Maturation Pathways of Cross-Reactive HIV-1 Neutralizing Antibodies

Abstract

Several human monoclonal antibodies (hmAbs) and antibody fragments, including the best characterized in terms of structure-function b12 and Fab X5, exhibit relatively potent and broad HIV-1 neutralizing activity. However, the elicitation of b12 or b12-like antibodies in vivo by vaccine immunogens based on the HIV-1 envelope glycoprotein (Env) has not been successful. B12 is highly divergent from the closest corresponding germline antibody while X5 is less divergent. We have hypothesized that the relatively high degree of specific somatic hypermutations may preclude binding of the HIV-1 envelope glycoprotein (Env) to closest germline antibodies, and that identifying antibodies that are intermediates in the pathways to maturation could help design novel vaccine immunogens to guide the immune system for their enhanced elicitation. In support of this hypothesis we have previously found that a germline-like b12 (monovalent and bivalent scFv as an Fc fusion protein or IgG) lacks measurable binding to an Env as measured by ELISA with a sensitivity in the μM range [1]; here we present evidence confirming and expanding these findings for a panel of Envs. In contrast, a germline-like scFv X5 bound Env with high (nM) affinity. To begin to explore the maturation pathways of these antibodies we identified several possible b12 intermediate antibodies and tested their neutralizing activity. These intermediate antibodies neutralized only some HIV-1 isolates and with relatively weak potency. In contrast, germline-like scFv X5 neutralized a subset of the tested HIV-1 isolates with comparable efficiencies to that of the mature X5. These results could help explain the relatively high immunogenicity of the coreceptor binding site on gp120 and the abundance of CD4-induced (CD4i) antibodies in HIV-1-infected patients (X5 is a CD4i antibody) as well as the maturation pathway of X5. They also can help identify antigens that can bind specifically to b12 germline and intermediate antibodies that together with Envs could be used as a conceptually novel type of candidate vaccines. Such candidate vaccines based on two or more immunogens could help guiding the immune system through complex maturation pathways for elicitation of antibodies that are similar or identical to antibodies with known properties.

Keywords: HIV; antibody; escape; germline; immune responses; vaccine.

Figure 1.

Binding characteristics of the scFv X5 in various forms including mature, germline (in this and subsequent figures the term germline is for convenience the representation of germline-like antibody in this study), and hybrids between various heavy and light chains. a and c. Gel analysis of purified mature, germline and hybrid scFv X5. In a, M, molecular weight marker, 1 and 2 are mature and germline scFv X5 respectively; In c, M, molecular weight marker, 1 is the hybrid scFv between mature X5 heavy chain and germline X5 light chain, and 2 is the hybrid between matured X5 heavy chain and matured b12 heavy chain. In b and d are bindings of purified proteins shown in a and c, respectively to bal gp120-CD4. Abbreviations in this and subsequenct figures are as following: math, matured heavy chain; germl, germline light chain; matl, matured light chain; and germh, germline heavy chain.

Figure 2.

Inhibition of pesudovirus infection by mature and germline scFv X5. Nine HIV Env-pseudotyped viruses were tested with a single concentration of both X5 original and germline as described in Materials and Methods. The concentration of scFv used is 600 nM. The names of the Envs used are shown on the X-axis and the numbers on the Y-axis represent the percentages of the pseudovirus activities. For each isolate, the bars represent the percentage of activities of the viruses treated with PBS only (left columns), scFv X5 original (middle columns), and germline (right columns).

Figure 3.

Determination of the IC50s of the mature and germline scFv X5 against representative HIV isolates. Viruses pseudotyped with Envs from M and T tropic viruses from B clade as well as one from A clade were used in the neutralization assay.

Figure 4.

Lack of binding by germline scFv and IgG b12 to multiple HIV Envs. Mature and germline b12 in both scFv (upper panel) and IgG (lower panel) formats, and an irrelevant IgG were tested for their bindings to four different HIV Envs as indicated on X axis. BSA was included as a negative control. The single concentration used for each antibody was 2.5 μM.

Figure 5.

Identification of b12 intermediate binders. Point mutations were introduced back to the H2 and adjacent frame work of germline b12. The resulted mutants were expressed, purified both as scFv and scFv-Fc. a. Gel analysis of the purified scFv and scFv-Fc. M, molecular weight marker. Numbers on the left indicate molecular weight in kd. Samples 1–9 are original, germline, A52P/G53Y, G53Y, math/germl, germh/matl, G53D, A52P/T57K, and A52P. Samples 10–15 are mature b12-Fc, germline b12-Fc, A52P/G53Y-Fc, G53Y-Fc, math/germl-Fc, and germh/matl-Fc. Sample 16 is the germline IgG b12. b. The bindings by the selected scFv were analyzed against bal gp120 in an ELISA (left panel). Two concentrations including 8 (bar closer to Y-axis) and 2.7 μM of each scFv were used. BSA as an antigen was included as a specificity control (right panel). The maximum value of Y-axis, which shows OD 405 from ELISA assay was set at 0.5 to reflect the weak bindings. Bindings of both mature b12 and A52P/G53Y reached saturation at both concentrations and were indicated.

Figure 6.

Determination of the strength of binding by various formats of b12. ELISAs were performed using bal gp120 as the antigen. Various scFv (a) and scFv-Fc (b) b12 as indicated were analyzed for their bindings. c. Competition ELISA between various scFv b12 and the original scFv-Fc b12. Fixed amount of various scFv b12 at 20 μg was pre-mixed with increasing amount of original scFv-Fc b12 in 100 μl of blocking buffer and applied to ELISA plate coated with bal gp120. The amount of bound scFv was measured using anti-his-HRP. d. Specific competition between sCD4 and various forms of b12 in binding to bal gp120. Fixed amount of sCD4 at 2 ug was mixed with increasing amount of various b12-Fc fusion protein in 100 μl of blocking buffer and added to ELISA plate coated with bal gp120. The bound sCD4 was detected with anti-his-HRP.

Figure 7.

Inhibition of pseudovirus infection by various scFv-Fc b12. Nine HIV Env-pseudotyped viruses were tested with a panel of scFv-Fc b12 variants. The original-Fc b12 was used at a concentration of 0.3 μM, while all the other b12–Fc fusion variants including the germline-Fc b12 were used at a concentration of 2 μM. For each isolate, the bars represent the percentage of activities of the viruses treated with PBS only, b12-Fc, math/germl-Fc, A52P/G53Y-Fc, germh/matl-Fc, G53Y-Fc, and germline-Fc sequentially, with the PBS treated sample closet to the Y-axis.

Figure 8.

Binding of various b12-Fc proteins to the surface antigens of three human cell lines. (a). Germline-Fc, G53Y-Fc, A52P/G53Y-Fc and mature-Fc b12 proteins were used at a concentration of 1 μM in the flow cytometry assay as described. (b). Germline-Fc, math/germl-Fc, germh/matl-Fc and original-Fc were further compared in the flow cytometry assay. The concentration used remained at 1 μM. The numbers on the X-axis represent the binding intensity and the numbers on the Y-axis represent the number of cells.