The activity spectrum of Vif from multiple HIV-1 subtypes against APOBEC3G, APOBEC3F, and APOBEC3H

Abstract

The APOBEC3 family comprises seven cytidine deaminases (APOBEC3A [A3A] to A3H), which are expressed to various degrees in HIV-1 susceptible cells. The HIV-1 Vif protein counteracts APOBEC3 restriction by mediating its degradation by the proteasome. We hypothesized that Vif proteins from various HIV-1 subtypes differ in their abilities to counteract different APOBEC3 proteins. Seventeen Vif alleles from seven HIV-1 subtypes were tested for their abilities to degrade and counteract A3G, A3F, and A3H haplotype II (hapII). We show that most Vif alleles neutralize A3G and A3F efficiently but display differences with respect to the inhibition of A3H hapII. The majority of non-subtype B Vif alleles tested presented some activity against A3H hapII, with two subtype F Vif variants being highly effective in counteracting A3H hapII. The residues required for activity were mapped to two residues in the amino-terminal region of Vif (positions 39F and 48H). Coimmunoprecipitations showed that these two amino acids were necessary for association of Vif with A3H hapII. These findings suggest that the A3H hapII binding site in Vif is distinct from the regions important for A3G and A3F recognition and that it requires specific amino acids at positions 39 and 48. The differential Vif activity spectra, especially against A3H hapII, suggest adaptation to APOBEC3 repertoires representative of different human ancestries. Phenotypic assessment of anti-APOBEC3 activity of Vif variants against several cytidine deaminases will help reveal the requirement for successful replication in vivo and ultimately point to interventions targeting the Vif-APOBEC3 interface.

Fig 1

A panel of 17 Vif alleles from different subtypes. (A) Phylogenetic relationships of subtype consensus sequences and selected Vif alleles from different subtypes were inferred using the neighbor-joining method. Bootstrap values of 70% or higher are shown next to the branches (1,000 replicates). Distances were computed using the maximum composite likelihood method. All positions containing gaps and missing data were eliminated from the data set (complete deletion option). There were a total of 568 positions in the final data set. Phylogenetic analyses were conducted in MEGA4. (B) Vif variants from different subtypes are expressed to comparable levels. pCRV1 Vif expression plasmids were transfected with GFP expression plasmid into HEK 293T cells. Lysates were separated by SDS-PAGE at 48 h posttransfection, transferred to PVDF membranes, and probed with antibodies against Vif, GFP, and GAPDH. (C) Relative Vif expression for each Vif variant is shown. Nonsaturated signals were acquired, background was subtracted, and values were normalized to GFP expression levels. NL4-3 Vif expression was set to 1. The average and standard deviations of three independent experiments are shown. Error bars represent standard deviations. α denotes anti.

Fig 2

Activity of Vif alleles against A3G and A3F. (A) Anti-A3G activities of the different Vif variants. Infectivity of pNL4-3ΔVif cotransfected with Vif alleles and A3G was assessed at 48 h posttransfection on TZM-bl reporter cells. Data shown are representative of three independent experiments. Error bars represent standard deviations. Unpaired t tests were computed to determine whether differences between LAI Vif infectivity and a given Vif variant reached significance (*, P = 0.05; **, P < 0.01; ***, P < 0.001 [Prism software]). (B) Degradation of A3G by the different Vif alleles. Transfected lysates were separated by SDS-PAGE at 48 h posttransfection, transferred to PVDF membranes, and probed with antibodies against FLAG, Vif, and GFP. One representative Western blot is shown. (C) Anti-A3F activities of the different Vif variants. Infectivity of pNL4-3ΔVif cotransfected with Vif alleles and A3G was assessed at 48 h posttransfection on TZM-bl reporter cells. Data shown are representative of three independent experiments. Error bars represent standard deviations. Unpaired t tests were computed to determine whether differences between LAI Vif infectivity and a given Vif variant reached significance (*, P = 0.05; **, P < 0.01; ***, P < 0.001 [Prism software]). (D) Degradation of A3F by the different Vif alleles. Transfected lysates were separated by SDS-PAGE at 48 h posttransfection, transferred to PVDF membranes, and probed with antibodies against FLAG, Vif, and GFP. Representative Western blots are shown.

Fig 3

Activity of Vif alleles against A3H hapII. (A) Anti-A3H hapII activities of the different Vif variants. Infectivity of pNL4-3ΔVif cotransfected with the Vif alleles and A3H hapII was assessed at 48 h posttransfection on TZM-bl reporter cells. Data shown are representative of three independent experiments. Error bars represent standard deviations. Unpaired t tests were computed to determine whether differences between LAI Vif infectivity and a given Vif variant reached significance (*, P = 0.05; **, P < 0.01; ***, P < 0.001 [Prism software]). No Vif/No A3H, 100% relative infectivity. (B) Degradation of A3H hapII by the different Vif alleles. Transfected lysates were separated by SDS-PAGE at 48 h posttransfection, transferred to PVDF membranes, and probed with antibodies against FLAG, Vif, and GFP. Representative Western blots of one out of three experiments are shown. No Vif/A3H, 100% A3H remaining. (C) A3H hapII degradation efficiency of different Vif variants. A3H hapII signals were quantified and the no-Vif control is set at 100%. Error bars represent standard deviations of three independent A3H hapII degradation assays. (D) Correlation between Vif-mediated rescue of viral infectivity in the presence of A3H hapII and Vif-mediated degradation. Gray symbols identify the controls (as described for A and B). (E) The pattern of A3H hapII-mediated degradation by Vif is independent of the FLAG tag. Selected active and nonactive Vif variants were transfected with FLAG-tagged and untagged A3H hapII expression plasmids. Transfected lysates were separated by SDS-PAGE at 48 h posttransfection and transferred to PVDF membranes. Proteins were detected with polyclonal rabbit serum directed against A3H or Vif. GFP was detected with a monoclonal antibody.

Fig 4

Comparison of the anti-APOBEC3 activities of different Vif alleles. A heat map representation illustrates the spectrum of infectivity obtained with each Vif variant in the presence of A3G, A3F, and A3H hapII. Normalized infectivity values are plotted. The controls for each deaminase are shown on the right side of the panel: the level of suppression achieved (%) in the absence of Vif is depicted (red), and the infectivity (%) in the absence of APOBEC3 is set to 100 (green). Relative infectivity values are the average of three independent experiments shown in Fig. 2 and 3. Max denotes maximum.

Fig 5

Identification of the residues important for activity of subtype F Vif variants against A3H hapII. (A) A schematic of the Vif N-terminal mutants generated to investigate activity against A3H hapII. Vif F1 and Vif F2 differ at positions 39, 48, and 61 to 63. (B) Infectivities of pNL4-3ΔVif cotransfected with Vif F1, Vif F2, and Vif mutants M1 to M8 and A3H hapII were assessed at 48 h posttransfection on TZM-bl reporter cells. Data shown are representative of three independent experiments. Error bars represent standard deviations of triplicate experiments. (C) Western blot depicting A3H hapII degradation in the presence of Vif F1, Vif F2, and Vif mutants M1 to M8. Transfected lysates were separated by SDS-PAGE at 48 h posttransfection, transferred to PVDF membranes, and probed with antibodies against FLAG, Vif, and GFP. Data shown are representative of three independent experiments. (D) Coimmunoprecipitation (Co-IP) of subtype F Vif variants with A3H hapII and A3G. Subtype F Vif F1, Vif F2, or empty control plasmid was cotransfected with HA-tagged A3H hapII, HA-tagged A3G, or control plasmid. APOBEC3 degradation was prevented by the addition of the proteasome inhibitor clasto-lactacystin β. Cleared lysates were incubated with anti-HA-coated beads and washed vigorously. Proteins were analyzed by Western blotting. (E) Coimmunoprecipitation of subtype F Vif variants with A3H hapII. Subtype F Vif F1, Vif F2, and Vif mutants M4 and M3 were cotransfected with HA-tagged A3H hapII. A3H hapII degradation was prevented by the addition of the proteasome inhibitor clasto-lactacystin β. Cleared lysates were incubated with anti-HA-coated beads and washed vigorously. Proteins were analyzed by Western blotting.

Fig 6

Characterization of the Vif/APOBEC3 interface that allows degradation of A3H hapII in the Vif F1 context. The anti-APOBEC3 activities of the 12 different alanine mutants in Vif F1 (alanine scanning mutagenesis of the region between residues 38 and 49) are shown. Infectivity of pNL4-3ΔVif cotransfected with the Vif mutants and A3G (A), A3F (B), and A3H hapII (C) was assessed 48 h posttransfection on TZM-bl reporter cells. Data shown are representative of three independent experiments. Error bars represent standard deviations. Western blots showing A3G, A3F, and A3H hapII degradation by the mutant F1 Vifs are directly under the respective infectivity values. (D) Transfected Vif F1 mutants are expressed to comparable levels. Representative Western blots are shown for Vif expression and the GFP transfection control. (E) Heat map summarizing the anti-A3G, -A3F, and -A3H hapII phenotypes associated with each Vif F1 alanine mutant.

Fig 7

Correlation between Vif genotype and activity against A3H hapII. The amino acid alignment of the 17 Vifs shows the region between residues 38 to 49 in the amino-terminal portion of Vif. Vif F1 serves as reference in this alignment. The Vifs are ranked in descending order of anti-A3H hapII activity (Fig. 4). Nine Vif variants (F1 to G1) rescued viral infectivity in the presence of A3H hapII to 20% or more. These infectivity values were significantly higher (P < 0.05, t test) from the no-Vif or the Vif C133S controls. The level of rescue is indicated as follows: +++, high, with infectivity >50% of the no-A3H control; ++, moderate, with 50 to 30% of the no-A3H control; and +, low, with 20 to 30% of the no-A3H control. Infectivity values of ≤20% of the no-A3H control were defined as lack of neutralization activity (—). The amino acid pairs at locations 39 and 48 are grouped as FH, ZH, and XN. Z denotes any residue except F, and X denotes any residue.