Essential role of Stat5 for IL-5-dependent IgH switch recombination in mouse B cells.

IL-5 stimulation of CD38-activated murine splenic B cells induces mu-gamma1 CSR at the DNA level leading to a high level of IgG1 production. Further addition of IL-4 in the system enhances IL-5-dependent mu-gamma1 CSR. Although some of the postreceptor signaling events initiated by IL-5 in activated B cells have been characterized, the involvement of Stat in IL-5 signaling has not been thoroughly evaluated. In this study, we examined the activation of Stat5 and activation-induced cytidine deaminase (AID) in CD38-activated murine splenic B cells by IL-5. The role of Stat5a and Stat5b in IL-5-induced mu-gamma1 CSR and also IgG1 and IgM production was documented, as IL-5 does not act on CD38-stimulated splenic B cells from Stat5a(-/-) and Stat5b(-/-) mice. Expression levels of CD38-induced germline gamma1 transcripts and AID in Stat5a(-/-) and Stat5b(-/-) B cells upon IL-5 stimulation were comparable to those of wild-type B cells. The impaired mu-gamma1 CSR by Stat5b(-/-) B cells, but not by Stat5a(-/-) B cells, was rescued in part by IL-4, as the addition of IL-4 to the culture of CD38- and IL-5-stimulated B cells induced mu-gamma1 CSR leading to IgG1 production. Analysis of cell division cycle number of wild-type B cells revealed that mu-gamma1 CSR was observed after five or six cell divisions. Stat5a(-/-) and Stat5b(-/-) B cells showed similar cell division cycles, but they did not undergo mu-gamma1 CSR. Our data support the notion that both Stat5a and Stat5b are essential for IL-5-dependent mu;-gamma1 CSR and Ig secretion; however, their major target may not be AID. Stat5a and Stat5b are not redundant, but rather are at least partially distinctive in their function.

Identification of amino acid residues of the T-cell epitope of Mycobacterium tuberculosis alpha antigen critical for Vbeta11(+) Th1 cells.

Stimulation of Mycobacterium tuberculosis-primed lymph node cells from C57BL/6 mice with alpha antigen (also known as antigen 85B and MPT59) induced cell proliferation, production of interleukin 2 and gamma interferon, and expansion of Vbeta11(+) CD4(+) T cells in conjunction with antigen-presenting cells in an I-A(b)-restricted manner. Using a series of 15-amino-acid peptides that overlapped each other by 5 amino acids and spanned the mature alpha antigen, we identified the antigenic epitope for alpha antigen-specific Vbeta11(+) Th1 cells. That peptide (peptide-25), which corresponds to amino acid residues 240 to 254 of alpha antigen, contains a motif that is conserved in I-A(b) and requires processing by antigen-presenting cells. Using peptide-25-reactive Vbeta11(+) T-cell clones and substituted peptide-25 mutants, we determined which amino acid residues within peptide-25 were critical for T-cell receptor (TCR) recognition. Our results showed that the amino acid residues at positions 245, 246, 248, 250, and 251 are important for recognition of TCRVbeta11 and that residues at positions 244, 247, 249, and 252 are I-A(b) contact residues. We also observed that active immunization of C57BL/6 mice with peptide-25 can lead to decreased bacterial load in the lungs of M. tuberculosis H37Rv-infected mice. These results should provide us with a useful tool for delineating the regulation of Vbeta11(+) Th1-cell development during M. tuberculosis infection and for developing a vaccine inducing a Th1-dominant immune response.

Mapping of V beta 11+ helper T cell epitopes on mycobacterial antigen in mouse primed with Mycobacterium tuberculosis.

Antigenic epitopes for Mycobacterium tuberculosis-reactive T cell immune responses have been mapped using the purified Mycobacterium protein antigen. Lymph node cells from C57BL/6 mice that had been immunized with heat-killed M. tuberculosis were cultured with various Mycobacterium protein antigens and their reactivity was monitored by proliferative response. Usage of the TCR beta chain repertoire was analyzed by flow cytometry. Stimulation of M. tuberculosis-primed lymph node cells with MPT59 (antigen 85B, alpha antigen) induced proliferative response, production of IL-2 and IFN-gamma, and the expansion of V beta 11+ CD4+ T cells in conjunction with antigen-presenting cells in an I-Ab-restricted manner. Lymph node cells from non-primed mice failed to proliferate in response to MPT59. Using peptides covering the complete mature 285 amino acids long MPT59 protein as 15-mer molecules overlapping by five amino acids, we identified the antigenic epitope for MPT59-specific V beta 11+ T cells. The 15-mer peptide, covering amino acid residues 240-254 of MPT59 [peptide-25 (amino acids 240-254)], contains the motif that is conserved for I-Ab and requires processing by antigen-presenting cells to trigger peptide-25-specific V beta 11+ CD4+ T cells. We conclude from these results that MPT59 and peptide-25 (amino acids 240-254) are not superantigens and require antigen processing in order to stimulate V beta 11+ Th1 cells. This experimental system will provide us with a useful tool for delineating the regulation of T cell development in a particular subset of M. tuberculosis infection and for developing antigenic peptides for Th1-dominant immune responses.

Identification of HTLV-1-specific CTL directed against synthetic and naturally processed peptides in HLA-B*3501 transgenic mice.

Previous studies of CTL responses to influenza peptides in HLA single transgenic mice resulted in the identification of at most one immunodominant epitope. Since HLA-B*3501 is known to present multiple HIV-1-specific T cell epitopes we tested the cellular immune response of HLA-B*3501 transgenic mice to synthetic HTLV-1 peptides mixed with the lipohexapeptide N-palmitoyl-S-[2,3-bis(palmitoyloxy)propyl]cysteinyl-seryl-lysyl-l ysyl- lysyl-lysine, which is a biocompatible, Th-epitopeindependent adjuvant. Eleven of 37 tested HLA-B*3501 binding peptides mounted a CTL response after three in vitro stimulations. The HLA-B*3501 affinity of peptides correlated with their ability to induce CTL in HLA-B*3501 transgenic mice. Seven peptides derived from env-gp46 (VPSPSSTPLL, VPSSSSTPL, YPSLALAPH, and YPSLALAPA), pol (QAFPQCTIL), gagp19 (YPGRVNEIL), and tax (GAFLTNVPY) proteins induced peptide-specific CTL Bulk CTL generated by four peptides derived from env-gp46 (SPPSTPLLY, VPSPSSTPLLY, and VPSPSSTPLL) and pol (QAFPQCTILQY) killed peptide-pulsed and recombinant vaccinia-infected target cells. The latter peptides therefore present T-cell epitopes and are vaccine candidates for our transgenic mouse model.

Excessive function of peripheral blood neutrophils from patients with Behçet's disease and from HLA-B51 transgenic mice.

OBJECTIVE: To elucidate the role played by HLA-B51 in the neutrophil hyperfunction of Behçet's disease, we determined the superoxide production by purified peripheral blood neutrophils from Behçet's disease patients, from HLA-B51 positive healthy individuals, and from HLA-B51 transgenic mice. METHODS: Neutrophil function was evaluated by flow cytometric analysis, detecting the conversion of 2',7'-dichlorofluorescin diacetate into dichloroflurescein, induced by superoxide in the neutrophils. RESULTS: A significant correlation between the neutrophil hyperfunction and the possession of HLA-B51 phenotype, regardless of the presence of the disease, was observed in humans. FMLP-stimulated neutrophils (without in vitro priming) from HLA-B51 transgenic mice, but not those from HLA-B35 transgenic mice or from nontransgenic mice, produced substantial amounts of superoxide. CONCLUSION: The HLA-B51 molecule itself may be responsible, at least in part, for neutrophil hyperfunction in Behçet's disease.

Antibody against interleukin-5 prevents antigen-induced eosinophil infiltration and bronchial hyperreactivity in the guinea pig airways.

Interleukin-5 (IL-5) induces proliferation, differentiation and activation of eosinophils. An animal model of local allergen (airways) sensitization was employed to study the effects of anti-IL-5 monoclonal antibody (mAb) on infiltration of eosinophils into inflammatory region, the development of antigen-induced late asthmatic response (LAR) and the increased bronchial responsiveness following LAR. Guinea pigs exposed to aerosolized ovalbumin (OVA) daily for 10 days developed an increase in the number of eosinophils in the tracheal wall 24 h after aerosolized OVA challenge. Furthermore, all animals developed an apparent LAR determined by the response with a 2-fold increase in respiratory resistance and showed an increase in bronchial responsiveness to acetylcholine 24 h after OVA challenge. In animals treated with anti-IL-5 mAb, however, eosinophil number in the tracheal wall dramatically decreased compared with animals treated with control antibody. The development of LAR was also remarkably suppressed by anti-IL-5 mAb treatment, although a similar magnitude of immediate bronchoconstriction was observed. Moreover, in anti-IL-5 antibody-treated guinea pigs, an increase in bronchial responsiveness to acetylcholine significantly decreased. Data demonstrate that IL-5 is involved in airway eosinophilia, development of LAR and an increase in bronchial responsiveness induced by allergen sensitization via the airways. Development of IL-5 synthesis inhibitors and/or receptor antagonists could provide another therapeutic class of anti-asthmatic drugs.

The role of the conserved residue in pocket A and the polymorphic residue in pocket E of HLA-B*3501 in presentation of human minor histocompatibility peptides to T cells.

We investigated T cell recognition for human minor histocompatibility (hmH) peptides using HLA-B*3501 restricted, hmH specific cytotoxic T lymphocytes (CTL) clones. These CTL clones killed C1R cells expressing HLA-B*3501 but not C1R cells expressing chimeric antigens between HLA-B*3501 and HLA-B*5101. They also failed to kill C1R cells expressing HLA-B*3501 mutants at residue 152 (B*3501-V152E) or at residue 171 (B*3501-Y171H). The CTL clone failed to kill C1R cells expressing these mutant molecules loaded with the hmH peptides isolated from C1R-B*3501 cells although it killed a self-B cell line expressing HLA-B3501 loaded with the specific hmH peptides. The CTL clone also failed to kill T2 cells expressing the mutant molecules loaded with the specific peptides whereas it killed T2 cells expressing HLA-B*3501 loaded with the specific peptide. On the other hand, naturally occurring specific hmH peptides were isolated from purified B*3501-V152E and B*3501-Y171H molecules, indicating that both HLA-B*3501-V152E and HLA-B*3501-Y171H molecules can bind the hmH peptides. These findings indicate that both the conserved residue 171 in pocket A and the polymorphic residue 152 in pocket E are critical in recognition of the T cells but not binding of the hmH peptides. Furthermore, these results provide the possibility that the TCR recognizes a conformational structure of hmH peptides bound to HLA-B*3501 molecules.

Different effects of substitutions at residues 224 and 228 of MHC class I on the recognition of CD8.

Previous studies indicated that weak xenoresponse to HLA class I by mouse T cells is due to the inefficient interaction of mouse CD8 with the alpha 3 domain of HLA class I. The present study using chimeric H-2Kb molecules with recombinant alpha 3 domain between H-2Kb and HLA-B7 as well as single amino acid mutants of H-2Kb demonstrated that each substitution at residues 224 and 228 affects recognition of CD8-dependent mouse CTL clones. On the other hand, reactivity of IL-2-producing H-2Kb-specific T cell hybridoma transfected with mouse CD8 alpha was abrogated by substitution at residue 224 but not by that at residue 228. This indicates that the substitution at residue 228 affects recognition of CD8-dependent CTL but does not critically affect binding of CD8 to MHC class I molecules, although residue 224 abrogates binding of CD8. The model structure of the alpha 3 domain of H-2Kb suggests that the substitution at residue 224 induces conformational change of CD8 binding loop, whereas minimum structure change by the substitution at residue 228 is expected. It is therefore speculated that minimum structure change of CD8 binding loop by substitution at residue 228 may influence binding affinity of CD8, which abrogates recognition of CD8-dependent CTL but not IL-2 production of the CD8-dependent T cell hybridoma.

Expression of human minor histocompatibility antigen on cultured kidney cells.

Incompatibility of human minor histocompatibility (hmH) antigens can induce rejection of grafts in organ transplantation and graft-versus-host reactions in bone marrow transplantation. In spite of their importance in clinical transplantation, hmH antigens are not well studied. Previous studies have demonstrated the expression of hmH antigens on T and B cells, hematopoietic progenitor cells and keratinocytes. We have for the first time demonstrated the expression of hmH antigens on cultured kidney cells using HLA-B35-restricted, hmH antigen-specific cytotoxic T lymphocyte (CTL) clones, which were previously established from a patient who rejected two kidneys from HLA-identical sisters. The CTL clones could not kill cultured kidney cells. Since cultured kidney cells expressed very low levels of HLA class I antigens it was thought that their failure to be killed by the CTL clones was due to lack of expression of HLA-B35 antigens. After induction of class I antigens on cultured kidney cells by interferon-gamma (IFN-gamma), the IFN-gamma-treated cultured kidney cells were killed by the CTL clones. Furthermore, we isolated hmH antigens as peptides from cultured kidney cells after treatment with IFN-gamma. These results indicate that cultured kidney cells express hmH antigens when HLA class I antigen is induced by IFN-gamma and hmH antigens on cultured kidney cells are recognized by T cells as peptides presented by HLA-B35 molecules.

Functional expression of HLA-C blank antigens on human blood lymphocytes.

The surface expression of two HLA-C blank Ag (Cb-1 and Cb-2) on PBL was investigated with Cb-1- and Cb-2-specific CTL clones generated by the stimulation of the HLA-C blank Ag on transfected Hmy2CIR cells. The Cb-1- and Cb-2-specific CTL clones could lyse EBV-transformed B cells and PHA-induced T cells from which the HLA-C blank genes were derived. Furthermore, the reactivity of these CTL clones with PHA-induced T cells was blocked by HLA class I monomorphic mAb. These results demonstrated that the HLA-C blank Ag are expressed on the surfaces of PBL. Thus, despite the fact that the HLA-C blank Ag are expressed on normal PBL, they are incapable of generating corresponding alloantibodies. On the other hand, the present study demonstrated that these Ag on normal PBL are able to induce specific CTL and that the capacity of these Ag to induce allogeneic CTL is almost identical to that of HLA-B Ag, indicating that they may function as alloantigens in vivo and play a significant role in the rejection of organ grafts and in the graft-versus-host reaction in bone marrow transplantation.

Presentation of human minor histocompatibility antigens by HLA-B35 and HLA-B38 molecules.

Cytotoxic T lymphocyte (CTL) clones specific for human minor histocompatibility antigens (hmHAs) were produced from a patient who had been grafted with the kidneys from his mother and two HLA-identical sisters. Of eight CTL clones generated, four recognized an hmHA (hmHA-1) expressed on cells from the mother and sister 3 (second donor); two recognized another antigen (hmHA-2) on cells from the father, sister 2 (third donor), and sister 3; and the remaining two clones recognized still another antigen (hmHA-3) on cells from the father and sister 3. Panel studies revealed that CTL recognition of hmHA-1 was restricted by HLA-B35 and that of hmHA-2 and hmHA-3 was restricted by HLA-B38. The HLA-B35 restriction of the hmHA-1-specific CTL clones was substantiated by the fact that they killed HLA-A null/HLA-B null Hmy2CIR targets transfected with HLA-B35 but not HLA-B51, -Bw52, or -Bw53 transfected Hmy2CIR targets. These data demonstrated that the five amino acids substitutions on the alpha 1 domain between HLA-B35 and -Bw53, which are associated with Bw4/Bw6 epitopes, play a critical role in the relationship of hmHA-1 to HLA-B35 molecules. The fact that the hmHA-1-specific CTLs failed to kill Hmy2CIR cells expressing HLA-B35/51 chimeric molecules composed of the alpha 1 domain of HLA-B35 and other domains of HLA-B51 indicated that eight residues on the alpha 2 domain also affect the interaction of hmHA-1 and the HLA-B35 molecules.

The structure and expression of genes encoding serologically undetected HLA-C locus antigens.

Approximately 20 to 50% individuals in every race are untypable by human alloantisera for at least one allele of HLA-C locus and the surface expression of HLA-C locus Ag in such an individual (HLA-C blank Ag) remains unknown. To investigate the structure and the surface expression of HLA-C blank Ag, two genes (Cb-1 and Cb-2) encoding HLA-C blank Ag were cloned and their primary structures were determined and compared with other HLA-C locus genes. The similarity of amino acids between Cb-1 and Cw1 was the highest among HLA-C locus genes previously published. Five amino acid substitutions between these molecules were shown to be located on the beta-strand of alpha 1 and alpha 2 domains, suggesting that they might change the conformational allodeterminants on the alpha-helical region of Cw1 which were recognized by antibodies. On the other hand, Cb-2 was the closest to Cw2.2. Six of nine amino acid substitutions between these molecules were observed on alpha 1 and alpha 2 domains, whereas three other substitutions were located on the leader peptide, the alpha 3 domain and the transmembrane. Two substitutions (residues 73 and 163) of the alpha-helical region of the alpha 1 and alpha 2 domains and one (residue 16) of exposed loop may make new allodeterminants which are not recognized by anti-Cw2 sera as well as other alloantisera. The surface expression of these genes was examined on transfected mouse L cells and human B cell line. Both gene products were expressed stably on the surface of these cells. These results suggest that HLA-C blank Ag are most probably expressed on cells in HLA-C blank individuals and that the primary structures of these Ag, which were not detectable by the available alloantisera, may be incapable of generating corresponding alloantibodies.

Ontogeny and function of B220+ L3T4+ T-cell subset of MRL/Mp-lpr/lpr mice.

T-cell-enriched populations obtained from lymph nodes (LNs) of 4-month-old MRL/Mp-lpr/lpr (MRL-lpr), C3H/HeJ-lpr/lpr (C3H-lpr), and C3H/HeJ-gld/gld (C3H-gld) mice were studied for the expression of B220, L3T4, and Lyt 2 antigens. A new B220+ L3T4+ phenotype was demonstrated in T-cell populations of these mice by two-color flow cytometry with phycoerythrin-conjugated monoclonal antibodies (MoAb) to L3T4 and FITC-anti-B220 MoAb. The generation of the T subset was apparently under the influence of the lpr or gld gene, since it could not be demonstrated in T-cell-enriched populations of MRL/Mp- +/+ and normal C3H mice. The expression level of L3T4 antigen on the T subset was lower than that on B220- L3T4+ cells, while the level of B220 antigen was similar to that of B220+ L3T4- or B220+ Lyt 2- cells. The B220+ L3T4+ phenotype appeared in LNs and spleens, but not in thymuses, of MRL-lpr mice as early as 2 months of age. Its proportion to whole LN T cells at this age was equivalent to that observed in 4-month-old mice. Functional studies on FACS-sorted cell populations revealed that the T subset similar to B220+ L3T4- cells possessed deficiencies in the IL-2-IL-2 receptor system. The appearance of the T subset with an intermediate phenotype and its functional defects suggests that lpr and gld genes in these autoimmune mice exert their influences on the ontogeny and function of L3T4+ T cells and contribute to the induction of early lupus.

Phenotypic and functional analyses on T-cell subsets in lymph nodes of MRL/Mp-lpr/lpr mice.

By means of killing and/or FACS sorting the double-negative (DN) Lyt2-, L3T4- cells, Lyt2+ or L3T4+ cells and B220- cell populations were separated from T-cell-enriched lymph node (LN) cells of 4- to 5-month-old MRL/Mp-lpr/lpr mice. These highly purified cell populations were examined for their proliferative responses, interleukin 2 (IL2) production and expression of IL2 receptor (IL2R) in response to phorbol myristate acetate (PMA) and the calcium ionophore A23187 (A2) or PMA plus concanavalin A. The DNT-cell population was unable to respond to the stimuli and did not express IL2R. Thus the DN T cells, the major population responsible for the lymphadenopathy, possess fundamental defects in signal transduction as well as in the IL2-IL2R-mediated function. On the other hand, Lyt2+ or L3T4+ T cells obtained by sorting or B220- cells purified by the sorting after killing B220+ cells, exhibited proliferative responses indistinguishable from that of LN cells of the congenic MRL/Mp-+/+(+/+) mice. These cells also expressed IL2R after stimulation, however, the amount of IL2 produced was significantly lower than that produced by congenic +/+ cells. This suggested that phenotypically normal Lyt2+ or L3T4+ T cells of lpr LNs also possess a partial defect in the signal transduction system for IL2 production under the influence of the lpr gene.

Studies on the role of the lpr gene in the development of immunological abnormalities and lupus nephritis. Analyses in F2 mice.

In order to clarify role of lpr gene of MRL/Mp-lpr/lpr (MRL/l) mice in the development of immunological abnormalities and lupus nephritis, F2 mice were obtained from crosses of (MRL/1 X AKR) F1 mice and examined with regard to lymph node enlargement, numbers of immunoglobulin (IgM and IgG) producing cells (PC) in the spleen, serum levels of anti-trinitrophenyl (TNP) and anti-DNA antibodies, amounts of circulating immune complexes (IC) and histopathological findings of the kidney. Out of total numbers of 234 F2 mice, 54 mice (F2[+]mice) (23.1%) were with lymphoproliferation and 180 mice (F2[-]mice) (76.9%) without it. As parental controls, 39 MRL/1, 27 AKR and 21 F1 mice were also checked. There were significant differences in the all of the parameters tested between F2 (+) and F2 (-) mice. Significant differences were also observed between F2 (+) and MRL/l mice in the weight of mesenteric lymph nodes, numbers of IgGPC, levels of IgG antibodies and amounts of IC, but not in the numbers of IgMPC and levels of IgM antibodies. All of F1 mice tested mimicked AKR mice in the values of various parameters described above. In MRL/l and F2 (+) mice, there were no significant correlations between the weight of mesenteric lymph nodes and other parameters. These data seem to show that the lpr gene behaves as a single recessive trait and there exists gene(s) which suppresses expression of the lpr gene.

One-way occurrence of graft-versus-host disease in bone marrow chimaeras between congenic MRL mice.

Bone marrow cells (BMCs) of MRL/Mp-lpr/lpr (MRL/l) mice, when infused into irradiated MRL/Mp-+/+ (MRL/n) mice, induced graft-versus-host disease (GVHD) and recipients died of wasting syndromes beginning around a few weeks later. Protracted appearance of GVHD was observed when (MRL/n X MRL/l) F1 mice were used as recipients of MRL/l BMSs. On the other hand, MRL/n BMCs did not elicit GVHD in irradiated MRL/l mice. Treatment of MRL/l BMCs with anti-Thy-l antiserum plus complement did not eliminate GVHD-provoking capacity. Thymectomy of the recipents reduced the incidence of GVHD. No apparent reaction was observed in mutual mixed lymphocyte culture of spleen cells from MRL/l and MRL/n mice. These data suggest that injected T cell precursors of MRL/l mice become mature through the host (MRL/n mice) thymus and appear in the periphery as functioning T cells, resulting in the appearance of GVHD.

Induction of immunoglobulin-secreting cells by 2-mercaptoethanol in in vitro culture of B cells from autoimmune mice.

Splenic B cells from older MRL/Mp-lpr/lpr (MRL/1) and male BXSB mice responded to 2-mercaptoethanol (2-ME) in in vitro culture and generated immunoglobulin-secreting cells (IgSC). Optimal concentration of 2-ME to induce IgSC was 5 X 10(-5) M. Kinetic studies revealed that the generation of IgSC was already apparent after 24 hr of culture and peak response was attained on the 2nd day. The response of B cells to 2 ME was enhanced in the presence of splenic T cells. Irradiation of B cells reduced the generation of IgSC. The B cell population of autoimmune mice responding to 2-ME to generate IgSC seems to be in a terminal stage of differentiation. This increased B cell differentiation was characteristic of autoimmune mice and assumed to have some significance in the development of autoimmune diseases.

Resistant nature of T cells of autoimmune mice to tolerance induction with human serum albumin.

Inducibility of immunological tolerance to a T-dependent antigen, human serum albumin, was studied in autoimmune mice (NZB, MRL/Mp-lpr/lpr and male BXSB mice), comparing with non-autoimmune mice (BALB/c, DBA/2, MRL/Mp-+/+ and female BXSB mice). Weekly injections of increasing doses of the tolerogen into DBA/2 mice induced T-cell tolerance without the participation of suppressor cells and without affecting B cells. All of the autoimmune mice tested were refractory to the tolerogen; partial tolerance (40%-60% responsiveness) was attained in mice aged 1 month, but not in older mice. By comparison, tolerance was induced in non-autoimmune mice to a similar degree irrespective of the age of the mice. The tolerance inducibility was also examined in reciprocal bone marrow cell (BMC) transfer experiments between NZB and DBA/2 mice and also BMC transfer into B10-D2 mice. It was shown that recipients of NZB BMC were more resistant to tolerance induction than those of DBA/2 BMC, while irradiated NZB mice receiving DBA/2 BMC were susceptible. Hence, it is suggested that the tolerance-resistant nature of the NZB mice was inherent in T-cell precursors in the bone marrow and not in the host environment. Similar findings in support of this notion were also observed in BSXB mice. The significance of tolerance resistance in the development of autoimmunity is discussed.