Target cell APOBEC3C can induce limited G-to-A mutation in HIV-1

Abstract

The evolutionary success of primate lentiviruses reflects their high capacity to mutate and adapt to new host species, immune responses within individual hosts, and, in recent years, antiviral drugs. APOBEC3G (A3G) and APOBEC3F (A3F) are host cell DNA-editing enzymes that induce extensive HIV-1 mutation that severely attenuates viral replication. The HIV-1 virion infectivity factor (Vif), expressed in vivo, counteracts the antiviral activity of A3G and A3F by inducing their degradation. Other APOBECs may contribute more to viral diversity by inducing less extensive mutations allowing viral replication to persist. Here we show that in APOBEC3C (A3C)-expressing cells infected with the patient-derived HIV-1 molecular clones 210WW, 210WM, 210MW, and 210MM, and the lab-adapted molecular clone LAI, viral G-to-A mutations were detected in the presence of Vif expression. Mutations occurred primarily in the GA context and were relatively infrequent, thereby allowing for spreading infection. The mutations were absent in cells lacking A3C but were induced after transient expression of A3C in the infected target cell. Inhibiting endogenous A3C by RNA interference in Magi cells prevented the viral mutations. Thus, A3C is necessary and sufficient for G-to-A mutations in some HIV-1 strains. A3C-induced mutations occur at levels that allow replication to persist and may therefore contribute to viral diversity. Developing drugs that inhibit A3C may be a novel strategy for delaying viral escape from immune or antiretroviral inhibition.

Figure 1. Expression of APOBEC mRNAs in Different Cells

Poly A+ RNA was isolated from individual cell types and APOBEC reading frames (A3B, A3C, A3F, and A3G) were amplified with primers specific for a single ORF. The specificity of the primers was verified by sequencing the PCR products. RT-PCR reactions for A3B, A3C, A3G, and A3F ORFs were analyzed by agarose gel electrophoresis. PCR templates were cDNAs for the indicated cell types or plasmids specific for A3B, A3C, A3G, or A3F (positive controls). Water was the nontemplate control (NTC), β-actin was the internal control, and reactions lacking reverse transcriptase (RT) were also used as control during cDNA preparation (unpublished data).

Figure 2. Inhibition of A3C Expression and G-to-A Mutation by siRNA 1

(A) Magi cells were transfected with siRNA 1 (50 nM) against A3C or with scrambled RNA. A FITC-conjugated oligo was cotransfected and FITC-positive cells were sorted 48 h following transfection. PolyA+ RNA was isolated, and RT-PCR was performed with primers specific for A3B (lanes 1 and 2), A3C (lanes 3 and 4), A3F (lanes 5 and 6), and A3G (lanes 7 and 8). β-Actin (lanes 9 and 10) was used as an internal control. A sample lacking template DNA (lane 11) was used as a negative control. Lane 12 is a molecular weight standard. (B) The transfected cells were infected with the VSV-G-pseudotyped 210WW and collected 24 h later. Viral DNA was amplified using the sensitive mutation assay and population sequencing was performed to analyze G-to-A mutation. The G-to-A mutation typically appeared as a mixture of G and A peaks at a given position, with G, the wild-type sequence, as the predominant peak. A change from G-to-A was considered a true mutation only if A represented at least 20% of the peak.

Figure 3. Effect of A3C Expression in SupT1

(A) The A3C gene cloned into the pTT-IRES-GFP lentiviral vector with an HA tag was expressed in 293T cells, and the protein detected with a polyclonal HA antibody. (B) This construct was used to transfect SupT1 cells. After 48 h, transfection efficiency of the construct was monitored by FACS analysis of GFP expression. (C) Transfected SupT1 cells were infected with the VSV-G-pseudotyped 210WW and NL4–3 HIV-1 and collected 24 h later. Viral DNA was amplified using the sensitive mutation assay, and population sequencing performed to analyze G-to-A mutation. The total number of G-to-A mutations in the protease region (positions 2255–2485) was counted separately for the GA and GG contexts. No mutations were detected after infection of SupT1 cells transfected with an empty vector. ND, non-detectable mutations.

Figure 4. G-to-A Mutation of Wild-Type Virus in PBMC and Different Cell Lines

(A) Growth kinetics of 210WW and NL4–3 in PBMC, CEMSS, and SupT1 cells. Cells were infected with p24-normalized viral particles produced from transfected 293T cells. Virus production was monitored by measuring HIV-1 p24 concentration in the culture supernatant. The data presented are one from four independent experiments, in which comparable results were obtained. (B) G-to-A mutation in HIV-1 strains 210WW and NL4–3. DNA was isolated from cell cultures 5 d after infection; the viral protease gene was amplified using the sensitive mutation assay for detecting G-to-A mutated templates. PCR products were then sequenced as a population. The total number of G-to-A mutations in the protease region (positions 2255–2485) was counted separately for the GA and GG contexts. ND, non-detected mutations. (C) Analysis of the GA and GG contexts of the protease gene in the different group of viruses after 7 d of H9 and CEMSS infections. The results presented are the mean and the standard deviations (error bars) of at least three independent infection experiments.