The prostate-specific G-protein-coupled receptor, PSGR, was first identified by three distinct laboratories using different experimental approaches (Xu et al. 2000; Xia et al. 2001; Yuan et al. 2001). The main purpose of the search was to find prostate-specific genes with potential diagnostic utility. PSGR was selected due to its prostate-specific expression and its overexpression in prostate cancer, which was suggestive of biomarker potential. Initial research on this molecule focused on its expression pattern, regulation, and probable role as a biomarker for prostate cancer. For an in-depth summary, please refer to the previous edition of this encyclopedia. Since then, research has emerged on the underlying mechanisms of its role in prostate cancer initiation and progression and its potential ligands. Herein we provide a review and update on the PSGR literature.
PSGR as a Molecule: Characterization
During its initial characterization, PSGR was described as a member of the olfactory subfamily of G-protein-coupled receptors (GPCRs) (Xia et al. 2001). It was localized to chromosome 11p15, with an open reading frame of 320 amino acids (Xu et al. 2000). It was also highly homologous (50% identity, 70% similarity) to GPCR odorant receptors, with the presence of seven olfactory motifs in the amino acid sequence (Yuan et al. 2001). Expression was localized to prostate epithelial cells and the metastatic, androgen-sensitive prostate cancer cell line, LNCaP (Xu et al. 2000). The receptor protein was specifically found in the human prostate and overexpressed in prostate cancer (Xia et al. 2001; Xu et al. 2000). PSGR expression was significantly increased in prostate tumor samples (62%) when compared to normal prostate tissue (Xu et al. 2000; Xia et al. 2001). Results of a comparison between human normal versus tumor tissue samples found overexpression of PSGR in both prostate intraepithelial neoplasia (PIN) and prostate carcinoma (PCa) (~76%) (Weng et al. 2005). Since then, PSGR mRNA was found in the metastatic, androgen-sensitive prostate cancer cell line MDA PCa 2a, but not in any of the other prostate cancer cell lines tested, including MDA PCa 2b (also metastatic, androgen sensitive), DU145, PC3, LAPC4, and VCaP (Rodriguez et al. 2014).
Two distinct exons were described for PSGR. The first was a short, noncoding exon (exon 1) and the second, a long, coding exon (exon 2), both separated by a 14.9 kb intron. In addition, two regulating promoters were also reported, one located within exon 1 and its upstream region, containing a TATA box sequence. The other was located upstream of exon 2, with no TATA box or GC-rich sequences, but diverse cis-elements, including two GATA factor-binding sites, a Lom2 site and a Freac-6 site (Weng et al. 2005), suggesting the presence of a shorter PSGR transcript. Luciferase assays containing the 5′ flanking regions of exon 1 and exon 2 tested the activity of both promoters. They showed that the TATA box at −31 was required for promoter activity and that exon 1 contains transcription factor-binding sites that enhance PSGR gene expression (Weng et al. 2005; Rodriguez et al. 2016a). Promoter activity for exon 2 was cell-type dependent, suggesting cell-specific functions (Weng et al. 2005). Both inflammatory (mediated by IL-6) and androgenic pathways showed potential regulatory function over PSGR, but further work remains in discerning the network of transcription factors that regulates its highly tissue-specific expression pattern (Rodriguez et al. 2016a). No further studies focused on transcriptional regulation of PSGR have been published since our previous edition.
PSGR as a Biomarker for Prostate Cancer
As part of the initial characterization studies, Weng et al. compared PSGR expression levels in human normal prostate with benign prostatic hyperplasia and PCa samples. Of these, 76% of PCa had increased PSGR expression, found predominantly in prostate epithelium and absent from the stroma (Weng et al. 2005), further supporting a potential role for PSGR as a PCa biomarker. Though no correlation of PSGR expression with factors such as clinical stage, patient age, or recurrence status was found, its high levels of expression made it potentially useful for early stage diagnosis of PCa (Ashida et al. 2004). This possibility was subsequently analyzed in conjunction with the already known diagnostic marker α-methylacyl-CoA racemase (AMACR) and the PSGR-related gene, PSGR2. The study was performed in biopsy samples, and therefore, did not address presurgical screening of PCa; however, here it was determined that alone PSGR was not a better diagnostic marker than AMACR, but in conjunction with PSGR2 and AMACR, all three genes were able to identify 80% of PCa samples, with no false positive diagnoses (Wang et al. 2006). For a presurgical screening approach, studies of PSGR expression in both blood and urine samples were performed. In blood samples, PSGR was not specific and was thus of little diagnostic use (Cardillo and Di Silverio 2006). In urine, PSGR was a significant predictor of PCa (sensitivity = 0.59 [0.47–0.70], specificity = 0.73 [0.65–0.80]), but again, had better diagnostic performance when combined with other PCa markers, such as prostate cancer gene 3 (PCA3) (sensitivity = 0.77 [0.65–0.85], specificity = 0.60 [0.5–0.68]) (Rigau et al. 2010, 2011). Overall, although blood screening showed little potential for clinical screening, PSGR detection in urine was able to improve on PCA3 detection and decreased the possibility of false negatives for PCa patients. Thus, it shows promise for the diagnosis of early-stage disease, such as high-grade PIN (HGPIN) and potential as a predictive marker for positive biopsy results (Sequeiros et al. 2015).
Mechanisms of Action
Since the previous edition, the majority of new findings for PSGR have been on its role in prostate cancer initiation and progression. While human PSGR has prostate-specific expression, the PSGR homologue in mice, olfr78 or MOR18-2, is not endogenously expressed in mouse prostate, but found in mouse brain and olfactory epithelium (Conzelmann et al. 2000; Yuan et al. 2001), as well as carotid glomus cells where it regulates breathing (Chang et al. 2015). To investigate its role in prostate carcinogenesis, human PSGR was overexpressed in mice under the control of the androgen-dependent probasin promoter in a FVB/N background (Rodriguez et al. 2014). PSGR transgenic mice developed an increased inflammatory response, with a reactive stroma, inflammatory cell infiltration, and epithelial cell shedding into the prostatic lumen, potentially explaining the presence of PSGR in urine samples described by Rigau et al. Transgenic mice also developed low-grade PIN (LGPIN) that did not progress into HGPIN or PCa. Mechanistically, PSGR overexpression increased cell proliferation, had no effect on cell apoptosis, and decreased androgen receptor (AR) expression in epithelial cells. Overexpression of PSGR in LnCAP cells resulted in a fivefold increase in xenograft prostate tumor mass after orthotopic injection, compared to control LnCAP cell xenograft tumors, suggesting that receptor number controls cell proliferation rates. PSGR overexpressing xenografts showed increased inflammatory cell infiltration as well. Overall, although PSGR was linked to an increased inflammatory response in the prostate, which potentially promotes carcinogenesis, no evidence was found that PSGR alone was sufficient to induce PCa (Rodriguez et al. 2014).
To study a potential role in prostate cancer progression, PSGR transgenic mice were crossed with a PTEN-deleted mouse model to generate PSGR-PTENΔ/Δ mice. Phosphatase and tensin homolog (PTEN) is a phosphatase that acts as a negative regulator of the PI3K-Akt signaling pathway and is a commonly deleted gene in PCa, supporting a tumor suppressor role (Rodriguez et al. 2016b). Double mutant mice had faster tumor progression and larger tumors than PTEN knockout alone. They showed decreased levels of epithelial E-cadherin expression and enhanced NFκB activation, suggesting a cooperative role in promoting invasion and metastasis. Of note, PTEN knockouts alone did not increase NFκB activation, suggesting a potential model where PTEN loss accentuates PSGR-driven prostate inflammation. Importantly, PTEN null prostates lost expression of both epithelial and stromal androgen receptor (AR); however double mutant mice regained expression of stromal AR (Rodriguez et al. 2016a, b). Reemergence of AR is a hallmark of castration-resistant prostate cancer (Shen and Abate-Shen 2010). Stromal AR expression was correlated with active NFκB, suggesting a link between PSGR, NFκB, and AR, a potential cooperation leading to hyperactivation of NFκB, and, possibly, a poor prognosis (Rodriguez et al. 2016b).
Ligands and Signaling
Previously, a potential ligand capable of activating PSGR was described (Neuhaus et al. 2009). This group identified androgen-related compounds as ligands for PSGR, the most potent being 1,4,6-androstatriene-3,17-dione (ADT) at ~10 nM, followed by the isoprenoid β-ionone and the steroid 6-dehydrotestosterone in the 100–200 μM range (Neuhaus et al. 2009). Treating LnCAP cells with these ligand concentrations, however, inhibited cell proliferation and increased invasion in vitro, suggesting the activation of distinct downstream signaling pathways from those previously described in the transgenic mouse models (Rodriguez et al. 2014, 2016b). In vivo, β-ionone had no effect on metastasis compared to vehicle control in a subcutaneous tumor model (Sanz et al. 2014). Antiproliferative effects after β-ionone treatment were also found in gastric cells not known to express PSGR (Dong et al. 2013), suggesting potential alternate mechanisms of β-ionone function.
GPCR downstream signaling pathway characterization is highly dependent on the endogenous ligand. For PSGR, a high-throughput screen using specialized HEK293 cells expressing a CRE-luciferase reporter was used to test a library of 93 odorants and identified propionic acid (EC50 ~ 100 μM) as a potential PSGR ligand (Saito et al. 2009). Curiously, the mouse homolog, olfr78, is activated by proprionate at lower potency (EC50 ~ 2 mM) (Pluznick et al. 2013). A separate study found that stimulation of LnCAP cells with 10% bovine serum activated inflammatory pathways in vitro and in vivo, suggesting the existence of a potential PSGR ligand in serum (Rodriguez et al. 2014). This unknown ligand led to NFκB activation and target gene expression that was blocked by the Akt inhibitor MK-2206, suggesting that Akt is a downstream target of PSGR (Rodriguez et al. 2014). The effector proteins linking PSGR to Akt remain to be clarified, but the evidence suggests that PSGR may have multiple ligands, with a predominant in vivo role linked to inflammation, which is independent of β-ionone stimulation.
The results shown in this chapter represent important steps forward in the understanding of the roles of PSGR. However, much work yet remains. Most importantly, finding the endogenous ligand for the receptor will allow for a more comprehensive study of the signaling pathways regulated by this receptor. Two major pathways have so far been described: first, initiated by binding of β-ionone to PSGR, results in inhibition of cell proliferation and increased cell invasion, with concomitant phosphorylation of p38 and SAPK/JNK kinases shortly after agonist binding and second, activated by an as yet unknown ligand present in bovine serum, increases cell proliferation, activates prostatic inflammatory pathways, and suggests a potential link between inflammation, androgen receptor, and PSGR. Alternatively, whether propionic acid is a ligand linked to either of these two pathways or capable of activating its own downstream signaling remains to be elucidated. Discerning the pathways downstream of the endogenous ligand will play an enormous role in the path research will follow for this receptor, as these pathways can possibly be used to improve upon PCa diagnostics or therapy.
- Ashida S, Nakagawa H, Katagiri T, Furihata M, Iizumi M, Anazawa Y, et al. Molecular features of the transition from prostatic intraepithelial neoplasia (PIN) to prostate cancer:genome-wide gene-expression profiles of prostate cancers and PINs. Cancer Res. 2004;64(17):5963–72.PubMedCrossRefGoogle Scholar