| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Medical Department M (Endocrinology and Diabetes) (E.T.V., C.B.D., J.G., S.N., N.M., J.O.L.J.), Aarhus University Hospital, DK-8000 Aarhus C, Denmark; Department of Biomedical Sciences (J.J.H.), the Panum Institute, University of Copenhagen, DK-2200 Copenhagen, Denmark; and Department of Pharmacology (O.S.), University of Aarhus, DK-8000 Aarhus, Denmark
Address all correspondence and requests for reprints to: Esben Thyssen Vestergaard, M.D., Medical Department M (Endocrinology and Diabetes), Aarhus University Hospital, DK-8000 Aarhus C, Denmark. E-mail: e.t.vestergaard{at}ki.au.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to study direct effects of ghrelin on substrate metabolism.
Design: This was a randomized, single-blind, placebo-controlled two-period crossover study.
Setting: The study was performed in a university clinical research laboratory.
Participants: Eight healthy men aged 27.2 ± 0.9 yr with a body mass index of 23.4 ± 0.5 kg/m2 were included in the study.
Intervention: Subjects received infusion of ghrelin (5 pmol·kg–1·min–1) or placebo for 5 h together with a pancreatic clamp (somatostatin 330 µg·h–1, insulin 0.1 mU·kg–1·min–1, GH 2 ng·kg–1·min–1, and glucagon 0.5 ng·kg–1·min–1). A hyperinsulinemic (0.6 mU·kg–1·min–1) euglycemic clamp was performed during the final 2 h of each infusion.
Results: Basal and insulin-stimulated glucose disposal decreased with ghrelin [basal: 1.9 ± 0.1 (ghrelin) vs. 2.3 ± 0.1 mg·kg–1·min–1, P = 0.03; clamp: 3.9 ± 0.6 (ghrelin) vs. 6.1 ± 0.5 mg·kg–1·min–1, P = 0.02], whereas endogenous glucose production was similar. Glucose infusion rate during the clamp was reduced by ghrelin [4.0 ± 0.7 (ghrelin) vs. 6.9 ± 0.9 mg·kg–1·min–1; P = 0.007], whereas nonesterified fatty acid flux increased [131 ± 26 (ghrelin) vs. 69 ± 5 µmol/min; P = 0.048] in the basal period. Regional lipolysis (skeletal muscle, sc fat) increased insignificantly with ghrelin infusion. Energy expenditure during the clamp decreased after ghrelin infusion [1539 ± 28 (ghrelin) vs. 1608 ± 32 kcal/24 h; P = 0.048], but the respiratory quotient did not differ. Minor but significant elevations in serum levels of GH and cortisol were observed after ghrelin infusion.
Conclusions: Administration of exogenous ghrelin causes insulin resistance in muscle and stimulates lipolysis; these effects are likely to be direct, although a small contribution of GH and cortisol cannot be excluded.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
More recently, the gastrointestinal peptide ghrelin was discovered as the endogenous ligand for the GH secretagogue receptor (2).
As expected, administration of ghrelin potently induces GH release (3), but the GH secretagogue receptor is present not only in the hypothalamus and the pituitary gland but also in many other organs and tissues (4, 5), suggesting that ghrelin may also possess peripheral, GH-independent effects.
Recognized effects of ghrelin include stimulation of ACTH secretion (3), and there is strong evidence to support that ghrelin also stimulates appetite, food intake (6, 7), and gastric motility (8, 9). In rodent models, ghrelin induces adiposity (10, 11, 12, 13), and ghrelin has been shown in vitro to stimulate the insulin-signaling cascade by phosphorylation of insulin receptor substrate 1 (14), phosphatidylinositol 3-kinase (14, 15), and protein kinase B/Akt (15, 16, 17).
By contrast, in clinical settings exogenous ghrelin increases plasma glucose (PG) and free fatty acids (18, 19, 20), and reduces glucose disposal rates (21) compatible with an impairment of insulin sensitivity. Thus, the impact of ghrelin on substrate metabolism is controversial, and, in particular, it has been difficult to dissect direct peripheral effects from indirect GH- and cortisol-mediated effects.
In the present study, we aimed to investigate the effects of ghrelin on glucose and lipid metabolism. Indeed, the GH release in response to a ghrelin bolus is blunted by concomitant somatostatin infusion (22), and the GH response to a continuous ghrelin infusion is most prominent during the first 60-min infusion, after which a decline toward baseline values is observed (18). In the present study, we doubled the somatostatin dose compared with the previous study by Di Vito et al. (22) to suppress GH secretion.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Eight healthy men aged 27.2 ± 0.9 yr with a body mass index of 23.4 ± 0.5 kg/m2 volunteered in this study. All had a normal physical examination.
The study was conducted in accordance with the Helsinki Declaration, and all subjects gave their oral and written informed consent to participate in the trial. The study protocol was approved by the local Ethics Committee of Aarhus County. According to the International Committee of Medical Journal Editors, the protocol was registered at Clinicaltrials.gov (ID NCT00512525).
Study protocol
The study protocol is illustrated in Fig. 1
. All participants were examined on two occasions separated by a minimum of 5 wk. The subjects were studied at 0700 h in a quiet, thermoneutral indoor environment after an overnight fast. The subjects fasted during the trial.
|
In a randomized, single-blind, placebo-controlled crossover design, we used a constant infusion of saline (as placebo) or human acylated ghrelin lasting 300 min at 5 pmol·kg body weight–1 · min–1. The period from 0–180 min is referred to as the basal period, and the period from 180–300 min as the clamp period.
At 0-min infusion of somatostatin (Ferring Pharmaceuticals, Copenhagen, Denmark) 330 µg/h, insulin (Actrapid; Novo Nordisk, Bagsværd, Denmark) 0.1 mU·kg–1·min–1, GH (Norditropin; Novo Nordisk) 2 ng·kg–1·min–1 was commenced. Infusion of glucagon (Glucagen; Novo Nordisk) 0.5 ng·kg–1·min–1 was added at 30 min. Euglycemia was intended, and if necessary, isotonic glucose was infused to maintain a PG level of approximately 5.0 mmol/liter to prevent somatostatin-induced hypoglycemia. At 180 min a hyperinsulinemic, euglycemic clamp (insulin 0.6 mU·kg–1·min–1) was performed. PG was clamped at 5.0 mmol/liter by adjusting the rate of infusion of 20% glucose according to PG measurements every 10 min. Insulin sensitivity was estimated by the level of glucose infusion rate (GIR) during the hyperinsulinemic, euglycemic clamp.
Tracers
A primed-continuous infusion of [3-3H]glucose (17 µCi bolus, followed by 0.17 µCi/min; New England Nuclear Life Science Products, Boston, MA) was initiated at 0 min and continued for 5 h. Glucose flux rates were calculated at 10-min intervals during the two steady-state periods (150–180 and 270–300 min, respectively) using Steele’s (23) nonsteady-state equations. During the clamp period, endogenous glucose production (EGP) was calculated by subtracting the rate of exogenous glucose infusion from the rate of appearance of [3-3H]glucose. [9,10-3H]palmitate (Danish Medicines Agency, Copenhagen, Denmark) was infused continuously at 0.3 µCi/min from 120–180 min, and blood samples were drawn for analysis of palmitate concentration and specific activity (SA). The steady state of SA was verified (150–180 min) for each individual. Plasma palmitate concentration and SA were determined by HPLC using [2H31]palmitate as an internal standard. Systemic palmitate flux (rate of appearance, µmol/min) was calculated using the [9,10-3H]palmitate infusion rate (dpm/min) divided by the steady-state palmitate SA (dpm/µmol).
Indirect calorimetry
Indirect calorimetry was performed at baseline before ghrelin and saline infusion, and for the last 30-min basal and clamp periods (Deltatrac; Datex Instruments, Helsinki, Finland), allowing measurements of energy expenditure (EE) and respiratory quotient (RQ) as described (24).
Microdialysis
After applying local analgesic of 1 ml lidocaine 1% superficial to the fascia of the gastrocnemius and the lateral vastus muscle, two microdialysis catheters (CMA-60; CMA, Stockholm, Sweden) were inserted into the muscles approximately 10 cm below and 14 cm above the patella at –60 min. Correct placement of the microdialysis catheters into the muscle was confirmed by the presence of muscle twitches during insertion. Two catheters were used to ensure viability of at least one because muscle twitches can disrupt the microdialysis membrane. Subsequently, two additional microdialysis catheters were positioned in the sc adipose tissue approximately 5 cm lateral of the umbilicus and in the femoral sc adipose tissue after 0.25 ml lidocaine sc before administration. The microdialysis catheters have a molecular cutoff of 20 kDa and a membrane length of 30 mm. Before insertion the catheters were manually flushed with perfusion fluid [Ringer Chloride, T1, CMA, Na+ 147 mmol/liter; K+ 1.4 mmol/liter; Ca2+ 2.3 mmol/liter; Cl– 156 mmol/liter (pH 6); osmolality 290 mosmol/kg] to allow for clearance of air bubbles from the microdialysis membranes. In addition, pre-wetting of the membranes in the perfusate media was performed as recommended by CMA. The microdialysis systems were perfused at a flow rate of 1 µl/min using CMA-107 perfusion pumps (CMA). The relative recovery of interstitial glycerol was assed by the internal reference method using 3H-glycerol (25). 3H-glycerol was added to the perfusate to obtain approximately 1000 cpm/µl. Perfusate and dialysate were counted using a Wallac 1450 liquid scintillation counter applying the Optiphase supermix scintillation fluid (Wallac, Turku, Finland) for muscle and Ultima Gold scintillation fluid (Packard Biosciences, Meriden, CT) for adipose tissue. Changes in interstitial glycerol concentration correspond to regional lipolysis (26).
Sampling of the interstitial fluid was commenced at –60 min, allowing for 60-min equilibration to minimize the influence of local edema and hemorrhage. The sampling was performed every 30 min and continued until 300 min.
Blood samples and measurements
PG was analyzed in duplicate using the glucose oxidase method (Beckman Instruments, Palo Alto, CA). Serum ghrelin (total levels) was measured in duplicate by an in-house assay as described previously (27). The assay measures immunoreactive levels of ghrelin using 125I-labeled bioactive ghrelin tracer and rabbit polyclonal antibodies raised against octanoylated human ghrelin. The assay recognizes the COOH terminal of ghrelin, and as such determines acylated as well as des-acylated ghrelin. Samples from each individual were analyzed in one assay. Serum GH was analyzed with a double-monoclonal immunofluorometric assay (Delfia, Wallac Oy, Turku, Finland). Serum cortisol was measured with a solid-phase, time-resolved fluoroimmunoassay (Delfia). Serum insulin and C peptide were measured with an immunoassay (Dako, Glostrup, Denmark). Plasma glucagon was measured by an in-house radioimmunoassay modified from Orskov et al. (28). Serum NEFA was determined by a colorimetric method using a commercial kit (Wako Chemicals, Neuss, Germany).
Glycerol in the microdialysis dialysate was measured in duplicate by an automated spectrophotometric kinetic enzymatic analyzer (CMA 600; CMA).
Statistical analysis
Results are expressed as mean ± SE. Systemic levels of ghrelin, GH, cortisol, glucagon, PG, insulin, C peptide, NEFA, and regional levels of glycerol were analyzed by two-way ANOVA. The interaction between time and treatment ("time x treatment") was considered the term of interest. The Bonferroni correction was used to account for multiple comparisons when appropriate. Glucose kinetics, palmitate fluxes, and indirect calorimetry were examined by Student’s two-tailed paired t test when appropriate. Effects of GH, cortisol, glucagon, insulin levels, and GIR (independent variables) on lipolysis (palmitate flux and NEFA increase) (dependent variables) were analyzed by comparing the changes in rates (GIR and palmitate flux) and areas under concentration-time curves (
AUC, GH, cortisol, glucagon, insulin, and NEFA) by backward regressions analyses. A P value less than 0.05 was considered significant. Statistical analysis was performed using SPSS version 14.0 for Windows (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Serum ghrelin was similar at baseline on both study days (P = 0.30). Infusion of ghrelin resulted in a 4.5-fold increase in circulating ghrelin levels during both the basal and clamp period. In the saline study, serum ghrelin levels decreased moderately and gradually with time (ANOVA P = 0.001; Fig. 2A).
|
). Serum cortisol increased approximately 1.9-fold after 90-min ghrelin exposure compared with placebo (Fig. 2C), followed by a decline toward placebo levels after 270 min. Overall, ANOVA revealed a significant difference with regards to glucagon levels between the 2 study days (P = 0.009), whereas post hoc analyses showed no differences between the individual time points for ghrelin and placebo treatment (all P > 0.05; Fig. 2D). Both serum C peptide (P = 0.93) and serum insulin levels remained similar during ghrelin infusion in the basal (P = 0.41) as well as the clamp period (P = 0.61; Fig. 2, E and F). Circulating metabolites
Euglycemia was intended, and glucose was infused from 0 min to ensure PG above 4.5 mmol/liter. Thus, PG was similar in the basal (P = 0.63) and at the end of the clamp period (P = 0.28; Fig. 2G). Overall, serum NEFA did not change, but post hoc pair-wise comparisons indicated a significant increase during ghrelin infusion at 180 and 210 min (P < 0.05; Fig. 2H).
Indirect calorimetry
The rates of EE and RQ were comparable in the basal period [EE: 1521 ± 43 (ghrelin) vs. 1552 ± 39 kcal/24 h, P = 0.30; and RQ: 0.89 ± 0.01 (ghrelin) vs. 0.88 ± 0.01, P = 0.53, respectively], whereas insulin-stimulated EE was significantly lower during ghrelin infusion as compared with placebo [EE: 1539 ± 28 (ghrelin) vs. 1608 ± 32 kcal/24 h; P = 0.048]. RQ was similar during the clamp period [0.92 ± 0.03 (ghrelin) vs. 0.93 ± 0.01; P = 0.65].
Glucose and palmitate metabolism
The rate of isotopically determined total glucose turnover during the basal [1.9 ± 0.1 (ghrelin) vs. 2.3 ± 0.1 mg·kg–1·min–1; P = 0.03] as well as the clamp period [3.9 ± 0.6 (ghrelin) vs. 6.1 ± 0.5 mg·kg–1·min–1; P = 0.02] was significantly lower during ghrelin administration as compared with placebo infusion (Fig. 3A). By contrast, EGP was similar during ghrelin and placebo administration in the basal period, and was suppressed to a similar degree during the hyperinsulinemic clamp (P nonsignificant; Fig. 3B). Significant lower GIR during the ghrelin study resulted in similar PG levels (Fig. 3C).
|
Insulin sensitivity, as assessed by GIR, was reduced during ghrelin infusion as compared with placebo infusion [4.0 ± 0.7 (ghrelin) vs. 6.9 ± 0.9 mg/kg·min; P = 0.007; Fig. 3C].
Correlations
Backward regressions analyses excluded AUC of glucagon, insulin, and GIR, and revealed that AUC of GH (P = 0.005) and cortisol (P = 0.002) were both significant predictors of AUC of NEFA. However, similar analyses, including palmitate flux, excluded all independent variables of the model as predictors of lipolysis.
Interstitial glycerol concentrations
Interstitial glycerol levels are provided in Table 1. In general, the hyperinsulinemic clamp suppressed interstitial glycerol levels in both muscle and adipose tissue during both ghrelin and placebo treatment. Statistical analysis revealed elevated glycerol levels in femoral sc fat during ghrelin and clamp conditions (P = 0.01), and overall, the regional glycerol release was elevated during ghrelin infusion, although it failed to reach statistical significance. Moreover, interstitial glycerol levels were less suppressed by hyperinsulinemia during ghrelin administration, which reached a significant difference in abdominal sc fat tissue (Table 1; P = 0.002).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
It is well recognized that ghrelin induces adiposity in rodents (10, 11, 12, 13) and stimulates the insulin-signaling cascade in various cell lines (14, 15, 16, 17). In humans, exogenous ghrelin increases PG and NEFA (18, 19, 20), and reduces glucose disposal rates (21) compatible with an impairment of insulin sensitivity. However, the direct metabolic effects of ghrelin in humans have been difficult to determine due to interference from GH and cortisol (18, 20, 21, 29, 30). In addition, ghrelin inhibits pancreatic β-cell function (29, 30), resulting in decreased glucose disposal rates per se.
Ghrelin enhances appetite (7), and circulating ghrelin increases in response to short as well as long-term nutritional shortage (6, 31). Increased ghrelin levels during fasting may have physiological implications via the reduction of peripheral insulin sensitivity and the concomitant increase in NEFA availability, both of which counteract peripheral glucose utilization. Theoretically, this may ensure glucose availability for oxidation in the central nervous system and other glucose-dependent tissues.
In our study the pancreatic clamp technique ensured constant and similar levels of insulin, C peptide, and glucose. Thus, the significant reductions of insulin sensitivity and glucose disposal observed are not attributable to decreased endogenous insulin secretion. Glucagon levels initially stabilized during the pancreatic clamp, but a small increase with time was recorded on both study days. A recent study documented that an insulin dose of 0.1 mU/kg·min has no effect on endogenous glucagon levels (32), and a previous study (33) showed that glucagon was not affected by exogenous ghrelin. The explanation for the observed increase in glucagon is not clear but could constitute a compensatory change in either clearance or secretion of glucagon. If anything, glucagon levels were lower on the ghrelin day, thus glucagon appears not to be involved in ghrelin-induced insulin resistance and reduced glucose disposal during ghrelin infusion. PG levels decreased during the basal period together with an initial increase of GIR, both of which could be attributed to insulin infusion during the pancreatic clamp. After this a decline in GIR in the basal period was observed, which probably was caused by GH.
One previous study designed to investigate the GH- and cortisol-independent effects of ghrelin reported intermediary metabolite levels only (19). In the present study, we provided more detailed information on turnover rates of glucose and NEFA, and the assessment of insulin sensitivity. Somatostatin strongly inhibits pituitary GH release, and in combination with a ghrelin bolus, somatostatin infusion blunts the GH increase (22). We have previously shown that ghrelin-induced GH release is most prominent during the first 60 min of a constant ghrelin infusion, followed by decline toward baseline values (18). Thus, to diminish the influence of GH in this study, we used a prolonged ghrelin infusion instead of a bolus administration in addition to a somatostatin infusion of approximately twice the dose reported by Di Vito et al. (22). Despite the concomitant somatostatin infusion, we did record a moderate but significant breakthrough of GH secretion in addition to cortisol release. However, this approach diminished the ghrelin-induced GH release by approximately 10-fold compared with earlier reports (18, 21) to levels similar to physiological GH levels recorded in the postabsorptive resting state (34). The low albeit significantly increased GH levels as compared with placebo may in part have contributed to the observed reduction in insulin sensitivity. The impact of physiological increments of GH on insulin sensitivity, glucose metabolism, and lipolysis has previously been investigated (35, 36, 37, 38). Initial insulin-like actions of GH increments (7.5-fold the levels of our present study) are followed by insulin-antagonistic effects (35). However, another study (2-h constant GH infusion; 5.9-fold GH levels as compared with our study) revealed no effects on glucose disposal or NEFA levels (36), and in a 1-h study, GH infusion (resulting in a GH elevation that was 
3.5-fold higher and more prolonged compared with the present reported) resulted in a 1-h reduction of GIR, an increase in NEFA levels, and a 12% reduction of glucose disposal (37). Hansen et al. (38) administered a low dose GH bolus (1 µg/kg) resulting in GH levels twice the levels of our study, and a slight increase of NEFA and peripheral lipolysis. In overall terms, the GH peak of our study is modest, transitory, and diminutive compared with integrated GH levels of the GH administration studies to our knowledge.
In the present study, cortisol levels during ghrelin infusion surpassed placebo levels from 60–240 min and reached approximately 2-fold the corresponding placebo after 90 min. The physiological significance of such a small and short-lived elevation in cortisol is uncertain, but probably minor. In two 6-h cortisol infusion studies combined with a pancreatic clamp, there was no effect of cortisol (increased to twice the levels of our study) on PG and GIR, but regional and whole body lipolysis increased (39, 40). In contrast, a prolonged infusion of cortisol (28 h; 2.4-fold cortisol levels compared with our study) decreased insulin sensitivity (41), whereas another prolonged infusion of cortisol (three times as high cortisol levels as in the present study) did not affect endogenous glucose release and glucose uptake compared with placebo levels (42).
Therefore, the observed reduction in insulin sensitivity in terms of a decrease of GIR, glucose disposal rate, and increased lipolysis appears more pronounced than would be expected by the moderate increments of GH and cortisol per se (35, 36, 37, 38, 39, 40, 41, 42). This notion is supported by a 17% reduction of GIR in gastrectomized patients receiving an equal ghrelin dose (but no somatostatin clamp) (21) compared with a 40% reduction of GIR in our study. GH-independent effects of ghrelin have been demonstrated in GH-deficient dwarf rats, in which ghrelin administration caused an increase in body weight and fat mass (10). In addition, direct effects of ghrelin have been investigated in vitro in adipocytes, but the results are ambiguous (15, 43). A potential method to obviate more completely the impact of GH and cortisol in human models could be to study hypopituitary patients with GH and ACTH deficiency.
Another limitation is the pharmacological nature of our study, which implies that the observed effects do not necessarily reflect physiological effects of ghrelin. Still, our data may be of clinical relevance because agonists as well as antagonists of ghrelin are currently being tested in animal and human models.
In conclusion, exogenous ghrelin overrides the inhibitory effect of concomitant somatostatin infusion with regards to GH and cortisol release. Nevertheless, our results indicate that ghrelin infusion acutely induces insulin resistance and lipolysis. The physiological relevance of ghrelin still remains to be determined, but we propose that ghrelin works in concert with GH to partition substrate metabolism during starvation toward NEFA utilization in peripheral tissues and thereby provision of glucose in tissues such as the central nervous system.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: E.T.V., C.B.D., J.G., S.N., N.M., J.J.H., and O.S. have nothing to declare. J.O.L.J. consults for and received lecture fees from Novo Nordisk and Pfizer.
First Published Online November 27, 2007
Abbreviations: AUC, Area under concentration-time curve; EE, energy expenditure; EGP, endogenous glucose production; GIR, glucose infusion rate; NEFA, nonesterified fatty acid; PG, plasma glucose; RQ, respiratory quotient; SA, specific activity.
Received September 7, 2007.
Accepted November 16, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. M Kiewiet, M. O van Aken, K. van der Weerd, P. Uitterlinden, A. P N Themmen, L. J Hofland, Y. B de Rijke, P. J D Delhanty, E. Ghigo, T. Abribat, et al. Effects of acute administration of acylated and unacylated ghrelin on glucose and insulin concentrations in morbidly obese subjects without overt diabetes Eur. J. Endocrinol., October 1, 2009; 161(4): 567 - 573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. T. Vestergaard, L. C. Gormsen, N. Jessen, S. Lund, T. K. Hansen, N. Moller, and J. O. L. Jorgensen Ghrelin Infusion in Humans Induces Acute Insulin Resistance and Lipolysis Independent of Growth Hormone Signaling Diabetes, December 1, 2008; 57(12): 3205 - 3210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gjedde, E. T. Vestergaard, L. C. Gormsen, A. L. D. Riis, J. Rungby, N. Moller, J. Weeke, and J. O. L. Jorgensen Serum Ghrelin Levels Are Increased in Hypothyroid Patients and Become Normalized by L-Thyroxine Treatment J. Clin. Endocrinol. Metab., June 1, 2008; 93(6): 2277 - 2280. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
