Back

Investigating the dynamics of proviral silencing in polyclonal HIV-1 infected Jurkat cell populations

Clark, S.; Atindaana, E.; Gopal, K.; Kidd, J. M.; Telesnitsky, A.

2026-02-26 microbiology
10.64898/2026.02.26.708239 bioRxiv
Show abstract

Previous work demonstrated that individual HIV-1 provirus-containing Jurkat cell clones maintain stable bimodal expression patterns throughout 8 to 9 days of culture. However, within polyclonal infected cell pools, the overall proportion of cells displaying HIV-1 expression declines with time. Here, we examined the processes underlying population silencing in polyclonal pools comprised of hundreds of individual barcoded proviral clones throughout culturing periods of 22 and 90 days. Monitoring HIV-1 LTR activity via a GFP reporter confirmed that initially most clones exhibited bimodal expression, with mixtures of transcriptionally active and inactive cells. Over time, however, the overall fraction of LTR-active cells declined substantially. High-throughput clonal tracking showed that this decline was not driven by uniform transcriptional silencing. Instead, population silencing resulted from a combination of mechanisms: selective expansion of clones with low LTR activity, reductions in expression within certain clones, and long-term maintenance of stable bimodal expression in others. Notably, over half of clones retained stable bimodal expression patterns after 22 days, and approximately 17% of the clones retained stable bimodal expression even after 90 days in an independent second experiment, despite ongoing phenotypic switching at the single-cell level. These findings demonstrate that population-level HIV-1 silencing emerges from heterogeneous, clone-specific behaviors rather than uniform transcriptional repression. These observations confirm stability for a large subset of clones and provide insight into how integrants with diverse replication and expression properties can act together during HIV-1 polyclonal population silencing.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.