Elite suppressors have low frequencies of intact HIV-1 proviral DNA

Summary

Elite controllers or suppressors (ES) control viral replication without antiretroviral therapy (ART). We used the intact proviral DNA assay to approximate the size of the inducible latent reservoir in ES and found that while the median frequency of both total and intact proviral DNA was markedly lower than the frequencies seen in chronic progressors (CP) on ART, there was no significant difference in the ratio of intact to total proviral DNA between ES and CP.

A cure for HIV can be achieved through complete eradication of the virus or by a functional cure where viral replication is controlled in the absence of antiretroviral therapy (ART).¹ Elite suppressors (ES) who maintain viral loads below the limit of detection of clinical assays are a model of a functional cure. While some studies suggest that infection with attenuated virus can lead to elite control and/or long term-non progression², we and others have shown that a subset of ES are infected with fully replication-competent virus^3–8. However, these subjects clearly have smaller inducible reservoirs compared to chronic progressors (CP) on (ART) as measured by the quantitative viral outgrowth assay (QVOA). Furthermore, the QVOA has been shown to underestimate the size of the reservoir in chronic progressors^9,10, but it is not known whether the same finding holds true in ES. A study based on more than 400 near full length sequences of provirus from 28 CP on ART demonstrated that only 2.4% of proviruses in CP did not contain large deletions or hypermutations¹⁰. The intact proviral DNA assay (IPDA) is a recently developed digital droplet PCR assay that uses primers developed from near full length proviral sequences to distinguish intact proviral DNA from most deleted and/or hypermutated proviral DNA¹¹. In this study we used the IPDA to approximate the frequency of cells with total and intact proviral DNA in a cohort of 9 ES and 10 CP. The CP had CD4 counts of greater than 500 cells/ul (median of 758) and had undetectable viral loads on ART for at least 1 year (median of 5.5 years). The frequency of total proviral DNA was significantly higher in the CP (median of 602.52 copies per million CD4+ T cells) than the ES (median of 30.4 copies per million CD4+ T cells, p= 0.0005, Figure 1A). As shown in Figure 1B, the frequency of intact proviral DNA was also significantly higher in CP (median of 36.95 copies per million CD4+ T cells) compared to ES (median of 1.61 copies per million CD4+ T cells, p= 0.0159). We next asked whether the proportion of total proviral DNA that was intact was different in the 2 groups of subjects (Figure 1C) and found that, there was no significant difference in the ratio of intact to total proviral DNA in CP versus ES.

Figure 1:

Total and intact provirus quantification. Frequency of total (A) and intact (B) proviruses were quantified, in 10 CP (red circles) and 9 ES (blue squares) using the intact proviral DNA assay (IPDA). The open symbol denotes a value that was below the limit of detection of the assay and represent the highest possible frequency of intact proviral DNA. The ratio of intact to total DNA is shown for CP and ES in C. Bars represent the median values with 95% confidence intervals. The frequency of total (D) and intact (E) proviruses were quantified in the CP10/ES9 transmission pair. The ratio of intact to total DNA is shown in 1F. The non-parametric, Kruskal-Wallis test was performed along with a Dunn’s multiple comparisons test to compare ES and CP values.

We previously described transmission of virus from CP10 to ES9⁵. The transmission occurred 20 years ago and ES9 has maintained elite control since then while CP10 has been on ART. The frequency of total proviral (Figure 1D) and intact proviral DNA (Figure 1E) was much higher in CP10 than in ES9, but the ratio of intact to total proviral DNA was similar in the 2 subjects (Figure 1F).

We compared the levels of intact proviral DNA to the frequency of cells with replication-competent virus as measured by the QVOA. ES24, who has the highest level of intact proviral DNA in our cohort (191 copies per million CD4+ T cells), also has a high frequency of cells positive for replication competent virus (4.57 infectious units per million IUPM)¹². In contrast, 5 ES who had less than 2 copies of intact proviral DNA per million had IUPM values of less than 0.06 (Supplementary Table 1), implying that low frequencies of intact proviral DNA were associated with low frequency of cells from which replication-competent virus can be cultured.

Prior studies have shown that ES have lower frequencies of total and integrated DNA^13–20, lower levels of baseline and inducible cell associated RNA^{21, 22}, and lower frequencies of cells positive for replication-competent virus^{3,6,8,16, 21,23}. These different assays measure different aspects of the viral reservoir, and while there have been studies comparing these assays in CP^11,24, this is the first time measurements of intact proviral DNA have been compared to measurements of the inducible replication-competent viral reservoir in ES. This is important as ES may represent a model of a functional cure and their small reservoir size may approximate what can be achieved with different curative interventions. We show here that while ES have much lower total and intact DNA levels than CP, the ratio of intact to total DNA is similar to the ratios present in CP. This would imply that ES do not have a higher proportion of defective viral DNA than CP. It remains unclear whether the small reservoir size is a cause or consequence of elite control, but lower peak viral loads and earlier control of viral replication has been reported in some ES²⁵. It is possible that this early control of viral replication limits the seeding of the reservoir as has been described in CP who are treated with ART during primary infection²⁶. However, ES24, had levels of intact DNA that were higher than the levels seen in our CP. This would suggest that a small reservoir is not an absolute requirement for elite control.

Our results have implications for HIV cure strategies as they further elucidate the viral reservoir in patients who may represent a model of a functional cure of HIV-1 infection.