HIV-1 Infection of DC: Evidence for the Acquisition of Virus Particles from Infected T Cells by Antigen Uptake Mechanism Narasimhan J. Venkatachari1, Sean Alber2, Simon C. Watkins2, Velpandi Ayyavoo1* 1 Department of Infectious Diseases and Microbiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America, 2 Department of Cell Biology and Physiology, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America Abstract Dendritic cells (DC) play a pivotal role in transmission and dissemination of HIV-1. Earlier studies reported that DC present at the site of infection trap virus particles via DC-SIGN and transfer the virus to the interacting na��ve �� T cells. This prompted us to ask the question whether DC could acquire virus from infected T cells during DC-T cell interaction. To address this, we investigated the likely transfer of virus from HIV-1 infected T cells to DC and the underlying mechanisms involved. Results indicate that DC acquire virus from infected T cells via antigen uptake mechanism and this results in infection of DC with expression of proteins directed by viral DNA. Further studies with HIV-1 lacking the Env protein also resulted in infection of DC. The use of antibodies against DC-SIGN and DC-SIGN-R ruled out a role for receptor in the infection of DC. Additional data show that DC infection is directly correlated with the ability of DC to take up antigen from infected T cells. Overall, these studies provide evidence to suggest that HIV-1, besides infecting immune cells, also utilizes immunological mechanism(s) to acquire and disseminate virus. Citation: Venkatachari NJ, Alber S, Watkins SC, Ayyavoo V (2009) HIV-1 Infection of DC: Evidence for the Acquisition of Virus Particles from Infected T Cells by Antigen Uptake Mechanism. PLoS ONE 4(10): e7470. doi:10.1371/journal.pone.0007470 Editor: Fu-Sheng Wang, Beijing Institute of Infectious Diseases, China Received April 17, 2009 Accepted September 22, 2009 Published October 15, 2009 Copyright: �� 2009 Venkatachari et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by grant R56 AI-50463 to VA from the NIAID. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: velpandi@pitt.edu Introduction HIV-1 infects macrophages, dendritic cells and T cells, which are also the key cells involved in inducing immune activation against invading pathogens [1,2,3]. HIV-1 transmission, infection and dissemination are facilitated by both cell-free and cell- associated virus in vitro and in vivo [4]. However, cell-associated virus transmission is more efficient than cell-free virus transmission [5,6]. Thus, HIV-1 has devised several strategies to utilize this pathway. One of the ways HIV-1 enhances viral transmission is by converting the immunological synapse to a virological synapse between the interacting antigen presenting cell (APC), T cell, and other immune cells. Dendritic cells (DC) are one of the first targets that encounter virus at the mucosal surface during transmission in vivo [7,8]. DC present under the mucosal membrane capture cell free virus as well as interact with infected donor cells, through the breached epithelial layer [1]. During this process DC capture virus particles and trans infect T cells efficiently as ������Trojan Horses������ [9]. In addition to the ability of DC to acquire virus in trans, a small percentage of DC are also infected in cis and support virus replication both in vivo and in vitro [10,11,12,13,14,15]. Thus, DC play a key role in infection, virus dissemination and pathogenesis. DC interact with pathogen infected/exposed cells in various tissue compartments as part of the immune surveillance function in vivo [16,17]. DC uptake antigens (from both cell membrane and cytoplasm) from infected cells, process and present them to na��ve �� and memory T cells. These studies indicate that there is sufficient interaction between DC and T cells during pathogen encounter. In HIV-1 infected individuals, activated CD4+ T lymphocytes are the major target cells for virus replication and infected T cells are present both in the periphery and in lymphoid organs [18,19]. Previous studies report that when an infected DC interacts with an uninfected T cell, captured virus in DC is transmitted to the T cells, that results in productive infection of T cells [10,20,21,22]. However, it is not known whether DC could acquire virus from an infected T cell resulting in infection of DC. To address this, we cocultured infected T cells with na��ve �� DC and evaluated the infection of DC by HIV-1. For this purpose, we used a HIV-1 reporter proviral plasmid that codes for EGFP before the nef open reading frame as described [23]. The reporter virus derived from the plasmid has allowed us to measure the expression and subcellular distribution of EGFP (driven by HIV-1 LTR) only in infected DC. Results presented here indicate that the cell-associated virus was taken by DC and infected DC as early as 12 hours and was maintained for more than six days, whereas cell free virus required 2���3 days to establish productive infection in DC. Infection of DC via infected T cell is dependent on T cell-DC contact and is independent of viral envelope and DC-SIGN. Furthermore, the percentage of DC infection is directly correlated with the ability of DC to acquire cell-associated antigen, suggesting DC could acquire virus from the infected T cells through the antigen uptake process. Collectively, these studies for the first time indicate that HIV-1 taken up by the DC through the antigen uptake mechanisms establishes cis infection in DC. PLoS ONE | www.plosone.org 1 October 2009 | Volume 4 | Issue 10 | e7470

Results Infection of DC mediated by cell associated virus DC generated as described in methods were cocultured with infected lymphocytes at a ratio of 2:9:1 (DC: uninfected PBL: infected PBL). Post coculture cells were stained for DC-SIGN, and EGFP+/ DC-SIGN+ cells were determined by flow cytometry. DC were gated based on side scatter and forward scatter followed by doublet discrimination gating (Fig. 1A). Single cells that are double positive for DC-SIGN+ and EGFP+ were considered as productively infected DC (Fig. 1A). Results from coculture experiment indicate that 7.6% of DC were infected at 12 hours post coculture with infected lymphocytes, whereas cell free virus did not infect DC (0%) at this time point (Fig. 1A). Addition of cycloheximide (CHX) (10 mg/ml) during coculture completely blocked infection of DC further confirming that EGFP expression in infected DC was due to de novo synthesis, and not due to cell conjugates or cell fusion. Comparison of Mean Fluorescence Intensity (MFI) of EGFP in infected DC and infected lymphocytes present in the same coculture (Fig. 1B), indicates that transcription of HIV-1 LTR driven EGFP in infected DC is significantly less compared to infected lymphocytes. DC infection was further confirmed by fluorescence microscopy where, DC-SIGN positive cells were EGFP also positive (Fig. 1C) as identified by the uniform subcellular distribution of EGFP that is indicative of de novo Figure 1. Reporter virus positive DC are the result of cis infection of DC. (A) DC were cocultured with HIV-1wt-EGFP reporter virus-infected PBL cells in the presence or absence of cycloheximide (10 mg/ml) or infected with cell-free virus. Post coculture (12 hrs), cells were stained for DC- SIGN. DC were gated based on side scatter and forward scatter followed by doublet discrimination (as shown in gating) and assessed for EGFP by flow cytometry. DC-SIGN and EGFP positive cells (%) are shown in the upper right quadrant. (B) Comparison of EGFP fluorescence (MFI) in infected lymphocytes and infected DC. Overlay of histogram of EGFP fluorescence in infected lymphocytes (green) and infected DC (red). (C) Detection of DC expressing EGFP by immunofluorescence microscopy. Red indicates DC-SIGN positive cells green represents EGFP positive cells Blue represents nuclear staining by DAPI. DC*, represents DC-SIGN and EGFP positive DC. (D) Detection of integrated HIV-1 proviral DNA in EGFP+ DC. DC were stained for DC-SIGN, and DC-SIGN+/EGFP+ DC were sorted by FACS. Integrated proviral DNA was assessed by real time Alu-LTR Taqman assay as described in Methods. To rule out contaminating lymphocytes in DC-SIGN+/EGFP+ sorted DC, mRNA from the sorted cells were evaluated for presence of CD28 mRNA by real-time PCR. RPLPO was used as control. Uninfected DC, Infected PBL controls were included. (E) DC were cocultured with HIV-1wt-EGFP reporter virus-infected PBL cells or with HIV-1wt-EGFP reporter virus-infected purified CD4+ T cells, or uninfected control cells. Twelve hours post coculture, cells were stained for DC-SIGN and assessed for EGFP positivity by flow cytometry. Cells (%) that are positive for DC-SIGN and EGFP are shown in the upper right quadrant. (F) DC were cocultured with either HIV-1wt-EGFP reporter virus-infected Jurkat T cells or with Jurkat cells expressing EGFP protein. Post coculture, the cells were stained for DC-SIGN and analyzed by flow cytometry or by (G) Immunofluorescence microscopy. DC-SIGN+/EGFP+ cells were gated based on the amount of EGFP in DC-SIGN+ cells to differentiate antigen uptake and productively infected DC. Results from multiple donors are shown in Fig. 1H, where 200 DC were counted scanning multiple fields (Fig. S2) for each culture. Figure represents one of 5���7 independent experiments with similar results. doi:10.1371/journal.pone.0007470.g001 Infection of DC PLoS ONE | www.plosone.org 2 October 2009 | Volume 4 | Issue 10 | e7470