High-fat diet-induced obesity (DIO) in rodents is associated with hyperleptinemia and resistance to leptin, but the response to agents acting downstream of leptin receptors remains unknown. We assessed the response of mice with DIO to treatment with MTII, an α-melanocyte-stimulating hormone analog. MTII delivered four times daily by intraperitoneal injection to C57BL/6J mice produced a dose-responsive effect on food intake, body weight, leptin, corticosterone, insulin, and free fatty acids. In DIO mice, administration of MTII 100 μg q.i.d. i.p. markedly suppressed feeding during the first 4 days of treatment, with food intake returning to control levels at day 5. Progressive weight loss also occurred over the first 4 days, after which weight plateaued at a level below control. After 8 days of treatment, MTII-treated DIO mice had major suppression of both leptin and insulin levels. Central administration of MTII for 4 days (10 nmol/day) in DIO mice significantly suppressed food intake, induced weight loss, and increased energy expenditure. These results indicate that 1) MTII administration to DIO mice causes suppression of food intake and body weight loss, and decreased food intake is primarily responsible for weight loss; 2) peripheral MTII improves insulin resistance in DIO mice; 3) “tachyphylaxis” to the effect of chronic MTII treatment on food intake occurs; and 4) at least some of the effects of MTII are exerted centrally. In conclusion, treatment with a melanocortin agonist is a promising therapeutic approach to DIO and associated insulin resistance.
Obesity is a complex disorder caused by the interaction of genes and the environment (1,2). The role of genes is best revealed through the identification of monogenic syndromes of obesity in mice and humans that involve defects in the adipocyte hormone leptin and components of the central neural circuits that are regulated by this hormone (1–3). Although genes play a major role in the determination of body weight, most obesity in rodents and humans is strongly influenced by diet availability and composition and is associated with hyperleptinemia and leptin resistance (1,2). The best-studied experimental model for this syndrome is C57BL/6J mice, which are susceptible to obesity when placed on a high-fat diet, a state referred to as diet-induced obesity (DIO) (4–7). Two defects that may cause leptin resistance have been described in these mice (6). The first involves a defect in leptin transport across the blood-brain barrier. This has been demonstrated by direct studies of reduced leptin transport in DIO mice (8 and our unpublished observations), as well as by evidence that food intake is more effectively suppressed when leptin is administered directly into the central nervous system (CNS) compared with the peripheral route (5). A second defect causing leptin resistance in DIO mice involves reduced leptin signaling to STAT3 in leptin-responsive neurons in the hypothalamus after peripheral or central administration (6). Although these defects in proximal leptin access and signaling are known, there is little or no information on the status of pathways downstream of leptin and its direct targets in DIO. Based on these observations, such information is critical for an understanding of the status of central pathways in DIO mice, one of the animal models that resembles human obesity. By studying the response of DIO, these experiments could also provide information on the response of common human obesity to chronic therapy with drugs aimed at this pathway.
The melanocortin (MC) pathway is important for the regulation of energy homeostasis, as revealed by genetic and pharmacological evidence (9–16), and MCs are an important target of leptin action within the brain (17,18). Leptin stimulates expression of pro-opiomelanocortin (POMC) in a population of arcuate neurons that process the peptide precursor to α-melanocyte-stimulating hormone (α-MSH), which then acts on central MC receptors MC3 and MC4R to inhibit food intake and increase energy expenditure (9). Leptin also inhibits expression of the endogenous MC antagonist agouti-related protein (AgRP) (13). Mutation in genes encoding POMC (14,15), the POMC-processing enzymes pc-1 (19,20) and carboxypeptidase E (20), and the MC3 (11,12,21) or MC4 (10,22,23) receptors cause obesity, as do overexpression of the MC antagonists agouti or AgRP (24–26). In addition, synthetic MC agonists acutely inhibit food intake in normal and ob/ob mice, and MC4 receptor antagonists increase food intake and limit the response to leptin (9,27,28). Based on these observations, a synthetic MC agonist would be a valuable reagent for examining the status of central pathways in DIO and predicting the response of DIO and perhaps common human obesity to chronic therapy with drugs aimed at this pathway.
In this study, we used the synthetic cyclic heptapeptide MTII (29), a nonspecific MC receptor agonist, to determine the response of both normal mice and mice with acquired obesity to peripheral and central pharmacological activation of this pathway.
RESEARCH DESIGN AND METHODS
Animals and diets.
C57BL/6J (3 and 8 weeks old) were obtained from The Jackson Laboratory (Bar Harbor, ME). All animals were individually housed and maintained at 25°C with a 12-h light/dark cycle. For the peripheral experiments, all animals were given unrestricted access to food (6.5% fat, 49% carbohydrate, and 23% protein wt/wt; 3.5 kcal/g Purina Rodent Chow 5008; Ralston-Purina, St. Louis, MO) and water, unless noted otherwise. A group of C57BL/6J mice was maintained on high-fat diet (44.9% fat; D12451; Research Diets, New Brunswick, NJ) for 10 months, starting after weaning, and was then used for the longitudinal experiments. For the central experiments, C57BL/6J (3 weeks old) were fed either chow diet (6.5% fat; Purina Rodent Chow 5008; Ralston-Purina) or a high-fat diet (42% fat; TD 88137; Harlan Teklad, Madison, WI) for 18 weeks. All animals were handled in accordance with the principles and guidelines of the National Institutes of Health, and the Institutional Animal Care and Use Committee of Beth Israel Deaconess Medical Center approved the protocol.
Experimental procedures
Short-term studies
Male C57BL/6J mice: 1-day experiment.
We studied the effect of different doses of MTII (5, 25, and 100 μg), administered intraperitoneally four times a day to C57BL/6J mice (n = 8 per group), on body weight, food intake, and circulating hormone concentrations. A PBS-treated group that was pair-fed with the highest dose of MTII (100 μg q.i.d.) was included.
2-Day experiment.
To assess whether prolonged MTII administration could have an effect on body weight, we administered either PBS or MTII (100 μg q.i.d.) for 48 h, following the same protocol as above.
Long-term studies
Male DIO-C57BL/6J mice.
To assess whether DIO mice that do not respond to peripheral leptin administration (6) would respond to prolonged MTII administration, we injected MTII peripherally and centrally. To study the peripheral effect of MTII, DIO mice were injected intraperitoneally for 8 days. In addition to PBS and MTII (100 μg q.i.d.) groups, another group, injected with PBS, was pair-fed with the MTII group. After the 8th day, injections were discontinued until mice reached their initial body weight (day 20). At that time, the PBS-treated group started receiving intraperitoneal PBS injections four times a day for 2 days, the MTII group received MTII (100 μg q.i.d.) for 1 day and leptin (1 μg/g body wt b.i.d.) the following day, whereas the previous pair-fed group was treated with leptin for 1 day and MTII the following day. To determine the central effects of MTII, low-fat- and high-fat-fed mice received either PBS or MTII (10 nmol/day) in the right lateral ventricle via minipumps (n = 8/group). An additional group pair-fed with the group MTII (10 nmol/day) (n = 8) was administered with PBS.
Intracerebroventricular administration of MTII.
Under ketamine (45 mg/kg; Abbott Laboratories, Chicago, IL) and xylazine (5 mg/kg; RBI, Natick, MA) anesthesia, a sterile cannula (Brain Infusion Kit II; Alzet, Mountain View, CA) connected to an osmotic minipump (model 1007D; Alzet) was stereotaxically implanted in the lateral ventricle (coordinates: 0.3 mm caudal, 1 mm lateral from bregma, and depth 2.3 mm, in accordance with KBJ Franklin and G Paxinos, Eds: The Mouse Brain in Stereotaxic Coordinates, Academic Press, San Diego, CA, 1997). The intracerebroventricular (ICV) cannula was then fixed in place using dental cement (Plastics One, Roanoke, VA). Pumps connected to the ICV cannulae were started overnight at 37 C in 0.9% saline and delivered PBS or MTII (10 nmol/day). At the time of surgery, mice were injected with the analgesic Buprenex (0.1 mg/kg; Reckritt and Colman, Richmond, VA).
Hormones and hormone assays.
Recombinant mouse leptin (Eli Lilly, Indianapolis, IN) and MTII (Bachem Bioscience, King of Prussia, PA) were dissolved in sterile PBS (pH 7.4). Blood glucose was assayed with a blood glucose meter (One Touch Profile; Lifescan, Milpitas, CA). For the ICV experiment, leptin and insulin were assayed by enzyme-linked immunosorbent assay (Crystal Chem, Chicago, IL). For other experiments, leptin, insulin, and corticosterone were assayed by radioimmunoassay using mouse leptin and rat insulin (Linco Research Institute, St. Louis, MO) and rat corticosterone (ICN, Costa Mesa, CA). Free fatty acids were assayed by an enzymatic colorimetric method (Waco Chemicals, Richmond, VA).
Oxygen consumption.
In vivo indirect open circuit calorimetry (Colombus Instruments, Columbus, OH) was performed in metabolic chambers, where mice were put for 4 h with free access to food and water at 25°C. Oxygen and carbon dioxide gas fractions were monitored at both the inlet and output ports, through which air flow of 0.56 l/min was passing, and were used to calculate oxygen consumption (Vo2), carbon dioxide production (Vco2), and respiratory quotient (RQ; ratio of Vco2 to Vo2), using Oxymax software (Colombus Instruments).
Dual-energy X-ray absorptiometry.
Dual-energy X-ray absorptiometry (DEXA) was used to determine in vivo body composition of mice treated centrally with MTII. After 4 days of MTII administration, mice were anesthetized with ketamine/xylazine at time of killing and scanned using a Lunar PIXImus densitometer (software version 1.45; Madison, WI).
Calculations and statistical analysis.
All results are presented as the means ± SE. Statistical significance was assessed by unpaired two-tailed Student’s t test, and ANOVA with Fisher’s post hoc test using Statview software (Abacus, CA). Differences were considered significant at the two-tailed P < 0.05 value.
RESULTS
Short-term experiments
Effects of MTII administration on food intake and body weight.
We studied the effect of increasing doses of MTII to induce weight loss and decrease food intake in normal C57BL/6J mice. Intraperitoneal administration of MTII for 1 or 2 days resulted in weight loss that was significantly greater than that induced by PBS (Table 1). A dose of MTII (100 μg q.i.d.) for 1 day resulted in ∼1.4 g more weight loss than the PBS-treated group, and it was similar to the weight loss of the pair-fed mice, indicating that the weight-reducing effect of MTII in normal mice is mainly due to decreased food consumption (Table 1).
Effects of MTII administration on hormonal parameters.
MTII administered four times daily resulted in a significant and dose-dependent decrease of serum insulin levels in normal C57BL/6J mice, despite unchanged serum glucose levels (Table 2). Pair-feeding to the highest dose administered resulted in a similar reduction of insulin levels, indicating that the effect of MTII administration in normal C57BL/6J mice is mainly a consequence of hypophagia. MTII administration also caused a dose-dependent decrease of serum leptin levels (Table 2). Administration of 100 μg of MTII to normal C57BL/6J mice for 24 and 48 h resulted in leptin levels being ∼30 and 50% of baseline, respectively (Table 2). However, the decrease in leptin levels was similar in magnitude to that induced by a decrease in food intake in the pair-fed group. Administration of increasing doses of MTII for 24 h resulted in increasing FFA levels that were significantly higher than PBS and pair-fed values (Table 2). Finally, MTII administration increased corticosterone levels, but the increase was not larger in magnitude than that induced by a similar decrease in food intake in the pair-fed group. Moreover, corticosterone levels did not appear to remain increased during the second day of MTII administration, when food intake returned toward baseline and was not significantly higher than in the PBS-treated group (Table 2).
Long-term effects of MTII treatment in DIO
Effects of prolonged peripheral MTII administration on body weight and food intake of leptin-resistant DIO mice.
Chronic administration of MTII (100 μg q.i.d.) resulted in weight loss significantly greater than that induced by PBS injections. Weight loss occurred during the first 4 days of MTII administration (Table 1) (Fig. 1). Use of a pair-fed group with the MTII group revealed that the weight loss induced by MTII administration was largely caused by decreased food intake (Table 1). However, the MTII-induced weight loss of DIO mice reached a plateau at approximately day 5. The treatment continued until day 8, at which point the MTII group had lost 3.20 g more than the PBS group (Table 1) (Fig. 1). Cumulative food intake of the MTII group was significantly lower than that of the PBS group for the first 8 days of treatment (Fig. 2B). Interestingly, daily food intake of the MTII-treated mice was suppressed during the first 4 days of treatment and then increased to approximately control levels between days 5 and 8 of treatment, suggesting tachyphylaxis to this effect of MTII (Fig. 2A). However, discontinuation of the treatment resulted in a progressive return of body weight toward baseline (Fig. 1). During the posttreatment period, food intake of the MTII group increased above that of the PBS groups for several days (Fig. 2A). These data indicate that MTII treatment is effective in this leptin-resistant animal model of obesity, but that some degree of tachyphylaxis to the effect of this medication develops after the 4th day of treatment. The compensatory feeding and weight response after cessation of MTII indicates that the drug is continuing to exert important suppressive effects on appetite during this period of partial tachyphylaxis. To examine the role of leptin expression in the waning effect of MTII at days 5–8 and the compensatory increase of food intake after day 8, we measured serum leptin levels of these mice at the time the treatment was discontinued and found that leptin levels were lower (see below). To prove that these mice were indeed leptin resistant, we treated them with exogenous leptin. After all mice reached their baseline weights, we treated the pair-fed group with leptin (intraperitoneally twice daily) for 1 day, followed by MTII treatment (100 μg i.p. q.i.d.) for another day. At the same time, the previously MTII-treated group was treated initially with MTII for 1 day and with leptin on the following day. During this 48-h period, the previously PBS-treated group was treated with PBS injections four times daily. As shown in Fig. 1, although MTII treatment resulted in significant weight loss, leptin treatment had absolutely no effect in these leptin-resistant mice.
Effect of peripheral MTII administration on leptin, insulin, and glucose levels of leptin-resistant DIO mice.
MTII administration in these mice resulted in a significant decrease of leptin levels to ∼50% of baseline (Table 2). Similarly, their glucose and insulin levels decreased significantly in response to MTII treatment. Importantly, both insulin and leptin levels of MTII-treated mice decreased to levels significantly lower than those of the mice pair-fed with the MTII-treated mice (Table 2). Finally, serum free fatty acid levels were similar in PBS- and MTII-treated and pair-fed mice after 8 days of treatment.
Effects of chronic central MTII.
C57BL/6J mice, fed either a low-fat or a high-fat diet for 17–18 weeks, received a central infusion of MTII (10 nmol/day) for 4 days. In mice fed a low-fat diet, central MTII administration induced a significantly greater body weight loss compared with PBS infusion. MTII-treated mice lost weight during the first 2 days of treatment compared with the PBS-treated mice. Their cumulative weight loss remained significantly greater through the 4 days of treatment (P = 0.012). MTII-treated mice also lost more weight than the pair-fed mice during the first 3 days of treatment (P < 0.05) (Fig. 3). The weight loss induced by MTII was largely due to a decrease in food intake. Cumulative food intake of the MTII group was significantly decreased compared with the PBS group during the first 2 days of administration (P < 0.05) (Fig. 3). MTII-treated mice ate 40% of the amount the PBS-treated group ate during the first day and ate less than the PBS-treated group for the first 2 days, after which their food intake increased to control levels, suggesting again that tachyphylaxis to this effect of MTII occurred. In comparison, high fat-fed mice treated with the same dose of MTII showed a more prolonged response than low fat-fed mice. MTII-treated DIO mice lost more weight than the PBS-treated mice during the first 3 days of treatment. Their cumulative body weight loss was significantly greater during the first 3 days compared with the PBS group (P < 0.05). MTII-treated DIO mice also lost more weight than the pair-fed mice (P < 0.05), suggesting that the weight loss in DIO mice is not due solely to a decreased food intake (Fig. 3). The effects of MTII on reducing food consumption lasted for 4 days. MTII-treated mice ate 50% of the amount the PBS group ate the first day and 70% the fourth day, and subsequently their cumulative food intake was decreased during the entire period (P < 0.05) (Fig. 3).
Effect of central MTII administration on leptin, insulin, and glucose levels of leptin-resistant DIO mice.
Adult mice fed a high-fat diet for 17–18 weeks are obese, normoglycemic, hyperinsulinemic, and hyperleptinemic compared with low fat-fed mice (Table 3). Central and chronic administration of MTII resulted in a significant decrease of leptin levels (P < 0.05) in mice fed a low-fat diet, whereas the leptin levels remained unchanged in the pair-fed group. Glucose and insulin levels in the fed state were not affected by MTII treatment. In mice fed a high-fat diet, central MTII administration induced a nonsignificant decrease of leptin and insulin levels (Table 3).
Effect of central MTII administration on energy expenditure and lean tissue weight of leptin-resistant DIO mice.
To determine whether body weight loss in DIO mice with ICV MTII treatment was caused solely by reduced food intake or involved increased energy expenditure as well, this parameter was determined in fed mice by recording Vo2 and Vco2 measurements for 4 h during the 3rd day of MTII administration. In mice fed a low-fat diet, energy expenditure (O2 consumption) was significantly increased with MTII administration compared with pair-fed mice (P < 0.01), but the RQ (calculated as Vco2/Vo2) remained unchanged, indicating that MTII-treated mice dissipated more energy, but both groups use the same ratio of carbohydrate to fat as fuel (Fig. 4). Energy expenditure after ICV MTII administration was increased by 33% in DIO mice (P < 0.001) compared with the pair-fed mice and by 8% (P < 0.05) compared with the PBS-treated group. MTII administration in mice fed the low-fat diet significantly decreased the weight of the epididymal fat pad (P < 0.05) compared with PBS-treated mice. In diet-induced obese mice, the weight of this tissue was not significantly decreased (Table 3). Total fat mass and lean mass were measured immediately before sacrificing by DEXA scan. MTII significantly decreased the weight of lean mass compared with PBS treatment in both low fat- and high fat-fed mice (Table 3). Although total fat mass measured between groups in either low fat- or high fat-fed mice tended to be lower in the MTII-treated groups, the observed difference did not achieve statistical significance at the conventional P = 0.05 level (Table 3). These data indicate that the loss of body weight induced by MTII is due to a decrease of both lean and fat mass, as previously described (30).
DISCUSSION
In this study, we have examined the efficacy of the MC agonist MTII as a novel therapeutic agent for obesity and associated insulin resistance in DIO mice.
MTII has robust effects on food intake and body weight in DIO mice.
DIO MTII mice treated peripherally with MTII (100 μg i.p. q.i.d.) displayed a substantial suppression of cumulative food intake (62%) and body weight (13%) during the first 4 days of treatment. In parallel, mice treated centrally by MTII showed a pronounced reduction of cumulative food intake and body weight. This represents the first evidence that an exogenous MC agonist may be effective at inducing weight loss in leptin-resistant mice with DIO.
Future experiments using groups of chow-fed mice of the same strain with MTII minipumped at the same doses used in the DIO mice to measure food intake and body weight would allow a comparison of efficacy to the DIO group and answer the question of whether this DIO model is associated with any MC “resistance.” Given that such resistance is a biological possibility (e.g., the agouti mouse), it seems important to address. This issue could also address how effective MC agonists will be in the human population as therapy and would allow for answers to mechanistic questions about whether this DIO model is associated with decreased or increased sensitivity to MC agonists.
Central administration of MTII in chow-fed normal and obese rodents has previously been described to dose-dependently attenuate food intake and decrease body weight. In all (9,31–33) but one study in POMC-deficient mice (15), feeding returned to control levels 8 h after a single administration (9,32), and cumulative feeding over 24 h was equal to control (32), suggesting the need for a suitably frequent rate of administration. Chronic peripheral administration of a MC agonist to obese POMC-deficient mice produced substantial weight loss (15), although once-daily administration to normal mice was without effect in that study (15) and accelerates weight loss during a fast in ob/ob mice (34). On the 3rd day of MTII treatment, DIO mice exhibited a higher energy expenditure as compared with pair-fed mice (33%), indicating that the MC system acts on body homeostasis not only by reducing food intake but also by increasing energy expenditure. Prior studies with various MC agonists and antagonists, largely performed in normal rodents, also revealed that the effects of the MC system on energy balance involves effects to both decrease food intake and increase energy expenditure through effects on autonomic output (35). Body composition analysis showed that MTII decreased both lean and fat mass and MTII-induced weight loss is not comprised preferentially of adipose tissue, as has been shown for leptin.
The hypophagic response to peripheral MTII wanes after 5 days, but a persistent suppression of appetite is made evident by the hyperphagic response to discontinuation of the drug.
In the period between 4 and 5 days after initiation of therapy, the continuously MTII-treated mice reverse the status of food consumption, going from a 62% reduction of food consumption to a level of food intake approximately equal to control. This change suggests that some degree of tachyphylaxis to this action of the drug has supervened. The increase in food consumption at day 5 of treatment is accompanied by a plateauing of the weight curve to a level parallel to, but below, that of control mice. We interpreted the return of food intake after 4 days of treatment to, but not above, control levels as implying that MTII had reduced but not totally lost its effects on food intake at this point. During the 10 days after cessation of MTII administration, compensatory hyperphagia was evident, and this was accompanied by a regaining of all of the lost weight. Thus, MC receptor stimulation by prolonged MTII administration in DIO mice induces major suppression of food intake and weight loss during the first 4 days of therapy, followed by partial loss of efficacy. We observed a waning effect on food intake in mice treated centrally with MTII as well, and this appears earlier in mice treated centrally than in mice treated peripherally. On day 3 of ICV infusion of MTII, DIO mice ate the same amount as the control mice. A waning in the anorectic effect of centrally administered α-MSH in normal male rats over a period of several days has very recently been described (30), but this phenomenon has not previously been examined in DIO. The nature of the partial waning of MTII action is a matter of considerable importance. It could result from MC receptor desensitization or from the superimposition of one or more adaptive responses to the initial treatment. It is also possible that increasing MC tone by long-term administration of MTII decreases the defended level of body weight rather than exerting a direct effect on food intake. In response to this decreased target weight, the mice could respond by eating less until that weight is achieved, and then they eat the appropriate amount of food to maintain the new level of body weight. The possible nature of such adaptations will be discussed further below and will be the focus of future research.
MTII administered peripherally has specific effects to suppress hyperinsulinemia and hyperleptinemia in DIO mice.
After 8 days of peripheral MTII treatment, when food intake was no longer suppressed below control levels (see below), DIO mice treated with MTII displayed a marked fall in insulin (6.13–0.98 ng/ml) and leptin levels (51.32–29.76 ng/ml). Importantly, both of these effects were greater than that produced by pair feeding. Thus, MC receptor activation for 8 days with MTII has a direct effect, independent of hypophagia or weight loss, to suppress leptin and insulin and, presumably, reduce insulin resistance. This provides further evidence that despite some loss of effect on appetite after 4 days of treatment, a major effect of MTII on metabolism persists through the full 8 days. Recent studies in normal, ob/ob, and MC4R-deficient mice suggest that central MC pathways exert direct effects on insulin secretion and sensitivity (36). In DIO mice treated for 3 days with ICV MTII, the small decrease of plasma leptin (1.32–1.18 ng/ml) and insulin levels (1.65–1.15 ng/ml) was not statistically significant. This could be due to either the fact that the duration of ICV treatment was shorter than the intraperitoneally administered MTII or the fact that the DIO mice, in whom the central effect of MTII was tested, were much less hyperleptinemic (13.24 vs. 51.32 ng/ml) and hyperinsulinemic (16.5 vs. 6.13 ng/ml) than the mice receiving peripheral administration of MTII.
The molecular basis for insulin resistance in obesity is a matter of major interest (37). A key molecule regulating insulin resistance in obesity is tumor necrosis factor-α (TNF-α), which acts by altering tyrosine phosphorylation of insulin receptor substrate-1 (IRS-1) (37,38). Because it has been shown that α-MSH alters TNF-α expression and action in macrophages mainly by MC1, but possibly also by MC3 and MC5 receptors (39), it is possible that MC peptides may improve insulin resistance by inhibiting TNF-α production in adipose tissue (39,40). Whether this mechanism contributes to the observed effect of MTII remains to be investigated. However, recent evidence suggests that leptin’s effect to reduce insulin resistance may be exerted, at least in part, through actions in the CNS (36). Our studies are the first to examine melanocortinergic therapy in mice with DIO and suggest that pharmacological efforts to increase MC tone through MC4 (and possibly MC3) receptors, most likely centrally, may be a useful strategy for reducing insulin resistance in obesity.
As leptin levels fall during MTII treatment, it is possible that AgRP expression would rise, producing an endogenous antagonist of the administered drug. These findings suggest that the responsiveness of human obesity to melanocortinergic agonists will depend, at least in part, on the state of POMC and AgRP expression in the hypothalamus of such patients, which is currently unknown.
Central versus peripheral sites of MTII action.
Most studies of MC agonists and antagonists as regulators of energy balance have used central administration paradigms. Such studies have demonstrated that, consistent with relatively selective expression of MC3 and MC4 receptors within the CNS (41), MC agonists have a clear central site of action, although some peripheral sites of action cannot be ruled out through such an approach. In any event, the development of a useful drug for human obesity would involve, almost certainly, a peripheral route of administration, which would activate peripheral receptors if they existed. For this reason, we chose to compare the peripheral versus the central route of administration in these studies. Prior work has revealed that peripherally administered α-MSH can activate central processes related to the regulation of fever in rodents (42). In addition, peripheral administration of an MC4 receptor agonist produced substantial weight loss in POMC-deficient mice (15). Thus, there is precedent for peripheral administration of MC agonists to produce hormonal and weight-reducing effects, but it is unclear whether body weight was influenced entirely by reduced food intake or increased energy expenditure, or in combination with a direct effect on adiposity (e.g., inhibition of free fatty acid uptake or stimulation of lipolysis) (15,43). MCs circulate in the serum, and their receptors are found in adipocytes (44), where they inhibit free fatty acid uptake and promote lipolysis (43). In contrast, the agouti protein, which competes for binding sites with α-MSH in the periphery, stimulates adipogenesis (45) and antagonizes MC-mediated lipolysis directly in adipocytes. In this study, we observed a significant effect of peripheral MTII to increase lipolysis during the first day of treatment of normal mice.
MTII is quite potent at peripherally expressed MC1 and MC5 receptors, in addition to activating the mainly centrally expressed MC3 and MC4 receptors (46). Recent studies have shown that a small fraction of peripherally administered MTII crosses the blood-brain barrier and affects CNS functioning (47,48). These studies are relevant to the use of a peripherally administered MC agonist in the treatment of obesity.
In conclusion, these studies provide strong evidence in support of the potential utility of MC agonist therapy in the treatment of obesity and its complications, insulin resistance and diabetes. Although many questions remain to be answered, MC receptor agonists are a promising new therapy for obesity and insulin resistance.
. | Groups . | Initial weight (g) . | Final weight (g) . | Weight loss (g) . | Cumulative food intake (g) . |
---|---|---|---|---|---|
Short-term study | |||||
C57BL/6J, 1 day | PBS | 24.08 ± 0.63 | 22.69 ± 0.55 | 1.39 ± 0.14* | 5.16 ± 0.36 |
5 μg q.i.d. MTII | 24.03 ± 0.62 | 22.25 ± 0.61 | 1.78 ± 0.13* | 3.94 ± 0.47 | |
25 μg q.i.d. MTII | 24.14 ± 0.59 | 21.78 ± 0.56 | 2.36 ± 0.08† | 3.52 ± 0.62‡ | |
100 μg q.i.d. MTII | 24.09 ± 0.48 | 21.34 ± 0.28 | 2.76 ± 0.21† | 2.46 ± 0.34† | |
Pair-fed | 24.87 ± 0.69 | 22.37 ± 0.62 | 2.50 ± 0.12† | 2.46 ± 0.01† | |
C57BL/6J, 2 days, | PBS | 21.95 ± 1.78 | 21.42 ± 1.61 | 0.44 ± 0.14 | 7.90 ± 0.95 |
100 μg q.i.d. MTII | 22.54 ± 0.65 | 20.86 ± 0.51 | 1.63 ± 0.23§ | 7.58 ± 0.34 | |
Longitudinal study | |||||
DIO | PBS | 47.84 ± 2.29 | 44.40 ± 2.96 | 3.44 ± 1.03 | See Fig. 2 |
100 μg q.i.d. MTII | 47.95 ± 2.43 | 40.30 ± 2.12 | 7.65 ± 0.58† | See Fig. 2 | |
Pair-fed | 45.02 ± 0.96 | 40.78 ± 0.55 | 4.24 ± 0.56 | See Fig. 2 |
. | Groups . | Initial weight (g) . | Final weight (g) . | Weight loss (g) . | Cumulative food intake (g) . |
---|---|---|---|---|---|
Short-term study | |||||
C57BL/6J, 1 day | PBS | 24.08 ± 0.63 | 22.69 ± 0.55 | 1.39 ± 0.14* | 5.16 ± 0.36 |
5 μg q.i.d. MTII | 24.03 ± 0.62 | 22.25 ± 0.61 | 1.78 ± 0.13* | 3.94 ± 0.47 | |
25 μg q.i.d. MTII | 24.14 ± 0.59 | 21.78 ± 0.56 | 2.36 ± 0.08† | 3.52 ± 0.62‡ | |
100 μg q.i.d. MTII | 24.09 ± 0.48 | 21.34 ± 0.28 | 2.76 ± 0.21† | 2.46 ± 0.34† | |
Pair-fed | 24.87 ± 0.69 | 22.37 ± 0.62 | 2.50 ± 0.12† | 2.46 ± 0.01† | |
C57BL/6J, 2 days, | PBS | 21.95 ± 1.78 | 21.42 ± 1.61 | 0.44 ± 0.14 | 7.90 ± 0.95 |
100 μg q.i.d. MTII | 22.54 ± 0.65 | 20.86 ± 0.51 | 1.63 ± 0.23§ | 7.58 ± 0.34 | |
Longitudinal study | |||||
DIO | PBS | 47.84 ± 2.29 | 44.40 ± 2.96 | 3.44 ± 1.03 | See Fig. 2 |
100 μg q.i.d. MTII | 47.95 ± 2.43 | 40.30 ± 2.12 | 7.65 ± 0.58† | See Fig. 2 | |
Pair-fed | 45.02 ± 0.96 | 40.78 ± 0.55 | 4.24 ± 0.56 | See Fig. 2 |
Data are means ± SE. Comparison of the effect of different doses of MTII administration on body weight and food intake in male C57BL/6J (n = 8 per group for the 1-day experiment and n = 4 per group for the 2-day experiment) and DIO (n = 4–5 per group) mice treated for 8 days.
P ≤ 0.001 vs. the pair-fed group by ANOVA with post hoc tests;
P < 0.001,
P < 0.05,
P ≤ 0.01 vs. the PBS-treated group by ANOVA with post hoc tests (or unpaired t test for experiments including two groups).
. | Groups . | Leptin (ng/ml) . | Corticosterone (ng/ml) . | Insulin (ng/ml) . | Glucose (mg/dl) . | Free fatty acids (mEq/l) . |
---|---|---|---|---|---|---|
Short-term study | ||||||
C57BL/6J, 1 days | PBS | 2.93 ± 0.74 | 183.88 ± 18.92 | 3.66 ± 0.92 | 169.5 ± 8.2 | 0.09 ± 0.06 |
5 μg q.i.d. MTII | 2.45 ± 0.71* | 321.13 ± 44.21† | 3.06 ± 0.89 | 173.1 ± 5.5 | 0.29 ± 0.11 | |
25 μg q.i.d. MTII | 1.64 ± 0.45 | 337.50 ± 28.58† | 2.04 ± 0.53 | 174.0 ± 5.7 | 0.36 ± 0.09*‡ | |
100 μg q.i.d. MTII | 0.83 ± 0.47†§ | 391.25 ± 39.54‖ | 0.47 ± 0.09‖¶ | 170.0 ± 10.1 | 0.52 ± 0.09†‡ | |
Pair-fed | 0.67 ± 0.21†§ | 365.50 ± 22.39‖ | 0.84 ± 0.26†§ | 183.2 ± 8.7 | 0.12 ± 0.05 | |
C57BL/6J, 2 days | PBS | 4.03 ± 0.72 | 193.00 ± 20.75 | ND | 191.6 ± 9.6 | 1.97 ± 0.14 |
100 μg q.i.d. MTII | 2.37 ± 0.15‡ | 200.25 ± 13.27 | 0.49 ± .09 | 184.3 ± 7.8 | 2.15 ± 0.71 | |
Longitudinal study | ||||||
DIO | PBS | 51.32 ± 11.67 | ND | 6.13 ± 2.02 | 134.5 ± 6.4 | 0.77 ± 0.05 |
100 μg q.i.d. MTII | 29.76 ± 3.02‡ | ND | 0.9 ± 0.33*‡ | 113.8 ± 4.2‡ | 0.73 ± 0.08 | |
Pair-fed | 47.20 ± 2.40 | ND | 4.01 ± 0.77 | 97.75 ± 7.4† | 0.62 ± 0.14 |
. | Groups . | Leptin (ng/ml) . | Corticosterone (ng/ml) . | Insulin (ng/ml) . | Glucose (mg/dl) . | Free fatty acids (mEq/l) . |
---|---|---|---|---|---|---|
Short-term study | ||||||
C57BL/6J, 1 days | PBS | 2.93 ± 0.74 | 183.88 ± 18.92 | 3.66 ± 0.92 | 169.5 ± 8.2 | 0.09 ± 0.06 |
5 μg q.i.d. MTII | 2.45 ± 0.71* | 321.13 ± 44.21† | 3.06 ± 0.89 | 173.1 ± 5.5 | 0.29 ± 0.11 | |
25 μg q.i.d. MTII | 1.64 ± 0.45 | 337.50 ± 28.58† | 2.04 ± 0.53 | 174.0 ± 5.7 | 0.36 ± 0.09*‡ | |
100 μg q.i.d. MTII | 0.83 ± 0.47†§ | 391.25 ± 39.54‖ | 0.47 ± 0.09‖¶ | 170.0 ± 10.1 | 0.52 ± 0.09†‡ | |
Pair-fed | 0.67 ± 0.21†§ | 365.50 ± 22.39‖ | 0.84 ± 0.26†§ | 183.2 ± 8.7 | 0.12 ± 0.05 | |
C57BL/6J, 2 days | PBS | 4.03 ± 0.72 | 193.00 ± 20.75 | ND | 191.6 ± 9.6 | 1.97 ± 0.14 |
100 μg q.i.d. MTII | 2.37 ± 0.15‡ | 200.25 ± 13.27 | 0.49 ± .09 | 184.3 ± 7.8 | 2.15 ± 0.71 | |
Longitudinal study | ||||||
DIO | PBS | 51.32 ± 11.67 | ND | 6.13 ± 2.02 | 134.5 ± 6.4 | 0.77 ± 0.05 |
100 μg q.i.d. MTII | 29.76 ± 3.02‡ | ND | 0.9 ± 0.33*‡ | 113.8 ± 4.2‡ | 0.73 ± 0.08 | |
Pair-fed | 47.20 ± 2.40 | ND | 4.01 ± 0.77 | 97.75 ± 7.4† | 0.62 ± 0.14 |
Data are means ± SE. Comparison of the effect of different doses of MTII administration on circulating leptin, corticosterone, insulin, glucose, and free fatty acid circulating levels in male C57BL/6J (n = 8 per group for the 1-day experiment and n = 4 per group for the 2-day experiment) and DIO (n = 4–5 per group) mice.
P < 0.05 vs. the pair-fed group by ANOVA;
P ≤ 0.01 and
P < 0.05 vs. the PBS-treated group by ANOVA with post hoc tests (or unpaired t tests for experiments including two groups);
P < 0.05 compared with 25 μg q.i.d. MTII group by ANOVA with post hoc tests;
P ≤ 0.001 vs. the PBS-treated group by ANOVA with post hoc tests (or unpaired t tests for experiments including two groups);
P < 0.01 compared with 25 μg q.i.d. MTII group by ANOVA with post hoc tests;
P < 0.01 vs. the pair-fed group by ANOVA.
. | Glucose (mg/dl) . | Insulin (ng/ml) . | Leptin (ng/ml) . | Epididymal fat pad (g) . | Fat mass (g) . | Lean mass (g) . |
---|---|---|---|---|---|---|
Chronic/low fat | ||||||
PBS | 136 ± 5 | 0.58 ± 0.07 | 1.63 ± 0.08 | 0.33 ± 0.03 | 3.80 ± 0.13 | 22.28 ± 0.30 |
MTII (10 nmol/day) | 126 ± 4 | 0.60 ± 0.10 | 1.03 ± 0.26* | 0.22 ± 0.02* | 3.39 ± 0.16 | 20.85 ± 0.29† |
Pair-fed | 121 ± 4* | 0.84 ± 0.10 | 1.39 ± 0.22 | 0.26 ± 0.04 | 3.23 ± 0.17* | 20.89 ± 0.56* |
Chronic/high fat | ||||||
PBS | 121 ± 3 | 1.65 ± 0.41 | 13.24 ± 0.67 | 1.66 ± 0.13 | 11.67 ± 1.45 | 21.83 ± 0.57 |
MTII (10 nmol/day) | 124 ± 5 | 1.15 ± 0.23 | 11.81 ± 0.98 | 1.67 ± 0.16 | 10.94 ± 0.97 | 20.06 ± 0.57* |
Pair-fed | 132 ± 6 | 1.48 ± 0.13 | 13.37 ± 0.34 | 1.73 ± 0.16 | 12.10 ± 0.79 | 21.36 ± 0.34 |
. | Glucose (mg/dl) . | Insulin (ng/ml) . | Leptin (ng/ml) . | Epididymal fat pad (g) . | Fat mass (g) . | Lean mass (g) . |
---|---|---|---|---|---|---|
Chronic/low fat | ||||||
PBS | 136 ± 5 | 0.58 ± 0.07 | 1.63 ± 0.08 | 0.33 ± 0.03 | 3.80 ± 0.13 | 22.28 ± 0.30 |
MTII (10 nmol/day) | 126 ± 4 | 0.60 ± 0.10 | 1.03 ± 0.26* | 0.22 ± 0.02* | 3.39 ± 0.16 | 20.85 ± 0.29† |
Pair-fed | 121 ± 4* | 0.84 ± 0.10 | 1.39 ± 0.22 | 0.26 ± 0.04 | 3.23 ± 0.17* | 20.89 ± 0.56* |
Chronic/high fat | ||||||
PBS | 121 ± 3 | 1.65 ± 0.41 | 13.24 ± 0.67 | 1.66 ± 0.13 | 11.67 ± 1.45 | 21.83 ± 0.57 |
MTII (10 nmol/day) | 124 ± 5 | 1.15 ± 0.23 | 11.81 ± 0.98 | 1.67 ± 0.16 | 10.94 ± 0.97 | 20.06 ± 0.57* |
Pair-fed | 132 ± 6 | 1.48 ± 0.13 | 13.37 ± 0.34 | 1.73 ± 0.16 | 12.10 ± 0.79 | 21.36 ± 0.34 |
Data are means ± SE. Comparison of the effect of central MTII administration (10 nmol/day) on hormonal parameters and weight of fat mass and lean mass in male C57BL/6J mice fed either a low- or high-fat diet for 18 weeks (n = 8 per group). Mice received centrally either PBS, MTII (10 nmol/day), or PBS and were pair-fed with the MTII group. Fat mass and lean mass were measured by DEXA scan.
P < 0.05 vs. PBS;
P < 0.01 vs. PBS.
Article Information
This work was supported by Grant DK-R37-28082 and DK-46930 from the National Institutes of Health (to J.S.F.) and by a grant from Eli Lilly (to J.S.F.) and the Swiss National Science Foundation (to D.D.P.).
REFERENCES
Address correspondence and reprint requests to Dr. Jeffrey S. Flier, Division of Endocrinology 325, Beth Israel Deaconess Medical Center, Harvard Medical School, 99 Brookline Ave., Boston, MA 02215. E-mail: [email protected].
D.D.P. and M.Z contributed equally to the work presented here.
Received for publication 20 July 2001 and accepted in revised form 12 February 2002.
AgRP, agouti-related protein; CNS, central nervous system; DEXA, dual-energy X-ray absorptiometry; DIO, diet-induced obesity; ICV, intracerebroventricular; IRS-1, insulin receptor substrate-1; MC, melanocortin; α-MSH, α-melanocyte-stimulating hormone; POMC, pro-opiomelanocortin; RQ, respiratory quotient; TNF-α, tumor necrosis factor-α.