B cell responses to HIV and the development of human monoclonal antibodies

Enriching for HIV-infected cells using anti-gp41 antibodies indirectly conjugated to magnetic microbeads

The isolation of a pure population of human immunodeficiency virus (HIV)-infected cells is highly desirable for evaluating the impact of HIV on cellular gene expression. Given that HIV gp41 transmembrane protein is anchored on the surface of HIV-infected cells, we evaluated the use of pooled anti-gp41 monoclonal antibodies (MAbs) and HIV immunoglobulins (HIV-Igs) indirectly conjugated to magnetic microbeads to positively select for infected cells. We demonstrate that pooled anti-gp41 monoclonal antibodies enriched for H9 cells infected with HIV IIIB by approximately 98%. Peripheral blood mononuclear cells (PBMCs) infected with a primary (HIV strain 302151) or laboratory-adapted (IIIB) strain were enriched by 54%-62%, depending on the initial viral inoculum. Using HIV-Ig in this magnetic positive-selection approach was also highly efficient for enriching for H9 cells infected with IIIB but less efficient for infected PBMCs. Both types of antibodies used in the selection process resulted in > 80 viability of selected HIV-infected cells. Analysis of interleukin 2 (IL-2) mRNA expression using real-time reverse transcription PCR (RT-PCR) of the HIV-enriched population demonstrated a higher level of IL-2 mRNA, by approximately four cycles, and an 8-fold increase in IL-2 expression, as evaluated by intracellular staining and flow cytometric analysis, in comparison to gp41-negative cells. Collectively, these data illustrate that antibodies targeting gp41 can be used to enrich for HIV-positive populations. This represents a novel approach for studying the impact of HIV on infected cells and on bystander/uninfected cells.

Immunoglobulin levels amongst persons with and without human immunodeficiency virus type 1 infection in Uganda and Norway

CD4+-cell count and viral load monitoring are expensive and unavailable to most human immunodeficiency virus (HIV)-infected people in Africa. In an attempt to evaluate alternative methods for monitoring antiretroviral (ARV) therapy, we measured concentrations of immunoglobulin (Ig)A, IgM, IgG and IgG1 amongst adults with and without HIV in Uganda and Norway. We adjusted for disease severity by stratifying HIV-positive subjects on CD4+-cell counts above and below 200 cells/ micro l. Median serum levels of IgG, IgG1 and IgA were significantly higher in HIV-positive persons compared with HIV-negative persons in both countries (P < 0.001 and P = 0.018 for IgA in Ugandan patients). Levels of IgA in Ugandan HIV-negative subjects were significantly lower than those in HIV-positive subjects with low CD4+ compared with those with high CD4+-cell counts (P < 0.001 and P = 0.069, respectively). IgM levels were different between the HIV-negative and the two HIV-positive groups in Norway (P < 0.001). The mean levels of IgM, IgG and IgG1 in HIV-negative and -positive African subjects were generally higher than those in comparable groups of Western subjects. Our results verify that levels of IgA, IgG and IgG1 vary between HIV-negative and -positive individuals in both study populations. Their determination may be useful in monitoring both disease progression and response to ARV therapy.

Association of phenotypic changes in B cell lymphocytes and plasma viral load in human immunodeficiency virus-infected patients

Many B cell abnormalities have been reported in human immunodeficiency virus (HIV)-infected patients, including changes in the expression of mu, gamma, and CD22 molecules on the cell surface. Phenotypic changes in these markers on B cells isolated from HIV-seropositive patients with high or low levels of plasma viremia were measured. The phenotypic changes in B cells isolated from such patients were compared with the markers on B cells isolated from HIV-seronegative individuals using three-color flow cytometry. HIV patients showed a reduction in the proportion of mature B cells isolated from peripheral blood mononuclear cells compared with B cells isolated from HIV-seronegative individuals. An increase in the proportion of B cells expressing both mu and gamma molecules on the cell surface was also seen in association with high-HIV plasma viremia. A low plasma viral load was accompanied by a reduction in the proportion of B cells expressing both mu and gamma molecules to a level comparable to those seen in HIV-seronegative individuals. HIV-seropositive individuals demonstrated an increase in the proportion of committed B cells, as indicated by an increase in the proportion of B cells expressing gamma molecules. This observation may explain the poor humoral response of HIV seropositive patients to neo-antigens. Our results demonstrate that phenotypic changes indicative of in vivo B cell activation and an increase in immature cells are associated with HIV infection, particularly with a high plasma viral load. Phenotypic changes in B cell markers may correlate with functional deficits of B cells.

Role of complement in HIV infection

In human plasma, HIV activates the complement system, even in the absence of specific antibodies. Complement activation would, however, be harmful to the virus if the reactions were allowed to go to completion, since their final outcome would be virolysis. This is avoided by complement regulatory molecules, which either are included in the virus membrane upon budding from the infected cells (e.g. DAF/CD55) or are secondarily attached to HIV envelope glycoproteins as in the case of factor H. By using this strategy of interaction with complement components, HIV takes advantage of human complement activation for enhancement of infectivity, for follicular localization, and for broadening its target cell range at the same time that it displays an intrinsic resistance against the lytic action of human complement. This intrinsic resistance to complement-mediated virolysis can be overcome by monoclonal antibodies inhibiting recruitment of human factor H to the virus surface, suggesting a new therapeutic principle.

Immunoglobulin and complement complexes in blood following infection with human immunodeficiency virus type 1

Freely soluble and complexed plasma immunoglobulin A (IgA), IgG, IgM, C1q, C3, and factor B in 36 human immunodeficiency virus type I (HIV-1)-seronegative controls, 69 asymptomatic HIV+ subjects, and 117 individuals with symptomatic HIV-associated disease were characterized. Levels of free and complexed IgG and IgA, and to a lesser extent free C1q and complexed IgM, increased with HIV-1 infection. In stark contrast, both HIV+ groups showed three- to sixfold declines in complexed C3, C1q, and factor B levels. The asymptomatic HIV+ population showed declines in levels of C3-bound IgA, IgG2, and IgG4 complexes. The asymptomatic group showed reductions in C3-complexed IgM, IgA, IgG2, and IgG4 levels. HIV infection is associated with complement-deficient immune complexes.

Primary in vitro immunization with multimeric synthetic peptides of HIV-1 envelope glycoproteins: generation of neutralizing human monoclonal antibodies

Peripheral blood lymphocytes from healthy HIV-1 seronegative donors were immunized in vitro with the following synthetic peptides: (i) an octameric poly-L-lysine conjugated peptide of the HIV-1MN V3 loop and (ii) a resin bound synthetic peptide aa642-665 of HIV-1 gp41. Lymphoblastoid cell lines (LCL) were obtained by immortalization with Epstein-Barr virus (EBV). We produced four LCL secreting human monoclonal antibodies (HuMoAbs) of the IgM isotype: three were directed against the V3 domain (FC10, FC81 and CF41) and one against aa642-665 (CA45C). Two of these HuMoAbs (FC81 and CA45C) reacted to viral surface antigen on HIV-1-infected cells. All the HuMoAbs inhibited 40-53% of cell fusion induced by HIV-1-infected H9 cells at 5 micrograms/ml. They also neutralized, at lower concentrations, cell-free infection with HIV-1MN, HIV-1IIIB and four primary clinical HIV-1 isolates. No enhancing activity of the HuMoAbs in the presence of complement was observed. The results presented here show the feasibility of generating neutralizing human monoclonal antibodies against HIV-1 by primary in vitro immunization with selected synthetic peptides of HIV-1 envelope glycoproteins. This approach has provided tools for further studies of synergistic neutralization assays, and generated potential immunoglobulin candidates for passive immunotherapy.

Human monoclonal or polyclonal antibodies recognize predominantly discontinuous epitopes on bee venom phospholipase A2

BACKGROUND

Two hybridomas, which secrete human monoclonal antibodies of IgG4 isotype specific for the main bee venom antigen/allergen phospholipase A2, were generated. The antigenic determinants recognized by these antibodies were mapped and compared with the binding sites of murine monoclonal and human polyclonal antibodies raised against the same antigen.

METHODS

Two hybridomas were developed by fusing heteromyelomas to Epstein-Barr virus immortalized B cells obtained from beekeepers. The cloned hybridomas were stable and secreted up to 40 mg/L of antibody into the culture supernatant. Phospholipase A2 specificity of the human monoclonal antibodies was confirmed by binding and inhibition ELISA and by Western blot analysis. Epitope mapping on phospholipase A2 was done with the PEPSCAN method and ELISA techniques.

RESULTS

The epitopes recognized by the human monoclonal antibodies were shown to be discontinuous and did not contain the sugar residue. Similar results were obtained with polyclonal antibodies of IgG4 isotype (from beekeepers) specific for phospholipase A2, which could also inhibit the binding of the human monoclonal antibodies to phospholipase A2. In contrast, antigen binding of the human monoclonal antibodies could not be inhibited by murine monoclonal antibodies against bee venom phospholipase A2.

CONCLUSIONS

The data indicate that the human monoclonal antibodies obtained are representative of a part of the polyclonal immune response to phospholipase A2 from beekeepers and may allow a more precise analysis of the humoral immune response to phospholipase A2 that is associated with protection.

Immunodeficiency and IL-6 production by peripheral blood monocytes in multicentric Castleman's disease

To study the pathogenesis of multicentric Castleman's disease (MCD), IL-6 producing cells and immune function were investigated in four MCD patients. The expression of IL-6 mRNA in one MCD lymph node was analysed by in situ hybridization. IL-6 mRNA expressing cells were scattered in the interfollicular areas and did not resemble plasma cells. Spontaneous IL-6 production was detected in the culture supernatants of peripheral blood mononuclear cells (PBMNC) from four patients. The IL-6 producing cells among the PBMNC were found to be monocytes by both in situ hybridization and immunohistochemistry. We evaluated immune function in four MCD patients. These studies show: (1) a negative PPD skin test in 3/4 patients, (2) decreased IL-2 production in 3/4 patients, (3) decreased T cell colony formation in 3/4 patients, (4) decreased NK activity and NK cell number in 2/4 patients, (5) increased soluble IL-2 receptor in 4/4 patients, and (6) decreased CD4/CD8 ratio in 3/4 patients. These results show that MCD resembles, in several ways, acquired immunodeficiency syndrome (AIDS).

An anti-gp41 human monoclonal antibody that enhances HIV-1 infection in the absence of complement

B lymphocytes from tonsillar tissue of an asymptomatic HIV-1-seropositive subject were transformed with Epstein-Barr virus (EBV) and tested for the production of HIV-1-specific antibodies by ELISA, using purified HIV-1SF2.2F11, a monoclonal antibody derived from a transformed line, is of the IgG1 subclass and recognizes an epitope in the conserved region of the envelope transmembrane glycoprotein gp41, which is expressed on the surface of HIV-infected T cells. The antibody does not mediate the lysis of infected T cells in antibody-dependent cellular cytotoxicity (ADCC) assays and does not neutralize the infectivity of HIV-1SF2 or the homologous isolate HIV-1TT2.2F11 appears to be the first anti-gp41 human monoclonal antibody that enhances the infectivity of an HIV-1 strain (i.e., SF128A) in the absence of complement.

Complement activation by human monoclonal antibodies to human immunodeficiency virus

It has been shown that the incubation of human immunodeficiency virus (HIV) with polyclonal antibodies from HIV-infected persons and complement results in complement-mediated neutralization due, at least in part, to virolysis. The current study was performed to determine whether any of a panel of 16 human monoclonal antibodies to HIV could activate complement and, if so, which determinants of the HIV envelope could serve as targets for antibody-dependent complement-mediated effects. Human monoclonal antibodies directed to the third variable region (V3 region) of HIVMN gp120 induced C3 deposition on infected cells and virolysis of free virus. Antibodies to two other sites on HIVMN gp120 and two sites on gp41 induced few or no complement-mediated effects. Similarly, only anti-V3 antibodies efficiently caused complement-mediated effects on the HIVIIIB isolate. In general, the level of C3 deposition on infected cells paralleled the relative level of bound monoclonal antibodies. As expected, pooled polyclonal antibodies from infected persons were much more efficient than monoclonal antibodies inducing C3 deposition per unit of bound immunoglobulin. Treatment of virus or infected cells with soluble CD4 resulted in increases in anti-gp41 antibody-mediated virolysis and C3 deposition but decreases in anti-V3 antibody-mediated virolysis and C3 deposition. In general, virolysis of HIV was more sensitive as an indicator of complement-mediated effects than infected-cell surface C3 deposition, suggesting the absence of or reduced expression of functional complement control proteins on the surface of free virus. Thus, this study shows that human monoclonal antibodies to the V3 region of gp120 are most efficient in causing virolysis of free virus and C3 deposition on infected cells. Elution of gp120 with soluble CD4 exposes epitopes on gp41 that can also bind antibody, resulting in virolysis and C3 deposition. These findings establish a serologically defined model system for the further study of the interaction of complement and HIV.