Glucocorticoids and hypothalamic obesity

Download Glucocorticoids and hypothalamic obesity

Post on 02-Jul-2016




2 download

Embed Size (px)


<ul><li><p>Neuroscience &amp; Biobehavioral Reviews, Vol. 12. pp. 29--37. Pergamon Press plc, 1988. Printed in the U.S.A. 0149-7634/88 $3.00 + .00 </p><p>Glucocorticoids and Hypothalamic Obesity </p><p>BRUCE M. KING </p><p>Department of Psychology, University of New Orleans, Lakefront, New Orleans, LA 70148 </p><p>Received 3 August 1987 </p><p>KING, B. M. Glucocorticoids and hypothalanlic obesity. NEUROSCI BIOBEHAV REV 12(1) 29-37, 1988.--Recent studies have demonstrated a role for adrenal glucocortico~d hormones in the hyperphagia and obesity which follow lesions of the ventromedial hypothalamus (VMH). Although VMH lesions elevate morning plasma corticosterone levels, it is concluded that this contributes little to the development of obesity. More importantly, animals with VMH lesions appear to be hyperresponsive to very low levels of circulating glucocorticoids. The overeating and obesity are both prevented and reversed by either complete adrenalectomy or complete hypophysectomy (i.e., resulting in plasma corticosterone levels of less than 1.0 p.g/dl) and restored by dosages of glucocorticoids that have no effect on feeding behavior and weight gain in nonlesioned adrenalectomized animals. Mineralocorticoid hormones have no effect on hypothalamic obesity. Judging by the time course of effects on feeding behavior in VMH-damaged mice of a single intracerebroventricular injection of a low dose of glucocorticoid, which has no effect when administered intraperitoneally, it is concluded that glucocorticoids exert their effect centrally in a permissive, rather than a regulatory, manner. Stimulation of the neighboring paraventricular nuclei (PVN) with norepinephrine or neuropeptide Y produces a rapid feeding response which is also abolished by adrenalectomy and restored with administration of glucocorticoids. However, it is unlikely that the PVN is the site at which glucocorticoids exert their effect in animals with VMH lesions, for PVN lesions or knife-cuts, or combination VMH-PVN lesions, also result in hyperphagia and obesity. It is concluded that adrenal glucocorticoid hormones exert their permissive effects on feeding behavior at brain sites other than the medial hypothalamus. The septo-hippocampal complex is suggested as a possible site. </p><p>Ventromedial hypothalamus Paraventricular nuclei Feeding behavior Obesity Adrenal glands </p><p>Pituitary Glucocorticoid hormones </p><p>DAMAGE to the area of the basomedial hypothalamus re- sults in obesity in a variety of species [3, 12, 15]. Obesity in humans resulting from tumors or injury in this region was first described in 1840 [88] and later became known as Frrhlich's syndrome [44]. It was initially attributed to an endocrine imbalance due to impaired pituitary function, but this was questioned by several investigators who observed obesity in humans or animals with hypothalamic, but no di- rect pituitary, damage [5, 38, 109]. In one of the first experi- ments to use the stereotaxic instrument after it was modified for use with rats, Hetherington and Ranson [59] reported that lesions of the ventromedial hypothalamus (VMH) were most effective in producing obesity. The abnormal weight gains resulting from such lesions were generally attributed to a disinhibition of feeding behavior [118] and the role of the endocrine system was minimized. </p><p>Later studies found that rats with VMH lesions gained more weight and/or fatty tissue than control animals during food restriction or pair-feeding [53, 54, 56], thus demonstrat- ing a primary (i.e., lesion-induced rather than feeding- induced) metabolic deficit. When VMH-lesioned rats were found to have greatly exaggerated levels of plasma insulin [52], even when pair-fed with controls [23, 48, 55], a hor- monal role in hypothalamic obesity was once again empha- sized. In fact, several investigators attributed VMH obesity almost entirely to exaggerated vagally-mediated insulin re- sponses [23, 66, 67, 96]. However, alterations in inter- mediary metabolism which enhance lipogenesis and lead to excessive fat accumulation had been demonstrated following VMH lesions even in a state of insulin deficiency [49]. In </p><p>addition, animals with VMH lesions made diabetic by pan- createctomy or injection of alloxan or streptozotocin were found to gain more weight than nonlesioned animals during controlled insulin administration [43,127, 133]. These results suggested that insulin was not the only factor important in lipid deposition in the VMH-iesioned animal. Based on results with vagally-transected animals, King and Frohman [73] concluded that hyperinsulinemia could account for only 40 percent of the weight gain after VMH lesions. </p><p>Adrenal glucocorticoids, which are also involved in carbohydrate metabolism, have long been known to be associated with various obesity syndromes. Hyperad- renocortical activity is associated with the hyperphagia and obesity in Cushing's syndrome [25], whereas adrenocortical insufficiency is associated with anorexia and weight loss in Addison's disease Ill2]. Elevating glucocorticoid levels in rodents has been found to produce abnormal weight gain and/or increased carcass fat, and increased gluconeogenesis and insulin production [62,86]. Genetically obese mice (ob/ob) and rats (fa/fa) and diabetic mice (db/db) have all been found to have elevated serum corticosterone levels [13, 20, 35, 57, 93, 94, 136], while adrenalectomy in these obese rodents slows or normalizes food intake and weight gain [11, 19, 100, 113, 114, 134]. </p><p>Despite this abundance of evidence with other models of obesity, glucocorticoid hormones have only recently been implicated in hypothalamic obesity. In fact, adrenalectomy appears to have an even more dramatic effect in animals with VMH lesions than in genetically obese animals, resulting in a complete reversal of the obesity rather than just a slowing of </p><p>29 </p></li><li><p>30 KING </p><p>the abnormal weight gain [18, 28, 30, 31, 71, 75]. This paper reviews the results of these studies and explores the possible mechanism(s) and site(s) at which glucocorticoids exert their effect. </p><p>VMH LESIONS AND ADRENOCORTICAL ACTIVITY </p><p>Nearly all vertebrate species display a circadian rhythm in adrenocortical activity. In rodents, which are nocturnal animals, plasma corticosterone levels reach their zenith within four hours of the transition to the dark phase of a 12 hour light/dark cycle and are at their nadir during the first few hours of the light phase of the cycle [129]. Rodents dis- play most of their physical activity and food and water in- gestion during the dark portion of the cycle, When main- tained under conditions of constant light or dark, corticoste- rone levels free-run (i.e., are not entrained to light-dark) with a period of about 24 hours unless the animals are kept on a restricted feeding schedule, in which case hormone levels become entrained around the presentation of food [76,92]. Plasma corticosterone levels rise during food presentation, or in anticipation of the presentation of food, and then rapidly decline during the course of a meal [21, 61, 129]. </p><p>Complete isolation of the mediobasal hypothalamus re- sults in disruption of the diurnal rhythm in corticosterone secretion [95] and inhibits the adrenocortical response to some, but not all stimuli [40, 50, 95]. Several early experi- ments reported that lesions of the mediobasal hypothalamus which included the median eminence impaired the pituitary- adrenocortical response to stress [16, 32, 45, 63, 87, 107]. There were no abnormal weight gains reported in any of these experiments, however, so it remained unknown whether typical hyperphagia-inducing VMH lesions affected pituitary-adrenocortical activity. A few subsequent experi- ments reported that obesity-inducing VMH lesions resulted in a significant reduction in adrenal weights [8, 9, 58, 130], but several others found no change [15, 17, 68, 85, 110, 119]. </p><p>The critical factor appears to be whether or not VMH lesions extend into the median eminence. Szent~gothai et al. [124] reported that damage to the median eminence that produced adrenal atrophy could impair the development of obesity in rats with medial hypothalamic lesions. King, Levine and Grossman [74] found that not only did rats with obesity-inducing VMH lesions which spared the median eminence have normal adrenal weights, but that the animals had significantly higher a.m. (i.e., first few hours of the light phase) plasma corticosterone levels. The adrenocortical re- sponse to shock-induced stress was normal. Several subse- quent experiments have also found that VMH lesions result in greater than normal a.m. plasma corticosterone levels [22, 72, 77, 99, 105]. Mice made obese by administration of gold thioglucose were reported not only to have markedly ele- vated corticosterone levels, but significantly larger adrenal weights as well [105]. </p><p>It is important to note that the lesion-induced elevations in plasma corticosterone are not observed throughout the day and night, but only in the first few hours of the light portion of the cycle (i.e., when levels in normal rats are at their nadir). VMH lesions appear to disrupt the diurnal rhythm in corticosterone levels [6,77], with glucocorticoid concentrations remaining relatively flat throughout the 24- hour cycle (i.e., intermediate between the zenith and nadir observed in normal animals). Dallman [26] has proposed that this is the result of the lesions interrupting fibers from the suprachiasmatic nuclei, lesions of which also abolish the cir- </p><p>1D \ </p><p>tu Z 0 (r w k- U~ 0 0 k- </p><p>0 </p><p>50 </p><p>40 </p><p>30 </p><p>20 </p><p>10 </p><p>I - </p><p>O UJ </p><p>O </p><p>450 </p><p>400 </p><p>350 </p><p>300 </p><p>250 </p><p>200 </p><p>o o Sham VMH </p><p>~ Lean VMH j I </p><p>o o Obese VMH </p><p>I I I I I 0 2 5 15 25 </p><p>u.I ,v, </p><p>I-- z </p><p>C~ O O U. </p><p>50 </p><p>40 </p><p>30 </p><p>20 </p><p>10 </p><p>I I I I 0 0 2 5 15 25 </p><p>DAYS </p><p>FIG. 1. Mean (_SEM) plasma corticosterone levels (top), body weight (middle), and food intake (bottom) of sham-operated rats (n=7). lean VMH rats that were food-restricted to their level of preoperative intake (n=6), and obese VMH rats that were fed ad lib (n=6). (Reproduced from King, Calvert, Esquerr6 Kaufman and Frohman [72].) </p></li><li><p>GLUCOCORTICOIDS AND HYPOTHALAMIC OBESITY 31 </p><p>O) </p><p>I - "I- (.9 UJ </p><p>0 113 </p><p>600 </p><p>550 </p><p>500 </p><p>450 </p><p>400 </p><p>350 </p><p>300 </p><p> VMH- -ADX o o VMH- -SADX - . . . . -~ SVMH-AOX O- . . . . -o SVMH- -SADX = - ADX- -VMH </p><p>. . . . -o ADX- -SVMH O </p><p>.t .t SADX- -VMH m b- . . . . ~ SADX- -SVMH </p><p>- n=6 O4 </p><p>o </p><p>,t,- </p><p>250 I I I I I I I I I I 1 I I I I I 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 </p><p>DAYS </p><p>FIG. 2. Mean body weight (g-+SE) of 8 groups of female rats (n=6/group) with combinations of adrenalectomy (ADX) or sham-adrenalectomy (SADX) and ventromedial hypothalamic lesions (VMH) or sham lesions (SVMH). (Reproduced from Bruce, King, Phelps and Veitia ll8].) </p><p>cadian rhythmicity of corticosterone in ad lib fed animals [1,91]. The role of the VMH in adrenocortical activity can not simply be inhibitory, however, for this does not adequately explain the less than normal p.m. corticosterone levels observed in VMH-lesioned rats [6,77]. </p><p>Unlike normal rats, or animals with suprachiasmatic le- sions, rats with VMH lesions on a restricted feeding schedule do not show a phase shift in activity or peak corti- costerone levels to the presentation of food [65,77]. Nor do they display the usual rapid decline in serum corticosterone levels with consumption of a meal [22]. It appears, therefore, that the VMH is the anatomical locus which mediates the circadian response to food synchronization [77]. </p><p>The critical question, of course, is just how important is the lesion-induced increase in a.m. plasma corticosterone levels to the hyperphagia and/or weight gain? King et al. [72] found the corticosterone elevations to be equally great in food-restricted (to normal body weight) and ad lib fed VMH rats on postoperative days 2 through 25 (see Fig. 1). It was concluded that the elevation in plasma glucocorticoids ob- served in VMH rats was a primary effect of the lesion and was independent of food intake or initial weight gain. There is some evidence that morning corticosterone levels return to normal after many weeks of food ad lib [74]. Obesity- inducing parasagittal knife cuts alongside the paraventricular nucleus have recently been reported to cause a decrease, rather than an increase, in corticosterone levels [121]. Ad- ministration of adrenocorticotropin (ACTH) reversed the ad- </p><p>renal deficit, but had no effect on food intake or weight gain [123]. Plasma concentrations and diurnal rhythmicity of cor- ticosterone have been reported to be normal in genetically obese rats [135]. It is concluded, therefore, that while the increase in a.m. corticosterone levels after VMH lesions may promote gluconeogenesis and abnormal weight gain, it is not essential for the development of obesity. </p><p>VMH LESIONS AND ADRENALECTOMY </p><p>Until recently, there had been very little attention de- voted to the effects of adrenalectomy on ventromediai hypo- thalamic obesity. In 1965, Bernardis and Skelton [8] reported that adrenalectomy had little or no effect on food intake and weight gain in weanling rats with VMH lesions. York and Bray [ 132] later claimed that adrenalectomy also had no effect in adult rats with VMH lesions, but the weight gains were uncommonly low for animals with hypothalamic lesions (about 85 g/42 days). Although a subsequent experiment by Mook, Fisher and Durr [90] reported that adrenalectomy partially suppressed abnormal weight gain in rats with VMH lesions, it was generally believed that the presence of adrenal corticosteroids was not essential for the development of hy- pothalamic obesity [1 I, 14]. </p><p>In the early 1980's, however, two independent series of experiments demonstrated that VMH obesity was critically dependent on adrenal glucocorticoids. Debons and his co- workers [28, 30, 31] reported that adrenalectomy completely </p></li><li><p>32 KING </p><p>prevented the development of obesity in mice treated with gold thioglucose, while King and his colleagues [18, 71, 75] found that removal of the adrenals completely prevented and reversed hyperphagia and obesity in rats given electrolytic VMH lesions (see Fig. 2). Adrenalectomy did not affect other motivated behaviors such as lever pressing for electri- cal stimulation of the brain [75]. Both groups of researchers verified completeness of adrenalectomy by directly measur- ing plasma levels of corticosterone. Adrenalectomies were considered to be complete by King eta / . [71,75], for exam- ple, only when stress-induced corticosterone levels were less than 1.0/zg/dl of plasma. Moderate weight gains were ob- served when corticosterone levels exceeded this criteria (see next section). Previous studies utilized only indirect verifi- cation tests, but residual adrenal tissue or accessory adrenal glands are common in rodents [111], which may account for the weight gains that were observed in these early experi- ments. </p><p>Hyperphagia and rapid weight gains were restored in both adrenalectomized gold thioglucose-treated mice and ad- renalectomized electrolytic lesio...</p></li></ul>


View more >