| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
School of Biological Sciences (K.R., Z.A., F.R.A.C., P.L., J.A.S., T.R.I., A.S.I.L.), University of Manchester, Manchester, United Kingdom M13 9PT; School of Biomedical Sciences (F.J.P.E.), University of Nottingham, United Kingdom NG7 2UH; Fachbereich Biologie/Zoologie (M.K.), Philipps Universtat Marburg, D-35043 Marburg, Germany
Address all correspondence and requests for reprints to: Andrew S. I. Loudon, School of Biological Sciences, 3.614 Stopford Building, University of Manchester, Oxford Road, Manchester M13 9PT, United Kingdom. E-mail: . andrew.loudon{at}man.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, attention has focused on the role played by the adipose tissue-derived hormone leptin, which is strongly implicated as one of the major peripheral signals controlling body fat reserves and appetite in mammals (6). In the Siberian hamster, leptin mRNA levels in adipose tissues reflect seasonal changes of fat reserves, showing maximal levels in fat hamsters housed in summerlike long-day (LD) conditions (7), and seasonal changes in leptin gene expression are associated with significant 2- to 4-fold increases in serum leptin concentrations on LD (8, 9, 10, 11). Conversely, the decline in body weight in SD is associated with a marked decrease in leptin levels. This seemingly paradoxical observation may reflect seasonal changes in leptin sensitivity or responsiveness. Our previous studies have investigated responsiveness of hamsters to leptin on long and short photoperiods, in which recombinant murine leptin was infused at physiological doses over a 7- to 14-d period (8). These studies revealed that LD-housed hamsters were refractory to leptin treatment, whereas SD animals exhibited further substantial weight and abdominal fat loss. Intriguingly, the leptin-induced reductions of body weight and abdominal fat in SD were not accompanied by changes in food intake, indicating a direct effect of leptin on energy balance. This species thus appears to be resistant to leptin in LD photoperiods, consistent with the observation that it gains weight and increases appetite at this phase of seasonal cycle.
The regulation of body weight and energy homeostasis involves many hypothalamic neuropeptide and receptor systems that are known to be regulated by leptin (6, 12, 13). Leptin receptors are localized in hypothalamic regions associated with appetite and energy metabolism, including the arcuate nucleus (ARC) in the ventral hypothalamus in which two competing neuropeptide systems are involved in energy balance regulation. Orexigenic peptides such as neuropeptide Y (NPY) and agouti-related peptide (AGRP) act on anabolic pathways, whereas a catabolic pathway operates via populations of proopiomelanocortin (POMC) and cocaine- and amphetamine-regulated transcript (CART) neurons in the ARC. These two pathways project to other hypothalamic structures involved in energy balance regulation, including the paraventricular nucleus and the lateral hypothalamus. Each peptide pair is known to be coexpressed in the same cell type in the ARC, and both POMC and NPY neurons are known to coexpress the leptin receptor (6, 12, 13).
Here we describe experiments designed to test the hypothesis that resistance to leptin in (fat) LD animals is a consequence of photoperiod-induced changes in ovarian steroid and leptin secretions. Our experiments were performed in ovariectomized estrogen-treated females as many studies have reported the influence of gonadal steroids on food intake and body weight by modulating the central nervous system effectors of energy homeostasis as well as the production of leptin (14). If the effects of photoperiod on leptin responsiveness are secondary to the effects of photoperiod on adiposity, it would be predicted that these changes would be mimicked by food-restricting animals to the body weight of each photoperiod. Thus, we examined leptin responsiveness on LD-housed females that were ovariectomized and given an estrogen implant and fed ad libitum, food restricted to the weight of SD-housed females, or food restricted and refed during the time of leptin infusion. Finally, we assessed the response of LD-housed hamsters to acute ip leptin treatment to determine whether LD-housed hamsters were entirely refractory to the effects of leptin.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hamsters were ovariectomized and a 31-µg estrogen implant was administered sc (OVX+E2) as previously described (8), so that effects of leptin, season, and food restriction could be assessed independently of seasonal changes in gonadal steroid feedback. All surgical procedures were undertaken at least 1 wk prior to the commencement of photoperiod or other treatments (experiments 1–3).
Leptin infusion protocols
Recombinant murine leptin (supplied by Amgen, Inc., Thousand Oaks, CA) was dissolved in 0.01 M PBS and administered via an osmotic minipump (100-µl capacity; Alzet model 1007D, Charles River Laboratories, Inc., Kent, UK) to deliver a dose of 15 µg/d per animal for a 7-d period. Control animals received PBS vehicle alone. Pumps were implanted sc in the scapular region by sterile surgical procedure under halothane (Fluothane, AstraZeneca Ltd., Cheshire, UK) anesthesia. New pumps were reimplanted on d 7 to provide a 14-d infusion period.
Experimental protocols
Experiment 1: defining the effects of photoperiod and leptin infusion on neuropeptide gene expression.
We tested the hypothesis that seasonal changes in body weight and fat reserves were mediated by photoperiodic modulation of neuropeptide gene expression in the ARC.
Twenty-four weight-matched 16-wk-old OVX+E2 female hamsters reared in LD were individually housed and exposed to LD or SD conditions for an 8-wk period, as previously described (8). At wk 9, each animal received an osmotic minipump containing either leptin or PBS (six animals per group per treatment). Pumps were replaced after 7 d for a further 1-wk period. On d 14 of treatment, hamsters were killed by cervical dislocation, and the brain was rapidly removed, frozen in dry ice, and stored at -80 C.
Experiment 2: defining the effects of food restriction and leptin treatment on gonadotropins and hypothalamic gene expression.
We tested the hypothesis that refractoriness to leptin action for LD-housed animals was a consequence of LD-induced elevation of leptin concentrations.
Before the initiation of food restriction (FR), daily food consumption/intake was measured during the preceding 5-d period for each individual animal to determine baseline levels before level of food restriction. Twenty-four weight-matched 12-wk-old OVX+E2 hamsters were kept on an LD photoperiod and either ad libitum fed (control group) or food restricted (by 40%) to reduce body weights to 26.0 ± 0.1 g, similar to those observed in animals exposed to SD for 8 wk [25.2 ± 0.7 g (8)]. This food restriction paradigm achieved SD-like body weights and fat reserves at different times ranging from 7–11 d following FR, according to an individual variation in the rate of response. As soon as an animal attained SD-like body weight, a 7-d osmotic minipump containing either leptin or PBS was implanted, as described above. On d 7, hamsters were anesthetized, and blood samples taken by cardiac puncture for serum leptin, LH, and FSH assays. Animals were killed by cervical dislocation, the brain rapidly removed, frozen in dry ice, and stored at -80 C. The abdominal fat pads were dissected and weighed.
Experiment 3: effects of refeeding of food-restricted animals and action of leptin.
We tested the hypothesis that leptin resistance was regulated by LD photoperiod by examining whether leptin was effective in suppressing the refeeding response of previously food-restricted LD-housed animals.
Baseline food intake was determined as described for Exp 2. Eight weight-matched 12-wk-old OVX+E2 hamsters were kept in LD and food restricted by 40% to reduce body weights to 8-wk SD levels. When an individual animal’s body weight had declined to a SD-like body weight, this animal received an osmotic minipump containing either leptin or PBS and was allowed to refeed ad libitum for 7 d. On d 7, hamsters were killed by cervical dislocation and the abdominal fat pads dissected and weighed.
Experiment 4: response of LD-housed hamsters to acute ip leptin treatment.
The above experiments revealed that LD-housed hamsters were not responsive to physiological doses of leptin. This experiment aimed to determine whether LD-housed animals exhibited relative or absolute refractoriness to leptin action.
For this acute study, seven LD-housed intact female hamsters were used. On d 1, all animals were fasted from 0700 h–1800 h, when lights were turned on. They were then fed for the next 6 h and food measurements made at intervals of 1 h, 3 h, and 6 h. Food was then withdrawn at 2400 h and animals fasted for the next 18 h. On the second day, animals were divided into two groups, one injected ip with leptin (14 mg/kg, n = 4) and the other one with PBS (n = 3), 30 min before lights off (1800 h). At lights off, food was placed in the food chambers and food intake determined at the same intervals as on d 1. Animals were then killed by cervical dislocation.
Hormone assays
LH.
A double-antibody RIA was used to determine serum LH concentration, as previously described for use in Syrian hamsters (15). RIA reagents were supplied by the National Hormone and Pituitary Program (NIDDK, Bethesda, MD) and employed rabbit antirat LH antibody S-11 and rat LH RP-3 reference preparation. The mean assay sensitivity was 0.125 ng/ml. The intraassay coefficients of variation were 1.52% at 2.96 ng/ml, 2.21% at 10.49 ng/ml, and 3.02% at 24.8 ng/ml. The interassay coefficients of variation were 6.52% at 2.81 ng/ml, 6.25% at 9.17 ng/ml, and 5.78% at 24.06 ng/ml. Serial dilutions of female Siberian hamster serum pools revealed concentration curves parallel to a standard LH curve in the RIA (data not shown). The LH concentration in serum was expressed as nanograms of LH per milliliter.
FSH.
A double-antibody RIA was used to determine serum FSH concentration, as previously described for use in Siberian hamsters (16, 17). RIA reagents were supplied by the National Hormone and Pituitary Program (NIDDK, Bethesda, MD) and employed rabbit antirat FSH antibody S-11 and rat FSH RP-2 reference preparation. The mean assay sensitivity was 0.75 ng/ml. The intraassay coefficients of variation were 4.93% at 12.8 ng/ml and 1.97% at 45.6 ng/ml. The interassay coefficients of variation were 11.34% at 11.02 ng/ml and 4.38% at 45.38 ng/ml. Serial dilutions of female Siberian hamster serum pools revealed concentration curves parallel to a standard FSH curve in the RIA (data not shown). The FSH concentration in serum was expressed as nanograms of FSH per milliliter.
Leptin.
Serum leptin concentrations were determined using the Linco multispecies kit (Linco Research, Inc., St. Charles, MO), as previously described for use in Siberian hamsters (9, 11). The limit of sensitivity was 2 ng/ml (50-µl sample size). The intraassay coefficient of variation was less than 1%. All samples were run in a single assay. Serum dilutions from hamsters housed in long photoperiod exhibited clear parallelism over the full range of physiologically detectable doses (data not shown). Serum leptin values were determined in a single RIA and expressed as nanograms of leptin per milliliter.
Hypothalamic neuropeptide gene expression
Oligonucleotide probes.
The 36-mer NPY probe [5'-GGA GTA GTA TCT GGC CAT GTC CTC TGC TGG CGC GTC-3'] was complementary to amino acids 40–51 of the rat preproNPY (18); its use in Siberian hamsters has been previously described (19, 20). The 42-mer POMC probe [5'-CAG GGC CCC TGA GCG ACT GTA GCA GAA TCT CGG CAT CTT CCA-3'] was complementary to a region completely conserved between rat and mouse sequences; its use in Siberian hamsters has been previously described (20). The 45-mer CART probe [5'-GGA CGC ATC ATCCAC GGC AGA GTA GAT GTC CAG GGC TCG GGG CTG-3'] was based on a conserved sequence in the rat and human. Its use in Siberian hamsters has been previously shown (21). Sequences were obtained and conserved regions identified using GenBank and BLAST software (http://www.ncbi.nlm.nih.gov/). The oligonucleotide probes (5 pmol) were end-labeled using terminal deoxynucleotide transferase (Roche, Lewes, UK) in the presence of 50 µCi 
-35S-dATP.
Ribonucleotide probes.
AGRP cDNA fragment was amplified from Siberian hamster hypothalamic cDNA by PCR (cycling conditions: single denaturation cycle of 94 C for 10 min, followed by 40 cycles of denaturation at 94 C for 45 sec, annealing at 55 C for 1 min, and extension at 72 C for 1 min, and then finally one cycle at 72 C for 10 min). Amplification products were ligated into PCRII-TOPO (CLONTECH Laboratories, Inc., Basingstoke, UK) plasmid, transformed into competent cells and sequenced. The 233-bp fragment of hamster AGRP was amplified using primers [5'-TCCCAGAGTTCCAGGTCTA-3'] and [5'-GCAAAAGGCATTGAAGAAGC-3'] from mouse sequence for the AGRP gene (GenBank U89484). Both primers were designed in conserved regions among species. The fragment was 84% identical with human AGRP mRNA and 93% to mouse AGRP mRNA. Sense and antisense riboprobes for hamster AGRP were prepared by linearizing the pCRII-TOPO plasmid containing the 233-bp hamster AGRP cDNA with HindIII and EcoRV, respectively, and labeled with 35S-labeled UTP.
In situ hybridization protocol.
Coronal sections containing the hypothalamus based on the rat brain atlas (22) were cut (15 µm thickness) on a cryostat. Because neuropeptide gene expression has been shown to be differentially regulated through the rostral-caudal projection of the ARC (23, 24), three approximate regions were defined: rostral from bregma -1.8 to -2.3 mm, mid from bregma -2.3 to -3.14 mm, and caudal from bregma -3.14 to -4.16 mm. These sections were thaw mounted onto APES-coated slides. They were then fixed in 4% paraformaldehyde for 15 min followed by one wash in PBS and then acetylated with 0.025 M acetic anhydride in 0.1 M triethanolamine/0.9% NaCl for 10 min to minimize nonspecific hybridization of the probes. After dehydration with increasing concentrations of ethanol (70%, 95%, and 100%), the sections were delipidated in chloroform for 5 min, rinsed in 95% ethanol, and air dried.
Oligoprobes (NPY, POMC, and CART).
Brain sections were hybridized at 37 C overnight with 5 x 105 cpm of 35S-dATP labeled oligoprobe in 30 µl hybridization buffer containing 4x saline sodium citrate (SSC), 50% deionized formamide, 1x Denhardt’s solution, 100 µg/ml salmon/herring sperm DNA, 10% dextran sulfate, 100 µg/ml poly A+ RNA, and 20 mM dithiothreitol. Following hybridization, sections were rinsed twice for 10 min each in 1x SSC at room temperature, then twice for 30 min in 1x SSC at 50–55 C, and washed for 5 min in 1x SSC at room temperature. Finally, sections were dehydrated in increasing concentrations of ethanol (70% and 95%).
Riboprobe (AGRP).
Brain sections were hybridized overnight at 45 C with 1 x 106 cpm of 35S-UTP-labeled hamster AGRP riboprobe in 30 µl hybridization buffer containing 2x SSC, 50% deionized formamide, 10 mM Tris-HCl, 1x Denhardt’s solution, 0.2% SDS (lauryl sulfate), 500 µg/ml salmon/herring sperm DNA, 250 µg/ml transfer RNA, 10% dextran sulfate, and 100 mM dithiothreitol. The following day, sections were rinsed twice for 15 min each in 4x SSC/20 mM mercaptoethanol, twice for 15 min in 2x SSC/20 mM mercaptoethanol, and twice for 15 min in 1x SSC/20 mM mercaptoethanol. Sections were then treated with 20 µg RNase A per ml of 0.5 M NaCl/1 mM EDTA/10 mM Tris, pH 8.0, for 30 min at 37 C. After rinsing for 15 min in 0.5 M NaCl/1 mM EDTA/10 mM Tris, pH8, slides were washed in 0.5x SSC for 30 min at 45 C. Finally, sections were dehydrated in increasing concentrations of ethanol (50% and 70%) containing 300 mM ammonium acetate, and finally 95% ethanol.
Slides were then air dried and exposed to autoradiographic film (Kodak Biomax MR-1, Amersham) at 4 C (11 d for NPY and POMC, 4 d for CART, and 4 wk for AGRP). Autoradiographic images were quantified using Image-Pro Plus Image analysis software (DataCell Ltd., Finchampstead, UK). Analysis was undertaken blind to the different treatment groups by the investigator. Results were expressed as relative OD (ROD), which was calculated by subtracting the relevant background value to the mean gray value measured within the ARC of each section. Sections pretreated with RNase A were used as controls (data not shown).
Statistical analysis
Data were analyzed by t tests, two-way ANOVA, or two-way repeated measures of ANOVA followed by post hoc Student-Newman-Keuls multiple comparison tests, using SigmaStat statistical software (SPSS, Inc., Chicago, IL). Results are presented as the mean ± SEM. Differences were considered statistically significant at the P less than 0.05 level.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). We have previously shown in these animals that short photoperiods induce a significant reduction in fat depot, food intake, and serum leptin and LH concentrations (8). Leptin treatment significantly reduced body weight in SD-housed animals but was without significant effect in hamsters on long photoperiods (Fig. 1b). This treatment also significantly reduced abdominal fat reserves in these SD-housed animals and not in the LD-housed hamsters, as previously reported (8).
|
and 3a).
|
|
and 3b).
AGRP gene expression (Figs. 2c and 3c).
There was no significant effect of photoperiod in the rostral (P = 0.085; two-way ANOVA), mid (P = 0.306; two-way ANOVA), and caudal (P = 0.511; two-way ANOVA) regions of the ARC. No significant effect of leptin infusion could be found in any of the ARC regions. There were also no significant interactions between photoperiod and leptin treatment in any of the ARC regions (rostral: P = 0.735; mid: P = 0.637; caudal: P = 0.286).
CART gene expression (Figs. 2d and 3d).
No significant effect of photoperiod could be observed in the rostral (P = 0.193; two-way ANOVA), mid (P = 0.494; two-way ANOVA), and caudal (P = 0.699; two-way ANOVA) regions of the ARC. There was no significant effect of leptin infusion in any of the ARC regions. Also, no significant interactions could be found between photoperiod and leptin treatment in any of the ARC regions (rostral: P = 0.609; mid: P = 0.409; caudal: P = 0.185).
Experiment 2: Effects of food restriction and leptin treatment on LD-housed animals
Serum leptin concentrations.
There was a significant effect of both leptin infusion (P < 0.01; two-way ANOVA) and FR (P < 0.01; two-way ANOVA) on serum leptin concentrations (Fig. 4).
|
]. Because animals achieved a SD-like body weight at different rates, the data in Fig. 5 are shown for d 1–9 FR and then for the subsequent 7-d infusion period.
|
; P = 0.227, two-way repeated measures of ANOVA). Body weights of LD ad libitum-fed animals remained stable throughout the experimental duration (18 d) and no effect of leptin was observed (Fig. 5). At the end of the experiment, there was a significant effect of FR on body weight (P < 0.05; by two-way ANOVA) but no action of leptin (P = 0.227; two-way ANOVA) (Fig. 6a).
|
). Food-restricted LD hamsters lost 0.79 ± 0.01 g of fat reserves. In contrast, there was no significant effect of leptin treatment on the rate of abdominal fat loss (P = 0.776, two-way ANOVA; Fig. 6b).
Reproductive hormones.
The ad libitum-fed animals had significantly higher serum LH concentrations than food-restricted animals (P < 0.05; two-way ANOVA; Fig. 7a). Leptin treatment did not reverse the suppressive effect of food restriction on LH concentrations.
|
). There was no significant action of leptin on FSH concentrations in either fed ad libitum or food-restricted animals.
Hypothalamic neuropeptide gene expression
NPY gene expression (Figs. 2a and 8a).
There was a significant increase in NPY mRNA levels following food restriction in the rostral (326%; P < 0.001; two-way ANOVA), mid (66%; P < 0.05; two-way ANOVA), and caudal (235%; P < 0.001; two-way ANOVA) regions of the ARC. Leptin infusion did not significantly change NPY levels in any of the ARC regions (rostral: P = 0.659; mid: P = 0.106; caudal: P = 0.499; two-way ANOVA). There were no statistically significant interactions between FR and leptin treatment in any regions of the ARC (rostral: P = 0.135; mid: P = 0.381; caudal: P = 0.388).
|
and 8b).
AGRP gene expression (Figs. 2c and 8c).
There was a significant increase in AGRP mRNA levels brought about by FR in the caudal region of the ARC (155%; P < 0.01; two-way ANOVA) but not in the mid (P = 0.151; two-way ANOVA) or rostral (P = 0.114; two-way ANOVA) regions. Leptin infusion did not induce any changes in AGRP expression in any of the ARC regions (rostral: P = 0.505; mid: P = 0.169; caudal: P = 0.412; two-way ANOVA). There were no statistically significant interactions between FR and leptin treatment in any regions of the ARC (rostral: P = 0.158; mid: P = 0.264; caudal: P = 0.230).
CART gene expression (Figs. 2d and 8d).
There was a significant reduction in CART gene expression by FR in the mid region of the ARC (21%; P < 0.05; two-way ANOVA) but not in the rostral (P = 0.165; two-way ANOVA) or caudal (P = 0.495; two-way ANOVA) regions. Leptin infusion did not induce any changes in CART expression in any of the ARC regions (rostral: P = 0.355; mid: P = 1.000; caudal: P = 0.322; two-way ANOVA). A significant interaction between FR and leptin treatment was found in the caudal ARC (P < 0.05; two-way ANOVA) but not in the other regions of the ARC (rostral: P = 0.821; mid: P = 0.622).
Experiment 3: Effects of FR/refeeding and leptin treatment
Here OVX+E2 female hamsters maintained in LD were food restricted by 40% to reach SD-like body weight, but then refed ad libitum while receiving either leptin or PBS for 7 d.
Food intake measured before FR did not exhibit significant differences between both groups (3.4 ± 0.2 g/d vs. 3.7 ± 0.2 g/d) (Fig. 9a). The 40% FR paradigm during 7–11 d achieved SD-like body weights of 25.8 ± 0.2 g. At the time of refeeding, animals of both groups increased their food intake (PBS treated: 3.7 ± 0.9 g/d vs. 2.1 ± 0.1 g/d during FR; leptin treated: 3.6 ± 0.5 g/d vs. 2.2 ± 0.1 g/d during FR). There were no significant differences in food intake at any time point during the leptin infusion/refeeding period between PBS and leptin groups (Fig. 9a).
|
), leptin-treated animals regained weight (25.7 ± 0.3 g before infusion vs. 29.7 ± 0.7 after the 7 d of infusion) at a similar rate as PBS-treated animals (25.8 ± 0.2 g before infusion vs. 30.2 ± 1.1 g after the 7 d of infusion). No significant differences in body weight were found at any time point during this period between PBS and leptin groups (Fig. 9b). Leptin treatment did not alter the abdominal fat depot weights (PBS treated: 1.1 ± 0.2 g vs. leptin treated: 1.07 ± 0.2 g) after the 7-d infusion period (Fig. 9c).
Experiment 4: Does acute leptin treatment blunt refeeding?
On d 1 (no leptin treatment), animals showed no significant differences in food intake at 1, 3, or 6 h after refeeding (Fig. 10a). On d 2, acute leptin treatment (14 mg/kg body weight) significantly reduced cumulative food intake by 37% 6 h after refeeding (P = 0.01; t test), compared with PBS-treated animals; no significant differences were observed 1 and 3 h after refeeding (Fig. 10b).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Following an 8-wk exposure to SD, expression was reduced by approximately 2-fold throughout the arcuate nucleus. Because these studies were undertaken in steroid-clamped animals, our results rule out the possibility that LD length influences energy balance and ARC POMC gene expression indirectly via changes in ovarian steroid secretion. These data contrast with the earlier studies of Syrian hamsters, in which SD-induced changes in POMC expression were dependent on seasonal changes in sex steroid secretion rather than photoperiod (24). Other studies in intact Siberian hamsters have also reported significant reductions in POMC expression after 8–18 wk of SD exposure (20, 25, 26, 27) and are 2consistent with our present data. In all of these studies, the direction of change of POMC mRNA levels is counterintuitive because gene expression is reduced when animals are in a catabolic state of reduced energy expenditure. Currently there is no direct evidence linking ARC POMC gene products to changes in overall seasonal energy metabolism, although products of this gene are potentially catabolic in the Siberian hamster (28). These data support the hypothesis that changes in plasma leptin concentrations and POMC gene expression are the result and not the cause of negative energy balance. It is possible that SD-induced down-regulation of POMC gene expression is counteracted by changes in other neuropeptide circuits to induce an overall state of weight loss.
|
Data on small seasonal rodents contrast with studies in seasonally breeding sheep. Here both NPY and POMC expression in the ARC have been shown to be modulated by photoperiod in ovariectomized ewes, with reduced NPY expression at the time of naturally reduced appetite and increased POMC expression in late summer (31). These seasonal changes in NPY are not dependent on changes in sex steroid secretion because they can be observed in castrates (32). There may therefore be considerable species differences in the seasonal expression of these hypothalamic peptides and in the relative role of sex steroids and photoperiod. In sheep, deer, and other large seasonal mammals, high amplitude seasonal cycles of food intake are common (2- to 4-fold) with small or nonsignificant seasonal changes in basal energy metabolism (33, 34). In rodents, seasonal changes in body fat content may be regulated primarily by changes in the pattern of energy expenditure including the onset of daily winter torpor cycles characteristic of small seasonal rodents (35), with less significant changes in food intake. Indeed, our earlier study (8) has shown that adipose tissue loss in SD leptin-treated animals was not associated with altered food intake, implying a direct action of leptin on energy metabolism.
We observed significant physiological effects of leptin on body weight and abdominal fat mass, which were confined to SD-housed animals. In contrast, this treatment elicited a small reduction in NPY expression only in the rostral region of the ARC for both LD- and SD-housed animals (see summary in Table 1). Because leptin decreased NPY expression to a similar extent on LD and SD photoperiods, it is most unlikely that leptin action on SD is mediated by changes in this peptide. Our later comparison of LD-housed ad libitum-fed animals showed no effect of leptin, and this may be attributable to the length of prior leptin treatment (14 and 7 d, respectively) in these two studies. The 14-d leptin infusion paradigm had no other detectable effect on gene expression in SD or LD-housed animals. Our data are consistent with the reported effects of acute leptin treatment on SD and LD-housed male hamsters (20). Here a single supraphysiological dose of about 6 mg/kg body weight failed to elicit significant changes in NPY or POMC gene expression in the mid-ARC region 6 h after treatment, despite inducing significant reductions in food intake. Taken together, these data suggest that both chronic and acute leptin treatment has relatively little effect on gene expression of these peptides in the ARC of Siberian hamsters. These results do not support the hypothesis that seasonally induced changes in sensitivity to leptin are mediated by photoperiod induced changes in NPY, POMC, CART, or AGRP gene expression.
Several mechanisms have been proposed for leptin resistance, including impaired leptin transport mechanism across the blood-brain barrier (36, 37), reduced leptin-receptor signal transduction (38, 39), or down-regulation of expression of leptin receptor (40, 41). In addition, a failure in one or more of the downstream neuronal systems responsive to leptin may also contribute to leptin resistance (42). Within the hypothalamus of the Siberian hamster, both forms of leptin receptor have been shown to hybridize to the ARC, ventromedial nucleus, dorsomedial nucleus, and lateral hypothalamic area (25), which is consistent with localization in other rodent species. Studies by Mercer et al. (25, 27) suggest that leptin receptor gene expression in the ARC is differentially regulated by acute 24-h food deprivation (increases expression) and chronic long-term responses to SD photoperiods (reduces expression). Indeed, the abundance of mRNA encoding the long form of the leptin receptor (Ob-Rb) in the ARC is decreased on SD at a time when these animals exhibit depressed leptin mRNA levels in adipose tissues and reduced serum leptin concentrations (25, 27), and our previous studies have revealed greater sensitivity to leptin treatment on such SD photoperiods (8, 10). Intriguingly, although chronic FR leads to reduced adiposity and leptin concentrations in plasma (this study, and Ref. 27), Ob-Rb expression is increased in food-restricted animals in both LD and SD photoperiods (27). Thus, Ob-Rb expression exhibits opposite responses for food-restricted LD- and SD-housed animals vs. SD-housed animals (which undergo natural declines in food intake), lending further weight to the argument that this receptor may be regulated by primary photoperiodic processes rather than circulating leptin concentrations per se.
Effects of FR and leptin infusion on LD-housed animals
FR in LD hamsters caused a significant reduction in body weight, fat mass, and circulating leptin similar to those observed after 8–10 wk in SD. This was accompanied by significant declines in gonadotropin concentrations but, unexpectedly, was not reversed by leptin treatment. It is possible that the regimen of FR adopted induced secondary effects on the hypothalamic-pituitary-adrenal axis (43), which are not measured here or in previous studies employing a similar regimen (10, 27). However, in a number of mammals, FR is known to induce decreased gonadotropin secretions (44) and also to reduce leptin concentrations (i.e. mice: 45, 46, 47 ; mare: 48 ; human: 49). Several studies have now shown that leptin administration can restore pulsatile LH secretion during fasting (i.e. rat: 50, 51, 52 ; mice: 46 ; rhesus macaque: 53).
Our work clearly shows that restoration of leptin concentrations to normal LD levels in food-restricted Siberian hamsters does not restore gonadotropins. A similar lack of action of leptin in reversing gonadotropin suppression has also been reported in chronically food-deprived ewes (54), and a study on rats reveals that leptin can reverse the effects of undernutrition on puberty following mid (20%) but not severe (30%) FR (55). In our study, animals were food restricted for relatively long periods of time before imposition of leptin treatment, but in the studies referred to above (47, 50, 51, 52, 53), animals were subjected to relatively short periods of food deprivation and simultaneously treated with leptin. It is possible that leptin is less effective or even ineffective in reversing nutritionally induced infertility once it has been established in chronically undernourished animals. The action of leptin on reproduction may be ultimately constrained by the availability of metabolic fuels from intake and/or mobilized fat stores (56). Evidence in support of this hypothesis derives from studies of fasting-induced anestrus in Syrian hamsters. Here leptin treatment is capable of reversing the acute effects of food restriction on estrous cyclicity, but simultaneous treatment with an inhibitor of glucose or fatty acid oxidation blocks the action of leptin (57). These studies are therefore consistent with the hypothesis that leptin may increase metabolic fuel availability but only when food intake and/or fat reserves are above a critical level. In extreme food deprivation (i.e. reduction to winter weights of LD-housed Siberian hamsters), these fuels would not be available for mobilization by leptin.
FR of LD-housed animals causes a significant elevation in NPY and AGRP and decrease in POMC and CART in the ARC. This is consistent with the results of other studies on this species (20, 25, 27, 29) and reveals that both the anabolic and catabolic pathways are sensitive to nutritional feedback effects. Similar observations have also been widely reported for a number of other mammalian species (i.e. mice: 58 ; rat: 23 , 59, 60, 61 ; sheep: 62). In many studies, leptin treatment has been reported to reverse the effects of food restriction on altered hypothalamic gene expression (i.e. rat: 23 , 60 ; mice: 58 ; sheep: 63). In marked contrast, our data clearly show that use of a leptin infusion paradigm, which we have shown induces significant effects on energy metabolism of SD-housed animals, is without effect in terms of altered pattern of hypothalamic gene expression in food-restricted LD-housed animals. Because our FR protocol also reduced leptin to SD concentrations, we concluded that the lack of responsiveness of food-restricted animals is not likely to be attributable to high LD-like endogenous circulating leptin concentrations.
Leptin administration at the time of refeeding in food-restricted animals does not modulate the pattern of food intake. This lack of responsiveness to leptin in LD-housed hamsters contrasts with similar studies undertaken in SD-housed hamsters (10). Here the refeeding response and recovery of body mass of hamsters was significantly blunted by administration of leptin (10). Together these data strongly suggest that leptin responsiveness of SD animals and the failure of LD-housed animals to exhibit reproductive, feeding behavioral, or neuropeptide gene expression responses to leptin may be attributable to a primary action of LD photoperiods rather than 2changes in endogenous leptin concentrations per se. It is also clear that apparent refractoriness to leptin on LD photoperiods is relative rather than absolute because we have also shown that a single supraphysiological ip dose of leptin is capable of suppressing the pattern of food intake in food-restricted hamsters housed on LD.
Does leptin play a role in the seasonal energy cycle of the Siberian hamster?
Our data support the hypothesis that responsiveness to leptin at physiological concentrations may be modulated by photoperiod. This raises an important question as to whether this hormone plays a role in regulating the seasonal sliding set-point observed in this species (1). We show in our study an apparent absence of response to leptin on LD photoperiods, and this argues against such a role. Further, studies of SD-housed animals in which leptin blunts the refeeding responses following food restriction also clearly show that such treatments do not reset the overall timing of seasonal changes in adiposity (10). It therefore appears that leptin action, although gated by photoperiod, is unlikely to play a role in timing seasonal set-point changes in the Siberian hamster. The converse argument is that photoperiod changes seasonal set point, such that animals exhibit a biologically appropriate response to endogenous concentrations at that time of year. This may explain the paradox of enhanced leptin responses on SD, a time when animals would never naturally be exposed to such an endocrine signal (8, 10, 11).
The ARC is rich in leptin receptors, and in Siberian hamsters, high expression levels have been reported (25). Remarkably, it appears that destruction of over 80–90% of the arcuate neurons in Siberian hamsters (using neonatal monosodium glutamate) does not block robust photoperiodic responses in terms of body weight changes and reproduction (19). A clear implication therefore is that seasonal photoperiodic regulation of both body weight and reproductive activity may depend on other hypothalamic pathways. In our study, we were able to observe significant changes in ARC gene expression in food-restricted animals for both orexigenic and anorexic peptides but not altered expression of these peptides using a leptin infusion regimen, and we did not observe changes in ARC gene expression in SD-housed leptin-treated animals, despite a significant physiological response. Together these data suggest that leptin action in the Siberian hamster may well target hypothalamic structures other than the ARC. In this respect, seasonal rodents may differ significantly from large seasonal ruminants such as sheep, in which, for instance, complete destruction of the ventral region of the hypothalamus including the ARC region results in loss of photoperiodic seasonal cycles of food intake and body weight (64).
In conclusion, our data strongly suggest that seasonal photoperiod change, not the change in sex steroids, body condition, or circulating leptin concentrations, is the important factor influencing leptin feedback action in the Siberian hamster.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This work was supported by a research grant awarded by the Biotechnology and Biological Sciences Research Council (to A.S.I.L. and F.R.A.C.) (34/S12833), a Biotechnology and Biological Sciences Research Council Ph.D. studentship (to Z.A.), and AstraZeneca Pharmaceuticals (Alderley Edge, Cheshire, UK).
Abbreviations: AGRP, Agouti-related peptide; ARC, arcuate nucleus; CART, cocaine- and amphetamine-regulated transcript; FR, food restriction; LD, long-day photoperiod; NPY, neuropeptide Y; OVX+E2, ovariectomized and a 31-µg estrogen implant administered; POMC, proopiomelanocortin; ROD, relative OD; SD, short-day photoperiod; SSC, saline sodium citrate.
Received January 22, 2002.
Accepted for publication April 22, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
