In 1986, Walker et al 1 demonstrated that CD8+ T cells can inhibit HIV-1 replication in vitro and proposed the importance of a soluble CD8+ T-cell-derived noncytolytic factor subsequently dubbed CD8 antiviral factor (CAF). These and other investigators have since contributed a growing body of work to characterize this factor or group of factors. The search for such factors produced seminal work by Cocchi et al, 2 demonstrating the ability of CD8+ T-cell-secreted β-chemokines to inhibit the replication of macrophage-tropic viral isolates. While this work has led to important insights into HIV pathogenesis and provided the foundation for new antiretroviral strategies, the inability of β-chemokines to perturb the replication of T-cell line–tropic (CXCR4 utilizing) viral isolates has clearly marked them as substances distinct from CAF. Other cytokines and chemokines including interleukin (IL)-16 3 and macrophage-derived chemokine 4 have also been considered as potential CAF candidates. However, to date no one factor or group of factors has been demonstrated to fulfill all the proposed characteristics attributed to CAF or accounted for all of the anti-HIV activity of CAF. 5–8 The identity of this factor has accordingly remained elusive. Nonetheless, the effort to discover the identity of CAF has continued to catalyze the discovery of novel factors with anti-HIV activity. 9,10
We recently demonstrated that human α-defensins 1, 2, and 3 contribute to the anti-HIV-1 activity of CAF. 11 In this work we concluded that α-defensins account for much of the anti-HIV activity of CAF not attributable to β-chemokines. Our laboratory has previously characterized the HIV-1-inhibitory activity of CD8+ T cells derived from a number of HIV-seropositive long-term nonprogressors (LTNPs) 12 (and unpublished observations). Our recent experiments 11 focused on purified CD8+ T cells from such LTNPs. In addition, CD8+ T cells from HIV-seronegative subjects and HIV-seropositive subjects with progressive immunodeficiency (progressors) were analyzed. These CD8+ T cells were stimulated in serum-free conditions with anti-CD3 and anti-CD28 antibodies, recombinant IL-2, phytohemagglutinin, and equal numbers of irradiated allogeneic peripheral blood mononuclear cells (PBMCs). A number of independent groups have also sought to induce the production of soluble CD8+ T-cell-derived non-cytolytic factors by subjecting CD8+ T cells to a stimulation protocol including irradiated allogeneic PBMCs or so-called feeder cells. 5,12–16
Subsequent efforts by our group have been aimed at precisely defining the subpopulation of CD8+ T cells that produce α-defensins. We have recently determined that in the absence of irradiated allogeneic PBMCs, stimulated CD8+ T-cell supernatants do not contain α-defensins. Although it could be argued that an allogeneic stimulus is a prerequisite for α-defensin production by CD8+ T cells, we considered it more likely that α-defensins were instead derived from cells residing within the feeder cell population. While we have not conclusively eliminated other α-defensin-producing mononuclear cell populations 17,18 as contributing sources, our data appear to implicate residual granulocytes within PBMC fractions as the major source of α-defensins. Furthermore, we have discovered that experimental conditions routinely used in immunofluorescence studies are responsible for the artifactual visualization of α-defensins within cells naturally devoid of these proteins.
Further work in our laboratory has extended and solidified the anti-HIV-1 activity of α-defensins 11 (and unpublished findings) and recent work by independent investigators 19 has also confirmed this activity. Our prior work made use of α-defensin-specific immunodepletion or neutralization strategies to demonstrate the elimination of the majority of anti-HIV-1 activity, not attributable to β-chemokines, from stimulated CD8+ T-cell supernatants. 11 This information and our new data therefore imply the absence or a limited quantity of a CD8+ T-cell-derived noncytolytic anti-HIV-1 factor, other than β-chemokines, in the context of our experimental conditions.
Blood Collection and Leukocyte Enrichment
Blood was collected from HIV-seronegative donors in ethylenediamine tetra-acetic acid (EDTA)-anticoagulated Vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ); samples were processed for analysis within 2 hours of blood collection. Alternatively, leukapheresis units were obtained from HIV-seronegative donors via the New York Blood Center; units were processed for analysis within 24 hours of leukapheresis and used in experiments without prior cryopreservation. PBMCs were isolated by standard Ficoll gradient density centrifugation (Lymphocyte Separation Medium; Cellgro, Herndon, VA). In some experiments freshly collected blood was enriched for leukocytes by elimination of erythrocytes with ammonium chloride lysis (ACL) buffer 20 or alternatively by centrifugation at 1000 ×g for 10 minutes to obtain a leukocyte buffy coat.
Irradiated allogeneic PBMCs used to stimulate CD8+ T cells were derived from PBMCs that had been obtained from HIV-seronegative leukapheresis units, irradiated (9000 rad) and cryopreserved until subsequent use. Leukapheresis units were obtained from LTNP HIV-seropositive donors and processed on the day of collection. PBMCs from LTNP HIV-seropositive individuals were derived via Ficoll gradient centrifugation and cryopreserved until subsequent use.
In experiments wherein α-defensin gene expression was to be assessed, blood was collected from HIV-seronegative donors in Vacutainer CPT tubes with sodium citrate (Becton Dickinson) and processed (as per the manufacturer’s instructions) to obtain PBMCs. PBMCs were isolated for downstream applications within 1 hour of blood collection.
Preparation of Enriched Mononuclear Cell Populations
T cells, CD8+ T cells, CD4+ T cells, or B cells were individually isolated from PBMCs by positive selection using CD2+-, CD8+-, CD4+-, or CD19+-targeted immunomagnetic beads (Dynal, Oslo, Norway), respectively, as per the manufacturer’s protocol. Beads were subsequently removed from the enriched cells using Detach-A-Bead (Dynal). The purity of the cells obtained by this procedure was typically >95% for all T cells and >92% for B cells as judged by flow cytometry. Monocytes were isolated from PBMCs by negative selection using immunomagnetic beads (Dynal). The purity of the cells obtained by this procedure was typically >50% as judged by flow cytometry. In some experiments, CD15+-targeted immunomagnetic beads (Dynal) were used to deplete granulocytes from the starting PBMC population, as per the manufacturer’s protocol. Depletion of granulocytes from PBMCs was verified by cell scatter characteristics and CD15- or CD66b-specific staining (as described below).
HeLa cells (American Type Culture Collection, Manassas, VA) were cultured in Dulbecco modified Eagle medium (Cellgro, Herndon, VA). HL60 cells (American Type Culture Collection, Manassas) were cultured in RPMI (Roswell Park Memorial Institute) 1640 medium (Cellgro). All media was supplemented with 10% fetal calf serum (FCS), 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid (HEPES) (10 m M), penicillin (100 U/mL), and streptomycin (100 μg/mL).
CD8+ T-Cell and Mononuclear Cell Stimulation
CD8+ T-cell stimulation was carried out as previously reported. 11 Briefly, isolated CD8+ T cells were stimulated for 3 days at 37° C in a humidified CO2 incubator with anti-CD3 (12F6; 0.1 μg/mL) (a gift from J. Wong of Massachusetts General Hospital) and anti-CD28 (2.5 μg/mL; Becton Dickinson, San Diego, CA) antibodies, together with an equal number of allogeneic irradiated PBMCs in serum-free RPMI 1640 medium (Cellgro, Herndon, VA), supplemented with phytohemagglutinin (2 μg/mL), recombinant IL-2 (20 U/mL), HEPES (10 m M), penicillin (100 U/mL), and streptomycin (100 μg/mL). Culture supernatants were collected and cleared by centrifugation at 3000 rpm for 5 minutes and were either used immediately or stored at −80°C until subsequent use. In subsequent experiments, whole PBMCs or T-cell, CD8+ T-cell, CD4+ T-cell, B-cell, and monocyte-enriched PBMCs were subjected to the same stimulation protocol in the absence of any allogeneic irradiated PBMCs. Identical experiments were performed with PBMC fractions that had been depleted of granulocytes.
Intracellular antigen specific antibodies: HNP1-3-specific DEF3 monoclonal antibody (IgG1 kappa; Serotech, Oxford, United Kingdom) was directly conjugated to Alexa-647 via an Alexa Fluor 647 Monoclonal Antibody Labeling Kit (Molecular Probes, Eugene, OR) as per the manufacturer’s instructions. An isotype-matched antibody (Mouse IgG1 kappa; Southern Biotech, Birmingham, AL) was similarly conjugated to Alexa-647 and used as an irrelevant control. Subsequent to conjugation and purification, fluorescence:protein ratios were determined to ensure equivalent labeling of both antibodies. Dose titration experiments were conducted to determine the concentration of antibody to be used for optimal sensitivity and specificity. In some experiments, HNP1-3-specific biotinylated D21 monoclonal antibody (HyCult Biotechnology, Uden, The Netherlands) was used. An isotype-matched antibody (Mouse IgG1-biotin; CalTag, Burlingame, CA) was used as an irrelevant control. Detection of both antibodies was accomplished by subsequent incubation with fluorescein isothiocyanate (FITC)-conjugated streptavidin (CalTag). Human myeloperoxidase and lactoferrin-specific antibodies (Combi-IC Monoclonal Antibodies; CalTag) were used in control experiments.
Cell surface antigen–specific antibodies: phycoerythrin-or FITC-conjugated CD3-, CD4-, CD8-, CD14-, CD15-, CD19-, and CD66b-specific monoclonal antibodies (BD Biosciences, San Jose, CA) were used for flow cytometric assessment of cell purity and in immunofluorescence studies as noted.
In some experiments (as indicated), cells were fixed in 4% formaldehyde in phosphate-buffered saline (PBS) (16% methanol-free formaldehyde; Polysciences, Inc., Warrington, PA) for 30 minutes at room temperature, with no additional permeabilization. In all other experiments, fixation and permeabilization of cells was performed according to the CalTag FIX&PERM protocol with minor modifications. Wash/resuspension buffer containing 0.5% bovine serum albumin and 10 m M sodium azide in PBS was used in all experiments. For each sample to be analyzed, 1 × 106 cells in 100 μL were pipetted into replicate 5-mL tubes. For staining of cell surface markers, an appropriate volume of conjugated antibody or isotype control was added to each tube. Each tube was mixed and incubated for 15 minutes at room temperature in the dark. Fixation medium (100 μL of Reagent A) was added to each tube and the tubes were incubated for a further 15 minutes at room temperature. Each tube was washed and then centrifuged for 5 minutes at 300 ×g. The supernatant was aspirated and the cell pellet resuspended by vortexing gently. Permeabilization medium (100 μL of Reagent B) was added to each tube concurrently with a conjugated intracellular antibody or iso-type control. The tubes were briefly vortexed, incubated at room temperature for 15–30 minutes, and then washed as before. Cell pellets were resuspended in 1% formaldehyde, stored at 4°C in the dark, and analyzed within 24 hours. Flow cytometric analysis was carried out using a dual-laser FACScalibur (BD Biosciences) and data were displayed using CELLQuest (BD Biosciences) software.
Samples of interest were also assessed by immunofluorescence microscopy. Cells were first counterstained using Hoechst 33342 (Molecular Probes), then affixed to microscopy slides using a cytocentrifuge. Prior to affixing cells, 1% FCS in PBS was applied to each slide. Cover slips were mounted using SlowFade Light antifade component A (Molecular Probes). Representative fluorescent cell images were acquired using a DeltaVision deconvolution microscopy system and the Resolve3D program (Applied Precision, Issaquah, WA). All images were captured at 60× magnification and with the same exposure time. Images were deconvolved, then analyzed using the Softworx program (Applied Precision). Output settings were scaled identically for all images shown.
Cell Mixing Experiments
In some immunofluorescence studies, ACL, buffy coat, or Ficoll-derived leukocyte fractions concurrently obtained from the same donors were mixed with roughly equal numbers of HeLa cells. HeLa cells were trypsinized, washed with complete media, and resuspended in PBS immediately prior to mixing with leukocytes. The resulting mixtures were resuspended in wash buffer prior to proceeding with fixation as described above. Cell scatter characteristics and α-defensin-specific immunofluorescence of leukocytes were identical in the absence or presence of HeLa cells. HeLa cells were easily distinguishable from leukocytes based on cell scatter characteristics.
Quantitative Real-Time RT-PCR
Quantitative real-time reverse transcriptase polymerase chain reaction (RT-PCR) was used to assess α-defensin (HNP1) gene expression in CD8+ T cells. CD8+ T cells were obtained by positive immunomagnetic selection (as described above) from Vacutainer CPT tube–derived PBMCs. Expression in residual PBMCs (collected after depletion of CD8+ T cells) and HL60 cells was also examined. RNA was isolated using the RNeasy Mini Kit (Qiagen, Valencia, CA) as per the manufacturer’s instructions. All samples were treated with on-column DNase digestion. Reverse transcription of RNA samples was carried out using Multiscribe Reverse Transcriptase (Applied Biosystems, Foster City, CA) as suggested by the manufacturer’s protocol. Parallel control reactions were set up with omission of the reverse transcriptase. Real-Time PCR was performed on an ABI 7700 Sequence Detection System (Applied Biosystems). HNP1-specific primers used were as follows: forward- 5′-GCTTGGCTCCAAAGCATCCA-3′ and reverse- 5′-TGCAGGTTCCATAGCGACGTTCTCC-3′. We designed and used a Taqman HNP1-specific probe, 5′-CATGGCCTGCTATTG - 3′, labeled at the 5′end with carboxyfluorescein (FAM) and the 3′ end with a nonfluorescent quencher (TaqMan MGB Probe, Applied Biosystems). PCR reactions were set up in 96-well plates in a final volume of 25 μL, containing 2.5 μL of template cDNA, TaqMan Universal PCR Master Mix (Applied Biosystems), 250 n M probe, and 900 n M of each primer. Eukaryotic elongation factor 2 (EEF2) gene expression was used as an endogenous control for each sample, to allow normalization and comparison between samples. Parallel reactions were therefore set up for each sample using TaqMan Universal PCR Master Mix and EEF2-specific primers and probe (Assays-On-Demand, Applied Biosystems). All reactions were set up in duplicate and with matching RT-negative controls. Thermal cycler conditions were as follows: 50°C for 2 minutes and 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 62°C for 1 minute. Plasmid-based standards, containing the HNP1 and EEF2 cDNA target sequences of interest, were used to generate standard curves and allow quantitation.
Ciphergen ProteinChip Analysis
We used surface enhanced laser description/ionization time of flight mass spectrometry (SELDI-TOF-MS) analysis and Ciphergen ProteinChip Arrays (Ciphergen Biosystems, Fremont, CA) to detect protein/peptide species present in culture supernatants, as previously described. 11 Culture supernatant (100 μL) was mixed with an equal volume of binding buffer (100 m M NaAc, pH 4.5; 0.2% Triton X-100 in PBS), and then applied onto a weak cation exchanger (WCX2) array and incubated at 4°C overnight with constant horizontal shaking. Unbound proteins/peptides were removed by washing with a buffer (50 m M NaAc, pH 4.5; 0.1% Triton X-100 in PBS) 3 times for 5 minutes each under identical conditions. The WCX2 arrays were then removed and rinsed in 1 m M HEPES (pH 4.5) for 30 seconds and air-dried before a saturated solution of 3,5-dimethoxy-4-hydroxycinnamic acid (SPA) in 50% acetonitrile and 0.5% trifluoroacetic acid was added to the chip surface. The arrays were analyzed with the Ciphergen ProteinChip Reader (model PBS II). In SELDI-TOF-MS analysis, a nitrogen laser (337 nm) desorbs the protein/SPA mixture from the array surface, enabling the detection of the proteins/peptides captured by the array. The mass spectra of proteins/peptides were generated using an average of 80 laser shots at a laser intensity of 220–240 arbitrary units. The mass to charge ratio (m/z) of each of the proteins/peptides captured on the array surface was determined according to externally calibrated standards.
Detection of α-Defensins in the Cytoplasm of Mononuclear Cells
The conclusion that stimulated CD8+ T cells produce α-defensins was bolstered by experiments employing immunofluorescence-based detection of intracellular α-defensins in a subpopulation of activated CD8+ T cells. 11 These supporting data appeared to be entirely consistent with a growing body of work indicating that α-defensin expression was not restricted to neutrophils. Indeed, in addition to expression in neutrophils, 21 and the myeloid-derived cell line HL60, 22,23 α-defensin expression has been reported in mononuclear cell populations including monocytes/macrophages, B cells, γδ T cells, and natural killer cells, 17,18 as well as in renal cell carcinoma malignant epithelial cells and renal cell carcinoma cell lines. 24
In a preliminary attempt to further define and characterize the CD8+ T-cell subpopulations we detected in our first study, we undertook a broader analysis of unstimulated mono-nuclear cell populations using flow cytometric analysis and immunofluorescence microscopy. We initially examined bulk leukocytes derived from whole blood by ACL of erythrocytes. Flow cytometric analysis revealed a differential pattern of expression (Fig. 1A). The highest level of expression was noted in granulocytes, followed by monocytes, and lastly by lymphocytes. Although this was somewhat in keeping with our previous findings, the detection of α-defensin in all leukocyte populations suggested the potential for artifactual staining. The visualization of HNP1-3 within neutrophils by immunohistochemical staining was first described in early work by Ganz et al. 21 Therein, cells were fixed with formalin without any subsequent permeabilization. Our previously published experiments 11 similarly used formalin fixation (1% formaldehyde), without any additional permeabilization prior to intracellular staining. In an effort to ensure adequate fixation we subsequently used 4% formaldehyde (as described in the “Methods” section). However, ensuing control experiments aimed at detecting myeloperoxidase and lactoferrin in leukocytes indicated that formaldehyde fixation alone did not allow optimal detection of intracellular antigens (data not shown). We therefore repeated these experiments with a permeabilization procedure concurrent with intracellular staining (CalTag FIX&PERM). Using this protocol, all leukocyte populations stained positive for relatively high levels of α-defensin (Fig. 1B). Visualization of intracellular α-defensins by immunofluorescence microscopy also suggested ubiquitous expression across leukocyte populations (data not shown). Similar results were obtained when staining was carried out with an alternate α-defensin-specific monoclonal antibody (D21).
Given the abundance of α-defensins in neutrophil azurophilic granules, 21,25,26 we reasoned that an in vitro process might trigger a spillage and uptake phenomenon, whereby α -defensins could exit the neutrophil, by degranulation or loss of cell membrane integrity, and subsequently be taken up by cells not natively producing α-defensins. ACL of erythrocytes to obtain whole leukocyte fractions can damage leukocyte and particularly granulocyte cell membranes. 27 We therefore conducted concurrent flow cytometric analysis of ACL-derived leukocytes, leukocyte buffy coat preparations, and PBMCs enriched by Ficoll gradient centrifugation. In a preliminary analysis, buffy coat–derived granulocytes stained more brightly, and lymphocytes and monocytes stained less brightly, relative to their counterparts in the ACL preparation, suggesting better retention of α-defensins within granulocytes in the absence of exposure to ACL buffer (data not shown). Ficoll-derived lymphocytes and monocytes stained variably among different donor PBMCs. We postulated that varying levels of granulocyte contamination might be responsible for the latter observation.
To further test this possibility, we conducted a series of mixing experiments utilizing the aforementioned leukocyte fractions and a number of cell lines determined to contain no detectable α-defensin. We reasoned that a cell line not expressing α-defensins could serve as an indicator for spillage and uptake of α-defensins. In the experiments shown, we incorporated HeLa cells into ACL, buffy coat, or Ficoll-derived leukocyte fractions concurrently obtained from the same donors. While no α-defensins could be detected in HeLa cells using flow cytometric analysis, seemingly abundant α-defensins were detectable in HeLa cells that had been mixed with ACL leukocytes prior to fixation and permeabilization (Figs. 2 and 3A). This was consistent across several experiments using leukocytes from several different donors. Independent experiments utilizing human embryonic kidney fibroblasts (293 cells) and a number of lymphoid-derived cell lines demonstrated a similar phenomenon (data not shown). Relatively lower levels of α-defensin were detected in HeLa cells mixed with buffy coat leukocytes (Figs. 2 and 3A). When HeLa cells mixed with Ficoll-derived PBMCs from the same donors were analyzed, even lower to negligible levels of α-defensin were detected (Figs. 2 and 3A). Consistent findings were noted when these cell mixtures were assessed by immunofluorescence microscopy (Fig. 4). Intriguingly, the level of α-defensin content in lymphocytes and monocytes within the different fractions (ACL leukocytes, buffy coat leukocytes, and Ficoll-derived PBMC) appeared to be directly related to levels in corresponding HeLa cells (Fig. 3A). Moreover, the level of α-defensin content in HeLa cells mixed with Ficoll-derived PBMCs was more pronounced when the source of the PBMCs was leukapheresis (Fig. 3B) units rather than freshly derived blood (Fig. 3A), with the former usually containing higher levels of contaminating granulocytes.
These results collectively suggested that α-defensins previously detected by immunofluorescence in CD8+ T cells 11 were in fact derived from a population, most likely contaminating granulocytes, residing within the PBMC population. To directly address this question, we removed most of residual granulocytes within PBMCs using immunomagnetic bead–based depletion and assessed their α-defensin content relative to nondepleted PBMCs. The success of granulocyte depletion of PBMCs was gauged by flow cytometric analysis. HeLa cells were mixed with each PBMC fraction. Lymphocytes (including CD8+ T cells) within depleted PBMCs contained little if any α-defensin (Figs. 3B and 5). In contrast, their counterparts within nondepleted PBMCs appeared to contain abundant α -defensin. Our ability to render lymphocytes virtually free of α-defensins was directly related to our capability to remove all traces of granulocyte contamination in the starting PBMC population. The same phenomenon was observed when we analyzed monocytes, as well as HeLa cells mixed with PBMCs (Figs. 3B and 5). Nonetheless, in some experiments negligible or minuscule amounts of α-defensin appeared to persist after CD15 depletion; it is unclear whether trace granulocyte contamination or another α-defensin-producing leukocyte contributed to this phenomenon.
Assessment of α-Defensin Gene Expression by Quantitative Real-Time RT-PCR
In parallel with immunofluorescence studies, we used quantitative real-time RT-PCR to analyze α-defensin gene expression in CD8+ T cells. RNA was obtained from 2 × 106 CD8+ T cells or 2 × 106 residual PBMCs (after CD8+ T-cell selection). RNA from 1 × 106 HL60 cells (a myeloid-derived cell line known to express α-defensins) served as a positive control. The abundance of α-defensin and EEF2 mRNA in each sample was determined in parallel. Quantitation of EEF2 mRNA was used to normalize α-defensin expression and allow comparison of samples. As expected, HL60 cells expressed abundant amounts of α-defensin mRNA (Fig. 6). No α-defensin expression could be detected in CD8+ T cells. In contrast, PBMCs obtained concurrently from the same donors did express α-defensin mRNA, though in small quantities relative to HL60 cells. These data appeared to corroborate our immunofluorescence findings suggesting that α-defensins visualized within lymphocytes were in fact an in vitro artifact.
In the Absence of Irradiated Allogeneic PBMCs, Stimulated CD8+ T-Cell Supernatants Do Not Contain α-Defensins
We previously described the detection of α-defensins in stimulated CD8+ T-cell culture supernatants derived from LTNP and HIV-seronegative donors. 11 In separate experiments we were unable to detect α-defensins in culture supernatants derived from several HIV-seropositive progressors’ CD8+ T cells. Follow-up experiments using CD8+ T cells from a separate set of HIV-seropositive progressors produced similar results. We were therefore reassured that α-defensin production was indeed specific to LTNP and HIV-seronegative donor CD8+ T cells. However, in light of our preliminary data from the aforementioned immunofluorescence studies, we proceeded with experiments wherein CD8+ T cells from LTNP or HIV-seronegative donors were stimulated in the presence or absence of irradiated allogeneic PBMCs. As expected, supernatants from cultured CD8+ T cells stimulated with our original protocol contained α-defensins (Fig. 7A). However, in the absence of irradiated allogeneic PBMCs in the culture system, no α-defensins were detectable in stimulated CD8+ T-cell culture supernatant (Fig. 7B). In parallel experiments, supernatants from cultured irradiated PBMCs stimulated in the absence of any heterologous CD8+ T cells contained α-defensins (Fig. 7C). These experiments further suggested that, in our experimental system, α-defensins were derived from a source residing within the irradiated allogeneic PBMC population.
Sources of α-Defensins Within PBMCs
To pursue the exact sources of α-defensins within PBMC fractions, we enriched individual cell populations from PBMCs by immunomagnetic bead–based selection and subjected them to the aforementioned stimulation conditions in the absence of any allogeneic exposure. In preliminary experiments, α-defensins were detected in the supernatants of CD4-, CD8-, CD14-, and CD19-enriched cells. Our ability to detect α-defensins in each specified cell fraction supernatant varied greatly among donor PBMCs. We reasoned that contaminating granulocytes might be contributing to these observations, as suggested by the immunofluorescence data. Accordingly, experiments were carried out wherein PBMCs were first treated with an anti-CD15 monoclonal antibody to eliminate residual granulocytes. Subsequent to granulocyte immunodepletion, lineage-specific immunoselection was carried out and the enriched mononuclear cell populations were stimulated in the absence of any allogeneic stimulus. The success of granulocyte depletion of PBMCs was gauged by flow cytometric analysis. α-Defensins were either barely or no longer detectable in the supernatants of cultured T- and B-cell-enriched PBMCs, respectively (Fig. 8). In addition, α-defensins were not detectable in CD4+ or CD8+ T cells derived from granulocyte-depleted PBMCs (Fig. 9). α-Defensins were variably detectable, however, in the supernatants of cultured monocyte-enriched PBMCs (Fig. 8). The latter finding might be explained by persistent granulocyte contamination and the relatively lower purity achieved by the negative selection method used to isolate monocytes (described in the “Methods” section). Alternatively, contamination with another α-defensin-producing leukocyte or production by monocytes themselves might have contributed to this observation. Taken together, these findings suggested that under our original experimental conditions even minor degrees of granulocyte contamination could result in the detection of α-defensins in the culture supernatant of enriched PBMC populations, including CD8+ T cells.
The assertion that α-defensins are produced by CD8+ T cells was bolstered by immunofluorescence studies demonstrating their presence within the cytoplasm of CD8+ T cells. 11 It is now apparent that in vitro conditions can promote the artifactual internalization of α-defensins by cells not natively producing these proteins, thus confounding our ability to define true α-defensin producer cells using immunofluorescence-based techniques. Independent studies, utilizing immunofluorescence to examine the subcellular localization of various protein transduction domain or cell penetrating peptides, have described a similar phenomenon. 28–30 Specifically, the cellular import and nuclear localization of HIV-1 Tat-derived peptide and polyarginine, 28 and HIV-1 Tat-derived peptide, HSV VP22-derived peptide, polylysine, or polyarginine fused to green fluorescent protein 29,30 appear to be induced by the fixation process. Strictly in the context of cells undergoing fixation with formaldehyde, the following conclusions could be drawn from our results: 1) ACL buffer exposure does contribute to the release of α-defensins from granulocytes; 2) in the absence of ACL buffer exposure, α-defensins could nevertheless be released into the extracellular milieu and subsequently be taken up by mononuclear cells natively producing little if any detectable α-defensin; and 3) the level of uptake of α-defensins by mononuclear cells is related to the relative quantities and integrity of granulocytes in the preparation. Most relevant to our earlier work, we have established that granulocyte contamination of Ficoll-derived PBMC preparations can result in the artifactual detection of α-defensins in mononuclear cells including CD8+ T cells. Analysis by quantitative real time RT-PCR also indicates that unstimulated CD8+ T cells do not express α-defensin mRNA.
Our current work also clearly establishes that in the absence of irradiated allogeneic PBMCs (feeder cells), stimulated CD8+ T-cell supernatants do not contain α-defensins. Allogeneic stimuli in the context of mixed lymphocyte reactions 31–34 or alloimmunization 35,36 have been previously demonstrated to elicit soluble noncytolytic anti-HIV-1 factors. It is therefore plausible that CD8+ T cells could be induced to produce α-defensins by a requisite allogeneic stimulus. Although such observations are intriguing, our data suggest that the true source of α-defensins lies within the feeder cell population. Our data are not conclusive, however, regarding all potential sources of α-defensins. Given the previously documented presence of α-defensin protein or mRNA in leukocyte sources other than myeloid-derived cells, 17,18 we cannot at this time conclude that α-defensins present in our stimulated CD8+ T-cell culture supernatants are exclusively derived from residual granulocytes present among irradiated allogeneic PBMCs. Indeed, despite granulocyte depletion, the negligible to minute levels of α-defensin occasionally detected by immunofluorescence and ProteinChip assays could be explained by persistent traces of granulocytes or another α-defensin-producing cell. During the preparation of this manuscript, Mackewicz et al 37 reported data consistent with our findings. Therein they report that in contrast to neutrophils, CD8+ T cells do not produce α-defensin. In keeping with previous reports, 17 they also note that α-defensins are expressed in monocytes. We are in the process of rigorously defining alternate sources of α-defensins within our experimental system. Moreover, any potential interplay between CD8+ T cells and the release of α-defensins derived from their source within the irradiated allogeneic PBMC population remains to be elucidated.
During the past 17 years the conditions used to induce and analyze CAF activity have been remarkably varied among laboratories. 38 The investigators who originally proposed the existence 1 and subsequently elucidated many of the characteristics of CAF have also generated and analyzed CD8+ T-cell-derived noncytolytic anti-HIV-1 activity via a variety of methods, 7,39 including stimulation with irradiated allogeneic PBMCs. 13 Beyond experimental conditions, the source of CD8+ T cells analyzed must also be taken into consideration. CD8+ T-cell-derived noncytolytic anti-HIV activity appears to be best elicited from certain asymptomatic HIV-seropositive subjects including LTNPs. 40,41 Such activity is also manifested by HIV-1 epitope-specific cytotoxic T lymphocytes. 42 Some have demonstrated that CD8+ T cells from exposed uninfected individuals 40,43,44 as well as vaccine-induced HIV-1-specific cytotoxic T cells from HIV-seronegative subjects 45 manifest a soluble noncytolytic inhibitory factor. In contrast, subjects with AIDS manifest strikingly lesser or absent activity. 41,46 In the context of an acute in vitro infection, seemingly incongruous results have been obtained with regards to the noncytolytic anti-HIV activity of CD8+ T cells from HIV-seronegative subjects. Whereas some studies have elicited significant activity from CD8+ T cells of HIV-seronegative subjects, 5,13,15,47,48 others have noted little or no such activity. 41,49 The diversity of methods and conclusions put forth by these and other investigators suggests that a number of different factors, alone or perhaps in various combinations, may possess noncytolytic antiretroviral activity.
Other laboratories have made headway in their search for and characterization of such anti-HIV factors by intently focusing on herpes virus Saimiri (HVS)-transformed CD8+ T-cell clones. 19,50–53 The characteristics of the noncytolytic antiretroviral activity elicited by this model appear to parallel those produced by mitogen-stimulated primary CD8+ T cells from some asymptomatic HIV-seropositive subjects. Nevertheless, it remains to be seen whether these are identical factors. Recent findings by Chang et al 19 are particularly relevant to our work, demonstrating that the anti-HIV factor derived from HVS-transformed CD8+ T cells is distinct from α-defensin-mediated anti-HIV activity.
We and other investigators have established the ability of human α-defensins to inhibit HIV-1 replication. 11,19 Moreover we have demonstrated that α-defensins account for the great majority of β-chemokine-independent antiretroviral activity in our CD8+ T-cell culture supernatants. 11 Taking into consideration our current findings, we are faced with the possibility that under our experimental conditions, such a soluble noncytolytic anti-HIV-1 factor is either not produced by CD8+ T cells or is produced in too small a quantity to be effective. Other investigators have demonstrated CAF activity using conditions that are to some extent analogous to ours, particularly with respect to the use of irradiated allogeneic PBMCs. 5,12–16 Given the modest differences in the experimental conditions employed among these investigators, we cannot speculate as to whether α-defensins or other factors with anti-HIV-1 activity were operative. With specific regard to CAF, given the panoply of methods used by investigators in the field, it remains to be seen whether independent groups are all describing the same factor or complex of factors. Undoubtedly, their endeavors will continue to precipitate the discovery of novel antiretroviral strategies and drive forward our understanding of HIV pathogenesis.
1. Walker CM, Moody DJ, Stites DP, et al. CD8+ lymphocytes can control HIV infection in vitro by suppressing virus replication. Science
2. Cocchi F, DeVico AL, Garzino-Demo A, et al. Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-suppressive factors produced by CD8+ T cells. Science
3. Baier M, Werner A, Bannert N, et al. HIV suppression by interleukin-16. Nature
4. Pal R, Garzino-Demo A, Markham PD, et al. Inhibition of HIV-1 infection by the beta-chemokine MDC. Science
5. Geiben-Lynn R, Kursar M, Brown NV, et al. Noncytolytic inhibition of X4 virus by bulk CD8(+) cells from human immunodeficiency virus type 1 (HIV-1)-infected persons and HIV-1-specific cytotoxic T lymphocytes is not mediated by beta-chemokines. J Virol
6. Greco G, Mackewicz C, Levy JA. Sensitivity of human immunodeficiency virus infection to various alpha, beta and gamma chemokines. J Gen Virol
7. Levy JA, Mackewicz CE, Barker E. Controlling HIV pathogenesis: the role of the noncytotoxic anti-HIV response of CD8+ T cells. Immunol Today
8. Mackewicz CE, Ortega H, Levy JA. Effect of cytokines on HIV replication in CD4+ lymphocytes: lack of identity with the CD8+ cell antiviral factor. Cell Immunol
9. Geiben-Lynn R, Brown N, Walker BD, et al. Purification of a modified form of bovine antithrombin III as an HIV-1 CD8+ T-cell antiviral factor. J Biol Chem
10. Geiben-Lynn R, Kursar M, Brown NV, et al. HIV-1 antiviral activity of recombinant natural killer cell enhancing factors, NKEF-A and NKEF-B, members of the peroxiredoxin family. J Biol Chem
11. Zhang L, Yu W, He T, et al. Contribution of human alpha-defensin 1, 2, and 3 to the anti-HIV-1 activity of CD8 antiviral factor. Science
12. Cao Y, Qin L, Zhang L, et al. Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection. N Engl J Med
13. Hsueh FW, Walker CM, Blackbourn DJ, et al. Suppression of HIV replication by CD8+ cell clones derived from HIV-infected and uninfected individuals. Cell Immunol
14. Toso JF, Chen CH, Mohr JR, et al. Oligoclonal CD8 lymphocytes from persons with asymptomatic human immunodeficiency virus (HIV) type 1 infection inhibit HIV-1 replication. J Infect Dis
15. Le Borgne S, Fevrier M, Callebaut C, et al. CD8(+)-cell antiviral factor activity is not restricted to human immunodeficiency virus (HIV)-specific T cells and can block HIV replication after initiation of reverse transcription. J Virol
16. Chun TW, Justement JS, Moir S, et al. Suppression of HIV replication in the resting CD4+ T cell reservoir by autologous CD8+ T cells: implications for the development of therapeutic strategies. Proc Natl Acad Sci U S A
17. Agerberth B, Charo J, Werr J, et al. The human antimicrobial and chemotactic peptides LL-37 and alpha-defensins are expressed by specific lymphocyte and monocyte populations. Blood
18. Obata-Onai A, Hashimoto S, Onai N, et al. Comprehensive gene expression analysis of human NK cells and CD8(+) T lymphocytes. Int Immunol
19. Chang TL, Francois F, Mosoian A, et al. CAF-mediated human immunodeficiency virus (HIV) type 1 transcriptional inhibition is distinct from alpha-defensin-1 HIV inhibition. J Virol
20. Robinson J. Current Protocols in Cytometry
. vol 1. New York: John Wiley & Sons; 1997.
21. Ganz T, Selsted ME, Szklarek D, et al. Defensins: natural peptide antibiotics of human neutrophils. J Clin Invest
22. Daher KA, Lehrer RI, Ganz T, et al. Isolation and characterization of human defensin cDNA clones. Proc Natl Acad Sci U S A
23. Bateman A, Singh A, Shustik C, et al. The isolation and identification of multiple forms of the neutrophil granule peptides from human leukemic cells. J Biol Chem
24. Muller CA, Markovic-Lipkovski J, Klatt T, et al. Human alpha-defensins HNPs-1, -2, and -3 in renal cell carcinoma: influences on tumor cell proliferation. Am J Pathol
25. Ganz T. Extracellular release of antimicrobial defensins by human polymorphonuclear leukocytes. Infect Immun
26. Rice WG, Ganz T, Kinkade JM Jr, et al. Defensin-rich dense granules of human neutrophils. Blood
27. van Oss CJ, Bronson PM, Dinolfo EA, et al. Two methods for the removal of erythrocytes from buffy coats for the production of human leukocyte interferon. Immunol Commun
28. Richard JP, Melikov K, Vives E, et al. Cell-penetrating peptides: a reevaluation of the mechanism of cellular uptake. J Biol Chem
29. Lundberg M, Johansson M. Is VP22 nuclear homing an artifact?Nat Biotechnol
30. Lundberg M, Wikstrom S, Johansson M. Cell surface adherence and endocytosis of protein transduction domains. Mol Ther
31. Bruhl P, Kerschbaum A, Zimmermann K, et al. Allostimulated lymphocytes inhibit replication of HIV type 1. AIDS Res Hum Retroviruses
32. Pinto LA, Sharpe S, Cohen DI, et al. Alloantigen-stimulated anti-HIV activity. Blood
33. Grene E, Pinto LA, Cohen SS, et al. Generation of alloantigen-stimulated anti-human immunodeficiency virus activity is associated with HLA-A*02 expression. J Infect Dis
34. Rugeles MT, Trubey CM, Bedoya VI, et al. Ribonuclease is partly responsible for the HIV-1 inhibitory effect activated by HLA alloantigen recognition. AIDS
35. Wang Y, Tao L, Mitchell E, et al. Allo-immunization elicits CD8+ T cell-derived chemokines, HIV suppressor factors and resistance to HIV infection in women. Nat Med
36. Wang Y, Underwood J, Vaughan R, et al. Allo-immunization elicits CCR5 antibodies, SDF-1 chemokines, and CD8-suppressor factors that inhibit transmission of R5 and X4 HIV-1 in women. Clin Exp Immunol
37. Mackewicz CE, Yuan J, Tran P, et al. Alpha-defensins can have anti-HIV activity but are not CD8 cell anti-HIV factors. AIDS
38. Tomaras GD, Greenberg ML. CD8+ T cell mediated noncytolytic inhibition of human immunodeficiency virus type I. Front Biosci
39. Mackewicz C, Levy JA. CD8+ cell anti-HIV activity: nonlytic suppression of virus replication. AIDS Res Hum Retroviruses
40. Levy JA. HIV pathogenesis and long-term survival. AIDS
41. Mackewicz CE, Ortega HW, Levy JA. CD8+ cell anti-HIV activity correlates with the clinical state of the infected individual. J Clin Invest
42. Yang OO, Kalams SA, Trocha A, et al. Suppression of human immunodeficiency virus type 1 replication by CD8+ cells: evidence for HLA class I-restricted triggering of cytolytic and noncytolytic mechanisms. J Virol
43. Stranford SA, Skurnick J, Louria D, et al. Lack of infection in HIV-exposed individuals is associated with a strong CD8(+) cell noncytotoxic anti-HIV response. Proc Natl Acad Sci U S A
44. Levy JA, Hsueh F, Blackbourn DJ, et al. CD8 cell noncytotoxic antiviral activity in human immunodeficiency virus-infected and -uninfected children. J Infect Dis
45. Bollinger RC, Quinn TC, Liu AY, et al. Cytokines from vaccine-induced HIV-1 specific cytotoxic T lymphocytes: effects on viral replication. AIDS Res Hum Retroviruses
46. Landay AL, Mackewicz CE, Levy JA. An activated CD8+ T cell phenotype correlates with anti-HIV activity and asymptomatic clinical status. Clin Immunol Immunopathol
47. Rosok B, Voltersvik P, Larsson BM, et al. CD8+ T cells from HIV type 1-seronegative individuals suppress virus replication in acutely infected cells. AIDS Res Hum Retroviruses
48. Barker TD, Weissman D, Daucher JA, et al. Identification of multiple and distinct CD8+ T cell suppressor activities: dichotomy between infected and uninfected individuals, evolution with progression of disease, and sensitivity to gamma irradiation. J Immunol
49. Walker CM, Thomson-Honnebier GA, Hsueh FC, et al. CD8+ T cells from HIV-1-infected individuals inhibit acute infection by human and primate immunodeficiency viruses. Cell Immunol
50. Chang TL, Mosoian A, Pine R, et al. A soluble factor(s) secreted from CD8(+) T lymphocytes inhibits human immunodeficiency virus type 1 replication through STAT1 activation. J Virol
51. Tomaras GD, Lacey SF, McDanal CB, et al. CD8+ T cell-mediated suppressive activity inhibits HIV-1 after virus entry with kinetics indicating effects on virus gene expression. Proc Natl Acad Sci U S A
52. Lacey SF, Weinhold KJ, Chen CH, et al. Herpesvirus saimiri transformation of HIV type 1 suppressive CD8+ lymphocytes from an HIV type 1-infected asymptomatic individual. AIDS Res Hum Retroviruses
53. Moriuchi H, Moriuchi M, Combadiere C, et al. CD8+ T-cell-derived soluble factor(s), but not beta-chemokines RANTES, MIP-1 alpha, and MIP-1 beta, suppress HIV-1 replication in monocyte/macrophages. Proc Natl Acad Sci U S A
Keywords:© 2004 Lippincott Williams & Wilkins, Inc.
HIV-1, CD8, lymphocytes; α-defensin; soluble; noncytolytic