Proviruses in CD4+ T cells reactive to autologous antigens contribute to nonsuppressible HIV-1 viremia

Abstract

Antiretroviral therapy (ART) halts human immunodeficiency virus-1 (HIV-1) replication, reducing plasma virus concentrations to below the limit of detection, but it is not curative because of a reservoir of latently infected CD4⁺ T cells. In some people living with HIV-1 (PLWH), plasma HIV-1 RNA becomes persistently detectable despite optimal ART. This nonsuppressible viremia (NSV) is characterized by identical, nonevolving HIV-1 RNA variants expressed from infected CD4⁺ T cell clones. The mechanisms driving persistent virus production from a specific population of infected cells are poorly understood. We hypothesized that proviruses in cells responding to chronic immunologic stimuli, including self-associated antigens, may drive viral gene expression and NSV. Here, we demonstrate that stimulation of CD4⁺ T cells with autologous cell lysates induced virus production in a major histocompatibility complex class II-dependent manner. In seven of eight participants with NSV, we recovered viral RNA released ex vivo in response to autologous cell lysates that matched plasma virus. This process involved both defective and replication-competent proviruses residing in conventional CD4⁺ T cells and was also observed in PLWH with undetectable viremia. These findings suggest that recognition of self-associated antigens is a potentially important cause of HIV-1 reservoir expression, which can contribute to persistent systemic inflammation and rebound upon ART interruption.

Figure 1.. NSV is driven by oligoclonal HIV-1 variants and does not correlate with the frequency of intact proviruses.

(A) Clinical characteristics and reservoir size measurements; “na” indicates not available. (B) Analysis of HIV-1 RNA plasma sequences; maximum likelihood trees are rooted on the HXB2 reference; variants matching sequences recovered from the quantitative viral outgrowth assay (QVOA) are indicated by black arrowheads; variants with known defects in the 5’-Leader are indicated with blue arrowheads. (C) Comparison of intact proviral DNA frequencies by IPDA from 400 people on ART with undetectable viremia (54)(left) and the participants in this study (right); circles in the right panel match colors as in (A) and (B); black error bars indicate the interquartile range of IPDA values; horizontal bars represent the range of plasma HIV-1 RNA during the period of persistent viremia, and circles indicate median values; circle size indicates the number of variants found in plasma; the red dashed line represents the limit of the detection of clinical assays of 20 copies/ml, and the gray area indicates the interquartile range of the frequency of intact proviruses.

Figure 2.. CD4⁺ T cell-DC co-culture can be used to investigate autologous lysate-induced virus production ex vivo.

(A) Experimental design and FACS gating strategy to isolate cells and generate autologous lysates. PBMCs from the 8 participants were sorted based on viability, CD19, CD20, and CD3, followed by lysis through freeze-thaw cycles and protein quantification. DCs were generated as previously described (56), and coculture was set up with CD4⁺ T cells from each participant. (B) Workflow to isolate HIV-1 RNA from virions from culture supernatant and cDNA synthesis for downstream assays. (C) Representative plots of HIV-1 RNA measurements from the ADK.d22 provirus for participant P1. CD4⁺ T cells were cultured with and without DCs. (D and E) Representative plots of HIV-1 PolyA RNA values for participants P2 (D) and P7 (E). Conditions with ɑMHC-II antibody or isotype control are indicated below the x-axis. Dots represent dPCR technical triplicate measurements, and error bars indicate the mean and standard error.

Figure 3.. Autologous lysates induce virion production in an MHC-II-dependent manner.

(A) Area under the curve of HIV-1 PolyA RNA across 7 days of culture for all participants with NSV; CD4⁺ T cell culture conditions are indicated at the top of the graph; “n/a” represents conditions that were not tested due to insufficient amount of cell lysate. Technical triplicates from dPCR measurements were used to calculate AUC values for each condition. (B) HIV-1 PolyA RNA AUCg values across all experiments grouped by type of stimulation; symbol colors indicate untreated CD4⁺ T cells (pink), coculture of CD4⁺ T cells with pulsed or unpulsed DCs (green), and CD4⁺ T cells treated with ɑCD3/CD28 (gray). (C) AUCg values obtained from CD4⁺ T cells with and without DCs were used to calculate the percentage of virus production compared to stimulation with ɑCD3/CD28. (D) AUCg values of virus production upon stimulation with autologous lysates from all participants using isotype control versus ɑMHC-II blocking antibody. Statistical significance was calculated by a two-tailed paired t-test. In (B), (C), and (D), errors bars indicate median and interquartile range.

Figure 4.. Autologous lysates induce HIV-1 variants matching virus contributing to NSV.

(A) Maximum likelihood tree analysis of each participant with NSV. Single genome sequences of plasma viruses, viral outgrowth isolates, and HIV-1 variants recovered from the autologous assay were analyzed based on the U5-gag (623–1806), P6-RT (1740–3410), or env (7050–7980) regions. Various shapes indicate viral variants obtained from different time points of the autologous assay. Trees were rooted to HXB2. (B) Pie charts of the percentage of HIV-1 variants in plasma that were detected by the CD4⁺ T cell-DC coculture. (C) Percentage of viral variants identical to predominant plasma clones (PPCs) for each participant, based on the type of CD4⁺ T cell stimulation. Black bars indicate the median for each group.

Figure 5.. Proviruses causing NSV are not compartmentalized in regulatory T cells.

(A) Sorting strategy for T_conv (CD127^hiCD25^low) and T_reg (CD127^lowCD25^hi). (B) Percentage of T_reg cells among CD4⁺ T cells in participants P1 to P4; error bars indicate mean and standard deviation. (C) Total LTR copies were quantified using a probe targeting the R-U5 junction. Proviruses of interest for P1 to P4 were quantified using integration site-specific probes. (D) Frequencies of specific proviruses, expressed as copies/10⁶ CD4⁺ T_conv or T_reg cells. Gray circles indicate values below the limit of detection, which is indicated in red text. Error bars indicate the standard error of the mean, resulting from 2 to 10 technical dPCR replicates.

Figure 6.. HIV-infected cells reactive to autologous lysates from replication-competent provirus.

(A) AUCg HIV-1 RNA copies from 7 days of culture for five participants with suppressed viremia (SV). Technical triplicate values from dPCR measurements were used to calculate AUC value for each condition. (B) Maximum likelihood trees for SV participants P9, P10, and P13, rooted to HXB2: U5-gag (623–1806) or env (7050–7980) regions were used for the sequence analysis. Symbol shapes indicate viral variants obtained from different culture time points. For P10, the HIV-1 variant matching a Gag-responding clone is shown with the blue arrow (see also fig. S4). No sequences were recovered for αCD3/CD28 stimulation in P13. (C) Number of HIV-1 variants obtained from each group of autologous lysates (DCs only, B, T, nBnT lysates) or CD4⁺ T cells with ɑCD3/CD28. Black bars represent the mean value from each group, and the color of the dots represents the study participants (P1 to P8, NSV; P9 to P13, SV). Participants without a response against autologous lysate or match to viral variants were excluded from this analysis.

Figure 7.. Virus production in response to autologous lysates and autoantibody reactome are similar between participants with and without NSV.

(A) AUCg of HIV-1 detected over time from CD4⁺ T cells alone, cultured with autologous DCs or treated with ɑCD3/CD28 between NSV and SV participants. Each dot represents AUCg value from an individual condition. We assigned a value of 1 to conditions without HIV-1 detection. The horizontal lines represent the geometric mean. (B) Breakdown of results from coculture experiments with CD4⁺ T cells and autologous DCs pulsed with different cell lysates. The horizontal lines represent the geometric mean. (C) Summary bar graph indicating the number and percentage of study participants with virus production; differences between the two groups were analyzed by Fisher’s exact test. (D and E) PhIP-seq analysis of antibodies binding to the virome or the human peptide libraries showing total numbers of reactive peptides (D) and number of proteins against which two or more non overlapping peptides were targeted by antibodies (E); each symbol indicates one study participant, and error bars represent mean and standard deviation. Differences between the two groups were tested by Mann-Whitney t-test. (F) PhIP-seq sub-analysis of proteins expressed by cells used for used for autologous, including DCs, monocytes (Mo), T cells, B cells, and NK cells lysates. Counts of total antibody responses were normalized by peptide library size for each cell type. Error bars indicate mean and standard deviation.