David, Chella S.

Abstract The HLA class II region on chromosome 6p21 accounts for majority of familial clustering in MS and is by far the major susceptibility locus. However due to multiple class II genes, high polymorphism and linkage disequilibrium, it has been difficult to understand the role of individual class II genes and interaction between class II genes in the disease process. Previously, we have shown that PLP91-110 induced classical EAE in DR3.Abo but not in DQ6.Abo transgenic (tg) mice, suggesting that DR3 is a disease susceptible gene in the context of PLP. In this study, we have generated double (tg) mice expressing HLA-DR3/DQ6 to simulate heterozygous haplotype of MS patients. Introduction of DQ6 onto DR3 tg mice led to decrease in disease incidence on immunization with PLP91-110 suggesting a protective role of DQ6. We further characterized that this protective effect was due to expression of high level of IFNγ by DQ6 restricted T cells, which suppressed proliferation of encephalitogenic DR3-restricted T cells by inducing apoptosis in antigen specific T cells. Our IFNγ blocking experiments confirmed an anti-inflammatory role of IFNγ. Our study suggests that DQ6 modifies the PLP91-110 specific T cell response in DR3 through IFNγ mediated activation induced cell death. Thus our double tg mouse model provides a novel tool to study linkage disequilibrium and explain how HLA-DQ molecule might modulate disease in MS-susceptible individual.

Abstract DRB1*0401 has been associated with predisposition to rheumatoid arthritis while DRB1*0402 is associated with resistance. We have generated transgenic mice lacking endogenous H2-A and E molecules (AEo) but expressing DRB1*0401 or DRB1*0402 molecules. All transgenic mice expressed functional DR molecules which could positively and negatively select various Vβ T cell repertoire profile. Transgenic mice were studied in vivo for induction of arthritis following immunization with type II collagen. DRB1*0401 mice are susceptible to collagen-induced arthritis while DRB1*0402 mice are resistant. Presence of DR4 leads to arthritis predominantly in female mice. CIA susceptible AEo.DQ8 mice have been shown to develop arthritis without a gender bias. However, AEo mice expressing DRB1*0401.DQ8 develop arthritis in female mice with a ratio of 3F:1M, similar to that in humans. DRB1*0402 mice did not show any defect in humoral response though they demonstrated a lower cellular response to CII and an increased activation-induced cell death compared to susceptible mice. In vitro T cell response with superantigen staphylococcal enterotoxin B and Concovalin A showed that DRB1*0402 mice are not defective in generating T cell response. The resistant mice showed an increased cell death in thymus suggesting deletion of autoreactive T cells at thymic level might be one reason for protection. The data suggest that DRB1 polymorphism lead to selection of TCR in thymus thus controlling the selection of autoreactive T cells available in periphery. Thus these transgenic mice may be useful to study the reasons for gender-bias observed in human patients develop therapeutic strategies.

Abstract Background/Aim: Dermatitis Herpetiformis is the autoimmune skin manifestation of celiac disease, and similar to celiac disease, is tightly associated with HLA-DQ2 and HLA-DQ8. Previously we have generated a mouse model of DH using DQ8 transgenic mice on an NOD background. Herein, we describe a similar DH-like blistering disease that spontaneously develops in DQ2 transgenic mice. Methods: Transgenic mice that express DQ2 on a background deficient in endogenous MHC II (Aeo) were raised on a gluten containing (GC) chow or a gluten free chow . Results: When raised and maintained on a gluten containing chow, approximately 15% of these Aeo DQ2 mice spontaneously developed blisters on the ears and tails at about six months of age. Control Aeo mice that were also raised on a gluten containing chow did not develop this blistering disease during their lifetime. Hematoxilyn and eosin staining revealed subepidermal blisters. Immunohistochemistry revealed IgA deposits at the dermal/epidermal junction in perilesional ear tissue. However, when these mice were raised and maintained on a gluten free diet, these mice did not develop blisters. Conclusions: This blistering disease is a gluten dependent disease that has many of the elements typically associated with Dermatitis Herpetiformis, most especially the association with the DQ2 gene.

Abstract Bacterial superantigens (BSAg) are produced by Staphylococcus aureus and Streptococcus pyogenes. BSAg bind directly to the MHC class II molecules without undergoing any processing and subsequently activate T cells in an antigen non-specific manner by interacting with the TCR Vβ region. BSAg have been implicated primarily in acute diseases such as food poisoning, toxic shock syndrome, etc. Since carrier strains of S. aureus and S. pyogenes can elaborate superantigens, there is a distinct possibility of chronic exposure to low doses of BSAg and it is imperative to study its outcome. As BSAg bind more avidly to human (HLA) than to murine (H-2) MHC class II, we delivered low concentrations (total of 10 μg/mouse) of BSAg to HLA-DR3 [staphylococcal enterotoxin (SE) B] or HLA-DQ8 [streptococcal pyrogenic exotoxin (SPE) A] transgenic mice through subcutaneously implanted miniosmotic pumps for 7 days. Continuous exposure to low concentrations of BSAg resulted in a significant protracted expansion of CD4+ T cells expressing TCR Vβ8 (for SEB); Vβ8 and Vβ14 (for SPEA) with a concomitant decrease in percentage of CD8+ T cells expressing these TCR. T cells bearing TCR Vβ6 showed an inverse correlation (slight decrease in CD4+ and increase in CD8+ compartments with SEB and SPEA). Histological examinations of major organs showed intense perivascular infiltration in lungs, kidneys and liver along with inflammatory infiltrates in the cardiac valvular endothelium in mice bearing pumps delivering BSAg but not PBS. In conclusion, we have shown the immunological outcome of chronic exposure to low levels of BSAg using the HLA class II transgenic mouse model.

Abstract Experimental autoimmune thyroiditis (EAT) is a model for Hashimoto’s thyroiditis which is characterized by mononuclear cell infiltration and follicular destruction. We previously reported on an Eα-transgenic strain (H2A−E+) that is highly susceptible to human (h) thyroglobulin (Tg)-induced EAT, but strongly tolerant to mouse (m)Tg. After mTg and adjuvant, traditional, resistant strains do not develop thyroiditis, nor detectable primed T cell proliferation, but do make anti-mTg Abs. By contrast, this H2Eb+ strain does not even produce mTg Abs. To identify what epitopes in the mouse thyroid hTg-primed, thyroiditogenic T cells recognize to evoke EAT, we first confirmed that hTg-primed cells do not respond to mTg, nor to other thyroid Ags. We next selected the three immunogenic epitopes on hTg known to be processed and presented by H2Eb, as the basis for synthesizing potential mTg epitopes. One 15-mer peptide, mTg409, did prime T cells, elicit Ab, and generate moderate thyroiditis. Thyroiditis incidence was increased after adoptive transfer of in vitro activated, mTg409-primed cells. Moreover, cells primed to hTg410, the corresponding, pathogenic hTg peptide, cross-reacted with mTg409, and vice versa. This mTg autoepitope, identified by using thyroiditogenic hTg epitopes, helps to explain the severe thyroiditis seen in this novel H2E+ transgenic model. (Supported by NIH grant DK45960)

Abstract Although association of HLA-B27 with human spondyloarthropathies has been known for several years, its role in disease pathogenesis is not understood. Recently, a few investigators have proposed that presentation of B27-derived peptides by MHC class II molecules may be the underlying mechanism. HLA-B27 transgenic rat and mouse models have provided a new tool for understanding the exact role of B27 in disease pathogenesis. HLA-B27 mice lacking endogenous β2-microglobulin (B27+β2mo) develop disease after they are transferred from the barrier facility to the conventional colony. This model was utilized to test the hypothesis that B27-derived peptide presented by MHC class II molecules is the cause of the disease. The MHC class II knockout gene, Aβo, was bred into our B27+β2mo mice, and disease manifestation was monitored. These mice develop spontaneous disease, demonstrating that MHC class II molecules do not play a major role in B27-related disease. Thus, the disease is not manifested by presentation of B27-derived peptides by class II molecules, since these mice are devoid of H2-A and H2-E molecules. Furthermore, in vivo treatment with mAb against the heavy chain of B27 reduced the incidence of disease in B27+β2mo mice. Our results clearly demonstrate that B27 heavy chains are directly involved in the disease process.

Abstract Among all the genetic factors linked with multiple sclerosis (MS), MHC class-II molecules show the strongest association. Generation of HLA class-II transgenic mice has helped to elucidate the role of HLA class-II gene in MS. We have shown that the human HLA-DR3 (DRβ1*0301) gene predisposes to proteolipid protein (PLP)-induced EAE, whereas HLA-DQβ1*0601 (DQ6) was resistant. We also showed that the DQ6 molecule protects from EAE in DR3DQ6 double transgenic mice by producing anti-inflammatory interferon gamma (IFNγ). To further confirm the role of IFNγ in protection, we generated DR3DQ8 mice lacking IFNγ (DR3DQ8.IFNγ-/-). Immunization with PLP91-110 peptide caused atypical EAE in DR3DQ8.IFNγ-/- mice characterized by ataxia and spasticity, hallmarks of brain-specific disease. Severe brain specific inflammation and demyelination in DR3DQ8.IFNγ-/- mice with minimal spinal cord pathology further confirmed brain-specific pathology. Severe disease in DR3DQ8.IFNγ-/- mice was due to increased encephalitogenicity of CD4 T cells and its ability to produce higher levels of IL-17 and GM-CSF compared to DR3DQ8 mice. Further, areas with demyelination showed increased presence of CD68+ inflammatory cells suggesting an important role for monocytes/microglia in causing brain pathology. Thus, our study supports a protective role for IFNγ in inflammatory and demyelinating disease through down regulation of IL-17 and induction of neuro-protective factors in the brain by monocytes/microglial cells.

Abstract HLA-DRB1*0401 has been associated with predisposition to develop rheumatoid arthritis (RA) while *0402 has been linked with resistance to develop RA. Using transgenic mice, we showed that *0401 mice are susceptible to develop collagen-induced arthritis (CIA) that mimics human disease in histopathology, production of autoantibodies and sex-bias while *0402 mice are not susceptible. In this study, we determined if *0401 mice have CD4 T cell repertoire that is predetermined to produce proinflammatory cytokines. The data shown here supports that both *0401 and *0402 mice can produce Th1/TH17 cytokines although conditions required for production of Th17 cytokines differ between the 2 strains. In context of CIA, *0402 mice generate a Th2 response that may explain its resistance to develop CIA. Further, our data supports that a significant subset of naïve CD4 T cells when activated in polarizing conditions can generate T regulatory cells in *0402 but not *0401 mice. We hypothesize that *0401 has been evolutionarily selected due to its ability to clear infection. Our data suggests that *0401 generates a storm of cytokines in response to bacterial products and may be more efficient in clearing infection than *0402. Autoimmunity is a bystander effect of the cytokine storm ensuing immune dysregulation along with the presence of lower number of T regulatory cells in *0401 mice.

Abstract The Ia.23 specificity is a private specificity of the H-2d haplotype which we have assigned to the I-E subregion. It is found on the same Ia molecule that bears Ia.7, a public specificity previously assigned to the I-C subregion of haplotypes H-2d, H-2k, H-2p and H-2r. We suggest a reassignment of specificity Ia.7 to the I-E subregion to simplify the interpretation of the data presented in this paper and elsewhere, and to permit an orderly description of that part of the I region to the right of I-J.

Abstract Bacterial superantigen-induced toxic shock syndrome (TSS) and sepsis are characterized by a profound systemic inflammatory response. Given the important role of NF¿B pathway in inflammation, we investigated whether a clinically approved proteasome inhibitor, Bortezomib, capable of inhibiting NF¿B pathway, be of therapeutic use in such conditions using HLA class II transgenic mice. Bortezomib significantly blocked staphylococcal enterotoxin B (SEB)-induced T cell activation and cytokine production in vitro. In SEB-induced TSS, in vivo administration of Bortezomib significantly reduced serum levels of several pro-inflammatory cytokines and chemokines. Nevertheless, 100% of SEB challenged mice treated with Bortezomib succumbed to TSS, and Bortezomib sensitized mice to TSS at concentrations of SEB which are otherwise non-lethal (i.e., as low as 2 μg/mouse), even when administered 24 hours after exposure to SEB. Sensitization to TSS by Bortezomib was not due to the pro-apoptotic effects of Bortezomib. Serum biochemical parameters and histopathological findings suggested acute liver failure as the possible cause of mortality. In sepsis model, whereas LPS alone (10 μg/mouse) was seldom lethal, it was uniformly lethal when administered along with Bortezomib. In conclusion, Bortezomib sensitizes mice to certain acute systemic inflammatory diseases. Supported by R01AI068741 and AI 72211