Orphan Receptor GPR50 Improves In�ammation and Insulin Signaling in 3T3-L1 Preadipocyte

Purpose: The goal of this study was to investigate the effect of orphan G Protein-Coupled Receptor 50 (GPR50) receptor on in�ammation and insulin signaling in 3T3-L1 preadipocyte. Subjects and Methods: A high-fat diet (HFD)-induced obesity-T2DM (Type 2 Diabetes Mellitus) mouse model was used in this research, and high expression of GPR50 in mouse adipose tissue was screened by microarray technology. Expression of GPR50 in 3T3-L1 cell line and obesity-T2DM mouse adipose tissue was con�rmed. To gain more insight into the potential role of this new target in obesity-associated IR development, a GPR50 knockout cell line was constructed in 3T3-L1 cell line. In�ammatory cytokine levels and insulin signaling pathways in the GPR50 knockout 3T3-L1 cell line were determined by quantitative real-time polymerase chain reaction analysis and western blot. Results: GPR50 expression was signi�cantly increased in adipose tissue of obesity-T2DM mice. GPR50 de�ciency increased in�ammation in 3T3-L1 cells. In addition, GPR50 de�ciency induced the phosphorylation of AKT and insulin receptor substrate (IRS)1. Furthermore, GPR50 knockout 3T3-L1 cell line had suppressed PPAR-γ expression. Conclusions: These data demonstrated a novel target GPR50 can affect in�ammation and insulin signaling in adipocytes. Furthermore, the effects are mediated through the regulation of insulin signaling and PPAR-γ expression.


Introduction
Type 2 diabetes(T2DM)is a metabolic disease which is widely prevalent in the world. It is characterized by insulin secretion de ciencies and systemic insulin resistant (IR) in adipose tissue, skeletal muscle, and liver. (Rachdaoui 2020) Although the mechanism of T2DM is not yet fully known, in ammation and insulin resistance play a central role in the pathogenesis of homeostasis through direct regulation of IRS-1 and glucose transporter type 4 (GLUT4). (Ahmadian et al. 2013) Loss of PPARγ in skeletal muscle, severe IR was shown. (Hevener et al. 2003) G protein-coupled receptors (GPCRs) are involved in endocrine and metabolic processes as well as many other physiological processions. Many medicinal drugs, including glycolipid metabolism drugs, target GPCRs.(Rask-Andersen, Almen, and Schioth 2011) GPR50 is an orphan GPCR, which shares highest sequence homology with melatonin receptors. (Emet et al. 2016) GPR50 has shown to be an important regulator of energy metabolism in GPR50 knockout mouse. (Ivanova et al. 2008) A sequence variant study suggests that GPR50 is related to mental disorders(Watkins and Orlandi 2020) and altered lipid metabolism. (Bhattacharyya et al. 2006) To investigate gene differential expression in mouse epididymal adipose tissue, we developed a HFD-induced obesity-T2DM mouse model. We observed a drastic increase in the expression of the orphan GPCR, GPR50, in the adipose tissue of HFD mice via gene array technology. We investigated the roles of GPR50 de ciency in obesity induced in ammation and insulin signaling. The association of GPR50 with in ammation and insulin signaling pathway was investigated by using GPR50 knockout 3T3-L1 cell line. The phosphorylation of IRS-1 and AKT were observed activated signi cantly in GPR50 knockout cell line. Also, we have shown that GPR50 regulates insulin signaling and PPAR-γ expression.

Cells, reagents, and treatments
The American Type Tissue Culture Collection provided the mouse preadipocyte 3T3-L1 cell line (Manassas, VA). Cells were maintained in high-glucose Dulbecco's Modi ed Eagle's Medium (DMEM) (Invitrogen, Carlsbad, CA) and 10% bovine calf serum (GIBCO, Carlsbad, CA) at 37 °C in a 5% CO2 incubator. In treatment experiments, 3T3-L1 cells were incubated in treatment experiments with 16.7 mM glucose or 0.25 mM palmitate for 48 h to mimic obese adipocytes.(Manna, Achari, and Jain 2017; Li et al. 2016) Antibody against IRS-1 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against phospho-IRS-1 Ser612 , phospho-Akt Ser473 , Akt, β-actin, GAPDH and PPAR-γ were purchased from Cell Signaling Technology (Danvers, MA). GPR50 antibody was purchased from Proteintech Group, Inc (Rosemont, IL). For insulin stimulation, GPR50 knockout 3T3-L1 cells or control cells were seeded in 6-well plates. Two days after reaching con uence, cells were treated with serum starvation in DMEM for 4 h, then washed three times with PBS. Cells were incubated in DMEM containing 100 nM insulin or DMEM alone for 30 min. Finally, cells were washed 3 times and proteins were extracted for Western blot detection.

Differentiation of 3T3-L1 adipocytes
Mouse 3T3-L1 preadipocytes were grown in Dulbecco's modi ed Eagle's medium (DMEM, Hyclone) with 10% fetal bovine serum (FBS, GIBCO). Two days after reaching con uence, the cells medium was modi ed into differentiation medium (StemPro™ Adipogenesis Differentiation Kit). The induction medium was replaced every three days until the 3T3-L1 had differentiated into mature adipocytes.

Mouse model
Eight weeks old C57BL/6J male mice were obtained from Guangdong Medical Laboratory Animal Center (Guangdong, China).
The mice were randomly divided into two groups after one-week adaptation: the HFD group (n=3 from 5) and the chow group (n=3 from 5). Mice were fed high-fat diet (45% fat, Research diet D12451, New Jersey) or chow diet ad libitum for 15 weeks before the symptoms of the mice were characteristic for T2DM. The body weight was assessed every two weeks. Animal care was in compliance with the Guide for the Care and Use of Laboratory Animals of Guangdong Province. All the procedures were under the supervision and approved by the Ethics Committee for Animal Research, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (Approval Number: SIAT-IRB-170401-YGS-RPG-A0312-01).

RNA Extraction
The epididymal fat tissues have been isolated from the mice. RNA samples were collected from the adipose tissue using Trizol reagent (Invitrogen Corporation, California) and were validated with Agilent Array platform for microarray assay and real-time PCR analysis.

Microarray Analysis
The study of microarray was performed by GMINIX (Shanghai, China) with Mouse Transcriptome Array 1.0 (Affymetrix). This microarray targets about 114,000 protein-coding transcripts. We puri ed RNA from three HFD and three chow mice in epididymal adipose tissues, and used a random priming process to transcribe the RNA into complementary DNA (cDNA). In the GeneChip® Hybridization Oven 645, cDNA was then fragmented, biotinylated and 5.5 μg of cDNA was hybridized to the GeneChip Mouse Transcriptome Array 1.0. The arrays were screened with the Gene Array Scanner 3000 7G (Affymetrix, California) after hybridization and washing. All the data were analyzed using the Robust Multichip Analysis (RMA) algorithm using Affymetrix default analysis settings and global scaling as a normalization process.
Quantitative real-time PCR The total RNA was extracted by the Trizol reagent from cells and mouse adipose tissue (Invitrogen Corporation, CA). Reverse transcription and RT-PCR were performed using RT-PCR Kit (Takara, Japan). For Q-PCR, a light Cycler (Roche, Switzerland) and SYBR Quantitative real-time PCR (Takara, Japan) kit were used. The endogenous control was used as GAPDH. Table 1 presents PCR primers used for mRNA quantitation and were synthesized from Sangon Biotech (Shanghai, China).
Enzyme-linked immunosorbent assay GPR50 knockout 3T3-L1 or control cells were incubated with16.7 mM glucose or 0.25 mM palmitate for 48 h as described above. Supernatants were collected centrifuge at 3 000 rpm for 5 min. The concentrations of IL-6 (Dakewe, CN), MCP-1(Sino biological, CN) and IL-1β (R&D Systems, Minneapolis) assayed with ELISA system kit according to manufacturers' instructions. Quantitative data are presented as average concentrations in pg/ml .

Construction of GPR50 knockout cell line
To establish stable GPR50-knockdout 3T3-L1 cell line, using an online CRISPR design tool (http://crispr.mit.edu), GPR50 small guide RNAs (sgRNAs) is designed. The sgRNA sequence is as following: GTTTCAGAGCTATGCTGGAAACAGCATAGCAAGTTGAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGGTGCT. For further analysis, the sequences were synthesized and annealed to form a gRNA duplex. BsmB I digested the lentiviral vector (Obio, Shanghai, China), resulting in sticky ends. The phosphorylated and annealed sgRNA was ligated to the lentiviral vector. Sequencing con rmed the right clone and the high purity plasmid was extracted. Shanghai OBiO Technology Co, Ltd completed the lentiviral packaging (Shanghai, China). The 3T3-L1 cell suspension was added to each well of a six-well plate with a density of 5×10 4 /mL, and the lentivirus was added to each well at the proper density. For monoclonal cell screening, 2 μg/mL puromycin was introduced after 72 hours of infection. 48 hours later, the screened and selected cells were divided and seeded into 96-well plates with 1 cell/well using ow cytometry. After about 2 weeks, single-cell colonies were obtained; a single cell pellet was picked up with microscope and inoculated into a new 24-well plate to begin the culture. Empty lentivector was used as a negative control. Cells infected with knockout lentivector (KO) were called KO cells, and the cells with empty lentivector infection were called Control cells. After selection, the e ciency of infection was veri ed by western blot.
Western blot analysis 3T3-L1 cells were lysed with a cell lysis buffer (Cell Signaling Technology, Danvers, MA) for western blot analysis and extracted protein concentration was determined by the BCA protein assay kit (Pierce, Rockford, IL). Protein (30μg) was subjected to 4-10% SDS-PAGE and transferred to a membrane of PVDF (Amersham Biosciences, UK). The membranes were then blocked for 2-3 hours at room temperature with 5% (w/v) BSA. The membranes were incubated overnight at 4°C with the indicated antibodies, and then for 1 hour with horseradish peroxidase conjugated secondary antibodies., followed by visualization with a chemiluminescence system (ECL, Amersham Biosciences, Buckinghamshire) on ChemiDoc MP System (Bio-Rad, USA), washings performed between incubations.

Statistical analysis
Quantitative real-time PCR results are presented as relative quantity to the control group, using β-actin as internal reference gene. Band densities were corrected for background and then normalized to the control (IRS1 for p-IRS1/AKT for p-AKT/βactin or GAPDH for PPAR-γ) signal of the same lane in western blot experiment. ELISA experiment results are normalized as relative quantity to the control group. All data were viewed as means SEM, and the differences between the two groups were measured using a two-way ANOVA, with a p<0.05 presented statistically signi cant. All the statistical analyses were conducted using GraphPad Prism7 (GraphPad Software, California). At least three biological replicates of each experiment were performed and analyzed.

Results
GPR50 as a novel candidate target in obesity-T2DM animal adipose tissue To systemically screen the new targets from the adipose tissue in obesity-T2DM mice, we compare two groups of male C57BL/6J mice fed standard chow diet or HFD (n=3 from 5 per group) for transcriptome analyses. The microarray information is available in the Omnibus Gene Expression (GSE100028). The results of our previously published paper showed that HFD mice exhibited obesity and T2DM symptoms.(Yao et al. 2019) As shown in Figure 1, we pick out 20 most signi cant differentially expressed genes between the chow and HFD groups. Among these genes, we have found the expression of an orphan GPCR named GPR50 was signi cantly increased in the adipose tissue of HFD group compare to the Chow group. Since the importance of GPCRs for discovering new targets for T2DM treatment and our experiment with GPCR study, we investigated the effect of GPR50 on T2DM.
Expression pro les of GPR50 in adipose tissue and differentiated 3T3-L1 cell line In epididymal adipose tissue, we con rmed the GPR50 expression increasing in the HFD mice at both mRNA and protein levels compare to the Chow mice, which was consistent with the microarray results (Figure 2A, 2B, 2C). Then, we studied GPR50 expression in 3T3-L1 pre-adipocytes during the lipogenic differentiation. The results revealed that after two days of adipogenic differentiation the expression of GPR50 increased in a time-dependent manner and reached peak at 8 days ( Figure  2D, 2E, 2F). These ndings showed that GPR50 may be involved in adipogenic differentiation.

Construction of GPR50 knockout 3T3-L1 cell line
To further investigate the function of GPR50 in T2DM, we constructed GPR50 knock-out cell line through CRISPR/Cas9 gene editing system. The full-length sequence of GPR50 was obtained from GeneBank database. According to the general principles of gRNA design, single-guide RNAs (sgRNAs) targeting GPR50 genes were built by an online software for CRISPR design (http://crispr.mit.edu). Methods and procedures have shown the SgRNA sequence. The transfer vector was successfully constructed by combining sgRNA after annealing connection. To improve transfection e ciency, the transfer vector was packaged in lentiviruses and successfully transfected into 3T3-L1 cells. The stable cell lines of GPR50 gene knockout were screened by puromycin and proved by western blot as shown in Figure 3.
In ammation increased in GPR50 de ciency 3T3-L1 cell line Obesity is related to the pathogenesis of IR in adipose tissue due to chronic low-grade in ammation. Selected markers of adipose in ammation in GPR50 de ciency 3T3-L1 cell line were measured. We used high glucose and palmitic acid (PA) to simulate high-fat diet stimulation cellularly. As shown in Figure 4A-D, at mRNA levels, the expression of proin ammatory cytokines IL-6, MCP-1and IL-1β which associated with IR, were signi cantly higher in GPR50-/-cells compared with control cells which transfected with control plasmid. Moreover, we analyzed the protein secretion of IL-6, MCP-1and IL-1β in the cell supernatant by ELISA. The results were not fully consistent with the Q-PCR. IL-6 protein secretion increased with palmitic acid (PA) stimulate; MCP1 was signi cantly higher in GPR50-/-cells used high glucose and palmitic acid (PA) to simulate. This suggests that MCP-1 maybe more important. And all these ndings indicated that GPR50 could improve adipose tissue in ammation induced by high glucose and PA.
Activation of insulin signaling in GPR50 de cient 3T3-L1 cell IR is a pathological condition frequently linked to T2DM. GPCRs are participating in the development of IR which can lead to T2DM induced by obesity. (Riddy et al. 2018) In order to evaluate the effect of GPR50 in insulin signaling pathway, protein levels of p-IRS-1/IRS-1 p-AKT/AKT were analyzed in GPR50 de ciency 3T3-L1 cells pretreated with glucose or PA by western blot. The protein expressions of p-IRS-1 was signi cantly enhanced in GPR50 de ciency 3T3-L1 cell line ( Figure 5A-B).
Although, high glucose and PA already induced phosphorylation of AKT, GPR50 de cient 3T3-L1 cell line still had signi cantly higher p-AKT than control group ( Figure 5A-C). High glucose and PA treatment led to an interruption of insulin pathway by reducing p-IRS-1, and this effect was reversed when GPR50 was removed ( Figure 5A, B). However, not like the case for p-IRS-1/IRS-1, glucose and PA increased p-AKT/AKT level signi cantly in 3T3-L1 cell, and even high in GPR50 knockout 3T3-L1 cell under the treatments. These ndings indicate that GPR50 might be able to improve high glucose and PA activated insulin signaling through IRS-1 pathway; but, might through other pathway related or not related to GPR50 to increase AKT phosphorylation dramatically which we didn't investigate yet ( Figure 5A, C). To further evaluate the effect of GPR50 on insulin pathway, we challenged the insulin signaling assay with insulin stimulation. As shown in Figure 7, insulin stimulation will increase phosphorylation levels on the AKT and insulin receptor substrate (IRS)1. But the protein expressions of p-IRS-1 and p-AKT was signi cantly decreased in GPR50 knockout 3T3-L1 cell line (Figure 7). Not like the PA and high glucose stimulation, p-IRS-1/IRS-1 and p-AKT/AKT level signi cantly induced in 3T3-L1 cell, and signi cantly inhibited in GPR50 knockout 3T3-L1 cell. These ndings also indicate that GPR50 might be able to inhibit insulin activated insulin signaling through IRS-1/AKT pathway, although PA and insulin stimulate the insulin pathway in opposite ways.
GPR50 enhanced transcription factor PPAR-γ expression PPAR-γ is an essential transcription factor expressed mainly in mammalian adipose tissue and other tissues for cell differentiation. PPAR-γ involves in modulating of insulin sensitivity. As shown in Figure 6, de ciency of GPR50 inhibited PPARγ expression at mRNA and protein level and also related to activated insulin signaling. However, treatment with high level glucose or PA had no effect on the expression of mRNA or protein levels of PPAR-γ in control 3T3-L1 cell. More interestingly, with insulin stimulation, we also found that PPAR-γ expression was inhibited in GPR50 knockout cell lines (Figure 7). These results showed that GPR50 could still regulate the expression of PPAR-γ under insulin stimulation, and the regulation was consistent with PA and high glucose stimulation, which were conducive to improving insulin signaling.

Discussion
GPR50 was screened in HFD-induced mice adipose tissue by gene microarray technology, suggesting that it may be functional in the metabolism and development of T2DM. Previous studies have shown that GPR50 is closely related to energy expenditure and feeding behavior. Finding the new mediators associated with adipocyte low-grade in ammation, su cient to induce IR in obesity and T2DM individuals, is an important work to do.
From this study, it seems that GPR50 has inhibitory effect on in ammation in adipocyte 3T3-L1, and that chronic tissue in ammation is an important cause of IR induced by obesity.(Soedono and Cho 2021) Therefore, we investigated the effects of GPR50 on insulin signaling pathways. Insulin receptor activation contributes to IRS1 phosphorylation, which initiates downstream signaling. When IRS1 is phosphorylated, its downstream signaling capacity is decreased.(Yang, Vijayakumar, and Kahn 2018; Copps and White 2012) According to our experimental results, GPR50 de ciency may signi cantly activate insulin signaling pathway in which IRS1 involved. As shown in Fig. 5, p-IRS1 was signi cantly triggered in the GPR50-de cient group relative to the control group. Interestingly, although the high glucose and PA treatment group reduced p-IRS1 expression by nearly 30 percent, high glucose and PA treatment did not downregulate p-IRS1 expression in GPR50 knockout cells as it did in 3T3-L1 cells. These results indicated that GPR50 can effectively regulate IRS1 phosphorylation, whereas stimulation by high glucose and PA could regulate IRS-1 through GPR50 directly. And we also found that GPR50 de ciency activated p-AKT, although stimulation of high glucose and PA activated p-AKT strongly, in GPR50 knockout 3T3-L1 cells, p-AKT was activated further. These results indicated that GPR50 might mediate the p-IRS1/AKT pathway. PPAR-γ regulates multiple genes transcription involved in the adipose precursor cell differentiation, regulates glucose uptake mediated by insulin, and increases insulin sensitivity. PPAR-γ is closely related to in ammation and insulin resistance. In our study, we found that knout out GPR50 signi cant inhibit PPAR-γ expression, this result suggests that GPR50 is conducive to the activation of PPAR-γ with increased plasma levels of free fatty acids (FFA). PPAR-γ activation in adipose tissue promotes lipid uptake and storage through induction of target genes include aP2, LPL, CD36 and so on. PPAR-γ also in uences the effects of in ammatory cytokines and adipokines, including adiponectin, MCP-1, and resistin. As well as modulating the expression of adipokines, these changes can have bene cial effects on systemic glucose metabolism, including suppressing hepatic glucose uptake and stimulating skeletal muscle glucose uptake. Considering the role in ammation, insulin signaling pathway and PPAR-γ expression play in the pathogenesis of insulin resistance, we tentatively believe GPR50 is involved in regulating in ammation and insulin signaling, therefore, it may contribute positively to insulin resistance.
During our experiments, we found GPR50 may affect 3T3-L1 differentiation ( Figure.A2 A, B unpublished data). After 12 days of lipogenic differentiation in GPR50-de cient 3T3L1 cell lines, intracellular lipid droplet accumulation was signi cantly less than to the control group. As we know, several GPCR activation in adipose tissue were linked to adipocyte function. (Chang et al. 2013) Such as GPR43 can promote 3T3-L1 cell differentiation and lipid droplet accumulation. (Neuhofer et al. 2013) And we know that lipogenic differentiation process affects PPAR-γ expression. So, we detected the expression of PPAR-γ and HSL ( Figure.A2 C, D) (hormone-sensitive lipase, HSL) in mRNA level. Although, we found GPR50 can affect lipogenic differentiation, but we are not yet able to clarify what and how GPR50 plays a role in lipogenic differentiation, and further experiments are needed to demonstrate that GPR50 can indeed in uence lipogenic differentiation and its effect on obesity-T2DM.

Conclusion
We rst identi ed GPR50 as a novel IR target candidate in adipose tissue. We demonstrate that GPR50 can improve high glucose and palmitic acid induced adipose tissue in ammation and may be able to inhibit insulin signaling by regulating the phosphorylation of IRS-1 and AKT. Expression of PPAR-γ transcription factor was also enhanced. Collectively, the above results suggest that GPR50 ameliorate in ammation and is involved in insulin signaling might via promote PPAR-γ expression, but there're still many details need to be elucidated (Figure 8).   Tables   Table 1 is not available with      representative. Western blots are shown along with relative densities, determined using Image J software. Data were analyzed using two-way ANOVA. Results are represented of three independent experiments. Values in bar graphs are means ± SEM. *, ****: p <0.05, p <0.0001 compared to Control; ###, ####: p <0.001, p <0.0001 compared to Control + Glucose, , p <0.0001 compared to Control + PA. &&&&, p <0.0001 compared to GPR50 knockout.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementaryMaterial.docx