1. Introduction

Since 1975, there has been a substantial increase in the global prevalence of obesity, with the magnitude nearly tripling. The World Health Organization projects that the prevalence of obesity among adults will exceed 20% by the year 2025 [1]. Currently, the management of excessive adiposity poses a paramount economic burden and healthcare predicament [2, 3]. In addition to the detrimental social and psychological implications, a multitude of studies have consistently demonstrated a significant association between obesity and an increased vulnerability to a range of health conditions, such as type 2 diabetes, cardiovascular diseases, and cancer [4–7].

Activating and maintaining the thermogenesis of beige/brown fat has been shown to be effective in treating obesity and related metabolic disorders in humans [8, 9]. As a well-established β-adrenergic GPCR, the β3-AR has been identified as a prominent target for stimulating adipose thermogenesis in mice. Regrettably, the clinical application of β3-AR has been impeded due to its low expression in human adipocytes and the cardiovascular risks associated with other adrenergic receptors [10, 11]. G protein-coupled receptors (GPCRs) are the most prevalent class of drug targets among all drugs approved by the U.S. Food and Drug Administration (FDA). They also play a crucial role in the clinical treatment of obesity [12–14]. Therefore, it is of clinical significance to identify novel GPCR targets that induce adipose thermogenesis.

ADGRA3 is classified as an orphan adhesion G protein-coupled receptor (aGPCR) and exhibits the typical domains found in aGPCRs within its N-terminal extracellular region (ECR), including a leucine-rich repeat (LRR), an immunoglobulin-like domain (Ig), a hormone-binding domain (HBD), and a GAIN domain [15]. ADGRA3 was initially discovered as a distinctive indicator of various spermatogonial progenitor cells [16, 17]. Recent studies have shown that the orphan status of receptors has posed challenges to the study of aGPCRs. However, these studies have also uncovered a conservative mechanism of aGPCR activation, which involves the use of tethered ligands in the GAIN domain [18, 19]. ADGRA3 has been previously identified as a receptor capable of auto-cleavage [20]. However, the functional activity of ADGRA3 in a constructive manner is still uncertain. A genome-wide association study (GWAS) demonstrated a significant correlation between single nucleotide polymorphisms (SNPs) of ADGRA3 and body weight in chickens [21].

Nevertheless, the precise role of ADGRA3 in the progression of obesity and adipose thermogenesis remains uncertain. This study aimed to investigate three main aspects: (1) the impact of ADGRA3 on browning of white adipose tissue (WAT) and brown adipose tissue (BAT), (2) the effects of ADGRA3 on metabolic homeostasis, and (3) the underlying mechanisms by which ADGRA3 induces adipose thermogenesis.

2. Material and methods

2.1. Mice

Wild-type (WT) C57BL/6J mice were obtained from the Center of Laboratory Animal at Sun Yat-sen University. All mice were housed in the Sun Yat-sen University Laboratory Animal Center, where they were subjected to a 12-hour light-dark cycle and maintained at a controlled environmental temperature of 21±1℃. Eight-week-old male C57BL/6J mice were fed with a normal chow diet (NCD) or a high fat diet (HFD, 60% kcal) for 12 weeks to render mice obese. With the exception of mice fed with a HFD, male mice at the age of eight weeks were utilized in all experimental procedures.

For the knockdown and over-expression experiments of Adgra3 in mice fed with a NCD, the following procedures were conducted: shAdgra3 (pLKO.1-U6-shAdgra3-2 plasmid encapsulated in nanomaterials) and shNC (pLKO.1-U6-shNC plasmid encapsulated in nanomaterials) were injected intraperitoneally (i.p.) for knockdown experiments, while Adgra3 OE (pLV3-CMV-Adgra3 (mouse)-3×FLAG plasmid encapsulated in nanomaterials) and CON (pLV3-CMV-MCS-3×FLAG plasmid encapsulated in nanomaterials) were injected i.p. for over-expression experiments. The frequency of the sessions was twice a week over a period of four weeks. For the knockdown and over-expression experiments of Adgra3 in mice fed with a HFD, the following procedures were conducted: shAdgra3 and shNC were injected intraperitoneally (i.p.) for knockdown experiments, while Adgra3 OE and CON were injected i.p. for over-expression experiments. The frequency of the sessions was twice a week over a period of 12 weeks. For the local over-expression of Adgra3 in BAT of mice fed with a NCD, Adgra3 OE and CON were injected locally into BAT once.

For the treatment with a selective β3-adrenoceptor agonist, CL-316,243 (hereafter referred to as CL), mice fed a HFD were injected intraperitoneally (i.p.) with CL (1 mg/kg daily) for 7 days. For the treatment with hesperetin (Hes), hesperetin is dissolved in drinking water (200mg/L) and water were available ad libitum.

Intraperitoneal injections of glucose (2g/kg for mice fed with a NCD and 1g/kg for mice fed with a HFD) or insulin (0.5U/kg for mice fed with a NCD and 1U/kg for mice fed with a HFD) were administered. At the designated time points of 0 minutes, intraperitoneal glucose or insulin tolerance tests were conducted on mice that had been fasted for six hours. After administration, the blood glucose concentration was assessed at specific time intervals using a OneTouch Ultra Glucometer. Finally, the animals were euthanized, followed by the collection of tissue samples. Cohorts of ≥3 mice per genotype or treatment were assembled for all in vivo studies. All in vivo studies were repeated 2-3 independent times. All procedures related to animal feeding, treatment and welfare were conducted at Sun Yat-sen University Laboratory Animal Center.

2.2. Stromal Vascular Fraction (SVF) and mature adipocytes isolation

SVF from inguinal white adipose tissue (iWAT) and BAT of WT male mice at 4 weeks of age were washed with PBS, minced and digested with 0.1% type Ⅱ collagenase in Dulbecco’s modified eagle medium (DMEM) containing 3% BSA and 25μg/ml DNase Ι for 30 min at 37℃. During the digestion, the mixed solution was shaken by a hand every 5 min. The mixed solution was filtered through a 70 μm cell strainer and then centrifuged at 500 g for 5 min at 4℃. The floating mature adipocytes were collected for subsequent analysis, and the pellets containing SVF were resuspended in red blood cell lysis buffer for 5 min at 37℃. Cells were centrifuged at 500 g for 10 min at 4℃ and the SVF pellets were collected for subsequent analysis.

2.3. Cell culture

3T3-L1 and 293T cell lines were purchased from the Cell Bank of the Chinese Academy of Sciences in Shanghai. Human adipose-derived mesenchymal stem cells (hADSCs) were purchased from the National Stem Cell Translational Resource Center. 3T3-L1 cells were culture and grown to confluence in high-glucose DMEM supplemented with 10% newborn calf serum (NCS). 293T cells were culture and grown to confluence in high-glucose DMEM supplemented with 10% FBS. Confluent 3T3-L1 pre-adipocytes were induced into mature beige-like adipocytes with 0.5 mM isobutyl methylxanthine (IBMX), 1 μM dexamethasone, 5 μg/ml insulin, 1 nM 3, 3’, 5-Triiodo-L-thyronine (T3), 125 μM indomethacin and 1 μM rosiglitazone in high-glucose DMEM containing 10% FBS for 2 days, then treated with high-glucose DMEM containing 5 μg/ml insulin, 1 nM T3, 1 μM rosiglitazone and 10% FBS for 6 days and cultured with high-glucose DMEM containing 10% FBS for 2 days. hADSCs were seeded on plates coated with 0.1% gelatin and culture and grown to confluence in human mesenchymal stem cells (hMSCs) specialized culture medium (ZQ-1320). Confluent hADSCs were induced into mature human adipocytes with adipogenic induction medium (PCM-I-004) according to the manufacturer’s instructions.

The shAdgra3, shGnas (pLKO.1-U6-shGnas plasmid encapsulated in nanomaterials) and shNC were added to 3T3-L1 mature beige-like adipocytes for 72 hours. The shADGRA3 (pLKO.1-U6-shADGRA3-(1/2/3) plasmid encapsulated in nanomaterials) and shNC were added to human adipocytes for 72 hours. The pcDNA3.1(+)-mGnas-6×His (mixture of pcDNA3.1(+)-Gnas(mouse)-6×His plasmid and transfection reagent), pcDNA3.1(+)-mGnai1-6×His (mixture of pcDNA3.1(+)-Gnai1(mouse)-6×His plasmid and transfection reagent), pcDNA3.1(+)-mGnaq-6×His (mixture of pcDNA3.1(+)-Gnaq(mouse)-6×His plasmid and transfection reagent), pcDNA3.1(+)-mGna12-6×His (mixture of pcDNA3.1(+)-Gna12(mouse)-6×His plasmid and transfection reagent), Adgra3 OE and CON were added to 3T3-L1 mature beige-like adipocytes or 293T for 48 hours. The ADGRA3 OE (pLV3-CMV-ADGRA3(human)-3×FLAG plasmid encapsulated in nanomaterials) and CON were added to human adipocytes for 48 hours. Hesperetin (10 μM) and PKAi (protein kinase A inhibitor, 20 μM H-89) was added to 3T3-L1 mature beige-like adipocytes for 48 hours. All in vitro studies were repeated 2-3 independent times.

2.4 Construction of plasmid

The pLV3-CMV-Adgra3(mouse)-3×FLAG, pLV3-CMV-ADGRA3(human)-3×FLAG, pLKO.1-U6-shADGRA3-(1/2/3), pLV3-CMV-MCS-3×FLAG, pcDNA3.1(+)-Gnas(mouse)-6×His, pcDNA3.1(+)-Gnai1(mouse)-6×His, pcDNA3.1(+)-Gnaq(mouse)-6×His and pcDNA3.1(+)-Gna12(mouse)-6×His plasmids were purchased from Shenzhen Yanming Biotechnology Co., LTD. The pLKO.1-U6-shAdgra3-(1/2/3) and pLKO.1-U6-shNC plasmids were purchased from Guangzhou Hanyi Biotechnology Co., LTD.

2.5. Temperature measurements

The body temperature was measured at 9:00 using a rectal probe connected to a digital thermometer.

2.6. Real-time Polymerase Chain Reaction (PCR)

Total RNA from tissue or cells was extracted with Trizol reagent. RNA concentration was measured by a NanoDrop spectrometer. 1000ng total RNA was reverse transcribed into cDNA by All-in-One RT SuperMix (G3337). Real-time PCR analysis using SYBR-Green fluorescent dye was performed with Step One Plus RT PCR System. Primers used for real-time PCR were listed in Table S1.

2.7. Histology and immunohistochemistry

Subcutaneous, epididymal white adipose tissue, interscapular brown adipose tissue and liver were fixed in 4% paraformaldehyde. Tissues were embedded with paraffin and sectioned by microtome. The slides were stained with hematoxylin and eosin (HE) using a standard protocol. For UCP1 and ADGRA3 immunohistochemistry, slides of various tissue were blocked with goat serum for 1h. Subsequently, the slides were incubated with anti-UCP1 (1:1000; ab10983) or anti-ADGRA3 (1:200; 11912-1-AP) overnight at 4℃ followed by detection with the EnVision Detection Systems. Hematoxylin was used as counterstain.

2.8. Western-blot

Tissues and cells were lysed in RIPA buffer supplemented with 1 mM PMSF and protease inhibitor cocktail (K1007). The protein concentration was measured by the BCA protein assay kit (BL521), and total cellular protein (25 μg) was subject to Western-blot analysis. The protein transferred to the PVDF membrane was probed with primary antibodies specific for HSP90 (1:1000; C45G), α-tubulin (1:10000; 666031-1-Ig), ADGRA3 (1:1000; 11912-1-AP), UCP1 (1:1000 for iWAT and cells or 1:10000 for brown adipose tissue (BAT); ab10983), FLAG-tag (HRP Conjugated, 1:2000; AF2855), HIS-tag (HRP Conjugated, 1:2000; AF2879), p-CREB (1:1000; AF5785) and CREB (1:1000; AF6566) overnight at 4 °C. Except FLAG-tag protein and HIS-tag protein, after being incubated with HRP conjugated secondary antibody, proteins were detected with chemiluminescence using Immobilon Western HRP Substrate on ChemiDoc MP Imaging System. The ImageJ software was used for gray scanning. For all Western-blots, each lane represented an independent sample and all experiments were replicated 2 to 3 times.

2.9. IP assay

HEK293T cells were transfected using PEI 40K transfection reagent (G1802) with indicated cDNAs and cultured using the manufacture’s protocol. Cells were lysed with IP lysis buffer (G2038) containing protease inhibitor cocktail (K1007). The lysates were precipitated with the FLAG-tag antibody (GB15938) or HIS-tag antibody (GB151251) in the presence of protein A+G agarose (P2055). The precipitants were washed five times with the IP lysis buffer and analyzed by immunoblot with the indicated antibodies.

2.10. Enzyme-linked immunosorbent assay (ELISA)

Mouse cAMP level was detected using a sensitive ELISA kit (MM-0544M2) purchased from Jiangsu Meimian Industrial Co., Ltd. Mouse IP1 level was detected using a sensitive ELISA kit (MM-0790M2) purchased from Jiangsu Meimian Industrial Co., Ltd. Mouse insulin level was detected using a sensitive ELISA kit (MM-0579M1) purchased from Jiangsu Meimian Industrial Co., Ltd. Mouse free tetraiodothyronine (fT4) level was detected using a sensitive ELISA kit (RXJ202449M) purchased from Quanzhou Ruixin Biological Technology Co., Ltd. All measurements were performed using the manufacture’s protocol.

2.11. Bodipy staining

For the lipid staining, the differentiated adipocytes were washed twice with PBS. The cells were then stained with 2 μM BODIPY staining solution (GC42959) for 15 min at 37℃, then washed three times with PBS according to manufacturer’s instructions. The stained cells were observed using a fluorescence microscope.

2.12. Mito-Tracker staining

The differentiated adipocytes were incubated with 100nM Mito-Tracker Red CMXRos (C1049) for 30 min according to manufacturer’s instructions. Then cells were washed with PBS and visualized under the confocal microscope.

2.13. Determination of 2-deoxy-D-glucose (2-NBDG) uptake

The differentiated adipocytes were washed twice with PBS. The cells were then incubated with 100 μM 2-NBDG staining solution (HY-116215) for 30 min at 37℃, then washed three times with PBS. The stained cells were observed using a fluorescence microscope.

2.14. Measurement of Triacylglycerol (TG)

The triacylglycerol in adipocytes, tissues and plasma was measured by using Triglyceride Assay Kit (A110-1-1) according to the manufacturer’s instructions.

2.15. Transmission electron microscopy

BAT sections were fixed in 2% (vol/vol) glutaraldehyde in 100mM phosphate buffer, pH 7.2 for 12 h at 4℃. The sections were then post-fixed in 1% osmium tetroxide, dehydrated in ascending gradations of ethanol and embedded in fresh epoxy resin 618. Ultra-thin sections (60-80 nm) were cut and stained with lead citrate before being examined on the FEI-Tecnai G2 Spirit Twin transmission electron microscope.

2.16. Differential expression analysis

The R package Linear Models for Microarray Data (limma) was used to analyze differential RNA-Sequencing expression. For screening high-expressed G-protein-coupled receptors in mouse BAT, limma was applied in the GSE118849 dataset to screen out BAT-elevated genes. For screening ADGRA3 high-expressed gene sets in human subcutaneous adipose, limma was applied in the human subcutaneous adipose dataset from GTEx Portal to screen out ADGRA3 high-expressed gene sets. Genes highly expressed in human adipocytes were obtained from the human protein atlas database. Genes with the cutoff criteria of |logFC| ≥ 1.0 and P < 0.05 were regarded as differentially expressed genes (DEGs). The DEGs of the GSE118849 dataset and the human subcutaneous adipose dataset were visualized as volcano plots by using the R package ggplot2.

2.17. Functional annotation for genes of interest

To explore DisGeNET, Gene Ontology (GO), WikiPathwas, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome of selected genes, Metascape was used to explore the functions among DEGs, with a cutoff criterion of p < 0.05. GO annotation that contains the biological process (BP) subontology, which can identify the biological properties of genes and gene sets for all organisms.

2.18. Gene set enrichment analysis (GSEA)

GSEA was performed to detect a significant difference in the set of genes expressed between the ADGRA3 high-expressed and ADGRA3 low-expressed groups in the enrichment of the KEGG collection.

2.19. Oxygen consumption rate (OCR)

The basal oxygen consumption rate of cells was measured using a BBoxiProbe R01 kit (BB-48211) according to the manufacturers’ instructions. The maximum oxygen consumption rate of cells was measured with the addition of FCCP (Trifluoromethoxy carbonylcyanide phenylhydrazone) with a final concentration of 1μM.

2.20. Infrared Thermography

BAT temperature was measured at room temperature by infrared thermography according to previous publications [22, 23]. The same batch of representative infrared images of mice were all captured using a thermal imaging camera (FLIR ONE PRO), measured at the same distance perpendicular to the plane on which the mice were located. To quantify interscapular region temperature, the average surface temperature from a region of the interscapular BAT was taken with FLIR Tools software.

2.21. Availability of Data and Materials

The transcriptomic dataset analyzed in this study can be accessed on the GTEx Portal database (https://gtexportal.org/home/multiGeneQueryPage), human protein atlas database (https://www.proteinatlas.org/) and GEO repository under accession number GSE118849. The PRESTO-Salsa dataset of ADGRA3 in this study can be accessed on the PRESTO-Salsa database (https://palmlab.shinyapps.io/presto-salsa/)[24]. All other data associated with this paper can be found in the main text or the Supplementary Materials.

2.22. Statistical Analysis

All data are presented as mean ± SEM. In this study, outliers that met the three-sigma rule were excluded from analysis, with the exception of those presented in Figure S1E. Given the possibility that the outliers in Figure S1E represent extreme expressions of the inherent variability within the population sample, we have chosen to retain these specific outliers for further analysis. Student’s t-test was used to compare two groups. One-way analysis of variance (ANOVA) or Two-way ANOVA was applied to compare more than two different groups on GraphPad Prism 9.0 software. For each parameter of all data presented, NS (No Significance), *p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001. p < 0.05 is considered significant.

3. Results

3.1. ADGRA3 is identified as a potential GPCR inducing the development of beige fat

We conducted a comprehensive analysis of three datasets to identify ADGRA3 as a potential GPCR target that promotes the development of beige fat (Figure 1A). To identify novel GPCRs that induce the biogenesis of beige fat, we conducted differential gene expression analysis (Figure 1B) and Venn diagram analysis (Figure 1C) using the GSE118849 dataset obtained from the Gene Expression Omnibus (GEO) database. Additionally, we utilized the human subcutaneous adipocytes dataset (Figure 1C, red) and human visceral adipocytes dataset (Figure 1C, purple) from the human protein atlas database to obtain genes that are highly expressed in human white adipocytes. The GSE118849 dataset comprises samples of brown adipose tissue (BAT) and inguinal white adipose tissue (iWAT) obtained from mice subjected to a 72-hour cold exposure at a temperature of 4℃.

Figure 1.

A total of 1134 differentially expressed genes (DEGs) that exhibited up-regulation in BAT compared to iWAT under cold stimulation were identified in the analysis, which might play a role in adipose thermogenesis. These DEGs were further screened to identify highly expressed GPCRs in BAT relative to WAT (Figure 1B, red). We conducted additional annotation on 1134 DEGs and identified that 27 of these genes were associated with the encoding of GPCRs (Table S2). Among the set of 27 genes, it was found that 24 genes were not present in the group of genes that exhibited high expression levels in human adipocytes, as determined by the human protein atlas database. Consequently, these 24 genes were excluded from further analysis. We conducted a comprehensive literature review and discovered that out of the three remaining GPCRs namely ADGRA3, ADRA1A, and ADRB1, only ADGRA3 has not been documented to have any association with brown fat. Therefore, our research subsequently shifted towards investigating the potential regulatory role of ADGRA3 in obesity and brown fat.

The findings indicated that the level of Adgra3 expression in mouse adipose tissue (Figure 1D) was comparatively lower than that of Adrb3, the coding gene for β3-AR. Conversely, in human adipose tissue, ADGRA3 expression was observed to be higher than that of ADRB3 (Figures 1E and S1E). We conducted an investigation to examine the regulatory effects of a high-fat diet on the transcription of Adgra3 and Ucp1 (Uncoupling protein 1, a functional protein and marker of beige/brown fat). The findings of the study demonstrated that a HFD had a significant inhibitory effect on the expression of ADGRA3 and UCP1 in iWAT and BAT, while CL robustly increased the expression of ADGRA3 and UCP1 in iWAT and BAT (Figures 1F-G and Figures S1A-D). Interestingly, in human subcutaneous fat, there was a moderate positive correlation between the expression level of ADGRA3 and the expression level of UCP1 (R=0.5, Figure 1H). On the other hand, the expression level of ADRB3 showed a weak positive correlation with the expression level of UCP1 (R=0.21, Figure S1F). The data presented in this study indicate that ADGRA3 is a GPCR that exhibits high expression levels in BAT and may participate in inducing adipose thermogenesis.

3.2. Adgra3 overexpression induces the biogenesis of beige adipocytes in vitro

To ascertain the predominant expression of ADGRA3, the isolation of stromal Vascular Fraction (SVF) and mature adipocytes from WAT and BAT was conducted for subsequent validation. The results showed that ADGRA3 is predominantly expressed in adipocytes. Furthermore, the expression level of ADGRA3 in BAT adipocytes was found to be higher compared to WAT adipocytes (Figure 1I). However, no significant difference was observed in the expression level of ADGRA3 in the SVF of WAT and BAT (Figure 1I). Moreover, it was observed that the modulation of the expression levels of Adgra3/ADGRA3 and Ucp1/UCP1 exhibited a similar pattern during the differentiation process between mouse and human adipocytes (Figure S1G-H).

To investigate the role of ADGRA3 in the biogenesis of beige adipocytes, we conducted an experiment where we transformed pre-adipocytes 3T3-L1 into mature beige-like adipocytes with a knockdown of Adgra3. Our findings indicate that the knockdown of Adgra3 resulted in a decrease in the expression of genes related to thermogenesis and lipolysis (Figure 2A). Western blot analysis and Mito-Tracker staining revealed a decrease in the expression of UCP1 (Figure 2B) and a reduction in the number of mitochondria (Figure 2C) following Adgra3 knockdown. Lipid droplet fluorescence staining and intracellular triglyceride assay were performed on adipocytes to assess the impact of Adgra3 knockdown. The results revealed a significant increase in the number of lipid droplets and intracellular triglyceride levels (Figures 2C-D) following Adgra3 knockdown. Moreover, the uptake of 2-deoxy-D-glucose (2-NBDG), a fluorescently-labeled deoxyglucose analog, by adipocytes was significantly inhibited following the knockdown of Adgra3 (Figure 2E). Furthermore, oxygen consumption rate (OCR) was detected to verify the effect of ADGRA3 on the oxygen consumption of adipocytes. The results indicated that the loss of ADGRA3 decreased the both basal and max OCR of adipocytes (Figures 2F-G).

Figure 2.

Following the overexpression of Adgra3, there was an observed up-regulation in the expression of UCP1 in 3T3-L1 mature beige-like adipocytes (Figures 2H-I). Additionally, Mito-Tracker staining revealed an increase in the quantity of mitochondria (Figure 2J). There was a notable reduction observed in the lipid droplets and intracellular triglyceride levels (Figures 2J-K) subsequent to the overexpression of Adgra3. Moreover, the findings indicated that the overexpression of Adgra3 resulted in an increased uptake of 2-NBDG by adipocytes (Figure 2L) and increased basal and maximum OCR (Figures 2M-N). The presented data suggest that ADGRA3 has the ability to stimulate the formation of beige adipocytes in vitro.

3.3. Adgra3 knockdown suppresses adipose thermogenic program and impairs metabolic homeostasis in vivo

To evaluate the role of ADGRA3 in the biogenesis of beige fat in vivo, mice fed with a NCD were injected with shNC or shAdgra3 for 28 days (Figure 3A). After knocking down Adgra3 in mice (Figures S2A-B), there was a significant increase in the weight of shAdgra3 mice (mice with Adgra3 knockdown) (Figure 3B). Furthermore, the food intake of shAdgra3 mice was elevated slightly (Figure S2C). Serum triacylglycerol (TG) levels (Figure S2D), weight of iWAT, epididymal white adipose tissue (eWAT) and BAT (Figure 3C) were significantly higher in shAdgra3 mice. Liver weight (Figure 3C) and TG levels in the liver (Figure S2E) did not show a significant difference between shNC mice and shAdgra3 mice. Meanwhile, hematoxylin-eosin staining showed that Adgra3 knockdown induced adipose expansion in iWAT (Figure S2F), eWAT (Figure S2G) and BAT (Figure S2F) but not lead to hepatic steatosis (Figure S2G).

Figure 3.

Remarkably, the knockdown of Adgra3 resulted in a significant reduction in both body temperature (Figure 3D) and BAT temperature (Figure 3E). Given the crucial influence of thyroid activity on thermogenesis, we measured the levels of serum free tetraiodothyronine (fT4) to evaluate the consequences of Adgra3 knockdown on thyroid activity, which indicated that the nanoparticle-mediated Adgra3 knockdown does not exert an inhibitory effect on thyroid activity (Figure S2H). The knockdown of Adgra3 resulted in a significant decrease in the expression of genes related to thermogenesis and lipolysis in both iWAT (Figure 3F) and BAT (Figure 3G). Moreover, the Western blot analysis (Figures 3H-I) and Immunohistochemical staining (Figure 3J) of UCP1 revealed comparable outcomes. Additionally, it was observed that the knockdown of Adgra3 resulted in an increase in the size of lipid droplets and a decrease in the number of mitochondria in BAT (Figure 3K). These findings indicate that ADGRA3 plays a crucial role as a receptor in the biogenesis of beige fat and the activation of BAT.

Moreover, the genes that were highly expressed in ADGRA3 high-expressed human subcutaneous adipose tissue (Figure S3A, red) exhibited enrichment in various biological processes. These processes included hyperinsulinism, obesity (Figure S3B), metabolic processes (Figure S3C), adipogenesis (Figure S3D), regulation of lipolysis in adipocytes (Figure S3E), and lipid metabolism (Figure S3F). GSEA was conducted to search the enriched KEGG pathways based on the expression level of ADGRA3 in human subcutaneous adipose dataset and human visceral adipose dataset from GTEx portal database. For ADGRA3 high-expressed group, both subcutaneous adipose dataset (Figure S3G) and visceral adipose dataset (Figure S3H) enriched in insulin signaling pathway, which indicates that ADGRA3 may be involved in the regulation of glucose metabolism in addition to its influence on lipid metabolism. Furthermore, it was observed that shAdgra3 mice exhibited significant disruptions in overall glycemic homeostasis (Figure 3L) and insulin sensitivity (Figure 3M). Moreover, the fasting serum insulin level was increased and the homeostasis model assessment of insulin resistance (HOMA-IR) showed an increase in shAdgra3 mice (Figures 3N-O). Hence, the findings of this study provide evidence that the knockdown of Adgra3 hampers adipose thermogenesis and disrupts metabolic homeostasis in vivo.

3.4. ADGRA3 activates the adipose thermogenic program and counteracts metabolic disease in vivo

To identify whether Adgra3 overexpression induces adipose thermogenesis and improves the metabolic homeostasis against obesity, Adgra3 OE and CON were injected i.p. into mice fed with a NCD (Figure S4A) and a HFD (Figure 4A), thereby establishing models of Adgra3-overexpressed mice (Figures 4F-G and Figures S4B-C). The growth of body weight of Adgra3 OE mice was alleviated (Figure 4B) during the HFD feeding accompanied with a slight decrease of food intake (Figure S5A). Furthermore, the weight of iWAT, eWAT, BAT and liver (Figure 4C) were significantly decreased in Adgra3 OE mice. The Adgra3 OE mice exhibited an elevation in both body temperature (Figures 4D and S4D) and BAT temperature (Figures 4E and S4E), while there was no difference in serum fT4 levels (Figures S4F and S5B). Meanwhile, Adgra3 overexpression decreased the TG level in serum and liver (Figures S5C-D) as well as the area of adipocytes in iWAT, eWAT, BAT and liver (Figures S4G and S5E-F).

Figure 4.

Moreover, the expression levels of thermogenic and lipolysis-related genes were elevated in iWAT (Figures 4F and S4H) and BAT (Figures 4G and S4I). Western-blot (Figures 4H-I and S4J-K) and immunohistochemical staining of UCP1 (Figures 4J and S4L) showed that the expression of UCP1 was increased dramatically in iWAT and BAT after Adgra3 overexpression. In addition, we found that after Adgra3 overexpression, BAT presented multiple thermogenesis fat features (Figure 4K). Furthermore, nanomaterials carrying Adgra3 OE were directly injected into BAT (Figure S5G), resulting in overexpression of Adgra3 and upregulation of thermogenic and lipolysis-related genes, as compared to BAT injected with CON (Figure S5H). These findings indicate that the overexpression of Adgra3 is capable of inducing the hallmarks of thermogenesis in mice, independently of other organs.

We then investigated the metabolic impact of ADGRA3. The glucose tolerance test (GTT) presented that Adgra3 overexpression improved the glucose homeostasis of HFD mice (Figure 4L). The insulin tolerance test (ITT) showed that Adgra3 overexpression alleviated the insulin resistance of HFD mice (Figure 4M). Moreover, the fasting serum insulin level was reduced after Adgra3 overexpression (Figure 4N) and the HOMA-IR also showed a robust improvement (Figure 4O) in Adgra3OE mice. Taken together, Adgra3 overexpression activates the adipose thermogenic program and improves the metabolic homeostasis in diet-induced obese mice against obesity and insulin resistance in vivo.

3.5. ADGRA3 activates the adipose thermogenic program via the G_s-PKA-CREB axis

To ascertain the ADGRA3-conjugated Gα protein, we conducted an overexpression of FLAG-labeled mouse ADGRA3 and four different types of His-labeled G_α proteins (G_s, G_i, G_q and G₁₂) in 293T cells. The lysate obtained from the 293T cells was then utilized for the subsequent co-immunoprecipitation (co-IP) analysis. The results of the co-IP experiment demonstrated that mouse ADGRA3 coupled to the G_s protein (Figures 5A-B), while no interaction was observed with the other three Gα proteins (G_i, G_q and G₁₂; Figures S6A-C) ADGRA3 exhibits intrinsic and auto-cleavable receptor activity, allowing it to signal even in the absence of an exogenous ligand [16, 20]. Hence, the overexpression of Adgra3 is capable of inducing cAMP production (Figure 5C), which serves as a second messenger indicating the activation of downstream signals mediated by G_s protein. This response is comparable to the effect of a ligand. However, there is no production of IP1, which is a metabolite of the downstream second messenger IP3 associated with G_q protein (Figure S6D). Additionally, our findings indicate that the effect of Adgra3 overexpression on cAMP production is dependent on G_s protein (Figures 5D and S6E). These results suggest that ADGRA3 is involved in the coupling of G_s protein, leading to the stimulation of downstream cAMP production.

Figure 5.

Hence, it was hypothesized that the overexpression of Adgra3 could potentially stimulate adipocyte thermogenesis by activating the PKA signaling pathway. As expected, the Western-blot analysis revealed that the overexpression of Adgra3 leads to an elevation in the phosphorylated form of CREB (p-CREB), indicating an increase in PKA-CREB signaling activity. This effect was observed in 3T3-L1 (Figure 5E), as well as in the iWAT and BAT (Figure 5F). Consistently, the knockdown of Adgra3 resulted in a decrease in the level of p-CREB in 3T3-L1 (Figure 5G), as well as in iWAT and BAT (Figure 5H). To investigate the potential role of Adgra3 overexpression in promoting the biogenesis of beige adipocytes and activating the PKA-CREB signaling pathway via G_s protein, we conducted an experiment using 3T3-L1 cells. The cells were treated with Adgra3 OE and shGnas, respectively. Adgra3 overexpression was found to be adequate in inducing the expression of UCP1 in 3T3-L1 cells. However, this effect was observed to be eliminated when Gnas was knocked down (Figure 5I and S6E). Furthermore, the utilization of PKAi (protein kinase A inhibitor, H-89) was employed to ascertain the dependence of the browning effect of Adgra3 overexpression on the PKA-CREB signal. The results showed that PKAi effectively inhibited the activation of PKA-CREB signaling and UCP1 expression induced by Adgra3 overexpression (Figures 5J and S6F). These results suggest that the observed browning effect resulting from Adgra3 overexpression is mediated through the PKA-CREB signaling pathway. Collectively, these findings indicate that ADGRA3 facilitates the biogenesis of beige adipocytes through the G_s-PKA-CREB signaling pathway.

3.6. Hesperetin: a screened ADGRA3 agonist that induces the biogenesis of beige adipocytes

Considering the difficulty of overexpressing ADGRA3 in clinical application, hesperetin was screened as a potential agonist of ADGRA3 by PRESTO-Salsa database (Figure 6A). The results showed that hesperetin treatment stimulates cAMP production (Figure 6B) and increases the expression level of UCP1 and p-CREB (Figures 6C-D). To verify whether hesperetin induces the biogenesis of beige adipocyte and activates PKA-CREB signal via ADGRA3, we treated 3T3-L1 with hesperetin and shAdgra3, respectively. We found that the induction effect of hesperetin on UCP1 and p-CREB is eliminated when Adgra3 is knocked down (Figures 6E-F). In addition, OCR was detected to verify the effect of hesperetin on the oxygen consumption of adipocytes. The results indicated that hesperetin increased the both basal and max OCR of adipocytes, which was ADGRA3-dependent (Figures 6G-H).

Figure 6.

Moreover, the results showed that the induction effect of hesperetin on UCP1 and p-CREB is attenuated after Gnas knocked down (Figures 6I-J), suggesting that hesperetin up-regulates UCP1 and activates PKA-CREB axis dependent on G_s. Furthermore, PKAi was used to verify whether the browning effect of hesperetin was dependent on PKA-CREB signal. The results revealed that hesperetin treatment resulted in the upregulation of UCP1 and p-CREB. However, this effect was found to be eliminated when PKAi was applied (Figures 6K-L), suggesting that the induction of UCP1 and p-CREB by hesperetin is dependent on PKA. These findings suggest that hesperetin exerts an induction effect on biogenesis of beige adipocytes via ADGRA3-G_s-PKA-CREB axis.

3.7. Hesperetin: a potential ADGRA3 agonist that activates the adipose thermogenic program and counteracts metabolic disease dependent on ADGRA3

To identify whether hesperetin induces adipose thermogenesis and improves the metabolic homeostasis against obesity via ADGRA3, shNC mice or shAdgra3 mice were treated with hesperetin and fed with a HFD (Figure 7A). Hesperetin was found to alleviate the growth of body weight (Figure 7B) during the HFD feeding and the weight of iWAT, eWAT, BAT and liver weight ratio (Figure 7C), which was dependent on ADGRA3. It is noteworthy that the food consumption of shAdgra3 mice slightly surpassed that of shNC mice, while the administration of hesperetin remained uninfluential on their dietary intake (Figure S7A). Hesperetin increased body temperature (Figure 7D) and BAT temperature (Figure 7E) in shNC mice, which were significantly blunted in shAdgra3 mice. The levels of serum fT4 were measured to evaluate the consequences of hesperetin treatment on thyroid activity, which indicated that hesperetin treatment does not activate thyroid activity (Figure S7B).

Figure 7.

Concurrently, hesperetin induced a decline in TG level in both serum (Figure S7C) and liver (Figure S7D), and also reduced the area of adipocytes in iWAT (Figure S7E) and BAT (Figure S7F). However, these effects were absent in shAdgra3 mice. Moreover, the expression level of UCP1 were elevated in both iWAT (Figures 7F and 7H) and BAT (Figures 7G and 7I) after hesperetin treatment in shNC mice but not in shAdgra3 mice. The results indicated that hesperetin treatment resulted in a substantial decrease in lipid droplet size and a significant increase in mitochondria quantity in BAT (Figure 7J). However, this browning effect was weakened in shAdgra3 mice. These findings suggest that hesperetin is sufficient to orchestrate the hallmarks of thermogenesis in mice, which is dependent on ADGRA3.

We then investigated the metabolic impact of hesperetin treatment. The GTT presented that hesperetin improved the glucose resistance of HFD mice which showed no effect in shAdgra3 mice (Figure 7K). The ITT showed that hesperetin alleviated the insulin resistance of HFD mice which showed no significance in shAdgra3 mice (Figure 7L). Moreover, the fasting serum insulin level was reduced after hesperetin treatment (Figure 7M) and the HOMA-IR also showed a moderate improvement (Figure 7N), which were dependent on ADGRA3. Taken together, hesperetin activates the adipose thermogenic program and improves the metabolic homeostasis in diet-induced obese mice against obesity and insulin resistance in vivo, which is ADGRA3 dependent.

3.8. ADGRA3 overexpression induces the biogenesis of human beige adipocytes in vitro

Given the elevated expression level of ADGRA3 compared to ADRB3 in human adipose tissue (Figures 1E and S1E), we induced human adipose-derived mesenchymal stem cells (hADSCs) into adipocytes to evaluate the effect of ADGRA3 on human adipocytes. The results showed that ADGRA3 knockdown led to a diminished expression of UCP1 (Figure 8A), whereas its overexpression elicited an enhancement in UCP1 expression (Figure 8B). Furthermore, Mito-Tracker and lipid droplet fluorescence staining illuminated a notable increase in lipid droplet count accompanied by a decrease in mitochondrial number following ADGRA3 knockdown (Figure 8C). Conversely, ADGRA3 overexpression resulted in a visible surge in mitochondrial quantity and a marked reduction in lipid droplet presence (Figure 8D).

Figure 8.

4. Discussion

In the present study, we have elucidated a novel role of ADGRA3 and hesperetin in inducing the development of beige adipocytes through the activation of the G_s-PKA-CREB signaling pathway. ADGRA3 is responsible for the activation of the adipose thermogenic program and plays a significant role in maintaining systemic glucose homeostasis. Additionally, the development of beige adipocytes induced by hesperetin is contingent upon the presence of ADGRA3. The novelty of this study is the discovery that ADGRA3 plays a role in the advancement of beige fat and the regulation of metabolic homeostasis. This suggests that targeting the ADGRA3-G_s-PKA-CREB signaling pathway could potentially be a therapeutic approach for obesity and related metabolic disorders.

The induction of beige fat has been investigated as a potentially effective therapeutic approach in combating obesity [25]. A clinical trial revealed that treatment with the chronic β3-AR agonist mirabegron leads to an increase in human brown fat, HDL cholesterol, and insulin sensitivity [26]. Subsequently, Blondin et al discovered that oral administration of mirabegron only elicits an increase in BAT thermogenesis when administered at the maximal allowable dose, indicating that human brown adipocyte thermogenesis is primarily driven by β2-adrenoceptor (β2-AR) stimulation [11]. Consistent with this finding, we found much higher levels of ADRB2 expression in human white adipose tissue than ADRB3 (Figure S1E). Furthermore, a recent study has demonstrated that simultaneous activation of β2-AR and β3-AR enhances whole-body metabolism through beneficial effects on skeletal muscle and BAT [27].

While the promotion of thermogenesis in brown and beige adipocytes in rodents was effectively achieved by the β3-adrenoceptor agonist, the clinical implications of this finding appear to be unfeasible in humans due to the low efficacy of β3-adrenoceptor agonists [28, 29]. It is of utmost importance to investigate alternative therapeutic targets that can effectively and selectively enhance beige adipogenesis in order to combat obesity and its related metabolic disorders. In this study, we have identified ADGRA3 as a novel GPCR therapeutic target that exhibits high expression in human adipocytes. In human adipose tissue, ADGRA3 is expressed at a lower level than ADRB2 (Figure S1E), which has been shown to be the main receptor mediating adrenergic activation of thermogenesis in human brown adipocytes [11]. Nevertheless, given ADRB2’s pivotal role in bronchodilation and vasodilation [30, 31], we believe that ADGRA3 has the potential to be an alternative target for inducing adipose thermogenesis. Overall, these findings suggest that ADGRA3, when overexpressed or stimulated by its potential agonist, hesperetin, can induce the biogenesis of beige fat.

Hesperetin has been reported to attenuate the age-related metabolic decline, reduce fat and improve glucose homeostasis in naturally aged mice [32]. Previous studies showed that hesperetin improved glycemic control [32, 33] and was involved in adipocyte differentiation [34], but whether hesperetin induces the biogenesis of beige adipocyte was uncertain. Previously, the influence of hesperetin on ADGRA3 has remained unreported. In this study, we screened hesperetin as a potential agonist for ADGRA3 by using the PRESTO-Salsa tool as well as discovered that hesperetin has an agonist effect on ADGRA3 through a series of experiments. This study focuses on the regulatory effect of hesperetin on adipose thermogenesis and explores whether this effect is dependent upon ADGRA3. As such, we refrained from conducting further investigations into other potential effects of hesperidin, including its potential role in antioxidant and in apoptosis.

In previous reports, male mice deficient in ADGRA3 showed obstructive azoospermia with high penetrance [15]. Moreover, a genome-wide association study (GWAS) identified a single nucleotide polymorphism (SNP) located in the downstream region of ADGRA3 as a genomic locus associated with body weight in chickens, suggesting that the ADGRA3 is a potential regulator of body weight [21]. Nevertheless, the agonist and the downstream signal axis of ADGRA3 remain unclear as well as the effects of ADGRA3 on adipose thermogenesis and glucose homeostasis have not been explored. Consequently, our study has confirmed that the knockdown of Adgra3 exacerbates obesity and disrupts glucose homeostasis. Additionally, both the overexpression of Adgra3 and the administration of hesperetin have been found to stimulate the biogenesis of beige adipocytes through the ADGRA3-G_s-PKA-CREB signaling pathway and improve glucose homeostasis.

Given the consideration that the non-targeted nanoparticle approach utilized in this study for modulating Adgra3 expression levels in vivo alter Adgra3 expression in tissues beyond adipose tissue (Figures S2A-B and S4B-C), notably the liver and skeletal muscle, the construction of Adgra3 adipose tissue-specific knockout/overexpression mouse models is imperative for a more nuanced understanding of the precise mechanisms underlying the influence of on adipose thermogenesis. We will employ more sophisticated models in subsequent studies to further elucidate the effects of ADGRA3 on adipose thermogenesis and metabolic homeostasis. Nevertheless, our findings underlie a potential therapeutic feature of ADGRA3 and hesperetin in obesity and the associated metabolic diseases from the thermogenic viewpoint of beige fat.

5. Conclusion

In conclusion, the activation of the G_s-PKA-CREB axis by ADGRA3 has been found to induce adipose thermogenesis, promote lipid metabolism, and alleviate lipid accumulation in adipose tissues. Furthermore, the induction of beige adipocyte biogenesis by hesperetin occurs through the ADGRA3-G_s-PKA-CREB axis. Given the importance of identifying signaling pathways that induce beige fat and alleviate obesity-related dysfunction in adipose tissue, our research findings suggest that hesperetin and activation of the intracellular signaling of ADGRA3 could serve as a promising and innovative therapeutic approach.

Acknowledgements

The authors are grateful for the GTEx Portal database, human protein atlas database, Gene Expression Omnibus and PRESTO-Salsa database. This research was funded by the Shenzhen Science and Technology Project (grant number: JCYJ20190807154205627), Fund of Shenzhen Key Laboratory of Systems Medicine for inflammatory diseases (grant number: ZDSYS20220606100803007) received by Zhonghan Yang. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Additional information

Compliance with Ethics Requirements

All the animal experiments were conducted with the approval of the Animal Care and Use Committee of Sun Yat-sen University (Approval ID: SYSU-IACUC-MED-2023-B082). This study was conducted in accordance with the ethical principles derived from the Declaration of Helsinki and Belmont Report and was approved by the review board of Sun Yat-sen University (Guangzhou, China).

CRediT authorship contribution statement

Zewei Zhao: Investigation, Conceptualization, Data curation, Formal analysis, Writing original draft, Writing-review & editing. Longyun Hu: Data curation, Formal analysis. Bigui Song: Data curation, Formal analysis. Tao Jiang: Data curation, Formal analysis. Qian Wu: Data curation, Formal analysis. Jiejing Lin: Data curation, Formal analysis. Xiaoxiao Li: Data curation, Formal analysis. Yi Cai: Data curation, Formal analysis. Jin Li: Data curation, Formal analysis. Bingxiu Qian: Data curation, Formal analysis. Siqi Liu: Data curation, Formal analysis. Jilu Lang: Conceptualization, Supervision, Writing-review & editing. Zhonghan Yang: Funding acquisition, Conceptualization, Supervision, Writing-review & editing.

Declaration of competing interests

The authors have declared no conflict of interest.

Abbreviations

• ANOVA: analysis of variance

• BAT: brown adipose tissue

• β2-AR: β2-adrenoceptor

• β3-AR: β3-adrenoceptor

• BMI: body mass index

• ADGRA3: adhesion G protein-coupled receptor A3

• CL: CL-316,243

• DEGs: differentially expressed genes

• edgeR: Empirical Analysis of Digital Gene Expression Data in R

• eWAT: epididymal white adipose tissue

• FDA: Food and Drug Administration

• GEO: Gene Expression Omnibus

• GPCRs: G protein-coupled receptors

• GSEA: Gene set enrichment analysis

• NCD: normal-chow diet

• HFD: high-fat diet

• GTT: glucose tolerance test

• HOMA-IR: homeostasis model assessment of insulin resistance

• ITT: insulin tolerance test

• iWAT: inguinal white adipose tissue

• Limma: Linear Models for Microarray Data

• Hes: Hesperetin

• PKAi: protein kinase A inhibitor

• NS: No Significance

• RPKM: Reads Per Kilobase per Million mapped reads

• TPM: Transcripts Per Kilobase Million

• WT: Wild-type

• GWAS: genome-wide association study

• SNP: single nucleotide polymorphism

• OCR: oxygen consumption rate

• SVF: Stromal Vascular Fraction

• hADSCs: human adipose-derived mesenchymal stem cells

Supplementary materials and methods

Table S1.

Table S2.

Supplementary figure

Figure S1.

Figure S2.

Figure S3.

Figure S4.

Figure S5.

Figure S6.

Figure S7.