| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Biology, Center for Neuroendocrine Studies, and Program in Neuroscience and Behavior, University of Massachusetts (E.L.M.-B., A.E.J., S.-i.M., J.F.M.), Amherst, Massachusetts 01003; the Department of Anatomy and Cell Biology, University of Cincinnati Medical School (M.N.L.), Cincinnati, Ohio 45267
Address all correspondence and requests for reprints to: Eric L. Bittman, Ph.D., Department of Biology and Program in Neuroscience and Behavior, University of Massachusetts, Amherst, Massachusetts 01003. E-mail: elb{at}bio.umass.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-sulfatoxymelatonin, were
not reestablished by behaviorally functional grafts. In Exp 2, SCN
lesions eliminated locomotor rhythmicity in adult male hamsters
maintained in DD. Seven to 12 weeks after restoration of locomotor
activity rhythms by fetal grafts, hosts and sham-lesioned controls were
decapitated at circadian times 4, 8, 12, 16, 20, or 24. Clear
circadian rhythms of both serum corticosterone and cortisol were seen
in sham-lesioned males, with peaks in late subjective day. No circadian
rhythms in either adrenal hormone were evident in serum from
lesioned-grafted males. Testicular regression, observed in intact and
sham-lesioned males maintained in DD, was absent not only in arrhythmic
SCN-lesioned hamsters given grafts of cerebral cortex, but also in
animals in which hypothalamic grafts had reinstated locomotor
rhythmicity. The pineal melatonin concentration rose sharply during the
late subjective night in control hamsters, but not in SCN-lesioned
animals bearing behaviorally effective transplants. Even though circadian rhythms of locomotor activity are restored by SCN transplants, circadian endocrine rhythms are not reestablished. Endocrine rhythms may require qualitatively different or more extensive SCN outputs than those established by fetal grafts.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Transplantation of the SCN can reinstate behavioral circadian
rhythms in rats and hamsters previously made arrhythmic by SCN lesions
(19, 20, 21, 22, 23, 24, 25). It is not known whether such grafts can also reinstate
circadian rhythms of neuroendocrine function in SCN-ablated animals.
The present studies were undertaken to determine whether SCN
transplants adequate to reinstate locomotor rhythms in SCN-lesioned
hamsters also restore estrous cycles, E2-induced LH surges,
or circadian rhythms of serum glucocorticoids, pineal melatonin
content, or the urinary melatonin metabolite, 6-sulfatoxymelatonin
(6MTS).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Surgery
Lesions. Hamsters were anesthetized with sodium
pentobarbital (65 mg/kg BW) and placed in a Kopf stereotaxic apparatus
(Kopf Instruments, Tujunga, CA). Bilateral electrolytic lesions of the
SCN were made using a tungsten electrode (0.5-mm diameter; A & M
Systems, Everett, WA), insulated with Epoxylite resin except at the
tip. Each lesion was made by passing a direct positive current (1.75
mA) for 20 sec using a Grass LM-5 lesionmaker (Grass Instrument Co.,
Quincy, MA). The electrode was positioned 0.35 mm on either side of the
midline, 0.8 mm rostral to bregma, and 7.8 mm ventral to the dura.
Grafting. Animals displaying an absence of circadian
rhythmicity for a period of at least 6 weeks of constant darkness (DD)
were selected as transplant recipients. Fetal tissue containing the SCN
or cerebral cortex (as a control) was obtained from embryonic day 15
hamster pups (two per host) and transplanted into the third ventricle
as previously described (22, 25). A drop of trypan blue was placed on
the chiasmatic region to facilitate visualization of the SCN area
during dissection and preparation of the tissue to be grafted. The
injection cannula (Wiretrol, Fisher Lab Products, Pittsburgh, PA)
containing the graft tissue (
5 µl) was lowered at the midline to
the same coordinates as the lesion. As the injection cannula was
raised, the graft was slowly ejected.
RIAs
After collection from males, whole blood was allowed to clot at
4 C for several hours before harvesting of serum by centrifugation at
1000 x g. Serum was stored at -20 C until RIAs were
performed. Cortisol and corticosterone concentrations were measured in
male hamsters using kits from ICN Biomedicals, Inc. (Costa
Mesa, CA). Serum aliquots were heated to 98 C for 10 min and allowed to
cool before addition of antibody and trace. Serial dilutions of
extracted hamster serum generated a slope that was parallel to
standard, and serum from adrenalectomized hamsters had undetectable
concentrations of immunoreactive cortisol and corticosterone. All
samples were run in a single RIA for each hormone; the intraassay
coefficients of variation (CVs) were 7.9% for cortisol and 7.4% for
corticosterone.
Pineal glands were sonicated, and melatonin was extracted using
methylene chloride as described previously (26, 27). Pineal and serum
samples were assayed for melatonin using previously described methods
(27). The mean intraassay CV (six assays) was 9%, and the interassay
CV was 14.9% at 50% binding. Urinary fractions were frozen at -20 C
until 6MTS was assayed by Stockgrand (Surrey, UK) using the method
of Arendt et al. (28).
LH was assayed in blood plasma collected from female hamsters, using a rat kit from the National Hormone and Pituitary Program. Reference preparation RP-3 was used as standard, NIDDK rLH I-9 (AFP-10250C) was used for iodination, and NIDDK S-10 anti-LH was used at a dilution of 1:200,000. Bound LH was precipitated using goat antirabbit IgG (1:20; Antibodies, Inc., Davis, CA). Standard and hamster serum produced parallel inhibition curves in this assay system. The interassay coefficient of variation was 8.4%, and the intraassay coefficient of variation was 7.0% (five assays).
Exp 1: female hamsters
Figure 1
illustrates the
procedures used in this experiment. To assess ovarian cyclicity, all
animals were inspected daily for a vaginal discharge 2–6 h after light
onset during all phases of the experiment in which the hamsters were
housed in a light-dark (LD) cycle (29). Hamsters exhibiting a minimum
of three regular estrous cycles (n = 47) received bilateral SCN
lesions. After 6–8 weeks in 14L:10D, hamsters were transferred to DD
for 3 weeks to confirm lesion efficacy. Hamsters whose behavior
remained rhythmic despite hypothalamic lesions were discarded.
Subsequent histology confirmed that complete destruction of the SCN
resulted in arrhythmicity (see below). Animals that continued to show
an absence of circadian rhythmicity of locomotor activity in DD (n
= 34) were returned to 14L:10D and grafted with fetal tissue.
Immediately before grafting, seven of these hamsters were
ovariectomized to rule out the possibility that the fetal grafts might
be exposed to gonadal steroid hormones that might influence their
subsequent capacity to sustain LH surges. Hamsters recovered for 2
weeks before they were returned to DD. The recovery of locomotor
rhythms was assessed over the ensuing 8–13 weeks. Hamsters were
returned to 14L:10D for the remainder of the experiment, and those that
had not been ovariectomized were inspected daily for vaginal discharge
to assess estrous cyclicity. Four weeks later, the remaining recipients
of SCN grafts were ovariectomized. An additional 12 hamsters were
ovariectomized at this point to serve as brain-intact controls.
|
MTS while housed in
14L:10D. Hourly urine samples were collected from female hamsters
(n = 45) housed in metabolic cages (Nalgene, Rochester, NY) for
48–72 h. Urine samples were taken from 13 experimental animals both
after the lesion was made and after the graft had reinstated locomotor
activity rhythms. In 4 additional cases, samples were collected both
after SCN lesions and after transplantation of fetal hypothalamic
tissue that failed to restore locomotor rhythms. The interval between
the production of complete SCN lesions and the onset of urine
collection ranged from 10 days to 14 weeks, and the interval between
neurotransplantation and collection of urine samples ranged between
9–13 weeks. In addition, 48–72 h of urine samples were collected from
8 SCN-lesioned hamsters between 4–14 weeks after surgery; these
females either received no fetal grafts or recovered behavioral
rhythmicity after transplantation but died before they were due to be
returned to the LD cycle. Urine samples were also collected from 10
intact control females. As animals did not urinate every hour,
fractions were pooled in 3-h bins for statistical analysis.
Blood collection. Chronically ovariectomized females were
anesthetized for implantation of a SILASTIC brand jugular cannula
(Dow Corning Corp., Midland, MI) to permit blood sampling.
On the following day, they were anesthetized with methoxyflurane
(Metofane, Mallinckrodt Veterinary, Inc., Mendelein, IL)
and implanted sc with three SILASTIC capsules (Dow Corning Corp.; 5-mm length, 1.45-mm id, 1.93-mm od) containing 1%
crystalline E2 mixed in cholesterol (Sigma Chemical Co., St. Louis, MO). Such capsules provide E2
concentrations comparable to those found at noon on the day of
proestrus (100 pg/ml serum; Jetton, A. E., Meyer, and
E. L. Bittman, unpublished data). Approximately 48 h after
implantation of E2 capsules, hamsters were transferred to
constant dim illumination (50 lux), and blood sampling was begun. In
6 grafted and 12 unlesioned control animals, blood samples (0.5 ml)
were collected every 2–3 h for a 24- to 36-h period. In 1 additional
control and 5 additional grafted hamsters, the cannulas became
occluded, but 2–3 samples were withdrawn by cardiac puncture under
light methoxyflurane anesthesia.
Blood cells were separated from plasma by centrifugation (1000 x g) at 4 C. Plasma was frozen at -20 C until assay. Cells were mixed with a corresponding volume of solution containing 0.015% human plasma substitute (Plasmanate, Cutter Laboratories, Berkeley, CA), 10 U heparin, and 0.5 mg dextrose/ml and slowly returned to the animal. Hamsters were returned to 14L:10D at the conclusion of the bleeding session.
Tract tracing and histological procedures. Approximately 2
weeks after blood sampling, grafted hamsters (n = 11) were
anesthetized with pentobarbital and placed in a stereotaxic apparatus.
Eight of these hamsters received a pressure injection of fluorescent
microspheres (0.5 µl; Lumafluour, NY) directed at either the
preoptic area (POA; 6.1 mm rostral to 
, 0.2 mm lateral to the
midsagittal sinus, and 7.7 mm below the dural surface) or the
paraventricular nucleus (PVN; 4.5 mm rostral to , 0.6 mm lateral,
7.3 mm ventral).
In three additional cases, grafted animals were given iontophoretic injections of the anterograde tracer, Phaseolus vulgaris leucoagglutinin (2.5% PHA-L; Vector Laboratories, Inc., Burlingame, CA; 0.1 M PBS, pH 8.0) applied through a glass micropipette (50-µm diameter) as described by Gerfen and Sawchenko (30). The micropipette was lowered at a 10° angle and aimed at the graft. A current of 7 µA was applied for 10 min, alternating between 7 sec on and 7 sec off.
One to 3 weeks later, all hamsters were deeply anesthetized with sodium pentobarbital and injected intracardially with 5000 U sodium heparin (Rugby Pharmaceuticals, Norcross, GA). Animals were transcardially perfused with 200 ml 0.01 M PBS followed by 350 ml 4% phosphate-buffered paraformaldehyde (Electron Microscopy Sciences, Fort Washington, PA). Brains were removed, postfixed overnight, and transferred to 30% sucrose for 1–3 days. Four series of coronal sections (40 µm) were cut on a rotary freezing microtome. Sections were stored in cryoprotectant (31) until the time of immunocytochemical processing.
Series of frozen sections through the graft and adjacent POA/hypothalamus were processed for immunocytochemical detection of vasoactive polypeptide (VIP; Incstar Corp., Stillwater, MN) and PHA-L (Vector Laboratories, Inc.) using previously described procedures (22, 32). Antigens were detected using a modified avidin-biotin-horseradish peroxidase procedure (22) with diaminobenzidine as the chromagen. Sections were mounted, dehydrated, cleared, coverslipped, and examined under bright- and darkfield illumination. Fluorescent microspheres were visualized using fluorescein and rhodamine filter cubes.
Exp 2: male hamsters
Figure 2 illustrates the animal
treatments used in this experiment. Male hamsters were entrained to
14L:10D for at least 2 weeks before they were subjected to SCN or sham
lesions. Two weeks after surgery, each hamster was transferred to DD
for 11 weeks, during which time lesion efficacy was assessed. Constant
dim red light (<0.5 lux at the cage level) was present to permit
animal maintenance and cage changes at approximately 3-week intervals.
Experimental animals exhibiting an absence of circadian rhythmicity in
their running wheel behavior were returned to 14L:10D for
transplantation of fetal hypothalamic tissue as described above. The
present report is based on 56 of these animals that became arrhythmic
after SCN lesions, recovered circadian rhythms after fetal hypothalamic
transplant, and survived to the end of the experiment. Control data
were obtained from animals with SCN lesions that received either no
graft (n = 8) or a temporal cortex graft (n = 7). Testes were
palpated at the time of transplantation to confirm that the SCN lesion
had blocked the gonadal regression that occurs in intact hamsters
maintained in DD (33). Two weeks postgraft, hamsters were returned to
DD for 11–13 weeks to assess the ability of the graft to reinstate
locomotor rhythmicity.
|
Coronal brain sections (20 µm) were cut on a cryostat and mounted onto subbed slides. These sections were processed for assessment of 2-[125I]iodomelatonin and [3H]DAMGO binding to evaluate rhythmicity of melatonin and opiate binding, respectively. The results of these analyzes will be reported elsewhere. After exposure of autoradiograms, sections were stained with cresyl violet and inspected to verify the completeness of lesions and the presence of viable grafts.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Three additional animals showed a
transient recovery or an increase in rhythmic organization without a
clear circadian pattern, and 10 others remained arrhythmic after the
transplant.
|
). In most cases, the vaginal smear record indicated a condition
of continuous diestrus without any evidence of a discharge over a
period of at least 6 weeks after reintroduction to 14L:10D following
verification of free running locomotor rhythmicity in DD.
LH surges. Twelve control hamsters were successfully
cannulated and bled for a period of at least 24 h, and 10 of these
animals experienced unambiguous LH surges (Fig. 4, solid lines). The mean
amplitude of these surges was 10.32 ± 2.30 ng/ml (mean ±
SEM, nadir to peak). LH surges uniformly reached a sharp
peak and never lasted more than 2 sampling intervals (4 h). Without
exception, these LH surges occurred during the mid- to late subjective
day in LL, with the LH peak falling at zeitgeber time 7.8 ±
0.3 h (mean ± SEM).
|
, dashed line). Although low amplitude excursions
(<2 ng/ml, nadir to peak) of LH were evident in blood samples obtained
by serial bleeding from several of the 6 cannulated, SCN-lesioned
female hamsters bearing grafts that reinstated locomotor rhythms, none
of these animals experienced a LH surge at CT8. Five of these animals
exhibited no LH surge at any time during the sampling period. The
remaining grafted hamster exhibited an aberrant pattern of LH
secretion: basal levels were higher than any of the control levels at
all of the bleeding times, suggesting a disruption of the negative
feedback effects of estradiol. Although there was no surge during the
subjective day, LH levels rose during the subjective night and surged
at CT24 (Fig. 4, dotted line; animal 249). In addition, LH surges were absent in blood samples collected by cardiac puncture from each of the hamsters whose cannulas became occluded during sampling. In three cases, cardiac puncture was performed at CT8, when LH surges occurred in the controls. The patterns of LH secretion in hamsters that were ovariectomized at the time of cannulation did not differ from those that were gonadectomized at the time the SCN transplant was performed. Thus, it is unlikely that the absence of LH surges is attributable to a masculinizing effect of ovarian hormones on the transplant.
Urinary 6MTS rhythms. Urinary concentrations of 6MTS
exhibited robust diurnal rhythms in intact controls, with
concentrations showing a 6-fold rise during the mid-dark period and
declining through the day (Fig. 5a). SCN
lesions eliminated the nocturnal rise in 6MTS, and this effect was
not reversed by SCN grafts that reinstated locomotor rhythms, either in
individual animals or in the mean values of the group (Fig. 5, b and
c). Urinary 6MTS concentrations in the lesioned animals remained
near the nadir values of intact controls throughout the 2-day sampling
period.
|
Of the eight animals that received retrograde tracer injections, only
three injections were restricted to the host POA. In these animals,
numerous retrogradely labeled cells were seen in the anterior and
medial hypothalamus of the host brain, but very few cells (two or three
per graft) were detected in the graft itself (Fig. 6, D and E). In the five remaining cases,
tracer from the injection site also spread into the graft (Fig. 6, A
and B). In these cases, in addition to retrogradely labeled cells in
the medial hypothalamus of the host, numerous labeled neurons were
found in the graft (Fig. 6, A, B, and F). Retrogradely labeled cells
were seen in all portions of the graft, including those corresponding
to the location of VIP cells characteristic of the donor SCN (Fig. 6C).
|
). Numerous
anterogradely labeled fibers could be traced from PHA-L deposits in the
POA to other areas of the host brain, including the lateral septum and
bed nucleus of the stria terminalis bilaterally, and the contralateral
POA. By contrast, relatively few host fibers entered the graft (Fig. 7B). When they did, anterogradely labeled fibers crossed the graft-host
border at attachment sites along the wall of the third ventricle where
the ependymal cell layer was absent.
|
). Hamsters were
included only in cases in which fetal hypothalamic transplants
reinstated circadian locomotor rhythms and in which running onsets were
sufficiently clear and reliable to allow determination of circadian
phase of death. None of the hamsters that received no graft or an
equivalent volume of fetal cortex experienced a restoration of
locomotor rhythms.
|
). SCN-lesioned
hamsters bearing behaviorally functional SCN grafts had significantly
heavier testes at the time of death, as did SCN-lesioned controls that
received grafts of temporal cortex (combined testis weight, 3.35
± 0.11 and 3.40 ± 0.26 g, respectively, mean ±
SEM; both P < 0.001 vs.
sham-lesioned controls; Fig. 9).
|
).
The pineal melatonin content in control hamsters was significantly
higher at CT16, 20, and 24 than at CT4 (P < 0.001). A
modest (2-fold) rise in serum melatonin concentrations occurred during
the subjective night in control hamsters, but did not achieve
statistical significance.
|
Glucocorticoid rhythms. Assay of serum cortisol and
corticosterone collected from groups of intact animals killed at
various circadian times indicates that both hormones display free
running rhythms that are phase locked to that of locomotor activity
(Fig. 11, top). Peak
concentrations of these hormones occurred during the late subjective
day. Corticosterone values at CT8 and 12 were significantly higher than
those at CT24, and values at CT8 were significantly higher than those
at CT4 (P < 0.05, by Duncan’s multiple range test).
Cortisol values at CT4 were significantly lower than those at CT12 and
20 (P < 0.05). In contrast, mean values for both
cortisol and corticosterone showed no significant effects of circadian
time among hamsters whose locomotor rhythms had been restored by SCN
transplants (Fig. 11, bottom).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Locomotor rhythmicity has provided a useful set of "hands" of the circadian clock in many previous studies (10, 11, 12, 13). Observation of the running activity records of intact hamsters enables one to predict with great accuracy the timing of the LH surge, ovulation, sexual behavior, pineal melatonin secretion, and adrenal corticosteroid release. The finding that hamsters in which exposure to constant light induces "splitting" of the activity rhythm have two LH surges, each phase-locked to one of the activity bouts (13), indicates the linkage of endocrine oscillators to the circadian clock that regulates activity. In light of such findings, the dissociation between the ability of grafts to reinstate rhythms of activity and endocrine secretion is surprising. The condition of the SCN-grafted hamster may be most comparable to that of animals sustaining knife cuts in the vicinity of the SCN that selectively interrupt particular sets of afferent or efferent projections of this nucleus. For example, semicircular knife cuts posterior to the SCN that eliminate estrous cycles spare circadian rhythms of drinking (34), and horizontal cuts that compromise photoperiodic responses and E2-induced LH surges may leave oscillations of locomotor activity intact (35, 36).
The observation that SCN grafts reinstate locomotor rhythms has raised the question of the source and nature of the signals by which the SCN expresses its pacemaker function. Our present tract tracing results are consistent with earlier indications that the graft establishes relatively few afferent or efferent connections with the host POA. Instead, graft neurons appear to project predominantly to other portions of the graft (22, 37, 38). Such evidence along with the ability of SCN grafts placed far from the basal anterior hypothalamus to restore activity rhythms and the effectiveness of interstitial transplants of dissociated SCN cells (39) argue for the existence of a humoral signal in the regulation of locomotor activity rhythms. The SCN releases neuropeptides rhythmically in vivo (40) and in vitro (41, 42), and the persistence of locomotor rhythmicity in hamsters after isolation of the SCN (43) and placement of encapsulated grafts (44) provides the most conclusive evidence for humoral output. However, neither the studies of hypothalamic islands nor those of encapsulated grafts used neuroendocrine rhythms as an end point. The present data indicate that if the output of the SCN graft responsible for locomotor activity is indeed humoral, it is qualitatively or quantitatively inadequate to restore neuroendocrine rhythms.
Recent studies are consistent with the hypothesis that SCN outputs important for neuroendocrine rhythms are axonal rather than humoral. Pharmacological evidence indicates that vasopressinergic projections to the paraventricular and dorsomedial hypothalamus suppress ACTH secretion, and that the rhythmicity of serum corticosterone results from fluctuations in activity in this pathway (45). The preovulatory surge of LH in hamsters and rats may be triggered by direct inputs of SCN efferents to GnRH- and/or estrogen receptor-immunoreactive cells (33, 46), and VIPergic SCN cells have been suggested to play an important role in phasic GnRH release. Similarly, rhythms of melatonin secretion in hamsters appear to depend upon SCN efferents to the region of the PVN, which, in turn, project to thoracic preganglionic neurons of the sympathetic nervous system (36). The failure of grafts to restore endocrine rhythms may reflect the inability of the grafted SCN to establish appropriate connections with neuroendocrine targets in the host, such as GnRH neurons and/or estrogen receptor-immunopositive cells (32, 46, 47). Some of these projections may need to be reciprocated for proper coordination of the LH surge or other endocrine rhythms. Although efferents from anterior hypothalamic xenografts form close contacts with GnRH cells in the host medial septum (48), it is not clear whether such outgrowth arises from the donor SCN in these grafts or instead from other anterior hypothalamic tissue (49). The POA and anterior hypothalamus, in which many hypophysiotropic cells reside, also receive dense input from the subparaventricular region, which is a target of the SCN (32, 50, 51, 52). It is likely that variable portions of the sub-PVN were included in the transplants, but we lack immunocytochemical markers with which to assess this. Furthermore, efferents of SCN grafts may travel short distances into the host brain, including the sub-PVN region. Although such input might have been expected to permit some relay of circadian information to neurosecretory cells, it must be insufficient to reinstate statistically significant neuroendocrine rhythmicity.
It is possible that the SCN lesions performed in this study eliminated not only the circadian pacemaker of the host, but also descending neurosecretory fibers that pass through the suprachiasmatic region en route to the median eminence. In this event, grafts that replace the pacemaker might have failed to repair damage to critical fibers of passage that may have arisen from either the electrolytic lesion or deformation of the host tissue by introduction of the transplant to the third ventricle. This explanation for the failure of SCN grafts to restore endocrine rhythms is unlikely for several reasons. First, fibers other than those of SCN cells that may be important to glucocorticoid and pineal rhythms are unlikely to have been damaged by the lesion. Second, the GnRH fibers that reach the rat median eminence from the POA traverse a variety of routes, including lateral trajectories that would be unlikely to have been compromised by the lesions (53, 54). The observation that GnRH-immunoreactive fibers are normally present in the intact SCN suggests, however, that reciprocal input between the POA and the pacemaker may be important to generation of the surge (46) (de la Iglesia, H. O., and E. L. Bittman, unpublished observations). As we did not examine the grafts for such contacts, we cannot evaluate the possibility that relevant host projections to the SCN were compromised by the surgery.
We have also considered the possibility that fetal grafts may have been masculinized by E2 released from polycystic ovaries of the host, thus eliminating the capacity of the graft to sustain GnRH surges. Although sexual dimorphism of the SCN has been reported (55), it is not established whether the male SCN is adequate to support LH surges. In an attempt to address this issue, we included in our study hosts in which ovariectomy was performed at the time of the graft to ensure that the transplant was not exposed to gonadal steroids. Hamsters treated in this way were no more likely to recover E2-induced LH surges than were animals that retained their ovaries until the time of cannulation a few days before the E2 challenge. Thus, it seems unlikely that masculinization of the SCN in the grafts can account for the failure of estrogen-induced LH surges to recover after transplantation.
Photoperiodic control of testicular function is clearly dependent upon the SCN. Melatonin receptors reside in the SCN of some photoperiodic species, and infusions of melatonin are particularly effective at regulating gonadal function when directed at structures in the SCN region of Siberian hamsters (56). In Syrian hamsters, however, injections or infusions of melatonin regulate gonadal function even after destruction of the SCN (57, 58). The present observation that SCN grafts were rarely if ever effective in reinstating nocturnal elevations of melatonin in the host pineal gland are consistent with the conclusion, drawn from earlier studies, that the role of the SCN in regulation of photoperiodic responses is attributable to its control of melatonin synthesis and secretion. Both the studies of females, in which the pattern of urinary melatonin metabolites was recorded in serial samples from individuals, and the data from males, in which grouped data were obtained by the death of free running hamsters at various circadian times, indicate that the endogenous melatonin patterns of the hosts were inadequate to trigger gonadal regression in DD.
The restoration of circadian rhythms by SCN grafts may not be an all or none phenomenon. Even the behavioral rhythms restored in this study, as in previous experiments, tend to show more variable onsets and less stable periods than those of intact hamsters. The melatonin contents of the pineal glands in the SCN-grafted male hamsters suggests that a partial restoration of this rhythm might have occurred; more than half of the hamsters killed at CT20 exhibited a partial rise in melatonin content. Even though the melatonin concentration in this group was significantly less than that in intact hamsters and did not differ from that of pineals collected during the subjective day, it is possible that the graft was adequate to induce a small nocturnal rise in melatonin in some of the hamsters. Similarly, one of the SCN grafted female hamsters (no. 249) exhibited a rise in LH. This apparent surge occurred at an inappropriate circadian time and was accompanied by apparent derangement of basal LH secretion. Furthermore, SCN grafts were uniformly incapable of reinstating estrous cycles. Nevertheless, data from this exceptional animal suggests that partial restoration of endocrine rhythmicity may occur, although it is insufficient to reinstate the functional response (ovarian cycles or gonadal regression in short days) upon transplantation of the SCN. Additional studies would be required to establish surgical procedures that might improve or maximize restoration of neuroendocrine rhythmicity.
We chose to study glucocorticoid rhythms in part because they seem most tightly linked to arousal and general activity. Thus, the inability of behaviorally effective grafts to reinstate adrenocortical rhythms was unexpected. On the other hand, it is rare that circadian rhythms of behavior in hamsters bearing SCN grafts are as regular and consistent as those in intact animals. Perhaps the absence of endocrine rhythms in SCN-grafted animals contributes to the imprecision of the restored behavioral oscillations. Projections or humoral signals of the graft that influence arousal or eating could generate periodic feedback signals (e.g. blood glucose, fatty acids, osmolality, pH, etc.) that could regulate endocrine secretion even if central pacemaker control over hypophysiotropic neurons was not restored. The absence of endocrine rhythms in hamsters bearing behaviorally effective grafts, however, suggests that such control by activity-generated feedback signals is weak. It will be interesting to examine the ability of SCN grafts to reinstate temperature rhythms, whose elimination by SCN lesions has been controversial (6, 7) and which may rely in part upon oscillations of thyroid hormone and interleukins. The contributions of gut hormones, insulin, thyroid hormones, GH, and vasopressin secretion to drinking and eating suggest that endocrine rhythmicity might facilitate the reappearance of behavioral rhythms after transplantation of the SCN. Additional experiments would be required to establish whether SCN grafts reinstate circadian patterns of secretion of these hormones.
In conclusion, the elimination of LH surges, melatonin, and glucocorticoid rhythms by SCN lesions is consistent with the findings of many previous studies in hamsters and rats and supports the role of this master pacemaker in regulation of rhythmic endocrine function. The inability of grafts adequate to restore locomotor rhythms to reverse the elimination of ovarian, adrenal, pituitary, and pineal rhythms and DD-induced testicular regression by SCN lesions suggests that SCN outputs necessary for neuroendocrine rhythms differ quantitatively or qualitatively from those that maintain locomotor behavior. For example, targets of SCN efferents that regulate endocrine function may require more extensive innervation than they receive from the transplants. In contrast, the maintenance of locomotor rhythmicity may require only a sparse, local neuronal outgrowth or a humoral signal from the graft.
|
Acknowledgments |
|---|
MTS assays. |
Footnotes |
|---|
2 Present address: Department of Neuroscience, University of
Pennsylvania Medical School, Philadelphia, Pennsylvania 19104.
3 Present address: Department of Biology, Middle Tennessee State
University, Murfreesboro, Tennessee 37132.
4 Present address: School of Medicine, Fukuoka University,
45–1,7-chome Nanakuma, Fukuoka 814–80, Japan.
5 Present address: St. Louis University School of Medicine, St.
Louis, Missouri 63119.
Received May 7, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Vujovic, A. J. Davidson, and M. Menaker Sympathetic input modulates, but does not determine, phase of peripheral circadian oscillators Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2008; 295(1): R355 - R360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Christian and S. M. Moenter Vasoactive Intestinal Polypeptide Can Excite Gonadotropin-Releasing Hormone Neurons in a Manner Dependent on Estradiol and Gated by Time of Day Endocrinology, June 1, 2008; 149(6): 3130 - 3136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Valenzuela, C. Torres-Farfan, H. G. Richter, N. Mendez, C. Campino, F. Torrealba, G. J. Valenzuela, and M. Seron-Ferre Clock Gene Expression in Adult Primate Suprachiasmatic Nuclei and Adrenal: Is the Adrenal a Peripheral Clock Responsive to Melatonin? Endocrinology, April 1, 2008; 149(4): 1454 - 1461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kals |
