The regulatory role of miRNAs on VDR in breast cancer

ABSTRACT

Triple negative breast cancer (TNBC) has been associated with the lack of three hormone receptors; estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2). However, a host of other steroid hormone receptors such as vitamin D receptor (VDR) is present in TNBC, and the role of these hormone receptors in breast tumorigenesis is unclear. The levels of microRNAs (miRNAs) are also expressed differently than in normal mammary epithelial cells. miRNAs are regulatory RNAs involved in various cellular functions, mainly gene silencing. Here, we reviewed the literature surrounding miRNAs in breast cancer, and performed in silico analysis to determine whether there was a correlation between levels of VDR in relation to miRNAs important in breast cancer development and tumorigenesis. We identified three miRNAs of interest, specifically, miR-23, miR-124, and miR-125. Through this research we determined the possibility that these miRNAs play an important role in controlling VDR activity and by virtue the development of breast cancer.

KEYWORDS:

• VDR
• miRNAs
• miR-125
• tumor suppressor
• biomarker
• treatment resistance
• breast cancer

Introduction

Triple negative breast cancer (TNBC) is defined as lacking three major steroid and growth receptors crucial in supporting breast cancer growth and development. Scientists have been investigating the etiology of this disease as well as the genetic drivers.⁴³ While TNBC is associated with the lack of ERα, PR, and HER2, TNBC does in fact express a variety of other hormone-receptors that may be culprits in disease onset or progression.^69,58 This is a relatively new concept and few investigators have thought of TNBC as a hormonally responsive tumor. This is because the clinical definition is derived from the concept that patients with these tumors do not respond to standard hormone-based therapies such as tamoxifen and raloxifene.^15,29 However, these treatments have been developed for estrogen-responsive tumors, and are based on the notion that the steroid-receptor pathways elucidated in the early 1960s promote breast tumor growth.^19,20 Given most breast cancers are positive for ERα and that in these tumors growth depends upon active ERα-induced mitogenic signaling,⁴⁸ the field developed a criterion for classifying breast cancer as ERα-positive hormone-sensitive vs. ERα-negative hormone-insensitive.⁵⁹

Triple-negative breast cancer is a subset of ERα-negative tumors that lacks ERα, progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2). The challenge with patients containing TNBC tumors is that they are notoriously difficult to treat. This is because TNBC is a particularly aggressive disease that presents earlier in life, and of course is hormone-insensitive.^9,16 Therefore, developing novel strategies to treat this patient group remains a challenge. Recent studies have highlighted that TNBC in fact expresses other steroid hormone receptors and raises the notion that TNBC much like other breast cancer are in fact hormone responsive,^71,63 and if the proper hormone-based treatment can be developed, more effective therapeutics outside of chemo and radiation based therapy can be developed to treat these patients.

Vitamin-D Receptor (VDR) is expressed in ∼80% to 90% of breast cancer; however, its role in breast cancer development is unclear.³⁹ This is evidenced by the finding that calcitriol and other agents such as Seocalcitol (EB 1089) used as a single agent in breast cancer patients are ineffective in reducing tumor burden,^50,28 without undue side effects such as extreme hypercalcemia and the development of renal stones.¹⁸ This may in fact speak to notion that calcitriol is still a weak vitamin D agonist, and that development of stronger therapeutics to regulate the expression of VDR will serve as more effective strategy in the future. VDR is a member of the nuclear class II receptor family,⁷⁹ which includes androgen receptor and estrogen receptor. The ligand for VDR is 1,25-dihydroxyvatimin D3, and recent studies have highlighted the role of this metabolite in breast cancer cell growth and differentiation in malignant breast tissue. Specifically, VDR promotes anti-proliferation effects in breast cancer, which is counterintuitive given the dogma for most breast cancer is that hormones such as estrogen promote the mitogenic activity in breast cancer. This is an interesting observation given the expression of VDR is elevated in breast cancer samples as compared with normal tissues. Yet, higher VDR expression is correlated to better prognosis in breast cancer patients.^23,24,8 Therefore, the expression of VDR versus the activity of VDR is an important factor in elucidating how to target this receptor for treatment of breast cancer. In this review, we highlight the role of noncoding RNAs, specifically, miRNAs, and how they may serve as potential therapeutic tools in TNBC from the perspective of targeting or regulating the expression of VDR.

miRNA biogenesis and function

miRNAs are approximately 19–22 nucleotide (nt) long RNAs that confer regulatory potential in cellular development and chronic disease by controlling gene expression at the post-transcriptional level.^80,6 miRNAs are transcribed as primary miRNAs in the nucleus of the cell, and are enzymatically processed in the nucleus by a family of RNAseIII enzymes including DROSHA and AGO2.^78,1,2 The residual 85-nt hairpin structures termed pre-miRNAs are exported into the cytoplasm. These pre-miRNAs are transported into the cytoplasm due to the interaction between Ran (a DTP-dependent protein) that binds with Exportin 5. This binding interaction induces an Exportin 5 conformational change in which the pre-miRNA can diffuse across the nuclear membrane.^21,22 In the cytoplasm, the pre-miRNA is processed by DICER1 into a 19–22 nt dsRNA duplex RNA that is incorporated into the AGO2/RISC complex, where strand-selection occurs such that a mature ssRNA guide strand is selected.^30,11 This mature miRNA then directs the RISC complex to mRNA targets by binding primarily to the 3′ untranslated region (UTR) of the respective gene. RISC complex binding to the 3′ UTR of messenger RNA (mRNA) induces one of two mutually exclusive processes, (1) cleavage of the mRNA to prevent translation into a protein or (2) stall the ribosomal machinery from completing translation of the mRNA so that it prevents the protein product of a specific gene from being generated.^7,36 The first case is evidenced by the finding that miRNAs alter the expression of a gene product at the RNA level, as verified by qPCR or array technologies, while the latter is evidenced by the finding that mRNA levels remain unchanged but the protein expression of the gene is reduced upon inducing miRNA-expression.^51,82 In fact, it was this latter observation that helped in forming the notion that small RNAs harbor regulatory function.

In recent studies, correlations have been observed between miRNA expression and VDR in breast cancer.²⁶ This has led to speculation that development of breast cancer may be the result miRNAs that control VDR expression and activity. VDR is a nuclear hormone receptor that regulates the pathways involved in cancer and immune response. This receptor functions as a transcription factor by binding directly to gene promoter regions. Studies involving agonists of vitamin D have illustrated that interrupting the VDR pathway can affect the growth of specific cell lines.¹⁷ Efforts to understand how VDR controls genetic networks that support tumorigenesis are of major interest. Moreover, given that the ligand of VDR is a metabolite, 25-hydroxyvitamin D (25[OH]D), this implies that metabolism, and subsequently the regulators of metabolic pathways, play a crucial role during tumorigenesis.^41,5 Therefore, understanding the regulatory role miRNAs impose upon the VDR pathway allows one to sufficiently identify targets of metabolic-related molecules for development in the clinic as biomarkers and therapeutics for cancer treatment.

The biology of VDR

VDR is a steroid hormone receptor that functions as a transcription factor and regulates numerous physiologic processes such as serum calcium levels, cell proliferation and differentiation, and immunomodulatory functions.⁶⁶ When VDR activity becomes dysregulated, several chronic disorders result, including diabetes, cardiac health, and cancer. This brings about an important question as to whether proper diet and vitamin intake alters one's risk for developing these chronic conditions. For instance, in high latitude areas, where sunlight is low, a greater percentage of the population has vitamin deficiency.⁵⁴ These insufficiencies in vitamin D are associated with enhanced rates of cardiovascular disease and cancer, as well as rickets and osteomalacia.²⁷

Understanding the structure of VDR, and the promoter sequences that VDR binds to in the genome, will help in understanding the function of VDR in certain genomic contexts. As with most steroid receptor hormones, VDR is a ligand-dependent transcription factor, and binds a dual hexameric motif that harbors a 3 base pair (bp) DR3 spacer motif.^55,65,64 This motif has been confirmed through Chip-Seq and biochemical assays, and VDR binds to these sites both in the presence or absence of liganded and unliganded VDR. Tuoresmaki and colleagues reported additional VDR binding motifs through genome-wide profiling, and identified distinct VDR binding patterns.⁷⁵ VDR harbors a modular protein structure comprising of A–E regions, and the proper folding of these domains is critical to the ligand-dependent function of VDR.^49,68 In general, 25[OH]D binds to VDR and induces a change in the structure of VDR such that VDR either activates or suppresses gene activity depending upon the associated co-regulatory proteins that VDR binds to. 25[OH]D is produced by the enzymatic processing of vitamin D from the diet in the liver by two major hydroxylases encoded by CYP27A1 and CYP2R1 to generate calcidiol.^76,12 Calcidiol can also be produced by from the conversion of 7-Dehydrocholesterol in the skin from UVB radiation to pre-vitamin B3. After spontaneous isomerization, pre-vitamin D3 converts to cholecalciferol and serves as a substrate for hepatocytic CYP2R1 and CYP27A1. 1,25(OH)₂D₃ is then produced in a tissue-specific manner by the CYP27B1 processing of calcidiol to calcitriol. 1,25(OH)₂D₃ and other natural vitamin D analogs promote the majority of regulatory functions associated with VDR. Liganded VDR predominately suppresses proliferation and promotes differentiation,⁵³ not by inhibiting canonical AP-1 activity, but rather due to promoter selectivity.

The role of unliganded VDR is less well understood; however, as with hormone receptors such as thyroid hormone (TH), unliganded VDR can still bind to regions in the genome and exert regulatory functions on those subsequent downstream gene targets,^31,73 As an example, unliganded VDR binds to the prolactin promoter and functions as a constitute activator. Prolactin is an important hormone for milk production, and is also a known estrogen-responsive gene. The VDR-mediated regulation of prolactin may be a requirement to sustain prolactin signaling under certain physiologic states, especially during time of low 17β-estradiol activity. Unliganded VDR also heterodimerizes with RXR, promoting nuclear retention of VDR and activation of a distinct gene network mutually exclusive of liganded VDR.^61,60 Interestingly, in breast cancer, 25-hydroxyvitamin D3–24-hydroxylase (CYP24A1) expression is inversely correlated with VDR protein levels. This is interesting since CYP24A1 converts calcidiol to the inactive vitamin D metabolite 24,25(OH)₂D₃, subsequently dampening the activity of vitamin D signaling.^34,70,4 Furthermore, the unliganded form of VDR binds to CYP24A1 repressing its expression and activity, in turn promoting an active vitamin D response by generating more of the active vitamin D metabolite 25[OH]D.

The role of VDR in breast cancer

In animal models, VDR signaling is observed to be a protective signaling pathway. Susceptibility to hyperplasia within the mammary gland is enhanced in VDR knockout mice. The anti-proliferative effect of VDR is so robust in the mammary gland that normal gland proliferation is reduced by high vitamin D intake.^40,33 VDR agonists including calcitriol have been tested in vitro and has been shown to reduce cell viability in a panel of breast and prostate cancer cell lines.⁷¹ VDR interacts with several genes and transcription factors such as SP1, HNF4a, FOXO4, and STAT3. Additionally, VDR interacts with NF-κB and cross talks with other immune-related pathways.¹³ This relationship is antagonistic in nature and has predominately been reported to occur in the colonic crypt cells as a mechanism to dampen immune-regulated responses. Therefore, VDR and the metabolic pathways that control VDR expression are linked to immunity and immune-regulated functions. Studying the role of VDR in breast cancer etiology is especially interesting, given vitamin D insufficiency is more prevalent in women, which has in part been associated with reduced time spent outdoors.^27,77 However, it is clear that there is cross-talk between VDR signaling and other steroid hormone receptor growth pathways such as estradiol signaling and epidermal growth factor signaling, both of which are important to the etiology and development of breast cancer. A recent study by Santagata and colleagues elegantly highlighted that most breast cancers express VDR, and more astoundingly, VDR is expressed in hormone-independent tumors such as TNBC.⁶³ In fact, based on immunohistochemistry assays on tissue microarrays, ER, PR, and VDR were expressed in a distinct bi-modal fashion in luminal epithelium. This expression pattern was distinct and separate from other hormone receptors, such as TRH, PTH1R, and RAR/RXR. Furthermore, VDR was commonly found to be co-expressed with AR or ER, indicating that crosstalk between these steroid hormone receptor pathways is common in mammary epithelial differentiation and breast tumorigenesis.

In some cases, disruption of the expression of the VDR protein product is a result of polymorphisms that interrupt the transcriptional process of co-regulatory genes. One is a polymorphism in Cdx2 which causes the promoter region of VDR to become disrupted, altering the transcription potential of VDR. Researchers analyzed the effects of this polymorphism on several cell lines, either ERα-negative or ERα-positive.⁶² In this analysis, investigators found that there was no correlation between Cdx2 status, VDR protein levels or VDR mRNA, when it concerned the ERα-positive breast cancer cell lines. In ERα-negative cell lines, however, there was a correlation between greater amounts of VDR protein expression and Cdx2 polymorphisms. Treatment of cell lines positive for the Cdx mutation with vitamin D altered the expression of downstream target genes, such as CYP24A1.⁶² However, in general, cell lines with Cdx2 polymorphisms were more sensitive to vitamin D treatment and had the greatest increase in upregulation of VDR. As an example, when treated with vitamin D, ERα-positive MCF-7 cells harbored a small inhibition of cell growth, while treatment of ERα-negative MDA-MB-468 cells harbored the greatest inhibition of cell growth. Overall this suggests that performing an analysis on breast cancer can provide information on any Cdx2 polymorphisms in VDR, and thus provide a marker for cancers more susceptible to vitamin D treatment.⁶²

Vitamin D3 signaling also regulates the activity and robustness of the innate immune system, which is best elucidated in macrophages. VDR signaling induces the anti-microbial peptide cathelicidin in macrophages that results in the killing of pathogenic bacteria.^38,35 The anti-inflammatory action of vitamin D is observed in several other tissues. This is relevant to breast cancer development because vitamin D3 regulates immune-regulatory pathways such as β-catenin. Specifically, Johnson et al. indicated that the RON receptor tyrosine kinase (MST1R) is overexpressed in breast cancer and associates with metastasis.⁴² VDR blocks the ability of RON to induce mammary tumors by decreasing active β-catenin levels. The constitutive expression of β-catenin in this model system overrides the growth suppressive effects of vitamin D3 signaling. VDR also regulates canonical Wnt signaling pathways that in turn controls immune responses in both colon and breast cancer models.³ The notion therefore is that developing vitamin D3 analogs that function in combination with RTK inhibitors such as Ron antagonists would be a novel way to treat breast cancer patients.

miRNAs in breast cancer

miRNAs are widely disrupted in cancer and regulate several tumorigenic processes including cell proliferation, apoptosis, and migration.^45,46 Numerous miRNAs function as oncogenes and tumor suppressors, and are currently under development for therapeutic gain. Common miRNAs studies over the past decade include miR-21, miR-155, and miR-181. These are broadly oncogenic in nature across several tumor types. In fact, given miR-155 regulates several immune regulated pathways, it is not surprising that miR-155 regulates VDR expression.^56,14 Common miRNAs such as miR-34a, let-7, and miR-206 are tumor-suppressive in nature, and these noncoding RNAs have been studied in breast cancer development extensively as well. Though some non-canonical miRNAs such as miR-627⁵⁷ and miR-98 and been shown to interact with VDR,^72,47 a question that remains, however, is how these miRNAs and others control VDR expression, and furthermore how this regulatory axis contributes to breast cancer etiology.

A recent study also highlighted the notion that noncoding RNAs such as H19 can interact with VDR, and this occurs through miR-675-5p. miR-675-5p in fact targets the VDR 3′ UTR at one site and reduces VDR protein expression.¹⁰ Interestingly, the overexpression of H19 also reduces VDR expression, in part by promoting the transcription of miR-675-5p. The mechanism of action for this action is that miR-675-5p sits in the first exon of H19, and from a gene-centric point of view H19 promotes its own expression since VDR negatively regulates this gene product.⁷⁴ By virtue of this H19-gene regulatory axis, miR-675-5p is co-transcribed with H19, and H19 uses miR-675-5p to abrogate VDR-induced silencing of H19. Additional miRNA interactions involved in breast cancer include miR-627 and JMJD1A, and miR-125b and NCOR2/SMRT.^57,74

In the miRNA field, bioinformatic tools such as TargetScan, a software program that provides prediction of the targets of miRNAs, have immensely helped in advancing the field.²⁵ We wanted to determine which miRNAs were putative candidates for VDR interaction. We used this and other bioinformatic programs to identify conserved sites of miRNAs across several organisms and ranked these miRNA candidates by context score, which rates the relevance of the miRNA to the desired gene and its interaction. Then, using the context score we selected three miRNAs (miR-23, miR-124, and miR-125) for further investigation, which may be of interest to the broader research community. We also focused on miR-125b, given there is a well-established regulatory potential between miR-125b and VDR.⁶⁷ While some of these miRNAs may have been previously assessed in correlation studies, they may not have been directly studied. We believe these miRNAs may serve as useful targets for therapy in breast cancer patients, and could in fact be used in combination with vitamin D3-based therapies. We then used another program, the USCS Genome Browser,³⁷ to interactively visualize the genomic data associated with the VDR gene. From this program, we determined the tissue-specific expression of VDR transcriptional activity, which helped us further infer which miRNAs could regulate VDR in breast cancer.

Epigenetic control of VDR expression

Using TargetScan as our first tool, we surveyed the context scores of miRNA sites; some preliminary research had been done, indicating a natural leaning toward the miRNAs previously entwined with VDR research. Of the originally chosen five miRNAs, we decided to further look at miR-23, miR-124, and miR-125. They have VDR-interaction context score of −0.16, −0.21, and −0.44, respectively. It should be noted that both miR-125-5p and miR-125-3p had a context score ranging from −0.44 to 0.12. We next decided that these were sufficient miRNAs to consider. We next investigated the roles these miRNAs play in either breast cancer or VDR. Of the three, miR-23 had the least research done in its relation to these topics. In the 2015, Huang and colleagues assessed 17β-estradiol-induced apoptosis via the activation of miR-23 and p53.³² Researchers introduced 17β-estradiol into SNU-387 cells, which is a liver cancer cell line. The upregulation of several miRNAs was seen and only a few were selected for further study, among them miR-23. The expression of miR-23 was assessed across six liver cancer cell lines, as determined by the UCSC genome browser. There was a noticeable difference of miR-23 expression across the six cell lines, which was originally attributed to the status of male or female cell lines. Upon closer inspection, it was determined that these cell lines differed in the status of the p53 gene, either as wild-type, mutant, or null. Specifically, cell lines L02 and HepG2 are p53 wild-type, and displayed a much greater expression of miR-23, while cell lines that were p53 mutated or p53 null (Hep3B, Huh7, SNU-387, and HepG2.2.15) displayed minimal miR-23 expression. To test the correlation between these concepts, researchers induced wild type p53 expression through a doxorubicin-inducible system, and the resulting miR-23 expression was observed.³² One question remaining was whether 17β-estradiol was responsible for activating p53. Indeed, in SNU-387 cells, researchers found that 10⁻⁸ M of 17β-estradiol resulted in the greatest increase of p53 expression. Furthermore, ERα mRNA expression was upregulated ∼3-fold, indicating that the regulation between the 17β-estradiol, p53 and miR-23 was mediated by ERα. The knockout of the receptor gave final evidence that ERa regulated p53 and miR-23, as these cells were unable to increase p53 or miR-23 expression upon 17b-estradiol stimulation. This miRNA has a strong correlation to oncogenic effects, but not yet to VDR regulation or correlation.

miR-124 was the next miRNA of interest. While miR-23 is easily associated with liver cancer cells, miR-124 was specific to breast cancer research. miR-124 was first confirmed to be at lower levels in breast cancer cell lines than normal breast cell lines.⁸¹ In the absence of miR-124, Beclin1, an autophagy-related protein producing gene, would have high expression levels. If miR-124 were increased, it would result in a decrease of Beclin1, indicating a negative regulation of Beclin1 by miR-124. Since breast cancer cell lines illustrated a decreased amount of miR-124 and therefore an increased amount of Beclin1, scientists investigated what would happen if there were an overexpression of miR-124 in those cell lines. They observed that the cell lines when treated with increase miR-124 would experience a reversal of autophagy of breast cancer cells.⁸¹ Beclin1 regulates autophagy in the cell and is helpful in maintaining a cell's homeostasis to prevent death. Thus, the increased levels of Beclin1 in breast cancer cells could be aiding the survival of those cancer cells. miR-124 and Beclin would be very interesting to look at in breast cancer cells to see if there is any interaction with the VDR pathways we are interested in.

miR-125 is the last miRNA of interest, and is the most heavily researched in relation to cancer and VDR. From previous research, miR-125 is known to target mRNAs ERBB2 and ERBB3, also known as HER2 and HER3.⁴⁴ These were mentioned earlier to play a role in TNBC. In fact, miR-125 can induce metastasis of breast cancer cells by targeting STARD13, but also functions as a tumor suppressor by targeting ERB2/ERBB3. The overexpression of miR-125 resulted in the suppression of both ERBB2 and ERBB3 expression, while the these genes actively suppress miR-125b levels. The overexpression of miR-125 was also seen to be correlated with abrogated cellular migration and invasion of cancerous cells. The researchers dive into the oncogenic and regulatory characteristics of miR-125 by looking at a germline mutation in the mature miRNA. The mutation was in fact a single nucleotide polymorphism (SNP) in which affected patients carried a T/A allelic alteration. The individuals associated with this mutation were more likely to develop breast cancer than those without it since loss of function of miR-125 could lead to upregulation or loss of regulation of its targets associated with the receptors commonly seen in breast cancer.⁴⁴ In another study, Mohri et al. identified several miRNAs regulating the vitamin D receptor including miR-125b. In this particular study, luciferase studies using a reporter with a miR-125b binding site in the VDR 3′ UTR indicated that miR-125 could bind and downregulate luciferase activity.⁵² This was done in both KGN and MCF-7 cells, indicating that there is active VDR signing in ERα-positive MCF-7 cells. Indeed, 1,25(OH)2D3 treatment of MCF-7 cells induced anti-proliferative effects in these cells, and moreover the overexpression of miR-125b abrogated this effect. Overall these studies suggest a link between VDR and miR-125 in breast cancer cell lines.

Discussion

Overall, we investigated the role of VDR in breast cancer, and furthermore investigated the correlation between miRNAs and VDR in breast cancer. TargetScan has allowed us to handpick miRNAs that could play a regulatory role in both VDR and breast cancer. We hope to continue this research with further experimentation. We would like to select cell lines of breast cancer and analyze their inhibition and growth with and without vitamin D treatment. Additionally, we would like to assess how miRNA expression and activity is altered in cell lines harboring various Cdx2 polymorphisms. These studies will answer an important question as to whether certain Cdx2 polymorphisms allow for greater susceptibility to particular vitamin D agonists or antagonists. Furthermore, we could analyze the miRNAs of interest are even seen in abundance in these cell lines by performing qPCR. From there we could perform more experiments that would analyze the effects on VDR by knocking out specific miRNAs or overexpressing them within particular cancer cell lines. A greater understanding of these regulatory molecules will aid in a better attempt at treatment and cure for the cancers that are notoriously difficult to treat. Overall, in this review, we highlighted the role of noncoding as it pertains to the regulation of VDR, and the implications of VDR in breast cancer development.

Disclosure of Potential Conflicts of Interest

B.D.A. holds patent interests with and consults with AUM LifeTech. The other author has no conflicts of interest to disclose.

Funding

This work was supported by grants to B.D.A. by the Research Foundation, and through start-up funds by the RNA Institute and the State of New York.