Abstract
Understanding cell proliferation mechanisms has been a long-lasting goal of the scientific community and specifically of cancer researchers. Previous genome-scale studies of cancer proliferation determinants have mainly relied on knockdown screens aimed to gauge their effects on cancer growth. This powerful approach has several limitations such as off-target effects, partial knockdown, and masking effects due to functional backups. Here we employ a complementary approach and assign each gene a cancer Proliferation Index (cPI) that quantifies the association between its expression levels and growth rate measurements across 60 cancer cell lines. Reassuringly, genes found essential in cancer gene knockdown screens exhibit significant positive cPI values, while tumor suppressors exhibit significant negative cPI values. Cell cycle, DNA replication, splicing and protein production related processes are positively associated with cancer proliferation, while cellular migration is negatively associated with it – in accordance with the well known “go or grow” dichotomy. A parallel analysis of genes' non-cancerous proliferation indices (nPI) across 224 lymphoblastoid cell lines reveals surprisingly marked differences between cancerous and non-cancerous proliferation. These differences highlight genes in the translation and spliceosome machineries as selective cancer proliferation-associated proteins. A cross species comparison reveals that cancer proliferation resembles that of microorganisms while non-cancerous proliferation does not. Furthermore, combining cancerous and non-cancerous proliferation signatures leads to enhanced prediction of patient outcome and gene essentiality in cancer. Overall, these results point to an inherent difference between cancerous and non-cancerous proliferation determinants, whose understanding may contribute to the future development of novel cancer-specific anti-proliferative drugs.
Highlights
Cancer is one of the leading causes of death worldwide and it is estimated that 12.7 million new cancer cases and 7.6 million cancer deaths occurred in 2008 [1]
Computing a cancer Proliferation Index We focus on studying the NCI-60 panel, consisting of 60 different cancer cell lines originating from nine different tumor types [14]
We find that genes associated with cholesterol metabolism have significantly negative cancer Proliferation Index (cPI) values in colon cancer, in accordance with studies showing that high density lipoprotein (HDL) cholesterol levels are inversely correlated with the risk of colon cancer [30]
Summary
Cancer is one of the leading causes of death worldwide and it is estimated that 12.7 million new cancer cases and 7.6 million cancer deaths occurred in 2008 [1]. In an attempt to find new anti-cancer drug targets, various studies in recent years used short hairpin RNA (shRNA) techniques and screened thousands of genes to find those that are essential for cancer growth and proliferation and are putative targets for clinical intervention in cancer [4,5,6,7] While this approach was shown to be powerful for the analysis of biological processes, shRNA screens have a variety of limitations such as off-target effects, partial knock down of the target genes and more [8,9,10]. As these are essentiality screens, they are less adequate to highlight genes involved in biological processes that have functional backups
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
