Immunomodulatory effects of HSV-2 infection on immature macaque dendritic cells modify innate and adaptive responses

Dendritic cells (DCs) are involved in innate and adaptive immunity.^1-5  DCs survey for pathogens to which they respond innately while also processing pathogens and presenting antigenic determinants to induce adaptive immune responses. DCs need to be activated or matured to stimulate potent adaptive immunity.⁶  Maturation involves the up-regulation of molecules on the DC surface and secretion of cytokines and chemokines that encourage the DC–T cell interactions needed to elicit strong immunity. Many pathogens trigger these pathways, modifying DC functions to encourage effective immune activation and clearance of infection.^7-10  Yet, other pathogens like immunodeficiency viruses (human [HIV] and simian [SIV])^11-15  and herpes simplex viruses (HSV)^16-18  exploit DC biology to facilitate infection and elicit immune responses incapable of preventing or eradicating infection. Moreover, there is a strong correlation between genital HSV (HSV-2) infection and the probability of acquiring HIV.¹⁹  Understanding how this might be orchestrated at the DC level is central to developing strategies to prevent DC-driven HIV spread.¹² 

Primary HSV-2 infection, occurring at the mucosal surfaces, is typically followed by the establishment of latency in the sacral root ganglia.²⁰  Neutralizing antibodies (Abs) and antiviral CD4⁺ and CD8⁺ T cells are induced,²⁰  which ultimately restrict virus replication at local sites to resolve (primary and reactivated) lesions.^21,22  DCs likely play a key role by orchestrating responses to HSV,^21,23-25  although low-level productive infection of DCs might also contribute to virus spread.¹⁸  HSV infection of human monocyte-derived DCs (moDCs) is cytopathic^23,26,27  and results in the down-modulation of several surface markers involved in the activation of T cells.^16-18  While this would result in mediocre anti-HSV immunity allowing the establishment of HSV infection (but not preventing infection), it must be sufficient to clear virus upon reactivation.^24,25  Herpetic lesions also comprise activated leukocyte infiltrates and enable direct blood contact, providing mechanisms for increased HIV spread. It is possible that HSV-2 infection of immature DCs additionally alters innate DC responses and compromises the ability of DCs to elicit potent adaptive responses to other pathogens, thereby further exacerbating HIV infection.

The initial step toward dissecting this biology in a relevant animal model was to validate HSV-2 infection of macaque DCs. We provide the first proof that immature macaque moDCs can be infected by HSV-2. We observed DC apoptosis upon exposure to infectious HSV-2 as well as reduced immunostimulatory functions in HSV-2–infected DCs. Notably, HSV-2–infected DCs stimulated weaker SIV-specific T-cell responses. These data will enable future studies in the macaque to explore how HSV-2 infection affects DC-driven HIV transmission and if the presence of coinfections, like HSV-2, influences the efficacy of preventative or therapeutic approaches.

Adult rhesus macaques (Macaca mulatta) were bred and housed at the Tulane National Primate Research Center ([TNPRC] Covington, LA). At the commencement of these studies all naive animals tested negative by polymerase chain reaction (PCR) for simian type D retroviruses, simian T-cell leukemia virus, and SIV. Healthy SHIV162P-infected macaques (infected about 6 years ago) were used as a source of virus-primed T cells for SIV-specific immune assays.¹⁵  Protocols were reviewed and approved by the Institutional Animal Protocols Care and Use Committee of the TNPRC. Animal care procedures were in compliance with the regulations detailed under the Animal Welfare Act²⁸  and in the “Guide for the Care and Use of Laboratory Animals.”²⁹ 

Peripheral blood mononuclear cells were isolated from heparinized macaque blood or human leukopacks (New York Blood Center, New York, NY) using Ficoll-Hypaque density gradient centrifugation (Amersham Pharmacia Biotech, Uppsala, Sweden). CD14⁺ monocytes were isolated using the nonhuman or human primate CD14 magnetic cell sorting (MACS) system (Miltenyi Biotec, Auburn, CA) and moDCs generated by culturing these cells in 100 U/mL recombinant human interleukin-4 (IL-4) (R&D Systems, Minneapolis, MN) and 1000 U/mL recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF) (Berlex Laboratories, Montville, NJ).^30,31  After 6 to 7 days, immature DCs were collected, an aliquot taken for flow cytometry, and the remainder used for HSV-2 infections. DC phenotype was assayed using fluorescein isothiocyanate (FITC)–conjugated anti–HLA-DR (clone L243; BD Biosciences, San Jose, CA) in combination with phycoerythrin (PE)–conjugated immunoglobulin G1 (IgG1) (clone X40, BD Biosciences Immunocytometry Systems, San Jose, CA), IgG2b (clone 27-35; BD Biosciences Pharmingen), anti-CD25 (clone 2A3; BD Biosciences Immunocytometry Systems), anti-CD80 (clone L307.4; BD Biosciences Immunocytometry Systems), anti-CD83 (clone HB15A; Immunotech, Marseille, France), or anti-CD86 (clone IT2.2; BD Biosciences Pharmingen).

Culture medium (R1) was RPMI 1640 (Cellgro; Fisher Scientific, Springfield, NJ) containing 2 mM L-glutamine (GIBCO Life Technologies, Grand Island, NY), 10 mM HEPES (N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid) (GIBCO Life Technologies), 50 μM 2-mercaptoethanol (Sigma Chemical, St Louis, MO), penicillin (100 U/mL) and streptomycin (100 μg/mL) (GIBCO Life Technologies), and 1% heparinized human plasma.

Viral stocks were propagated in Vero cells (American Type Culture Collection [ATCC] Manassas, VA), titered by plaque formation on Vero cells,³²  and aliquots stored at –80°C. Immature moDCs were exposed to HSV-2 (strain G; ATCC) at 1 or 5 plaque-forming units (pfu) per cell for 1 hour at 37°C. Controls were set up alongside where DCs were exposed to the medium in which the viral stocks were grown (Dulbecco modified Eagle medium [DMEM] containing 2% fetal bovine serum [FBS]). Typically, 10⁶ cells were added to Eppendorf tubes (precoated with RPMI 1640 containing 10% fetal calf serum [FCS] on ice for 5 to 10 minutes) in a final volume of 100 μL R1 containing 1 to 5 pfu HSV-2 per cell. After incubation, cells were washed 5 times with 500 μL cold R1 (3 minutes, 3000 rpm) to remove unbound virus. Cells were recounted, resuspended in medium with 1000 U/mL GM-CSF and 100 U/mL IL-4, and plated at 0.5 × 10⁶ to 1 × 10⁶ cells per well of a 24-well plate (2 mL) or 2 × 10⁵ to 5 × 10⁵ cells per well of a 48-well plate (250 μL to 1 mL). UV-inactivated virus preparations were used to verify the effects of infectious virus. Aliquots (100 to 200 μL) of virus were plated in a 96-well flat-bottomed tray on ice and exposed to UV light (Blak Ray Lamp, Model UVL-21; UVP, Upland, CA) for 4 hours. Virus inactivation (4 to 5 log reduction in titer) was verified using the Vero cell plaque assay.

DCs (1 × 10⁶) were collected after reculture, washed with 500 μL of 1× Dulbecco phosphate-buffered saline (DPBS) containing 2% FCS (Cellgro; Fisher Scientific), resuspended in 250 μL per tube of Fixation/Permeabilization Solution (BD Biosciences Pharmingen), and incubated for 20 minutes at 4°C in the dark. Cells were washed twice with 500 μLof1 × Perm/Wash Buffer (BD Biosciences), resuspended in the same buffer, and plated into a 96-well V-bottomed tray (about 1 × 10⁵ cells per well). Cells were stained with 1 μg/mL anti–HSV-2 ICP8 monoclonal antibody (mAb) (IgG2a isotype; Virusys, North Berwick, ME) or 1 μg/mL mouse IgG2a isotype control (R&D Systems). After 30 minutes of incubation (room temperature in the dark), cells were washed 4 times with 1 × DPBS (100 μL per well). Bound Abs were detected with biotin-SP (Spacer)–conjugated AffiniPure Donkey Anti-mouse IgG (5 μg/mL; Jackson ImmunoResearch Labs, West Grove, PA) and incubated for 30 minutes at room temperature in the dark. Cells were then washed (as above). Cellbound biotinylated Abs were detected by adding Qdot 525 streptavidin conjugate (10 nM final concentration; Quantum Dot, Hayward, CA) and incubated (30 minutes at room temperature).

For double staining with PE-conjugated Abs to monitor DC phenotype, the cells were then incubated (15 minutes, 4°C) with 5% normal mouse serum (Jackson ImmunoResearch Labs) before being incubated with PE-conjugated IgG2a (BD), anti-IgG2b (BD Pharmingen), anti–HLA-DR (BD), anti-CD40 (BD Pharmingen), anti-CD80, anti-CD83, or anti-CD86 for 30 minutes (4°C in the dark). Cells were washed 4 times with 1 × DPBS. Ab dilutions were made in the Qdot incubation buffer (Quantum Dot). The cells were acquired on a FACSCalibur flow cytometer (BD Biosciences) and data analyzed using FlowJo software (Tree Star, Ashland, OR).

Recultured DCs were collected into 1.5 mL Eppendorf tubes, washed 4 times in serum-free RPMI 1640, and brought to 1 × 10⁶ cells per mL; 25 μL (2.5 × 10⁴ cells) was placed onto a spot of an Alcian blue–coated slide^30,33  and allowed to adhere for 8 minutes at 4°C. After confirming cells had adhered, the supernatant was carefully aspirated and the adhered cells washed 5 times with 1 × DPBS. Cells were then fixed with a drop of freshly prepared 4% paraformaldehyde (PFA) (Electron Microscopy Sciences, Washington, PA) in 1 × DPBS (20 minutes, room temperature). PFA was carefully aspirated and cells washed 5 times with 1 × DPBS. To quench endogenous peroxidase activity, a drop of 0.3% H₂O₂ in 1 × DPBS (made fresh each time) was added to each spot of cells for 1 hour (room temperature). The slides were then washed 5 times with 1 × DPBS and the last drop of 1 × DPBS left on for 5 minutes; 1 × DPBS was then aspirated, and nonspecific binding was blocked with the blocking buffer (30 minutes, room temperature) (TSA Fluorescein System; Perkin Elmer, Boston, MA). Cells were washed 5 times with 1 × DPBS, and 25 μL of the diluted anti-HSV Abs was added to each spot of cells for 30 minutes (room temperature): 1 μg/mL anti–HSV-2 ICP8 and mouse IgG2a isotype control; 5 μg/mL anti-HSV ICP5 mAb (IgG1 isotype; Virusys), anti–HSV-2 glycoprotein D (gD) mAb (IgG1 isotype; Biodesign International, Saco, ME), anti–HSV-1 and -2 envelope glycoprotein mAb (IgG1 isotype; United States Biological, Swampscott, MA), and mouse IgG1 isotype control (R&D Systems). Cells were washed 5 times with 1 × DPBS before 25 μL of an 8 μg/mL solution of peroxidase-conjugated AffiniPure [Heavy + Light chain] Donkey Anti-Mouse IgG (H+L, Jackson ImmunoResearch Labs) was added and the slides incubated for 30 minutes (room temperature). After washing, bound peroxidase was then labeled with fluorescein tyramide, diluted 1:50 in the amplification diluent (TSA Fluorescein System) for 4 minutes (room temperature; 25 μL per spot). The TSA (Tyramide Signal Amplification) Fluorescein System amplifies signals allowing sensitive visualization of proteins by microscopy that are difficult to detect by other staining methods.³³  Cells were washed 5 times with 1 × DPBS. Nuclei were stained with DAPI (4′, 6′-diamidino-2-phenylindole) (1.75 ng/mL, 25 μL per spot; Molecular Probes, Eugene, OR) for 2 minutes (room temperature) before the last 5 washes with 1 × DPBS. Slides were mounted in Aqua-Poly/Mount (Polysciences, Warrington, PA) and were examined using an Olympus AX70 epifluorescence microscope (Universal Imaging, West Chester, PA) with Plan Apo 60×/1.40 oil, DIC H, α/0.17, WD, 0.21 lenses (Nikon, Tokyo, Japan). The images were acquired using Image-Pro Plus 5.1 (MediaCybernetics, San Diego, CA) and analyzed using Photoshop 6.0 (Adobe Systems, San Jose, CA).

Untreated DCs and DCs exposed to infectious or UV-treated HSV-2 were stained with annexin V–FITC and propidium iodide (PI) (BD Biosciences Pharmingen). Cells were washed twice with cold 1 × DPBS and then resuspended in 1 × binding buffer (BD Biosciences Pharmingen) at a concentration of 1 × 10⁶/mL. A total of 100 μL of the solution (1 × 10⁵ cells) was left unstained or incubated with annexin V–FITC 1:20 and PI 1:20 for 15 minutes at room temperature in the dark. A total of 400 μL of 1 × binding buffer was added to each tube, and samples were analyzed by flow cytometry within 1 hour of labeling using a BD FACSCalibur.

Mixed leukocyte reactions (MLRs) were set up using human T cells.³⁴  Immature moDCs (exposed or not to live versus UV-treated HSV-2) were collected after 4 hours of reculture and viable cells recounted and resuspended in fresh medium. Human T cells were prepared from leukopacks by depleting CD11b⁺CD14⁺CD16⁺HLA-DR⁺ cells (using Miltenyi MACS beads). T-cell purity was typically more than 95% as verified by fluorescence-activated cell sorter (FACS) analysis of cells stained with mAbs against CD3 versus HLA-DR, CD11b, CD14, CD16, or CD20. Triplicate cultures of 2 × 10⁵ T cells mixed with different doses of macaque or allogeneic human DCs (input doses normalized as viable large cells/DCs) were set up in 96 well flat-bottomed trays (200 μL final). Controls of DCs (highest dose) and T cells cultured in medium alone were included. The plate was incubated for 5 days and then pulsed with tritiated thymidine (³H-TdR; 1 μCi [2.7 × 10^–17 Bq] per well) for 8 hours before being harvested using a cell harvester (Brander, Gaithersburg, MD). The amount of ³H-TdR incorporated by the cells was measured using the Liquid Scintillation & Luminescence Counter, 1450 microbeta Wallac Jet (PerkinElmer Life and Analytical Sciences, Boston, MA).

Cell-free tissue culture supernatants were collected from the MLRs (highest DC/T cell ratio) and the recultured DCs, spun at 1000 g for 3 minutes, and stored at –20°C prior to testing. Cytokine and chemokine levels were measured using the Beadlyte Human 22-plex Detection System according to the manufacturer's instructions (Upstate Biotechnology, Lake Placid, NY), washing the plates between steps with a MultiScreen Vacuum Manifold (Millipore, Billerica, MA). The plate was read on a Luminex 100 Instrument (Luminex, Austin, TX). Data were analyzed using STarStation software (Applied Cytometry System, Sacramento, CA). IL-12 levels were also monitored in the DC supernatants using the macaque-reactive IL-12 (p70 and p40) enzyme-linked immunosorbent assay (ELISA) (BioSource International, Camarillo, CA), and interferon-α (IFN-α) levels were measured using the monkey–cross-reactive Human IFN-α ELISA Kit (BioSource), according to the manufacturer's instructions.

Blood from healthy, SHIV162P-infected animals was used as a source of SIV-primed T cells¹⁵  to monitor the presentation of SIV antigens by autologous DCs.^15,31,34  Peripheral blood leukocytes were separated into CD14^– and CD14⁺ fractions,³¹  and the CD14⁺ monocytes were cultured in GM-CSF/IL-4 to generate immature DCs (as described under “Generation of immature moDCs”). CD14^– cells were cultured in R1 (at least 2 × 10⁷/mL) with 5 mL fresh R1 being added to the CD14^– cultures every 2 days when the DCs were fed with cytokines. On day 6, the DCs were collected and infected with 5 pfu per cell of HSV-2 or not (as described under “HSV-2 infection”). Medium-treated versus HSV-2–infected DCs were then recultured (2.5 × 10⁵ DCs in 250 μL of medium with GM-CSF/IL-4 in a 48-well plate) for 4 hours at 37°C in the presence or absence of aldrithiol 2 (AT-2)–inactivated SIV E11S (30 ng p27 per 10⁵ DCs).^31,33,35,36  The DCs were collected 4 hours later, washed 5 times in medium, and the viable large cells recounted. Coincidently, the CD14^– cells were collected and the T cells purified by depleting HLA-DR⁺CD11b⁺CD16⁺ cells (Miltenyi MACS system).

A range of doses of medium or AT-2 SIV-pulsed HSV-2–infected and uninfected DCs were plated with 2 × 10⁵ autologous T cells in an anti-IFN-γ–coated enzyme-linked immunospot (ELISPOT) plate (triplicates per condition).¹⁵  T-cell and DC controls cultured alone were included. For each experiment, 2.5 × 10³ DCs and 5 × 10⁴ T cells were cultured with (or without) 1 μg/mL phytohemagglutinin-P (PHA-P; Sigma Chemical) versus medium to control for T-cell and DC functionality. Prior to developing the ELISPOT plates after overnight incubation,¹⁵  cell-free supernatants were collected. Supernatants were analyzed immediately or after storage at –20°C for the presence of additional soluble factors using the Luminex assay (as described under “Luminex and ELISA analysis of cytokines and chemokines”). ELISPOT plates were counted using an AID ELISPOT reader (Cell Technology, Columbia, MD) using once optimized settings.

HSV-2–exposed immature macaque DCs were analyzed immediately or after reculture for the expression of viral proteins (Figure 1). Within 4 hours of reculture, we observed cytoplasmic expression of the major HSV-2 capsid protein (ICP5) and gD as well as the products of the env gene (Figure 1A). The HSV-2 single-strand DNA binding protein, ICP8, was detected after overnight reculture of HSV-2–exposed DCs (Figure 1A) but not earlier (not shown). ICP8 was predominantly expressed in a nuclear pattern (Figure 1A-B), with points of more intense staining sometimes at the periphery of the nucleus. No HSV-2 proteins were detected immediately following the virus pulse (not shown), and all IgG controls were negative (Figure 1A, lower row).

Monitoring ICP8 expression as a stringent measure of infection, flow cytometry was used to quantify the numbers of ICP8⁺ DCs (Figure 1C-D). Unlike medium-treated DCs (less than 1% ICP8⁺), ICP8 expression was evident after infection and this increased with time (Figure 1C). This result was confirmed in 15 experiments using cells from 7 different animals (Figure 1D), averaging 4.36% ICP8⁺ (range, 1.43%-8.11%) after 4 hours and 14.69% ICP8⁺ (range, 6.37%-34.5%) after overnight reculture. HSV-2–exposed human moDCs showed similar ICP8 expression (Figure 1D), typically in a nuclear localization (not shown). Thus, exposure of immature macaque DCs to HSV-2 resulted in the increased expression of HSV-2 proteins over time, indicative of viral infection—not uptake of virus particles.

HSV infection leads to apoptosis of DCs from other species.^26,27,37  Microscopic examination of HSV-2–infected macaque DCs suggested that these cells were unhealthy. FACS data revealed that many of the HSV-2–exposed DCs, especially the ICP8⁺ cells, were smaller in size, based on their forward scatter profiles (Figure 1C), further indicative of poor cell health. There was a considerable reduction in the numbers of trypan blue–negative viable cells recovered after overnight culture (48.6% uninfected versus 22.13% infected; Figure 2A).

Cells were also stained with annexin V/PI to identify apoptotic and necrotic cells. Increases in the percentages of early apoptotic (annexin V–positive PI-negative; 58.9% versus 9.49%) as well as late apoptotic/necrotic (annexin V–positive PI-positive; 28.6% versus 10.1%) cells were evident in the HSV-2–infected DCs (versus medium-treated controls; Figure 2B). There was a slight increase in the percentage of apoptotic cells after 4 hours of reculture, but this was greater after overnight incubation of HSV-2–infected macaque DCs (average of 45.72% versus 16.66% in HSV-2–infected versus uninfected cells; Figure 2C). The percentages of cells in late apoptosis/necrosis increased even further over time (average of 82.15% versus 30.61% in HSV-2–infected versus uninfected cells; Figure 2C). This was due to infection, because exposure to the same amount of UV-inactivated HSV-2 did not induce DC death (similar to medium-exposed controls). This mirrors results for human DCs (Figure 2C).

Initially, we observed that HLA-DR and CD86 expression were reduced in the HSV-2–infected macaque DCs after overnight reculture but not after 4 hours of reculture (Figure 3A-B). More extensive examination revealed that HLA-DR, CD40, CD80, CD83, and CD86 were decreased on cells expressing the highest levels of ICP8 (defined as ICP8⁺; Figure 3C). These markers were largely unaffected or even slightly elevated on ICP8^– cells within the infected macaque DC cultures (uninfected or low-level infection) compared with the medium-cultured controls. Interestingly, cells from 3 animals infected with only 1 pfu of HSV-2 per cell (average, 1.66% ICP8⁺) exhibited increased levels of several of these surface markers on the ICP8^– cells, but the levels of costimulatory molecule expression were still lower in the ICP8⁺ fraction (Figure 3C). These cells were clearly far healthier than cells exposed to 5 pfu of HSV-2 per cell (microscopic examination and forward scatter profile; not shown). We confirmed that the down-modulation of the costimulatory molecules requires infectious HSV-2, that exposure to UV-treated HSV-2 tended to increase the expression of these markers (like ICP8^– cells in the low-level infected cultures), and that these responses reflected those seen in human moDCs (Figure 3D).

As a functional measure of DC activity after exposure to infectious HSV-2 we evaluated the production of certain soluble factors over time. Infectious (not UV-treated) HSV-2 induced IL-6, tumor necrosis factor-α (TNF-α), macrophage inflammatory protein-1α (MIP-1α) (CCL3), and RANTES (regulated on activation normal T cells expressed and secreted) (CCL5) secretion by macaque DCs during the overnight reculture (Figure 4). UV-treated HSV-2 induced (at best) low-level IL-6, CCL3, and CCL5 release by macaque DCs. Some IL-6, TNF-α, and CCL3 were detected within 4 hours of infection, with the TNF-α response being maintained after overnight reculture. Contrastingly, HSV-2 induced little IL-12 p40 secretion by macaque DCs (Figure 4), and this was verified by an IL-12 ELISA that detects the p40 and p70 subunits (not shown). IL-8 (CXCL8) produced by macaque DCs was reduced upon HSV-2 infection. Unlike HSV-1 infection of human moDCs,³⁸  we did not detect any increase in IFN-α release by HSV-2–infected macaque moDCs (not shown). Much like monkey DCs, HSV-2 infection induced human DCs to secrete more TNF-α (10.8 ± 1.8 versus 133.0 ± 39.9), negligible (overall less) IL-12 (111.8 ± 36.51 versus 35.4 ± 11.4), and consistently less CXCL8 (1191.7 ± 179.2 versus 727.1 ± 201.4; mean pg/mL ± SEM after overnight reculture, 3 uninfected versus HSV-2–infected DC donors). In 2 of the 3 human donors, infectious HSV-2 induced only low-level IL-6 release (about 1.2-fold more than medium-treated DCs), and this was evident only after 4 to 6 hours of reculture. HSV-2 did not induce CCL3 and CCL5 production by human DCs within this period. In 3 of 3 human DCs and 13 of 21 macaque DCs tested, MCP-1 (CCL2) was decreased after HSV-2 infection (not shown).

Because HSV-2 infection modulated the expression of costimulatory molecules and soluble factors by macaque DCs, we investigated whether this influenced their ability to stimulate an MLR.³⁴  Because HSV-2 infection resulted in dramatic cell death after overnight reculture, we tested the immunostimulatory activity of HSV-2–infected versus uninfected DCs after 4 hours of reculture (when cells were still healthy; Figure 2).

Although the same numbers of viable DCs were compared, HSV-2–infected immature DCs stimulated considerably less T-cell proliferation than medium- or UV-HSV-2–exposed DCs (Figure 5). Representative responses induced by macaque versus human DCs are shown (Figure 5A). Most assays were performed using fresh (unfixed) DCs (Figure 5A-B). Thus, the lower proliferative responses activated by infected DCs might simply have reflected the loss of DCs (and, hence, immunostimulatory activity) due to virus-induced DC death during the MLR cultures. Therefore, we compared PFA-fixed and unfixed HSV-2–infected and uninfected macaque DCs. The overall levels of T-cell proliferation were reduced when the DCs were PFA-fixed (not shown), as would be expected because the fixed DCs could not mature further in culture (upon T-cell contact). However, the difference between the magnitudes of the responses induced by HSV-2–infected and uninfected DCs persisted (Figure 5C).

Luminex analyses revealed consistently lower responses in MLRs stimulated by HSV-2–infected macaque (Figure 6) and human (not shown) DCs. Although IL-2 levels were comparable after 5 days, examination after 1 day of culture confirmed that more IL-2 was produced earlier in the cultures stimulated by uninfected DCs, coinciding with greater T-cell proliferation (Figure 5). All other factors were highest in the 5-day cultures (not shown), and the levels of factors released in cultures stimulated by HSV-2–infected DCs were consistently lower than those induced by uninfected DCs (Figure 6). Comparable patterns (of lower magnitude) were seen in the supernatants collected from the cultures set up with PFA-fixed DCs (not shown). Comparing medium-, UV-treated HSV-2–, and live HSV-2–exposed DCs side by side, we verified that these decreased responses were dependent upon DC exposure to infectious HSV-2 (not shown).

To extend our examination of the impact of HSV-2 infection on DC functions and to begin to understand how HSV-2 might impact SIV infection, we measured the capacity of HSV-2–infected moDCs to stimulate SIV-specific T cells. We achieved this by using DCs and T cells from healthy infected macaques and chemically inactivated AT-2 SIV as the antigen.^15,31,34 

Preliminary flow cytometric analyses using fluorescently labeled viruses³⁹  confirmed that comparable numbers of HSV-2–infected DCs captured SIV compared with their uninfected counterparts (not shown; 30 ng p27 per 10⁵ DCs as used for the antigen stimulation assays). Despite this, AT-2 SIV-loaded HSV-2–infected DCs induced lower levels of SIV-specific IFN-γ release compared with uninfected DCs (Figure 7A,C). The reduced stimulatory function of the HSV-infected DCs was less apparent (but detected) in PHA-stimulated DC–T cell mixtures (Figure 7B-C). Luminex analysis of the supernatants revealed that the low levels of SIV-specific TNF-α, CCL3, and CCL5 produced by cultures of healthy uninfected DCs and T cells were not detected in the HSV-2–infected DC–T cell mixtures (Figure 7D). PHA-induced TNF-α, CCL3, and CCL5 secretion was also lower in the HSV-2–infected DC–T cell cultures.

Epidemiologic studies demonstrated that HSV-2 increases the frequency of HIV-1 transmission by 2- to 4-fold.¹⁹  Although mucosal epithelial cells (mainly keratinocytes) and sensory neurons are primary targets for HSV infection,^40-42  DCs can become infected by HSV and their function down-modulated as a result.^16,17,23,37  In addition to herpetic lesions affording HIV direct access to the blood and the activated leukocytes recruited to the lesions, we hypothesize that HSV-2 infection of DCs compromises their innate and adaptive activities to further exacerbate HIV transmission. Exploring ways to use the macaque system to study the spread of HIV in the face of sexually transmitted diseases (STDs), we provide the first evidence that HSV-2 infection of macaque DCs modifies their capacity to stimulate strong immunity, underscoring the potential for this animal model.

Similar to reports on murine³⁷  and human^18,23,27,38  DCs along with the human data herein, immature macaque DCs can be infected by HSV-2 (Figure 1). Discrepancies between the kinetics of infection and protein expression in each system probably reflect differences in the DCs used (bone marrow–derived DCs versus moDCs versus Langerhans cells in different species), the virus isolates examined (HSV-1 versus HSV-2), the different assays employed to monitor infection, and the time points after infection when the cells were analyzed. However, the predominantly nuclear localization of ICP8 in infected macaque DCs agrees with the fact that the single-strand DNA binding protein ICP8 is important for HSV-2 replication, accumulating in replication complexes within the nuclei.^43,44 

In all species, infectious HSV is cytotoxic and induces apoptosis of DCs (Figure 2).^26,27,37,38  Inconsistencies in the numbers of infected macaque DCs (up to 35% ICP8⁺) versus the numbers of apoptotic (about 46%) or apoptotic/necrotic (about 82%) cells suggest that the methods employed to detect infection underestimated infected DC numbers. Alternatively, this might reflect bystander killing of noninfected DCs within the HSV-2–infected macaque DC cultures. HSV triggers apoptosis in immature human DCs by a caspase-dependent mechanism,^26,27  but the mechanism of HSV-2–driven apoptosis of macaque DCs was not examined. TNF-α released in response to HSV by macaque (Figure 4) and human DCs²⁶  could contribute to TNF-α–mediated apoptosis of infected DCs and bystander noninfected DCs that might also be mediated via a caspase-dependent mechanism.⁴⁵ 

HSV infection of murine,³⁵  human (Figures 3, 4, 5),^{16-18,23,27,38}  and macaque DCs (Figures 3, 4, 5, 6, 7) also caused considerable modifications of their innate and adaptive functions. The earliest reports revealed that HSV infection of immature^16,37  and mature¹⁷  DCs dampened their function by down-regulating expression of surface molecules important in DC–T-cell communication and DC immunostimulatory capacities. HSV infection even diminished the responses of immature human DCs to maturation stimuli.¹⁶  HSV-2 infection of macaque and human DCs similarly coincided with reduced HLA-DR, CD40, CD80, CD83, and CD86 expression. Notably, during early or low-level infection, there was less impact on the expression of these molecules. In fact, levels were often increased on cells expressing little HSV-2 protein (apparent in less aggressive infections after exposure to a low inoculum) or after exposure to UV-treated HSV-2.

Adding to human DC data,²⁶  HSV-2–infected macaque DCs secreted increased levels of IL-6, TNF-α, CCL3, and CCL5 (Figure 4) but little IL-12 or IFN-α. Type I IFN released by HSV-1–infected human DCs might be important in the establishment of immune responses through the induction of maturation of bystander noninfected DCs.³⁸  It is possible during the early stages of infection, when DCs are exposed to smaller amounts of infectious virus and/or noninfectious particles in the infecting inoculum, that at least low-level HSV-2–specific responses could be stimulated by the maturing DCs. IL-6 plays a significant role in limiting DC maturation.⁴⁶  Even low-level IL-6 production combined with the paucity of type I IFN and IL-12 release by HSV-2–infected macaque (and human) DCs would favor the induction of less robust effector immune responses, allowing the onset of HSV-2 infection and potentially impacting responses against other antigens.

Predictably, HSV-2–infected macaque DCs possessed a generalized impairment of their immunostimulatory function (Figures 5, 6, 7), supporting evidence from other species (Figure 5A).^16,23,37,38  This was evident at the level of T-cell proliferation and the magnitudes of cytokine and chemokine responses in DC–T cell cocultures. Decreased immunostimulatory activity was evident in macaque and human DCs within 4 hours of HSV-2 infection when membrane phenotypic changes were negligible but functional modifications were already apparent. This emphasizes the profound and rapid impact HSV-2 infection has on both macaque and human DCs. The most notable differences observed between human and macaque DCs were the kinetics of some of the cytokine responses and the lack of detectable CCL3 and CCL5 release by HSV-2–infected human DCs within the time frame examined. Importantly, however, DCs from both species manifested similarly impaired immunostimulatory function. Moreover, while the release of soluble factors was lower in HSV-2–infected (human and macaque) DC–T cell mixtures, there was no shift from T helper-1 (Th1) to Th2 type responses. Therefore, although responses were less robust, Th1 responses would still be able to be induced by HSV-2–infected DCs to ultimately control infection. This parallels the delayed or suboptimal responses described in the human system, which can be offset by uninfected DCs within an infected population cross-presenting HSV antigens.^23,27,47  Innate activation of DCs through TLR stimulation by HSV also likely contributes to controlling HSV infection.⁴⁸ 

Such compromised immunostimulatory function of macaque DCs also extended to SIV-specific responses (Figure 7). Even though HSV-2–infected macaque DCs captured SIV, their ability to stimulate SIV-specific IFN-γ, TNF-α, CCL3, and CCL5 release was diminished, much like the decreased antigen-specific responses induced by HSV-1–infected human DCs.²³  This supports the hypothesis that HSV-2–infected immature DCs stimulate poor SIV-specific immune responses and may have important implications for HIV transmission.

This represents the first report demonstrating the modified immune functions of HSV-2–infected macaque DCs, validating the similarities between macaques and humans, and that HSV-2 infection compromises the capacity of DCs to stimulate SIV-specific responses. Whether this influences how HIV/SIV modifies immature moDCs^12,13,49,50  remains to be determined and can now be explored in vivo in the macaque system. Being able to model HSV-2 infection in macaques, we are studying the impact of HSV-2 infection on transmission and testing how HSV-2 alters the efficacy of microbicide or vaccine strategies in preventing HIV spread. Identifying approaches that function effectively in the face of other infections, especially those that down-modulate immune function, will be critical in preventing the spread of HIV and other STDs.

The authors thank Dr Anthony Cunningham, Lidija Bosnjak, and members of the laboratory of Dr Melissa Pope for critical discussions on the manuscript. Additional thanks go to Kanani Titchen and Jose Romero (in Dr David Phillips' laboratory at the Population Council) for additional technical support.

The opinions expressed herein are those of the authors and do not necessarily reflect the views of the U.S. Agency for International Development.