GPR116, an Adhesion G-Protein–Coupled Receptor, Promotes Breast Cancer Metastasis Via the Gaq-P63rhogef-Rho Gtpase Pathway

Published OnlineFirst September 5, 2013; DOI: 10.1158/0008-5472.CAN-13-1049

Cancer Molecular and Cellular Pathobiology Research

GPR116, an Adhesion G-Protein–Coupled Receptor, Promotes Breast Cancer Metastasis via the Gaq-p63RhoGEF-Rho GTPase Pathway

Xiaolong Tang1, Rongrong Jin1, Guojun Qu1, Xiu Wang1, Zhenxi Li1, Zengjin Yuan1, Chen Zhao1, Stefan Siwko2, Tieliu Shi1, Ping Wang1, Jianru Xiao1, Mingyao Liu1,2, and Jian Luo1

Abstract Adhesion G-protein–coupled receptors (GPCR), which contain adhesion domains in their extracellular region, have been found to play important roles in cell adhesion, motility, embryonic development, and immune response. Because most adhesion molecules with adhesion domains have vital roles in cancer metastasis, we speculated that adhesion GPCRs are potentially involved in cancer metastasis. In this study, we identified GPR116 as a novel regulator of breast cancer metastasis through expression and functional screening of the adhesion GPCR family. We found that knockdown of GPR116 in highly metastatic (MDA-MB- 231) breast cancer cells suppressed cell migration and invasion. Conversely, ectopic GPR116 expression in poorly metastatic (MCF-7 and Hs578T) cells promoted cell invasion. We further showed that knockdown of GPR116 inhibited breast cancer cell metastasis in two mammary tumor metastasis mouse models. Moreover, GPR116 modulated the formation of lamellipodia and actin stress fibers in cells in a RhoA- and Rac1-dependent manner. At a molecular level, GPR116 regulated cell motility and morphology through the Gaq-p63RhoGEF-RhoA/Rac1 pathway. The biologic significance of GPR116 in breast cancer is substantiated in human patient samples, where GPR116 expression is significantly correlated with breast tumor progression, recurrence, and poor prognosis. These findings show that GPR116 is crucial for the metastasis of breast cancer and support GPR116 as a potential prognostic marker and drug target against metastatic human breast cancer. Cancer Res; 73(20); 1–13. 2013 AACR.

Introduction leading to metastasis (2). The Rho family GTPases, including Breast cancer is the leading cause of cancer-related mor- RhoA, Rac1, and Cdc42, are essential for cytoskeletal dynam- tality in females worldwide (1). Death caused by breast ics, especially cancer cell migration (3, 4). Among them, fi cancer primarily results from cancer cells metastasizing to RhoAregulatestheactincytoskeletoninstress ber forma- fi distal organs such as lung, bone, liver, or brain. Cell motility tion (5). Rac1 modulates a meshwork of actin laments at is regulated by the cytoskeleton, and disruptions in cyto- the cell periphery to produce lamellipodia and membrane fl skeletal regulation are a means by which cancer cells devel- ruf es (5). Activated Rho GTPases induce multiple down- op inappropriate migratory and invasive characteristics, stream signaling pathways during cancer cell migration, such as ROCK1/2 signaling (4). All Rho GTPases cycle between an inactive guanosine diphosphate (GDP)-bound Authors' Affiliations: 1East China Normal University and Shanghai and an active guanosine triphosphate (GTP)-bound state. Changzheng Hospital Joint Research Center for Orthopedic Oncology, The process is accelerated by a large family of Rho guanine Shanghai Key Laboratory of Regulatory Biology, Institute of Biomedical Sciences and School of Life Sciences, East China Normal University, nucleotide exchange factors (Rho GEF; ref. 6). One such Rho Shanghai, China; and 2Center for Cancer and Stem Cell Biology, Alkek GEF is p63RhoGEF (also called GEFT), which is known as an Institute of Biosciences and Technology, Texas A&M University Health – Science Center, Houston, Texas effector of the heterotrimeric guanine nucleotide binding protein GaqandtherebylinksGaq-coupled receptors Note: Supplementary data for this article are available at Cancer Research a Online (http://cancerres.aacrjournals.org/). (GPCR) to the activation of the Rho GTPases (6). G q- p63RhoGEF-Rho GTPase has been reported as a protein X. Tang and R. Jin contributed equally to this work. complex (6), which is involved in multiple physiologic func- J. Xiao and M. Liu are co-senior authors of this article. tions such as vascular smooth muscle contractility (7) and cell movement (8). Corresponding Author: Jian Luo, The Institute of Biomedical Sciences, – School of Life Sciences, East China Normal University, 500 Dongchuan G-protein coupled receptors (GPCR or GPR) are integral Road, Shanghai 200241, China. Phone: 86-212420-6947; Fax: 86-215434- membrane proteins participating in the transmission of 4922; E-mail: [email protected] signals from the extracellular environment to the cytoplasm. doi: 10.1158/0008-5472.CAN-13-1049 A variety of external stimuli, including neurotransmitters, 2013 American Association for Cancer Research. hormones, phospholipids, growth factors, and proteases, can

www.aacrjournals.org OF1

Tang et al.

activate GPCRs (9, 10). Activated GPCRs typically transduce were purchased from Santa Cruz Biotechnology; and mono- signals to effector proteins through their heterotrimeric G clonal anti-Rac1 (23A8) antibody was obtained from Millipore, proteins. These signaling pathways represent important and GPR116 (Ab111169) antibody used to conduct immuno- specific targets for a variety of physiologic functions and histochemistry on human breast cancer tissue was purchased therapeutic approaches, ranging from the control of blood from Abcam. pressure, allergic response, kidney function, and hormonal disorders to neurologic diseases (11). As a result, approxi- GPR116 mouse monoclonal antibody production mately 50% of marketed pharmaceuticals target human The fragment of 27 amino acid residues from the N terminal GPCRs or their signaling pathways (12). However, there are region of human GPR116, with the sequence of "WNYES- few GPCRs that are cancer therapeutic targets. The adhesion TIHPLSLHEHEPAGEEALRQKR," was synthesized and used as GPCRs are the second largest GPCR family with 33 members, immunogen. BALB/c mice were immunized by intraperitoneal most of which are orphan receptors (13). This family is very injection with 100 mg of immunogen (dissolved in PBS) in a 50% special because of the so-called adhesion domains in their emulsion with complete Freunds adjuvant on day 1 and long N-terminal extracellular regions. Adhesion domains, boosted by intraperitoneal injection on days 21 and 36 with which are thought to have adhesive properties, mainly exist immunogen in incomplete Freunds adjuvant. After 60 days, in some adhesion molecules such as integrins, cadherins, freshly harvested spleen cells obtained from the immunized and selectins. However, limited studies have shown that mice were prepared for cell fusion to generate hybridoma lines, adhesion GPCRs are involved in the regulation of cell which were subsequently screened by ELISA and immuno- adhesion, motility, embryonic development, and immune blotting. The specificity of the antibody for immunoblotting response (12–14). GPR116, named Ig-hepta in the rat, is a was examined using a competition strategy with the antigen member of the adhesion GPCR family. Previous studies peptide. showed that rat Gpr116 forms a dimer and is cleaved at multiple sites in the N-terminal region yielding several Immunofluorescence staining fragments with unknown functions (15–17). Recent evidence After plating on coverslips and culturing in 24-well plates, from an adipocyte-specific GPR116-knockout mouse model cells were fixed in 4% paraformaldehyde for 15 minutes, shows that GPR116 plays a critical role in adipocyte biology permeabilized with 0.1% Triton/PBS for 5 minutes at room and systemic energy homeostasis (18). Further research on temperature, and blocked with 1% bovine serum albumin for lung development indicates that GPR116 regulates pulmo- 30 minutes. Then the cells were incubated with primary nary surfactant pool size and is important for lung surfac- antibody for 1 hour, followed by incubation with the appro- tant homeostasis (19, 20). These discoveries implicate the priate secondary antibody or Alexa Fluor 594 phalloidin for 30 physiologic and pathologic functions of GPR116. minutes at room temperature. Finally, cells were photographed As the adhesion molecules (integrins, cadherins, and selec- with a confocal laser scanning microscope (Leica TCS SP5). tins, etc.) have an important function in cancer metastasis (12), The concentrations of antibodies or dyes were: anti-paxillin it is reasonable to speculate that adhesion GPCRs also have (BD, 610051), 1 mg/mL; Alexa Fluor 488–conjugated anti-rabbit vital functions in cancer progression and metastasis. In this IgG (Invitrogen), 1 mg/mL; and Alexa Fluor 594 phalloidin study, we screened the adhesion GPCR family and identified (Invitrogen), 1:500 dilutions. GPR116 as a novel regulator of breast cancer cell migration and invasion in vitro and metastasis in vivo by modulation of the GTPase activity assay Gaq-p63RhoGEF-Rho GTPases signaling pathway. Moreover, RhoA activity in cell lysates was measured using the fusion GPR116 is strongly correlated with breast cancer progression, protein of glutathione S-transferase (GST) and the RhoA- metastasis and poor prognosis by analysis of human breast binding domain of Rhotekin (RBD). For Rac1 and Cdc42 cancer clinical samples. Our results suggest that GPR116 is a activities, GST-PBD, a GST fusion protein with the binding potential regulator for cancer metastasis that may have value domain of PAK, was used. Briefly, both fusion proteins as a prognostic marker. expressed in BL21 were purified with glutathione Sepharose beads (GE Healthcare). Cell lysates were prepared with RhoA lysis buffer [50 mmol/L Tris–HCl, pH 7.4, 150 mmol/L NaCl, Materials and Methods 5 mmol/L MgCl2, 1 mmol/L dithiothreitol, 1 mmol/L phenyl- Cell lines, reagent, and antibodies methylsulfonyl fluoride (PMSF), 1% Triton X-100, proteinase All human breast cancer cell lines were obtained from the inhibitor cocktail] or Rac1 lysis buffer (20 mmol/L HEPES- Chinese Academy of Sciences Committee Type Culture Col- NaOH, pH 7.5, 120 mmol/L NaCl, 10% glycerol, 2 mmol/L lection Cell Bank. The cell lines were authenticated by the EDTA, 0.5% Nonidet P-40, 1 mmol/L PMSF, proteinase inhib- Chinese Academy of Sciences Committee Type Culture Col- itor cocktail] and incubated with GST-RBD or GST-PBD pre- lection Cell Bank. All the culture media and reagents were from coupled to glutathione Sepharose beads at 4C for 45 minutes Gibco. Antibodies against RhoA (67B9), Cdc42 (11A11), MLC2 to precipitate GTP-bound RhoA, Rac1, or Cdc42. The precip- (#3672S), p-MLC2 (#3671S), Cofilin (#3312), and p-Cofilin itant was boiled in Laemmli sample buffer, and eluted product (77G2) were obtained from Cell Signaling Technology; mono- was separated on a 12% SDS-PAGE gel. Immunoblotting clonal anti-b-actin (A2228) and anti-Flag (F7425) antibodies was used to detect the protein level of total and GTP-bound were from Sigma; antibodies to GNAq (E17) and GNA11 (D17) RhoA, Rac1, or Cdc2.

OF2 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

Wound-healing migration and Transwell invasion GAATGCGGCCCGGAGGTCACGTAGGTTGG -30. The PCR assays product was cloned into the lentiviral vector pLVX-IRES- For the migration assay, cells were cultured in 12-well plates ZsGreen1. The constitutively active mutants of RhoA, Rac1, coated with rat collagen type 1 (10 mg/mL; BD). When 90% and Gq were cloned as previously described (21). Full-length confluent, cells were starved for 12 hours. Then, a wound RGS2 and the RGS domain of p115RhoGEF (including amino was scratched in the center of the cell monolayer by a sterile acid residues 1–252) were amplified by PCR and cloned into plastic pipette tip. The debris was removed by washing with the EcoR1/BamH1 site of p3FLAG-CMV-10. PBS twice. After 12-hour incubation, the wound was photographed under an inverted microscope. For the invasion assay, siRNA and short hairpin RNA preparation and Transwell filter membranes were coated with 50 mL Matrigel transfection (BD Biosciences; diluted 1:1 in DMEM) and incubated at The nontargeting control short hairpin RNA (shRNA; room temperature for 30 minutes before cells were seeded. shNTC, 50GTTCTCCGAACGTGTCACGTT30) and 2 GPR116 A total of 5 104 cells suspended in medium without FBS were shRNAs (shGPR116 #1, 50CGGCTGAAGAATACACTGTTA- plated on the upper chamber membranes (8-mm pore size, C30; shGPR116 #2, 50GGACTACAACTCCTTTCAAGCC30) 6.5-mm diameter; Corning). The insert was then transferred targeting GPR116 mRNA (ENSEMBL ID: ENST00000283296 to a well containing 600 mL medium with 10% FBS. Cells were and ENST00000265417) were cloned into the lentiviral vec- incubated for 8 to 24 hours at 37C in a humidified incubator. tor pLKO.1-puro. The nontargeting control siRNA (scram- Noninvasive cells were removed by swiping the top of mem- bled siRNA, 50UUCUCCGAACGUGUCACGUTT30) and GPR116 brane with cotton swabs, and invasive cells were stained with siRNA oligos (GPR116 siRNA, 50CCUUGUGUUCCAUAUCA- crystal violet and counted. UATT30; si30-UTR, 50GACAGACCCUGUCAUAUAU30) against GPR116 were obtained from GenePharma. Cells were trans- Cell proliferation and apoptosis assays fected with appropriate plasmids or siRNAs using Lipofecta- Cell growth was monitored by cell counting on the indicated mine reagent (Invitrogen) according to the manufacturer's days after seeding. Briefly, cells were plated in 6-well plates at a protocol. density of 3 104 cells per well. On the indicated day, cell number was counted. Cellular apoptosis was monitored by the Lentivirus preparation, lentiviral infections, and stable FITC Annexin V Apoptosis Detection Kit (BD) according to the cell lines manufacturer's protocol. Briefly, cells stained with fluorescein To produce viral constructs, 10 mg of lentiviral vector isothiocyanate (FITC) Annexin V and propidium iodide (PI) (pLKO.1-puro and pLVX-IRES-ZsGreen1), 10 mg of pSPAX2, were detected by fluorescence-activated cell sorting (FACS). and 5 mg of pMD2G were cotransfected into human embryonic The PI and FITC double-negative cells were counted as healthy kidney 293T cells using Lipofectamine reagent (Invitrogen). cells. Supernatants were collected 48 hours after transfection and filtered through a 0.45-mm membrane (Millipore). Then, cells Reverse transcription PCR and real-time reverse were directly infected using 8 mg/mL of polybrene for 6 to transcription PCR 24 hours. To get stable knockdown cells, growth medium Total RNA was isolated from cells and cDNA was synthe- containing 1 mg/mL of puromycin (Sigma) was added for sized using the Prime Script RT-PCR Kit from Takara. Two-step 1 week before use in experiments. To obtain a high-expression real-time PCR was carried out using SYBR Green Mix (Takara). population, cells infected with vector pLVX-IRES-ZsGreen1 The expression of the b-actin gene in each sample was used as were sorted by FACS in the GFP channel. an internal control. All the primers used are listed in Supple- mentary Table S1. Breast cancer cell migration following ectopic GPR116 expression Immunoblotting Breast cancer MDA-MB-231 cells were transfected with Samples prepared in Laemmli buffer were separated using si30-UTR, which targets the 30-untranslated region (UTR) of SDS-PAGE. Briefly, the separated proteins were transferred to GPR116 transcripts (ENSEMBL ID: ENST00000283296 and nitrocellulose membranes (Millipore). After blocking for 1 hour ENST00000265417). After 12 hours, the cells were infected in 5% non–fat dried milk and PBS (w/v), membranes were with a lentivirus for expression of the GPR116 coding se- incubated with primary antibody at 4C overnight. The mem- quence and then were sorted by FACS in the GFP channel to branes were washed three times with PBST and incubated with ensure 100% infection because the expression vector (pLVX- infrared dye–labeled secondary antibody for 45 minutes. IRES-ZsGreen1) expresses an extra GFP protein. Transwell Immunoblotting was visualized using the LI-COR Odyssey chamber assay was then conducted as previously described. System. In vivo mouse experiments and analyses Plasmid constructs All animal work was conducted in accordance with accepted Full-length coding DNA sequence of human GPR116 standards of the Ethics Committee of East China Normal (ENSEMBL ID: ENST00000283296 and ENST00000265417) was University. The mouse tail vein injection tumor lung metastasis amplified using the forward primer 50-CCGCTCGAGTGGA- model and the mouse left ventricle injection tumor bone CAGGCCAACCAACTC-30 and the reverse primer: 50-ATAA- metastasis model were conducted as described previously

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF3

Tang et al.

(22). Paraformaldehyde-fixed mouse tissues were embedded in adhesion domains in their N-terminus. To identify potential paraffin and sectioned (4 mm). The paraffinized sections of adhesion GPCRs involved in cancer cell migration, we screened mouse tissues and human tissue microarray were treated with the expression of various adhesion GPCRs in MDA-MB-231, a 0.3% hydrogen peroxide/methanol and incubated with mono- highly metastatic breast cancer cell line, using reverse tran- clonal antibodies in Immunol Staining Primary Antibody scription PCR (RT-PCR; Fig. 1A). Fourteen adhesion GPCRs Dilution Buffer (Beyotime, #P0103) followed by incubation were found to be highly expressed in MDA-MB-231 cells with the Histostain-Plus IHC Kit reagents (Rabbit Primary, (Supplementary Fig. S1). To explore their roles in breast can- Mt-bio#LHK612) according to the manufacturer's instructions. cer cell migration, we used siRNAs to downregulate the expression of each of the 14 adhesion GPCRs and used a Bioluminescence measurement wound-healing assay to examine the effect of knockdown on In vivo bone and lung metastasis assay was monitored by migration. Our data showed that silencing GPR116 by specific bioluminescence imaging (Caliper Life Sciences) as previously siRNA (Supplementary Fig. S2A) significantly reduced cell described (23). Briefly, to detect metastases of MDA-MB-231 migration (Fig. 1B), suggesting that GPR116 is a novel regulator breast cancer cells stably expressing luciferase at indicated of breast cancer cell migration. days, anesthetized mice were injected with 150 mg/kg D-lucif- To confirm the role of GPR116 in breast cancer cell migra- erin potassium salt (115144-35-9, AOK chem). Biolumines- tion, we made 2 stable GPR116-knockdown cell lines using cence was assayed, and photon per second was quantified by lentiviral transduction of 2 different specific shRNAs in MDA- software (Living Image3.2, Caliper). MB-231 cells. The knockdown of GPR116 expression was confirmed by Western blotting in the 2 shRNAs stably trans- Tissue microarray and survival analysis in patient fected cells (Supplementary Fig. S2B). As shown in Fig. 1C, sample knockdown of GPR116 markedly reduced breast cancer cell The human tissue microarray was bought from Cybrdi Bio- invasion as measured by a Transwell invasion assay. To further technology, including #BR8010, #CC08-11, #BR10010, #BR1503, confirm that shRNA effects were due to specific knockdown #BR2082, and #BR721. All the pathologic diagnosis was sup- of GPR116, we transduced an expression vector with only plied in the manufacturer's instructions of the product. Tissue the coding sequence of GPR116 into the cells expressing fi 0 core staining was classi ed as either negative or positive. The siRNA targeting a noncoding sequence in the 3 -UTR of positive cores were further subdivided by the intensity of GPR116. The results showed that re-introduction of GPR116 GPR116 staining into either low or high groups by 2 research- cDNA into the GPR116-knockdown cells rescued the invasive ers, respectively. GPR116 expression levels in tumor samples ability of the cells (Fig. 1D) by increasing the expression of for survival analysis derived from the discovery cohort were GPR116 (Supplementary Fig. S2D), indicating that GPR116 determined by microarray analysis using Affymetrix U133B rather than off-target shRNA effects is responsible for the Genechips (Affymetrix). The raw database is available at the invasive phenotype of MDA-MB-231 cells. GEO (GSE6532; ref. 24). To exclude an effect of GPR116 knockdown on cell proliferation and apoptosis, we examined cell growth and Statistical analysis apoptosis in MDA-MB-231 cells with stable knockdown of Experiments were carried out with 3 or more replicates. The GPR116. Our data show that GPR116 knockdown had little results were reported as mean SD or mean SEM and are fi effect on cell growth over 5 days (Supplementary Fig. S3A). indicated in the gure legends. Data were analyzed using SPSS Furthermore, little apoptosis was observed in GPR116- 13.0 for Windows version (Chicago). For comparisons of knockdown breast cancer cells by Annexin V apoptosis central tendencies, normally distributed data sets were ana- assay (Supplementary Fig. S3B and S3C), suggesting that lyzed using unpaired 2-sided Student t tests under assumption the effects of GPR116 on breast cancer cell migration and of equal variance. Non-normally distributed data sets were – U invasion are not mediated by changes in cell growth or analyzed using nonparametric Mann Whitney tests. The apoptosis. normality of data sets was evaluated visually using Q–Q plots fi – To further con rm the effects of GPR116 on breast can- and tested statistically using Shapiro Wilk tests. Survival cer cell migration and invasion, we examined whether over- curves were analyzed according to the Kaplan–Meier method, P expression of GPR116 promoted cell invasion in breast cancer and for differences between curves, the value was calculated cells (MCF-7 and Hs578T) that express low levels of GPR116 c2 by the log-rank test. The test was applied to analyze the (Supplementary Fig. S4). Our data showed that overexpres- relationship between GPR116 expression status and clinico- sion of GPR116 remarkably promoted cell invasion of MCF-7 pathologic status. Differences were considered as statistically and Hs578T, respectively (Fig. 1E and F), indicating that the significant data when P < 0.05. Immunoblotting density was fi expression levels of GPR116 directly correlated with breast quanti ed by ImageJ software. cancer cell migration and invasion.

Results Knockdown of GPR116 in breast cancer cells reduces Identiﬁcation of GPR116 as a regulator of breast cancer lung and bone metastasis in vivo cell migration and invasion To examine whether GPR116 regulates breast cancer We speculated that adhesion GPCRs may play important cell metastasis in vivo, we investigated lung metastasis of roles in cancer metastasis because most of them contain MDA-MB-231 cells using mouse tail vein injection (Fig. 2A).

OF4 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

Figure 1. GPR116 regulates migration and invasion of breast cancer cells. A, schematic representation of screening adhesion GPCRs for regulators of breast cancer cell migration. B, knockdown of GPR116 in breast cancer cells reduces migration properties. MDA-MB-231 cells were transfected with nontargeting control siRNA (scrambled siRNA) or GPR116 siRNA (GPR116 siRNA) for wound-healing assay. Representative images at time points 0 and 12 hours (top). Breast cancer cells migrating back into the scratched surface were counted after 12 hours (bottom). Magnification, 100. Columns show the means of experiments conducted in triplicate; bars show SD. , P < 0.001. C, knockdown of GPR116 in breast cancer cells reduces invasive properties. Two shRNA-mediated stable knockdown MDA-MB-231 cells (shGPR116 #1 and #2) were used in the Boyden chamber invasion assay. Six hours after seeding, invasive cells were stained, photographed (left), and counted (right). Magnification, 200. Columns show the means of experiments conducted in triplicate; bars show SD., P < 0.01. D, loss of invasion ability induced by knockdown of endogenous GPR116 (si30-UTR) was rescued through GPR116 re-introduced (si30-UTR þ GPR116) by lentivirus in breast cancer cells MDA-MB-231. Cells were stained and photographed (left), and invasive cells were counted (right). Magnification, 200. Columns show the means of experiments conducted in triplicate; bars show SD. , P < 0.01. E and F, overexpression of GPR116 promotes breast cancer cell invasion. Vector control cell (zsgreen) and stably overexpressing GPR116 (GPR116) MCF-7 cells (E) or Hs578T cells (F) were used in the Boyden chamber invasion assay. Cancer cells were allowed to invade for 16 hours for MCF-7 cells or 6 hours for Hs578T cells. The crystal violet–stained invasive cells were photographed (left) and counted (right). Magnification, 200. Columns show the means of experiments conducted in triplicate; bars show SD. , P < 0.05, , P < 0.001.

As expected, the luciferase-labeled control MDA-MB-231 and E) and strikingly decreased the osteolytic lesion area (Fig. cells efficiently metastasized to the lung region after 24 days. 2D and F). As a result, decreased GPR116 expression was In contrast, knockdown of GPR116 significantly reduced the associated with a significant prolongation in survival for tu- ability of the 2 stable knockdown breast cancer cells (Fig. 2A mor-bearing mice (Fig. 2G). Taken together, these data suggest and B) to form lung metastases, suggesting a role of GPR116 that GPR116 regulates breast tumor cell metastases in vivo. in breast cancer cell metastasis to the lung tissue. To further examine the role of GPR116 in tumor meta- GPR116 regulates lamellipodium and actin stress fiber stasis, we also investigated bone metastases of breast cancer formation in breast cancer cells cells using the mouse left ventricle injection tumor metastasis To examine the mechanisms underlying GPR116 regula- model, an extensively used model for tumor bone metastasis tion of cell migration and invasion, we investigated cell (22, 23). As shown in Fig. 2C and D, control tumor cells were morphology in GPR116-knockdown MDA-MB-231 cells. As predominantly detected in the hind legs and strongly induced shown in Fig. 3A, knockdown of GPR116 led to remarkable osteolytic bone metastasis. The knockdown of GPR116 signi- cell morphology changes, characterized by an elongated ficantly reduced the development of bone metastases (Fig. 2C cellular morphology (Fig. 3A). Similar results were obtained

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF5

Tang et al.

Figure 2. GPR116 downregulation reduces the metastatic properties of human breast cancer cells in vivo. A and B, GPR116 knockdown reduces breast cancer cell lung metastasis in the mouse tail vein injection tumor metastasis model. Two shRNA-mediated stable GPR116 knockdown MDA-MB- 231 cells (shGPR116 #1 and #2) or control cells (shNTC; 1 106 cells) were injected into the tail vein of 4- to 5-week-old female nu/nu mice (n ¼ 10 mice in each group). A, bioluminescent imagings of representative mice in each experimental group at the indicated days after the cancer cell injection are shown in the same color scale. B, quantitative analysis of metastatic cells in lungs by bioluminescence analysis is shown as means SD. , P < 0.001. C–G, GPR116 knockdown inhibits breast cancer cell bone metastasis in the mouse left ventricle injection tumor metastasis model. Two shRNA- mediated stable GPR116- knockdown MDA-MB-231 cells (shGPR116 #1 and #2) or control cells (shNTC; 2 105 cells) were injected directly into the left ventricle of 4- to 5-week-old female nu/nu mice (n ¼ 10 mice in each group). Bioluminescent (C) and radiographic (D) imaging of representative mice in each experimental group at the indicated days are shown in the same color scale. White arrows indicate osteolytic lesions (D). Quantitative analysis of metastatic cells in bone by bioluminescence (E) and lesion area (F) analysis. Columns, means; bar, SD. , P < 0.01; , P < 0.001. Kaplan– Meier survival curve of mice (n ¼ 10) inoculated with shNTC or GPR116 knockdown (shGPR116 #1 and #2) cells (G). P ¼ 0.0006 by the log-rank test (shGPR116 #1 vs. shNTC), P < 0.0001 by the log-rank test (shGPR116 #2 vs. shNTC).

from immunofluorescence staining (Fig. 3A, right). The GPR116 regulates breast cancer cell migration through numbers of elongated cells were remarkably increased from RhoA- and Rac1-dependent pathways 10.3% (control cells) to 50.0% and 47.5% in the 2 stable To examine the mechanism by which GPR116 regulates the GPR116-knockdown cell lines, respectively (Fig. 3B). More- formation of lamellipodia and stress fibers, we investigated the over, knockdown of GPR116 ledtomuchnarrowerlamelli- activation of Rac1 and RhoA, respectively (4). As shown in Fig. podia spreading in more than 90% of GPR116-knockdown 3D, knockdown of GPR116 expression significantly decreased cells (Fig. 3A). Conversely, overexpression of GPR116 in RhoA and Rac1 activity but did not affect the activity of Cdc42 MCF-7 cells dramatically increased the formation of stress in MDA-MB-231 breast cancer cells. Consistently, overexpres- fibers and induced lamellipodia (Fig. 3C). These results sion of GPR116 in MCF-7 cells and 293T cells increased the indicate that GPR116 regulates the formation of lamellipodia activity of RhoA and Rac1, but not of Cdc42 (Fig. 3E and and stress fibers in breast cancer cells. Supplementary Fig. S5), supporting that GPR116 regulation of

OF6 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

Figure 3. GPR116 regulates breast cancer cells morphology and migration by modulating small GTPases. A and B, knockdown of GPR116 in breast cancer cells leads to elongated cellular morphology, reduced formation of actin stress fibers, and narrower lamellipodia spreading. A, left, two shRNA-mediated stable GPR116-knockdown MDA-MB-231 cells (shGPR116 #1 and #2) or control cells (shNTC) were photographed by phase-contrast microscopy. Yellow arrows indicate leading edge lamellae of the cell. Magnification, 200. Cells were fixed and stained for F-actin with Alexa Fluor 594 phalloidin (red), for nuclei with DAPI (blue), and for paxillin (green). Right, the images were photographed with a confocal laser scanning microscope. White arrows indicate the stress fibers. Objective lens magnification, 63. B, number of elongated cell was counted. , P < 0.01, two shRNAs versus shNTC. C, overexpression of GPR116 in MCF-7 cells affects cell morphology. MCF-7 cells were stained by phalloidin (red). White arrows indicate the lamellae and stress fibers. Magnification, 400. D, knockdown of GPR116 in breast cancer cells downregulates the activation of RhoA and Rac1 but not Cdc42 in GTPase activity assays. E, overexpression of GPR116 in breast cancer cells upregulates the activation of RhoA and Rac1 but not Cdc42 in GTPase activity assays. MCF-7 cells overexpressing GPR116 by lentiviral transduction were lysed, and cell lysates were subjected to pull down (for activated Rho/Rac/Cdc42) or immunoblotting using indicated antibodies. F, knockdown of GPR116 in breast cancer cells decreases the phosphorylation of cofilin and MLC2 by immunoblotting using indicated antibodies. G, GPR116 knockdown decreases the phosphorylation of cofilin in tumor sections of tumor metastasis mouse model. Thirty-six days after GPR116 knockdown, MDA-MB-231 cells or control cells (1 106 cells) were injected into tail vein as described in Fig. 2A, mouse lungs were removed, sectioned, and subjected to immunohistochemical staining using the indicated antibodies. Magnification, 100. H and I, RhoA or Rac1 rescues the migration ability inhibited by knockdown of GPR116. Stable GPR116-knockdown MDA-MB-231 cells (shGPR116) were transduced to overexpress constitutively active mutant RhoA Q63L (shGPR116 þ CA RhoA) or Rac1 Q61L (shGPR116 þ CA Rac1) by lentiviral transduction, and effects on migration were determined by wound-healing migration assay. H, representative images at time points 0 and 12 hours. Magnification, 200. I, breast cancer cells migrating back into the scratched surface were counted after 12 hours. Columns show the means of experiments conducted in triplicate; bars show SD. , P < 0.01; , P < 0.05.

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF7

Tang et al.

RhoA and Rac1 activity is independent of cell type. To further according to a sequence analysis using the bioinformatics confirm the effect of GPR116 on RhoA and Rac1 activity, we prediction tool (32), we found that GPR116 would only examined the phosphorylation of MLC-2 and cofilin, which is couple to Gaq/11. Taken together, these data suggest that induced by the activation of small GTPases. As shown in Fig. 3F, the activation of RhoA and Rac1 by GPR116 is Gaq/11- knockdown of GPR116 significantly reduced the phosphoryla- dependent. To further confirm this result, we overexpressed tion of MLC-2 and cofilin in MDA-MB-231 breast cancer cells. a constitutively active mutant of Gaq(CAGq;ref.33)in Similar results were obtained from sections of MDA-MB-231 GPR116-knockdown MDA-MB-231 cells. As shown in Fig. 4C, metastases in mice. Compared with control mouse metastases, knockdown of GPR116 reduced the activation of RhoA and knockdown of GPR116 inhibited the phosphorylation of cofilin Rac1, whereas expression of CA Gq completely restored the (Fig. 3G). activity of RhoA and Rac1 disrupted by the GPR116 knock- To examine whether GPR116-mediated breast cancer cell down. These results showed that the regulation of RhoA and migration is dependent on RhoA and Rac1, we overexpressed Rac1 by GPR116 is Gaq-dependent. constitutively active mutants of RhoA or Rac1 in GPR116- To determine whether GPR116 regulation of breast cancer knockdown breast cancer cells. As shown in Fig. 3H and I, cell migration and morphology is dependent on Gaq, we knockdown of GPR116 suppressed breast cancer cell migra- expressed CA Gq in GPR116-knockdown cells. Expression of tion, whereas ectopic expression of active mutant RhoA and CA Gq rescued cell migration (Fig. 4D) as well as cell mor- Rac1 partially restored cell migration (Fig. 3H and I), suggest- phology in GPR116-knockdown MDA-MB-231 cells (Fig. 4E). ing that GPR116-mediated breast cancer cell migration is These results indicated that GPR116 regulation of breast RhoA- and Rac1-dependent. cancer cell migration is Gaq-dependent. To determine whether RhoA and Rac1 are directly involved in GPR116 mediated formation of cell lamellipodia Regulation of RhoA and Rac1 activity by GPR116 is and stress fibers, we investigated whether overexpression through p63RhoGEF of RhoA or Rac1 in GPR116-knockdown cells could rescue Gaq coupling to p63RhoGEF is important for GPCR- lamellipodium and stress fiber formation. As shown in mediated RhoA activation (6, 33, 34). In addition, p63Rho- Supplementary Fig. S6, knockdown of GPR116 reduced stress GEF/GEFT has been reported to activate Rac1 and mediate fiber formation and induced an elongated cell shape in MDA- tumorigenesis, tumor metastasis, and myogenesis (8, 35, 36). MB-231 breast cancer cells, whereas overexpression of con- To investigate whether p63RhoGEF is required for the acti- stitutively active mutant RhoA rescued the stress fiber vation of RhoA and Rac1 in breast cancer cells, we over- formation in GPR116-knockdown cells and overexpression expressed p63RhoGEF in MDA-MB-231 cells. Our data show- of constitutively active mutant Rac1 promoted strong lamel- ed that p63RhoGEF activated both RhoA and Rac1 in breast lipodia formation in GPR116-knockdown cells. And both cancer cells (Fig. 4F and Supplementary Fig. S8A). To further mutants partially rescued the elongated cell shape pheno- examine whether p63RhoGEF is required for GPR116-medi- type (Supplementary Fig. S6), indicating that the regulation ated activation of RhoA and Rac1, we overexpressed a cata- of actin stress fiber and lamellipodia formation by GPR116 is lytic domain–deleted mutant p63RhoGEF (p63RhoGEF DN) RhoA- and Rac1-dependent. in GPR116-overexpressing 293T cells. Our results showed that p63RhoGEF DN significantly abrogated the GTPase activation GPR116 regulates RhoA and Rac1 activity by coupling to induced by GPR116 (Fig. 4G and Supplementary Fig. S8B), Gaq/11 suggesting that GPR116 regulation of RhoA and Rac1 activity Previous studies suggest that Ga12/13 and Gaq/11 are the is p63RhoGEF-dependent. 2 most common Ga proteins directly activating Rho family GTPases to regulate cell migration and morphology (25–28). GPR116 is associated with malignant breast cancer in We first examined whether GPR116 regulates RhoA and Rac1 humans activity through Ga12/13 or Gaq/11. In GPR116-overexpres- To further investigate the biologic significance of GPR116 in sing 293T cells, we co-expressed GPR116 with the regulator breast cancer, we used 3 separate approaches to analyze the of G-protein signaling 2 (RGS2)orRGSdomainsofp115Rho- expression of GPR116 in patient samples. First, we analyzed the GEF, which selectively block the activity of Gaq/11 or Ga12/ mRNA expression trend of GPR116 by 2 gene expression 13, respectively (29, 30), and then examined the activation datasets, GDS2250 and GDS2415, from Gene Expression Omni- of RhoA and Rac1. Our data showed that the expression of bus (GEO). As shown in Fig. 5A, the gene expression trend of RGS2,whichblocksGaq/11 activation, abrogated the acti- GPR116 was signifi cantly upregulated in breast tumors (n ¼ vation of RhoA and Rac1 induced by GPR116 (Fig. 4A), 47) and recurrent tumors (n ¼ 59) compared with normal whereas expression of the RGS domain of p115RhoGEF, tissue (Fig. 5A, top) and no recurrent tumor (Fig. 5A, bottom), which blocks Ga12/13 activation, had little effect on the respectively. Second, using immunohistochemistry, we exam- activation of RhoA and Rac1 (Supplementary Fig. S7A). ined the protein expression of GPR116 in human breast cancer Moreover, knockdown of Gaq/11 blunted the GPR116- tissues (n ¼ 490) and in normal human breast tissues (n ¼ 67). dependent activation of RhoA and Rac1 (Fig. 4B). In addi- Consistent with mRNA expression analysis, most of the normal tion, the C-terminal domain of the Gaqprotein(GqCT) breast tissues had no signal or pale staining of GPR116 protein markedly attenuated GPR116 induced activation of RhoA (Fig. 5B and C and Table 1); however, the staining of GPR116 and Rac1 (Supplementary Fig. S7B; refs. 6, 31). Finally, protein remarkably increased with progression through

OF8 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

Figure 4. GPR116 initiates the Gaq/11 signaling pathway in breast cancer cell migration. A, RGS2 blocks activation of RhoA and Rac1 induced by overexpression of GPR116. After ectopic expression of GPR116 and Flag-tagged RGS2 in 293T cells, cell lysates were subjected to the GTPase activity assay for RhoA and Rac1 activity. B, knockdown of Gaq/11 inhibits activation of RhoA and Rac1 induced by overexpression of GPR116. 293T cells were transfected with GPR116 or Gaq/11 shRNA. The cell lysates were subjected to the GTPase activity assay for RhoA and Rac1 activity. C, constitutively active mutant Gaq rescues the activity of RhoA and Rac1 in GPR116-knockdown breast cancer cells. GPR116-knockdown MDA-MB-231 cells (shGPR116) overexpressing the constitutively active mutant Gaq R183C (CA Gq) by lentiviral transduction were lysed. The cell lysates were subjected to the GTPase activity assay for RhoA and Rac1 activity. D, constitutively active mutant of Gaq rescues the migration ability in GPR116-knockdown breast cancer cells. GPR116-knockdown MDA-MB-231 cells (shGPR116) were transduced to overexpress constitutively active mutant Gaq R183C (shGPR116 þ CA Gq). Wound-healing migration assay was conducted for 12 hours. The migrated cells were quantified by counting. Columns, means; bar, SD. , P < 0.01. E, constitutively active mutant Gaq partially rescues cell morphology in GPR116-knockdown breast cancer cells. GPR116- knockdown MDA-MB-231 cells (shGPR116) were transduced to overexpress constitutively active mutant Gaq R183C (shGPR116 þ CA Gq), and then the cells were fixed and stained for F-actin with Alexa Fluor 594 phalloidin (red) and nuclei were counterstained with DAPI (blue). Magnification, 400. F, p63RhoGEF increases the activity of RhoA and Rac1 in breast cancer cells. MDA-MB-231 cells were transfected with or without p63RhoGEF. After 24 hours, the cell lysates were subjected to the pull-down assay for the activation of RhoA and Rac1 (left). Quantification of active RhoA and Rac1 (right). Columns, means; bar, SD. , P < 0.05. n ¼ 3. G, catalytic domain–deleted mutant of p63RhoGEF (p63RhoGEF DN) blocks the activity of RhoA and Rac1 induced by GPR116. 293T cells were transfected with GPR116 or p63RhoGEF DN. After 24 hours, the activation of RhoA and Rac1 was analyzed by GTPase activity assay (left). Quantification of active RhoA and Rac1 (right). Columns, means; bar, SD. , P < 0.05. n ¼ 3. clinical stages, from benign breast lump (n ¼ 38), through in with high GPR116 expression levels was highest in tumors with situ carcinoma (n ¼ 364), to distant metastatic tumors (n ¼ distant metastases, suggesting that the expression of GPR116 88; Fig. 5C and Table 1). Especially, the percentage of samples was correlated with breast cancer progression and metastatic

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF9

Tang et al.

Figure 5. Expression of GPR116 is significantly associated with malignant breast cancer progression, metastasis, and poor prognosis. A, gene expression trend shows GPR116 was significantly upregulated in breast tumor and recurrent tumor compared with normal tissue (top) and no recurrent tumor (bottom), respectively. Two gene expression datasets, GDS2250 and GDS2415, from GEO were collected and used for evaluating the gene expression trend of GPR116. A t test was applied to detect differential expression between 2 groups of breast samples: tumor versus normal (n ¼ 47, P < 0.01, top) and recurrent tumor versus no recurrent tumor (n ¼ 59, P < 0.01, bottom). The numbers of upregulated or downregulated samples were counted and the coverage percentages (%) are shown. The red box indicates the coverage percentage of upregulated samples; the green box represents the coverage percentage of downregulated samples. Bar, SD. B and C, GPR116 protein expression levels are significantly associated with malignant breast cancer progression. GPR116 levels in representative human breast tissues of normal tissues (n ¼ 67) or tumor tissues (n ¼ 490) stratified by immunostaining expression level of GPR116. No GPR116, negative GPR116 staining for all tumor cells; low GPR116, hematoxylin signal (blue) is stronger than GPR116 signal (brown); high GPR116, GPR116 signal (brown) is stronger than hematoxylin signal (blue). The proportion of each expression level was calculated in 4 different breast cancer stage samples: normal tissues (n ¼ 67), benign breast lump (n ¼ 38), in situ carcinoma (n ¼ 364), and distant metastatic tumor (n ¼ 88). Magnification, 100. D, Kaplan–Meier analysis of recurrence-free survival (left) and distant metastasis-free survival (right) according to GPR116 mRNA expression data in 340 patients with breast cancer acquired from microarray analysis (GEO accession # GSE6532; discovery cohort; n ¼ 340). P values were calculated by the log-rank test.

status. The multivariate analysis also indicated that the expres- survival analysis of the specimens revealed that the GPR116- sion of GPR116 was positively associated with tumor stage and high expression group (n ¼ 170) had signiﬁcantly shorter histologic grade (P < 0.0001, Table 1). Finally, we examined the recurrence-free survival (P ¼ 0.046) and distant metastasis- prognostic value of GPR116 expression in breast cancer by free survival (P ¼ 0.013) as compared with the GPR116-low microarray-based gene expression data from primary breast group (n ¼ 170, Fig. 5D). Therefore, our data indicate that tumor specimens resected at time of diagnosis (GEO accession GPR116 is signiﬁcantly correlated with breast cancer progres- # GSE6532; discovery cohort; n ¼ 340; ref. 24). Kaplan–Meier sion, metastasis, and poor prognosis.

OF10 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

Table 1. Association between GPR116 expression and clinicopathologic parameters in breast cancer

Characteristic No GPR116 Low GPR116 High GPR116 P n (%) 57 (10.2) 131 (23.5) 369 (66.2) Age at diagnosis (mean) 43.1 (57) 42.5 (131) 48.27 (369) Menopausal status at diagnosis,a n (%) Age (45) 30 (52.6) 80 (61.1) 147 (39.8) <0.0001 Age (>45) 27 (47.4) 51 (38.9) 222 (60.2) Stage (tumor),a,b n (%) 0 2 (11.1) 5 (5.6) 29 (10.1) 0.003 I 8 (44.4) 18 (20.2) 33 (11.5) II 6 (33.3) 44 (49.4) 166 (57.8) III 2 (11.1) 22 (24.7) 59 (20.6) pN stageþ,a,c n (%) N0 12 (66.7) 43 (48.9) 200 (69.4) 0.009 N1 5 (27.8) 29 (33.0) 55 (19.0) N2 1 (5.6) 16 (18.2) 33 (11.5) Histologic grade,a,d n (%) G1 8 (72.7) 34 (41.0) 18 (6.9) <0.0001 G2 2 (18.8) 45 (54.2) 201 (76.7) G3 1 (9.0) 4 (4.8) 43 (16.4) Tumor size,a n (%) T < 2 cm 11 (64.7) 26 (29.5) 65 (22.6) 0.002 2cm T < 5 cm 6 (35.3) 43 (48.9) 156 (54.2) T 5 cm 0 (0) 19 (21.6) 67 (23.3) Histologic type Normal 36 (63.2) 26 (19.8) 5 (14) <0.0001 Benign breast lump 4 (7.0) 20 (15.2) 14 (3.8) In situ carcinoma 15 (26.3) 72 (55.0) 277 (75.1) Distant metastatic tumor 2 (3.5) 13 (9.9) 73 (19.8)

NOTE: The differences between the two groups were evaluated by c2 tests. an for these characteristics does not add up to total studied (n ¼ 557) because of missing information. bStage I, tumor size 2 cm and no nodal involvement; stage II, tumor size 2 cm and 1–3 nodes, 2–4 cm and 0–3 nodes, or 4þ cm and 0 nodes; stage III, tumor size 2 cm and 4þ nodes, 2–4 cm and 4þ nodes, or >4 cm and 1 þ nodes. cpN stage N0, regional lymph node metastases; N1, metastases to movable ipsilateral level I, II axillary lymph node(s); N2, metastases in ipsilateral level I, II axillary lymph nodes that are clinically ﬁxed or matted. dHistologic grade 1, the cancer cells look similar to normal cells and grow very slowly; G2, the cancer cells look more abnormal and are slightly faster growing; G3, the cancer cells look very different from normal cells and tend to grow quickly.

Discussion downregulation of GPR116 markedly decreased the activity of In the present study, we identified GPR116 as a regulator of the small GTPases RhoA and Rac1, but not Cdc42. Conversely, breast cancer metastasis through screening the adhesion ectopic expression of GPR116 in MCF-7 and 293T cells led to GPCR family. GPR116 regulates breast cancer cell migration activation of RhoA and Rac1, but not Cdc42. We further and invasion in both highly and poorly metastatic breast showed that RhoA and Rac1 were responsible for GPR116- cancer cells and modulated the formation of actin stress fibers induced cancer cell migration and morphology changes. Pre- and lamellipodia through the Gaq-p63RhoGEF-Rho GTPase vious studies have shown that RhoA and Rac1 regulate the signaling pathway. Moreover, loss of GPR116 largely inhibited formation of an elongated cell morphology in macrophages breast cancer metastasis in 2 mouse metastasis models. Fur- (37) and regulate the motility of diverse tumor cells (38–40), thermore, by examining human breast cancer tissues and which are consistent with our observations. Second, we found clinicopathologic parameters, we found that the expression that GPR116 promoted breast cancer cell migration and inva- of GPR116 is significantly associated with tumor stage, metas- sion via Gaq signaling. We showed that inhibiting Gaqin tasis, and poor prognosis, providing pathologic support for the GPR116-overexpressing cells significantly blocked RhoA and biologic function of this receptor in vitro and in vivo. Rac1 activation and ectopically expressing a constitutively We also clarified the molecular mechanism how GPR116 active mutant of Gaq markedly rescued the signaling and cell regulates breast cancer cell migration. First, we found that phenotype in GPR116-knockdown breast cancer cells. Third,

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF11

Tang et al.

p63RhoGEF, an important Gaq effector, has been reported to rat GPR116 has been reported to form dimers and to be mediate RhoA, Rac1, or Cdc42 activation in different cell types processed multiple times to generate several fragments (15), (35, 41, 42). Our data showed that GPR116 mediated RhoA and which suggested that the cleaved fragments of GPR116 may Rac1 activation by Gaq and p63RhoGEF signaling. function as ligands. Our results also showed that GPR116 is Our observations indicate that GPR116 was involved in strongly correlated with human breast cancer progression, breast cancer metastasis regulation through the classical metastasis, and poor prognosis and its expression even starts Ga protein–mediated signaling pathway. Previous reports in benign breast lumps. Therefore, GPR116 may serve as a bio- have shown that some adhesion GPCRs are associated with marker for breast cancer diagnosis through detecting the tumor progression, such as EMR2 in breast cancer (43), cleaved fragments in body fluid using appropriate antibodies. GPR56 in pancreatic cancer and glioblastoma (44, 45), and In conclusion, for the first time, we show that GPR116 plays a CD97 in thyroid and colorectal cancer (28, 46). Interestingly, critical role in promoting breast cancer progression and adhesion GPCRs frequently signal through Ga proteins. metastasis through coupling to Gaq and the p63RhoGEF- GPR56 has been reported to couple to Ga12/13 and Gaq/11 RhoA/Rac1 signaling pathway. Furthermore, the strong cor- (25, 27), GPR133 couples to Gas, and CD97 couples to relation between the expression of GPR116 and breast cancer Ga12/13 (28, 47). Our study supports the idea that adhesion malignancy supports its role as a metastasis promoter in GPCRs also activate the classical receptor/G-protein signal- humans. Therefore, GPR116 could be used as a potential ing pathway. therapeutic target and diagnostic marker for breast cancers. During cell migration, Rac1 and RhoA have a reciprocal feedback interaction over controlling the plasticity of cell Disclosure of Potential Conflicts of Interest migration. Activated Rac1 has been shown to inhibit RhoA No potential conflicts of interest were disclosed. activity by the Rac1 downstream effector WAVE2 (4). RhoA also inhibits the activity of Rac1 through regulating the Rac- GAP (Rac GTPase–activating protein) ARHGA22 (4). In our Authors' Contributions Conception and design: R. Jin, G. Qu, J. Xiao, M. Liu, J. Luo study, we showed that GPR116 regulated cancer cell migration Development of methodology: R. Jin, G. Qu and metastasis by activating both RhoA and Rac1. Though Acquisition of data (provided animals, acquired and managed patients, the precise molecular mechanism still remains to be elucidat- provided facilities, etc.): X. Tang, R. Jin, G. Qu, X. Wang, Z. Li Analysis and interpretation of data (e.g., statistical analysis, biostatistics, ed, there are several facets to interpreting this conflict. computational analysis): X. Tang, R. Jin, G. Qu, X. Wang, Z. Yuan, C. Zhao, First, p63RhoGEF has been shown to activate different small T. Shi, J. Luo Writing, review, and/or revision of the manuscript: X. Tang, R. Jin, S. Siwko, GTPases including RhoA, Rac1, and Cdc42 in different cell T. Shi, M. Liu, J. Luo types (35, 41, 42), suggesting that GPR116 regulates both RhoA Administrative, technical, or material support (i.e., reporting or orga- and Rac1 via the activation of p63RhoGEF in breast cancer nizing data, constructing databases): P. Wang, J. Xiao Study supervision: M. Liu, J. Luo cells. Second, it has been reported that at the leading edge, activated RhoA antagonizes Rac1 during cell migration. There- fore, during cell migration, how GPR116 spatiotemporally Grant Support regulates RhoA and Rac1 will need further investigation. This work was supported by grants from the National Basic Research Program of China (2012CB910402, 2012CB910404, and 2010CB529704), the National Last, both RhoA and Rac1 have high expression and activity Natural Science Foundation of China (30930055, 81071437, 81330059, and levels in breast cancer (4). It is reasonable to speculate about 81272911), and the Science and Technology Commission of Shanghai Munici- pality (11DZ2260300 and 12XD1406100). the existence of a common signaling pathway that enhances The costs of publication of this article were defrayed in part by the the expression and activity of both RhoA and Rac1 in breast payment of page charges. This article must therefore be hereby marked cancer. In fact, previous studies have shown that Ras can acti- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this vate Rac1 (48), which can, in turn, activate RhoA (49). fact. Overexpression of GPR116-induced breast cancer cell mig- Received April 10, 2013; revised July 29, 2013; accepted August 15, 2013; ration and stimulated the activation of RhoA and Rac1. And published OnlineFirst September 5, 2013.

References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer a pathway for the activation of RhoA by GPCRs. Science 2007; statistics. CA Cancer J Clin 2011;61:69–90. 318:1923–7. 2. Hall A. The cytoskeleton and cancer. Cancer Metastasis Rev 2009; 7. Momotani K, Artamonov MV, Utepbergenov D, Derewenda U, Dere- 28:5–14. wenda ZS, Somlyo AV. p63RhoGEF couples Galpha(q/11)-mediated 3. Fukata M, Nakagawa M, Kaibuchi K. Roles of Rho-family GTPases signaling to Ca2þ sensitization of vascular smooth muscle contrac- in cell polarisation and directional migration. Curr Opin Cell Biol tility. Circ Res 2011;109:993–1002. 2003;15:590–7. 8. Swenson-Fields KI, Sandquist JC, Rossol-Allison J, Blat IC, Wen- 4. Sahai E, Marshall CJ. RHO-GTPases and cancer. Nat Rev Cancer nerberg K, Burridge K, et al. MLK3 limits activated Galphaq 2002;2:133–42. signaling to Rho by binding to p63RhoGEF. Mol Cell 2008;32: 5. Hall A. Rho GTPases and the actin cytoskeleton. Science 1998;279: 43–56. 509–14. 9. Lagerstrom MC, Schioth HB. Structural diversity of G protein-coupled 6. Lutz S, Shankaranarayanan A, Coco C, Ridilla M, Nance MR, Vettel receptors and signiﬁcance for drug discovery. Nat Rev Drug Discov C, et al. Structure of Galphaq-p63RhoGEF-RhoA complex reveals 2008;7:339–57.

OF12 Cancer Res; 73(20) October 15, 2013 Cancer Research

GPR116 Promotes Metastases through Gaq Pathway

10. Dorsam RT, Gutkind JS. G-protein-coupled receptors and cancer. Nat 31. Chillar A, Wu J, Cervantes V, Ruan KH. Structural and functional Rev Cancer 2007;7:79–94. analysis of the C-terminus of Galphaq in complex with the human 11. Luo J, Zhou W, Zhou X, Li D, Weng J, Yi Z, et al. Regulation of bone thromboxane A2 receptor provides evidence of constitutive activity. formation and remodeling by G-protein-coupled receptor 48. Devel- Biochemistry 2010;49:6365–74. opment 2009;136:2747–56. 32. Jiang Z, Guan C, Zhou Y. Computational prediction of the coupling 12. Tang XL, Wang Y, Li DL, Luo J, Liu MY. Orphan G protein-coupled specificity of g protein-coupled receptors. Appl Biochem Biotechnol receptors (GPCRs): biological functions and potential drug targets. 2007;141:109–18. Acta Pharmacol Sin 2012;33:363–71. 33. Lutz S, Freichel-Blomquist A, Yang Y, Rumenapp U, Jakobs KH, 13. Yona S, Stacey M. Adhesion-GPCRs: structure to function. New York, Schmidt M, et al. The guanine nucleotide exchange factor p63Rho- Austin, TX: Springer ScienceþBusiness Media; Landes Bioscience; GEF, a specific link between Gq/11-coupled receptor signaling and 2010. RhoA. J Biol Chem 2005;280:11134–9. 14. Aust G. Adhesion-GPCRS in tumorigenesis. Adv Exp Med Biol 34. Niu J, Profirovic J, Pan H, Vaiskunaite R, Voyno-Yasenetskaya T. G 2010;706:109–20. Protein betagamma subunits stimulate p114RhoGEF, a guanine 15. Fukuzawa T, Hirose S. Multiple processing of Ig-Hepta/GPR116, nucleotide exchange factor for RhoA and Rac1: regulation of cell a G protein-coupled receptor with immunoglobulin (Ig)-like repeats, shape and reactive oxygen species production. Circ Res 2003;93: and generation of EGF2-like fragment. J Biochem 2006;140: 848–56. 445–52. 35. Smith TK, Hager HA, Francis R, Kilkenny DM, Lo CW, Bader DM. Bves 16. Abe J, Fukuzawa T, Hirose S. Cleavage of Ig-Hepta at a "SEA" module directly interacts with GEFT, and controls cell shape and movement and at a conserved G protein-coupled receptor proteolytic site. J Biol through regulation of Rac1/Cdc42 activity. Proc Natl Acad Sci U S A Chem 2002;277:23391–8. 2008;105:8298–303. 17. Abe J, Suzuki H, Notoya M, Yamamoto T, Hirose S. Ig-hepta, a novel 36. Wuertz CM, Lorincz A, Vettel C, Thomas MA, Wieland T, Lutz S. member of the G protein-coupled hepta-helical receptor (GPCR) family p63RhoGEF–a key mediator of angiotensin II-dependent signaling that has immunoglobulin-like repeats in a long N-terminal extracellular and processes in vascular smooth muscle cells. FASEB J 2010;24: domain and defines a new subfamily of GPCRs. J Biol Chem 1999; 4865–76. 274:19957–64. 37. Bhavsar PJ, Vigorito E, Turner M, Ridley AJ. Vav GEFs regulate 18. Nie T, Hui X, Gao X, Li K, Lin W, Xiang X, et al. Adipose tissue deletion of macrophage morphology and adhesion-induced Rac and Rho acti- Gpr116 impairs insulin sensitivity through modulation of adipose vation. Exp Cell Res 2009;315:3345–58. function. FEBS Lett 2012;586:3618–25. 38. Wu D, Asiedu M, Wei Q. Myosin-interacting guanine exchange factor 19. Yang MY, Hilton MB, Seaman S, Haines DC, Nagashima K, Burks (MyoGEF) regulates the invasion activity of MDA-MB-231 breast CM, et al. Essential regulation of lung surfactant homeostasis by cancer cells through activation of RhoA and RhoC. Oncogene the orphan G protein-coupled receptor GPR116. Cell Rep 2013;3: 2009;28:2219–30. 1457–64. 39. Gulhati P, Bowen KA, Liu J, Stevens PD, Rychahou PG, Chen M, et al. 20. Bridges JP, Ludwig MG, Mueller M, Kinzel B, Sato A, Xu Y, et al. mTORC1 and mTORC2 regulate EMT, motility, and metastasis of Orphan G protein-coupled receptor GPR116 regulates pulmonary colorectal cancer via RhoA and Rac1 signaling pathways. Cancer Res surfactant pool size. Am J Respir Cell Mol Biol 2013;49:348–57. 2011;71:3246–56. 21. Cho SG, Wang Y, Rodriguez M, Tan K, Zhang W, Luo J, et al. 40. Zhao P, Zhang W, Wang SJ, Yu XL, Tang J, Huang W, et al. HAb18G/ Haploinsufficiency in the prometastasis Kiss1 receptor Gpr54 delays CD147 promotes cell motility by regulating annexin II-activated RhoA breast tumor initiation, progression, and lung metastasis. Cancer Res and Rac1 signaling pathways in hepatocellular carcinoma cells. Hepa- 2011;71:6535–46. tology 2011;54:2012–24. 22. Fang Y, Chen Y, Yu L, Zheng C, Qi Y, Li Z, et al. Inhibition of breast 41. Mitchell DC, Bryan BA, Liu L, Hu XH, Huang XQ, Ji WK, et al. GEFT, A cancer metastases by a novel inhibitor of TGFbeta receptor 1. J Natl Rho family guanine nucleotide exchange factor, regulates lens differ- Cancer Inst 2013;105:47–58. entiation through a Rac1-mediated mechanism. Curr Mol Med 23. Li Z, Xiao J, Wu X, Li W, Yang Z, Xie J, et al. Plumbagin inhibits breast 2011;11:465–80. tumor bone metastasis and osteolysis by modulating the tumor-bone 42. Bryan BA, Mitchell DC, Zhao L, Ma W, Stafford LJ, Teng BB, et al. microenvironment. Curr Mol Med 2012;12:967–81. Modulation of muscle regeneration, myogenesis, and adipogenesis by 24. Middelbeek J, Kuipers AJ, Henneman L, Visser D, Eidhof I, van Horssen the Rho family guanine nucleotide exchange factor GEFT. Mol Cell Biol R, et al. TRPM7 is required for breast tumor cell metastasis. Cancer Res 2005;25:11089–101. 2012;72:4250–61. 43. Stacey M, Chang GW, Davies JQ, Kwakkenbos MJ, Sanderson RD, 25. Iguchi T, Sakata K, Yoshizaki K, Tago K, Mizuno N, Itoh H. Orphan G Hamann J, et al. The epidermal growth factor-like domains of the protein-coupled receptor GPR56 regulates neural progenitor cell human EMR2 receptor mediate cell attachment through chondroitin migration via a G alpha 12/13 and Rho pathway. J Biol Chem 2008; sulfate glycosaminoglycans. Blood 2003;102:2916–24. 283:14469–78. 44. Shashidhar S, Lorente G, Nagavarapu U, Nelson A, Kuo J, Cummins J, 26. Cotton M, Claing A. G protein-coupled receptors stimulation and the et al. GPR56 is a GPCR that is overexpressed in gliomas and functions control of cell migration. Cell Signal 2009;21:1045–53. in tumor cell adhesion. Oncogene 2005;24:1673–82. 27. Luo R, Jeong SJ, Jin Z, Strokes N, Li S, Piao X. G protein-coupled 45. Huang Y, Fan J, Yang J, Zhu GZ. Characterization of GPR56 protein receptor 56 and collagen III, a receptor-ligand pair, regulates cortical and its suppressed expression in human pancreatic cancer cells. Mol development and lamination. Proc Natl Acad Sci U S A 2011;108: Cell Biochem 2008;308:133–9. 12925–30. 46. Steinert M, Wobus M, Boltze C, Schutz A, Wahlbuhl M, Hamann J, et al. 28. Ward Y, Lake R, Martin PL, Killian K, Salerno P, Wang T, et al. CD97 Expression and regulation of CD97 in colorectal carcinoma cell lines amplifies LPA receptor signaling and promotes thyroid cancer pro- and tumor tissues. Am J Pathol 2002;161:1657–67. gression in a mouse model. Oncogene 2013;32:2726–38. 47. Bohnekamp J, Schoneberg T. Cell adhesion receptor GPR133 couples 29. Shi CS, Sinnarajah S, Cho H, Kozasa T, Kehrl JH. G13alpha-mediated to Gs protein. J Biol Chem 2011;286:41912–6. PYK2 activation. PYK2 is a mediator of G13alpha -induced serum 48. Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A. The small response element-dependent transcription. J Biol Chem 2000;275: GTP-binding protein rac regulates growth factor-induced membrane 24470–6. ruffling. Cell 1992;70:401–10. 30. Martin CB, Mahon GM, Klinger MB, Kay RJ, Symons M, Der CJ, et al. 49. Nobes CD, Hall A. Rho, rac, and cdc42 GTPases regulate the assembly The thrombin receptor, PAR-1, causes transformation by activation of of multimolecular focal complexes associated with actin stress fibers, Rho-mediated signaling pathways. Oncogene 2001;20:1953–63. lamellipodia, and filopodia. Cell 1995;81:53–62.

www.aacrjournals.org Cancer Res; 73(20) October 15, 2013 OF13

GPR116, an Adhesion G-Protein−Coupled Receptor, Promotes Breast Cancer Metastasis via the G α q-p63RhoGEF-Rho GTPase Pathway

Xiaolong Tang, Rongrong Jin, Guojun Qu, et al.

Cancer Res Published OnlineFirst September 5, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-13-1049

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2013/09/10/0008-5472.CAN-13-1049.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2013/10/04/0008-5472.CAN-13-1049. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.