Methylation Profile of X-Chromosome-Related Genes in Male Breast Cancer

Abstract

Background: Androgen receptor (AR) has been described to play a prominent role in male breast cancer (MBC). It maps on chromosome X, and recent reports indicate that X-chromosome polysomy is frequent in MBC. Since the response to anti-androgen therapy may depend on AR polysomy and on its overexpression similarly to prostate cancer, the aim of the present study was to investigate the DNA methylation level of AR and its coregulators, especially those mapped on the X-chromosome, that may influence the activity of AR in MBC. Methods: The DNA methylation level of AR, MAGEA2, MAGEA11, MAGEC1, MAGEC2, FLNA, HDAC6, and UXT, mapped on the X-chromosome, was evaluated by quantitative bisulfite-NGS. Bioinformatic analysis was performed in a Galaxy Project environment using BWA-METH, MethylDackel, and Methylation Plotter tools. The study population consisted of MBC (41 cases) compared with gynecomastia (17 cases). Results: MAGEA family members, especially MAGEA2, MAGEA11, MAGEC, and UXT and HDAC6 showed hypomethylation of several CpGs, reaching statistical significance by the Kruskal-Wallis test (p < 0.01) in MBC when compared to gynecomastia. AR showed almost no methylation at all. Conclusions: Our study demonstrated for the first time that MAGEA family members mapped on the X-chromosome and coregulators of AR are hypomethylated in MBC. This may lead to their overexpression, enhancing AR activity.

Keywords: DNA methylation; FLNA; HDAC6; MAGE family; UXT; X-chromosome; androgen receptor; male breast cancer.

Figure 1

AR shows a strong positivity by immunohistochemistry in most neoplastic cells in male breast cancer.

Figure 2

Methylation plotter of MAGEA2, MAGEA11, and MAGEC2, showing statistically significant differences by Kruskal–Wallis U Test (CpG with KW < 0.05 are highlighted by an asterisk) between male breast cancer (MBC) and gynecomastia (Gyn).

Figure 3

Principle component analysis of all cases included in this study. Unit variance scaling is applied to rows; singular value decomposition (SVD) with imputation is used to calculate principal components. X and Y axes show principal component 1 and principal component 2 that explain 13.1 and 10.5% of the total variance, respectively. Prediction ellipses are such that with probability 0.95, a new observation from the same group will fall inside the ellipse. N = 58 data points.

Figure 4

HeatMap using the whole CpGs coming from seven genes evaluated in this study. Rows are centered; unit variance scaling is applied to rows. Both rows and columns are clustered using correlation distance and average linkage; 92 rows, 58 columns.