Multiplexed, targeted profiling of single-cell proteomes and transcriptomes in a single reaction

Follow up from the “single-cell” paper (Darmanis et al 2016). Again looking at protein and RNA levels at the single cell level, but here they took a methodology development step by integrating the PEA and RNA analyses into the same assay, using a Fludigm chip/instrument. This was achieved by using reverse transcriptase instead of DNA polymerase to carry out the extension step in PEA, enabling parallel PEA/RT-PCR analysis from the same sample, using the IFC for a Fluidigm C1 instrument. They examined 96 RNAs and 38 overlapping protein markers using this method. After initial evaluation using recombinant protein dilutions, 27 PEA assays and 89 RT-PCR assays showed good linear detection above background down to single-cell level. As a test system, MCF-7 breast cancer cultures were chemically treated with PMA, which is known to activate protein kinase C, inhibit cell growth and induce apoptosis in this cell line. As a pre-test, 4 protein/RNA markers were also analyzed using standard in situ staining techniques, and these showed good qualitative agreement with the PEA/RT-PCR assays for patterns of heterogeneity. PCA of the complete dataset showed that the 27 proteins and 89 RNAs distinguished PMA-treated from control cells. Despite using significantly fewer markers, the PEA analysis gave clearer separation than the RNA. As seen in the previous publication, the correlation between equivalent proteins and RNAs at the single cell level wasn’t very strong. The more highly expressed RNAs tended to be better correlated with their proteins in untreated cells, whereas those with lower expression levels were better correlated in treated cells. One interesting individual RNA/protein was MET, which had almost no correlation in untreated cells (p=0.03) but a very strong positive correlation (p=0.53) in treated cells. Closer investigation led them to speculate that MET may be primarily regulated at the translation level rather than at the transcription level. Overall this study showed that this is a potentially powerful new tool for performing true parallel RNA/Protein analysis at the single cell level, and preliminary data supports the earlier propostion that protein markers may be the most informative indicator of cell heterogeneity.