A suppressive role of guanine nucleotide-binding protein subunit beta-4 inhibited by DNA methylation in the growth of anti-estrogen resistant breast cancer cells

Breast cancer is the most common malignancy found in women worldwide and the second leading cause of cancer-related deaths among North American women [1]. Globally, it is estimated that 1.67 million new cases were diagnosed, and 522,000 women died from this disease in 2012 (GLOBOCAN 2012). Approximately 60% of these deaths were contributed by less developed countries. Although the exact etiology of breast cancer is currently unknown, the estrogen/estrogen-receptor (ER) signaling may play a crucial role in the development of this disease [2–5]. Furthermore, sustained estrogenic exposure may also increase the risk of breast, ovarian, and endometrial cancers [6].

Estrogen receptor alpha (ERα) and beta (ERβ), two ligand-inducible transcription factors, are members of the steroid/thyroid receptor superfamily that primarily mediate estrogen’s biological function through binding [7]. ERα and ERβ are encoded by either ESR1 or ESR2 genes that are composed of 595 and 530 amino acids, respectively. Both eventually form an N-terminal domain (NTD), a DNA-binding domain (DBD), and a ligand-binding domain (LBD) [8]. Sequence analysis indicates that ERα and ERβ display ~ 97% similarity in the DBD and 59% in the LBD, whereas they display only 16% similarity in the NTD [8]. This implicates a functional similarity and difference between these two ERs. For instance, once activated via estrogen binding, both dimerized ERs can either bind to the estrogen-response element (ERE) in the DNA or interplay with other transcription factors, such as AP1, Sp1, and NF-κB [9], eventually influencing the transcription of genes. However, ERα may predominantly bind to ERE elements [10], while ERβ may primarily interact with AP1 sites [11]. Furthermore, as demonstrated, ERα is a key player in promoting cell growth and proliferation [12, 13], whereas ERβ plays an important role in anti-proliferation, differentiation, and apoptosis in human malignancies, including breast cancer [14, 15].

Because ERα is expressed in 70% of breast cancers [16], and the proliferation of these ERα-positive breast cancers is largely dependent on estrogen/ERα signaling [17], the endocrine therapy that targets estrogen/ERα signaling has been well established as an effective adjuvant treatment for patients with ERα-positive breast cancers [18]. The endocrine-therapy agents that are currently used for ERα-positive breast cancer include fulvestrant (also known as ICI 182,780 and faslodex, the ER downregulator that selectively downregulates and/or degrades ERα), tamoxifen (the ER modulator that selectively antagonizes ERα function), and aromatase inhibitors (e.g. letrozole and anastrozole, which inhibit estrogen production by attenuating aromatase activity) [17, 19]. As an important adjuvant therapy, continuing 10-year tamoxifen treatment, when compared with 5-year exposure, has been shown to further reduce the risk of disease recurrence and mortality in a randomized trial of women with ER-positive breast cancers [20]. Unfortunately, long-term exposure may eventually lead to the development of acquired resistance to these drugs [21–23], which is a serious clinical problem in hormonal therapy. However, the underlying mechanisms are not completely understood.

In this study, we globally analyzed genomic DNA methylation, correlated with gene expression profiling, and identified GNB4 that was silenced by DNMT3B-mediated DNA methylation in both fulvestrant-resistant (MCF-7/182^R-6) and tamoxifen-resistant (MCF-7/TAM^R-1) breast cancer cell lines. Ectopic expression of GNB4 enhanced proliferation of MCF-7/182^R-6 and MCF-7/TAM^R-1 cell lines in response to either fulvestrant or tamoxifen, while it shortened G2 and S phases in the cell cycle. We also noted that the ectopic expression of GNB4 induced apoptosis in the MCF-7/182^R-6 cell line, whereas it attenuated the induction of apoptosis in the MCF-7/TAM^R-1 cell line. Cell-cycle and apoptosis regulators were aberrantly expressed in these cell lines in response to the ectopic GNB4 expression. In contrast, siRNA-mediated knockdown of GNB4 inhibited proliferation of two resistant cell lines in the presence of either fulvestrant or tamoxifen, and induced either S phase arrest or apoptosis. Our results provide novel insight into the role of GNB4 in the growth of both antiestrogen-resistant and sensitive breast cancer cells and may represent a target for treatment of breast cancer.

Total RNA was isolated from S05, 182^R-6, and TAM^R-1 cells using an Illustra RNAspin mini kit according to the manufacturer’s instructions (GE Healthcare Life Sciences). Quantification, purity, and integrity of the RNA samples were measured using a NanoDrop 2000c spectrophotometer (Thermo Scientific) and an Agilent 2100 bioanalyzer (Santa Clara). RNA samples with RIN values of seven or higher were used for further analysis. Procedures for library preparation, hybridization, detection, BeadChip statistical analysis, and data processing have been described previously [19]. Heatmaps were generated by Dr. Yaroslav Ilnytskyy for genes that were differentially expressed between any of the groups (ANOVA type analysis with p.adjusted < 0.001) and for top 1000 most variable probes in DNA methylation.

Although the well-established and effective endocrine therapy has provided millions of women with ER+ breast cancer with targeted treatment option [20, 23], most patients with metastatic disease would unfortunately and inevitably develop resistance to the drugs [28, 29], which has become a major clinical challenge in the treatment of this disease. As demonstrated, the reactivation (reestablishment) of ER and growth-factor signaling through crosstalk has been proposed as a primary mechanism for the development of antiestrogen resistance. A central role of upregulated EGFR/ErbB and the sustained activation of the EGFR/ErbB/ERK signaling pathway has been strongly indicated in the maintenance of antiestrogen-resistant breast cancer cell growth [30–32]. Although ERα is downregulated in tamoxifen-resistant breast cancer cell lines, the receptor is highly activated (phosphorylated) [32]. Importantly, the activated ER has been shown to promote the expression of insulin-like growth factor II (IGF2) in tamoxifen-resistant cell lines, resulting in the reactivation of insulin-like growth factor 1 receptor (IGF1R) signaling that activates EGFR via c-src-dependent phosphorylation [33]. Interestingly, the activated IGF2/IGF1R signaling may therefore in turn contribute to the phosphorylation of ER in tamoxifen-resistant cell lines [33], hence forming a positive-feedback proliferation loop. In addition, changes in the sensitivity of antiestrogen-resistant breast cancer cells to estradiol (E2) may also play a pivotal role in the development of antiestrogen resistance. Several lines of evidence have demonstrated that long-term estrogen deprivation (LTED) enhances the sensitivity of breast cancer cells to low levels of E2 (~ 10,000-fold reduction) [34–36]. The biological effect induced by E2 hypersensitivity, such as proliferation, was also mediated by the activation (phosphorylation) of ER and ERK1/2 (extracellular signal-regulated kinase 1/2) signalings. More recently, several studies highlighted a key role of protein kinases in the development of antiestrogen resistance, such as tyrosine kinase FYN and aurora kinase A [37, 38]. Interestingly, the tamoxifen-resistant breast cancer cells may derive from cancer stem-like cells [39].

To our knowledge, this study has revealed, for the first time, that GNB4 was epigenetically silenced in antiestrogen-resistant breast cancer cells, and it has highlighted an important role of GNB4 in the growth of antiestrogen resistant breast cancer cells. Heterotrimeric G proteins which are composed of an alpha subunit, a beta subunit (e.g. GNB4), and a gamma subunit function as molecular switches that play a crucial role in signal transduction from a cell surface receptor to internal effectors in a G protein-coupled receptor (GPCR) pathway [40]. Although the GPCR signaling pathway has been extensively studied, very little is known about GNB4. However, since the GPCR pathway plays a pivotal role in many biologic and pathologic processes, including tumorigenesis, it may also reflect a role of GNB4 in these processes. Evidence has demonstrated that GNB-isoforms are essential for chemokine-induced GPCR downstream signaling [41]. GNB4 may be one of the key genes that cause Charcot-Marie-Tooth disease, a heterogeneous group of the inherited neuropathies [42, 43]. GNB4 has also been linked to cancer. Haplotypes of GNB4 intron-1 have been shown to be associated with the survival rate of patients with colorectal and urothelial bladder carcinomas [44, 45]. We showed here that GNB4 was downregulated in both fulvestrant- and tamoxifen-resistant breast cancer cell lines, and that this was attributed to DNMT3B-mediated DNA methylation, because knockdown of DNMT3B by siRNA significantly elevated the expression of GNB4 at both mRNA and protein levels (Fig. 2). GNB4 has been reported to be downregulated in progressive breast cancer due to the acquired tamoxifen resistance [46], which is consistent with our results. Importantly, we found that the ectopic expression of GNB4 remarkably promoted the proliferation of antiestrogen-resistant breast cancer cells in response to antiestrogen drugs (Fig. 3). We also noted that the enforced expression of GNB4 caused cell cycles G2 and S to undergo phase acceleration (Fig. 4a and b), which may contribute to the GNB4-mediated proliferation of antiestrogen-resistant cells (Fig. 3b and c). Interestingly, the ectopic expression of GNB4 completely abolished apoptosis in tamoxifen-resistant TAM^R-1 cells, while it significantly induced apoptosis in fulvestrant-resistant 182^R-6 cells (Fig. 4c and d). However, the dramatic effects of GNB4 on apoptosis in both cell lines may all contribute to GNB4-mediated cellular proliferation (Fig. 3b and c). Recently, several interesting studies have indicated that apoptotic cells could promote the proliferation of surrounding cells (apoptosis-induced proliferation) due to the mitogenic signals released by the apoptotic cells [47–49]. siRNA-mediated GNB4 knockdown, however, suppressed proliferation of 182^R-6 and TAM^R-1 cells in the presence of antiestrogen drugs, and induced S-phase arrest of 182^R-6 cells and apoptosis of TAM^R-1 cells (Fig. 6a and b), further validating a crucial role of GNB4 in the development of antiestrogen resistance of breast cancer cells. GNB4 siRNA had no effect on GNB4 expression in 182^R-6 cells at 72 h after transfection (Additional file 2: Figure S2), may reflect a longer half-life time of GNB4 in this cell line, since in another independent experiment, GNB4 was noted to be downregulated at 96 h after transfection (Additional file 3: Figure S4A and B).

GNB4 is a key component of heterotrimeric G proteins, which play an essential role in the transduction of GPCR-mediated signaling. Because of the crucial role of GPCR in the activation of AKT (v-AKT murine thymoma viral oncogene homolog) and ERK1/2 pathways [50, 51], an increase in phosphorylated AKT and/or phosphorylated ERK1/2 was expected to be seen in antiestrogen-resistant cells in response to the ectopic GNB4 expression. As expected, the phosphorylated AKT and phosphorylated ERK1/2 were elevated in 182^R-6 cells in response to the GNB4 expression (Fig. 5b). This may contribute to GNB4-mediated cellular proliferation (Fig. 3b). However, the enforced expression of GNB4 caused a reduction in the phosphorylated AKT and had no effect on the phosphorylated ERK1/2 in TAM^R-1 cells (Fig. 5b). The mechanism involved is unclear. It may be interesting to look at the crosstalk with other pathways, such as ER. Importantly, the GNB4-induced upregulation of BAX in 182^R-6 cells and the downregulation in TAM^R-1 cells may contribute to GNB4’s effect on apoptosis in these cells (Fig. 5b and Fig. 4c and d). We also noted that GNB4 caused an induction in BCL2 in 182^R-6 cells, while it had no effect in TAM^R-1 cells (Fig. 5b), implicating that the functional effect of GNB4 on apoptosis in 182^R-6 cells may be due to the balance between proapoptotic (BAX) and antiapoptotic (BCL2) proteins.

Cyclins and cyclin-dependent kinases control cell-cycle progression and transitions. Cyclin A interacts with cyclin-dependent kinase 2 (CDK2) or CDK1 to form a complex that governs the S phase of the cell cycle [52]. Cyclin D interacts with CDK4 or CDK6 to form a complex that controls the G1 phase of the cell cycle [49]. However, in a complex with CDK2, cyclin E controls the S-phase progression and G1-S transition [52, 53]. The results we presented here showed that cyclin E and CDK2 were upregulated in 182^R-6 cells in response to GNB4 (Fig. 5a), which may contribute to the shortened S phase (Fig. 4a). The ectopic GNB4, however, attenuated the expression of CDK2 and cyclin A and E in TAM^R-1 cells (Fig. 5a), which may contribute to the S-phase arrest (Fig. 4a). Although the ectopic GNB4 caused a profound induction in both cyclin D1 and CDK6 in 182^R-6 cells (Fig. 5a), this induction had no effect on the G1 phase of the cell cycle (Fig. 4a). The GNB4-induced upregulation of CDK6 and downregulation of cyclin D1 had no effect on the G1 phase of TAM^R-1 cells (Fig. 5a and Fig. 4b). We also noted that the ectopic GNB4 had no effect on the expression of p21, an inhibitor of cyclin D/CDK6 and cyclin E/CDK2 complexes [54]. However, the mechanism underlying the shortened GNB4-induced G2 phase of the cell cycle in TAM^R-1 cells is still unclear. Although we did not measure the expression of CDC25 and speedy/ringo C in these cells, it has been demonstrated that these two molecules play an important role in controlling G2-phase progression [55, 56].

In summary, although the observed reduced level of GNB4 in the two antiestrogen resistant cell lines is not the underlying cause of antiestrogen resistance, GNB4 is important for growth of both antiestrogen resistant and antiestrogen sensitive breast cancer cells and thereby a target for treatment of breast cancer.