in cells from LD32 males as compared to agematched hamsters in short days for 12 weeks (SD12;
P<0.005). Reproductively photorefractory hamsters in the
SD32 group had intermediate blastogenesis by lymphocytes
which did not differ from that of LD32 or SD12 hamsters
(P>0.10, both comparisons; Fig. 5A).
Because some traits of photorefractory hamsters require
>32 weeks to fully revert to the long-day phenotype, we also
assessed spontaneous proliferation 8 weeks later (i.e. at week
40) in a subset of hamsters that had been exposed to short
days continuously from week 0 (SD40). In these hamsters,
basal lymphocyte proliferation was significantly elevated
compared to that of the SD12 group (P<0.0001) and also
significantly exceeded values of SD32 and LD40 hamsters
(P<0.01 and P<0.05, respectively; Fig. 5A). To account for
putative effects of increased age on spontaneous proliferation
of lymphocytes in short days, lymphocyte proliferation values
were calculated relative to age-matched long-day control
hamsters (i.e. relative to LD32 at week 32 and relative to
LD40 at week 40); this permitted direct comparisons of proliferation rates among short-day groups that differed in age.
Expressed as a percentage of long-day control values, lymphocyte proliferation was significantly reduced after 12 weeks of
exposure to short days (P<0.01 versus LD32), intermediate
after 32 weeks in short days (P<0.15 versus LD32, P>0.20
versus SD12), and significantly elevated by week 40 (P<0.01
versus LD40, P<0.0001 versus SD12, P<0.0001 versus SD32;
Fig. 5C). Proliferation values after 40 weeks of exposure
to short days were significantly higher than those obtained
after either 12 or 32 weeks in short days (P<0.05, both
comparisons; Fig. 5C).
Castrated hamsters: In castrated hamsters, the spontaneous
proliferation of splenic lymphocytes in vitro differed in response to short day treatments (F=3.30, d.f.=2,34, P<0.05;
Fig. 5B). At week 32, spontaneous proliferation of lymphocytes was greater in cells extracted from the xLD32 group as
compared to xSD12 hamsters (P<0.05). After 32 weeks in
short days, reproductively photorefractory xSD32 hamsters
also exhibited reduced proliferation values relative to xLD32
hamsters (P<0.05); these values were comparable to those
of xSD12 hamsters (P>0.95; Fig. 5B). As with intacts, a subset
of castrated hamsters was kept in short days until week 40
(xSD40) along with a group of long-day controls (xLD40). At
week 40, proliferative values of castrated refractory hamsters
were still significantly suppressed relative to long-day values
0
0
(B)
photorefractory, photosensitive, and long-day hamsters in the
relative proportion of monocytes on week 32 (P>0.25, all
comparisons). In castrated hamsters, pairwise comparisons of
samples obtained at week 32 did not yield significant effects of
photoperiod treatments on mean values of any of the three
lymphocyte populations (monocytes: F=0.83, d.f.=2,26,
P>0.40, Fig. 4A; granulocytes: F=1.83, d.f.=2,26, P>0.15,
Fig. 4B; B-like lymphocytes: F=2.01, d.f.=2,26, P>0.15;
Fig. 4C).
600
% Relative to control (LD32)
Intacts
% Relative to control (LD32)
600
LD32 SD12 SD32
xLD32 xSD12 xSDR32
Intacts
Castrates
160
140
140
120
120
100
100
80
80
60
60
40
40
20
20
% Relative to control (LD32)
% Relative to control (xLD32)
(A)
0
0
LD32 SD12 SD32
xLD32 xSD12 xSDR32
FIG. 4. Mean (tSEM) (A) circulating monocytes, (B) granulocytes and (C)
B-like lymphocytes in intact and castrated (x) hamsters in Experiment 2.
All measures were obtained on week 32. Group abbreviations as in Fig. 2.
Data are presented as a percentage of long-day control values (xLD32 for
intact hamsters; LD32 for castrated hamsters) to normalize variance.
*Pf0.05 versus all other photperiod treatment groups within surgical
treatment.
hamsters (P>0.10). The proportion of B-like lymphocytes
was significantly elevated after exposure to short days
for 12 weeks (P=0.001), and also returned to long-day
values after 32 weeks in short days (P>0.45 versus longday control value). No significant differences existed among
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Recrudescence of immune function 325
(A)
(B)
0.8
Intact
0.7
Absorbance units (OD)
Absorbance units (OD)
0.8
0.6
0.5
0.4
0.3
Castrated
0.7
0.6
0.5
0.4
0.3
LD32
SD12
SD32
LD40
SD40
xLD32
(C)
xSD32
xLD40
32
40
xSD40
(D )
Intact
30
% Control (xLD) value
% Control (LD) value
30
xSD12
20
10
0
–10
–20
Castrated
20
10
0
–10
–20
–30
–30
12
32
40
12
Fig. 5. (A,B) Mean (tSEM) basal (0 mg/ml ConA) in vitro proliferation (expressed as absorbance units at 490 nm) of splenic lymphocytes from intact
(A) and castrated (B) hamsters in Experiment 2. (C,D) Mean (tSEM) in vitro lymphocyte proliferation in intact (C) and castrated (D) hamsters after 12, 32,
and 40 weeks of exposure to short days. Data in (C) and (D) are expressed as a percentage of age-appropriate long-day control values (LD or xLD) to
permit comparison of proliferative responses obtained at different time points in the study. Group abbreviations as in Fig. 2. †P=0.05 versus long-day
control value on a given week. ‡Pf0.05 versus week 40 value. *Pf0.05 versus xLD32.
(P<0.05). However, age appeared to significantly affect basal
lymphocyte proliferation, as proliferative values of castrated
long-day control hamsters were significantly higher at week 40
than at week 32 (P=0.05; Fig. 5B). Thus, in vitro lymphocyte
proliferation values were calculated relative to age-matched
long-day control values. As with intact hamsters, when
expressed as a percentage of age-specific long-day control
values, lymphocyte proliferation in castrated hamsters was
significantly reduced after 12 weeks of exposure to short days
(P<0.05 versus xLD32); however, basal proliferation in
castrated hamsters remained significantly reduced after 32
(P<0.05 versus xLD32, P>0.95 versus xSD12; Fig. 6D) and
after 40 (P<0.05 versus xLD40, P>0.80 versus xSD12,
P>0.80 versus xSD32; Fig. 5D) weeks in short days. In
castrated hamsters, proliferation values after 40 weeks of
exposure to short days did not differ from those obtained after
12 or 32 weeks in short days (P>0.50, both comparisons;
Fig. 5D).
Experiment 3: Lymphocyte responses to melatonin
Because basal lymphocyte proliferation differed significantly
among photoperiod treatment groups, proliferative responses
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
to melatonin were expressed as a percentage of basal (0 pg/ml
melatonin) values. In both long-day control hamsters and
photorefractory hamsters, proliferative responses to in vitro
melatonin treatments did not differ between lymphocytes
harvested at weeks 32 and 40 (long-day controls: F=1.2,
d.f.=3,36, P>0.30; photorefractory hamsters: F=0.79,
d.f.=3,54, P>0.50); data from LD32 and LD40 hamsters
were therefore pooled, as were data from SD32 and SD40,
for subsequent analyses. Similarly, in castrated hamsters, no
differences were observed between week 32 and 40 proliferative responses to melatonin (castrated long-day control hamsters: F=1.67, d.f.=3,42, P>0.15; castrated photorefractory
hamsters: F=0.32, d.f.=3,51, P>0.80). Thus, data on proliferative responses to melatonin from xLD32 and xLD40
were pooled, as were data from xSD32 and xSD40 hamsters.
Intact hamsters
Lymphocyte responses to in vitro incubation with 50 pg/ml
melatonin depended on photoperiod treatments (F=5.97,
d.f.=2,47, P<0.005; Fig. 6A). Lymphocytes of SD12 hamsters
exhibited higher proliferative responses to in vitro melatonin
relative to both LD32/40 hamsters (P<0.005) and SD32/40
hamsters (P<0.005; Fig. 6A). Melatonin (50 pg/ml) treatment
326 Recrudescence of immune function
(A)
120
to that obtained at lower melatonin concentrations: the
response to melatonin of lymphocytes from photorefractory
(SD32/40) hamsters differed significantly from that of SD12
hamsters (P<0.05), and SD12 values tended to differ from
those of long-day control (LD32/40) hamsters (P=0.07).
LD32/LD40
SD12
SD32/SD40
Intact
% Basal proliferation
110
Castrated hamsters
Photoperiod treatments did not significantly affect lymphocyte proliferation in the presence of melatonin at any
concentration used (50 pg/ml: F=0.50, d.f.=2,48, P>0.60;
500 pg/ml: F=0.56, d.f.=2,48, P>0.55; 1000 pg/ml: F=0.40,
d.f.=2,48, P>0.65; Fig. 6B).
100
90
80
Discussion
These experiments addressed the control of seasonal changes
in immune function by photoperiod and photorefractoriness.
In Experiment 1, 3 weeks of exposure to long days did not
enhance cell-mediated immune function (spontaneous in vitro
splenic lymphocyte proliferation) in short-day-adapted male
Siberian hamsters. These data indicate that photostimulation
of immune function is not an instantaneous phenomenon, but
rather, in common with seasonal transitions in reproductive
function (21), may be a lengthy process that requires weeks-tomonths to occur. Given such a latency to stimulation by long
days, we hypothesized that seasonal transitions in immune
function may be governed in part by refractoriness mechanisms analogous to those that regulate the lengthy seasonal
transitions in reproductive function (19). To test this hypothesis, in Experiment 2 male hamsters were exposed to short
days for 12 weeks (an interval sufficient to induce shortday adjustments in immune function) (12, 14, 15) or for
32–40 weeks (an interval sufficient to permit reproductive
photorefractoriness to manifest). Intact hamsters exposed to
short days for 12 weeks exhibited a decrease in spontaneous
in vitro splenic lymphocyte proliferation, a reduction in
the number of circulating granulocytes, and an increase in the
number of circulating B-like lymphocytes. In contrast,
hamsters exposed to short days for 32–40 weeks failed to
exhibit these short-day changes in splenic lymphocyte
proliferation and circulating lymphocyte phenotypes. The
experimental design assessed immune function in age-matched
controls (i.e. SD12 hamsters had been in short days for
12 weeks at a time when SD32 hamsters had been in short
days for 32 weeks), thus permitting the conclusion that
changes in immune function observed in SD32 hamsters were
not a result of age per se, or interactions between age and
photoperiod, but rather were a consequence of the extended
duration of exposure to short days. These data confirm
previous reports (12–15) that short days affect several aspects
of immune function in this species, but indicate that these
effects are not permanent: after 32–40 weeks, short-day
induced changes in immune function exhibit a ‘spontaneous’
return to long-day values. These data are consistent with
the hypothesis that, along with reproductive function, select
aspects of immune function become refractory to short
days (23).
The present data extend the conclusions of a related study
(23) which addressed the issue of immunological photorefractoriness in deer mice (Peromyscus maniculatus), a species
70
60
50
500
1000
[Melatonin] pg/ml
(B)
120
xLD32/xLD40
xSD12
xSD32/xSD40
Castrated
% Basal proliferation
110
100
90
80
70
60
50
500
1000
[Melatonin] pg/ml
FIG. 6. Mean (tSEM) in vitro proliferation of splenic lymphocytes from
intact (A) and castrated (B) hamsters in Experiment 2. Lymphocytes were
incubated for 48 h in the presence of 0, 50, 500 or 1000 pg/ml melatonin.
Proliferative responses to melatonin are expressed as a percentage of basal
(0 pg/ml melatonin) proliferation. Data from week 32 and week 40 longday controls were combined, as were data from week 32 and week 40
photorefractory hamsters (see Results). Group abbreviations as in Fig. 2.
*Pf0.05 versus all other groups. †Pf0.05 versus SD32/SD40 group.
had a comparable effect on proliferation of LD32/40 and
SD32/40 lymphocytes (P=0.80). A similar pattern of responses was observed in lymphocytes treated with 500 pg/ml
melatonin (F=3.23, d.f.=2,48, P<0.05; Fig. 6A). At a concentration of 1000 pg/ml, melatonin treatments tended to
affect changes in lymphocyte proliferation (F=2.71, d.f.=
2,48, P=0.08; Fig. 6A). The pattern of changes in basal
proliferation in response to 1000 pg/ml melatonin was similar
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Recrudescence of immune function 327
that exhibits enhanced mitogen-stimulated proliferation of
splenic lymphocytes after 10–12 weeks of exposure to
short days (9). Long-term (32 weeks) exposure of male
deer mice to short days restored the long-day (immunocompromised) phenotype, suggesting that a ‘spontaneous
regression’ of this measure of immune function had occurred
(23). Photoperiodic control of immune function in deer mice
occurs independent of gonadal hormone secretion (33), implying that refractoriness-associated changes in immune function
in this species were not simply a consequence of gonadal
recrudescence (23). However, gonadal hormones have significant immunomodulatory effects in Siberian hamsters (16),
and the present data suggest that gonadal hormone secretion
figures prominently in the expression of immunological photorefractoriness in this species. In castrated hamsters, short days
likewise inhibited basal in vitro lymphocyte proliferation, but
failed to significantly alter numbers of circulating granulocytes and B-like lymphocytes, indicating that changes in
gonadal hormone secretion are not required for short days to
inhibit lymphocyte proliferation, but are necessary for day
length to affect circulating leucocyte phenotypes. Gonadal
hormones are not without effect on lymphocyte proliferation in this species (16), but the present data suggest that, in
common with deer mice (33), photoperiodic modulation of
lymphocyte proliferation can occur in the absence of the
gonads. In contrast to intact animals, however, spontaneous
in vitro lymphocyte proliferation in castrated hamsters was
not restored to long-day-like values after prolonged (32 or
40 weeks) exposure to short days. Short days thus inhibit
lymphocyte proliferation independent of the inhibitory effects
of short days on gonadal hormone secretion, but this gonadal
inhibition of immune function persists at a time when immune
function in intact hamsters has returned to the long-day
phenotype (i.e. undergone recrudescence). Together, these
observations suggest that, in castrated hamsters, this measure
of immune function does not become refractory to short days.
Moreover, recrudescence of lymphocyte proliferation in intact
but not in castrated hamsters suggests that spontaneous
recrudescence of the gonads is necessary for spontaneous
recrudescence of this measure of immune function to occur.
The observed recrudescence of several measures of immune
function in intact hamsters appears analogous to the spontaneous recrudescence observed in the reproductive system of
this, and other, photoperiodic species (34). Whereas the neural
mechanism that mediates spontaneous recrudescence of the
reproductive axis remains essentially unknown (6, 19), data
from the present study permit limited insight into the mechanism by which ‘spontaneous’ recrudescence of immune function may occur. Earlier work in Siberian hamsters indicated
that in vitro lymphocyte proliferation is inhibited by castration and enhanced by in vivo testosterone (in males) or
oestradiol (in females) treatment (16). The present data
suggest that there exist both gonadal hormone-dependent
and gonadal hormone-independent effects of day length on
lymphocyte proliferation. In the intact hamster, photoperioddriven changes in gonadal hormone secretion may be sufficient, but do not appear necessary, to induce short-day-like
lymphoproliferative activity. With prolonged exposure to
inhibitory photoperiods, the brain becomes refractory to short
days, initiating recrudescence of the gonads and a gradual
#
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
restoration of elevated gonadal hormone secretion (35). Vernal
recrudescence of gonadal hormone secretion may, in turn,
be necessary to restore long-day-like lymphocyte function.
Seasonal changes in lymphocyte proliferative capacity thus
may be driven, or masked, by seasonal changes in gonadal
hormone secretion, ensuring synchrony of seasonal transitions
in both reproductive and immune function.
Lymphocytes from photoperiodic mammals provide a model
system in which to test whether changes in responsiveness to
melatonin are associated with the induction of refractoriness
to short days. Mammalian lymphocytes possess functional
melatonin receptors (36–39), and in vitro melatonin treatment
enhances or inhibits lymphocyte proliferation, depending on
species and sex (14, 15, 40). Consistent with previous work
in photoperiodic rodents, in vitro treatment of lymphocytes
with physiological concentrations of melatonin (50 pg/ml and
500 pg/ml) enhanced basal proliferation of SD12 lymphocytes
relative to long-day controls. The functional significance of
this enhancement is unclear, but perhaps, in vivo, the expanded
nocturnal duration of circulating melatonin may enhance the
otherwise inhibited basal rate of lymphocyte proliferation
in short days, effectively permitting maintenance of elevated
immune function in the short days of winter. Regardless of
its physiological significance, the observed photic modulation
of lymphocyte responsiveness to melatonin yields a model
system with which to test the hypothesis that photorefractoriness of immune function is associated with a change in
responsiveness of lymphocytes to melatonin. The results of
Experiment 3 indicated that, in contrast to SD12 lymphocytes, proliferation of lymphocytes extracted from chronic
(32–40 week) short-day hamsters was not enhanced by
treatment with melatonin but, instead, was comparable to
that of long-day control animals. Thus, long-term exposure to
short days is associated with a loss of responsiveness of
lymphocytes to melatonin. In the reproductive axis, induction
of photorefractoriness is a result of changes in responsiveness
of thalamic and hypothalamic melatonin target tissues to
inhibitory patterns of melatonin secretion (6). The loss of
responsiveness to melatonin in lymphocytes of photorefractory hamsters suggests formal similarities between melatoninresponsive cells of the immune system which mediate host
defence, and those in the brain which mediate reproductive
responses (5, 6, 41).
Formal similarities notwithstanding, these data do not
permit the conclusion that lymphocytes per se become
refractory to melatonin. Indeed, photoperiod treatments
were completely without effect on lymphocyte responses to
melatonin in castrated hamsters. Castrated hamsters did,
however, exhibit changes in body weight and fur colour (data
not shown) after exposure to short days, and these traits
became refractory to short days in the absence of gonadal
hormone secretion. Photoperiodic regulation of body weight
and fur colour is pineal-dependent (42, 43), and neural
mechanisms that regulate these traits become refractory to
melatonin (6) independent of gonadal hormone secretion. The
absence of both photoperiodic and refractoriness-associated
differences in lymphocyte responsiveness to melatonin in
castrated hamsters implies that ancillary factors, likely
photoperiod-driven changes in gonadal hormone secretion, are
required to mediate such effects. Taken together, the present
328 Recrudescence of immune function
7
data suggest that the loss of lymphocyte responsiveness to
melatonin in intact hamsters after prolonged exposure to
short days is a by-product of reproductive photorefractoriness. Seasonal changes in gonadal hormone secretion
may activate mechanisms which in turn permit differential
responsiveness to melatonin depending on the phase in the
annual reproductive cycle.
In summary, 12 weeks of exposure to short days induced
changes in several aspects of immune function of male
Siberian hamsters. Gonadal hormone-dependent, short-dayinduced decreases in circulating granulocytes and increases
in circulating B-like lymphocytes spontaneously returned
to long-day values after 32 weeks in short days. Gonadal
hormone-independent, short-day-induced decreases in spontaneous proliferation of splenic lymphocytes in vitro were
restored to long-day levels after 32–40 weeks of exposure to
short days in intact but not in castrated hamsters, suggesting
that refractoriness of the reproductive system to short days
drives spontaneous recrudescence of immune function.
In vitro melatonin treatments enhanced basal proliferation
of lymphocytes from gonad-intact hamsters exposed to short
days for 12 weeks, but melatonin treatments were without
effect on lymphocytes from photorefractory hamsters or longday hamsters, and were ineffective in castrated hamsters
regardless of photoperiod history. In common with neural
melatonin targets that regulate the reproductive axis, splenic
lymphocytes become unresponsive to melatonin after prolonged exposure to short days; however, photoperiod- and
refractoriness-induced changes in gonadal hormone secretion
may ultimately dictate this change in responsiveness to
melatonin.
8
9
10
11
12
13
14
15
16
17
18
19
Acknowledgements
20
We thank Judy Holbeck, Anna Colacino, Long Tran, Brian Spar,
P.J. Ianaconi, Jim Eisner and Inga Gurevich for technical assistance and
animal care, and Staci Bilbo, Dr David Freeman, Dr Andrew Hotchkiss
and Dr Irving Zucker for advice on the manuscript. This investigation was
supported by National Institutes of Health, National Research Service Award
12875 from the National Institutes of Mental Health, NIH grants MH 57535
and NS 40254, National Science Foundation IBN 00–0854, and the Natural
Sciences and Engineering Research Council of Canada.
21
22
23
Accepted 8 January 2002
24
25
References
1
2
3
4
5
6
26
Bronson FH. Mammalian Reproductive Biology. Chicago: University of
Chicago Press, 1989.
Saunders DS. An Introduction to Biological Rhythms. Glasgow: Blackie,
1977.
Illnerova H, Hoffmann K, Vanecek J. Adjustment of pineal melatonin
and N-acetyltransferase rhythms to change from long to short photoperiod in the Djungarian hamster Phodopus sungorus. Neuroendocrinology
1984; 38: 226–231.
Illnerova H, Sumova A. Photic entrainment of the mammalian rhythm in
melatonin production. J Biol Rhythms 1997; 12: 547–555.
Badura LL, Goldman BD. Central sites mediating reproductive
responses to melatonin in juvenile male Siberian hamsters. Brain Res
1992; 598: 98–106.
Freeman DA, Zucker I. Refractoriness to melatonin occurs independently at multiple brain sites in Siberian hamsters. Proc Natl Acad Sci
USA 2001; 98: 6447–6452.
27
28
29
30
#
Bartness TJ, Powers JB, Hastings MH, Bittman EL, Goldman BD. The
timed infusion paradigm for melatonin delivery: what has it taught us
about the melatonin signal, its reception, and the photoperiodic control
of seasonal responses? J Pineal Res 1993; 15: 161–190.
Goldman BD. Mammalian photoperiodic system: formal properties
and neuroendocrine mechanisms of photoperiodic time measurement.
J Biol Rhythms 2001; 16: 283–301.
Nelson RJ, Demas GE. Seasonal changes in immune function. Q Rev Biol
1996; 71: 511–548.
Nelson RJ, Demas GE, Klein SL, Kriegsfeld LJ. Seasonal Patterns of
Stress, Immune Function, and Disease Processes. New York: Cambridge
University Press, 2002.
Nelson RJ, Demas GE, Klein SL, Kriegsfeld LJ. The influence of season,
photoperiod, and pineal melatonin on immune function. J Pineal Res
1995; 19: 149–165.
Yellon SM, Fagoaga OR, Nehlsen-Cannarella SL. Influence of
photoperiod on immune cell functions in the male Siberian hamster.
Am J Physiol 1999; 276: R97–R102.
Yellon SM, Teasley LA, Fagoaga OR, Nguyen HC, Truong HN,
Nehlsen-Cannarella SL. Role of photoperiod and the pineal gland in
T cell-dependent humoral immune reactivity in the Siberian hamster.
J Pineal Res 1999; 27: 243–248.
Drazen DL, Kriegsfeld LJ, Schneider JN, Nelson RJ. Leptin, but not
immune function, is linked to reproductive responsiveness to photoperiod. Am J Physiol 2000; 278: 1401–1407.
Prendergast BJ, Yellon SM, Tran LT, Nelson RJ. Photoperiod
modulates the effect of in vitro melatonin on lymphocyte proliferation
in female Siberian hamsters. J Biol Rhythms 2001; 16: 224–233.
Bilbo SD, Nelson RJ. Sex steroid hormones enhance immune function in male and female Siberian hamsters. Am J Physiol 2001; 280:
R207–R213.
Hoffmann K. Effect of short photoperiods on puberty, growth and
moult in the Djungarian hamster (Phodopus sungorus). J Reprod Fertil
1978; 54: 29–35.
Gorman MR, Zucker I. Seasonal adaptations of Siberian hamsters. II.
Pattern of change in day length controls annual testicular and body
weight rhythms. Biol Reprod 1995; 53: 116–125.
Watson-Whitmyre K, Stetson MH. Reproductive refractoriness in
hamsters: environmental and endocrine etiologies. In: Stetson MH, ed.
Processing of Environmental Information in Vertebrates. New York:
Springer-Verlag, 1988: 203–218.
Stetson MH, Watson-Whitmyre M, Matt KS. Termination of photorefractoriness in golden hamsters – photoperiodic requirements. J Exp
Zool 1977; 202: 81–88.
Simpson SM, Follett BK, Ellis DH. Modulation by photoperiod of
gonadotrophin secretion in intact and castrated Djungarian hamsters.
J Reprod Fertil 1982; 66: 243–250.
Clarke JR. Physiological problems of seasonal breeding in eutherian
mammals. Oxford Rev Reprod Biol 1981; 3: 244–312.
Prendergast BJ, Nelson RJ. Spontaneous regression of immune function
in a photoperiodic rodent, Peromyscus maniculatus. Proc Roy Soc Lond
Series B Biol Sci 2001; 268: 1–8.
Duncan MJ, Goldman BD. Hormonal regulation of the annual pelage
color cycle in the Djungarian hamster, Phodopus sungorus. II. Role of
prolactin. J Exp Zool 1984; 230: 97–103.
Herbert TB, Cohen S. Stress and immunity in humans: a meta-analytic
review. Psychosom Med 1993; 55: 364–379.
Janeway CA, Travers P, Walport M, Capra JD. Immunobiology. New
York: Garland Publishing, 1999.
Azcona-Olivera JI, Ouyang YL, Warner RL, Linz JE, Pestka JJ. Effects
of vomitoxin (deoxynivalenol) and cyclohexamide on IL-2, 4, 5 and 6
secretion and mRNA levels in murine CD4+ cells. Food Chem Toxicol
1995; 33: 433–441.
Demas GE, Nelson RJ. Exogenous melatonin enhances cell-mediated,
but not humoral, immune function in adult male deer mice (Peromyscus
maniculatus). J Biol Rhythms 1998; 13: 245–252.
Cory AH, Owen TC, Barltrop JA, Cory JG. Use of an aqueous soluble
tetrazolium/formazan assay for cell growth assays in culture. Cancer
Commun 1991; 3: 207–212.
Kitamura T, Tange T, Terasawa T, Chiba S, Kuwaki T, Miyagawa K,
Piao YF, Miyazono K, Urabe A, Takaku F. Establishment and
characterization of a unique human cell line that proliferates dependently
on GM-CSF, IL-3, or erythropoietin. J Cell Physiol 1989; 140: 323.
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
Recrudescence of immune function 329
32
33
31
34
35
36
#
Nieman DC, Fagoaga OR, Butterworth DE, Warren BJ, Utter A,
Davis JM, Henson DA, Nehlsen-Cannarella SL. Carbohydrate
supplementation affects blood granulocyte and monocyte trafficking but not function after 2.5 h of running. Am J Clin Nutr 1997;
66: 153–159.
Demas GE, Nelson RJ. Short-day enhancement of immune function is
independent of steroid hormones in deer mice (Peromyscus maniculatus).
J Comp Physiol [B] 1998; 168: 419–426.
Gieni RS, Li Y, HayGlass KT. Comparison of [3H]thymidine
incorporation with MTT- and MTS-based bioassays for human and
murine IL-2 and IL-4 analysis. Tetrazolium assays provide markedly
enhanced sensitivity. J Immunol Meth 1995; 187: 85–93.
Prendergast BJ, Nelson RJ, Zucker I. Mammalian seasonal rhythms:
behavior and neuroendocrine substrates. In: Pfaff D, Arnold A,
Fahrbach S, Rubin R, ed. Hormones, Brain and Behavior. San Diego:
Academic Press, 2002.
Matt KS, Stetson MH. Hypothalamic–pituitary–gonadal interactions
during spontaneous testicular recrudescence in golden hamsters. Biol
Reprod 1979; 20: 739–746.
Calvo JR, Rafii-el-Idrissi M, Pozo D, Guerrero JM. Immunomodulatory
role of melatonin: specific binding sites in human and rodent lymphoid
cells. J Pineal Res 1995; 18: 119–126.
2002 Blackwell Science Ltd, Journal of Neuroendocrinology, 14, 318–329
37
38
39
40
41
42
43
Pozo D, Delgado M, Fernandez-Santos JM, Calvo JR, Gomariz RP,
Martin-Lacave I, Ortiz GG, Guerrero JM. Expression of the
Mel1a-melatonin receptor mRNA in T and B subsets of lymphocytes
from rat thymus and spleen. FASEB J 1997; 11: 466–473.
Barjavel MJ, Mamdouh Z, Raghbate N, Bakouche O. Differential
expression of the melatonin receptor in human monocytes. J Immunol
1998; 160: 1191–1197.
Konakchieva R, Manchev S, Pevet P, Masson-Pevet M. Autoradiographic detection of 2-(125I)-iodomelatonin binding sites in immune tissue
of rats. Adv Exp Med Biol 1999; 460: 411–415.
Kriegsfeld LJ, Drazen DL, Nelson RJ. In vitro melatonin treatment
enhances cell-mediated immune function in male prairie voles (Microtus
ochrogaster). J Pineal Res 2001; 30: 193–198.
Bittman EL. Hamster refractoriness: the role of insensitivity of pineal
target tissues. Science 1978; 202: 648–650.
Bartness TJ, Goldman BD. Effects of melatonin on long-day responses
in short-day housed adult Siberian hamsters. Am J Physiol 1988; 255:
R823–R830.
Badura LL, Goldman BD. Prolactin-dependent seasonal changes
in pelage: role of the pineal gland and dopamine. J Exp Zool 1992;
261: 27–33.