Kai W. Wucherpfennig - US grants

This project continues the general theme of the program project and contributes to the program by 1. identifying DQ8 binding peptides and 2. examining the peptide specificity of islet infiltrating T cells. The project has strong collaborative interactions with Project 1 (Structure of DQ8/peptide complexes) and with Project 2 (DQ8 transgenic mouse model). It has been very difficult to identify the MHC/peptide specificity of tissue infiltrating T cells because specific probes have not been available. The goal of this project is to generate such probes and to define the MHC/peptide specificity of T cells in early islet infiltrates. The approach is to express multivalent, soluble MHC class II molecules and to load these molecules with defined peptides. Multivalent molecules are used because the affinity of monovalent MHC/peptide complexes for the TCR is too low. Classical studies with IgG and IgM antibodies have demonstrated that the 'functional affinity' of antibody binding to antigen is greatly enhanced by multivalent attachment. HLA-DQ8 is associated with the highest risk for the development of insulin dependent diabetes. Multivalent DQ8 molecules are expressed in the Drosophila Schneider cell system using three different designs (DQ8-IgG and DQ8-IgM fusion proteins, DQ8/streptavidin tetramers). These designs differ by the number of MHC/peptide arms per complex (2, 4 and 10) and by the presence of effector domains (Fc portion of IgG or IgM which fix complement). We will evaluate which design is most effective 1. for the in situ detection of cells and 2. for the depletion of peptide specific T cells. Experiments with a human DQ8 restricted T cell clone will examine the affinity of these probes for the TCR as well as the mechanism(s) by which target cells are killed following attachment of multivalent DQ8/peptide complexes. Little is presently known about peptide binding to DQ8, despite its importance in the pathogenesis of diabetes. DQ8 binding peptides will be identified based on an analysis of the DQ8 peptide binding motif. T cell responses to these peptides will be examined in collaboration with Dr. Lipes. The peptide specificity of T cells in early islet infiltrates will be examined in NOD and in DQ8-transgenic NOD mice, using multivalent MHC/peptide complexes for in situ detection. In situ analysis with multivalent MHC/peptide complexes may prove to be a powerful approach for characterizing the specificity of T cells in early autoimmune lesions.

Activated but not resting T cells can cross the blood brain barrier and initiate a T cell mediated autoimmune response against the central nervous system myelin. Previous studies have demonstrated that human myelin basic protein (MBP) specific T cell clones are activated by viral peptides that are quite distinct in their primary sequence from the MBP peptide. The goal of this project is to examine the pathogenicity of such viral peptides and to define the structural basis for the recognition of viral peptides by MBP specific T cell receptors (TCRs). The encephalitogenicity of viral peptides will be examined for a conserved, immunodominant T cell epitope of human MBP (residues 88- 102) that induces experimental autoimmune encephalomyelitis in several well established animal models. The induction of CNS autoimmunity will be examined by transfer of virus specific T cell lines from virus- infected to naive animals and by immunization with viral peptides that activate MBP(88-102) specific T cells. The activation of human MBP specific T cells by virus infected cells will be examined, in an effort to determine which viruses activate MBP specific T cells from patients with multiple sclerosis and other inflammatory CNS diseases. The activation of autoreactive T cells by foreign peptides may be important for the induction of T cell mediated autoimmunity by microbial antigens. However, the sequence diversity of peptides that are recognized by autoreactive T cell clones is not known, since previous approaches have assumed that such microbial peptides would only represent minor variations of self-pep tide sequences. Peptides that are recognized by MBP specific TCRs will be sought using a peptide library in which all peptide positions, except two MHC anchor residues, will be randomized. To investigate the structural basis for the recognition of DR/peptide complexes by a TCR, soluble MBP specific TCRs will be expressed using an coli expression system that recently allowed the crystallization of a MHC class I/peptide/TCR complex. MBP specific TCRs will be co-crystallized with DR2/MBP and other cross-reactive peptides to examine the structural basis for TCR cross-reactivity by X-ray crystallography.

DESCRIPTION: (Adapted from the applicant's abstract) - The major histocompatibility complex (MHC) is an important susceptibility locus for many human autoimmune diseases. Susceptibility to pemphigus vulgaris (PV), an autoimmune disease of the skin, is strongly associated with MHC class II alleles that are rare in the general population. In PV patients, autoantibodies against a keratinocyte adhesion molecule, desmoglein 3, inhibit desmoglein 3-mediated kearatinocyte adhesion and induce severe blister formation. The mechanisms by which MHC class II genes confer susceptibility to this antibody-mediated autoimmune disease will be studied in a DQ1 transgenic mouse model and in patients with PV. The PV-associated DQ1 molecule differs from a common DQ1 subtype that does not confer susceptibility to PV only at a single position of the DQbeta chain (aspartic acid/valine at DQbeta 57). This disease-associated polymorphism introduces a negative charge into the DQ1 peptide binding site. The functional consequences of the DQ(beta) 57 polymorphism will be examined using soluble DQ1 molecules that have been expressed in Drosphila Schneider cells. Peptides from the desmoglein 3 autoantigen that bind to the PV-associated DQ1 molecule will be sought, and the pathogenicity of DQ1 peptide ligands will be assessed. Transgenic mice that express the PV-associated DQalpha and DQbeta genes (DQA1*0101, DQB1*0503) will be generated as an animal model for PV; the bareskin (Bsk) mutation will be introduced into these mice to allow visual assessment of skin lesions. To determine if susceptibility in this model faithfully replicates the genetics of human disease, mice that express a DQ1 subtype that is not associated with PV will also be generated. DQ1 transgenic mice will be used to define the DQ1 peptide ligands that induce skin autoimmunity and to determine which T cell population(s) induce the production of pathogenic autoantibodies by B cells. NK1.1+ T cells are of particular interest because they are the major source of IL-4 for the differentiation of naive T cells into Th2 cells. The relevance of findings in the transgenic mouse model for the pathogenesis of the human disease will be investigated. It will be determined whether peptide(s) that induce disease in the transgenic mouse model are recognized by T cells from PV patients and whether these T cells induce the production of desmoglein 3 autoantibodies. The five specific aims are designed to study the interaction of DQ1-bound peptides in DQ1 transgenic mice and by studies in PV patients. Aim 1 is "to generate transgenic mice that express PV associated DQ1 molecule". Mice will be generated which express DQA1*0101 and DQB1*0503 and control transgenic animals with DQA1*0101 and DQB1*0501. This will be done in animals that have had the MHC class II knocked out (Ab superscript o/o). In addition, the bareskin mutation (Bsk) will also be moved into these animals so that the lesions of pemphigus can be directly observed. The second specific aim is to "find the structural requirements for peptide binding by the PV-associated DQ1 molecule and to identify desmoglein 3 peptides that are bound by DQ1." Peptide expression libraries will be constructed and an effort will be made to identify a motif for the peptide sequence which binds to this particular class II molecule. Soluble molecules have been generated after molecular engineering binding of the alpha and beta chains to one another in place of the transmembrane domains. A third specific aim is "to examine the induction of autoimmunity by desmoglein 3 peptides and DQ1 transgenic mice. Specific aim four is "to examine the role of NK1.1+/- T cells and other T cell subsets in the induction of a Th2-mediated autoimmune response against desmoglein 3." Cell transfer experiments are planned, as well as the use of T cell receptor alpha chain knock out mice. The fifth aim is "to examine the relevance of findings in the transgenic mouse model for the pathogenesis of the human disease." Here desmoglein 3 specific B cell lines will be developed as well as T cell lines that recognize desmoglein 3 peptides as presented by DQ1. The author hopes to reconstruct a system in which T cell clones induce production of desmoglein 3 autoantibodies. A number of cytokines will be explored in this system.

DESCRIPTION (provided by applicant): This PPG was initiated in 1998 in response to a RFA for "Human Immunology Centers of Excellence" and has a well-defined, central focus on the mechanisms of antigen presentation and T cell recognition in human autoimmune diseases, in particular multiple sclerosis (MS) and type 1 diabetes (T1D). Antigen presentation is an integral component of every autoimmune disease process and thus represents an important scientific and clinical problem. The six investigators who come together in this PPG have highly complementary areas of expertise and have formed a cohesive, multidisciplinary program under the guidance of the Program Director. During the present funding period this group has been highly productive, documented by numerous joint publications that have had a significant impact on the field. The overarching hypothesis is that the development and progression of autoimmune diseases are controlled by specialized populations of antigen presenting cells that serve distinct roles in tolerance induction versus propagation of autoimmunity. There are four major themes that connect the individual projects of this PPG. A previously unrecognized population of lymph node stromal cells with functional similarities to thymic medullary epithelial cells induces peripheral tolerance by expression of a wide variety of peripheral tissue antigens, and the biology of this cell population and potential therapeutic applications will be investigated (Projects 1 &2). The role of self-reactive B cells as efficient antigen presenting cells for T cells with the same specificity will be examined in a new mouse model that resembles a severe subtype of MS. New technologies, including a nanowell technique for functional definition of B cell populations at a single cell level and fluorescent self-antigen tetramers, will be used to define the functional properties of self-reactive B cells in patients with MS and T1D (Projects 3 &4). Particular emphasis will be placed on definition of the antigen presentation mechanisms responsible for T cell differentiation into regulatory and effector T cell subsets (Projects 1-3) and on characterization of the unique recognition and signaling properties of self-reactive T cells isolated from patients with MS and T1D (Projects 1 &4). During the past two funding cycles, the program has already had an impact on the development of therapeutics, and the areas of investigation for the next funding period offer a significant number of new opportunities.

T cell receptor sequence analysis indicates that myelin specific T cells are expanded in multiple sclerosis (MS) patients, but adequate techniques for the enumerization of antigen-specific T cells in relationship to the disease process are not available. We hypothesize that the frequency of T cells specific for myelin-derived peptides has been greatly underestimated by limiting dilution techniques. Analysis of antigen-specific CD4+ T cells with sensitive techniques will have important implications for understanding the pathogenesis of MS and for developing adequate tools for immune monitoring. The analysis of CD8+ T cells in viral infections has been revolutionized by the creation of tetrameric forms of MHC class I/peptide complexes. Massive expansion of antigen-specific CD8+ T cells has been documented in both acute and chronic viral infections. The introduction of this technology led to the realization that the frequency of antigen specific CD8+ T cells in viral infections had been greatly underestimated. The generation of tetrameric forms of human MHC class II molecules may have a major impact on the investigation of human autoimmune disease. An expression system for soluble HLA-DR2 was previously developed for crystallization of HLA-DR2 with a bound peptide form human myelin basic protein. Based on this expression system, tetramers of HLA-DR2/peptide complexes have been generated with fluorescently labeled streptavidin. These molecules will be used to determine the frequency, activation state and cytokine profile of antigen specific T cells in the cerebrospinal fluid and blood of MS patients with the HLA-DR2 haplotype. The tetramer approach will also be used to generate multivalent TCRs using cDNAs derived from myelin specific T cell clones. MS plaque tissue will be stained with these multivalent TCRs to examine surface expression of HLA-DR2-bound myelin peptides by antigen presenting cells in the target organ.

This project focuses on the recognition of HLA-DR2/peptide complexes by myelin basic protein (MBP) specific T cell receptors (TCRs) that have been isolated from multiple sclerosis (MS) patients. It was previously thought that TCRs are highly specific for particular foreign or self-peptides. However, a number of studies performed over the past several years have demonstrated that there is a substantial degree of degeneracy in TCR recognition of MHC/peptide complexes. Microbial peptides can activate autoreactive T cell clones, despite having relatively little sequence homology with the self-peptide. The structural and functional basis of degenerate peptide recognition by TCRs will be examined in the proposed experiments. Combinatorial peptide libraries will be used to examine the sequence diversity of peptides that activate autoreactive T cell clones specific for myelin basic protein (Aim 1). MBP specific TCRs will be expressed as soluble proteins in the Baculovirus system and the binding affinity of these TCRs for different HLA-DR2/peptide complexes will be examined (Aim 2). A soluble MBP specific TCR will be co-crystallized with the HLA-DR2/MBP peptide complex and a crossreactive HLA-DR2/peptide complex. The structures of these TCR/HLA-DR2/peptide complexes will be determined by X-ray crystallography (Aim 3). Analysis of the structural requirements for peptide binding and TCR recognition may have important implications for understanding the pathogenesis of MS and for designing peptide-based therapies.

[unreadable] DESCRIPTION (provided by applicant): Despite significant progress in recent years in elucidating the intracellular biochemical consequences of T cell activation via the T cell receptor (TCR), the precise mechanism driving the initial communication of a signal across the membrane upon receptor ligation remains unknown. TCR clustering, coreceptor recruitment and conformational changes in the receptor have all been proposed to play important roles in signal initiation, but no widely accepted model has emerged. The structure and stoichiometry of the complete TCR/CD3 complex expressed at the T cell surface also remains an open question, and models involving a TCR valency of greater than one have been proposed. The goal of this project is to define the structural features that drive the assembly of the TCR/CD3 complex, and to examine the triggering mechanisms based on an understanding of the subunit interactions and the stoichiometry of the complex at the cell surface. An in vitro translation and assembly system will be one of the major tools for these studies, and the investigators have developed new methods for the preparation of ER microsomes that permit complex assembly experiments as well as approaches for isolation of key assembly intermediates. The preliminary data demonstrate unusual interactions between basic and acidic transmembrane residues of the TCR and associated CD3 chains. Assembly of each of the three CD3 dimers with the TCR requires two acidic CD3 and one basic TCR residue in the membrane, and in each case these three transmembrane residues appear to interact. The major hypothesis of this project is that these distinctive interactions among potentially charged transmembrane residues are critical not only in the assembly of the alpha-beta, gamma-delta and pre-TCR, but also in guiding the conformational changes that are induced upon receptor triggering. Specifically, these interaction points between the TCR and CD3 subunits in the membrane may be at the center of conformational movements of both extracellular and cytoplasmic domains that bring the signaling modules into close proximity upon receptor engagement. Such a conformational change may be induced by recruitment of TCR/CD3 complexes to lipid rafts, which have an increased thickness relative to the surrounding plasma membrane due to the presence of sphingolipids and cholesterol. [unreadable] [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): The MHC class II region is a principal susceptibility locus for multiple sclerosis (MS) and many other autoimmune diseases, indicating that peptide presentation to CD4 T cells is critical in the pathogenesis. The major goal of this project is develop novel strategies for the treatment of MS and other autoimmune diseases that target the MHC class II antigen presentation pathway. This program is a collaborative effort between the PI's lab and the Harvard Center for Neurodegeneration and Repair (HCNR) that has established a drug discovery program within our academic environment. During the previous funding period, we developed a novel real time peptide binding assay and identified several groups of small molecules that modulate peptide binding by MHC class II. The most interesting group of compounds is represented by a small molecule termed J10 that accelerates peptide loading more than 70-fold. This small molecule has functional similarities with HLA-DM, the protein that catalyzes loading of peptides onto MHC class II molecules in a late endosomal compartment. MHC class II-based therapeutics are inefficiently loaded onto MHC molecules because they are exposed to endosomal proteases for extended periods of time before they reach the HLA-DM compartment. J10 is active over a broad pH range, indicating that it may enable MHC class II binding of therapeutics in early endosomal compartments or at the cell surface. New data demonstrate that a J10 derivative is active in vivo and that it substantially enhances peptide display. Furthermore, covalent attachment of the J10 catalyst to the C-terminus of peptides substantially increases the efficiency of peptide loading both in vitro and in vivo. Our hypothesis is that the efficacy of MHC class II-based therapeutics can be substantially enhanced with a small molecule that catalyzes rapid binding to MHC class II molecules. The goals for the next funding period are to define their precise mechanisms of action and to test therapeutic applications in humanized mouse models of MS. In Aim 1, we will investigate the structural and functional mechanisms by which these small molecules accelerate peptide loading. We will map the J10 binding site by site-directed mutagenesis of candidate regions and co-crystallize J10 with DR/peptide complexes for structure determination by X-ray crystallography. In Aim 2, we will examine whether the therapeutic efficacy of three different classes of MHC class II-based therapeutics (Copaxone, tolerogenic self-peptides, MHC class II inhibitors) can be increased by co- administration or covalent attachment of J10. The medicinal chemistry component of this program will continue to synthesize derivatives with improved activity and drug-like properties for these. PUBLIC HEALTH RELEVANCE A number of therapeutic approaches for the treatment of autoimmune diseases require binding of the therapeutic to MHC class II molecules. A major issue is that loading of these compounds into the MHC class II binding site is inefficient because the natural catalyst that accelerates peptide loading is confined to a late endosomal compartment. We have discovered small molecules that substantially accelerate peptide binding to MHC class II molecules and will now examine their mechanism of action as well as therapeutic applications in humanized mouse models of MS.

[unreadable] DESCRIPTION (provided by applicant): Antigen presentation by MHC class II molecules represents a key step in the development of T cell mediated autoimmune diseases. The peptide repertoire presented to CD4 T cells is shaped by the action of HLA-DM, a protein with an MHC-like fold that acts as an enzyme which accelerates peptide exchange. The major goals of the project are to define the mechanism by which DM catalyzes peptide exchange and to examine whether low affinity peptides involved in autoimmune processes can be selectively removed by increasing the editing function of DM. We have identified four small molecules that accelerate DM-catalyzed peptide exchange and increase the activity of this enzyme. These small molecules represent unique tools to examine relevant structural and biological properties of DM. The focus of Aim 1 is to determine the structural basis of DM-catalyzed peptide exchange. We aim to determine the structure of DM with one of the small molecule accelerators in order to define the small molecule binding site and the conformational change that increases the activity of the enzyme. This structural information will be used to design functional experiments on the catalytic mechanism. We also aim to crystallize the complex of DM with DR/peptide and will examine whether the small molecule accelerators promote the formation of crystals suitable for structure determination by shifting the equilibrium to an active conformation of DM. The focus of Aim 2 is to determine whether low affinity peptides can be selectively removed from MHC class II molecules by increasing the editing function of DM with small molecule accelerators. This question is relevant because T cells specific for low affinity self-peptides can escape tolerance induction and cause autoimmune diseases. For a systematic assessment of the peptide repertoire, we will analyze peptides eluted from MHC class II molecules by mass spectrometry and determine the binding properties of peptides whose abundance is significantly affected by the small molecules. In the B10.PL mouse model of multiple sclerosis, disease is mediated by CD4 T cells specific for the low affinity Ac(1-11 ) peptide of myelin basic protein (MBP) since T cells specific for high affinity determinants are subjected to central and peripheral tolerance mechanisms. T cell experiments will assess whether small molecule accelerators of DM prevent presentation of this self peptide to CD4 T cells. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): This project focuses on the recognition of HLA-DR2/peptide complexes by myelin basic protein (MBP) specific T cell receptors (TCRs) that have been isolated from multiple sclerosis (MS) patients. It was previously thought that TCRs are highly specific for particular microbial or self-peptides. However, analysis of human MBP specific TCRs demonstrated a substantial degree of crossreactivity in TCR recognition of MHC-bound peptides. As a result, microbial peptides can activate these autoreactive T cell clones even though they only have limited sequence similarity with the self-peptide. The structural basis and biological significance of TCR crossreactivity will be examined in the proposed experiments. The goal of Aim 1 is to determine the three-dimensional structure of a human TCR bound to the HLA-DR2/MBP peptide complex, and to compare this structure to a complex with a crossreactive microbial peptide instead of the MBP peptide. The structural mechanisms of TCR crossreactivity will be further probed in mutagenesis experiments based on the structural data. The goal of Aim 2 is to define the extent and in vivo significance of TCR crossreactivity. The total number of peptides that can be recognized by MBP-specific T cell clones will be defined with peptide libraries in which each bead carries a distinct sequence. The in vivo activity of microbial peptides and the corresponding microbial proteins will be tested in a humanized mouse model of MS. This model is based on expression of HLA-DR2 and the human TCR that is the subject of the structural studies. Definition of the structural and functional basis of TCR crossreactivity has important implications for understanding the pathogenesis of MS and other chronic inflammatory diseases.

Anodes; Autoimmune Diseases; CRISP; Cell Communication and Signaling; Cell Signaling; Central Nervous System; Chronic; Complex; Computer Retrieval of Information on Scientific Projects Database; Data; Demyelinations; Disease; Disorder; Funding; Genes, Class II; Genes, HLA Class II; Genes, MHC Class II; Grant; HLA-DR2; HLA-DR2 Antigen; Human; Human, General; INFLM; Immune; Inflammation; Institution; Intracellular Communication and Signaling; Investigators; Ligands; MHC Class II; MHC Class II Genes; MHC Receptor; MS (Multiple Sclerosis); Major Histocompatibility Complex Receptor; Man (Taxonomy); Man, Modern; Mice, Transgenic; Multiple Sclerosis; NIH; National Institutes of Health; National Institutes of Health (U.S.); Nervous System, CNS; Neuraxis; Peptides; Proteins; Reaction; Receptor Protein; Receptors, Antigen, T-Cell; Research; Research Personnel; Research Resources; Researchers; Resolution; Resources; Sclerosis, Disseminated; Signal Transduction; Signal Transduction Systems; Signaling; Source; Structure; T-Cell Receptor; T-Cells; T-Lymphocyte; Thymus-Dependent Lymphocytes; Time; Transgenic Mice; United States National Institutes of Health; Visit; autoimmune disorder; base; beamline; biological signal transduction; detector; disease/disorder; gene product; improved; insular sclerosis; receptor; thymus derived lymphocyte

ATGN; Acute; Affinity; Anabolism; Animal Model; Animal Models and Related Studies; Antibodies; Antibody Affinity; Antigen Presentation; Antigens; Assay; Autoantibodies; Autoantigens; Autoimmune Diseases; Autoimmune Status; Autoimmunity; Autologous Antigens; B blood cells; B cell receptor; B-Cells; B-Lymphocytes; Binding; Binding (Molecular Function); Binding Sites; Bioassay; Biologic Assays; Biological Assay; Blood Serum; Bursa-Dependent Lymphocytes; Bursa-Equivalent Lymphocyte; CD4 Positive T Lymphocytes; CD4 T cells; CD4 lymphocyte; CD4+ T cell; CD4+ T-Lymphocyte; CD4-Positive Lymphocytes; CLIP peptide; Capsid; Cell Communication and Signaling; Cell Signaling; Cells; Cells, CD4; Childhood; Chimera Protein; Chimeric Proteins; Class; Class II Antigens; Class II Major Histocompatibility Antigens; Clinical; Combining Site; Complex; Consensus; Demyelinating Diseases; Demyelinating Disorders; Detection; Development; Diabetes Mellitus, Brittle; Diabetes Mellitus, Insulin-Dependent; Diabetes Mellitus, Juvenile-Onset; Diabetes Mellitus, Ketosis-Prone; Diabetes Mellitus, Sudden-Onset; Diabetes Mellitus, Type 1; Diabetes Mellitus, Type I; EAE; Encephalomyelitis; Encephalomyelitis, Allergic; Environment; Experimental Allergic Encephalitis; Experimental Allergic Encephalomyelitis; Experimental Autoimmune Encephalitis; Experimental Autoimmune Encephalomyelitis; Funding; Fusion Protein; Generations; Genes, Class II; Genes, HLA Class II; Genes, MHC Class II; Histocompatibility Antigens Class II; Human; Human, General; I-A Antigen; IDD; IDDM; Ia Antigens; Ia-Like Antigens; Imagery; Immune Response Antigens; Immune-Response-Associated Antigens; Immunoassay; Immunologic Monitoring; Immunosurveillance; In Vitro; Individual; Insulin-Dependent Diabetes Mellitus; Intracellular Communication and Signaling; Investigators; Knock-in; Knock-in Mouse; Label; Ligands; Lymph node proper; MAG Protein; MHC Class II; MHC Class II Genes; MHC Class II Molecule; MHC Class II Protein; MHC class II antigen; MOG glycoprotein; Major Histocompatibility Complex Class II; Mammals, Mice; Man (Taxonomy); Man, Modern; Methods; Methods and Techniques; Methods, Other; Mice; Microsomes; Modeling; Molecular Interaction; Monitoring, Immune; Monitoring, Immunologic; Monitoring, Immunological; Murine; Mus; Myelin; Myelin Associated Glycoprotein; Myeloencephalitis; Numbers; Patients; Peptides; Pleural Glycoproteins; Population; Preparation; Procedures; Production; Programs (PT); Programs [Publication Type]; Proteins; R01 Mechanism; R01 Program; RPG; Radio; Radiolabeled; Reactive Site; Reagent; Receptors, Antigen, B-Cell; Recovery; Reporting; Research Grants; Research Personnel; Research Project Grants; Research Projects; Research Projects, R-Series; Researchers; Reticuloendothelial System, Lymph Node; Sampling; Self-Antigens; Serum; Signal Transduction; Signal Transduction Systems; Signaling; Solutions; Staging; Strepavidin; Streptavidin; Stromal Cells; Structure; System; System, LOINC Axis 4; T-Cells; T-Lymphocyte; T1 diabetes; T1D; T1DM; T4 Cells; T4 Lymphocytes; Techniques; Technology; Thymus-Dependent Lymphocytes; Time; Translations; Type 1 diabetes; Viral; Visualization; Work; antigen antibody affinity; autoimmune antibody; autoimmune disorder; autoimmune encephalomyelitis; autoreactive B cell; base; biological signal transduction; biosynthesis; coat (nonenveloped virus); experiment; experimental research; experimental study; gene product; helper T cell; human disease; immunogen; insulin dependent diabetes; interest; juvenile diabetes; juvenile diabetes mellitus; ketosis prone diabetes; lymph gland; lymph nodes; model organism; mouse model; myelin glycoprotein; myelin oligodendrocyte glycoprotein; new approaches; novel; novel approaches; novel strategies; novel strategy; oligodendrocyte-myelin glycoprotein; pediatric; programs; radiolabel; radiotracer; research study; self reactive antibody; self recognition (immune); thymus derived lymphocyte; tool; type I diabetes

Dr. Wucherpfennig will continue to be responsible for coordinating the research efforts among all participating labs at Harvard Medical School, the Massachusetts Institute of Technology and New York University, and for administrative oversight of the entire program. He has worked closely with all investigators involved in this PPG during the current funding period, investigators who are presently funded by this PPG (Hafler and Ploegh) and members who joined the PPG for this renewal (Kuchroo and Turley). The close interactions among the laboratories are facilitated by the close involvement of Drs. Wucherpfennig, Turley, Kuchroo and Hafler in the Immunology Program at Harvard Medical School which the program members serve in many different capacities (i.e. Kuchroo & Wucherpfennig - seminar committee, Hafler & Wucherpfennig - executive committee, Turley & Wucherpfennig - graduate committee). Dr. Ploegh was until recently Director of the Immunology Graduate Program at Harvard Medical School and remains closely involved in the activities of the program. The many informal interactions among the investigators greatly facilitate joint research efforts. Dr. Wucherpfennig will continue to collaborate with all members of this PPG and will identify new opportunities for joint research activities. These will be discussed at joint meetings, which in the past have emphasized the application of new technologies to the investigation of autoimmune diseases in both animal models and patients, as evidenced by the joint development of novel approaches by Drs. Wucherpfennig & Hafler (antigen tetramers) and Drs. Ploegh & Hafler (nanowell technology for investigation of B cell function). Dr. Wucherpfennig will also continue to be in charge of all administrative aspects, and will be assisted by his administrator (Kimberly Foemmel) and Dana-Farber administrators who specialize in grants management (Ms. Jane Macdonald) and financial management (Ms. Linda Shehu). No funding is requested for these Dana-Farber administrators.

APC; ATGN; Affect; Affinity; Antigen Presentation; Antigen Targeting; Antigen-Presenting Cells; Antigens; Antirejection Therapy; Autoantigens; Autoimmune; Autoimmune Diseases; Autoimmune Process; Autoimmune Status; Autoimmunity; Autologous Antigens; Binding; Binding (Molecular Function); Binding Sites; Biochemical; Body Tissues; CD4 Positive T Lymphocytes; CD4 T cells; CD4 lymphocyte; CD4+ T cell; CD4+ T-Lymphocyte; CD4-Positive Lymphocytes; CLD; Cell Communication and Signaling; Cell Signaling; Cells; Cells, CD4; Central Nervous System; Chimera Protein; Chimeric Proteins; Chromosome Pairing; Collaborations; Combining Site; Complex; DRA; Data; Deep; Dephosphorylation; Depth; Development; Disease; Disorder; Dose; EAE; Elements; Encephalomyelitis, Allergic; Epithelial Cells; Event; Experimental Allergic Encephalitis; Experimental Allergic Encephalomyelitis; Experimental Autoimmune Encephalitis; Experimental Autoimmune Encephalomyelitis; Funding; Fusion Protein; Goals; HLA-DR2; HLA-DR2 Antigen; Human; Human, General; Hydrogen Bonding; Image; Immunologic Accessory Cells; Immunosuppression Effect; Immunosuppressions (Physiology); Immunosuppressive Effect; Immunosuppressive Therapy; In Vitro; Incidence; Influenza HA; Influenza Hemagglutinin; Intracellular Communication and Signaling; Investigation; Kinetic; Kinetics; L-Tyrosine; Lead; Lentivirinae; Lentivirus; Ligands; Lymph node proper; MHC binding peptide; MVB; Mammals, Mice; Man (Taxonomy); Man, Modern; Mass Spectrum; Mass Spectrum Analysis; Membrane Proteins; Membrane-Associated Proteins; Methods and Techniques; Methods, Other; Mice; Mice, Transgenic; Modeling; Molecular Interaction; Monocytes / Macrophages / APC; Multivesicular Body; Murine; Mus; Myelin; N-terminal; NH2-terminal; Natural immunosuppression; Nervous System, CNS; Neuraxis; Numbers; Pathogenicity; Pathway interactions; Patients; Pb element; Peptide antibodies; Peptide-MHC; Peptide-Major Histocompatibility Protein Complex; Peptide/MHC Complex; Peptides; Peripheral; Personal Satisfaction; Photometry/Spectrum Analysis, Mass; Population; Programs (PT); Programs [Publication Type]; Proliferating; Property; Property, LOINC Axis 2; Protein Degradation, Metabolic; Protein Degradation, Regulatory; Protein Dephosphorylation; Protein Turnover; Proteins; Reactive Site; Relapse; Reticuloendothelial System, Lymph Node; Reticuloendothelial System, Thymus; Role; SLC26A3; SLC26A3 gene; Self-Antigens; Signal Pathway; Signal Transduction; Signal Transduction Systems; Signaling; Signaling Molecule; Signaling Protein; Site; Spectrometry, Mass; Spectroscopy, Mass; Spectrum Analyses, Mass; Spectrum Analysis, Mass; Stromal Cells; Structure; Subfamily lentivirinae; Surface; Surface Proteins; Synapses; Synapsis; Synapsis, Chromosomal; Synaptic; T-Cell Activation; T-Cell Receptor Interaction; T-Cell Subsets; T-Cells; T-Lymphocyte; T-Lymphocyte Subsets; T4 Cells; T4 Lymphocytes; TCR Activation; TCR Interaction; TYR; Techniques; Therapeutic immunosuppression; Therapy, Anti-Rejection; Thymus; Thymus Gland; Thymus Proper; Thymus-Dependent Lymphocytes; Tissues; Transgenic Mice; Transgenic Organisms; Tyrosine; Tyrosine, L-isomer; Viral; Virus-Lenti; Work; accessory cell; anti-microbial; antimicrobial; artificial immunosuppression; autoimmune disorder; autoimmune encephalomyelitis; base; biological signal transduction; calcium flux; calcium mobilization; density; disease/disorder; experiment; experimental research; experimental study; gene product; heavy metal Pb; heavy metal lead; helper T cell; imaging; immunogen; immunological synapse; immunosuppression; in vivo; lymph gland; lymph nodes; mouse model; novel; pMHC; para-Tyrosine; pathway; programs; protein degradation; release of sequestered calcium ion into cytoplasm; research study; response; self recognition (immune); site targeted delivery; social role; synapse formation; synaptogenesis; targeted delivery; thymus derived lymphocyte; transcription factor; transgenic; ubiquitin ligase; well-being

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. The goal of the project is to determine the mechanisms of peptide loading onto human MHC class II molecules. MHC class II molecules bind peptides very tightly, with half-lives of several days to weeks. Empty MHC class II molecules have a tendency to aggregate and they assemble in the ER with invariant chain, which occupies the binding groove like a peptide. In the endosomal peptide loading compartment, invariant chain is degraded by a set of proteases, leaving the CLIP peptide of invariant chain in the groove. Removal of the CLIP peptide requires the action of HLA-DM and makes the binding groove vacant for binding of microbial peptides. We have defined the molecular requirements for the interaction of MHC class II molecules with HLA-DM. We found that HLA-DM binding required release of the N-terminal part of the peptide. Based on this insight, we are now attempting to characterize this interaction at a structural level.

DESCRIPTION (provided by applicant): Recent work has convincingly shown that the degree of T cell infiltration plays a critical role in the natural progression of many cancers. A landmak study found that the type, density, and location of cytotoxic T cells within tumors enabled better prediction of patient survival than histopathological methods currently used to stage cancer. However, the tumor microenvironment strongly inhibits expansion and effector functions of tumor-specific T cells. The goal of this project is to identify novel targes for therapy by pinpointing the key genetic and biochemical defects within tumor-infiltrating T cells that restrain their effector function. The identification of such therapeutic targts is of broad relevance in oncology because T cell mediated immune responses have the potential to eradicate cancers. Towards this goal, we propose a novel in vivo shRN discovery approach that enables identification of critical genes and pathways in the relevant microenvironment. Our hypothesis is that shRNAs which target critical inhibitors within dysfunctional T cells can reprogram them to undergo substantial expansion in tumors. T cells will be genetically modified with shRNA pools and then transferred into tumor-bearing mice so that enrichment of particular shRNAs within tumors can be quantified by Illumina sequencing of the shRNA cassette. This in vivo approach will also allow us to address a second related problem in oncology, the identification of combination therapies that act in a highly synergistic manner on defined cellular pathways. We will approach this issue using a lentiviral vector with two shRNA cloning sites, so that an active shRNA can be tested for synergy against a pool of shRNAs. We will determine which shRNA combinations optimize T cell activityin vivo in terms of proliferation, cytokine production and anti-tumor cytotoxic actio. These therapeutic approaches will be tested in a mouse model in which melanomas spontaneously develop based on genetic lesions found in the human disease. A central goal of this effort is to translate these discoveries into clinical application through collaboration with a clinical investigator with expertise in adoptive T cell therapy. Adoptive transfer of T cells that express 'chimeric antigen receptors' (CARs) has emerged as a promising approach because the antibody-like extracellular domain of a CARbinds with high affinity to a surface molecule on tumor cells, while the cytoplasmic domain induces T cell activation. Co-expression of shRNAs and CARs through the same lentiviral vector into T cells could greatly enhance T cell survival and expansion within tumors. This approach will be tested using human T cells in mice bearing human melanomas, as an important step towards clinical translation.

DESCRIPTION (provided by applicant): The MHC locus contributes significantly to the genetic risk for type 1diabetes (T1D), but the molecular mechanisms are not well understood. The central problem is that current experimental methods for characterization of self-reactive T cell populations are highly inadequate. These approaches, including tetramer labeling, intracellular cytokine staining (ICS) and ELISpot assays, enable sensitive detection of high- affinity microbe-specific T cell populations but are suboptimal for self-reactive T cells which tend to have low affinities for their peptide-MHC ligands. We will use a novel single-cell technology that enables sensitive detection of self-reactive T cells and generates a comprehensive body of data on surface phenotype, cytokine release and other functions, such as proliferation and cytotoxicity. A dense, elastomeric array of wells with subnanoliter volumes (nanowells) is generated by replica molding, and individual T cells are co-cultured with autologous mature dendritic cells pulsed with antigen. Many different cytokines are captured on a glass slide and quantified on a microarray scanner, while CD8 T cell cytotoxicity is quantified by imaging lysis of co- cultured target cells. T cells of interest can also be isolated for subsequent clonal expansion. This system provides a rapid and high-throughput method for ex vivo characterization of lymphocytes. Preliminary data show that this approach greatly increases the sensitivity of detection for self-reactive T cells and enables comprehensive assessment of their ex vivo functions. We will use this novel technology to address three fundamental questions on the mechanisms by which MHC genes confer susceptibility and resistance to T1D. First, it remains unknown whether there are important differences in CD4 and/or CD8 T cell functions in patients with distinct MHC haplotypes that confer different degrees of risk. We will compare cytokine patterns and CD8 T cell cytotoxicity in response to B cell antigens in patients who carry either DR3-DQ2 (DQ2) or DR4- DQ8 (DQ8) haplotypes or the highest risk DQ2/DQ8 haplotype. Distinct, yet complementary functions could account for the high risk conferred by heterozygosity for DQ2 and DQ8 haplotypes. Second, genetic data suggest that DQ trans-dimers (encoded in trans by different haplotypes) contribute to the high risk of patients with heterozygosity for DQ2/DQ8 haplotypes. We will directly test this hypothesis by cloning CD4 T cells from nanowells and testing their MHC-peptide specificity. Third, several haplotypes are known to confer dominant protection from T1D. Particularly important is the DR15-DQ6 haplotype because it reduces risk more than 30- fold and is common in populations with a high incidence of T1D. We will assess three possible mechanisms for dominant protection: epitope capture, deletion of particular effector T cell populations, or induction of b cell- specific T cels with regulatory functions that can control effector T cell responses. This highly novel approach will thus allow us to address central questions on the function of MHC genes in T1D. PUBLIC HEALTH RELEVANCE: This project focuses on the mechanisms for MHC-linked susceptibility to T1D. A novel single cell approach will be used to interrogate the function of ¿ cell specific CD4 and CD8 T cell populations in patients and normal subjects with different predisposing or protective MHC genotypes. These studies have the potential to impact the design of future prevention approaches and to improve prediction of T1D in susceptible populations.

Dr. Wucherpfennig has provided scientific and administrative leadership for this PPG since 1999, and this effort has been productive as evidenced by the large number of collaborative publications, many of which have been in high-impact journals. He will continue to closely interact with all members ofthe team to foster collaboration and identify new scientific opportunities. Dr. Wucherpfennig will organize regular formal scientific meetings at which recent progress will be presented and new opportunities for collaboration will be discussed. All project leaders work in the Boston area, Drs. Wucherpfennig, Turtey and Kuchroo on the Harvard Medical School (HMS) campus, and Dr. Love at the Koch Institute of MIT. Drs. Turley and Wucherpfennig are neighbors at the Dana-Farber Cancer Institute, and Drs. Kuchroo, Turtey and Wucherpfennig frequently meet at seminars and as part of the Program for Immunology at HMS. Drs. Kuchroo and Wucherpfennig also jointly teach an annual quarter course on Autoimmunity' for graduate students (since 1999). These frequent interactions foster dialogue and collaboration. In addition. Dr. Wucherpfennig will organize annual meetings with the Scientific Advisory Board (SAB) at which each project leader will present recent progress and discuss directions for the coming year. All members of our SAB work at HMS, and we will therefore also have many opportunities to consult them on an informal basis. Dr. Wucherpfennig will also continue to be in charge of all administrative aspects and will be assisted by this administrator and Dana-Farber grants and financial management specialists.

Genome-wide analyses of MS susceptibility loci have emphasized the importance of MHC class II genes, strongly implicating antigen presentation to CD4 T cells as a key process in the pathogenesis of MS. However, it has been very difficult to study myelin-specific CD4 T cells from MS patients ex vivo, due to low T cell receptor affinities for the relevant peptide-MHC complexes. Therefore, two fundamental questions remain unresolved: First, which molecular properties distinguish myelin-specific T cells in patients with MS from those in healthy individuals? Second, are there distinguishing properties of myelin-specific and virusspecific T cells that could be exploited for therapeutic gain? In Aim 1, we will study TCR recognition by myelin-specific T cells from MS patients. Imaging experiments identified alterations in immunological synapse formation by myelin-specific compared to virus-specific T cells, and structural studies demonstrated unusual binding topologies by some myelin-specific TCRs isolated from MS patients. The Garcia lab recently developed a yeast peptide-MHC display approach to identify entirely novel peptide ligands for a given TCR and showed that one ofthe peptides was recognized with a non-conventional TCR topology and failed to induce signaling. This means that TCR binding topology can have a profound effect on signaling. This approach will be used to examine the irnpact of TCR topology on the function of myelin-specific T cells and determine whether administration of such peptides can prevent spontaneous disease in human TCR/MHC transgenic mice. In Aim 2, we will collaborate with Drs. Love and Hafler to study myelin and virus-specific T cells from MS patients and control subjects identified with the nanowell device. Antigenresponsive T cells will be isolated from nanowells with a robotic device, which enables detailed molecular characterization of myelin-specific T cell populations. The transcriptome of myelin-specific T cells from MS patients and control subjects will be examined, with an emphasis on the expression of genes associated with susceptibility to MS, transcription factors and cytokine signaling molecules. Clonal expansion of myelinspecific will be assessed by sequencing of TCR chains, and the frequency of myelin-reactive TCR sequences in peripheral blood and CSF T cells will be determined by lllumina sequencing ofthe TCR repertoire. These studies will determine whether myelin-specific T cells with particular cytokine proflles and/or gene expression programs are preferentially expanded in MS patients compared to healthy subjects.

DESCRIPTION (provided by applicant): Antigen presentation is an integral component of every autoimmune disease process, and thus represents an important scientific and clinical problem. The seven investigators who come together in this PPG have highly complementary areas of expertise and have formed a cohesive, multidisciplinary program. The overarching hypothesis is that the development and progression of autoimmune diseases are controlled by specialized populations of antigen-presenting cells (APCs) that serve distinct roles in the induction of different effector and regulatory T cell programs. The team emphasizes direct investigation of APC - T cell interactions in patients with autoimmune diseases, in particular multiple sclerosis (MS), and integrates these human immunological studies with in-depth mechanistic studies in relevant animal models. During the previous funding period, the group developed a novel nanowell-based technology platform for multiplexed investigation of T cell function in autoimmune diseases. The technology enables co-culture of single T cells with mature dendritic cells in wells of subnanoliter volume for multi-dimensional characterization of cytokine secretion and surface markers. Furthermore, responding T cells can be recovered with a robotic device for characterization of transcriptional programs. This technique will be used by all investigators to examine the functional consequences of T cell interactions with distinct populations of APC. The team will address a long-standing challenge in the field and define the functional and molecular differences between self-reactive T cells in patients with MS and healthy subjects. Studies in MS CNS lesions and animal models will examine how the interaction of T cells with different populations of APCs results in the formation of chronic inflammatory microenvironments in the target organ. Of particular interest is the complex interplay between T cells, B cells and stromal cells that results in the formation of ectopic lymphoid follicles in the CNS. Studies during the previous funding period have shown that Th17 cells express podoplanin (PDPN), a surface molecule that interacts with CLEC-2 on B cells and mature dendritic cells. Antibody-based blockade of PDPN function prevents formation of ectopic lymphoid follicles, and the function of these molecules will now be studied in MS lesions and conditional knock-out mice. The program is highly synergistic based on our focus on an important problem in the autoimmunity field, and our highly collaborative approach integrates a unique team of investigators with expertise in molecular and cellular immunology, biophysics, and engineering to investigate disease mechanisms in autoimmunity with cutting-edge technologies.

DESCRIPTION (provided by applicant): The Major Histocompatibility Complex (MHC) represents the most important susceptibility locus for many common human autoimmune diseases, including type 1 diabetes, rheumatoid arthritis and multiple sclerosis, indicating that antigen presentation to CD4 T cells represents an important step in their pathogenesis. HLA-DM (DM) plays a central role in the MHC class II (MHCII) antigen presentation pathway: it induces dissociation of the invariant chain-derived CLIP peptide and edits the peptide repertoire, favoring presentation of high-affinity peptides. The central goals of this project are to define th mechanism of DM action at a structural level and to advance our understanding of self-antigen presentation in autoimmune diseases. Preliminary studies have shown that the interaction of DM and HLA-DR (DR) molecules is controlled by the occupancy state of the DR peptide binding groove. We found that DM cannot bind to DR proteins when the peptide binding groove is fully occupied by peptides. Rather, DM binds to an unstable transition state of DR proteins in which the peptide N-terminus has partially dissociated from the groove. We utilized this insight to develop a strategy for crystallization of the complex of DM bound to DR1. Aim 1 Since submission of the original proposal, we have been able to determine the crystal structure of the DR1-DM complex at a resolution of 2.6Å. The structure suggests a novel mechanism for rapid selection of high-affinity peptides (also referred to as 'editing'). In the DM-bound state, three D residues have moved into a critical part of the groove (P1 pocket and P2 site), rendering it initially inaccessible to peptides. Peptides need to compete for access to these sites, and this energetic barrier drives selection of peptides with the highest affinities. We have also recently developed a novel approach to measure rapid binding of peptides to empty DR1-DM complexes. This technique will now be used to define critical mechanistic steps in editing of the peptide repertoire by DM. Aim 2 is based on novel observations from the DR1-DM structure. HLA-DQ8 (DQ8) confers susceptibility to type 1 diabetes while HLA-DQ6 (DQ6) induces dominant protection. The DR1-DM structure shows that a substantial number of residues located at the DR1-DM interface are polymorphic in DQ8 and DQ6. Furthermore, preliminary data show substantial differences in the peptide binding properties of DQ8 and DQ6. We will use a series of functional approaches to dissect the impact of these DQ polymorphisms on peptide presentation by DQ8 and DQ6. Also, the novel strategy that enabled crystallization of the DR1-DM complex will be used for crystallization trials of a DQ-DM complex. We anticipate that these studies will have a significant impact on our understanding of DM function and MHCII antigen presentation in autoimmune diseases.

Project Summary This training program is designed to train the next generation of scientists in Cancer Immunology at the Dana- Farber/Harvard Cancer Center, a Harvard-wide, NCI-designated Comprehensive Cancer Center. Recent clinical trials have demonstrated that cancer immunotherapy can induce durable responses in patients with diverse types of advanced cancer. The discovery that targeting of a single inhibitory receptor on immune cells can unleash durable anti-tumor immunity has far-reaching implications for oncology and raises many fascinating questions about the biological mechanisms that regulate anti-tumor immunity. The resulting rapid growth in the field of cancer immunology has created a significant need for scientists with in depth expertise at the interface of immunology and tumor biology. To address this need, we have built a comprehensive program supported by faculty who bring a wide spectrum of expertise in basic and translational cancer immunology. The preceptors of this grant have made a series of important contributions to cancer immunology and immunotherapy. We propose to appoint 2 predoctoral and 6 postdoctoral scientists per year. Our predoctoral trainees will be selected from a pool of students who have enrolled in the Immunology and Basic Biomedical Sciences Graduate Programs at Harvard Medical School. These students will be appointed after their second year of study once they have chosen one of our mentors as their thesis advisor and committed to a research topic in cancer immunology. The postdoctoral positions will be awarded to a select group of recent recipients of PhD and MD/PhD degrees and support their training in cancer immunology. Laboratory training in cancer immunology will be complemented by a didactic program that gives students the required background in cancer immunology and prepares them for independent careers in academic research, biotechnology or a broad range of other job opportunities. The program directors are highly committed to teaching and mentoring and have the required expertise to lead this effort. The Postdoctoral and Graduate Student Affairs Offices at Dana-Farber and Harvard Medical School provide a wide range of programs to all of our trainees designed to enhance their training experience and address individual needs. Retreats organized by the Harvard Immunology Program and the Department of Cancer Immunology and Virology at Dana-Farber will help to create a sense of community and give our trainees the opportunity to learn from other trainees and faculty.

PROJECT SUMMARY Despite enormous success in treating several types of cancer, immune checkpoint blocker (ICB) therapy still only shows efficacies in a subset of patients. Identifying novel regulators of immunotherapy response as well as improving the response rate of cancer immunotherapies remain open questions. Recently, we used CRISPR screens in mouse models to investigate T-cell infiltration, proliferation, and killing efficacy, and identified PBAF of the SWI/SNF chromatin remodeling complex as one novel regulator of T-cell mediated cytotoxicity. We also developed a computational model, TIDE, to identify gene signatures of CD8 T-cell dysfunction in immune hot tumors and T-cell exclusion in immune cold tumors. The resulting signatures, computed from tumor profiles in non-immunotherapy setting, show promising results in predicting melanoma and lung cancer patient response to immune checkpoint blockade based on pre-treatment tumor expression profiles. This proposed project aims to improve the TIDE biomarkers, identify novel regulators, and elucidate their mechanisms underlying ICB response. In Aim 1, we will develop machine learning approaches on large collection of clinical tumor transcriptome profiles from non-ICB settings to refine the TIDE predictive biomarker of ICB response, and develop a web server to comprehensively evaluate different ICB response biomarkers in all the available ICB cohorts. In Aim 2, we will conduct in vivo CRISPR screens in mouse syngeneic tumor models to identify cancer-cell intrinsic regulators of ICB response, which can serve as novel targets to improve ICB response. In Aim 3, we will elucidate the mechanism underlying two novel regulators of ICB response and characterize their effects on the tumor immune microenvironment using single-cell RNA-seq, single-cell ATAC- seq, and computational modeling. Our investigative team has combined expertise in computational methodology immunotherapy. immunology and big data mining, functional genomics profiling Our proposed studies, if successfully executed, and translational benefits to cancer immunotherapy. and screening, cancer immunology and could provide new insights into cancer

Project Summary MICA and MICB (MICA/B) are stress proteins that are frequently expressed by diverse types of human cancer as a consequence of genomic damage, but are rarely expressed by healthy cells. MICA/B serve as ligands for the NKG2D receptor expressed by all cytotoxic lymphocytes, including CD8 T cells, ?? T cells, NKT cells and NK cells, enabling recognition and elimination of stressed and transformed cells. Proteolytic shedding of MICA/B is a major immune evasion mechanism from NKG2D-mediated tumor immunity in many human cancers. This shedding process involves unfolding of the MICA/B ?3 domain by the action of the disulfide isomerase ERp5 which enables MICA/B cleavage by proteases belonging to the ADAM and MMP families. It is not feasible to inhibit shedding in vivo with small molecule inhibitors because the relevant proteases have broad substrate specificities. We developed an approach to inhibit MICA/B shedding by designing antibodies that sterically block the shedding site in the MICA/B ?3 domain. These antibodies potently inhibit MICA/B shedding by a diverse panel of human cancer cell lines and thereby substantially increase the cell surface density of these stimulatory NKG2D ligands. As a consequence, MICA/B antibodies induce strong killing of human tumor cells by NK cells. These antibodies also induce immunity in mouse models of metastasis. Single-cell RNA-seq data show that a MICA/B antibody induces a striking shift among metastasis-infiltrating NK cells to an activated and cytotoxic state. We have also validated these antibodies in a humanized mouse model in which human NK cells target metastases formed by human tumor cells. Many human cancers are resistant to immunotherapy with checkpoint blockade through loss of MHC class I expression. However, MHC class I protein expression is not required for anti-tumor immunity mediated by innate T cell populations (NKT cells, ???T cells) and NK cells that all express the NKG2D receptor. The major goal of this project is to develop MICA/B antibodies as a therapeutic strategy for MHC class I deficient tumor cells that are resistant to conventional CD8 T cells. We have developed an integrated approach to study this important question in fully immunocompetent mouse models (Aim 1) as well as humanized mouse models and human tumor metastases (Aim 2). In Aim 1, we will examine the contribution of innate T cell and NK cell populations to MICA/B antibody mediated immunity against spontaneous metastases. In Aim 2, we will perform an in depth single-cell RNA-seq analysis of NKG2D-expressing innate T cell and NK cell population in human melanoma metastases. We will also use a humanized mouse model to develop combination therapies with established cancer therapeutics that enhance MICA/B expression and may therefore act synergistically with MICA/B antibodies. These studies will significantly advance the cancer immunotherapy field by developing novel approaches to target human cancers resistant to current immunotherapies.

Abstract Cytotoxic T cells play a central role in tumor immunity based on their ability to target tumor cells with a high degree of specificity. Checkpoint blockers that target the inhibitory CTLA-4 and PD-1 receptors on T cells have made a major impact on the treatment of melanoma and other human cancers. However, primary and secondary resistance to checkpoint blockade remain major clinical problems, and most melanoma patients still fail to benefit from these therapeutic agents. The goal of this project is to define the transcriptional and epigenetic mechanisms that render melanomas resistant to T cell-mediated immunity. We recently performed a genome-scale screen and discovered a large number of genes (n=128) that mediate resistance by melanoma cells to cytotoxic T cells. Of particular interest are three major transcriptional and epigenetic regulators, the PBAF form of the SWI/SNF complex as well as the SOX4 and NF-kB p65 transcription factors. Aim 1 will focus on the role of PBAF complex in resistance to immunotherapy. The PBAF complex is relevant in human melanoma because the ARID2 gene is frequently mutated. Preliminary data demonstrate that inactivation of the genes encoding the three unique subunits of the PBAF complex (Pbrm1, Arid2 and Brd7) sensitizes murine melanoma cells to cytotoxic T cells. We will investigate the impact of the PBAF complex on the tumor microenvironment in human melanoma as well as murine and zebrafish models of melanoma. Mechanistic studies will focus on the epigenetic enzymes and transcription factors that cooperate with the PBAF complex to inhibit T cell-mediated tumor immunity. Aim 2 will focus on the two major transcription factors ? NF-kB p65 and SOX4 ? that were discovered in the genetic screen. We will investigate whether the NF-kB and SOX4 transcription factors cooperate with the PBAF complex to confer resistance to cytotoxic T cells. The TNF??? NF-kB pathway was identified as the top resistance pathway in the screen, and we will investigate whether TNF? secreted by activated T cells renders neighboring tumor cells more resistant to T cell-mediated killing. The TNF? ? NF-kB pathway can induce either cell survival or cell death, and a number of gene products identified in the screen inhibit TNF? induced cell death, including the kinase TBK1. We will therefore investigate whether the outcome of TNF?-mediated signaling can be switched from tumor cell resistance to cell death using available TBK1 inhibitors. Expression of the transcription factor SOX4 is induced by TGF? following activation of latent TGF? by integrin ?V. Both genes encoding SOX4 (Sox4) and integrin ?V (Itgav) were identified as major resistance mechanisms in our genetic screen, and we will investigate whether this resistance pathway can be targeted with a mAb that inhibits integrin ?V mediated TGF? activation. This project will make a substantial contribution to this P01 through major collaborations with Drs. Fisher and Zon (melanoma models), Drs. Rodig, Hodi and Fisher (human melanoma) and Drs. Liu, Zon and Fisher (epigenetic mechanisms).

Abstract T cells are central effector cells of protective anti-tumor immunity, but little is currently known about these important immune cells in human GBM. We have generated single-cell RNA-seq data on tumor-infiltrating T cell populations from GBM patients at initial diagnosis or relapse. We used these full-length RNA-seq data to identify clonally expanded T cell populations based on their TCR? and ? chain sequences and then examined which genes were overexpressed by such expanded T cells. This analysis highlighted the KLRB1 gene which encodes the CD161 receptor that was previously shown to inhibit NK cell-mediated cytotoxicity. The CD161 ligand, CLEC2D, is expressed at the cell surface of human GBM cells. We therefore hypothesize that the CD161 ? CLEC2D pathway inhibits the anti-tumor function of both CD8 and CD4 effector T cell populations in GBM. Preliminary data show that inactivation of the KLRB1 gene in primary human T cells greatly enhances their effector function in a humanized mouse model of GBM. Our preliminary data also demonstrate that several other inhibitory receptors are expressed by substantial populations of GBM-infiltrating T cells, including CD96 and two prostaglandin E2 receptors (EP2 and EP4). Aim 1 will focus on the analysis of tumor-infiltrating T cells in GBM patients enrolled in the phase 1b NeoVax plus PD-1 antibody trial described in Project 1. These studies will primarily focus on the expression of inhibitory receptors by T cells and their ligands by tumor cells and myeloid cells. Expression of inhibitory receptors and their ligands will be examined by 16-color spectral flow cytometry and single-cell RNA-seq (in collaboration with Cores 1 and 2), with an emphasis on CD161, PD-1, CD96 and prostaglandin E2 receptors. We will investigate paired tumor samples from the same patient obtained at initial surgery and relapse in order to determine how expression of these inhibitory receptors and their ligands changes following immunotherapy with NeoVax plus PD-1 antibody. In collaboration with Project 1, we will also examine the spatial distribution of T cells that express CD161 and other inhibitory receptors. Aim 2 will investigate the therapeutic significance of the CD161 ? CLEC2D pathway. We will first use a genetic approach to study this inhibitory receptor by inactivating the KLRB1 gene in primary T cells. Blocking mAbs specific for human CD161 will also be used to examine the therapeutic potential of these findings. We will also examine combination therapies (collaboration with Projects 2, 4 and Core 3) involving the inhibitory receptors identified by single-cell RNA-seq in human GBM infiltrating T cells, with a particular focus on CD161, PD-1, CD96 and the prostaglandin E2 receptors. These studies will significantly advance our understanding of T cell function in GBM and characterize important inhibitory receptor ? ligand interactions that constrain effector T cell function.

Therapeutic Targeting of a Major Immune Resistance Pathway in Triple-negative Breast Cancer Abstract Checkpoint blockade results in moderate survival benefit in a subset of patients with triple-negative breast cancer (TNBC) but most patients currently fail to benefit from immunotherapy. We recently discovered that the genes encoding the integrin ?V and SOX4 proteins (ITGAV and SOX4 genes) render tumor cells resistant to killing by cytotoxic T cells. Integrin ?V?6 and SOX4 form a resistance pathway that is particularly relevant to TNBC: Integrin ?V?6 releases TGF? from an inactive latent complex by a force-dependent mechanism, and active TGF? induces expression of the SOX4 transcription factor that inhibits T cell-mediated tumor immunity. The integrin ?V?6 heterodimer is expressed at a low level by healthy epithelial cells, but its expression is highly upregulated in many epithelial cancers, including TNBC. TGF? is an important immunosuppressive cytokine in human tumors, but has been difficult to target due to its pleiotropic biology in different cell types and tissues. This approach provides an opportunity for more selective targeting of TGF? for cancer immunotherapy. Our preliminary data demonstrate that an integrin ?V?6 blocking mAb inhibits SOX4 expression and sensitizes TNBC cells to cytotoxic T cells. This antibody confers a substantial survival benefit in two aggressive mouse models of TNBC that are resistant to checkpoint blockade. In Aim 1, we will study the integrin ?V?6 ? SOX4 resistance pathway as an immunotherapy target in aggressive and highly metastatic mouse models of TNBC. In particular, we will examine how inhibition of integrin ?V?6 induces a substantial influx of CD8 T cells into TNBCs that are poorly infiltrated by T cells. In Aim 2, we will investigate the molecular mechanisms by which the integrin ?V?6 ? SOX4 pathway inhibits T cell-mediated tumor immunity. Preliminary data demonstrate that the SOX4 transcription factor inhibits expression of genes from multiple innate immune pathways in tumor cells, including the cytosolic dsRNA and dsDNA sensing pathways as well as the type 1 interferon response pathway. We will define the direct transcriptional targets of SOX4 and study how SOX4 cooperates with other transcription factors to render tumor cells resistant to T cell-mediated tumor immunity. In Aim 3, we will study the significance of the integrin ?V?6 ? SOX4 pathway in human TNBC. Our hypothesis is that this resistance pathway inhibits T cell infiltration in human TNBC, and we will therefore examine the spatial relationship between integrin ?V?6/SOX4 expression and T cell infiltration in human TNBC specimens. High-affinity integrin ?V?6 antibodies and a small molecule inhibitor are already being tested in clinical trials for fibrosis indications. The studies described here could thus provide the scientific rationale for testing of such inhibitors in TNBC and other human cancers of epithelial origin.

Abstract Single cell RNA-sequencing (scRNA-seq) has emerged as a powerful approach to dissect tumors and their ecosystems by determining the state of individual cells and inferring partial genetic information from expressed transcripts. For infiltrating T cells, this provides a comprehensive approach to define T cell states in human tumors. For malignant cells, scRNA-seq paves the way to characterizing cellular states and their associated genotypes at cellular resolution. The Molecular Immunology Core (Core 1) provides standardized and reproducible methods to profile and characterize immune cells and malignant cells by scRNA-seq in glioblastoma in both patients and mouse models. Core 1 also provides expert personnel, facilitates the sharing of knowledge within the program, and maintain the sophisticated equipment necessary to generate data for all Projects. The Molecular Immunology Core will be led Dr. Mario Suvà and Dr. Kai Wucherpfennig who will contribute their expertise in single-cell technologies and immune cell isolation.