Chromatin interaction analysis reveals changes in small chromosome and telomere clustering between epithelial and breast cancer cells

Abstract

BackgroundHigher-order chromatin structure is often perturbed in cancer and other pathological states. Although several genetic and epigenetic differences have been charted between normal and breast cancer tissues, changes in higher-order chromatin organization during tumorigenesis have not been fully explored. To probe the differences in higher-order chromatin structure between mammary epithelial and breast cancer cells, we performed Hi-C analysis on MCF-10A mammary epithelial and MCF-7 breast cancer cell lines.ResultsOur studies reveal that the small, gene-rich chromosomes chr16 through chr22 in the MCF-7 breast cancer genome display decreased interaction frequency with each other compared to the inter-chromosomal interaction frequency in the MCF-10A epithelial cells. Interestingly, this finding is associated with a higher occurrence of open compartments on chr16–22 in MCF-7 cells. Pathway analysis of the MCF-7 up-regulated genes located in altered compartment regions on chr16–22 reveals pathways related to repression of WNT signaling. There are also differences in intra-chromosomal interactions between the cell lines; telomeric and sub-telomeric regions in the MCF-10A cells display more frequent interactions than are observed in the MCF-7 cells.ConclusionsWe show evidence of an intricate relationship between chromosomal organization and gene expression between epithelial and breast cancer cells. Importantly, this work provides a genome-wide view of higher-order chromatin dynamics and a resource for studying higher-order chromatin interactions in two cell lines commonly used to study the progression of breast cancer.Electronic supplementary materialThe online version of this article (doi:10.1186/s13059-015-0768-0) contains supplementary material, which is available to authorized users.

SMARCA4 regulates gene expression and higher-order chromatin structure in proliferating mammary epithelial cells

“…S4A-J). The shSCRAM and shSMARCA4 Hi-C data sets exhibited high correlation (R 2 > 0.95) with the previously published wild-type MCF-10A Hi-C data set (Barutcu et al 2015). The scaling plot curves of genomic interaction frequencies along genomic distance showed similar trends of decay and an increase at distances larger than >200 Mb, which we previously showed to be a characteristic of the MCF-10A genome (Supplemental Fig.…”

“…Moreover, we identified large blocks of inter-chromosomal interactions between Chr 3 and Chr 5, Chr 3 and Chr 9, and Chr 6 and Chr 19, which represent previously known translocations in the MCF-10A genome (Supplemental Fig. S5; Marella et al 2009;Barutcu et al 2015). Taken together, these analyses reflect the high quality and reproducibility of the Hi-C data.…”

“…The scaling plot curves of genomic interaction frequencies along genomic distance showed similar trends of decay and an increase at distances larger than >200 Mb, which we previously showed to be a characteristic of the MCF-10A genome (Supplemental Fig. S4C; Barutcu et al 2015). The Hi-C data sets displayed similar cis/trans interaction ratios (an average of ∼25%) (Supplemental Fig.…”

“…4A; Barutcu et al 2015;Crane et al 2015). At multiple resolutions (10-Mb to 40-kb resolution genome-wide), SMARCA4 knockdown resulted in the disruption of existing interactions and the emergence of novel specific contacts throughout each chromosome ( Fig.…”

“…Compartmentalization of the genome is correlated with gene expression (Barutcu et al 2015;Dixon et al 2015). To understand the link between compartment switching and gene expression upon SMARCA4 knockdown, we plotted the shSMARCA4/ shSCRAM log 2 fold change RNA-seq expression levels of the genes that were located either within unchanged compartments or within regions that switched compartments (Fig.…”