Global GPR75 deficiency protects against diet-induced obesity through central and peripheral mechanisms Authors: Xiaoyun Cao, Vanna Chhay, Thi Tun Thi, Hui Li, Yu Shi, Yifan Wang, Yilin Min, Renhua Song, Kai Yan Cheng, Yang Yang, Justin Jong-Leong Wong, Haojie Yu Published: 2026 ABSTRACT Loss-of-function variants of GPR75 are associated with protection against obesity in humans; however, the mechanisms through which GPR75 regulates whole-body energy homeostasis remain incompletely understood. Here, we generated Gpr75 global knockout (GKO) mice and performed comprehensive metabolic phenotyping under chow and high-fat diet (HFD) conditions. Despite high expression of GPR75 in the brain, male GKO mice exhibited unchanged food intake during the period when body weights began to diverge. Instead, GPR75 deficiency resulted in reduced whole-body energy expenditure and an elevated respiratory exchange ratio, indicating decreased fat utilization. Notably, male, but not female, GKO mice displayed markedly impaired intestinal lipid absorption, reduced chylomicron output, and increased fecal lipid excretion, consistent with limited dietary lipid assimilation under HFD feeding.To define the tissue-specific contributions of GPR75, we next examined hepatocyte-, intestinal epithelial-, endothelial/hematopoietic-, and AgRP neuron-specific Gpr75 knockout mice. Deletion of Gpr75 in hepatocytes (Alb-Cre), intestinal epithelium (Vil-Cre), or endothelial/hematopoietic lineages (Tie2-Cre) did not reproduce the pronounced protection against HFD-induced weight gain observed in GKO mice. In contrast, AgRP neuron-specific deletion reduced body weight by approximately 10%–12% in male and female mice and decreased white adipose depot mass by 27%–50%, establishing AgRP neurons as a specific neuronal population through which GPR75 contributes to body-weight regulation. However, the magnitude of this phenotype remained substantially smaller than that observed in global GPR75-deicient mice.Together, our findings demonstrate that GPR75 regulates diet-induced obesity through both central and peripheral mechanisms. AgRP-neuronal GPR75 contributes to body-weight regulation in both sexes, whereas global GPR75 deficiency additionally impairs intestinal lipid assimilation in males through a mechanism not recapitulated by deletion in the tested peripheral cell populations. These complementary effects likely contribute to the pronounced resistance to diet-induced obesity and hepatic steatosis observed in global GPR75-deficient mice and further support GPR75 as a therapeutic target for obesity and associated metabolic liver disease. Introduction Obesity, characterized by excessive fat accumulation, is a major global public health issue, affecting approximately 15% of the world's population [1]. As a significant risk factor for chronic diseases such as type 2 diabetes (T2D), cardiovascular disease (CVD), metabolic dysfunction-associated steatotic liver disease (MASLD), and cancer, among others, there is an urgent need for innovative therapeutic strategies to combat the rising prevalence of obesity across the world. GPR75 is an orphan G protein-coupled receptor that belongs to Class A (Rhodopsin) family. Large-scale human exome sequencing studies have uncovered that individuals carrying GPR75 loss of function (LOF) variants are protected against obesity [[2], [3], [4]]. In line with human genetic findings, global depletion of GPR75 in mice protects against diet-induced obesity and associated metabolic syndromes, including insulin resistance and MASLD [2,[4], [5], [6], [7], [8], [9], [10]]. However, despite this strong genetic and physiological evidence, the precise tissues and mechanisms through which GPR75 regulates whole-body energy homeostasis remain incompletely understood. GPR75 is highly expressed in the central nervous system, including agouti-related peptide (AgRP) and neuropeptide Y (Npy) neurons, which play critical roles in appetite regulation and energy balance [4]. Recent work using Nestin-Cre-mediated neuronal deletion demonstrated that neuron-specific loss of Gpr75 is sufficient to protect against HFD-induced obesity, primarily through reduced food intake and modest increases in energy expenditure [8]. However, under prolonged high-fat feeding, these Nestin-Cre-mediated knockout mice exhibited depot-specific adipose tissue remodeling, characterized by unchanged subcutaneous fat mass but significantly increased visceral fat mass and largely preserved adipocyte size in both depots [8]. These phenotypes differ substantially from those observed in global knockout models, which display marked reductions in overall adiposity and adipocyte size. Together, these findings indicate that neuronal GPR75 signaling plays an important role in energy balance, although the specific neuronal populations mediating these effects remain incompletely defined. Moreover, neuronal GPR75 deficiency may not fully account for the systemic metabolic adaptations observed in global knockout models, suggesting additional contributions from non-neuronal tissues and/or coordinated inter-organ mechanisms. In addition to the brain, GPR75 is expressed in peripheral tissues, including the intestine [4], [7]. The intestine plays a central role in nutrient absorption and chylomicron production, processes essential for dietary lipid assimilation. Whether peripheral GPR75 signaling contributes to energy extraction and adiposity has not been clearly defined. In this study, we performed a comprehensive physiological characterization of CRISPR-generated Gpr75 global knockout (GKO) mice under chow and high-fat diet conditions. We found that, despite high brain expression, global GPR75 deficiency did not alter food intake during the early divergence of body weight. Instead, GKO mice exhibited reduced energy expenditure, increased respiratory exchange ratio (RER), markedly reduced intestinal chylomicron output, and increased fecal lipid excretion, indicating impaired dietary lipid assimilation under HFD feeding. To determine whether this phenotype could be attributed to a specific tissue, we generated multiple tissue-specific knockout models targeting hepatocytes (Alb-Cre), intestinal epithelial cells (Vil-Cre), endothelial/hematopoietic lineages (Tie2-Cre), and AgRP neurons (AgRP-Cre). Deletion of Gpr75 in hepatocytes, intestinal epithelial cells, or endothelial/hematopoietic lineages did not alter HFD-induced body weight gain. In contrast, AgRP neuron-specific deletion reduced body weight by approximately 10–12% in both male and female mice, but this effect remained substantially smaller than the reduction observed in global knockout mice. Collectively, these findings suggest that while neuronal GPR75 contributes to energy balance, the full protection against diet-induced obesity observed in global GPR75 deficiency likely reflects coordinated, multi-organ regulation rather than a single dominant tissue-autonomous mechanism. Importantly, our study identifies AgRP neurons as a specific neuronal population through which GPR75 regulates body weight and adiposity, while supporting a broader role for GPR75 as a systemic regulator acting through integrated central and peripheral mechanisms. Result GPR75 promotes obesity and associated metabolic syndromes To elucidate the physiological and molecular function of GPR75 with respect to systemic energy homeostasis, we employed CRISPR/Cas9 system to generate whole-body knockout (Gpr75−/−) mouse model (Fig. S1A and S1B). The deletion of Gpr75 was confirmed by measuring mRNA levels across multiple tissues (Fig. S1C). Gpr75−/− mice were viable, fertile and normal in appearance without noticeable developmental defects. Mating between Gpr75+/− mice produced offspring in the expected genotype distribution ratio of 1:2:1. We next measured the body weight of male and female Gpr75−/− mice and their control littermates (Gpr75+/+) fed a chow diet. Male and female Gpr75−/− mice show similar body weight as Gpr75+/+ littermates upon weaning (Figure 1A,B). Of note, male Gpr75−/− mice exhibit a reduced body weight starting at five weeks of age compared to their control littermates (Figure 1A). This lean phenotype persisted into adulthood, stabilizing at a weight difference of approximately 10–12% (Figure 1A). Consistently, chow-fed GPR75-deficient mice exhibited significantly reduced weights of inguinal (iWAT) and epididymal (eWAT) adipose tissue, along with a trend of reduced liver weight, compared to their wild-type controls (Figure 1C). Furthermore, although significant body weight difference was not observed in female mice fed chow, weight of eWAT in GPR75-deficient mice were significantly less than those in wild-type littermates (Figure 1B,D).Figure 1GPR75 promotes obesity and related metabolic syndromes. Body weight curve of Gpr75−/− and Gpr75+/+(A) male mice (n = 12–14 mice per group) and (B) female mice fed normal chow diet (n = −9-10 mice per group). Quantification of liver and adipose tissues' weight from Gpr75−/− and Gpr75+/+(C) male mice and (D) female mice fed normal chow diet. “Chow” refers to the Teklad Global 18% Protein Rodent Diet (Diet 2918, ENVIGO), as defined in the Methods (E) Body weight curve of male Gpr75−/− and Gpr75+/+ mice fed HFD for 16 weeks (n = 8 mice per group). Data represented as mean ± SEM. (F) Representative images and (G) weight of Liver, eWAT, ingWAT and iBAT from male Gpr75−/− and Gpr75+/+ mice fed HFD for 16 weeks. (H) Body weight curve of female Gpr75−/− and Gpr75+/+ mice fed HFD for 18 weeks (n = 9 mice per group). Data represented as mean ± SEM. (I) Representative images and (J) weight of Liver, eWAT, ingWAT and iBAT from female Gpr75−/− and Gpr75+/+ mice fed HFD for 18 weeks. (K) Representative H&E staining of eWAT, ingWAT, and iBAT from male Gpr75−/− and Gpr75+/+ littermates on HFD for 16 weeks. Scale bar, 100 μm for eWAT and ingWAT, and 50 μm for BAT. (L) Quantification of adipocyte size in eWAT and ingWAT. AdipoSoft software is used to quantify 4 regions with an average of 80–100 adipocytes per region. (M) Fasting plasma TG, (N) fasting plasma TC and (O) fasting plasma FFA measurement in male Gpr75−/− and Gpr75+/+ mice fed normal chow and HFD. (P) Representative H&E staining of liver from male Gpr75−/− and Gpr75+/+ control littermates on HFD for 16 weeks. Scale bar, 50 μm. (Q) Hepatic TG and (R) hepatic TC levels of male and female Gpr75−/− and Gpr75+/+ control littermates on HFD for 16 weeks and 18 weeks respectively. (S) Plasma ALT activity from male and female Gpr75−/− and Gpr75+/+ mice fed on HFD for 16 weeks and 18 weeks respectively. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. GPR75 promotes obesity and related metabolic syndromes. Body weight curve of Gpr75−/− and Gpr75+/+(A) male mice (n = 12–14 mice per group) and (B) female mice fed normal chow diet (n = −9-10 mice per group). Quantification of liver and adipose tissues' weight from Gpr75−/− and Gpr75+/+(C) male mice and (D) female mice fed normal chow diet. “Chow” refers to the Teklad Global 18% Protein Rodent Diet (Diet 2918, ENVIGO), as defined in the Methods (E) Body weight curve of male Gpr75−/− and Gpr75+/+ mice fed HFD for 16 weeks (n = 8 mice per group). Data represented as mean ± SEM. (F) Representative images and (G) weight of Liver, eWAT, ingWAT and iBAT from male Gpr75−/− and Gpr75+/+ mice fed HFD for 16 weeks. (H) Body weight curve of female Gpr75−/− and Gpr75+/+ mice fed HFD for 18 weeks (n = 9 mice per group). Data represented as mean ± SEM. (I) Representative images and (J) weight of Liver, eWAT, ingWAT and iBAT from female Gpr75−/− and Gpr75+/+ mice fed HFD for 18 weeks. (K) Representative H&E staining of eWAT, ingWAT, and iBAT from male Gpr75−/− and Gpr75+/+ littermates on HFD for 16 weeks. Scale bar, 100 μm for eWAT and ingWAT, and 50 μm for BAT. (L) Quantification of adipocyte size in eWAT and ingWAT. AdipoSoft software is used to quantify 4 regions with an average of 80–100 adipocytes per region. (M) Fasting plasma TG, (N) fasting plasma TC and (O) fasting plasma FFA measurement in male Gpr75−/− and Gpr75+/+ mice fed normal chow and HFD. (P) Representative H&E staining of liver from male Gpr75−/− and Gpr75+/+ control littermates on HFD for 16 weeks. Scale bar, 50 μm. (Q) Hepatic TG and (R) hepatic TC levels of male and female Gpr75−/− and Gpr75+/+ control littermates on HFD for 16 weeks and 18 weeks respectively. (S) Plasma ALT activity from male and female Gpr75−/− and Gpr75+/+ mice fed on HFD for 16 weeks and 18 weeks respectively. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. Once start HFD (60% Kcal from fat) feeding, the body weight of male Gpr75+/+ and Gpr75−/− littermates began to diverge rapidly while female mice exhibited similar body weights during the first six weeks of HFD feeding (Figure 1E,H). After HFD feeding for 16 weeks, male Gpr75−/− mice display 29% less body weight compared to Gpr75+/+ mice (33.4 ± 1.4g vs 46.8 ± 5.0g) (Figure 1E). Furthermore, weights of liver, eWAT, ingWAT and BAT were profoundly reduced in male Gpr75−/− mice by 34%, 64%, 66%, and 43%, respectively, compared to control littermates (Figure 1F,G). We further performed histological analysis showing significantly smaller adipocytes in both eWAT and iWAT from GPR75-deficient mice compared to control littermates (Figure 1K,L). Moreover, there is substantially reduced fat accumulation in iBAT from Gpr75−/− mice compared to control littermates (Figure 1K). Similarly, female GPR75-deficient mice showed 19% less body weight than Gpr75+/+ littermates after HFD feeding for 16 weeks (Figure 1H), with significantly reduced eWAT and ingWAT weight, but not liver and BAT, by 57% and 44%, respectively, upon GPR75 depletion. (Figure 1I,J). We next examined whether GPR75 deficiency protect against obesity-associated metabolic complications. Plasma triglyceride (TG) and total cholesterol (TC) levels were significantly reduced in Gpr75−/− mice compared to Gpr75+/+ littermates fed HFD (Figure 1M,N, Fig. S1D and S1E). Of note, GPR75 depletion did not affect plasma free fatty acid (FFA) levels under either chow or HFD feeding at any time measured (Figure 1O). Female GPR75-deficient mice did not show changes in plasma TC and FFA levels but showed a trend toward reduced plasma TG levels compared control mice under HFD feeding for 18 weeks (Fig. S1F–S1H). In addition, overnight fasting glucose levels were similar between Gpr75+/+ and Gpr75−/− mice (Fig. S1I and S1J). Consistent with previous studies [4,7], depletion of GPR75 reduced liver TG contents by 58% and 47%, respectively, in male and female mice compared to their control littermates upon HFD feeding (Figure 1P,Q). Liver cholesterol levels were significantly reduced in female, but not male, mice upon GPR75 depletion (Figure 1R). However, plasma ALT levels were only found to be significantly reduced in male, but not in female, knockout mice compared to control littermates (Figure 1S). Collectively, these data suggest that depletion of GPR75 protects against diet-induced obesity and excessive fat accumulation in the liver. Male Gpr75 knockout mice display unchanged food intake, reduced energy expenditure, and an elevated respiratory exchange ratio The protective effect against HFD-induced fat accumulation in adipose tissue and liver is drastic upon GPR75 depletion. However, there is still debate about whether GPR75 regulates energy balance by modulating food intake or systemic energy expenditure [[4], [5], [6],[8], [9], [10]]. To address this question, we employed indirect calorimetry to evaluate the energy balance and substrate utilization in male Gpr75+/+ and Gpr75−/− littermates (10 weeks old) fed chow or HFD for 2 weeks. At this stage, their body weights just began to diverge rapidly, making it more likely to uncover the causal mechanism rather than adaptive effects (Figure 2A).Figure 2Male Gpr75 knockout mice display unchanged food intake, reduced energy expenditure, and an increased respiratory exchange ratio. (A) Diagrammatic representation of indirect calorimetry (n = 8 male mice per group). (B–E) Food intake in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). (F–I) Energy expenditure in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). (J–M) Respiratory exchange ratio (RER) in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). Graph and analysis were generated by CalR, a web application for indirect calorimetry analysis (https://calrapp.org/). Food intake and energy expenditure were analyzed using the CaIR GLM/ANCOVA framework, whereas RER was analyzed by one-way ANOVA. Data are presented as mean ± SEM. ∗P ≤ 0.05, ∗∗P ≤ 0.01. Male Gpr75 knockout mice display unchanged food intake, reduced energy expenditure, and an increased respiratory exchange ratio. (A) Diagrammatic representation of indirect calorimetry (n = 8 male mice per group). (B–E) Food intake in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). (F–I) Energy expenditure in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). (J–M) Respiratory exchange ratio (RER) in Gpr75−/− and Gpr75+/+ mice fed on CD (left) and HFD (right). Graph and analysis were generated by CalR, a web application for indirect calorimetry analysis (https://calrapp.org/). Food intake and energy expenditure were analyzed using the CaIR GLM/ANCOVA framework, whereas RER was analyzed by one-way ANOVA. Data are presented as mean ± SEM. ∗P ≤ 0.05, ∗∗P ≤ 0.01. To our surprise, we observed no differences in food intake under both chow and HFD feeding (Figure 2B–E), despite GPR75 is highly expressed across the brain [4,7]. Instead, we observed a significant reduction in both daytime and nighttime energy expenditure in the knockout mice compared to wild-type littermates under chow, with a trend toward reduced energy expenditure under HFD feeding (Figure 2F–I). Consistently, the body temperature of GPR75-deficient mice was significantly lower than control littermates upon overnight fasting (Fig. S2A). Importantly, the respiratory exchange ratio (RER) was significantly increased in Gpr75−/− mice, particularly during the light phase (Figure 2J-2M and Fig. S2B–S2I), indicating a shift toward greater carbohydrate utilization and reduced fat oxidation compared with Gpr75+/+ mice. Additionally, while total distance traveled was comparable between genotypes in chow-fed mice, Gpr75−/− mice showed a trend toward greater distance traveled at nighttime under HFD-fed conditions (Fig. S2J-S2Q). Interestingly, we also found that Gpr75−/− mice consumed significantly more water than their control littermates at nighttime under both chow and HFD feeding conditions (Fig. S2R-S2U). Taken together, given that GPR75-dificient mice exhibited unchanged food intake, reduced energy expenditure, and increased RER, we hypothesized that the lean phenotype of GPR75 knockout mice is likely caused by attenuated dietary fat absorption in the gastrointestinal (GI) tract. GPR75 deficiency significantly attenuates intestinal dietary fat absorption in male mice To test our hypothesis, we next examined intestinal absorption of dietary fat by conducting a lipid tolerance test, in which an olive oil bolus was administered to mice via oral gavage, followed by measuring plasma TG levels over the course of 2 h (Figure 3A). In 7-week-old chow-fed male mice, plasma TG increased at a similar rate between Gpr75+/+ and Gpr75−/− mice after olive oil gavage (Fig. S3A). It's well established that 10%–12% of intestinal TG output after a meal is derived from lipid stores accumulated from previous dietary fat intake [11]. These stored lipids are retained in the intestine within intracellular compartment, such as enterocytes, or extracellular compartment, such as the lamina propria and lacteals of the mesenteric lymphatic system, and are released in response to a second meal or other stimulatory factors [12]. To distinguish between stored lipids and newly absorbed fatty acids from olive oil bolus, we added BODIPY-labelled fatty acid (BODIPY™ 500/510 C1, C12 (4,4-Difluoro-5-Methyl-4-Bora-3a,4a-Diaza-s-Indacene-3-Dodecanoic Acid) to the olive oil for gavage, allowing us to directly quantify intestinal fatty acids absorption by measuring the BODIPY intensity in the plasma. Consistently, there was no difference in plasma BODIPY signals in the young male mice between with and without GPR75 (Fig. S3B). When the chow-fed mice were older (12 weeks old), we observed a trend of reduction in both blood TG and BODIPY levels in GPR75-deficient mice compared to control littermates (Fig. S3C–S3E).Figure 3GPR75 depletion attenuates intestinal fat absorption and enhances fecal lipid excretion in male mice. (A) Schematic of oral gavage in Gpr75−/− and Gpr75+/+ male mice with 200 μl olive oil with or without 30 μg C1-BODIPY 500/510 C12 (fluorescent fatty acid analog; Life Technologies #D3823). Blood collected via tail vein before gavage (T0), 60min (T60) and 120min (T120) after gavage. (B) Plasma TG levels after olive oil gavage in Gpr75−/− and Gpr75+/+ mice fed HFD for 8 weeks (n = 8 male mice per group). Data represented as mean ± SEM. (C) Change in Plasma BODIPY (Δ plasma BODIPY) 2h after olive oil oral gavage in Gpr75−/− and Gpr75+/+ mice fed HFD (n = 6 male mice per group). (D) Western blotting analyses of Plasma APOB48 and APOB100 of Gpr75−/− and Gpr75+/+ mice fed HFD (n = 6 male mice per group). (E) Relative quantification of plasma APOB48 and (F) plasma APOB100 by ImageJ of male Gpr75−/− and Gpr75+/+ mice fed HFD. (G) Daily weight and (H) representative images of feces collected from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks (n = 8 mice per group). (I) Number of fecal pellets of male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (J) Total fecal TG, (K) TC, and (L) FFA levels from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (M) Total fecal protein level from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (N) Representative images and (O) quantification of small intestine (SI) length and (P) colon length from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 and 8 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. GPR75 depletion attenuates intestinal fat absorption and enhances fecal lipid excretion in male mice. (A) Schematic of oral gavage in Gpr75−/− and Gpr75+/+ male mice with 200 μl olive oil with or without 30 μg C1-BODIPY 500/510 C12 (fluorescent fatty acid analog; Life Technologies #D3823). Blood collected via tail vein before gavage (T0), 60min (T60) and 120min (T120) after gavage. (B) Plasma TG levels after olive oil gavage in Gpr75−/− and Gpr75+/+ mice fed HFD for 8 weeks (n = 8 male mice per group). Data represented as mean ± SEM. (C) Change in Plasma BODIPY (Δ plasma BODIPY) 2h after olive oil oral gavage in Gpr75−/− and Gpr75+/+ mice fed HFD (n = 6 male mice per group). (D) Western blotting analyses of Plasma APOB48 and APOB100 of Gpr75−/− and Gpr75+/+ mice fed HFD (n = 6 male mice per group). (E) Relative quantification of plasma APOB48 and (F) plasma APOB100 by ImageJ of male Gpr75−/− and Gpr75+/+ mice fed HFD. (G) Daily weight and (H) representative images of feces collected from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks (n = 8 mice per group). (I) Number of fecal pellets of male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (J) Total fecal TG, (K) TC, and (L) FFA levels from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (M) Total fecal protein level from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 weeks. (N) Representative images and (O) quantification of small intestine (SI) length and (P) colon length from male Gpr75−/− and Gpr75+/+ mice fed HFD for 4 and 8 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. After HFD feeding, we found that Gpr75−/− mice displayed a substantially attenuated increase of plasma TG levels compared to control littermates over the 2-h time course following olive oil gavage (Figure 3B). Furthermore, plasma BODIPY measurements revealed significantly reduced plasma BODIPY intensities in Gpr75−/− mice 2 h post-gavage, suggesting that the absorption of luminal free fatty acids was impaired upon GPR75 depletion in male mice (Figure 3C). In contrast, no significant difference was observed in female mice under the same conditions (data not shown). Consistent with the reduced intestinal chylomicron-TG output, western blot analysis showed significantly reduced plasma ApoB48, but not Apob100, in Gpr75−/− mice compared to Gpr75+/+ littermates 2 h after olive oil gavage (Figure 3D–F). Of note, we also conducted lipid tolerance test at early time point of HFD feeding (4 weeks), and GPR75 depletion consistently led to an attenuation of intestinal chylomicron-TG output compared to wild-type control littermates (Fig. S3F). The reduced plasma TG and BODIPY levels in the lipid tolerance assay might be attributed to increased peripheral LPL activities due to absence of GPR75. To explore this possibility, we measured post-heparin peripheral LPL activities and found no significant difference between GPR75-deficient mice and control littermates (Fig. S3G). Taken together, these data suggest that GPR75 is likely to promote dietary fat absorption and chylomicron output in the intestine. Depletion of GPR75 results in increased fecal lipid excretion in male mice Given the unaffected food intake and reduced fat absorption upon GPR75 depletion, we next investigated whether fecal lipid excretion was increased in male Gpr75−/− mice compared to Gpr75+/+ littermates. We collected feces every 24 hours for three consecutive days from individually housed male mouse after four weeks of HFD feeding, a stage where fat mass expands rapidly. Our data collected from two independent cohorts showed that total number and weight of daily expelled feces were significantly increased by 26% and 32%, respectively, in GPR75-deficient mice compared to control littermates (Figure 3G–I). Water content in the feces did not appear to be affected, as the dried feces weight was significantly increased by 29% upon GPR75 depletion (Fig. S3H). Further characterization of the collected feces revealed that the width and length of individual feces were similar between the two groups, indicating sufficient transit and processing along the gastrointestinal tract (Fig. S3I). Next, we measured fecal lipid (TG, TC, and FFA) contents and found that relative lipid content (mg lipid/g feces) remains similar between Gpr75+/+ and Gpr75−/− mice (Fig. S3J-S3L). However, absolute excreted TG, TC, and FFA amount every 24 hours (mg lipid/day) were significantly higher in mice lacking GPR75 compared to their control littermates (Figure 3J-3L). Additionally, no differences in TG:FFA ratio was observed between Gpr75+/+ and Gpr75−/− mice, suggesting that the hydrolysis of dietary TG via pancreatic lipase was not significantly affected (Fig. S3M). Of note, we assessed protein/peptide levels in the feces and found that absolute excreted protein/peptide levels every 24 hours (mg/day) were significantly higher in mice lacking GPR75 compared to their control littermates, while the relative content (mg/g feces) remains similar (Figure 3M and Fig. S3N). These data suggest that GPR75 deficiency in male mice might result in a systemic defect in intestinal lipid and protein absorption, which likely contributes to reduced whole-body energy expenditure and an increased utilization of carbohydrates as an energy source. Given the substantially increased number of stools expelled per mouse during the 24 h of collection, we next asked if the GI transit time is accelerated upon GPR75 deficiency. To measure whole GI transit, we gavage male mice with carmine red, a non-absorbable red dye with no nutritional value, after which we recorded the time for the first appearance of feces with red color. We observed no significant difference in whole GI transit time between GPR75-deficient mice and control littermates, suggesting that GPR75 does not play a role in regulating GI motility (Fig. S3O). Overall, the major mechanism responsible for reduced fat mass and body weight in Gpr75−/− male mice fed HFD is likely due to attenuated dietary lipids absorption along with increased lipid fecal excretion. Depletion of GPR75 leads to shortened intestine length in male mice To interrogate if the differences in lipid absorption is caused by impaired structure of intestine, we performed histological analysis of intestine from male mice under HFD feeding for 4 weeks. H&E and alcian blue staining revealed comparable villi length, crypt depth, and goblet cell numbers per villi between Gpr75−/− mice and control littermates (Fig. S3R–S3V). Lacteal length also appears to be comparable between Gpr75−/−and Gpr75+/+ mice fed HFD (Fig. S3P and S3Q). Surprisingly, the total length of intestine, but not colon, was significantly reduced in male Gpr75−/− mice compared to Gpr75+/+ littermates fed with HFD (Figure 3N-3P). Overall, these data suggest that while depletion of GPR75 did not alter villus structure, it did result in a reduction in intestine length. This reduction might contribute to the lean phenotype of GPR75-deficient mice or could be secondary to decreased body weight, as intestine length in humans is correlated with body weight [13]. Single-cell RNA-Seq and histochemical profiling of GPR75 expression in the intestine Given the GI tract phenotype, we next profiled GPR75 mRNA expression in the intestine in both humans and mice. In addition to the brain where GPR75 expression is widely distributed across various regions, its expression is detected in the intestine [4,7]. We first interrogated where GPR75 is expressed within intestine by employing a single-cell RNA sequencing (RNA-seq) dataset that consists of human intestine (Duodenum, jejunum, and ileum) and colon [14]. Among 160371 intestine cells and 119740 colon cells we have analyzed, GPR75 was detected in 0.063% and 0.115%, respectively, of these cells in the dataset (Figure 4A). Gpr75 expression in the intestine and colon is heterogeneous, across a variety of regions and cell types including enterocytes, immature enterocytes, enteroendocrine cells, goblet cells, immune cells, stem cells, Paneth cells, transit amplifying cells (TA), myofibroblasts, endothelial cells, etc. (Figure 4B–D, Fig. S4A).Figure 4Single-cell RNA-Seq and histochemical profiling of GPR75 expression in the intestine. (A) The percentage of GPR75-positive cells across four healthy human donors (including both male and female donors) in different locations of intestinal tract. GPR75 expression in epithelial, stromal, and immune compartments of the human (B) small intestine and (C) human colon. (D) Violin plot showing GPR75 expression among cell types within the epithelial, stromal, and immune compartments of the human small intestine. (E) Expression of Gpr75, as detected by RNAScope (red), in jejunum of the small intestine. Scale bar, 100 μm. (F) Volcano plot depicting differentially expressed genes in the jejunum from Gpr75−/− and Gpr75+/+ male mice. Red dots represent genes upregulated in Gpr75−/− mice while blue dots represent downregulated genes. (G) Gene sets significantly enriched (FDR <0.25) in the comparison of jejunum from male Gpr75−/− mice and Gpr75+/+ mice using the gene ontology (GO) annotation, ordered by the increasing NES. Single-cell RNA-Seq and histochemical profiling of GPR75 expression in the intestine. (A) The percentage of GPR75-positive cells across four healthy human donors (including both male and female donors) in different locations of intestinal tract. GPR75 expression in epithelial, stromal, and immune compartments of the human (B) small intestine and (C) human colon. (D) Violin plot showing GPR75 expression among cell types within the epithelial, stromal, and immune compartments of the human small intestine. (E) Expression of Gpr75, as detected by RNAScope (red), in jejunum of the small intestine. Scale bar, 100 μm. (F) Volcano plot depicting differentially expressed genes in the jejunum from Gpr75−/− and Gpr75+/+ male mice. Red dots represent genes upregulated in Gpr75−/− mice while blue dots represent downregulated genes. (G) Gene sets significantly enriched (FDR <0.25) in the comparison of jejunum from male Gpr75−/− mice and Gpr75+/+ mice using the gene ontology (GO) annotation, ordered by the increasing NES. Due to great limitations of single-cell RNA-Seq in profiling mRNAs of low-expression GPCRs, we next performed in situ hybridization (RNAScope) to map Gpr75 mRNA expression within the intestine. In agreement with single cell RNA-Seq data, we found that GPR75 is widely distributed throughout the mucosa, submucosa, and muscularis externa, with the highest expression observed in epithelium, villus crypt, and laminar propria (Figure 4E). Taken together, these profiling suggest that Gpr75 is heterogeneously expressed and widely distributed in the intestine. Transcriptomic profiling of jejunum upon GPR75 depletion To further explore the mechanism of how GPR75 regulates intestinal chylomicron output and fecal lipid excretion, we performed transcriptome analysis of jejunum from male mice fed HFD for 4 weeks. In total 455 differentially expressed genes (217 upregulated vs 238 downregulated) were uncovered upon depletion of GPR75 (Figure 4F). Unbiased gene set enrichment analysis (GSEA) by using three different databases, namely KEGG pathway gene sets, GO gene sets, and reactome pathway gene sets, revealed that GPR75 deficiency led to upregulation of pathways related to mucosa immune response and defense response to bacterium (Figure 4G, Fig. S4B and S4C, Fig. S5A and S5B). In addition, cell–cell junction assembly, regulation of glucose transmembrane transport, ERK1 and ERK2 cascade, and potassium channel complex were significantly upregulated in intestine of GPR75-deficient mice compared to control littermates (Figure 4G, Fig. S4B and S4C, Fig. S5A and S5B). In contrast, pathways related to cell cycle, triglyceride metabolic process, proteinogenic amino acid biosynthetic process, peroxisome, and fatty acid catabolic process were among the top downregulated pathways in the absence of GPR75 (Figure 4G, Fig. S4B and S4C, Fig. S5A and S5B). To interrogate if the unregulated immune response is associated with impaired intestinal mucosa barrier integrity upon depletion of GPR75, we performed assay to examine intestine permeability by oral gavage of FITC-dextran in HFD-fed male mice. We found that GPR75-mice exhibited significantly reduced intestinal permeability compared to control mice after 8 weeks of HFD feeding (Fig. S4D). GPR75-deficient mice exhibit a significant change in microbiome composition Given the GI tract phenotypes observed in male GPR75-deficient mice, we next asked whether these changes also led to alterations in microbiome composition. 16S rRNA gene sequencing analysis of fecal samples from male mice fed a HFD for 4 weeks revealed a significant reduction in Actinobacteria and an increase in Bacteroidetes in GPR75-deficient mice compared to control mice (Figure 5A- 5D, Fig. S6). Additionally, GPR75 depletion showed a trend toward reduced Firmicutes, the most abundant phylum in the gut microbiome, compared to control littermates (Figure 5A–D). In both humans and mice, Bacteroidetes is negatively correlated with obesity and fat mass, while Firmicutes is positively correlated with these traits. Studies have demonstrated that microbiota transplants from western diet-induced obese mice to lean germ-free recipients resulted in substantially increased fat deposition compared to transplants from chow-fed lean donors [15,16]. This effect is attributed to enhanced fermentation and an increased capacity for energy harvest in the microbiome of obese mice, likely driven by the increased abundancy of Firmicutes [15,16]. Therefore, the observed microbial shift in GPR75-deficient mice is consistent with reduced energy harvest capacity and may represent either a contributing factor or a downstream adaptation to impaired dietary lipid assimilation.Figure 5Male GPR75-deficient mice exhibit a significant change in microbiome composition. (A) Phylum-level relative abundance in Gpr75+/+ and Gpr75−/− male mice, with percentages indicated for the major phyla. (B) Relative abundance of Actinobacteria, Bacteroidetes and Firmicutes. (C) Genus-level relative abundance in Gpr75+/+ and Gpr75−/− mice. (D) Genus-level changes in relative abundance between Gpr75+/+ and Gpr75−/− mice. Fold changes were calculated as log2 (mean relative abundance in Gpr75−/− mice/mean relative abundance in Gpr75+/+ mice). Blue and orange bars indicate decreased and increased relative abundance in Gpr75−/− mice, respectively (n = 8 mice per genotype). Data are presented as mean ± SD. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. Male GPR75-deficient mice exhibit a significant change in microbiome composition. (A) Phylum-level relative abundance in Gpr75+/+ and Gpr75−/− male mice, with percentages indicated for the major phyla. (B) Relative abundance of Actinobacteria, Bacteroidetes and Firmicutes. (C) Genus-level relative abundance in Gpr75+/+ and Gpr75−/− mice. (D) Genus-level changes in relative abundance between Gpr75+/+ and Gpr75−/− mice. Fold changes were calculated as log2 (mean relative abundance in Gpr75−/− mice/mean relative abundance in Gpr75+/+ mice). Blue and orange bars indicate decreased and increased relative abundance in Gpr75−/− mice, respectively (n = 8 mice per genotype). Data are presented as mean ± SD. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. Deletion of Gpr75 in selected peripheral tissues fails to recapitulate the global knockout phenotype While global Gpr75 deletion robustly protects against diet-induced obesity and is associated with impaired dietary lipid assimilation and altered gut microbiota, these findings do not establish whether the phenotype arises from GPR75 signaling within a specific tissue or from coordinated inter-organ mechanisms. Given the prominent gastrointestinal alterations observed in GPR75-deficient mice, we first asked whether deletion of Gpr75 in intestinal cell types directly involved in nutrient absorption and chylomicron production could recapitulate the global phenotype. Enterocytes are responsible for dietary lipid uptake and packaging into chylomicrons, while lacteal endothelial cells facilitate chylomicron transport into the circulation. To test whether impaired lipid assimilation resulted from cell-autonomous effects in these compartments, we generated intestinal epithelial-specific (Vil-Cre) (Figure 6A) and endothelial/hematopoietic-specific (Tie2-Cre) (Fig. S7A) Gpr75 knockout mice. However, deletion of Gpr75 in either enterocytes (Figure 6B–E) or endothelial lineages (Fig. S7B–S7E) did not alter body weight gain under high-fat diet feeding, indicating that loss of GPR75 in these individual intestinal compartments is insufficient to reproduce the lean phenotype observed in global knockout mice.Figure 6Deletion of Gpr75 in enterocytes does not affect body weight. (A) Intestinal and liver Gpr75 mRNA expression in Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice measured by real-time PCR. (B) Body weight curve of male Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice fed HFD for 10 weeks. (n = 8 mice per group). (C) Liver weights of male Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice fed HFD for 7 weeks. (n = 8 mice per group). (D) Body weight curve of female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 5 mice per group) fed HFD for 21 weeks. (E) Weights of Liver, eWAT, ingWAT and iBAT from female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 5 mice per group) fed HFD for 21 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. Deletion of Gpr75 in enterocytes does not affect body weight. (A) Intestinal and liver Gpr75 mRNA expression in Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice measured by real-time PCR. (B) Body weight curve of male Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice fed HFD for 10 weeks. (n = 8 mice per group). (C) Liver weights of male Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice fed HFD for 7 weeks. (n = 8 mice per group). (D) Body weight curve of female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 5 mice per group) fed HFD for 21 weeks. (E) Weights of Liver, eWAT, ingWAT and iBAT from female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 5 mice per group) fed HFD for 21 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests. Because Gpr75-deficient mice exhibit marked protection against hepatic steatosis, and hepatocytes play a central role in systemic lipid handling, we next generated hepatocyte-specific knockout mice using Alb-Cre (Fig. S8A). Despite efficient deletion, Alb-Cre-mediated Gpr75 deficiency did not affect body weight gain (Fig. S8B and S8D), liver weight or adipose depot mass (Fig. S8C and S8E), nor hepatic triglyceride content (Fig. S8F) in either male or female mice during high-fat feeding. Notably, a recent study reported that AAV8-mediated hepatic Gpr75 depletion protected mice from diet-induced hepatic steatosis [17]. As liver histology was not assessed in our cohort, hepatic steatosis cannot be conclusively evaluated. Nevertheless, under our HFD conditions, our findings suggest that the reduced body-weight gain observed in global GPR75-deficient mice is unlikely to be primarily attributable to hepatic Gpr75 deficiency. GPR75 acts in AgRP neurons to control HFD-induced weight gain Gpr75 is highly expressed in the central nervous system, including the hypothalamus, and has been reported to be present in a subset of NPY/AgRP neurons [4]. Given the high expression of Gpr75 in AgRP neurons and the established role of these neurons in regulating energy balance, we generated Agrp-Cre–mediated Gpr75 conditional knockout mice and the knockout efficiency was confirmed (Fig. S9A–C). Under HFD feeding, male AgRP-specific Gpr75 knockout mice exhibited a significant reduction in body weight gain compared with control littermates (Figure 7A). In female mice, body weight showed a similar downward trend, although the difference did not reach statistical significance (Figure 7C). Consistent with the body weight phenotype, AgRP-specific Gpr75 knockout mice of both sexes displayed reduced adipose tissue mass, with an even greater effect in females. In males, eWAT, ingWAT, and iBAT weights were reduced by 28%, 30%, and 20%, respectively, whereas in females the reductions reached 64%, 45%, and 41%, respectively (Figure 7B,D). Histological analysis further revealed significantly smaller adipocytes in both eWAT and ingWAT from knockout mice compared with controls (Figure 7E), supporting a reduction in adiposity at the cellular level. We next examined whether AgRP neuronal Gpr75 deficiency confers protection against obesity-associated metabolic abnormalities. In male mice, plasma TG and TC levels, as well as hepatic TG and TC contents, were not significantly altered (Figure 7F,G). In females, AgRP-specific Gpr75 deletion significantly reduced TG levels in both plasma and liver, whereas TC levels remained unchanged in both compartments (Figure 7H,I). Together, these findings establish AgRP neurons as a causal neuronal population through which GPR75 regulates body weight and adiposity. However, the magnitude of the phenotype is smaller than that observed in global Gpr75 knockout mice, indicating that AgRP-neuronal GPR75 accounts for only part of the overall metabolic effects of GPR75 deficiency.Figure 7AgRP neuron-specific deletion of Gpr75 reduces HFD-induced weight gain and adiposity. (A) Body weight curve of male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (n = 7–8 mice per group). (B) Weight of Liver, eWAT, ingWAT and iBAT from male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (n = 7–8 mice per group). (C) Body weight curve of female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice (n = 6 mice per group) fed HFD for 17 weeks. (D) Weights of Liver, eWAT, ingWAT and iBAT from female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 6 mice per group) fed HFD for 17 weeks. (E) Representative H&E staining of eWAT and ingWAT from male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice on HFD for 17 weeks. Scale bar, 50 μm. Quantification of adipocyte size in eWAT and ingWAT. (F) Plasma TG and TC in male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (G) Hepatic TG and TC in male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks (H) Plasma TG and TC in female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (I) Hepatic TG and TC in female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests or two-way analysis of variance (ANOVA) with Bonferroni's multiple comparisons test. AgRP neuron-specific deletion of Gpr75 reduces HFD-induced weight gain and adiposity. (A) Body weight curve of male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (n = 7–8 mice per group). (B) Weight of Liver, eWAT, ingWAT and iBAT from male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (n = 7–8 mice per group). (C) Body weight curve of female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice (n = 6 mice per group) fed HFD for 17 weeks. (D) Weights of Liver, eWAT, ingWAT and iBAT from female Gpr75f/f and Vil1-cre+/−, Gpr75f/f mice (n = 6 mice per group) fed HFD for 17 weeks. (E) Representative H&E staining of eWAT and ingWAT from male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice on HFD for 17 weeks. Scale bar, 50 μm. Quantification of adipocyte size in eWAT and ingWAT. (F) Plasma TG and TC in male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (G) Hepatic TG and TC in male Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks (H) Plasma TG and TC in female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. (I) Hepatic TG and TC in female Gpr75f/f and Agrp-cre+/−, Gpr75f/f mice fed HFD for 17 weeks. Data are presented as mean ± SD, unless otherwise stated. ∗P ≤ 0.05, ∗∗P ≤ 0.01. All p values were calculated by unpaired two-tailed student's t tests or two-way analysis of variance (ANOVA) with Bonferroni's multiple comparisons test. Discussion Many lines of evidence, including human genetic association studies and research in animal models, have unequivocally established GPR75 as a key regulator of diet-induced obesity [2,[4], [5], [6], [7], [8], [9], [10]]. However, the mechanism underlying the lean phenotype upon GPR75 inhibition remains largely unknown. In this study, we generated global Gpr75 knockout mice using the CRISPR/Cas9 system and found that GPR75-deficient mice are protected against diet-induced obesity and excessive fat accumulation in the liver. To define the tissue-specific basis of this phenotype, we further generated conditional knockout models targeting hepatocytes, intestinal epithelial cells, endothelial/hematopoietic lineages, and AgRP neurons. Deletion of Gpr75 in the tested peripheral tissues failed to recapitulate the global knockout phenotype, whereas AgRP neuron-specific deletion significantly reduced body weight and adiposity in both sexes, identifying AgRP neurons as a functionally relevant neuronal population through which GPR75 regulates energy balance. However, the magnitude of this phenotype was substantially smaller than that observed in global knockout mice, indicating that additional central and peripheral mechanisms contribute to the full metabolic effects of GPR75 deficiency. Despite the pronounced protection against obesity, food intake remained unchanged in global Gpr75 knockout mice during the period when body weight trajectories began to diverge. Unexpectedly, whole-body energy expenditure was reduced rather than increased, accompanied by lower fasting body temperature and an elevated respiratory exchange ratio, consistent with reduced fat utilization. Further investigation revealed a sex-dependent peripheral phenotype: male, but not female, Gpr75 knockout mice exhibited impaired intestinal lipid absorption, reduced chylomicron output, and increased fecal lipid excretion during HFD feeding. These findings indicate that reduced dietary lipid assimilation contributes to the protection against diet-induced obesity in male mice, while the overall phenotype of global GPR75 deficiency likely reflects coordinated actions across multiple tissues and mechanisms. Where does GPR75 control energy balance: brain, GI tract, or both? GPR75 is most highly expressed in the brain in both humans and mice, with widespread distribution across various brain regions [4]. This expression pattern in the brain differs from other well-known appetite-regulating genes, such as Agrp and Mc4r, which are more regionally localized [18,19]. The widespread distribution of GPR75 in the brain suggests that GPR75 may regulate energy balance through integrated neural circuitry rather than a single defined hypothalamic node. Recent work using Nestin-Cre-mediated neuronal deletion demonstrated that brain-specific loss of Gpr75 is sufficient to protect against HFD-induced obesity, primarily through reduced food intake and modest increases in energy expenditure [8]. However, these neuronal knockout mice also exhibited a marked reduction in lean mass and depot-specific alterations in adipose tissue during prolonged high-fat feeding, characterized by unchanged subcutaneous fat mass but significantly increased visceral fat mass, with largely preserved adipocyte size [8]. These body composition changes differ substantially from those observed in global GPR75-deficient mice, which display robust reductions in fat mass with preserved lean mass and decreased adipocyte size. Thus, while broad neuronal GPR75 signaling clearly contributes to body weight regulation, neuronal deletion alone does not fully recapitulate the systemic metabolic phenotype of global deficiency. To further dissect the neuronal contribution, we selectively deleted Gpr75 in AgRP neurons, a key hypothalamic population controlling appetite and energy balance [20]. AgRP-Cre-mediated deletion produced only a modest reduction in male body weight gain (∼10% after 12 weeks of HFD), markedly smaller than the ∼29% reduction observed in male global knockout mice. These findings identify AgRP neurons as a functionally relevant site of GPR75 action but indicate that they account for only part of the global phenotype. Additional neuronal populations and/or coordinated interactions between central and peripheral tissues are therefore likely to contribute to the full anti-obesity effect of GPR75 deficiency. In addition to the brain, GPR75 is expressed in other tissues, including intestine, white adipose tissue, and kidney, although at much lower levels [4]. Because our data demonstrate that reduced fat mass in global GPR75-deficient mice is associated with impaired intestinal lipid absorption and diminished chylomicron output, we hypothesized that GPR75 might directly regulate dietary lipid handling within the gut. In situ hybridization revealed broad expression of GPR75 across the intestinal epithelium, lamina propria, villi crypts, and muscularis externa, supporting the possibility of local intestinal regulation. However, genetic interrogation of specific intestinal compartments did not support a dominant tissue-autonomous role. Deletion of Gpr75 in enterocytes (Vil-Cre) or endothelial/hematopoietic lineages (Tie2-Cre) failed to alter body weight gain under HFD feeding. Similarly, hepatocyte-specific deletion (Alb-Cre) did not affect body weight gain or hepatic lipid accumulation. Taken together with the modest AgRP-Cre phenotype, these findings indicate that deletion of Gpr75 in any single tested tissue is insufficient to phenocopy the robust global knockout. Collectively, our data support a model in which GPR75 regulates diet-induced obesity through coordinated multi-organ interactions rather than a dominant tissue-autonomous mechanism. One possibility is that GPR75 signaling in the brain modulates intestinal lipid absorption through brain-to-gut communication pathways. Emerging evidence supports the existence of brain-to-gut communications that controls intestinal fat absorption [21]. Although the gastrointestinal tract possesses intrinsic neural circuitry, central nervous system inputs critically regulate gut motility, mucosal secretion, and intestinal blood flow [22,23]. Thus, the impaired intestinal lipid assimilation observed in global GPR75-deficient mice may reflect integrated neuron-gut signaling rather than isolated epithelial dysfunction. Importantly, during rapid HFD-induced fat mass expansion, diminished intestinal fat absorption and increased fecal excretion, rather than reduced food intake, appear to contribute to the reduced adiposity of GPR75-deficient mice. However, whether GPR75 controls these gastrointestinal phenotypes through central, local intestinal, or coordinated brain-gut mechanisms remains to be determined. Causal mechanisms underlying reduced fat mass upon GPR75 inhibition: less food intake, more fecal excretion, or both? Under a chow diet, GPR75-deficient mice exhibit significantly reduced body weight early in life, around 5–7 weeks old in our study. Previous study has shown that GPR75-deficinet mice consumed less food immediately post-weaning, when these mice were becoming thin. However, food intake reached to control mice levels in older adult GPR75-deficient mice after the thin phenotypes was established [6]. In agreement with this finding, our measurement of food intake of 10-week-old mice under both chow and HFD feeding did not reveal any significant difference between knockout and control mice, suggesting that hypophagia likely contributes to reduced fat mass in the early life stages of GPR75-deficient mice. Indeed, the body weight difference (approximately 9–10%) between GPR75-deficient mice and control littermates remains relatively stable from 7 to 15 weeks of age under chow feeding, which explains why food intake difference was no longer detectable in adult mice. However, given the substantial fat mass expansion during HFD feeding (approximately 7-fold increase in absolute fat mass after 16-week HFD feeding) and the drastically increased body weight difference between GPR75-deficient and control mice-growing from 10% under chow to almost 30% after 16 weeks of HFD feeding-other mechanisms likely contribute to the lean phenotype significantly in addition to the initial reduction in food intake during early life. Our data indicates that during the period of rapid fat mass accumulation under HFD feeding, there is no difference in food intake between GPR75-deficient and control mice. Instead, GPR75-deficient mice exhibit significantly diminished fat absorption and increased fecal excretion, which appear to be contributing to the lean phenotype during this phase of HFD-induced fat mass expansion. GPR75 regulates GI tract phenotypes Our data demonstrated that reduced chylomicron output likely explains the lean phenotype, diminished energy expenditure, and increased RER in GPR75-deficient mice. Intestinal chylomicron output consists of several steps, including dietary fat absorption, chylomicron assembly in enterocytes, trafficking and exit from the enterocytes, transport across the lamina propria, entry into lacteals, and circulation in lymphatics and blood [12,24]. A reduction in chylomicron output could result from dysregulation at any of these stages. Given that GPR75 is broadly expressed in the villi, including enterocytes, lamina propria stromal cells and lacteal-endothelial cells, the exact site and mechanisms by which GPR75 regulates chylomicron secretion require further investigation. In line with reduced intestinal chylomicron output, daily absolute fecal TG, TC, and FFA excretion (mg/day) were substantially increased in GPR75-deficient mice compared to control littermates fed HFD. Important to note that this increase in daily fecal lipid excretion is due to a greater total fecal output (g feces/day), while relative fecal lipid levels (mg lipid/g feces) remained similar between GPR75-deficient mice and control littermates. In addition, daily fecal protein excretion was significantly higher in GPR75-deficient mice, suggesting that the impaired nutrient absorption extends beyond dietary fat when GPR75 is inhibited. Importantly, the unchanged relative fecal TG levels (mg TG/g feces) in GPR75-deficient mice contrast with findings from Orlistat (a pancreatic lipase inhibitor) treatment, where relative fecal TG levels increased approximately 10-fold compared to placebo [25,26]. All these taken together suggest that GPR75 deficiency leads to a systemic defect in intestinal lipid and protein absorption, which likely contributes to reduced whole-body energy expenditure and an increased utilization of carbohydrates as an energy source. The increased absolute fecal excretion (g feces/day) suggests a potential increase in gut motility. However, we did not find a significant difference in total GI transit time between GPR75-deficient mice and control littermates. Consistent with this, water content in the stools was not affected upon depletion of GPR75. Furthermore, the width and length of individual stools were similar between GPR75-deficient mice and control littermates. Notably, we observed a significant decrease in intestine length in male GPR75-deficient mice fed HFD, suggesting that GPR75 and its downstream signaling may have intestinotrophic effects. Whether this shortened intestine length contributes to the increased fecal excretion remains unknown. Of note, GSEA analysis revealed that expression of several defensins was upregulated and enriched in the pathway related to defensive response to bacterial. Since defensins are anti-microbial peptides primarily produced by Paneth cells in the crypts, it is possible that GPR75 may play a role in regulating defensin synthesis in these cells. Despite transcriptome analysis indicated an upregulation of pathways related to immune response and anti-bacterial response, GPR75-deficient mice exhibited a significant reduction in gut permeability compared to control littermates. Furthermore, fecal microbiome analysis demonstrated that the microbial composition in GPR75-deficient mice was associated with a lean phenotype: characterized by a reduction in Firmicutes and an increase in Bacteroidetes compared to control littermates. An intriguing question remains how GPR75 is involved in shaping the microbiome and whether this microbial shift contributes to the lean phenotype upon GPR75 inhibition. Limitations of the study Several limitations of this study should be acknowledged. First, although the lean phenotype of GPR75-deficient mice was associated with reduced dietary lipid assimilation and increased fecal lipid loss, we did not fully define the molecular mechanisms linking GPR75 deficiency to these changes. The relationship between impaired intestinal fat absorption and increased fecal lipid excretion is likely complex, and the causal sequence remains unresolved. Second, metabolic-cage studies were performed only in male mice, limiting our ability to determine whether the observed changes in energy expenditure and substrate utilization are shared across sexes. Notably, the intestinal fat-absorption defect was detected in male but not female mice, indicating a sex-dependent component of the peripheral phenotype. Third, although conditional deletion of Gpr75 in hepatocytes (Alb-Cre), intestinal epithelium (Vil-Cre), or endothelial/hematopoietic lineages (Tie2-Cre) did not affect body weight gain within the experimental period, these findings demonstrate only that deletion in the individually tested peripheral cell populations is insufficient to recapitulate the robust global knockout phenotype. They do not exclude smaller contributions that might become apparent with a more prolonged metabolic challenge, nor contributions from other peripheral cell types or coordinated inter-organ mechanisms. Fourth, the 60% kcal-high-fat diet used in this study provides a robust model of diet-induced obesity but exceeds the dietary fat content typically consumed by humans and therefore may not fully recapitulate the nutritional context of human obesity [27]. Finally, while AgRP neuron-specific deletion identified AgRP neurons as a functionally relevant neuronal population, we did not further elucidate the downstream molecular mechanisms through which GPR75 acts in these neurons. Defining these cell-type-specific signaling pathways and their integration with peripheral metabolic regulation will require further investigation. In summary, global GPR75 deficiency protects against diet-induced obesity and hepatic steatosis through coordinated central and peripheral mechanisms, including reduced dietary lipid assimilation under HFD feeding. Importantly, we identify AgRP neurons as a specific neuronal population through which GPR75 regulates body weight and adiposity. These findings support GPR75 as a promising therapeutic target for obesity and associated metabolic disease. Methods Mice and diet All animal care and experimental procedures used in this study were approved by the Institutional Animal Care and Use Committee at the National University of Singapore (Protocol number R20-1518). Mice were housed under a 12h light–dark cycle in a temperature and humidity-controlled environment (21–23 °C, 45–55%). Mice were maintained on an irradiated grain-based laboratory diet (Teklad Global 18% Protein Rodent Diet, diet 2918; Envigo, Madison, WI, USA), hereafter referred to as “chow,” or high-fat diet (HFD, 60% fat; TD.06414; Envigo) and with free access to water [28]. Both male and female 6- to 37-week-old mice were used in this study as reported. Gpr75+/−, loxp/loxp knockin mice (Gpr75fl/fl), and AgRP-Cre mice (Stock No: C001249) were obtained from Cyagen Biosciences at 8 weeks of age. Albumin-Cre (The Jackson Laboratory, RRID: IMSR_JAX:003574), Tie2-Cre (The Jackson Laboratory, RRID: IMSR_JAX:008863), Villin-Cre (The Jackson Laboratory, RRID: IMSR_JAX:004586) mice were obtained from Jackson Laboratory at 8 weeks of age. Gpr75+/+ and Gpr75−/− littermates were generated by breeding Gpr75+/− mice. Liver-specific, enterocyte-specific, endothelial/hematopoietic-specific and AgRP neuron-specific knockout mice were generated by crossing Gpr75fl/fl with Albumin-Cre, Villin-Cre, Tie2-Cre and Agrp-Cre transgenic mice, respectively. Indirect calorimetry All experiments were performed at room temperature of 21–23 °C on a 12h light–dark cycle in a PhenoMaster System (TSE systems, Bad Homburg, Germany). The experimental design followed current consensus recommendations for indirect calorimetry studies [29]. Mice were individually housed in open-circuit metabolic chambers and allowed to acclimatize to the new housing condition for 1 week prior to experiment. Weight of food and volume of water consumed, gas exchange, and activity levels were measured consecutively for 4 days under chow diet and 3 days under HFD. Prior to measurement under HFD, the mice were allowed to recover for 7 days, followed by HFD habituation for another 7 days. Data were analyzed using CaIR [30]. Body temperature measurement Mice were fasted overnight (14–16h) and refed for 6 h. Body temperature was measured using RET-3 rectal probe (Physitemp) attached to a Type-J/K/T/R/S Handheld Thermocouple Thermometer (Digi-Sense). Oral lipid tolerance assay Mice were fasted for 4h followed by 200 μl of highly refined, low-acidity olive oil (Sigma–Aldrich, O1514) or 30ug BODIPY-labelled fatty acid (BODIPY™ 500/510 C1, C12 (4,4-Difluoro-5-Methyl-4-Bora-3a,4a-Diaza-s-Indacene-3-Dodecanoic Acid) in 200uL olive oil via oral gavage. The olive oil contained predominantly oleic acid (C18:1). Blood samples were taken from tail vein right before and at different time points after oral gavage as reported. Plasma triglyceride (TG) was then isolated to measure TG levels using Infinity Triglycerides Reagent (Thermo Scientific, TR22421) or Triglyceride Colorimetric Assay Kit (Elabscience, E-BC-K261-M). BODIPY level in the plasma was measured using a plate reader (Varioskan Flash, Thermo Scientific) with an excitation wavelength of 492 nm and emission wavelength of 520 nm. Whole gut transit time assay Mice were fasted overnight (14–16h) and gavage with 200 μl of 6% Carmine Red (Sigma–Aldrich, C1022) dissolved in 0.5% methylcellulose (Sigma–Aldrich, M0512) solution. Subsequently, the mice are individually housed in a clean cage with free access to food and water. Whole gut transit time is calculated as the time interval between the gavage and the first observed red fecal pellet. Intestinal permeability assay Mice were fasted overnight (14–16h) and gavage with 200 μl of 10 mg/ml FITC-Dextran 4 KDa (Sigma–Aldrich, 46944) in PBS. Blood samples were collected from tail vein right before and at different time points after oral gavage as reported. Plasma FITC-Dextran levels were measured using a plate reader (Varioskan Flash, Thermo Scientific) with an excitation wavelength of 485 nm and emission wavelength of 535 nm. Blood lipid assays Blood was collected from tail tips after 4h of fasting and plasmas were isolated via centrifugation. Triglyceride (TG) was measured using Infinity Triglycerides Reagent (Thermo Scientific, TR22421) or Triglyceride Colorimetric Assay Kit (Elabscience, E-BC-K261-M). Total cholesterol (TC) was measured using Infinity Cholesterol Reagent (Thermo Scientific, TR13421) and free fatty acid (FFA) was measured by Free Fatty Acid Assay Kit (Abcam, ab65341). Blood glucose analysis Mice were fasted for 4h or overnight (14–16h) as specified. Blood was collected from tail tips and blood glucose was measured using handheld glucometer, Accu-Chek Instant (Roche). Western blotting analysis Mouse plasma samples were analysed for ApoB-48 and ApoB-100 level. 2uL of serum subjected to SDS-PAGE using standard protocol. Blots were blocked for 1h with 5% Blotting-Grade Blocker (Bio-Rad, 1706404) in TBST (1X TBS and 0.1% Tween-20) and were incubated overnight with anit-apoB antibody (Santa Cruz Biotechnology, sc-393636, 1:1000) diluted in 5% Blotting-Grad Blocker in TBST at 4 °C. Blots were washed three times with TBST, followed by HRP-conjugated secondary antibody (Anti-mouse IgG, Cell Signalling, 7076, 1:2000) incubation for 1h at room temperature. Blots were visualised using ECL detection reagent (Clarity Western ECL Substrate, Bio-Rad, 1705061). Quantification was performed using Fiji ImageJ software [31]. Tissues and fecal lipids analysis Tissues were excised and snap-frozen in liquid nitrogen immediately and subsequently stored at −80 °C. Fecal pellets were picked up from the cage using forceps, and stored at −80 °C. The protocol for lipid extraction uses folch's extraction method with minor modifications [32]. Briefly, 50 mg of snap-frozen tissues (liver, duodenum, jejunum, and ileum) were homogenized in 500 μl of ice-cold PBS. 200 μl of the homogenate was transferred into 1200 μl of chloroform: methanol (2:1; v/v) mixture followed by vigorous vortex for 30s, followed by addition of 100 μl of ice-cold PBS into the mixture and mixed vigorously for 15s. Similarly, 50 mg of dried, pulverized feces were mixed with 200 μl of ice-cold PBS, and vortex vigorously for 60s. The mixture is then transferred into 1200 μl of chloroform: methanol (2:1; v/v) mixture. The mixture was then centrifuged at 4,200 rpm for 10 min at 4 °C. 200 μl of the organic phase (bottom layer) was transferred into a new tube and evaporated in the fume hood till completely dry. The dried pellet is dissolved in 100 μl 1% Triton X-100 in ethanol using constant rotation for 1h, and subsequently dried using a speedVac vacuum concentrator before dissolving in 100 μl of 1% triton X-100 in PBS. TG and TC levels were measured using Infinity Triglycerides Reagent (Thermo Scientific, TR22421) or Triglyceride Colorimetric Assay Kit (Elabscience, E-BC-K261-M) and Infinity Cholesterol Reagent (Thermo Scientific, TR13421) respectively, according to manufacturer's instructions. FFA level was measured by Free Fatty Acid Assay Kit (Abcam, ab65341), based on manufacturer's instructions. For tissue BODIPY measurement, 50 mg of snap frozen small intestinal tissues (duodenum, jejunum, and ileum) were homogenised in 500 μl of ice-cold RIPA buffer. The mixture was then centrifuged at 4,200 rpm for 10 min at 4 °C. Measure the BODIPY fluorescent signal in the supernatant using a plate reader (Varioskan Flash, Thermo Scientific) with an excitation wavelength of 492 nm and emission wavelength of 520 nm. Fecal protein analysis To analyze fecal protein level, 100 mg of feces were mixed in 500 μl PBS containing 0.1% Tween 20 and protease/phosphatase inhibitor (Thermo Scientific, 78430). The mixture is vortex for 15 min to get homogenous fecal suspension, and subsequently centrifuged at 4,200 rpm for 10 min at 4 °C. Protein concentration in the supernatant is measured by BCA assay protocol as per manufacturer protocol (Thermo Scientific, 23225). Histology Tissues (Liver, adipose tissues, and jejunum of small intestine) were fixed in 10% formalin for 24–48 h at 4 °C, dehydrated, embedded into paraffin and sectioned (5 μm) for H&E staining. Adipocyte size quantification was performed on H&E-stained sections of eWAT and ingWAT pads, using AdipoSoft software. 80 to 100 adipocytes per fat pad section and four sections per mouse were analyzed. Alcian blue staining protocol was used to stain goblet cells. Quantification of villi length, and crypt depth was performed using Fiji ImageJ software measuring 25 villi and 40 crypts per mice [31]. Similarly, 20 villi and 60 crypts were used to quantify goblet cell count per villi and crypt respectively. For oil red O staining, jejunum sections were fixed in 10% formalin for 24–48 h at 4 °C, cryopreserved with 30% sucrose solution, and embedded in OCT compound (Sakura, 4583). 10 μm was used to stain with Oil red O solution (Sigma-Adrich, O1391). Immunofluorescence Whole-mount Jejunum IF staining protocol was adapted from Bernier-Latmani and Petrova, 2016 with modifications [33]. Briefly, Jejunum was cut into 1 cm sections and fixed in 10% formalin for 24–48 h at 4 °C, followed by overnight incubation with 30% sucrose solution. Jejunum tissue sections was blocked in 2% BSA blocking buffer containing 3% horse serum in 0.3% Triton X-100 and 1% Tween20, for 2h at 4 °C, and subsequently incubated with Rat anti-LYVE1 antibody (Thermo Scientific, 14-0443-82), overnight at 4 °C. Jejunum tissue sections were washed in 0.3% Triton X-100 in PBS for 2h at 4 °C with buffer changed every 30 min, followed by Goat anti-rat secondary antibody, Alexa Fluor 647 (Thermo Scientific, A-21247) overnight incubation at 4 °C. Jejunum tissue sections were then thoroughly washed for 100 min with buffer changes every 10 min, then fixed in 10% formalin for 24h. The jejunum tissue sections were cut into strips containing 2–3 villi wide and mounted onto slides. Images were taken by confocal imaging (Zeiss LSM710). 20 lacteals per mice were quantified using Fiji ImageJ software [31]. RNAscope ISH analysis RNAscrope assay was performed using according to manufacturer's protocol. Briefly, Jejunum sections were fixed in 10% formalin for 24–48 h at 4 °C, dehydrated, embedded into paraffin and sectioned at 5 μm thickness, and air-dried at room temperature overnight. Slides were treated and detected by fluorescence staining using Gpr75 probe (Advanced Cell Diagnostics, 318281) and RNAscope Multiplex Fluorescent Reagent Kit v2 (Advanced Cell Diagnostics, 323270). Negative control probe targeting DapB gene, and positive control probe targeting Polr2a gene was performed for each RNAscope ISH experiment. Images were taken by confocal imaging (Olympus FV3000). Microbiota profiling Bacterial genomic DNA was extracted from feces of mice and 30 ng of purified DNA was used for PCR amplification. Sequencing was performed with BGI's proprietary DNBSEQ™ sequencing technology platforms. Single-nucleus RNAseq analysis The snRNA-seq data were obtained from the previous published study [14] and processed using Seurat [34]. Given the low and sparse expression of GPR75 in the human intestine, nuclei expressing ≥0.5 UMI count for GPR75 gene were identified as GPR75+ cells. The violin plots were generated using the Seurat package and ggplot2. RNAseq and data analysis For the RNA-seq assay, total mRNA was isolated from mouse jejunum samples using Trizol (Invitrogen, 15596018). The extracted total RNA was quantified using a Nanodrop 2000 spectrophotometer (Thermo Fisher) and assessed for quality using a Bioanalyzer. cDNA library preparation and mRNA sequencing were performed by Novogene, utilizing the state-of-the-art Illumina NovaSeq platforms and a paired-end 150 bp sequencing strategy. For data analysis, the raw reads were initially processed with FastQC and Cutadapt to eliminate adapters, over-represented sequences, and low-quality sequences. Subsequently, the alignment of the reads was conducted using the HISAT2 software. SAMtools was utilized for sorting and converting the mapped reads to the BAM format. Following that, reads per kilobase per million (RPKM) values were calculated for each gene, and differentially expressed genes (DEGs) were identified using DESeq2, considering those with a fold change of more than 0.5 and a p value of less than 0.01. To further explore the biological implications of the DEGs, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome enrichment analysis was carried out employing the clusterProfiler package. Gene sets with a P value below 0.05 were considered statistically significant. Quantification and statistical analysis Statistical analyses were performed using Prism (Graphad). No statistical methods were used to predetermine sample size. Values are presented as means ± SD or mean ± SEM, as indicated in figure legends. Statistical significance of differences was determined by unpaired, two-tailed student's t test (two groups), with p values < 0.05 representing statistically significant. For indirect calorimetry, analysis is performed by CalR programme (https://calrapp.org/). CRediT authorship contribution statement Xiaoyun Cao: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Vanna Chhay: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Thi Tun Thi: Validation, Methodology, Investigation. Hui Li: Methodology, Investigation, Data curation. Yu Shi: Software, Formal analysis. Yifan Wang: Methodology, Formal analysis. Yilin Min: Methodology. Renhua Song: Methodology, Conceptualization. Kai Yan Cheng: Visualization, Resources. Yang Yang: Methodology. Justin Jong-Leong Wong: Writing – review & editing, Conceptualization. Haojie Yu: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.