Downregulation of histone H2A and H2B pathways is associated with anthracycline sensitivity in breast cancer

Abstract

Background: Drug resistance in breast cancer is the major obstacle to effective treatment with chemotherapy. While upregulation of multidrug resistance genes is an important component of drug resistance mechanisms in vitro, their clinical relevance remains to be determined. Therefore, identifying pathways that could be targeted in the clinic to eliminate anthracycline-resistant breast cancer remains a major challenge.

Methods: We generated paired native and epirubicin-resistant MDA-MB-231, MCF7, SKBR3 and ZR-75-1 epirubicin-resistant breast cancer cell lines to identify pathways contributing to anthracycline resistance. Native cell lines were exposed to increasing concentrations of epirubicin until resistant cells were generated. To identify mechanisms driving epirubicin resistance, we used a complementary approach including gene expression analyses to identify molecular pathways involved in resistance, and small-molecule inhibitors to reverse resistance. In addition, we tested its clinical relevance in a BR9601 adjuvant clinical trial.

Results: Characterisation of epirubicin-resistant cells revealed that they were cross-resistant to doxorubicin and SN-38 and had alterations in apoptosis and cell-cycle profiles. Gene expression analysis identified deregulation of histone H2A and H2B genes in all four cell lines. Histone deacetylase small-molecule inhibitors reversed resistance and were cytotoxic for epirubicin-resistant cell lines, confirming that histone pathways are associated with epirubicin resistance. Gene expression of a novel 18-gene histone pathway module analysis of the BR9601 adjuvant clinical trial revealed that patients with low expression of the 18-gene histone module benefited from anthracycline treatment more than those with high expression (hazard ratio 0.35, 95 % confidence interval 0.13-0.96, p = 0.042).

Conclusions: This study revealed a key pathway that contributes to anthracycline resistance and established model systems for investigating drug resistance in all four major breast cancer subtypes. As the histone modification can be targeted with small-molecule inhibitors, it represents a possible means of reversing clinical anthracycline resistance.

Trial registration: ClinicalTrials.gov identifier NCT00003012 . Registered on 1 November 1999.

Figures

Fig. 1

Characterisation of epirubicin-resistant cell lines. Native and resistant cells were exposed to drug concentrations ranging from 0.3 to 3000 nM. Cell viability was determined 72 h later by Cell Counting Kit-8 assay. a Percentage of live cells relative to dimethyl sulphoxide control was plotted against epirubicin concentration. Black = native cells, magenta = resistant cells. b Half-maximal inhibitory concentration (IC50) values in nanomolar concentrations ± standard deviation. Resistance factor is shown in parentheses and represents resistant IC50/native IC50

Fig. 2

Expression of conventional breast cancer biomarkers and selected multidrug resistance genes. Cell lysates were prepared in radioimmunoprecipitation assay buffer supplemented with cOmplete Mini protease inhibitor and PhosSTOP phosphatase inhibitor. Quantities (10–50 μg) of total protein were run on a 10 % gel (MDR1), 4–20 % precast gels (epidermal growth factor receptor [EGFR], oestrogen receptor [ER], progesterone receptor [PR], type II topoisomerase [TOPO IIα]) and Any kD Mini-PROTEAN TGX Precast Protein Gels (human epidermal growth factor receptor 2 [HER2], HER3), transferred onto polyvinylidene fluoride membrane and developed using chemiluminescence substrate. Nat native, Epi-R epirubicin-resistant, GAPDH glyceraldehyde-3-phosphate dehydrogenase

Fig. 3

Resistant cell lines overcome epirubicin-induced G2/M arrest. a–d Cells were synchronised by a double-thymidine block and treated with dimethyl sulphoxide or epirubicin at selection doses established for each resistant cell line: 25 nM epirubicin to MDA-MB-231, 30 nM epirubicin to MCF7, 15 nM epirubicin to SKBR3 and 15 nM epirubicin to ZR-75-1. Epirubicin concentration was increased to 100 nM for MCF7 and SKBR3 cells because G2/M block was not observed at the lower doses of epirubicin. Cells were collected at 48 h, stained with propidium iodide and analysed by flow cytometry. Debris was gated out. GAPDH glyceraldehyde-3-phosphate dehydrogenase

Fig. 4

Network-based analysis of epirubicin-resistant (EpiR) cell lines. a Venn diagram of genes with significant changes in expression in breast cancer cell lines. b Histone module identified in Functional Interaction network analysis. Coloured rings denote genes demonstrating consistent changes across all four cell lines. Red rings = upregulated genes, green rings = downregulated genes, diamonds = linker genes. c Quantitative real-time polymerase chain reactions performed on RNA isolated from native and epirubicin-resistant cell lines. Bar graphs indicate average quantitative means, while error bars represent standard error of the mean. p Values were calculated using unpaired t test. ns non-significant. d Immunoblotting of total H2A and H2B histone proteins in native and epirubicin-resistant cell lines. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a housekeeping control. e Reactome pathways significantly enriched within the module shown in (b)

Fig. 5

Histone gene knockdown is not sufficient to resensitise breast cancer cells to epirubicin. a A total of 7 × 104 ZR75-1 epirubicin-resistant (EpiR) cells and MDA-MB-231 EpiR cells were transfected with 30 nM of each siRNA (Dharmacon; GE Healthcare, Lafayette, CO, USA) targeting HIST1H2AC and HIST1H2BK (individual knockdowns not shown for simplicity). Negative controls included media only, Lipofectamine only or mock transfection with non-targeting siRNA. Percentage gene expression knock-down is shown in Additional file 1: Table S3. b IC50 values were generated using non-linear regression analysis, and average values of two independent experiments are graphed. Error bars represent standard deviation. c Fold changes in gene expression of each histone variant relative to the housekeeping gene, RPL37A. d-e Histone knockdown effects on cell cycle and apoptosis for MDA-MB-231 and ZR-75-1. Tables show average percentages of cells in G1, S and G2 stages of cell cycle, and precent live versus dead cells in each experimental and control condition. Numbers in parenthesis indicate standard deviation between two experiments

Fig. 6

Histone module is a biological marker for anthracycline therapy. High expression and low expression of histone module were tested for association with distant recurrence-free survival (DRFS) and overall survival (OS) in the BR9601 trial, in which patients were treated with standard chemotherapy (CMF) or anthracycline-containing chemotherapy (E-CMF). a DRFS and OS for patients treated with E-CMF vs. CMF split into high or low histone gene expression groups. b Multivariate, treatment by marker analysis after adjustment for human epidermal growth factor receptor 2 (HER2) status, oestrogen receptor (ER) status, nodal status, grade and age. HR hazard ratio, CI confidence interval

Fig. 7

Histone deacetylase inhibitors (HDACi) induce cytotoxicity in epirubicin-resistant (EpiR) cell lines. a An example of the effect of panobinostat on all cell lines. Half-maximal inhibitory concentration values for panobinostat and the remaining HDACi are shown in Additional file 1: Table S4. b A schematic indicating that HDACi could offer a viable treatment option for patients who do not respond to anthracyclines (high histone score). In addition, patients with low histone score who would initially benefit from epirubicin treatment may in due course develop de novo resistance, and, if diagnosed with “high histone score” at recurrence, may be offered HDACi as a subsequent treatment option. c Working models of molecular mechanisms involved in epirubicin resistance. There are three proposed mechanisms by which HDACi sensitise cells to epirubicin: (1) by transcriptional activation of repressors and pro-apoptotic genes, (2) by repression of resistance genes and (3) due to increased accessibility to DNA