The vertebrate immune response to invading pathogenic microbes requires coordination between early sensing of microbial products by innate immune cells and subsequent activation of antigen-specific lymphocytes that make up the adaptive immune system. Our laboratory studies the regulatory mechanisms that govern the development of distinct classes of lymphocytes, their migration to lymphoid organs and sites of inflammation, and their differentiation into cells with specialized effector functions following interaction with innate immune system cells.
We also investigate how the composition of the commensal microbiota influences host defense against pathogens and susceptibility to autoimmune disease. Microbial pathogens have adopted numerous strategies designed to circumvent host immune responses. We study mechanisms by which the human immunodeficiency virus (HIV) interacts with host cells to subvert normal immune defenses to its advantage, and we seek to understand how HIV and other retroviruses activate host innate immunity, with the aim of developing better vaccine strategies.
The majority of mature T lymphocytes fall into one of two functional categories: helper cells, which react with peptides complexed to major histocompatibility complex (MHC) class II molecules on antigen-presenting cells, and cytotoxic cells, which recognize peptides bound to MHC class I molecules. These cells are distinguished on the basis of surface expression of the CD4 or CD8 coreceptors, which are coexpressed on immature double-positive (DP) thymocytes but are singly expressed upon maturation on thymocytes with T cell antigen receptors (TCRs) specific for class II and class I, respectively.
In the thymus, DP cells with TCRs of desired affinity for self-antigen undergo "positive selection" and migration to secondary lymphoid organs. The selection is coupled to transcriptional shutoff of CD4 or CD8 and to commitment to the cytotoxic or helper lineage, respectively. To understand the mechanism of lineage specification, we have studied the transcriptional regulation of the coreceptor genes. We identified a region within the Cd4 gene that is required for silencing expression of CD4 in the cytotoxic T cell lineage, and we showed that the transcription factor Runx3 and its partner protein, CBFβ, bind to motifs in the silencer and are required for initiation, but not maintenance, of epigenetic silencing. The Runx3-CBFβ complex also contributes to the activation of CD8 expression and to the activation or repression of a large number of genes in CD8-lineage cells, and we are working on identifying direct targets.
Another transcription factor, ThPOK, directs positively selected cells to the CD4 lineage and prevents specification of CD8-lineage T cells. We found that ThPOK, which is up-regulated only in MHC IIselected thymocytes, blocks expression of Runx3 in such cells after the CD4 lineage has already been specified. Rémy Bosselut's group (National Cancer Institute) showed that the transcription factor GATA-3 is up-regulated before ThPOK in class IIselected cells and is required for subsequent expression of ThPOK, while Ichiro Taniuchi (RIKEN Research Center for Allergy and Immunology, Japan) and colleagues have shown that ThPOK functions in a positive autoregulatory loop that secures commitment to the CD4 lineage. Thus, it appears that GATA-3 is a specifying factor for T helper cell differentiation while ThPOK subsequently reinforces lineage commitment by preventing Runx3 expression and diversion to the CD8 lineage. An important question that remains to be addressed is how TCR signals during and after positive selection initiate transcriptional programs that specify the alternate cell fates.
We are also identifying additional transcription factors that are involved in Cd4 silencer function and that may also contribute to lineage specification and commitment. We have also shown that the enhancer required to initiate CD4 expression in DP thymocytes establishes a heritable state of active gene expression that is maintained in mature T helper cells, and we are studying additional linked genetic elements that may communicate with and be "instructed" by the enhancer. Thus, Cd4 is regulated by heritable ON and OFF marks, and we seek to identify the trans-acting factors involved in initiation and maintenance of these heritable states. We are focusing, in particular, on noncoding RNAs to determine if any are involved in lineage specification.
In related studies, we have generated mice with a conditional mutant allele of the RNAseIII enzyme Drosha, which is required for processing of microRNAs (miRNAs). T cells defective for Drosha or Dicer, which is involved in RNA interference and miRNA biogenesis, have been examined for small-RNA composition and for functional defects. Both enzymes were found to be required for acquisition of regulatory T cell (Treg) anti-inflammatory function, and we discovered novel activities distinct from the canonical microRNA pathway. Drosha was found to also regulate protein levels by direct cleavage of coding mRNAs with secondary stem-loop structures, and multiple miRNAs independent of Drosha, but dependent on another RNase activity and Dicer, were identified in several lineages. Thus, Drosha and at least one other related enzyme have multiple activities in post-transcriptional regulation of gene expression.
In our studies of thymocyte development, we identified the transcription factor RORγt as a regulator of the life span of DP cells. RORγt is required for the development of lymphoid tissue inducer (LTi) cells that, in turn, are essential for morphogenesis of lymph nodes and gut-associated lymphoid tissues. We found that RORγt is expressed in a subset of T helper cells that express the proinflammatory cytokines interleukin-17 (IL-17) and IL-22 and are referred to as Th17 cells. These cells are enriched in the mucosa of the small intestine and require specific components of the commensal bacterial flora for their differentiation. In mice defective for RORγt, Th17 cells and other lymphoid cells that produce IL-17 and IL-22 are absent, and the animals are resistant to multiple models of autoimmune disease, including autoimmune encephalomyelitis (EAE), a mouse disease that resembles multiple sclerosis, and inflammatory bowel disease.
RORγt expression is induced when T helper cells are exposed to either IL-6 or transforming growth factor β (TGFβ), the two cytokines that, in combination, direct the differentiation of Th17 cells. We found that although RORγt is sufficient to induce expression of Th17-lineage cytokines, its function is blocked by the transcription factor Foxp3, which is also up-regulated by TGFβ and directs the differentiation of Tregs. IL-6 and other inflammatory cytokines (e.g., IL-23) overcome the inhibitory effect of Foxp3, thus favoring differentiation of the Th17 lineage. In the intestine, Th17 cells and Tregs appear to differentiate from common precursors, and their proportions depend on the composition of the commensal microbiota and, presumably, the balance of cytokines, including TGFβ, IL-6, and IL-23.
Germ-free mice lack Th17 cells in the intestinal mucosa, but they acquire such cells when they are colonized with segmented filamentous bacteria (SFB), a species of Gram-positive bacteria. Mice colonized with SFB have enhanced protection from enteropathogenic bacteria, but they are also more susceptible to a model of Th17-dependent spontaneous arthritis. Thus, specific bacteria such as SFB contribute to maintaining a balance between protective and pathogenic activities of proinflammatory T helper cells. Efforts are under way to identify SFB products and host signaling components required for the accumulation of intestinal Th17 cells.
RORγt is an orphan nuclear hormone receptor, and identification of a ligand and of small chemical compounds that modulate its activity as a transcriptional activator offers the possibility of manipulating Th17 cell differentiation and reducing severity of inflammation in autoimmune diseases. We have developed insect cell-based screens to identify genes required for RORγt transcriptional activity and small molecules that modulate activity. Compounds that specifically inhibit RORγt and not other nuclear receptors have been shown to prevent mouse and human Th17 cell differentiation in vitro and EAE in mice. In addition, we have characterized a biosynthetic pathway involved in the generation of a natural ligand for RORγt, and in collaboration with David Mangelsdorf (HHMI, University of Texas Southwestern Medical Center at Dallas) we have isolated candidate molecules with ligand activity and are elucidating their structure.
We are pursuing additional studies to characterize transcriptional and post-transcriptional regulatory networks involved in Th17 cell differentiation. In addition to RORγt, there are at least eight other transcription factors known to be required for induction of IL-17 or IL-22. By combining expression profiling in T cells that lack any one of these factors with genome-wide chromatin immunoprecipitation (RNA-seq and ChIP-seq), we aim to determine the targets for each transcription factor and characterize the transcriptional network for Th17 cells. These studies may help identify novel targets for modulation of inflammatory lymphocyte function.
Dendritic cells (DCs), which are the major innate sensors of microbial pathogens, present antigen to T cells while they also produce diverse cytokines that influence the effector function of the differentiating T cells. In the intestine, multiple DC subsets can be found, and these selectively induce the differentiation of Tregs or Th17 cells. We are developing approaches to deplete individual DC subsets and study their functions in vivo. Although DCs mediate innate immune defenses against mucosal pathogens, they are also exploited by pathogens during infection and dissemination in humans. For example, infection of T cells with HIV is substantially enhanced if the virus first interacts with DCs. HIV infects helper T cells through the interaction of its envelope glycoprotein with CD4 and the chemokine receptors CCR5 or CXCR4. When DCs are present, there is dramatic enhancement of viral infectivity in the absence of replication in the DCs. The enhanced infectivity requires the internalization of HIV into a specialized recycling vesicular compartment in DCs.
Characterization of the molecular mechanism by which DCs enhance HIV infectivity may result in new therapies to block viral transmission and latency. Our current efforts are therefore aimed at identifying DC host genes required for enhanced delivery of HIV to T cells, characterizing the basis for DC resistance to HIV infection, and testing the hypothesis that DCs can serve as a reservoir for nonreplicating virus in vivo.
In the course of studying DC resistance to HIV replication, we found that when the block is circumvented, DCs mount a strong innate response, producing type I interferon and acquiring potent activity for stimulating naive T cells. This innate response, which is unique to HIV-infected DCs, involves binding of newly synthesized HIV capsid protein to host cyclophilin A and subsequent phosphorylation of IRF3, which directly regulates interferon transcription. We are exploring the possibility that this DC-specific pathway could be exploited for development of therapeutic or prophylactic vaccines and that "elite controllers," HIV-infected individuals whose immune system holds the virus in check, may have naturally enhanced innate responses to the virus.
Grants from the National Institutes of Health provided support for the work on HIV entry and pathogenesis and on the RORγt ligand. Grants from the Sandler Program in Asthma Research and the National Multiple Sclerosis Society have supported the studies on Th17 and regulatory T cells in inflammation.
As of June 09, 2010