Evaluating the Intactness of Persistent Viral Genomes in Simian Immunodeficiency Virus-Infected Rhesus Macaques after Initiating Antiretroviral Therapy within One Year of Infection

Abstract

The major obstacle to more-definitive treatment for HIV infection is the early establishment of virus that persists despite long-term combination antiretroviral therapy (cART) and can cause recrudescent viremia if cART is interrupted. Previous studies of HIV DNA that persists despite cART indicated that only a small fraction of persistent viral sequences was intact. Experimental simian immunodeficiency virus (SIV) infections of nonhuman primates (NHPs) are essential models for testing interventions designed to reduce the viral reservoir. We studied the viral genomic integrity of virus that persists during cART under conditions typical of many NHP reservoir studies, specifically with cART started within 1 year postinfection and continued for at least 9 months. The fraction of persistent DNA in SIV-infected NHPs starting cART during acute or chronic infection was assessed with a multiamplicon, real-time PCR assay designed to analyze locations that are regularly spaced across the viral genome to maximize coverage (collectively referred to as "tile assay") combined with near-full-length (nFL) single-genome sequencing. The tile assay is used to rapidly screen for major deletions, with nFL sequence analysis used to identify additional potentially inactivating mutations. Peripheral blood mononuclear cells (PBMC) from animals started on cART within 1 month of infection, sampled at least 9 months after cART initiation, contained at least 80% intact genomes, whereas those from animals started on cART 1 year postinfection and treated for 1 year contained intact genomes only 47% of the time. The most common defect identified was large deletions, with the remaining defects caused by APOBEC-mediated mutations, frameshift mutations, and inactivating point mutations. Overall, this approach can be used to assess the intactness of persistent viral DNA in NHPs.IMPORTANCE Molecularly defining the viral reservoir that persists despite antiretroviral therapy and that can lead to rebound viremia if antiviral therapy is removed is critical for testing interventions aimed at reducing this reservoir. In HIV infection in humans with delayed treatment initiation and extended treatment duration, persistent viral DNA has been shown to be dominated by nonfunctional genomes. Using multiple real-time PCR assays across the genome combined with near-full-genome sequencing, we defined SIV genetic integrity after 9 to 18 months of combination antiretroviral therapy in rhesus macaques starting therapy within 1 year of infection. In the animals starting therapy within a month of infection, the vast majority of persistent DNA was intact and presumptively functional. Starting therapy within 1 year increased the nonintact fraction of persistent viral DNA. The approach described here allows rapid screening of viral intactness and is a valuable tool for assessing the efficacy of novel reservoir-reducing interventions.

Keywords: SIV; genome intactness; genomic integrity; persistent viral reservoir; single-genome sequencing.

FIG 1

Viral load data. Plasma viral load data are shown for rhesus macaques starting cART early (day 10 [d10] [A] and d27 [B]) and late (C). Each group is color coded by cART start date, and arrows indicate sample times and numbers of animals sampled for tile profiling and sequencing (in parentheses).

FIG 2

Low percentage of deleted genomes and G-to-A-hypermutated genomes in SIV-infected rhesus macaques with early (i.e., day 10) initiation of cART. A total of 185 individual genomes from 12 animals with day 10 initiation of cART were analyzed with single-genome amplification and Sanger sequencing. Maps of individual SIV DNA viral genomes are shown. nt, nucleotide.

FIG 3

Real-time qPCR tile assay design and evaluation. Wild-type SIVmac239 (A) and 6 deleted mutants (B to G) were generated by restriction enzyme digestion and religation. A total of 50,000 copies of each plasmid were used in each reaction to assess the 23-tile assay. C_T values are plotted to allow visualization of the deleted region (i.e., no C_T value) in each plasmid, with the dashed line showing the deletion site. There was complete agreement between deletion boundaries as determined through tile assay profiling in plasmids B to F and sequence analysis of plasmid clones. For panel G, the short deletion was located in an 86-nucleotide region between tiles 16 and 17 and was not detected by tile assay profiling. Tiles 1 to 4 represent locations in gag, 5 to 11 locations in pol, 12 to 14 locations in accessory genes, 15 to 20 locations in env, and 21 to 23 locations in nef and in the 3′ LTR; gray dashed lines highlight these gene regions.

FIG 4

Correlation between tile assay profiling and nFL sequencing of individual SIV genomes. (Left) The round 2 near-full-length PCR products from 101 individual genomes from animals with day 27 initiation of cART were analyzed with single-genome amplification/Sanger sequencing (left) and tile assay profiling (representative profiles are shown to the right). (Right) Panels A, B, and D show evidence of higher-than-expected C_T profile values (corresponding to confirmed G-to-A-hypermutated genomes). Panel C and E to K highlight deleted genomes.

FIG 5

SIV genome integrity in animals treated during chronic infection. Data represent results of mutation analysis of all PBMC-derived and LN-derived SIV DNA genomes from animals treated with cART for 1 year. For animals T028 (A) and T159 (B), only PBMCs were analyzed, while for T154 (C) and T158 (D), both PBMCs and LN cells were assessed. Sequences that contained two or more types of mutation were categorized in the following order: large deletions, APOBEC mutations, frameshift insertions/deletions, and alterations in stop codons.

FIG 6

Expanded clones as predicted by tile assay profiling. Genomes derived from potentially expanded clones were predicted by tile assay deletion patterns (i.e., genomes in the same clone demonstrated identical tile assay profiling patterns). Two potentially expanded clones were identified in T158 (A and B) and two more in T028 (C and D). The triplex qPCR assay was used with subsequent sequencing confirming identical break points and additional polymorphisms in all sequences from each putative clonal group.