Activation of an Olfactory Receptor Inhibits Proliferation of Prostate Cancer Cells

Abstract

Olfactory receptors (ORs) are expressed not only in the sensory neurons of the olfactory epithelium, where they detect volatile substances, but also in various other tissues where their potential functions are largely unknown. Here, we report the physiological characterization of human OR51E2, also named prostate-specific G-protein-coupled receptor (PSGR) due to its reported up-regulation in prostate cancer. We identified androstenone derivatives as ligands for the recombinant receptor. PSGR can also be activated with the odorant β-ionone. Activation of the endogenous receptor in prostate cancer cells by the identified ligands evoked an intracellular Ca2+ increase. Exposure to β-ionone resulted in the activation of members of the MAPK family and inhibition of cell proliferation. Our data give support to the hypothesis that because PSGR signaling could reduce growth of prostate cancer cells, specific receptor ligands might therefore be potential candidates for prostate cancer treatment. Olfactory receptors (ORs) are expressed not only in the sensory neurons of the olfactory epithelium, where they detect volatile substances, but also in various other tissues where their potential functions are largely unknown. Here, we report the physiological characterization of human OR51E2, also named prostate-specific G-protein-coupled receptor (PSGR) due to its reported up-regulation in prostate cancer. We identified androstenone derivatives as ligands for the recombinant receptor. PSGR can also be activated with the odorant β-ionone. Activation of the endogenous receptor in prostate cancer cells by the identified ligands evoked an intracellular Ca2+ increase. Exposure to β-ionone resulted in the activation of members of the MAPK family and inhibition of cell proliferation. Our data give support to the hypothesis that because PSGR signaling could reduce growth of prostate cancer cells, specific receptor ligands might therefore be potential candidates for prostate cancer treatment. Excessive signaling by G-protein-coupled receptors (GPCRs) 3The abbreviations used are: GPCRG-protein-coupled receptorPSGRprostate-specific G-protein-coupled receptorPSAprostate-specific antigenORolfactory receptorCKcytokeratinMAPmitogen-activated proteinMAPKMAP kinaseSAPKstress-activated protein kinaseJNKc-Jun NH2-terminal kinasesiRNAsmall interfering RNADHTdihydrotestosteroneRTreverse transcriptionADT1,4,6-androstatriene-3,17-dioneGFPgreen fluorescent protein. such as endothelin A receptor (1Godara G. Cannon G.W. Cannon Jr., G.M. Bies R.R. Nelson J.B. Pflug B.R. Prostate. 2005; 65: 27-34Crossref PubMed Scopus (32) Google Scholar), bradykinin 1 receptor (2Taub J.S. Guo R. Leeb-Lundberg L.M. Madden J.F. Daaka Y. Cancer Res. 2003; 63: 2037-2041PubMed Google Scholar), follicle-stimulating hormone receptor (3Ben-Josef E. Yang S.Y. Ji T.H. Bidart J.M. Garde S.V. Chopra D.P. Porter A.T. Tang D.G. J. Urol. 1999; 161: 970-976Crossref PubMed Scopus (114) Google Scholar), and thrombin receptor (4Chay C.H. Cooper C.R. Gendernalik J.D. Dhanasekaran S.M. Chinnaiyan A.M. Rubin M.A. Schmaier A.H. Pienta K.J. Urology. 2002; 60: 760-765Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar, 5Cooper C.R. Chay C.H. Gendernalik J.D. Lee H.L. Bhatia J. Taichman R.S. McCauley L.K. Keller E.T. Pienta K.J. Cancer. 2003; 97: 739-747Crossref PubMed Scopus (156) Google Scholar) is known to occur in prostate cancers due to strong overexpression of the respective receptors. Activation of some of these GPCRs results in androgen-independent androgen receptor activation, thus promoting the transition of prostate cancer cells from an androgen-dependent to an androgen-independent state (6Dai J. Shen R. Sumitomo M. Stahl R. Navarro D. Gershengorn M.C. Nanus D.M. Clin. Cancer Res. 2002; 8: 2399-2405PubMed Google Scholar, 7Lee L.F. Guan J. Qiu Y. Kung H.J. Mol. Cell. Biol. 2001; 21: 8385-8397Crossref PubMed Scopus (169) Google Scholar). G-protein-coupled receptor prostate-specific G-protein-coupled receptor prostate-specific antigen olfactory receptor cytokeratin mitogen-activated protein MAP kinase stress-activated protein kinase c-Jun NH2-terminal kinase small interfering RNA dihydrotestosterone reverse transcription 1,4,6-androstatriene-3,17-dione green fluorescent protein. The prostate-specific G-protein-coupled receptor (PSGR) is a class A GPCR that was initially identified as a prostate-specific tumor biomarker (8Wang J. Weng J. Cai Y. Penland R. Liu M. Ittmann M. Prostate. 2006; 66: 847-857Crossref PubMed Scopus (45) Google Scholar, 9Weng J. Wang J. Cai Y. Stafford L.J. Mitchell D. Ittmann M. Liu M. Int. J. Cancer. 2005; 113: 811-818Crossref PubMed Scopus (51) Google Scholar, 10Xu L.L. Stackhouse B.G. Florence K. Zhang W. Shanmugam N. Sesterhenn I.A. Zou Z. Srikantan V. Augustus M. Roschke V. Carter K. McLeod D.G. Moul J.W. Soppett D. Srivastava S. Cancer Res. 2000; 60: 6568-6572PubMed Google Scholar). It is specifically expressed in prostate epithelial cells, and its expression increases significantly in human prostate intraepithelial neoplasia and prostate tumors, suggesting that PSGR may play an important role in early prostate cancer development and progression (9Weng J. Wang J. Cai Y. Stafford L.J. Mitchell D. Ittmann M. Liu M. Int. J. Cancer. 2005; 113: 811-818Crossref PubMed Scopus (51) Google Scholar, 11Xu L.L. Sun C. Petrovics G. Makarem M. Furusato B. Zhang W. Sesterhenn I.A. McLeod D.G. Sun L. Moul J.W. Srivastava S. Prostate Cancer Prostatic. Dis. 2006; 9: 56-61Crossref PubMed Scopus (36) Google Scholar). Although expression of the human PSGR was found to be prostate-specific (10Xu L.L. Stackhouse B.G. Florence K. Zhang W. Shanmugam N. Sesterhenn I.A. Zou Z. Srikantan V. Augustus M. Roschke V. Carter K. McLeod D.G. Moul J.W. Soppett D. Srivastava S. Cancer Res. 2000; 60: 6568-6572PubMed Google Scholar, 12Yuan T.T. Toy P. McClary J.A. Lin R.J. Miyamoto N.G. Kretschmer P.J. Gene. 2001; 278: 41-51Crossref PubMed Scopus (46) Google Scholar), mRNA can also be detected in the olfactory zone and the medulla oblongata of the human brain (12Yuan T.T. Toy P. McClary J.A. Lin R.J. Miyamoto N.G. Kretschmer P.J. Gene. 2001; 278: 41-51Crossref PubMed Scopus (46) Google Scholar). Human PSGR shares 93% amino acid homology to the respective mouse and rat homologues, which are also expressed in the brain (12Yuan T.T. Toy P. McClary J.A. Lin R.J. Miyamoto N.G. Kretschmer P.J. Gene. 2001; 278: 41-51Crossref PubMed Scopus (46) Google Scholar). Interestingly, PSGR has numerous sequence motifs in common with the large superfamily of olfactory receptors (ORs), which build the largest class of human GPCRs and allow the recognition of a wide range of structurally diverse molecules in the nasal epithelium (13Buck L. Axel R. Cell. 1991; 65: 175-187Abstract Full Text PDF PubMed Scopus (3695) Google Scholar, 14Firestein S. Sci. STKE. 2004; 2004: pe15PubMed Google Scholar, 15Malnic B. Godfrey P.A. Buck L.B. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 2584-2589Crossref PubMed Scopus (415) Google Scholar). Recently, also the steroid hormones androstenone and androstadienone were identified as OR ligands (16Keller A. Zhuang H. Chi Q. Vosshall L.B. Matsunami H. Nature. 2007; 449: 468-472Crossref PubMed Scopus (466) Google Scholar). In addition to their role in the sensory neurons of the nose, ORs have been found in different tissues throughout the body (17Feldmesser E. Olender T. Khen M. Yanai I. Ophir R. Lancet D. BMC Genomics. 2006; 7: 121Crossref PubMed Scopus (194) Google Scholar, 18Zhang X. De la Cruz O. Pinto J.M. Nicolae D. Firestein S. Gilad Y. Genome Biol. 2007; 8: R86Crossref PubMed Scopus (130) Google Scholar). Their function(s) in these extranasal locations are questionable except for in a few cases where functional studies have been performed in spermatozoa (19Fukuda N. Yomogida K. Okabe M. Touhara K. J. Cell Sci. 2004; 117: 5835-5845Crossref PubMed Scopus (177) Google Scholar, 20Spehr M. Gisselmann G. Poplawski A. Riffell J.A. Wetzel C.H. Zimmer R.K. Hatt H. Science. 2003; 299: 2054-2058Crossref PubMed Scopus (596) Google Scholar) and in enterochromaffin cells of the gastrointestinal tract (21Braun T. Voland P. Kunz L. Prinz C. Gratzl M. Gastroenterology. 2007; 132: 1890-1901Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar). Here, we report the identification of steroid ligands of heterologously expressed PSGR and investigate the functional relevance of PSGR expression in prostate tissue. Steroid hormones elicited rapid Ca2+ responses in the LNCaP prostate cancer cell line and in primary human prostate epithelial cells. Moreover, activated PSGR causes phosphorylation of p38 and stress-activated protein kinase/c-Jun NH2-terminal kinase (SAPK/JNK) mitogen-activated protein kinases (MAPKs), resulting in reduced proliferation rates in prostate cancer cells. Reagents for cell culture use were purchased from Invitrogen, unless stated otherwise. HEK293 cells were maintained under standard conditions in minimum Eagle's medium supplemented with 10% fetal bovine serum, 100 units/ml penicillin and streptomycin, and 2 mml-glutamine. LNCaP cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum and 100 units/ml penicillin and streptomycin. PC-3 cells were maintained in Ham's F12/RPMI 1640 (1:1) supplemented with 10% fetal bovine serum and 100 units/ml penicillin and streptomycin. HEK293 cell transfections with the PSGR-containing plasmid were performed using a standard calcium phosphate precipitation technique; for siRNA experiments, LNCaP cells were transiently transfected with either targeted or scrambled siRNAs using Exgen 500 (Fermentas). 2 days after transfection, the growth medium was removed and replaced with standard Ringer solution. Prostate cancer epithelial cells were isolated from freshly collected prostate tissue, which was obtained from radical prostatectomy specimens (adenocarcinoma of the prostate pT2c, pN0, Gleason score 4 + 4) after written consent. The tissue pieces were minced to 1-mm3 pieces and digested for 30 min at 37 °C in Ringer solution containing 0.1% trypsin-EDTA. The tissue was dissociated by trituration and washed, and single cell suspensions were prepared by centrifugation of the remaining tissue pieces. Cells were seeded in the serum-free keratinocyte medium, supplemented with 50 ng/ml human recombinant epidermal growth factor 1–53 and 50 mg/ml bovine pituitary extract and kept at 37 °C in a humidified incubator with 5% CO2. When they reached 70–80% confluence, the cells were trypsinized and subcultured either in Petri dishes for Ca2+-imaging experiments or in 96-well plates for cell proliferation. Cell morphology was checked with Zeiss Axioskop 2 microscope and viewed with ×20 magnification. The following primary antibodies were used: (a) rabbit polyclonal antibodies against p44/42 MAPK and against phosphorylated p44/42 MAPK (New England Biolabs); (b) rabbit polyclonal antibodies against p38 MAPK and phosphorylated p38 MAPK (New England Biolabs); (c) rabbit polyclonal antibodies against SAPK/JNK and phosphorylated SAPK/JNK (New England Biolabs); and (d) rabbit polyclonal against PSGR (Abcam). Secondary goat anti-rabbit antibodies conjugated to horseradish peroxidase (Bio-Rad) or Alexa Fluor 546 or Alexa Fluor 488 (Invitrogen) were used. LNCaP cells were treated with 500 mm β-ionone for the indicated times, harvested, pelleted, and homogenized in lysis buffer (50 mm Tris HCl, pH 7.4, 150 mm NaCl, 1 mm EDTA, 1% Triton X-100) with protease inhibitors (Roche Applied Science Complete® protease inhibitor mixture). Sample aliquots of the cells were mixed with Laemmli buffer (30% glycerol, 3% SDS, 125 mm Tris/Cl, pH 6.8), resolved by 10% SDS-PAGE, and transferred to nitrocellulose membrane (Protran; Schleicher & Schuell). The nitrocellulose membranes were stained with Ponceau S (Sigma), blocked with TBST (150 mm NaCl, 50 mm Tris-Cl, Tween 20, pH 7.4) containing 5% nonfat dried milk (Bio-Rad), and incubated with primary antibodies diluted in 3% dry milk in TBST. After washing and incubation with horseradish peroxidase-coupled secondary antibodies, detection was performed with ECL plus or ECL Advance (Amersham Biosciences) on Hyperfilm ECL (Amersham Biosciences). Cells were incubated (45 min/37 °C) in Ringer solution containing 3 mm Fura-2-AM (Molecular Probes). After removal of extracellular Fura-2, cells were washed with extracellular solution. For the inhibitor experiments, cells were treated with 10 μm flutamide (Sigma). Ca2+ was performed as using a Zeiss microscope for were from of and were Exposure to was using a used were a of T. and J. Steroid hormones were purchased from and for potential activation of PSGR were in at experiments in HEK293 cells and for activation of LNCaP cells as ligands to Ca2+ responses in different experiments they not Ca2+ in μm was as the at the of in HEK293 cells. Human PSGR was from human by using specific that the and for the plasmid was by PSGR targeted and scrambled siRNA were with siRNA were by and the to the The siRNA sequence of PSGR was the to were and The following scrambled of the siRNA sequence was used as and LNCaP cells and primary human prostate epithelial cells were in 96-well plates at a of After at 37 °C with 5% cells were treated with different of β-ionone dihydrotestosterone 10 or with a of β-ionone and 100 and cells were with β-ionone and of inhibitors for p38 and Cell proliferation was after 3 and days using the cell proliferation RNA of human olfactory epithelium, LNCaP and human primary prostate cells was isolated with digested with and with of mRNA with was by using reverse (Invitrogen) and was performed with 2 of and specific for PSGR prostate-specific antigen androgen cytokeratin and cytokeratin The were for 1 investigate the of PSGR in prostate tissue, we to a for G-protein-coupled receptor. We therefore transiently expressed PSGR in HEK293 cells and its by the cell responses to a of using Ca2+ to the we used to for GPCRs M. Gisselmann G. Poplawski A. Riffell J.A. Wetzel C.H. Zimmer R.K. Hatt H. Science. 2003; 299: 2054-2058Crossref PubMed Scopus (596) Google Scholar, A. Zhang W. J. Hatt H. 2006; PubMed Scopus Google Scholar, C.H. M. C. M. Gisselmann G. Hatt H. J. 1999; PubMed Google Scholar). In the used for a of 100 steroid hormones from different we identified as ligands was at different for activation of PSGR and found to as a in the range The of an at with at at 4 and which are in and were for PSGR ligands of the at of the steroid were the of the at in the of the to PSGR single substances, which were and found to be were and were in a of 10 mm in HEK293 cells and not Ca2+ responses and PSGR has sequence of an olfactory receptor and because ORs can be activated by ligands with different receptor we the transiently expressed PSGR in HEK293 cells with a of odorant using Ca2+ 100 and which was used to for ORs M. Gisselmann G. Poplawski A. Riffell J.A. Wetzel C.H. Zimmer R.K. Hatt H. Science. 2003; 299: 2054-2058Crossref PubMed Scopus (596) Google Scholar, A. Zhang W. J. Hatt H. 2006; PubMed Scopus Google Scholar, C.H. M. C. M. Gisselmann G. Hatt H. J. 1999; PubMed Google Scholar). the we found β-ionone to be an in elicited Ca2+ responses at as as of the were as single and not in Ca2+ in HEK293 cells. In HEK293 cells, β-ionone not Ca2+ We that the of expressed PSGR to β-ionone can be by the structurally which not expressed PSGR and we PSGR is expressed in prostate epithelial cells. The prostate of epithelium in a The epithelium is of different a that a and is for the development of the epithelial cells, and the which is for the of and other of the C. J.M. Clin. Sci. 2005; PubMed Scopus Google Scholar). The of androgen-dependent cells, which and the androgen receptor C. J.M. Clin. Sci. 2005; PubMed Scopus Google Scholar), as as PSGR (10Xu L.L. Stackhouse B.G. Florence K. Zhang W. Shanmugam N. Sesterhenn I.A. Zou Z. Srikantan V. Augustus M. Roschke V. Carter K. McLeod D.G. Moul J.W. Soppett D. Srivastava S. Cancer Res. 2000; 60: 6568-6572PubMed Google Scholar). We human prostate epithelial cells from freshly collected prostate tissue, obtained from a radical prostatectomy we the LNCaP cell which is from a prostate cancer and was also to PSGR (10Xu L.L. Stackhouse B.G. Florence K. Zhang W. Shanmugam N. Sesterhenn I.A. Zou Z. Srikantan V. Augustus M. Roschke V. Carter K. McLeod D.G. Moul J.W. Soppett D. Srivastava S. Cancer Res. 2000; 60: 6568-6572PubMed Google Scholar). that primary to the LNCaP cell expression of the prostate epithelial cell and androgen receptor Moreover, cell PSGR prostate cancer cell not PSGR protein can be detected in LNCaP prostate cells, but not in HEK293 cells as by LNCaP cells to the of β-ionone with an in the intracellular calcium which could be by the of the specific inhibitor the of not an in the calcium and The of β-ionone not an in the intracellular calcium in PC-3 cells, which were not to PSGR and The primary cell culture the morphology of prostate epithelial cells the PSGR and β-ionone to prostate cancer cells elicited a in the intracellular Ca2+ to the Ca2+ responses in the LNCaP cell these results that PSGR is expressed in primary prostate epithelial cells and can be activated by the identified that the Ca2+ is not or by androgen receptor activation, we performed Ca2+ in the of the androgen receptor inhibitor which not the or the of the Moreover, that are known to the androgen and but not the heterologously expressed elicited Ca2+ increases in LNCaP cells increases in intracellular Ca2+ at in the range that have been identified as PSGR ligands in HEK293 cells were to an in the intracellular Ca2+ in LNCaP cells, at or 4 in not Ca2+ We to that of prostate cells with β-ionone and intracellular Ca2+ activation of PSGR the PSGR β-ionone to LNCaP cells elicited a in the intracellular Ca2+ that the Ca2+ was due to PSGR activation, we reduced the expression using a plasmid for in expression of interfering as an fluorescent which the of the and of Ca2+ responses specifically in the cells. that cells of PSGR Ca2+ in LNCaP cells were with the Ca2+ in cells not and of the Ca2+ that siRNA expression reduced the Ca2+ The responses from the cells in but on significantly in cells. of not the Ca2+ the expression of scrambled siRNA not also a Ca2+ in LNCaP cells, in the range and the of PSGR signaling in prostate cells, we treated the LNCaP and PC-3 cell and the primary prostate cells with μm β-ionone for and which resulted in a of cell proliferation in prostate cells, with a inhibition at for the LNCaP cell line A and Moreover, the of could be in LNCaP or primary cells by β-ionone A and were the cells with or The not proliferation rates of HEK293 cells or prostate cancer cells not PSGR Moreover, the of β-ionone on LNCaP cells could be by the of the specific inhibitor the intracellular of β-ionone signaling in LNCaP cells, we the phosphorylation of members of the MAP kinase which are known to be in the of cell and in prostate cells. LNCaP cells of the extracellular kinase after not but a in phosphorylation of p38 and SAPK/JNK which to 30 min of of LNCaP cells with inhibitors of p38 and SAPK/JNK kinases to with β-ionone the of β-ionone on cell proliferation The p38 MAPK inhibitor the proliferation at the only of the SAPK/JNK inhibitor 1 for and for were only not The inhibitors not cell proliferation of LNCaP cells at the used We the PSGR and identified steroid hormones and the odorant β-ionone as Moreover, we identified β-ionone as specific inhibitor for the receptor. PSGR is therefore a membrane receptor for of on intracellular signaling steroid have been identified in cells intracellular steroid receptors and in the of and for which GPCRs been identified as the of the receptor that these responses E. M. A. K. M. 2003; PubMed Scopus Google Scholar). In the we the of G-protein-coupled membrane steroid which to the superfamily of odorant receptors. to the E. M. J. Clin. 2000; PubMed Scopus Google Scholar), responses to be at steroid not be by inhibitors of transcription or and be in the of for the receptors. of these for the activation of PSGR in LNCaP cells by the androstenone derivatives that we identified as PSGR of the steroid ligands we identified as a PSGR is which can be by of is by cells of the and is the androgen in can a of and to is known to the on androgen receptors. was to be to by a to the family M. D. D. K. A. PubMed Scopus Google Scholar, K. Google Scholar). The that the of is the which the role in steroid hormone the steroid of rat that might have physiological relevance J.A. 1991; PubMed Scopus Google Scholar). The other steroid that we on to β-ionone and are not known to occur in the human body as is known to as an inhibitor the of to and is used in of cancer of the prostate A.M. C.H. J. Steroid PubMed Scopus Google Scholar). on sequence PSGR to the superfamily of odorant receptors. In ORs allow the recognition of a wide range of in the olfactory sensory neurons of the L.B. PubMed Scopus Google Scholar). Recently, human OR was identified as a steroid the steroid hormones androstenone and androstadienone in a recombinant expression (16Keller A. Zhuang H. Chi Q. Vosshall L.B. Matsunami H. Nature. 2007; 449: 468-472Crossref PubMed Scopus (466) Google Scholar). Although activation of the endogenous receptor by steroid hormones was not is that receptor also rapid steroid because in the receptor the of the respective by the of and may the OR signaling in the sensory neurons (16Keller A. Zhuang H. Chi Q. Vosshall L.B. Matsunami H. Nature. 2007; 449: 468-472Crossref PubMed Scopus (466) Google Scholar). the that ORs were found to be expressed in different tissues (17Feldmesser E. Olender T. Khen M. Yanai I. Ophir R. Lancet D. BMC Genomics. 2006; 7: 121Crossref PubMed Scopus (194) Google Scholar, 18Zhang X. De la Cruz O. Pinto J.M. Nicolae D. Firestein S. Gilad Y. Genome Biol. 2007; 8: R86Crossref PubMed Scopus (130) Google Scholar, X. M. H. Zhang X. Zou Liu J. M. G.M. Firestein Proc. Natl. Acad. Sci. U. S. A. 2004; 101: PubMed Scopus Google Scholar, P. S. G. M. Genomics. PubMed Scopus Google Scholar), the functions of these receptors due to the of the expression could only be for human M. Gisselmann G. Poplawski A. Riffell J.A. Wetzel C.H. Zimmer R.K. Hatt H. Science. 2003; 299: 2054-2058Crossref PubMed Scopus (596) Google Scholar) and mouse (19Fukuda N. Yomogida K. Okabe M. Touhara K. J. Cell Sci. 2004; 117: 5835-5845Crossref PubMed Scopus (177) Google Scholar), which are in Here, we that to the OR superfamily can be activated by identified specific receptor ligands in prostate cells. Moreover, PSGR could also be detected in human olfactory epithelium, of the role of receptor also in the olfactory Our data give support for the hypothesis that at some expressed ORs may have functions and a role for in to its sequence to odorant PSGR can be activated by the odorant a of is an found in and as a of Science. PubMed Scopus Google Scholar). have been to cancer growth and development inhibition of D.M. P. H. Proc. Biol. 1999; PubMed Scopus Google Scholar, H. J. 1999; PubMed Scopus (156) Google Scholar), but the β-ionone and have been to by a that is not to D. A. M.C. 2004; PubMed Scopus Google Scholar). The and cell of β-ionone and are therefore by the detection of a β-ionone OR in prostate epithelial cells, we a by which and other can growth of tumor cells. In the we that PSGR is a G-protein-coupled membrane steroid receptor in prostate cells. PSGR to the superfamily of ORs and a for the of at some In we found that intracellular signaling in cell are activated by which might be used in the as a in prostate cancer We H. and J. for B. of for on the and J. and T. for with