Disruption of Gpr45 causes reduced hypothalamic POMC expression and obesity

Abstract

A rise in the occurrence of obesity has driven exploration of its underlying genetic basis and potential targets for intervention. GWAS studies have identified obesity susceptibility pathways involving several neuropeptides that control energy homeostasis, suggesting that variations in the genes that regulate food intake and energy expenditure may contribute to obesity. In this study, we identified 5 additional obesity loci, including a neuronal orphan GPCR called Gpr45, in a forward genetic screen of mutant mice generated by piggyBac insertional mutagenesis. Disruption of Gpr45 led to increased adiposity at the time of weaning and increases in body mass, fat content, glucose intolerance, and hepatic steatosis with advancing age. Mice with disruptions in Gpr45 also displayed a reduction in expression of the metabolic regulator POMC and less energy expenditure prior to the onset of obesity. Mechanistically, we determined that GPR45 regulates POMC expression via the JAK/STAT pathway in a cell-autonomous manner. Consistent with this finding, intraventricular administration of melanotan-2, an analog of the POMC derivative α-MSH, suppressed adult obesity in Gpr45 mutants. These results reveal that GPR45 is a regulator of POMC signaling and energy expenditure, which suggests that it may be a potential intervention target to combat obesity.

Figure 1. A pilot genetic screen for obese mutations.

(A) The screen pipeline. (B) Body weight distribution of screened mutants. Numbers of lines with homozygotes/heterozygotes (hom/het) body weight ratio in a given range are illustrated as red dots. The histogram data were normalized by Gaussian distribution and plotted as an overlay (blue line). Hom/het body weight ratios of overweight mutants used in the secondary screen are also shown on the overlay in red (obese or overweight) or black (nonobese or nonoverweight). (C) Body weight (BW), fat/lean ratio (F/L), fasted plasma glucose (FPG) and the area under the glucose tolerance test curve (AUC) for obese and overweight lines are compared with those of the wild-type littermates and shown as a heat map. Red, significantly high; blue, significantly low; no color, no significant difference. (D) Histogram presentations of fold changes in BW, F/L, FPG, and AUC of both 12-week-old mutants and their wild-type littermates, the values of which were arbitrarily defined as 1. Asterisks indicate statistically significant differences calculated with raw numbers. Empty bars, +/+ (n ≥ 3); filled bars, PB/PB (n ≥ 3). All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test.

Figure 2. Disruption of Gpr45 causes obesity.

(A) Photographs of three 1-year-old littermates of indicated Gpr45 genotypes. (B) Cartoon illustration of the position of 2 independent PB insertions in Gpr45. Both insertions are mapped in the only intron of Gpr45, upstream of its coding exon (E2). The filled black box indicates the coding sequence. Arrowheads indicate the position of primers for real-time RT-PCR analysis. (C) Real-time RT-PCR analysis showing decreased Gpr45 expression in brains of P5 mutant mice. Empty bars, +/+ (n = 5); dashed bars, PB1/+ (n = 4); filled bars, PB1/PB1 (n = 4). Wild-type data served as the statistical controls. The bar patterns of genotypes also applies to D, G, H and I. (D) Average liver mass of 6-month-old females (+/+, n = 8; PB1/PB1, n = 9) and males (+/+, n = 4; PB1/PB1, n = 4). (E) Representative Oil red O and hematoxylin staining showing hepatic steatosis in a 6-month-old male mouse. Scale bar: 0.05 mm. (F) Body weight curves of female (+/+, n = 15; PB1/+, n = 22; PB1/PB1, n = 16) and male (+/+, n = 10; PB1/+, n = 16; PB1/PB1, n = 14) littermates. Wild-type data served as the statistical controls. (G and H) Average fat/lean ratio (G) and lean mass (H) of Gpr45 littermates (n ≥ 10 for each group at each time point). (I) Average fat pad mass of 6-month-old female (+/+, n = 8; PB1/PB1, n = 9) and male (+/+, n = 4; PB1/PB1, n = 4) mice. gWAT, perigonadal fat; rWAT, retroperitoneal fat; BAT, interscapular brown fat. All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01 and ***P < 0.001 by Student’s t test; ^#P < 0.05 and ^###P < 0.001, PB1/+ versus +/+ in F.

Figure 3. Disruption of Gpr45 impairs hormone and glucose homeostasis.

(A and D) ELISA analysis revealed elevated plasma leptin levels in both 12-week-old female (A) and male (D) mutants. (B and E) ELISA analysis revealed elevated plasma insulin levels in both 12-week-old female (B) and male (E) mutants. At least 4 mice were used for each genotype. (C and F) Insulin tolerance test (ITT) results of 15-week-old female (C, +/+, n = 15, 267.6 ± 3.9 U insulin; PB1/+, n = 21, 273.1 ± 4.9 U insulin; PB1/PB1, n = 15, 375.3 ± 11.1 U insulin) and male (F, +/+, n = 8, 345.3 ± 3.5 U insulin; PB1/+, n = 10, 350.9 ± 3.3 U insulin; PB1/PB1, n = 12, 441.7 ± 13.5 U insulin) mice. Wild-type data were used as the statistical control. (G and H) Glucose tolerance test (GTT) results of 12-week-old female (G, +/+, n = 12; PB1/+, n = 12; PB1/PB1, n = 12) and male (H, +/+, n = 8; PB1/+, n = 8; PB1/PB1, n = 8) mice. Wild-type data were used as the statistical control. (I) Glucose infusion rate (GIR) determined by hyperinsulinemic-euglycemic clamp on 15-week-old male mice (n = 5). All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test; ^#P < 0.05, PB1/+ versus +/+ in F.

Figure 4. Obesity is the primary defect in Gpr45 mutants.

(A) Daily fat/lean ratio of female (+/+, n = 9; PB1/+, n = 18; PB1/PB1, n = 8) and male (+/+, n = 9; PB1/+, n = 18; PB1/PB1, n = 8) mice revealed early-onset obesity before weaning. Data from wild-type mice were used as the control for the P value calculation. (B) Bar graphs showing similar plasma leptin levels found in 2-week-old mutant and wild-type mice. At least 7 female or 6 male mice were analyzed for each genotype. (C) Bar graphs showing higher plasma leptin levels in 4-week-old mutant mice than those in wild-type littermates. Number of mice: female, +/+, 7; PB1/+, 11; PB1/PB1, 11; male +/+, 7; PB1/+, 8; PB1/PB1, 6. (D) Oil red O and hematoxylin staining detecting no hepatic steatosis in a 4-week-old male mouse. Scale bar: 0.05 mm. (E) Bar graphs showing normal plasma insulin levels in 4-week-old mutant mice. Number of mice: female, +/+, n = 7; PB1/+, n = 11; PB1/PB1, n = 10; male, +/+, n = 5; PB1/+, n = 4; PB1/PB1, n = 6. (F) Bar graphs showing no difference in fasted plasma glucose (FPG) levels of 9-week-old female (+/+, n = 5; PB1/PB1, n = 4) and male (+/+, n = 5; PB1/PB1, n = 4) mice. All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test; ^###P < 0.001, PB1/+ versus +/+ in A.

Figure 5. Disruption of Gpr45 reduces energy expenditure.

(A) ANCOVA-estimated daily food intake of female (+/+, n = 8; PB1/PB1, n = 7) and male (+/+, n = 9; PB1/PB1, n = 6) mice during the ages between P21 and P33. Bar graphs showing similar daily food intake in mutants estimated at a common body weight of 12.58 g and 15.85 g for female and male mice, respectively. (B) ANCOVA-estimated 24-hour total energy expenditure (EE) of female (+/+, n = 22; PB1/PB1, n = 22) and male (+/+, n = 22; PB1/PB1, n = 22) mice during the ages between P21 and P33 at room temperature (22°C). Bar graphs showing decreased EE in mutants estimated at a common lean mass of 9.76 g and 10.155 g for females and males, respectively. (C) Bar graphs showing reduced 24-hour travel distance but extended sleeping time in P21 female (+/+, n = 6; PB1/PB1, n = 7) and male (+/+, n = 8; PB1/PB1, n = 6) mutants. m, distance traveled in meters. (D) Body temperatures of P14 female (+/+, n = 6; PB1/PB1, n = 6) and male (+/+, n = 22; PB1/PB1, n = 22) mice after cold exposure (4°C). All data are shown as the mean ± SEM. ***P < 0.001 by ANCOVA analysis (A and B). *P < 0.05 and **P < 0.01 by Student’s t test (C and D).

Figure 6. Disruption of Gpr45 decreases hypothalamic POMC expression and spontaneous firing rate of POMC neurons.

(A) Results of RT-PCR analysis showing the presence of Gpr45 mRNA in neuronal tissues and the testes of adult mice. (B) Real-time RT-PCR analysis revealed decreased Pomc, but similar expression of other hypothalamic neuropeptides in P14 mice (n ≥ 4 for each genotype). Expression of Gapdh serves as the internal control to calculate relative expression levels. (C) Western blots show decreased POMC levels in hypothalami of P14 mice (n = 3 for each genotype). A quantitative comparison is listed below. Expression of GAPDH serves as the internal control to calculate relative expression levels. (D) Bar graphs showing significantly reduced Gpr45 and Pomc expression in primary hypothalamic cells from the mutants. Summarized real-time RT-PCR results from 3 independent wells are shown, with Gapdh serving as the internal control. (E) Knockdown of Gpr45 in primary hypothalamic cells caused down regulation of Pomc. Real-time RT-PCR data were collected with the same procedure in (D), except for the use of 5 wells in each group. (F) Representative recordings of resting membrane potential from wild-type (n = 16 from 6 mice) and PB1/PB1 (n = 14 from 3 mice) animals. (G and H) Statistics for resting membrane potential (G) and spontaneous firing rate (H) in these recordings. All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test.

Figure 7. GPR45 regulates POMC for its role in energy expenditure.

(A and B) Western blots shows decreased phosphorylation of JAK2 and STAT3 in primary mutant hypothalamic cells (A) and hypothalami of P14 mutants (B). Quantitative comparisons are shown on the right, with total JAK2 or STAT3 protein as the internal control. n = 3 for each genotype. (C and D) Physical interaction between JAK2 and GPR45. Immunoprecipitation (C) by anti-JAK2 antibody showing that GPR45 interacts with JAK2 in the brain of adult HA-GPR45 transgenic mice. Anti-HA antibody was used to recognize HA-tagged GPR45. In vitro binding assay (D) shows that JAK2 interacts with the carboxyl terminus (GPR45-C tail), but not the third intracellular loop of GPR45 (GPR45-i3). Anti-GST antibody was used to recognize GST-tagged GPR45 fragments. (E) Chromatin immunoprecipitation by anti-STAT3 antibody revealed reduced STAT3 recruitment to the Pomc promoter in hypothalami of P14 mutants (n = 4). (F and H) Intra-third-ventrical administration of MTII reduced both bodyweight (F) and fat mass (H) in adult male mutants. gWAT, perigonadal fat; rWAT, retroperitoneal fat; BAT, interscapular brown fat. (G) MTII treatment reduced food intake in Gpr45 mutants and their wild-type littermates. Five mice were used for each group. Data from saline-treated wild-type mice serve as the baseline for statistics. All data are shown as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test; ^#P < 0.05, ^##P < 0.01, and ^###P < 0.001, MTII-treated PB1/PB1 vs. saline-treated +/+; ^§P < 0.05 and ^§§P < 0.01, MTII-treated +/+ vs. saline-treated +/+ in F and G.