Abstract
Human T cells responding to viral pathogens after HSCT can be dysfunctional and ineffective even when present in adequate numbers. The basis for this qualitative, rather than quantitative, T cell defect is not known. We hypothesized that impaired differentiation of naive T cells into memory T cells causes a qualitative T cell defect. We, therefore, studied CD8 T cell memory differentiation in 2 murine models of LCMV infection: an acute infection in which virus-specific T cells are fully functional and a chronic infection model in which T cells are present in normal numbers but are dysfunctional. In order to define how the differentiation state of dysfunctional T cells in chronic infection differ from those in acute infection, we generated gene expression profiles from naive (day 0), effector (day 7 [d7]), and memory (day 30 [d30]) T cells using oligonucleotide microarrays. In acute infection, T cell states were so distinct that 100% of naive, effector, and memory samples could be correctly classified, using a k-Nearest-Neighbor prediction classifier (kNN). Next, we hypothesized that the signature of normal memory T cells would contain gene-sets known to be important in memory differentiation. We found that Stat-5 target genes were the most enriched gene-sets in memory cells relative to naive cells of >400 gene-sets tested (P = .004 by permutation testing). This is consistent with the central role of Stat-5 signaling in memory development. Lastly, we determined where dysfunctional d30 T cells from chronic infection lie relative to the “differentiation-space” of acute infection. None of the chronic d30 T cell samples were classified as memory by kNN prediction; all were misclassified as effectors. However, chronic d30 T cells were not simply persistent effectors because chronic d30 T cells showed marked differences in gene expression pattern compared to d7 effectors. Thus, dysfunctional T cells in chronic viral infection fail to complete normal memory differentiation and are arrested in a distinct, effector-like phase. Our results show that the disruption of memory differentiation causes qualitative T cell dysfunction, and that this altered differentiation can be detected by gene-expression profiling. We are now generating a global map of antigen-specific T cell differentiation in humans. This gene-expression-based approach to evaluating immune reconstitution will identify the mechanisms limiting the recovery of T cell memory after HSCT.