SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

There is a breakdown of tolerance to neutrophil components during systemic vasculitis, which is marked by autoantibodies and T cells with specificity for proteinase 3 or myeloperoxidase, expressed on the surface of apoptotic neutrophils. This study was undertaken to investigate the effects of human apoptotic and necrotic neutrophils on human dendritic cell (DC) phenotype and ability to stimulate allogeneic T cell proliferation.

Methods

DCs were generated from human peripheral blood mononuclear cells and allowed to interact with human apoptotic and necrotic neutrophils in the presence or absence of tumor necrosis factor α (TNFα). Effects on DC phenotype and ability to stimulate T cell proliferation were observed.

Results

Immature DCs engulfed apoptotic and necrotic neutrophils, resulting in up-regulation of CD83 and class II major histocompatibility complex molecules, but down-regulation of CD40, CD80, and CD86, and a decreased ability to stimulate T cell proliferation. When TNFα was added in combination with apoptotic neutrophils, the inhibitory effects were overcome to some extent.

Conclusion

Our results suggest that DC uptake of apoptotic or necrotic neutrophils alone does not shift the immune response from tolerance to autoimmunity in systemic vasculitis. However, cytokines found at sites of inflammation in vasculitis patients may act as maturation factors for DCs, and in combination with apoptotic neutrophils, may lead to an autoimmune phenotype.

Dendritic cells (DCs) are potent antigen-presenting cells, which are present in low numbers in many body tissues (1). Immature DCs can ingest antigen, but express low levels of molecules required for antigen presentation and T cell stimulation, e.g., class I and II major histocompatibility complex (MHC) molecules, and costimulatory molecules (2). Following antigen uptake, DCs migrate to secondary lymphoid organs, where they receive maturation signals, which include components of bacteria and viruses, e.g., lipopolysaccharide (LPS), tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), and interferon-α (3–5). Mature DCs are less able to ingest antigen but have increased expression of MHC and costimulatory molecules, enabling them to activate antigen-specific T cells and induce primary immune responses (2).

In the immune system, cells usually die by apoptosis, but necrosis may occur in cases of severe injury. Apoptotic and necrotic cells can both be engulfed and degraded by immature DCs, but the outcome with regard to the immune response may vary depending on the type of cell death and whether maturation signals are received (6–12). If apoptotic cells are engulfed by DCs in the absence of a subsequent maturation signal, T cells may become tolerant to the peptides displayed on the surface of the DCs (13). During a viral infection, cells die by apoptosis and necrosis. Both infected apoptotic and necrotic cells are ingested by DCs, and maturation signals received from the infected necrotic cells induce expression of DC-costimulatory molecules, leading to the presentation of viral antigenic peptides from apoptotic cells, T cell activation, and immunity (7, 9).

Gallucci et al have demonstrated in a murine model that, in the absence of foreign substances, signals from necrotic fibroblasts, but not apoptotic or healthy fibroblasts, lead to DC maturation (7). In contrast, Salio et al found that neither necrotic nor apoptotic melanoma cell lines induce human DC maturation (14). A further level of complexity has been added by the suggestion that even without additional maturation signals, high numbers of apoptotic cells (at a ratio of 5 apoptotic cells to 1 DC) are capable of inducing DC maturation and the presentation of intracellular antigens from these apoptotic cells, whereas low numbers of apoptotic cells are disposed of in a noninflammatory manner (15). Furthermore, DCs challenged with low numbers of anti–β2-glycoprotein I–opsonized apoptotic cells present antigen with high efficiency and secrete proinflammatory and maturation factors, e.g., IL-1β and TNFα (16). Opsonization also substantially increases the percentage of DCs engaged in phagocytosis (17). Thus, there is no clearly predictable effect of apoptotic or necrotic cells on DC maturation that can be defined from the current literature, despite the importance such responses may have for development (or lack of development) of tolerance or autoimmunity.

Microscopic polyangiitis (MPA) and Wegener's granulomatosis (WG) are two forms of small-vessel vasculitis in which autoimmune responses are directed toward neutrophil enzymes (18–20). Antineutrophil cytoplasmic antibodies (ANCAs) develop and may bind to proteinase 3 (PR3) (particularly in patients with WG) or to myeloperoxidase (MPO) (particularly in patients with MPA) (21–23). These antibodies, PR3 ANCA and MPO ANCA, are believed to be involved in the initial development of neutrophil-mediated endothelial injury since they can bind to their target antigens when these are expressed on the surface of TNFα-primed neutrophils (24, 25) and also on the surface of apoptotic neutrophils (26). Ligation of PR3 and MPO on primed neutrophils induces neutrophil activation, with a respiratory burst and release of granule contents (27), factors that promote endothelial damage (28, 29). ANCAs also induce accelerated and dysregulated apoptosis of TNFα-primed neutrophils, resulting in a “reduced window of opportunity” for recognition and phagocytosis by macrophages before disintegration (30).

We have been interested in the initial breakdown of tolerance to neutrophil components during vasculitis. We hypothesized that in the event apoptotic or necrotic neutrophils are not cleared by macrophages, DCs may be recruited. In WG, granulomas containing large numbers of freshly activated, apoptotic, and necrotic neutrophils are present in upper airways early in the development of vasculitis (31–33). As neutrophils undergo apoptosis, they express PR3 and MPO on their surface (26, 34). Apoptotic neutrophils injected into rats induce production of ANCA (35). The aim of the present study was to establish how DCs might handle apoptotic or necrotic neutrophils in terms of uptake and effects on DC maturation and T cell–stimulatory ability. Specifically, we investigated whether DC uptake of high numbers of apoptotic neutrophils (15), of antibody–opsonized apoptotic cells (16), or of necrotic cells (7) would increase DC maturation and ability to stimulate T cell proliferation.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Approval for this study was obtained from the institutional ethics review committee, and the Declaration of Helsinki principles were followed.

DC culture.

DCs were generated from peripheral blood mononuclear cells (PBMCs) as previously described (36). Briefly, freshly drawn heparinized peripheral blood from healthy volunteers was mixed 1:1 with phosphate buffered saline (PBS; Sigma-Aldrich, Poole, UK). PBMCs were isolated by density centrifugation on Ficoll-Paque PLUS (Amersham Pharmacia Biotech, Little Chalfont, UK) and resuspended to 5 × 106/ml in RPMI 1640 (Sigma-Aldrich) supplemented with 10% heat-inactivated fetal calf serum (First Link, Brierly Hill, UK), 2 mML-glutamine, 1% nonessential amino acids, 15 mM HEPES buffer, and 100 units/ml penicillin + 100 μg/ml streptomycin (all from Sigma-Aldrich). Monocytes were allowed to adhere to 6-well plates (3 ml/well; BD Biosciences, Cowley, UK) for 2 hours at 37°C. The nonadherent cells were removed and the adherent cells were resuspended to 3 ml/well in fresh medium containing 800 units/ml recombinant human granulocyte–macrophage colony-stimulating factor (GM-CSF) (Leucomax; Sandoz, Surrey, UK) and 500 units/ml recombinant human IL-4 (R&D Systems, Abingdon, UK). On days 3 and 6, 1 ml of medium was removed and replaced with fresh medium containing 2,400 units/ml GM-CSF and 1,500 units/ml IL-4. DCs were harvested on day 8 of culture for subsequent experiments.

Neutrophil isolation.

Human neutrophils were isolated as previously described (37). Neutrophils were washed and resuspended to 5 × 106/ml in Iscove's modified Dulbecco's medium (IMDM; Sigma-Aldrich) supplemented with penicillin/streptomycin and 10% autologous serum. In some experiments, neutrophil membranes were dyed red, using the PKH26 red fluorescent cell linker kit according to the instructions of the manufacturer (Sigma-Aldrich).

Induction of neutrophil apoptosis and necrosis.

Neutrophils were cultured at 37°C for 18 hours, after which 80 ± 1.4% of the cells (mean ± SEM) were apoptotic as assessed by morphologic analysis using Jenner-Giemsa (BDH, Poole, UK) staining of cytospins. Apoptosis was confirmed by staining for annexin V (30, 34). Fewer than 5% of cells were necrotic as assessed by trypan blue dye exclusion and propidium iodide incorporation. Nevertheless, preparations were washed to remove necrotic debris prior to use. Necrosis was achieved in these cells with 4 cycles of freezing (liquid nitrogen) and thawing (37°C), leading to complete fragmentation of the cells, with <1% intact cells remaining. In some experiments freshly isolated neutrophils were made necrotic by freezing and thawing. Secondary necrosis was achieved by culturing neutrophils at 37°C for 40 hours, by which time >90% of the cells were necrotic as assessed by trypan blue staining.

Opsonization of apoptotic neutrophils.

Apoptotic neutrophils were resuspended to 5 × 106 cells/ml in IMDM containing penicillin/streptomycin, but without serum. IgG was isolated from serum of normal healthy volunteers or from patients with PR3 ANCA or MPO ANCA, using a Hi-trap protein G affinity column (Amersham Pharmacia Biotech) as previously described (37). The neutrophils were opsonized with 200 μg/ml IgG at 37°C for 45 minutes. They were then washed and incubated with fluorescein isothiocyanate (FITC)–conjugated mouse anti-human IgG (Fab monomer; kindly supplied by The Binding Site, Birmingham, UK) at 4°C for 30 minutes, washed, and fixed using 2% paraformaldehyde (BDH). Opsonization of neutrophils with ANCA IgG relative to normal IgG was confirmed by flow cytometry (FACScan; Becton Dickinson, Oxford, UK) using Cell Quest software (Becton Dickinson) and analyzed using winMDI software (available from The Scripps Research Institute, La Jolla, CA: http://facs.scripps.edu/facsindex.html).

Quantitation of phagocytosis of neutrophils by DCs.

Uptake of apoptotic neutrophils by DCs was assessed using an assay with microscopic quantification to determine macrophage ingestion of apoptotic neutrophils (38, 39). Briefly, on day 8 of culture, DCs were resuspended to 5 × 105 cells/ml. Cells (250 μl) were added to a 24-well plate (BD Biosciences) and allowed to adhere at 37°C for 30 minutes. Apoptotic neutrophils (opsonized and nonopsonized) were added to the DCs to yield ratios of neutrophils to DCs ranging from 0.1:1 to 20:1. Interaction was allowed for 1 hour at 37°C. In some experiments phagocytosis was prevented by the addition of cytochalasin D (Sigma-Aldrich) to the DCs at the beginning of the experiment, or by performing the experiment at 4°C. After 1 hour, noningested neutrophils were washed out of the wells with ice-cold saline, with care taken not to dislodge the DCs. The cells were fixed by addition of ice-cold 2% paraformaldehyde and stained for MPO using hydrogen peroxide and dimethoxybenzidine (o-diansidine; Sigma-Aldrich). The neutrophils were 100% positive for MPO staining, whereas the DCs were 100% negative. The number of neutrophils ingested by DCs was counted, and the phagocytic index was calculated as the number of phagocytosed neutrophils divided by the total number of DCs (expressed as a percentage).

Coculture of DCs and neutrophils.

DCs on day 8 of culture were resuspended to 1 × 106 cells/ml in IMDM containing penicillin/streptomycin. Cells (100 μl) were added to a 96-well plate (BD Biosciences). Apoptotic, opsonized apoptotic, or necrotic neutrophils (100 μl) were added to the DCs to yield neutrophil to DC ratios of 1:1, 5:1, and 20:1. In some cases, 300 units/ml TNFα was added along with the neutrophils. DCs were cultured without neutrophils as a reference, and with 300 units/ml TNFα as a positive control for DC maturation. The cells were incubated for 24 hours (overnight) at 37°C.

Flow cytometric analysis of DCs.

The phenotype of the DCs after coculture was assessed. Cells were incubated with FITC-conjugated mouse anti-human CD40 (Alexis, Nottingham, UK), CD80 (BD PharMingen, Oxford, UK), CD83 (BD PharMingen), CD86 (Alexis), class II MHC (Dako, Ely, UK), and an IgG1 isotype control (Dako) for 30 minutes at 4°C. Cells were fixed with 2% paraformaldehyde, assessed by flow cytometry (FACScan) using CellQuest software, and analyzed using winMDI software.

Mixed lymphocyte reaction (MLR).

The membrane dye PKH26 has previously been used to quantitate the proliferation of cells by flow cytometry (40, 41). Use of this method avoids the need to purify mixed populations of cells and yields a cumulative measure of proliferation. DCs and neutrophils were cocultured for 24 hours as described above, and the DCs were then analyzed for their T cell–stimulatory capacity. Responder cells were monocyte-depleted PBMCs, membranes of which had been dyed red using the PKH26 red fluorescent cell linker kit. These cells (100 μl) were plated in 96-well plates at 1 × 106 cells/ml. DCs cocultured as above were added at various ratios to the T cells and cultured at 37°C for 5 days. On day 5, the cells were stained with FITC-conjugated mouse anti-human CD3 (Dako) and proliferation relative to cultures of T cells alone was assessed by flow cytometry (FACScan) using CellQuest software and analyzed using winMDI software. The amount of proliferation seen is proportional to the dye-negative population.

Necrotic tumor cell line.

An Epstein-Barr virus–transformed B lymphocyte tumor cell line (B-LCL; R-Biopharm, Darmstadt, Germany) that was negative for Mycoplasma by immunofluorescence was kindly provided by Chris Dawson (Cancer Research Centre, University of Birmingham).

Statistical analysis.

The nonparametric Mann-Whitney test was used for comparisons of unpaired groups, and Wilcoxon's signed rank test was used for comparisons with values obtained using DCs cultured alone (which were assigned a value of 1).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Dendritic cell uptake of apoptotic neutrophils.

Neutrophils were cultured for 18 hours to routinely achieve >60% apoptosis (range 62–91%, mean ± SEM 80 ± 1.4%). Apoptosis was verified by morphologic analysis (Figure 1A), which revealed that apoptotic cells were shrunken in volume and showed characteristic condensation of their nuclei. Apoptosis was verified by flow cytometric analysis of phosphatidylserine exposure (annexin V binding) and propidium iodide incorporation (Figure 1B). We first investigated the ability of DCs to engulf increasing numbers of apoptotic neutrophils after 1 hour of culture, using a microscopic quantitation assay in which ingested apoptotic neutrophils are stained for MPO. Phagocytosis of apoptotic neutrophils increased as the ratio of neutrophils to DCs increased (Figure 1C). At the highest ratio of neutrophils to DC (20:1), the phagocytic index (mean ± SEM) was 24.9 ± 4.5%. Uptake was decreased at least 3.6-fold when apoptotic neutrophils were cocultured with DCs in the presence of cytochalasin D or at 4°C (Figures 1D and E), suggesting that cells were not bound nonspecifically.

thumbnail image

Figure 1. Dendritic cell (DC) uptake of apoptotic neutrophils. A, Cytospin of neutrophils aged for 18 hours and stained using Jenner-Giemsa. Healthy neutrophils are large with multilobed nuclei (open arrow); apoptotic neutrophils are shrunken in volume and show characteristic condensation of their nuclei (solid arrow). B, Flow cytometric plot showing the percentage of neutrophils that are apoptotic (annexin V positive) but not necrotic (propidium iodide negative). FITC = fluorescein isothiocyanate; PE = phycoerythrin. C–E, Phagocytic index (expressed as a percentage), calculated as the number of phagocytosed neutrophils divided by the total number of DCs after 1-hour culture of DCs with apoptotic neutrophils, and measured C, in experiments with increasing ratios of apoptotic neutrophils to DCs, D, in the presence or absence of cytochalasin D (CCD), and E, in experiments performed at 37°C or 4°C. Values are the mean ± SEM.

Download figure to PowerPoint

The uptake of membrane-dyed neutrophils by DCs can also be seen by fluorescence microscopy. Dyed neutrophils aged overnight had characteristic condensed nuclei associated with apoptosis (Figure 2A). DCs, shown in phase contrast, were dye negative (Figure 2B). DCs cultured for 1 hour in the presence of dyed neutrophils acquired the red dye from the neutrophils (Figure 2C). The distribution of dye through the DCs suggests that fragments, rather than whole cells, were eaten.

thumbnail image

Figure 2. Fluorescence microscopy, showing dendritic cell (DC) uptake of dyed apoptotic neutrophils. Neutrophils were dyed with a red membrane dye and allowed to age for 18 hours. Apoptotic neutrophils were then cocultured with DCs for 1 hour. Cytospins of neutrophils only, DCs only, and DC and neutrophil cocultures were examined by fluorescence microscopy. A, Dyed neutrophils, showing the characteristic condensed nuclei associated with apoptosis. B, Nondyed DCs in phase contrast (large cells are DCs, smaller cells are contaminating lymphocytes). C, DCs and neutrophils cocultured for 1 hour. Arrows show DCs that have acquired the red dye from the neutrophils.

Download figure to PowerPoint

Effects of exposure to high numbers of apoptotic neutrophils on DC phenotype.

The effects of increasing numbers of apoptotic neutrophils on the maturation of DCs were studied. DCs were cocultured with increasing numbers of apoptotic neutrophils (1:1–20:1) overnight (24 hours), after which the DCs were collected, stained for markers indicative of maturation, and the mean fluorescence intensity (MFI) was analyzed on a FACScan flow cytometer (Figure 3A). The phenotype of DCs cultured overnight without the addition of neutrophils was taken as the reference point. DCs cultured alone were assigned a value of 1. Addition of apoptotic neutrophils at neutrophil to DC ratios of 1:1 and 20:1 caused a significant increase in the MFI of CD83 (Figure 3B), a marker specific for DC maturation, compared with DCs cultured alone, which expressed little or no CD83. This increase was greatest (mean ± SEM 1.32 ± 0.10–fold; P = 0.016) when the highest ratio of neutrophils to DCs (20:1) was used.

thumbnail image

Figure 3. Effect of apoptotic neutrophils on dendritic cell (DC) phenotype. A, Flow cytometric plots of DCs cultured overnight (24 hours) in the absence of neutrophils (open graphs), showing the fluorescence intensity of the DC markers compared with the intensity obtained with isotype-matched negative control (shaded graphs). B–E, The phenotype of DCs for B, CD83, C, class II major histocompatibility complex (MHC), D, CD86, and E, CD40 was examined on day 8 of culture (DC), after overnight culture in a fresh well without the addition of neutrophils (O/N), and after overnight culture with neutrophils at apoptotic neutrophil to DC ratios of 1:1, 5:1, and 20:1. CD80 was not tested in this set of experiments, but was included later. Mean fluorescence intensity (MFI) results are expressed as a ratio compared with DCs cultured overnight alone, which were assigned an arbitrary value of 1. Individual symbols represent individual experiments; bars show the mean ± SEM. ∗ = P < 0.05; ∗∗ = P < 0.01 versus DC cultured overnight alone, by Wilcoxon's signed rank test.

Download figure to PowerPoint

When apoptotic neutrophils were added to DCs at ratios of 1:1 and 20:1, the MFI of class II MHC was significantly increased compared with that observed with DCs alone, and this was greatest (2.01 ± 0.34–fold; P = 0.016) at the highest ratio of neutrophils to DCs (Figure 3C). High numbers of apoptotic neutrophils (20 neutrophils to 1 DC) caused a significant decrease in the MFI of the costimulatory molecule CD86 (1.51 ± 0.05–fold; P = 0.008) (Figure 3D), and they also caused a significant decrease in the MFI of CD40, which is required for T cell activation (1.62 ± 0.11–fold; P = 0.016) (Figure 3E). Low numbers of apoptotic neutrophils (1 neutrophil to 1 DC) actually caused an increase in the MFI of CD86 and CD40; however, these results were not statistically significant. The process of transferring DCs to a fresh well before culturing overnight in the absence of apoptotic neutrophils led to a spontaneous up-regulation of CD83 and CD40, but no change in CD86 and class II MHC. Addition of apoptotic neutrophils allowed further up-regulation of all of these markers, suggesting that full maturation was not induced.

In further experiments, apoptotic neutrophils that had been opsonized with PR3 ANCA or MPO ANCA IgG were added to DCs. The phagocytic index for nonopsonized apoptotic neutrophils at the highest ratio (20 neutrophils to 1 DC) was 24.9 ± 4.5% (mean ± SEM), compared with 34.8 ± 3.4% for PR3 ANCA IgG–opsonized neutrophils (P = 0.114) and 44.3 ± 5.4% for MPO ANCA IgG–opsonized neutrophils (P = 0.029) (Figure 4A). Because opsonization increased uptake of apoptotic neutrophils, we anticipated that effects on DC phenotype might be more marked, or different. However, opsonization of apoptotic neutrophils with MPO ANCA IgG did not result in any different phenotypic changes of the DCs when compared with nonopsonized apoptotic neutrophils (Figures 4B–E).

thumbnail image

Figure 4. DC uptake of opsonized apoptotic neutrophils (polymorphonuclear cells [PMN]). A, DCs were cultured with apoptotic neutrophils that were nonopsonized or had been opsonized for 1 hour with proteinase 3 antineutrophil cytoplasmic antibody (PR3 ANCA) or myeloperoxidase ANCA (MPO ANCA) IgG. The phagocytic index was calculated as the number of phagocytosed neutrophils divided by the total number of DCs (expressed as a percentage). Values are the mean ± SEM. ∗ = P < 0.05 versus nonopsonized apoptotic neutrophils, by nonparametric Mann-Whitney test. B–E, The phenotype of DCs for B, CD83, C, class II MHC, D, CD86, and E, CD40 was examined after overnight culture with nonopsonized neutrophils at apoptotic neutrophil to DC ratios of 1:1 (1), 5:1 (2), and 20:1 (3), or MPO-opsonized neutrophils at apoptotic neutrophil to DC ratios of 1:1 (4), 5:1 (5), and 20:1 (6). CD80 was not tested in this set of experiments, but was included later. MFI results are expressed as a ratio compared with DCs cultured alone, which were assigned an arbitrary value of 1. Individual symbols represent individual experiments; bars show the mean ± SEM. There were no statistically significant differences in DC phenotype after uptake of nonopsonized versus MPO-opsonized apoptotic neutrophils at any ratio, for any of the phenotypic markers examined. See Figure 3 for other definitions.

Download figure to PowerPoint

Exposure to necrotic neutrophils affects DC phenotype in a similar manner to apoptotic neutrophils.

Next, the effects of necrotic neutrophils on the phenotype of the DCs were examined. Necrotic neutrophils were generated by 4 cycles of freeze–thawing to cause complete cellular fragmentation. DCs were cocultured with apoptotic or necrotic neutrophils with or without TNFα, at 5 neutrophils to 1 DC (a ratio selected for reasons of feasibility). After 24 hours the DCs were collected, stained for markers indicative of maturation, and analyzed on a FACScan flow cytometer. DCs cultured with TNFα were used as a positive control, and DCs cultured alone were used as the reference, and assigned a value of 1. Figure 5A shows a representative FACS analysis from DCs cultured alone compared with DCs cultured with TNFα and DCs cultured with apoptotic neutrophils. Addition of necrotic neutrophils to DCs caused changes to the DC phenotype similar to those observed with apoptotic neutrophils, i.e., up-regulation of CD83 and class II MHC and down-regulation of CD86 and CD80 (Figures 5B–E). Addition of necrotic neutrophils to DCs caused a more significant decrease in CD40 expression than was seen in experiments using apoptotic neutrophils (mean ± SEM 4.08 ± 0.06–fold versus 1.59 ± 0.05–fold; P = 0.008) (Figure 5F).

thumbnail image

Figure 5. Effect of apoptotic or necrotic neutrophils, with or without addition of tumor necrosis factor α (TNFα), on DC phenotype. A, Flow cytometric plots of DCs cultured overnight (24 hours) in the absence of neutrophils (shaded graphs), showing the fluorescence intensity of the DC markers compared with the intensity obtained with DCs cultured in the presence of TNFα (open graphs, solid lines) and in the presence of apoptotic neutrophils (open graph, dotted line). B–F, The phenotype of DCs for B, CD83, C, class II MHC, D, CD86, E, CD80, and F, CD40 was examined after overnight culture in a fresh well without the addition of neutrophils (1), with TNFα (2), with apoptotic neutrophils at a ratio of 5 neutrophils to 1 DC (3), with apoptotic neutrophils at a ratio of 5 neutrophils to 1 DC and with TNFα (4), with necrotic neutrophils at a ratio of 5 neutrophils to 1 DC (5), and with necrotic neutrophils at a ratio of 5 neutrophils to 1 DC and with TNFα (6). MFI results are expressed as a ratio compared with DCs cultured alone, which were assigned an arbitrary value of 1. Individual symbols represent individual experiments; bars show the mean ± SEM. ∗ = P < 0.05; ∗∗ = P < 0.01, by nonparametric Mann-Whitney test. See Figure 3 for other definitions.

Download figure to PowerPoint

The up-regulation of CD83 and class II MHC seen when apoptotic and necrotic neutrophils were added to DCs was enhanced when TNFα was added concurrently. TNFα added in combination with apoptotic neutrophils was more effective than when added in combination with necrotic neutrophils, at overcoming the down-regulation of CD80, CD86, and CD40. The down-regulation of CD40 seen when apoptotic neutrophils were added to DCs was reversed when TNFα was added in combination (1.59-fold decrease to 1.14-fold increase; P = 0.056) (Figure 5F).

Neutrophils that had been allowed to age for 40 hours, indicating progression through apoptosis to secondary necrosis, showed effects on DC phenotype similar to those seen with apoptotic neutrophils (data not shown). Neutrophils subjected to 1 cycle of freeze–thawing, resulting in mainly intact, trypan blue-positive cells, exerted similar effects on DC phenotype as neutrophils subjected to 4 cycles of freeze–thawing (data not shown); Sauter et al (9) reported that supernatants of tumor cells subjected to 1 cycle of freeze–thawing did not cause up-regulation of CD83, whereas those subjected to 4 cycles did.

Effect of exposure to apoptotic and necrotic neutrophils on the ability of DCs to stimulate the MLR.

Mature DCs are potent stimulators of allogeneic T cell proliferation in the MLR (42). To determine the effects of apoptotic and necrotic neutrophils on the ability of DCs to stimulate the MLR, DCs were cocultured with apoptotic or necrotic neutrophils (with or without TNFα) at a ratio of 5 neutrophils to 1 DC for 24 hours. DCs stimulated with TNFα were used as a positive control, DCs cultured alone were used as a reference, and T cells alone were used as a negative control. DCs and red-dyed responder T cells were cocultured for 5 days. Proliferation was assessed as the percentage of T cells that had become dye negative. There was no significant loss of dye from T cells cultured alone on day 5 compared with T cells freshly isolated on day 0. After 5 days, the highest amount of proliferation occurred when DCs were stimulated with TNFα as the positive control (Figure 6A). DCs cultured with apoptotic neutrophils were less able to stimulate the MLR than DCs alone, and this reduction in T cell proliferation was even more marked when DCs were cocultured with necrotic neutrophils.

thumbnail image

Figure 6. Effect of exposure to apoptotic and necrotic neutrophils on the ability of dendritic cells (DCs) to stimulate allogeneic T cell proliferation in a mixed lymphocyte reaction. DCs were added to allogeneic red membrane-dyed T cells at DC to T cell ratios decreasing from 1:4 to 1:64. After 5 days of culture, the T cells were identified with a fluorescein isothiocyanate-conjugated CD3 antibody, and the percentage proliferation was determined as the percentage of CD3-positive, red dye-negative T cells. A, DCs cultured overnight with no addition, with tumor necrosis factor α (TNFα), with apoptotic neutrophils (apop) at a ratio of 5 neutrophils to 1 DC, or with necrotic neutrophils (nec) at a ratio of 5 neutrophils to 1 DC. B, DCs cultured overnight with no addition, with TNFα, with apoptotic neutrophils at a ratio of 5 neutrophils to 1 DC (with or without TNFα), or with necrotic neutrophils at a ratio of 5 neutrophils to 1 DC (with or without TNFα). C, DCs cultured overnight with no addition, with TNFα, or with apoptotic neutrophils (polymorphonuclear cells [PMN]) at ratios of 1, 5, and 20 neutrophils to 1 DC. D, DCs cultured overnight with no addition, with TNFα, with apoptotic neutrophils, with secondarily necrotic neutrophils (2nd Nec), or with necrotic neutrophils after 1 or 4 cycles of freeze–thawing (FT) (at ratios of 5 neutrophils to1 DC). E, DC cultured overnight with no addition, with TNFα, or with a necrotic B lymphocyte tumor cell line (nec B-LCL) (at a ratio of 5 cells to 1 DC). Values are the mean ± SEM.

Download figure to PowerPoint

Coculture of DCs with apoptotic neutrophils and TNFα caused proliferation of responder T cells to increase above that seen with DCs cultured alone (Figure 6B). Although addition of TNFα to necrotic neutrophils prior to coculture with DCs did cause some increase in T cell proliferation, this still remained below the level seen when DCs were cultured alone (Figure 6B).

The effects on the ability of DCs to induce T cell proliferation seen in studies using freshly isolated neutrophils treated with 4 cycles of freeze–thawing to cause necrosis were similar to those observed with already apoptotic neutrophils treated with 4 cycles of freeze–thawing to cause necrosis (data not shown). Although the MLR experiments in Figures 6A and B were carried out with DCs that had been exposed to apoptotic or necrotic neutrophils at a ratio of 5 neutrophils to 1 DC rather than 20 neutrophils to 1 DC (to facilitate conduct of these experiments), significant reduction of the ability of DCs to stimulate T cells was still apparent. At a ratio of 20 apoptotic neutrophils to 1 DC, the ability of the DCs to induce T cell proliferation was greatly inhibited (to a similar degree as was observed with necrotic neutrophils in Figure 6A) (Figure 6C).

The nature of neutrophil death affects the ability of DCs to stimulate T cell proliferation in a stepwise manner. DCs cocultured with apoptotic neutrophils stimulated the highest amount of T cell proliferation, followed by secondarily necrotic neutrophils, then by neutrophils subjected to 1 cycle of freeze–thawing, while neutrophils subjected to 4 cycles of freeze–thawing were least effective (Figure 6D). On the other hand, a necrotic B-LCL tumor cell line subjected to 4 cycles of freeze–thawing and then cocultured with DCs induced DC maturation as assessed by an increase in CD40, CD80, and CD86 expression on the DCs (data not shown) and by the ability of DCs to stimulate T cell proliferation (Figure 6E). These results are consistent with those reported by Sauter et al (9).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Immature DCs are capable of efficiently phagocytosing both apoptotic and necrotic cells (7, 9, 14). This study is the first to provide direct evidence that immature human monocyte-derived DCs can ingest human apoptotic and necrotic neutrophils. We also hypothesized that uptake of either high numbers of apoptotic neutrophils, antibody-opsonized apoptotic neutrophils, or necrotic neutrophils might sensitize DCs for T cell stimulation. These hypotheses were of interest with regard to the development of antineutrophil-directed autoimmune responses in systemic vasculitis, a group of disorders in which aberrant interactions between neutrophils and ANCAs lead to endothelial cell and vascular injury (28, 29). Our results did not confirm these hypotheses since they did not provide evidence that aberrant uptake of neutrophils by DCs triggers an autoimmune response leading to generation of ANCAs. The data also did not support the contention that ANCAs may help perpetuate an immune response through their previously demonstrated effects on accelerating neutrophil apoptosis. ANCAs induce accelerated and dysregulated apoptosis of TNFα-primed neutrophils, resulting in a reduced likelihood of recognition and phagocytosis by macrophages before disintegration through secondary necrosis (30). However, uptake of secondarily necrotic neutrophils by DCs led to a reduced, not enhanced, capacity of DCs to stimulate an MLR.

As has previously been described (15), if DCs are fed with increasing numbers of apoptotic cells they take up increasing numbers of cells, and this is also true for apoptotic neutrophils. We tested whether high numbers of apoptotic or necrotic neutrophils could induce DC maturation. High numbers of apoptotic T cells from a murine T cell line have been shown to induce DC maturation as assessed by secretion of IL-1β and TNFα and up-regulation of class II MHC, CD86, and CD40 (15). In the present study, the addition of apoptotic neutrophils to DCs did cause an increase in CD83 (a DC maturation-specific marker) and class II MHC expression, which would suggest maturation. However, high numbers of apoptotic neutrophils actually caused a down-regulation of the costimulatory molecules CD80 and CD86, and of CD40, which is required for T cell activation. This down-regulation corresponds to a decreased ability of DCs treated with high numbers of apoptotic neutrophils to stimulate T cell proliferation in an MLR. These results are consistent with the findings of previous studies showing that when DCs are cocultured with apoptotic cells prior to the addition of a maturation stimulus, they fail to mature phenotypically, and in fact their surface expression of CD83 and CD86 was lower than that in control DCs, as was the ability to induce allogeneic T cell responses (43, 44).

Sauter et al have found that certain tumor cell lines that have undergone 4 cycles of freeze–thawing are able to stimulate DC maturation as assessed by induction of CD83 and lysosome-associated membrane glycoprotein expression, up-regulation of CD86, HLA, and CD40 expression, and ability to stimulate allogeneic and superantigen-stimulated syngeneic T cell proliferation, but with no release of TNFα or IL-1β (9). Subjecting neutrophils to 4 cycles of freeze–thawing resulted in complete cellular fragmentation. Addition of necrotic neutrophils did cause DC maturation as assessed by CD83 and class II MHC expression; however, there was down-regulation of CD80 and CD86 as seen with apoptotic neutrophils, and an even greater down-regulation of CD40. Sauter and colleagues demonstrated DC maturation only when necrotic cell lines were used, and not with necrotic primary cells. Furthermore, they did not examine whether primary necrotic cells can cause a down-regulation of costimulatory molecules and CD40 (9). Corresponding to the very low level of CD40 expression on DCs following uptake of necrotic neutrophils, DCs were less able to stimulate T cells in an MLR when treated with necrotic cells compared with apoptotic neutrophils.

These results would suggest that neutrophils, whether apoptotic, secondarily necrotic, or 4 freeze–thawed necrotic, do not per se shift the outcome from tolerance to autoimmunity in ANCA-associated vasculitis. In accordance with the findings of Sauter et al (9), we found that necrotic B-LCL did induce DC maturation as assessed by an increase in expression of CD40, CD80, and CD86 on the DC surface, and also had an increased ability to stimulate allogeneic T cell proliferation. Urban et al showed that necrotic cells derived from primary cell isolates did not lead to DC maturation, but unlike apoptotic cells, they also did not prevent LPS-induced maturation (43).

In the presence of apoptotic peripheral blood lymphocytes, PBMCs and monocytes have been found to produce increased amounts of antiinflammatory cytokines and decreased amounts of proinflammatory cytokines (45). Macrophage responses have been shown to differ after uptake of apoptotic or necrotic neutrophils. Phagocytosis of apoptotic neutrophils induced macrophages to secrete antiinflammatory cytokines and actively inhibited the production of proinflammatory cytokines (46). Similarly, late apoptotic neutrophils ingested by macrophages did not trigger the release of proinflammatory cytokines (47), whereas necrotic neutrophils induced macrophages to secrete inflammation mediators (48) and were able to efficiently costimulate T cells due to rapid up-regulation of CD40 (49). Clearly, these results and the results of the current study emphasize the differential effects of apoptotic and necrotic cells on macrophages and DCs and highlight the need for study of other types of primary cells.

Opsonization of low numbers of apoptotic cells has been shown to substantially increase the percentage of DCs engaged in phagocytosis (17) and to induce DC maturation as assessed by secretion of IL-1β and TNFα (16). Opsonization of apoptotic neutrophils with IgG from ANCA-positive vasculitis patients increased the numbers of neutrophils phagocytosed by the DCs but did not induce DC maturation. In fact, the results obtained with opsonized apoptotic neutrophils were equivalent to those obtained with nonopsonized apoptotic neutrophils.

This study shows that the negative effects of apoptotic or necrotic neutrophils on DC phenotype and function can be partially counteracted by TNFα, a proinflammatory cytokine present at sites of inflammation. The down-regulation of CD80, CD86, and CD40 seen when necrotic neutrophils were added to DCs was overcome very slightly by concurrent addition of TNFα, but the levels of expression were still below those seen when DCs were untreated (i.e., immature DCs). This was mirrored in the ability of DCs to stimulate T cell proliferation, which was still reduced. It is conceivable that future studies that could be undertaken with varying dose ratios of necrotic neutrophils to TNFα might yield results that would highlight more subtle influences here. The down-regulation of CD80, CD86, and CD40 expression on DCs following addition of apoptotic neutrophils was overcome following addition of TNFα and apoptotic neutrophils, and addition of TNFα and apoptotic neutrophils caused a slight enhancement in the expression of CD40. Once again the level of CD40 expression correlated with T cell proliferation since the proliferation seen was greater than that observed with DCs only.

Thus, TNFα was able to partially overcome the effects of apoptotic cells on DCs, but was unable to rescue the necrotic cell effect, suggesting that apoptotic and necrotic cells may have subtle differences in recognition mechanisms or in strength of suppressive signal. In addition, TNFα may act on the small population of living cells in the apoptotic fraction, leading to the release of proinflammatory agents, which activate DCs.

There is evidence for enhanced transcription of the TNFα gene in PBMCs from patients with systemic vasculitis compared with controls (50). Furthermore, TNFα and IL-1β have been detected, by immunocytochemistry, polymerase chain reaction, and in situ hybridization, in the kidneys of patients with ANCA-positive systemic vasculitis, demonstrating that cytokines are produced in these patients (51). Circulating T cells from patients with active WG secrete increased amounts of TNFα compared with T cells from healthy controls (52). The production of TNFα, IL-1, and IL-8 by macrophages is significantly up-regulated when they are incubated with ANCA-opsonized apoptotic neutrophils compared with normal IgG–opsonized or nonopsonized apoptotic neutrophils (34, 53). TNFα levels prior to development of vasculitis have not been studied, but, since many patients have a history of chronic infection before the onset of vasculitis (54) and intercurrent infection often precedes relapse (55), it is likely that synthesis of TNFα is increased. If this is the case, it is conceivable that uptake of apoptotic neutrophils by DCs within an environment rich in TNFα might encourage development of an autoimmune response toward neutrophil PR3 or MPO.

Contrary to our hypotheses prior to these studies, uptake of apoptotic and especially necrotic neutrophils appears to have a tolerogenic effect on DCs. However, cytokines found in vivo at sites of inflammation, such as airway granulomata prior to the development of overt autoimmune systemic vasculitis, may act as maturation factors for DCs, allowing the uptake of apoptotic neutrophils to push the outcome toward autoimmunity instead of tolerance. Clearly, future studies need to address the ability of healthy or patient-derived DCs to take up apoptotic or necrotic neutrophils in the presence or absence of TNFα and to present PR3 or MPO antigens to antigen-specific T cells.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We thank Peter Hewins for isolating IgG samples from patients and controls, Peter Nightingale for advice on the statistical analyses, and John Owen for help with the fluorescence microscopy and advice on the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES