Dataset Information

Stromal-associated cytokines bias the interplay between gene expression and DNA methylation in human breast cancers.

ABSTRACT: In human tumours, the crosstalk between cancer cells and their microenvironment is involved in tumour progression, metastasis and resistance to anti-cancer therapies. Among the factors involved in this exchange of information pro-inflammatory cytokines seem to play a crucial role. We observed that a group of pro-inflammatory cytokines, interleukin 6 (IL6), interleukin 1-beta (IL1b), and tumour necrosis factor-alpha (TNFa), preferentially activated genes exhibiting a high basal methylation level at their transcription start sites, in the human breast cancer cell line MCF7. In human breast tumours, these responding genes were also hypermethylated, and some of them (N = 104) were differentially methylated across human breast tumour samples (The Cancer Genome Atlas cohort). While their expression was positively correlated with the stromal content of the tumours and the expression of stromal-associated pro-inflammatory cytokines, the expression of this subset of genes was negatively correlated with their methylation level at their 5' end. Nevertheless, while the methylation level of this subset of genes was not correlated with the stromal cell content of the tumours, this negative correlation was partially lost in tumours with high stromal cell content. Consistently, we observed that the methylation level in this subset of genes influenced the correlation between gene expression and stromal cell content. Thus, these data indicated that the stromal component of breast tumours should be taken into account for DNA methylation and gene expression studies and suggest an additional pathway, via DNA methylation, in the cross-talk between cancer cells and their microenvironment in human breast cancers.

SUBMITTER: Hernandez-Vargas H

PROVIDER: S-EPMC7188392 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Breast cancer is the most common cause of cancer death among women worldwide. Localized breast cancer can be cured by surgery and adjuvant therapy, but mortality remains high for tumors that metastasize early. Type IV collagen is a basement membrane protein, and breach of this extracellular matrix structure is the first step of cancer invasion. Type IV collagen is found in the stroma of many cancers, but its role in tumor biology is unclear. Here, expression of type IV collagen in the stroma of small breast cancers was analyzed, correlated to clinically used prognostic biomarkers and patient survival. The findings were further validated in an independent gene expression data cohort. Tissue samples from 1,379 women with in situ and small invasive breast cancers (≤15 mm) diagnosed in 1986-2004 were included. Primary tumor tissue was collected into tissue microarrays. Type IV collagen expression in tissues was visualized using immunohistochemistry. Gene expression data was extracted from the Cancer Genome Atlas database. Out of 1,379 women, 856 had an invasive breast cancer and type IV collagen staining was available for 714 patients. In Kaplan-Meier analysis high type IV collagen expression was significantly associated (p = 0.026) with poorer breast cancer specific survival. There was no correlation of type IV collagen expression to clinically used prognostic biomarkers. High type IV collagen expression was clearly associated to distant metastasis (p = 0.002). In an external validation cohort (n = 1,104), high type IV collagen mRNA expression was significantly (p = 0.041) associated with poorer overall survival, with overexpression of type IV collagen mRNA in metastatic tissue. Stromal type IV collagen expression in the primary tumor correlates to poor breast cancer specific survival most likely due to a higher risk of developing distant metastasis. This ECM protein may function as biomarker to predict the risk of future metastatic disease in patients with breast cancers.

Project description:IntroductionAberrant DNA methylation has been found frequently in human breast cancers, associated with the loss of expression of a number of regulatory genes for growth and correlated to clinical outcomes. The present study was undertaken to determine whether methylation of a set of growth-suppressor genes would correlate to the expression of estrogen receptors (ERs) and progesterone receptors (PRs).MethodsWe used a pyrosequencing methylation analysis to study the methylation of 12 known growth-suppressor genes in 90 pairs of malignant/normal breast tissues. We also examined the expression of ERs and PRs in those specimens by immunohistochemistry. Mutations of p53 in tumor cells were detected by direct sequencing.ResultsTwelve tumor-suppressor genes: ARHI, RASSF1A, HIN-1, RARbeta2, hMLH1, 14-3-3 sigma, RIZ1, p16, E-cadherin, RIL, CDH13, and NKD2 were selected for this methylation study. Five of them (RIL, HIN-1, RASSF1A, CDH13, and RARbeta2) were frequently methylated in breast cancers (57%, 49%, 58%, 44%, and 17%, respectively) but not the normal breast (0-4%). Two panels of methylation profiles were defined. The methylation of the HIN-1/RASSFIA panel strongly correlated to the expression of ERs, PRs, and hormone receptors (HRs; which were defined as 'positive' if ERs and/or PRs were positive; p < 0.001). Conversely, the methylation of the RIL/CDH13 panel strongly correlated to negative ER, PR, and HR expression (p = 0.001, 0.025, and 0.001, respectively). The subset of triple-negative breast cancers (in other words, those with negative ER, PR, and HER-2/neu status) was positively associated with the methylation of the RIL/CDH13 panel and negatively associated with the HIN-1/RASSF1A panel. Mutations of p53 were found in nine breast tumors (11%), seven of which lacked methylation in both panels.ConclusionWe have defined two panels (HIN-1/RASSFIA, and RIL/CDH13) of methylation profiles, which correlated, either positively or negatively, to HR status.

Dataset Information

Stromal-associated cytokines bias the interplay between gene expression and DNA methylation in human breast cancers.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets