Abstract
Background/Aim: Triple negative cancer (TNBC) is a subtype of breast cancer that is highly aggressive, with poor prognosis and responds differently to treatments. This study investigated the role of vorinostat and indole-3-carbinol (I3C) on regulating critical receptors that are not normally expressed in TNBC. Materials and Methods: Using real-time PCR, immunostaining, and western blots, the re-expression of estrogen receptor α (ER), progesterone receptor (PR) and human epidermal growth factor receptor-2 (HER2) receptors was examined in four different TNBC cell types. Results: ERα was re-expressed in three subtypes using vorinostat and I3C. Re-expression of the PR by vorinostat was also detected. Neither vorinostat nor I3C resulted in re-expression of the HER2 receptor. A significant decrease in growth and sensitivity to tamoxifen was also noted. Conclusion: The results of this study show that vorinostat and I3C modulate the re-expression of critical receptors in certain subtypes of TNBC through several pathways and these effects can be influenced by the molecular profiles of TNBCs.
Triple negative breast cancer (TNBC) is one of the most aggressive subtypes of breast cancer (1). Although about 85% of breast cancers are estrogen-positive, about 15-20% fall into the category of TNBC. This subtype of cancer lacks targeted therapy receptors, such as the estrogen receptor (ER), the progesterone receptor (PR), and the human epidermal growth factor receptor-2 (HER2) (2). The estrogen receptor status is important, both as an independent indicator of prognosis and as a basis for selecting a treatment. Specifically, patients whose tumors are characterized as ER-positive are candidates for anti-estrogen therapy, such as tamoxifen (or similar agents), while patients with ER-negative tumors are often given cytotoxic chemotherapy as a standard of care (2). In addition to estrogen's role in breast cancer, a recent study (3) examined the role of progesterone in TNBC and recurring TNBC tumors that often have a tendency to develop resistance to current therapeutics. TNBC patients are often limited to cytotoxic chemotherapies with harsh side effects. In addition, this type of cancer has the tendency to occur in younger women and is associated with risk factors, including: i) being of African descent (4), ii) BRCA1 mutation (5-7), iii) a strong family history of breast cancer (8), iv) lifestyle, and v) environmental factors (9-11). Additional risk factors involving life-style include: i) poor diet (12, 13), ii) high alcohol intake (14, 15), iii) smoking (16, 17), iv) lack of breastfeeding (18, 19), v) lack of exercise (20), and vi) obesity. Although mutations in BRCA 1 and 2 are known to be involved in the initiation of TNBC (21-25), research has revealed that these genes are regulated by epigenetic factors (26, 27). A number of environmental and lifestyle-related factors can contribute to the modulation of receptors' expression that are involved in signaling pathways through epigenetic mechanisms (28). The modification of histones is a known epigenetic mechanism involved in the regulation of gene expression. Acetylation is associated with gene expression while deacetylation is associated with gene repression (29).
In cancer certain epigenetic modifications may result in reversible changes in gene expression of critical signaling pathways. Targeting the mechanisms that control such epigenetic modifications could potentially be used in cancer therapy (29, 30). Considerable interest has developed with regards to understanding the role of histone deacetylases (HDACs) in cancer. HDACs are part of a superfamily that is divided into different classes based on their mode of action (31), such as i) cancer progression (32), ii) self-renewal and expansion of stem cells (33), and iii) epithelial to mesenchymal transition (EMT) (34). Although studies have also shown that inactivation of certain HDACs may be involved in tumorigeneses (35), increased HDAC activity has been noted in a number of cancers (36, 37), including TNBC (38). HDACs are involved in a number of cellular events in cancer, such as i) tissue differentiation (39), ii) autophagy (40), iii) apoptosis (41), iv) migration (42), and v) mitosis (39). These enzymes deacetylate histone and non-histone proteins (41) and are involved in the regulation of the cell cycle (42), DNA damage response and autophagy (43), metastasis, and angiogenesis (44).
Epigenetic drugs, such as histone deacetylases inhibitors (HDACis), are emerging as promising therapies for various cancers, including TNBC. Because of their ability to target different pathways, HDACis are known as multifunctional agents. Inhibition of certain HDACs has been associated with the induction of apoptosis and cell cycle arrest (45). Recently, studies showed that HDACis, such as vorinostat, modulate critical HDACs involved in cancer stem cell progression, such as increases in HDAC7, which is also associated with poor prognosis (46). Studies have also shown the ability of HDACi to re-express critical genes producing tumor suppressors (47) and critical receptors, such as ERα in TNBC (48).
Progesterone (P) is also known to suppress TNBC cells (49). Progesterone plays an important role in mammary epithelial cell proliferation by binding to its receptor (PR), followed by receptor dimerization, nuclear localization, and binding to progesterone-responsive elements in target genes (50). High levels of progesterone are considered a risk for breast cancer under certain conditions (51), with a yet unidentified role in TNBC. However, studies have shown that progesterone-α receptors are expressed in TNBC and that progesterone suppresses growth and invasion of tumors or TNBC cells expressing high levels of progesterone-α receptors (52).
The lack of expression of HER-2 is another feature of TNBC (53). Excessive expression of HER-2 stimulates cell growth (54). Drugs targeting the HER-2 protein, such as Herceptin (trastuzumab), are given to patients with HER-2 positive breast cancer (55), but these do not have any effect on TNBC patients. To date, no drug is known to re-express HER-2 in TNBC.
Improvement on the treatment of women with TNBC is greatly needed. These cancers are very invasive and the associated mortality rate is extremely high, affecting also women under 40, during their reproductive years (1). Understanding whether ER, PR, or HER2 receptors can be reactivated in specific subtypes will greatly enhance our understanding of epigenetic regulation in these highly deadly cancers and can, thus, expand the treatment potential of the currently approved targeted therapies for these subtypes of TNBC. This study investigated the role of epigenetic mechanisms controlling the re-expression of major drug targeting receptors (ER, PR, and HER-2) in TNBC subtypes. The effect of treatments using vorinostat as an epigenetic drug and indole-3-carbinol as a dietary agent were assessed alone or in combination in cells representing different TNBC subtypes to determine if re-expressing critical receptors could increase the cells' sensitivity to targeted therapies.
Materials and Methods
Chemicals, drugs and antibodies. Vorinostat, 4-hydroxytamoxifen, and indole-3-carbinol (I3C) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Estrogen (17-β estradiol) was purchased from Sigma-Aldrich. MTS kits (Cell Titer 96 Cell Proliferation Assay) were purchased from Promega (Madison, WI, USA). BioCoat Invasion chambers were purchased from Thermo Fisher Scientific (Waltham, MA, USA).
Cell lines and treatment. MDA-MB-231, HCC70, HCC1806, BT549, HCC1143, and MCF7 cells were purchased from American Type Culture Collection (Manassas, VA, USA). IC50s were determined for vorinostat in the triple-negative cell lines MDA-MB-231 (ATCC® HTB-26™) (M: Mesenchymal stem cell-like), BT-549 (ATCC® HTB-122™) (M: mesenchymal-like), HCC1806 (ATCC® CRL-2335™) and HCC70 (ATCC® CRL-2315™) are both basal-like (BL2), HCC 1143 (ATCC® CRL-2321™), and the breast cancer cell line MCF7 (ATCC® HTB-22™). BT549, MB-231, MCF10A (ATCC® CRL-1037™) and MCF7 cells were grown in GIBCO MEM media (Thermo Fisher Scientific) supplemented with fetal bovine serum, L-glutamine, and penicillin-streptomycin. HCC70 cells were grown in GIBCO RPMI media (Thermo Fisher Scientific) supplemented with fetal bovine serum (Atlanta Biologicals-Premium select, Minneapolis, MN, USA), HEPES buffer, pen strep, sodium pyruvate, and L-glutamine (all purchased from Life Technologies, Carlsbad, CA, USA). HCC1806 and HCC1143 cells were grown in RPMI media supplemented with pen strep and L-glutamine. For the treatment, all cell lines were grown in phenol-free, charcoal-stripped serum (Atlanta Biological-Premium select, Minneapolis, MN, USA) for 3-5 days prior to treatment with 17-β-estradiol (100 nM), vorinostat (both from Sigma-Aldrich, St. Louis, MO, USA) (10 μM) (doses were determined from a dose range study and IC50), or I3C (300 μM). For all studies discussed in this manuscript, cells were exposed to treatment agents for 24 h. Whole cell lysates, DNA and total RNA were collected after 6, 12, 24, and 48 h of treatment.
mRNA. Total RNA was extracted using the Qiagen RNeasy Micro kit (QIAGEN, Germantown, MD, USA) and cDNA was synthesized using SuperScript III (Bio-Rad Laboratories, Hercules, CA, USA). The expression levels of mRNA were measured using TaqMan® Gene Expression arrays for ERα, PR, HER2 and β-actin. Real Time PCR was run with the QuantStudio system from Thermo Fisher Scientific. Results were analyzed using Applied Biosystems qPCR Analysis Module within the Thermo Fisher Cloud.
Immunostaining (ER and PR). All cells were grown in 24-well plates (Thermo Fisher Scientific). Cells were seeded and treated with the drugs, vorinostat, 17-β-estradiol or I3C. Following treatment, the cells were washed with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde for 10 min at room temperature. Then the cells were permeabilized with 0.2% Triton X-100TM Surfact-AmpsTM Detergent Solution (Thermo Fisher Scientific; Catalog number: 85111) for 10 min at room temperature, followed by washing with ice-cold PBS. To block unspecific antigen epitopes, the cells were blocked with cold fish serum (Thermo Fisher Scientific; Catalog number: 37527) for 1 h at room temperature. Subsequently, cells were incubated with a primary antibody [monoclonal anti-ER alpha (Abcam, Cambridge, MA, USA)] in 1% fish serum overnight on a shaker at 4 °C. The cells were washed 3 times with PBS-T20 (Pierce 20X PBS Tween-20, diluted to 1X with water which contains 10 mM sodium phosphate, 0.15M NaCl, 0.05% Tween™ 20, pH 7.5, (Thermo Fisher Scientific; Catalog number 28352) 10 min each time and incubated with a secondary antibody [goat polyclonal secondary antibody to Rabbit IgG-H&L (Alexa Fluor® 555), ab150078 (Abcam)] in 1% cold fish serum in the dark for 30 min at 37°C on a shaker. Cells were washed three times with PBS-T20 for 10 min each time and then the nuclei were stained using DAPI [NucBlue™ Live ReadyProbes™ Reagent, Catalog number: R37605, (Thermo Fisher Scientific)] for 5 min before taking images.
MTS assay. Cancer cells from each subtype were plated in 96-well plates (10,000 cells per well) and were treated with vorinostat (10 μM) for 1, 3, 5, and 7 days. Procedures were followed as outlined in the MTS protocol (Promega, Madison, WI, USA).
Cell migration and invasion assay. To determine the invasive potential of the TNBC cell lines before and after each treatment, cells were evaluated using the BD Biocoat™ Matrigel™ Invasion Chambers (BD Biosciences, San Jose, CA, USA; Catalog Number 354480). Each cell culture invasion chamber contains an 8-micron pore size PET membrane treated with Matrigel Matrix. An appropriate number of TNBC cells (10,000) per invasion chamber was seeded in the presence or absence of each treatment. The medium in the upper chamber was serum-free, while the medium in the lower chamber contained 5% FBS as a chemoattractant. Non-invasive cells were removed with a cotton swab and cells that migrated through the Matrigel Matrix were stained on the lower surface of the membrane with toluidine stain (Thermo Fisher Scientific) and scored for% invasion or inhibition of invasion.
Protein. Whole cell protein lysates were prepared using RIPA lysis buffer (Thermo Fisher Scientific). Total protein (40 μg/lane) was separated using Bolt 4-12% Bis-Tris precast gels (Life Technologies) in Bolt MES and were transferred onto PDF membranes using the iBlot Dry Blotting system (Thermo Fisher Scientific). Western blot analysis was performed using the iBind Western Blot System (Thermo Fisher Scientific) with two primary antibodies: i) rabbit monoclonal anti-ER alpha antibody, and ii) rabbit polyclonal anti-α-tubulin (ab52866), 1:6000, and a secondary anti-rabbit HRP-conjugated antibody (both from Abcam). Protein levels were detected using Supersignal West Femto (Thermo Fisher Scientific) and the ChemiDoc Touch Imaging System (Bio-Rad). Images were analyzed using the Image Lab 6.0.1 software (Bio-Rad).
Nuclear extraction and HDAC total activity. EpiQuik™ nuclear extraction kits (Epigentek, Farmingdale, NY, USA) were used on the treated cells to assess the HDAC activity. Procedures were followed according to the manufacturers' instructions. The Epigenase™ HDAC Activity/Inhibition Direct assay kit (Epigentek) was used for the measurement of the HDAC activity/inhibition. Briefly, in this assay an acetylated histone HDAC substrate is stably coated onto microplate wells. Active HDACs bind to the substrate and remove acetyl groups from it. The HDAC-deacetylated products can be recognized using a high affinity acetylated HDAC histone capture-antibody provided in the kit. The ratio or amount of deacetylated products, which is proportional to the enzyme activity, can then be colorimetrically measured using a Cytation 3 cell imaging reader (BioTek Instruments, Winooski, VT, USA) and reading at 450 nm. The activity of the HDAC enzyme is proportional to the OD intensity measured.
Statistical analyses. One-way ANOVA analyses were performed using the Prism software Version 6.0 (GraphPad, San Diego, CA, USA). To compare values, p-Values of <0.05 were considered significant. Graphs represent the mean±SD.
Results
Re-expression of ERα and PR receptors (RNA and immunostaining). We investigated the effects of the small-molecule vorinostat (V), also known as SAHA (suberoylanilide hydroxamic acid), on the re-expression of ERα and PR receptors in TNBC cells. Vorinostat induced the re-expression of ERα and PR receptors in TNBC cells with a basal-like 2 phenotype, as well as in HCC70 and HCC1806 (African American cell lines), when treated for 24 h (Figures 1A and B). Treatment with the dietary agent, I3C, also caused the re-expression of ERα in HCC70 cells and PR in HCC1806 cells.
Figure 2 demonstrates the effect of treatment with vorinostat and I3C on two other subtypes: i) HCC1143 (basal-like 1) (Figure 2A), and ii) BT549 (M) (Figure 2B). Vorinostat increased the expression of PR in both HCC1143 and BT549 cells. Vorinostat also increased the expression of ERα in BT549 cells but not in HCC1143 cells. The dietary agent, I3C, increased the expression of ERα significantly in both cell subtypes. I3C had no effect on PR expression in BT549 cells but it did increase PR in HCC1143 cells.
Immunostaining confirmed the re-expression of ERα in HCC70 cells (Figure 3). The control cells showed no nuclear ERα staining (Figure 3A) compared to the vorinostat-treated cells (Figure 3B), I3C-treated cells (Figure 3C), or the combination of vorinostat-IC3-treated cells (Figure 3D). Immunostaining confirmed the nuclear re-expression of ERα in HCC1806 TNBC cells (Figure 4).
Another subtype of TNBC, the MDA-MB231 cells, showed re-expression of the ERα after 6 h as both RNA (5-fold, p≤0.0001) and protein (2-fold) (Figure 5A, left and right panel, respectively). The sensitivity to tamoxifen treatment was also tested in this subtype, where, in combination with vorinostat, cell growth was further inhibited (p≤0.0001) and all other treatments significantly inhibited growth compared to the control (Figure 5B). Vorinostat's effect on this cell line occurred sooner compared to the other subtypes.
Sensitivity to Tamoxifen in Basal-2 subtypes of TNBC. The lack of effective treatment using targeted therapies is one of TNBC challenges. However, here, cells of the most aggressive TNBC BL2 subtype responded to tamoxifen when treated with vorinostat (Figure 6). Both of these subtypes, HCC70and HCC1806 are from two different African American women, as identified by their unique ATCC number. One exciting finding was that tamoxifen alone significantly inhibited the growth of these two basal-2 TNBC subtype cells. This was not noted in MB-231 cells, which is a different TNBC subtype, as shown in Figure 5B. Furthermore, vorinostat and I3C alone and in combination decreased growth in these two TNBC cell lines.
Invasion assay. TNBC is not only aggressive but also highly invasive. Figure 7 shows a representative figure of the most aggressive type of the two basal-2 cell lines in our study, HCC70. The invasiveness of this line was inhibited by vorinostat and I3C. Similar results were seen for all other cell lines. We also noted that treatment with progesterone significantly increased the invasiveness in this cell line; however, combined treatment with vorinostat and progesterone decreased the invasive capacity of HCC70 cells.
HDAC total activity. We observed inhibition of HDAC activity in two subtypes of triple negative breast cancer cell lines: i) in the mesenchymal BT549 tamoxifen treatment (Figure 8A, p≤0.013) and ii) the basal-like 2 HCC70 line (Figure 8C). Results are also shown for MCF7 (Figure 8D) and MCF10A (Figure 8E), which is a non-malignant cell line. Vorinostat, as well as I3C and tamoxifen inhibited the overall HDAC activity more in the HCC70 cell line (Figure 8C), even though these three treatments similarly affected HDAC activity in the MC7 cell line (Figure 8D), a non-triple negative cell line.
HDAC7 inhibition by vorinostat. Increased HDAC7 has been associated with increases in cancer stem cell proliferation (56), thus, the possibility that vorinostat and I3C may be acting as HDACis is highly significant since HDACs are considered to play a critical role in chemoresistance (57). To determine if vorinostat affected HDAC7 in the HCC70 cell line, we performed western blot and real-time PCR analysis. Our investigation revealed a significant decrease in HDAC7 at both RNA and protein levels. Specifically, western blots analysis in MDA-MB-231 and HCC70 TNBC cells following treatment with vorinostat showed that HDAC7 was significantly decreased in HCC70 compared to MDA-MB-231 cells (Figure 9A). Similarly, in Figure 9B real-time analysis showed decreased HDAC7 expression in HCC70 cells following treatment with vorinostat.
Discussion
TNBC is an aggressive cancer that is molecularly characterized to an extent (1, 58). Here we demonstrated that the TNBC cell subtypes not only bear different molecular signatures but also show different responses to treatment. The lack of therapies for TNBC is an unmet medical need. This is particularly true for African Americans and Latin women who have higher rates of TNBC compared to other women (59). There is growing evidence that TNBC-related disparities may actually drive the aggressive biology of this type of cancer, observed in African-American TNBC patients (4, 8). Other treatment barriers, such as screening, stage at diagnosis, income and biological factors have been identified as contributing factors to the disparities of TNBC in minority populations (6-13). In most cases, standard chemotherapy is usually the systemic treatment for advanced TNBC however, emerging treatments are currently being investigated, including epigenetic therapies. Epigenetic “signatures” have been shown in patients' molecular profiles with TNBC and the possibility of re-expressing specific genes through epigenetic drugs that could be then targeted for therapy could be very promising (11, 12). In addition to the reversal of epigenetic mechanisms by drugs, a number of studies have succeeded in modulating the expression of specific genes involved in cancer development through dietary agents (60, 61), some of which have shown anti-cancer effects in breast cancer (62, 63). Such an example involves cruciferous vegetables containing I3C, and these are being investigated in clinical trials (64-66). I3C can affect TNBC through estrogen-independent actions, including blocking cell cycle progression and cell growth (67), increasing apoptosis (68) and inhibiting metastasis (69). Here, we showed that I3C enhanced the anti-cancer effects of vorinostat. In the basal-2 TNBC cell line, HCC70, IC3 significantly decreased their HDAC activity. Combining I3C and vorinostat had an effect on several cellular pathways, suggesting promising solutions for treating TNBC.
Our study demonstrated differences with regards to the re-expression of critical receptors in subtypes of TNBC cell lines, providing evidence for the importance of molecular profiling of individual cancers (or an individual's cancer), especially for this type of breast cancer. TNBC, particularly the aggressive genotypes, cannot be subjected to targeted therapies due to the absence of expression of specific receptors (ER, PR and Her2). The typical available treatments are surgery, chemotherapy, and/or radiation, while in recent years several new therapeutic options are used in TNBC (70). These options include treatment with poly-(ADP-ribose)-polymerase (PARP) inhibitors (71), angiogenesis inhibitors (72), rapamycin (mTOR) inhibitors (73), and androgen inhibitors (74), however, the efficacy of these drugs remains a problem. Clinical trials are investigating targeted therapies as single agents or in combination with other drugs (75), such as epigenetic drugs (76). Some of these drugs have already been approved by FDA for lymphoma, multiple myeloma and colon cancer (77, 78).
Among the targeted therapies investigated in TNBC, HDACis appear very promising (31, 38). HDACis exert their effect through many mechanisms, such as by inhibition of proliferation and angiogenesis, as well as enhancement of T-cell-mediated tumor immunity (39-41, 79). Some studies have shown these agents' ability to inhibit the capacity of cancer stem cells for self-renewal, leading to a decrease in the relapse of tumors' aggressive behavior (32, 33).
TNBC lacks the expression of the critical receptors ER, PR and HER2, which are targets for therapies in other types of breast cancer. Our study demonstrated the ability of vorinostat to re-express some of these critical receptors in different subtypes of TNBC. Aggressive subtypes of TNBC were able to re-express ERα and PR, which sensitized cells to the anti-estrogen drug, tamoxifen. Only certain subtypes of TNBC responded to vorinostat when receptors were re-expressed. Our study confirms other studies (80) in which vorinostat re-expressed ERα in some but not all TNBC cell lines. Our study clearly demonstrated that different cell subtypes bear distinct molecular signatures that play a critical role in drug response. Precision medicine is extremely important, and molecular profiling of patients' tumors may prove to be one of the most effective ways for handling aggressive cancers, such as TNBC. One exciting finding here was the fact that tamoxifen alone significantly inhibited the growth of a certain cell subtype, particularly of the two African American lines of the basal 2 subtype. Recent findings have demonstrated the presence of ERβ in subtypes of TNBC (81), while this receptor has been found in a number of African American TNBCs (82, 83). By analogy, this could suggest that tamoxifen may be useful in such TNBCs expressing ERβ.
TNBC is not only an aggressive cancer, but a highly resistant one to most current therapies. Studies have shown that expression of HDAC7 is associated with cancer stem cell progression and a resistant phenotype (84). Vorinostat significantly inhibited HDAC7 expression at both RNA and protein levels. HDAC inhibitors, such as vorinostat, are currently in clinical trials on their own or in combination for a number of other cancers (78) and are proving very promising with only few side effects.
In conclusion, the HDAC inhibitor, vorinostat, has shown specificity for different subtypes of TNBC, which could potentially be used as an efficient treatment in patients. Although this manuscript focused largely on vorinostat, the use of the dietary component, I3C, showed that it has the ability to enhance the vorinostat anticancer effects in certain TNBC subtypes. More research is needed to further investigate the mechanism through this new emerging agent and its ability to tackle an aggressive cancer, such as TNBC.
Footnotes
This article is freely accessible online.
Disclaimer: The views presented in this manuscript do not necessarily reflect those of the US Food and Drug Administration.
Funding
This study was funded to Beverly Lyn-Cook through a grant from the FDA-Office of Women's Health.
- Received March 10, 2020.
- Revision received June 15, 2020.
- Accepted June 17, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved