Abstract 3491: Rapid and ultra-sensitive single-cell transcript profiling with droplet digital PCR (ddPCR): Application to cell cycle analysis

Abstract

Abstract The recognition that tumors and other apparently homogeneous cell populations are often heterogeneous and display stochastic expression of their genes, has motivated the development of methods to profile gene expression in single cells. Among these, RNA-Seq is the most powerful for broad-scale profiling and discovery of genes and pathways influencing biological processes and states of interest. However, given the extent of sample manipulation required to prepare each RNA-Seq library – and the trade-off between sensitivity and sample throughput – it is important to validate discovery findings by orthogonal digital methods with comparable or greater sensitivity and accuracy. Droplet Digital PCR is such a method that can rapidly and cost-effectively replicate (or challenge) RNA-Seq results with minimal sample processing when defined targets are to be verified (Chen et al, 2012). Furthermore, ddPCR is performed in 96-well plates and is well-suited to high throughput studies of focused sets of genes in large numbers of single cells (e.g. hundreds in a day). Here we present a simple and robust workflow for profiling multiplexed, transcript targets in flow-sorted, Jurkat single-cells using Taqman 5’ probe hydrolysis assays. We demonstrate that Bio-Rad's QX100 & QX200 Droplet Digital PCR systems provide absolute counts of transcripts from >100,000 copies to <10 copies per cell, using unpurified and unamplified cDNA synthesized in 30’ in the original wells used for flow-sorting. The results are highly reproducible and the method is currently being used to explore the contribution of the cell-cycle to expression differences among single-cell populations. More than 30 cell-cycle genes have been evaluated by ddPCR in exponentially growing Jurkat cells. Several genes, including CCNB1, CDK1 and DNM2, show an unusually broad range of transcript levels (over 2 logs), suggestive of their known phase-specific cell-cycle roles. Jurkat cells have been FACS sorted for either G0/G1 (“G1”) or G2/M (“G2”) single-cells using DyeCycle Orange to stain for DNA content, and are being assessed for phase-specific expression profiles. Preliminary results indicate that G2 single-cells are enriched for higher CCNB1 and CDK1 transcript levels as expected from their role in promoting early events in mitosis; however, even G2 cells still show expression levels spanning ∼2 logs, suggesting cell-cycle phase heterogeneity and/or bursty transcription within the sorted G2 population. While the average mRNA level of even the house-keeping gene, GAPDH, is several-fold higher in G2 cells as compared to G1 cells, transcripts for CCNB1 and CDK1 average ∼6-fold higher in G2 than in G1. In this work, we demonstrate both the analytical performance and the biological utility of droplet digital PCR for single-cell analysis. This same workflow could be used for fingerprinting and study of sub-populations in tumors. Citation Format: George Karlin-Neumann, Shenglong Wang, Camille Troup, Yann Jouvenot, Eli Hefner. Rapid and ultra-sensitive single-cell transcript profiling with droplet digital PCR (ddPCR): Application to cell cycle analysis. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3491. doi:10.1158/1538-7445.AM2014-3491