Glucose tolerance and serum insulin levels in an animal model of obesity induced by the antipsychotic drug, sulpiride

Antipsychotics and glucose metabolism: how brain and body collide

Since the serendipitous discovery of the first antipsychotic (AP) drug in the 1950s, APs remain the cornerstone of treatment for schizophrenia. A shift over the past two decades away from first-generation, conventional APs to so-called "atypical" (or 2nd/3rd generation) APs parallels acknowledgment of serious metabolic side-effects associated in particular with these newer agents. As will be reviewed, AP drugs and type 2 diabetes are now inextricably linked, contributing to the three- to fivefold increased risk of type 2 diabetes observed in schizophrenia. However, this association is not straightforward. Biological and lifestyle-related illness factors contribute to the association between type 2 diabetes and metabolic disease independently of AP treatment. In addition, APs have a well-established weight gain propensity which could also account for elevated risk of insulin resistance and type 2 diabetes. However, compelling preclinical and clinical evidence now suggests that these drugs can rapidly and directly influence pathways of glucose metabolism independently of weight gain and even in absence of psychiatric illness. Mechanisms of these direct effects remain poorly elucidated but may involve central and peripheral antagonism of neurotransmitters implicated not only in the therapeutic effects of APs but also in glucose homeostasis, possibly via effects on the autonomic nervous system. The clinical relevance of studying "direct" effects of these drugs on glucose metabolism is underscored by the widespread use of these medications, both on and off label, for a growing number of mental illnesses, extending safety concerns well beyond schizophrenia.

Atypical antipsychotics and effects on feeding: from mice to men

RATIONALE

So-called atypical antipsychotics (AAPs) are associated with varying levels of weight gain and associated metabolic disturbances, which in patients with serious mental illness (SMI) have been linked to non-compliance and poor functional outcomes. Mechanisms underlying AAP-induced metabolic abnormalities are only partially understood. Antipsychotic-induced weight gain may occur as a result of increases in food intake and/or changes in feeding.

OBJECTIVE

In this review, we examine the available human and preclinical literature addressing AAP-related changes in feeding behavior, to determine whether changes in appetite and perturbations in regulation of food intake could be contributing factors to antipsychotic-induced weight gain.

RESULTS

In general, human studies point to disruption by AAPs of feeding behaviors and food consumption. In rodents, increases in cumulative food intake are mainly observed in females; however, changes in feeding microstructure or motivational aspects of food intake appear to occur independent of sex.

CONCLUSIONS

The findings from this review indicate that the varying levels of AAP-related weight gain reflect changes in both appetite and feeding behaviors, which differ by type of AAP. However, inconsistencies exist among the studies (both human and rodent) that may reflect considerable differences in study design and methodology. Future studies examining underlying mechanisms of antipsychotic-induced weight gain are recommended in order to develop strategies addressing the serious metabolic side effect of AAPs.

Characterisation of olanzapine-induced weight gain and effect of aripiprazole vs olanzapine on body weight and prolactin secretion in female rats

RATIONALE

Atypical antipsychotic drug (APD)-induced weight gain causes non-compliance, increasing the risk of relapse and medical complications.

OBJECTIVES

In an animal model, we assessed body weights, food intake, body fat/lean body mass contents and blood serum levels of glucose and lipids in female rats treated with olanzapine (Experiment 1). Also, we investigated the effect of aripiprazole vs olanzapine treatment on weight gain (WG) and plasma prolactin secretion in two strains (Wistar and Sprague-Dawley) and in two housing conditions (singly and group housed; Experiment 2).

METHODS

In Experiment 1, Wistar females received either vehicle or olanzapine (5.0 mg kg(-1), p.o.) twice daily for 14 days. In Experiment 2, female rats (Wistar or Sprague-Dawley), housed singly or in groups, received either vehicle, aripiprazole (2.0-8.0 mg kg(-1), p.o.), or olanzapine (1.0-10 mg kg(-1), p.o.) twice daily for 7 days. Body weights and food intake were assessed daily. Body composition and blood assays were analyzed at the end of the treatment.

RESULTS

WG induced by chronic olanzapine treatment was characterised by hyperphagia, increased body fat, and serum free fatty acid content and reduced lean tissue and serum glucose content. Subchronic aripiprazole treatment resulted in rapid and robust WG similar to those observed with olanzapine. In spite of similar effects on body weight, aripiprazole and olanzapine stimulated markedly different patterns of prolactin secretion. Body weight changes and prolactin secretion induced by these APDs were significantly modulated by housing and by strain.

CONCLUSION

Assessment of body weight in the present model may not have predictive validity, and other measures may be needed to differentiate between WG-inducing and weight-neutral drugs.

Comparative effects of the antipsychotics sulpiride and risperidone in female rats on energy balance, body composition, fat morphology and macronutrient selection

Previous studies showed that the antipsychotic drugs (APDs) sulpiride (SUL) and risperidone (RIS) induced body weight gain (BWG), hyperphagia, and increased serum levels of leptin, prolactin and corticosterone in female rats. Neither SUL nor RIS increased BWG or food intake (FI) in male rats. To further develop the animal model of APD-induced obesity, SUL (20 mg/kg/sc), RIS (0.5 mg/kg/sc) or vehicle (1 cm(3)/kg/sc) were administered to female Wistar rats for 10 or 12 days. Body composition, fat tissue morphology, energy expenditure and food efficiency were assessed in animals fed a high-fat diet. In another experiment, macronutrient selection was evaluated in animals fed with pure diets. SUL and RIS significantly increased BWG and FI, with a stronger effect of SUL. Both drugs increased fat gain and food efficiency, and did not modify energy expenditure. Obesity was due to adipocyte hyperplasia. SUL-treated rats significantly decreased fat intake (p = 0.039), showed a tendency to increase protein intake and did not modify carbohydrate consumption. No differences were observed between the RIS and the vehicle group. The macronutrient selection pattern differs from that observed in obese people undergoing APD treatment and in most animal models of obesity. Those findings suggest that SUL administration does not properly model APD treatment in humans. Results on macronutient selection in RIS-treated rats must be considered as preliminary, since in this experiment the animals did not gain weight significantly. Other diet protocols and lower APD doses must be tested to further characterize the RIS model.

Valproate-associated reproductive and metabolic abnormalities: are epileptic women at greater risk than bipolar women?

OBJECTIVE

Evidence indicates that valproate (VPA) may have an adverse impact on reproductive endocrine and metabolic functions in women with epilepsy. This study explores whether the association of VPA with reproductive endocrine abnormalities is applicable to women with bipolar disorder (BD) or is unique to women with epilepsy.

METHODS

Thirty female patients aged 18-40 years with a DSM-IV diagnosis of BD (15 on lithium monotherapy and 15 on VPA monotherapy or VPA in combination with lithium therapy) and 15 with idiopathic generalized epilepsy (IGE) on VPA monotherapy were evaluated for reproductive endocrine functioning and metabolic parameters.

RESULTS

The menarche age, mean length of menstrual cycle and mean length of menses were not significantly different between groups. None of the bipolar patients on lithium, three (20%) of the bipolar patients on VPA and seven (47%) of the epileptic patients on VPA reported menstrual disturbances. Hirsutism scores of the epilepsy group were significantly higher than those bipolar women, regardless of treatment. Serum total testosterone levels were significantly higher in patients (both with BD and with IGE) treated with VPA than in those treated with lithium. Serum FSH levels were significantly lower and LH-to-FSH ratio was significantly higher in patients with epilepsy than in patients with BD, regardless of treatment. The weight parameters and lipid values investigated did not differ significantly between the groups.

CONCLUSION

The study supports the conclusion that VPA may be associated with menstrual abnormalities and increased total testosterone levels in both bipolar and epileptic patients although women with BD did not show clinical features of hyperandrogenism (menstrual abnormalities, hirsutism and truncal obesity) as did frequently as women with epilepsy.

The antipsychotic drug sulpiride does not affect bodyweight in male rats. Is insulin resistance involved?

Previous studies have shown that prolonged administration of antipsychotic drugs induces obesity in female but not in male rats. To explore the mechanisms involved in this sex-dependent effect, we administered the dopamine antagonist sulpiride (20 mg/kg i.p.) or vehicle (0.1 N HCl) to adult male rats during 21 days and daily assessed bodyweight and food intake. Then, we evaluated the glucose tolerance and the serum levels of insulin, leptin, total testosterone, dehydroepiandrosterone-sulfate (DHEA-S), thyroid hormones and blood lipids. In another experiment, food intake and water intake were assessed after acute injections of sulpiride or vehicle into the perifornical lateral hypothalamus. Lastly, the dopamine metabolites dihydroxyphenylacetic acid (DOPAC) and homovanilic acid (HVA) in the lateral hypothalamus were assessed by in vivo microdialysis after acute systemic injections of sulpiride and vehicle. Chronic sulpiride administration did not affect bodyweight gain and food intake. However, prolactin levels and the area under the glucose and insulin curves were significantly elevated. Acute sulpiride significantly increased food intake, water intake, DOPAC and HVA levels. The acute effects of sulpiride show that this drug is active at the perifornical lateral hypothalamus, which is a brain area where blockade of dopamine receptors stimulates feeding. However, after prolonged administration, sulpiride did not affect body weight. This lack of effect may be related to the impairment of insulin sensitivity, which may prevent body weight gain, and counteract other effects of sulpiride that promote adiposity such as hyperprolactinemia. These findings noticeably contrast with those observed in sulpiride-treated female rats that appear to display enhanced insulin sensitivity. The changes in insulin sensitivity do not appear related to a decrease in androgenic activity, because testosterone and DHEA-S levels were not affected by sulpiride. However, these results should be considered as preliminary because other relevant endocrine variables such as free testosterone, steroid binding globulin and pituitary gonadotrophin levels were not evaluated. Since the same sex-dependent effect on body weight and food intake in rats has been observed during administration of risperidone, which has a different pharmacological profile than sulpiride, future studies must evaluate other neurotransmitters involved in food intake regulation such as serotonin, noradrenaline and histamine.

Incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat is suggestive of a role in endothelium-derived hyperpolarizing factor-mediated responses

Although the chemical nature of endothelium-derived hyperpolarizing factor (EDHF) remains elusive, electrophysiological evidence exists for electrical communication between smooth muscle cells and endothelial cells suggesting that electrotonic propagation of hyperpolarization may explain the failure to identify a single chemical factor as EDHF. Anatomical evidence for myoendothelial gap junctions, or the sites of electrical coupling, is, however, rare. In the present study, serial-section electron microscopy and reconstruction techniques have been used to examine the incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat where EDHF responses have been reported to vary. Myoendothelial gap junctions were found to be very small in the mesenteric arteries, the majority being <100 nm in diameter. In addition, they were significantly more common in the distal compared with the proximal regions of this arterial bed. Pentalaminar gap junctions between adjacent endothelial cells were much larger and were common in both proximal and distal mesenteric arteries. These latter junctions were frequently found near the myoendothelial gap junctions. These results provide the first evidence for the presence of sites for electrical communication between endothelial cells and smooth muscle cells in the mesenteric vascular bed. Furthermore, the relative incidence of these sites suggests that there may be a relationship between the activity of EDHF and the presence of myoendothelial gap junctions.

Antipsychotic drug-induced obesity in rats: correlation between leptin, insulin and body weight during sulpiride treatment

Sulpiride (SUL, 20 mg kg-1 day-1) induces weight gain, hyperphagia, hyperprolactinemia, hypogonadism, and perhaps increased insulin sensitivity in rats. Leptin seems to signal the brain about the size of body fat stores and nutrient metabolism. We evaluated the basal serum leptin levels in rats after acute (1 h) or prolonged SUL or vehicle administration (10, 20 and 30 days). At days 10 and 30 leptin was also assessed during a glucose overload test. As the maximal weight gain during SUL administration is observed at days 10-15 of treatment, leptin was measured in a comparison group of insulin-treated rats (5 IU day-1 for 10 days). SUL-treated rats significantly gained weight. However, leptin levels were not significantly increased at any time-point of treatment. SUL did not affect insulin levels either. By contrast, leptin levels were significantly elevated after insulin administration, along with weight gain and hyperinsulinemia. An opposite correlation was also observed at day 10: leptin and insulin correlated negatively in the SUL group and positively in the insulin group. In addition, leptin and the magnitude of weight gain tended to correlate positively after SUL treatment, but negatively after insulin administration. SUL-treated rats, thus, appear to exhibit an unusual type of weight gain, characterized by normal circulating leptin and insulin levels. Such a particular leptin profile may be related to hyperprolactinemia, hypogonadism or lack of hyperinsulinemia. Molecular Psychiatry (2000) 5, 70-76.