| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Amylin Pharmaceuticals, Inc., San Diego, California 92121
Address all correspondence and requests for reprints to: Jonathan D. Roth, Ph.D., Amylin Pharmaceuticals, Inc., San Diego, California 92121. E-mail: jonathan.roth{at}amylin.com.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The effects of amylin on food intake and body weight have to date been evaluated only in lean animals consuming a low-fat formulation of chow. Rats consuming high-fat diets become hyperphagic and obese and develop reduced sensitivity to hormonal signals involved in body weight regulation, including leptin (8) and insulin (9). Whether diet-induced obesity (DIO) would render animals resistant to amylins weight-loss effects has not been determined. To test this, the effects of amylin (300 µg/kg·d) on food intake and body weight were assessed in lean Harlan Sprague Dawley (HSD) rats consuming a low-fat diet (standard chow) and then in DIO-prone rats on a moderately high-fat diet. Moreover, the mechanism(s) by which body weight changed after amylin administration in previous studies remains largely unknown. For example, the reduction in body weight could reflect the consequence of the reduced food intake of the animals due to the anorectic properties of amylin, or it could be due to the expression of amylins metabolic effects combined with, or as an alternative explanation to, its anorexigenic effects. Therefore, the studies in DIO-prone rats also incorporated a pair-fed control group, and further mechanistic evaluation of whether amylins effects on body weight and composition, plasma hormones/metabolites, energy expenditure, hypothalamic neuropeptide expression, and other metabolic parameters were independent of its effects on food intake.
| Materials and Methods |
|---|
|
|
|---|
DIO-prone rats were divided into one of three treatment groups (vehicle, pair fed, or amylin) and were implanted with sc minipumps that contained either vehicle (50% dimethylsulfoxide for vehicle and pair-fed rats) or 300 µg/kg·d (76 nmol/kg·d) rat amylin. In previous dose-response work in out-bred DIO rats, 300 µg/kg·d is a near maximal dose for reducing weekly food intake across a 4-wk infusion period (Ref. 11 and our internal observations). Food intake (corrected for spillage) and body weight was recorded daily. Whereas vehicle and amylin-treated rats continued to have ad libitum access to food during the minipump period, the pair-fed rats were restricted to the mean daily intake of the amylin-treated rats. Pair-fed controls were allowed access to food starting approximately 12 h before the onset of the dark cycle. On the final day of the experiment, food was removed from all groups approximately 3 h before the animals were killed. Animals were deeply anesthetized using isoflourane and blood was collected by cardiac puncture, and various tissues were harvested for different end point analyses. Liver and gastrocnemius muscle were freeze clamped in liquid nitrogen. Weights of epididymal, retroperitoneal, sc, and perirenal fat pads (all unilateral) were determined postmortem. Rats used for chemical analyses of body composition (n = 8/group; performed by Covance Laboratories, Madison, WI) were killed by isoflourane overdose; plasma or tissues were not taken from these rats. Brains (n = 4/group) were rapidly removed and frozen in a dry ice bath in freezing media (Tissue Tek, Elkhart, IN; O.C.T.).
Body composition by nuclear magnetic resonance (NMR)
After removal of the minipumps, rats were briefly placed (
1 min) in a well-ventilated plexiglass tube that was then inserted into a specialized rodent NMR machine (Echo Medical Systems, Houston, TX). Rats were scanned before pump implantation and on the final day of the experiment. The change in actual grams of fat and dry lean tissue was calculated (e.g. grams of body fat after treatment grams of body fat at baseline = change in grams of body fat). Dry lean tissue content was calculated by subtracting the values provided by the NMR (lean content water content = dry lean).
Plasma analyses
All determinations were performed using plasma that was collected from rats that had been fasted for approximately 3 h (postabsorptive). Plasma triglycerides, glucose, and cholesterol were measured using a COBAS Mira plasma analyzer (Roche, Stockholm, Sweden). Plasma leptin and insulin were measured using Linco Research kits (St. Charles, MO). Plasma amylin concentrations were quantified using a two-site sandwich immunoenzymetric assay with fluorescent detection.
Indirect calorimetry
Rats were placed in an indirect calorimeter (Columbus Instruments Oxymax equal flow system, Columbus, OH) during the first (n = 8/group; d 46) and third week (d 1621) of treatment (n = 1415/group). Rats were allowed to habituate in the Oxymax cages for 24 h before testing. Measurements were taken over a 24-h period during which the animals had ad libitum access to food (except for pair-fed animals). Respiratory quotient (RQ) and energy expenditure (EE) were calculated and averaged across the 24-h measurement session. Rats were placed in the NMR after the final Oxymax session, dry lean mass was calculated, and EE was calculated per kilogram per dry lean mass. Brown fat pads were also removed from these animals for analyses of uncoupling protein (UCP)-1 mRNA expression.
Liver, muscle, and fat biochemistry
Tissue triglycerides were extracted in chloroform-methanol, using published protocols and quantified using a colorimetric assay (Pointe Scientific, Inc.) (12). Tissue glycogen levels were measured using the amyloglucosidase method (adapted from Ref. 13). A colorimetric microplate assay was used to quantify glucose produced in the presence of glucose oxidase (Pointe Scientific, Inc., Canton, MI). Lipolytic activity (in epididymal fat) was quantified by glycerol release into the incubation medium (adapted from Ref. 14). Basal and isoproterenol-stimulated lipolysis was quantified from fat explants after 4 (vehicle and amylin groups; n = 5/group) or 23 d (amylin, pair-fed and vehicle groups; n = 5/group) of in vivo treatment. Samples were incubated for 60 min in the presence of 0 (basal), 0.3, or 3.0 µM isoproterenol, and a 10-µl sample was collected from the incubation buffer and assayed using a commercially available kit (Sigma Diagnostics, St. Louis, MO).
Liver and muscle gene expression
Liver and epididymal fat samples that had been stored in RNAlater (Ambion, Austin, TX) were homogenized using a FastPrep instrument (Qbiogene, Cambridge, UK). Total cellular RNA was extracted (RNeasy mini and RNeasy lipid kits; QIAGEN, Valencia, CA), DNase digested (DNAfree kit; Ambion) and cDNA was generated using the SuperScript III one-step RT-PCR system (Invitrogen, Carlsbad, CA) for real-time PCR. Oligonucleotide primers and probes specific for acetyl CoA carboxylase (ACC)-1, fatty acid synthase (FAS), ACC2, lipoprotein lipase (LPL), carnitine palmitoyl transferase (CPT)-1, hydroxymethylglutaryl-CoA synthase (HMG-CoA), and 18S were obtained from PE Applied Biosystems (Foster City, CA). The cDNAs (50 ng) were used as a template for each sample. Real-time PCR was completed using standard conditions (PE Applied Biosystems) on an ABI PRISM 7900 sequence detection system.
In situ hybridization (ISH) for neuropeptide Y (NPY), proopiomelanocortin (POMC), and melanin-concentrating hormone (MCH)
To compare central nervous system gene expression across amylin- and pair-fed-treated rats that had exhibited a similar magnitude of weight loss, four animals from each group were selected for ISH, counterbalanced for their final body weight on the day the animals were killed. The group means ± SE for percent baseline body weight for the selected animals were: vehicle 110.5 ± 0.84, pair-fed 99.9 ± 1.0, amylin 98.3 ± 1.1. For the ISH experiments, probe design (NPY, POMC, and MCH), cryomicrotome sectioning, hybridization, and imaging were performed by the custom service Phylogeny, Inc. (Columbus, OH). Briefly, sections (57 µm) containing the arcuate nucleus and the lateral hypothalamic nucleus were identified for ISH. The gene-specific cDNAs for NPY, POMC, and MCH (designed and amplified by RT-PCR from rat brain total RNA), were cloned into pGEMT Easy (Promega, Madison, WI). The specificity of each probe was verified by Northern blot. cDNAs were transcribed to generate 35S-uridine 5-triphosphate-labeled cRNA probes for ISH.
The cRNA transcripts were synthesized according to the manufacturers conditions (Ambion) and labeled with 35S-uridine 5-triphosphate (>1000 Ci/mmol; Amersham, Aylesbury, UK). Sections were hybridized overnight at 52 C in solutions containing 50% deionized formamide and 50,000 to 75,000 cpm/µl 35S-labeled cRNA probe. The tissue was subjected to stringent washes and treated with 20 µg/ml RNase A at 37 C for 30 min. The slides were washed, dehydrated, dipped in NTB-2 nuclear track emulsion (Kodak, Rochester, NY), and exposed for 12 wk in light tight boxes with desiccant at 4 C. Photographic development was carried out in Kodak D-19. Slides were counterstained lightly with toluidine blue and analyzed using both light- and dark-field optics of a Axiophot microscope (Carl Zeiss, New York, NY). Identification of the nuclei of interest was based on relationships to surrounding structures such as the optic chiasma, mamillary bodies, and hippocampus. Sense control cRNA probes (identical with the mRNAs) were run on parallel sections to assess background levels of hybridization signal. Gene expression in NPY, POMC (in the arcuate), and MCH (in the lateral hypothalamus) ISH dark-field images was quantified using integrated OD (IOD; Image-Pro Plus 4.1 analysis software; Media Cybernetics, L.P., Silver Spring, MD) in a consistent region of interest. Mean IOD values were calculated for antisense (two sections/rat) and sense (two sections/rat); final IOD = mean antisense sense. Sections were analyzed at a x10 magnification by an observer blind to the treatment conditions.
Acute food intake in agouti mice
Eight- to 10-wk-old obese agouti KK/Upj-Ay/aJ mice (n = 24) and their lean KK/Upj-a/aJ counterparts (n = 24) were obtained from Jackson Laboratories (Bar Harbor, ME) and housed two per cage at Amylin Pharmaceuticals. At the time of testing, mice had mean body weights of 24.5 g (wild type) and 32.0 g (agouti). Mice were fasted overnight and administered ip injections of either vehicle (saline) or amylin (100 µg/kg). A preweighed portion of food was placed in their cage and cumulative intake (corrected for spillage) was measured 30, 60, 120, and 180 min after injection.
Statistical analysis
Changes in food intake and body weight were analyzed using a repeated-measures ANOVA with post hoc comparisons where indicated. Unless otherwise specified, one-way ANOVA was used to compare other end point analyses. P < 0.05 was considered significant. Graphs were generated using Prism 4 for Windows (GraphPad Software, San Diego, CA). Data are presented as mean ± SE.
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
DIO in rodents is accompanied by characteristic comorbidities of human obesity [increases in fat mass, plasma hormones/metabolites, decreased metabolic rate, etc. (18)]. The in-bred DIO-prone vehicle control animals used in the present study had high levels of carcass fat (>30% by chemical analyses), triglycerides (
400 mg/dl), and leptin (
13 ng/ml). DIO-prone rats also develop hyperinsulinemia, leptin resistance (8), and impaired central insulin signaling (9). Hyperamylinemia has been reported in models of DIO [ranging from 520 pM in lean animals to 1275 pM in obese animals; (19)]. In line with these findings, amylin levels in untreated animals were about 2-fold higher in DIO-prone rats (
10 pM), compared with lean HSD rats (
5 pM). Our results clearly demonstrate that at pharmacological levels, amylin elicits a similar magnitude of vehicle corrected body weight loss in DIO-prone rats as it does in lean HSD rats (both
10%). In DIO-prone rats, amylin therapy also reduced triglycerides by 31%, cholesterol by 10%, insulin by 49%, and leptin by 59%.
Whereas the effects of amylin on food intake and body weight in the present studies were achieved with high levels of amylin (
1000 pM), data from three different acute behavioral assays support that the effects of amylin on food intake in rodents may not be due to malaise or the induction of competing behaviors. First, amylin does not produce a taste aversion across a wide range of doses tested (up to 200 µg/kg) (16, 20, 21). Second, doses of amylin up to 300 µg/kg fail to elicit pica (kaolin consumption), a behavior elicited by agents with known emetic properties such as cisplatin (22). Third, doses of amylin that decrease food intake (0.1100 µg/kg) do not induce competing locomotor activities (i.e. hypo- or hyperactivity) (23). The doses administered in these studies likely achieved higher plasma amylin levels (at least transiently) than our minipump dose (e.g. 300 µg per 24 h), although it is difficult to generalize from assays of acute intake and locomotor behavior to the long-term effects on intake and body weight observed herein. It is notable that in the clinic, approximately 76% of the patients who received the amylinomimetic pramlintide achieved a statistically significant reduction in body weight and never reported any nausea (24).
A key finding in the present studies is that in the face of equal body weight loss in amylin-treated rats and their pair-fed controls, amylin altered body composition by mechanisms not entirely explained by total caloric intake. These changes in composition were not evident in the pair-fed group, as determined by end point chemical carcass composition studies. Furthermore, analysis of body composition before and after treatment (by NMR) confirmed that, over time, caloric restriction alone reduced lean tissue in addition to body fat, whereas amylin treatment produced greater fat loss than caloric restriction alone, without compromising lean tissue. One interpretational caveat with the pair-feeding paradigm used is that over the course of the treatment period pair-fed controls likely consumed larger, less frequent meals relative to amylin-treated rats. Additionally, because amylin slows the rate of gastric emptying, nutrient absorption may also have varied across the groups. To what extent these effects may have contributed to the observed changes in body composition is difficult to discern. Nevertheless, having consumed the same total number of calories across a given period of treatment, amylin-treated rats had a lower percentage of fat and a higher percentage of protein relative to vehicle controls. Whole-animal calorimetry, ex vivo biochemical, and tissue mRNA expression studies were conducted to interrogate whether amylin-induced changes in body composition occurred through increased fat breakdown and/or decreased fat production.
Reduced food intake is expected to decrease RQ and EE. Indeed, during the first week of treatment, amylin decreased RQ to the same extent as pair feeding, indicating equivalent increases in fat oxidation in these groups. EE also tended to be lower in pair-fed controls and higher in amylin-treated rats. When EE was assessed once the rate of body weight change had stabilized, RQ had returned to vehicle control levels; however, EE expressed as a function of total body weight was significantly elevated in amylin-treated rats. These differences in EE were no longer evident when the data were expressed per kilogram dry lean mass because proportionally, amylin-treated rats had more metabolically active tissue than their pair-fed counterparts. Changes in body composition and energetic status were not associated with changes in brown fat UCP-1 mRNA, further supporting the concept that changes in EE in the amylin group were attributable to their relatively higher lean mass. The potential contribution of amylin-induced changes in spontaneous physical activity to the observed changes in energy expenditure was not directly assessed in the present studies. Upon central administration, amylin has been shown to decrease locomotor activity, an effect that would more likely be consistent with a reduction, rather than an enhancement, in energy expenditure (17). In contrast, acute peripheral administration of amylin (up to 300 µg/kg) does not appear to alter locomotor activity (23). Collectively, these findings support the notion that a counterregulatory decrease in EE does not accompany amylin-induced body weight loss.
In isolated adipocyte preparations, amylin stimulates neither lipolysis (25) nor basal or insulin-stimulated rate of glucose incorporation into either CO2 or triacylglycerol (26). We hypothesized that if amylin regulates fat metabolism, these effects are indirect or require sustained exposure. When epididymal fat pads were excised from animals that had been exposed to amylin, the basal rate of lipolysis did not differ across the treated groups. Sustained in vivo administration of amylin also failed to augment the ex vivo lipolytic sensitivity of adipose tissue to adrenergic stimulation. Thus, at least in epididymal fat, amylin did not regulate basal or stimulated lipolysis. We cannot rule out whether lipolysis was altered in retroperitoneal fat, a fat pad that was reduced by amylin treatment. Likewise, we were unable to account for the changes in body composition by differential regulation of genes involved in lipid metabolism or energy balance (ACC1, FAS, ACC2, LPL, CPT-1, HMG-CoA) in peripheral tissues after 3 wk of treatment. One cannot exclude the possibility that changes in these genes may have occurred earlier (i.e. during the active phase of weight loss). At present, the specific biochemical and molecular mechanisms underlying amylins effects on body composition remain to be characterized.
Acute (single) injection of amylin in lean animals binds receptors in the area postrema, activating a transsynaptic circuit comprised of neurons in the nucleus of the solitary tract, the lateral parabrachial nucleus, and the central nucleus of the amygdala (27) Activation of this pathway by amylin inhibits neurons in the lateral hypothalamic area, which are activated during fasting, and down-regulates mRNA expression levels of the intake-stimulating peptide, orexin (6). These observations are consistent with amylins role as a centrally acting satiating hormone. The present studies extend these results to include the modulation of arcuate hypothalamic signals by sustained administration of amylin in DIO-prone animals. Amylin-induced weight loss was accompanied by parallel increases in mRNA levels of NPY and POMC in the arcuate nucleus. That amylin retained its effects on body weight in the face of increased NPY mRNA is consistent with the observation that amylin completely inhibited NPY-stimulated food intake (28). Whereas the changes in NPY could be explained by changes in food intake (similar elevations in pair-fed controls), the changes in POMC were unique to amylin-treated rats. Arguably, these changes in arcuate gene expression could be direct or indirect. Amylin binding sites in the arcuate region have been demonstrated using ex vivo autoradiography (29). Furthermore, arcuate neurons modulate their rate of firing after the application of amylin (30).
The observation that amylin inhibited acute food intake in agouti mice (in which obesity develops due to the central blockade by agouti of hypothalamic melanocortin-4 receptors), together with reports on amylins ability to suppress acute food intake in leptin-deficient (ob/ob) and leptin receptor-resistant (db/db) (31) models, points to circuitry not necessarily dependent on intact leptin and melanocortin signaling. However, further comparison of the effects of amylin across a wider range of doses in agouti mice relative to wild-type mice could be useful in uncovering more subtle interactions between these two systems. Amylin may also have regulated arcuate gene expression in an indirect manner, through activation of the area postrema, nucleus of the solitary tract, lateral parabrachial nucleus, and central nucleus of the amygdala, regions that share rich bidirectional projections with the arcuate nucleus (6). The histaminergic system may also play a contributory role. In rodents, genetic deletion of histamine H1 receptors and the direct infusion of H1 antagonists into the ventral medial hypothalamus both decrease responsiveness to the peripheral effects of amylin on food intake (32, 33). Dissecting out the relative role(s) of each of these interacting neural circuits/signals will require further experimentation. In the face of elevated NPY mRNA levels, amylin retained its effects on body weight and was associated with increased POMC gene expression not entirely explained by changes in food intake and prevailing body weight.
Collectively, the aforementioned effects of amylin (e.g. decreased fat mass, preservation of lean mass, no decrease in EE, and increased POMC) resemble those observed after leptin administration (15). It is tempting to speculate that amylin may help restore or at least increase leptin sensitivity. Given that leptin levels correlate strongly with body adiposity, the greater fat loss evident with amylin treatment suggests that, relative to pair-fed controls, leptin levels were actually better maintained in the presence of amylin (34). In turn, this may have helped drive the changes in body composition, EE, and changes in hypothalamic gene expression evident with amylin treatment. Further studies are warranted to dissect the mechanistic basis for these effects.
In summary, our studies demonstrated that peripheral administration of amylin reduced food intake, slowed body weight gain, and selectively reduced body fat in a rodent model of obesity. Amylin exerted these effects through multiple mechanisms including metabolic alterations and modulation of hypothalamic activity. Finally, the anorexigenic and weight-lowering properties of the amylin analog pramlintide have been demonstrated in obese humans, further supporting the evaluation of amylinomimetics as weight-lowering agents (35).
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online August 24, 2006
Abbreviations: ACC, Acetyl CoA carboxylase; CPT, carnitine palmitoyl transferase; DIO, diet-induced obesity; EE, energy expenditure; FAS, fatty acid synthase; HMG-CoA, hydroxymethylglutaryl-CoA synthase; HSD, Harlan Sprague Dawley; IOD, integrated OD; ISH, in situ hybridization; LPL, lipoprotein lipase; MCH, melanin-concentrating hormone; NMR, nuclear magnetic resonance; NPY, neuropeptide Y; ODU, unit of OD; POMC, proopiomelanocortin; RQ, respiratory quotient; UCP, uncoupling protein.
Received March 28, 2006.
Accepted for publication August 15, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. L. Trevaskis, T. Coffey, R. Cole, C. Lei, C. Wittmer, B. Walsh, C. Weyer, J. Koda, A. D. Baron, D. G. Parkes, et al. Amylin-Mediated Restoration of Leptin Responsiveness in Diet-Induced Obesity: Magnitude and Mechanisms Endocrinology, November 1, 2008; 149(11): 5679 - 5687. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. T. Bello, M. H. Kemm, and T. H. Moran Salmon calcitonin reduces food intake through changes in meal sizes in male rhesus monkeys Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2008; 295(1): R76 - R81. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Roth, B. L. Roland, R. L. Cole, J. L. Trevaskis, C. Weyer, J. E. Koda, C. M. Anderson, D. G. Parkes, and A. D. Baron Leptin responsiveness restored by amylin agonism in diet-induced obesity: Evidence from nonclinical and clinical studies PNAS, May 20, 2008; 105(20): 7257 - 7262. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. Hwang, J. L. Chan, G. Ntali, D. Malkova, and C. S. Mantzoros Leptin Does Not Directly Regulate the Pancreatic Hormones Amylin and Pancreatic Polypeptide: Interventional studies in humans Diabetes Care, May 1, 2008; 31(5): 945 - 951. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Roth, T. Coffey, C. M. Jodka, H. Maier, J. R. Athanacio, C. M. Mack, C. Weyer, and D. G. Parkes Combination Therapy with Amylin and Peptide YY[3 36] in Obese Rodents: Anorexigenic Synergy and Weight Loss Additivity Endocrinology, December 1, 2007; 148(12): 6054 - 6061. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. K. Chelikani, A. C. Haver, and R. D. Reidelberger Effects of intermittent intraperitoneal infusion of salmon calcitonin on food intake and adiposity in obese rats Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2007; 293(5): R1798 - R1808. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Mack, J. Wilson, J. Athanacio, J. Reynolds, K. Laugero, S. Guss, C. Vu, J. Roth, and D. Parkes Pharmacological actions of the peptide hormone amylin in the long-term regulation of food intake, food preference, and body weight Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2007; 293(5): R1855 - R1863. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Roth, H. Hughes, T. Coffey, H. Maier, J. L. Trevaskis, and C. M. Anderson Effects of prior or concurrent food restriction on amylin-induced changes in body weight and body composition in high-fat-fed female rats Am J Physiol Endocrinol Metab, October 1, 2007; 293(4): E1112 - E1117. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. D. Strader, H. Shi, R. Ogawa, R. J. Seeley, and O. Reizes The Effects of the Melanocortin Agonist (MT-II) on Subcutaneous and Visceral Adipose Tissue in Rodents J. Pharmacol. Exp. Ther., September 1, 2007; 322(3): 1153 - 1161. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Aronne, K. Fujioka, V. Aroda, K. Chen, A. Halseth, N. C. Kesty, C. Burns, C. W. Lush, and C. Weyer Progressive Reduction in Body Weight after Treatment with the Amylin Analog Pramlintide in Obese Subjects: A Phase 2, Randomized, Placebo-Controlled, Dose-Escalation Study J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 2977 - 2983. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Smith, J. E. Blundell, C. Burns, C. Ellero, B. E. Schroeder, N. C. Kesty, K. S. Chen, A. E. Halseth, C. W. Lush, and C. Weyer Pramlintide treatment reduces 24-h caloric intake and meal sizes and improves control of eating in obese subjects: a 6-wk translational research study Am J Physiol Endocrinol Metab, August 1, 2007; 293(2): E620 - E627. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||