Humanized GPRC6AKGKY is a gain-of-function polymorphism in mice

GPRC6A is proposed to regulate energy metabolism in mice, but in humans a KGKY polymorphism in the third intracellular loop (ICL3) is proposed to result in intracellular retention and loss-of-function. To test physiological importance of this human polymorphism in vivo, we performed targeted genomic humanization of mice by using CRISPR/Cas9 (clustered regularly interspaced short palindromic repeats-CRISPR associated protein 9) system to replace the RKLP sequence in the ICL3 of the GPRC6A mouse gene with the uniquely human KGKY sequence to create Gprc6a-KGKY-knockin mice. Knock-in of a human KGKY sequence resulted in a reduction in basal blood glucose levels and increased circulating serum insulin and FGF-21 concentrations. Gprc6a-KGKY-knockin mice demonstrated improved glucose tolerance, despite impaired insulin sensitivity and enhanced pyruvate-mediated gluconeogenesis. Liver transcriptome analysis of Gprc6a-KGKY-knockin mice identified alterations in glucose, glycogen and fat metabolism pathways. Thus, the uniquely human GPRC6A-KGKY variant appears to be a gain-of-function polymorphism that positively regulates energy metabolism in mice.

To understand the function of this uniquely human KGKY polymorphism in vivo, we have knocked-in the KGKY sequence to replace the ancestral RKLP sequence in the 3rd intracellular loop of the mouse gene. The transgenic mouse model establishes the KGKY insertion/deletion as a gain-of-function polymorphism in vivo.

Results
"Humanized" Gprc6a-KGKY-knockin transgenic mice have enhanced metabolic functions. Our previous data found that the insertion of the KGKY sequence in the mouse GPRC6A (i.e., humanized mouse GPRC6A ICL3_KGKY ) acquires characteristics of the human GPRC6A ICL3_KGKY , namely predominate location to endosome-like intracellular punctuate structures and gain-of-function of mTOR signaling, an evolutionarily conserved pathway in endosomal nutrient signaling 13,32 . We extend these data, by showing that osteocalcin induces an increase in the magnitude and duration of mTOR signaling in cells transfected with the mGPRC6A ICL3_KGKY mutant compared to the wild-type mouse GPRC6A ICL3_RKLP cDNA in vitro ( Figure S1A). In contrast, activation of ERK signaling by osteocalcin was not different between transfected mGPRC6A ICL3_KGKY mutant compared to the wild-type GPRC6A ICL3_RKLP ( Figure S1B).
To test the function of this insertion/deletion in vivo, we "humanized" the mouse Gprc6a gene by using CRISPR/Cas9 system to replace the ICL3_RKLP sequence in the mouse with the ICL3_KGKY sequence (Fig. 1a). Gprc6a-KGKY-knockin mice had similar body weights as Gprc6a-RKLP wild type controls (Fig. 1b). Gprc6a-KGKY-knockin Figure 1. Creation of Gprc6a-KGKY-knockin mice. (a) Structure of Gprc6a exon 6, knockin sequence and genotype primers and Dra I restricted enzyme location. The orange box shows the KY location in blue arrow. The yellow arrow shows the ssOBD-HDR location. The green arrow shows sgRNA location. (b) Comparison of the body weight in wild type and Gprc6a-KGKY-knockin mice at age from 8 to 54 weeks. (c) Site specific brown and white fat content in 20 week-old male mice. Interscapular brown fat (iBAT) and white inguinal fat (iFAT) and epididymal fat (eFAT). Values represent the mean ± SEM. *Significant difference between wild type mouse Gprc6a-RKLP and Gprc6a-KGKY-knockin mice (P < 0.05, Student's t test; n = 6).
Gprc6a-KGKY-knockin mice also had an improved glucose tolerance test (GTT) as shown by lower blood glucose concentrations compared to controls at different time points after glucose injection (Fig. 3a). Accordingly, the net area under the curve, which represents the variation in glucose concentration from baseline over the test duration, was smaller in Gprc6a-KGKY-knockin mice.
Gprc6a-KGKY-knockin mice lowered blood glucose in response to insulin, but they maintained a higher blood glucose compared to controls throughout most of the insulin tolerance test 33 (Fig. 3b). The pyruvate tolerance test (PTT) assesses the effects of glucose production through gluconeogenesis. Gprc6a-KGKY-knockin mice had a greater increase in blood glucose in response to pyruvate (Fig. 3c) compared to wild-type (Gprc6a-RKLP ) mice, consistent with higher gluconeogenesis in Gprc6a-KGKY-knockin mice.
Next, we compared the serum insulin and FGF-21 response in Gprc6a-KGKY-knockin mice with the pharmacological effects of GPRC6A activation by its ligand Ocn. We found that Ocn administration resulted in increases in both insulin and FGF-21 circulating levels ( Fig. 4a, b). In addition, Ocn stimulated FGF-21 message levels in the liver (Fig. 4c).
Since FGF-21 is produced by the liver, we next examined additional hepatic parameters in Gprc6a-KGKY-knockin mice. We found slight reductions in liver cholesterol content but no differences in liver triglyceride or glycogen content in Gprc6a-KGKY-knockin mice (Fig. 5a-c). Liver glucose-6-P levels were significantly elevated in in Gprc6a-KGKY-knockin mice compared to wild-type mice (Fig. 5d), consistent with GPRC6A regulating glucose metabolism in the liver.   Table S1 for genes shown in the heat map in Fig. 6b). Biological process (GO), KEGG and mammalian phenotype enrichment analysis of the DEGs revealed that Gprc6a-KGKY-knockin in liver resulted in differences in lipid and glucose metabolism ( Fig. 6c-e). Genes induced in Gprc6a-KGKY-knockin mice included genes involved in lipid homeostasis (6 genes), localization (20 genes), modification (7 genes), transport (14 genes), storage (10 genes), and fatty metabolic process (13 genes) (Fig. 6c). The complete list of up and down regulated genes are shown in Tables S2 and S3 in Supplemental materials.

Discussion
In the current studies, we used fine-scale genomic humanization of the GPRC6A mouse 3rd ICL to create a transgenic mouse with knock-in of the uniquely human KGKY polymorphism (Gprc6a-KGKY-knockin mice) to determine if this polymorphism has a critical impact on the function of this G-protein coupled receptor in vivo. We found that Gprc6a-KGKY-knockin mice differed from wild type mice expressing the ancestral RKLP variant by the presence of a lower circulating blood glucose, improved glucose tolerance test, increased serum insulin and FGF-21 concentrations. We also observed higher glucose levels in Gprc6a-KGKY-knockin after either pyruvate or insulin administration. Thus, the GPRC6A-KGKY polymorphism may enhance liver gluconeogenesis and glucose production as well as increase peripheral glucose utilization. We also observed a decrease white fat mass in Gprc6a-KGKY-knockin mice, consistent with either direct or indirect effects of GPRC6A to regulate adipocyte function. Overall, these findings are consistent with the KGKY variant being a gain-of-function polymorphism. Indeed, many of these phenotypic features mimic the effects of pharmacological activation of GPRC6A by Ocn treatment, which has been shown to stimulate insulin production by pancreatic β-cells, to enhance glucose and FGF-21 production by hepatocytes, to enhance glucose uptake by skeletal muscle without affecting insulin sensitivity, and to attenuate high fat diet induced hepatosteatosis in mice [21][22][23]29,[34][35][36] . A gain-of-function is also consistent with the recent in vitro findings showing that both hGPRC6A ICL3_KGKY and humanized mouse mGPRC6A ICL3_KGKY are retained intracellularly in ligand naive cells, but exhibit enhanced signaling responses in response to ligand stimulation 13,32,36 .
GPRC6A is expressed in several tissues, including β-cells, skeletal muscle, adipocytes and liver 17,29 , where it has direct effects to regulate glucose and fat metabolism. In this regard, GPRC6A is shown to have function in β-cell 8,9,18,19 , skeletal muscle 19 , testes 21 , skin 11 , adipocytes 6,19 , and fibroblasts 37 . Expression of the GPRC6A-KGKY Our data suggests that GPRC6A also has important functions in the liver. Though additional studies in hepatocytes are needed, the hepatic transcriptome analysis suggests that the GPRC6A-KGKY variant may directly affect hepatic glucose metabolism by decreasing glycolysis and increasing gluconeogenesis. GPRC6A also likely regulates hepatic glycogen metabolism by increasing glycogenesis and glycogenolysis. Our findings also implicate GPRC6A in hepatic fat metabolism where the alterations in gene expression suggests an effect to decrease fatty acid uptake and fatty acid synthesis, and to increase β-oxidation. Overall, these changes on glucose and lipid metabolism in the liver are consistent with the effects of Ocn activation of GPRC6A in the liver to prevent high-fat diet induced hepatosteatosis 17,29,36,38 and effects of Ocn to stimulate hepatic glucose production and skeletal muscle glucose uptake and utilization through GPRC6A-dependent mechanisms 22 . Another mechanism whereby GPRC6A regulates systemic energy homeostasis is through the release of an ensemble of hormones, including insulin secretion in pancreatic β-cells 3,7,9,[17][18][19] , testosterone (T) production in Leydig cells 2,4,21,32 , IL-6 secretion in skeletal muscle 20,22 , adiponectin 6   www.nature.com/scientificreports/ in Gprc6a-KGKY-knockin are consistent with GPRC6A stimulation of insulin secretion. In addition, we show for the first time that the hormone FGF-21 is regulated by GPRC6A. We have the novel finding of elevated FGF-21 levels in Gprc6a-KGKY-knockin mice. Moreover, treatment of wild type mice with the GPRC6A ligand Ocn resulted increased FGF-21. Since Ocn is purported to be the cognate ligand for GPRC6A [4][5][6][7][8][9]21 , these findings are consistent with direct hepatic effects of GPRC6A to regulate FGF-21 production. Some of the metabolic alterations in Gprc6a-KGKY-knockin mice may be due to FGF-21. FGF-21 has paracrine effects in the liver to regulate hepatic lipid oxidation, triglyceride clearance, ketogenesis, and gluconeogenesis 42,43 , as well as systemic effects to increase fat browning, glucose and fatty acid utilization and insulin sensitivity in muscle, to increase insulin synthesis 44 , and central nervous system actions to regulated energy intake and sugar consumption 38 . Limitations of our studies include the lack of information on the effects of dietary fat and calorie intake on the observed phenotype in Gprc6a-KGKY-knockin mice. The current study also addresses the controversy surrounding the function of the ICL3_KGKY polymorphism in GPRC6A in humans, which has been reported to be a hypomorphic by some investigators 26,31 . Our findings showing a gain-of-function in GPRC6A-KGKY mice are consistent with in vitro data showing that the endogenous human GPRC6A containing the KGKY variant in the 3rd intracellular loop is a gain-of-function polymorphism in PC-3 cells in vitro 13,32 . The GPRC6A-KGKY variant as a risk factor for metabolic syndrome, however, has not been well studied in human populations, and case controlled genetic associative studies have been inadequately powered to reach a definitive conclusion 26 ; although the P91S (rs2274911) SNP is associated with insulin resistance 30,45 . Clinical association studies, however, showing that levels of Ocn, are inversely associated with glycemic status and insulin 46 , body mass index, fasting glucose and insulin, triglycerides, and leptin, and positively correlated with adiponectin in humans 33,47 , suggest that GPRC6A is functional in humans. Additional studies are needed to understand the differential functions of GPRC6A-KGKY GPRC6A-RKLP variants in humans. If GPRC6A is an important regulator of glucose and fat metabolism in humans, the unequally distributed RKLP and KGKY polymorphisms may impact racial differences in energy metabolism and help to explain the large variations in serum FGF-21 levels 48 .
In conclusion, the emerging direct functions of GPRC6A in the liver and other tissues, its ability to coordinate the release of an ensemble of metabolically active hormones, and the ability of multiple ligands, including amino acids, cations, Ocn, T and certain natural products to activate GPRC6A, suggests activation of this G-protein coupled receptor provides a new schema for understanding and manipulating energy metabolism. Collectively, the functions of GPRC6A may be to directly and indirectly coordinate the anabolic responses of liver, muscle, fat in response to diverse hormonal and environmental factors. If so, activation of GPRC6A may provide a target to treat metabolic syndrome (MetS), type 2 diabetes (T2D), and non-alcoholic fatty liver disease (NAFLD). To this end, small molecule agonists for GPRC6A have recently been developed that lower glucose in mice 49 that may serve as therapeutic leads to develop GPRC6A agonists.

Methods
Animals. Gprc6a-KGKY-knockin mice were generated by sgRNA/CRISPR-Cas9 target site in proximity to the RKLP → KY mutation site in The University of Alabama at Birmingham. A sgRNA targeting the site immediately preceding the RKLP region (GCT TTG TAT TTG CATT-CAAG<GGG>) was selected for co-injection with Cas9 protein and a single stranded DNA oligonucleotide (ssODN-HDR; MmGprc6aHDRssODN: GGT AGT ATA GAC AGG GAT GAA TGT GAT CCA AGC TAT GAA GTA AAT CAG CAT CCC AAA GGT CAG AAC TTG GCT TCG TTG TAA TTC TCgtaTTT CCC tTTaAAT GCA AAT ACA AAG CAA ATG AAG GCC AGA ACT GTG ATG TAG CCC AGC ATG GTA CCA AAT GCC AGT GCT GAC CCC TCC TCA CAT TCC AGG AT) repair template. The ssODN was modified to replace the mouse-specific RKLP coding sequence to the human-specific KGKY coding sequence in ICL3, and also to ensure that the modified sequence is not targeted by the CRISPR/sgRNA. PCR genotyping using MmGprc6a-gen-F2 and MmGprc6a-gen-R1 primer set. The knockin of the human GPRC6A KY polymorphism PCR product cut by Dra I resulted in 335 and 242 bp bands.
Mice were maintained and used in accordance with recommendations as described (National Research Council. 1985; Guide for the Care and Use of Laboratory Animals DHHS Publication NIH 86-23, Institute on Laboratory Animal Resources, Rockville, MD, USA) and following guidelines established by the University of Tennessee Health Science Center Institutional Animal Care and Use Committee. The animal study protocol was approved by the institutional review boards at University of Tennessee Health Science Center Institutional Animal Care and Use Committee.
Site-directed mutagenesis. Site-directed mutagenesis was conducted using the QuickChange mutagenesis kit (Stratagene) according to the protocols of the manufacturer using a primer set including mGPRC6A. KYfor: gcattcaagggcaaatatgagaattacaacgaagcc and mGPRC6A.KYrev: ggcttcgttgtaattctcatatttgcccttgaatgc. Mutant was constructed on wild type mouse GPRC6A (mGPRC6A ICL3_KGRKLP in pcDNA3 vector) background. The mGPRC6A ICL3_KGKY was confirmed by DNA sequencing. HEK-293 cells were transfected with mGPRC6A ICL3_KGRKLP or mGPRC6A ICL3_KGKY plasmids using TransFast Transfection Reagent (Promega) for 48 h, then the transfected cells were selected by G418.
Measurement of total and phospho-eRK and -mtoR by elisa analysis. Briefly, HEK-293 cells transfected with wild type mouse GPRC6A (mGPRC6A ICL3_KGRKLP ) and mGPRC6A ICL3_KGKY mutant cDNA plasmids were starved by overnight incubation in serum-free DMEM/F12 containing 0.1% bovine serum albumin (BSA) and stimulated with various ligands at different doses. ERK activation were assessed the time as indicated after treatment by using ERK1/2 (phospho-T203/Y204) ELISA Kit (Invitrogen) corrected for the amount of total Library preparation and sequencing. The library preparation and sequencing were carried out by Novogene Co., Ltd. (Chula Vista, CA, USA). Briefly, mRNA was first enriched using oligo 52 beads and fragmented randomly by adding fragmentation buffer. Then the cDNA was synthesized by using mRNA template and random hexamers primer, after which a custom second-strand synthesis buffer (Illumina; Mountain View, CA, USA), dNTPs, RNase H, and DNA polymerase I were added to initiate the second-strand synthesis. Second, after terminal repair, a ligation and sequencing adaptor ligation, the double-stranded cDNA library was completed through size selection and PCR enrichment. The library quality was accessed by Qubit 2.0, Agilent 2100, and Q-PCR. The DNA from the qualified libraries are fed into Illumina sequencers at an average depth of 42 million reads per sample.

RNA-seq data analysis.
Raw reads were quality filtered with NGS QC Toolkit version 2.3 53 to remove adaptor contaminated reads or reads containing > 20% low-quality (Q < 20) bases. Filtered reads were aligned to the mouse reference sequence (GRCm38/mm10) using STAR aligner version 2.5.0a 54 . Raw read count was quantified across all annotated mm10 transcript using FeatureCounts version 1.6.3 implemented in Subread package 55 , then submitted to DeSeq2 version 1.10.1 56 to identify the differentially expressed genes between Gprc6a-KGKY-knockin and WT groups (three replicates for each group). Differentially expressed genes were defined as having an adjusted P value < 0.05. Gene set enrichment analysis for Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Mammalian Phenotype Ontology were analyzed with WebGestalt 57 (https ://www.webge stalt .org/) with default parameters. A threshold of FDR < 0.05 was used to determine the significant enriched terms. Volcano plot and heatmaps were drawn using R program.