T cell tolerance by clonal elimination in the thymus

Alteration of idiotypic connectivity in prenatally tolerized neonatal mice

Prenatal tolerization with trinitrobenzenesulphonic acid (TNBS) leads to expansion of trinitrophenyl (TNP)-specific B cells, the majority of which become refractory to stimulation during postnatal development. One possible explanation could be that they belong to the repertoire of naturally activated B cells which are limited in expansion after antigenic stimulation due to a high degree of idiotypic connectivity. To evaluate this hypothesis, 59 thymus- and 490 spleen-derived B-cell hybridomas from 6-day-old prenatally untreated and prenatally TNBS-treated mice were tested for reactivity against 33 arbitrarily chosen clones derived from the same fusions, 17 being derived from control and 16 from tolerized litters. Two major points could be deduced: (1) Idiotypic connectivity, including connectivity of TNP- and anti-TNP-reactive monoclonal antibodies (MoAb), was maintained after prenatal tolerization. This accounted for thymus- and spleen-derived MoAb. (2) Only TNP- and anti-TNP-reactive MoAb derived from prenatally untreated and prenatally tolerized mice displayed significantly distinct idiotypic profiles. Differences were pronounced, especially with thymus-derived MoAb. Thus, TNP-specific B cells in prenatally tolerized newborns do not behave like B cells of adult mice stimulated by external antigen, but rather like a part of the naturally activated, idiotypically connected B-cell repertoire of the newborn. This could explain B-cell unresponsiveness at older age as a consequence--at least partly--of their high idiotypic connectivity.

Correlation between the V beta 4+ CD8+ T-cell population and the H-2d haplotype

The V beta 4+ T-cell population was examined with a newly established antibody, KT4, specific for V beta 4. Between 4.8% and 19.4% of CD3+ peripheral T cells from various inbred strains of mice or F1 hybrids expressed V beta 4. The CD4 T-cell population had higher numbers of V beta 4+ T cells (5.5%-20.6%) than the CD8 T-cell population (2.5%-10.7%). Deletion of certain V beta-expressing T cells due to the presence of the Mlsa antigen and/or the absence of certain Tcrb-V genes increased relative numbers of V beta 4+ T cells. The data suggest that V beta 4+ CD8+ T cells might be positively selected by H-2d molecules.

Analysis of type II collagen-reactive T cells in the mouse. I. Different regulation of autoreactive vs. non-autoreactive anti-type II collagen T cells in the DBA/1 mouse

The T cell reactivity against type II collagen (CII) was analyzed in the collagen-induced arthritis-susceptible mouse strain DBA/1. It was shown that the proliferative response in lymph node cells from rat CII-immunized mice was mainly directed against a foreign determinant present on all heterologous CII tested but not on autologous CII. A T cell line with this reactivity reacted with high sensitivity with CII and the determinant was mapped to the CB11 fragment of CII. A weak autoreactive response could be detected in the primary cultures using high concentrations of mouse CII and this reactivity remained after several stimulations with high concentrations of rat CII but not with low concentrations of rat CII. A similar response against mouse CII but with only limited cross-reactivity to rat CII was seen when culturing the cells with mouse CII as antigen. The optimal concentration for the autoreactive response was always more than 100-fold higher than for the response of the T cells specific for heterologous CII. An anti-CII T cell response could also be detected in spleen cells from unimmunized mice and the strongest response was obtained using autologous CII. These results suggest that T cells recognizing self CII are normally activated in the DBA/1 mouse and possibly as a consequence exhibit a clonal anergy pattern with a weak proliferative response only at high concentrations of CII.

Cell components required for deletion of an autoreactive T cell repertoire

T cells become tolerant to self antigens during their development in the thymus. Clonal deletion of thymocytes bearing T cell receptor (TcR) which recognize self antigens is a major mechanism for generating tolerance. In the present study we have used allogeneic bone marrow (BM) chimeras, prepared with various combinations of mouse strains and focusing especially on expressions of I-E molecules and Mls-1a antigens on the cell surface, to investigate both immunohistochemically and by flow cytometry the cell components that contribute to the clonal deletion of T cells positive for V beta 6 TcR. The V beta 6 TcR expression is strongly associated with T cell recognition of both I-E and Mls-1a antigens. We found that I-E+ cells derived from donor BM (and thus not of recipient lineage) represented a primary requirement for deletion of Mls-1a-reactive thymocytes which bear V beta 6 TcR. Immunohistochemical analysis revealed that the donor-derived I-E+ cells were distributed mainly to the thymic medulla and that the V beta 6+ cells were eliminated from the thymic medulla between 2 and 3 weeks following BM transplantation. In contrast, Mls-1a+ cells of either donor or recipient origin might be responsible for the deletion, even though cortical epithelial cells appeared not to express Mls-1a antigens.

Evidence for the existence of distinct heterogeneity among the peripheral CD4-CD8- T cells from MRL-lpr/lpr mice based on the expression of the J11d marker, activation requirements, and functional properties

Autoimmune-susceptible, MRL-lpr/lpr (lpr) mice develop a profound lymphadenopathy resulting from the accumulation of CD4-CD8- (double-negative, DN) cells in peripheral lymphoid organs. The source and the mechanism of this abnormal accumulation of cells is still unknown. Recently, we reported that a significant number (approximately 35%) of the CD4-CD8- cells expressed J11d, a marker expressed by immature thymocytes but not by mature functional peripheral T cells. In the present study, we investigated the phenotype, growth requirements, and functional properties of purified J11d+ and J11d- subpopulations. Using the mAb, F23.1, which recognizes a TCR determinant encoded by the V beta 8 gene family, it was observed that approximately 30% of the J11d+ and J11d- DN cells expressed this determinant. Further studies on the thymus revealed that J11d+ DN cells from lpr thymus also contained F23.1+ cells (approximately 25%), whereas, similar cells from normal MRL(-)+/+mice were all F23.1-, consistent with earlier reports in other normal strains. Further phenotypic studies revealed that the peripheral J11d+ and J11d- cells from lpr mice were similar in expressing CD3, Ly-5 (B220), and Ly-24 (Pgp-1) determinants. When stimulated with phorbol myristic acetate (PMA) and recombinant IL-2 (rIL-2), only J11d- cells but not J11d+ cells responded by proliferation. However, in the presence of calcium ionophore (A23187) and PMA, both J11d+ and J11d- subpopulations proliferated by producing and responding to endogenous IL-2 but not IL-4. The lymph node T cells from 1-month-old MRL-lpr/lpr mice responded strongly when stimulated with PMA + rIL-4 or PMA + rIL-6. In contrast both J11d+ and J11d- subpopulations failed to respond when similarly stimulated. The J11d+ but not J11d- cells demonstrated spontaneous cytotoxic activity against the NK-sensitive YAC-1 tumor targets. The J11d- cells did not exhibit cytotoxic potential in spite of culture with PMA + rIL-2. Even after repeated culture in vitro with PMA + A23187 or PMA + rIL-2, both J11d+ and J11d- subpopulations failed to express the mature phenotype bearing CD4 and/or CD8 antigens. The present study demonstrates the expansion of unique J11d+, alpha beta-TCR+, DN T cells with cytotoxic potential in lpr mice and further suggests the existence of phenotypic and functional heterogeneity among the abnormal lpr DN cells.

Death and transfiguration of cortical thymocytes: a reconsideration

The fate of most cells in the mammalian thymus, namely CD4+CD8+ (double positive) cortical cells, is unknown. A fraction of the cells in this population serve as developmental intermediates, but the vast majority will die. In this article Ellen Rothenberg discusses how and when the cells are condemned or promoted to maturity, in the light of intriguing new data that suggests that they may be able to be rescued for longer than was previously thought.

Inhibition of T cell development in thymic organ culture: implications for the mechanism of action of cyclosporin A

We have examined the effects of the immunosuppressive drug cyclosporin A (CsA) on the phenotypic maturation of T cells in thymic organ cultures begun at day 16 of gestation. CsA specifically inhibited the generation of cells expressing high levels of alpha/beta TcR/CD3 complexes and a mature phenotype defined by CD4 and CD8 surface markers. Adding interleukin (IL) 1 beta, IL 2 or IL 4 failed to reverse the effects of CsA, and major histocompatibility complex class II expression in the thymic medulla was preserved. Possible mechanisms of CsA-mediated inhibition of T cell development are discussed.

Origin and selection of peripheral CD4-CD8- T cells bearing alpha/beta T cell antigen receptors in autoimmune gld mice

We have analyzed the origin and development of unusual CD4-CD8- alpha/beta T cell receptor-positive peripheral T cells produced in large numbers by mice homozygous for the gld mutation (C3H-gld/gld). These mice may be an important model for investigating processes controlling T cell development. Bone marrow transfers demonstrated that the gld defect was intrinsic to bone marrow-derived cells. Clonal deletion of potentially autoreactive cells was observed in peripheral gld CD4-CD8-, CD4+CD8-, and CD4-CD8+ T cells, as well as mature thymocytes. This suggests that gld CD4-CD8- T cells have passed through the thymus in ontogeny and that gld autoimmunity does not result from a general defect in elimination of self-reactive thymocytes. These observations, combined with demethylation of the CD8 gene in the CD4-CD8- population, support prior expression of CD4 and/or CD8 in gld CD4-CD8- T cell ontogeny, perhaps at a CD4+CD8+ stage. Steroid sensitivity of gld thymocytes and CD4-CD8- T cells was normal. Therefore, we found no gross abnormalities in two major mechanisms of inducible cell death in the gld thymus, the clonal deletion process associated with tolerance and the steroid-inducible endogenous endonuclease thought to be involved in apoptosis of unselected thymocytes. The data suggest that if gld CD4-CD8- T cells arise via escape from normal elimination in the thymus, they must do so by a novel defect in thymic selection (perhaps related to aberrant positive signals) and/or are expanded by an extrathymic process which allows clonal deletion to occur.

Kinetics of mature T-cell development in the thymus

We have reexamined the balance between cell birth, cell maturation, and cell death in the thymus by labeling dividing thymocytes and their progeny in vivo with [3H]-thymidine, isolating clearly defined subpopulations by fluorescence-activated cell sorting, and determining the distribution of label by autoradiography. When mature thymocytes were precisely defined (as CD4+CD8- CD3+ or CD4-CD8+ CD3+) and separated from immature single positives (CD4+CD8- CD3- and CD4-CD8+ CD3-), a lag was observed in the rate of entry of [3H]thymidine into mature cells. Thus, many of the mature thymocytes appear to derive from a small nondividing cortical thymocyte pool, rather than originating directly from the earliest dividing CD4+CD8+ blasts. There was little evidence for cell division during or after mature thymocyte formation, suggesting a one-for-one differentiation from cortical cells rather than selective clonal expansion. The rate of production of mature single positive thymocytes agreed closely with estimates of the rate of export of mature T cells from the thymus and was only 3% of the rate of production of double-positive cortical thymocytes. This was compatible with a stringent selection process and extensive intrathymic cell death and suggested that no extensive negative selection occurred after the mature cells were formed.

Strong T cell tolerance in parent----F1 bone marrow chimeras prepared with supralethal irradiation. Evidence for clonal deletion and anergy

T cell tolerance induction was examined in long-term H-2-heterozygous parent----F1 chimeras prepared with supralethal irradiation (1,300 rad). Although these chimeras appeared to be devoid of host-type APC, the donor T cells developing in the chimeras showed marked tolerance to host-type H-2 determinants. Tolerance to the host appeared to be virtually complete in four assay systems: (a) primary mixed lymphocyte reactions (MLR) of purified lymph node (LN) CD8+ cells (+/- IL-2); (b) primary MLR of CD4+ (CD8-) thymocytes; (c) skin graft rejection; and (d) induction of lethal graft-vs.-host disease by CD4+ cells. Similar tolerance was observed in chimeras given double irradiation. The only assay in which the chimera T cells failed to show near-total tolerance to the host was the primary MLR of post-thymic CD4+ cells. In this assay, LN CD4+ cells regularly gave a significant antihost MLR. The magnitude of this response was two- to fourfold less than the response of normal parental strain CD4+ cells and, in I-E(-)----I-E+ chimeras, was paralleled by approximately 70% deletion of V beta 11+ cells. Since marked tolerance was evident at the level of mature thymocytes, tolerance induction in the chimeras presumably occurred in the thymus itself. The failure to detect host APC in the thymus implies that tolerance reflected contact with thymic epithelial cells (and/or other non-BM-derived cells in the thymus). To account for the residual host reactivity of LN CD4+ cells and the incomplete deletion of V beta 11+ cells, it is suggested that T cell contact with thymic epithelial cells induced clonal deletion of most of the host-reactive T cells but spared a proportion of these cells (possibly low affinity cells). Since these latter cells appeared to be functionally inert in the thymus (in contrast to LN), we suggest that the thymic epithelial cells induced a temporary form of anergy in the remaining host-reactive thymocytes. This anergic state disappeared when the T cells left the thymus and reached LN.

Ek alpha transgene in B10 mice suppresses the development of myasthenia gravis

Mice bearing the H-2b haplotype are susceptible to the development of experimental autoimmune myasthenia gravis (EAMG), induced by acetylcholine receptor (AChR) autoimmunity. One of the genes influencing EAMG susceptibility has been mapped to the Ab locus of the major histocompatibility complex, and the A beta chain has been implicated in the pathogenesis. Mice of the H-2b haplotype, including C57BL/10 (B10), have a genomic deletion of the E alpha gene and therefore fail to express the E molecule on their cell surface. To test the hypothesis that failure to express the cell surface E molecule in B10 mice contributes to EAMG pathogenesis, Ek alpha transgenic B10 mice expressing the E molecule were examined. Expression of the E molecule in Ek alpha transgenic B10 mice partially prevented the development of EAMG.

Tolerance and diabetes in transgenic mice over-expressing class I histocompatibility molecules in pancreatic beta cells

The Class I gene, H-2Kb, was linked to the rat insulin promoter and the construct inoculated into fertilized mouse eggs to produce lines of transgenic mice. Mice which expressed the Class I molecule in the beta cells of the pancreas developed diabetes and progressive loss of their pancreatic beta cells. This occurred whether the transgene product was syngeneic or allogeneic with respect to its host. No lymphocytic infiltration was ever seen in transgene expressing mice, even in those deliberately immunized with H-2Kb-bearing cells. When the transgene product was allogeneic, spleen cells from the transgenic mice stimulated in vitro with irradiated B10.A(5R) cells (KbDd), could kill H-2d targets in vitro, but not targets bearing H-2Kb. Responsiveness of spleen cells to H-2Kb targets returned with advancing age, as the severity of diabetes increased. The results indicate that diabetes in this model occurs independently of the immune system, and point to an extra-thymic mechanism of tolerance induction dependent on the continuous presence of antigen.

T-cell receptor beta-chain DNA polymorphism frequencies in healthy HLA-DR homozygotes

In view of numerous recent reports of T-cell receptor (TCR) beta-chain/disease associations with HLA-associated diseases, we tested the possibilities that associations might exist directly between these two gene complexes at the level of the germline DNA. We determined frequencies of five TCR-beta DNA polymorphisms in 33 HLA-DR2/2 homozygotes, 29 HLA-DR3/3 homozygotes and 42 HLA-DR4/4 homozygotes. The control population (n = 74) was chosen without "bias toward" their HLA-DR genes. We selected DR2, DR3 and DR4 homozygotes because they have been the most frequently involved in HLA-DR associated diseases. Our results indicate that the recent reports in the literature of TCR-beta/disease associations can not be explained by a significantly different distribution of TCR-beta genes in HLA-DR2+, -DR3+, or -DR4+ subpopulations. Our results also suggest that if co-evolution between TCR-beta and MHC haplotypes does exist, the selective pressures in recent generations have not been strong enough to significantly alter the germline TCR-beta gene frequencies in HLA-DR2+, -DR3+, or -DR4+ subpopulations.

BK virus terminates tolerance to dsDNA and histone antigens in vivo

In order to characterize the immune response to BK virus, a human polyomavirus containing dsDNA and host cell histones, we followed the appearance of antibodies in five outbred rabbits after intravenous inoculation with purified infectious BK virus without any adjuvant. The animals were followed for 15 weeks after the first inoculation and booster doses were given after four and eight weeks. Antibodies were studied by ELISA techniques with the BK virus particle, dsDNA, ssDNA or the individual histones as test antigens. Antibodies to BK virus structural proteins were detected in all rabbits. Two out of five rabbits produced antibodies to dsDNA, ssDNA, nucleosomes and histones H1 and H3. Even a weak reactivity to H2B was detected in one serum. The autoantibody response was transient as it declined after a few weeks, but it reappeared after a second boost in one of the rabbits. The other animals did not respond in the same manner. The specificity of the antibodies against dsDNA was ascertained by inhibition studies employing S1 nuclease treated DNA as inhibitor. Furthermore, the dsDNA used as coating antigen was not recognized by a human reference serum with known specificity for ssDNA. The rabbit antisera did not show any reactivity to a panel of other (in this context irrelevant) autoantigens. This suggests that the anti-DNA and -histone antibodies are not a result of non-specific polyclonal B cell activation. Thus, inoculation of dsDNA viruses may represent a new model that allows us to investigate mechanisms responsible for circumvention of tolerance to self molecules.

Novel origin of lpr and gld cells and possible implications in autoimmunity

The lpr and gld mutations are prime examples of single-gene defects associated with expansion of a unique double-negative (CD4-8-), T-cell receptor alpha:beta + cell population and heightened polyclonal and autoimmune responses. The exact origin of these autoimmunity-inducing/enhancing T cells remains controversial. Here, we review the characteristics of the lpr and gld mutations, and speculate on the possible relationship of these cells to normal thymic differentiation pathways. We argue that mounting evidence now supports the existence of a CD4/CD8-loss pathway of late thymic differentiation, responsible for the origin of both normal and lpr/gld double-negative alpha:beta + cells. We further speculate that downregulation of CD4 and CD8 accessory molecules on thymocytes with moderately autoreactive T-cell receptors is involved in selecting cells, including lpr/gld precursors for this pathway. Escape of a large number of such autoreactive cells from thymic elimination might be an important contributory factor to the pathogenesis of autoimmunity.

Characterization of V beta-bearing cells in athymic (nu/nu) mice suggests an extrathymic pathway for T cell differentiation

In the present article, the expression of the T cell receptor (TcR) beta chain and other T cell molecules was evaluated in surface immunoglobulin-negative spleen cell populations of young and old BALB/c and C57BL/6 nude mice, using a panel of monoclonal antibodies. The results obtained show that in young nude mice, most Thy-1high cells do not express other T cell markers. These mice have, however, a sizable population of Thy-1low cells with the same phenotype of alpha/beta+, CD4-CD8- thymocytes or MRL/lpr peripheral T cells, expressing predominantly genes of the V beta 8 family. The evolution of alpha/beta+ cells in aging nudes is strongly suggestive of an extrathymic pathway of differentiation of these cells since (a) the acquisition of high density TcR and CD3, as well as Thy-1 or CD4CD8 antigens at the cell surface of nude V beta+ T cells is not simultaneous; (b) alpha/beta+ cells in nude mice co-express other T cell markers at random and, even in old mice, they never completely resemble to the predominant high Thy-1+ CD3+ TcR alpha/beta+, CD4+CD8+ cells of euthymic controls; and (c) BALB/c nude T cells express V beta 11 genes, that are deleted in euthymic BALB/c mice. This latter finding may also indicate differences in the mechanisms of selection of T cells specificities in the thymus vs. the peripheral pools.

CD4+CD8+ thymocytes are susceptible to DNA fragmentation induced by phorbol ester, calcium ionophore and anti-CD3 antibody

Stimulation of murine thymocytes with phorbol ester or calcium ionophore for 18-24 h resulted in 70%-80% fragmentation of DNA into 180-200-bp multiples, followed by cell death. Experiments with fractionated subpopulations by panning or flow cytometry revealed that DNA fragmentation was selectively observed in CD4+CD8+ cells and in a portion of CD4-CD8+ cells. To investigate whether DNA cleavage is also inducible via antigen-specific receptors, thymocytes were incubated in wells precoated with anti-CD3 antibody. An approximately 20% increase of DNA fragmentation was constantly observed when unseparated thymocytes were stimulated with anti-CD3 antibody. In this anti-CD3-induced DNA degradation, CD4+CD8+ cells are probably the target cells, since (a) fetal thymocytes at day 18 of gestation were found vulnerable to anti-CD3-induced DNA cleavage and (b) flow cytometry analysis of viable cells recovered after cultivation in the anti-CD3-coated wells revealed that CD4+CD8+ cells were preferentially decreased. Further experiments with purified CD4+CD8+ cells, however, could not define a clear-cut increase of DNA fragmentation when isolated CD4+CD8+ cells were stimulated with anti-CD3 antibody. Addition of interleukin (IL) 1, IL 2, IL 3, IL 4 or interferon-gamma to the CD4+CD8+ cell cultures failed to yield a DNA cleavage similar to that of unseparated thymocytes.

Limit of T cell tolerance to self proteins by peptide presentation

Cytotoxic T lymphocytes (CTLs) recognize foreign peptides bound to major histocompatibility complex (MHC) class I molecules. MHC molecules can also bind endogenous self peptides, to which T cells are tolerant. Normal mice contained CTLs specific for self peptides that were from proteins of ubiquitous or tissue-restricted expression. In vivo, these endogenous self peptides are not naturally presented in sufficient density by somatic cells expressing MHC class I molecules. They can, however, be presented if added exogenously. Thus, our data imply that CTLs are only tolerant of those endogenous self peptide sequences that are presented by MHC class I-positive cells in a physiological manner.

Thymic epithelium tolerizes for histocompatibility antigens

The role of thymic epithelium in the establishment of tissue tolerance was analyzed with a murine chimeric system. All T cells differentiated from birth onward in a thymus comprising allogeneic epithelium and syngeneic hematopoietic cells. Embryonic thymic rudiments that contained no hematopoietic cells from C3H (H-2k) donors were grafted to newborn athymic (nude) BALB/c (H-2d) mice. Chimeras that had normal T cell numbers and function rejected third-party skin grafts, but permanently accepted grafts syngeneic to the thymic epithelium. In vitro functional assays did not always correlate with the state of tolerance in vivo. Thus, pure thymic epithelium induces tolerance to histocompatibility antigens.

Thymic epithelium tolerizes for histocompatibility antigens

The role of thymic epithelium in the establishment of tissue tolerance was analyzed with a murine chimeric system. All T cells differentiated from birth onward in a thymus comprising allogeneic epithelium and syngeneic hematopoietic cells. Embryonic thymic rudiments that contained no hematopoietic cells from C3H (H-2k) donors were grafted to newborn athymic (nude) BALB/c (H-2d) mice. Chimeras that had normal T cell numbers and function rejected third-party skin grafts, but permanently accepted grafts syngeneic to the thymic epithelium. In vitro functional assays did not always correlate with the state of tolerance in vivo. Thus, pure thymic epithelium induces tolerance to histocompatibility antigens.

Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs

p56lck, a lymphocyte-specific member of the src family of cytoplasmic protein-tyrosine kinases, is associated noncovalently with the cell surface glycoproteins CD4 and CD8, which are expressed on functionally distinct subpopulations of T cells. Using transient coexpression of p56lck with CD4 or CD8 alpha in COS-7 cells, we show that the unique N-terminal region of p56lck binds to the membrane-proximal 10 and 28 cytoplasmic residues of CD8 alpha and CD4, respectively. Two cysteine residues in each of the critical sequences in CD4, CD8 alpha, and p56lck are required for association. Our results suggest a novel role for cysteine-mediated interactions between unrelated proteins and provide a model for the association of other src-like cytoplasmic kinases with transmembrane proteins.

Self-tolerance to transgenic gamma delta T cells by intrathymic inactivation

During their intrathymic differentiation, T lymphocytes expressing alpha beta T-cell receptors (TCR) are negatively and positively selected. This selection contributes to the establishment of self-tolerance and ensures that mature CD4+ and CD8+ cell populations are restricted by the self major histocompatibility complex. Little is known, however, about gamma delta T-cell development. To investigate whether selection operates in the establishment of the gamma delta T-cell class, we have generated transgenic mice using gamma- and delta-transgenes encoding a TCR that is specific for a product of a gene in the TL-region of the TLb haplotype. Similar numbers of thymocytes expressing the transgenic TCR were generated in mice of TLb and TLd haplotypes. But gamma delta thymocytes from TLb and TLd transgenic mice differed in cell size, TCR density and in their capacity to respond to TLb stimulator cells or interleukin-2 (IL-2). In contrast to gamma delta T cells from TLd transgenic mice, gamma delta T cells from TLb transgenic mice did not produce IL-2 and did not proliferate in response to TLb stimulator cells, but they did proliferate in the presence of exogenous IL-2. These results indicate that functional inactivation of self-antigen-specific T cells could contribute to the establishment of self-tolerance to thymic determinants.

Stimulation of all T cells bearing V beta 1, V beta 3, V beta 11 and V beta 12 by staphylococcal enterotoxin A

To determine the molecular mechanisms of T cell stimulation by staphylococcal enterotoxin A (SEA), we examined the expression of T cell receptor (TcR) V beta on the T cells from four strains of mice stimulated in vitro with SEA, using flow cytometric analysis for the number of T cells bearing V beta 3, V beta 6, V beta 8, V beta 11 and RNA blotting analysis for the amount of transcripts of V beta 1, V beta 5 and V beta 12. The number of T cell blasts bearing V beta 1, V beta 3, V beta 1 or V beta 12 were increased in the T cell blasts proliferating in vitro in response to SEA in C57BL/6 mice. In AKR/J mice, which contain few V beta 11- or V beta 12-bearing T cells due to a tolerance to the self-MHC class II IE-antigens, T cells bearing V beta 1 or V beta 3 responded to SEA. SEA enriched only V beta 1-bearing T cells in BALB/c mice carrying Mls-2a which lack Mls-1a-reactive V beta 3-bearing T cells as well as V beta 11- and V beta 12-bearing T cells. In spite of the presence of V beta 1-bearing T cells, C3H/He T cells exhibited a very low responsiveness to SEA. T cell repertoires skewed by clonal deletion of self-reactive T cells may in part account for the different sensitivity to SEA among the different strains. A tolerance to SEA can be established in C57BL/6 mice which have been primed i.v. with SEA and treated i.p. with 200 mg/kg of cyclophosphamide 2 days later. All mature T cells bearing V beta 3 or V beta 11 were virtually abolished in the periphery of tolerant mice. These results suggest that most T cells reactive to SEA bear V beta 1, V beta 3, V beta 11 or V beta 12 and that clonal deletion of mature T cells reactive to SEA may account for the cellular mechanisms for cyclophosphamide-induced tolerance to SEA.

Autoimmunity and the pathogenesis of glomerulonephritis

Self-tolerance is maintained by: thymic influences on developing T cells; peripheral mechanisms that can tolerise post-thymic T cells; and to a variable extent the tolerisation of potentially autoreactive B cells. The presence of autoreactive T cells in normal individuals suggests that mechanisms to control the activity of such cells may be important. Failure of any of these processes may lead to autoimmunity. The relationship between glomerulonephritis and the mechanisms leading to breakdown of self-tolerance remains elusive. An important observation is that autoimmune diseases are strongly associated with particular products of the major histocompatibility complex (MHC). This association may reflect the intimate involvement of the MHC in thymic T cell education. Another explanation is that T cells only recognise antigens presented in the context of MHC molecules. Although there has been progress in identifying the targets recognised by autoantibodies in vasculitis and anti-GBM disease, nothing is known about the T cells involved. Despite our ignorance, therapy aimed specifically at the T cell can be effective. This approach is being supplemented by attempts to engage immunoregulatory mechanisms, such as idiotype-antiidiotype interactions. The hope is that such treatments, or combinations thereof, will allow a more focused suppression of the autoimmune response, in contrast to the non-specific (and therefore potentially dangerous) methods of immunosuppression available at present.

A novel autoregulatory cytokine is required for the regulation of autoaggressive responses

Limiting dilution studies indicate that cells with the potential to lyse autologous target cells exist in the peripheral blood of all normal individuals. In contrast to allocytotoxic cells, autocytolytic cells are down-regulated by a second less frequent cell population. When recombinant interleukin 2 is substituted for crude lymphocyte conditioned medium in these limiting dilution experiments, autocytotoxicity develops normally. Under these conditions, however, the autocytotoxic response is not down-regulated. Mixing crude lymphocyte-conditioned medium together with recombinant interleukin 2 restores the regulation of autocytotoxicity normally seen at high responder cell dose. These findings indicate that a second soluble factor present in the conditioned medium is necessary either for the activation, growth, or differentiation of the regulatory cell population or alternatively, to render the cytotoxic population responsive to the activity of regulatory cells. Gel filtration studies indicate that the molecular weight of this factor is between 60 and 80 kd. This factor appears to be distinct from known immunologically active cytokines. It is conceivable that deficiencies of this cytokine may be relevant to the pathogenesis of autoimmune diseases or graft-versus-host reactions.

Intrathymic T cell repertoire after prenatal trinitrobenzene-sulfonic acid-treatment

It is still a matter of debate, whether tolerance toward self-non-MHC antigens is due to intrathymic deletion or to regulatory processes in the periphery. To further pursue this question, responsiveness toward TNP and an anti-TNP monoclonal antibody (Sp6) carrying a recurrent idiotype was evaluated in prenatally trinitrobenzenesulfonic acid (TNBS)-treated mice. In prenatally untreated as well as in TNBS-treated mice, thymocytes proliferating in the absence of nominal antigen were double negative (L3T4-/Lyt2-), but antigen-specific thymocytes were single positive (L3T4+/Lyt2- or L3T4-/Lyt2+). TNBS-treated mice differed from controls inasmuch as in their first week of life T cells proliferating in response to TNP were found in the thymus and detected at increased frequencies in the spleen. The frequency of TNP-specific thymocytes and spleen cells declined rapidly, finally reaching in the spleen a level of 20-30% of controls. Furthermore, after antigenic stimulation, the frequency of thymocytes and spleen cells proliferating in response to TNP was found to be increased in control mice, but TNP-specific T cell were no more recovered in the thymus or the spleen of tolerized mice. The same accounted for thymic and splenic T cells proliferating in response to Sp6. They were expanded in control mice after antigenic stimulation, but were undetectable in TNBS-treated mice. Thus, T cells with specificity for an internal (Sp6) and an external (TNP) antigen, provided the latter was present during ontogeny, were detected in the thymus of control and, transiently, in the thymus of tolerized mice. But, the fate of antigen-specific thymocytes was different in prenatally untreated and TNBS-treated mice. The data are interpreted in the sense that tolerance toward non-MHC antigens may be acquired subsequently to tolerance toward self-MHC antigens and possibly after imprinting of antigen specificity.

The role of germline gene expression and somatic mutation in the generation of autoantibodies to DNA

Several distinctive features of anti-DNA autoantibodies have been identified by a detailed analysis of the available heavy and light chain sequences. They include unique VH gene segments that are not normally expressed in antibodies to external antigens, somatic mutations which may serve to change the antigenic specificity as well as to increase affinity, a less stringent choice of light chains, and a unique basic peptide in the heavy chain CDR3. It is proposed that in the majority of cases, the regulatory mechanism of self-tolerance in the healthy animal operates via VH gene expression to prevent the synthesis of potentially high affinity anti-DNA autoantibodies.

Sequence elements required for activity of a murine major histocompatibility complex class II promoter bind common and cell-type-specific nuclear factors

We have examined the sequence elements and corresponding DNA-binding factors required for transient expression of the A alpha d promoter fused to the bacterial chloramphenicol acetyltransferase reporter gene in a variety of cultured cell lines. Deletion analysis demonstrated that only about 110 nucleotides of sequence 5' of the transcription start site are required for constitutive expression in the murine B-lymphoma cell line A20 or for gamma interferon-induced expression in the murine monocytic cell line WEHI-3. Linker-scanner mutation of this region indicated that at least three sequence elements are required for promoter activity. These elements correspond to the conserved sequence elements found in other human and mouse class II genes, the X box, the Y box, and the H box. Analysis of DNA-binding activity showed that the three most predominant factors present in extracts from WEHI-3, A20, or L cells (which do not express the class II genes) are actually a family of factors that bind to a fourth sequence element, overlapping the 3' end of the X-box sequence, that is homologous to the cyclic AMP-responsive enhancer element. A single common factor that binds to the Y box was detected in extracts from all cells tested, as has been seen with the Y-box elements of other class II genes. Another common factor was found that binds to the more conserved 5' region of the X-box element, although A20 extracts contained a second, distinct binding activity for this region. A common binding factor for the H-box element was detected in extracts from WEHI-3 and L cells. However, this activity was absent in A20 cell extracts. Instead, two different H-box-binding activities were detected, suggesting that different components are involved in class II gene expression in B cells and macrophages. Finally, gamma interferon treatment did not significantly alter the DNA-binding activity in WEHI-3 cells for any of the sequence elements shown to be required for induced chloramphenicol acetyltransferase expression.

Requirement for cotolerogenic gene products in the clonal deletion of I-E reactive T cells

T cells that express the T cell receptor V beta 5.2 domain react with the class II major histocompatibility complex (MHC) molecule I-E, and V beta 5.2+ T cells are deleted in mouse strains that express I-E glycoproteins. By examination of genetically defined recombinant inbred (RI) mouse strains, it was found that the deletion was dependent on the expression of I-E and one of a limited number of non-MHC gene products (cotolerogens). The gene encoding one of these cotolerogens maps to chromosome 12 and is linked to the endogenous provirus Mtv-9. These observations suggest that the I-E-mediated and minor lymphocyte-stimulating antigen (Mls)-mediated deletions of alpha beta T cells from the repertoire are similar; both require the expression of a class II MHC glycoprotein and a second non-MHC gene product.

Protein kinase-C involvement in thymocyte apoptosis induced by hydrocortisone

The involvement of protein kinase-C in thymocytes death induced by hydrocortisone was studied. Thymus cells were incubated 6 hr or in the presence of hydrocortisone, labeled with Acridine orange, and the DNA content of each nuclei was estimated by cytofluorimetry. The results indicate that hydrocortisone-induced DNA fragmentation can be prevented by adding the protein kinase-C inhibitor H-7 to the cell suspension. Incubation of the H-A 1004, an inhibitor of c-AMP-dependent protein kinase, with low effect on on protein kinase-C, did not interfere with the cortisone-mediated DNA fragmentation. Therefore, it can be concluded that protein kinase-C plays an important role in the process of lympholysis mediated by corticoids.

CD8 is required during positive selection of CD4-/CD8+ T cells

Interactions between self-MHC molecules and T cells are necessary for the proper development of mature T cells, in part due to an absolute requirement for self-MHC-TCR interactions. Recently, we showed that CD4-mediated interactions also participate in shaping the T cell repertoire during thymic maturation. We now examine the possible role of the CD8 molecule during in vivo T cell development. Our results demonstrate that perinatal thymi treated with intact anti-CD8 mAb fail to generate CD8 single-positive T cells, while the generation of the other main phenotypes remains unchanged. Most importantly, the use of F(ab')2 anti-CD8 mAb fragments gave identical results, i.e., lack of generation of CD4-/CD8+ cells, with no effect on the generation of CD4+/CD8+. Furthermore, selective blocking of one CD8 allele with F(ab')2 mAbs in F1 mice expressing both CD8 alleles did not interfere with the development of CD4-/CD8+ cells, demonstrating that the absence of CD8+ T cells in homozygous mice is not due to depletion, but rather is caused by a lack of positive selection. This is most likely attributable to a deficient CD8-MHC class I interaction. Our findings strongly advocate that CD8 molecules are vital to the selection process that leads to the development of mature single-positive CD8 T cells.

The toxicity of staphylococcal enterotoxin B in mice is mediated by T cells

Staphylococcal enterotoxin B (SEB) has been shown in the past to be a potent T cell stimulant in mouse or man. The toxin acts as a superantigen that is, it binds to class II MHC proteins and, as such a complex, stimulates T cells bearing particular V beta s as part of their receptors. The toxin also has several pathological effects, causing, in mice, rapid weight loss, thymus atrophy, immunosuppression, and, at high doses, death. The data in this paper show that at least one of these effects, weight loss, is T cell mediated. Staphylococcal enterotoxin-mediated weight loss is MHC dependent, and is almost absent in animals expressing MHC class II molecules, which, complexed with SEB, are poor T cell stimulants. Also, mice that lack T cell function, genetically or because of cyclosporin A treatment, lose no or less weight than controls in response to SEB. Finally, animals bred such that they express few T cells bearing V beta s with which SEB can interact lose much less weight in response to the toxin than littermate controls that have higher numbers of reactive T cells. It is therefore suggested that the pathological effects of the staphylococcal, T cell-stimulating toxins in mouse and man may be partially or wholly the consequence of massive T cell stimulation.

Sequence elements required for activity of a murine major histocompatibility complex class II promoter bind common and cell-type-specific nuclear factors

We have examined the sequence elements and corresponding DNA-binding factors required for transient expression of the A alpha d promoter fused to the bacterial chloramphenicol acetyltransferase reporter gene in a variety of cultured cell lines. Deletion analysis demonstrated that only about 110 nucleotides of sequence 5' of the transcription start site are required for constitutive expression in the murine B-lymphoma cell line A20 or for gamma interferon-induced expression in the murine monocytic cell line WEHI-3. Linker-scanner mutation of this region indicated that at least three sequence elements are required for promoter activity. These elements correspond to the conserved sequence elements found in other human and mouse class II genes, the X box, the Y box, and the H box. Analysis of DNA-binding activity showed that the three most predominant factors present in extracts from WEHI-3, A20, or L cells (which do not express the class II genes) are actually a family of factors that bind to a fourth sequence element, overlapping the 3' end of the X-box sequence, that is homologous to the cyclic AMP-responsive enhancer element. A single common factor that binds to the Y box was detected in extracts from all cells tested, as has been seen with the Y-box elements of other class II genes. Another common factor was found that binds to the more conserved 5' region of the X-box element, although A20 extracts contained a second, distinct binding activity for this region. A common binding factor for the H-box element was detected in extracts from WEHI-3 and L cells. However, this activity was absent in A20 cell extracts. Instead, two different H-box-binding activities were detected, suggesting that different components are involved in class II gene expression in B cells and macrophages. Finally, gamma interferon treatment did not significantly alter the DNA-binding activity in WEHI-3 cells for any of the sequence elements shown to be required for induced chloramphenicol acetyltransferase expression.

Role of CD8+ T cells in mercury-induced autoimmunity or immunosuppression in the rat

In Brown-Norway (BN) rats mercuric chloride induces an autoimmune disease characterized by an increase in serum IgE concentration, and by the production of anti-glomerular basement membrane antibodies responsible for a glomerulonephritis with a heavy proteinuria. (i) This disease results from a B-cell polyclonal activation probably due to frequent anti-class II T cells. (ii) The self limitation observed in this model is associated with both a decrease in the frequency of anti-class II T cells and the emergence of CD8+ T cells able to suppress these autoreactive T cells. (iii) In Lewis (LEW) rats which do not develop autoimmunity, HgCl2 provokes the appearance of non-antigen-specific CD8+ T cells responsible for a depression of T-cell functions. The aim of this work was to test the effect of treatment with an anti-CD8 monoclonal antibody (MoAb) in both BN and LEW rats. Anti-CD8 MoAb-treated rats were effectively depleted in CD8+ T cells. However, neither the induction nor regulation phases of mercury-induced autoimmunity were modified in BN rats. Mercury-induced immunosuppression in LEW rats was abrogated; however, depletion in CD8+ T cells did not allow the disease to occur in that strain. Finally, CD8 depletion induced in normal BN rats the appearance of rare anti-class II T cells showing that these cells are normally present in that strain but negatively controlled by suppressor T cells.

Allelic variations in the human T cell receptor V beta 6.7 gene products

Polymorphisms of human TCR gene products have been suggested by the description of a mAb, OT145, that identifies a subset of TCRs in some individuals but not in others (6). Here we demonstrate that this mAb detects a TCR allotype of the V beta 6.7 gene. Two allelic products of this V gene differ by two nonconservative amino acid substitutions. The mAb OT145 appears to react with V beta 6.7 a gene products ("+" allele), but not with V beta 6.7b gene products ("-" allele). This represents the first direct demonstration that TCR V gene allotypes exist and provides a possible explanation for immune responses under the control of TCR V genes and for disease associations with TCR V genes.

Thymic dendritic cells and B cells: isolation and function

The thymus is the primary organ in which T cells undergo rearrangement of T cell receptor alpha and beta genes, positive selection for affinity to self MHC products, and elimination (negative selection) of reactivity to self antigens. These events require an interaction of the developing T cell with other cell types in the thymus. The latter include epithelial cells, macrophages, dendritic cells, and the recently described thymic B cells the majority of which are CD5+. Here we review the identification and isolation of thymic dendritic cells and CD5+ B cells. We consider phenotype, ontogeny, and function, including possible contributions to the induction of self tolerance. Thymic dendritic cells are similar to spleen dendritic cells, but are larger and exhibit a few differences in phenotype. Dendritic cells from both organs are equally potent accessory cells for the MLR and lectin-induced, T cell proliferation. Thymic dendritic cells have higher levels of Fc receptors and support anti-CD3 dependent mitogenesis. Thymic CD5+ B cells share phenotypic features with peritoneal CD5+ B cells. However thymic B cells neither proliferate nor form antibody producing cells in response to the stimulation with LPS or anti-IgM plus IL-4, but do respond to stimulation with MHC class II-restricted helper T cells. Thymic dendritic cells and CD5+ B cells both appear at a similar time in ontogeny, about 14 d of gestation, which is the time T cell differentiation begins to take place. Dendritic cells from spleen, which are potent activators for peripheral T cells, are also potent inactivators for thymic-derived cytotoxic T cells. A correlation between reactivity to MIs products and the expression of TCR-V beta genes is well documented, and B cells are the primary APC for this antigen. Therefore, thymic CD5+ B cells may be a good tool for the investigation of tolerance to M1s products.

Experimental models of autoimmune diseases

Several experimental models of autoimmune diseases have been studied which often mimic the human situation. Autoreactive T cells that emerge either spontaneously or after immunization have been identified in several situations. The main lesson from these models is that these autoreactive T cells are negatively controlled in the normal situation and that a defect either inherited or acquired in this regulatory circuit is responsible for autoimmunity.