PELP1: A novel therapeutic target for hormonal cancers

Summary

Recent studies implicate the estrogen receptor (ER) coregulator proline-, glutamic acid-, and leucine-rich protein (PELP) 1 as playing critical roles in ER-genomic, ER-nongenomic and ER-signaling cross talk with growth factor signaling pathways. PELP1 expression is deregulated in hormonal cancers and recent studies further elucidated the molecular mechanisms by which PELP1 regulates hormone therapy response. Although PELP1 is important for normal functions of the ER, the possibility to target ER-PELP1 axis appears to be an effective strategy for preventing hormonal carcinogenesis and therapy resistance. Thus, PELP1 may be useful as prognostic marker for hormonal cancers and PELP1 signaling may be useful to generate targeted therapeutics to overcome hormonal therapy resistance.

INTRODUCTION

The human estrogen receptor (ER) is a key transcriptional regulator in breast cancer biology. The majority of the human breast cancers start out as hormone-dependent. (1;2). Endocrine therapy using tamoxifen, a selective estrogen receptor modulator (3) and aromatase inhibitors, which ablate estrogen (E2) biosynthesis, has been found to substantially improve disease-free survival (4). Despite these positive effects, initial or acquired resistance to endocrine therapies frequently occurs. Emerging evidence suggests that ER action is complex, involves genomic as well as nongenomic signaling events (5-7), and requires functional interactions with coregulators (8). As a modulator of ER functions, ER-coregulators are likely to play a role in breast cancer progression and emerging data suggest that the level and activity of coregulators can specify hormonal sensitivity of tumors. Advances in research during the past decade have identified several novel proteins as being ER coregulators (9). Even though coregulators modulate ER functions, each coregulator appears to play an important and a non-overlapping function in vivo (10-12). One of the ER coregulators, proline-, glutamic acid-, leucine-rich protein-1 (PELP1) (13), (also previously known as modulator of the nongenomic actions of the estrogen receptor, MNAR) (14), is unique because it plays important roles in both genomic (15) and nongenomic actions of the ER (16;17). In this review we have summarized the current perspective on the significance of PELP1 signaling axis in ER biology and the novel possibility of including PELP1 as a therapeutic target in treatment regimens along with current endocrine therapies.

PELP1 MOLECULAR ASPECTS

PELP1 contains several motifs and domains that are commonly present in many transcriptional co-activators, including 10 nuclear receptor (NR)-interacting boxes (LXXLL motifs), a zinc finger motif, a glutamic acid-rich domain, and 2 proline-rich domains (13). Accordingly, PELP1 is shown to interact with and modulate functions of several nuclear receptors and transcriptional activators including AR (18;19), ERR (20), GR (21;22), PR (13), RXR (23), FHL2 (19), AP1 (20;24) and STAT3 (25). A distinctive feature of PELP1 is the presence of an unusual stretch of 70 acidic amino acids in the C-terminus that functions as a histone-binding region (15;24). PELP1 is localized both in the nuclear and cytoplasmic compartments and had unique functions in both compartments. PELP1 directly interacts with several cytosolic kinases including the p85 subunit of PI3K (17), SH3 domain of c-Src via its N-terminal PXXP motif (26). PELP1 is shown to be phosphorylated by several kinases including PKA (27), HER2 (17), Src (26), CDKs (28) and its phosphorylation is modulated by estrogen and growth factors (25).

PELP1 IN ER GENOMIC SIGNALING

PELP1 functions as a coactivator of both ERα (13) and ERβ (29) and modulates their transactivation functions. PELP1 co-localizes and interacts with the acetyltransferases CBP and p300 (15). PELP1 promotes and maintains the hypoacetylated state of histones at the target chromatin, and ER binding reverses its role to hyperacetylate histones (24). PELP1 is also associated with deacetylases, including components of nucleosome remodeling and the histone deacetylation complex (30), and inhibition of deacetylase activity increases the PELP1 residency time at the target gene promoter (15). Researchers have found reduced levels of linker histone H1 in chromatin surrounding the region targeted by PELP1 (15). Furthermore, PELP1 was found to exist in a complex with histone modifying enzymes like histone methyltransferases and demethylases (31). Our ongoing studies shows that PELP1 interacts with the Histone H3 methylase (SETDB1/ESET) and lysine demethylase (KSD1) proteins and has potential to modulate enzymatic activities of SETDB1 and KSD1 (32). Taken together, these findings suggest that PELP1 contributes to and perhaps directs ER-mediated alterations in the chromatin that is required for the optimal transcriptional response by regulating the activities of chromatin-modifying enzymes and by facilitating optimal histone code at ER target genes.

PELP1 IN ER NON-GENOMIC SIGNALING

PELP1 participates in ER cytoplasmic and membrane-mediated signaling (MIS) events by coupling the ER with several cytosolic kinases (33). PELP1 modulates the interaction of ERs with cSrc, stimulating cSrc enzymatic activity, leading to the activation of the mitogen activated protein kinase (MAPK) pathway (34). PELP1 directly interacts with the p85 subunit of PI3K and enhances PI3K activity (17). Greger et al., reported a direct correlation between PELP1 expression levels and E2-induced activation of PI3 and Akt kinases. Mechanistic studies found that E2 treatment induced complex formation between PELP1, ERα, cSrc, and p85, the regulatory subunit of PI3K. The interaction between p85 and PELP1 requires activation of cSrc and PELP1 phosphorylation on Tyr 920 (26). These findings implicate that PELP1 acts as a scaffolding protein promoting ER interactions with intracellular kinases and this facilitate activation of ER-nongenomic signaling pathways.

PELP1 IN ER SIGNALING CROSS TALK

Growth factors promote tyrosine and serine phosphorylation of PELP1 (17) (35). PELP1 interacts with several growth factor receptors, including EGFR and HER2 (17;36). Growth factor-induced phosphorylation of PELP1 is shown to affect subnuclear localization of PELP1 and also influence PELP1-mediated ER transactivation functions (27). PELP1 interacts with signal transducers and activators of transcription 3 (STAT3), and PELP1-STAT3 interactions play a mechanistic role in the positive regulation of STAT3 transcription (37). Overexpression of PELP1 potentiates phosphorylation of STAT3 at Ser 727 and plays an important role in the growth factor mediated STAT3 transactivation functions. Such regulatory interactions of PELP1 have important functional implications in the cross-talk of estrogen receptor and growth factor signaling (37). PELP1 also interacts with hepatocyte growth factor receptor regulated tyrosine kinase substrate (HRS) and HRS was found to sequester PELP1 in the cytoplasm, leading to the activation of MAPK in a manner that is dependent on the epidermal growth factor receptor (38). Collectively, these findings suggest that growth factor signals promote phosphorylation of PELP1. Since growth factor signals play key roles in hormonal therapy resistance, PELP1 interactions with growth factor axis may have functional implications in breast tumors with deregulated growth factor signaling.

PELP1 IN CELL CYCLE PROGRESSION

PELP1 signaling plays an important role in E2-mediated cell cycle progression (26;39). PELP1 is a Retinoblastoma (pRb)-interacting protein and PELP1 deregulation is shown to promote cyclin D1 expression (39). Mechanistic studies found that PELP1 plays a permissive role in E2-mediated cell cycle progression by enhancing E2 mediated G1-S progression. Increased PELP1 expression found in mammary glands during pregnancy also supports a physiological role for PELP1 in E2-mediated cell cycle progression in mammary glands (13). Evidence also indicates that PELP1 plays a role in meiosis by interacting with the androgen receptor (AR) and appears to mediate inhibition of meiosis via Gβγ signaling. Ligand activation and binding to the AR overcomes this inhibition, which allows maturation to occur (40). Recently completed work in our laboratory demonstrates that PELP1 is a novel substrate of both CDK4 and CDK2. PELP1 is recruited to several E2F target genes; functions as an E2F coregulator and mutation of CDK phosphorylation sites in PELP1 abolishes PELP1-mediated E2F transactivation functions (28). CDK phosphorylation of PELP1 may confer growth advantage to breast epithelial cells and thus contribute towards tumorigenesis by accelerating cell cycle progression.

PELP1 DEREGULATION IN HORMONAL CANCERS

Several lines of evidence implicate PELP1 as a potential proto-oncogene and its expression is deregulated in a wide variety of hormone-driven tumors including breast (17;41-43), endometrial (29), ovarian (44) and prostate cancer (19). Overexpression of PELP1 in fibroblasts and epithelial model cells results in cellular transformation and PELP1 over expression in breast cancer model cells potentiates rapid tumor growth in xenograft studies (43). PELP1 interacts with and modulates functions of several proto-oncogenes, including c-Src, STAT3, HER2 and EGFR (45). PELP1 has been reported to be widely expressed in breast cancer cells (13;46) and is shown to be up regulated in a subset of breast tumors (13;47). A very recent study identified PELP1 as a cell fate determination factor (DACH1)-binding protein. DACH1 functions as an endogenous inhibitor of ERα transcriptional activity. Using 2,200 breast tumor samples, Popov et al., found that DACH1 expression is lost during breast cancer progression (48). Using mechanistic studies, they also demonstrated that DACH1 and PELP1 colocalize in the nucleus in ~80% of cells and that the relative balance of DACH1 and PELP1 in breast cancer cells has implications in ER signaling as DACH1 loss can potentiate PELP1 coactivation functions of ERα (48). Salivary duct carcinoma is a high-grade neoplasm with morphology similar to that of mammary duct carcinoma. Interestingly, these salivary tumors express PELP1 and ER, and PELP1 signaling may play a role in salivary tumorigenesis (49). Grivas et al., reported recently that PELP1 expression was higher in epithelial cells of colon carcinomas than in normal mucosa and that PELP1 overexpression in epithelial cells was found to be an independent favorable prognostic factor (50). PELP1 expression in myofibroblasts from normal mucosa as well as in carcinomas suggests that deregulated expression of co-regulators in both epithelial cells and myofibroblasts may contribute to the initiation and progression of colorectal carcinoma (51). Marquez-Garban et al., reported deregulated PELP1 expression in lung tumors (52). These emerging findings suggest that PELP1 is a novel proto-oncogene and its deregulation can potentially contribute to oncogenesis in hormone-driven cancers.

ROLE OF PELP1 IN METASTASIS

PELP1 interacts with several proteins involved in cytoskeleton remodeling, including Src kinase, PI3K, four and a half LIM protein 2, and ILK1 (17;53-55). PELP1 over expression uniquely enhances E2-mediated ruffles and filopodium-like structures and model cells that over express PELP1 exhibit increased cell motility (43). PELP1 modulates functions of metastasis-associated protein 1 (MTA1), via its interactions with MTA1-associated co-activator MICOA and promotes ERα-transactivation functions in a synergistic manner (30). Additionally, PELP1 modulates expression of MTA3, a gene implicated in the invasive growth of human breast cancers (56). Using breast tumor prognostic arrays, Rajhans et al., found that node-positive and metastatic tumors had higher PELP1 expression than the expression in node-negative specimens. Node-positive and metastatic tumors exhibited a greater expression of PELP1 than did node-negative tumors (P = 0.003) (43). Our ongoing studies have also shown that PELP1 mediated ER-nongenotropic actions play a role in cell motility/metastasis (55). Habashy et al., investigated the clinical and biological relevance of PELP1 protein expression in 1,162 patients with invasive breast cancers and found that increased PELP1 expression is associated with tumor clinical parameters, shorter breast cancer-specific survival (BCSS) and disease-free incidence (DFI), implicating PELP1 protein expression as an independent prognostic predictor of shorter BCSS and DFI in breast cancer (41). The ability of PELP1 to interact with various enzymes that modulate the cytoskeleton and its putative deregulation in metastatic breast tumors suggest that PELP1 signaling plays a role in tumor cell migration and metastasis.

ROLE OF PELP1 IN LOCAL ESTROGEN SYNTHESIS

In situ estrogen synthesis has been implicated in tumor cell proliferation through autocrine or paracrine mechanisms especially in postmenopausal women. Several studies have reported a significant increase in activity of aromatase (Cyp19), a key enzyme involved in E2 synthesis in breast tumors (57). Recent studies from our laboratory suggested that PELP1 deregulation contributes to increased expression of aromatase and local E2 synthesis, and that PELP1 cooperates with growth factor signaling components in the activation of the aromatase gene. PELP1 deregulation uniquely contributed to this via activation of the aromatase promoter I.3/II (20). Mechanistic studies found that PELP1 interacts with LSD1 (32), that PELP1-LSD1-mediated histone epigenetic modifications play a role in the local aromatase expression and that PELP1 deregulation could play a role in modulating histone methylation at the aromatase promoter region (58). Immunohistochemistry (IHC) analysis of breast tumor arrays revealed that substantial number of PELP1 overexpressing breast tumors also overexpressed aromatase when compared with PELP1 low-expressing tumors (20), suggesting that PELP1 regulation of aromatase represents a novel mechanism for in situ estrogen synthesis, leading to tumor proliferation by an autocrine loop and thus opening a new avenue for ablating local aromatase activity in breast tumors.

ROLE OF PELP1 IN HORMONAL THERAPY RESISTANCE

Although the mechanisms for hormone therapy resistance remains elusive, recent studies suggest that the presence of alternative signaling pathways including ER nongenomic signaling and ER-cross talk with the growth factor components contributes to hormone therapy resistance (16;54;59). PELP1 deregulation appears to play an important role in the development of hormone therapy resistance (25). PELP1 plays an essential role in ER-nongenomic actions by coupling the ER with Src and PI3K pathways (16;54). In addition, PELP1 interacts with growth factor signaling components and participates in the ligand-independent activation of ER (25). Although PELP1 is predominantly localized in the nucleus of hormonally responsive tissues (13;41), it exhibits cytoplasmic localization in a subset of tumors from breast and endometrial tissues and salivary glands (17;29;35;37;49). Further, breast cancer model cells mimicking PELP1 cytoplasmic expression showed resistance to tamoxifen via excessive activation of the c-Src signaling axis (17). Recently, a transgenic (Tg) mouse model that uniquely expresses PELP1 in the cytoplasm (PELP1-cyto) was developed (60). By 12 weeks of age, mammary glands of these Tg mice developed widespread hyperplasia with increased cell proliferation and exhibited to resistance to tamoxifen therapy. In another study, IHC analysis of tumors from human patients indicated that cytoplasmic localization of PELP1 was an independent prognostic marker for determining response to tamoxifen (60). These findings suggest that PELP1 localization could be used as a determinant of hormone sensitivity or resistance. This evidence strongly suggests that PELP1 deregulation has potential to confer resistance of breast tumors to hormonal therapy.

THERAPEUTIC TARGETING OF PELP1

With several studies hinting towards the therapeutic benefits of targeting PELP1 in hormonal cancer initiation, progression and metastasis, it has become increasingly necessary to develop strategies to effectively interfere with PELP1 signaling or to silence the expression of PELP1 in cancers. Currently, there are no pharmacological inhibitors of PELP1. Therefore, targeting critical components of PELP1 signaling axis such as PELP1-Src axis, PELP1-CDK2, and PELP1-LSD1 axis could be used as a therapeutic option. Src kinase play an essential role in the activation of PELP1-mediated non-genomic signaling leading to cell migration, metastasis, and local E2 synthesis. Pharmacological inhibition of Src kinase using dasatinib (BMS-354825) significantly inhibits activation of PELP1-mediated tumorigeneic functions (61). Therefore, the Src inhibitor dasatinib holds a therapeutic promise in blocking the PELP1 signaling axis. PELP1 is novel substrate of CDKs and inhibition of CDK function by roscovitine is effective in reducing PELP1-mediated therapeutic potential (62). Tumor cells overexpressing PELP1 in the cytoplasm are distinctly sensitive to TNFα-induced apoptosis. Therefore, TNFα-based therapeutics could also be used as a strategy for tumors with cytoplasmic PELP1 expression (63). Pargyline is a selective monoamine oxidase inhibitor that blocks LSD1 activity (64) and is approved by FDA for treatment of moderate to severe hypertension. Since PELP1 interacts with LSD1 and promotes epigenetic changes at ER target genes, inhibition of PELP1-LSD1 axis by pargyline could also be explored as an important therapy option. A recently conducted study in our laboratory indicated that treatment of breast cancer cell lines which overeexpress PELP1 with pargyline substantially inhibited PELP1 tumorigeneic functions, thus using pargyline may have therapeutic utility in PELP1 deregulated tumors (58). Ohshiro et al., found that resveratrol, a well-established phytoestrogen and chemopreventive agent, promotes PELP1 localization to autophagosomes. This association may be important in the process of cell death by autophagy in the cancer cells and suggests that resveratrol could be used to induce cell death via autophagy induction in PELP1-deregulated tumors (65). We recently developed PELP1 siRNA nanoparticles using chitosan (66), a biocompatible and biodegradable polymer to down regulate PELP1 expression. We tested and successfully confirmed the novel possibility of using PELP1siRNA nanoparticles to silence PELP1 expression in breast cancer cells. Our ongoing studies are using this approach to examine whether PELP1 downregulation sensitize therapy resistant breast cancer cells to hormone therapy (62;67).

CONCLUSIONS AND FUTURE DIRECTIONS

Accumulating evidence strongly suggests that PELP1 is a novel proto-oncogene, whose expression is deregulated in many hormonal cancers. Molecular studies implicate PELP1 as a molecular scaffold that allows the ER to form distinct complexes to facilitate signal transduction. Deregulation of PELP1 expression or functions will have wide implications in ER signaling and therapy response. For substantial improvement in cancer therapy, nonsingular approaches combining more than one treatment agent into one delivery vehicle shows to be quite promising. Therefore, combining the chemotherapeutic drugs together with PELP1 axis targeting drugs appears to be promising for effective control of hormone therapy resistant tumors. Future studies elucidating the molecular mechanism of PELP1 actions and the molecular composition of PELP1-associated complexes in tumor cells compared to normal cells, characterizing the physiological function of PELP1 using Tg/knockout mouse models, and identifying genome-wide PELP1 target genes will assist in the development of novel therapeutic targets to block PELP1 signaling axis. Since PELP1 expression is deregulated in hormonal tumors, understanding the mechanisms that contribute to PELP1 deregulation will provide novel avenues for therapeutic intervention. Future studies are also warranted to extend siRNA therapeutics to specifically down regulate PELP1 expression in hormonal cancers by using a targeted delivery approach. PELP1 appears to have a potential application in assessing the clinical outcome of patients with ER-positive breast cancer. Since few published studies show that PELP1 expression correlates with therapy resistance and survival, future studies using a large number of advanced and therapy-resistant tumors are needed to establish PELP1 as a prognostic marker for hormonal tumors.

Figure 1.

Schematic representation of the current understanding of PELP1 signaling pathway. PELP1 appears to function as a scaffolding protein that facilitates formation of various signaling complexes and its expression is deregulated in hormonal cancers. PELP1 regulates multiple signaling pathways including ER-genomic signaling by facilitating epigenetic changes, ER-nongenomic signaling by activating Src-MAPK pathway, local estrogen synthesis via ERRα - aromatase signaling and cell cycle progression via pRb/E2F pathway. PELP1 deregulation in hormonal tumors is likely to contribute to the development of tumorigenesis, metastasis and hormonal independence by activating multiple signaling pathways and by facilitating ER signaling cross talk. Insert in the middle shows PELP1 expression in advanced breast tumors as analyzed by IHC.