The T-cell repertoire is heavily influenced by tolerance to polymorphic self-antigens

The effects of chronic infection with a superantigen-producing virus

C3H/HeJ mice transmit a mouse mammary tumor virus from mother to pup in milk. The retrovirus infects mice shortly after birth and, when expressed in recipient mice, produces a V beta 14-specific superantigen. The consequences of such expression on V beta 14-bearing T cells are examined in this paper. Most cells bearing V beta 14 and either CD4 or CD8 are eliminated in the thymus. Some V beta 14-bearing cells escape to the periphery, however. Those bearing CD8 are unaffected by expression of the viral superantigen. The percentage of peripheral CD4+ T cells bearing V beta 14 drops with time after birth. In large part this seems to be due to the fact that many of these cells become anergic because of exposure to the viral superantigen. Unlike normal T cells, these anergic cells cannot undergo peripheral postthymic expansion. Consequently, they drop in percentage even during a time when their total numbers are constant.

lpr T cells vaccinate against lupus in MRL/lpr mice

MRL/MP-lpr/lpr mice are homozygous for the lpr mutation that results in the accumulation of phenotypically abnormal cells (CD3+CD4+CD8-) in all lymphoid issues. Although no major abnormalities in the T cell receptor repertoire expressed by such lpr cells have been reported, the lpr mutation is a major disease-accelerating factor. Finally, intravenous administration of irradiated lpr cells recovered from the hyperplastic lymph nodes of adult diseased animals to young MRL/Mp-lpr/lpr mice resulted in a highly significant amelioration of disease parameters. This "T cell vaccination" approach resulted in a selective depletion of cells expressing products of the V beta 8.2 subfamily among lymph node T cells, in addition to eliciting a surge in peripheral T cells capable of conferring disease protection in adoptive transfer experiments. Thus, a strategy aimed at specifically reducing the frequency of lpr cells proved successful in mitigating the autoimmune process. These findings add to the involvement of lpr cells in the autoimmune process and constitute the first report that T cell vaccination may be beneficial to a spontaneously occurring autoimmune disease.

Altered course of visceral leishmaniasis in mice expressing transgenic I-E molecules

Previous studies had shown that the outcome of infection with Leishmania donovani was exquisitely sensitive to the influence of the major histocompatibility complex. In this study, we have examined the course of infection in non-obese diabetic (NOD) and NOD-E-3 mice, the latter expressing an I-E molecule as a result of transgenic introduction of the wild-type Ed alpha gene. Introduction of this transgene significantly altered the course of infection allowing for enhanced parasite multiplication in the viscera from day 14 to day 28. This was associated with both a delayed and reduced tissue granulomatous response in NOD-E-3 mice. In vitro, spleen cells from these mice produced equivalent levels of interferon (IFN)-gamma during the early phase of infection but this originated from populations having a different balance of T cells subsets. In NOD mice CD8+ T cells contribute substantially to the total levels of IFN-gamma produced, but in transgenic mice the contribution from this subset is significantly decreased. This is reflected in a reduction in the proportion of Leishmania-specific CD8+ T cells, which could only partially be accounted for by deletion of V beta 5- and V beta 3-expressing CD8+ T cells in NOD-E-3 mice. This study highlights the impact of the introduction of a class II gene product on disease outcome and unexpectedly on the functional potential of CD8+ T cells.

Direct activation of murine T cells by staphylococcal enterotoxins

The capacity of staphylococcal enterotoxins to stimulate all T cells bearing certain TCR variable region alleles has generated a great deal of interest. This stimulation appears to involve specific binding of the toxin to class II molecules and subsequent stimulation of the T cell via the TCR V beta elements. Recent studies from our laboratory have focused on the ability of staphylococcal enterotoxins to directly activate purified lymph node T cells and a panel of T cell clones and hybridomas. A T cell costimulation assay was performed to assess cellular activation requirements and cytokine receptor expression. Activation of highly purified lymph node T cells by staphylococcal enterotoxin B (SEB) required costimulatory signals which could be provided by IL-1, IL-2, IL-4, or IL-6, whereas SEB alone demonstrated no significant proliferative response. Using a panel of TH1 and TH2 cell clones and T cell hybridomas possessing various responsive and nonresponsive V beta alleles, it was possible to demonstrate that SEA and SEB costimulate T cells via the TCR complex. Additionally, enterotoxin-pretreated T cells demonstrated a significant proliferative response upon exposure to class II-bearing accessory cells, suggesting that these toxins bind directly to T cells. Highly purified T cells cultured with both SEB and IL-1 exhibit significantly increased levels of IL-2 receptor, whereas cells cultured with SEB or IL-1 alone demonstrated low levels of this receptor. These results do not exclude an association of the staphylococcal enterotoxins with class II molecules in a manner which results in a high avidity binding to the TCR required for transduction of the appropriate activation signals. In the absence of class II molecules, however, these superantigens can still bind to T cells, and the activation signal is delivered in the presence of cytokines that trigger T cell growth and lymphokine production.

Negative selection of Tcra-V8+CD8+ T cells by MHC class I molecules

Tcrb-V-specific positive and negative selection of T cells has been well documented. In contrast, nothing is known about Tcra-V-specific selection. Using Tcra-V8-specific KT50 antibody Tcra-V8-specific selection of T cells has been examined. The CD8+ T cell subpopulation bearing Tcra-V8 are shown to be negatively selected by major histocompatibility complex (MHC) class I H-2Kd and H-2Dd/Ld molecules. Furthermore, percentages of these T cells are also influenced by Tcra-V haplotypes. Involvement of non-H-2 self (super)antigens in this MHC class I restricted negative selection, however, remains to be determined.

Preferential association of alpha and beta chains of the T cell antigen receptor

The cytotoxic T lymphocyte (CTL) clone B6.2.16 expresses a V beta 8.2/J beta 2.3/C beta 2-encoded T cell receptor (TcR) beta chain and an alpha chain that is encoded by a novel V alpha gene segment, J alpha 27 and C alpha. While expression of V alpha B6.2.16 and J alpha 27 is not detectable in lymph node cells of normal C57BL/6 mice, expression of these gene segments was readily seen in transgenic mice expressing the rearranged beta chain gene of the B6.2.16 T cell clone. This finding indicates that only a limited number of alpha chains can associate with the B6.2.16 beta chain and strongly suggests that the size of the TcR repertoire of mature T cells is not only limited by TcR ligand-mediated thymic selection but also by restrictions in alpha-beta combinatorial chain association.

Distinct T-cell receptor V beta gene usage by human T lymphocytes stimulated with the streptococcal pyrogenic exotoxins and pep M5 protein

A number of streptococcal products, including the streptococcal pyrogenic exotoxin (SPE) types A, B, and C as well as a 22-kDa fragment of M type 5 protein (pep M5), are potent stimulants of human T-lymphocyte blastogenesis and belong to the newly designated family of superantigens. The V beta usage of human T cells stimulated with these toxins was investigated by using the polymerase chain reaction. We demonstrate that SPE A, B, and C as well as pep M5 stimulate the proliferation of T cells in a dose-dependent manner. pep M5 stimulates cells bearing V beta 2, 4, and 8 elements of the T-cell receptor (TCR), whereas SPE A stimulates TCR V beta 2-, 12-, 14-, and 15-bearing cells. SPE B stimulated only cells expressing TCR V beta 8 elements, while SPE C stimulated cells expressing V beta 1, 2, 5.1, and 10. These studies reveal that the preferential usage of particular V beta elements is distinct for these different superantigens, which may be important in the pathogenesis of various streptococcal diseases.

Mapping T-cell receptor-peptide contacts by variant peptide immunization of single-chain transgenics

To test models of T-cell recognition, mice transgenic for T-cell receptor alpha or beta chain have been immunized with variant peptides that force changes in the resulting T-cell response. In particular, charge substitutions on the peptide often elicit reciprocal charges in the junctional (CDR3) sequences of T-cell receptor V alpha or V beta chains, indicating direct T-cell receptor-peptide contact, and allowing derivation of a topology for the T-cell receptor-MHC interaction. At one position on the peptide, variants transformed a homogeneous V beta response into a very heterogeneous one.

The T cell receptor/CD3 complex is composed of at least two autonomous transduction modules

Recent studies have demonstrated that the CD3-zeta subunit of the T cell antigen receptor (TCR) complex is involved in signal transduction. However, the function of the remaining invariant subunits, CD3-gamma, -delta, and epsilon, is still poorly understood. To examine their role in TCR function, we have constructed TCR/CD3 complexes devoid of functional zeta subunit and showed that they are still able to trigger the production of interleukin-2 in response to antigen or superantigen. These data, together with previous results, indicate that the TCR/CD3 complex is composed of at least two parallel transducing units, made of the gamma delta epsilon and zeta chains, respectively. Furthermore, the analysis of partially truncated zeta chains has led us to individualize a functional domain that may have constituted the building block of most of the transducing subunits associated with antigen receptors and some Fc receptors.

Tcrb-V3+ T-cell deletion and a new mouse mammary tumor provirus, Mtv-44

Genes encoding endogenous superantigens causing Tcrb-V3+ T-cell deletion co-segregate with mouse mammary tumor proviruses (Mtv), Mtv-3, Mtv-6, and Mtv-13. In addition Mtv-1 has been implicated in deletion of these T cells. We have examined levels of Tcrb-V3+ T cells and Mtv integrations in the following offspring and their parental strains, [(CBA-T6 x NZW)F1 x CBA], [(CBA x C3H/He)F1 x CBA], and [(B10.S (9R) x NOD]F1 mice. We show that a new Mtv (Mtv-44) from NZW mice and Mtv-1 from C3H/He mice cosegregate with genes encoding ligands for partial deletion of Tcrb-V3+ T cells and that some NOD mice have an additional Mtv (Mtv-45) which is closely linked to Mtv-3.

The open reading frames in the 3' long terminal repeats of several mouse mammary tumor virus integrants encode V beta 3-specific superantigens

Mice expressing the minor lymphocyte stimulation antigens, Mls-1a, -2a, or -3a, singly on the B10.BR background have been generated. Mls phenotypes correlate with the integration of mouse mammary tumor viruses (MTV) in the mouse genome. The open reading frames within the 3' long terminal repeats of the integrated MTVs 1, 3, 6, and 13 encode V beta 3-specific superantigens. Sequence data for these viral superantigens is presented, indicating that it is the COOH-terminal portion of the viral superantigen that interacts with the T cell receptor V beta element.

Molecular characterization of Mls-1

Recently a series of endogenous and exogenous superantigens have been described which have one common feature, namely, they lead to in vivo deletion and in vitro stimulation of T cells expressing particular T cell receptor V beta genes. The Mls antigens represent the prototypes of these molecules. We have mapped Mls-1 to the endogenous mammary tumor virus (MMTV) Mtv-7, while other SAG have also been associated with various MMTV. The open reading frame gene of the MMTV encodes the SAG. Thus, the new terminology MMTV sag has been proposed for this gene. Transfection experiments suggest that the expression of MMTV sag is tightly controlled, probably by a negative acting factor encoded within the open reading frame. Furthermore, a pronounced IL-4 effect is seen in the functional detection of the transfected Mtv-7 sag. Since this lymphokine does not influence the mRNA level of the endogenous or transfected MMTV genes, it is likely that it exerts its effect by increasing transcription of MHC class II genes, whose products are required for functional detection of Mls. We have identified one mouse strain, MA/MyJ, which has an Mls-1 phenotype but does not contain Mtv-7. The SAG activity of this strain was mapped to a new mammary tumor provirus, Mtv-43, not seen in other inbred strains. Sequence analyses revealed that the predicted amino acid sequences of the Mtv-7 and the Mtv-43 sag genes are very similar. This is particularly striking in the C-terminus, where all other MMTV sag sequences differ 100%. Thus, this region of the molecule seems to control the V beta specificity of SAG molecules. It is likely that the SAG expression provides an advantage for the infectious MMTV, probably by facilitating its transmission by T cells from the site of primary residence in the gut to its final destination, the mammary glands.

Retroviral super-antigens and T cells

For many years immunologists have been intrigued by a series of potent antigens encoded in the murine genome. These antigens, originally termed minor lymphocyte stimulating (Mls) antigens, are capable of inducing extremely strong T cell proliferative responses when presented in the context of MHC class II molecules. Recently, Mls antigens have been shown to stimulate T cells bearing particular T cell receptor V beta elements, leading to the designation of super-antigens. The endogenous expression of these super-antigens in mice results in the clonal elimination of large numbers of T cells in order to maintain self-tolerance. In this review we discuss the recent identification of endogenous super-antigens as retroviral gene products. In addition, we analyze the role of class II MHC molecules in the presentation of endogenous super-antigens to T cells. Finally, we discuss the dramatic effect of retroviral super-antigens on the T cell repertoire.

Preferential use of T-cell receptor V genes in human autoimmune thyroid disease

We review here our analyses of hTcR V gene activity within the thyroid glands of patients with autoimmune thyroid disease. Our data, based on thyroid aspiration specimens, indicate that early in the onset of Graves' disease there is a marked restriction in both hTcR V alpha and V beta gene families utilized by intrathyroidal T-cells. Later, however, and as seen in surgical thyroid specimens from patients with long term disease, there appears to be a loss of V beta restriction for unclear reasons. In contrast, patients with Hashimoto's thyroiditis are usually diagnosed later in the natural history of the disease and appear to show much less hTcR V gene family restriction. The mechanisms driving the use of few and many hTcR V genes within the thyroid gland are likely to be complex and may reveal important insights into disease pathogenesis.

Cross-species transplantation tolerance: rat bone marrow-derived cells can contribute to the ligand for negative selection of mouse T cell receptor V beta in chimeras tolerant to xenogeneic antigens (mouse + rat----mouse)

Mixed xenogeneic bone marrow reconstitution (mouse + rat----mouse) results in stable mixed lymphopoietic chimerism (1-48% rat), long-term survival, and the induction of stable functional donor-specific transplantation tolerance to xenoantigens in vivo. To examine the role of negative selection of potentially xenoreactive T lymphocytes during tolerance induction across a species barrier, mixed xenogeneic chimeras (mouse + rat----mouse) were prepared and analyzed using a mixture of mouse and rat bone marrow cells for relative T cell receptor (TCR)-V beta expression on mouse T cells. In mixed xenogeneic chimeras (B10 mouse + rat----B10 mouse), T cell maturation proceeded normally in the presence of rat bone marrow-derived elements, and functional donor-specific tolerance to rat xenoantigens was present when assessed by mixed lymphocyte reactivity in vitro. V beta 5, which is expressed at high (undeleted) levels in normal B10 mice, was consistently deleted in B10 recipients of Wistar Furth (WF), but not F344 rat bone marrow, whereas the coadministration of either F344 rat or WF rat bone marrow with B10 mouse bone marrow cells resulted in a significant decrease in expression of TCR-V beta 11. Taken together, these data demonstrate for the first time that rat bone marrow-derived cells can contribute in a strain-specific manner to the ligand for negative selection of specific mouse TCR-V beta during tolerance induction across a species barrier.

Endogenous ligands selecting T cells expressing particular V beta elements

It has recently become clear that the minor lymphocyte stimulatory antigens (Mls) and other endogenous ligands which lead to the partial or total deletion of T cells bearing particular V beta segments are encoded by mouse mammary tumor virus (MMTV). We review here the genetic analyses of multiple V beta 11 and V beta 3 deletion ligands and demonstrate the involvement of MMTV in all examples. Several features of Mls and the V beta 11/V beta 3 deleting ligands identify them as members of the superantigen family. Bacterial superantigens are known to bind both MHC class II and the TCR in regions distinct from conventional peptide antigens. Within the MMTV genome, the 3' LTR has been identified as encoding superantigen function. We present data demonstrating that in vitro translation identifies the major product of the open reading frame (ORF) within the 3' LTR as a type II integral membrane glycoprotein. It is proposed that the type II membrane glycoprotein interacts with MHC and TCR in a manner analogous to the bacterial superantigens and distinct from conventional peptide antigen. Several unanswered questions regarding superantigen action remain; what determines total or partial deletion? How is Mls transferred between cells? These questions are addressed in the discussion.

Low affinity interaction of peptide-MHC complexes with T cell receptors

The interaction of antigen-specific T cell receptors (TCRs) with their ligands, peptides bound to molecules of the major histocompatibility complex (MHC), is central to most immune responses, yet little is known about its chemical characteristics. The binding to T cells of a labeled monoclonal antibody to the TCR was inhibited by soluble class II MHC heterodimers complexed to different peptides. Inhibition was both peptide- and TCR-specific and of low affinity, with a KD = 4 x 10(-5) to 6 x 10(-5) M, orders of magnitude weaker than comparable antibody-antigen interactions. This finding is consistent with the scanning nature of T cell recognition and suggests that antigen-independent adhesion precedes TCR engagement.

Analysis of class II MHC structure in thymic nurse cells

During the course of thymocyte maturation, the processes of positive selection and tolerance induction are mediated by interactions between thymocyte T-cell receptors and MHC molecules on thymic stromal cells. The means by which these seemingly contrary processes can be mediated by interactions between the same molecules has long been a source of controversy. One idea which has been put forward is that the MHC molecules in different microenvironments of the thymus are not the same. We have tested this hypothesis by examining class II transcripts derived from thymic cortical epithelial cells known as thymic nurse cells, reasoning that alternative splicing of primary transcripts might give rise to a positively selecting MHC molecule. However, we found no evidence for alternative splicing of these transcripts. These results are presented and discussed with regard to implications for possible mechanisms of positive selection.

Thymic stromal cells in culture. 2. Binding of normal thymocytes to a cloned thymic stromal cell line

Thymic stromal cell line TS-9 was found to selectively bind a subpopulation of normal murine thymocytes. Selective binding allowed the isolation and phenotypic characterization of the adherent and nonadherent subpopulations of thymocytes. Flow cytometric analysis of fluorescently labeled thymocytes revealed that the adherent and nonadherent populations differ in maturity, with the adherent population enriched in immature thymocytes of the PNAhi, Thy-1hi, CD3-/lo, and CD4+/CD8+ double positive surface phenotype. A quantitative microwell assay was developed to measure the binding of thymocytes to TS-9. Thymocytes labeled with vital DNA stain Hoechst 33342 were allowed to bind to TS-9 in microwells and the intense fluorescence of this label was readily detected with a scanning fluorometer. The binding was trypsin-sensitive and hyaluronidase and PI-PLC resistant. The binding was also temperature dependent and sensitive to cytochalasin B. A panel of monoclonal antibodies to cell surface antigens including CD2, LFA-I/ICAM-I, and Thy-1 was screened in a quantitative binding assay for their ability to inhibit the binding of thymocytes to TS-9. The binding was partially inhibited by the C3C12 monoclonal antibody which recognizes the recently identified and apparently unique gp23,gp45 complex expressed on murine stromal cells.

bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship

Early death is the fate of most developing T lymphocytes. Because bcl-2 can promote cell survival, we tested its impact in mice expressing an E mu-bcl-2 transgene within the T lymphoid compartment. The T cells showed remarkably sustained viability and some spontaneous differentiation in vitro. They also resisted killing by lymphotoxic agents. Although total T cell numbers and the rate of thymic involution were unaltered, the response to immunization was enhanced, consistent with reduced death of activated T cells. No T cells reactive with self-superantigens appeared in the lymph nodes, but an excess was found in the thymus. These observations, together with previous findings on B cells, suggest that modulated bcl-2 expression is a determinant of life and death in normal lymphocytes.

Profound alteration in an alpha beta T-cell antigen receptor repertoire due to polymorphism in the first complementarity-determining region of the beta chain

Amino acid residues that are critical in maintaining the framework structure of immunoglobulin heavy- and light-chain variable (V) regions are strongly conserved in the V alpha and V beta proteins of the alpha beta T-cell antigen receptor (TCR alpha beta). Consequently, it has been proposed that TCR alpha beta has a conformation similar to that of an immunoglobulin Fab fragment and that the regions of the TCR homologous to the three immunoglobulin complementarity-determining regions (CDRs 1, 2, and 3) bind to the peptide antigen-major histocompatibility complex (MHC) molecule ligand. A single amino acid substitution in the predicted CDR1 of the V beta 3 protein of certain mouse strains dramatically altered TCR alpha beta usage in an antigen-specific MHC-restricted immune response but did not abrogate V beta 3 specificity for the superantigens minor lymphocyte stimulatory locus (Mls)c and staphylococcal enterotoxin A (SEA). The results confirm the importance of the V beta CDR1 in antigen-MHC molecule recognition, supporting the Fab-like structural model of TCR alpha beta, and provide further evidence that conventional antigen-MHC recognition and superantigen recognition are mediated by distinct regions of the TCR beta chain. They also suggest that allelic polymorphism may be a significant source of diversity in the TCR repertoire.

Identification of a T cell receptor beta chain variable region, V beta 20, that is differentially expressed in various strains of mice

A cDNA library of TCR beta chain transcripts from BALB/c thymocytes was constructed using anchored polymerase chain reaction (PCR). Screening of this library led to the identification of a V beta gene segment, V beta 20, structurally related to V beta 3 and V beta 17. Genomic analysis of mice displaying deletions in their V beta loci, together with mapping of cosmid clones, situated V beta 20 2.5 kb beside V beta 17. The expression of V beta 20 was estimated by PCR in mice of different H-2 and Mls types. Peripheral T cells from H-2k and H-2d mice did not express V beta 20, whereas in I-E-negative mice (C57Bl/6 and SJL), V beta 20 transcripts were detected. The lack of V beta 20 transcripts in (C57Bl/6 x CBA/J)F1, (C57Bl/6 x BALB/c)F1, and in congenic B6.H-2k mice suggests that the differential use of V beta 20 is due to an I-E-mediated clonal deletion process. The involvement of the Mls super antigens was excluded by analysis of all Mls type combinations. The nature of the V beta 20-deleting element(s) is discussed in the context of the I-E/superantigen systems controlling the expression of V beta 11 and V beta 17.

Mls--a retrovirus exploits the immune system

The identity of minor lymphocytes stimulating (Mls) antigens, endogenous superantigens that can activate, or induce the deletion of, large portions of the T-cell repertoire, has recently been revealed: they are encoded by mouse mammary tumor viruses (MMTV) that have integrated into the germ line as DNA proviruses. As Hans Acha-Orbea and Ed Palmer point out, Mls-mediated modulation may be only the tip of the retrovirus iceberg; already murine leukemia virus (MuLV), with similar superantigen properties, has been discovered.

Mechanisms of IgE tolerance: dual regulatory T cell lesions in perinatal IgE tolerance

The mechanisms of genetic control of IgE responses are exercised at different immuno-physiological levels. This study centered upon the development of IgE lineage-specific regulatory T cells. Herein, we demonstrate the following points: (a) perinatal administration of soluble self IgE molecule or self IgE complexed with foreign antigen induces IgE tolerance as manifested by antigen-specific IgE unresponsiveness and a generalized IgE immunodeficiency, and the induction of IgE tolerance does not affect antigen-specific IgG1, IgG, and IgA responses; (b) inducibility of perinatal IgE tolerance is correlated with complete absence of endogenously secreted IgE in the neonates; and the state of persistent IgE tolerance also does not correlate with the presence of high levels of circulating anti-IgE autoantibodies; (c) The lesions induced during the ontogeny of IgE antibody system do not appear to result from an imbalance of production of interleukin 4 and interferon-gamma by T helper Th2 and Th1 cells in antigen-stimulated cultures; the dual immunoregulatory lesions in T cell subsets are demonstrated: clonal anergy/deletion of CD4+ IgE Th cells and the presence of CD8+ IgE suppressor cells induced by perinatal IgE treatment. We propose that antigen/interleukin 4 activated B cells are controlled by IgE lineage-specific regulatory T cells which recognize self IgE determinant(s) on IgE committed B cells. Life-long IgE tolerance ensues as a result of a new steady state of IgE lineage-specific CD4+ and CD8+ T cell subsets.

A method for production of antibodies to human T-cell receptor beta-chain variable regions

Mouse T-cell hybridomas bearing human V beta elements were produced by transfection of human/mouse hybrid T-cell receptor beta-chain genes into a mouse T-cell hybridoma lacking an endogenous beta-chain gene. These hybridomas were entirely mouse in origin except for the human V beta region. These cells were used to immunize mice against human V beta elements. Mouse monoclonal antibodies have thus been generated against human V beta 13.1 and -13.2. We expect that the method outlined in this paper will be useful in the production of monoclonal antibodies specific for other human V beta or V alpha elements.

Co-segregation of a gene encoding a deletion ligand for Tcrb-V3+ T cells with Mtv-3

A gene encoding the endogenous superantigen Mlsc, which deletes Tcrb-V3+ T cells in the NOD inbred mouse strain, was found to co-segregate with Mtv-3 on chromosome 11. This identifies a fourth gene encoding a deletion ligand for Tcrb-V3+ T cells and extends recently published observations in support of the hypothesis that a number of endogenous superantigens are the products of Mtv proviruses.

Analysis of the immune system with transgenic mice: T cell development

Transgenic mice carrying functionally rearranged T cell receptor genes have contributed significantly to our knowledge of T cell development and thymic positive and negative selection processes. In addition, TCR-transgenic mice have been used to investigate mutations affecting thymocyte development, like scid and lpr. Gene targeting by homologous recombination will allow to analyze more specifically the molecular mechanisms underlying thymic selection and peripheral tolerance.

Involvement of both T cell receptor V alpha and V beta variable region domains and alpha chain junctional region in viral antigen recognition

We have studied the lymphocytic choriomeningitis virus (LCMV)-specific cytotoxic T cell response in transgenic mice expressing either the T cell receptor (TcR) alpha (V alpha 2/J alpha TA31) or the corresponding TcR beta (V beta 8.1/D beta/J beta 2.4) chain originally isolated from the LCMV glycoprotein specific (residues 32-42), H-2Db-restricted T cell clone P14. The expression of single transgenic TcR chains did not influence the corresponding endogenous TcR V gene usage in unstimulated T cells indicating that one particular TcR alpha or beta chain can randomly pair with different V beta or V alpha chains without any obvious bias. However, upon infection with LCMV, reactive cytotoxic T lymphocytes (CTL) from P14 beta-transgenic mice were predominantly V alpha 2+ whereas CTL from P14 alpha-transgenic mice preferentially expressed V beta 8.1 and unexpectedly also V beta 8.3 (but not V beta 8.2). Correspondingly, the LCMV-specific CTL response in both alpha and beta TcR-transgenic mice was strongly biased to the original P14 T cell epitope (LCMV glycoprotein residues 32-42). Sequence analysis of a large panel of LCMV-reactive "half-transgenic" TcR from P14 single receptor chain-transgenic mice revealed a highly conserved VJ alpha and a more diverse VDJ beta junctional region. This report demonstrates that the antigen specificity of the studied TcR depends on the specific combination of both TcR alpha and beta chains which implies that amino acids located in the TcR V alpha and V beta segments as well as in the junctional region are involved in binding of the viral antigenic fragment.

Generation of monoclonal antibodies against soluble human T cell receptor polypeptides

One approach to the diagnosis and therapy of T cell-mediated diseases is to develop reagents specific for T cell receptor (TcR) variable (V) regions. To date, however, TcR expressed on the surface of antigen-specific T lymphocytes have proven to be poorly immunogenic. As a result, few monoclonal antibodies (mAb) recognizing human variable regions are available. In this report, we have used the "phosphatidylinositol linkage" strategy to generate soluble forms of two human allogeneic TcR derived from human cytotoxic T lymphocytes (CTL) known to be specific for HLA-A2 and HLA-Aw68/HLA-Aw69, respectively. Monomeric TcR alpha and beta chains from the HLA-A2-specific CTL were purified in large quantities from CHO cells and each was used to immunize mice to generate mAb. In particular, the anti-beta chain mAb, denoted anti-V beta 13, stain a significant (approximately 5%) fraction of human peripheral blood alpha/beta T lymphocytes, immunoprecipitate native anti-A2 TcR molecules, and activate T cells transfected with the relevant alpha and beta chain cDNA. Anti-alpha chain mAb were also obtained against a constant region determinant which can immunoprecipitate detergent-solubilized polypeptides. In general, we find that immunizations with soluble protein are far superior to those with cells bearing TcR chimeras or in combination with the purified protein.

An additional hypervariable region encoded by V gene segments occurs in Tcr V beta at a location compatible with its involvement in Tcr active site--a general model for alloreactivity

The number of V alpha and V beta sequences of T cell receptors now available allows a meaningful analysis of their variability profiles. Variability plots were derived using a modified form of Wu and Kabat's algorithm: variability is not computed as a proportion of the number of different residues occurring at a position, but rather proportionally to the physicochemical differences between the different residues. Results show that the classical hypervariable regions occurring in immunoglobulins also occur in T cell receptors at equivalent positions. Contrary to immunoglobulins the framework of Tcr V regions displays many relatively variable regions and positions. This phenomenon can be connected with the genetic organization of V genes of T cell receptors which seem to avoid any framework homogenization and the resulting gene conversion. More importantly an additional hypervariable region was detected in V beta but not in V alpha. This fourth hypervariable region is located between the second and the D hypervariable CDR. The predicted three-dimensional location of this additional hypervariable region is compatible with a possible role in antigen recognition and therefore also in positive and/or negative selection. Furthermore our data suggest that this fourth hypervariable region is involved in the recognition of superantigens like bacterial enterotoxins. Indeed this additional hypervariable region is not detected when variability is derived using an alignment of the V beta subgroups stimulated by one toxin of S. aureus. Finally we propose a new and simple molecular model to explain alloreactivity as crossreactivity between the universe of shapes (isomers of conformation) of different MHC haplotypes.

Autoimmunity in non-human primates: the role of major histocompatibility complex and T cells, and implications for therapy

Two autoimmune disease models were studied in rhesus monkeys: type II collagen-induced arthritis (CIA) and experimental allergic encephalomyelitis (EAE). Unrelated outbred animals were used in these studies. In both models disease resistant and susceptible individuals could be identified. Susceptibility correlated with in vitro cellular responsiveness to antigen in the CIA model. In both models resistant as well as susceptible individuals developed a humoral response to the inducing antigen. However, there is an indication that IgM antibodies play a crucial role in the induction of CIA. No clear association between major histocompatibility complex (MHC) type and disease incidence was found although a higher frequency of a certain DR type was observed in EAE susceptible monkeys. It is likely that both the antigen binding capacity of the MHC class II molecules and the T-cell repertoire play an important role in determining whether disease will develop or not.

T-lymphocyte dysfunctions occurring together with apical gut epithelial cell autoantibodies

Gut epithelial cell autoantibodies have been considered a hallmark of autoimmune enteropathy, a disorder occurring in children with protracted diarrhea of unknown etiology. Four patients (two male and two female) with such autoantibodies were studied. Immunofluorescence analysis showed two different disjunctive staining patterns: complement-fixing apical (three of four) and cytoplasmic (the remaining fourth one), which are shown to be directed against different structures. All three patients positive for complement-fixing apical gut epithelial cell autoantibodies had abnormal T-cell responses in vitro, one of them with an immunoglobulin G2 immunoglobulin deficiency and another with an immunoglobulin A deficiency. An immunoglobulin A deficiency without T-cell alterations was also diagnosed in the cytoplasmic gut epithelial cell autoantibody-positive patient. These findings suggest that different immunologic alterations (either a T-cell abnormality or immunoglobulin deficiency) may favor the appearance of gut epithelial cell autoantibodies (complement-fixing apical or cytoplasmic, respectively). Furthermore, these autoantibodies should not be considered a specific marker of autoimmune enteropathy, because they may not always be associated with such a disease: two patients with apical gut epithelial cell autoantibodies showed no signs of intestinal lesion or diarrhea.

A unified model for T cell antigen recognition and thymic selection of the T cell repertoire

Positive selection of T cells during thymic differentiation predisposes mature T cells to recognize glycoproteins encoded by "self" alleles of the major histocompatibility complex (MHC) as "restricting elements" for antigen presentation. Yet, negative selection also occurs during thymic differentiation resulting in the clonal deletion of T cells reactive with "self" MHC glycoproteins. Thymic processes of positive and negative selection represent a paradox because the "altered self" view of T cell receptor (TcR) recognition indicates that the same type of TcR-MHC glycoprotein binding interaction mediates both positive and negative selection of the T cell repertoire. Most contemporary models of thymic selection attempt to explain these paradoxical observations by quantitative differences of affinity. That is, TcR interactions with MHC ligands that are of medium affinity lead to positive selection whereas those that are of high affinity lead to negative selection. The purpose of this manuscript is to provide an alternative model of thymic selection based on the efficacy (the ability of a ligand to catalyze receptor mediated biological activity) of TcR-MHC ligand interactions. The "efficacy" model predicts that among those thymocytes exhibiting affinity for self MHC ligands, some clones bind self MHC ligands without efficacy whereas others bind these ligands with efficacy. Immature T cells that bind MHC ligands without efficacy do not undergo TcR mediated activation and thereby escape clonal deletion. Instead, these T cells compete for growth--promoting sites on thymic antigen presenting cells (APC) based upon their clonotypic TcR affinity for self MHC ligands. These T cells experience positive thymic selection and eventually dominate a repertoire of mature T cells predisposed to exhibit non-efficacious binding to "self" MHC ligands. In contrast, immature T cells that exhibit efficacious binding to self MHC ligands are deleted from the T cell repertoire during thymic maturation. By this mechanism, the mature T cell repertoire is selected so that clonotypic T cells are predisposed to bind the very sites on MHC glycoproteins responsible for antigen presentation without risk of autoimmunity. Given the clonotypic diversity of the mature repertoire, complexes of foreign peptides and self MHC glycoproteins would be recognized by clones of the appropriate specificity as highly efficacious ligands. In summary, the "efficacy" model is entirely consistent with the "altered self" concept of T cell antigen recognition and readily accounts for both positive and negative selection of the T cell repertoire.

Evidence for the effects of a superantigen in rheumatoid arthritis

While studying the alpha beta T cell receptor repertoire in rheumatoid arthritis (RA) patients, we found that the frequency of V beta 14+ T cells was significantly higher in the synovial fluid of affected joints than in the peripheral blood. In fact, V beta 14+ T cells were virtually undetectable in the peripheral blood of a majority of these RA patients. beta-chain sequences indicated that one or a few clones dominated the V beta 14+ population in the synovial fluid of individual RA patients, whereas oligoclonality was less marked for other V beta's and for V beta 14 in other types of inflammatory arthritis. These results implicate V beta 14-bearing T cells in the pathology of RA. They also suggest that the etiology of RA may involve initial activation of V beta 14+ T cells by a V beta 14-specific superantigen with subsequent recruitment of a few activated autoreactive v beta 14+ T cell clones to the joints while the majority of other V beta 14+ T cells disappear.

Resistance to collagen-induced arthritis in Biozzi mice is not associated with T cell receptor V beta gene polymorphism

H-I and H-II high antibody-responder Biozzi mice, which express the H-2q permissive haplotype, were shown to be sensitive and refractory to collagen-induced arthritis, respectively. To assess a possible role of T cell receptor (TcR) V beta gene deletion or polymorphism in the resistance to arthritis in H-II mice, the germinal structure of TcR V beta genes and their expression at the membrane level were compared in both lines. In contrast to H-2q refractory SWR mice, which exhibit a deletion of about 50% of TcR V beta gene segments. H-I and H-II lines have an identical and complete set of V beta genes and exhibit no difference in the average expression of V beta 6, V beta 8 and V beta 11 gene products on T cell surface. These results indicate that mechanisms other than V beta gene deletion or polymorphism can be involved in the resistance of H-2q-positive mice to experimental arthritis.

The T cell receptor repertoire influences V beta element usage in response to myoglobin

T cell clones recognizing the sperm whale myoglobin (SpWMb) epitope 110-121 in association with H-2d major histocompatibility complex class II molecules display a very limited heterogeneity of T cell receptor (TCR) V beta usage in DBA/2 mice. All clones previously tested used the same V beta 8.2 gene segment and very restricted junctional regions. To investigate the significance of this observation in vivo, we immunized DBA/2 mice with the intact SpW Mb protein or peptide 110-121. Only the V beta 8+ T cells showed any significant response to the 110-121 epitope. The response to peptide 110-121 was then analyzed in mice which, either as a consequence of antibody depletion or through genetic deletion of TCR V beta genes, lacked V beta 8+ peripheral T cells. DBA/2 mice depleted of V beta 8+ T cells by antibody treatment responded poorly to the 110-121 peptide, and only at high antigen concentrations. In contrast, DBA/2V beta a mice (homozygous for a deletion of multiple V beta gene segments including the V beta 8 family) made a response at least as great as that made by DBA/2 mice, even though the DBA/2V beta a mice had a very restricted TCR V beta repertoire compared with DBA/2 mice. Mechanisms which might determine differences in the 110-121 specific response of DBA/2, DBA/2V beta a and F23.1-treated DBA/2 mice are discussed.

Characterization of the progeny of pre-T cells maintained in vitro by IL-3: appearance in the periphery and V beta utilization in vivo

We have recently demonstrated that bone marrow-resident cells, which are able to repopulate the thymus of irradiated recipient mice (pre-T cells), can be maintained in vitro for at least 2 weeks in the presence of exogenous IL-3. Because this marrow culture system can be applied to the study of early T cell differentiation, it is important to ascertain the extent to which in vitro culture of the pre-T cells might alter the T cell progeny which can develop from them. In previous work, we showed that the progeny of cultured pre-T cells appeared to develop in a kinetically normal fashion within the thymus of recipients and that the acquisition of key developmental markers (IL-2R and CD3) was identical in the progeny of fresh and cultured pre-T cells. Here, we report the results of experiments carried out to characterize the progeny of cultured pre-T cells which were found in the peripheral lymphoid tissues several weeks following intrathymic transfer to irradiated recipients. We found no remarkable differences between the progeny of cultured or fresh marrow cells with respect to the timing of their appearance in the periphery nor their expression of CD4 or CD8. By studying the patterns of utilization of five different V beta gene products by the T cells derived from fresh or cultured bone marrow, we were able to test the susceptibility of both sets of progeny to both positive and negative selection pressures during their in vivo maturation. These experiments established that the progeny of cultured marrow cells were equally susceptible to TCR repertoire selection, as were the progeny of fresh bone marrow cells, and that the process of in vitro growth did not alter the potential TCR repertoire of the pre-T cells.

The T-cell repertoire is not dictated by self antigens alone

The influence of genetic and environmental factors on the functional cytotoxic T cell (CTL) allorepertoire was studied by comparing the CTL precursor frequencies against the same HLA alloantigens in 116 sibling pairs. A significantly different precursor frequency was found in 68%, 61% and 59% of siblings sharing 0, 1 and 2 HLA haplotypes, respectively. These data show that, although HLA is important in determining the T-cell repertoire, this is hardly reflected in the allorepertoire. Even 50% of the monozygotic twin pairs showed a significant disparity in their CTL allorepertoire, indicating that environmental factors play a role as well. The non-inherited maternal HLA antigens could be identified as one of the environmental factors shaping the CTL allorepertoire.

Human X-box-binding protein 1 is required for the transcription of a subset of human class II major histocompatibility genes and forms a heterodimer with c-fos

A complementary DNA encoding a member of the leucine-zipper class of proteins (human X-box-binding protein, hXBP-1) that binds to the 3' end of the conserved X box (X2) of the HLA-DRA major histocompatibility complex gene was recently described. Further gel-retardation analysis has demonstrated that hXBP-1 also binds to HLA-DPB X2 but not to other X2 sequences. Transient transfection of a mammalian expression vector with the hXBP-1 cDNA inserted in the antisense orientation represses the surface expression of HLA-DR and HLA-DP in Raji cells. Cotransfection of the antisense hXBP-1 vector with a HLA-DRA/chloramphenicol acetyltransferase (but not a HLA-DQB/chloramphenicol acetyltransferase) reporter plasmid decreases chloramphenicol acetyltransferase activity in Raji cells and in gamma-interferon-treated HeLa cells relative to cells cotransfected with a control antisense vector. Moreover, hXBP-1 is shown to form a stable heterodimer with the product of the c-fos protooncogene. These data suggest that the hXBP-1 c-fos heterodimer is critical for the transcription of a subset of the human class II major histocompatibility complex genes and that the regulatory mechanisms for the different class II genes are distinct.

In vivo modulation of the distribution of thymocyte subsets: effects of estrogen on the expression of different T cell receptor V beta gene families in CD4-, CD8- thymocytes

Estrogen treatment of mice has been shown to deplete CD4+, CD8+ double-positive (DP) thymocytes and to alter the relative proportion of CD4+ and CD8+ single-positive (SP) thymocytes. In this work, we have studied the effect of the steroid hormone 17 beta-estradiol (E2) on the different subsets of CD4-/CD8- double-negative (DN) thymocytes by analyzing the expression of CD5, CD3-epsilon and of several V beta gene family products of the T cell antigen receptor (TCR). After in vivo administration of E2 a significant decrease in the number and proportion of dull CD5+, CD3-, beta-TCR- DN thymocytes was observed. In contrast E2 treatment significantly increased the proportion of bright CD5+, CD3+, beta-TCR+ DN cells. The E2-induced increase in DN/TCR+ cells was observed for subsets expressing V beta 6, V beta 8, and V beta 11, but not V beta 3 gene products of the TCR. Thus, estrogen administration results in a selective inbalance of the DN thymocyte subsets by depleting an immature, dull CD5+, CD3-, TCR beta- DN subset, while enriching a mature, bright CD5+, CD3+, TCR beta+ DN subset of cells. In addition to TCR beta+ DN thymocytes, an increased proportion of CD4+ and CD8+ SP thymocytes expressing V beta 8, V beta 6, and V beta 11, but not V beta 3, TCR proteins was also observed after E2 administration. An involvement of intrathymic cytokine production in mediating the hormone action is suggested by the ability of estrogen to increase the levels of IL-1 alpha mRNA of intact thymus. Our data suggest that estrogen exerts its effects on a broad range of immature cells, including dull CD5+, CD3-, beta-TCR- DN and DP thymocytes.

Analysis of the interaction site for the self superantigen Mls-1a on T cell receptor V beta

Superantigen bound to major histocompatibility complex (MHC) products have been shown to stimulate T cells in a V beta-specific manner. Mouse T cells bearing V beta 8.1 usually respond to the self superantigen, Mls-1a, whereas T cells bearing V beta 8.2a do not. Previously, using site-directed mutational analysis, we identified the residues of natural variants of T cell receptor (TCR) V beta 8.2 that conferred Mls-1a reactivity. These residues are predicted to lie on a beta-pleated sheet of the TCR V beta element, well away from the expected binding site for antigen and MHC proteins. This study was undertaken to determine the effect of glycosylation on this beta-pleated sheet on Mls-1a reactivity and to map the extent of the interaction site on V beta 8.2 for Mls-1a. to Mls-1a, as well as to peptides derived from the conventional protein antigen, chicken ovalbumin. Here we demonstrate that first, N-linked carbohydrate on the lateral surface of V beta blocks the interaction of the TCR V beta with the self superantigen, Mls-1a, but has no effect on the TCR interaction with peptide antigen and MHC, second, that the interaction site for Mls-1a extends over the surface of the solvent-exposed beta-pleated sheet on the side of the TCR, and third, that mutations which affect both superantigen and peptide antigen reactivity lie at the beginning of the first complementarity determining region of V beta, consistent with models of the trimolecular complex of TCR-peptide-MHC.

Alteration of the T-cell receptor repertoire in A.CA mice expressing an Ead transgene

In an effort to generate an A.CA mouse expressing Ed, the Ead gene has been introduced into A.CA mice which lack the major histocompatibility complex (MHC) class II E molecule. Flow cytometric analysis shows cell surface expression of the E alpha chain on lymphocytes and macrophages in the transgenic mice. Analysis of T-cell receptor (Tcr) genes deleted in some E-expressing mouse strains demonstrates that T cells expressing Tcrb-V5 are partially deleted in these transgenic mice while those expressing Tcrb-V8 and Tcrb-11 are not. In addition, the expressed E alpha d chain can promote Mycoplasma arthriditis mitogen (MAM)-induced T-cell proliferation. The expression of the E alpha chain, presumably as an A beta fE alpha d heterodimer, can alter the peripheral T-cell repertoire and T-cell reactivity to a microbial superantigen.

Human HLA-DQ beta chain presents minor lymphocyte stimulating locus gene products and clonally deletes TCR V beta 6+, V beta 8.1+ T cells in single transgenic mice

Minor lymphocyte stimulating locus (Mls) gene products in association with mouse major histocompatibility complex (MHC) class II molecules are known to determine the repertoire of T-cell receptor (TCR) in mature T cells. In order to test whether human class II molecules can present mouse Mls, HLA-DQ beta transgenic mice were generated. The expression and function of the DQ beta transgene were studied in the progeny of one selected founder which was H-2f and H-2E negative. In these mice, DQ beta molecules pairing with mouse A alpha chain and invariant chain are expressed on the cell surface in a tissue-specific manner. When the DQ beta gene was bred into the Mls-1a strain DBA/1 (H-2q), T cells bearing V beta 6 and V beta 8.1 TCR were clonally deleted in the thymus of DQ beta+ transgenics but not in DQ beta-negative full sibs. Thus, the data presented here clearly demonstrate that the human MHC DQ beta chain can present Mls in the clonal deletion of T cells. Our results also suggest the requirement for an interaction between CD4 and class II molecules (alpha chain) for clonal deletion of T cells to occur.

Thymic stromal cells in culture. I. Establishment and characterization of a line which is cytotoxic for normal thymocytes and produces hematopoietic growth factor(s)

Lines of thymic stromal cells have been established. One of these, designated TS-9, has been cloned and studied extensively. This line expresses both acid and alkaline phosphatases. Despite repeated cloning, TS-9 cells remain morphologically heterogeneous. The origin of these cells is not clear. They express low levels of immunologically identifiable cytokeratins, produce laminin, a basement membrane protein, but express antigens typically found on bone marrow stromal cells. The TS-9 cells are MHC Class I+ but Class II-. They express the Thy-1, Pgp-1, and Mac-2 antigens but not other lineage markers of T cells or macrophages. Coculturing TSC with normal thymocytes or with the CTLL-1 cell line leads to a profound inhibition of lectin-induced and/or IL-2 induced T cell proliferation. This requires direct cell-cell contact and ultimately results in the death of the bound lymphocytes. It cannot be reproduced by culturing the thymocytes with TSC culture supernatants. These supernatants do contain hematopoietic growth factor(s) which augment the growth of some T lineage cells and support the growth of monocytic colonies in semi-solid culture medium. Both normal thymocytes and a variety of T cell tumors bind to TSC but only the normal cells are killed as a consequence of this interaction. Neither the binding nor the killing appear to be MHC restricted. We suggest that this killing may provide a model for the effector mechanism of the negative selection imposed by the thymus on developing T cells.

The genetic basis of negative selection of Tcrb-V11+ T cells

Non-H-2 genes responsible for negative selection of Tcrb-V11+ T cells were examined using backcross mice of various strains with C58, which does not delete Tcrb-V11+ T cells. Two independently segregating genes were found: one leading to partial deletion was closely linked to Ly-2/Ly-3 on chromosome 6, and the second giving virtually complete deletion has not yet been mapped. The A strain had only the former, whereas BALB/c, BALB.K, B10.BR, CBA-T6, C3H/He, and DBA/2 expressed both of these genes. Although a gene(s) of the NIH strain led only to partial deletion, the chromosomal localization of the gene(s) has not yet been determined: no informative polymorphic molecules are expressed from genes on chromosome 6 of this strain.

Ablation of "tolerance" and induction of diabetes by virus infection in viral antigen transgenic mice

To address the mechanisms of tolerance to extrathymic proteins, we have generated transgenic mice expressing the lymphocytic choriomeningitis viral (LCMV) glycoprotein (GP) in the beta islet cells of the pancreas. The fate of LCMV GP-specific T cells was followed by breeding the GP transgenic mice with T cell receptor transgenic mice, specific for LCMV and H-2Db. These studies suggest that "peripheral tolerance" of self-reactive T cells does not involve clonal deletion, clonal anergy, or a decrease in the density of T cell receptors or accessory molecules. Instead, this model indicates that self-reactive cytotoxic T cells may remain functionally unresponsive, owing to a lack of appropriate T cell activation. Infection of transgenic mice with LCMV readily abolishes peripheral unresponsiveness to the self LCMV GP antigen, resulting in a CD8+ T cell-mediated diabetes. These data suggest that similar mechanisms may operate in several so-called "T cell-mediated autoimmune diseases."

A virus-encoded "superantigen" in a retrovirus-induced immunodeficiency syndrome of mice

The development of an immunodeficiency syndrome of mice caused by a replication-defective murine leukemia virus (MuLV) is paradoxically associated with a rapid activation and proliferation of CD4+ T cells that are dependent on the presence of B cells. The responses of normal spleen cells to B cell lines that express the defective virus indicated that these lines express a cell surface determinant that shares "superantigenic" properties with some microbial antigens and Mls-like self antigens. This antigen elicited a potent proliferative response that was dependent on the presence of CD4+ T cells and was associated with selective expansion of cells bearing V beta 5. This response was markedly inhibited by a monoclonal antibody specific for the MuLV gag-encoded p30 antigen.