A multi-tiered map of EMT defines major transition points and identifies vulnerabilities

Abstract

TGFβ mediated epithelial to mesenchymal transition (EMT) proceeds through hybrid "E/M" states. A deeper understanding of these states and events which regulate entry to and exit from the E/M states is needed for therapeutic exploitation. We quantified >60,000 molecules across ten time points and twelve omic layers in mammary epithelial cells. Proteomes of whole cells, phosphoproteins, nucleus, extracellular vesicles, secretome and membrane resolved major shifts, E→E/M and E/M→M during EMT, and defined state-specific signatures. Metabolomics identified early activation of arachidonic acid pathway and an enzyme-mediated switch from Cytochrome P450 to Cyclooxygenase / Lipoxygenase branches during E→E/M. Single-cell transcriptomics identified GLIS2 as an early modulator of EMT. Integrative modeling-predicted combinatorial inhibition of AURKB, PP2A and SRC exposed vulnerabilities at E→E/M juncture. Covariance analysis revealed remarkable discordance between proteins and transcripts, and between proteomic layers, implying insufficiency of current approaches. Overall, this dataset provides an unprecedented resource on TGFβ signaling, EMT and cancer.

Spatial transcriptomics reveals distinct and conserved tumor core and edge architectures that predict survival and targeted therapy response

“…CSTA has been identified as a tumor suppressor gene involved in mesenchymal-epithelial-transition (MET), while MT2A is a key regulator of EMT. 38,39 These findings predict that the TC undergos increased differentiation retaining an epithelial-like phenotype, while the LE becomes increasingly mesenchymal. Other differentially spliced driver genes in the TC and LE included proto-oncogenes and tumor suppressor genes (Fig.…”

Spatial transcriptomics reveals distinct and conserved tumor core and edge architectures that predict survival and targeted therapy response

“…CSTA has been identified as a tumor suppressor gene involved in mesenchymal-epithelial-transition (MET), while MT2A is a key regulator of EMT. 38,39 These findings suggest that the tumor core is predicted to undergo increased differentiation retaining an epithelial phenotype, while the leading edge becomes increasingly mesenchymal. Other differentially spliced driver genes in the tumor core and leading edge included proto-oncogenes and tumor suppressor genes (Fig.…”

DESP: Demixing Cell State Profiles from Dynamic Bulk Molecular Measurements

“…H2AFX is a core histone protein variant that both contributes to chromatin remodeling and is post-translationally modified during EMT 17,18 . Notably, H2AFX is hyper-phosphorylated during the earliest timepoints of EMT 13 , suggesting a change in this protein's functional activity during the initial intermediate cell states (Fig. 5e).…”

“…We used data from (Paul et al, 2021) 13 . In that study, cells of the human mammary epithelial cellline MCF10A were treated with TGF-Beta to induce epithelial-to-mesenchymal transition (EMT) and samples were aliquoted at ten timepoints for multi-omics profiling.…”

“…EMT is a complex biological process by which epithelial cells lose their adhesion and gradually differentiate into mobile mesenchymal cells and is of particular importance to the biomedical community due to its integral role in wound healing, tissue development, and cancer progression 12 . We generated bulk quantitative proteomics and scRNA-Seq data generated at eight consecutive timepoints representing MCF10A mammary epithelial cells undergoing EMT 13 , and we show that DESP is able to demix the corresponding cellular proteomes and recover expected patterns of canonical EMT markers. Additionally, DESP goes further by identifying the proteomic profile of transient cell states key to the differentiation process that were not detected at the transcriptomics level and whose signal is significantly diluted in the bulk proteomics due to the confounding presence of other cell states in the mixture.…”