Microvascular diseases, such as retinopathies, neuropathies, and nephropathies, are a devastating consequence of type 1 and type 2 diabetes. The etiology of diabetes-associated microvascular dysfunction is poorly understood, and, likewise, treatment modalities for these disorders are limited. Interestingly, proinsulin C-peptide has been shown to play a protective role against diabetes-associated complications in experimental animals and in diabetic humans and is thus an attractive therapeutic target. However, an important step in the development of C-peptide-based therapeutics is identification of the C-peptide receptor, which is likely a G protein-coupled receptor (GPCR). Using a unique Deductive Ligand-Receptor Matching Strategy, we sought to determine whether one of the known orphan GPCRs is essential for C-peptide signaling. Knockdown of GPR146, but not GPR107 or GPR160, blocked C-peptide-induced cFos expression in KATOIII cells. Furthermore, stimulation with C-peptide caused internalization of GPR146, and examples of punctate colocalization were observed between C-peptide and GPR146 on KATOIII cell membranes. These data indicate that GPR146 is likely a part of the C-peptide signaling complex and provide a platform for the elucidation of the C-peptide signalosome.
Diabetes is associated with the development of microvascular complications, such as nephropathy, neuropathy, and retinopathy, which can severely and negatively impact quality of life (Wahren et al. 2012). The etiology of diabetes-associated microvascular dysfunction is complex and poorly understood, and thus, therapeutic interventions to prevent or treat these disorders are limited. However, proinsulin C-peptide has emerged as a promising target for the treatment of diabetes-associated microvascular complications (Luppi et al. 2011, Wahren et al. 2012). Indeed, treatment with exogenous C-peptide was shown to prevent or reverse retinopathies (Ido et al. 1997), neuropathies (Wahren et al. 2012), and nephropathy (Wahren et al. 2012) in experimental animals and in diabetic patients. Importantly, plasma levels of C-peptide strongly and negatively correlated with the incidence of microvascular complications in diabetic patients (Bo et al. 2012), indicating that endogenous C-peptide may play a protective role in the vasculature. However, an essential step in the development of C-peptide-based therapeutics is identification of the C-peptide receptor (CpepR).
Although C-peptide has been shown to bind specifically to human cell membranes (Rigler et al. 1999, Luppi et al. 2011), the receptor of C-peptide remains unknown. C-peptide has been shown to initiate intracellular signaling cascades that are usually associated with the activation of a G protein-coupled receptor (GPCR), such as the activation of protein kinase C and the mobilization of calcium (Wahren et al. 2012). Furthermore, the effects of C-peptide were shown to be sensitive to pertussis toxin (Luppi et al. 2011, Wahren et al. 2012), strongly suggesting that the CpepR is a GPCR, perhaps one of the 136 orphan GPCRs cataloged by IUPHAR (Sharman et al. 2013). We have developed a Deductive Ligand-Receptor Matching Strategy that we recently used to identify the cognate receptor of the peptide hormone, neuronostatin (Samson et al. 2008), to be GPR107 (Yosten et al. 2012). In these initial studies, we sought to employ the same methodology to determine whether any of the known orphan GPCRs were essential for transmitting the C-peptide signal in vitro and thus provide a platform for elucidation of the C-peptide signalosome.
Materials and methods
KATOIII cells (ATCC, Manassas, VA, USA) were maintained in Iscove's Modified Dulbecco's Medium (IMDM; ATCC) with 20% fetal bovine serum (FBS) and 1% penicillin/streptomycin. HEK293 cells (a kind gift from M Rauchman, Saint Louis University) were cultured in DMEM (Sigma–Aldrich) with 20% FBS and 1% penicillin/streptomycin. TF-1 cells (ATCC) were maintained in RPMI-1640 (Sigma–Aldrich) with 10% FBS, 1% penicillin/streptomycin, and 2 ng/ml GM-CSF (Sigma–Aldrich). FBS was purchased from ATCC, and penicillin/streptomycin was purchased from Sigma–Aldrich.
Total RNA was collected from KATOIII, HEK293, and TF-1 cells using a Qiagen RNeasy RNA Isolation Kit. RNA was used as a template to produce cDNA with oligo(dT) and MMLV reverse transcriptase (Promega). Primers targeting orphan GPCRs were designed using PrimerQuest Software (Integrated DNA Technologies, Coralville, IA, USA) and specificity was confirmed using NCBI PrimerBLAST (www.ncbi.nlm.nih.gov/tools/primer-blast) and USC Genome Browser (genome.ucsc.edu). Primers (Supplementary Table 1, see section on supplementary data given at the end of this article) were purchased from Integrated DNA Technologies. PCRs were performed in an Applied Biosystems GeneAmp 2400 using the following protocol: 94 °C for 3 min, 94 °C for 30 s, 56 °C for 45 s, 72 °C for 45 s, GOTO Step 2×25 cycles, and 72 °C for 10 min.
Knockdown of orphan GPCRs
KATOIII cells (∼80% confluent) were plated on 12-well, tissue culture-treated plates in IMDM with 20% FBS and no antibiotics and incubated at 37 °C under 5% CO2 for 48 h. Cells were transfected with vehicle (media with Lipofectamine 2000 alone), or vehicle containing 1, 10, or 100 nM siRNA directed against GPR107 (Yosten et al. 2012), GPR146 (Yosten et al. 2012), GPR160, or eGFP as a control (Yosten et al. 2012), using Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions. Constructs of siRNA were designed and validated by Integrated DNA Technologies (Supplementary Table 1, see section on supplementary data given at the end of this article). Five hours later, the medium was changed to serum-free IMDM in preparation for cell response assays, which were performed 24 h following transfection. Concentrations of siRNA used in these studies were determined in preliminary experiments using standard concentration-response protocols to identify the most effective concentrations (i.e. the lowest concentration of siRNA that consistently led to at least 50–60% knockdown; Yosten et al. 2012).
Cell response assays
KATOIII cells were serum-starved overnight (∼18 h) and then exposed to vehicle (serum-free IMDM), 1 nM C-peptide in vehicle, or media containing 20% FBS (positive control) for 1 h at 37 °C under 5% CO2. Cells were lysed and RNA was collected using the Qiagen RNeasy RNA Isolation Kit. Changes in cFos mRNA expression were evaluated by real-time PCR (RT-PCR). Experiments were performed at least three times in triplicate. Human C-peptide was purchased from Phoenix Pharmaceuticals (Burlingame, CA, USA).
Cells treated with C-peptide (see above section, Cell response assays) were lysed and RNA was collected using the Qiagen RNeasy RNA Isolation Kit. cDNA was produced using oligo(dT) and MMLV Reverse Transcriptase (Promega). RT-PCRs were performed using Bio-Rad iQ SybrGreen MasterMix, according to the manufacturer's directions, and primers specific for human cFos and GAPDH (Samson et al. 2008, Yosten et al. 2012) diluted 1:10 in nuclease-free water. RT-PCR was conducted using a Bio-Rad CFX96 Real Time thermocycler according to the following protocol: 95 °C for 5 min, 94 °C for 15 s, 60 °C for 1 min, GOTO 2×30, 72 °C for 10 min, melt curve 65–96 °C; and increment 1 °C for 1 s (Yosten et al. 2012). Data were analyzed using the ΔΔCt method, as described previously (Yosten et al. 2012).
KATOIII cells were cultured on chambered coverslips (Fisher Scientific, Pittsburgh, PA, USA) and treated with 1 or 5 nM C-peptide for 1, 5, 10, 15, 20, or 30 min or left unstimulated. Cells were washed in 1×PBS and fixed with 4% paraformaldehyde for 10 min followed by blocking with 10% donkey and human sera (each 10%, Sigma–Aldrich) in 1×PBS with 0.5% BSA for 30 min. Cells were stained with either antibodies to GPR146 (rabbit polyclonal, Antibodies Online, cat #ABIN213365) at 10 μg/ml, insulin receptor β (Abcam, Cambridge, MA, USA, cat #ab983) at 10 μg/ml, and/or C-peptide (mouse monoclonal, Abcam, cat #ab8297) at 2 μg/ml for 1 h followed by donkey anti-mouse conjugated with Dylight 549 or donkey anti-mouse conjugated with Dylight 488 secondary antibodies (Jackson Immunoresearch, West Grove, PA, USA) also for 1 h. In some cases, wheat germ agglutinin (Molecular Probes, Invitrogen) was used to visualize the plasma membrane. All antibodies were diluted in blocking buffer.
Immunostained KATOIII cells were visualized on a LSM510 confocal microscope platform (Zeiss, Germany) in the Molecular Microbiology Imaging Facility at Washington University in St Louis. Individual cells or groups of cells were visualized both in single plane images as well as Z stacks captured at 0.2 μm intervals. Differential interference contrast was used in some images to visualize the structure of the cells. Images were post processed either in ImageJ (NIH, Bethesda, MD, USA) using global brightness/contrast adjustments or in the Huygens Essentials package (SVI, Hilversum, The Netherlands) using deconvolution algorithms. Where appropriate, Z stacks were reconstructed and a ray-traced surface projection of the three-dimensional relationship of the stacked data was produced (Huygens Essentials software package, SVI). Object colocalization was determined by the number of objects with greater than or equal to two coincident voxels also using the Huygens Essentials software package (SVI).
Data were analyzed using a Student's t-test, ANOVA, or a nonparametric test (Mann–Whitney U test), as appropriate (Zar 1984). The tests used for each experiment are reported in the figure legends. Statistical analyses were performed using SPSS Software (IBM).
In order to identify CpepR candidates, we utilized our Deductive Ligand-Receptor Matching Strategy (Yosten et al. 2012). First, we identified three human cell lines (KATOIII, HEK293, and TF-1) that responded to treatment with C-peptide with an increase in cFos expression (Fig. 1A). These cell lines were screened for the expression of orphan GPCRs (Sharman et al. 2013). Of the 136 orphan GPCRs, 24 were expressed by all the three cell lines (Fig. 1B). In order to generate a ‘short list’ of candidate receptors from the 24 orphan GPCRs identified in our initial screen, we evaluated those 24 receptors based on homology and predicted function using bioinformatic databases, including NCBI (www.ncbi.nlm.nih.gov/), UniProt (www.uniprot.org/), and the PDSP GPCR database (pdsp.med.unc.edu/). Through this process, we eliminated four receptors that had been deorphanized (GPR31 (NM_005299.2; Guo et al. 2011); LGR4 (NM018490; Glinka et al. 2011); LPHN1 (NM_014921; Silva et al. 2011); and Mas1 (NM_002377; Santos et al. 2013)). Eleven receptors were eliminated because they exhibited high homology to structural proteins (GPR125 (NM_145290; Bjarnadottir et al. 2004)), amino acid receptors (MRGPRD (NM_198923) and TAAR6 (NM_175067)), chemokine receptors (FY/DARC (NM_002036) and CMKLR1 (NM_004072)), proton sensors (GPR65 (NM_003608)), sphingosine receptors (GPR63 (NM_030784.2)), and sensory or nociceptive proteins (OPN3 (NM_014322); MRGPRX1 (NM_147199; Dong et al. 2001); MRGPRX4 (NM_054032; Dong et al. 2001); and TAAR8 (NM_053278); Sharman et al. 2013). We next compared the known expression profiles of the remaining orphan GPCRs (Regard et al. 2008) to the expected sites of expression, given C-peptide's known actions (i.e. kidney, fat, skeletal muscle, and pancreas), and generated a short list that included GPR160 and GPR146 as the top two candidates. These receptors, as well as the neuronostatin receptor, GPR107 (Yosten et al. 2012; expressed by KATOIII, HEK293, and TF-1 cells), were knocked down individually in KATOIII cells using siRNA (Yosten et al. 2012). Vehicle-treated (Lipofectamine 2000, no siRNA) or control siRNA (eGFP)-treated cells responded to treatment with C-peptide with a significant increase in cFos expression (Fig. 1C). Knockdown of either GPR107 or GPR160 did not affect C-peptide-induced cFos expression; however, knockdown of GPR146 completely blocked the ability of C-peptide to stimulate cFos expression in those cells (Fig. 1C). Knockdown of GPR146 also abrogated FBS-induced cFos expression in KATOIII cells (Fig. 1D).
KATOIII cells were stained using an antibody directed against the intercellular domain of GPR146. GPR146 could be detected throughout the cell membranes of KATOIII cells (Fig. 2A). Exposure to 1 nM C-peptide appeared to induce internalization of GPR146 over the course of 30 min (Fig. 2B). Furthermore, cells that were exposed to C-peptide exhibited punctate regions of colocalization between C-peptide and GPR146 (Fig. 2C) that increased in number as a function of incubation time (Fig. 2D). Studies from several groups have indicated that C-peptide- and insulin-initiated signaling cascades functionally interact (Wahren et al. 2012). However, KATOIII cells exposed to C-peptide alone exhibited only infrequent, punctate colocalization between GPR146 and the insulin receptor (Fig. 2E).
GPR146 is expressed in three C-peptide-responsive cells lines and appears to be essential for C-peptide-induced cFos expression in KATOIII cells. Furthermore, knockdown of GPR146 reduced FBS-induced cFos expression, indicating that GPR146 is essential for transmitting the signal of a factor present in FBS, such as C-peptide. Stimulation with C-peptide induced internalization of GPR146, and punctate colocalization was observed between C-peptide and GPR146 on KATOIII cell membranes. These initial observations indicate that GPR146 may be part of the C-peptide signalosome and is a potential candidate receptor for proinsulin C-peptide. It should be noted that while a functional interaction between C-peptide and GPR146 exists, we have not demonstrated a direct physical interaction between the two proteins, except by co-immunofluorescence. For this reason, confirmatory experiments must be performed in order to ascertain the specific role of GPR146 in transmitting the C-peptide signal. These necessary confirmatory experiments must include radioligand binding studies, co-immunoprecipitation, and in vivo functional assays. Although our initial observations do not conclusively demonstrate that GPR146 is a CpepR, the identification of GPR146 as essential for C-peptide-dependent signaling provides a platform for elucidation of the C-peptide signaling complex.
GPR146 may contribute to C-peptide signaling in several ways. GPR146 may act as a co-receptor of the CpepR(s), as part of a C-peptide signaling complex or signalosome, or could act as a CpepR itself. Alternatively, GPR146 could be a promiscuous GPCR (activated by multiple ligands) or may interact with modulating proteins, such as RAMPs (Kadmiel et al. 2012), thus modifying receptor specificity. However, knockdown of GPR146 did not alter the responsiveness of KATOIII cells to neuronostatin (Yosten et al. 2012), thus suggesting specificity of GPR146 for C-peptide.
Little is known about GPR146. GPR146 was originally identified as a 150-amino acid, partial sequence, termed PGR8 (Gloriam et al. 2004). In humans, the gene for GPR146, now known to be a 333-amino acid protein, is encoded by a single exon located at 7p22.3 (Gloriam et al. 2004). Orthologs of GPR146 have been identified in monkey, mouse, rat, chicken, zebrafish, and fugu (Gloriam et al. 2004). The rodent orthologs of GPR146 share 74% homology with the human sequence (Gloriam et al. 2004). Interestingly, human and rodent C-peptide sequences only exhibit ∼70% homology. Although over 150 single-nucleotide polymorphisms have been identified in the human GPR146 gene (Sherry et al. 2001), no clinical associations have been reported thus far.
Several lines of evidence suggest that C-peptide- and insulin-initiated signaling cascades functionally interact (Luppi et al. 2011, Wahren et al. 2012). Some studies have reported that the insulin receptor interacts with a GPCR coupled to Gαq (Kisfalvi et al. 2007), and, interestingly, the actions of C-peptide are likely mediated by Gαq (Wahren et al. 2012). These data suggest that the insulin receptor and the CpepR might physically interact. However, we observed only infrequent colocalization of the insulin receptor and GPR146 following stimulation with C-peptide alone. Co-stimulation with C-peptide and insulin may be necessary to induce oligomerization of insulin receptor and GPR146, or GPR146 may interact with another membrane protein to transmit the C-peptide signal. Regardless, elucidation of the C-peptide signalosome is crucial for the development of C-peptide-based therapeutics for the treatment of diabetes-associated microvascular dysfunction. Although the studies presented here do not unequivocally demonstrate that GPR146 is a CpepR, they are an important first step in the elucidation of the C-peptide signaling complex.
This is linked to the online version of the paper at http://dx.doi.org/10.1530/JOE-13-0203.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This work was funded by NIH HL-066023 to W K S and by NEI 5K12 EY016336 to G R K.
Author contribution statement
G R K planned and performed the experiments, contributed to the discussion of the results, and aided in the preparation of the manuscript. L J R planned and performed the experiments. W K S planned the experiments, contributed to the discussion of the results, and edited the manuscript. G L C Y planned and performed the experiments, contributed to the discussion of the results, and prepared the manuscript.
The authors would like to thank Drs Darcy Denner and Michael Rauchmann for their kind gift of HEK 293 cells, as well as for their advice on culturing those cells.
BoSGentileLCastiglioneAPrandiVCanilSGhigoECicconeG2012C-peptide and the risk for incident complications and mortality in type 2 diabetic patients: a retrospective cohort study after a 14-year follow-up. European Journal of Endocrinology167173–180. (doi:10.1530/EJE-12-0085)
KisfalviKReyOYoungSHSinnett-SmithJRozengurtE2007Insulin potentiates Ca2+ signaling and phosphatidylinositol 4,5-bisphosphate hydrolysis induced by Gq protein-coupled receptor agonists through an mTOR-dependent pathway. Endocrinology1483246–3257. (doi:10.1210/en.2006-1711)
SilvaJPLelianovaVGErmolyukYSVysokovNHitchenPGBerninghausenORahmanMAZangrandiAFidalgoSToneviskyAG2011Latrophilin 1 and its endogenous ligand Lasso/teneurin-2 form a high-affinity transsynaptic receptor pair with signaling capabilities. PNAS10812113–12118. (doi:10.1073/pnas.1019434108)
Zar JH 1984 In Biostatistical Analysis