TY - JOUR
AU1 - Appleyard, Suzanne, M.
AU2 - Hayward,, Michael
AU3 - Young, Juan, I.
AU4 - Butler, Andrew, A.
AU5 - Cone, Roger, D.
AU6 - Rubinstein,, Marcelo
AU7 - Low, Malcolm, J.
AB - Proopiomelanocortin (POMC) neurons in the hypothalamus are direct targets of the adipostatic hormone leptin and contribute to energy homeostasis by integrating peripheral and central information. The melanocortin and β-endorphin neuropeptides are processed from POMC and putatively coreleased at axon terminals. Melanocortins have been shown by a combination of pharmacological and genetic methods to have inhibitory effects on appetite and body weight. In contrast, pharmacological studies have generally indicated that opioids stimulate food intake. Here we report that male mice engineered to selectively lack β-endorphin, but that retained normal melanocortin signaling, were hyperphagic and obese. Furthermore, β-endorphin mutant and wild-type mice had identical orexigenic responses to exogenous opioids and identical anorectic responses to the nonselective opioid antagonist naloxone, implicating an alternative endogenous opioid tone to β-endorphin that physiologically stimulates feeding. These genetic data indicate that β-endorphin is required for normal regulation of feeding, but, in contrast to earlier reports suggesting opposing actions of β-endorphin and melanocortins on appetite, our results suggest a more complementary interaction between the endogenously released POMC-derived peptides in the regulation of energy homeostasis. PROOPIOMELANOCORTIN (POMC) neurons in the hypothalamic arcuate nucleus (Arc) contribute to energy homeostasis by integrating peripheral and central information related to caloric balance and metabolism (1–5). These neurons are direct targets of the adipostatic hormone leptin, receive afferent signals from the medullary dorsal vagal complex, and project to other hypothalamic and brainstem nuclei that control feeding and autonomic responses (1, 2, 4, 5). POMC is processed stoichiometrically to melanocortins and β-endorphin, and both classes of neuropeptides are putatively coreleased at axon terminals (6). The role of melanocortins in the regulation of appetite and metabolism has been recently defined by pharmacological and genetic methods (7–11); however, the actions of endogenously released β-endorphin in energy homeostasis have not been characterized. Pharmacological studies have generally indicated that opioids stimulate food intake (12, 13), opposite to the anorectic effect of melanocortins. However, it is impossible to mimic the actions of individual endogenous opioid pathways by pharmacological manipulation. β-Endorphin has a relatively high affinity for the μ, δ, and κ subtypes of opioid receptors (14), and exogenously applied β-endorphin can act at sites where it is not normally released, but that are the targets of other opioid peptide signaling pathways. Conversely, application of subtype-selective opioid receptor antagonists will interfere with multiple endogenous peptide signaling pathways. For these reasons we chose to differentiate the effects of β-endorphin on food intake from other endogenous opioid peptides, such as enkephalin and dynorphin, by genetic removal of β-endorphin. Materials and Methods Mice We generated β-END−/− and β-END+/+ mice either from heterozygous β-END+/− mating pairs or from homozygous β-END−/− or β-END+/+ mating pairs. Mice were genotyped by PCR as described previously (15) and were N10 congenic to the C57BL/6J strain (The Jackson Laboratory, Bar Harbor, ME). The transgenic (Tg) rescue experiments were carried out in N5 congenic C57BL/6J animals; Tg2 was previously termed phal*, and Tg13 was previously termed pomc*27 (16, 17). The compound Mc4r−/−/β-END−/− mice were generated by mating Mc4r−/− males (8) (N3 C57BL/6J) and β-END+/− (N10 C57BL/6J) females. Mice were housed under constant temperature in murine specific pathogen-free animal facilities with ad libitum access to water and rodent chow pellets (5% fat, 19% protein, and 5% fiber by weight; 3.4 g/kcal). Experimental animals All mice were housed in animal rooms dedicated to use by the Tg facility and located within the Department of Comparative Medicine. All procedures conformed to USPHS guidelines and were approved by the institutional animal care and use committee. Growth curves, carcass analysis and dual energy x-ray absorptiometry (DEXA) scan Mice were weighed weekly from weaning at 3 wk of age. At age 5 months mice were euthanized, and liver, spleen, kidney, heart, testis, and all fat pads were removed and weighed. A portion of the inguinal fat pads was placed in 4% paraformaldehyde, then dehydrated, embedded in paraffin. Fifteen-micrometer-thick sections were stained with hematoxylin and eosin. The percent body fat was determined by analyzing the carcass composition of mice by DEXA (PIXImus mouse densitometer, Lunar Corp., Madison, WI). Food intake measurements Mice were housed individually at 5 wk of age. They had ad libitum access to water and food pellets (5% fat, 19% protein, and 5% fiber; 3.4 g/kcal) provided in wire cage tops. Food was weighed daily, and average daily intake was calculated from consecutive measurements over 7 d after a 10- to 14-d acclimatization period to individual housing. Oxygen consumption measurements Oxygen consumption was determined for multiple animals simultaneously by indirect calorimetry using an Oxymax system (Columbus Instruments, Columbus, OH) as previously described (10). Plasma insulin, glucose, leptin, and T4 measurements Mice were killed after a 16- to 20-h fast (fasted) or after ad libitum access to food (fed), and trunk blood was collected. T4, insulin, and leptin levels were determined by RIA (Linco Research, Inc., St. Charles, MO) from sera. Glucose levels were measured from blood using a Basic One Touch glucometer (Lifescan, Johnson \|[amp ]\| Johnson, Milpitas, CA). All blood samples were obtained in the fed state from 5-month-old mice unless otherwise stated. Glucose tolerance test Mice were fasted for 16 h and then given an ip injection of 2 g/kg glucose. Glucose levels were determined from tail bleeds at 0, 15, 30, 45, 60, 90, 120, 150, and 180 min. Insulin sensitivity test Mice were fasted for 16 h and then given an ip injection of 0.75 U/kg insulin. Glucose levels were determined from tail bleeds at 0, 30, 60, and 90 min. Immunocytochemistry Mice were perfused with 4% paraformaldehyde, and hypothalamic (24-μm thick) and pituitary (12-μm thick) sections were prepared for immunostaining as described previously. A rabbit polyclonal antibody to β-endorphin (18) was diluted 1:5000 in 1 mg/ml BSA (fraction V) and 0.3% Triton X-100 in PBS for incubation with free-floating hypothalamic sections. The β-endorphin antibody was used in the same buffer at 1:1000 for staining slide-mounted pituitary sections. After washing the sections and incubating with biotinylated goat antirabbit IgG (Vector Laboratories, Inc., Burlingame, CA), the reaction product was visualized with an ABC Elite kit (Vector Laboratories, Inc.) and diaminobenzidine, followed by a methyl green counterstain. β-Endorphin RIA β-Endorphin levels were measured in both the hypothalamus and pituitary using a rat β-endorphin RIA kit (Phoenix Pharmaceuticals, Inc., Belmont, CA). The tissue was prepared as described in the manufacturer’s instructions. Intracerebroventricular (icv) injections A 27-gauge stainless steel cannula was surgically placed into a lateral ventricle of anesthetized mice as described previously (9). The mice were allowed to recover for at least 1 wk. Neuropeptide Y (NPY; 1.5 μg; Peninsula Laboratories, Inc., Belmont, CA), 0.5 μg β-END (gift from Dr. A. Parlow), or saline in a 1.5-μl volume was injected through the cannula. Naloxone (10 mg/kg) or 0.9% saline vehicle was given ip 15 min before the icv injection. Food intake was measured each hour for 3 h after the injection. This short time period within the light phase of the animals’ diurnal cycle was inadequate to reveal the basal difference between genotypes that was primarily evident during nocturnal feeding bouts. Naloxone effects on food intake Mice were fasted for 24 h and then given an ip injection of 10 mg/kg naloxone or 0.9% saline. After 15 min, food was returned, and food intake was measured for 1 h. Statistics Statistical analyses were performed using PRISM (GraphPad Software, Inc., San Diego, CA) or StatView (SAS Institute, Inc., Cary, NC). Growth curves, insulin sensitivity, and glucose tolerance tests were analyzed using a two-factor ANOVA with time as the repeated measure. All other data were analyzed using t test or ANOVA, followed by Bonferroni/Dunn post hoc analysis for paired group comparisons. Results Male mice lacking β-endorphin (β-END−/−) are significantly heavier and have greater adiposity than wild-type (β-END+/+) male mice To examine the effect of β-endorphin deficiency on weight homeostasis we weighed both β-END−/− and β-END+/+ mice over time. These growth curves demonstrated that male β-END−/− mice weighed significantly more than male β-END+/+ mice, starting at 4 wk of age and continuing into adulthood [two-way repeated measures ANOVA: significant main effect of genotype (P < 0.001), time (P < 0.0001), and genotype × time interaction (P < 0.0001); df of genotype = 1, time = 9, genotype × time = 9; F values: genotype = 16.6, time = 1213, genotype × time = 4.5; n = 15–20; Fig. 1A]. In contrast, the weights of female β-END−/− mice differed only transiently from those of female β-END+/+ mice between 4–8 wk of age (P < 0.05, by t test at each time point; n = 10–11; Fig. 1A). Body length and the weights of various organs, such as liver, spleen, kidney, heart, and testis, were not changed in either sex of β-END−/− mice (data not shown). However, both the inguinal/gonadal and retroperitoneal/perirenal white fat stores of male β-END−/− mice were 2-fold heavier than those of male β-END+/+ mice (P < 0.01, by t test; n = 8), whereas intrascapular brown fat was not altered (Fig. 1B). Furthermore, male β-END−/− mice had 50% greater total body fat as measured by DEXA scan (P < 0.05, by t test; n = 8; Fig. 1C), and histological examination suggested hypertrophy of the adipocytes (Fig. 1D). The fat stores of female β-END−/− mice were not increased (data not shown). An identical sexually dimorphic phenotype with equivalent or greater male pattern obesity was observed in β-END−/− mice crossed onto either a 129S6/SV (6 months old: β-END+/+ 33.9 ± 2.1 g, n = 8; β-END−/−, 49.6 ± 1.9 g, n = 13) or outbred Swiss albino background. These data indicate that the development of obesity in β-endorphin-deficient mice is independent of the known genetic predisposition of C57BL/6 mice to gain excessive weight and adipose mass. Figure 1. Open in new tabDownload slide Weight, fat mass, food intake, and oxygen consumption of β-END−/− and β-END+/+ mice. A, Growth curves of sibling male and female β-END−/− and β-END+/+ mice reared by β-END+/− parents (n = 11–24) (male mice, P < 0.001 for genotype, by repeated measure ANOVA). B, Weights of inguinal/gonadal (IG), retroperitoneal/perirenal (PR), and intrascapular brown (BR) fat pads in male β-END−/− and β-END+/+ mice (n = 8; *, P < 0.01, by t test). C, Body fat percentage from DEXA scan analysis of male β-END−/− and β-END+/+ mice (n = 8; *, P < 0.05, by t test). D, Morphology of inguinal fat from male β-END−/− and β-END+/+ mice. Scale bar, 50 μm. E, Average daily food intake of β-END+/+/Mc4r+/+, β-END−/−/Mc4r+/+, β-END+/+/Mc4r−/−, and β-END−/−/Mc4r−/− male mice (n = 6–8; *, P < 0.05 compared with wild-type mice; #, P < 0.05 compared with all other groups, by t test). F, Oxygen consumption in β-END−/− and β-END+/+ mice. All data are expressed as the mean ± sem. Figure 1. Open in new tabDownload slide Weight, fat mass, food intake, and oxygen consumption of β-END−/− and β-END+/+ mice. A, Growth curves of sibling male and female β-END−/− and β-END+/+ mice reared by β-END+/− parents (n = 11–24) (male mice, P < 0.001 for genotype, by repeated measure ANOVA). B, Weights of inguinal/gonadal (IG), retroperitoneal/perirenal (PR), and intrascapular brown (BR) fat pads in male β-END−/− and β-END+/+ mice (n = 8; *, P < 0.01, by t test). C, Body fat percentage from DEXA scan analysis of male β-END−/− and β-END+/+ mice (n = 8; *, P < 0.05, by t test). D, Morphology of inguinal fat from male β-END−/− and β-END+/+ mice. Scale bar, 50 μm. E, Average daily food intake of β-END+/+/Mc4r+/+, β-END−/−/Mc4r+/+, β-END+/+/Mc4r−/−, and β-END−/−/Mc4r−/− male mice (n = 6–8; *, P < 0.05 compared with wild-type mice; #, P < 0.05 compared with all other groups, by t test). F, Oxygen consumption in β-END−/− and β-END+/+ mice. All data are expressed as the mean ± sem. Male β-END−/− mice have increased food intake, but no overt change in basal metabolic rate To determine the underlying mechanism for the increased weight and adiposity of the β-END−/− mice, we examined both their food intake and their basal metabolic rate. The average daily food intake of 7- to 8-wk-old male β-END−/− mice was significantly increased compared with that of β-END+/+ males (P < 0.05, by t test; n = 8; Fig. 1E). To determine whether the increased feeding was due to attenuated αMSH biosynthesis or signaling, we generated compound homozygote mice lacking both β-endorphin and the MC4 receptor, the primary receptor mediating the anorectic effects of αMSH. The average daily food intake of β-END−/−/Mc4r−/− mice was significantly higher than that of either β-END−/−/Mc4r+/+ or β-END+/+/Mc4r−/− mice (P < 0.05, by t test; n = 6–8; Fig. 1E). β-END−/−/Mc4r−/− mice were also significantly heavier than β-END+/+/Mc4r−/− mice between 4 and 9 wk of age [two way repeated measures ANOVA: significant main effect of genotype (P < 0.05) and time (P < 0.005); df of genotype = 1, time = 3; F values: genotype = 6.78, time = 109.4; n = 7]. The weights of the mice (grams) were: at 4 wk: β-END+/+/Mc4r−/−, 17.6 ± 1.3; β-END−/−/Mc4r−/−, 21.5 ± 0.7; 7 wk: β-END+/+/Mc4r−/−, 27.7 ± 1.1; β-END−/−/Mc4r−/−, 31.9 ± 0.9; and 9 wk: β-END+/+/ Mc4r−/−, 36.5 ± 1.0; β-END−/−/Mc4r−/−, 39.0 ± 1.0. The difference in weights was not significant after 10 wk of age, probably due to the large metabolic defect seen in the Mc4r−/− mice that overshadowed the additive genotype effect on food intake. In contrast to the augmented food intake, there were no significant differences between β-END−/− and β-END+/+ mice in their basal metabolic rate, as measured by oxygen consumption or respiratory quotient (Fig. 1F), or serum T4 levels (β-END+/+, 3.9 ± 0.1 ng/ml, n = 8; β-END−/−, 3.5 ± 0.4 ng/ml, n = 7). These data are consistent with our previous findings that the basal core temperature and activity levels of the β-END−/− and β-END+/+ male mice were also not significantly different (15, 19). β-END−/− mice are hyperinsulinemic and hyperleptinemic and have altered glucose homeostasis Previous studies have suggested that β-endorphin plays a role in the regulation of insulin and glucose homeostasis (20, 21). We therefore examined insulin and glucose homeostasis in β-END−/− mice. We found that 5-month-old β-END−/− male mice had elevated basal insulin levels [Fig. 2A; one-way ANOVA F(3,17) = 7.17; P < 0.05 for β-END+/+ fed mice compared with β-END+/+ fasted mice; P < 0.05 for β-END−/− fed mice compared with β-END+/+ fed mice and β-END−/− fasted mice) and exhibited an attenuated hypoglycemic response to insulin administration [Fig. 2B; two way repeated measures ANOVA: significant main effect of genotype (P < 0.01) and time (P < 0.001); df of genotype = 1, time = 2, genotype × time = 2; F values: genotype = 8.8; time = 5.1, genotype × time = 1.0; n = 8]. Furthermore, male β-END−/− mice were not able to clear glucose as efficiently as β-END+/+ mice in a glucose tolerance test (Fig. 2C; two-way repeated measures ANOVA: significant main effect of genotype (P < 0.005), time (P < 0.0001), and genotype × time interaction (P < 0.001); df of genotype = 1, time = 12, genotype × time = 12; F values: genotype = 13.7, time = 141.0, and genotype × time F = 3.4; n = 6–7], although basal fasted and postprandial glucose levels were normal. In contrast, the nonobese female β-END−/− mice had normal glucose tolerance tests (Fig. 2D). However, nonobese male β-END−/− mice at 4 wk of age had normal glucose and insulin levels (weight: β-END+/+, 19.4 ± 0.5 g, n = 6; β-END−/−, 19.6 ± 0.4 g, n = 5; glucose: β-END+/+, 143 ± 4.9 mg/ml, n = 6; β-END−/−, 141 ± 11.8 mg/ml, n = 5; insulin: β-END+/+, 0.7 ± 0.1 ng/ml, n = 6; β-END−/−, 0.8 ± 0.1 ng/ml, n = 5). These results suggest that the modest hyperinsulinemia in adult males is probably secondary to the increased fat mass rather than due to a direct effect of the loss of β-endorphin. Examination of the levels of the adipostatic hormone leptin showed that adult male β-END−/− mice were also hyperleptinemic (β-END+/+, 3.9 ± 0.7 ng/ml, n = 9; β-END−/−, 8.3 ± 0.9 ng/ml, n = 10; P < 0.01, by t test) consistent with their increased adiposity. Figure 2. Open in new tabDownload slide Insulin and glucose homeostasis in β-END−/− and β-END+/+ mice. A, Insulin levels in male β-END−/− and β-END+/+ mice in the fed and fasted states (n = 5–6; *, P < 0.05 compared with β-END+/+ fasted mice; #, P < 0.05 compared with β-END+/+ fed mice and β-END−/− fasted mice, by one-way ANOVA). B, Insulin sensitivity test in male β-END−/− and β-END+/+ mice (n = 8; P < 0.01 for genotype, by repeated measure ANOVA). C, Glucose tolerance test in male β-END−/− and β-END+/+ mice (n = 6); P < 0.005 for genotype, by repeated measure ANOVA). D, Glucose tolerance test in female β-END−/− and β-END+/+ mice. All data are expressed as the mean ± sem. Figure 2. Open in new tabDownload slide Insulin and glucose homeostasis in β-END−/− and β-END+/+ mice. A, Insulin levels in male β-END−/− and β-END+/+ mice in the fed and fasted states (n = 5–6; *, P < 0.05 compared with β-END+/+ fasted mice; #, P < 0.05 compared with β-END+/+ fed mice and β-END−/− fasted mice, by one-way ANOVA). B, Insulin sensitivity test in male β-END−/− and β-END+/+ mice (n = 8; P < 0.01 for genotype, by repeated measure ANOVA). C, Glucose tolerance test in male β-END−/− and β-END+/+ mice (n = 6); P < 0.005 for genotype, by repeated measure ANOVA). D, Glucose tolerance test in female β-END−/− and β-END+/+ mice. All data are expressed as the mean ± sem. Tissue-specific Tg rescue of β-endorphin in β-END−/− mice To determine which site of β-endorphin release was responsible for the obesity phenotype in β-END−/− mice, we used a tissue-specific Tg rescue approach (Fig. 3A). Tg2 contains 2 kb of the mouse Pomc promoter previously shown to be sufficient to express the gene in the pituitary, but not the central nervous system (CNS; Refs. 16 and 17). In contrast, Tg13 has all the regulatory elements of the mouse Pomc gene required for Tg expression in the pituitary and CNS (3, 16). Both constructs have an in-frame oligonucleotide insertion in exon 3 that disrupts the biological activity of αMSH and ACTH (Fig. 3A). The two strains of Tg mice were crossed to β-END−/− mice, and progeny were examined by immunohistochemistry. Tg2 rescued β-endorphin expression only in pituitary anterior lobe corticotrophs and intermediate lobe melanotrophs, whereas Tg13 restored accurate cell-specific β-endorphin expression in both lobes of the pituitary and Arc neurons (Fig. 3B). The β-endorphin content in the pituitary (combined anterior and neurointermediate lobes) was restored to wild-type levels by both transgenes (β-END+/+, 257 ± 19 ng/gland, n = 5; β-END−/−, <0.16 ng/gland, n = 6; Tg2/β-END−/−, 197 ± 8 ng/gland, n = 5; Tg13/β-END−/−, 276 ± 42 ng/gland, n = 4). The levels of β-endorphin in β-END+/+, Tg2/β-END−/−, and Tg13/β-END−/− mice were not statistically different from each other. Tg13 also restored quantitatively normal levels of hypothalamic β-endorphin in all rescued mice, whereas β-endorphin remained undetectable in four of five hypothalami examined from the Tg2 rescued mice, consistent with the immunohistochemical data (β-END+/+, 484 ± 61 pg/hypothalamus, n = 5; β-END−/−, <160 pg/hypothalamus, n = 7; Tg2/β-END−/−, <160 pg/hypothalamus, n = 4; Tg13/β-END−/−, 387 ± 27 ng/hypothalamus, n = 4). The one other Tg2/β-END−/− mouse had levels of immunoreactive β-endorphin just above RIA sensitivity, possibly due to trace contamination of the hypothalamic block by a fragment of the underlying pituitary intermediate lobe. Figure 3. Open in new tabDownload slide Tg rescue of β-endorphin expression restores body and fat pad weights of 5-month-old male β-END−/− mice. A, Structure of Pomc transgenes. Tg2 is a 10.2-kb genomic fragment that contains the entire transcriptional unit of the mouse Pomc gene and 2 kb of 5′- and 3′-flanking sequences. Tg13 is a 27-kb genomic fragment that contains the entire transcriptional unit of the mouse Pomc gene and 13 kb of 5′- and 8 kb of 3′-flanking sequences. Both constructs have an in-frame 33-bp heterologous sequence (arrow) inserted into a blunted NcoI site in exon 3 that disrupts the coding sequence for ACTH and αMSH by the addition of 11 amino acids (CLSLSSLLSRM), but does not alter the processing or sequence of the carboxyl-terminal β-endorphin. The boxes indicate exon sequences and the black-shaded regions indicate untranslated sequences. B, Immunocytochemical localization of β-endorphin in the pituitary (left) and Arc (right) of β-END+/+, β-END−/−, and β-END−/− mice expressing the Tg2 or Tg13 transgene. The arrows indicate β-endorphin-immunoreactive pituitary corticotrophs and melanotrophs and hypothalamic neurons. IL, Intermediate lobe; III, third ventricle. Scale bars, 200 μm for pituitary sections; 50 μm for Arc sections. C, Body weights (n = 6–15; by one-way ANOVA: *, P < 0.001 compared with both β-END+/+ and β-END−/−/Tg13; # P < 0.01 compared with β-END+/+, β-END−/−, and β-END−/−/Tg13 mice). D, Weights of inguinal/gonadal (IG) fat pads (n = 6–15; by one-way ANOVA: *, P < 0.01 compared with both β-END+/+ and β-END−/−/Tg13). E, Weights of retroperitoneal/perirenal (PR) fat pads (n = 6–15; by one-way ANOVA: *, P < 0.001 compared with both β-END+/+ and β-END−/−/Tg13; #, P < 0.05 compared with β-END+/+, β-END−/−, and β-END−/−/Tg13). All data are expressed as the mean ± sem. Figure 3. Open in new tabDownload slide Tg rescue of β-endorphin expression restores body and fat pad weights of 5-month-old male β-END−/− mice. A, Structure of Pomc transgenes. Tg2 is a 10.2-kb genomic fragment that contains the entire transcriptional unit of the mouse Pomc gene and 2 kb of 5′- and 3′-flanking sequences. Tg13 is a 27-kb genomic fragment that contains the entire transcriptional unit of the mouse Pomc gene and 13 kb of 5′- and 8 kb of 3′-flanking sequences. Both constructs have an in-frame 33-bp heterologous sequence (arrow) inserted into a blunted NcoI site in exon 3 that disrupts the coding sequence for ACTH and αMSH by the addition of 11 amino acids (CLSLSSLLSRM), but does not alter the processing or sequence of the carboxyl-terminal β-endorphin. The boxes indicate exon sequences and the black-shaded regions indicate untranslated sequences. B, Immunocytochemical localization of β-endorphin in the pituitary (left) and Arc (right) of β-END+/+, β-END−/−, and β-END−/− mice expressing the Tg2 or Tg13 transgene. The arrows indicate β-endorphin-immunoreactive pituitary corticotrophs and melanotrophs and hypothalamic neurons. IL, Intermediate lobe; III, third ventricle. Scale bars, 200 μm for pituitary sections; 50 μm for Arc sections. C, Body weights (n = 6–15; by one-way ANOVA: *, P < 0.001 compared with both β-END+/+ and β-END−/−/Tg13; # P < 0.01 compared with β-END+/+, β-END−/−, and β-END−/−/Tg13 mice). D, Weights of inguinal/gonadal (IG) fat pads (n = 6–15; by one-way ANOVA: *, P < 0.01 compared with both β-END+/+ and β-END−/−/Tg13). E, Weights of retroperitoneal/perirenal (PR) fat pads (n = 6–15; by one-way ANOVA: *, P < 0.001 compared with both β-END+/+ and β-END−/−/Tg13; #, P < 0.05 compared with β-END+/+, β-END−/−, and β-END−/−/Tg13). All data are expressed as the mean ± sem. Tg rescue of β-endorphin to sites in the CNS is required for the full rescue of the obesity phenotype Re-expression of β-endorphin to the pituitary by Tg2 partially rescued both the body weight and fat pad weights of β-END−/− mice (Fig. 3, C–E). However, re-expression of β-endorphin to the pituitary and CNS by Tg13 completely normalized body weight and fat mass of the β-END−/− mice (Fig. 3, C–E), suggesting that both central and pituitary β-endorphin expressions contribute to energy homeostasis [one way ANOVA: body weights (n = 6–15): F(3,38) = 22.7, P < 0.001 for β-END+/+ mice compared with β-END−/− mice, P < 0.001 for Tg13/β-END−/− compared with β-END−/− mice, P < 0.01 for Tg2/β-END−/− compared with β-END−/−, β-END+/+, and Tg13/β-END−/− mice; inguinal fat pad weights (n = 6–15): F(3,38) = 11.6, P < 0.001 for β-END+/+ mice compared with β-END−/− mice, P < 0.001 for Tg13/β-END−/− compared with β-END−/− mice, P < 0.01 for Tg2/β-END−/− compared with β-END+/+ and Tg13/β-END−/− mice; retroperitoneal/perirenal fat pad weights (n = 6–15): F(3,38) = 11.9, P < 0.001 for β-END+/+ mice compared with β-END−/− mice, P < 0.01 for Tg13/β-END−/− compared with β-END−/−, P < 0.01 for Tg2/β-END−/− mice compared with β-END+/+ and β-END−/− mice]. NPY, β-endorphin, and naloxone effects on food intake in β-END−/− mice We next examined whether the response of β-END−/− mice to food intake stimulated by opioids or NPY was altered. β-Endorphin injected icv stimulated an equivalent increase in food intake in wild-type and β-END−/− mice (Fig. 4A). Interestingly, the orexigenic effects of NPY were slightly increased in male β-END−/− mice (Fig. 4B), but were unaltered in female β-END−/− mice, consistent with their normal weight and feeding behavior (data not shown). NPY-stimulated food intake was inhibited equivalently by the nonselective opioid antagonist naloxone in both genotypes (Fig. 4B). Both 10 and 1 mg/kg (data not shown) naloxone attenuated NPY-induced food intake. These doses were used because they have been shown previously to attenuate food intake (22–25). Naloxone has also been shown to decrease food intake after 24-h food deprivation (23, 26, 27). We therefore tested whether this refeeding was altered in β-END−/− mice. However, naloxone inhibited feeding to the same extent in previously food-restricted β-END−/− and wild-type mice (Fig. 4C). Figure 4. Open in new tabDownload slide Effects of β-endorphin, NPY, and naloxone on food intake in male β-END+/+ and β-END−/− mice. A, Cumulative intake of food during a 3-h period after icv injection of either saline or 0.5 μg β-endorphin 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 6–10; *, P < 0.05 vs. saline, no effect of genotype, by t test). B, Cumulative food intake during a 3-h period after an icv injection of either saline or 1.5 μg NPY 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 10–15; *, P < 0.01 vs. saline-treated β-END+/+ mice; #, P < 0.01 vs. saline-treated β-END−/− mice and P < 0.05 vs. NPY-treated β-END+/+ mice, by t test). C, Food intake during a 1-h period after a 24-h fast and 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 6–10; *, P < 0.05 vs. saline, by t test). All data are expressed as the mean ± sem. Figure 4. Open in new tabDownload slide Effects of β-endorphin, NPY, and naloxone on food intake in male β-END+/+ and β-END−/− mice. A, Cumulative intake of food during a 3-h period after icv injection of either saline or 0.5 μg β-endorphin 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 6–10; *, P < 0.05 vs. saline, no effect of genotype, by t test). B, Cumulative food intake during a 3-h period after an icv injection of either saline or 1.5 μg NPY 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 10–15; *, P < 0.01 vs. saline-treated β-END+/+ mice; #, P < 0.01 vs. saline-treated β-END−/− mice and P < 0.05 vs. NPY-treated β-END+/+ mice, by t test). C, Food intake during a 1-h period after a 24-h fast and 15 min after an ip injection of either saline or 10 mg/kg naloxone (n = 6–10; *, P < 0.05 vs. saline, by t test). All data are expressed as the mean ± sem. Discussion The main finding of this study was that mice lacking the endogenous opioid peptide β-endorphin have a sexually dimorphic obesity phenotype, present in only male mice. The increased adiposity of β-END−/− mice appears to be due to changes in caloric intake and not energy utilization. Furthermore, the effect of the opioid antagonist naloxone on food intake is unchanged in β-END−/− mice, suggesting that an alternative endogenous opioid tone physiologically stimulates feeding. Our results also suggest that the effects of β-endorphin on food intake are independent of αMSH, which is proposed to be coreleased with β-endorphin from POMC neuronal terminals (6). αMSH is a potent anorexigenic peptide that plays an important role in energy homeostasis through actions primarily at the MC4 receptor (8). Here we report that the increase in food intake seen in mice lacking the MC4 receptor and that in mice lacking β-endorphin are additive, as the food intake of male mice lacking both β-endorphin and the MC4 receptor was significantly greater than that of Mc4r−/− mice. Together with our previous observations of normal tissue content of αMSH in β-END−/− mice (15), these data make it unlikely that the obesity phenotype of the β-END−/− mice is due to changes in αMSH biosynthesis or MC4 receptor signaling. We did not generate mice lacking both β-endorphin and the MC3 receptor, as the phenotype of MC3 receptor knockout mice is significantly different from that of the β-endorphin-deficient mice described here and shows a reversed sexual dimorphism (10, 11). It is therefore unlikely that the obesity phenotype of β-END−/− mice is due to changes in γMSH biosynthesis or MC3 receptor signaling. POMC is expressed in both the Arc and the caudal nucleus of the solitary tract of the brainstem (28), two regions implicated in energy homeostasis (1, 4, 13). However, β-endorphin is also synthesized and released from the pituitary into the circulation, where it can act on peripheral sites to potentially alter energy homeostasis (1, 4, 13, 17). Our results suggest that re-expression of β-endorphin to its central sites of release is required to fully rescue the obesity phenotype. The mechanism by which re-expression of β-endorphin to the pituitary only partially rescues the obesity is not clear. However, it may reflect either a contribution of circulating β-endorphin on peripheral sites of action or transport of β-endorphin across the blood-brain barrier and subsequent activation of central opioid receptors (29). The obesity phenotype seen in the β-END−/− mice appears somewhat paradoxical, as exogenous opioid agonists have previously been shown to increase food intake, potentially by modulating the rewarding aspects of food (12). Indeed, we also found that exogenous β-endorphin stimulated food intake in both β-END−/− and β-END+/+ mice despite the obesity resulting from the selective genetic deficiency of endogenous β-endorphin. However, a critical difference between the pharmacological and genetic manipulations of opioid action is that genetic ablation removes β-endorphin signaling from its physiological site of release only. As discussed in the introduction, this effect is impossible to mimic by pharmacological approaches. Therefore, the genetic removal of β-endorphin may reveal an anorectic effect of the peptide. Alternatively, the apparent discrepancy between the pharmacology and this genetic study could reflect the different effects of opioid agonists on short- and long-term food intake. Opioid agonists have been shown to increase short-term food intake; however, they do not increase long-term food intake (30, 31). Therefore, the genetic removal of β-endorphin may reveal a long-term anorectic action of the peptide compared with potential short-term orexigenic effects of this peptide. Alternatively, removal of β-endorphin from the hypothalamus may disrupt normal signaling and lead to an increase in orexigenic tone. Indeed, there was a very mild increase in NPY-stimulated food intake in male, but not female, β-END−/− mice. However, the nonspecific antagonist naloxone attenuated NPY-induced food intake to the same degree in both β-END−/− and β-END+/+ mice, suggesting that an endogenous opioid peptide other than β-endorphin is required for the opioid-dependent actions of NPY. Consistent with this idea of an alternative opioid tone involved in the orexigenic effects of opioids, naloxone also attenuated food intake after food deprivation to the same extent in both β-END−/− and β-END+/+ mice. It is also possible that this endogenous opioid tone is a consequence of compensatory increases that occur due to the removal of β-endorphin, although we have no direct evidence for altered enkephalin or dynorphin expression in β-END−/− mice. Recent evidence from our laboratory demonstrated that male mice lacking β-endorphin have reduced levels of operant responses (e.g. lever presses) for food reinforcers when maintained under ad libitum feeding conditions, but not when food-restricted (32). These data suggest that β-endorphin modulates the hedonic value of food independently of energy homeostasis and support a role for endogenous β-endorphin in the rewarding aspects of food intake. Together with the data presented here, these results imply multiple actions for β-endorphin in the regulation of food intake. However, as male mice lacking β-endorphin are hyperphagic and mildly obese, the overriding effect of the genetic removal of β-endorphin under ad libitum access to food appears to be an increase in orexigenic tone despite a reduction in the hedonic value of food intake. β-Endorphin has been proposed to produce its feeding effects predominantly through activation of the μ opioid receptor (33), although it has affinity for μ, δ, and κ opioid receptors. No compensatory changes have been detected in the levels of μ, δ, or κ opioid receptors in β-END−/− mice (34, 35). Interestingly, MOP−/− mice have a sexually dimorphic obesity phenotype markedly similar to that of β-END−/− mice (Kieffer, B., unpublished observations). The fact that deficiencies in either the endogenous opioid ligand β-endorphin or in the μ opioid receptor cause such similar weight phenotypes indicates that the obesity, although milder than that associated with some other monogenic syndromes, reflects a real alteration in physiological function due to loss of the β-endorphin tone. Although the sexually dimorphic nature of the weight phenotype observed in β-END−/− mice remains to be fully explored, it is consistent with previous studies showing that the expression and effects of β-endorphin are sexually dimorphic (36–39). In conclusion, β-END−/− male mice have increased weight and adiposity that appear to result primarily from increased food intake. Our results suggest the endogenous opioid β-endorphin has an unexpected anorexic effect in regulating energy homeostasis. Furthermore, these data implicate an alternative endogenous opioid system in some of the orexigenic effects of opioids previously attributed to β-endorphin. Acknowledgments We thank A. Parlow and the National Hormone and Pituitary Program for the β-endorphin, R. Klein and R. Turner for assistance with the dual energy x-ray absorptiometry, and J. Williams for comments on the manuscript. This work was supported by NIH Grants PO1-DK-55819, F32-DK-10082, F32-DA-05841, and R03-TW-01233; the International Scholar Program of the Howard Hughes Medical Institute; and Agencia Nacional de Promoción Científica y Tecnológica Grant PICT98 03167. Abbreviations Arc Arcuate nucleus; CNS central nervous system; DEXA dual energy x-ray absorptiometry; icv intracerebroventricular; NPY neuropeptide Y; POMC proopiomelanocortin; Tg transgenic. 1 Elmquist JK , Maratos-Flier E , Saper CB , Flier JS 1998 Unraveling the central nervous system pathways underlying responses to leptin. Nat Neurosci 1 : 445 – 450 Google Scholar Crossref Search ADS PubMed WorldCat 2 Kalra SP , Dube MG , Pu S , Xu B , Horvath TL , Kalra PS 1999 Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 20 : 68 – 100 Google Scholar PubMed WorldCat 3 Cowley MA , Smart JL , Rubinstein M , Cerdan MG , Diano S , Horvath TL , Cone RD , Low MJ 2001 Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 411 : 480 – 484 Google Scholar Crossref Search ADS PubMed WorldCat 4 Flier J , Maratos-Flier E 2000 Energy homeostasis and body weight . Curr Biol 10 : R215 – R217 Google Scholar Crossref Search ADS PubMed WorldCat 5 Schwartz MW , Woods SC , Porte Jr D , Seeley RJ , Baskin DG 2000 Central nervous system control of food intake. Nature 404 : 661 – 671 Google Scholar PubMed WorldCat 6 Castro MG , Morrison E 1997 Post-translational processing of proopiomelanocortin in the pituitary and in the brain. Crit Rev Neurobiol 11 : 35 – 57 Google Scholar Crossref Search ADS PubMed WorldCat 7 Yeo GS , Farooqi IS , Challis BG , Jackson RS , O’Rahilly S 2000 The role of melanocortin signalling in the control of body weight: evidence from human and murine genetic models. Q J Med 93 : 7 – 14 Google Scholar Crossref Search ADS WorldCat 8 Huszar D , Lynch CA , Fairchild-Huntress V , Dunmore JH , Fang Q , Berkemeier LR , Gu W , Kesterson RA , Boston BA , Cone RD , Smith FJ , Campfield LA , Burn P , Lee F 1997 Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 88 : 131 – 141 Google Scholar Crossref Search ADS PubMed WorldCat 9 Fan W , Boston BA , Kesterson RA , Hruby VJ , Cone RD 1997 Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385 : 165 – 168 Google Scholar Crossref Search ADS PubMed WorldCat 10 Butler AA , Kesterson RA , Khong K , Cullen MJ , Pelleymounter MA , Dekoning J , Baetscher M , Cone RD 2000 A unique metabolic syndrome causes obesity in the melanocortin-3 receptor-deficient mouse. Endocrinology 141 : 3518 – 3521 Google Scholar Crossref Search ADS PubMed WorldCat 11 Chen AS , Marsh DJ , Trumbauer ME , Frazier EG , Guan XM , Yu H , Rosenblum CI , Vongs A , Feng Y , Cao L , Metzger JM , Strack AM , Camacho RE , Mellin TN , Nunes CN , Min W , Fisher J , Gopal-Truter S , MacIntyre DE , Chen HY , Van Der Ploeg LH 2000 Inactivation of the mouse melanocortin-3 receptor results in increased fat mass and reduced lean body mass. Nat Genet 26 : 97 – 102 Google Scholar Crossref Search ADS PubMed WorldCat 12 Glass MJ , Billington CJ , Levine AS 1999 Opioids and food intake: distributed functional neural pathways? Neuropeptides 33 : 360 – 368 Google Scholar Crossref Search ADS PubMed WorldCat 13 Kalra SP , Horvath TL 1998 Neuroendocrine interactions between galanin, opioids, and neuropeptide Y in the control of reproduction and appetite. Ann NY Acad Sci 863 : 236 – 240 Google Scholar Crossref Search ADS PubMed WorldCat 14 Kieffer BL 1995 Recent advances in molecular recognition and signal transduction of active peptides: receptors for opioid peptides. Cell Mol Neurobiol 15 : 615 – 635 Google Scholar Crossref Search ADS PubMed WorldCat 15 Rubinstein M , Mogil JS , Japon M , Chan EC , Allen RG , Low MJ 1996 Absence of opioid stress-induced analgesia in mice lacking β-endorphin by site-directed mutagenesis. Proc Natl Acad Sci USA 93 : 3995 – 4000 Google Scholar Crossref Search ADS PubMed WorldCat 16 Young JI , Otero V , Cerdan MG , Falzone TL , Chan EC , Low MJ , Rubinstein M 1998 Authentic cell-specific and developmentally regulated expression of pro-opiomelanocortin genomic fragments in hypothalamic and hindbrain neurons of transgenic mice. J Neurosci 18 : 6631 – 6640 Google Scholar PubMed WorldCat 17 Rubinstein M , Mortrud M , Liu B , Low MJ 1993 Rat and mouse proopiomelanocortin gene sequences target tissue-specific expression to the pituitary gland but not to the hypothalamus of transgenic mice. Neuroendocrinology 58 : 373 – 380 Google Scholar Crossref Search ADS PubMed WorldCat 18 Dave JR , Rubinstein N , Eskay RL 1985 Evidence that β-endorphin binds to specific receptors in rat peripheral tissues and stimulates the adenylate cyclase-adenosine 3′,5′-monophosphate system. Endocrinology 117 : 1389 – 1396 Google Scholar Crossref Search ADS PubMed WorldCat 19 Grahame NJ , Low MJ , Cunningham CL 1998 Intravenous self-administration of ethanol in β-endorphin-deficient mice. Alcohol Clin Exp Res 22 : 1093 – 1098 Google Scholar Crossref Search ADS PubMed WorldCat 20 Ramabadran K , Bansinath M 1990 Glucose homeostasis and endogenous opioid peptides. Int J Clin Pharmacol Ther Toxicol 28 : 89 – 98 Google Scholar PubMed WorldCat 21 Fulghesu AM , Ciampelli M , Guido M , Murgia F , Caruso A , Mancuso S , Lanzone A 1998 Role of opioid tone in the pathophysiology of hyperinsulinemia and insulin resistance in polycystic ovarian disease. Metabolism 47 : 158 – 162 Google Scholar Crossref Search ADS PubMed WorldCat 22 Dawson Jr R 1983 Naloxone-induced suppression of food intake is potentiated by neonatal administration of monosodium glutamate to mice. Neurobehav Toxicol Teratol 5 : 523 – 526 Google Scholar PubMed WorldCat 23 Brown DR , Holtzman SG 1979 Suppression of deprivation-induced food and water intake in rats and mice by naloxone. Pharmacol Biochem Behav 11 : 567 – 573 Google Scholar Crossref Search ADS PubMed WorldCat 24 Hayward MD , Low MJ 2001 The effect of naloxone on operant behavior for food reinforcers in DBA/2 mice. Brain Res Bull 56 : 537 – 543 Google Scholar Crossref Search ADS PubMed WorldCat 25 Rudski JM , Grace M , Kuskowski MA , Billington CJ , Levine AS 1996 Behavioral effects of naloxone on neuropeptide Y-induced feeding. Pharmacol Biochem Behav 54 : 771 – 777 Google Scholar Crossref Search ADS PubMed WorldCat 26 Thornhill JA , Taylor B , Marshall W , Parent K 1982 Central, as well as peripheral naloxone administration suppresses feeding in food-deprived Sprague-Dawley and genetically obese (Zucker) rats. Physiol Behav 29 : 841 – 846 Google Scholar Crossref Search ADS PubMed WorldCat 27 Levine AS , Weldon DT , Grace M , Cleary JP , Billington CJ 1995 Naloxone blocks that portion of feeding driven by sweet taste in food-restricted rats . Am J Physiol 268 : R248 – R252 Google Scholar PubMed WorldCat 28 Gee CE , Chen CL , Roberts JL , Thompson R , Watson SJ 1983 Identification of proopiomelanocortin neurones in rat hypothalamus by in situ cDNA-mRNA hybridization. Nature 306 : 374 – 376 Google Scholar Crossref Search ADS PubMed WorldCat 29 Banks WA , Kastin AJ 1990 Peptide transport systems for opiates across the blood-brain barrier . Am J Physiol 259 : E1 – E10 Google Scholar PubMed WorldCat 30 Levine AS , Billington CJ 1989 Opioids. Are they regulators of feeding? Ann NY Acad Sci 575 : 209 – 219 Google Scholar Crossref Search ADS PubMed WorldCat 31 de Zwaan M , Mitchell JE 1992 Opiate antagonists and eating behavior in humans: a review. J Clin Pharmacol 32 : 1060 – 1072 Google Scholar PubMed WorldCat 32 Hayward MD , Pintar JE , Low MJ 2002 Selective reward deficit in mice lacking β-endorphin and enkephalin. J Neurosci 22 : 8251 – 8258 Google Scholar PubMed WorldCat 33 Silva RM , Hadjimarkou MM , Rossi GC , Pasternak GW , Bodnar RJ 2001 Beta-endorphin-induced feeding: pharmacological characterization using selective opioid antagonists and antisense probes in rats. J Pharmacol Exp Ther 297 : 590 – 596 Google Scholar PubMed WorldCat 34 Mogil JS , Grisel JE , Hayward MD , Bales JR , Rubinstein M , Belknap JK , Low MJ 2000 Disparate spinal and supraspinal opioid antinociceptive responses in β-endorphin-deficient mutant mice. Neuroscience 101 : 709 – 717 Google Scholar Crossref Search ADS PubMed WorldCat 35 Slugg RM , Hayward MD , Ronnekleiv OK , Low MJ , Kelly MJ 2000 Effect of the μ-opioid agonist DAMGO on medial basal hypothalamic neurons in β-endorphin knockout mice. Neuroendocrinology 72 : 208 – 217 Google Scholar Crossref Search ADS PubMed WorldCat 36 Govoni S , Yang HY 1981 Sex differences in the content of β-endorphin and enkephalin-like peptides in the pituitary of obese (ob/ob) mice. J Neurochem 36 : 1829 – 1833 Google Scholar Crossref Search ADS PubMed WorldCat 37 Krzanowska EK , Bodnar RJ 2000 Analysis of sex and gonadectomy differences in β-endorphin antinociception elicited from the ventrolateral periaqueductal gray in rats. Eur J Pharmacol 392 : 157 – 161 Google Scholar Crossref Search ADS PubMed WorldCat 38 Krzanowska E , Bodnar RJ 2000 Sex differences in locomotor activity following β-endorphin in the ventrolateral periaqueductal gray. Physiol Behav 68 : 595 – 598 Google Scholar Crossref Search ADS PubMed WorldCat 39 Islam AK , Beczkowska IW , Bodnar RJ 1993 Interactions among aging, gender, and gonadectomy effects upon naloxone hypophagia in rats. Physiol Behav 54 : 981 – 992 Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2003 by The Endocrine Society
TI - A Role for the Endogenous Opioid β-Endorphin in Energy Homeostasis
JF - Endocrinology
DO - 10.1210/en.2002-221096
DA - 2003-05-01
UR - https://www.deepdyve.com/lp/oxford-university-press/a-role-for-the-endogenous-opioid-endorphin-in-energy-homeostasis-pjtNegUwrA
SP - 1753
VL - 144
IS - 5
DP - DeepDyve
ER -