A New Type of Nonsuppressible Viremia Produced by HIV-Infected Macrophage

Abstract

Background: HIV-1 RNA typically declines rapidly after initiation of antiretroviral therapy (ART); often reaching undetectable levels within a few weeks and remaining undetectable by standard assays. However, some patients on ART have persistent nonsuppressible viremia (NSV) that does not respond to treatment optimization or intensification. NSV can emerge at the time of ART initiation (primary NSV) or after being ART-suppressed (secondary NSV). Here, we examine mechanisms producing primary NSV in four people on ART.

Methods: Blood samples were collected from four participants who, despite being adherent to ART, required approximately a year or more to become virologically suppressed. Viral RNA and proviral DNA genomes were sequenced to examine HIV-1 drug resistance, genome intactness and genetic diversity. The ability of HIV-1 Envs to facilitate efficient entry into cells expressing low levels of CD4 (a proxy for macrophage tropism) was assessed.

Results: Before ART, the blood contained HIV-1 RNA genomes that were adapted to replication in CD4+ T cells and rapidly decayed after ART initiation. During ART, the blood contained HIV-1 genomes that were drug sensitive, genetically diverse, macrophage-tropic, not evolving and often had defects in vpr.

Conclusions: Our results suggest that in individuals with primary NSV, ART stopped virus replication, but large pools of long-lived, HIV-infected macrophage continued to produce virus. This is mechanistically distinct from secondary NSV produced by CD4+ T cell clones. In addition, defects in vpr independently accumulation in macrophage-tropic lineages found in three participants, suggesting that vpr may impact survival of, or virus production from, HIV-infected macrophage.

Figure 1:. Participants that slowly suppressed viral RNA after initiating ART with drug sensitive virus.

Viral loads (HIV-1 RNA cp/mL) and CD4+ T cell (cells/μL; red lines) are shown for all four participants (A-D). Limits of quantification for viral load tests are marked as black, dashed lines. Timepoints when plasma was sampled and used for viral RNA sequencing are represented by circles (⚫) and PBMC samples are represented by squares (▄). Colors indicate when each sample was collected. Phylogenetic trees of partial HIV-1 env (V1V3) sequences (identical sequences collapsed) are shown and slow decay lineages are marked with stars.

Figure 2:. Diversity of intact HIV-1 env genes and phenotypic assay for macrophage tropism (M tropism).

Phylogenetic trees of intact HIV-1 env genes from all Participants are shown (A-D). Amplicons of envs were generated at limiting dilution from blood plasma RNA (circles ⚫) and proviral DNA in PBMCs (squares ▄). Plasma lineages that decayed slowly after ART initiation or did not decay are noted with a star and sequences are colored according to their sample time. Luciferase reporter viruses were generated from envs (selected envs marked with arrows in phylogenetic trees) in both lineages that slowly decayed after ART initiation (E) and lineages that decayed rapidly after ART initiation (F). M tropism was assessed based on the ability of reporter viruses to enter Affinofile cells expressing a low surface density of CD4 (similar to levels on macrophage) relative to their ability to enter Affinofile cells expressing a high density of CD4 (similar to levels on CD4+ T cells). M- and T-tropic controls are shown. The envs derived from plasma RNA are shown with solid bars and envs from proviral DNA are shown with crosshatched bars.

Figure 3:. Intactness of near full-length sequences from P1 plasma RNA.

Sequences were generated at limiting dilution as two half genomes, and analyzed for intactness of open reading frames, hypermutation, and large deletions. Stars indicate nonsynonymous changes including frameshifts. Sequences are grouped by both their cellular tropism (M-tropic, blue bar on left; T-tropic, red bar on left) and the timepoint when they were sampled. Created in BioRender. Moeser, M. (2025) https://BioRender.com/62gzait.

Figure 4:. Changes in the frequency of HIV-1 variants with defects in vpr and M-tropic HIV-1.

1. Longitudinal deep sequencing of viral RNA in plasma was performed to separately assess vpr intactness (solid black bars) and M tropism (crosshatched bars) in participants with longitudinal plasma samples. Proportions determined by Miseq Primer ID sequencing.