Sangeetha Satheesan, Haitang Li, John C. Burnett, Mayumi Takahashi, Shasha Li, Shiny Xiaqin Wu, Timothy W. Synold, John J. Rossi, Jiehua Zhou

Although current combinatorial antiretroviral therapy (cART) is therapeutically effective in the majority of HIV patients, interruption of therapy can cause a rapid rebound in viremia, demonstrating the existence of a stable reservoir of latently infected cells. HIV latency is therefore considered a primary barrier to HIV eradication. Identifying, quantifying, and purging the HIV reservoir is crucial to effectively curing patients and relieving them from the lifelong requirement for therapy. Latently infected transformed cell models have been used to investigate HIV latency; however, they cannot accurately represent the quiescent cellular environment of primary latently infected cells in vivo. For this reason, in vivo humanized murine models have been developed for screening antiviral agents, identifying latently infected T-cells, and establishing treatment approaches for HIV research. Such models include humanized bone marrow/liver/thymus (BLT) mice and SCID-hu-thy/liv mice, which are repopulated with human immune cells and implanted human tissues through laborious surgical manipulation. However, no one has utilized the human hematopoietic stem cell (HSC)-engrafted NOD/SCID/IL2rγnull (NSG) model (hu-NSG) for this purpose. Therefore, in the present study we used the HIV-infected hu-NSG mouse to recapitulate the key aspects of HIV infection and pathogenesis in vivo. Moreover, we evaluated the ability of HIV-infected human cells isolated from HIV-infected hu-NSG mice on suppressive cART to act as a latent HIV reservoir. Our results demonstrate that the hu-NSG model is an effective surgery-free in vivo system in which to efficiently evaluate HIV replication, antiretroviral therapy, latency and persistence, and eradication interventions.IMPORTANCE HIV can establish a stably integrated, non-productive state of infection at the level of individual cells, known as HIV latency, which is considered a primary barrier to curing HIV. A complete understanding of the establishment and role of HIV latency in vivo would greatly enhance attempts to develop novel HIV purging strategies. An ideal animal model for this purpose should be easy to work with, should have a shortened disease course so that efficacy testing can be completed in a reasonable time, and should have immune correlates that are easily translatable to humans. We therefore describe a novel application of the HSC-transplanted humanized NSG model for dynamically testing antiretroviral treatment, supporting HIV infection, establishing HIV latency in vivo. The hu-NSG model could be a facile alternative to humanized BLT or SCID-hu-thy/liv mice in which laborious surgical manipulation and time-consuming human cell reconstitution is required.