This site will redirect to thermofisher.com on April 7. Get details >


Reveal gene expression heterogeneity at the single cell level

Cellular heterogeneity is present in any biological sample, be it from our less-than-perfect ability to subset cells, or subtle gene expression differences from cell to cell. Paradoxically, most of our understanding of gene expression is based upon bulk population averages. This bulk analysis, although informative, often leads to conclusions that assume ensemble averages reflect the dominant biological mechanism operating within the entire population. Using such measurements and assumptions can mask the presence of rare or small subpopulations of cells or bimodal cellular behaviors, and ignores essential cell-to-cell differences. To fully understand if cellular heterogeneity contributes to biological function or contains relevant information, a single-cell approach must be applied. The hidden story behind qRT-PCR’s inability to look beyond the population as a whole, and a bimodal response upon stimulation, is illustrated with a comparison to PrimeFlow® RNA Assay.

Revealing a hidden story beneath bulk masking

Aims
mRNA is commonly assessed by quantitative RT-PCR (qRT-PCR) where cells are isolated after stimulation so that total RNA can be extracted and then undergo TaqMan® or SYBR® Green RT-PCR reactions. This results in amplified bulk measurements that mask the stochastic gene expression that occurs at the single cell level. In this example, PrimeFlow® RNA data are compared to qRT-PCR.

Results
The same cell populations of normal human peripheral blood mononuclear cells (PBMCs) stimulated with PMA and Ionomycin for 0-5 hours were analyzed by qRT-PCR and PrimeFlow® RNA in parallel. By qRT-PCR, it would appear that IFNγ and TNFα are initially induced at similar levels and that IFNγ plateaus while TNFα declines slowly. However, with the PrimeFlow® RNA assay, it is observed that TNFα mRNA is rapidly and highly induced at 1 hour, dramatically declines between 1-2 hours, and then declines more slowly over 3-5 hours after stimulation. In contrast, IFNγ is maintained the entire time course. As shown earlier, the PrimeFlow® RNA data can be further analyzed into CD8+ or CD8- cells which reveal additional sample heterogeneity that qRT-PCR cannot achieve from a single sample. Even though qRT-PCR shows that IFNγ is expressed at higher than TNFα, PrimeFlow® RNA shows that the percentage of cells expressing IFNγ is actually less than the number of cells that express TNFα.

Conclusions
Examination of stimulated cells over time using current qRT-PCR technology shows similar kinetics as with PrimeFlow® RNA. However, PrimeFlow® RNA uncovers finer details of the kinetics at a single cell level and allows users to study multiple parameters within the same sample, thereby eliminating the bulk averages previously masking cellular heterogeneity.

Figure 1
RNA Expression by qPCR

Figure 2
Cytokine mRNA Signal Intensity by FlowRNA

Figure 3
Percent of Cells Expressing Cytokine by FlowRNA

Figure 4

Kinetics of  TNFα transcription and translation measured by PrimeFlow® RNA Assay

Legend: Normal human peripheral blood mononuclear cells were stimulated with PMA and Ionomycin for 0-5 hours. Cells' RNA was exacted and was analyzed by qRT-PCR (Figure 1). Using PrimeFlow® RNAassay, the cells were fixed, permeabilized, and intracellularly stained with antibodies for CD8, IFN-gamma, TNF-alpha. Next, cells underwent a series of hybridization steps to label mRNA for IFN-gamma and TNF-alpha. Viable cells in the lymphocyte gate were analyzed for relative cytokine mRNA medium fluorescent intensity (Figure 2), for percentage of cytokine mRNA producing cells (Figure 3), and were further gated upon CD8+ and CD8- and analyzed for percentage of TNF-alpha mRNA producing cells (Figure 4).