Functional and molecular characterization of human monoclonal antibody reactive with the immunodominant region of HIV type 1 glycoprotein 41.

The immunoreactivity, functional activity, and molecular features of a human monoclonal antibody (HMAb), F240, from an HIV-1-infected individual have been studied. Flow cytometric analysis demonstrated that F240 is reactive with cells infected with a broad range of laboratory isolates but not with uninfected cells. Reactivity of F240 is greatly enhanced by preincubation of infected cells with soluble CD4, and to a much lesser extent, with F105, an HMAb reactive with the CD4-binding site of gp120. This enhancement is temperature dependent, with maximum enhancement observed at 37 degrees C, and suggests that the F240 epitope may be more accessible after gp120 has bound to CD4 in vivo. Immunoblot analysis reveals antigen specificity of F240 for gp41 or its precursor gp160. F240 specificity is mapped to the immunodominant region of the gp41 ectodomain by Pepscan analysis. This epitope has been implicated in eliciting nonprotective antibodies that enhance infection in the presence of complement. Consistent with this, F240 failed to neutralize laboratory isolates and enhanced viral infection in a complement-dependent manner. The F240 VH demonstrates extensive somatic mutations compared with the product of its closest homologous germline gene VH3-3.11. Most amino acid substitutions occur in CDR2, characteristic of an antigen-driven response, and in FR3, a phenomenon observed in other anti-HIV-1 envelope HMAbs. Primary structure analysis of the F240 heavy chain revealed strong homology in the CDR domains to an HMAb (3D6) reactive with the same gp41 region, which suggests that these HMAbs could define a potential human antibody clonotype.

Influence of heavy chain constant regions on antigen binding and HIV-1 neutralization by a human monoclonal antibody.

F105, a neutralizing IgG1 kappa human mAb, is reactive with a discontinuous epitope within the gp120 CD4 binding site. Because isotype usage may affect Ab function, we examined the effect of isotype on Ag/Ab interactions and HIV-1 neutralization. An IgG3 kappa Ab was prepared by linking the variable regions of F105 to cloned human kappa and gamma 3 constant regions. Immunoreactivity of F105 IgG1 and IgG3 with IIIB-, MN-, and RF-infected cells was equivalent. Inhibition of binding and fusion of IIIB to uninfected cells and neutralization of IIIB virus was comparable for F105 IgG1 and IgG3, with 14 to 23 micrograms/ml required for 90% neutralization. In contrast, F105 IgG3 was marginally more effective at inhibition of MN binding/fusion and significantly more effective at neutralization of MN virus (62 micrograms/ml for IgG3 and > 100 micrograms/ml for IgG1 to achieve 90% neutralization). Despite high affinity binding to RF-infected cells, F105 IgG1 minimally neutralizes free RF virus. F105 IgG3 is dramatically more effective against the RF isolate, with 2 to 20 micrograms/ml of Ab required for 50% neutralization. Both isotypes were relatively ineffective at inhibition of RF binding/fusion. Thus, whereas affinity with native Ags on the surface of HIV-1-infected cells was unaffected by heavy chain constant regions, Ab isotype can strongly influence virion neutralization. Structural changes in gp120, as a result of increased flexibility conferred by the elongated IgG3 hinge region, are suggested as a possible mechanism to increase neutralization of selected HIV-1 isolates. These results may have significant implications in the design of immunotherapeutic and vaccine agents.

Plasma pharmacokinetics and biological activity of a human immunodeficiency virus type 1 neutralizing human monoclonal antibody, F105, in cynomolgus monkeys.

The IgG1 kappa human monoclonal antibody (HMab), F105 reacts with a discontinuous epitope on the CD4 binding site (CD4BS) of human immunodeficiency virus type 1 (HIV-1)/gp120 and has broad neutralizing activity. F105 HMab (60 mg/kg bolus) was administered intravenously to four monkeys and serum was collected at intervals to determine pharmacokinetics in a primate model. Average serum F105 concentrations, as determined by enzyme-linked immunosorbent assay, were analyzed with MINSQ software using a two-compartment, first-order model. The half-life for the alpha phase of the distribution curve is 6.7 h and for the beta elimination phase, 9.6 days. The volume of distribution is 0.65 L/kg and the rate of clearance 2 ml/kg/h. Serum levels of 1.3-1.6 mg/ml of F105 were maintained for 24 h. When monkey serum from day 15 postdose was tested, total serum F105 was 230 +/- 79 micrograms/ml and was immunoreactive with cells infected with the MN and IIIB strains of HIV-1 as determined by flow cytometry. Binding activity was identical to that obtained with stock F105 HMab. Identical neutralizing activity between the injected and uninjected antibody was also observed. Thus, serum neutralizing titers (90%) of 1:2000 at peak and 1:30 at day 15 postdose for MN virus were observed. These data indicate that high in vivo levels of HMab F105 can be attained by single bolus administration with full retention of biological activity. Of importance, levels of antibody necessary for effective neutralization can be achieved and maintained.

Effect of antibody valency on interaction with cell-surface expressed HIV-1 and viral neutralization.

F(ab) and F(ab')2 fragments of the human mAb, F105, were compared to intact IgG1 for binding to the CD4 binding site of HIV-1/gp120 on the surface of infected cells and viral neutralization. F105 IgG1 and F(ab')2 bound to IIIB, MN, and RF infected cells and neutralized these strains in an identical fashion, whereas strain-specific differences were observed in F(ab) activity. Although F105 F(ab) bound with equivalent affinity to IIIB-infected cells, there was a 4- to 10-fold decrease in the neutralization of IIIB by monovalent F(ab) compared to the bivalent molecules. F105 F(ab) demonstrated both diminished binding and neutralization of the MN strain and failed to bind or neutralize the RF strain. When cooperativity of V3 loop antibody (V3ab) with F105 IgG and fragments was examined, the binding of F105 IgG and F(ab')2 to IIIB-, MN-, or RF-infected cells was modestly enhanced by V3ab; viral neutralization was substantially enhanced by the combination of V3ab and F105 IgG and F(ab')2. The combination of F105 F(ab) with V3ab also resulted in significant cooperative neutralization of IIIB and MN, but the lack of F105 F(ab) binding and neutralization of RF was unaltered by V3ab. These results suggest that bivalent interaction may be important in binding and neutralization of virus, and support the notion that this interaction may depend on conformational changes in oligomeric gp120 on intact virions and cell surface rather than on affinity or steric effects.

Loss of serum antibodies to a conformational epitope of HIV-1/gp120 identified by a human monoclonal antibody is associated with disease progression.

Serum antibody reactive with epitopes within the CD4 binding site (CD4BS) of HIV-1/gp120 on infected cells was measured by inhibition of binding of a human monoclonal antibody, F105, which recognizes a conformational epitope within this region. Serum samples from 27% of ARC/AIDS patients blocked binding of F105 to this epitope, while samples from 100% of asymptomatic seropositive patients blocked binding. F105 blocking activity increased in 87% of asymptomatic donors who maintained stable disease over a 3-6 year period and decreased in 50% of asymptomatic patients with progressive disease. Moreover, serum samples from patients with stable disease exhibited higher titers of F105 blocking activity. The presence of F105 blocking activity also correlated with serum neutralization of virus. When diluted 1:640, serum with low F105 blocking activity neutralized only 20-30% of viral cytopathic effect (CPE), while serum with high F105 blocking activity neutralized > 80%. Serum neutralization was enhanced by the addition of F105. Seroreactivity to infected cells was detectable within 6 months of seroconversion, but F105 blocking activity was delayed by an additional 6-12 months, as was the development of high titers of neutralizing antibody. These data support the notion that the humoral response to the CD4BS on gp120 may be important in the maintenance of health.

Human monoclonal antibodies to the V3 loop of HIV-1 gp120 mediate variable and distinct effects on binding and viral neutralization by a human monoclonal antibody to the CD4 binding site.

Interactive effects between human monoclonal antibodies specific for the V3 loop (257-D and 447-D) and an epitope within the CD4 binding site (F105) of HIV-1 gp120 were evaluated for neutralization of viral cytopathogenicity and binding to HIV-infected cells. Regardless of antibody pair, only additive effects were observed in neutralization of MN and SF2 virus though each antibody alone had potent neutralizing activity on these strains. Significant cooperativity was observed between F105 and 447-D in neutralization of RF. Relatively high concentrations (> 100 micrograms/ml) of each individual antibody are required for partial neutralization (25--40%) of RF. Coincubation with 10 micrograms/ml of each antibody increased neutralization activity 3--4-fold more than predicted for additive effects alone. No enhancement was seen upon coincubation of F105 with 257-D which does not neutralize RF. Antibody interactions with native antigen on HIV-infected cells was measured by flow cytometry. Results were consistent with neutralization results in the majority of flow cytometry experiments; however, enhanced binding did not necessarily predict enhanced neutralization. These data support the notion that either a conformational change occurs with binding of V3 loop antibodies which enhances the binding and neutralizing activity of antibodies directed to the CD4 binding site of gp120 or vice versa, or new antigenic sites are exposed by the V3 loop antibodies on cell surfaces and virions. Of importance, cooperativity is observed even at very low antibody concentrations.

Neutralization of HIV-1 by F105, a human monoclonal antibody to the CD4 binding site of gp120.

The functional ability of the human monoclonal antibody (HMab) F105 to neutralize commonly available laboratory strains and a selection of primary isolates of human immunodeficiency virus (HIV)-1 was studied. F105 is representative of a class of human antibodies that react with conformational epitopes within the discontinuous CD4 binding site on HIV-1 envelope glycoprotein gp120. F105 binds with relatively similar affinities to native antigen expressed on the surfaces of cells infected with each of five laboratory isolates tested (IIIB, SF2, MN, RF, and CC) and neutralizes SF2, IIIB, and MN with concentrations of antibody ranging from 140 ng to 10 micrograms/ml. Nonetheless, neutralization by F105 alone of RF and CC is poor at modest antibody concentrations despite high affinity binding to surface gp120 on infected cells. Neutralization of HIV-1 strains by F105 is unaffected by normal sera and cooperativity is observed with serum samples from HIV-1 infected patients. Of significance, neutralization of RF and MN by F105 is enhanced by some HIV-seropositive sera at low concentrations. F105 also neutralized 30% of HIV-1 primary isolates in lymphocyte cultures. Although it is unclear how relevant in vitro studies will be to in vivo events, these data allow comparison of F105 with other HMabs to the CD4 binding site and V3 loop and provide an in vitro reference for in vivo activity. These studies demonstrate that antibody interactions among different classes of antibodies may be important in in vivo neutralization of HIV-1.