Professor, Departments of Cell Biology
Professor, Department of Microbiology & Immunology
Chanin Bldg., Room 407
Studies on the Structural Basis for T Cell Activation and Regulation
Destruction of invading pathogenic organisms and transformed cells occurs through the specific recognition of these threats by the adaptive immune system. In order to recognize such a diverse and constantly changing set of targets, the immune system must exhibit broad reactivity, yet to avoid destruction of self tissues, activation must be tightly regulated. Broad reactivity arises from generation of a polyclonal T-cell population by recombination of the T-cell receptor (TCR) protein to allow the recognition of diverse antigenic peptide-Major Histocompatibility complex (MHC) protein complexes. No single receptor protein, however, is responsible for regulation of the immune response, (costimulation) and instead arises from action of both inhibitory and stimulatory signals delivered through both the TCR and diverse coreceptors present within the immunological synapse.
The research program of the Nathenson lab has followed a simple philosophy, structure follows from function and conversely we can extend structural insights into a testable functional hypothesis. Historically we have used this guiding principle to study the role of MHC in T-cell activation. Biochemical studies using in vitro complex formation of MHC/peptide were used to evaluate the precise mechanism of peptide binding, and in conjunction with the x-ray structure, to determine the structural rules, which govern the selection and binding of such peptides to the MHC molecule. Further biochemical studies were carried out to start defining interactions that are important for recognizing the peptide and the MHC complex by the TCR.
Our initial studies on T cell regulation in collaboration with the Almo laboratory focused on the stimulatory coreceptor CD28 and the inhibitory coreceptor CTLA-4, which, despite their difference in function, bind to the same ligands B7-1 and B7-2. Structural studies suggested that CTLA-4 is capable of forming an extended oligomeric structure, which led to the hypothesis that formation of an extended lattice is necessary for CTLA-4 function but not for CD28. Further biophysical studies using FRET on CTLA-4 and heterodimeric mutants are currently being employed to test this mechanism.
Subsequently we have investigated additional members of the CD28 and B7 receptor families, most notably co inhibitory proteins PD-1 and PD-L1/2. Mutagenesis studies guided by the crystal structure of PD-1 allowed us to generate mutants which either have enhanced or disrupted binding to both the PD-L1 and PD-L2 ligands and more interestingly one mutant that differentially affects ligand binding. Studies with the Nosanchuk lab demonstrated that in the fungal disease Histoplasmosis, the PD-1 pathway is up regulated in order to evade the immune response. Anti PD-1 antibodies reverse the defect in immunity. The mutant PD-1 proteins provide not only reagents to elucidate the signaling pathway but also as a potential therapeutic.
We have recently determined the structures of a number of non-CD28/B7 members of the immunoglobulin and TNF superfamilies with costimulatory activities that have been localized or been postulated to function within the Immunological Synapse. The (SLAM) family includes homophilic and heterophilic receptors that modulate both adaptive and innate immune responses. These receptors share a common ectodomain organization: a membrane proximal IgC domain and a membrane distal IgV domain that is responsible for ligand recognition. We found that three members of the family, NTB-A. CD84, and LY-9 self-associate with a Kd in the nanomolar to sub-micromolar range. These data, in combination with previous reports, demonstrate that the SLAM family homophilic affinities span at least three orders of magnitude, and suggest that differences in the affinities may contribute to the distinct signaling behavior exhibited by the individual family members. These structural data also suggest that, like NTB-A, all SLAM family homophilic dimers adopt a highly kinked organization spanning an end-to-end distance of ~140 Å and are of particular interest in that this kinked geometry can be accommodated by either formation of the same cell (cis) or opposing cells (trans). We believe that the change from cis to trans binding can act as a biological switch to prevent signaling from occurring outside a fully formed immunological synapse.
The Tim family of receptors regulates effector CD4+ T cell functions, and polymorphisms are implicated in autoimmune and allergic diseases. The ectodomains of Tim family members are composed of a membrane distal IgV domain, which represents the ligand recognition domain, and membrane proximal mucin domains. Examination of the sequence identified four cysteines, which are invariant within the Tim receptor family and structural studies demonstrate they form two non-canonical disulfide bonds that support formation of a surface (“cleft”) not present in other IgSF superfamily members. Binding and mutagenesis studies demonstrate that this unique structural feature mediates a previously unidentified galectin-9-independent binding site. Biochemical and biophysical studies indicate that the unique structural features observed in Tim-3 are conserved among the entire Tim family.
GITRL (Glucocorticoid-induced TNF receptor) ligand, a recently identified member of the TNF superfamily, binds to its receptor GITR on both effector and regulatory T cells and generates positive costimulatory signals implicated in a wide range of T cell functions. Structural studies demonstrate that hGITRL ectodomain self-assembles into a homotrimer that differs from conventional TNF family members by adopting an atypical "open flower-like" assembly and, due too the decreases of interacting residues, hGITRL exhibits a relatively weak tendency to trimerize. Such a unique assembly behavior has direct implications for hGITRL:GITR signaling, as this dynamic equilibrium appears to reduce the strength of the signal transduced through the hGITRL:GITR pathway to a biologically optimal level, as opposed to the maximal achievable level.
As more components of the immunological synapse are discovered we will expand our program to not only look at individual receptor proteins but how the network of coreceptors within the immunological synapse interact to ultimately control immunity.