Impaired growth hormone response to insulin-induced hypoglycaemia in obese patients: restoration blocked by ritanserin after fenfluramine administration

The role of leptin in regulating bone metabolism

Leptin was initially best known for its role in energy homeostasis and regulation of energy expenditure. In the past few years we have realized that leptin also plays a major role in neuroendocrine regulation and bone metabolism. Here, we review the literature the indirect and direct pathways through which leptin acts to influence bone metabolism and discuss bone abnormalities related to leptin deficiency in both animal and human studies. The clinical utility of leptin in leptin deficient individuals and its potential to improve metabolic bone disease are also discussed. We are beginning to understand the critical role leptin plays in bone metabolism; future randomized studies are needed to fully assess the potential and risk-benefit of leptin's use in metabolic bone disease particularly in leptin deficient individuals.

Ontogeny of diet-induced obesity in selectively bred Sprague-Dawley rats

Outbred Sprague-Dawley rats selectively bred for their propensity to develop diet-induced obesity (DIO) become heavier on low-fat diet than those bred to be diet resistant (DR) beginning at approximately 5 wk of age. Here we assessed the development of metabolic and neural functions for insights into the origins of their greater weight gain. From week 5 to week 10, chow-fed DIO rats gained 15% more body weight and ate approximately 14% more calories but had only slightly greater adiposity and plasma leptin than DR rats. From day 3 through week 10, DIO and DR rats had similar mRNA expression of arcuate nucleus neuropeptide Y, proopiomelanocortin, agouti-related peptide, and all splice variants of the leptin receptor (OB-R). When fed a high-energy (HE; 31% fat) diet, 7-wk-old DIO rats had a 240% increase in plasma leptin levels after only 3 days. Despite this early leptin rise, they maintained a persistent hyperphagia and became more obese than chow-fed DIO rats and DR rats fed chow or HE diet. Their failure to reduce caloric intake, despite high levels of leptin, suggests that selectively bred DIO rats might have reduced leptin sensitivity similar to that seen in the outbred DIO parent strain.

CNS sensing and regulation of peripheral glucose levels

It is clear that the brain has evolved a mechanism for sensing levels of ambient glucose. Teleologically, this is likely to be a function of its requirement for glucose as a primary metabolic substrate. There is no question that the brain can sense and mount a counterregulatory response to restore very low levels of plasma and brain glucose. But it is less clear that the changes in glucose associated with normal diurnal rhythms and feeding cycles are sufficient to influence either ingestive behavior or the physiologic responses involved in regulating plasma glucose levels. Glucosensing neurons are clearly a distinct class of metabolic sensors with the capacity to respond to a variety of intero- and exteroceptive stimuli. This makes it likely that these glucosensing neurons do participate in physiologically relevant homeostatic mechanisms involving energy balance and the regulation of peripheral glucose levels. It is our challenge to identify the mechanisms by which these neurons sense and respond to these metabolic cues.

Direct modification of somatotrope function by long-term leptin treatment of primary cultured ovine pituitary cells

Leptin is produced primarily in adipocytes and regulates body energy balance. A close link between leptin and pituitary hormones, including GH, has been reported. The mechanisms employed by leptin to influence somatotropes are not clear, however. Here we report a direct action of recombinant ovine leptin on primary cultured ovine somatotropes by analyzing the levels of mRNA encoding for GH or the receptors for GHRH (GHRH-R) and GH-releasing peptides (GHRP). Treatment of ovine somatotropes with leptin (10(-7)-10(-9) M) for 1-3 d reduced the mRNA levels encoding GH and GHRH-R, but increased GHRP receptor mRNA levels in a time- and dose-dependent manner. Three-day treatment of cells with leptin decreased the GH response to GHRH stimulation, but the GH response to GHRP-2 stimulation was increased. The combined effect of GHRH and GHRP-2 on GH secretion was not altered by treatment of the cells with leptin. These results demonstrated a direct action of leptin on ovine pituitary cells, leading to a reduced sensitivity of somatotropes to GHRH. It is also suggested that GHRP may be useful to correct the decrease in GHRH-induced GH secretion by leptin.

Effect of the serotonin antagonists ritanserin and ketanserin in Cushing's disease

Central serotonergic regulation could have a role in the course of pituitary-dependent Cushing's disease. We studied the effects of ritanserin and ketanserin, two related selective 5HT2 receptor antagonists, in 11 patients with Cushing's disease. Treatment lasted from 1 month to 1 year (up to 4 years in one patient). Daily doses were 10-15 mg for ritanserin, and 40-80 mg for ketanserin. Since the two drugs share the same mechanism of action and no qualitative or quantitative differences in response to their administration were observed, the results were pooled together. Patients were assessed by clinical and hormonal evaluation. Urinary cortisol and ACTH were considered the parameters of interest. Short-term response: after 1 month, there was a significant decrease of urinary cortisol from 781 (160) to 331 (215) nmol/d (P < 0.02) while ACTH was 9.8 (1.5) pmol/L baseline and again 8.8 (2.2) pmol/L at 1 month (P = NS). For 9 patients, hormonal parameters were available after 1 week of treatment. In this case, also ACTH levels were significantly decreased (from 9.6 (1.7) to 5.2 (1.3) pmol/L; P < 0.01) together with urinary cortisol (from 781 (194) to 372 (165) nmol/d; P < 0.01). Long-term response: in 3 patients, hormonal parameters failed to respond to serotonin receptor antagonists, which were thus discontinued. An improvement was recorded in the remaining 8 patients, that was prolonged in 3, and transient in 5. In 3 of these latter patients, a marked increase of ACTH was observed before treatment discontinuation. Ketanserin was given to 2 patients with Nelson's syndrome, with only transient ACTH decrease in one, and no changes in ACTH response to CRH after 1 month treatment in both cases. An inhibitory effect of ritanserin and ketanserin on ACTH and cortisol production in Cushing's disease appeared to be limited both in terms of duration of response and number of patients with a satisfactory outcome. However, the results may provide a better understanding of serotonergic modulation in Cushing's disease and lead to therapeutic developments.

Metabolic imprinting on genetically predisposed neural circuits perpetuates obesity

There is an obesity epidemic in the industrialized world that is not simply explained by excess energy intake and decreased energy expenditure. Persistent obesity develops when genetically predisposed individuals are in a chronic state of positive energy balance. Once established, the obese body weight is avidly defended against both over- and underfeeding. Animal studies have shown that lean individuals who are genetically predisposed toward obesity have abnormalities of neural function that prime them to become obese when caloric density of the diet is raised. These neural abnormalities are gradually "corrected" as obesity becomes fully developed, suggesting that obesity is the normal state for such individuals. Thus, defense of the obese body weight may be perpetuated by the formation of new neural circuits involved in energy-homeostasis pathways that are not then easily abolished. Such neural plasticity can occur in both adult life and during nervous-system development. Early pre- and postnatal metabolic conditions (maternal diabetes, obesity, undernutrition) can lead genetically predisposed offspring to become even more obese as adults. This enhanced obesity is associated with altered brain neural circuitry, and these changes can then be passed on to subsequent generations in a feed-forward cycle of ever-increasing body weight. Thus, the metabolic perturbations associated with obesity during both brain development and adult life can produce "metabolic imprinting" on genetically predisposed neural circuits involved in energy homeostasis. Drugs that reduce body weight decrease the defended body weight and alter neural pathways involved in energy homeostasis but have no permanent effect on body weight or neural function in most individuals. Thus, early intervention in mothers, infants, children, and adults may be the only way to prevent the formation of permanent neural connections that promote and perpetuate obesity in genetically predisposed individuals.

Defense of body weight against chronic caloric restriction in obesity-prone and -resistant rats

Half of Sprague-Dawley rats develop and defend diet-induced obesity (DIO) or diet resistance (DR) when fed a high-energy (HE) diet. Here, adult male rats were made DIO or DR after 10 wk on HE diet. Then half of each group was food restricted for 8 wk on chow to maintain their body weights at 90% of their respective baselines. Rate and magnitude of weight loss were comparable, but maintenance energy intake and the degree of sympathetic activity (24-h urine norepinephrine) inhibition were 17 and 29% lower, respectively, in restricted DR than DIO rats. Restricted DIO rats reduced adipose depot weights, plasma leptin, and insulin levels by 35%. Restricted DR rats reduced none of these. When fed ad libitum, both DR and DIO rats returned to the body weights of their respective chow-fed phenotype controls within 2 wk. This was associated with increased adipose mass and leptin and insulin levels only in DIO rats. Thus DR rats appear to alter primarily their lean body mass, whereas DIO rats primarily alter their adipose mass during chronic caloric restriction and refeeding.

Three-month treatment with metformin or dexfenfluramine does not modify the effects of diet on anthropometric and endocrine-metabolic parameters in abdominal obesity

Abdominal obesity is connoted by hyperinsulinism and insulin insensitivity, a trend toward glucose intolerance, hypoactivity of GH/IGF-I axis and alterations of hypothalamo-pituitary-adrenal (HPA) axis. It has been hypothesized that treatment with metformin (MET) and dexfenfluramine (DEX) could counteract those endocrine-metabolic alterations. Thus, we studied the effects of 3-month treatment with MET or DEX on anthropometric (BMI, WHR, FM and FFM), metabolic (basal and OGTT-induced glucose) and hormonal variables (IGF-I, DHEA-S, androstendione, testosterone, fT3, fT4, TSH, basal and OGTT-induced insulin) as well as on blood pressure in 28 normotensive patients with abdominal obesity (OB, 3 M, 25 F; 47.5+/-1.5 yr [mean+/-SE], BMI 35.4+/-1.1 kg/m2, WHR 0.98+/-0.04 and 0.86+/-0.07, in M and F, respectively). All patients were on balanced hypocaloric diet (1400 Kcal/day). Patients were randomly assigned to treatment with MET (no.=10, 500 mg twice daily po) or DEX (no.=10, 15 mg thrice daily po) or placebo (no.=8). Before treatment all groups had similar anthropometric, metabolic and hormonal values. After 3-month treatment with MET, DEX or placebo, weight, BMI and WHR reductions were similar in all groups (p<0.05 vs baseline in either group). In each group FFM/FM ratio showed non significant trend toward increase. No significant variations in metabolic and endocrine variables were recorded in each group after 1 and 3-month treatment. However, glucose tolerance, OGTT-induced insulin response, glucose/insulin ratio showed a similar trend toward improvement in all groups, while IGF-I, 24 h urinary cortisol, DHEA-S, androstendione, testosterone, thyroid hormone and TSH levels did not show any variation. Significant (p<0.02) and similar reductions of DBP, but not of SBP, levels were found in all groups. In conclusion, our findings demonstrate that, at least after 3-month treatment, metformin and dexfenfluramine do not modify the effects of diet on anthropometric, metabolic and hormonal parameters as well as on blood pressure in patients with abdominal obesity.

A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction

The adipocyte-specific hormone leptin, the product of the obese (ob) gene, regulates adipose-tissue mass through hypothalamic effects on satiety and energy expenditure. Leptin acts through the leptin receptor, a single-transmembrane-domain receptor of the cytokine-receptor family. In rodents, homozygous mutations in genes encoding leptin or the leptin receptor cause early-onset morbid obesity, hyperphagia and reduced energy expenditure. These rodents also show hypercortisolaemia, alterations in glucose homeostasis, dyslipidaemia, and infertility due to hypogonadotropic hypogonadisms. In humans, leptin deficiency due to a mutation in the leptin gene is associated with early-onset obesity. Here we describe a homozygous mutation in the human leptin receptor gene that results in a truncated leptin receptor lacking both the transmembrane and the intracellular domains. In addition to their early-onset morbid obesity, patients homozygous for this mutation have no pubertal development and their secretion of growth hormone and thyrotropin is reduced. These results indicate that leptin is an important physiological regulator of several endocrine functions in humans.

The influence of chronic administration of the serotonin agonist dexfenfluramine on responsiveness to corticotropin releasing hormone and growth hormone-releasing hormone in moderately obese people

Evidence exists that the serotoninergic system influences the hypothalamo-pituitary-adrenal axis and pituitary GH secretion. The effect was investigated of chronic (9 weeks) administration of the serotonin receptor agonist dexfenfluramine (DF) (15 mg twice daily) versus placebo on both systems in 24 moderately obese patients with an overconsumption of snacks. Before as well as after treatment, 100 micrograms GHRH and 100 micrograms CRH was administered iv. Weight loss after 9 weeks was 3.1 +/- 2.3 kg in the DF group and 0.1 +/- 1.2 in the placebo group (p < 0.001). No significant difference between the changes in the apparently normal ACTH and cortisol response in both groups was found. The apparently low GH response did not increase or normalize, but on the contrary decreased further after DF. No changes were shown in urinary free cortisol excretion and serum IGF-I levels. It can be concluded that in moderately obese patients the use of DF during a 9-week period induced a significant weight loss, did not influence the responsiveness of the pituitary gland to CRH nor the overall activity of the pituitary adrenal axis and finally, did not normalize the blunted GH response after GHRH.

The effect of hypocaloric diet with and without D-fenfluramine treatment on growth hormone release after growth hormone-releasing factor stimulation in patients with android obesity

Basal and stimulated growth hormone (GH) secretions are impaired in obesity, and partial restoration of the GH response to various stimuli is observed after weight loss. The aim of the present study was to investigate whether D-fenfluramine, a serotoninergic agent, would increase the GH response to growth hormone-releasing factor (GRF) as compared with placebo in obese android patients. The subjects were 17 patients with android obesity (four men and 13 women) aged 21 to 58 years with a body mass index (BMI) ranging from 32.0 to 52.2 kg/m2 and an abdominal-gluteal ratio greater than 1.0. The following four GRF (1-44) tests were performed: T-30 (control), T0 (after 30 days of a hypocaloric diet), T1 (after 30 days of either placebo or D-fenfluramine 15 mg twice daily), and T2 (after 30 additional days of placebo or D-fenfluramine). The hypocaloric diet was maintained during the T1 and T2 periods. At each test, the serum GH response to GRF was measured at frequent intervals, and the peak GH response and the GH area under the curve were calculated. Serum insulin concentrations were also assayed before GRF stimulation, and the insulin to GH ratio was obtained. The D-fenfluramine-treated group had a mean +/- SEM GH peak level after GRF significantly higher at T1 (43.3 +/- 8.2 micrograms/L) and T2 (50.9 +/- 9.2 micrograms/L) compared with the placebo group. Likewise, the mean integrated areas of GH response were significantly higher for the D-fenfluramine-treated group as compared with the placebo group at both T1 and T2 of therapy.(ABSTRACT TRUNCATED AT 250 WORDS)

Reproducibility of glycaemic thresholds for activation of counterregulatory hormones and hypoglycaemic symptoms in healthy subjects

Nine healthy subjects were studied on two separate occasions, at least two weeks apart, using the glucose clamp technique to produce a gradual hypoglycaemia. Glucose thresholds for neuroendocrine and symptom responses varied up to 1.5 mmol/l between subjects. There was a significant correlation between individual glucose thresholds on day 1 and 2 for adrenaline (p = 0.0008), growth hormone (p = 0.007) and pancreatic polypeptide (p = 0.02), and for autonomic (p = 0.018) and neuroglycopoenic (p = 0.023) symptoms, whereas no significant correlations were found for glucagon and cortisol. The mean intra-individual differences in glucose thresholds between day 1 and 2 were 0.22 mmol/l for the hormones and 0.25 mmol/l for the symptoms. We conclude that healthy subjects differ in hypoglycaemic thresholds, and that the difference reflects individual variation.

Serotoninergic receptor activation by dextrofenfluramine enhances the blunted pituitary-adrenal responsiveness to corticotropin-releasing hormone in obese subjects

To explore the interrelationships between the serotoninergic system and the hypothalamic-pituitary-adrenal (HPA) axis in human obesity, we evaluated cortisol and adrenocorticotropic hormone (ACTH) response to synthetic human corticotropin-releasing hormone (hCRH, 1 microgram/kg intravenously [IV]) before and after stimulation of the serotoninergic system by dextrofenfluramine (d-FF, 30 mg/d for 3 months) in nine obese women. These responses were compared with a CRH test (1 microgram/kg) carried out in nine age-matched normal-weight women. Plasma cortisol of obese subjects did not significantly increase after CRH (peak value 127.1 +/- 11.2 ng/mL v 104.1 +/- 9.5 ng/mL). This response was lower (P less than .005) than in the controls, in whom the basal cortisol value of 120.6 +/- 11.8 ng/mL reached a peak value of 221.2 +/- 13.4 ng/mL. However, after administration of d-FF, CRH significantly increased (P less than .0001) plasma cortisol (peak value 170.6 +/- 18.0 ng/mL v 111.5 +/- 10.8 ng/mL) and the response was enhanced (P less than .05) as compared with that obtained before d-FF. The ACTH levels of our patients showed a small increment after CRH injection (peak value 13.5 +/- 1.7 pg/mL v 9.6 +/- 1.1 pg/mL), but the hormonal response was lower (P less than .005) than in controls (peak value 38.1 +/- 5.5 pg/mL v 13.8 +/- 0.8 pg/mL). However, after d-FF, CRH induced a significant increment (P less than .05) in plasma ACTH at 30 minutes (20.4 +/- 3.7 pg/mL v 10.9 +/- 0.9 pg/mL) and 45 minutes (18.0 +/- 2.6 pg/mL), even though this response was not significantly different from that observed before d-FF administration.(ABSTRACT TRUNCATED AT 250 WORDS)

Blunted growth hormone (GH) responsiveness to GH-releasing hormone in obese patients: influence of prolonged administration of the serotoninergic drug fenfluramine

The aim of the present study was to ascertain if reduced central serotoninergic activity might contribute to the well-known blunted growth hormone (GH) response to GH-releasing hormone (GHRH) in obese patients. Thus, we studied the effect of prolonged stimulation of the serotoninergic system by fenfluramine (FF; 60 mg twice daily for 7 days) on GHRH-induced GH release in nine obese and seven normal subjects. In controls, GHRH (100 micrograms intravenously [IV]) injection increased GH levels from 2.3 +/- 1.8 (+/- SE) to 18.5 +/- 2.8 mU/L, P less than .002. FF administration enhanced both basal and GHRH-stimulated GH levels (peak, 38.4 +/- 8.3 v 6.9 +/- 2.6 mU/L, P less than .002). This response was significantly higher (P less than .02) than in pretreatment. In obese patients, GH responsiveness to GHRH was slight (peak, 7.1 +/- 2.0 v 0.6 +/- 0.18 mU/L, P less than .01) and lower (P less than .01) than in controls. FF administration did not affect this response. In controls, the enhanced FF-induced GH release after a maximal dose of GHRH indicates that serotoninergic activation influences GH secretion and that the mechanism involved is independent of endogenous GHRH. In obese patients, we found a blunted GH responsiveness to GHRH that was not affected by FF, thus supporting the hypothesis that the serotoninergic control on GH release is impaired.