Abstract
Following activation by antigen, costimulation, and inflammation, naïve CD8+ T cells initiate a program of clonal expansion and differentiation resulting in wide-spread changes in expression of genes involved in cell-cycle, metabolism, effector function, apoptosis, and homing. Although, several key transcription factors (TFs) have been shown to be important in effector CD8+ T cell differentiation, the precise transcriptional regulation of this differentiation program remains poorly understood. The AP-1 family member BATF plays an important role in regulating differentiation and function in CD4+ Th17 cells, CD4+ follicular helper T cells, and in Ig class switching in B cells. We now show that BATF is also required for effector CD8+ T cell differentiation and regulates a core program of genes involved in effector differentiation. We found that BATF expression is rapidly up-regulated during effector CD8+ T cell differentiation in the mouse model of lymphocytic choriomeningitis virus (LCMV) infection. To examine the role of BATF in effector differentiation, we studied congenically distinct wild type (WT) and BATF knockout (KO) naïve P14 TCR transgenic CD8+ T cells co- transferred into a WT host. Upon infection, the BATF KO cells exhibited a profound, cell-intrinsic defect in effector CD8+ T cell differentiation, with a ∼400-fold decrease in peak number of effector cells. BATF KO effectors showed sustained activation and increased cell death by the mid-expansion phase of the immune response. To address the question of how loss of BATF causes such a severely diminished antigen-specific response, we profiled the binding sites of BATF throughout the genome by chromatin immunoprecipitation and high-throughput sequencing (ChIP-seq) in primary CD8+ effector cells. We found that BATF bound to regulatory regions in many genes critical for effector differentiation, including transcription factors (e.g. Tbx21, Eomes, Prdm1), genes involved in cytokine signaling (e.g. Il12rb2, Il2ra), homing (e.g. Sell, Selp, Ccr9), effector function (e.g. Gzmb, Ifng, Il2), apoptosis (e.g. Bcl2, Bcl2l1, Mcl1), and T cell activation (e.g. Ctla4, Cd247, Tnfrsf4), suggesting a major role for BATF in effector CD8+ T cell differentiation. Indeed, we found that genes bound by BATF were highly significantly overrepresented among genes that changed as a result of naïve CD8+ T cells differentiating into effectors in vivo (P = 10-27). Comparison of gene expression in in vitro WT and BATF KO effectors confirmed that BATF bound genes were perturbed by BATF loss of function. Analysis of the kinetics of gene expression during the first 72 hours of effector differentiation showed that loss of BATF perturbed the temporal sequence of expression of critical transcription factors, such as T-bet and Eomes, and resulted in inappropriately early cytokine expression. This suggests that BATF may be required to coordinate the earliest events in CD8+ T cell effector differentiation. To test this hypothesis, we used in vivo CFSE tracking to follow the early CD8+ T cell response during LCMV infection. We found that while BATF KO CD8+ T cells initiate cell division, there was a dramatic collapse in the ability to sustain proliferation and differentiation as early as day 3 post-infection. These results indicate that BATF ensures the orderly progression of a program of genes required by effector cells, restraining the expression of some and promoting the expression of others. More broadly, our results suggest that BATF may provide a common regulatory infrastructure for the development of effector cells in all T cell lineages.
Disclosures:
Wherry: Genentech: Patents & Royalties.
The NK receptor KLRG1 is dispensable for virus‐induced NK and CD8
+
T‐cell differentiation and function
in vivo