Transcriptomic profiles of aging in naive and memory CD4+ cells from mice

Background: CD4+ T cells can be broadly divided into naive and memory subsets, each of which are differentially impaired by the aging process. It is unclear if and how these differences are reflected at the transcriptomic level. We performed microarray profiling on RNA derived from naive (CD44(low)) and memory (CD44(high)) CD4+ T cells derived from young (2-3 month) and old (28 month) mice, in order to better understand the mechanisms of age-related functional alterations in both subsets. We also performed follow-up bioinformatic analyses in order to determine the functional consequences of gene expression changes in both of these subsets, and identify regulatory factors potentially responsible for these changes. Results: We found 185 and 328 genes differentially expressed (FDR <= 0.05) in young vs. old naive and memory cells, respectively, with 50 genes differentially expressed in both subsets. Functional annotation analyses highlighted an increase in genes involved in apoptosis specific to aged naive cells. Both subsets shared age-related increases in inflammatory signaling genes, along with a decrease in oxidative phosphorylation genes. Cis-regulatory analyses revealed enrichment of multiple transcription factor binding sites near genes with age-associated expression, in particular NF-kappa B and several forkhead box transcription factors. Enhancer associated histone modifications were enriched near genes down-regulated in naive cells. Comparison of our results with previous mouse and human datasets indicates few overlapping genes overall, but suggest consistent up-regulation of Casp1 and Il1r2, and down-regulation of Foxp1 in both mouse and human CD4+ T cells. Conclusions: The transcriptomes of naive and memory CD4+ T cells are distinctly affected by the aging process. However, both subsets exhibit a common increase inflammatory genes and decrease in oxidative phosphorylation genes. NF-kappa B, forkhead box, and Myc transcription factors are implicated as upstream regulators of these gene expression changes in both subsets, with enhancer histone modifications potentially driving unique changes unique to naive cells. Finally we conclude that there is little overlap in age-related gene expression changes between humans and mice; however, age-related alterations in a small subset of genes may be conserved.