Abstract
Estrogen responsive breast cancer cell lines have been extensively studied to characterize transcriptional patterns in hormone-responsive tumors. Nevertheless, due to current technological limitations, genome-wide studies have typically been limited to population averaged data. Here we obtain, for the first time, a characterization at the single-cell level of the states and expression signatures of a hormone-starved MCF-7 cell system responding to estrogen. To do so, we employ a recently proposed model that allows for dissecting single-cell states from time-course microarray data. We show that within 32 hours following stimulation, MCF-7 cells traverse, most likely, six states, with a faster early response followed by a progressive deceleration. We also derive the genome-wide transcriptional profiles of such single-cell states and their functional characterization. Our results support a scenario where estrogen promotes cell cycle progression by controlling multiple, sequential regulatory steps, whose single-cell events are here identified.
Highlights
Cellular responses to estrogens are characterized by a transcriptional activation and/or repression of specific subsets of genes, whose characterization will provide essential information on the molecular and genomic pathways of the hormone-responsive breast cancer (BC) phenotype
The dynamics of a cell is defined by the transition rates, wj,jz1, between all pairs of consecutive states j and jz1, and the single-cell states by their expression signatures: we indicate with bij the expression of gene i in state j
Number of States Our first aim was to quantify the most likely number of states that MCF-7 cells visit across 32 hours after estrogen stimulation
Summary
Cellular responses to estrogens are characterized by a transcriptional activation and/or repression of specific subsets of genes, whose characterization will provide essential information on the molecular and genomic pathways of the hormone-responsive breast cancer (BC) phenotype. To this aim, estrogen responsive BC cell lines are useful model systems because of their deep transcriptional similarities with ERa-expressing breast tumors [1,2]. A genome-wide quantitative analysis of the system at the single cell level is still lacking This is related to an intrinsic limitation of current major time course genome-wide assays. While such high-throughput techniques allow for a genome-wide characterization of the transformation of the population, they do not directly provide information on the cell states and expression signatures at the single-cell level
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
