Fig 1.
Design and cell characteristics of in vitro culture model.
(A) Schematic of primary CD4 T cell culture system. Primary CD4 T cells were isolated by negative selection, infected with the NL4-3 clone of HIV-1, and stimulated through TCR cross-linking with plate-bound immobilized anti-CD3 + anti-CD28 antibodies. After 4 days incubation, cells were recovered and transferred to new culture wells that lacked TCR stimulus. During the expansion phase of culture, cells were maintained in medium supplemented with exogenous cytokines, to enhance proliferation. IL-2 was added on day 4; and IL-15 was added on days 4 and 7. During the contraction phase of culture, interferon-beta (IFNβ) was added to the medium on days 10 and 12, to reduce the level of activation-induced cell death. Antiretroviral drugs were added (Nevirapin, at day 5 or Indinavir, at day 7) to prevent further spread of viral infection (refer to details in Materials and Methods for rationale in drug timing). Resting, infected cells were obtained at the end of culture on day 14. (B) Isolated CD4 T cells were either infected with NL4-3 (MOI = 0.01 TCID50) or mock infected (Uninfected) under similar experimental conditions and cultured in parallel for 14 days, according to the schema depicted in A. Indinavir was added on day 7 and replenished every two days until the end of culture (horizontal hatched bar). Cell number and viability were measured by trypan blue dye exclusion, along with culture volume, throughout the culture period to calculate the total number of viable cells present at each indicated time point (Days 4, 7, 10, 14). Relative cell numbers were derived by normalization to the initial cell count for each culture condition (infected vs. uninfected control). Results show the mean ± SEM (error bars) from 5 independent experiments, using cells from different donors. Refer to Materials and Methods for complete details of the culture system. (C) Flow cytometry analysis of cell cycle status (7-AAD staining of DNA content) and level of active DNA synthesis (bromodeoxyuridine uptake, BrdU) was performed on aliquots of infected and uninfected CD4 T cells, taken 14 days after culture initiation. For maximal sensitivity, cells were allowed to incorporate BrdU overnight, prior to fixation, permeabilization, and staining. Depiction of events in each cell cycle phase, based on DNA content, has been enhanced by coloration: G0/1, green; S, dark blue; G2/M, fuchsia. The presence of fragmented DNA (<G0/1), found with apoptosis is shown in red. For comparison between the uninfected control and infected cultures, the percentage of cells in S and G2/M cell cycle phases (lower right) and the total percentage of BrdU-positive cells (upper right) are shown in each dot plot. Representative results from one of 3 independent experiments.
Fig 2.
Kinetics of HIV replication induced during in vitro cell culture.
(A) Productive HIV replication was analyzed in parallel with the induction of cell proliferation (Fig 1B). The NL4-3 infected CD4 T cells (Infected/Stim) and uninfected control cells (Uninfected) were stimulated and maintained in culture for 14 days, according to the experimental design shown in Fig 1A. The horizontal hatched bar indicates the addition of antiretroviral Indinavir (HIV protease inhibitor) to culture (days 7–14). A separate portion of infected cells was maintained in culture medium alone, without the addition of stimulating agents, cytokines or Indinavir (Infected/US). Aliquots of culture supernatant were removed on days 4, 7, 10, 14; and assayed for levels of secreted p24 by ELISA. Results shown: cumulative mean ± SEM for 5 independent experiments (cells from same donors used for Fig 1B). (B) Representative experiment showing serial analysis of intracellular expression of HIV Gag (p24/55) in a culture of infected, TCR-stimulated CD4 T cells; Indinavir added on day 7. Results depicted by flow cytometry dot plots of forward scatter, FSC (cell size) versus intracellular Gag (ICp24/55) staining. The positive cursor was set according to the fluorescence threshold of the uninfected cell control, stained in parallel with anti-p24/55 antibody. The percentage of cells, expressing intracellular Gag (ICp24/55) at each time point is given in the upper right corner of each plot. Representative results from one of 5 independent experiment, using cells from different donors. (C) Infected and stimulated CD4 T cells were cultured, with Indinavir present from day 7 through day 14. Cell aliquots were taken on days 4, 7 and 14 for RT-qPCR analysis of five different species of HIV transcripts. RNA copy numbers were normalized to 25 ng total RNA input. Results show the mean ± SEM of 3 experiments, using cells from different donors.
Table 1.
Quantification of HIV latent infection at the end of culture.
Fig 3.
Cell stimulation status at the time of infection and the amount of cell proliferation following infection impact the establishment of HIV latency.
Aliquots of freshly isolated primary CD4 T cells were infected with HIV at different time points relative to the addition of activated stimulus: before (day 0) or following (days 4, 7, 10) initial TCR stimulation. Antiretroviral drugs were not added to culture in these experiments. (A) Productive virus replication, following infection at varying time points after cell stimulation (day 4) was monitored sequentially over time in culture (days 0 to 14) for the levels of secreted p24 production by ELISA analysis. (B) Cell aliquots were taken at the end of culture (day 14), to extract genomic DNA and perform qPCR analysis for the quantification of integrated HIV DNA copies. Sample legend (A and B) for time of HIV infection relative to the addition of activation stimulus: uninfected control, Uninf; infected on day 0, Id0; on day 4, Id4; on day 7, Id7; on day 10, Id10. Data shown (A and B) represent the mean ± SEM from 3 independent experiments, using cells from different donors. In a separate set of cell-sorting experiments (C, D and E), our standard 14-day cell model of HIV latency (Fig 1A) was used to examine the influence of infected cell proliferation on the capacity to establish latent infection. Prior to cell infection and stimulation, the isolated CD4 T lymphocytes were stained with CFSE dye to track progressive cell divisions during culture. (C) Diagram of experimental design for the identification and sorting of cell subsets with different proliferation profiles (representative graphics taken from one of 4 replicate experiments). Nevirapine (RT inhibitor) was added on day 5. At the end of culture (day 14), cells were sorted, based on their CFSE content and proliferation profile (C), into subsets that: i) had divided only a few times (CFSE hi+mi / far-right top panel), using a sorting gate depicted by the green bar in the Day 14 histogram (right panel), which was determined by the cell division profile seen on Day 5 (middle panel histogram); or ii) had divided many times following day 5 (CFSE lo / far-right bottom panel), using the red bar sorting gate depicted in Day 14 histogram. (D) Aliquots of infected cell subsets, recovered at day 14 from 4 replicate sorting experiments were analyzed for copies of integrated HIV DNA per 500 ng total genomic DNA (80,000 cell equivalents). (E) Additional aliquots from the recovered infected cell subsets were restimulated using plate-bound anti-CD3 + anti-CD28. Induction of soluble p24 release (left graph) and percent intracellular Gag (ICp24/55) expression (right graph) were quantified at 7 days following the secondary reactivation by TCR stimulus. Individual data points, with mean and SEM, are shown for 4 experiments using cells from different donors.
Fig 4.
Latent infection is established in CD4 T cells that do not divide in response to TCR stimulation.
(A) Representative example of the experimental design for identification and sorting of cell subsets with different proliferation profiles. CD4 T lymphocytes were prepared and cultured as described in Fig 3C. Nevirapine, RT inhibitor was added on day 5. At the end of culture (day 14), cells were sorted based on their CFSE content and proliferation profile into subsets that had: i) not divided (CFSEhi); ii) divided only a few times after day 5 (CFSEmi); or iii) divided many times after day 5 (CFSElo). (B) Infected cell subsets, recovered from sorting on day 14, were washed and restimulated with immobilized anti-CD3 + anti-CD28 to induce a second round of productive virus replication. Each recovered cell subset was analyzed by flow cytometry for expression of intracellular Gag (ICp24/55) at 2 and 7 days, following reactivation. Representative data, from one of 4 experiments, are shown. (C) Cell subsets isolated at the end of the initial culture phase (day 14) were tested for their capacity to respond to a second TCR stimulation. On day 4 following restimulation, aliquots from the uninfected control culture and each infected cell subset were analyzed by flow cytometry for cell status (7-AAD staining of DNA content) and level of active DNA synthesis (2 hr pulse with BrdU). Representative data, from one of 3 experiments, are shown. (D) Bar graph of infected (black) and uninfected (gray) cell subpopulations that were stained on day 14 for CD45RA and CD27 co-expression. Bars show the proportions of different T cell maturation phenotypes found in the CFSEhi subset of cultured cells that had not proliferated. Results depict the mean + SD of 4 experiments, using cells from different donors.
Table 2.
Integrated HIV in sorted CFSE-stained cell subsets.
Fig 5.
Characterization of CD4 T cell subset responses following TCR stimulation.
Aliquots of isolated CD4 T cells were stained with CFSE dye, infected with HIV, and activated with immobilized anti-CD3 + anti-CD28 antibodies, as described in Fig 4A, except no antiretrovirals were added. After 4 days, the cells were removed from TCR stimulus, washed, resuspended in fresh medium supplemented with IL-2 and IL-15, and transferred to new culture plates. Cell samples were collected daily on days 4 to 7, for flow cytometry expression analysis of antigen markers associated with activation (CD69, CD25 and CD38) and exhaustion (PD-1, TIGIT). (A) The gating strategy based on number of cell divisions. The far-right cell population peak (bottom panel) was set as the non-dividing subpopulation (CFSEhi), based on the unstimulated cell profile (top panel). Cells that divided 1–4 times composed the CFSEmi subset, and cells that went through more than 4 divisions composed the CFSElo subset (bottom panel). Percentages of cells expressing each activation (B) and exhaustion (C) associated marker were determined for each of the three defined subpopulations (CFSEhi, CFSEmi, and CFSElo) throughout the 7-day time course. Results depict the mean values ± SD (error bars) of 3 experiments, using cells from different donors.
Fig 6.
Potential influence of Vpr in the establishment of latent HIV infection in primary CD4 T cell cultures, following TCR stimulation.
Experimental design, based on that described in Fig 5. Aliquots of isolated CD4 T cells were stained with CFSE dye, infected with isogenic HIV clones, and activated with immobilized anti-CD3 + anti-CD28 antibodies. Cell samples were infected in parallel (MOI, 0.01 IU) with wild type (wt) NL4-3 and its vpr-mutated clone (vpr-). After 4 days of activation, the cells were removed from TCR stimulus, washed, resuspended in fresh medium supplemented with IL-2 and IL-15, and transferred to new culture plates. No antiretroviral drugs were added. Cell samples were collected daily on days 4 to 7 to evaluate HIV replication, using flow cytometric analyses with the gating strategy and identification of CFSE-stained cell subset proliferation profiles, as depicted in Fig 5A. (A) Proportions of non-dividing cells (CFSEhi) in the infected cell population that expressed Gag (ICp24/55) were compared between the cultures infected with wild type (wt) and vpr-mutated clone (vpr-) of NL4-3, throughout the 7-day time course. (B) Percentages of cells expressing intracellular Gag were determined for each of the three defined cell subpopulations (CFSEhi, CFSEmi, CFSElo), comparing the wt and vpr- infected cultures over time and increasing cell divisions. (C) In the non-dividing cell subpopulation (CFSEhi) that was present throughout culture (6B), the relative amount of Gag protein expressed per cell (mean fluorescence intensity, MFI) was measured initially and followed through subsequent cell divisions in samples taken on days 4 to 7, comparing the wt and vpr- infected cultures. Results depict the mean values ± SD (error bars) of 3 experiments, using cells from different donors.
Fig 7.
Direct establishment of latent HIV infection in non-dividing “bystander” CD4 T cells during co-culture with productively infected cells.
(A) Schematic diagram of experimental design. After isolation of CD4 T lymphocytes, a portion of the cells was stained with CFSE dye, and the remaining cells were left unstained. The unstained cells were infected, stimulated, and cultured according to the standard in vitro model protocol (condition #3 in diagram). Following 4 days in culture, the unstimulated, uninfected, CFSE-stained cells (condition #2 in diagram) were mixed with the infected, stimulated cells in a 1:1 ratio. The cell mixture was cultured for an additional 10 days in the presence of cytokines; Indinavir was added at day 7. For comparison, a separate portion of the CFSE-stained cells was infected, stimulated, and cultured alone in parallel (condition #1 in diagram). (B) The cells were monitored for intracellular Gag (p24/55) expression and degree of cell division (CFSE content) until the end of culture. Top row: CFSE-stained, infected, and stimulated cells. Bottom row: CFSE-stained, unstimulated cells (green subpopulation) mixed at day 4 with infected, stimulated, unstained cells (red subpopulation). Representative results are shown from one of 3 independent experiments, using cells from different donors.
Table 3.
Integrated HIV copies and cell-associated infectious units in non-dividing (CFSE+) vs dividing (CFSE-) subsets.
Table 4.
Integrated HIV DNA, IUPM, and integrants/IU in latently infected bystander CD4 cells derived from the in vitro co-culture model.
Fig 8.
Optimized model of bystander cell infection and the importance of cell-to-cell contact.
(A) Diagram of the optimized in vitro bystander cell model to generate latently infected primary CD4 T cells. A portion of isolated primary CD4 T lymphocytes were labeled with CFSE dye, held overnight, washed, and then acutely infected (6–20 hr) with NL4-3. Following infection, the cells were cultured in flat-bottom plates with immobilized anti-CD3 + anti-CD28 antibodies to stimulate cell proliferation and virus replication. Four days following activation, the infected CSFE-labeled cells were removed from the TCR stimulus and mixed with unstimulated, uninfected autologous CD4 T cells (unlabeled); and the cell mixture, cultured in fresh medium containing exogenous rIL-2 and rIL-15 for 3 days. On day 7 of culture, the non-dividing CFSE-negative subpopulation (“bystander cells”) was isolated by FACS. The recovered resting cells, which carry non-productive latent HIV infection, were cultured in fresh medium without cytokines for 2–3 days, before use in subsequent experiments. (B) Schematic representation of the transwell membrane culture system. (C) HIV integration was quantified by ddPCR in the target bystander cells, following infection in transwell cultures or direct co-culture cell mixtures. Results shown as HIV integrants per 80,000 cell equivalents. (D) Representative example of flow cytometry gating strategy to quantify percentages of cultured cells that expressed intracellular Gag (p24/25), following TCR stimulation for 48 hr. Note that percentage of Gag-positive cells, following establishment of latent infection, was lower in the transwell cultures than in the direct mixed cell co-cultures. (E) Fold increases (TCR stimulated / baseline) in the percentages of cells expressing intracellular Gag (ICp24/25), following 48 hr TCR stimulation. Results for C. and E. depict the mean values ± SD (error bars) of 3 experiments, using cells from different donors.