BACKGROUND: microRNAs (miRNAs) are emerging as key regulators of the immune system, but their role in CD8+ T cell differentiation is not well explored. Some evidence suggests that signals from cell surface receptors influence the expression of miRNAs in CD8+ T cells, and may have consequent effects on cell phenotype and function. We set out to investigate whether common gamma chain cytokines modulated human CD8+ T cell expression of miR-146a, which previous studies have associated with different stages of CD8+ differentiation. We also investigated how changes in miR-146a related to other miRNAs that alter with CD8+ differentiation status. METHODS: We treated human CD8+ T cells with the cytokines IL-2, IL-7 or IL-15 either at rest or after stimulation with anti-CD3 and anti-CD28. For some experiments we also purified human CD8+ T cell subsets ex vivo. Flow cytometry was used in parallel to assess cell surface memory marker expression. Total RNA from these cells was subjected to microarray analysis and real-time PCR for miRNA expression. Nucleofection studies were performed to assess potential mRNA targets of miR-146a. RESULTS: We find that miR-146a is up-regulated in naïve CD8+ T cells exposed to IL-2 or IL-15, even in the absence of an activating T cell receptor stimulus, but not when IL-7 is also present. miR-146a expression correlates with a memory phenotype in both ex vivo and in vitro cultured cells although in our hands overexpression of miR-146a was not sufficient alone to drive a full memory phenotype. In ex vivo analysis, miR-146a was one of a small number of miRNAs that was differentially expressed between naïve and memory CD8+ T cells. CONCLUSIONS: miR-146a is emerging as a critical regulator of immune system. Our data shows that miR-146a expression is strongly influenced by the cytokine milieu even in the absence of a T cell receptor stimulus. Our results have implications for studies designed to assess the function of miR-146a, help to define a fingerprint of miRNA expression in CD8+ T cell subsets and may be useful when designing optimal protocols for T cell expansion as efficacy of T cell immunotherapy is correlated with an 'early' memory phenotype.