Eli E. Sercarz

Restricted use of T cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy

Experimental allergic encephalomyelitis (EAE) is a paralytic autoimmune disease induced in susceptible animals by active immunization with myelin basic protein (MBP) or by passive transfer of MBP-specific T helper (TH) lymphocytes. We have analyzed the T cell receptor genes of 33 clonally distinct TH cells specific for a nonapeptide of MBP inducing EAE in B10.PL (H-2u) mice. All 33 TH cells used two alpha variable gene segments (V alpha 2.3, 61%; V alpha 4.2, 39%), the same alpha joining gene segment (J alpha 39), and two V beta and J beta gene segments (V beta 8.2-J beta 2.6, 79%; V beta 13-J beta 2.2, 21%). The anti-V beta 8 monoclonal antibody F23.1 was found to block completely recognition of the nonapeptide by V beta 8 TH cells in vitro and to reduce significantly the susceptibility of B10.PL mice to peptide-induced EAE.

Encephalitogenic T cells in the B10.PL model of experimental allergic encephalomyelitis (EAE) are of the Th-1 lymphokine subtype

T helper cells reactive to myelin basic protein are clearly implicated in the pathogenesis of murine EAE. We have developed a T cell line, BML-1 that (1) is reactive to the encephalitogenic amino terminal nonapeptide (1-9NAC) of MBP, (2) is I-Au restricted, and (3) induces relapsing EAE in B10.PL (H-2u) mice. Measurement of the lymphokine profile of BML-1 revealed secretion of IL-2, interferon-gamma and lymphotoxin but not IL-4. This profile is consistent with the Th1/DTH subtype. Coculture of BML-1 with MBP-primed B cells shows that BML-1 does not provide significant helper function in vitro. In addition, BML-1 secretion of interferon-gamma was found to inhibit LPS-induced anti-MBP antibody responses. This suggested that anti-MBP antibodies may not be necessary for induction of EAE. Sera from mice, in which severe disease was induced with the 1-9NAC peptide and Bordetella pertussis, showed no development of serum antibodies to MBP. These data show that MBP-reactive Th cells of the Th-1/DTH subtype can induce EAE and do not provide Th function for anti-MBP responses and that serum anti-MBP antibodies are not found in peptide 1-9NAC-induced disease. T cell lines specific for encephalitogenic epitopes and characterized for lymphokine secretion will provide a useful tool for understanding the role of T cells in the induction of EAE.

The Choice of T‐Cell Epitopes Utilized on a Protein Antigen Depends on Multiple Factors Distant from, as well as at the Determinant Site

The amino acid sequence of hen eggwhite lysozyme (HEL) has many differences when compared to the mouse lysozyme sequence, and so it could be predicted that there are many potential T-cell determinants on the HEL molecule. However, analysis of the T-cell response to HEL in several mouse strains has revealed that some regions are favored at the expense of others (Maizels et al. 1980, Katz et al. 1982, Bixler et al. 1985). Limitation of the T-cell response to one or a few determinants has been found in the immune response to other proteins, e.g. sperm whale myoglobin (Berkower et al. 1984), cytochrome c (Solinger et al. 1979), )9-galactosidase (Krzych et al. 1982) and staphylococcal nuclease (Finnegan et al. 1986), and appears to be a general property of the immune system. However, we have found by immunization with HEL peptides that there are additional determinants on the protein that were undetected following immunization with the native protein alone. Similarly, as T-cell responses to other proteins have been investigated in greater detail, more determinants have been uncovered (Livingstone & Fathman 1987). Rather than the T-cell response being limited by

Induction of anti-self-immunity to cure cancer

Navreet K. Nanda and Eli E. Sercan Department of Microbiology and Molecular Genetics University of California, Los Angeles Los Angeles, California 90024-1489 A quiet but profound revolution in theoretical framework, a paradigm shift, now has infused the consideration of therapies in autoimmune diseases and cancers, as exem- plified at the Jennifer Jones Simon Foundation workshop on Tumor Antigens as Self-Antigens, held in Los Angeles, California, on January 28-27, 1995. It is becoming accepted that a large set of antigenic determinants of the self have not induced self-tolerance (reviewed by Sercarz et al., 1993) and that these peptide determinants furnish target structures for autoimmune at- tack (reviewed by Lanzavecchia, 1995) and could provide potential targets for immune responses directed against tumors. It is not necessary to seek mysterious nonself- or neoself-antigens expressed by the tumor. The bulk of the lively and open discussion, characteristic of these free format workshops with no fixed presentations, was modu- lated very capably by A. Tobin (University of California, Los Angeles), a neurobiologist, and focused on how to initiate, maintain, and regulate antitumor autoimmunity, which could translate into effective treatment in cancer clinics. Are there in fact certain tumor-related determinants that can be rendered into crucial targets of attack by the im- mune system? The pertinent focus is on suitable antigen processing and subsequent presentation by major histo- compatibility complex (MHC) class I and class II molecules expressed by tumor cells. Any of the peptides that is bound in the groove of an MHC molecule on tumor cells provides a potential target determinant for attack by the immune system. The peptides bound by the MHC molecules of all ceils, including tumor cells, are derived from endogenous cellular (or viral) proteins, and the antigen-processing ma- chinery of the cells manages to display certain antigen determinants to ambient T cells. Tumor cells are distinct in that they possess additional oncoproteins that either are overexpressed owing to dysregulation or are mutated and have thereby conferred the tumor phenotype to these cells, to developmental antigens reexpressed during the process of tumorigenesis, or to passenger mutations in nononcogenic proteins that result from the loss of mecha- nisms that maintain genomic stability. Using T cell clones that are specifically able to kill the tumor as detectors for immunogenic tumor-derived determinants bound to MHC molecules of the tumor, several investigators have shown that T cells indeed recognize peptides from endogenous normal (and occasionally mutant) self-proteins. T. Boon and colleagues (van der Bruggen et al., 1991), in their pioneering studies, drew a page from bacterial genetics and cloned genes encoding tumor antigens recognized by CD8 T cells specific for human melanomas. Melanoma antigen 1 (MAGE-1) to MAGE-3 were the original family of human melanoma-specific antigens that were molecularly defined in this way and were followed by isolation of deter- minants on tyrosinase, gp100, and Melan-A-MART-1 rec- ognized by antimelanoma T cells (reviewed by Houghton, 1994). It is interesting that the MAGE gene family is not expressed in any normal adult tissue except testis, but is expressed in a large proportion of other human tumors (including small cell lung carcinoma, breast cancers, and colon carcinoma) and perhaps represents developmental antigens reexpressed during the process of tumorigene- sis. Tyrosinase, gplO0, and Melan-A-MART-1 are normal self-proteins specific to the melanocyte lineage and T cells specific for determinants on each of these antigens can be found in a large majority of melanoma patients (re- viewed by Houghton, 1994; Pardoll, 1994). It is thus be- coming clear that there is no special group of proteins that can be dubbed tumor antigens, and the distinction between self-antigens and tumor antigens is rapidly van- ishing. Thus, as summed up by Tobin, we have entered an era in which “we either know the tumor antigens or know how to know them.” But can the immune system make a T cell response against all the peptides bound to MHC molecules of a tumor cell? Part of the answer lies in the availability of the Tcell repertoire membership directed against MHC-bound determinants and the proportion of T cells rendered toler- ant in each individual. It is now thought that only well- expressed self-determinants are efficient in tolerance in- duction. In addition, on every self-antigen, there are sequestered determinants that do not succeed in inducing tolerance but that, under the circumstances of severe in- flammation and its attendant cytokine milieu, can be dis- played in an immunogenic context. Among the diversity of self-reactive T cells that evade negative selection, tu- mor-specific members can be mobilized under conditions of heightened awareness by the immune system-in which MHC molecules, surface adherence costimulators are up-regulated and become available for possibly killing interaction along with newly displayed, pre- viously “cryptic” self-antigenic determinants (reviewed by Sercarz et al., 1993; Lanzavecchia, 1995). Autoimmune disease on one hand and the existence of tumor-specific cytolytic T cells recognizing self-peptides in the cancer patient on theother are atestimonyto theenormouspoten- tial resources inherent in the positively selected T cell rep- ertoire, directed to the cryptic self. For example, T cells specific for peptides of self-protein tyrosinase can be iso- lated from normal individuals, which can attack and kill melanoma cells from human lymphocyte antigen (HLA)- matched cancer patients (Visseren et al., 1995). What may be critical is the density of the upregulated peptide MHC complexes, as well the expression of costimula- tory signals that influence activation of the otherwise silent tumor-specific T cell repertoire existent in the cancer pa- tient. The task at hand is how to generate and manipulate

Clustering of T Cell Ligands on Artificial APC Membranes Influences T Cell Activation and Protein Kinase C θ Translocation to the T Cell Plasma Membrane

T cell activation is associated with active clustering of relevant molecules in membrane microdomains defined as the supramolecular activation cluster. The contact area between these regions on the surface of T cells and APC is defined as the immunological synapse. It has been recently shown that preclustering of MHC-peptide complexes in membrane microdomains on the APC surface affects the efficiency of immune synapse formation and the related T cell activation. Disruption of such clusters may reduce the efficiency of stimulation. We describe here an entirely artificial system for Ag-specific, ex vivo stimulation of human polyclonal T cells (artificial APC (aAPC)). aAPC are based on artificial membrane bilayers containing discrete membrane microdomains encompassing T cell ligands (i.e., appropriate MHC-peptide complexes in association with costimulatory molecules). We show here that preclustering of T cell ligands triggered a degree of T cell activation significantly higher than the one achieved when we used either soluble tetramers or aAPC in which MHC-peptide complexes were uniformly distributed within artificial bilayer membranes. This increased efficiency in stimulation was mirrored by increased translocation from the cytoplasm to the membrane of protein kinase θ, a T cell signaling molecule that colocalizes with the TCR within the supramolecular activation cluster, thus indicating efficient engagement of T cell activation pathways. Engineered aAPC may have immediate application for basic and clinical immunology studies pertaining to modulation of T cells ex vivo.

Mhc-guided processing: binding of large antigen fragments.

Ever since the emergence of models for the processing and presentation of antigenic determinants by MHC class II molecules, the main view has been that proteins are unfolded, enzymatically cleaved into peptide lengths of about 12–25 amino acids and then loaded onto MHC class II molecules. There is, however, an alternative model stating that partially intact unfolding antigens are first bound by MHC class II molecules and then trimmed to fragments of a smaller size while remaining bound to the MHC class II molecule. In this analysis, we make the case that a considerable portion of the elutable peptide cargo belongs to this latter class.

The Adaptor Protein AP-3 Is Required for CD1d-Mediated Antigen Presentation of Glycosphingolipids and Development of Vα14i NKT Cells

Relatively little is known about the pathway leading to the presentation of glycolipids by CD1 molecules. Here we show that the adaptor protein complex 3 (AP-3) is required for the efficient presentation of glycolipid antigens that require internalization and processing. AP-3 interacts with mouse CD1d, and cells from mice deficient for AP-3 have increased cell surface levels of CD1d and decreased expression in late endosomes. Spleen cells from AP-3–deficient mice have a reduced ability to present glycolipids to natural killer T (NKT) cells. Furthermore, AP-3–deficient mice have a significantly reduced NKT cell population, although this is not caused by self-tolerance that might result from increased CD1d surface levels. These data suggest that the generation of the endogenous ligand that selects NKT cells may also be AP-3 dependent. However, the function of MHC class II–reactive CD4+ T lymphocytes is not altered by AP-3 deficiency. Consistent with this divergence from the class II pathway, NKT cell development and antigen presentation by CD1d are not reduced by invariant chain deficiency. These data demonstrate that the AP-3 requirement is a particular attribute of the CD1d pathway in mice and that, although MHC class II molecules and CD1d are both found in late endosomes or lysosomes, different pathways mediate their intracellular trafficking.

Immune tolerance to autoantibody-derived peptides delays development of autoimmunity in murine lupus.

Mechanisms that initiate and maintain autoantibody (autoAb) production in individuals with autoimmune diseases like SLE are poorly understood. Inadequate suppression of autoreactive T cells and/or unusual activation of T and B cells may underlie the persistence of pathogenic autoAbs in lupus. Here, we examine the possibility that in mice with lupus, autoAb molecules may be upregulating their own production by activating self-reactive T cells via their own processed peptides; downregulation of this circuit may decrease autoAb production and delay the development of lupus. We found that before the onset of clinical disease, lupus-prone (NZB/NZW) F1 [BWF1] (but not MHC-matched nonautoimmune mice) developed spontaneous T cell autoimmunity to peptides from variable regions of heavy chains (VH) of syngeneic anti-DNA mAbs but not to peptides from the VH region of an mAb to an exogenous antigen. Tolerizing young BWF1 mice with intravenous injections of autoAb-derived determinants substantially delayed development of anti-DNA antibodies and nephritis and prolonged survival. Thus, in such an autoAb-mediated disease, the presence of autoreactive T cells against VH region determinants of autoAbs may represent an important mechanism involved in the regulation of autoimmunity. Our findings show that tolerizing such autoreactive T cells can postpone the development of an autoimmune disease like SLE.

The dominant self and the cryptic self: shaping the autoreactive T-cell repertoire

Deletion of autoreactive T cells during the induction of self tolerance has been directly demonstrated. However, it is still relatively easy to detect self reactivity in normal healthy animals. In this article, Guy Gammon, Eli Sercarz and Gilles Benichou speculate on which T cells may escape tolerance induction and discuss how these cells could subsequently be involved in autoimmunity.

Regulatory T Cells Control Autoimmunity In Vivo by Inducing Apoptotic Depletion of Activated Pathogenic Lymphocytes

Clinical autoimmunity requires both activation of self-reactive T cells as well as a failure of peripheral tolerance mechanisms. We previously identified one such mechanism that involves regulatory T cells recognizing TCR Vβ8.2 chain-derived peptides in the context of MHC. How this regulation affects the fate of target Vβ8.2+ T lymphocytes in vivo that mediate experimental autoimmune encephalomyelitis has remained unknown. The present study using immunoscope and CFSE-labeling analysis demonstrates that the expansion of regulatory CD4 and CD8 T cells in vivo results in apoptotic depletion of the dominant, myelin basic protein-reactive Vβ8.2+ T cells, but not subdominant Vβ13+ T cells. The elimination of only activated T cells by this negative feedback mechanism preserves the remainder of the naive Vβ8.2+ T cell repertoire and at the same time results in protection from disease. These studies are the first in clearly elucidating the fate of myelin basic protein-specific encephalitogenic T cells in vivo following regulation.