A major effort in this laboratory is directed towards understanding the function of CD1 molecules and the NK T cells that recognize them. CD1 molecules constitute a third family of antigen presenting molecules, but they have a specialized function: the presentation of glycolipid antigens. NK T cells are a distinct T lymphocyte sublineage with a number of unusual features, including auto reactivity for CD1d, a nearly identical or invariant T cell antigen receptor (TCR) alpha chain, and the ability to produce copious quantities of cytokines following TCR stimulation. There is evidence suggesting that NK T cells are important for immune regulation and tolerance, the prevention of autoimmune disease, host defense, and the surveillance for tumors. The immune response of NK T cells is highly stimulated by the marine sponge-derived glycosphingolipid, alpha-galactosyl ceramide (aGalCer, also known as KRN 7000), when presented by CD1d. The Kirin Pharmaceutical Research Corporation is developing aGalCer as a potential immunomodulatory agent. aGalCer, is not widely distributed in nature, however, and therefore is unlikely to be the natural ligand recognized by NK T cells. Nevertheless, this synthetic glycolipid provides an ideal system for the exploration of CD1-mediated responses. We are exploring the following issues:

1. What is the pathway for the development of the NK T cell sublineage?

We have shown that NKT cells develop in the thymus and they go through the typical CD4+ CD8+ or double positive stage. Nevertheless, mutation in a variety of genes affect NKT cell but not conventional T cell development, including mutations in the lymphotoxin b receptor and its down stream effectors. We now wish to determine the factors that direct the unique developmental pathway of NK T cells, and if recognition of a TCR agonist is required for the development of these cells. A variety of approaches, including in vitro organ culture, the use of CD1d tetramers for flow cytometry and enrichment of NK T cells, and the use of genetically engineered mouse strains, are being used to address this issue.

2. How do mature NK T cells regulate their cytokine synthesis?

We have found that NKT cells contain IL-4 and IFN cytokine mRNA before activation, and that a Th1 or Th2 pattern of cytokine production is difficult to induce in these cells. We are studying how signals not mediated through the TCR, such as toll-like receptor or cytokine receptor signals modulate IL-4 and IFN production by NKT cells. We are also determining how the communication of NKT cells with DC may influence a Th1 or Th2 polarization of the adaptive immune response, despite the presence of preformed cytokine mRNAs.

3. How are complexes of CD1d/aGalCer and TCRs formed?

To address this issue collaborators with expertise in organic chemistry are synthesizing novel aGalCer analogs for use in biochemical and immunology-based assays. Soluble versions of CD1d and the TCR have been produced and used in real time binding studies by surface plasmon resonance. We are also identifying natural, microbial antigens for NKT cells.

4. Pathways leading to glycolipid antigen processing and presentation.

We are determining how CD1d trafficks to lysosomal compartments to acquire the lipid antigens that stimulate NKT cells. Candidate molecules that interact with the CD1d cytoplasmic tail have been identified by yeast-two-hybrid screening and a proteomics based approach.

Mucosal Immune SystemColitis.

When sorted CD4+, CD45RBhigh T cells are transferred to immune deficient mice, the recipients develop a severe colitis between 4 - 10 weeks post transfer. Pathogenesis depends upon a normal intestinal flora, and it is mediated by the secretion of IFN gamma and TNF. Colitis can be prevented by co-transfer of CD4+ CD25+ regulatoyr T cells (Treg), through mechanisms that depend on the induction of IL-10 and TGFß synthesis.

We have studied the roles of ß7 integrins, Ig family costimulatory molecules, TNF family cytokines in the pathogenesis and protection from disease, and in each case we have obtained surprising results. First, although ß7 integrins, which are known to be important for mucosal homing, are required for the pathogenesis of CD4+, CD45RBhigh T cells, they are irrelevant for Treg function. Our data indicate that Treg prevent disease by acting in organized lymphoid tissues rather than in the intestine itself. Second, we found that immune deficient RAG-/- mice that lack either B7-1 or B7-2 develop disease more rapidly; disease does not occur in the absence of both B7 molecules. A partial decrease in co-stimulation therefore increases inflammation. Preliminary data indicate that the synergistic interaction of both B7-1 (CD80) and B7-2 (CD86) is required for the full level of IL-2 synthesis by activated T cells, which induces CTLA-4 expression, thereby down regulating T cell expansion and pathogenesis. We are now studying a genetic model of altered co-stimulation, transgenic mice that express a soluble B7-2 IgFc fusion protein. These mice spontaneously develop colitis, but in the context of a complete immune system. We are determining if the fusion protein acts as a costimulation antagonist, or by some other mechanism.

Third, we have studied the role of the TNF family cytokine LIGHT. Preliminary data indicate LIGHT has a pro-inflammatory effect when expressed by T cells. For example, transgenic mice that over express LIGHT develop multi-organ inflammation in the intestine. By contrast, it has an anti-inflammatory effect when expressed by DC. Transfer of CD4+, CD45RBhigh T cells into LIGHT-/- RAG-/- mice causes greatly accelerated disease. In vitro studies demonstrate that LIGHT deficient DC stimulate greatly increased levels of IFN release during T cell priming. We will investigate the receptors for LIGHT responsible for these effects, the cell types that express the receptors, and the mechanism by which LIGHT modulates the course of disease.

Our work is supported by NIH grants RO1 CA52511, RO1 AI40617, RO1 AI45053, RO1 AI65016, PO1 DK46763, U54 GM61502, and support for individual post doctoral fellows from a variety of agencies.

Gisen Kim, Ph.D., Postdoctoral Research Fellow

Crohn's disease and ulcerative colitis (inflammatory bowel disease: IBD) are chronic immune disorders that cause severe inflammation in the digestive tract. It is estimated that as many as one million Americans especially adolescents and young adults have IBD.

We have been investigating the pathogenesis of IBD using a transfer model of mouse colitis that resembles human Crohn's disease in the histological findings and the responsive cytokines that induce chronic inflammation. In this model, a reconstitution of immune deficient mice (Rag1-/- mice) with CD4+ T cells lacking a regulatory population, such as CD4+CD25+ T cells, induces chronic colitis by 6 to 10 weeks after the transfer. Our project has been focused on the activation and the regulation of donor na?e CD4+ T cells in the days immediately after transfer. We have found that a synergistic costimulatory signal by two B7 molecules is critical for the regulation of T cell proliferation and differentiation, as well as activation. Surprisingly, colitis induction was dramatically accelerated in the recipients that are deficient for either B7-1 or B7-2, although B7-1 and B7-2 double deficient recipients did not develop colitis. We concluded that an insufficient costimulation with a single costimulatory molecule failed to upregulate the regulatory molecule CTLA-4, and this caused a dysregulated proliferation and differentiation of the donor T cells. Costimulation by a single B7 molecule induced a comparable proliferation of na?e CD4+ T cells, despite a failure to upregulate IL-2 production, which is critical for upregulating CTLA-4.

The spontaneous colitis that develops in the B7-2 Fc Ig transgenic mice (B7-2 Fc Tg) is another model that we have been investigating to understand the importance of B7 costimulation in regulating CD4+ T cell dependent colitis. The B7-2 Fc soluble fusion protein that is expressed in this transgenic mouse line, under the control of apo-lipoprotein E promotor, causes chronic spontaneous colitis by the age of 10 weeks. A phenotypical analysis demonstrated marked activation of B cells and CD11b+ cells in the spleen and the mesenteric lymph nodes, as well as increased levels of serum IgA and IgM. Interestingly, B7-2 Tg mice bred to the B7-1/B7-2 double deficient background developed more severe colitis, and exhibited a higher disease penetrance, probably because of reduced numbers of CD4+CD25+ regulatory T cells. At the same time, severe colitis in the B7 deficient background suggests an agonistic effect of fusion protein against one of its receptors, CD28. Furthermore, colitis induction was dramatically accelerated in the CD4+CD45RBhigh T cell transfer model, either by injecting the B7-2 Fc fusion protein with the donor cells, or by using B7-2 Tg Rag deficient mice as recipients.

These two models provide strong tools to understand the basic mechanisms how B7 costimulatory molecules regulate T cell mediated autoimmunity, especially colitis, as well as to understand the roles of regulatory systems, such as activation induced regulation and CD4+CD25+ regulatory T cells, in maintaining the integrity of immune systems. Furthermore, these two models will be very useful for investigating the therapeutic applications to treat chronic severe colitis.