We studied the effects of photoperiod on metabolic profiles, adiposity, and gene expression of hypothalamic appetite-regulating peptides in gonad-intact and castrated Soay rams. Groups of five to six animals were studied 6, 18, or 30 wk after switching from long photoperiod (LP: 16 h of light) to short photoperiod (SP: 8 h of light). Reproductive and metabolic indexes were measured in blood plasma. Expression of neuropeptide Y (NPY), proopiomelanocortin (POMC), and leptin receptor (ObRb) in the arcuate nucleus was measured using in situ hybridization. Testosterone levels of intact animals were low under LP, increased to a peak at 16 wk under SP, and then declined. Voluntary food intake (VFI) was high under LP in both intact and castrated animals, decreased to a nadir at 12–16 wk under SP, and then recovered, but only in intact rams as the reproductive axis became photorefractory to SP. NPY gene expression varied positively and POMC expression varied negatively with the cycle in VFI, with differences between intact and castrate rams in the refractory phase. ObRb expression decreased under SP, unrelated to changes in VFI. Visceral fat weight also varied between the intact and castrated animals across the cycle. We conclude that 1) photoperiodic changes in VFI reflect changes in NPY and POMC gene expression, 2) changes in ObRb gene expression are not necessarily determinants of changes in VFI, 3) gonadal status affects the pattern of VFI that changes with photoperiod, and 4) in the absence of gonadal factors, animals can eat less but gain adiposity.
- Soay ram
- neuropeptide Y
- leptin receptor
- sex steroids
many mammals show cycles in voluntary food intake (VFI), body weight, and energy metabolism as an adaptation to life in a seasonal climate with predictable food abundance in summer and scarcity in winter. The annual cycle in day length (photoperiod) is utilized as a time cue to synchronize this physiology in a range of species, including sheep. Such photoperiodic control of energy conservation is seen in ovariectomized (OVX) ewes under natural conditions (14) and in rams under controlled lighting (4, 13, 35, 36, 38). Exposure to a change from long photoperiod (LP) to short photoperiod (SP) in sheep (13, 14, 35, 36, 38) reduces VFI and activates reproductive function (3, 34). Expression of the gene for neuropeptide Y (NPY) in the arcuate nucleus (ARC) of the ovine brain (13, 14) is high under LP and is closely linked to the time when VFI is highest. Since this neuropeptide is orexigenic (25, 46), there is a strong indication that this drives the seasonal cycle in VFI. On the other hand, testosterone treatment of castrated rams increases NPY gene expression under LP (17), so VFI can be influenced by testosterone and/or photoperiod.
Cells that express proopiomelanocortin (POMC) in the ARC also play a major role in energy balance (15, 25), so expression of this gene may be important in the seasonal regulation of metabolism. POMC encodes for melanocortins and β-endorphin (25), which act to either decrease or increase VFI, respectively. It is thought that the anorectic properties of melanocortins act in an opposite manner to NPY (25). In OVX ewes, the expression of the POMC gene appeared to be unrelated to the seasonal appetite cycle (14), but in Soay rams, expression was higher under SP, a condition when testosterone levels are high and VFI is low (13). In the ram model, therefore, our earlier observations suggest that lower VFI under SP could be due to an increased POMC expression, driven by increased testosterone levels and/or photoperiod.
NPY-producing cells express leptin, insulin (23, 25), and ghrelin receptors (9, 58), allowing reception of peripheral metabolic information and leptin and insulin to reduce NPY expression (25, 45). POMC-expressing cells also possess leptin (7, 25) and insulin receptors (7), and these circulating factors increase POMC expression. Accordingly, these two cell types are well placed to act as “first-order” neurons in the central recognition of peripheral metabolic status (46). In addition to their ability to sense metabolic status, subsets of both NPY- and POMC-producing cells in the ovine ARC express estrogen receptors (53), at least in the ewe, so expression of the relevant genes can be influenced by gonadal steroids. Whether these cells also express androgen receptors is not known (47).
The aim of the present study was to investigate the interaction between photoperiod and gonadal status in the control of VFI and energy balance, using our highly seasonal Soay ram model. The animals were preconditioned under LP and then exposed to constant SP for 30 wk. This regimen was selected to induce a cycle of testicular activation and increasing testosterone secretion lasting ∼12–16 wk (photoinductive phase) followed by testicular regression and declining plasma testosterone levels, lasting a further 12–16 wk (photorefractory phase) (3, 34) as occurs naturally outdoors during the seasonal transition from summer to winter. We studied age-matched intact and castrated animals to assess the relative importance of photoperiod and gonadal status in the regulation of VFI, BW, adiposity, metabolic factors, and hypothalamic NPY, POMC, and leptin receptor (ObRb) gene expression.
MATERIALS AND METHODS
Animals in this study were treated under a Project License issued by the United Kingdom Home Office in accordance with the Animals Act (Scientific Procedures) of 1986.
Animals and Lighting Regimen
Soay rams were used for the study because of their marked seasonal physiology similar to that of the wild-type mouflon (32). The animals were sexually mature and 1 yr old at the commencement of the study (body weight 22.1 ± 0.6 kg, mean ± SE). They were brought indoors in February into light-controlled rooms, individually penned with visual and tactile contact between neighbors, and fed a standardized diet of commercial dried grass pellets (Vitagrass; Vitagrass Farms, Cumbria, UK). Hay and water were provided ad libitum. The environmental temperature varied between 10 and 20°C.
The animals were initially preconditioned to LP (16-h light:8-h dark) for 16 wk. The light was turned on at 0800, and light intensity was ∼160 lux at the animal's eye level. After 4 wk, half of the animals were surgically castrated under a general anesthesia (castrated group, n = 16) and the remainder were untreated (intact group, n = 15). Following the 16-wk pretreatment under LP, the lighting was abruptly switched to prolonged SP (8-h light:16-h dark) for 30 wk. This lighting regimen was predicted to induce testicular reactivation and increasing testosterone secretion during the first half of the study (termed the photoinduction phase), followed by testicular regression and withdrawal of testosterone (termed the photorefractory phase) in the intact rams. This would mimic the reproductive transition that occurs naturally from summer to winter outdoors (34). The testosterone cycle would be absent in the castrated rams, thus allowing a direct comparison with the age- and time-matched intact control rams.
At weekly intervals, VFI was measured over a period of 24 h for each individual by providing 2 kg of the standard dried grass pellets and subtracting the weight of food remaining at the same time the following day. The food hoppers were designed to minimize food wastage. Every 2 wk, testis diameter was measured through the scrotum with the use of calipers, and every 4 wk, the animals were weighed using a large animal cage balance. The animals were fully habituated to the handling procedures.
To record changes in hormone levels and other metabolic parameters, a twice-weekly blood sample was taken between 1000 and 1200 by jugular venipuncture 2 h after the time of feeding. The blood samples were heparinized, and the plasma was separated by centrifugation within 30 min and stored at −20°C as three equal aliquots to avoid repeated refreezing for the multiple assays.
On three occasions at 6, 18, and 30 wk under SP, a cohort of five intact and five to six castrated rams were euthanized by an overdose of pentobarbital sodium (Euthatal; Rhone Merieux, Essex, UK). The brains were perfused with 4% paraformaldehyde and prepared for in situ hybridization histochemistry as described previously (13). At autopsy, we recorded total body weight and the weight of abdominal fat and rumen/reticulum, abomasum, intestines, and reproductive organs. Subtraction of the gut from the total body weight gave an indication of the extent to which a difference in body weight was due to a difference in gastrointestinal food content and VFI.
In Situ Hybridization
Frozen brain tissues were cut at 20-μm thickness in coronal plane and were kept at −20°C. In situ hybridization was performed as previously described (13, 52) using 35S-dUTP-labeled riboprobe (Amersham Pharmacia Biotech, Sydney, Australia). To quantify the level of gene expression, one section per animal per peptide was taken between mid- and caudal ARC for NPY and mid-ARC for POMC and ObRb. Slides were exposed under photographic emulsion at 4°C (9 days for NPY, 11 days for POMC, and 9 wk for ObRb). Emulsion-dipped slides were analyzed at a cellular level by counting the number of labeled cells and silver grains per cell. The number of labeled cells was counted at ×200 magnification. For number of silver grains per cell, 10 cells in each of the five areas of the ARC were randomly selected for analysis of silver grain density at ×400 magnification with the use of a microcomputer imaging device (MCID) microimaging system from Imaging Research (Brock University, St. Catherines, Ontario, Canada). If there were fewer than 10 cells/area, then every cell was counted.
To assess the reproductive status of the rams, the blood concentrations of follicle-stimulating hormone (FSH) and testosterone were measured using routine radioimmunoassays validated for sheep plasma. The FSH assay (40) used NIDDK-FSH-RP2 as a standard with intra- and interassay coefficients of variation of <8.0%. Concentrations of testosterone were measured using a method not involving chromatography and modified for an iodinated tracer (51).
Assay of Metabolic Indicators
For statistical comparisons, we used the method of two-way analysis of variance after checking for homogeneity of variance. Comparisons between means were performed by post hoc testing using the least significant differences method.
Food Intake, Body Weight, and Adipose Tissue Weight
VFI of both gonad-intact and castrated animals was high and increased during the pretreatment period under LP (Fig. 1A). The switch to SP resulted in a decline in VFI in both groups to a nadir between 12 and 18 wk. VFI then recovered from weeks 20 to 30 in the gonad-intact animals but remained low in the castrated animals. This produced a significant difference in VFI between the two groups (Fig. 1, A and B) over the time when the intact rams were showing testicular regression and declining blood testosterone concentrations due to the development of photorefractoriness (see below).
The gonad-intact animals were slightly (P < 0.05) heavier than castrates, but total body weight remained relatively stable across the study (Fig. 2A). The weight of the rumen and intestine (including food within) followed the same trend as VFI (Fig. 2B). Total body weight less gut weight increased significantly at weeks 18 (P < 0.05) and 30 (P < 0.01), compared with week 6 in castrated animals (Fig. 2C), and a similar trend toward increased body weight was seen at week 30 in the gonad-intact animals. Figure 2Dshows the longitudinal changes in subtracted body weight in the castrated animals.
In the castrated animals, total abdominal fat weight was greater at weeks 18 (P < 0.05) and 30 (P < 0.001) than at week 6, but there was no parallel change in the gonad-intact animals (Fig. 2E). A similar trend was seen for omental fat weight and retroperitoneal fat (Fig. 2, F and G). The level of adiposity was higher in the castrated animals than in the intact animals at weeks 18 and 30 (Fig. 2, E–G). Within the castrated and gonad-intact groups, there was no change across time in plasma leptin concentrations, although levels were higher (P < 0.05) in castrated than in intact animals at week 18 (Fig. 2H).
Plasma NEFA concentrations in gonad-intact animals were significantly higher at week 18 than at weeks 6 (P < 0.001) and 30 (P < 0.01) (Fig. 3A). In castrated animals, NEFA concentrations were significantly higher at weeks 18 (P < 0.01) and 30 (P < 0.001) than at week 6. NEFA concentrations in gonad-intact animals were significantly higher than those in castrated animals at weeks 6 (P < 0.01) and 18 (P < 0.05). Plasma urea concentrations in gonad-intact animals were significantly lower at week 18 than at weeks 6 (P < 0.01) and week 30 (P < 0.05) (Fig. 3B), following a similar trend to VFI (Fig. 1, A and B). In castrated animals, plasma urea concentrations were significantly higher at week 6 than at weeks 18 (P < 0.01) and 30 (P < 0.05), also following the pattern of change in VFI. Plasma glucose concentrations in gonad-intact animals were significantly lower at week 6 than at weeks 18 (P < 0.01) and 30 (P < 0.001) (Fig. 3C). In castrated animals, glucose concentrations were significantly higher at week 30 than at weeks 6 (P < 0.01) and 18 (P < 0.05). Plasma glucose concentrations were significantly (P < 0.05) higher in the castrated animals than in the gonad-intact animals at week 30. Plasma insulin concentrations were similar within and between the groups of animals (Fig. 3D).
Gonad-intact rams displayed reactivation of testicular function before the end of the LP pretreatment photoperiod as normally occurs outdoors (long-day refractoriness), accounting for the elevation in plasma FSH concentrations and increased testicular size within 6 wk of SP. By week 30, testis weight had reduced significantly (P < 0.001) (Fig. 4A). Parallel changes were seen in plasma testosterone levels, increasing from week 6 to a maximum at week 18 and then declining significantly by week 30 (P < 0.01) (Fig. 4B). Seminal vesicle weights reflected testosterone levels in both castrates and gonad-intact animals (Fig. 4C). Plasma FSH levels were predictably higher in castrated than in gonad-intact animals (Fig. 4D), and the decline in plasma FSH concentrations in gonad-intact animals by week 18 (P < 0.001) reflected the negative feedback effect of testosterone (and inhibin), which is maximal at this time.
Gene Expression for NPY, POMC, and ObRb in ARC
In gonad-intact animals, the number of cells expressing NPY was similar at each of the three sampling time points under SP (Fig. 7A), but the level of expression per cell was lower (P < 0.05) at week 18 than at weeks 6 and 30 (Fig. 7B). In castrated animals, the number of cells and the level of expression per cell were lower (P < 0.001) at week 18 than at week 6 and remained low until week 30. There was a highly significant (P < 0.001) rebound in expression of NPY at week 30 in gonad-intact animals (Fig. 7, A and B).
POMC expression was lowest at week 6, showing a significant rise by week 18 in both gonad-intact and castrated rams (Fig. 7, C and D). In castrated but not in gonad-intact animals, there was a further significant increase in cell number and in expression level per cell (P < 0.05 and 0.01, respectively) between weeks 18 and 30. This effect was such that there was a significant difference in cell number (P < 0.05) and in expression per cell (P < 0.01) between castrated and gonad-intact animals at week 30.
In the castrated animals, both cell number and level of expression per cell were lower at weeks 18 and 30 than at week 6 (Fig. 7, E and F). In the gonad-intact animals, a similar pattern was observed, but this reached statistical significance for the level of expression per cell only.
We have used the highly seasonal Soay ram model to examine the interaction between photoperiod and gonadal status in the regulation of VFI, energy balance, and the expression of relevant hypothalamic appetite-regulating peptide genes in the ARC. In the absence of gonadal hormones, the reduction in VFI under SP that was displayed in the castrated animals demonstrates a specific effect of photoperiod. This SP effect was associated with reduced NPY and ObRb gene expression in the ARC and a parallel increase in POMC gene expression. This is consistent with the view that photoperiod acts through the hypothalamic appetite regulatory centers to regulate VFI and energy balance. Most remarkably, VFI was reduced in gonad-intact animals by a switch from LP to SP but increased again spontaneously following the decline in testicular activity under prolonged exposure to SP.
In all cases, VFI was coupled to levels of expression of the NPY gene in the ARC. Thus NPY expression was lower when VFI was reduced and increased in the gonad-intact animals, when VFI rose in this group. POMC expression also was consistent with the notion that melanocortins derived from this gene act to reduce food intake (25), being elevated in both castrated and gonad-intact animals at week 18 under SP. On the other hand, the increase in VFI in gonad-intact animals between weeks 18 and 30 was not associated with a reduction in POMC expression, suggesting that the predominant driver of VFI is NPY, rather than a reduction in melanocortin production. In the castrated animals, there was a further increase in POMC expression in concert with reduced NPY expression, which can account for the maintenance of low VFI in this group. Since both NPY and POMC cells are regulated by leptin, the levels of ObRb may be relevant to the changes in expression that were seen. Expression of this receptor was reduced by SP in both gonad-intact and castrated groups between weeks 6 and 18, which does not account for lowered NPY and increased POMC expression at these times. The continued low level of ObRb in the gonad-intact animals between weeks 18 and 30 (when NPY expression and VFI increased in this group), however, is inconsistent with the reduction in NPY and VFI. If it were accepted that leptin acts as a satiety factor, reducing NPY expression and increasing POMC expression, then a reduction in the signaling apparatus (ObRb) under SP might facilitate an increase in VFI, but this was clearly not the case. In fact, SP caused reduced VFI, even though ObRb expression was reduced. It should be noted, however, that gene expression may not be reflected in expressed receptor levels. Accordingly, we conclude that the changes in NPY and POMC gene expression are not simply explained by changes in ObRb expression and are more likely to be due to the central effects of photoperiod acting through melatonin to influence expression of genes such as those for NPY and POMC. It would be of considerable interest to ascertain whether the appetite-regulating cells of the brain express melatonin receptors, because the central administration of melatonin using cerebral microimplants has been shown to affect the body weight cycle in Soay rams under LP (33).
The differing pattern of VFI in the castrated and gonad-intact animals demonstrates that the cycle of testicular activity influences the appetite cycle. This was not evident during the first 18 wk of SP, however, suggesting that the predominant effect at this stage was photoperiod with no additional effect related to the presence of gonadal hormones. This is consistent with another study, which showed no effect of testosterone on NPY expression in castrated rams under SP (17). Following this, plasma testosterone levels fell and VFI rose in the gonad-intact animals, indicating an interactive effect of testosterone on central mechanisms controlling appetite. Such a change was not seen in the castrated animals, so a cyclic pattern of gonadal activation and quiescence appears necessary for this phenomenon. Since this is the pattern that normally occurs, the data suggest that the gonadal cycle modulates the photoperiod-induced VFI cycle. This effect could be due to the action of gonadal androgens or estrogens, since testosterone can be aromatized to estrogen by the brain. Nutritional status does not change the expression of aromatase activity in the ovine brain (50). Subsets of NPY- and POMC-expressing cells possess estrogen receptor-α in the female ovine brain (53), allowing for direct action of this steroid, but it is unknown whether these cell types possess androgen receptors in either sex (48). A high level of androgen receptor expression is found in the ARC and premammillary nucleus (47), so these receptors could be in the appetite-regulating cells or in nearby cells. It is notable that the premammillary nucleus is the site where melatonin acts to mediate photoperiod effects on the reproductive axis (37).
Estrogen reduces food intake and body weight in a range of species, such as rats (57, 59), mice (21), and monkeys (27), but the only available data in male sheep (1) showed no effect of low-dose estrogen implants on feeding status in castrated rams. In regard to the effects of androgen on VFI in sheep, nothing is known, but both testosterone and dihydrotestosterone reduce VFI in other species (43). The results in rams are consistent with the view that androgens suppress VFI at the peak of sexual activity, and androgen withdrawal provides a cue to activate feeding after the rut. This may favor a recovery of body condition after the mating season, which would be before midwinter in rams living under natural conditions outdoors.
Our data show that a cycle of testicular activity in some way “programs” the appetite cycle of rams such that a reduction in VFI is caused by SP coincidentally with maximal testicular function, but appetite is restored when testosterone levels fall. This is not seen in castrated animals, in which there is no cycle of testicular activity. Photoperiod can have inductive effects, but refractoriness is often seen when a particular photoperiodic schedule is prolonged. For example, under constant SP, plasma levels of luteinizing hormone in estrogen-implanted OVX ewes “escape” from the breeding season pattern (44). Another example of photorefractoriness is the pattern of prolactin secretion seen in Soay rams under constant photoperiod (30). The present study clearly showed a cycle of testicular activity that was induced by SP, but the reproductive axis subsequently became refractory to this photoperiod, and testicular activity declined. In the case of VFI, we saw an apparent photorefractory response in gonad-intact animals but not in castrated animals, implying that the testicular cycle is an important determinant of the appetite cycle.
Even though there were differences in body weight between gonad-intact and castrated animals in the current study, the longitudinal changes in body weight in either gonad-intact or castrated animals were unremarkable. This was despite major excursions in VFI. When the weight of the gut was taken into account, it was clear that both gonad-intact and castrated animals were gaining weight throughout the experiment as the animals became more physically mature. The most interesting feature of the change in body weight was how it related to changes in body fat. In particular, the visceral adiposity of the castrated animals increased throughout the study, whereas this did not occur in the gonad-intact animals. This effect was apparent even though the castrated animals displayed reduced VFI across the same period. In other words, the castrated animals gained fat over the period when they were eating less. The reason that this occurred in castrated animals and not in gonad-intact animals could be due to the actions of testicular steroids in the latter. Testosterone increases muscle mass, protein synthesis, RNA synthesis, and glycogen accumulation in muscle (6), and intact male ungulates have greater muscle mass and less fat than castrates (18, 42). Muscle mass has a greater energy requirement than fat tissue (29), and functional testes have a large energy requirement (49). Overall, gonad-intact rams have a greater energy requirement than castrates (22), which would prevent the deposition of fat. The effects of testosterone on sexual and aggressive behavior also may have prevented accumulation of adipose in the gonad-intact animals, since the animals display these behaviors to a greater extent than the castrates, and this changes with the cycle in testosterone secretion (31). Studies in Soay rams living under feral conditions have shown that castrated rams spend more time feeding, whereas intact rams spend more time moving and showing sexual and aggressive behavior during the mating season (54), and the lack of the energetic reproductive behavior in castrates is reflected in a significantly longer life span (26).
Both castrated and gonad-intact animals gained body weight under SP conditions, which could be due to an effect of photoperiod to lower overall energy expenditure. Since the breed evolved under harsh environmental conditions, the energetic costs associated with reproduction, growth, and maintenance need to be coordinated (16). Accordingly, one study showed a marked photoperiodic effect on metabolic rate in Soay rams, which fell by 25% when day length was reduced from 18 to 6 h (5). Whether a similar photoperiodic effect is seen in castrated animals is not known, but the reduction in energy expenditure seen under SP (despite the demands of the testes) would be sufficient to ensure that the animals in the present study remained in positive energy balance. This would lead to growth of muscle in the gonad-intact animals and the deposition of adipose tissue in the castrated animals. The photoperiodic effect to reduce metabolic rate may be more marked in the castrated animals, since the demands of the testes and steroid-driven process are not operative.
Metabolic indices in plasma of our sheep presented a complex picture relating to adiposity and energy balance. In castrated animals, NEFA levels increased with increasing adiposity, probably reflecting mobilization of energy stores during a period of reduced VFI. In gonad-intact animals, NEFA levels increased to a greater extent between weeks 6 and 18 under SP, perhaps reflecting a greater mobilization of fat stores with reduced food intake, which could be related to activation of the reproductive cycle and the higher energy requirements of rams with active gonads. Plasma NEFA concentrations are a reflection of whole body lipolysis so that, in the case of energy demand, levels are a reflection of fat mobilization (19, 20). By week 18, the gonad-intact animals had higher NEFA concentrations than the castrated animals, which may be an indication that the intact animals were mobilizing fat, whereas castrates were not. By the end of the experiment, however, NEFA concentrations had fallen in the gonad-intact animals, perhaps indicating that the animals were deriving most of their energy from ingested food, which had increased by this time. Although energy balance is the major determinant of NEFA concentrations in plasma, higher levels also are seen in animals of greater adiposity (39). Accordingly, NEFA levels continued to rise throughout the study in the castrated animals, as a reflection of their increasing adiposity. Changes in plasma glucose levels occurred without significant alterations in insulin status. In general, glucose levels rose across the experimental period, indicating increased gluconeogenesis. The increase in plasma urea levels between weeks 18 and 30 in the gonad-intact animals was most likely due to their increase in dietary protein intake; such a shift did not occur in the castrates, since their food intake remained stable over this time. These small, albeit significant, changes in metabolic indicators show that the animals were never in a state of negative energy balance, even though excursions in food intake were substantial. These changes in metabolic indices provide no clue as to how castrated animals eat less food yet gain visceral adipose tissues, but it seems likely to reflect the lowered energy expenditure in these sex steroid-deficient animals and perhaps a prolonged reduction in maintenance requirements during SP. Greater energy demand in the gonad-intact animals and the preferential partitioning of nutrient to muscle would prevent accumulation of adipose tissue in the gonad-intact animals.
Leptin is synthesized from adipocytes, and we expected changes in plasma levels to parallel adiposity, but this was not the case. Leptin levels were higher in castrated animals than in gonad-intact animals at the 18-wk time point, which may have been due to the maximal difference in plasma testosterone levels occurring at this time. Previous studies showed that plasma leptin concentrations per fat cell mass are negatively correlated with circulating concentrations of androgens (10). The effect of estrogen on serum leptin levels is controversial (11, 28, 55, 56). We conclude that gonadal steroids, probably testosterone and/or estrogen, most likely affect leptin production in rams, although this has not been tested in an explicit way. ObRb transmits the satiety effect of leptin to the brain (25), but the lack of changes in levels of leptin in plasma and in ObRb expression (vide supra), relative to the significant changes in hypothalamic NPY and POMC gene expression, strongly suggest that altered leptin status does not account for the season cycle in VFI. The question as to what extent leptin status dictates VFI under different photoperiods is further confounded by two issues. First, a recent study (2) has shown that leptin transport into the brain and leptin levels in cerebrospinal fluid of rams are higher under LP than SP. Thus leptin would be expected to exert greater satiety effect during LP, but VFI is higher under this photoperiod. Second, we found relatively little shift in plasma leptin levels, despite changes in abdominal adiposity. Leptin production is greater in subcutaneous fat of humans (41) but greater in visceral fat of rats (60). It is not known which fat depot produces most leptin in sheep. Finally, it should be noted that metabolic signals to the brain include many other factors apart from leptin (insulin, ghrelin, among others), and the levels of leptin per se may have relatively little impact in terms of signaling to appetite centers, except in times of deficit (12).
In conclusion, we have shown that the pattern of change in VFI that is induced in Soay rams by a change from LP to constant SP is different in gonad-intact and castrated animals. In particular, the biphasic pattern of VFI response that we observed in gonad-intact animals associated with the sequence of reproductive activation and regression suggests that the testicular cycle programs the seasonal cycle of VFI. The changes in VFI can be substantially explained by alterations in the level of expression of the genes for NPY and POMC, without any significant influence of the leptin system. Interestingly, in the absence of gonads, the animals display reduced food intake under prolonged SP but gain adipose tissue associated with a lethargic lifestyle.
The long-term funding for the Soay sheep project was provided by the Medical Research Council, UK. C. Anukulkitch was the recipient of scholarships from the Thai Government and the Faculty of Medicine Siriraj Hospital, Mahidol University, Thailand.
We thank the late Norah Anderson and the staff at the Marshall Building for care of the animals and collection of the blood samples and VFI data, and Margaret Blackberry, Iain Swanston, and Irene Cooper for expert help with the hormone assays.
Present address of C. Anukulkitch, A. Rao, and I. J. Clarke: Dept. of Physiology, Building 13F, Monash University, Clayton VIC 3800, Australia.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society