Abstract Virus-specific CD8 T cells exert potent antiviral activity against HIV-1 and SIV both in vitro and in vivo. Nevertheless, despite abundant CD8 T cell responses in HIV-1-infected humans and SIV- infected macaques, they are unable to fully suppress virus replication. This is likely due to the majority of HIV-1 and SIV replication occurring in CD4+ T cells concentrated within B-cell follicles in secondary lymphoid tissues; where virus-specific CD8 T cells are relatively few in number. In fact, we found that in vivo effector virus-specific CD8 T cell to target SIV RNA+ cell ratios (E:T) were over 40-fold lower inside compared to outside of B cell follicles in lymphoid tissues during SIV infection in rhesus macaques. These findings indicate that B cell follicles are an immune privileged site in which low levels of virus-specific CD8 T cells permit ongoing viral replication. Furthermore, we found that the majority of virus-specific CD8 T cells fail to express the follicular homing molecule CXCR5, likely explaining low levels of virus-specific CD8 T cells localizing to and surveilling B cell follicles. Taken together these data suggest that the inability of HIV- and SIV- specific CD8 T cells to fully suppress virus replication may be due to a deficiency of virus-specific CD8 T cells in B-cell follicles. These findings have led us to our central hypothesis that targeting HIV-specific T cells to B cell follicles will lead to durable remission of HIV infection. In support of this hypothesis we have shown, in SIV-infected rhesus macaques, that increased levels of virus-specific CD8 T cells in B cell follicles is associated with lower viral loads. To test this hypothesis, we propose to evaluate a T cell immunotherapy product that targets virus- specific CD8 T cells to B cell follicles. We are calling this product CD4-MBL-CAR/CXCR5 T cell immunotherapy. Our long-term goal is to develop an intervention that will lead to durable remission of HIV infection. To test this central hypothesis, we propose the following two specific aims. 1) To determine the in vivo localization, persistence and antiviral efficacy of autologous CD4-MBL- CAR/CXCR5 T cells infused into antiretroviral drug (ART) suppressed rhesus macaques before and after treatment interruption. 2) To determine whether preconditioning with CD20 antibodies (instead of cyclophosphamide) improves the abundance, persistence, or antiviral efficacy of autologous CAR/CXCR5-transduced CD8 T cells infused into ART suppressed rhesus macaques before and after treatment interruption. Our proposed studies targeting virus-specific CAR T cells to follicles will have a broad impact on the field by providing insights into cell trafficking, persistence, and pre-conditioning regimens for immunotherapy approaches. Most importantly, these studies could result in an effective strategy to induce long-term sustained remission of HIV.