The Effect of the Melanocortin Agonist, MT-II, on the Defended Level of Body Adiposity
http://www.100md.com
《内分泌学杂志》
Department of Psychiatry (R.J.S., K.A.W.), Genome Research Institute, University of Cincinnati, and Procter & Gamble Pharmaceuticals (M.L.B., C.C.Ma., O.R., C.C.Mc., D.P.T., M.C.G., R.J.S.), Mason, Ohio 45040
Abstract
A wide range of experimental evidence implicates a critical role for melanocortin signaling in the control of food intake and body adiposity. Melanocortin receptor agonists such as MT-II potently reduce food intake and body weight, making such agonists potential therapeutics for obesity. The critical concept addressed by the present experiments is whether the homeostatic effects of melanocortin agonists directly regulate food intake or whether the effects on food intake are secondary, with the primary effects being the regulation of body weight and adiposity. To investigate this, we compared the effect of various doses of MT-II given via osmotic minipump for 28 d to alter food intake, body weight, and body fat in dietary-induced obese rats. In addition, before the implantation of the minipump, dietary-induced obese rats were weight reduced by differing amounts using varying levels of food restriction. The results show that in food-restricted rats, MT-II-treated rats consume significantly more calories than those receiving MT-II after ad libitum access to food. More importantly, regardless of the widely differing levels of body fat among the different dietary treatments employed, body fat at the end of the study was determined exclusively by the dose of MT-II, with MT-II-treated rats having less body fat than vehicle-treated rats. These experiments support the hypothesis that melanocortin signaling primarily regulates total body adiposity and that food intake is adjusted as necessary to achieve a specific level of body adiposity.
Introduction
GORDON KENNEDY, in the 1950s, proposed that body adiposity is defended by the accurate matching of caloric intake to caloric expenditure over time (1). To do so requires that the central nervous system (CNS) have the ability to monitor the status of adipose stores and to use this information to adjust both caloric intake and energy expenditure. Convincing evidence indicates that hormones play this critical role as "adiposity signals" and provide negative feedback about how body fat stores are changing (2, 3). One hormone hypothesized to be such an adiposity signal is the adipocyte-derived hormone, leptin. Leptin circulates in proportion to the amount of adipose tissue and crosses the blood-brain barrier where it interacts with a known receptor that is located in several regions of the CNS, including densely within the arcuate nucleus of the hypothalamus (4, 5). CNS administration of leptin results in decreased food intake and weight loss, whereas genetic leptin deficiency and genetic leptin resistance results in increased food intake and greatly increased levels of body adiposity (6). Leptin may not be the only adiposity signal. Insulin may also function as an adiposity signal, inasmuch as it also circulates in proportion to body adipose mass. Like leptin, insulin crosses the blood-brain barrier where it interacts with its receptor that is expressed heavily within the arcuate nucleus of the hypothalamus (7). Most critically, CNS administration of insulin results in decreased food intake and body weight, whereas reduced CNS insulin action via central administration of insulin antibodies or by targeted genetic disruption of the CNS insulin receptor results in enhanced food intake and weight gain (8, 9). Thus, diverse evidence points to both leptin and insulin as critical signals from the periphery to the CNS about the status of peripheral energy stores.
An important conceptual point is that adiposity signals influence food intake but do so to influence the amount of body adipose mass. That is to say, food intake is reduced by administration of these signals but only to achieve a lower body adipose mass. One piece of evidence in support of this hypothesis is that continuous infusion of either leptin or insulin does not result in sustained reductions in food intake. Rather, food intake is suppressed during active weight loss but returns to normal once a new level of body weight (and presumably body adiposity) is achieved (10, 11). Thus, it would appear that once the animal has achieved a level of body adiposity that is appropriate to the degree of leptin or insulin signaling, leptin and insulin lose their ability to suppress food intake. A more direct test of the hypothesis that the degree of suppression of food intake elicited by an adiposity signal depends on the current level of body adiposity employs a paradigm first developed by Powley and Keesey (12). In this paradigm, animals are food restricted until they achieve a level of body weight equivalent or below that predicted to be achieved after exogenous administration of the signal. If the signal merely suppressed food intake, its effect should be the same in the weight-reduced state as it was in the ad libitum-fed state. Alternatively, if the signal is actually a component of an adiposity signaling pathway, then food intake after the infusion of the substance should be suppressed only in the ad libitum-fed condition and less so or not at all in the weight-reduced state. In the case of insulin, the data are quite clear. Insulin infusion in the weight-reduced state was much less effective at suppressing food intake than in the ad libitum-fed condition (11). Even more importantly, the ultimate level of body weight achieved to insulin administration was identical in both the weight-reduced and ad libitum-fed conditions. Such data strongly support the notion that insulin is an adiposity signal that influences food intake only to the extent that is necessary to achieve a new defended level of body adiposity.
Within the CNS, the control of energy balance is regulated by a complex neural network involving a number of systems both within and beyond the hypothalamus. Some of these systems are direct targets for the actions of adiposity signals. In particular, the central melanocortin system appears to be an important mediator of the actions of both leptin and insulin (13). The endogenous melanocortin agonist, -MSH, is produced within neurons of the arcuate nucleus from the precursor peptide proopiomelanocortin. Exogenous administration of melanocortin agonists elicits a strong reduction in food intake and body weight and either pharmacological antagonists or genetic disruption of the melanocortin-4 receptor produces increased food intake and concomitant weight gain (14, 15). These -MSH-producing neurons express both leptin and insulin receptors, and both leptin and insulin stimulate expression of proopiomelanocortin, the precursor to -MSH (16). Moreover, blockade of melanocortin receptors suppresses the ability of both insulin and leptin to suppress food intake and produce weight loss, supporting that both adiposity factors mediate their food intake effects by activating central melanocortin pathways (16, 17).
Because the melanocortin system is critically regulated by the actions of adiposity signals, it is tempting to hypothesize that exogenous administration of melanocortin receptor agonists also alters the defended level of body adiposity, influencing food intake only secondarily to achieve a specific level of body adiposity. To test this hypothesis, we employed the Powley and Keesey strategy (12) of comparing the ability of the stable -MSH analog, MT-II, to suppress food intake and cause weight loss in rats at three different levels of body adiposity. If activity of the melanocortin system is tied to a specific level of energy intake directly, we would expect that previous food restriction should have little impact on the amount of calories consumed during MT-II treatment and this would be accompanied by differing levels of body adiposity. Alternatively, if the activity of the melanocortin system is tied to the defended level of body adiposity, the amount of calories consumed during MT-II treatment should be different between the ad libitum-fed and food-restricted conditions but they should work to achieve identical levels of body adiposity.
Materials and Methods
Study design
Seventy-two male Wistar rats (Charles-River, Portage, MI) with an approximate average starting weight of 275 g were all placed on a calorically dense 45% high-fat (HF) diet (Research Diets, Inc., New Brunswick, NJ; D12541) for 16 wk (see Fig. 1). At the end of this period, rats weighed approximately 675 g. At this point, rats were divided into three weight-matched groups. One group continued to receive ad libitum access to the HF diet (AD LIB group). The other two groups were food restricted either by 30% (RES-30) or 60% (RES-60) of their baseline food intake for a period of 14 d. At this point, rats in each group were randomized to one of three treatment conditions: vehicle (50 mM sodium acetate, pH = 5.0), MT-II (Bachem, Inc., Torrance, CA) at 0.3 mg/kg·d, or MT-II at 3.0 mg/kg·d. MT-II or vehicle was delivered sc via surgically implanted 28-d minipumps (model 2ML4; Alza, Palo Alto, CA). During the next 28 d, all rats had ad libitum access to HF diet, after which they were euthanized by CO2 inhalation followed by exsanguination via a cardiac puncture. The terminal blood sample was collected into heparinized vials and centrifuged to produce plasma that was stored at –80 C for subsequent analysis.
Body composition analysis
Body composition of rats was determined at several points in the course of the study using a QDR-4500A dual x-ray absorptiometry (DXA) scanner (Hologic, Inc., Bedford, MA). Briefly, whole body composition analysis was accomplished in isoflurane-anesthetized rats (2% isoflurane delivered with 100% oxygen) using software version 10.0 and protocols that were provided by the manufacturer. On each scan day, quality control and tissue calibration standards were scanned first according to the manufacturer’s instructions. Rats were scanned in the prone position, using a laser starting point that was positioned at the midline of the rat body just distal to the nose. Each DXA scan was acquired using the rat whole body protocol with a 7.1-in. width and a 14.2-in. length, with a total scan time of about 2.4 min per rat.
Statistical analyses
A one-way ANOVA model was fit to body weight, DXA fat mass, and DXA lean mass data (see Fig. 2). This model included the term for group [vehicle, MT-II (0.3 mg/kg·d), or MT-II (3.0 mg/kg·d)] and was fit by day. Where appropriate, the ANOVA was followed by pair-wise Student’s t tests. Additional body weight and food intake analyses were done in a similar fashion: an appropriate ANOVA model was fit to the endpoint and was followed by pair-wise Student’s t tests to perform the comparisons of interest. In all comparisons, statistical significance was assigned as P < 0.05. A two-way ANOVA model was fit to DXA fat mass and DXA lean mass, including terms for day (0, 14, or 28), group (vehicle, MT-II 0.3 mg/kg·d, or MT-II 3.0 mg/kg·d), and the day*group interaction (see Fig. 4). The ANOVA was followed by pair-wise Student’s t tests for comparisons between days by group, and between groups by day.
Results
All rats gained significant weight over the course of the 16-wk exposure to the HF diet, and all groups had the same amount of total body fat before the food restriction paradigms began (see Fig. 2B; d –14). As expected, food restriction caused significant weight loss in both the RES-30 and RES-60 groups, with the greatest weight loss in the RES-60 group (see Fig. 2A). Also as expected, both the RES-30 and RES-60 groups lost significant amounts of body fat to a degree that was proportional to the level of food restriction when compared with the AD LIB group (see Fig. 2B). Food restriction also resulted in a small, but statistically significant reduction in DXA-derived lean mass that again was proportional to the degree of food restriction (see Fig. 2C).
After rats were implanted with the minipump, the AD LIB rats that were treated with 0.3 mg/kg·d or 3.0 mg/kg·d MT-II showed similar weight loss that was different from that seen with the vehicle-treated rats (see Fig. 3A). Consistent with the weight loss, both doses of MT-II suppressed food intake to a similar degree (see Fig. 3B). The food intake suppression induced by either dose of MT-II was strongest during the first few days of infusion, with a slow return of food intake toward baseline levels (approximately 20 g/d) over the first 2-wk period. It is important to note that during the last 2 wk of the MT-II infusion, food intake had returned to the level of the vehicle-infused group despite the maintenance of the weight loss.
In both the RES-30 and RES-60 groups, vehicle-treated rats were significantly hyperphagic for the first 2 wk of ad libitum feeding after being released from the food constraint on d 0; normal food intake levels resumed during the final 2 wk of vehicle treatment (see Fig. 3, D and F). RES-30 and RES-60 rats treated with vehicle also showed significant weight gain over the 28 d that brought them to a body weight approaching that of the AD LIB vehicle-treated rats by d 28 (see Fig. 3, C and E). RES-30 or RES-60 food-restricted rats treated with MT-II showed a dose-dependent blunting of the hyperphagia compared with the food intake of vehicle-treated animals. In the first 14 d after minipump implantation, the MT-II was more effective at suppressing food intake at the higher dose than at the lower dose, although as in the AD LIB setting, during the last 14 d of treatment there were no significant differences in daily food intake among the MT-II- and vehicle-treated rats (see Fig. 3, D and F). In both RES-30 and RES-60 conditions, both doses of MT-II caused a similar degree of sustained body weight loss over the 28-d infusion period compared with the vehicle treatment (see Fig. 3, C and E).
A key aspect of these studies is the comparison of the effects of MT-II between the different dietary conditions. Although MT-II clearly suppressed food intake at all three levels of food restriction, the actual amount of calories consumed during MT-II treatment differed across the three levels of restriction. In the two-way ANOVA (dose x diet), the cumulative amount consumed either 1, 7, or 14 d after minipump implantation was significantly influenced by both the dose of MT-II (P < 0.0001) and the level of food restriction (P < 0.0001). On the first day, there was also a significant interaction between these two variables (P = 0.028) that was not apparent on cumulative intake after 7 or 14 d. The implication of this analysis is that, to predict the amount of food consumed, we would need to know both the degree of food restriction as well as the dose of MT-II delivered. This analysis also implies that the effect size of the MT-II on food intake is different across the dietary groups only on the first day but not on cumulative intake over 7 and 14 d. There was also a difference in the nature of the dose-effect curve across the three restriction levels. In AD LIB rats, there was no difference between the 0.3 and 3.0 dose of MT-II, although there was a difference in both the RES-30 and RES-60 conditions.
Body weights of MT-II-treated rats were not significantly different regardless of starting nutritive state of rats (AD LIB, RES-30, or RES-60; see Table 1). The effects of 28 d of MT-II treatment on body composition of rats at 14 and 28 d of treatment are shown in Fig. 4. MT-II treatment at either the 0.3-mg/kg·d or 3.0-mg/kg·d dose level caused similar degrees of fat mass loss in AD LIB-, RES-30-, and RES-60-treated rats (see Fig. 4, A, C, and E). Interestingly, the majority of fat loss due to MT-II treatment was observed by d 14 of treatment, with only minimal further fat loss induced by an additional 14 d of treatment. This temporal pattern of fat loss corresponds with the temporal pattern of food intake suppression induced by MT-II in both food-restricted and ad libitum-feeding rats (see Fig. 3). Although food restriction induced a significant reduction of lean mass as mentioned previously, MT-II treatment failed to significantly affect lean mass when compared with vehicle-treated rats in either food-restricted or ad libitum conditions (see Fig. 4, B, D, and F). In fact, lean mass loss due to food restriction in RES-60 rats appeared to recover over the subsequent 28 d of ad libitum feeding, even in MT-II-treated rats (see Fig. 4F).
Among the most important comparisons in the study was the examination of the effects of MT-II treatment between the ad libitum and food-restricted groups. Table 1 displays the terminal body weight and body fats for each of the nine groups. In the vehicle-treated rats, both food-restricted groups showed rapid weight gain once ad libitum access to food was allowed, and by the end of the 28-d period, their body weights and body fats had almost returned to the level of the nonrestricted group. Most interesting is the comparison of the body weight achieved by the MT-II-treated groups. In the rats treated with 0.3 mg/kg·d dose of MT-II, both the food-restricted and ad libitum-fed rats achieved nearly identical body weights and body fats. Thus for this dose of MT-II, the starting point of the animal was not important in determining the final body fat. In the rats treated with the 3.0 mg/kg dose of MT-II, the data are similar. Both the 30 and 60% restriction groups treated with MT-II achieve identical body weights, but both groups trended for a lower body weight than the ad libitum-fed rats, although this trend did not achieve statistical significance. Body fats between all three treatments receiving the high dose of MT-II were clearly not different. Thus, the trend for lower body weight with higher dose of MT-II in food-restricted vs. ad libitum rats was driven not by differences in body fat, but instead was due to an enhanced lean mass effect in the ad libitum rats. This was probably a reflection of the lower lean mass attained during the food restriction period in nutritionally restricted rats, which was maintained during MT-II treatment. The results with body fat are exactly paralleled by those of plasma leptin levels. There was no difference between dietary groups in plasma leptin levels measured at the conclusion of the experiment, but plasma leptin levels are reduced in all of the MT-II-treated groups (see Table 1).
Discussion
These data examine the effect of chronic administration of a melanocortin receptor agonist to influence food intake, body weight, and body composition. The first finding of these studies is that chronic sc infusion of MT-II is quite effective to produce reductions in food intake, body weight, and body fat even in rats made quite obese by prolonged exposure to a HF diet. This result agrees with other work using melanocortin agonists in both mouse and rat and separates that work from what has been generally reported with peripheral administration of leptin that appears to lose most of its efficacy in rats made obese after exposure to a HF diet (18, 19, 20).
The second finding of these studies is that several aspects of the effects of MT-II are altered by prior food restriction. MT-II-treated rats in either of the restricted groups consumed more than MT-II-treated rats that were ad libitum fed. However, this analysis does not take into account the different baseline intakes of the ad libitum and restricted groups. When one compares the effect size of MT-II to its vehicle-treated comparison group, our statistical analysis shows that the effect size is different only on the first day after minipump implantation but not on 7 or 14 d cumulative intake.
Importantly, the ability of MT-II to induce weight loss is much less evident in the restricted rats that weighed less at the beginning of the experiment than the ad libitum-fed rats. Thus, the effect of MT-II to induce body weight loss is highly dependent on the state at which the rats began the experiment. From one perspective this implies that MT-II was less effective in the restricted rats. In the case of the melanocortin system, activity at CNS melanocortin receptors is regulated not just by the presence of endogenous agonists but also by endogenous antagonists/inverse agonists. Agouti-related peptide (AgRP) is produced in arcuate nucleus cell bodies and is carefully regulated by peripheral adiposity signals such as leptin (13, 21, 22). Thus, when leptin levels are reduced, such as in the restricted groups in this experiment, hypothalamic AgRP levels should be elevated. Increased AgRP in the synapse would work to block melanocortin-4 receptors and decrease the ability of exogenous MT-II to inhibit food intake (23). Consequently, we can point to a straightforward hypothesis about why MT-II would be less effective to reduce food intake and induce weight loss in the restricted groups.
The greater number of calories consumed in restricted rats treated with MT-II stands in contrast to the effects on body fat. In this regard, the body fat levels achieved by rats treated with MT-II are similar regardless of whether the animal started the dosing regimen at a lower body fat because of food restriction. That is to say, in the animals whose body fat was reduced by enforced food restriction, the effect of MT-II was to maintain that body fat loss at a level identical to the body fat loss induced by the same dose of MT-II in ad libitum-fed rats with higher body fats.
This finding has two implications. The first is that melanocortin agonists that might be used to treat human obesity should be equally effective when used either to produce weight loss or to maintain weight loss already achieved. Thus, melanocortin agonists could be used to help individuals maintain weight loss that was achieved by other means and improve compliance to dietary regimens designed to maintain lost weight.
The second implication is more theoretical and concerns the parameters that are primarily regulated by the melanocortin system. To predict how much an animal will eat in these experiments you would need to know both the dose of MT-II it was receiving and the degree of prior food restriction. This stands in contrast to the degree of body adiposity. To predict the amount of body fat at the end of the experiment one would need to only know whether the animal received MT-II or vehicle, because prior restriction had no impact on the achieved level of body fat. Our conjecture is that this outcome supports the hypothesis that melanocortin signaling primarily regulates body adiposity. From this perspective, changes in food intake are simply a means to an end, and that end is the maintenance of a level of body fat consistent with the amount of melanocortin signaling. Further support for this hypothesis comes from the fact that, in MT-II-treated rats, food intake had returned to normal by d 14. By d 14, rats had already achieved a level of body fat that would be maintained through d 28. Consequently, the dissipating effect of MT-II on food consumption is not "tachyphalaxis" but rather is the result of animals now consuming the appropriate number of calories to maintain a specific level of adiposity consonant with their degree of melanocortin signaling. As a consequence, it might be more appropriate to term the melanocortin system as an effector pathway for regulating adiposity, rather than as an anorexigenic or satiety pathway per se.
Footnotes
Abbreviations: AgRP, Agouti-related peptide; CNS, central nervous system; DXA, dual x-ray absorptiometry; HF, high fat.
References
Kennedy GC 1953 The role of depot fat in the hypothalamic control of food intake in the rat. Proc R Soc Lond (Biol) 140:579–592
Porte DJ, Baskin DG, Schwartz MW 2002 Leptin and insulin action in the central nervous system. Nutr Rev 60:S20–S29
Friedman JM 2002 The function of leptin in nutrition, weight, and physiology. Nutr Rev 60:S1–S14;discussion S68–S84, S85–S87
Saper CB, Chou TC, Elmquist JK 2002 The need to feed: homeostatic and hedonic control of eating. Neuron 36:199–211
Ahima RS, Saper CB, Flier JS, Elmquist JK 2000 Leptin regulation of neuroendocrine systems. Leptin regulation of neuroendocrine systems. 21:263–307
Seeley RJ, Woods SC 2003 Monitoring of stored and available fuel by the CNS: implications for obesity. A cortical-hippocampal system for declarative memory. 4:901–909
Woods SC, Seeley RJ, Baskin DG, Schwartz MW 2003 Insulin and the blood-brain barrier. Insulin and the blood-brain barrier 9:795–800
McGowan MK, Andrews KM, Grossman SP 1992 Chronic intrahyphothalamic infusions of insulin or insulin antibodies alter body weight and food intake in the rat. Physol Behav 51:753–766
Brüning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, Klein R, Krone W, Müller-Wieland D, Kahn CR 2000 Role of brain insulin receptor in control of body weight and reproduction. Science 289:2122–2125
Seeley RJ, van Dijk G, Campfield LA, Smith FJ, Nelligan JA, Bell SM, Baskin DG, Woods SC, Schwartz MW 1996 The effect of intraventricular administration of leptin on food intake and body weight in the rat. Horm Metab Res 28:664–668
Chavez M, Kaiyala K, Madden LJ, Schwartz MW, Woods SC 1995 Intraventricular insulin and the level of maintained body weight in rats. Behav Neurosci 109:528–531
Powley TL, Keesey RE 1970 Relationship of body weight to the lateral hypothalamic feeding syndrome. Absence of satiety during sham feeding in the rat. 70:25–36
Seeley RJ, Drazen DL, Clegg DJ 2004 The critical role of the melanocortin system in the control of energy balance. Annu Rev Nutr 24:133–149
Cone RD, Cowley MA, Butler AA, Fan W, Marks DL, Low MJ 2001 The arcuate nucleus as a conduit for diverse signals relevant to energy homeostasis. Int J Obes Relat Metab Disord 25(Suppl 5):S63–S67
Cone RD 1999 The central melanocortin system and energy homeostasis. Melanin-concentrating hormone receptor: an orphan receptor fits the key. Trends Endocrinol Metab 10:211–216
Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, Woods SC 2002 The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 22:9048–9052
Seeley RJ, Yagaloff KA, Fisher SL, Burn P, Thiele TE, van Dijk G, Baskin DG, Schwartz MW 1997 Melanocortin receptors in leptin effects. Nature 390:349
Pierroz DD, Ziotopoulou M, Ungsunan L, Moschos S, Flier JS, Mantzoros CS 2002 Effects of acute and chronic administration of the melanocortin agonist MTII in mice with diet-induced obesity. Diabetes 51:1337–1345
Clegg DJ, Benoit SC, Air EL, Jackman A, Tso P, D’Alessio D, Woods SC, Seeley RJ 2003 Increased dietary fat attenuates the anorexic effects of intracerebroventricular injections of MTII. Endocrinology 144:2941–2946
Li G, Zhang Y, Wilsey JT, Scarpace PJ 2004 Unabated anorexic and enhanced thermogenic responses to melanotan II in diet-induced obese rats despite reduced melanocortin 3 and 4 receptor expression. J Endocrinol 182:123–132
Wilson BD, Ollmann MM, Barsh GS 1999 The role of agouti-related protein in regulating body weight. Mol Med Today 5:250–256
Bagnol D, Lu XY, Kaelin CB, Day HE, Ollmann M, Gantz I, Akil H, Barsh GS, Watson SJ 1999 Anatomy of an endogenous antagonist: relationship between Agouti-related protein and proopiomelanocortin in brain. J Neurosci 19:RC26
Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg HT, Woods SC, Seeley RJ 2000 Long-term orexigenic effects of AgRP-(83–132) involve mechanisms other than melanocortin receptor blockade. Am J Physiol 279:R47–R52(Randy J. Seeley, Melissa )
Abstract
A wide range of experimental evidence implicates a critical role for melanocortin signaling in the control of food intake and body adiposity. Melanocortin receptor agonists such as MT-II potently reduce food intake and body weight, making such agonists potential therapeutics for obesity. The critical concept addressed by the present experiments is whether the homeostatic effects of melanocortin agonists directly regulate food intake or whether the effects on food intake are secondary, with the primary effects being the regulation of body weight and adiposity. To investigate this, we compared the effect of various doses of MT-II given via osmotic minipump for 28 d to alter food intake, body weight, and body fat in dietary-induced obese rats. In addition, before the implantation of the minipump, dietary-induced obese rats were weight reduced by differing amounts using varying levels of food restriction. The results show that in food-restricted rats, MT-II-treated rats consume significantly more calories than those receiving MT-II after ad libitum access to food. More importantly, regardless of the widely differing levels of body fat among the different dietary treatments employed, body fat at the end of the study was determined exclusively by the dose of MT-II, with MT-II-treated rats having less body fat than vehicle-treated rats. These experiments support the hypothesis that melanocortin signaling primarily regulates total body adiposity and that food intake is adjusted as necessary to achieve a specific level of body adiposity.
Introduction
GORDON KENNEDY, in the 1950s, proposed that body adiposity is defended by the accurate matching of caloric intake to caloric expenditure over time (1). To do so requires that the central nervous system (CNS) have the ability to monitor the status of adipose stores and to use this information to adjust both caloric intake and energy expenditure. Convincing evidence indicates that hormones play this critical role as "adiposity signals" and provide negative feedback about how body fat stores are changing (2, 3). One hormone hypothesized to be such an adiposity signal is the adipocyte-derived hormone, leptin. Leptin circulates in proportion to the amount of adipose tissue and crosses the blood-brain barrier where it interacts with a known receptor that is located in several regions of the CNS, including densely within the arcuate nucleus of the hypothalamus (4, 5). CNS administration of leptin results in decreased food intake and weight loss, whereas genetic leptin deficiency and genetic leptin resistance results in increased food intake and greatly increased levels of body adiposity (6). Leptin may not be the only adiposity signal. Insulin may also function as an adiposity signal, inasmuch as it also circulates in proportion to body adipose mass. Like leptin, insulin crosses the blood-brain barrier where it interacts with its receptor that is expressed heavily within the arcuate nucleus of the hypothalamus (7). Most critically, CNS administration of insulin results in decreased food intake and body weight, whereas reduced CNS insulin action via central administration of insulin antibodies or by targeted genetic disruption of the CNS insulin receptor results in enhanced food intake and weight gain (8, 9). Thus, diverse evidence points to both leptin and insulin as critical signals from the periphery to the CNS about the status of peripheral energy stores.
An important conceptual point is that adiposity signals influence food intake but do so to influence the amount of body adipose mass. That is to say, food intake is reduced by administration of these signals but only to achieve a lower body adipose mass. One piece of evidence in support of this hypothesis is that continuous infusion of either leptin or insulin does not result in sustained reductions in food intake. Rather, food intake is suppressed during active weight loss but returns to normal once a new level of body weight (and presumably body adiposity) is achieved (10, 11). Thus, it would appear that once the animal has achieved a level of body adiposity that is appropriate to the degree of leptin or insulin signaling, leptin and insulin lose their ability to suppress food intake. A more direct test of the hypothesis that the degree of suppression of food intake elicited by an adiposity signal depends on the current level of body adiposity employs a paradigm first developed by Powley and Keesey (12). In this paradigm, animals are food restricted until they achieve a level of body weight equivalent or below that predicted to be achieved after exogenous administration of the signal. If the signal merely suppressed food intake, its effect should be the same in the weight-reduced state as it was in the ad libitum-fed state. Alternatively, if the signal is actually a component of an adiposity signaling pathway, then food intake after the infusion of the substance should be suppressed only in the ad libitum-fed condition and less so or not at all in the weight-reduced state. In the case of insulin, the data are quite clear. Insulin infusion in the weight-reduced state was much less effective at suppressing food intake than in the ad libitum-fed condition (11). Even more importantly, the ultimate level of body weight achieved to insulin administration was identical in both the weight-reduced and ad libitum-fed conditions. Such data strongly support the notion that insulin is an adiposity signal that influences food intake only to the extent that is necessary to achieve a new defended level of body adiposity.
Within the CNS, the control of energy balance is regulated by a complex neural network involving a number of systems both within and beyond the hypothalamus. Some of these systems are direct targets for the actions of adiposity signals. In particular, the central melanocortin system appears to be an important mediator of the actions of both leptin and insulin (13). The endogenous melanocortin agonist, -MSH, is produced within neurons of the arcuate nucleus from the precursor peptide proopiomelanocortin. Exogenous administration of melanocortin agonists elicits a strong reduction in food intake and body weight and either pharmacological antagonists or genetic disruption of the melanocortin-4 receptor produces increased food intake and concomitant weight gain (14, 15). These -MSH-producing neurons express both leptin and insulin receptors, and both leptin and insulin stimulate expression of proopiomelanocortin, the precursor to -MSH (16). Moreover, blockade of melanocortin receptors suppresses the ability of both insulin and leptin to suppress food intake and produce weight loss, supporting that both adiposity factors mediate their food intake effects by activating central melanocortin pathways (16, 17).
Because the melanocortin system is critically regulated by the actions of adiposity signals, it is tempting to hypothesize that exogenous administration of melanocortin receptor agonists also alters the defended level of body adiposity, influencing food intake only secondarily to achieve a specific level of body adiposity. To test this hypothesis, we employed the Powley and Keesey strategy (12) of comparing the ability of the stable -MSH analog, MT-II, to suppress food intake and cause weight loss in rats at three different levels of body adiposity. If activity of the melanocortin system is tied to a specific level of energy intake directly, we would expect that previous food restriction should have little impact on the amount of calories consumed during MT-II treatment and this would be accompanied by differing levels of body adiposity. Alternatively, if the activity of the melanocortin system is tied to the defended level of body adiposity, the amount of calories consumed during MT-II treatment should be different between the ad libitum-fed and food-restricted conditions but they should work to achieve identical levels of body adiposity.
Materials and Methods
Study design
Seventy-two male Wistar rats (Charles-River, Portage, MI) with an approximate average starting weight of 275 g were all placed on a calorically dense 45% high-fat (HF) diet (Research Diets, Inc., New Brunswick, NJ; D12541) for 16 wk (see Fig. 1). At the end of this period, rats weighed approximately 675 g. At this point, rats were divided into three weight-matched groups. One group continued to receive ad libitum access to the HF diet (AD LIB group). The other two groups were food restricted either by 30% (RES-30) or 60% (RES-60) of their baseline food intake for a period of 14 d. At this point, rats in each group were randomized to one of three treatment conditions: vehicle (50 mM sodium acetate, pH = 5.0), MT-II (Bachem, Inc., Torrance, CA) at 0.3 mg/kg·d, or MT-II at 3.0 mg/kg·d. MT-II or vehicle was delivered sc via surgically implanted 28-d minipumps (model 2ML4; Alza, Palo Alto, CA). During the next 28 d, all rats had ad libitum access to HF diet, after which they were euthanized by CO2 inhalation followed by exsanguination via a cardiac puncture. The terminal blood sample was collected into heparinized vials and centrifuged to produce plasma that was stored at –80 C for subsequent analysis.
Body composition analysis
Body composition of rats was determined at several points in the course of the study using a QDR-4500A dual x-ray absorptiometry (DXA) scanner (Hologic, Inc., Bedford, MA). Briefly, whole body composition analysis was accomplished in isoflurane-anesthetized rats (2% isoflurane delivered with 100% oxygen) using software version 10.0 and protocols that were provided by the manufacturer. On each scan day, quality control and tissue calibration standards were scanned first according to the manufacturer’s instructions. Rats were scanned in the prone position, using a laser starting point that was positioned at the midline of the rat body just distal to the nose. Each DXA scan was acquired using the rat whole body protocol with a 7.1-in. width and a 14.2-in. length, with a total scan time of about 2.4 min per rat.
Statistical analyses
A one-way ANOVA model was fit to body weight, DXA fat mass, and DXA lean mass data (see Fig. 2). This model included the term for group [vehicle, MT-II (0.3 mg/kg·d), or MT-II (3.0 mg/kg·d)] and was fit by day. Where appropriate, the ANOVA was followed by pair-wise Student’s t tests. Additional body weight and food intake analyses were done in a similar fashion: an appropriate ANOVA model was fit to the endpoint and was followed by pair-wise Student’s t tests to perform the comparisons of interest. In all comparisons, statistical significance was assigned as P < 0.05. A two-way ANOVA model was fit to DXA fat mass and DXA lean mass, including terms for day (0, 14, or 28), group (vehicle, MT-II 0.3 mg/kg·d, or MT-II 3.0 mg/kg·d), and the day*group interaction (see Fig. 4). The ANOVA was followed by pair-wise Student’s t tests for comparisons between days by group, and between groups by day.
Results
All rats gained significant weight over the course of the 16-wk exposure to the HF diet, and all groups had the same amount of total body fat before the food restriction paradigms began (see Fig. 2B; d –14). As expected, food restriction caused significant weight loss in both the RES-30 and RES-60 groups, with the greatest weight loss in the RES-60 group (see Fig. 2A). Also as expected, both the RES-30 and RES-60 groups lost significant amounts of body fat to a degree that was proportional to the level of food restriction when compared with the AD LIB group (see Fig. 2B). Food restriction also resulted in a small, but statistically significant reduction in DXA-derived lean mass that again was proportional to the degree of food restriction (see Fig. 2C).
After rats were implanted with the minipump, the AD LIB rats that were treated with 0.3 mg/kg·d or 3.0 mg/kg·d MT-II showed similar weight loss that was different from that seen with the vehicle-treated rats (see Fig. 3A). Consistent with the weight loss, both doses of MT-II suppressed food intake to a similar degree (see Fig. 3B). The food intake suppression induced by either dose of MT-II was strongest during the first few days of infusion, with a slow return of food intake toward baseline levels (approximately 20 g/d) over the first 2-wk period. It is important to note that during the last 2 wk of the MT-II infusion, food intake had returned to the level of the vehicle-infused group despite the maintenance of the weight loss.
In both the RES-30 and RES-60 groups, vehicle-treated rats were significantly hyperphagic for the first 2 wk of ad libitum feeding after being released from the food constraint on d 0; normal food intake levels resumed during the final 2 wk of vehicle treatment (see Fig. 3, D and F). RES-30 and RES-60 rats treated with vehicle also showed significant weight gain over the 28 d that brought them to a body weight approaching that of the AD LIB vehicle-treated rats by d 28 (see Fig. 3, C and E). RES-30 or RES-60 food-restricted rats treated with MT-II showed a dose-dependent blunting of the hyperphagia compared with the food intake of vehicle-treated animals. In the first 14 d after minipump implantation, the MT-II was more effective at suppressing food intake at the higher dose than at the lower dose, although as in the AD LIB setting, during the last 14 d of treatment there were no significant differences in daily food intake among the MT-II- and vehicle-treated rats (see Fig. 3, D and F). In both RES-30 and RES-60 conditions, both doses of MT-II caused a similar degree of sustained body weight loss over the 28-d infusion period compared with the vehicle treatment (see Fig. 3, C and E).
A key aspect of these studies is the comparison of the effects of MT-II between the different dietary conditions. Although MT-II clearly suppressed food intake at all three levels of food restriction, the actual amount of calories consumed during MT-II treatment differed across the three levels of restriction. In the two-way ANOVA (dose x diet), the cumulative amount consumed either 1, 7, or 14 d after minipump implantation was significantly influenced by both the dose of MT-II (P < 0.0001) and the level of food restriction (P < 0.0001). On the first day, there was also a significant interaction between these two variables (P = 0.028) that was not apparent on cumulative intake after 7 or 14 d. The implication of this analysis is that, to predict the amount of food consumed, we would need to know both the degree of food restriction as well as the dose of MT-II delivered. This analysis also implies that the effect size of the MT-II on food intake is different across the dietary groups only on the first day but not on cumulative intake over 7 and 14 d. There was also a difference in the nature of the dose-effect curve across the three restriction levels. In AD LIB rats, there was no difference between the 0.3 and 3.0 dose of MT-II, although there was a difference in both the RES-30 and RES-60 conditions.
Body weights of MT-II-treated rats were not significantly different regardless of starting nutritive state of rats (AD LIB, RES-30, or RES-60; see Table 1). The effects of 28 d of MT-II treatment on body composition of rats at 14 and 28 d of treatment are shown in Fig. 4. MT-II treatment at either the 0.3-mg/kg·d or 3.0-mg/kg·d dose level caused similar degrees of fat mass loss in AD LIB-, RES-30-, and RES-60-treated rats (see Fig. 4, A, C, and E). Interestingly, the majority of fat loss due to MT-II treatment was observed by d 14 of treatment, with only minimal further fat loss induced by an additional 14 d of treatment. This temporal pattern of fat loss corresponds with the temporal pattern of food intake suppression induced by MT-II in both food-restricted and ad libitum-feeding rats (see Fig. 3). Although food restriction induced a significant reduction of lean mass as mentioned previously, MT-II treatment failed to significantly affect lean mass when compared with vehicle-treated rats in either food-restricted or ad libitum conditions (see Fig. 4, B, D, and F). In fact, lean mass loss due to food restriction in RES-60 rats appeared to recover over the subsequent 28 d of ad libitum feeding, even in MT-II-treated rats (see Fig. 4F).
Among the most important comparisons in the study was the examination of the effects of MT-II treatment between the ad libitum and food-restricted groups. Table 1 displays the terminal body weight and body fats for each of the nine groups. In the vehicle-treated rats, both food-restricted groups showed rapid weight gain once ad libitum access to food was allowed, and by the end of the 28-d period, their body weights and body fats had almost returned to the level of the nonrestricted group. Most interesting is the comparison of the body weight achieved by the MT-II-treated groups. In the rats treated with 0.3 mg/kg·d dose of MT-II, both the food-restricted and ad libitum-fed rats achieved nearly identical body weights and body fats. Thus for this dose of MT-II, the starting point of the animal was not important in determining the final body fat. In the rats treated with the 3.0 mg/kg dose of MT-II, the data are similar. Both the 30 and 60% restriction groups treated with MT-II achieve identical body weights, but both groups trended for a lower body weight than the ad libitum-fed rats, although this trend did not achieve statistical significance. Body fats between all three treatments receiving the high dose of MT-II were clearly not different. Thus, the trend for lower body weight with higher dose of MT-II in food-restricted vs. ad libitum rats was driven not by differences in body fat, but instead was due to an enhanced lean mass effect in the ad libitum rats. This was probably a reflection of the lower lean mass attained during the food restriction period in nutritionally restricted rats, which was maintained during MT-II treatment. The results with body fat are exactly paralleled by those of plasma leptin levels. There was no difference between dietary groups in plasma leptin levels measured at the conclusion of the experiment, but plasma leptin levels are reduced in all of the MT-II-treated groups (see Table 1).
Discussion
These data examine the effect of chronic administration of a melanocortin receptor agonist to influence food intake, body weight, and body composition. The first finding of these studies is that chronic sc infusion of MT-II is quite effective to produce reductions in food intake, body weight, and body fat even in rats made quite obese by prolonged exposure to a HF diet. This result agrees with other work using melanocortin agonists in both mouse and rat and separates that work from what has been generally reported with peripheral administration of leptin that appears to lose most of its efficacy in rats made obese after exposure to a HF diet (18, 19, 20).
The second finding of these studies is that several aspects of the effects of MT-II are altered by prior food restriction. MT-II-treated rats in either of the restricted groups consumed more than MT-II-treated rats that were ad libitum fed. However, this analysis does not take into account the different baseline intakes of the ad libitum and restricted groups. When one compares the effect size of MT-II to its vehicle-treated comparison group, our statistical analysis shows that the effect size is different only on the first day after minipump implantation but not on 7 or 14 d cumulative intake.
Importantly, the ability of MT-II to induce weight loss is much less evident in the restricted rats that weighed less at the beginning of the experiment than the ad libitum-fed rats. Thus, the effect of MT-II to induce body weight loss is highly dependent on the state at which the rats began the experiment. From one perspective this implies that MT-II was less effective in the restricted rats. In the case of the melanocortin system, activity at CNS melanocortin receptors is regulated not just by the presence of endogenous agonists but also by endogenous antagonists/inverse agonists. Agouti-related peptide (AgRP) is produced in arcuate nucleus cell bodies and is carefully regulated by peripheral adiposity signals such as leptin (13, 21, 22). Thus, when leptin levels are reduced, such as in the restricted groups in this experiment, hypothalamic AgRP levels should be elevated. Increased AgRP in the synapse would work to block melanocortin-4 receptors and decrease the ability of exogenous MT-II to inhibit food intake (23). Consequently, we can point to a straightforward hypothesis about why MT-II would be less effective to reduce food intake and induce weight loss in the restricted groups.
The greater number of calories consumed in restricted rats treated with MT-II stands in contrast to the effects on body fat. In this regard, the body fat levels achieved by rats treated with MT-II are similar regardless of whether the animal started the dosing regimen at a lower body fat because of food restriction. That is to say, in the animals whose body fat was reduced by enforced food restriction, the effect of MT-II was to maintain that body fat loss at a level identical to the body fat loss induced by the same dose of MT-II in ad libitum-fed rats with higher body fats.
This finding has two implications. The first is that melanocortin agonists that might be used to treat human obesity should be equally effective when used either to produce weight loss or to maintain weight loss already achieved. Thus, melanocortin agonists could be used to help individuals maintain weight loss that was achieved by other means and improve compliance to dietary regimens designed to maintain lost weight.
The second implication is more theoretical and concerns the parameters that are primarily regulated by the melanocortin system. To predict how much an animal will eat in these experiments you would need to know both the dose of MT-II it was receiving and the degree of prior food restriction. This stands in contrast to the degree of body adiposity. To predict the amount of body fat at the end of the experiment one would need to only know whether the animal received MT-II or vehicle, because prior restriction had no impact on the achieved level of body fat. Our conjecture is that this outcome supports the hypothesis that melanocortin signaling primarily regulates body adiposity. From this perspective, changes in food intake are simply a means to an end, and that end is the maintenance of a level of body fat consistent with the amount of melanocortin signaling. Further support for this hypothesis comes from the fact that, in MT-II-treated rats, food intake had returned to normal by d 14. By d 14, rats had already achieved a level of body fat that would be maintained through d 28. Consequently, the dissipating effect of MT-II on food consumption is not "tachyphalaxis" but rather is the result of animals now consuming the appropriate number of calories to maintain a specific level of adiposity consonant with their degree of melanocortin signaling. As a consequence, it might be more appropriate to term the melanocortin system as an effector pathway for regulating adiposity, rather than as an anorexigenic or satiety pathway per se.
Footnotes
Abbreviations: AgRP, Agouti-related peptide; CNS, central nervous system; DXA, dual x-ray absorptiometry; HF, high fat.
References
Kennedy GC 1953 The role of depot fat in the hypothalamic control of food intake in the rat. Proc R Soc Lond (Biol) 140:579–592
Porte DJ, Baskin DG, Schwartz MW 2002 Leptin and insulin action in the central nervous system. Nutr Rev 60:S20–S29
Friedman JM 2002 The function of leptin in nutrition, weight, and physiology. Nutr Rev 60:S1–S14;discussion S68–S84, S85–S87
Saper CB, Chou TC, Elmquist JK 2002 The need to feed: homeostatic and hedonic control of eating. Neuron 36:199–211
Ahima RS, Saper CB, Flier JS, Elmquist JK 2000 Leptin regulation of neuroendocrine systems. Leptin regulation of neuroendocrine systems. 21:263–307
Seeley RJ, Woods SC 2003 Monitoring of stored and available fuel by the CNS: implications for obesity. A cortical-hippocampal system for declarative memory. 4:901–909
Woods SC, Seeley RJ, Baskin DG, Schwartz MW 2003 Insulin and the blood-brain barrier. Insulin and the blood-brain barrier 9:795–800
McGowan MK, Andrews KM, Grossman SP 1992 Chronic intrahyphothalamic infusions of insulin or insulin antibodies alter body weight and food intake in the rat. Physol Behav 51:753–766
Brüning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, Klein R, Krone W, Müller-Wieland D, Kahn CR 2000 Role of brain insulin receptor in control of body weight and reproduction. Science 289:2122–2125
Seeley RJ, van Dijk G, Campfield LA, Smith FJ, Nelligan JA, Bell SM, Baskin DG, Woods SC, Schwartz MW 1996 The effect of intraventricular administration of leptin on food intake and body weight in the rat. Horm Metab Res 28:664–668
Chavez M, Kaiyala K, Madden LJ, Schwartz MW, Woods SC 1995 Intraventricular insulin and the level of maintained body weight in rats. Behav Neurosci 109:528–531
Powley TL, Keesey RE 1970 Relationship of body weight to the lateral hypothalamic feeding syndrome. Absence of satiety during sham feeding in the rat. 70:25–36
Seeley RJ, Drazen DL, Clegg DJ 2004 The critical role of the melanocortin system in the control of energy balance. Annu Rev Nutr 24:133–149
Cone RD, Cowley MA, Butler AA, Fan W, Marks DL, Low MJ 2001 The arcuate nucleus as a conduit for diverse signals relevant to energy homeostasis. Int J Obes Relat Metab Disord 25(Suppl 5):S63–S67
Cone RD 1999 The central melanocortin system and energy homeostasis. Melanin-concentrating hormone receptor: an orphan receptor fits the key. Trends Endocrinol Metab 10:211–216
Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, Woods SC 2002 The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 22:9048–9052
Seeley RJ, Yagaloff KA, Fisher SL, Burn P, Thiele TE, van Dijk G, Baskin DG, Schwartz MW 1997 Melanocortin receptors in leptin effects. Nature 390:349
Pierroz DD, Ziotopoulou M, Ungsunan L, Moschos S, Flier JS, Mantzoros CS 2002 Effects of acute and chronic administration of the melanocortin agonist MTII in mice with diet-induced obesity. Diabetes 51:1337–1345
Clegg DJ, Benoit SC, Air EL, Jackman A, Tso P, D’Alessio D, Woods SC, Seeley RJ 2003 Increased dietary fat attenuates the anorexic effects of intracerebroventricular injections of MTII. Endocrinology 144:2941–2946
Li G, Zhang Y, Wilsey JT, Scarpace PJ 2004 Unabated anorexic and enhanced thermogenic responses to melanotan II in diet-induced obese rats despite reduced melanocortin 3 and 4 receptor expression. J Endocrinol 182:123–132
Wilson BD, Ollmann MM, Barsh GS 1999 The role of agouti-related protein in regulating body weight. Mol Med Today 5:250–256
Bagnol D, Lu XY, Kaelin CB, Day HE, Ollmann M, Gantz I, Akil H, Barsh GS, Watson SJ 1999 Anatomy of an endogenous antagonist: relationship between Agouti-related protein and proopiomelanocortin in brain. J Neurosci 19:RC26
Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg HT, Woods SC, Seeley RJ 2000 Long-term orexigenic effects of AgRP-(83–132) involve mechanisms other than melanocortin receptor blockade. Am J Physiol 279:R47–R52(Randy J. Seeley, Melissa )