To determine whether latency can be established and reversed in both proliferating and nonproliferating CD4+ T cells in the same model in vitro.

Activated CD4+ T cells were infected with either a nonreplication competent, luciferase reporter virus or wild-type full-length enhanced green fluorescent protein (EGFP) reporter virus and cultured for 12 days. The cells were then sorted by flow cytometry to obtain two distinct T-cell populations that did not express the T-cell activation markers, CD69, CD25 and human leukocyte antigen (HLA)-DR: CD69CD25HLA-DR small cells (nonblasts) that had not proliferated in vitro following mitogen stimulation and CD69CD25HLA-DR large cells (which we here call transitional blasts) that had proliferated. The cells were then reactivated with latency-reversing agents and either luciferase or EGFP quantified.

Inducible luciferase expression, consistent with latent infection, was observed in nonblasts and transitional blasts following stimulation with either phorbol-myristate-acetate/phytohemagglutinin (3.8 ± 1 and 2.9 ± 0.5 fold above dimethyl sulfoxide, respectively) or romidepsin (2.1 ± 0.6 and 1.8 ± 0.2 fold above dimethyl sulfoxide, respectively). Constitutive expression of luciferase was higher in transitional blasts compared with nonblasts. Using wild-type full-length EGFP reporter virus, inducible virus was observed in nonblasts but not in transitional blasts. No significant difference was observed in the response to latency-reversing agents in either nonblasts or transitional blasts.

HIV latency can be established in vitro in resting T cells that have not proliferated (nonblasts) and blasts that have proliferated (transitional blasts). This model could potentially be used to assess new strategies to eliminate latency.