Regulation of T cell Differentiation

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications & trials

Abstract

In the series of studies to be described we define the mechanism of induction and the function of a new type of IL-10-producing regulatory T cell that we have termed Tr2 T cells. New data generated during the current period provides evidence that stimulation of TORC1 activity and ensuing generation of a transcription factor that is essential to IL-10 transcription in Tr2 cells is critically dependent on a metabolic process known as glutaminolysis. In initial studies we showed that Tr2 cells are induced by dendritic cells (DCs) stimulated by zymogen-depleted yeast extracts (ZD) and by the hyphal form of C. albicans, both of which express 1,3-beta glucan, the ligand of Dectin-1. The T cells so stimulated undergo two interlocking molecular processes that together result in Tr2 cells. The first involves activation of GATA3, a factor that binds to the IL-10 promoter at two sites, i.e., at a distal site where it acts as a direct transcription factor and at the proximal site where it acts indirectly on transcription as a epigenetic factor that augments histone acetylation. The second involves activation of the TORC1 arm of the mTOR signaling pathway and the expression of a particular C/EBP isoform. These conclusions were initiated by micro-array analyses of Tr2 cells in which we showed that Tr2 gene expression was distinct from Tr1 and Th2 gene expression and that C/EBP signaling was among the several signaling pathways that could underlie this distinct expression pattern. In follow-up studies we showed that stimulation of T cells from mice with targeted deletion of C/EBP stimulated under Tr2 conditions led to greatly decreased IL-10 production as compared to similarly stimulated WT cells. In addition, we showed that T cells from C/EBP-deficient mice stimulated under Tr2 conditions in which C/EBP levels were partially repleted by retroviral expression of isoforms of C/EBP led to recovery of IL-10 production, but only if the repleting retrovirus expressed the LIP isoform of C/EBP but not the LAP isoform of C/EBP. Finally, we showed that C/EBP processing into LIP or LAP isoforms was dependent on mTOR signaling in that phosphorylation of eukaryote initiation factor ((elf)-4E) resulting from TORC1 activity regulated C/EBP translation into LIP and LAP and is necessary for LIP expression. Thus, in the absence of TORC-1 signaling due to the presence of rapamycin, LIP translation from C/EBP is virtually absent and, as a result, IL-10 production is greatly inhibited. Whereas the above findings established that mTOR (TOTRC1)activation was central to the induction of IL-10 synthesis in Tr2 cells, further explanation was necessary for how such activation was induced. Recent studies have shown that the metabolic profile of a T cell is one key determinant of its subset differentiation pathway. Accordingly, we performed metabolome studies to identify a metabolic profile that might be unique to Tr2 cells and influence its differentiation. Such studies revealed that Tr2 cells were clearly different from Th2 and Th0 cells according to partial least-squares discriminant analysis (PLS-DS) and showed a distinct metabolic signature . Among the various metabolic pathways unique to Tr2 cells, we focused on glutamate metabolism due to its known connection to the mTOR activation pathway. It is known that glutamine taken up by T cells is converted into glutamate by a process known as glutaminolysis and is then converted into alpha-ketoglutarate; the latter then either enters the TCA cycle for support of mitochondrial respiration or translocates to the lysosome to induce mTOR activity. On this basis, we first measured the total glutamine consumption as well as the glutaminolysis-linked oxygen consumption rate (OCR) and ATP production of Tr2 and Th2 cells generated in vitro. Whereas we found no difference between these two subsets with respect to total consumption, Tr2 cells exhibited a significantly lower level of OCR and ATP production than Th2 cells. This suggested that in Tr2 cells glutaminolysis-generated alpha-KG is utilized for a process other than mitochondrial respiration, most likely for the activation of the mTOR pathway. We next determined the mTOR activity (as evaluated by the phosphorylation status of S6, a downstream signaling component of mTOR) of comparable Th2 and Tr2 cell populations and found that Tr2 cells exhibited stronger mTOR activity than Th2 cells by this criterion. In addition, Tr2 cells exhibited greater ablation of IL-10 production than Th2 cells when exposed to a glutaminase inhibitor, 6-Diazo-5-oxo-L-norleucine (DON), caused specific inhibition of C/EBPbeta-LIP expression in Tr2 cells. Finally, we found that cell-permeable dimethyl alpha-KG (DMK) induced IL-10 production in Tr2 cells previously subjected to prior glutamine-deprivation. Taken together, these data strongly suggested that the mTOR pathway is activated in Tr2 cells by glutaminolysis and its downstream induction of alpha-KG. In parallel studies, we investigated the mechanism of how LIP regulates IL-10 production in Tr2 cells. These initially centered around studies with an IL-10 promoter-luciferase construct already alluded to above and showed that promoter activity was maximally stimulated by the presence of plasmids expressing CREB1 and LIP and in fact deletion of binding sites for these factors led to greatly reduced promoter activity. Since the CREB1 and LIP binding sites in the promoter are adjacent to one another and CREB1 had been shown previously to bind to C/EBP we reasoned that the LIP1/CREB1 cooperativity was due to facilitated binding of one or both factors to the IL-10 promoter. This hypothesis was subsequently supported by EMSA studies that showed that CREB1-LIP protein complexes extracted from the nucleus of HEK293 cells (pre-transfected with CREB1 and LIP expressing plasmids) bound to the DNA sequence found in the IL-10 promoter binding these transcription factors under physiologic conditions; in contrast, a similarly obtained CREB1-LAP complex had a poor capacity to bind to this sequence. These findings were accompanied by studies showing that C/EBP and CREB1 binding to the IL-10 promoter in Tr2 cells as determined by CHiP studies was enhanced in cells expressing LIP and LAP as compared to cells expressing only LAP, indicating the CREB1 binding is enhanced by complex formation with LIP. These studies support the conclusion that TORC1 signaling in nascent Tr2 cells leads to high IL-10 production because such signaling generates LIP-CREB1 complexes and augmented binding of these transcription factors to the IL-10 promoter. In initial studies of the regulatory function of Tr2 cells we showed that renal infection caused by C. albicans is accompanied by renal infiltration of IL-10 producing CD4 T cells and that mice that have CD4 T cells that cannot produce IL-10 (i.e., mice lacking Tr2 cells) exhibit better survival of infection than wild type mice. Thus, the generation of Tr2 cells limits the pro-inflammatory (protective) effect of C. albicans infection. This seemingly negative effect of T2 cells could provide selective advantage during chronic fungal infection were response to infection is the main case of morbidity. In further exploration of the regulatory function of Tr2 cells we conducted studies of Tr2 cell regulation of experimental asthma induced by house-dust mite antigen (HDM. We found that ZD administration (IP) during initial asthma induction by HDM gives rise to a dramatic reduction in total BAL cells, BAL eosinophils and CD4-positive cells; in addition, total IgE and HDM-specific IgE in the circulation are dramatically reduced. These studies thus showed that Tr2 can be induced by ZD during a Th2-driven inflammation such as asthma and may therefore have efficacy in treating asthma.