Back

RelB proximity proteomics and CRISPR screening define chromatin regulators of noncanonical NF-κB control of HIV latency and reactivation

Bussey-Sutton, C. R.; Gray, C. N.; Wu, B. J.; Peterson, J. J.; Burgos, S. D.; Mills, C. A.; Herring, L. E.; Browne, E. P.; Turner, A.-M. W.; Margolis, D. M.; Emerman, M.; Strahl, B. D.

2026-08-12 cell biology
10.64898/2026.08.11.744217 bioRxiv
Show abstract

Activation of the non-canonical NF-{kappa}B pathway via RelB/p52 signaling by SMAC mimetics such as AZD5582 is a promising strategy to induce HIV expression from latency, but the chromatin mechanisms linking RelB/p52 to proviral regulation remain poorly defined. Here, we combine RelB BioID proteomics, targeted HIV-CRISPR screening, and pharmacologic validation to identify ncNF-{kappa}B-associated regulators of HIV expression. RelB BioID revealed an extensive interaction network of chromatin and transcriptional regulators in basal and AZD5582-activated states. Basally associated RelB proteins included NSD2, SWI/SNF components, UHRF1, and DNMT1, while AZD5582-enriched proteins included p300, USP7, LSD1/KDM1A, NuRD components, SIN3A, and HBO1/KAT7. Functional screening using a custom guide RNA library targeting all BioID-identified factors identified regulators that promote HIV reactivation, including HBO1/KAT7, NSD2, and SIN3A, and regulators that restrict HIV expression, including p300, CHD4, and USP7. Because KAT7/HBO1 and p300 encode acetyltransferases with opposing screen phenotypes, we tested whether their catalytic activities contribute to HIV transcriptional regulation and found that, consistent with the CRISPR screen, KAT7/HBO1 inhibition reduced AZD5582-induced reactivation, whereas p300 inhibition enhanced it. Together, these data define a resource linking the RelB-associated chromatin landscape to HIV latency and reactivation.

Matching journals

The top 5 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.