Endogenously Expressed nef Uncouples Cytokine and Chemokine Production from Membrane Phenotypic Maturation in Dendritic Cells

Culture medium

RPMI 1640 (Cellgro; Fisher Scientific, Springfield, NJ) was supplemented with 2 mM l-glutamine (Life Technologies, Grand Island, NY), 50 μM 2-ME (Sigma-Aldrich, St. Louis, MO), 10 mM HEPES (Life Technologies), penicillin (100 U/ml)-streptomycin (100 μg/ml; Life Technologies), and 1% human plasma (heparinized).

Animals

Adult macaques (Macaca mulatta) were housed in the Tulane Regional Primate Research Center. Animal care operations were in compliance with the regulations detailed under the animal welfare act, and in the Guide for the Care and Use of Laboratory Animals. Before use, all animals used in this study tested negative for Abs to SIV, type D retroviruses, and simian T cell leukemia virus type 1. Male and female adult macaques were used for this study.

SIV isolates

The cloned viruses SIVmac239 (wild type) and SIVmac239 delta nef (delta nef) (36) were grown as previously described (37).

Generation of DCs

DCs were generated from PBMCs isolated from healthy macaques or HIV seronegative human donors. Donors were not screened for anti-Adeno immunity. Buffy coat units were purchased from the New York Blood Center to generate human DCs. Peripheral blood was collected by standard venipuncture using heparinized Vacutainers (BD Bioscience, San Jose, CA) from healthy SIV-seronegative rhesus macaques (anesthetized with 10 mg/kg ketamine HCl). The mononuclear cell fraction was isolated by Ficoll-Hypaque (Amersham Pharmacia Biotech, Uppsala, Sweden) density gradient (38). DCs were generated from either adherent PBMCs as previously described (27) or CD14⁺ monocytes isolated using the anti-human CD14 MACS system (Miltenyi Biotech, Auburn, CA) (6) and were plated at 3 × 10⁶ cells/well in a six-well tray (3 ml/well) in the presence of 100 U/ml recombinant human IL-4 (R&D Systems, Minneapolis, MN) and 1000 U/ml recombinant human GM-CSF (Immunex, Seattle, WA). Cytokines were added to the cultures on days 0, 2, 4, and 6 in culture. After 6–7 days in culture, immature DCs were harvested for infection.

Analysis of DC phenotype

The DC phenotype was monitored by flow cytometry for each experiment. At least 1 × 10⁴ DCs were resuspended in PBS/5% FCS/0.1% sodium azide (staining buffer; 100 μl/well of a V-bottom 96-well tray (ICN Biomedicals, Aurora, OH)). The cultures were stained with the appropriate PE-conjugated IgGs (BD Biosciences; Ancell, Bayport, MN; R&D Systems) or with PE-labeled mAbs against HLA-DR, CD25, CD58, CD80, (BDIS), CD4, CD40, CD86, CD206, CXCR4, CCR5 (BD PharMingen, San Diego, CA), CD83 (Immunotech, Marseilles, France), HLA-ABC (DAKO, Carpinteria, CA), CD74 (Ancell), CD209 (used for both human and macaque cells; R&D Systems), or CCR5 (used for macaque cells; R&D Systems). Indirect staining with goat F(ab′)₂ anti-mouse Ig (BioSource, Camarillo, CA) was used to detect CD205 expression using the mouse anti-human CD205 clone 38-2 (39), CD209 staining using either mAb 507 (provided by Dr. V. KewalRamani) (40) for macaque cells or DC28 for human cells (AIDS Research and Reference Reagent Program, National Institutes of Health, Bethesda, MD). The necessary unconjugated IgGs were included as negative controls for the indirect staining. The cells were incubated with the Abs for at least 20 min at 4°C, then were washed four times with staining buffer and fixed in 10% formalin in PBS (pH 7.2–7.4). For the indirect stains, the cells were washed after the primary Ab staining, incubated with the secondary goat-anti-mouse Ig, and washed again before fixation. Stained cells were examined by flow cytometry using a FACScan (BD Biosciences), and the data were analyzed with CellQuest software (BD Biosciences).

Isolation and phenotyping of CD4⁺ T cells

The CD14-negative fractions of the macaque and human PBMCs were cultured at 0.5–1 × 10⁷/ml for 6–7 days with the DCs, with additional fresh medium being added to the cultures every 2 days. On the day of the assay, the CD4⁺ T cells were further purified by negative selection using anti-CD8 and anti-HLA-DR MACS beads (Miltenyi Biotech) (6). The resulting cell preparations were at least 99% viable by trypan blue dye exclusion. Comparative infection studies (using macaque cells that are usually more fragile than human cells) confirmed that such cultured T cells were functional and behaved just like T cells freshly isolated from PBMCs on the day of the experiment, avoiding the need to re-bleed the donors (data not shown). The purity of the CD3⁺CD4⁺ T cells was verified as >95% by direct staining flow cytometry for membrane expression of CD8 and HLA-DR (vs the simultest isotype control). The activation state of the isolated CD4⁺ T cells was also monitored each time before performing the experiments by direct staining with FITC-conjugated anti-CD3 (BD PharMingen) combined with PE-conjugated anti-CD69, -CD25, or -HLA-DR (BD Biosciences, Fig. 5⇓F).

Adeno isolates and infection of DCs

Adenovirus expressing the HIV-1 SF2 nef allele (Adeno-nefHIV), a mutant in the PxxP region of SF-2 nef (Adeno-nefHIV_PA), or Adeno expressing the green fluorescent protein (Adeno-gfp) were previously described by Swingler et al. (22). Adeno expressing gfp and the HIV-1 SF2 nef allele was constructed using the Adeno-quest kit from Quantum Biotech (Qbiogene, Carlsbad, CA) and their pQBI-AdCMV5GFP shuttle vector. This virus was used for the extensive FACS analyses, allowing us to gate on gfp⁺ cells in both the control and nef-loaded populations. Recombinant Adeno encoding the SIVmac239 nef allele (Adeno-nefSIV) was derived from Adeno type 5, deleted of E1 and E3 regions to generate a replication-deficient Adeno (29), and purchased from Quantum Biotech (Durham, NC). For Adeno infection, 10⁵ immature DCs were infected in 50 μl medium with 100-1000 PFU of virus in a 96-well, round-bottom tray (ICN Biomedicals) for 2 h at 37°C. After 2 h fresh medium containing 1% human plasma was added. For T cell activation experiments 10²–10⁴ immature DCs were infected in 50 μl medium, and after 2 h 10⁵ autologous T cells were added to the DCs. Adeno-infected DCs or DCs mixed with T cells were cultured for various lengths of time before the indicated parameters were measured. Adeno stocks were monitored for endotoxin contamination using the Single Test Limulus Amebocyte Lysate Assay (N289-06; BioWhittaker, Walkersville, MD). Endotoxin levels in the Adeno preparations were routinely <0.06 endotoxin units/ml.

Immunoperoxidase staining of cells for Nef

Cytospins of DCs were prepared using a cytocentrifuge (Shandon, Pittsburgh, PA), loading ∼2 × 10⁴ cells/slide (41). Slides were air-dried for 1 h at room temperature and subsequently fixed in absolute acetone for 10 min at room temperature. The slides were incubated with a mouse mAb recognizing SIVmac251 Nef (17.2, AIDS Research and Reference Reagent Program) for 30 min at room temperature, washed five times with staining buffer, and incubated for 30 min with HRP-conjugated donkey anti-mouse Ig (DAM-HRP; 1/300 dilution; Jackson ImmunoResearch, West Grove, PA) for 30 min at room temperature. Non bound HRP-conjugated donkey anti-mouse Ig was washed off, and bound HRP was detected with stable 3,3′-diaminobenzidine tetrahydrochloride dihydrate (Research Genetics, Huntsville, AL). The slides were mounted in PBS/glycerol and analyzed using an Olympus AX70 microscope (Melville, NY).

Determination of T cell activation

To assess levels of T cell activation and proliferation, 10²–10⁴ immature DCs (infected or not with Adeno) were mixed with 1 × 10⁵ CD4⁺ T cells and cultured in a round-bottom, 96-well tray for up to 6 days. Tritiated thymidine ([³H]TdR; 1 μCi/well; NEN, Boston, MA) was added to the cells for the final 8 h of culture, and the [³H]TdR incorporated by proliferating cells was measured using a Wallac 1205 Betaplate liquid scintillation counter (Gaithersburg, MD). Responses are reported as mean counts per minute of [³H]TdR incorporated by duplicate or triplicate cultures (±SEM). T cell activation was also monitored by flow cytometry, where additional wells containing replicate DC-T cell mixtures were collected after 3 days and stained for T cell activation markers (PE-CD25, CD69, or HLA-DR) in combination with CD3 (FITC). The T cells were analyzed by gating on small and large activated lymphocytes by forward scatter excluding the DCs and DC-T cell conjugates. The gate marker was set relative to the isotype controls for each condition, and the percentage of cells stained above this cut-off point is given. Expression of the nuclear activation Ag Ki-67 was also monitored as a measure of T cell activation after 1–3 days of DC-T cell coculture by immunoperoxide staining of acetone-fixed cells (above) using the anti-Ki-67 mAb (MIB-1, AMAC, Westbrook, ME) (41). The percentage of Ki-67⁺ cells was calculated by counting the numbers of Ki-67⁺ cells in five independent fields (averaging 100 total cells in each) and is expressed as mean ± SEM over the five fields.

Measurement of chemokines and cytokines

The supernatants of Adeno-infected DCs (vs mock controls) were collected, by taking 50 μl from the 200-μl cultures and replacing it with 50 μl fresh medium at the indicated time points. Aliquots were stored at −20°C before analysis of the chemokine and cytokine content by ELISA. Recombinant anti-human CCL3 (macrophage inflammatory protein-1α (MIP-1α)), CCL4 (MIP-1β), CCL5 (RANTES), CXCL8 (IL-8), IL-10, and IL-6 ELISAs were obtained from R&D Systems. CCL3, CCL4, CCL5, CXCL8, and IL-6 ELISAs cross-reacted with macaque chemokines and cytokines. Anti-human IL-12 ELISA was purchased from Endogen (Woburn, MA) and cross-reacted with macaque IL-12. Recombinant anti-monkey ELISAs for TNF-α, IL-10, and IFN-γ were purchased from U-CyTech (Utrecht, The Netherlands).

In vitro SIV infection of DC-T cell cultures

After infection of immature DCs with Adeno as described above, 10⁵ T cells were added to 10⁴ DCs (per well) in a 96-well, round-bottom tray. The cocultures were subsequently pulsed with 5 × 10³ 50% T cell infectious dose of SIVmac239 wild type or delta nef/10⁵ cells for 1.5 h at 37°C as previously described (27). Every 2 days of a 15-day coculture, 50-μl aliquots were collected and stored at −20°C before the p27 content was analyzed by ELISA (Zeptometrix, Buffalo, NY). To determine whether the addition of maturation stimuli would rescue delta nef replication in immature DC-T cell cultures, the cultures were infected (27), after which the indicated maturation stimuli were added. These included 50% monocyte-conditioned medium (MCM) (42), soluble CD40 ligand (CD40L; 1/100 of a baculovirus stock) (6), or a combination of PGE₂ (10⁻⁷ M, PGE₂; Sigma P6532) with TNF-α (50 ng/ml; R&D Systems). Two or 3 days postinfection, DC maturation in the infected cultures was verified by FACS, measuring the expression of CD25, CD86, CD83, and HLA-DR. In some experiments 5 ng/ml staphylococcal enterotoxin B (SEB) or allogeneic CD4⁺ T cells were added to provide an activated T cell environment.

Statistical analysis

The statistical significance of the nef-induced chemokine and cytokine production was analyzed by comparing the peak values of chemokine or cytokine production in Adeno-nef- vs Adeno-gfp-infected DCs. The nonparametric SIGN test was used to directly compare the two groups. There were no statistically significant differences between the mock and Adeno-gfp controls, and p values (to three decimal places) are shown for the Adeno-gfp vs Adeno-nef cultures. A value of p < 0.05 was considered statistically significant.

Nef does not induce changes in the membrane phenotype of immature DCs

To measure nef-induced changes in immature DCs, Adeno vectors were used to express nef in the DCs (Adeno-nef). Infection of immature DCs with Adeno expressing the green fluorescent protein (gfp; Adeno-gfp) or Adeno expressing gfp and the HIV-1 SF2 nef allele (Adeno-gfp-nefHIV) enabled measurement of infection efficiency by FACS by monitoring the percentage of gfp-positive DCs (Fig. 1⇓A). The infection efficiency in different donors ranged from 5–90% (>10 experiments), averaging 40% for human DCs and 53% for macaque DCs. Similarly, the levels of gfp expression varied between donors and the various Adeno preparations, with the Adeno-gfp infections sometimes appearing more efficient than the Adeno-gfp-nefHIV infections, as shown in Fig. 1⇓A. However, this did not alter the nef effects observed (below). Both the levels of expression and the numbers of expressing cells typically peaked around 48 h postinfection with either Adeno-gfp-nefHIV or Adeno-gfp, and therefore gfp expression was routinely measured at this time. The levels of Nef expression (immunoperoxidase staining; Fig. 1⇓B) were variable between cells, with some Nef-positive cells staining very strongly for the Nef protein (dark reaction product) and others quite faintly, but above background (asterisks). The amount of Nef-expressing DCs and that of gfp-expressing DCs correlated closely. Nef expression in Adeno-nefHIV-infected DCs was also confirmed by Western blot (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Expression of gfp and nef in immature DCs infected with recombinant Adeno. A, FACS analysis of immature human or macaque DCs infected with Adeno-gfp-nefHIV (nef) or Adeno-gfp (gfp) or uninfected (mock) monitored 48 h postinfection (peak infection). The dot plots showing forward size scatter (FSC; x-axes) vs log green fluorescence intensity (gfp; y-axes) were gated on large cells and indicate the gfp-positive Adeno-infected DCs. B, Macaque DCs were spun onto glass slides 48 h after mock infection or infection with Adeno-nefSIV (nef) or Adeno-gfp (gfp). The cells were immunoperoxidase stained (dark reaction product) for the expression of SIV Nef. Faintly stained Nef-positive cells readily visible above background staining of Adeno-gfp-infected or uninfected cells and the isotype controls (not shown) are indicated by asterisks. The original magnification was ×40. Typical results from one of more than five experiments are provided.

To test our hypothesis that nef creates a mature DC we first analyzed whether the expression of nef in DCs alters the DC membrane phenotype. Our initial studies (>15 experiments) revealed that neither Adeno infection (35) nor the presence of nef (Fig. 2⇓) up-regulated the expression of common maturation markers (HLA-DR, CD25, CD83, CD86). The same was true when SIVmac239 nef was tested in both human and macaque DCs (data not shown), but more detailed analyses have been performed with the HIV construct that we have available in house. Advances in the DC field highlight many molecules that are expressed by DCs at different stages of activation, some of which may be important in HIV/SIV replication and/or DC-T cell communication (reviewed in Ref. 9). Thus, a wide panel of markers was monitored to more extensively assess the effects of nef on immature DC membrane phenotype (Fig. 2⇓). Uninfected controls run in parallel (not shown) with those cells in Fig. 2⇓ confirmed that Adeno infection has no impact on the expression of these markers by DCs (35). Notably, the DC phenotype was identical in the Adeno-gfp- and Adeno-gfp-nef-infected DC cultures (Fig. 2⇓). Despite some variability between both human and macaque donors (Fig. 2⇓B), there was no consistent up- or down-modulation of any markers in response to nef, even when the gfp-positive fractions (Adeno-infected cells) were directly compared (Fig. 2⇓). Importantly there were no changes in the expression of known HIV-binding molecules such as CD4, CCR5, CXCR4, CD206 (macrophage mannose receptor), or CD209 (DC-SIGN) in response to nef.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Phenotypic analysis of Adeno-infected immature human and macaque DCs. Human and macaque immature DCs were infected with Adeno-gfp-nefHIV (nef) or Adeno-gfp (gfp). Forty-eight hours postinfection the DCs were analyzed for expression of the indicated markers by staining with PE-conjugated mAbs or PE-conjugated goat anti-mouse Ig to detect unlabeled primary Abs. Indirect stains for CD205 (human and monkey, mAb 38-2) and CD209 (human, mAb DC28) are shown. Macaque DCs shown were stained with PE-labeled mAb 12507 (anti-CD209), but gave the same results with indirect staining. A, PE staining (log PE) of large gfp-positive cells (forward scatter (FSC) vs log PE) data are shown from one representative experiment of at least three similar experiments (different donors) for both monkey and human DCs. Isotype controls for direct stains exhibited mean fluorescence intensities of <1 log and indirect stains 1–1.5 logs. B, The PE relative fluorescence intensity (RFI) of the gfp-gated cells was calculated for each donor. The RFI for each phenotypic marker was determined by subtracting the mean fluorescence intensity of the appropriate isotype control. Symbols represent the same donor in each graph. ▪, ▴, and •, Three or four individual macaques; □, ▵, and ○, three human donors. Staining equivalent to the isotype control was assigned a value of 0.1 to accommodate log-scale analysis.

Nef induces chemokine and inflammatory cytokine production in immature DCs

Another characteristic feature of DC maturation is the secretion of chemokines and inflammatory cytokines (6, 43, 44). It has been shown that nef triggers macrophages to secrete CCL3 (MIP-1α) and CCL4 (MIP-1β) (22) and that exogenous recombinant Nef induces human DCs to secrete several cytokines and chemokines (26). We specifically investigated whether nef expressed in an immature DC could induce changes in chemokine or cytokine expression. Human and macaque DCs were analyzed for HIV nef-induced production of chemokines (Figs. 3⇓ and 4⇓). Nef expression in DCs induced the secretion of β-chemokines CCL3, CCL4, and CCL5 (RANTES). The secretion of CCL3 and CCL4 in human immature DCs peaked at around 6–18 h. CCL5 secretion tended to peak later (18–48 h) and at consistently lower levels. When immature macaque DCs were infected with Adeno-nefHIV, chemokine production was induced, but was delayed compared with that of human DCs, with the peak production occurring after 48–72 h (Fig. 3⇓A). The peak chemokine levels from independent experiments using DCs from 6–12 different human (upper row) or macaque (lower row) donors are summarized in Fig. 4⇓A. The p values indicate the significance of the nef-induced secretion in each case (Adeno-nef vs Adeno-gfp populations). While nef-dependent CCL3 and CCL4 secretion was statistically significant, CCL5 production was not; it was generally secreted at lower levels and less reproducibly between donors. CCL3 secretion was induced by both HIV and SIV nef in macaque (Fig. 3⇓B) and human DCs (data not shown). The kinetics of CCL4 production by HIV or SIV nef-bearing DCs were similar to those of CCL3 (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

SIV and HIV nef-induced chemokine production in human and monkey DCs. A, Immature human and macaque DC were left uninfected (mock) or were infected with Adeno-nefHIV (nefHIV) or Adeno-gfp (gfp). The supernatants were collected at the indicated time points after infection and assayed for CCL3, CCL4, and CCL5. The results represent one of >10 experiments. B, Immature macaque DCs were left uninfected (mock) or were infected with Adeno-SIVnef (nefSIV), Adeno-HIVnef (nefHIV), and Adeno-gfp (gfp), and supernatants taken at 24, 48, and 72 h were assayed for CCL3 production. C, Supernatants from human and macaque DCs uninfected (mock) or infected with Adeno-nefHIV (nefHIV), Adeno-nefHIV_PA (nefHIV_PA), and Adeno-gfp (gfp) were sampled at the indicated times, and CCL3 was detected by ELISA. The results shown in B and C are representative of at least three independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Peak cytokine and chemokine production induced by nef in immature DCs. Data represent the concentrations of chemokines and cytokines at the peak of production for 10⁵ DCs. Values are plotted for 6–12 independent experiments, where each dot represents the results from an individual donor. A, Immature Human (upper row) or macaque (lower row) DCs were mock-infected or infected with Adeno-nefHIV (nef) or Adeno-gfp (gfp). The supernatants were analyzed for CCL3, CCL4, CCL5, and IL-12 levels by ELISA. B, Supernatants from immature human DCs after mock infection or infection with Adeno-nefHIV or Adeno-gfp were analyzed for TNF-α, CXCL8, and IL-6 production by ELISA. The p values shown in each graph indicate the significance of the difference between nef- and gfp-expressing DCs cultures.

Two biochemical interactions of Nef have been described that putatively link Nef to signal transduction pathways: SH3 domains of protein tyrosine kinases of the Src family involving the conserved proline-rich (PxxP)₄ repeat in Nef (45, 46, 47), and a serine kinase (48) implicated in the effect of Nef on signal transduction in T cells. Both interactions have been mapped to the highly conserved core of the Nef molecule (49). Human and monkey DCs were infected with Adeno-nefHIV, Adeno-nefHIV_PA (PxxP mutant), and Adeno-gfp. Just like wild-type nef, the PxxP mutant of nef induced CCL3 (Fig. 3⇑C), CCL4, and CCL5 (data not shown) production in immature DCs. Therefore, the proline-rich motif is not required for nef-induced chemokine production by DCs.

We also observed that nef triggered statistically significant secretion of IL-12, TNF-α, CXCL8 (IL-8), and IL-6 in immature human (Fig. 4⇑) and macaque (Fig. 4⇑A and data not shown) DCs. On the average, cytokine release by macaque DCs was lower than that produced by human DCs (e.g., IL-12 (Fig. 4⇑A) and TNF-α (data not shown)) and in the case of TNF-α did not reach statistical significance. There was almost no background cytokine secretion in uninfected or Adeno-gfp-infected immature DCs of TNF-α, IL-6, or IL-12. In contrast, high levels of CXCL8 secretion were observed in uninfected immature DCs, as previously described (6). However, the nef-induced CXCL8 production was statistically significant. No nef-induced IL-1β, IL-10, or IFN-γ production was detected (data not shown).

Adeno-nef-infected DCs activate autologous T cells in coculture

The secretion of inflammatory cytokines and chemokines by nef-expressing DCs (Figs. 3⇑ and 4⇑) is a characteristic feature of DC maturation, yet the membrane phenotype of nef-expressing DCs remains that of an immature DC (Fig. 2⇑). Since mature DCs are more potent in T cell stimulation than immature DCs, and this usually coincides with elevated costimulatory molecule expression (1), we were interested to ascertain what impact nef had (if any) on the ability of immature DCs to activate autologous CD4⁺ T cells. Cocultures of nef-expressing DCs (vs control DCs) and autologous T cells were analyzed for T cell activation by several methods. T cell proliferation was monitored by [³H]TdR incorporation (Fig. 5⇓, A–D), expression of the cell cycle nuclear Ag Ki-67 (Fig. 5⇓E), and up-regulation of activation markers by FACS (Fig. 5⇓F). The negligible expression of CD69 and CD25 by the purified T cells before culture with nef-bearing vs control DCs is evident in Fig. 5⇓F (top row).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Nef expression in immature DCs induces T cell activation. A, Immature human or macaque DCs (10⁴ cells/well) were mock-infected (mock) or infected with Adeno-nefHIV (nefHIV), Adeno-nefSIV (nefSIV), or Adeno-gfp (gfp). After infection, resting autologous 10⁵ CD4⁺ T cells were added. T cell proliferation was assessed after 3 days (human) or 6 days (macaque) by measuring the amount of [³H]TdR incorporated. The mean counts per minute ± SEM [³H]TdR incorporated in triplicate cultures of one experiment that is representative of three are shown. B, Immature (Imm) and mature (Mat) DCs were generated from healthy macaque blood and cultured with CD4⁺ T cells isolated from the same (Autologous) or a different (Allogeneic) animal (10⁴ DC with 10⁵ T cells). The amount of [³H]TdR incorporated during the last 8 h of a 5-day culture is shown (mean counts per minute ± SEM from triplicate cultures from one of at least four similar experiments). C, Graded doses of mock-, nefHIV-, or gfp-infected immature human DCs were added to autologous CD4⁺ T cells (10⁵ cells/well) to obtain DC/T cell ratios of 1/10, 1/30, and 1/100. Three days after culture T cell proliferation was analyzed by measuring [³H] TdR uptake. Results are representative of two experiments, and values shown are the means (±SEM) of duplicate wells. D, Purified CD4⁺ T cells (10⁵/well) were cultured in medium (med) or in the presence of DC-conditioned supernatants (mock, nefHIV, or gfp). Supernatants were collected from mock-, nefHIV-, or gfp-infected immature DCs, and each was diluted relative to adding the nef-DC supernatant at a final of 1 ng/ml CCL3 (MIP-1α). CD4⁺ T cells were cultured with PMA (10 ng/ml; Sigma) and ionomycin (1 μM; Sigma) as a positive control for T cell activation (PMA/Ion). [³H]TdR incorporation was measured after 3 days of culture, and data from one of two identical experiments are shown. E, Mock-, nefHIV-, or gfp-loaded immature DCs were mixed with autologous CD4⁺ T cells (1 DC/10 T cells) and cultured for 1–3 days. At each time point the cells were collected, and cytospins were prepared. The acetone-fixed cells were then immunoperoxidase-stained for Ki-67 expression, and the numbers of Ki-67⁺ cells were counted. The mean percentages (±SEM) of Ki-67⁺ cells were calculated by counting five separate fields of at least 100 cells each. Data are shown from one of three comparable experiments. F, CD4⁺ T cells were purified (>95%) by depleting CD8⁺HLA-DR⁺ cells, and their activation status was monitored before coculture with the DCs (T cells). Mock-, nefHIV-, nefSIV-, or gfp-loaded immature human DCs were cultured for 3 days with autologous CD4⁺ T cells (1 DC/10 T cells). Before and after culture, the T cells were assayed for expression of the early activation marker CD69 as well as CD25 (for nefHIV where more extensive analyses were performed) by staining with FITC-conjugated anti-CD3 and PE-conjugated anti-CD69 or anti-CD25. The FITC-Ig and PE-Ig controls were included and used to set the marker gates shown in the DC-T cell cocultures. Fluorescence data are shown for cells gated on forward scatter and side scatter to encompass resting and activated T cells excluding DCs or DC-T cell conjugates. The percentages of PE⁺ cells above the indicated markers are shown in each DC-T cell panel. One example, representative of three independent experiments, is shown. Identical results were obtained with human and macaque cells.

Infection of immature DCs with either Adeno-nefHIV or Adeno-nefSIV provoked CD4⁺ T cell proliferation in human and macaque autologous DC-T cell cultures (Fig. 5⇑, A and C). In the human system T cells cultured with Adeno-nefSIV-infected DCs showed lower [³H]TdR uptake compared with Adeno-nefHIV-infected DCs. This was not apparent, however, in the macaque system. These levels of increased autologous T cell proliferation stimulated by the nef-expressing immature DCs reflected responses induced by mature DCs, but were significantly lower than responses seen in allogeneic systems (Fig. 5⇑B). The increased autologous T cell proliferation observed in the presence of nef-expressing DCs was dependent on the numbers of DCs added to the cocultures (Fig. 5⇑C). Differences in T cell activation were still evident at a dose of 1 DC/30 T cells, but not at the 1/100 ratio. The background autologous MLRs induced by the mock and gfp-loaded DCs similarly decreased with lower DC numbers. Further evidence that DCs are needed to augment T cell activation was provided by the fact that addition of nef-triggered DC supernatants to purified CD4⁺ T cells resulted in minimal T cell proliferation (<500 cpm [³H]TdR uptake compared with 100–200 cpm for the untreated, mock, or gfp-DC supernatant controls vs the stronger responses seen in PMA/ionomycin-positive controls; Fig. 5⇑D). Thus, soluble factors released by nef-triggered DCs alone are not sufficient to drive the T cell proliferation seen in the nef-DC-T cell cultures.

Expression of the nuclear activation Ag Ki-67 was also increased in nef-bearing DC-T cell cocultures (Fig. 5⇑E). In addition, FACS analysis revealed an increase in the expression of the activation markers CD69 and CD25 by T cells cocultured with nef-bearing autologous DCs (Fig. 5⇑F). The percentage of CD69⁺ T cells went from 26–29%, and that of CD25⁺ T cells from 14–18% in the mock or gfp-infected DC-T cell cultures to ∼45 and 26%, respectively, in the nef-loaded DC-T cell mixtures (Fig. 5⇑F). Peak T cell activation levels are shown (Fig. 5⇑). Hence, paralleling the elevated secretion of chemokines and cytokines induced by nef, nef-loaded immature DCs also stimulate greater autologous MLRs.

Nef, but not activation with maturation stimuli, drives delta nef replication in immature DC-T cell cultures

Mature, but not immature, DCs facilitate delta nef amplification in concert with autologous CD4⁺ T cells, whereas wild-type virus replicates strongly in the presence of either DC subset (27). We have hypothesized that nef in the wild-type virus can modulate immature DCs to act more like their mature counterparts to drive virus growth. While we detected nef-induced secretion of soluble factors by immature DCs as well as a nef-dependent augmentation of their ability to stimulate autologous MLRs, this did not coincide with changes in DC membrane phenotype. Thus, we set out to delineate whether activating DCs via traditional maturation stimuli vs nef would promote delta nef replication in the immature DC-T cell milieu.

Immature macaque DC-T cell cocultures were pulsed with delta nef (27), and the maturation stimulus, MCM, was added after the infection to mature the DCs in the culture. After 2–3 days the DCs had matured (FACS confirmed up-regulated CD25, CD80, CD83, and CD86; data not shown); however, replication of delta nef was consistently lower than that of wild-type virus (Fig. 6⇓A). Since MCM contains chemokines (50) that may interfere with the spread of infection in the DC-T cell mixtures, other maturation stimuli that are known to activate macaque DCs (PGE₂/TNF-α or CD40L) (6) were examined. Just as with MCM-activated DCs, maturation of the DC with either PGE₂/TNF-α or CD40L (confirmed by FACS analysis after 2–3 days; data not shown) did not rescue theimpaired replication of the delta nef virus (Fig. 6⇓B). As expected, wild-type virus grew normally in the immature DC-T cell mixtures, and activation of T cells with SEB rescued the delta nef replicative capacity. The slight decrease in virus production upon addition of the maturation stimuli was seen at most time points in all experiments, but was not significant.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Nef expression in DCs enhances delta nef replication in immature DC-T cell cultures. Immature DCs and CD4⁺ T cells were obtained from healthy macaques, and DC-T cell cocultures were infected with wild-type (wt) or delta nef (delta) SIVmac239 (27 ). After infection the cells were washed, and the indicated stimuli were added. The various DC-T cell mixtures were then cultured for up to 15 days, during which time supernatants were collected every 2 days, and productive infection was monitored by p27 ELISA. A, Fresh medium (med.) or fresh medium with 50% MCM was added to the cultures after infection. Peak virus production levels detected after 9 days are shown. B, After infection, fresh medium (med.), MCM, PGE₂ and TNF-α (P/T), CD40L, and SEB were added to the delta nef-infected cultures. Cocultures infected with wild-type virus (wt) were included as a control. The peak p27 levels seen on day 10 of culture are provided. C, After infection of immature DC-T cell cultures with wild-type virus (wt; ▪ and ▴) or delta nef (delta; □ and ▵), SEB was added immediately (day 0 (d0)) or 2 days later (d2) to the cultures. Virus replication was assayed over the following 11 days. D, Immature DCs were pulsed with either delta nef (□ and ▵) or wild-type virus (▪ and ▴), and the virus was washed away before DCs were mixed with allogeneic CD4⁺ T cells immediately (d0) or after an overnight incubation (d1). Productive infection was monitored over the next 15 days. E, Immature macaque DCs were infected with Adeno-nefSIV (nef) or Adeno-gfp (gfp) or were mock-infected as a control (mock). After 2 h autologous CD4⁺ T cells were added to the DCs, and the cocultures were immediately infected with delta nef. As a positive control for virus replication SEB was added to the delta nef-infected DC-T cell cultures. F, Immature macaque DCs were pulsed with Adeno-nefHIV (Expt. 1), Adeno-nefSIV (Expt. 2), or Adeno-gfp (gfp) or were mock-infected (mock) as a control. Autologous CD4⁺ T cells were then added just before the cocultures were infected with delta nef. The bars show peak p27 levels from two independent experiments. These data are representative of at least three separate experiments.

Although the DCs matured in response to the stimuli, it was possible that the 1–2 days it took for DCs to fully mature were too late to rescue the infectious delta nef persisting in these cultures. To test this, we stimulated the T cells after 1–2 days, since immediate activation of the T cells readily reveals the presence of infectious delta nef (Fig. 6⇑B) (27). Addition of SEB to delta nef-infected cocultures immediately or 2 days later rapidly amplified delta nef in these cultures (Fig. 6⇑C). Similarly, adding delta nef-pulsed DCs to allogeneic CD4⁺ T cell immediately or 1 day after the virus pulse also resulted in robust delta nef replication and underscored how DCs retain infectious delta nef virus for considerable lengths of time (Fig. 6⇑D). Therefore, infectious virus persists for up to 2 days and can be rescued upon encountering activated CD4⁺ T cells, but not simply by maturation of virus-carrying DCs within the cultures.

Since nef appears to have unique influence over immature DCs, triggering certain pathways of DC activation, but not others, we were interested in determining whether nef-bearing DCs could drive virus growth in the immature DC-T cell milieu. Immature DCs carrying nef (or not) were mixed with autologous CD4⁺ T cells and immediately exposed to delta nef. The expression of either SIV (Fig. 6⇑, E and F) or HIV (Fig. 6⇑F) nef in DCs facilitated SIV delta nef replication in immature DC-T cell cultures (Fig. 6⇑, E and F). Taken together these results stress how HIV and SIV nef are interchangeable and signal selective pathways in immature DCs that drive virus replication in immature DC-T cell cultures.