Gene Expression Profiles from Peripheral Blood Mononuclear Cells Are Sensitive to Short Processing Delays

Abstract

In the analysis of peripheral blood gene expression, timely processing of samples is essential to ensure that measurements reflect in vivo biology, rather than ex vivo sample processing variables. The effect of processing delays on global gene expression patterns in peripheral blood mononuclear cells (PBMCs) was assessed by isolating and stabilizing PBMC-derived RNA from 3 individuals either immediately after phlebotomy or after a 4 h delay. RNA was labeled using NuGEN Ovation labeling and probed using the Affymetrix HG U133 Plus 2.0 GeneChip(®). Comparison of gene expression levels (≥2-fold expression change and P < 0.05) identified 307 probe sets representing genes with increased expression and 46 indicating decreased expression after 4 h. These differentially expressed genes include many that are important to inflammatory, immunologic, and cancer pathways. Among others, CCR2, CCR5, TLR10, CD180, and IL-16 have decreased expression, whereas VEGF, IL8, SOCS2, SOCS3, CD69, and CD83 have increased expression after a 4 h processing delay. The trends in expression patterns associated with delayed processing were also apparent in an independent set of 276 arrays of RNA from human PBMC samples with varying processing times. These data indicate that the time between sample acquisition, initiation of processing, and when the RNA is stabilized should be a prime consideration when designing protocols for translational studies involving PBMC gene expression analysis.

FIG. 1.

Processing effects of gene expression signatures. (A) Samples clustered according to 353 probe sets identified (P < 0.05 and fold-change ≥2-fold) representing genes differentially expressed between samples processed immediately or after a 2 or 4 h delay. (B) Samples from this study were clustered according to gene expression data related to 2082 probe sets identified by Baechler et al. as differentially expressed after overnight shipment of peripheral blood before RNA extraction. Each column represents 1 sample and each row represents 1 probe set. In the heatmap, expression levels are indicated by color (yellow, median; red, increased; blue, decreased). Bar at the bottom indicates class membership: red, immediate processing; yellow, 2 h delay; blue, 4 h delay. See online article at www.liebertonline.com for color figure.

FIG. 2.

Average expression levels of time-sensitive probe sets. In a pilot experiment, 307 probe sets were identified with increased expression and 46 with decreased expression after a 4 h processing delay. Average expression of probe sets with increased (solid line) or decreased (dotted line) expression is shown in 267 independent samples. Samples are arranged according to time to processing.