Concurrent Transcript and Protein Quantification in a Massive Single Cell Array Enables Population-Wide Observation of Oncogene Escape

Abstract

The cancer stem cell (CSC) model focuses on the key role these cells can play in drug resistance, since CSCs are believed to reconstitute the cancer after intense chemotherapy treatment. In order to effectively identify and profile CSCs within a heterogeneous tumor population, we are investigating quantitative correlation of messenger RNAs and their translated proteins as a distinctive parameter of the CSC population. However, previous research on mRNA expression and protein abundance has shown the correlations between the two are weak or only “stochastically meaningful” due to the significant level of experimental error originating from ensemble observations. Here, we demonstrate a robust microwell-based method to minimize these errors by monitoring both mRNA expression and the corresponding protein abundance from individual cancer cells. Simultaneous observation of membrane protein expression by immunostaining and detection of mRNA transcripts directly from individual cells using a one-step, reverse transcription polymerase chain reaction have been integrated in a massive single-cell array platform. The proposed experimental scheme was initially tested and validated in three established lung cancer cell lines by correlating mRNA transcript and protein expression levels of individual cells and quantitation of heterogeneity. Responses at the individual cell level to known transcriptional and translational inhibitors, as well as to EGFR-specific inhibitors were evaluated, providing quantitative measures of the heterogeneous response of non-small cell lung cancer cells to the inhibitors. Results showed that drug-treated cell lines displayed oncogene escape due to expunction of drug-sensitive subpopulations in the cell lines. Furthermore, correlation of c-MET mRNA and protein levels revealed unique response patterns in different EGFR-mutated cell lines. Thus, these results demonstrate the potential for molecular profiling at the single cell level to prospectively identify the CSCs subpopulation for effective combinatorial treatments.