Catalog Home Page

Analysis of HIV-1 hypermutation and its relationship with APOBEC3G and vif genetic variation in vivo

Mallal, S., Pace, C., Nolan, D., Keller, J., Moore, C., Carvalho, F., Patterson, A., Sutton, K. and Gaudieri, S. (2005) Analysis of HIV-1 hypermutation and its relationship with APOBEC3G and vif genetic variation in vivo. In: 12th Conference on Retroviruses and Opportunistic Infections, 22 - 25 February 2005, Hynes Convention Center, Boston, MA.

PDF (Poster)
Download (210kB)


Background: Editing of HIV-1 DNA and RNA by APOBEC family cytidine deaminases to produce hypermutated sequences has emerged as a potent host antiviral response. This effect is modulated by the presence of vif that opposes APOBEC activity. Here we examine the clinical relevance and extent of hypermutation within proviral DNA, the sequence context of G®A hypermutation, and the contributions of genetic variation in host APOBEC3G and HIV-1 vif.

Methods: Full-length HIV-1 proviral DNA sequences from 227 antiretroviral-naive patients were examined, including 24,675 G®A substitutions. Hypermutation was classified by 3 criteria and validated by analyses of the population distribution of G®A substitutions. In individuals with hypermutation, APOBEC3G exonic and putative 5’ regulatory region sequences were compared with control data (n = 200). Analyses of vif amino acid sequences were also undertaken, and pre-treatment viral load was assessed in group comparisons.

Results: All 3 criteria for hypermutation were satisfied in sequences obtained from 8 (3.5%) individuals, with 1 criterion met by 21 (9.3%). Viral load (log10 copies/mL) was significantly lower in the hypermutated group using the broader definition (4.19 vs 4.72, p = 0.01) but not the restrictive definition (4.56 vs 4.68, p = 0.38). Compared with background rates of 6.9 ± 4.0% G®A substitution, hypermutated sequences demonstrated 20.3 ± 6.2% G®A substitutions by a restrictive definition (broad = 15.5 ± 1.2%) (both p < 0.001). Hypermutated G®A substitutions demonstrated a preference for G at the +1 position (45% vs 23%) and enrichment of T at position –1 (34% vs 23%) (p < 0.001) consistent with an APOBEC3G plus strand DNA target motif TGG (corresponding to CCA on the minus strand DNA). Expected strong selection against CpG dinucleotides was also universally noted. APOBEC3G sequencing of the 8 subjects with definite hypermutation revealed no novel variants. Analysis of vif sequence however, revealed that 100% of the hypermutated group had defective vif, defined as absence of appropriate start and stop codons and/or presence of in-frame stop codons.

Conclusions: This in vivo study demonstrates that hypermutation is not infrequent at a population level and is associated with lower pretreatment viral load, in keeping with a host antiviral effect potentially mediated by APOBEC3G.

Publication Type: Conference Item
Murdoch Affiliation: Centre for Clinical Immunology and Biomedical Statistics
Conference Website:
Notes: Poster presentation
Item Control Page Item Control Page


Downloads per month over past year