Impact of V2 mutations on escape from a potent neutralizing anti-V3 monoclonal antibody during in vitro selection of a primary human immunodeficiency virus type 1 isolate

Abstract

KD-247, a humanized monoclonal antibody to an epitope of gp120-V3 tip, has potent cross-neutralizing activity against subtype B primary human immunodeficiency virus type 1 (HIV-1) isolates. To assess how KD-247 escape mutants can be generated, we induced escape variants by exposing bulked primary R5 virus, MOKW, to increasing concentrations of KD-247 in vitro. In the presence of relatively low concentrations of KD-247, viruses with two amino acid mutations (R166K/D167N) in V2 expanded, and under high KD-247 pressure, a V3 tip substitution (P313L) emerged in addition to the V2 mutations. However, a virus with a V2 175P mutation dominated during passaging in the absence of KD-247. Using domain swapping analysis, we demonstrated that the V2 mutations and the P313L mutation in V3 contribute to partial and complete resistance phenotypes against KD-247, respectively. To identify the V2 mutation responsible for the resistance to KD-247, we constructed pseudoviruses with single or double amino acid mutations in V2 and measured their sensitivity to neutralization. Interestingly, the neutralization phenotypes were switched, so that amino acid residue 175 (Pro or Leu) located in the center of V2 was exchanged, indicating that the amino acid at position 175 has a crucial role, dramatically changing the Env oligomeric state on the membrane surface and affecting the neutralization phenotype against not only anti-V3 antibody but also recombinant soluble CD4. These data suggested that HIV-1 can escape from anti-V3 antibody attack by changing the conformation of the functional envelope oligomer by acquiring mutations in the V2 region in environments with relatively low antibody concentrations.

FIG. 1.

Selection of neutralization-resistant virus variants against KD-247. (A) The selection was carried out in PM1/CCR5 cells, as described in Materials and Methods. (B) Sensitivity of MOKW5p(200) and MOKW9p(2000) virus to KD-247 as determined by MTT assay. PM1/CCR5 cells (2 × 10³) were exposed to 100 TCID₅₀ of MOKW9p(−), MOKW5p(200), or MOKW9p(2000) virus and were cultured in the presence of various concentrations of KD-247. The IC₅₀ values were determined by MTT assay on day 7 of culture. The assay was conducted in duplicate. The values shown are representative of three separate experiments.

FIG. 2.

Amino acid sequences of gp120 from the supernatants of MOKW-infected PM1/CCR5 cells passaged in the presence or absence of KD-247. Viral RNA from the cell culture supernatants at several concentrations of KD-247 was reverse transcribed. After the obtained cDNAs were subjected to PCR amplification and cloning, the env regions in the viruses passaged in the presence or absence of KD-247 were sequenced. The V2, V3, and C3 regions are indicated. The top amino acid sequence represents one of the major sequences from supernatants of MOKW-infected PBMCs. The locations and numbers of specific amino acids, based on the HXB2 sequence, are shown above the consensus line. The number of clones with the listed sequence among the total number of clones tested is given after each designation. For each set of clones, the deduced amino acid sequence of the gp120 was aligned by the single amino acid code. Dots denote sequence identity.

FIG. 3.

Schematic representation of recombinant MOKW env genes used for analysis of the genetic basis for resistance to KD-247. MOKW-RDP, MOKW-KNL/C3m, and MOKW-KNL/V3m env genes were amplified from passaged MOKW virus-infected PM1/CCR5 cells in the absence or presence of KD-247. MOKW-KNL, MOKW-RDP/V3m, and MOKW-RDP/C3m env genes were constructed by replacing each region of MOKW-RDP with the corresponding MOKW-KNL/C3m or MOKW-KNL/V3m sequence. MOKW-KNP, MOKW-RDL, MOKW-KDL, and MOKW-RNL viruses were constructed by site-directed mutagenesis. Construction of the clones and mutagenesis procedures are described in Materials and Methods. The locations and numbers of specific amino acids, based on the HXB2 sequence, are shown above the MOKW-RDP sequence.

FIG. 4.

Neutralization sensitivities of pseudoviruses with env genes from passaged MOKW viruses to MAbs, rsCD4, and CCR5 inhibitors. Pseudoviruses with the envelope sequences listed on the figure were prepared as described in Materials and Methods. KD-247, 447-52D, rsCD4, and 17b were preincubated with 100 TCID₅₀ of each MOKW pseudotype virus for 15 min, followed by the addition of the mixtures to the target cells (GHOST-hi5). The target cells were treated with TAK-779 and 2D7 for 15 min, followed by inoculation of the pseudotype clones. The inhibitory effects were determined by measuring the luciferase activities on day 2 of culture. Conc, concentration.

FIG. 5.

Comparison of antibody binding to cell surface-expressed MOKW Env proteins. (A) 293T cells transfected with MOKW Env expression vectors were harvested at 24 h posttransfection and stained with KD-247. Flow cytometry data for binding of the KD-247 (black lines) to cell surface MOKW Env proteins are shown among GFP-gated 293T cells along with the control antibody (normal human IgG; dotted lines). The number at the top right of each graph is the MFI. (B) Each bar indicates the relative binding of KD-247, 447-52D, and 17b to MOKW Env-expressing cell surfaces. Data were normalized to each antibody's MFI for MOKW-RDP virus. FL3-H, relative fluorescence.

FIG. 6.

Neutralization sensitivities of pseudoviruses with env genes from MOKW9C(−) virus with selected V2 mutations to MAbs, rsCD4, and CCR5 inhibitors. Pseudoviruses that have envelope sequences with the selected V2 mutations listed on Fig. 4 were prepared as described in Materials and Methods. KD-247, 447-52D, rsCD4, and IgGb12 were preincubated with 100 TCID₅₀ of each MOKW pseudotype virus for 15 min, followed by addition of the mixtures to the target cells (GHOST-hi5). Target cells were treated with TAK-779 and RPA-T4 for 15 min, followed by inoculation of the pseudotype clones. Inhibitory effects were determined by measuring the luciferase activities on day 2 of culture. Conc, concentration.

FIG. 7.

Binding affinity of anti-V3 MAbs to monomeric gp120. Viral lysates for each MOKW pseudovirus were used. gp120 was captured onto microtiter wells using a sheep polyclonal antibody specific for the C terminus of gp120. Serial dilutions of KD-247 or 447-52D were tested for binding by ELISA. Because of differences in the amount of bound gp120, optical density at 405 nm (OD₄₀₅) values were normalized to saturating levels of antibody (5 μg/ml) for comparison. Conc, concentration.

FIG. 8.

Comparison of antibody binding to cell surface-expressed MOKW Env proteins with V2 mutations. (A) 293T cells transfected with MOKW Env-expression vectors were harvested at 24 h posttransfection and stained with KD-247. Flow cytometry data for binding of the KD-247 (black lines) to cell surface MOKW Env proteins are shown for GFP-gated 293T cells along with data for the control antibody (normal human IgG; dotted lines). The number at the top right of each graph is the MFI. (B) Each bar indicates relative binding of KD-247, 447-52D, and IgGb12 to MOKW Env-expressing cell surfaces. Data were normalized to each antibody's MFI for MOKW-RDP virus. FL3-H, relative fluorescence.

FIG. 9.

Replication kinetics of infectious molecular clones NL-MOKW-RDL and NL-MOKW-KNL. PM1/CCR5 cells were exposed to NL-MOKW-RDL (open square) or NL-MOKW-KNL (filled square) and cultured for 10 days. Virus replication was monitored by measuring the amounts of p24 Gag protein produced in the culture supernatants. The data are representative of the results from two independent experiments.