Nuclear levels of NF-kappaB correlate with syncytium-forming capacity of 8e51 cells, expressing a defective HIV virus.

The double NF-kappaB site identified in the LTR of the human immunodeficiency virus-1 (HIV-1) has been demonstrated to be necessary for efficient viral transcription. In this report we present the characterisation of NF-kappaB subunits engaged in complexes binding to the HIV-1 NF-kappaB site in human 8e51 T-cells, that harbour a defective HIV-1. At least four different specific NF-kappaB complexes are present in the nucleus of these cells. With the use of specific antibodies we have determined the composition of each complex using electrophoretic mobility shift assays. The results show the presence of several NF-kappaB family members, with the transactivating RelA being engaged in multiple complexes. The importance of NF-kappaB complexes in viral functions has been established comparing the level of NF-kappaB DNA-binding complexes with syncytia-forming activity of 8e51 cells. In fact, 8e51 cells that had almost lost their syncytia-forming capacity were found to contain at least 10 times less active NF-kappaB DNA-binding complex than the actively fusing cells. The correlation is specific as the level of at least three other transcription factors did not change.

Role of HLA class I in HIV type 1-induced syncytium formation.

Neutralization of HIV-1 in vitro by anti-HLA class I antibodies suggests that class I molecules are involved in HIV-1 infection. HIV-infected cells can fuse with uninfected cells in a process that leads to the formation of multinucleated syncytia, involving an interaction between host and viral antigens expressed at the cell surfaces. We used a syncytium assay between the 8E5 cell line chronically infected with a pol-defective variant of LAV IIIb, and the CD4-positive cell line MOLT3, to study the role of HLA class I in HIV-1-induced cell fusion. By probing cells with a panel of anti-HLA monoclonal antibodies (MABs) we demonstrated that the fusion process is modulated specifically by C alleles of HLA class I expressed on uninfected cells but not by that on already infected cells. Addition of beta 2-microglobulin to the cocultures resulted in a dose-dependent enhancement in both the number and size of syncytia, whereas exogenous HLA-C-restricted peptides inhibited syncytium formation, implying that only certain conformational states of HLA class I are permissive for syncytium formation. Treatment of cocultures with HLA-Cw4-restricted peptides containing amino acid substitutions in the anchor residues showed that syncytium inhibition was dependent on conventional binding of the peptide inside the groove. The data indicate that HLA class I, in a conformation free of peptide but associated with beta 2-microglobulin, can directly influence virus-induced cell fusion.

Autoantibodies to CD4 in HIV type 1-exposed seronegative individuals.

The aim of this study was to investigate the presence and the fine specificity of anti-CD4 autoantibodies in seronegative subjects sexually exposed to HIV-1. Anti-CD4 autoantibodies were previously detected in a fraction of HIV-1-seropositive individuals. Whole sera, purified IgG fractions, and supernatants of EBV-transformed lymphoblastoid cell lines were analyzed by means of ELISA, Western blot, and by competition assays using monoclonal antibodies with known fine specificities. Anti-CD4 antibodies were found in 6 of 18 individuals exposed to HIV-1 infection and who have been persistently seronegative. These antibodies inhibited HIV-1-driven syncytium formation, did not interfere with the CD4-gp120 interaction, and competed for CD4 binding with two of three anti-CD4 monoclonals with known fine specificities. Moreover, autoantibodies with the same fine specificities were found in the supernatants of oligoclonal EBV-transformed B cell lines derived from these individuals. At variance, in the HIV-1-positive patients included in our study, the anti-CD4 antibody response was directed to a broader panel of epitopes, including those involved in CD4-gp120 interactions. In conclusion, anti-CD4 antibodies specific for defined epitopes of the CD4 molecule are generated in the course of an early immune response to HIV-1 antigens in the absence of other signs of infection, as they can be detected by conventional methods. These autoantibodies may play a protective role either alone or in association with other cellular and humoral factors.

Human immunodeficiency virus type 1 (HIV-1)-seronegative injection drug users at risk for HIV exposure have antibodies to HLA class I antigens and T cells specific for HIV envelope.

The question of whether persistently seronegative persons at high risk for human immunodeficiency virus type 1 (HIV-1) infection exhibit HIV-1-specific T cell responses and antibodies to HIV-1 envelope epitopes shared with selected HLAs was assessed. These antibodies are not detectable by conventional serologic methods. Envelope-specific helper T (Env-Th) cell responses and antibodies specific for the HIV/HLA epitopes were studied in 21 HIV-1-negative injection drug users (IDUs). HIV/HLA antibodies were detected in 7 (33.3%) of 21 IDUs and 4 (4.3%) of 94 low-risk controls. Env-Th cell responses were detected in 16 (76.2%) of 21 IDUs and in 2 (3.1%) of 65 low-risk controls. All HIV/HLA antibody-positive IDUs also had Env-Th cell responses. These findings confirm the presence of HIV-1-specific immunity in conventionally seronegative individuals. Further characterization of these responses could provide the basis for new preventive strategies.

Expression of distinct conformations of free HLA-Cw4 heavy chains in transfected neuroblastoma cells.

Epitope mapping of HLA-Cw4 indicates that the two monoclonal antibodies (mAbs) L31 and M38, specific for beta 2-microglobulin (beta2m)-free HLA-C heavy chains, react preferentially with the KYK motif, located in the binding groove (alpha1 domain). Transfection of HLA-Cw4 cDNA into a neuroblastoma cell line, which normally expresses negligible HLA class I, resulted in the constitutive surface expression of molecules displaying different reactivities with the two mAbs. This cellular system was used to determine whether L31 and M38 recognize distinct conformations of beta2m-free HLA-C proteins, and to investigate their mechanism of expression. Interferon-gamma greatly enhanced the expression of L31-reactive free chains, while abolishing that of M38-reactive molecules. The cytokine-induced expression of L31-reactive molecules was inhibited by anti-sense oligonucleotides specific for beta2m mRNA, while constitutive expression of L31-reactive molecules was only partially affected. Exogenous beta2m resulted in a reduction of constitutive L31 reactivity, and in a concomitant increase of M38 reactivity. These results indicate that: 1) at the cell surface, L31 and M38 react with two distinct conformations of HLA-Cw4 beta2m-free heavy chains, of which the L31-reactive conformation is the least folded; 2) the expression of both conformers can be modulated by endogenous or exogenous beta2m; and (3) L31-reactive molecules exposed at the cell surface are likely to derive from the dissociation of empty HLA-Cw4/beta2m complexes.

Human antibodies to immunodominant C5 region of HIV-1 gp120 cross-react with HLA class I on activated cells.

Cross-reactive antibodies to HLA class I and HIV-1 gp120 were detected in the sera of HIV-1-positive individuals, and were found to share the same epitope specificity as a gp120-HLA class I cross-reactive monoclonal antibody (M38). The amino acid residues of HLA recognized by both M38 and the patient antibodies occur as a clustered pair in the HLA-C alpha 1 domain. These sequences (KYKR and RKLR) are shared by almost all HLA-C alleles and are available to antibody binding only on beta 2-microglobulin-dissociated HLA heavy chains expressed on activated cells. Similar to M38, the antibody-binding sites on HIV-1 gp120 were mapped to two noncontiguous stretches of amino acids (KYK and KAKR), which flank a hydrophobic area of the immunodominant C5 region involved in gp41 binding. Serum antibodies immunoaffinity purified on synthetic HLA and gp120 peptides representing the M38-reactive regions were shown to bind to both HLA and gp120 in Western blot, as well as to membrane-bound HLA heavy chains, and to exhibit selective complement-mediated lysis of activated T cells. No serum antibodies could be detected that bind to the gp120 C5 region (peptide IEPLGVAPT) flanked by the two HLA-like sequences.

Cross-reactive response to human immunodeficiency virus type 1 (HIV-1) gp120 and HLA class I heavy chains induced by receipt of HIV-1-derived envelope vaccines.

Autoantibodies specific to HLA class I antigens were detected in the sera of persons vaccinated with human immunodeficiency virus type 1-derived recombinant vaccines by using synthetic peptides representing the amino acid sequences recognized by an HLA class I/gp120 cross-reactive monoclonal antibody. Study subjects received recombinant vaccinia gp160 (vacc-env) alone, vacc-env followed by one dose of recombinant gp160 (rgp160, 640 micrograms), or four doses of rgp160 alone (640 or 80 micrograms). All sera from vacc-env/rgp160-vaccinated subjects contained HLA/gp120 cross-reactive antibodies, as did all sera from recipients of the rgp160 alone at 640 micrograms/dose. In contrast, none of the sera from subjects who received either the vacc-env alone or the 80 micrograms/dose rgp160 alone contained detectable HLA cross-reactive antibodies, and these same sera showed little or no envelope reactivity on Western blot. The results showed a striking correlation between immunogenicity and the induction of cross-reactive antibodies by the rgp160 vaccine.

Association of methotrexate treatment with a decrease of double negative (CD4-CD8-) and gamma/delta T cell levels in patients with juvenile rheumatoid arthritis.

OBJECTIVE: It has been suggested that double negative (CD4-CD8-) (DN) and gamma/delta T cells may be involved in some autoimmune diseases. We investigated peripheral blood DN and gamma/delta T cell levels in patients with active juvenile rheumatoid arthritis (JRA). METHODS: DN and gamma/delta T cell levels were measured in 42 patients with active JRA and in 10 healthy controls comparable for age by an immunofluorescence double staining procedure. RESULTS: All 3 JRA onset types had DN and gamma/delta T cell levels not significantly different from those of controls, although a wide scattering of data was present. No correlation was found between DN or gamma/delta T cell levels and erythrocyte sedimentation rate values or the number of active joints. When patients were divided according to treatment, we found that DN and gamma/delta T cell levels were significantly lower (p = 0.001, p = 0.02, respectively) in patients receiving methotrexate (MTX) than in patients not receiving MTX. The association of MTX treatment with a decrease in DN and gamma/delta T cell levels was also confirmed in a followup study of individual patients. Among patients not receiving MTX, patients with systemic JRA presented DN T cell levels significantly higher than those of controls. In 5 patients with pauciarticular JRA DN and gamma/delta T cell levels were higher in synovial fluid than in the peripheral blood. CONCLUSIONS: We found an increase in peripheral blood DN T cell levels in systemic JRA; treatment with MTX appears to be associated with a decrease in DN and gamma/delta T cell levels.

Serum interleukin-6 levels and joint involvement in polyarticular and pauciarticular juvenile chronic arthritis.

We measured serum and synovial fluid interleukin 6 (IL-6) levels in patients with polyarticular-onset and pauciarticular-onset juvenile chronic arthritis (JCA), using the hybridoma cell line B9. During active disease, but not during remission, serum IL-6 levels were significantly elevated in patients with polyarticular (p < 0.0001 vs controls) and in patients with pauciarticular JCA (p < 0.01 vs controls). In patients with active polyarticular disease (polyarticular and extended pauciarticular) serum IL-6 levels correlated with the extent and severity of joint involvement (p < 0.05), with erythrocyte sedimentation rate values (p < 0.005) and with C reactive protein concentrations (p < 0.05). In patients with persistent pauciarticular JCA, serum IL-6 levels correlated with the joint swelling score (p < 0.05). Synovial fluid IL-6 levels, measured in 5 patients with pauciarticular JCA, were markedly elevated (range 600-24,000 HGF U/ml). In conclusion, our data suggest that IL-6 is an important pathogenic mediator in polyarticular and pauciarticular JCA.

Correlation of serum interleukin-6 levels with joint involvement and thrombocytosis in systemic juvenile rheumatoid arthritis.

We measured interleukin-6 (IL-6) levels in 70 serum samples obtained from 25 patients with systemic-onset juvenile rheumatoid arthritis (JRA), using the hybridoma cell line B9. Patients with systemic-onset JRA had significantly elevated serum IL-6 levels during active disease (mean +/- SD 92.1 +/- 75.1 hybridoma growth factor units/ml; P less than 0.00001 versus healthy age-matched controls), but not during remission. Serum IL-6 levels correlated with the extent and severity of joint involvement (P less than 0.001) and with platelet counts (P less than 0.05). Our data suggest that IL-6 plays a significant role in the pathogenesis of systemic-onset JRA.