Kimitaka Sagawa

Immunopathological mechanisms of human T cell lymphotropic virus type 1 (HTLV-I) uveitis. Detection of HTLV-I-infected T cells in the eye and their constitutive cytokine production.

The immunopathology of human T cell lymphotropic virus type 1 (HTLV-I) uveitis was addressed by using T cell clones (TCC) established from the intraocular fluid of patients with HTLV-I uveitis. Proviral DNA of HTLV-I was identified in 55 out of 94 (59%) or 13 out of 36 (36%) TCC from the ocular fluid or the peripheral blood of these patients, respectively. Most of HTLV-I-infected TCC had a CD3+ CD4+ CD8- phenotype. HTLV-I infection on TCC was confirmed by analysis of the viral mRNA, nucleotide sequence, virus-associated proteins, and virus particles. HTLV-I-infected TCC, but not HTLV-I negative TCC, constitutively produced high amounts of IL-6 (1,336 +/- 1,050 pg/ml) and TNF-alpha (289 +/- 237 pg/ml) in the absence of any stimuli. HTLV-I-infected TCC from the ocular lesion also constitutively produced high amounts of IL-1 alpha (12,699 pg/ml), IL-2 (61 pg/ml), IL-3 (428 pg/ml), IL-8 (1,268 pg/ml), IL-10 (28 pg/ml), IFN-gamma (5,095 pg/ml), and GM-CSF (2,886 pg/ml). Hydrocortisone, a drug effective in vivo for the treatment of HTLV-I uveitis, severely depressed cytokine production in vitro in most cases. In summary, the results demonstrated direct evidence of HTLV-I infection of the eye and suggest that cytokines produced by HTLV-I-infected T cells are responsible for the intraocular inflammation in patients with HTLV-I uveitis.

The distribution of DR5, MT2, and MB3 specificities on human Ia subsets.

Human Ia molecules were isolated from cells of LG-38, an HLA-homozygous lymphoid cell line of DR5 specificity. Three Ia subsets could be distinguished and separated by using specific alloantisera and a monoclonal antibody with polymorphic reactivity. These subsets carried the specificities DR5 and MT2, MT2 alone and MB3 alone. The structure of the three molecular species was analyzed by microfingerprinting. The subset carrying only MT2 was similar, in both the component α and β chains, to the major subset carrying DR5 and MT2, whereas the subset carrying MB3 was distinct in both chains from the other two subsets. These data are compatible with our previous findings obtained for the products of two Ia loci closely linked to the DR locus and provisionally called DC and BR; they also support the conclusion that the subset carrying only MT2 is an allelic product of the BR locus, whereas the MB3 subset is an allelic product of the DC locus. MT2 appears to be a shared specificity of DR and BR loci products.

HTLV-I uveitis

Human T-cell lymphotropic virus type I (HTLV-I) is known to cause adult T-cell leukemia/T-cell lymphoma and tropical spastic paraparesis/HTLV-I-associated myelopathy. Recent seroepidemiologic, clinical, and virologic studies indicate that the virus is also related to a certain type of uveitis, which has been classified as uveitis without defined etiologies or idiopathic uveitis. According to the seroepidemiologic survey, the seroprevalence of HTLV-I in patients with idiopathic uveitis was significantly higher than that of two control groups, that is, patients with uveitis with defined etiologies and patients with nonuveitic ocular diseases. Clinically, the uveitis seen in HTLV-I carriers is characterized by moderate to severe cellular infiltration in the eye and by moderate retinal vasculitis, and the intraocular inflammation responds well to corticosteroid therapy. Interestingly, 25% of female patients with the disease had a previous history of Graves disease with hyperthyroidisms. The following virologic, molecular biologic findings suggest that cytokines produced by HTLV-I-infected T cells in the eye play the central role in the pathogenic mechanisms of the uveitis: (a) the virus load in the peripheral blood monocytes analyzed by the quantitative polymerase chain reaction methods was significantly greater in patients with the uveitis than in asymptomatic carriers, (b) the proviral DNA of HTLV-I and the gene expression of the virus at the mRNA level was detected in the infiltrating cells from the eyes of the patients, (c) the virus particles were detected by electron-microscopic examination in the T-cell clones established from the intraocular fluid of the patients, and (d) the HTLV-I-infected T cells produced a variety of cytokines without any stimuli, such as interleukin (IL)-1 alpha, IL-2, IL-3, IL-6, IL-8, IL-10, tumor necrosis factor alpha, interferon-gamma, and granulocyte-macrophage colony-stimulating factor. Based on the seroepidemiologic, clinical, and virologic data, the uveitis seen in HTLV-I carriers is considered to be a distinct clinical entity related to HTLV-I infection, and the disease is designated as HTLV-I uveitis.

Immunopathology and cytokine detection in the skin lesions of patients with Kawasaki disease

To investigate the pathogenesis of Kawasaki disease, we examined biopsy specimens from the skin lesions of 10 Japanese patients at the site of the polymorphous exanthem on the hip and at the site of an inoculation with BCG (bacillus Calmette-Gruérin) vaccine of two of them. Ten normal patients were also examined for control purposes. The histopathologic features of the skin lesions were characterized by extensive edema with dilation of the small blood vessels in the papillary dermis; the site of BCG vaccine inoculation showed much greater inflammatory change; neutrophil emigration was weak and most of the infiltrates were CD4+ T lymphocytes and CD13+ macrophages. The expression of the DR locus of human leukocyte antigen was detected not only on the epidermal keratinocyte surface but also on the walls of the small blood vessels and the infiltrating cells around these blood vessels. We also detected cytokines in the skin lesions: (1) interleukin-1 alpha and tumor necrosis factor alpha were strongly positive in all patients with acute Kawasaki disease, (2) interleukin-2 and interferon gamma were weakly or partially positive, (3) no cytokines were detected in the convalescent phase, and (4) the amounts of cytokines at the site of BCG vaccine inoculations were larger than those at the site of the polymorphous exanthem. These findings suggest that CD4+ T lymphocytes and CD13+ macrophages are activated and interleukin-1 alpha and tumor necrosis factor alpha may be involved in the pathogenesis of the inflammation of acute Kawasaki disease.

Cytokine Production by T Cells Infiltrating in the Eye of Uveitis Patients

The capacity of T cells to produce cytokines was investigated using T-cell clones (TCCs) established from infiltrating cells in the aqueous humor (AH) or peripheral blood mononuclear cells (PBMC) of patients with Vogt-Koyanagi-Harada (VKH) disease or sarcoidosis. The cytokines produced and tested in the study were interleukin (IL)-1alpha, IL-6, IL-8, interferon (IFN)-gamma, tumor necrosis factor (TNF)-alpha, and granulocyte monocyte colony stimulating factor (GM-CSF). All TCCs (n = 9) from AH of VKH patients spontaneously produced significantly larger amounts of IL-6, IL-8, and IFN-gamma than TCCs from healthy donor PBMC. All TCCs (n = 9) from AH of the sarcoidosis patient spontaneously produced significantly larger amounts of IL-1alpha, IL-6, and IL-8 than TCCs from healthy donor PBMC. In addition, the effects of antiinflammatory drugs on the cytokine production by the TCCs were investigated. Hydrocortisone significantly suppressed the production of IL-6, IL-8, and GM-CSF by TCCs from AH of VKH patients. Tacrolimus also significantly suppressed the production of IL-8 and GM-CSF by the TCCs. FTY720, an experimental drug, suppressed only GM-CSF production by TCCs from AH of VKH patients. Diclofenac failed to suppress the production of any cytokines by any TCCs. All tested drugs did not suppress the production of cytokines by TCCs from the sarcoidosis patient. These results thus suggest that cytokines produced by T cells infiltrating in the eye may play an important role in the pathogenesis of uveitis.

Soluble Fas ligand and soluble Fas in ocular fluid of patients with uveitis

AIMS To investigate the presence of soluble Fas ligand (sFasL) and soluble Fas (sFas) in ocular fluid of patients with uveitis. METHODS Samples of aqueous humour (AH, n=17), vitreous fluid (n=9), and serum (n=60) were collected from patients with uveitis which included Behçets disease, Vogt–Koyanagi–Harada disease, sarcoidosis, human T lymphotropic virus type 1 (HTLV-I) uveitis, sympathetic ophthalmia, HLA-B27 associated acute anterior uveitis, and ocular toxoplasmosis. The AH of patients with age related cataract without uveitis obtained during cataract surgery was used as controls (n=20). The amounts of sFasL and sFas were measured by enzyme linked immunosorbent assay. RESULTS Significant amounts of sFasL were detected in AH of patients with age related cataract (non-uveitis group). sFasL was also detected in AH of patients with uveitis, though the amounts were slightly lower than those in the non-uveitis group. On the other hand, the levels of sFas in AH of patients with uveitis were significantly higher than those in controls. As for the disease activity, the levels of sFasL and sFas in the vitreous fluid of patients with active uveitis were significantly higher than those in inactive uveitis. sFasL in the serum of healthy donors and patients with uveitis was below detectable levels, except for patients with HTLV-I uveitis who had significant amounts of sFasL in the serum. The levels of sFas in the serum of patients with Behçets disease, sarcoidosis, and HTLV-I uveitis were significantly higher than those of healthy donors. CONCLUSIONS sFasL is present in the AH of non-uveitic eyes with age related cataract. Intraocular levels of sFasL and sFas are significantly increased in uveitis, particularly in active uveitis. These data suggest that intraocular sFasL and sFas may have a regulatory role in uveitis.

Characterization of CD4+ T helper cells in patients with Kawasaki disease (KD): preferential production of tumour necrosis factor‐alpha (TNF‐α) by Vβ2− or Vβ8− CD4+ T helper cells

KD is an acute febrile illness in children characterized by coronary arteritis accompanied by aneurysm and thrombotic occlusion. The etiology of KD is unknown. It has been recently reported that KD is associated with the selective expansion of Vβ2+ and Vβ8.1+ T cells in peripheral blood lymphocytes (PBL), by studying the T cell receptor (TCR) repertoire of in vitro activated T cells. KD may therefore be caused by a superantigen [1–3]. To understand better the immunopathology of KD, we investigated TCR Vβ2 and Vβ8.1 expression on both the T cells of freshly isolated PBL and T cell clones (TCC) from patients with KD. Cytokine production by TCC was also studied. Blood samples were obtained from patients with acute (n= 20) and convalescent (n= 20) KD, agematched children with non‐infectious diseases (n= 18), and healthy adults (n= 20). Among these four groups, there were no significant differences in the percentages of either Vβ2+ or Vβ8.1+. T cells of freshly isolated PBL. The same was true for the CD4+ or CD8+ T cell subsets. One hundred and five TCC (98 CD3+ CD4+ CD8− and seven CD3+ CDA− CD8+) established from the affected skin, lymph node or PBL of six patients with KD were also negative for either Vβ2 or Vβ8.1 TCR. Sixty‐eight of 105 TCC (65%) produced detectable levels (>5 pg/ml) of TNF‐α (6–1016 pg/ml), in the absence of any stimuli. In contrast, only 11 (10%) of 105 TCC or 7(7%) of 97 TCC produced detectable levels of IL‐2 or IL‐6, respectively, in the absence of any stimuli. Stimulation with phytohaemagglutinin (PHA) and phorbol myristate acetate (PMA) induced most TCC to produce higher amounts of TNF‐α, IL‐2 and IL‐6. These results suggest that CD4+ T helper cells expressing TCR‐β other than Vβ2 or Vβ8 receptor, primarily through TNF‐α production, are involved in the immunopathology of KD.

Induction of MAGE Genes in Lymphoid Cells by the Demethylating Agent 5-Aza-2′-deoxycytidine

MAGE genes encoding tumor antigens recognized by cytotoxic T lymphocytes are appropriate target molecules for specific immunotherapy of cancer. We have investigated whether the demethylating agent 5‐aza‐2′‐deoxycytidine (DAC) induces MAGE‐1, ‐2, ‐3, and ‐6 in normal and malignant lymphoid cells. DAC induced these MAGE genes in both PHA/interleukin‐2 (IL‐2)‐activated T cells from healthy donors and MAGE‐negative T and B cell leukemias in most cases. It also induced MAGE‐1 in IL‐2‐dependent T cell clones and all MAGE genes tested in Epstein‐Barr virus‐transformed B cell lines. Expression of MAGE‐1 protein in the cells was confirmed by western blot analysis with anti‐MAGE‐1 polyclonal antibody. Therefore, demethylation is a potent stimulus to induce MAGE genes in both normal and malignant lymphoid cells.

Co-Stimulation with anti-CD28 (Kolt-2) enhances DNA synthesis by defective T cells in common variable immunodeficiency

In normal T cells, an ann‐CD28 MoAb (Kolt‐2) will synergize with the mitogenic stimuli Phytohaemagglutinin (PHA). anti‐CD3 (OKT3) or a combination of anti‐CD2 antibodies (OKT11 and GT2) in the induction of DNA synthesis. A subgroup of patients with common variable immunodeficiency (CVID) show a defect in DNA synthesis by T cells stimulated in vitro with the above mitogens. We have now investigated whether anti‐CD28 will correct the defect. This strategy partially restored DNA synthesis, providing evidence that the CD28 co‐stimulatory pathway in CVID T cells is normal. Ligation of CT28 acts through co‐stimulating IL‐2 secretion. The natural ligand (B7) for CD28 on antigen‐presenting cells from CVID patients is expressed normally. We conclude that the defect in CVID T cells lies in pathways that lead to transcription of the IL‐2 gene other than that induced by ligation of CD2K with Kolt‐2.

Increased expression of interleukin-2 receptor α on peripheral blood mononuclear cells in HTLV-I tax/rex mRNA-positive asymptomatic carriers

Using the double-nested reverse transcription-polymerase chain reaction, we assayed human T-cell leukemia virus type I (HTLV-I) tax/rex-encoded mRNA in the peripheral blood mononuclear cells (PBMCs) of asymptomatic carriers as an index of the expression of HTLV-I in vivo in relation to the proviral DNA level. HTLV-I tax/rex mRNA was detected in only 1 (3.3%) of 30 samples with medium or lower proviral DNA levels, but it was detected in 11 (39.3%) of 28 samples with high HTLV-I proviral DNA levels, estimated as equal to or more than the proviral DNA of 10 ng of HUT102 (i.e., HUT102 cells were used as positive controls). The mean number of interleukin-2 receptor alpha (IL-2R alpha)-positive cells as a percentage of the total number of PBMCs was higher (13.2%) in the tax/rex mRNA-positive carriers with high proviral DNA levels than in the carriers who were mRNA negative (8.4%) (p = 0.004, Wilcoxon test). These results suggest that virus activation as indicated by the presence of tax/rex mRNA in asymptomatic carriers with high proviral DNA levels is associated with an elevation of the IL-2R alpha-positive cells in vivo.