A cell line model for the study of CD4-negative HIV-1 infection and latent virus reservoirs

Abstract

Although controversial, limited publications support the notion that HIV-1 can infect CD4-negative cells. The objective of this study was to provide a comprehensive investigation of a universally available CD4-negative cell line model system that can be infected with X4 and R5 HIV-1 to generate integrated proviral DNA and serve to study latent viral reservoirs. The reason that HIV-1 infection of CD4-negative cells has become less investigated is due to a lack of a fully characterized model for the study of this unusual pathway. To address this critical need, human osteosarcoma (HOS) cells, engineered to express either CD4, CCR5 or CXCR4, and easily available from a commercial source were used. CD4 expression was examined using western immunoblot, flow cytometry, anti-CD4 blocking antibody and mRNA expression. Cells were infected with HIV-1 pseudo-enveloped viruses bearing either JR-FL (R5-tropic) or HXB2 (X4-tropic) envelopes, constructed on NL4-3 luciferase/GFP backbone. Infection was monitored by luciferase readout and visualized by GFP immunofluorescence. Raltegravir was used to inhibit integration, and AMD3100 and maraviroc used to block chemokine coreceptors, CXCR4 and CCR5, respectively. Productive versus latent infection was quantified by dual-fluorescence readouts using HI.fate.E. We confirmed that HOS cells lack CD4. HOS cells expressing only CCR5 or CXCR4 supported HIV-1 infection, although infection was significantly lower than in matched CD4-positive controls. Raltegravir treatment blocked proviral integration in all instances. Coreceptor antagonism and envelope-deficient viruses revealed that infection of CD4-negative CXCR4 cells remained CXCR4-dependent, whereas CD4-negative CCR5 cells showed evidence of CCR5-independent infection. Dual-reporter HI.fate.E assays indicated that CD4-negative cells could support both productive and latent infection. These studies establish a universally available cell line model for the study of CD4-negative HIV-1 infection. This cell line model will provide insight into the question of how CD4-negative cells can be infected with HIV-1 and whether CD4-negative cells can provide latent viral reservoirs in HIV/AIDS. Author summarySince the first description of HIV/AIDS in 1981 and the recognition that CD4 was a primary receptor for HIV-1 in 1983, a limited number of reports have suggested that cells lacking CD4 could be infected with HIV-1. These reports continued even when it was shown in 1996 that co-receptors, CXCR4 and CCR5, were also required for HIV-1 infection of CD4 T-helper cells. Indeed, crystallography studies showed that CD4 was required to interact with the HIV-1 envelope gp120 in order to cause conformational changes in the envelope to expose the binding motif for chemokine co-receptor engagement, required for additional conformational changes to expose the gp41 fusion protein, allowing for entry and infection. However, reports continued that cells lacking CD4 could be infected which raised questions as to how this can happen. To address this critical gap, we have identified a cell line, HOS, that is commercially available, having expression of CD4, CXCR4 and/or CCR5. Using these HOS cell lines, we have been able to confirm that HIV-1, either X4 or R5 enveloped viruses, can infect CD4-negative cells. We have also confirmed that infection is productive and allows for latent proviral integration. Our findings provide a system for further studies of the mechanism(s) of HIV-1 infection of CD4-negative cells using a consistent model and may aid in elucidating establishment of viral reservoirs.