Search bioRxiv⌕ Search

Biology subjects

Foekens, J. A.

Publications and source records attributed to Foekens, J. A..

2 recordsLinked to original sources

Uncovering the signaling landscape controlling breast cancer cell migration identifies novel metastasis driver genes

Metastasis is the major cause of death in cancer patients and migration of cancer cells from the primary tumor to distant sites is the prerequisite of metastasis formation. Here we applied an imaging-based RNAi phenotypic cell migration screen using two highly migratory basal breast cancer cell lines (Hs578T and MDA-MB-231) to provide a repository for signaling determinants that functionally drive cancer cell migration. We screened ~4,200 individual target genes covering most cell signaling components and discovered 133 and 113 migratory modulators of Hs578T and MDA-MB-231, respectively, of which 43 genes were common denominators of cell migration. Interaction networks of candidate migratory modulators were in common with networks of different clinical breast cancer prognostic and metastasis classifiers. The splicing factors PRPF4B and BUD31 and the transcription factor BPTF were amplified in human primary breast tumors and the expression was associated with metastasis-free survival. Depletion of PRPF4B, BUD31 and BPTF caused primarily down-regulation of genes involved in focal adhesion and ECM-interaction pathways. PRPF4B was essential for triple negative breast cancer cell migration and critical for breast cancer metastasis formation in vivo, making PRPF4B a candidate for further drug development. Our systematic phenotypic screen provides an important repository of candidate metastasis drug targets.

cancer biology↗

Partially methylated domains are hypervariable in breast cancer and fuel widespread CpG island hypermethylation

Global loss of DNA methylation and CpG island (CGI) hypermethylation are regarded as key epigenomic aberrations in cancer. Global loss manifests itself in partially methylated domains (PMDs) which can extend up to megabases. However, the distribution of PMDs within and between tumor types, and their effects on key functional genomic elements including CGIs are poorly defined. Using whole genome bisulfite sequencing (WGBS) of breast cancers, we comprehensively show that loss of methylation in PMDs occurs in a large fraction of the genome and represents the prime source of variation in DNA methylation. PMDs are hypervariable in methylation level, size and distribution, and display elevated mutation rates. They impose intermediate DNA methylation levels incognizant of functional genomic elements including CGIs, underpinning a CGI methylator phenotype (CIMP). However, significant repression effects on cancer-genes are negligible as tumor suppressor genes are generally excluded from PMDs. The genomic distribution of PMDs reports tissue-of-origin of different cancers and may represent tissue-specific silent regions of the genome, which tolerate instability at the epigenetic, transcriptomic and genetic level.

cancer biology↗