 |
 |
 |
|
|
Excessive body weight gain (BWG) induced by antipsychotic drugs (APs) was reported soon after the introduction of chlorpromazine in psychiatry (1). However, interest in this problem increased after 1990 because of the strong propensity of some new atypical agents to induce BWG (2–4). Multiple mechanisms are probably involved in AP-induced BWG, and most research has focused on AP effects on brain histamine, serotonin, dopamine, and central and peripheral acetylcholine, as well as on the metabolic-endocrine effects of hyperprolactinemia (1,5). Pollmächer and others (6) and Melkerson and others (7,8), have suggested a relation between AP-induced BWG and cytokines (mainly tumour necrosis factor alpha [TNF-a ]) and leptin. Authorities in the field now frequently mention this relation (4,9), and the implicit assumption is that leptin or TNF-a may cause excessive BWG. In this paper, we review the evidence relating leptin and TNF-a to AP-induced BWG. We first describe the main findings on leptin and TNF-a activity in cases of primary obesity. Second, we examine the studies that associate leptin and TNF-a to AP-induced obesity. Our discussion is based on the current theories of causality in chronic diseases. Leptin, TNF-a , and Primary ObesityLeptin is a protein synthesized in the adipose tissue and, in minor proportions, in the placenta, stomach, and muscles (10). A few people and rodents lack leptin receptors in key tissues or display very low leptin production, along with severe obesity and diverse endocrine-metabolic abnormalities (11,12). Administration of the peptide to obese rodents lacking leptin decreases body weight (BW) and restores fertility (13). Leptin administration also decreases BW in obese or nonobese mice and rats (14,15) and in people with obesity not related to a leptin deficiency (16). In most humans and rodents, leptin serum levels correlate positively with body mass index (BMI: weight [kg] / height [m2]), with percentage of body fat, and with basal serum insulin levels (17,18). As an example, subjects with obesity or anorexia nervosa (AN) display higher and lower serum leptin levels, respectively, than do people with a normal weight. Importantly, these high or low leptin levels tend to normalize when an adequate BW is reached (19). Leptin levels are also higher in women than in men, even after correction for BMI (20). Leptin is believed to be a messenger from the adipose tissue that signals the brain about the extent of body fat through a negative feedback regulatory mechanism. In turn, the nervous and endocrine systems should trigger a cascade of reactions to correct the increment or decrement in adipose tissue (11,12). However, an immediate question arises in relation to this model: Why don’t the increased leptin levels in obese patients correct excessive BW? It has been proposed that obesity is accompanied by a “leptin resistance” (10), a concept analogous to the insulin resistance observed in patients with diabetes and obesity. Such leptin resistance may be related to a decreased transport of the peptide through the blood–brain barrier and to postreceptor abnormalities (10). TNF-a is a protein of the cytokine family. It is mainly synthesized by adypocites and macrofages, but there is also local synthesis in the brain. It plays a prominent role in the mechanisms of tissue growth, inflammation, and immunity (6,21), and it is also involved in BW regulation. Intracerebral or systemic injections of TNF-a decreased food intake and BW in rodents (22). In addition, severe BW loss and anorexia were observed in mice transfected with the human TNF-a gene (23). In the adypocite, TNF-a stimulates thermogenesis and lipolysis and decreases lipogenesis. Thus, it collectively decreases body fat and protein mass (21). In the cachectic states accompanying diverse diseases, TNF-a is massively synthesized in the macrophages and is one of the main factors mediating the severe anorexia observed in these conditions. Interestingly, high circulating TNF-a levels are found in obese subjects and correlate positively with the BMI and insulin and leptin levels (24). Converging evidence demonstrates that TNF-a impairs insulin sensitivity and is a critical factor in the insulin resistance and diabetes mellitus associated with obesity (25). Argiles and others developed a model for TNF-a in BW regulation that resembles in some ways that of leptin (21). According to these authors, TNF-a functions in healthy subjects as an adipostat and assists the brain in preventing excessive fluctuations in BW. Hence, during cachexia, the activity of TNF-a is abnormally increased, whereas in obesity its action may be impaired (21). Data obtained in some animal experiments suggest that TNF-a may promote BWG (26,27), since nonobese male mice with a targeted disruption of this cytokine gene (TNF-a -/-) displayed less weight than their intact control littermates (TNF-a +/+) (27). In addition, when the 2 groups of animals were exposed to a high-fat diet that induces obesity, the fat pads in the TNF-a +/+ group weighed significantly more than in the TNF-a -/- group. As well, the BWG was nonsignificantly higher in the TNF-a +/+ group (26). However, using another model of obesity in mice (that is, the obese–obese model) and targeted mutations in 2 receptors for TNF-a , these authors found no difference in BWG and body composition between the obese mice with, or without, the receptor mutation (26). In a third model of obesity (induced by gold-thioglucose, which causes chemical ablation of the ventromedial hypothalamus and induces hyperphagic obesity), a similar degree of obesity was obtained in TNF-a +/+ and TNF-a-/- subjects (27). As a whole, these results show that TNF-a is not necessary to observe obesity in mice, even though the cytokine modulates BW in nonobese mice. Collectively, these data suggest that leptin and TNF-a are hormones normally involved in physiological mechanisms regulating BW. However, once obesity is established, both hormones display deleterious effects on glucose homeostasis, blood pressure, and immunity. Regarding the relation between leptin and the pathological regulation of BW, investigators are particularly attentive in describing subjects with abnormal or atypical correlations between leptin and the physiological variables that are relevant for BW regulation. It is thus of considerable scientific interest to find people in whom serum leptin levels do not correlate positively with BMI, with percentage of body fat, and with basal insulin levels or in whom changes in BW do not induce the expected changes in leptin levels. For example, subjects with hyperthyroidism or bulimia display lower leptin levels than would be expected for their BMI (28,29). The initial conceptualization of the role of leptin in human physiology has evolved. It is now acknowledged that the peptide may be involved in numerous regulatory systems close or distal to signalling the body-fat size, such as the systems regulating nutrient availability, reproduction, hematopoiesis, the immune system, and the function of the adrenal and thyroid axis, among others (10). Initially, it also seemed implicit that the elevated leptin levels observed in cases of obesity were innocuous. This assumption has also been challenged, and converging evidence points to a role for leptin in the development of myocardial infarction (30), hypertension (31), insulin resistance (10), and prostatic cancer (32).
|
||||||