Suppression of HIV-1 Infection by APOBEC3 Proteins in Primary Human CD4+ T Cells Is Associated with Inhibition of Processive Reverse Transcription as Well as Excessive Cytidine Deamination
-
Fig 1Comparison of CD4+ T-cell-derived wild-type and Vif-deficient HIV-1 for APOBEC3G packaging (A), single-cycle infectivity (B), and cDNA synthesis (C and D). CD4+ T cells isolated from three different donors were infected with wild-type or ΔVif HIV-1. The released viral particles were isolated, purified, and analyzed. (A) Wild-type and ΔVif HIV-1 virions corresponding to 10 ng p24Gag were analyzed by immunoblotting using antibodies specific for A3G and p24Gag (loading control). (B) Virion infectivity was determined using TZM-bl reporter cells challenged with wild-type or ΔVif viruses corresponding to 5 ng p24Gag, followed by the measurement of β-galactosidase activity in cell lysates. Background β-galactosidase activity (2,808 counts) from uninfected control cell lysates was subtracted. (C and D) Equivalent amounts of wild-type and ΔVif virions (5 ng p24Gag) were used for in vitro endogenous reverse transcription assays. Total DNA was harvested at the indicated times, and cDNA levels were measured using qPCR with primer-probe sets monitoring production of strong stop (C) or first-strand transfer (D) cDNA. wt, wild type. Statistical analysis was performed applying a 2-sample, unequal-variance t test: *, P < 0.05; **, P < 0.01; ***, P < 0.001.
-
Fig 2Effect of packaged APOBEC3 proteins on cDNA synthesis in infected CD4+ T cells. The six CD4+ T-cell-derived virus stocks characterized for Fig. 1 were used to infect fresh cultures of CD4+ T cells from the same three donors (corresponding to the three different graphs). Total DNA was harvested at the indicated times after infection, and the levels of strong stop (A) or first-strand transfer (B) cDNA were measured using qPCR. Levels of cDNA were normalized to the amount of total DNA extracted and are shown as proportions of the peak accumulation detected in each set of reactions, and mean values are indicated. For statistical analysis of the differences in cDNA levels between wild-type and ΔVif infections, data available from all donors were combined for each time point and a 2-sample, unequal-variance t test was carried out. For strong stop cDNA (A), P < 0.01 at 4 h, P < 0.001 at 8 h, and P < 0.01 at 24 h. For first-strand transfer cDNA (B), P < 0.001 at 4 h, P < 0.001 at 8 h, and P < 0.01 at 24 h.
-
Fig 3Inhibition of endogenous reverse transcription by ectopically expressed APOBEC3G. (Top) Immunoblotting was used to compare wild-type and ΔVif virions derived from infected primary CD4+ T cells (lanes 1 and 2, respectively), with virions generated by transfection of 293T cells with pIIIB Δvif and increasing quantities of pcDNA3.1 A3G (lanes 3 to 8, A3G/provirus transfection ratios of 0:1, 1:81, 1:27, 1:9, 1:3, and 1:1, respectively). (Bottom) Virions analyzed in lanes 3 to 8 of panel A were assessed in ERT assays. Total DNA was harvested at the indicated times, and levels of strong stop cDNA were measured using qPCR.
-
Fig 4Impact of APOBEC3G on length of HIV-1 cDNA synthesized by endogenous reverse transcription. Virions characterized for Fig. 3 and generated by transfection of 293T cells with pcDNA3.1 A3G and pIIIB Δvif (0:1, 1:27, and 1:1 ratios) were purified and subjected to ERT reactions. Nucleic acids were extracted at the indicated times, and cDNA products were tailed, amplified, cloned, and sequenced. The lengths of between 22 and 46 reverse transcripts per time point and A3G concentration are shown. cDNA length is plotted on the ordinate, with the abscissa corresponding to individual cDNAs sorted by length. Each graph corresponds to one A3G/provirus ratio and demonstrates progression of reverse transcription over four time points (30, 60, 120, and 240 min).
-
Fig 5Analysis of mutational load imposed on HIV-1 cDNA by endogenous APOBEC3 proteins in CD4+ T cells. A 500-nt nef-R fragment was amplified by limiting dilution, followed by PCR from DNA samples extracted from the 18 independent infections previously described in Fig. 2, and then sequenced. (A) Distribution of mutation types in wild-type (top) and ΔVif (bottom) HIV-1 from all sequences. The total numbers of available nucleotides are indicated. (B) Percentage of G residues mutated over time for individual wild-type (at 4 h and 24 h) and ΔVif (at 4 h, 8 h, and 24 h) infections, with the mean values indicated. (C) Proportion of sequenced viral cDNAs from wild-type-infected (at 4 h and 24 h) and ΔVif-infected (at 4 h, 8 h, and 24 h) cells carrying at least one G-to-A mutation, with the mean values indicated. (D) Percentage of sequences with the indicated numbers of G-to-A mutations at various time points in ΔVif and wild-type virus infections.
-
Fig 6Dinucleotide sequence preferences for G-to-A mutations in HIV-1 cDNA from wild-type- and ΔVif-infected CD4+ T cells. Sequencing data described in Fig. 5 were analyzed for dinucleotide sequences at mutated sites. (A) Total number and percentage of mutated G residues in all four possible dinucleotide contexts for wild-type (top) and ΔVif (bottom) infections. (B) Correlation between the percentage of mutated G residues in the 5′-GG dinucleotide context and the total number of mutated G residues per viral sequence. Each data point represents the average of the percentage of mutations in the 5′-GG context for all cDNAs harboring the same number of G mutations. The Spearman rank correlation coefficient (rS) and the corresponding P value were determined for the two variables before averaging of the values.
- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
