High-mannose-specific deglycosylation of HIV-1 gp120 induced by resistance to cyanovirin-N and the impact on antibody neutralization

Abstract

HIV-1 uses glycans on gp120 to occlude its highly immunogenic epitopes. To better elucidate escape mechanisms of HIV-1 from carbohydrate-binding agents (CBA) and to understand the impact of CBA-escape on viral immune evasion, we generated and examined the biological properties of HIV-1 resistant to cyanovirin-N (CV-N) or cross-resistant to additional CBAs. Genotypic and phenotypic characterization of resistant env clones indicated that 3-5 high-mannose residues from 289 to 448 in the C2-C4 region of gp120 were mutated and correlated with the resistance levels. The specificity and minimal requirements of deglycosylation for CV-N resistance were further assessed by mutagenesis study. The sensitivity of resistant variants to a range of CBAs, immunoglobulins, sera and monoclonal antibodies (MAb) were investigated. For the first time, our data have collectively defined the high-mannose residues on gp120 affecting CV-N activity, and demonstrated that CBA-escape HIV-1 has increased sensitivity to immunoglobulins and sera from HIV patients, and particularly to V3 loop-directed MAbs. Our study provides a proof-of-concept that targeting HIV-1 glycan shields may represent a novel antiviral strategy.

Fig. 1. CV-N resistant viral strains and the cross-resistance to plant and red alga lectins

Anti-HIV-1 activity was assessed in TZM-bl cells by infecting with HIV-1_IIIB, or CV-N resistant isolates CV or GCV in the presence of serially diluted (A) CV-N, (B) GNA and (C) GRFT. The infectivity of each virus in medium alone culture was arbitrarily set to 100%. Data are representative of at least 3 independent experiments, with each determination performed in triplicate (mean ± SD).

Fig. 2. Genotypic and phenotypic characterization of CV-N resistant env molecular clones

(A) Alignment of the glycosylation changes in resistant HIV-1 viruses and their env clones. Nonsynonymous sequence polymorphisms in env PCR products, containing both the wild and the mutated amino acids, are indicated by assigning “?” to the position. (B) Fusogenic activity of IIIB Env in the presence of serially diluted CV-N or GNA was determined using the Env mediated cell-cell fusion assay. (C) The fusogenic activity of Env encoded by each env molecular clone in the presence or absence of 100 nM CV-N or 400 nM GNA. The fusogenic activity of each Env in medium alone culture was arbitrarily set to 100%. (D) The infectivity of IIIB Env, CV1 or GCV4 pseudotyped virus in the presence or absence of 100 nM CV-N or 400 nM GNA were assessed in TZM-bl cells. The infectivity of each virus in medium alone culture was arbitrarily set to 100%. Data are representative of 3 independent experiments, with each determination performed in triplicate (mean ± SD).

Fig. 3. Defining high-mannose residues on gp120 that affect CV-N antiviral activity

(A) The schematic of N-linked glycosylation sites on IIIB and HXB2 gp120 according to Leonard et al (Leonard et al., 1990) and Gallaher et al (Gallaher et al., 1995). The constant regions C1, C2, C3, C4 and C5, and the variable regions V1, V2, V3, V4 and V5 are shown. Arrows indicate sites that were changed in CV-N resistant viruses. The fusogenic activity of (B) HXB2 Env and its mutants, (C) IIIB Env, GCV4 and GCV4 mutants, and (D) IIIB Env and its mutants, in the presence or absence of 100 nM CV-N, 400 nM GNA or 25 μg/ml MAb 2G12. The fusogenic activity of each Env in medium alone culture was arbitrarily set to 100%. Data are representative of 3 independent experiments, with each determination performed in triplicate (mean ± SD).

Fig. 4. Sensitivity of CV-N resistant viruses to immunoglobulins and sera from HIV-1 positive individuals

Anti-HIV-1 activity was assessed in TZM-bl cells by infecting with HIV-1_IIIB, CV or GCV in the presence of serially diluted (A) HIV negative serum, (B) HIV-IgG, (C) HIV serum 1 and (D) HIV serum 2. The infectivity of each virus in the absence of indicated serum was arbitrarily set to 100%. Data are representative of 2-5 independent experiments, with each determination performed in triplicate (mean ± SD). * P<0.01, compared with HIV-1_IIIB.

Fig. 5. CV-N resistant viruses demonstrate increased sensitivity to V3 loop-directed MAbs

Anti-HIV-1 activity was assessed in TZM-bl cells by infecting with HIV-1_IIIB, CV or GCV in the presence of serially diluted (A) MAb 2G12, (B) MAb 447-52D and (C) MAb 1101. The infectivity of each virus in medium alone culture was arbitrarily set to 100%. Data are representative of 3 independent experiments, with each determination performed in triplicate (mean ± SD). * P<0.01, compared with HIV-1_IIIB.