GPR162 is a beta cell CART receptor

Summary Cocaine and amphetamine-regulated transcript (CART) is expressed in pancreatic islet cells and neuronal elements. We have previously established insulinotropic actions of CART in human and rodent islets. The receptor for CART in the pancreatic beta cells is unidentified. We used RNA sequencing of Cartpt knockdown (KD) INS-1 832/13 cells and identified GPR162 as the most Cartpt-regulated receptor. We therefore tested if GPR162 mediates the effects of CART in beta cells. Binding of CART to GPR162 was established using proximity ligation assay, radioactive binding, and co-immunoprecipitation, and KD of Gpr162 mRNA caused reduced binding. Gpr162 KD cells had blunted CARTp-induced exocytosis, and reduced CARTp-induced insulin secretion. Furthermore, we identified a hitherto undescribed GPR162-dependent role of CART as a regulator of cytoskeletal arrangement. Thus, our findings provide mechanistic insight into the effect of CART on insulin secretion and show that GPR162 is the CART receptor in beta cells.


INTRODUCTION
In 1995, cocaine-and amphetamine-regulated transcript (CART) was identified as a transcript being upregulated in rat striatum in response to psychostimulant drugs. 1 However, much earlier, a fragment of the CART peptide was identified in extracts of sheep hypothalamus. 24][5] Initially, CART was found to have anorexigenic effects. 6Since then, multiple biological effects of CART have been established. 4,7Among them is the effect of CART on islet hormone secretion from the pancreatic islets.We have established that CART increases insulin secretion, but reduces glucagon secretion, both in vivo and in vitro/ex vivo in rodent and human islets. 8However, the understanding of the molecular basis for the effects has been hampered by the lack of an identified CART receptor.Vicentic et al. suggested in 2005 that the CART receptor is a G proteincoupled receptor as they found radiolabeled CART to be displaced by CARTp55-102, but not by non-active forms of CARTp, and that binding was reduced in the presence of a GTP analog in AtT20 cells. 9Furthermore, by using radiolabeled CART peptide, Kuhar and colleagues were able to detect high-affinity-specific binding in PC12 cells. 10This binding was found to increase upon differentiation of the cells and was inhibited by actinomycin D or cycloheximide treatment, suggesting that the observed binding of CART is dependent on RNA and protein synthesis.It was also reported that PACAP6-38 acts as a CART receptor antagonist and that CARTp-mediated phosphorylation of ERK is inhibited by pertussis toxin. 10hen studying neuropathic pain, Yosten et al. 11 found GPR160 to be necessary for CARTp-mediated cell activation in cell culture.RNA interference of Gpr160 prevented CART from activating c-FOS expression and phosphorylation of ERK in vitro in KATO III cells and PC-12 cells, respectively. 11Also, intrathecal injections of CART induced phosphorylation of ERK and CREB in the dorsal horn of the spinal cord. 11Furthermore, intracerebroventricular administration of a GPR160 antibody to the 4 th ventricle increased food intake in rats. 12However, whether GPR160 is a receptor for CART in the beta cell was not investigated by Yosten et al. 11 or Haddock et al. 12 and subsequently the identity of the beta cell CART receptor remains to be resolved.It should be noted that the data by Yosten et al. and Haddock et al. were recently challenged. 13n the present study, we set out to identify a potential CART receptor in beta cells, using a transcriptomics approach, binding studies, as well as secretion studies, gene expression, and studies showing lack of CART effects after small interfering RNA (siRNA) knockdown.The data presented here point to GPR162 being the beta cell receptor for CART.

RESULTS
RNA sequencing of Cartpt KD INS-1 832/13 cells reveals GPR162 to be a candidate receptor for CART In an attempt to understand the mechanisms behind the insulinotropic effects of CART in pancreatic beta cells, RNA sequencing of INS-1 832/13 cells, subjected to siRNA-mediated knockdown (KD) of Cartpt, was performed.In total, 5027 genes were found to be differentially expressed (2532 genes were upregulated and 2495 genes were downregulated) (Tables S1 and S2).Reasoning that KD of a ligand could affect the expression of its cognate receptor, we primarily focused on orphan receptors that were affected by Cartpt KD.This way we identified Gpr162 as the orphan receptor to be most significantly affected by Cartpt KD and which had the highest fold change in its expression (p = 6.9*10À90 , log fold-change: À1.4).Reduced Gpr162 mRNA expression after Cartpt KD was confirmed with qPCR in INS-1 832/13 cells (Figure 1A) and by Western blotting (Figure 1B).Using a targeted approach, we previously established that Cartpt KD affects a wide range of genes crucial for beta cell function and insulin secretion and production. 14To assess the overall impact of Cartpt KD on cellular function, systematic Gene Ontology (GO) term analysis was performed.This revealed that, among others, processes related to cytoskeletal rearrangement were affected in Cartpt KD cells.Genes including Rhob, Sec23a, Rab3a, Rab7a, Iqgap1, Rabgef1, and Arhgap27 were upregulated by Cartpt KD, while Arhgap36, Rhoq, Iqgap2, Rhobtb3, Rab15, and Rab12 were downregulated by Cartpt KD.Other processes affected by Cartpt KD included sulfur compound biosynthetic process (GO:0044272, upregulated), glycerolipid biosynthesis (GO:0045017, upregulated), phospholipid biosynthetic process (GO: 0008654, upregulated), glycerophospholipid metabolic process (GO:0006650, upregulated), synapse organization (GO:0050808, downregulated), cell junction organization (GO:0034330, downregulated), cell morphogenesis involved in differentiation (GO: 0000904, downregulated), and cell morphogenesis (GO:0000902, downregulated) (Table S3).

Binding assays confirm the interaction between CART and GPR162
Based on the findings from the RNA sequencing of Cartpt KD INS-1 832/13 cells, we set out to investigate the possibility that GPR162 acts as a CART receptor in beta cells.Since CART is known to affect human and murine secretion of islet hormones, we first confirmed GPR162 mRNA expression (Figure 1A) and protein expression (Figures 1B and 1C) in INS-1 832/13 cells.We also show that Cartpt KD in INS-1 832/13 cells affects Gpr162 mRNA expression and GPR162 protein expression (Figures 1A, 1B, and 1C, respectively).Furthermore, we confirmed GPR162 expression in mouse and human islets (Figure 1D).In line with CART affecting both insulin and glucagon secretion, 8,15 GPR162 was localized to beta cells (Figure 1D) and alpha cells (Figure S1D).Moderate GPR162 immunoreactivity was evident in the cytoplasm and more intense immunoreactivity was seen in hitherto unidentified cellular structures.In human islets 82% G 13% of the beta cells were GPR162 immunoreactive and the corresponding number was 74% G 20% for mouse beta cells.A similar expression pattern in beta cells was seen using two different GPR162 antibodies (Figure S1A), and pre-absorption experiments further added to the specificity of the antibody (Figure S1B).Furthermore, Gpr162 KD (using two different siRNAs) caused a robust reduction in GPR162 immunoreactivity in INS-1 832/13 (Figure S1C).Having established the presence of GPR162 in alpha and beta cells, we then employed a multitechnique strategy to test whether CART binds to GPR162 using INS-1 832/13 cells as a model for beta cells.Firstly, we performed coimmunoprecipitation which revealed a stronger binding for GPR162 in the CARTp-treated cells than the binding observed in control cells (Figure 1E).Furthermore, we showed that radioactive 125 I-CARTp55-102 binds to INS-1 832/13 cells, and that the degree of binding was lower in cells treated with Gpr162 siRNA (Figure 1F).Reduced GPR162 protein level after Gpr162 KD was verified by western blot (Figure S1D).Moreover, proximity ligation assay (PLA) confirmed an interaction between GPR162 and CART (Figure 1G).Notably, cells treated with Gpr162 siRNA displayed 90% lower CART-GPR162 interaction, compared with control cells (Figures 1H and 1I).The specificity of the PLA is shown in Figure S2; in a control experiment, no interaction was found between CART and the GIP receptor, a receptor with a known ligand.While these studies were in progress, GPR160 was presented as a CART receptor in KATOIII and PC-12 cells. 11Although GPR160 was not among the differentially expressed genes after Cartpt KD (Tables S1A and S1B), we examined possible expression of GPR160 in islets and interaction between GPR160 and CART in beta cells.We failed to detect GPR160 immunoreactive cells in human islets or GPR160-CART interaction through PLA in INS-1 832/13 cells (Figures S4 and S5, respectively).The specificity of the GPR160 antibody is indicated in Figures S3C-S3F which shows robust expression of GPR160 in tissues previously reported to express GPR160.

Knockdown of Gpr162 reduces CART-induced exocytosis
Having established that CART binds to GPR162, we next tested whether CART-induced biological effects could be reduced after KD of Gpr162 mRNA.First, we employed whole-cell capacitance measurements in INS-1 832/13 cells treated with CARTp55-102 for 1 h.In agreement with our previous findings, 8 control cells responded to 10 À7 M CARTp55-102 with an approximately 1.3-fold increase in capacitance (Figures 2A and 2B).Notably, when Gpr162 mRNA was knocked down, the CART-induced increase in capacitance was abolished (Figures 2A  and 2B).

Knockdown of Gpr162 abolishes the effect of CART on insulin secretion
Gpr162 KD per se resulted in increased insulin secretion at 16.7 mM glucose (with or without IBMX) in INS-1 832/13 cells (Figure 2C; replicated with another siRNA in Figure S1F).Furthermore, the previously established 8 insulinotropic effect of CARTp55-102 on glucose and IBMX-stimulated insulin secretion was abolished in Gpr162 KD cells (Figure 2D).Gpr162 KD per se was also found to increase Ins1, Ins2, and Cartpt  mRNA in INS-1 832/13 cells (Figure 2E; replicated with additional siRNA Figure S1E).The effect on insulin transcription was replicated in human islets treated with siRNA against GPR162 (Figure 2F).

CART and GPR162 knockdown affects density of actin filaments in the beta cell
As Cartpt KD affected genes related to regulation of cytoskeletal processes (Tables S1A, S1B, and S2), we investigated a potential role for CART in regulating actin filament arrangement by culturing INS-1 832/13 cells for 24 h in the presence of 10 À7 M CARTp55-102.Control cells responded to the CARTp-incubation with an approximately 2-fold increase in actin staining, while CARTp was without effect in Gpr162 KD cells (Figures 3A and 3B).Thus, our data show that CART binds to GPR162 and that beta cell effects of CART are reduced or abolished after Gpr162 KD, suggesting a role for GPR162 in CARTp-mediated signaling in beta cells.
To shed light on the function of Gpr162 in beta cells, we performed RNA sequencing of Gpr162 KD INS-1 832/13 cells and cells treated with scrambled siRNA as controls.In total, 4929 genes were affected by Gpr162 KD; 2461 genes were upregulated and 2468 were downregulated (Tables S3A and S3B).Corroborating our qPCR data, Cartpt was markedly upregulated in Gpr162 KD INS-1 832/13 cells (padj = 3.12*10 À5 , 2.25-fold upregulated).In keeping with our hypothesis that GPR162 is a receptor for CART (i.e., a peptide known to affect insulin secretion), the GO term insulin secretion (GO:0050796) was enriched in Gpr162 KD cells (3.1-fold enriched, padj = 0.0106).GO term analysis also revealed processes related to cytoskeletal architecture (e.g., regulation of actin cytoskeleton organization GO:0032956, regulation of actin filamentbased process GO:0032970, regulation of cytoskeleton organization GO: 0051493, and actin cytoskeleton organization GO:0030036) to be enriched by Gpr162 KD.These GO terms mirror the Cartpt KD RNA-seq data and support our observations of CART-induced actin rearrangement.In particular the GO term cytoskeletal protein binding (GO:0008092) contained a number of important cytoskeletal genes (e.g., Afap1, Ckap5, Gsn, Rab3a, Rab3b, Rab11b, and Arpc4) (Table S6).
In an attempt to identify signaling molecules downstream of GPR162, we searched for Cartpt KD-and Gpr162 KD-affected mediators of GPCR signaling in both the RNA-seq dataset.Of interest, G protein alpha 13 (Gna13) was found to be upregulated both by Gpr162 KD (padj<0.05;1.2-fold upregulated) and by Cartpt KD (padj<0.005;1.68-fold upregulated).Notably, GNA13 was found to be co-localized with GPR162 in human beta cells (Figure 4A) and PLA in INS-1 832/13 cells confirmed an interaction between GNA13 and GPR162 (Figure 4B).Gna13 KD caused 1.5-fold increased Cartpt mRNA expression and trends toward increased Ins1 and Ins2 mRNA expression (p = 0.07 and p = 0.08, respectively; Figure 4C).

DISCUSSION
We have previously shown that CART increases insulin secretion. 8However, the identity of the CART receptor mediating this effect in beta cells is unknown to date. 15Here, we provide evidence for GPR162 as a CART receptor in beta cells.We also demonstrate that the insulinotropic effect of CART is GPR162 dependent.
GPR162 was identified as the most CART-affected orphan receptor in INS-1 832/13 cells.Perhaps counterintuitive, Cartpt KD caused reduced expression of GPR162.On the other hand, Gpr162 KD triggered a massive increase in Cartpt mRNA expression.By showing that CART binds to GPR162 and that biological effects of CART are reduced or ablated after Gpr162 KD, we propose that GPR162 acts as a CART receptor in beta cells.We also established expression of GPR162 in a majority of human and mouse beta cells.Moderate GPR162 immunoreactivity was evident across the cytoplasm and more intense immunoreactivity was seen in cellular structures that remain to be deciphered.GPR162 is an orphan receptor belonging to the rhodopsin (class A) family of G protein-coupled receptors, 16 previously shown to be expressed in the brain 17 as well as in the heart and kidney. 18It is highly conserved between mammalian species, and human GRP162 has 95% amino acid identity with rodent orthologs. 19The identification of the CART receptor is long in the making.More than 25 years after the discovery of CART in 1995, 1 no specific CART receptor has been put forward.A few clues as to the properties of this elusive receptor have been put forward over the years.Lakatos et al. presented evidence that exogenous CART was capable of activating AtT20 (a pituitary cell line) 20 and studies using pertussis toxin suggested that the signaling of the CART receptor occurs through G i/o proteins. 20,21G i/o -dependent signaling in beta cells was later questioned as CART was shown to increase cAMP in INS-1 832/13 cells. 22,23n the brain, it was recently suggested through a series of experiments that GPR160, another member of the class A family of GPCRs located on a different chromosome than GPR162, is a CART receptor in the brain. 11,12Information on GPR160 is scarce, but it has been reported to be a potential marker for prostate cancer, 24 where it is involved in apoptosis and cell-cycle arrest. 25From an insulin perspective, it should be noted that it has previously been reported that GPR160 has no effect on C-peptide-induced cFOS expression in KATOIII cells. 26ecently, the notion of GPR160 being the CART receptor in the brain was put into question by Freitas-Lima et al. who investigated the ability of radiolabeled CART to bind in cell lines (THP1 cells and GPR160-transfected U2OS and U-251 MG cells) with high expression of GPR160. 13Also, these authors failed to detect Gpr160 mRNA expression or GPR160 immunoreactivity in the PC12 cell line.The data presented by Freitas-Lima et al. are in line with our inability to detect GPR160 immunoreactivity in human islets or an interaction between CART and GPR160 in INS-1 832/ 13 cells. 13Sanchez-Navarro et al. recently reported that AAV-mediated knockdown of Gpr160 in the dorsal vagal complex in the rat brain partially attenuates the anorexigenic effects of CART on meal microstructure. 27While we do not oppose the possibility that GPR160 is a CART receptor in the brain, further studies are needed to completely resolve the issue of the central CART receptor.In the present study, we propose that GPR162 serves as a CART receptor in the beta cell.This is based on the presence of GPR162 immunoreactivity in islet cells (in both rodent and human pancreas), an interaction between CART and GPR162, and reduced binding after Gpr162 KD.The phenomenon of hormones having multiple, and sometimes even tissue-specific, receptors is not unusual; the plethora of 5-HT receptors being one example 28 and the family of somatostatin receptors being another. 29Further studies are needed to understand whether GPR162 is also involved in mediating CART effects in the brain.It has been shown that GPR162 is widely expressed in, among other neurons, GABAergic neurons in the mouse hippocampus. 17Extensive expression has also been reported in other brain areas, such as the hypothalamus and amygdala, which regulate energy homeostasis and hedonic feeding. 17Of note, the hypothalamus is a brain region with significant expression of CART. 30Furthermore, sequence structure-based phylogeny has suggested GPR162 to be a receptor for adrenaline and noradrenaline. 31However, to our knowledge, this has not yet been functionally proven.
Interestingly, one report presents evidence in line with GPR162 being involved in regulation of glucose metabolism; Caruso et al. showed that GPR162 genetic variants are associated with impairments in glucose homeostasis. 32Genotyping of four single-nucleotide polymorphisms of GPR162 in obese and normal-weight subjects of different ages revealed that variant rs2071081 in intron 4 is associated with impaired insulin levels and HOMA-IR in children. 32t has also been reported that intracerebroventricular injections of a Gpr162 antisense oligodeoxynucleotide for 7 days results in decreased food intake in rats. 32Furthermore, it has been shown that diabetes affects the expression of Gpr162 mRNA expression 18 and that streptozotocin-treated rats displayed reduced Gpr162 mRNA expression in the right ventricle of the heart and in the kidney, while the expression was increased in the brain. 18Also, GPR162 has been shown to increase the radiation-induced DNA damage response and is involved in the activation of the type I interferon system in carcinoma cell lines. 33ur observation that CART affects cytoskeletal rearrangement is novel.5][36][37] In the present study, a large number of genes involved in cytoskeletal rearrangement were affected by both Cartpt KD and Gpr162 KD.We have previously demonstrated that CART causes increased synchronization of Ca 2+ oscillations between beta cells in different parts within the same islet. 8Furthermore, Cartpt KD also affects genes crucial for regulation of exocytosis, 14 a process indeed dependent on actin architecture.The observed effect of CARTp on actin architecture in INS-1 832/13 cells was lost in Gpr162 KD cells, suggesting that CARTp exerts its cytoskeleton-modulating effects through Gpr162.
Along with de-orphanizing GPR162, our data also suggest that GPR162 signals downstream via GNA13.This is based on an observed GPR162-GNA13 interaction in INS-1 cells as well as co-localization of GPR162 and GNA13 in rodent and human beta cells.In addition, knockdown of Gpr162 mRNA or Gna13 mRNA affected CARTp-induced effects. 22GNA13 is reportedly more expressed in the liver than in other insulin target tissues. 38Mice with a liver-specific deletion of Gna13 have been reported to display impaired glucose tolerance and noticeable insulin resistance when fed a high-fat diet, 39 suggesting the involvement of GNA13 in glucose regulatory processes.
We also identified an interaction between CART and GPR162 using three different methods.The effect of Gpr162 KD on this interaction ranged from 25% using radiolabeled CART to 90% using PLA.The observed difference is likely related to different sensitivity of the approaches.On the other hand, Gpr162 KD caused a near-complete abolishment of the CART-induced effect on exocytosis, insulin secretion, and actin rearrangement.The PLA data is in accordance with the observed lack of biological effect, but we have no ready explanation for why we only observed a moderate reduction in CART-GPR162 interaction using radiolabeled CART.However, it is not inconceivable that it is related to the timing of siRNA treatments and/or incubation time during the binding assay.
In summary, we provide evidence for GPR162 as a CART receptor in beta cells.This is based on the expression of GPR162 in beta cells, interaction between CART and GPR162 using several approaches, and reduced or ablated effect of CART after Gpr162 KD.In light of insulinotropic and glucagonostatic actions of CART, the identification of GPR162 as a beta cell CART receptor may open up avenues for novel strategies for the treatment of T2D.

Limitations of the study
It should be mentioned that the present study has a few limitations.Firstly, the majority of experiments were performed in a cell line (INS-1 832/13 cells).Secondly, RNA sequencing does not reveal regulation on the protein level, and as such caution should be exercised when interpreting and extrapolating results obtained on the mRNA level to protein level.However, in this case, extrapolation from mRNA to protein seems to hold true, as mRNA levels and protein levels change in the same direction after knockdown.While the effects of exogenous CARTp were clearly reduced or abolished after Gpr162 KD, Gpr162 KD in itself caused increased insulin secretion.Thus, the potential existence and influence of other contributors and mediators of CART signaling in the beta cell cannot be ruled out.

STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:

Tissue handling
Human or mouse pancreata were either fixed in 4% paraformaldehyde for 24 h (then placed in 70% ethanol) or fixed in Stefanini's solution (2% paraformaldehyde and 0.2% picric acid in 0.1 M PBS, pH 7.2) for 24 h (then rinsed thoroughly in Tyrode's solution containing 10% sucrose).Mice were lightly anaesthesized with Isofluran (Abbott, Berkshire, UK) and then cervically dislocated.Pancreas was dissected out and placed in either Stefanini's or 4% paraformaldehyde.Sections were cut to a thickness of 6 mm.All procedures were approved by the Animal Ethics Committee in Lund and Malmo ¨.
Pancreatic sections were incubated with primary antibodies overnight at 4 C, followed by rinsing in PBS with 0.25% Triton X-100 for 2 X 10 min.Thereafter, secondary antibodies (donkey anti-rabbit Cy2 for GNA13, GPR160 and GPR162, donkey anti-goat Texas Red for GPR162 and donkey anti-guinea pig Alexa Fluor 594 for insulin and glucagon) were applied for 1 h at room temperature.Sections were again rinsed in PBS with 0.25% Triton X-100 and mounted in PBS:glycerol (1:1).Secondary antibodies were purchased from Jackson Immunoresearch (West Grove, PA) (Cy2) and Life Technologies (Alexa Fluor 594).Cy2 and Alexa Fluor 594 were used at a dilution of 1:400.

Radiolabeling of CARTp55-102
125 I was incorporated into the meta-position of tyrosine side chains of CARTp55-102 using oxidative iodination.20 mL of a 300 mM phosphate buffer (pH 7.4) was used to resuspend 1 nmol of the CARTp55-102 peptide, after which 0.4 mCi 125 I (NEZ033A, PerkinElmer, Waltham, MA) was added.A volume of 30 mL 3 mg/mL Chloramine-T in phosphate buffer was added gradually while stirring over 6 min, and the reaction was quenched by adding 400 mL H 2 O with 0.1% trifluoroacetic acid (TFA).The radiolabeled peptide was purified by reversed-phase HPLC on a C18 column and an acetonitrile column.Elution fractions of 1 mL were collected and it was determined where the radiolabeled peptide eluted by measuring g radiation intensity on a PerkinElmer 2470 Automatic Gamma Counter (PerkinElmer).

Radioactive binding
INS-1 cells were seeded to 24-well plates and transfected with Gpr162 siRNA (or control siRNA).Two days after transfection cells were assayed for competition binding for 3 h at 4 C using 10-15 p.m. 125 I CARTp55-102 or unlabeled CARTp55-102 in 0.4 mL of 50 mM HEPES buffer (pH7.4 supplemented with 1 mM CaCl 2 , 5 mM MgCl 2 and 0.5% (w/v) bovine serum albumin).After incubation cells were washed two times in binding buffer supplemented with 0.5 M NaCl at 4 C. Non-specific binding was determined as the binding in the presence of 0.1mM unlabeled CARTp55-102.

Coimmunoprecipitation
INS-1 832/13 cell lysates were incubated with 10 À7 M CARTp55-102 for 30 min at 4 C and then cross-linked with 1 mM BS 3 for 30 min at room temperature, then precipitated using biotin-labelled CART antibody (code H003-62, dilution 1:5000, Phoenix Pharmaceuticals) bound to streptavidin-coated magnetic Dynabeads (Thermo Fisher).Cells lysates were then washed five times in ice-cold PBS and membrane proteins were extracted using MemPER Plus (Thermo Scientific).Protein concentration was then measured using BCA assay (BioRad, Hercules, CA) and 20 mg loaded on a 10% Mini-PROTEAN TGX Stain-Free gel (BioRad).After electrophoresis (200 V for 40 min) and transfer, the PVDF membrane siRNA knockdown in human islets Islets were picked clean and transferred to 35-mm dishes with fresh medium (CMRL1066 supplemented with 10 nM nicotinamide, 10 mM HEPES, 0.25 mg/mL fungizone, 50 mg/mL gentamicin, 2 mM L-glutamine, 10 mg/mL ciprofloxacin and 10% v/v heat-inactivated human serum).On the day of transfection islets (100 islets/dish) were transferred to dishes containing transfection media (CMRL1066 without supplementation).60 nM of siRNA and Lipofectamine RNAiMAX was used.The following day, cells were transferred to fresh supplemented medium.RNA was extracted 96 h after transfection using the NucleoSpin XS kit (Nagel-Macherey).The siRNA for GPR162 (s26015) was purchased from Life Technologies.The sequences for scrambled siRNA were sense: 5 0 -GAGACCCUAUCCGUGAUUAtt-3 0 and antisense: 5 0 -UAAUCACG GAUAGGGUCUCtt-3' (Silencer Select customer designed siRNA, Ambion, Life Technologies).TaqMans used were from Life Technologies and were: Hs00973214 for GPR162, Hs00355773 for INS and Hs004427620 for TBP.HPRT1 (4326321E-1010014) was from Applied Biosystems.Gene expression levels were determined using the 2 ÀDDCt method.

QUANTIFICATION AND STATISTICAL ANALYSIS
Images were quantified using ImageJ and CellSens Elements (Olympus) software.One-way ANOVA with Dunnett's test or Student's t-test was used where appropriate.Data are presented as mean G SEM. Differences were considered statistically significant if p < 0.05.

Figure 3 .
Figure 3. CARTp affects the cytoskeleton in INS-1 832/13 cells in a GPR162-dependent manner (A) Actin staining in INS-1 832/13 incubated for 24 h with CARTp (10 À7 M) (green; nuclei counterstained with DAPI) (panel i and ii).(panel iii and iv) Gpr162 KD INS-1 cells incubated for 24 h with CARTp (10 À7 M) (green; nuclei counterstained with DAPI).(B) Quantification of (A).Scale bars 20 mm n = 6 biological replicates.Data are presented as mean G SEM. * above column indicates comparison with the NC + vehicle group, * in brackets above two columns indicate the comparison made.NC, negative control; KD, knockdown.Data were analyzed using one-way ANOVA with Dunnett's test (Figure 3B).