Microbial Antigen Triggers Rapid Mobilization of TNF- to the Surface of Mouse Neutrophils Transforming Them into Inducers of High-Level Dend
http://www.100md.com
免疫学杂志 2005年第8期
Abstract
Neutrophils play a critical role in early immunity to many microbial pathogens, and this may in part be due to their ability to release immunoregulatory cytokines and chemokines during infection. Here, we demonstrate by flow cytometric analysis that mouse polymorphonuclear leukocytes (PMN) up-regulate surface expression of TNF- within 10 min of stimulation with LPS, and that this is followed by gradual loss over a period of 18 h. Early increases in surface TNF- expression correlated with loss of intracellular pools of preformed TNF-. Nevertheless, extended incubation with LPS resulted in increased levels of TNF- mRNA synthesis and replenishment of intracellular cytokine. After triggering with LPS, PMN acquired the ability to induce dendritic cell (DC) TNF- and IL-12 production. Transwell assays demonstrated that high-level DC TNF- production induced by LPS-triggered neutrophils was dependent upon cell-to-cell contact and neutrophil TNF-, but neither was required for neutrophil instruction of DC IL-12 synthesis. The data suggest that microbial Ag-triggered mouse PMN acquire the capacity to deliver potent DC-activating signals through elaboration of cytokines and direct interactions at the cell surface.
Introduction
Cellular components of innate immunity are critical in sensing infection and initiating responses leading to long-term protection through acquired immunity (1). Dendritic cells (DC), 3 through their ability to capture Ag, migrate to secondary lymphoid organs, present antigenic peptide with costimulation, and release proinflammatory cytokines, are important in driving Th1 differentiation (2, 3, 4). Nevertheless, the cellular and molecular events involved in DC activation during microbial infection are not fully understood.
Neutrophils are often the first cell type recruited to site of inflammation, and they are considered to be terminally differentiated effectors involved in phagocytosis and intracellular killing of pathogens (5, 6). More recent evidence demonstrates the ability of PMN to produce several important proinflammatory cytokines and chemokines (7, 8). In a Toxoplasma gondii infection model, we previously demonstrated that parasite-triggered mouse neutrophils secrete DC and macrophage chemotactic factors including CCL2, CCL3, CCL4, CCL5, and CCL20 (9, 10). PMN coincubated with live tachyzoites also release factors that induce DC up-regulation of costimulatory molecules and secretion of IL-12 as well as TNF- (10, 11, 12) Mice lacking CXCR2, a chemokine receptor involved in neutrophil recruitment, display increased susceptibility to Toxoplasma (13). Furthermore, mice depleted of Gr-1+ cells are unable to survive acute toxoplasmosis (12, 14, 15). This lack of resistance is associated with defective type 1 cytokine responses, in particular decreased TNF- and IL-12 expression by splenic DC (10, 16). These results suggest that PMN play a role in orchestrating early immunity through production of DC-activating factors.
To further clarify the role of neutrophils as immunomodulatory cells in responses driven by microbial Ag, we investigated the ability of mouse PMN to trigger DC activation after stimulation with LPS. Use of LPS rather than a highly complex microbial lysate such as soluble tachyzoite Ag, which we employed in previous studies, provides the advantage that a single molecule with a single receptor (TLR4) is involved, facilitating dissection of PMN-DC interactions. LPS induces neutrophil activation both in vivo and in vitro, as measured by release of cytokines, degranulation, increased adhesion potential, and oxidative burst activity (17, 18, 19, 20, 21).
In this study, we present data showing that mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface upon LPS stimulation. Transmembrane TNF- expression is accompanied by up-regulation of surface CD40L and CD11b/CD18. LPS-triggered mouse neutrophils are potent inducers of DC TNF- when cells are placed in direct contact with each other. This response itself is dependent upon PMN TNF- expression. Endotoxin-activated mouse neutrophils also induce DC IL-12 production, but here neither PMN TNF- nor direct contact was required. The data suggest microbial Ag endow PMN with potent DC-activating capability, a property dependent upon increased expression of neutrophil membrane TNF-.
Materials and Methods
Mice
C57BL/6, TNF receptor (TNFR) 1/2 double knockout (TNFR1/2–/–), TNF- knockout and wild-type (WT) counterparts (B6129SF2/J) were obtained from The Jackson Laboratory. C3H/HeN and C3H/HeJ were purchased from Taconic Farms. Female mice between 6 and 8 wk of age were used throughout. Genotype of knockout animals was confirmed by RT-PCR according to protocols obtained from The Jackson Laboratory. Animals were housed in filter-covered isolator cages in the animal facility of the College of Veterinary Medicine at Cornell University, which is accredited by the American Association for Accreditation of Laboratory Care.
Reagents and Ab
Ultrapure LPS (Escherichia coli strain 0111:B4) was purchased from List Biological Laboratories. FITC-conjugated goat anti-hamster IgG Ab was obtained from eBioscience. Allophycocyanin-conjugated Ab specific for Ly6G, PE-conjugated Ab specific for TNF- (MP6-XT22), CD40L (MR1), CD11b (M1/170), CD14 (rmC5-3), TNFR type II (p75) (TR75-89), purified rat anti-mouse TNF- (MP6-XT22), and hamster anti-mouse TNF- (TN3-19.12) were obtained from BD Pharmingen. Purified antiserum specific for CD68 (T-16) and PE-conjugated TNF- anti-goat Ab were obtained from Santa Cruz Biotechnology. Normal hamster Ig and normal mouse serum (NMS) were purchased from Jackson ImmunoResearch Laboratories. Cycloheximide was obtained from Calbiochem, and recombinant mouse GM-CSF was purchased from PeproTech.
Bone marrow PMN
Neutrophils were isolated from mouse bone marrow following a previously published protocol (22). Briefly, single cell suspensions of bone marrow cells were collected from femur and tibia and resuspended in DMEM supplemented with 5% FCS (Hyclone Laboratories), 100 U/ml penicillin, and 100 mg/ml streptomycin (PenStrep; Invitrogen Life Technologies). Cells were then centrifuged at 500 x g for 7 min at 4°C and resuspended in Ca2+-free HBSS supplemented with 0.38% sodium citrate. The cell suspension was layered on top of a step gradient consisting of 52, 65, and 75% Percoll diluted in Ca2+-free HBSS, and centrifuged at 1500 x g for 30 min at 4°C. Neutrophils were recovered at the interface between the 65 and 75% Percoll layers. The proportion of neutrophils, determined by Diff-Quik staining of cytospin preparations, was routinely >90%.
Peritoneal neutrophils
Mice were i.p. injected with 1 ml of 10% thioglycolate (Difco). Peritoneal exudate cells were obtained by lavage 18 h later with ice-cold PBS. Cells were washed in PBS, passed through a 70-μm nylon cell strainer, and erythrocytes in the suspension were lysed using Red cell lysis buffer (Sigma-Aldrich). Cells were then washed, resuspended at 2 x 107 cells/ml in buffer composed of Dulbecco’s PBS (Invitrogen Life Technology) containing 0.5% BSA (Sigma-Aldrich), 1 mM EDTA (Fisher Scientific), and incubated for 15 min at 4°C with anti-MHC class II magnetic microbeads. After washing, the mixture was transferred to columns installed within a magnetic apparatus to remove MHC class II-expressing cells, according to the manufacturer’s instructions (Miltenyi Biotec).
Bone marrow-derived DC (BMDC)
Generation of BMDC was accomplished following a previously published protocol (23). Briefly, single cell bone marrow preparations were obtained as described above, cells were washed in RPMI 1640 (Fisher Scientific) and resuspended at 2 x 105 cells/ml in DC medium composed of RPMI 1640 supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin, 10% FCS, 5 x 10–5 M 2-ME, and 20 ng/ml GM-CSF. Cells were plated on 100 x 15 mm standard sterile polystyrene petri dishes (Fisher Scientific) and cultured for 9 days at 37°C in 5% CO2. Fresh DC medium, containing 20 ng/ml GM-CSF, was added on days 3, 6, and 8 after culture initiation. On day 9, cells were resuspended in cDMEM composed of DMEM (Invitrogen Life Technologies) supplemented with 10% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin, 5 x 10–5 M 2-ME, 10 mM HEPES (Invitrogen Life Technologies), 100 μM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), alone or in the presence of different stimuli at 37°C in 5% CO2. Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C.
PMN culture
Peritoneal PMN (2 x 106/ml) were cultured in cDMEM alone or in the presence of 1 μg/ml LPS at 37°C in 5% CO2 in a 96-well tissue culture plate (Corning Costar). Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C. In the coculture experiments, neutrophils were cultured in medium or LPS (1 μg/ml) for 3 h followed by three washes; then they were either plated together with immature BMDC in 12-well tissue culture plates (Corning Costar) or placed in the upper well of a Transwell system (Corning Costar), while BMDC were added to the lower well. In some experiments PMN were fixed for 15 min on ice in 1% formaldehyde solution. In experiments involving blocking of new protein synthesis, neutrophils were incubated in medium containing 30 μg/ml cycloheximide for 15 min followed by a stimulation with 1 μg/ml LPS.
RT-PCR
RNA was isolated, reverse transcribed and subjected to PCR amplification as described (12). The primer sequences used were: -actin, TGACGGGGGTCACCCACACTGTGCCCATCTA (sense), CTAGAAGCATTGCGGTGGACGATGGAGGG (antisense); TNF-, CAGCCTCTTCTCATTCCTGCTTGTG (sense), CTGGAAGACTCCTCCCAGGTATAT (antisense). The cDNA was amplified 31 cycles.
Flow cytometry
To analyze surface markers on PMN, Fc receptors were blocked in FACS buffer (PBS, 1% BSA, and 0.1% sodium azide) containing 5 μg/ml anti-mouse CD16/CD32 (BD Pharmingen) and 10% NMS for 15 min at 0°C, then cells were stained with optimal concentrations of allophycocyanin-conjugated anti-Gr-1(Ly-6G) in combination with PE-conjugated antisera specific for TNF-, CD40L, CD11b, CD14, TNFR type II (p75), and unlabeled goat anti-CD68 followed by PE-conjugated anti-goat Ab for 30 min at 0°C. For intracellular cytokine detection, peritoneal cells were blocked as described above, stained with allophycocyanin-conjugated anti-Gr-1 Ab and unlabeled anti-TNF- Ab, then cells were fixed in 3% paraformaldehyde (Sigma-Aldrich), 0.1 mM CaCl2, and 0.1 mM MgCl2 for 30 min at 0°C. Cells were subsequently washed in permeabilization buffer (PBS with 0.075% saponin) and incubated for 15 min at 0°C in permeabilization buffer containing 10% NMS. After two washes in permeabilization buffer, unlabeled hamster anti-TNF- was added, and cells were incubated for 30 min at 0°C. After washing in permeabilization buffer, FITC-conjugated anti-hamster was added followed by a 30-min incubation at 0°C and subsequently washed for flow cytometric analysis. Data was acquired on FACSCalibur system (10,000 events per sample) and analyzed with CellQuest software (BD Biosciences Immunocytometry Systems).
Cytokine ELISA
IL-12p40 was measured as previously described (12), and TNF- was measured using a commercially obtained kit (BD Pharmingen).
Statistical Analysis
The statistical significance of the data was analyzed using SD and unpaired two-tailed Student’s t test.
Results
LPS triggers rapid changes in peritoneal PMN surface membrane-associated molecules
We began by examining the influence of endotoxin on expression of neutrophil cell surface markers. Flow cytometric analysis revealed early LPS-induced up-regulation of surface TNF-, CD40L, and CD11b, followed by a time-dependent down-regulation relative to cells cultured in medium alone (Fig. 1A). The surface expression of these molecules in medium alone at 30 min was equivalent to staining of freshly isolated, nonstimulated peritoneal neutrophils (data not shown). We also examined expression of CD14, TNFR type II, and CD68, which are markers of exocytosis of secretory vesicles, specific and azurophilic granules, respectively (24, 25, 26, 27). Both CD14 and CD68 were rapidly mobilized to the cell surface, consistent with exocytosis of secretory vesicles and azurophilic granules. Interestingly, the influence of LPS on TNFR type II expression was distinct, in that LPS triggered early down-regulation. This was followed by a gradual increase in TNFR type II surface expression that required 18 h to reach maximal levels (Fig. 1B). These effects are likely mediated by TLR4, because increased expression of TNF- and other molecules did not occur in peritoneal PMN from the LPS nonresponder strain C3H/HeJ (Fig. 1C and data not shown).
FIGURE 1. LPS triggers rapid changes in the pattern of molecules expressed on the neutrophil surface. Thioglycolate-elicited peritoneal exudate cells from C57BL/6 (A and B), and C3H/HeN/C3H/HeJ (C) mice were stimulated with LPS (thick lines) or incubated in medium (thin lines) for varying times, then stained with Ab to Gr-1 and the indicated markers. The data show expression profiles gated on Gr-1high cells. C, Surface TNF- expression 60 min after treatment with LPS or medium. The numbers in A and C indicate the percentage of positive cells after stimulation with LPS (top) vs incubation in medium (bottom). This experiment was repeated three times with essentially identical results.
Control of neutrophil transmembrane TNF-
We next focused on surface expression of TNF- by peritoneal PMN. TNF- is synthesized as a transmembrane 26-kDa protein that is transported to the cell surface where it is cleaved by a metalloproteinase (TACE) to release the soluble 17-kDa protein (28, 29, 30, 31, 32, 33, 34). Therefore, increased TNF- levels on the cell surface (Fig. 1A) could be the result of higher levels of transmembrane cytokine or, alternatively, the result of soluble TNF- binding to neutrophil TNF receptors. To distinguish these possibilities, we stimulated WT or TNFR1/2–/– peritoneal neutrophils with LPS and analyzed surface expression of TNF-. Both WT (Fig. 2A) and TNFR1/2–/– (Fig. 2B) peritoneal PMN up-regulated surface TNF- expression after 1 h of stimulation with LPS. This result argues that transmembrane 26-kDa TNF- is up-regulated on the PMN cell surface, rather than a situation in which soluble TNF- binds to neutrophil TNF receptors. In addition, preincubation of cells with unlabeled anti-TNF- Ab prevented binding of the PE-conjugated Ab, confirming the specificity of Ab binding (Fig. 2C).
FIGURE 2. LPS induces neutrophil surface TNF- expression independently of TNF receptors. Thioglycolate-induced peritoneal exudate cells from WT (A and C) and TNFR1/2–/– (B) mice were cultured in medium (thin lines) or stimulated with LPS (thick lines) for 1 h and stained for surface TNF-. C, Cells were stained with an unlabeled blocking anti-TNF- Ab followed by staining for TNF- as in A and B. The profiles shown are gated on the Gr-1high population.
We previously found that mouse PMN contain an intracellular pool of IL-12 (35). Because we observed surface expression of TNF- very early after LPS stimulation in our experiments, we investigated the possibility that mouse PMN also contain preformed TNF-. To address this question, we first pretreated peritoneal PMN with cycloheximide to block new protein synthesis before stimulation of cells with LPS. Cycloheximide had no effect on early LPS-induced up-regulation of surface TNF-. However, after 1 h of LPS stimulation, cycloheximide-treated cells displayed a marginal but reproducible decrease in TNF- surface expression, suggesting that de novo protein synthesis may contribute to surface expression at later time points (Fig. 3A). To further examine the presence of preformed TNF-, we blocked surface TNF-, permeabilized cells, and stained for intracellular cytokine. In this case, bone marrow and peritoneal exudate PMN were strongly positive for intracellular TNF- (Fig. 3B).
FIGURE 3. Mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface after LPS stimulation. A, Early LPS-triggered PMN surface TNF- expression does not require new protein synthesis. Thioglycolate-elicited cells (PEC) were preincubated with cycloheximide then stimulated for the indicated times with LPS. Cells were stained with Ab to Gr-1 and TNF-. The profiles shown are gated on Gr-1high cells. Green line, LPS with no cycloheximide; red line, LPS with cycloheximide; black line, medium with cycloheximide. No difference was observed between medium and medium with cycloheximide (data not shown). B, PMN contain intracellular TNF-. Bone marrow cells and PEC were subjected to surface TNF- blocking using unlabeled rat anti-mouse TNF- Ab, then permeabilized and stained with hamster anti-mouse TNF- Ab. Data shown are gated on Gr-1high cells. Red line, anti-TNF- Ab; black line, isotype control. C, Changes in intracellular and membrane bound TNF- during LPS stimulation. Thioglycolate-elicited cells were incubated in medium or stimulated with LPS for the indicated times. Cells were subsequently stained with Gr-1 Ab and surface TNF- using a rat anti-mouse TNF-. After permeabilization and blocking, cells were stained for intracellular TNF- using a hamster anti-mouse Ab. Profiles shown are gated on Gr-1high cells. Green and gray lines, intracellular TNF- expression after incubation with LPS and medium, respectively. Red and black lines, surface TNF- expression after incubation with LPS and medium, respectively. Dotted lines, isotype control staining. Numbers in each histogram show percentage of positive followed by geometric mean fluorescence intensity for LPS-stimulated cells (top) and cells incubated in medium alone (bottom). D, RT-PCR analysis of purified peritoneal exudate PMN incubated in medium alone or stimulated for the indicated times with LPS.
We next examined modulation of surface vs intracellular expression of TNF- in LPS-stimulated peritoneal PMN. At 10 min after LPS stimulation, when compared with unstimulated cells, we observed a decrease in intracellular TNF- and a parallel increase in surface expression of this cytokine (Fig. 3C). However, after 1 h of exposure to LPS, both surface and intracellular TNF- expression were increased (Fig. 3C). These results suggest that de novo protein synthesis replenishes the pool of intracellular TNF- that is initially mobilized to the neutrophil surface. Evidence that LPS induces increased TNF- gene transcription is shown by RT-PCR in Fig. 3D. Increased levels of TNF- mRNA were apparent within 15 min, and maximal levels were attained by 60 min of stimulation. The combined results argue that both preformed pools and newly synthesized TNF- contribute to high level expression of this cytokine on the cell surface.
LPS stimulated mouse peritoneal neutrophils promote dendritic cell cytokine production
To determine whether LPS-triggered peritoneal PMN can directly stimulate TNF- and IL-12 production by BMDC, we began by measuring cytokine levels in PMN-DC cocultures. When BMDC were cultured with LPS or cocultured with LPS-activated peritoneal PMN, high levels of both TNF- and IL-12p40 were released into the coculture supernatants (Fig. 4A). In contrast, when nonstimulated mouse peritoneal PMN were used in the cocultures, no TNF- and minimal amounts of IL-12p40 were detected (Fig. 4A). To test whether the triggering effect of peritoneal PMN was due to bona fide factors produced by the latter cells or simply to LPS carryover, we conducted identical experiments employing BMDC derived from LPS nonresponder animals. In this case, although direct DC stimulation with LPS failed to elicit a cytokine response, coculture with LPS-stimulated peritoneal neutrophils triggered robust TNF- and IL-12p40 responses (Fig. 4B). This result demonstrates that the cytokine responses were triggered by neutrophil factors rather than residual LPS in the cocultures.
FIGURE 4. LPS-triggered PMN express factors that trigger DC TNF- and IL-12 production. BMDC were incubated with medium (open bars), LPS (black bars; 1 μg/ml), unstimulated PMN (gray bars), and LPS-stimulated PMN (cross-hatched bars). The ratio of PMN to DC was 2:1. After 18 h of incubation (37°C, 5% CO2), DC culture supernatants were collected for TNF- and IL-12p40 ELISA. A, BMDC from C3H/HeN (LPS responder) and B, BMDC from C3H/HeJ (LPS nonresponder) mice. Levels of cytokine detected in PMN culture supernatants after 3 h incubation in medium alone or in the presence of LPS followed by 18 h in medium, respectively, were: 31 ± 2 pg/ml; 109 ± 7 pg/ml for TNF- and <1 pg/ml and 147 ± 1 pg/ml for IL-12p40. n.d., Not detected. Data are representative of three different experiments.
Direct cell-to-cell contact and peritoneal PMN TNF- are required for optimal DC production of TNF- but not IL-12p40
We next compared the role of soluble vs membrane-bound PMN-derived factors in promoting cytokine release by BMDC. Accordingly, we exposed LPS nonresponder BMDC to cell-free supernatants from 18-h LPS-stimulated mouse peritoneal PMN. In parallel, we cocultured 3-h LPS-stimulated peritoneal PMN with the BMDC as in Fig. 4B. Although DC exposed to soluble factors from LPS-triggered peritoneal PMN released both TNF- and IL-12p40, the response was enhanced when BMDC were cocultured with LPS-stimulated peritoneal PMN (Fig. 5). This was particularly true for TNF- production, where there was a >6-fold increase in the amount of cytokine released in a direct cell-to-cell contact situation. We next used a transwell system to further probe the requirement for optimal DC cytokine production. As shown in Fig. 6A, TNF- production was severely curtailed when LPS-triggered WT mouse peritoneal PMN, and DC were separated by a membrane. In contrast, WT DC IL-12p40 production in response to LPS-stimulated WT peritoneal PMN was minimally affected (Fig. 6A).
FIGURE 5. Cell-to-cell contact is required for optimal DC TNF- but not IL-12p40 production. C3H/HeJ (LPS nonresponder) BMDC were incubated 18 h with supernatants from unstimulated (open bars) or 18 h with LPS-stimulated (black bars) PMN. In addition, DC were coincubated with PMN that had been exposed to medium (gray bars) or LPS (cross-hatched bars) for 3 h. A 2:1 ratio of PMN to DC was used in the coincubation experiments. After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. PMN alone stimulated with LPS released 311 ± 16 pg/ml (TNF-) and 3490 ± 139 pg/ml (IL-12p40). The experiment was repeated twice with similar results. *, p < 0.01 (DC + 18 h PMN supernatants vs DC-PMN coculture). n.d., Not detected.
FIGURE 6. Requirements for PMN triggering of DC cytokine production. Thioglycolate-elicited peritoneal PMN were preincubated 3 h with LPS, then added directly to BMDC (black bar) or separated in a Transwell system (open bar). Supernatants were collected 18 h later for TNF- and IL-12p40 ELISA. A, WT and TNF-–/– PMN were compared for ability to trigger WT DC. B, WT PMN were used to trigger WT and TNFR1/2–/– DC. C, WT neutrophils were used to trigger WT and CD40–/– DC. No cytokine release was detected when nonstimulated PMN were used in each of these experiments (data not shown). This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs Transwell). **, p < 0.01 (direct cell-to-cell contact between WT DC and WT PMN vs direct cell-to-cell contact between WT DC and TNF–/– PMN). n.d., Not detected.
We previously demonstrated that supernatants from T. gondii-triggered PMN induce DC up-regulation of costimulatory molecules CD40 and CD86 and that this response depends upon TNF- (10). In this study, we investigated the involvement of PMN-derived TNF- in promoting cytokine production in PMN-BMDC coculture experiments. Although LPS-stimulated WT peritoneal PMN were able to trigger TNF- production by BMDC, LPS-stimulated TNF–/– neutrophils were defective in driving such a response (Fig. 6A). However, both LPS-stimulated WT and TNF–/– peritoneal PMN were able to promote DC IL-12p40 production in cocultures and when separated in a Transwell chamber system (Fig. 6A). Similarly, using TNFR1/2–/– DC, we found that LPS-stimulated peritoneal PMN no longer elicited DC TNF- release while leaving IL-12p40 production unaffected (Fig. 6B). Last, DC lacking CD40 were examined. Here, there was no evidence for involvement of this costimulatory molecule in DC cytokine production (Fig. 6C). Collectively, the results in Fig. 6 show that direct cell-to-cell contact is required for optimal TNF- but not IL-12p40 production, and that DC TNF- but not IL-12p40 release depends upon neutrophil TNF- production.
Fixed mouse peritoneal neutrophils retain the ability to stimulate DC TNF- release
The results presented so far indicate that cell-to-cell contact between DC and PMN directs TNF- production by DC, but they do not rule out additional involvement of neutrophil soluble factors. To address this issue, peritoneal neutrophils were subjected to paraformaldehyde fixation before DC stimulation. This procedure results in loss of ability of PMN to secrete factors, but surface membrane molecules will be retained. As shown in Fig. 7A, fixation had no effect on the ability of LPS-triggered peritoneal neutrophils to drive DC TNF- production. In contrast, fixation ablated the ability of PMN to drive DC IL-12p40 production, consistent with a role for soluble factors in this situation (Fig. 7B). These results demonstrate that direct contact between DC and PMN is necessary and sufficient to promote TNF- production by DC, but that soluble factors induce DC IL-12 production.
FIGURE 7. LPS-stimulated fixed mouse neutrophils retain the ability to stimulate DC TNF- production. Peritoneal PMN were incubated for 3 h in medium or LPS, washed, fixed or left untreated, and added directly to BMDC (black bars) or separated in a Transwell system (open bars). After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. Untreated PMN alone stimulated with LPS released 143 ± 1 pg/ml (TNF-) and 395 ± 2 pg/ml (IL-12). No cytokine release was detected when nonstimulated PMN were used (data not shown). In addition, no cytokine was detected in supernatants of LPS-stimulated fixed neutrophils, and no DC IL-12 was detected using fixed neutrophils as stimulators. This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs transwell). n.d., Not detected.
Discussion
The results of this study demonstrate that mouse neutrophils triggered with microbial Ag become licensed activators of DC. Although PMN themselves serve as a source of proinflammatory cytokines IL-12p40 and TNF-, the amount of cytokine released by neutrophil-activated DC is 1–2 orders of magnitude higher (Figs. 4–6). Importantly, by using DC with nonfunctional TLR4, it was possible to rule out involvement of endotoxin carryover in the effects of LPS-triggered neutrophils on DC. The DC TNF- release was triggered by neutrophil TNF-, and this was optimally driven by direct cell-to-cell contact.
Neutrophil TNF- was mobilized to the membrane surface within 15 min of endotoxin stimulation relative to cells incubated for the same length of time in medium alone. Surface TNF- up-regulation occurred in the presence of cycloheximide indicating that de novo protein synthesis was not required. Furthermore, nonstimulated mouse neutrophils contained an intracellular pool of TNF- that displayed a transient decrease concomitant with increased surface TNF- expression. In addition to TNF-, PMN have previously been shown to contain preformed IL-12, IL-6, and MIP-2 (35, 36, 37). Both TNF- and IL-12 are present in non-elicited mouse neutrophils. It is, therefore, likely that these cytokines are expressed in PMN during the course of differentiation in the bone marrow. Thus, in addition to their well-known role as microbicidal effectors, neutrophils may also serve an important function as vehicles that rapidly deliver preformed cytokines to sites of inflammation and infection.
Whereas intracellular TNF- pools were depleted 10 min after LPS stimulation, after 60 min, intracellular TNF- was replenished. LPS triggering also induced increased levels of TNF- mRNA. The data argue that, although mouse PMN possess an intracellular pool of preformed cytokine, the cells also display the ability to up-regulate cytokine gene induction, as shown here for TNF- and elsewhere for IL-12 (35). Sustained PMN production of inflammatory cytokines, particularly for the case of GM-CSF, is likely to be important for delaying onset of apoptosis that these cells are programmed to undergo (38, 39). In this way, immunoregulatory and microbicidal effects of neutrophils at sites of inflammation and infection are likely to be prolonged.
We, and others, have suggested that neutrophils, by virtue of their pattern of cytokine expression, may have important in vivo immunoregulatory effects on DC (10, 40, 41, 42). In addition to triggering DC TNF- and IL-12 production, activated PMN exert other effects on these cells. For example, mouse neutrophil release of chemokines CCL3, CCL4, CCL5, and CCL20 together mediate potent chemotactic activity on immature DC (10). Furthermore, PMN induce potent DC up-regulation of costimulatory molecules such as CD40 and CD86, an effect that is in part dependent upon neutrophil TNF- release (10). Other recent studies show that neutrophils assume type 1 and type 2 cytokine profiles, depending on the activating stimulus (43, 44, 45). Together, these findings provide evidence for a key role for neutrophils in controlling immunoregulatory effects of DC.
An unexpected conclusion emerging from our studies is that neutrophil-directed production of DC TNF- and IL-12 is non-identical. Thus, DC IL-12 production does not rely on neutrophil TNF-. In contrast, DC TNF- production requires TNF- expression by peritoneal neutrophils. Furthermore, PMN-induced release of DC IL-12, in contrast to TNF-, occurs equally efficiently across a transwell membrane relative to when cells are in direct contact. CD40 ligation is well known to be involved in DC activation (46), and indeed we show here that LPS induces up-regulation of PMN CD40L expression. Nevertheless, using CD40–/– DC, we were able to exclude a role for CD40:CD40L in effects of PMN on DC. Likewise, IL-12 has been implicated as a priming signal for high-level DC IL-12 production (47), but experiments employing IL-12–/– PMN demonstrate lack of involvement of this cytokine (unpublished observations). The potential role of other neutrophil-derived IL-12-inducing mediators, for example bradykinin, is currently under investigation (48).
Our studies here and elsewhere (9, 10, 12) employ thioglycolate-elicited peritoneal and bone marrow neutrophils. In principle, these populations are non-identical since the former has responded to an inflammatory stimulus. Nevertheless, we have yet to detect differences in these populations, particularly with regard to cytokine/chemokine production. In addition, it can be argued that elicited PMN are the most relevant to study because neutrophils arriving at a site of infection do so in response to inflammatory signals.
The biological significance of DC TNF- production is presently unclear. The cells are more well-known as an IL-12 source that, together with Ag and costimulation, play a central role in the generation of Th1 effector lymphocytes. Nevertheless, DC subsets have been identified in vivo that mediate innate immune defense against microbial infection by producing both NO and TNF- (49). The role of TNF- in this situation may be to promote, along with other inflammatory mediators, macrophage and DC acquisition of microbicidal function.
In vivo evidence for the importance of neutrophils in DC function is provided by observations that depletion of Gr-1(Ly6G)+ cells results in decreased TNF- and IL-12 production by splenic CD11c+ cells and increased susceptibility during infection with microbial pathogens such as T. gondii (10). In this regard, activated PMN that rapidly accumulate at a site of infection or inflammation may play an important role in driving acquisition of DC microbicidal function. The fact that direct cell contact is required for optimal DC TNF- production may be a way to focus the response solely at inflammatory sites of infection and neutrophil accumulation.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Barbara Butcher, Laura Del Rio, and Matthias Hesse for constructive criticism and insightful discussion during the course of this work.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI47888.
2 Address correspondence and reprint requests to Dr. E. Denkers, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853-6401. E-mail address: eyd1{at}cornell.edu
3 Abbreviations used in this paper: DC, dendritic cell; PMN, polymorphonuclear leukocyte; BMDC, bone marrow-derived DC; NMS, normal mouse serum; WT, wild type.
Received for publication October 12, 2004. Accepted for publication January 10, 2005.
References
Janeway, C. A., Jr, R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197.
Palucka, K., J. Banchereau. 2002. How dendritic cells and microbes interact to elicit or subvert protective immune responses. Curr. Opin. Immunol. 14:420.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Belaaouaj, A., R. McCarthy, M. Baumann, Z. Gao, T. J. Ley, S. N. Abraham, S. D. Shapiro. 1998. Mice lacking neutrophil elastase reveal impaired host defense against gram negative bacterial sepsis. Nat. Med. 4:615.
Jackson, S. H., J. I. Gallin, S. M. Holland. 1995. The p47phox mouse knockout model of chronic granulomatous disease. J. Exp. Med. 182:751.
Cassatella, M. A.. 1999. Neutrophil-derived proteins: selling cytokines by the pound. Adv. Immunol. 73:369
Del Rio, L., B. A. Butcher, S. Bennouna, S. Hieny, A. Sher, E. Y. Denkers. 2004. Toxoplasma gondii triggers MyD88-dependent and CCL2 (MCP-1) responses using distinct parasite molecules and host receptors. J. Immunol. 172:6954.
Bennouna, S., S. K. Bliss, T. J. Curiel, E. Y. Denkers. 2003. Cross-talk in the innate immune system: neutrophils instruct early recruitment and activation of dendritic cells during microbial infection. J. Immunol. 171:6052.
Bliss, S. K., A. J. Marshall, Y. Zhang, E. Y. Denkers. 1999. Human polymorphonuclear leukocytes produce IL-12, TNF-, and the chemokines macrophage-inflammatory protein-1 and -1 in response to Toxoplasma gondii antigens. J. Immunol. 162:7369
Bliss, S. K., Y. Zhang, E. Y. Denkers. 1999. Murine neutrophil stimulation by Toxoplasma gondii antigen drives high level production of IFN--independent IL-12. J. Immunol. 163:2081.
Del Rio, L., S. Bennouna, J. Salinas, E. Y. Denkers. 2001. CXCR2 deficiency confers impaired neutrophil recruitment and increased susceptibility during Toxoplasma gondii infection. J. Immunol. 167:6503.
Scharton-Kersten, T., G. Yap, J. Magram, A. Sher. 1997. Inducible nitric oxide is essential for host control of persistent but not acute infection with the intracellular pathogen Toxoplasma gondii. J. Exp. Med. 185:1.
Sayles, P. C., L. J. Johnson. 1997. Exacerbation of toxoplasmosis in neutrophil depleted mice. Nat. Immun. 15:249.
Bliss, S. K., L. C. Gavrilescu, A. Alcaraz, E. Y. Denkers. 2001. Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69:4898.
Wright, S. D., R. A. Ramos, A. Hermanowski-Vosatka, P. Rockwell, P. A. Detmers. 1992. Activation of the adhesive capability of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J. Exp. Med. 173:1281.
Guthrie, L. A., L. C. McPhail, P. M. Henson, R. B. J. Johnston. 1984. Priming of neutrophils enhanced release of oxygen metabolites by bacterial lipopolysaccharide: evidence for increased activity of the superoxide-producing enzyme. J. Exp. Med. 160:1656.
Surette, M. E., R. Palmantier, J. Gosselin, P. Borgeat. 1993. Lipopolysaccharides prime whole human blood and isolated neutrophils for the increased synthesis of 5-lipoxygenase products by enhancing arachidonic acid availability: involvement of CD14 antigen. J. Exp. Med. 178:1347.
Dubravec, D. B., D. R. Spriggs, J. A. Mannick, M. L. Rodrick. 1990. Circulating human peripheral blood granulocytes synthesize and secrete tumor necrosis factor . Proc. Natl. Acad. Sci. USA 87:6758
Cassatella, M. A., L. Meda, S. Gasperini, A. D‘Andrea, X. Ma, G. Trinchieri. 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25:1
Kiefer, F., J. Brumell, N. Al-Alawi, S. Latour, A. Cheng, A. Veillette, S. Grinstein, T. Pawson. 1998. The Syk protein tyrosine kinase is essential for Fc receptor signaling in macrophages and neutrophils. Mol. Cell Biol. 1998. 18:4209.
MacDonald, A. S., A. D. Straw, B. Bauman, E. J. Pearce. 2001. CD8– dendritic cell activation status plays an integral role in influencing Th2 response development. J. Immunol. 167:1982.
Borregaard, N., J. B. Cowland. 1997. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89:3503.
Bainton, D. F.. 1999. Distinct granule populations in human neutrophils and lysosomal organelles identified by electron microscopy. J. Immunol. Methods 232:153.
Gullberg, U., N. Bengtsson, E. Bulow, E. Garwicz, A. Lindmark, I. Olsson. 1999. Processing and targeting of granule proteins in human neutrophils. J. Immunol. Methods 232:201.
Beutler, B., A. Cerami. 1989. The biology of cachectin/TNF- primary mediator of the host response. Annu. Rev. Immunol. 7:625.
Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, et al 1997. A metalloproteinase disintegrin that releases tumor-necrosis factor- from cells. Nature 385:729.
Grohmann, U., M. L. Belladonna, R. Bianchi, C. Orabona, E. Ayroldi, M. C. Fioretti, P. Puccetti. 1998. IL-12 acts directly on DC to promote nuclear localization of NF-B and primes DC for IL-12 production. Immunity 9:315.
Aliberti, J., J. P. Viola, A. Vieria-de-Abreu, P. T. Bozza, A. Sher, J. Scharfstein. 2003. Cutting Edge: bradykinin induces IL-12 production by dendritic cells: a danger signal that drives Th1 polarization. J. Immunol. 170:485.
Serbina, N. V., T. P. Salazar-Mathar, C. A. Biron, W. A. Kuziel, E. A. Pamer. 2003. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19:59.
Abstract
Neutrophils play a critical role in early immunity to many microbial pathogens, and this may in part be due to their ability to release immunoregulatory cytokines and chemokines during infection. Here, we demonstrate by flow cytometric analysis that mouse polymorphonuclear leukocytes (PMN) up-regulate surface expression of TNF- within 10 min of stimulation with LPS, and that this is followed by gradual loss over a period of 18 h. Early increases in surface TNF- expression correlated with loss of intracellular pools of preformed TNF-. Nevertheless, extended incubation with LPS resulted in increased levels of TNF- mRNA synthesis and replenishment of intracellular cytokine. After triggering with LPS, PMN acquired the ability to induce dendritic cell (DC) TNF- and IL-12 production. Transwell assays demonstrated that high-level DC TNF- production induced by LPS-triggered neutrophils was dependent upon cell-to-cell contact and neutrophil TNF-, but neither was required for neutrophil instruction of DC IL-12 synthesis. The data suggest that microbial Ag-triggered mouse PMN acquire the capacity to deliver potent DC-activating signals through elaboration of cytokines and direct interactions at the cell surface.
Introduction
Cellular components of innate immunity are critical in sensing infection and initiating responses leading to long-term protection through acquired immunity (1). Dendritic cells (DC), 3 through their ability to capture Ag, migrate to secondary lymphoid organs, present antigenic peptide with costimulation, and release proinflammatory cytokines, are important in driving Th1 differentiation (2, 3, 4). Nevertheless, the cellular and molecular events involved in DC activation during microbial infection are not fully understood.
Neutrophils are often the first cell type recruited to site of inflammation, and they are considered to be terminally differentiated effectors involved in phagocytosis and intracellular killing of pathogens (5, 6). More recent evidence demonstrates the ability of PMN to produce several important proinflammatory cytokines and chemokines (7, 8). In a Toxoplasma gondii infection model, we previously demonstrated that parasite-triggered mouse neutrophils secrete DC and macrophage chemotactic factors including CCL2, CCL3, CCL4, CCL5, and CCL20 (9, 10). PMN coincubated with live tachyzoites also release factors that induce DC up-regulation of costimulatory molecules and secretion of IL-12 as well as TNF- (10, 11, 12) Mice lacking CXCR2, a chemokine receptor involved in neutrophil recruitment, display increased susceptibility to Toxoplasma (13). Furthermore, mice depleted of Gr-1+ cells are unable to survive acute toxoplasmosis (12, 14, 15). This lack of resistance is associated with defective type 1 cytokine responses, in particular decreased TNF- and IL-12 expression by splenic DC (10, 16). These results suggest that PMN play a role in orchestrating early immunity through production of DC-activating factors.
To further clarify the role of neutrophils as immunomodulatory cells in responses driven by microbial Ag, we investigated the ability of mouse PMN to trigger DC activation after stimulation with LPS. Use of LPS rather than a highly complex microbial lysate such as soluble tachyzoite Ag, which we employed in previous studies, provides the advantage that a single molecule with a single receptor (TLR4) is involved, facilitating dissection of PMN-DC interactions. LPS induces neutrophil activation both in vivo and in vitro, as measured by release of cytokines, degranulation, increased adhesion potential, and oxidative burst activity (17, 18, 19, 20, 21).
In this study, we present data showing that mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface upon LPS stimulation. Transmembrane TNF- expression is accompanied by up-regulation of surface CD40L and CD11b/CD18. LPS-triggered mouse neutrophils are potent inducers of DC TNF- when cells are placed in direct contact with each other. This response itself is dependent upon PMN TNF- expression. Endotoxin-activated mouse neutrophils also induce DC IL-12 production, but here neither PMN TNF- nor direct contact was required. The data suggest microbial Ag endow PMN with potent DC-activating capability, a property dependent upon increased expression of neutrophil membrane TNF-.
Materials and Methods
Mice
C57BL/6, TNF receptor (TNFR) 1/2 double knockout (TNFR1/2–/–), TNF- knockout and wild-type (WT) counterparts (B6129SF2/J) were obtained from The Jackson Laboratory. C3H/HeN and C3H/HeJ were purchased from Taconic Farms. Female mice between 6 and 8 wk of age were used throughout. Genotype of knockout animals was confirmed by RT-PCR according to protocols obtained from The Jackson Laboratory. Animals were housed in filter-covered isolator cages in the animal facility of the College of Veterinary Medicine at Cornell University, which is accredited by the American Association for Accreditation of Laboratory Care.
Reagents and Ab
Ultrapure LPS (Escherichia coli strain 0111:B4) was purchased from List Biological Laboratories. FITC-conjugated goat anti-hamster IgG Ab was obtained from eBioscience. Allophycocyanin-conjugated Ab specific for Ly6G, PE-conjugated Ab specific for TNF- (MP6-XT22), CD40L (MR1), CD11b (M1/170), CD14 (rmC5-3), TNFR type II (p75) (TR75-89), purified rat anti-mouse TNF- (MP6-XT22), and hamster anti-mouse TNF- (TN3-19.12) were obtained from BD Pharmingen. Purified antiserum specific for CD68 (T-16) and PE-conjugated TNF- anti-goat Ab were obtained from Santa Cruz Biotechnology. Normal hamster Ig and normal mouse serum (NMS) were purchased from Jackson ImmunoResearch Laboratories. Cycloheximide was obtained from Calbiochem, and recombinant mouse GM-CSF was purchased from PeproTech.
Bone marrow PMN
Neutrophils were isolated from mouse bone marrow following a previously published protocol (22). Briefly, single cell suspensions of bone marrow cells were collected from femur and tibia and resuspended in DMEM supplemented with 5% FCS (Hyclone Laboratories), 100 U/ml penicillin, and 100 mg/ml streptomycin (PenStrep; Invitrogen Life Technologies). Cells were then centrifuged at 500 x g for 7 min at 4°C and resuspended in Ca2+-free HBSS supplemented with 0.38% sodium citrate. The cell suspension was layered on top of a step gradient consisting of 52, 65, and 75% Percoll diluted in Ca2+-free HBSS, and centrifuged at 1500 x g for 30 min at 4°C. Neutrophils were recovered at the interface between the 65 and 75% Percoll layers. The proportion of neutrophils, determined by Diff-Quik staining of cytospin preparations, was routinely >90%.
Peritoneal neutrophils
Mice were i.p. injected with 1 ml of 10% thioglycolate (Difco). Peritoneal exudate cells were obtained by lavage 18 h later with ice-cold PBS. Cells were washed in PBS, passed through a 70-μm nylon cell strainer, and erythrocytes in the suspension were lysed using Red cell lysis buffer (Sigma-Aldrich). Cells were then washed, resuspended at 2 x 107 cells/ml in buffer composed of Dulbecco’s PBS (Invitrogen Life Technology) containing 0.5% BSA (Sigma-Aldrich), 1 mM EDTA (Fisher Scientific), and incubated for 15 min at 4°C with anti-MHC class II magnetic microbeads. After washing, the mixture was transferred to columns installed within a magnetic apparatus to remove MHC class II-expressing cells, according to the manufacturer’s instructions (Miltenyi Biotec).
Bone marrow-derived DC (BMDC)
Generation of BMDC was accomplished following a previously published protocol (23). Briefly, single cell bone marrow preparations were obtained as described above, cells were washed in RPMI 1640 (Fisher Scientific) and resuspended at 2 x 105 cells/ml in DC medium composed of RPMI 1640 supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin, 10% FCS, 5 x 10–5 M 2-ME, and 20 ng/ml GM-CSF. Cells were plated on 100 x 15 mm standard sterile polystyrene petri dishes (Fisher Scientific) and cultured for 9 days at 37°C in 5% CO2. Fresh DC medium, containing 20 ng/ml GM-CSF, was added on days 3, 6, and 8 after culture initiation. On day 9, cells were resuspended in cDMEM composed of DMEM (Invitrogen Life Technologies) supplemented with 10% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin, 5 x 10–5 M 2-ME, 10 mM HEPES (Invitrogen Life Technologies), 100 μM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), alone or in the presence of different stimuli at 37°C in 5% CO2. Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C.
PMN culture
Peritoneal PMN (2 x 106/ml) were cultured in cDMEM alone or in the presence of 1 μg/ml LPS at 37°C in 5% CO2 in a 96-well tissue culture plate (Corning Costar). Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C. In the coculture experiments, neutrophils were cultured in medium or LPS (1 μg/ml) for 3 h followed by three washes; then they were either plated together with immature BMDC in 12-well tissue culture plates (Corning Costar) or placed in the upper well of a Transwell system (Corning Costar), while BMDC were added to the lower well. In some experiments PMN were fixed for 15 min on ice in 1% formaldehyde solution. In experiments involving blocking of new protein synthesis, neutrophils were incubated in medium containing 30 μg/ml cycloheximide for 15 min followed by a stimulation with 1 μg/ml LPS.
RT-PCR
RNA was isolated, reverse transcribed and subjected to PCR amplification as described (12). The primer sequences used were: -actin, TGACGGGGGTCACCCACACTGTGCCCATCTA (sense), CTAGAAGCATTGCGGTGGACGATGGAGGG (antisense); TNF-, CAGCCTCTTCTCATTCCTGCTTGTG (sense), CTGGAAGACTCCTCCCAGGTATAT (antisense). The cDNA was amplified 31 cycles.
Flow cytometry
To analyze surface markers on PMN, Fc receptors were blocked in FACS buffer (PBS, 1% BSA, and 0.1% sodium azide) containing 5 μg/ml anti-mouse CD16/CD32 (BD Pharmingen) and 10% NMS for 15 min at 0°C, then cells were stained with optimal concentrations of allophycocyanin-conjugated anti-Gr-1(Ly-6G) in combination with PE-conjugated antisera specific for TNF-, CD40L, CD11b, CD14, TNFR type II (p75), and unlabeled goat anti-CD68 followed by PE-conjugated anti-goat Ab for 30 min at 0°C. For intracellular cytokine detection, peritoneal cells were blocked as described above, stained with allophycocyanin-conjugated anti-Gr-1 Ab and unlabeled anti-TNF- Ab, then cells were fixed in 3% paraformaldehyde (Sigma-Aldrich), 0.1 mM CaCl2, and 0.1 mM MgCl2 for 30 min at 0°C. Cells were subsequently washed in permeabilization buffer (PBS with 0.075% saponin) and incubated for 15 min at 0°C in permeabilization buffer containing 10% NMS. After two washes in permeabilization buffer, unlabeled hamster anti-TNF- was added, and cells were incubated for 30 min at 0°C. After washing in permeabilization buffer, FITC-conjugated anti-hamster was added followed by a 30-min incubation at 0°C and subsequently washed for flow cytometric analysis. Data was acquired on FACSCalibur system (10,000 events per sample) and analyzed with CellQuest software (BD Biosciences Immunocytometry Systems).
Cytokine ELISA
IL-12p40 was measured as previously described (12), and TNF- was measured using a commercially obtained kit (BD Pharmingen).
Statistical Analysis
The statistical significance of the data was analyzed using SD and unpaired two-tailed Student’s t test.
Results
LPS triggers rapid changes in peritoneal PMN surface membrane-associated molecules
We began by examining the influence of endotoxin on expression of neutrophil cell surface markers. Flow cytometric analysis revealed early LPS-induced up-regulation of surface TNF-, CD40L, and CD11b, followed by a time-dependent down-regulation relative to cells cultured in medium alone (Fig. 1A). The surface expression of these molecules in medium alone at 30 min was equivalent to staining of freshly isolated, nonstimulated peritoneal neutrophils (data not shown). We also examined expression of CD14, TNFR type II, and CD68, which are markers of exocytosis of secretory vesicles, specific and azurophilic granules, respectively (24, 25, 26, 27). Both CD14 and CD68 were rapidly mobilized to the cell surface, consistent with exocytosis of secretory vesicles and azurophilic granules. Interestingly, the influence of LPS on TNFR type II expression was distinct, in that LPS triggered early down-regulation. This was followed by a gradual increase in TNFR type II surface expression that required 18 h to reach maximal levels (Fig. 1B). These effects are likely mediated by TLR4, because increased expression of TNF- and other molecules did not occur in peritoneal PMN from the LPS nonresponder strain C3H/HeJ (Fig. 1C and data not shown).
FIGURE 1. LPS triggers rapid changes in the pattern of molecules expressed on the neutrophil surface. Thioglycolate-elicited peritoneal exudate cells from C57BL/6 (A and B), and C3H/HeN/C3H/HeJ (C) mice were stimulated with LPS (thick lines) or incubated in medium (thin lines) for varying times, then stained with Ab to Gr-1 and the indicated markers. The data show expression profiles gated on Gr-1high cells. C, Surface TNF- expression 60 min after treatment with LPS or medium. The numbers in A and C indicate the percentage of positive cells after stimulation with LPS (top) vs incubation in medium (bottom). This experiment was repeated three times with essentially identical results.
Control of neutrophil transmembrane TNF-
We next focused on surface expression of TNF- by peritoneal PMN. TNF- is synthesized as a transmembrane 26-kDa protein that is transported to the cell surface where it is cleaved by a metalloproteinase (TACE) to release the soluble 17-kDa protein (28, 29, 30, 31, 32, 33, 34). Therefore, increased TNF- levels on the cell surface (Fig. 1A) could be the result of higher levels of transmembrane cytokine or, alternatively, the result of soluble TNF- binding to neutrophil TNF receptors. To distinguish these possibilities, we stimulated WT or TNFR1/2–/– peritoneal neutrophils with LPS and analyzed surface expression of TNF-. Both WT (Fig. 2A) and TNFR1/2–/– (Fig. 2B) peritoneal PMN up-regulated surface TNF- expression after 1 h of stimulation with LPS. This result argues that transmembrane 26-kDa TNF- is up-regulated on the PMN cell surface, rather than a situation in which soluble TNF- binds to neutrophil TNF receptors. In addition, preincubation of cells with unlabeled anti-TNF- Ab prevented binding of the PE-conjugated Ab, confirming the specificity of Ab binding (Fig. 2C).
FIGURE 2. LPS induces neutrophil surface TNF- expression independently of TNF receptors. Thioglycolate-induced peritoneal exudate cells from WT (A and C) and TNFR1/2–/– (B) mice were cultured in medium (thin lines) or stimulated with LPS (thick lines) for 1 h and stained for surface TNF-. C, Cells were stained with an unlabeled blocking anti-TNF- Ab followed by staining for TNF- as in A and B. The profiles shown are gated on the Gr-1high population.
We previously found that mouse PMN contain an intracellular pool of IL-12 (35). Because we observed surface expression of TNF- very early after LPS stimulation in our experiments, we investigated the possibility that mouse PMN also contain preformed TNF-. To address this question, we first pretreated peritoneal PMN with cycloheximide to block new protein synthesis before stimulation of cells with LPS. Cycloheximide had no effect on early LPS-induced up-regulation of surface TNF-. However, after 1 h of LPS stimulation, cycloheximide-treated cells displayed a marginal but reproducible decrease in TNF- surface expression, suggesting that de novo protein synthesis may contribute to surface expression at later time points (Fig. 3A). To further examine the presence of preformed TNF-, we blocked surface TNF-, permeabilized cells, and stained for intracellular cytokine. In this case, bone marrow and peritoneal exudate PMN were strongly positive for intracellular TNF- (Fig. 3B).
FIGURE 3. Mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface after LPS stimulation. A, Early LPS-triggered PMN surface TNF- expression does not require new protein synthesis. Thioglycolate-elicited cells (PEC) were preincubated with cycloheximide then stimulated for the indicated times with LPS. Cells were stained with Ab to Gr-1 and TNF-. The profiles shown are gated on Gr-1high cells. Green line, LPS with no cycloheximide; red line, LPS with cycloheximide; black line, medium with cycloheximide. No difference was observed between medium and medium with cycloheximide (data not shown). B, PMN contain intracellular TNF-. Bone marrow cells and PEC were subjected to surface TNF- blocking using unlabeled rat anti-mouse TNF- Ab, then permeabilized and stained with hamster anti-mouse TNF- Ab. Data shown are gated on Gr-1high cells. Red line, anti-TNF- Ab; black line, isotype control. C, Changes in intracellular and membrane bound TNF- during LPS stimulation. Thioglycolate-elicited cells were incubated in medium or stimulated with LPS for the indicated times. Cells were subsequently stained with Gr-1 Ab and surface TNF- using a rat anti-mouse TNF-. After permeabilization and blocking, cells were stained for intracellular TNF- using a hamster anti-mouse Ab. Profiles shown are gated on Gr-1high cells. Green and gray lines, intracellular TNF- expression after incubation with LPS and medium, respectively. Red and black lines, surface TNF- expression after incubation with LPS and medium, respectively. Dotted lines, isotype control staining. Numbers in each histogram show percentage of positive followed by geometric mean fluorescence intensity for LPS-stimulated cells (top) and cells incubated in medium alone (bottom). D, RT-PCR analysis of purified peritoneal exudate PMN incubated in medium alone or stimulated for the indicated times with LPS.
We next examined modulation of surface vs intracellular expression of TNF- in LPS-stimulated peritoneal PMN. At 10 min after LPS stimulation, when compared with unstimulated cells, we observed a decrease in intracellular TNF- and a parallel increase in surface expression of this cytokine (Fig. 3C). However, after 1 h of exposure to LPS, both surface and intracellular TNF- expression were increased (Fig. 3C). These results suggest that de novo protein synthesis replenishes the pool of intracellular TNF- that is initially mobilized to the neutrophil surface. Evidence that LPS induces increased TNF- gene transcription is shown by RT-PCR in Fig. 3D. Increased levels of TNF- mRNA were apparent within 15 min, and maximal levels were attained by 60 min of stimulation. The combined results argue that both preformed pools and newly synthesized TNF- contribute to high level expression of this cytokine on the cell surface.
LPS stimulated mouse peritoneal neutrophils promote dendritic cell cytokine production
To determine whether LPS-triggered peritoneal PMN can directly stimulate TNF- and IL-12 production by BMDC, we began by measuring cytokine levels in PMN-DC cocultures. When BMDC were cultured with LPS or cocultured with LPS-activated peritoneal PMN, high levels of both TNF- and IL-12p40 were released into the coculture supernatants (Fig. 4A). In contrast, when nonstimulated mouse peritoneal PMN were used in the cocultures, no TNF- and minimal amounts of IL-12p40 were detected (Fig. 4A). To test whether the triggering effect of peritoneal PMN was due to bona fide factors produced by the latter cells or simply to LPS carryover, we conducted identical experiments employing BMDC derived from LPS nonresponder animals. In this case, although direct DC stimulation with LPS failed to elicit a cytokine response, coculture with LPS-stimulated peritoneal neutrophils triggered robust TNF- and IL-12p40 responses (Fig. 4B). This result demonstrates that the cytokine responses were triggered by neutrophil factors rather than residual LPS in the cocultures.
FIGURE 4. LPS-triggered PMN express factors that trigger DC TNF- and IL-12 production. BMDC were incubated with medium (open bars), LPS (black bars; 1 μg/ml), unstimulated PMN (gray bars), and LPS-stimulated PMN (cross-hatched bars). The ratio of PMN to DC was 2:1. After 18 h of incubation (37°C, 5% CO2), DC culture supernatants were collected for TNF- and IL-12p40 ELISA. A, BMDC from C3H/HeN (LPS responder) and B, BMDC from C3H/HeJ (LPS nonresponder) mice. Levels of cytokine detected in PMN culture supernatants after 3 h incubation in medium alone or in the presence of LPS followed by 18 h in medium, respectively, were: 31 ± 2 pg/ml; 109 ± 7 pg/ml for TNF- and <1 pg/ml and 147 ± 1 pg/ml for IL-12p40. n.d., Not detected. Data are representative of three different experiments.
Direct cell-to-cell contact and peritoneal PMN TNF- are required for optimal DC production of TNF- but not IL-12p40
We next compared the role of soluble vs membrane-bound PMN-derived factors in promoting cytokine release by BMDC. Accordingly, we exposed LPS nonresponder BMDC to cell-free supernatants from 18-h LPS-stimulated mouse peritoneal PMN. In parallel, we cocultured 3-h LPS-stimulated peritoneal PMN with the BMDC as in Fig. 4B. Although DC exposed to soluble factors from LPS-triggered peritoneal PMN released both TNF- and IL-12p40, the response was enhanced when BMDC were cocultured with LPS-stimulated peritoneal PMN (Fig. 5). This was particularly true for TNF- production, where there was a >6-fold increase in the amount of cytokine released in a direct cell-to-cell contact situation. We next used a transwell system to further probe the requirement for optimal DC cytokine production. As shown in Fig. 6A, TNF- production was severely curtailed when LPS-triggered WT mouse peritoneal PMN, and DC were separated by a membrane. In contrast, WT DC IL-12p40 production in response to LPS-stimulated WT peritoneal PMN was minimally affected (Fig. 6A).
FIGURE 5. Cell-to-cell contact is required for optimal DC TNF- but not IL-12p40 production. C3H/HeJ (LPS nonresponder) BMDC were incubated 18 h with supernatants from unstimulated (open bars) or 18 h with LPS-stimulated (black bars) PMN. In addition, DC were coincubated with PMN that had been exposed to medium (gray bars) or LPS (cross-hatched bars) for 3 h. A 2:1 ratio of PMN to DC was used in the coincubation experiments. After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. PMN alone stimulated with LPS released 311 ± 16 pg/ml (TNF-) and 3490 ± 139 pg/ml (IL-12p40). The experiment was repeated twice with similar results. *, p < 0.01 (DC + 18 h PMN supernatants vs DC-PMN coculture). n.d., Not detected.
FIGURE 6. Requirements for PMN triggering of DC cytokine production. Thioglycolate-elicited peritoneal PMN were preincubated 3 h with LPS, then added directly to BMDC (black bar) or separated in a Transwell system (open bar). Supernatants were collected 18 h later for TNF- and IL-12p40 ELISA. A, WT and TNF-–/– PMN were compared for ability to trigger WT DC. B, WT PMN were used to trigger WT and TNFR1/2–/– DC. C, WT neutrophils were used to trigger WT and CD40–/– DC. No cytokine release was detected when nonstimulated PMN were used in each of these experiments (data not shown). This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs Transwell). **, p < 0.01 (direct cell-to-cell contact between WT DC and WT PMN vs direct cell-to-cell contact between WT DC and TNF–/– PMN). n.d., Not detected.
We previously demonstrated that supernatants from T. gondii-triggered PMN induce DC up-regulation of costimulatory molecules CD40 and CD86 and that this response depends upon TNF- (10). In this study, we investigated the involvement of PMN-derived TNF- in promoting cytokine production in PMN-BMDC coculture experiments. Although LPS-stimulated WT peritoneal PMN were able to trigger TNF- production by BMDC, LPS-stimulated TNF–/– neutrophils were defective in driving such a response (Fig. 6A). However, both LPS-stimulated WT and TNF–/– peritoneal PMN were able to promote DC IL-12p40 production in cocultures and when separated in a Transwell chamber system (Fig. 6A). Similarly, using TNFR1/2–/– DC, we found that LPS-stimulated peritoneal PMN no longer elicited DC TNF- release while leaving IL-12p40 production unaffected (Fig. 6B). Last, DC lacking CD40 were examined. Here, there was no evidence for involvement of this costimulatory molecule in DC cytokine production (Fig. 6C). Collectively, the results in Fig. 6 show that direct cell-to-cell contact is required for optimal TNF- but not IL-12p40 production, and that DC TNF- but not IL-12p40 release depends upon neutrophil TNF- production.
Fixed mouse peritoneal neutrophils retain the ability to stimulate DC TNF- release
The results presented so far indicate that cell-to-cell contact between DC and PMN directs TNF- production by DC, but they do not rule out additional involvement of neutrophil soluble factors. To address this issue, peritoneal neutrophils were subjected to paraformaldehyde fixation before DC stimulation. This procedure results in loss of ability of PMN to secrete factors, but surface membrane molecules will be retained. As shown in Fig. 7A, fixation had no effect on the ability of LPS-triggered peritoneal neutrophils to drive DC TNF- production. In contrast, fixation ablated the ability of PMN to drive DC IL-12p40 production, consistent with a role for soluble factors in this situation (Fig. 7B). These results demonstrate that direct contact between DC and PMN is necessary and sufficient to promote TNF- production by DC, but that soluble factors induce DC IL-12 production.
FIGURE 7. LPS-stimulated fixed mouse neutrophils retain the ability to stimulate DC TNF- production. Peritoneal PMN were incubated for 3 h in medium or LPS, washed, fixed or left untreated, and added directly to BMDC (black bars) or separated in a Transwell system (open bars). After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. Untreated PMN alone stimulated with LPS released 143 ± 1 pg/ml (TNF-) and 395 ± 2 pg/ml (IL-12). No cytokine release was detected when nonstimulated PMN were used (data not shown). In addition, no cytokine was detected in supernatants of LPS-stimulated fixed neutrophils, and no DC IL-12 was detected using fixed neutrophils as stimulators. This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs transwell). n.d., Not detected.
Discussion
The results of this study demonstrate that mouse neutrophils triggered with microbial Ag become licensed activators of DC. Although PMN themselves serve as a source of proinflammatory cytokines IL-12p40 and TNF-, the amount of cytokine released by neutrophil-activated DC is 1–2 orders of magnitude higher (Figs. 4–6). Importantly, by using DC with nonfunctional TLR4, it was possible to rule out involvement of endotoxin carryover in the effects of LPS-triggered neutrophils on DC. The DC TNF- release was triggered by neutrophil TNF-, and this was optimally driven by direct cell-to-cell contact.
Neutrophil TNF- was mobilized to the membrane surface within 15 min of endotoxin stimulation relative to cells incubated for the same length of time in medium alone. Surface TNF- up-regulation occurred in the presence of cycloheximide indicating that de novo protein synthesis was not required. Furthermore, nonstimulated mouse neutrophils contained an intracellular pool of TNF- that displayed a transient decrease concomitant with increased surface TNF- expression. In addition to TNF-, PMN have previously been shown to contain preformed IL-12, IL-6, and MIP-2 (35, 36, 37). Both TNF- and IL-12 are present in non-elicited mouse neutrophils. It is, therefore, likely that these cytokines are expressed in PMN during the course of differentiation in the bone marrow. Thus, in addition to their well-known role as microbicidal effectors, neutrophils may also serve an important function as vehicles that rapidly deliver preformed cytokines to sites of inflammation and infection.
Whereas intracellular TNF- pools were depleted 10 min after LPS stimulation, after 60 min, intracellular TNF- was replenished. LPS triggering also induced increased levels of TNF- mRNA. The data argue that, although mouse PMN possess an intracellular pool of preformed cytokine, the cells also display the ability to up-regulate cytokine gene induction, as shown here for TNF- and elsewhere for IL-12 (35). Sustained PMN production of inflammatory cytokines, particularly for the case of GM-CSF, is likely to be important for delaying onset of apoptosis that these cells are programmed to undergo (38, 39). In this way, immunoregulatory and microbicidal effects of neutrophils at sites of inflammation and infection are likely to be prolonged.
We, and others, have suggested that neutrophils, by virtue of their pattern of cytokine expression, may have important in vivo immunoregulatory effects on DC (10, 40, 41, 42). In addition to triggering DC TNF- and IL-12 production, activated PMN exert other effects on these cells. For example, mouse neutrophil release of chemokines CCL3, CCL4, CCL5, and CCL20 together mediate potent chemotactic activity on immature DC (10). Furthermore, PMN induce potent DC up-regulation of costimulatory molecules such as CD40 and CD86, an effect that is in part dependent upon neutrophil TNF- release (10). Other recent studies show that neutrophils assume type 1 and type 2 cytokine profiles, depending on the activating stimulus (43, 44, 45). Together, these findings provide evidence for a key role for neutrophils in controlling immunoregulatory effects of DC.
An unexpected conclusion emerging from our studies is that neutrophil-directed production of DC TNF- and IL-12 is non-identical. Thus, DC IL-12 production does not rely on neutrophil TNF-. In contrast, DC TNF- production requires TNF- expression by peritoneal neutrophils. Furthermore, PMN-induced release of DC IL-12, in contrast to TNF-, occurs equally efficiently across a transwell membrane relative to when cells are in direct contact. CD40 ligation is well known to be involved in DC activation (46), and indeed we show here that LPS induces up-regulation of PMN CD40L expression. Nevertheless, using CD40–/– DC, we were able to exclude a role for CD40:CD40L in effects of PMN on DC. Likewise, IL-12 has been implicated as a priming signal for high-level DC IL-12 production (47), but experiments employing IL-12–/– PMN demonstrate lack of involvement of this cytokine (unpublished observations). The potential role of other neutrophil-derived IL-12-inducing mediators, for example bradykinin, is currently under investigation (48).
Our studies here and elsewhere (9, 10, 12) employ thioglycolate-elicited peritoneal and bone marrow neutrophils. In principle, these populations are non-identical since the former has responded to an inflammatory stimulus. Nevertheless, we have yet to detect differences in these populations, particularly with regard to cytokine/chemokine production. In addition, it can be argued that elicited PMN are the most relevant to study because neutrophils arriving at a site of infection do so in response to inflammatory signals.
The biological significance of DC TNF- production is presently unclear. The cells are more well-known as an IL-12 source that, together with Ag and costimulation, play a central role in the generation of Th1 effector lymphocytes. Nevertheless, DC subsets have been identified in vivo that mediate innate immune defense against microbial infection by producing both NO and TNF- (49). The role of TNF- in this situation may be to promote, along with other inflammatory mediators, macrophage and DC acquisition of microbicidal function.
In vivo evidence for the importance of neutrophils in DC function is provided by observations that depletion of Gr-1(Ly6G)+ cells results in decreased TNF- and IL-12 production by splenic CD11c+ cells and increased susceptibility during infection with microbial pathogens such as T. gondii (10). In this regard, activated PMN that rapidly accumulate at a site of infection or inflammation may play an important role in driving acquisition of DC microbicidal function. The fact that direct cell contact is required for optimal DC TNF- production may be a way to focus the response solely at inflammatory sites of infection and neutrophil accumulation.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Barbara Butcher, Laura Del Rio, and Matthias Hesse for constructive criticism and insightful discussion during the course of this work.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI47888.
2 Address correspondence and reprint requests to Dr. E. Denkers, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853-6401. E-mail address: eyd1{at}cornell.edu
3 Abbreviations used in this paper: DC, dendritic cell; PMN, polymorphonuclear leukocyte; BMDC, bone marrow-derived DC; NMS, normal mouse serum; WT, wild type.
Received for publication October 12, 2004. Accepted for publication January 10, 2005.
References
Janeway, C. A., Jr, R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197.
Palucka, K., J. Banchereau. 2002. How dendritic cells and microbes interact to elicit or subvert protective immune responses. Curr. Opin. Immunol. 14:420.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Belaaouaj, A., R. McCarthy, M. Baumann, Z. Gao, T. J. Ley, S. N. Abraham, S. D. Shapiro. 1998. Mice lacking neutrophil elastase reveal impaired host defense against gram negative bacterial sepsis. Nat. Med. 4:615.
Jackson, S. H., J. I. Gallin, S. M. Holland. 1995. The p47phox mouse knockout model of chronic granulomatous disease. J. Exp. Med. 182:751.
Cassatella, M. A.. 1999. Neutrophil-derived proteins: selling cytokines by the pound. Adv. Immunol. 73:369
Del Rio, L., B. A. Butcher, S. Bennouna, S. Hieny, A. Sher, E. Y. Denkers. 2004. Toxoplasma gondii triggers MyD88-dependent and CCL2 (MCP-1) responses using distinct parasite molecules and host receptors. J. Immunol. 172:6954.
Bennouna, S., S. K. Bliss, T. J. Curiel, E. Y. Denkers. 2003. Cross-talk in the innate immune system: neutrophils instruct early recruitment and activation of dendritic cells during microbial infection. J. Immunol. 171:6052.
Bliss, S. K., A. J. Marshall, Y. Zhang, E. Y. Denkers. 1999. Human polymorphonuclear leukocytes produce IL-12, TNF-, and the chemokines macrophage-inflammatory protein-1 and -1 in response to Toxoplasma gondii antigens. J. Immunol. 162:7369
Bliss, S. K., Y. Zhang, E. Y. Denkers. 1999. Murine neutrophil stimulation by Toxoplasma gondii antigen drives high level production of IFN--independent IL-12. J. Immunol. 163:2081.
Del Rio, L., S. Bennouna, J. Salinas, E. Y. Denkers. 2001. CXCR2 deficiency confers impaired neutrophil recruitment and increased susceptibility during Toxoplasma gondii infection. J. Immunol. 167:6503.
Scharton-Kersten, T., G. Yap, J. Magram, A. Sher. 1997. Inducible nitric oxide is essential for host control of persistent but not acute infection with the intracellular pathogen Toxoplasma gondii. J. Exp. Med. 185:1.
Sayles, P. C., L. J. Johnson. 1997. Exacerbation of toxoplasmosis in neutrophil depleted mice. Nat. Immun. 15:249.
Bliss, S. K., L. C. Gavrilescu, A. Alcaraz, E. Y. Denkers. 2001. Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69:4898.
Wright, S. D., R. A. Ramos, A. Hermanowski-Vosatka, P. Rockwell, P. A. Detmers. 1992. Activation of the adhesive capability of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J. Exp. Med. 173:1281.
Guthrie, L. A., L. C. McPhail, P. M. Henson, R. B. J. Johnston. 1984. Priming of neutrophils enhanced release of oxygen metabolites by bacterial lipopolysaccharide: evidence for increased activity of the superoxide-producing enzyme. J. Exp. Med. 160:1656.
Surette, M. E., R. Palmantier, J. Gosselin, P. Borgeat. 1993. Lipopolysaccharides prime whole human blood and isolated neutrophils for the increased synthesis of 5-lipoxygenase products by enhancing arachidonic acid availability: involvement of CD14 antigen. J. Exp. Med. 178:1347.
Dubravec, D. B., D. R. Spriggs, J. A. Mannick, M. L. Rodrick. 1990. Circulating human peripheral blood granulocytes synthesize and secrete tumor necrosis factor . Proc. Natl. Acad. Sci. USA 87:6758
Cassatella, M. A., L. Meda, S. Gasperini, A. D‘Andrea, X. Ma, G. Trinchieri. 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25:1
Kiefer, F., J. Brumell, N. Al-Alawi, S. Latour, A. Cheng, A. Veillette, S. Grinstein, T. Pawson. 1998. The Syk protein tyrosine kinase is essential for Fc receptor signaling in macrophages and neutrophils. Mol. Cell Biol. 1998. 18:4209.
MacDonald, A. S., A. D. Straw, B. Bauman, E. J. Pearce. 2001. CD8– dendritic cell activation status plays an integral role in influencing Th2 response development. J. Immunol. 167:1982.
Borregaard, N., J. B. Cowland. 1997. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89:3503.
Bainton, D. F.. 1999. Distinct granule populations in human neutrophils and lysosomal organelles identified by electron microscopy. J. Immunol. Methods 232:153.
Gullberg, U., N. Bengtsson, E. Bulow, E. Garwicz, A. Lindmark, I. Olsson. 1999. Processing and targeting of granule proteins in human neutrophils. J. Immunol. Methods 232:201.
Beutler, B., A. Cerami. 1989. The biology of cachectin/TNF- primary mediator of the host response. Annu. Rev. Immunol. 7:625.
Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, et al 1997. A metalloproteinase disintegrin that releases tumor-necrosis factor- from cells. Nature 385:729.
Grohmann, U., M. L. Belladonna, R. Bianchi, C. Orabona, E. Ayroldi, M. C. Fioretti, P. Puccetti. 1998. IL-12 acts directly on DC to promote nuclear localization of NF-B and primes DC for IL-12 production. Immunity 9:315.
Aliberti, J., J. P. Viola, A. Vieria-de-Abreu, P. T. Bozza, A. Sher, J. Scharfstein. 2003. Cutting Edge: bradykinin induces IL-12 production by dendritic cells: a danger signal that drives Th1 polarization. J. Immunol. 170:485.
Serbina, N. V., T. P. Salazar-Mathar, C. A. Biron, W. A. Kuziel, E. A. Pamer. 2003. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19:59.(Soumaya Bennouna and Eric)
Neutrophils play a critical role in early immunity to many microbial pathogens, and this may in part be due to their ability to release immunoregulatory cytokines and chemokines during infection. Here, we demonstrate by flow cytometric analysis that mouse polymorphonuclear leukocytes (PMN) up-regulate surface expression of TNF- within 10 min of stimulation with LPS, and that this is followed by gradual loss over a period of 18 h. Early increases in surface TNF- expression correlated with loss of intracellular pools of preformed TNF-. Nevertheless, extended incubation with LPS resulted in increased levels of TNF- mRNA synthesis and replenishment of intracellular cytokine. After triggering with LPS, PMN acquired the ability to induce dendritic cell (DC) TNF- and IL-12 production. Transwell assays demonstrated that high-level DC TNF- production induced by LPS-triggered neutrophils was dependent upon cell-to-cell contact and neutrophil TNF-, but neither was required for neutrophil instruction of DC IL-12 synthesis. The data suggest that microbial Ag-triggered mouse PMN acquire the capacity to deliver potent DC-activating signals through elaboration of cytokines and direct interactions at the cell surface.
Introduction
Cellular components of innate immunity are critical in sensing infection and initiating responses leading to long-term protection through acquired immunity (1). Dendritic cells (DC), 3 through their ability to capture Ag, migrate to secondary lymphoid organs, present antigenic peptide with costimulation, and release proinflammatory cytokines, are important in driving Th1 differentiation (2, 3, 4). Nevertheless, the cellular and molecular events involved in DC activation during microbial infection are not fully understood.
Neutrophils are often the first cell type recruited to site of inflammation, and they are considered to be terminally differentiated effectors involved in phagocytosis and intracellular killing of pathogens (5, 6). More recent evidence demonstrates the ability of PMN to produce several important proinflammatory cytokines and chemokines (7, 8). In a Toxoplasma gondii infection model, we previously demonstrated that parasite-triggered mouse neutrophils secrete DC and macrophage chemotactic factors including CCL2, CCL3, CCL4, CCL5, and CCL20 (9, 10). PMN coincubated with live tachyzoites also release factors that induce DC up-regulation of costimulatory molecules and secretion of IL-12 as well as TNF- (10, 11, 12) Mice lacking CXCR2, a chemokine receptor involved in neutrophil recruitment, display increased susceptibility to Toxoplasma (13). Furthermore, mice depleted of Gr-1+ cells are unable to survive acute toxoplasmosis (12, 14, 15). This lack of resistance is associated with defective type 1 cytokine responses, in particular decreased TNF- and IL-12 expression by splenic DC (10, 16). These results suggest that PMN play a role in orchestrating early immunity through production of DC-activating factors.
To further clarify the role of neutrophils as immunomodulatory cells in responses driven by microbial Ag, we investigated the ability of mouse PMN to trigger DC activation after stimulation with LPS. Use of LPS rather than a highly complex microbial lysate such as soluble tachyzoite Ag, which we employed in previous studies, provides the advantage that a single molecule with a single receptor (TLR4) is involved, facilitating dissection of PMN-DC interactions. LPS induces neutrophil activation both in vivo and in vitro, as measured by release of cytokines, degranulation, increased adhesion potential, and oxidative burst activity (17, 18, 19, 20, 21).
In this study, we present data showing that mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface upon LPS stimulation. Transmembrane TNF- expression is accompanied by up-regulation of surface CD40L and CD11b/CD18. LPS-triggered mouse neutrophils are potent inducers of DC TNF- when cells are placed in direct contact with each other. This response itself is dependent upon PMN TNF- expression. Endotoxin-activated mouse neutrophils also induce DC IL-12 production, but here neither PMN TNF- nor direct contact was required. The data suggest microbial Ag endow PMN with potent DC-activating capability, a property dependent upon increased expression of neutrophil membrane TNF-.
Materials and Methods
Mice
C57BL/6, TNF receptor (TNFR) 1/2 double knockout (TNFR1/2–/–), TNF- knockout and wild-type (WT) counterparts (B6129SF2/J) were obtained from The Jackson Laboratory. C3H/HeN and C3H/HeJ were purchased from Taconic Farms. Female mice between 6 and 8 wk of age were used throughout. Genotype of knockout animals was confirmed by RT-PCR according to protocols obtained from The Jackson Laboratory. Animals were housed in filter-covered isolator cages in the animal facility of the College of Veterinary Medicine at Cornell University, which is accredited by the American Association for Accreditation of Laboratory Care.
Reagents and Ab
Ultrapure LPS (Escherichia coli strain 0111:B4) was purchased from List Biological Laboratories. FITC-conjugated goat anti-hamster IgG Ab was obtained from eBioscience. Allophycocyanin-conjugated Ab specific for Ly6G, PE-conjugated Ab specific for TNF- (MP6-XT22), CD40L (MR1), CD11b (M1/170), CD14 (rmC5-3), TNFR type II (p75) (TR75-89), purified rat anti-mouse TNF- (MP6-XT22), and hamster anti-mouse TNF- (TN3-19.12) were obtained from BD Pharmingen. Purified antiserum specific for CD68 (T-16) and PE-conjugated TNF- anti-goat Ab were obtained from Santa Cruz Biotechnology. Normal hamster Ig and normal mouse serum (NMS) were purchased from Jackson ImmunoResearch Laboratories. Cycloheximide was obtained from Calbiochem, and recombinant mouse GM-CSF was purchased from PeproTech.
Bone marrow PMN
Neutrophils were isolated from mouse bone marrow following a previously published protocol (22). Briefly, single cell suspensions of bone marrow cells were collected from femur and tibia and resuspended in DMEM supplemented with 5% FCS (Hyclone Laboratories), 100 U/ml penicillin, and 100 mg/ml streptomycin (PenStrep; Invitrogen Life Technologies). Cells were then centrifuged at 500 x g for 7 min at 4°C and resuspended in Ca2+-free HBSS supplemented with 0.38% sodium citrate. The cell suspension was layered on top of a step gradient consisting of 52, 65, and 75% Percoll diluted in Ca2+-free HBSS, and centrifuged at 1500 x g for 30 min at 4°C. Neutrophils were recovered at the interface between the 65 and 75% Percoll layers. The proportion of neutrophils, determined by Diff-Quik staining of cytospin preparations, was routinely >90%.
Peritoneal neutrophils
Mice were i.p. injected with 1 ml of 10% thioglycolate (Difco). Peritoneal exudate cells were obtained by lavage 18 h later with ice-cold PBS. Cells were washed in PBS, passed through a 70-μm nylon cell strainer, and erythrocytes in the suspension were lysed using Red cell lysis buffer (Sigma-Aldrich). Cells were then washed, resuspended at 2 x 107 cells/ml in buffer composed of Dulbecco’s PBS (Invitrogen Life Technology) containing 0.5% BSA (Sigma-Aldrich), 1 mM EDTA (Fisher Scientific), and incubated for 15 min at 4°C with anti-MHC class II magnetic microbeads. After washing, the mixture was transferred to columns installed within a magnetic apparatus to remove MHC class II-expressing cells, according to the manufacturer’s instructions (Miltenyi Biotec).
Bone marrow-derived DC (BMDC)
Generation of BMDC was accomplished following a previously published protocol (23). Briefly, single cell bone marrow preparations were obtained as described above, cells were washed in RPMI 1640 (Fisher Scientific) and resuspended at 2 x 105 cells/ml in DC medium composed of RPMI 1640 supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin, 10% FCS, 5 x 10–5 M 2-ME, and 20 ng/ml GM-CSF. Cells were plated on 100 x 15 mm standard sterile polystyrene petri dishes (Fisher Scientific) and cultured for 9 days at 37°C in 5% CO2. Fresh DC medium, containing 20 ng/ml GM-CSF, was added on days 3, 6, and 8 after culture initiation. On day 9, cells were resuspended in cDMEM composed of DMEM (Invitrogen Life Technologies) supplemented with 10% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin, 5 x 10–5 M 2-ME, 10 mM HEPES (Invitrogen Life Technologies), 100 μM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), alone or in the presence of different stimuli at 37°C in 5% CO2. Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C.
PMN culture
Peritoneal PMN (2 x 106/ml) were cultured in cDMEM alone or in the presence of 1 μg/ml LPS at 37°C in 5% CO2 in a 96-well tissue culture plate (Corning Costar). Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C. In the coculture experiments, neutrophils were cultured in medium or LPS (1 μg/ml) for 3 h followed by three washes; then they were either plated together with immature BMDC in 12-well tissue culture plates (Corning Costar) or placed in the upper well of a Transwell system (Corning Costar), while BMDC were added to the lower well. In some experiments PMN were fixed for 15 min on ice in 1% formaldehyde solution. In experiments involving blocking of new protein synthesis, neutrophils were incubated in medium containing 30 μg/ml cycloheximide for 15 min followed by a stimulation with 1 μg/ml LPS.
RT-PCR
RNA was isolated, reverse transcribed and subjected to PCR amplification as described (12). The primer sequences used were: -actin, TGACGGGGGTCACCCACACTGTGCCCATCTA (sense), CTAGAAGCATTGCGGTGGACGATGGAGGG (antisense); TNF-, CAGCCTCTTCTCATTCCTGCTTGTG (sense), CTGGAAGACTCCTCCCAGGTATAT (antisense). The cDNA was amplified 31 cycles.
Flow cytometry
To analyze surface markers on PMN, Fc receptors were blocked in FACS buffer (PBS, 1% BSA, and 0.1% sodium azide) containing 5 μg/ml anti-mouse CD16/CD32 (BD Pharmingen) and 10% NMS for 15 min at 0°C, then cells were stained with optimal concentrations of allophycocyanin-conjugated anti-Gr-1(Ly-6G) in combination with PE-conjugated antisera specific for TNF-, CD40L, CD11b, CD14, TNFR type II (p75), and unlabeled goat anti-CD68 followed by PE-conjugated anti-goat Ab for 30 min at 0°C. For intracellular cytokine detection, peritoneal cells were blocked as described above, stained with allophycocyanin-conjugated anti-Gr-1 Ab and unlabeled anti-TNF- Ab, then cells were fixed in 3% paraformaldehyde (Sigma-Aldrich), 0.1 mM CaCl2, and 0.1 mM MgCl2 for 30 min at 0°C. Cells were subsequently washed in permeabilization buffer (PBS with 0.075% saponin) and incubated for 15 min at 0°C in permeabilization buffer containing 10% NMS. After two washes in permeabilization buffer, unlabeled hamster anti-TNF- was added, and cells were incubated for 30 min at 0°C. After washing in permeabilization buffer, FITC-conjugated anti-hamster was added followed by a 30-min incubation at 0°C and subsequently washed for flow cytometric analysis. Data was acquired on FACSCalibur system (10,000 events per sample) and analyzed with CellQuest software (BD Biosciences Immunocytometry Systems).
Cytokine ELISA
IL-12p40 was measured as previously described (12), and TNF- was measured using a commercially obtained kit (BD Pharmingen).
Statistical Analysis
The statistical significance of the data was analyzed using SD and unpaired two-tailed Student’s t test.
Results
LPS triggers rapid changes in peritoneal PMN surface membrane-associated molecules
We began by examining the influence of endotoxin on expression of neutrophil cell surface markers. Flow cytometric analysis revealed early LPS-induced up-regulation of surface TNF-, CD40L, and CD11b, followed by a time-dependent down-regulation relative to cells cultured in medium alone (Fig. 1A). The surface expression of these molecules in medium alone at 30 min was equivalent to staining of freshly isolated, nonstimulated peritoneal neutrophils (data not shown). We also examined expression of CD14, TNFR type II, and CD68, which are markers of exocytosis of secretory vesicles, specific and azurophilic granules, respectively (24, 25, 26, 27). Both CD14 and CD68 were rapidly mobilized to the cell surface, consistent with exocytosis of secretory vesicles and azurophilic granules. Interestingly, the influence of LPS on TNFR type II expression was distinct, in that LPS triggered early down-regulation. This was followed by a gradual increase in TNFR type II surface expression that required 18 h to reach maximal levels (Fig. 1B). These effects are likely mediated by TLR4, because increased expression of TNF- and other molecules did not occur in peritoneal PMN from the LPS nonresponder strain C3H/HeJ (Fig. 1C and data not shown).
FIGURE 1. LPS triggers rapid changes in the pattern of molecules expressed on the neutrophil surface. Thioglycolate-elicited peritoneal exudate cells from C57BL/6 (A and B), and C3H/HeN/C3H/HeJ (C) mice were stimulated with LPS (thick lines) or incubated in medium (thin lines) for varying times, then stained with Ab to Gr-1 and the indicated markers. The data show expression profiles gated on Gr-1high cells. C, Surface TNF- expression 60 min after treatment with LPS or medium. The numbers in A and C indicate the percentage of positive cells after stimulation with LPS (top) vs incubation in medium (bottom). This experiment was repeated three times with essentially identical results.
Control of neutrophil transmembrane TNF-
We next focused on surface expression of TNF- by peritoneal PMN. TNF- is synthesized as a transmembrane 26-kDa protein that is transported to the cell surface where it is cleaved by a metalloproteinase (TACE) to release the soluble 17-kDa protein (28, 29, 30, 31, 32, 33, 34). Therefore, increased TNF- levels on the cell surface (Fig. 1A) could be the result of higher levels of transmembrane cytokine or, alternatively, the result of soluble TNF- binding to neutrophil TNF receptors. To distinguish these possibilities, we stimulated WT or TNFR1/2–/– peritoneal neutrophils with LPS and analyzed surface expression of TNF-. Both WT (Fig. 2A) and TNFR1/2–/– (Fig. 2B) peritoneal PMN up-regulated surface TNF- expression after 1 h of stimulation with LPS. This result argues that transmembrane 26-kDa TNF- is up-regulated on the PMN cell surface, rather than a situation in which soluble TNF- binds to neutrophil TNF receptors. In addition, preincubation of cells with unlabeled anti-TNF- Ab prevented binding of the PE-conjugated Ab, confirming the specificity of Ab binding (Fig. 2C).
FIGURE 2. LPS induces neutrophil surface TNF- expression independently of TNF receptors. Thioglycolate-induced peritoneal exudate cells from WT (A and C) and TNFR1/2–/– (B) mice were cultured in medium (thin lines) or stimulated with LPS (thick lines) for 1 h and stained for surface TNF-. C, Cells were stained with an unlabeled blocking anti-TNF- Ab followed by staining for TNF- as in A and B. The profiles shown are gated on the Gr-1high population.
We previously found that mouse PMN contain an intracellular pool of IL-12 (35). Because we observed surface expression of TNF- very early after LPS stimulation in our experiments, we investigated the possibility that mouse PMN also contain preformed TNF-. To address this question, we first pretreated peritoneal PMN with cycloheximide to block new protein synthesis before stimulation of cells with LPS. Cycloheximide had no effect on early LPS-induced up-regulation of surface TNF-. However, after 1 h of LPS stimulation, cycloheximide-treated cells displayed a marginal but reproducible decrease in TNF- surface expression, suggesting that de novo protein synthesis may contribute to surface expression at later time points (Fig. 3A). To further examine the presence of preformed TNF-, we blocked surface TNF-, permeabilized cells, and stained for intracellular cytokine. In this case, bone marrow and peritoneal exudate PMN were strongly positive for intracellular TNF- (Fig. 3B).
FIGURE 3. Mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface after LPS stimulation. A, Early LPS-triggered PMN surface TNF- expression does not require new protein synthesis. Thioglycolate-elicited cells (PEC) were preincubated with cycloheximide then stimulated for the indicated times with LPS. Cells were stained with Ab to Gr-1 and TNF-. The profiles shown are gated on Gr-1high cells. Green line, LPS with no cycloheximide; red line, LPS with cycloheximide; black line, medium with cycloheximide. No difference was observed between medium and medium with cycloheximide (data not shown). B, PMN contain intracellular TNF-. Bone marrow cells and PEC were subjected to surface TNF- blocking using unlabeled rat anti-mouse TNF- Ab, then permeabilized and stained with hamster anti-mouse TNF- Ab. Data shown are gated on Gr-1high cells. Red line, anti-TNF- Ab; black line, isotype control. C, Changes in intracellular and membrane bound TNF- during LPS stimulation. Thioglycolate-elicited cells were incubated in medium or stimulated with LPS for the indicated times. Cells were subsequently stained with Gr-1 Ab and surface TNF- using a rat anti-mouse TNF-. After permeabilization and blocking, cells were stained for intracellular TNF- using a hamster anti-mouse Ab. Profiles shown are gated on Gr-1high cells. Green and gray lines, intracellular TNF- expression after incubation with LPS and medium, respectively. Red and black lines, surface TNF- expression after incubation with LPS and medium, respectively. Dotted lines, isotype control staining. Numbers in each histogram show percentage of positive followed by geometric mean fluorescence intensity for LPS-stimulated cells (top) and cells incubated in medium alone (bottom). D, RT-PCR analysis of purified peritoneal exudate PMN incubated in medium alone or stimulated for the indicated times with LPS.
We next examined modulation of surface vs intracellular expression of TNF- in LPS-stimulated peritoneal PMN. At 10 min after LPS stimulation, when compared with unstimulated cells, we observed a decrease in intracellular TNF- and a parallel increase in surface expression of this cytokine (Fig. 3C). However, after 1 h of exposure to LPS, both surface and intracellular TNF- expression were increased (Fig. 3C). These results suggest that de novo protein synthesis replenishes the pool of intracellular TNF- that is initially mobilized to the neutrophil surface. Evidence that LPS induces increased TNF- gene transcription is shown by RT-PCR in Fig. 3D. Increased levels of TNF- mRNA were apparent within 15 min, and maximal levels were attained by 60 min of stimulation. The combined results argue that both preformed pools and newly synthesized TNF- contribute to high level expression of this cytokine on the cell surface.
LPS stimulated mouse peritoneal neutrophils promote dendritic cell cytokine production
To determine whether LPS-triggered peritoneal PMN can directly stimulate TNF- and IL-12 production by BMDC, we began by measuring cytokine levels in PMN-DC cocultures. When BMDC were cultured with LPS or cocultured with LPS-activated peritoneal PMN, high levels of both TNF- and IL-12p40 were released into the coculture supernatants (Fig. 4A). In contrast, when nonstimulated mouse peritoneal PMN were used in the cocultures, no TNF- and minimal amounts of IL-12p40 were detected (Fig. 4A). To test whether the triggering effect of peritoneal PMN was due to bona fide factors produced by the latter cells or simply to LPS carryover, we conducted identical experiments employing BMDC derived from LPS nonresponder animals. In this case, although direct DC stimulation with LPS failed to elicit a cytokine response, coculture with LPS-stimulated peritoneal neutrophils triggered robust TNF- and IL-12p40 responses (Fig. 4B). This result demonstrates that the cytokine responses were triggered by neutrophil factors rather than residual LPS in the cocultures.
FIGURE 4. LPS-triggered PMN express factors that trigger DC TNF- and IL-12 production. BMDC were incubated with medium (open bars), LPS (black bars; 1 μg/ml), unstimulated PMN (gray bars), and LPS-stimulated PMN (cross-hatched bars). The ratio of PMN to DC was 2:1. After 18 h of incubation (37°C, 5% CO2), DC culture supernatants were collected for TNF- and IL-12p40 ELISA. A, BMDC from C3H/HeN (LPS responder) and B, BMDC from C3H/HeJ (LPS nonresponder) mice. Levels of cytokine detected in PMN culture supernatants after 3 h incubation in medium alone or in the presence of LPS followed by 18 h in medium, respectively, were: 31 ± 2 pg/ml; 109 ± 7 pg/ml for TNF- and <1 pg/ml and 147 ± 1 pg/ml for IL-12p40. n.d., Not detected. Data are representative of three different experiments.
Direct cell-to-cell contact and peritoneal PMN TNF- are required for optimal DC production of TNF- but not IL-12p40
We next compared the role of soluble vs membrane-bound PMN-derived factors in promoting cytokine release by BMDC. Accordingly, we exposed LPS nonresponder BMDC to cell-free supernatants from 18-h LPS-stimulated mouse peritoneal PMN. In parallel, we cocultured 3-h LPS-stimulated peritoneal PMN with the BMDC as in Fig. 4B. Although DC exposed to soluble factors from LPS-triggered peritoneal PMN released both TNF- and IL-12p40, the response was enhanced when BMDC were cocultured with LPS-stimulated peritoneal PMN (Fig. 5). This was particularly true for TNF- production, where there was a >6-fold increase in the amount of cytokine released in a direct cell-to-cell contact situation. We next used a transwell system to further probe the requirement for optimal DC cytokine production. As shown in Fig. 6A, TNF- production was severely curtailed when LPS-triggered WT mouse peritoneal PMN, and DC were separated by a membrane. In contrast, WT DC IL-12p40 production in response to LPS-stimulated WT peritoneal PMN was minimally affected (Fig. 6A).
FIGURE 5. Cell-to-cell contact is required for optimal DC TNF- but not IL-12p40 production. C3H/HeJ (LPS nonresponder) BMDC were incubated 18 h with supernatants from unstimulated (open bars) or 18 h with LPS-stimulated (black bars) PMN. In addition, DC were coincubated with PMN that had been exposed to medium (gray bars) or LPS (cross-hatched bars) for 3 h. A 2:1 ratio of PMN to DC was used in the coincubation experiments. After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. PMN alone stimulated with LPS released 311 ± 16 pg/ml (TNF-) and 3490 ± 139 pg/ml (IL-12p40). The experiment was repeated twice with similar results. *, p < 0.01 (DC + 18 h PMN supernatants vs DC-PMN coculture). n.d., Not detected.
FIGURE 6. Requirements for PMN triggering of DC cytokine production. Thioglycolate-elicited peritoneal PMN were preincubated 3 h with LPS, then added directly to BMDC (black bar) or separated in a Transwell system (open bar). Supernatants were collected 18 h later for TNF- and IL-12p40 ELISA. A, WT and TNF-–/– PMN were compared for ability to trigger WT DC. B, WT PMN were used to trigger WT and TNFR1/2–/– DC. C, WT neutrophils were used to trigger WT and CD40–/– DC. No cytokine release was detected when nonstimulated PMN were used in each of these experiments (data not shown). This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs Transwell). **, p < 0.01 (direct cell-to-cell contact between WT DC and WT PMN vs direct cell-to-cell contact between WT DC and TNF–/– PMN). n.d., Not detected.
We previously demonstrated that supernatants from T. gondii-triggered PMN induce DC up-regulation of costimulatory molecules CD40 and CD86 and that this response depends upon TNF- (10). In this study, we investigated the involvement of PMN-derived TNF- in promoting cytokine production in PMN-BMDC coculture experiments. Although LPS-stimulated WT peritoneal PMN were able to trigger TNF- production by BMDC, LPS-stimulated TNF–/– neutrophils were defective in driving such a response (Fig. 6A). However, both LPS-stimulated WT and TNF–/– peritoneal PMN were able to promote DC IL-12p40 production in cocultures and when separated in a Transwell chamber system (Fig. 6A). Similarly, using TNFR1/2–/– DC, we found that LPS-stimulated peritoneal PMN no longer elicited DC TNF- release while leaving IL-12p40 production unaffected (Fig. 6B). Last, DC lacking CD40 were examined. Here, there was no evidence for involvement of this costimulatory molecule in DC cytokine production (Fig. 6C). Collectively, the results in Fig. 6 show that direct cell-to-cell contact is required for optimal TNF- but not IL-12p40 production, and that DC TNF- but not IL-12p40 release depends upon neutrophil TNF- production.
Fixed mouse peritoneal neutrophils retain the ability to stimulate DC TNF- release
The results presented so far indicate that cell-to-cell contact between DC and PMN directs TNF- production by DC, but they do not rule out additional involvement of neutrophil soluble factors. To address this issue, peritoneal neutrophils were subjected to paraformaldehyde fixation before DC stimulation. This procedure results in loss of ability of PMN to secrete factors, but surface membrane molecules will be retained. As shown in Fig. 7A, fixation had no effect on the ability of LPS-triggered peritoneal neutrophils to drive DC TNF- production. In contrast, fixation ablated the ability of PMN to drive DC IL-12p40 production, consistent with a role for soluble factors in this situation (Fig. 7B). These results demonstrate that direct contact between DC and PMN is necessary and sufficient to promote TNF- production by DC, but that soluble factors induce DC IL-12 production.
FIGURE 7. LPS-stimulated fixed mouse neutrophils retain the ability to stimulate DC TNF- production. Peritoneal PMN were incubated for 3 h in medium or LPS, washed, fixed or left untreated, and added directly to BMDC (black bars) or separated in a Transwell system (open bars). After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. Untreated PMN alone stimulated with LPS released 143 ± 1 pg/ml (TNF-) and 395 ± 2 pg/ml (IL-12). No cytokine release was detected when nonstimulated PMN were used (data not shown). In addition, no cytokine was detected in supernatants of LPS-stimulated fixed neutrophils, and no DC IL-12 was detected using fixed neutrophils as stimulators. This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs transwell). n.d., Not detected.
Discussion
The results of this study demonstrate that mouse neutrophils triggered with microbial Ag become licensed activators of DC. Although PMN themselves serve as a source of proinflammatory cytokines IL-12p40 and TNF-, the amount of cytokine released by neutrophil-activated DC is 1–2 orders of magnitude higher (Figs. 4–6). Importantly, by using DC with nonfunctional TLR4, it was possible to rule out involvement of endotoxin carryover in the effects of LPS-triggered neutrophils on DC. The DC TNF- release was triggered by neutrophil TNF-, and this was optimally driven by direct cell-to-cell contact.
Neutrophil TNF- was mobilized to the membrane surface within 15 min of endotoxin stimulation relative to cells incubated for the same length of time in medium alone. Surface TNF- up-regulation occurred in the presence of cycloheximide indicating that de novo protein synthesis was not required. Furthermore, nonstimulated mouse neutrophils contained an intracellular pool of TNF- that displayed a transient decrease concomitant with increased surface TNF- expression. In addition to TNF-, PMN have previously been shown to contain preformed IL-12, IL-6, and MIP-2 (35, 36, 37). Both TNF- and IL-12 are present in non-elicited mouse neutrophils. It is, therefore, likely that these cytokines are expressed in PMN during the course of differentiation in the bone marrow. Thus, in addition to their well-known role as microbicidal effectors, neutrophils may also serve an important function as vehicles that rapidly deliver preformed cytokines to sites of inflammation and infection.
Whereas intracellular TNF- pools were depleted 10 min after LPS stimulation, after 60 min, intracellular TNF- was replenished. LPS triggering also induced increased levels of TNF- mRNA. The data argue that, although mouse PMN possess an intracellular pool of preformed cytokine, the cells also display the ability to up-regulate cytokine gene induction, as shown here for TNF- and elsewhere for IL-12 (35). Sustained PMN production of inflammatory cytokines, particularly for the case of GM-CSF, is likely to be important for delaying onset of apoptosis that these cells are programmed to undergo (38, 39). In this way, immunoregulatory and microbicidal effects of neutrophils at sites of inflammation and infection are likely to be prolonged.
We, and others, have suggested that neutrophils, by virtue of their pattern of cytokine expression, may have important in vivo immunoregulatory effects on DC (10, 40, 41, 42). In addition to triggering DC TNF- and IL-12 production, activated PMN exert other effects on these cells. For example, mouse neutrophil release of chemokines CCL3, CCL4, CCL5, and CCL20 together mediate potent chemotactic activity on immature DC (10). Furthermore, PMN induce potent DC up-regulation of costimulatory molecules such as CD40 and CD86, an effect that is in part dependent upon neutrophil TNF- release (10). Other recent studies show that neutrophils assume type 1 and type 2 cytokine profiles, depending on the activating stimulus (43, 44, 45). Together, these findings provide evidence for a key role for neutrophils in controlling immunoregulatory effects of DC.
An unexpected conclusion emerging from our studies is that neutrophil-directed production of DC TNF- and IL-12 is non-identical. Thus, DC IL-12 production does not rely on neutrophil TNF-. In contrast, DC TNF- production requires TNF- expression by peritoneal neutrophils. Furthermore, PMN-induced release of DC IL-12, in contrast to TNF-, occurs equally efficiently across a transwell membrane relative to when cells are in direct contact. CD40 ligation is well known to be involved in DC activation (46), and indeed we show here that LPS induces up-regulation of PMN CD40L expression. Nevertheless, using CD40–/– DC, we were able to exclude a role for CD40:CD40L in effects of PMN on DC. Likewise, IL-12 has been implicated as a priming signal for high-level DC IL-12 production (47), but experiments employing IL-12–/– PMN demonstrate lack of involvement of this cytokine (unpublished observations). The potential role of other neutrophil-derived IL-12-inducing mediators, for example bradykinin, is currently under investigation (48).
Our studies here and elsewhere (9, 10, 12) employ thioglycolate-elicited peritoneal and bone marrow neutrophils. In principle, these populations are non-identical since the former has responded to an inflammatory stimulus. Nevertheless, we have yet to detect differences in these populations, particularly with regard to cytokine/chemokine production. In addition, it can be argued that elicited PMN are the most relevant to study because neutrophils arriving at a site of infection do so in response to inflammatory signals.
The biological significance of DC TNF- production is presently unclear. The cells are more well-known as an IL-12 source that, together with Ag and costimulation, play a central role in the generation of Th1 effector lymphocytes. Nevertheless, DC subsets have been identified in vivo that mediate innate immune defense against microbial infection by producing both NO and TNF- (49). The role of TNF- in this situation may be to promote, along with other inflammatory mediators, macrophage and DC acquisition of microbicidal function.
In vivo evidence for the importance of neutrophils in DC function is provided by observations that depletion of Gr-1(Ly6G)+ cells results in decreased TNF- and IL-12 production by splenic CD11c+ cells and increased susceptibility during infection with microbial pathogens such as T. gondii (10). In this regard, activated PMN that rapidly accumulate at a site of infection or inflammation may play an important role in driving acquisition of DC microbicidal function. The fact that direct cell contact is required for optimal DC TNF- production may be a way to focus the response solely at inflammatory sites of infection and neutrophil accumulation.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Barbara Butcher, Laura Del Rio, and Matthias Hesse for constructive criticism and insightful discussion during the course of this work.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI47888.
2 Address correspondence and reprint requests to Dr. E. Denkers, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853-6401. E-mail address: eyd1{at}cornell.edu
3 Abbreviations used in this paper: DC, dendritic cell; PMN, polymorphonuclear leukocyte; BMDC, bone marrow-derived DC; NMS, normal mouse serum; WT, wild type.
Received for publication October 12, 2004. Accepted for publication January 10, 2005.
References
Janeway, C. A., Jr, R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197.
Palucka, K., J. Banchereau. 2002. How dendritic cells and microbes interact to elicit or subvert protective immune responses. Curr. Opin. Immunol. 14:420.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Belaaouaj, A., R. McCarthy, M. Baumann, Z. Gao, T. J. Ley, S. N. Abraham, S. D. Shapiro. 1998. Mice lacking neutrophil elastase reveal impaired host defense against gram negative bacterial sepsis. Nat. Med. 4:615.
Jackson, S. H., J. I. Gallin, S. M. Holland. 1995. The p47phox mouse knockout model of chronic granulomatous disease. J. Exp. Med. 182:751.
Cassatella, M. A.. 1999. Neutrophil-derived proteins: selling cytokines by the pound. Adv. Immunol. 73:369
Del Rio, L., B. A. Butcher, S. Bennouna, S. Hieny, A. Sher, E. Y. Denkers. 2004. Toxoplasma gondii triggers MyD88-dependent and CCL2 (MCP-1) responses using distinct parasite molecules and host receptors. J. Immunol. 172:6954.
Bennouna, S., S. K. Bliss, T. J. Curiel, E. Y. Denkers. 2003. Cross-talk in the innate immune system: neutrophils instruct early recruitment and activation of dendritic cells during microbial infection. J. Immunol. 171:6052.
Bliss, S. K., A. J. Marshall, Y. Zhang, E. Y. Denkers. 1999. Human polymorphonuclear leukocytes produce IL-12, TNF-, and the chemokines macrophage-inflammatory protein-1 and -1 in response to Toxoplasma gondii antigens. J. Immunol. 162:7369
Bliss, S. K., Y. Zhang, E. Y. Denkers. 1999. Murine neutrophil stimulation by Toxoplasma gondii antigen drives high level production of IFN--independent IL-12. J. Immunol. 163:2081.
Del Rio, L., S. Bennouna, J. Salinas, E. Y. Denkers. 2001. CXCR2 deficiency confers impaired neutrophil recruitment and increased susceptibility during Toxoplasma gondii infection. J. Immunol. 167:6503.
Scharton-Kersten, T., G. Yap, J. Magram, A. Sher. 1997. Inducible nitric oxide is essential for host control of persistent but not acute infection with the intracellular pathogen Toxoplasma gondii. J. Exp. Med. 185:1.
Sayles, P. C., L. J. Johnson. 1997. Exacerbation of toxoplasmosis in neutrophil depleted mice. Nat. Immun. 15:249.
Bliss, S. K., L. C. Gavrilescu, A. Alcaraz, E. Y. Denkers. 2001. Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69:4898.
Wright, S. D., R. A. Ramos, A. Hermanowski-Vosatka, P. Rockwell, P. A. Detmers. 1992. Activation of the adhesive capability of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J. Exp. Med. 173:1281.
Guthrie, L. A., L. C. McPhail, P. M. Henson, R. B. J. Johnston. 1984. Priming of neutrophils enhanced release of oxygen metabolites by bacterial lipopolysaccharide: evidence for increased activity of the superoxide-producing enzyme. J. Exp. Med. 160:1656.
Surette, M. E., R. Palmantier, J. Gosselin, P. Borgeat. 1993. Lipopolysaccharides prime whole human blood and isolated neutrophils for the increased synthesis of 5-lipoxygenase products by enhancing arachidonic acid availability: involvement of CD14 antigen. J. Exp. Med. 178:1347.
Dubravec, D. B., D. R. Spriggs, J. A. Mannick, M. L. Rodrick. 1990. Circulating human peripheral blood granulocytes synthesize and secrete tumor necrosis factor . Proc. Natl. Acad. Sci. USA 87:6758
Cassatella, M. A., L. Meda, S. Gasperini, A. D‘Andrea, X. Ma, G. Trinchieri. 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25:1
Kiefer, F., J. Brumell, N. Al-Alawi, S. Latour, A. Cheng, A. Veillette, S. Grinstein, T. Pawson. 1998. The Syk protein tyrosine kinase is essential for Fc receptor signaling in macrophages and neutrophils. Mol. Cell Biol. 1998. 18:4209.
MacDonald, A. S., A. D. Straw, B. Bauman, E. J. Pearce. 2001. CD8– dendritic cell activation status plays an integral role in influencing Th2 response development. J. Immunol. 167:1982.
Borregaard, N., J. B. Cowland. 1997. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89:3503.
Bainton, D. F.. 1999. Distinct granule populations in human neutrophils and lysosomal organelles identified by electron microscopy. J. Immunol. Methods 232:153.
Gullberg, U., N. Bengtsson, E. Bulow, E. Garwicz, A. Lindmark, I. Olsson. 1999. Processing and targeting of granule proteins in human neutrophils. J. Immunol. Methods 232:201.
Beutler, B., A. Cerami. 1989. The biology of cachectin/TNF- primary mediator of the host response. Annu. Rev. Immunol. 7:625.
Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, et al 1997. A metalloproteinase disintegrin that releases tumor-necrosis factor- from cells. Nature 385:729.
Grohmann, U., M. L. Belladonna, R. Bianchi, C. Orabona, E. Ayroldi, M. C. Fioretti, P. Puccetti. 1998. IL-12 acts directly on DC to promote nuclear localization of NF-B and primes DC for IL-12 production. Immunity 9:315.
Aliberti, J., J. P. Viola, A. Vieria-de-Abreu, P. T. Bozza, A. Sher, J. Scharfstein. 2003. Cutting Edge: bradykinin induces IL-12 production by dendritic cells: a danger signal that drives Th1 polarization. J. Immunol. 170:485.
Serbina, N. V., T. P. Salazar-Mathar, C. A. Biron, W. A. Kuziel, E. A. Pamer. 2003. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19:59.
Abstract
Neutrophils play a critical role in early immunity to many microbial pathogens, and this may in part be due to their ability to release immunoregulatory cytokines and chemokines during infection. Here, we demonstrate by flow cytometric analysis that mouse polymorphonuclear leukocytes (PMN) up-regulate surface expression of TNF- within 10 min of stimulation with LPS, and that this is followed by gradual loss over a period of 18 h. Early increases in surface TNF- expression correlated with loss of intracellular pools of preformed TNF-. Nevertheless, extended incubation with LPS resulted in increased levels of TNF- mRNA synthesis and replenishment of intracellular cytokine. After triggering with LPS, PMN acquired the ability to induce dendritic cell (DC) TNF- and IL-12 production. Transwell assays demonstrated that high-level DC TNF- production induced by LPS-triggered neutrophils was dependent upon cell-to-cell contact and neutrophil TNF-, but neither was required for neutrophil instruction of DC IL-12 synthesis. The data suggest that microbial Ag-triggered mouse PMN acquire the capacity to deliver potent DC-activating signals through elaboration of cytokines and direct interactions at the cell surface.
Introduction
Cellular components of innate immunity are critical in sensing infection and initiating responses leading to long-term protection through acquired immunity (1). Dendritic cells (DC), 3 through their ability to capture Ag, migrate to secondary lymphoid organs, present antigenic peptide with costimulation, and release proinflammatory cytokines, are important in driving Th1 differentiation (2, 3, 4). Nevertheless, the cellular and molecular events involved in DC activation during microbial infection are not fully understood.
Neutrophils are often the first cell type recruited to site of inflammation, and they are considered to be terminally differentiated effectors involved in phagocytosis and intracellular killing of pathogens (5, 6). More recent evidence demonstrates the ability of PMN to produce several important proinflammatory cytokines and chemokines (7, 8). In a Toxoplasma gondii infection model, we previously demonstrated that parasite-triggered mouse neutrophils secrete DC and macrophage chemotactic factors including CCL2, CCL3, CCL4, CCL5, and CCL20 (9, 10). PMN coincubated with live tachyzoites also release factors that induce DC up-regulation of costimulatory molecules and secretion of IL-12 as well as TNF- (10, 11, 12) Mice lacking CXCR2, a chemokine receptor involved in neutrophil recruitment, display increased susceptibility to Toxoplasma (13). Furthermore, mice depleted of Gr-1+ cells are unable to survive acute toxoplasmosis (12, 14, 15). This lack of resistance is associated with defective type 1 cytokine responses, in particular decreased TNF- and IL-12 expression by splenic DC (10, 16). These results suggest that PMN play a role in orchestrating early immunity through production of DC-activating factors.
To further clarify the role of neutrophils as immunomodulatory cells in responses driven by microbial Ag, we investigated the ability of mouse PMN to trigger DC activation after stimulation with LPS. Use of LPS rather than a highly complex microbial lysate such as soluble tachyzoite Ag, which we employed in previous studies, provides the advantage that a single molecule with a single receptor (TLR4) is involved, facilitating dissection of PMN-DC interactions. LPS induces neutrophil activation both in vivo and in vitro, as measured by release of cytokines, degranulation, increased adhesion potential, and oxidative burst activity (17, 18, 19, 20, 21).
In this study, we present data showing that mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface upon LPS stimulation. Transmembrane TNF- expression is accompanied by up-regulation of surface CD40L and CD11b/CD18. LPS-triggered mouse neutrophils are potent inducers of DC TNF- when cells are placed in direct contact with each other. This response itself is dependent upon PMN TNF- expression. Endotoxin-activated mouse neutrophils also induce DC IL-12 production, but here neither PMN TNF- nor direct contact was required. The data suggest microbial Ag endow PMN with potent DC-activating capability, a property dependent upon increased expression of neutrophil membrane TNF-.
Materials and Methods
Mice
C57BL/6, TNF receptor (TNFR) 1/2 double knockout (TNFR1/2–/–), TNF- knockout and wild-type (WT) counterparts (B6129SF2/J) were obtained from The Jackson Laboratory. C3H/HeN and C3H/HeJ were purchased from Taconic Farms. Female mice between 6 and 8 wk of age were used throughout. Genotype of knockout animals was confirmed by RT-PCR according to protocols obtained from The Jackson Laboratory. Animals were housed in filter-covered isolator cages in the animal facility of the College of Veterinary Medicine at Cornell University, which is accredited by the American Association for Accreditation of Laboratory Care.
Reagents and Ab
Ultrapure LPS (Escherichia coli strain 0111:B4) was purchased from List Biological Laboratories. FITC-conjugated goat anti-hamster IgG Ab was obtained from eBioscience. Allophycocyanin-conjugated Ab specific for Ly6G, PE-conjugated Ab specific for TNF- (MP6-XT22), CD40L (MR1), CD11b (M1/170), CD14 (rmC5-3), TNFR type II (p75) (TR75-89), purified rat anti-mouse TNF- (MP6-XT22), and hamster anti-mouse TNF- (TN3-19.12) were obtained from BD Pharmingen. Purified antiserum specific for CD68 (T-16) and PE-conjugated TNF- anti-goat Ab were obtained from Santa Cruz Biotechnology. Normal hamster Ig and normal mouse serum (NMS) were purchased from Jackson ImmunoResearch Laboratories. Cycloheximide was obtained from Calbiochem, and recombinant mouse GM-CSF was purchased from PeproTech.
Bone marrow PMN
Neutrophils were isolated from mouse bone marrow following a previously published protocol (22). Briefly, single cell suspensions of bone marrow cells were collected from femur and tibia and resuspended in DMEM supplemented with 5% FCS (Hyclone Laboratories), 100 U/ml penicillin, and 100 mg/ml streptomycin (PenStrep; Invitrogen Life Technologies). Cells were then centrifuged at 500 x g for 7 min at 4°C and resuspended in Ca2+-free HBSS supplemented with 0.38% sodium citrate. The cell suspension was layered on top of a step gradient consisting of 52, 65, and 75% Percoll diluted in Ca2+-free HBSS, and centrifuged at 1500 x g for 30 min at 4°C. Neutrophils were recovered at the interface between the 65 and 75% Percoll layers. The proportion of neutrophils, determined by Diff-Quik staining of cytospin preparations, was routinely >90%.
Peritoneal neutrophils
Mice were i.p. injected with 1 ml of 10% thioglycolate (Difco). Peritoneal exudate cells were obtained by lavage 18 h later with ice-cold PBS. Cells were washed in PBS, passed through a 70-μm nylon cell strainer, and erythrocytes in the suspension were lysed using Red cell lysis buffer (Sigma-Aldrich). Cells were then washed, resuspended at 2 x 107 cells/ml in buffer composed of Dulbecco’s PBS (Invitrogen Life Technology) containing 0.5% BSA (Sigma-Aldrich), 1 mM EDTA (Fisher Scientific), and incubated for 15 min at 4°C with anti-MHC class II magnetic microbeads. After washing, the mixture was transferred to columns installed within a magnetic apparatus to remove MHC class II-expressing cells, according to the manufacturer’s instructions (Miltenyi Biotec).
Bone marrow-derived DC (BMDC)
Generation of BMDC was accomplished following a previously published protocol (23). Briefly, single cell bone marrow preparations were obtained as described above, cells were washed in RPMI 1640 (Fisher Scientific) and resuspended at 2 x 105 cells/ml in DC medium composed of RPMI 1640 supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin, 10% FCS, 5 x 10–5 M 2-ME, and 20 ng/ml GM-CSF. Cells were plated on 100 x 15 mm standard sterile polystyrene petri dishes (Fisher Scientific) and cultured for 9 days at 37°C in 5% CO2. Fresh DC medium, containing 20 ng/ml GM-CSF, was added on days 3, 6, and 8 after culture initiation. On day 9, cells were resuspended in cDMEM composed of DMEM (Invitrogen Life Technologies) supplemented with 10% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin, 5 x 10–5 M 2-ME, 10 mM HEPES (Invitrogen Life Technologies), 100 μM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), alone or in the presence of different stimuli at 37°C in 5% CO2. Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C.
PMN culture
Peritoneal PMN (2 x 106/ml) were cultured in cDMEM alone or in the presence of 1 μg/ml LPS at 37°C in 5% CO2 in a 96-well tissue culture plate (Corning Costar). Supernatants from the cultures were recovered 18 h later and either used immediately or stored at –80°C. In the coculture experiments, neutrophils were cultured in medium or LPS (1 μg/ml) for 3 h followed by three washes; then they were either plated together with immature BMDC in 12-well tissue culture plates (Corning Costar) or placed in the upper well of a Transwell system (Corning Costar), while BMDC were added to the lower well. In some experiments PMN were fixed for 15 min on ice in 1% formaldehyde solution. In experiments involving blocking of new protein synthesis, neutrophils were incubated in medium containing 30 μg/ml cycloheximide for 15 min followed by a stimulation with 1 μg/ml LPS.
RT-PCR
RNA was isolated, reverse transcribed and subjected to PCR amplification as described (12). The primer sequences used were: -actin, TGACGGGGGTCACCCACACTGTGCCCATCTA (sense), CTAGAAGCATTGCGGTGGACGATGGAGGG (antisense); TNF-, CAGCCTCTTCTCATTCCTGCTTGTG (sense), CTGGAAGACTCCTCCCAGGTATAT (antisense). The cDNA was amplified 31 cycles.
Flow cytometry
To analyze surface markers on PMN, Fc receptors were blocked in FACS buffer (PBS, 1% BSA, and 0.1% sodium azide) containing 5 μg/ml anti-mouse CD16/CD32 (BD Pharmingen) and 10% NMS for 15 min at 0°C, then cells were stained with optimal concentrations of allophycocyanin-conjugated anti-Gr-1(Ly-6G) in combination with PE-conjugated antisera specific for TNF-, CD40L, CD11b, CD14, TNFR type II (p75), and unlabeled goat anti-CD68 followed by PE-conjugated anti-goat Ab for 30 min at 0°C. For intracellular cytokine detection, peritoneal cells were blocked as described above, stained with allophycocyanin-conjugated anti-Gr-1 Ab and unlabeled anti-TNF- Ab, then cells were fixed in 3% paraformaldehyde (Sigma-Aldrich), 0.1 mM CaCl2, and 0.1 mM MgCl2 for 30 min at 0°C. Cells were subsequently washed in permeabilization buffer (PBS with 0.075% saponin) and incubated for 15 min at 0°C in permeabilization buffer containing 10% NMS. After two washes in permeabilization buffer, unlabeled hamster anti-TNF- was added, and cells were incubated for 30 min at 0°C. After washing in permeabilization buffer, FITC-conjugated anti-hamster was added followed by a 30-min incubation at 0°C and subsequently washed for flow cytometric analysis. Data was acquired on FACSCalibur system (10,000 events per sample) and analyzed with CellQuest software (BD Biosciences Immunocytometry Systems).
Cytokine ELISA
IL-12p40 was measured as previously described (12), and TNF- was measured using a commercially obtained kit (BD Pharmingen).
Statistical Analysis
The statistical significance of the data was analyzed using SD and unpaired two-tailed Student’s t test.
Results
LPS triggers rapid changes in peritoneal PMN surface membrane-associated molecules
We began by examining the influence of endotoxin on expression of neutrophil cell surface markers. Flow cytometric analysis revealed early LPS-induced up-regulation of surface TNF-, CD40L, and CD11b, followed by a time-dependent down-regulation relative to cells cultured in medium alone (Fig. 1A). The surface expression of these molecules in medium alone at 30 min was equivalent to staining of freshly isolated, nonstimulated peritoneal neutrophils (data not shown). We also examined expression of CD14, TNFR type II, and CD68, which are markers of exocytosis of secretory vesicles, specific and azurophilic granules, respectively (24, 25, 26, 27). Both CD14 and CD68 were rapidly mobilized to the cell surface, consistent with exocytosis of secretory vesicles and azurophilic granules. Interestingly, the influence of LPS on TNFR type II expression was distinct, in that LPS triggered early down-regulation. This was followed by a gradual increase in TNFR type II surface expression that required 18 h to reach maximal levels (Fig. 1B). These effects are likely mediated by TLR4, because increased expression of TNF- and other molecules did not occur in peritoneal PMN from the LPS nonresponder strain C3H/HeJ (Fig. 1C and data not shown).
FIGURE 1. LPS triggers rapid changes in the pattern of molecules expressed on the neutrophil surface. Thioglycolate-elicited peritoneal exudate cells from C57BL/6 (A and B), and C3H/HeN/C3H/HeJ (C) mice were stimulated with LPS (thick lines) or incubated in medium (thin lines) for varying times, then stained with Ab to Gr-1 and the indicated markers. The data show expression profiles gated on Gr-1high cells. C, Surface TNF- expression 60 min after treatment with LPS or medium. The numbers in A and C indicate the percentage of positive cells after stimulation with LPS (top) vs incubation in medium (bottom). This experiment was repeated three times with essentially identical results.
Control of neutrophil transmembrane TNF-
We next focused on surface expression of TNF- by peritoneal PMN. TNF- is synthesized as a transmembrane 26-kDa protein that is transported to the cell surface where it is cleaved by a metalloproteinase (TACE) to release the soluble 17-kDa protein (28, 29, 30, 31, 32, 33, 34). Therefore, increased TNF- levels on the cell surface (Fig. 1A) could be the result of higher levels of transmembrane cytokine or, alternatively, the result of soluble TNF- binding to neutrophil TNF receptors. To distinguish these possibilities, we stimulated WT or TNFR1/2–/– peritoneal neutrophils with LPS and analyzed surface expression of TNF-. Both WT (Fig. 2A) and TNFR1/2–/– (Fig. 2B) peritoneal PMN up-regulated surface TNF- expression after 1 h of stimulation with LPS. This result argues that transmembrane 26-kDa TNF- is up-regulated on the PMN cell surface, rather than a situation in which soluble TNF- binds to neutrophil TNF receptors. In addition, preincubation of cells with unlabeled anti-TNF- Ab prevented binding of the PE-conjugated Ab, confirming the specificity of Ab binding (Fig. 2C).
FIGURE 2. LPS induces neutrophil surface TNF- expression independently of TNF receptors. Thioglycolate-induced peritoneal exudate cells from WT (A and C) and TNFR1/2–/– (B) mice were cultured in medium (thin lines) or stimulated with LPS (thick lines) for 1 h and stained for surface TNF-. C, Cells were stained with an unlabeled blocking anti-TNF- Ab followed by staining for TNF- as in A and B. The profiles shown are gated on the Gr-1high population.
We previously found that mouse PMN contain an intracellular pool of IL-12 (35). Because we observed surface expression of TNF- very early after LPS stimulation in our experiments, we investigated the possibility that mouse PMN also contain preformed TNF-. To address this question, we first pretreated peritoneal PMN with cycloheximide to block new protein synthesis before stimulation of cells with LPS. Cycloheximide had no effect on early LPS-induced up-regulation of surface TNF-. However, after 1 h of LPS stimulation, cycloheximide-treated cells displayed a marginal but reproducible decrease in TNF- surface expression, suggesting that de novo protein synthesis may contribute to surface expression at later time points (Fig. 3A). To further examine the presence of preformed TNF-, we blocked surface TNF-, permeabilized cells, and stained for intracellular cytokine. In this case, bone marrow and peritoneal exudate PMN were strongly positive for intracellular TNF- (Fig. 3B).
FIGURE 3. Mouse neutrophils contain preformed TNF- that is rapidly mobilized to the cell surface after LPS stimulation. A, Early LPS-triggered PMN surface TNF- expression does not require new protein synthesis. Thioglycolate-elicited cells (PEC) were preincubated with cycloheximide then stimulated for the indicated times with LPS. Cells were stained with Ab to Gr-1 and TNF-. The profiles shown are gated on Gr-1high cells. Green line, LPS with no cycloheximide; red line, LPS with cycloheximide; black line, medium with cycloheximide. No difference was observed between medium and medium with cycloheximide (data not shown). B, PMN contain intracellular TNF-. Bone marrow cells and PEC were subjected to surface TNF- blocking using unlabeled rat anti-mouse TNF- Ab, then permeabilized and stained with hamster anti-mouse TNF- Ab. Data shown are gated on Gr-1high cells. Red line, anti-TNF- Ab; black line, isotype control. C, Changes in intracellular and membrane bound TNF- during LPS stimulation. Thioglycolate-elicited cells were incubated in medium or stimulated with LPS for the indicated times. Cells were subsequently stained with Gr-1 Ab and surface TNF- using a rat anti-mouse TNF-. After permeabilization and blocking, cells were stained for intracellular TNF- using a hamster anti-mouse Ab. Profiles shown are gated on Gr-1high cells. Green and gray lines, intracellular TNF- expression after incubation with LPS and medium, respectively. Red and black lines, surface TNF- expression after incubation with LPS and medium, respectively. Dotted lines, isotype control staining. Numbers in each histogram show percentage of positive followed by geometric mean fluorescence intensity for LPS-stimulated cells (top) and cells incubated in medium alone (bottom). D, RT-PCR analysis of purified peritoneal exudate PMN incubated in medium alone or stimulated for the indicated times with LPS.
We next examined modulation of surface vs intracellular expression of TNF- in LPS-stimulated peritoneal PMN. At 10 min after LPS stimulation, when compared with unstimulated cells, we observed a decrease in intracellular TNF- and a parallel increase in surface expression of this cytokine (Fig. 3C). However, after 1 h of exposure to LPS, both surface and intracellular TNF- expression were increased (Fig. 3C). These results suggest that de novo protein synthesis replenishes the pool of intracellular TNF- that is initially mobilized to the neutrophil surface. Evidence that LPS induces increased TNF- gene transcription is shown by RT-PCR in Fig. 3D. Increased levels of TNF- mRNA were apparent within 15 min, and maximal levels were attained by 60 min of stimulation. The combined results argue that both preformed pools and newly synthesized TNF- contribute to high level expression of this cytokine on the cell surface.
LPS stimulated mouse peritoneal neutrophils promote dendritic cell cytokine production
To determine whether LPS-triggered peritoneal PMN can directly stimulate TNF- and IL-12 production by BMDC, we began by measuring cytokine levels in PMN-DC cocultures. When BMDC were cultured with LPS or cocultured with LPS-activated peritoneal PMN, high levels of both TNF- and IL-12p40 were released into the coculture supernatants (Fig. 4A). In contrast, when nonstimulated mouse peritoneal PMN were used in the cocultures, no TNF- and minimal amounts of IL-12p40 were detected (Fig. 4A). To test whether the triggering effect of peritoneal PMN was due to bona fide factors produced by the latter cells or simply to LPS carryover, we conducted identical experiments employing BMDC derived from LPS nonresponder animals. In this case, although direct DC stimulation with LPS failed to elicit a cytokine response, coculture with LPS-stimulated peritoneal neutrophils triggered robust TNF- and IL-12p40 responses (Fig. 4B). This result demonstrates that the cytokine responses were triggered by neutrophil factors rather than residual LPS in the cocultures.
FIGURE 4. LPS-triggered PMN express factors that trigger DC TNF- and IL-12 production. BMDC were incubated with medium (open bars), LPS (black bars; 1 μg/ml), unstimulated PMN (gray bars), and LPS-stimulated PMN (cross-hatched bars). The ratio of PMN to DC was 2:1. After 18 h of incubation (37°C, 5% CO2), DC culture supernatants were collected for TNF- and IL-12p40 ELISA. A, BMDC from C3H/HeN (LPS responder) and B, BMDC from C3H/HeJ (LPS nonresponder) mice. Levels of cytokine detected in PMN culture supernatants after 3 h incubation in medium alone or in the presence of LPS followed by 18 h in medium, respectively, were: 31 ± 2 pg/ml; 109 ± 7 pg/ml for TNF- and <1 pg/ml and 147 ± 1 pg/ml for IL-12p40. n.d., Not detected. Data are representative of three different experiments.
Direct cell-to-cell contact and peritoneal PMN TNF- are required for optimal DC production of TNF- but not IL-12p40
We next compared the role of soluble vs membrane-bound PMN-derived factors in promoting cytokine release by BMDC. Accordingly, we exposed LPS nonresponder BMDC to cell-free supernatants from 18-h LPS-stimulated mouse peritoneal PMN. In parallel, we cocultured 3-h LPS-stimulated peritoneal PMN with the BMDC as in Fig. 4B. Although DC exposed to soluble factors from LPS-triggered peritoneal PMN released both TNF- and IL-12p40, the response was enhanced when BMDC were cocultured with LPS-stimulated peritoneal PMN (Fig. 5). This was particularly true for TNF- production, where there was a >6-fold increase in the amount of cytokine released in a direct cell-to-cell contact situation. We next used a transwell system to further probe the requirement for optimal DC cytokine production. As shown in Fig. 6A, TNF- production was severely curtailed when LPS-triggered WT mouse peritoneal PMN, and DC were separated by a membrane. In contrast, WT DC IL-12p40 production in response to LPS-stimulated WT peritoneal PMN was minimally affected (Fig. 6A).
FIGURE 5. Cell-to-cell contact is required for optimal DC TNF- but not IL-12p40 production. C3H/HeJ (LPS nonresponder) BMDC were incubated 18 h with supernatants from unstimulated (open bars) or 18 h with LPS-stimulated (black bars) PMN. In addition, DC were coincubated with PMN that had been exposed to medium (gray bars) or LPS (cross-hatched bars) for 3 h. A 2:1 ratio of PMN to DC was used in the coincubation experiments. After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. PMN alone stimulated with LPS released 311 ± 16 pg/ml (TNF-) and 3490 ± 139 pg/ml (IL-12p40). The experiment was repeated twice with similar results. *, p < 0.01 (DC + 18 h PMN supernatants vs DC-PMN coculture). n.d., Not detected.
FIGURE 6. Requirements for PMN triggering of DC cytokine production. Thioglycolate-elicited peritoneal PMN were preincubated 3 h with LPS, then added directly to BMDC (black bar) or separated in a Transwell system (open bar). Supernatants were collected 18 h later for TNF- and IL-12p40 ELISA. A, WT and TNF-–/– PMN were compared for ability to trigger WT DC. B, WT PMN were used to trigger WT and TNFR1/2–/– DC. C, WT neutrophils were used to trigger WT and CD40–/– DC. No cytokine release was detected when nonstimulated PMN were used in each of these experiments (data not shown). This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs Transwell). **, p < 0.01 (direct cell-to-cell contact between WT DC and WT PMN vs direct cell-to-cell contact between WT DC and TNF–/– PMN). n.d., Not detected.
We previously demonstrated that supernatants from T. gondii-triggered PMN induce DC up-regulation of costimulatory molecules CD40 and CD86 and that this response depends upon TNF- (10). In this study, we investigated the involvement of PMN-derived TNF- in promoting cytokine production in PMN-BMDC coculture experiments. Although LPS-stimulated WT peritoneal PMN were able to trigger TNF- production by BMDC, LPS-stimulated TNF–/– neutrophils were defective in driving such a response (Fig. 6A). However, both LPS-stimulated WT and TNF–/– peritoneal PMN were able to promote DC IL-12p40 production in cocultures and when separated in a Transwell chamber system (Fig. 6A). Similarly, using TNFR1/2–/– DC, we found that LPS-stimulated peritoneal PMN no longer elicited DC TNF- release while leaving IL-12p40 production unaffected (Fig. 6B). Last, DC lacking CD40 were examined. Here, there was no evidence for involvement of this costimulatory molecule in DC cytokine production (Fig. 6C). Collectively, the results in Fig. 6 show that direct cell-to-cell contact is required for optimal TNF- but not IL-12p40 production, and that DC TNF- but not IL-12p40 release depends upon neutrophil TNF- production.
Fixed mouse peritoneal neutrophils retain the ability to stimulate DC TNF- release
The results presented so far indicate that cell-to-cell contact between DC and PMN directs TNF- production by DC, but they do not rule out additional involvement of neutrophil soluble factors. To address this issue, peritoneal neutrophils were subjected to paraformaldehyde fixation before DC stimulation. This procedure results in loss of ability of PMN to secrete factors, but surface membrane molecules will be retained. As shown in Fig. 7A, fixation had no effect on the ability of LPS-triggered peritoneal neutrophils to drive DC TNF- production. In contrast, fixation ablated the ability of PMN to drive DC IL-12p40 production, consistent with a role for soluble factors in this situation (Fig. 7B). These results demonstrate that direct contact between DC and PMN is necessary and sufficient to promote TNF- production by DC, but that soluble factors induce DC IL-12 production.
FIGURE 7. LPS-stimulated fixed mouse neutrophils retain the ability to stimulate DC TNF- production. Peritoneal PMN were incubated for 3 h in medium or LPS, washed, fixed or left untreated, and added directly to BMDC (black bars) or separated in a Transwell system (open bars). After 18 h, supernatants were collected for TNF- (A) and IL-12p40 (B) ELISA. Untreated PMN alone stimulated with LPS released 143 ± 1 pg/ml (TNF-) and 395 ± 2 pg/ml (IL-12). No cytokine release was detected when nonstimulated PMN were used (data not shown). In addition, no cytokine was detected in supernatants of LPS-stimulated fixed neutrophils, and no DC IL-12 was detected using fixed neutrophils as stimulators. This experiment was repeated twice with similar results. *, p < 0.01 (direct cell-to-cell contact vs transwell). n.d., Not detected.
Discussion
The results of this study demonstrate that mouse neutrophils triggered with microbial Ag become licensed activators of DC. Although PMN themselves serve as a source of proinflammatory cytokines IL-12p40 and TNF-, the amount of cytokine released by neutrophil-activated DC is 1–2 orders of magnitude higher (Figs. 4–6). Importantly, by using DC with nonfunctional TLR4, it was possible to rule out involvement of endotoxin carryover in the effects of LPS-triggered neutrophils on DC. The DC TNF- release was triggered by neutrophil TNF-, and this was optimally driven by direct cell-to-cell contact.
Neutrophil TNF- was mobilized to the membrane surface within 15 min of endotoxin stimulation relative to cells incubated for the same length of time in medium alone. Surface TNF- up-regulation occurred in the presence of cycloheximide indicating that de novo protein synthesis was not required. Furthermore, nonstimulated mouse neutrophils contained an intracellular pool of TNF- that displayed a transient decrease concomitant with increased surface TNF- expression. In addition to TNF-, PMN have previously been shown to contain preformed IL-12, IL-6, and MIP-2 (35, 36, 37). Both TNF- and IL-12 are present in non-elicited mouse neutrophils. It is, therefore, likely that these cytokines are expressed in PMN during the course of differentiation in the bone marrow. Thus, in addition to their well-known role as microbicidal effectors, neutrophils may also serve an important function as vehicles that rapidly deliver preformed cytokines to sites of inflammation and infection.
Whereas intracellular TNF- pools were depleted 10 min after LPS stimulation, after 60 min, intracellular TNF- was replenished. LPS triggering also induced increased levels of TNF- mRNA. The data argue that, although mouse PMN possess an intracellular pool of preformed cytokine, the cells also display the ability to up-regulate cytokine gene induction, as shown here for TNF- and elsewhere for IL-12 (35). Sustained PMN production of inflammatory cytokines, particularly for the case of GM-CSF, is likely to be important for delaying onset of apoptosis that these cells are programmed to undergo (38, 39). In this way, immunoregulatory and microbicidal effects of neutrophils at sites of inflammation and infection are likely to be prolonged.
We, and others, have suggested that neutrophils, by virtue of their pattern of cytokine expression, may have important in vivo immunoregulatory effects on DC (10, 40, 41, 42). In addition to triggering DC TNF- and IL-12 production, activated PMN exert other effects on these cells. For example, mouse neutrophil release of chemokines CCL3, CCL4, CCL5, and CCL20 together mediate potent chemotactic activity on immature DC (10). Furthermore, PMN induce potent DC up-regulation of costimulatory molecules such as CD40 and CD86, an effect that is in part dependent upon neutrophil TNF- release (10). Other recent studies show that neutrophils assume type 1 and type 2 cytokine profiles, depending on the activating stimulus (43, 44, 45). Together, these findings provide evidence for a key role for neutrophils in controlling immunoregulatory effects of DC.
An unexpected conclusion emerging from our studies is that neutrophil-directed production of DC TNF- and IL-12 is non-identical. Thus, DC IL-12 production does not rely on neutrophil TNF-. In contrast, DC TNF- production requires TNF- expression by peritoneal neutrophils. Furthermore, PMN-induced release of DC IL-12, in contrast to TNF-, occurs equally efficiently across a transwell membrane relative to when cells are in direct contact. CD40 ligation is well known to be involved in DC activation (46), and indeed we show here that LPS induces up-regulation of PMN CD40L expression. Nevertheless, using CD40–/– DC, we were able to exclude a role for CD40:CD40L in effects of PMN on DC. Likewise, IL-12 has been implicated as a priming signal for high-level DC IL-12 production (47), but experiments employing IL-12–/– PMN demonstrate lack of involvement of this cytokine (unpublished observations). The potential role of other neutrophil-derived IL-12-inducing mediators, for example bradykinin, is currently under investigation (48).
Our studies here and elsewhere (9, 10, 12) employ thioglycolate-elicited peritoneal and bone marrow neutrophils. In principle, these populations are non-identical since the former has responded to an inflammatory stimulus. Nevertheless, we have yet to detect differences in these populations, particularly with regard to cytokine/chemokine production. In addition, it can be argued that elicited PMN are the most relevant to study because neutrophils arriving at a site of infection do so in response to inflammatory signals.
The biological significance of DC TNF- production is presently unclear. The cells are more well-known as an IL-12 source that, together with Ag and costimulation, play a central role in the generation of Th1 effector lymphocytes. Nevertheless, DC subsets have been identified in vivo that mediate innate immune defense against microbial infection by producing both NO and TNF- (49). The role of TNF- in this situation may be to promote, along with other inflammatory mediators, macrophage and DC acquisition of microbicidal function.
In vivo evidence for the importance of neutrophils in DC function is provided by observations that depletion of Gr-1(Ly6G)+ cells results in decreased TNF- and IL-12 production by splenic CD11c+ cells and increased susceptibility during infection with microbial pathogens such as T. gondii (10). In this regard, activated PMN that rapidly accumulate at a site of infection or inflammation may play an important role in driving acquisition of DC microbicidal function. The fact that direct cell contact is required for optimal DC TNF- production may be a way to focus the response solely at inflammatory sites of infection and neutrophil accumulation.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank Barbara Butcher, Laura Del Rio, and Matthias Hesse for constructive criticism and insightful discussion during the course of this work.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI47888.
2 Address correspondence and reprint requests to Dr. E. Denkers, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853-6401. E-mail address: eyd1{at}cornell.edu
3 Abbreviations used in this paper: DC, dendritic cell; PMN, polymorphonuclear leukocyte; BMDC, bone marrow-derived DC; NMS, normal mouse serum; WT, wild type.
Received for publication October 12, 2004. Accepted for publication January 10, 2005.
References
Janeway, C. A., Jr, R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197.
Palucka, K., J. Banchereau. 2002. How dendritic cells and microbes interact to elicit or subvert protective immune responses. Curr. Opin. Immunol. 14:420.
Guermonprez, P., J. Valladeau, L. Zitvogel, C. Thery, S. Amigorena. 2002. Antigen presentation and T cell stimulation by dendritic cells. Annu. Rev. Immunol. 20:621.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Belaaouaj, A., R. McCarthy, M. Baumann, Z. Gao, T. J. Ley, S. N. Abraham, S. D. Shapiro. 1998. Mice lacking neutrophil elastase reveal impaired host defense against gram negative bacterial sepsis. Nat. Med. 4:615.
Jackson, S. H., J. I. Gallin, S. M. Holland. 1995. The p47phox mouse knockout model of chronic granulomatous disease. J. Exp. Med. 182:751.
Cassatella, M. A.. 1999. Neutrophil-derived proteins: selling cytokines by the pound. Adv. Immunol. 73:369
Del Rio, L., B. A. Butcher, S. Bennouna, S. Hieny, A. Sher, E. Y. Denkers. 2004. Toxoplasma gondii triggers MyD88-dependent and CCL2 (MCP-1) responses using distinct parasite molecules and host receptors. J. Immunol. 172:6954.
Bennouna, S., S. K. Bliss, T. J. Curiel, E. Y. Denkers. 2003. Cross-talk in the innate immune system: neutrophils instruct early recruitment and activation of dendritic cells during microbial infection. J. Immunol. 171:6052.
Bliss, S. K., A. J. Marshall, Y. Zhang, E. Y. Denkers. 1999. Human polymorphonuclear leukocytes produce IL-12, TNF-, and the chemokines macrophage-inflammatory protein-1 and -1 in response to Toxoplasma gondii antigens. J. Immunol. 162:7369
Bliss, S. K., Y. Zhang, E. Y. Denkers. 1999. Murine neutrophil stimulation by Toxoplasma gondii antigen drives high level production of IFN--independent IL-12. J. Immunol. 163:2081.
Del Rio, L., S. Bennouna, J. Salinas, E. Y. Denkers. 2001. CXCR2 deficiency confers impaired neutrophil recruitment and increased susceptibility during Toxoplasma gondii infection. J. Immunol. 167:6503.
Scharton-Kersten, T., G. Yap, J. Magram, A. Sher. 1997. Inducible nitric oxide is essential for host control of persistent but not acute infection with the intracellular pathogen Toxoplasma gondii. J. Exp. Med. 185:1.
Sayles, P. C., L. J. Johnson. 1997. Exacerbation of toxoplasmosis in neutrophil depleted mice. Nat. Immun. 15:249.
Bliss, S. K., L. C. Gavrilescu, A. Alcaraz, E. Y. Denkers. 2001. Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69:4898.
Wright, S. D., R. A. Ramos, A. Hermanowski-Vosatka, P. Rockwell, P. A. Detmers. 1992. Activation of the adhesive capability of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J. Exp. Med. 173:1281.
Guthrie, L. A., L. C. McPhail, P. M. Henson, R. B. J. Johnston. 1984. Priming of neutrophils enhanced release of oxygen metabolites by bacterial lipopolysaccharide: evidence for increased activity of the superoxide-producing enzyme. J. Exp. Med. 160:1656.
Surette, M. E., R. Palmantier, J. Gosselin, P. Borgeat. 1993. Lipopolysaccharides prime whole human blood and isolated neutrophils for the increased synthesis of 5-lipoxygenase products by enhancing arachidonic acid availability: involvement of CD14 antigen. J. Exp. Med. 178:1347.
Dubravec, D. B., D. R. Spriggs, J. A. Mannick, M. L. Rodrick. 1990. Circulating human peripheral blood granulocytes synthesize and secrete tumor necrosis factor . Proc. Natl. Acad. Sci. USA 87:6758
Cassatella, M. A., L. Meda, S. Gasperini, A. D‘Andrea, X. Ma, G. Trinchieri. 1995. Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25:1
Kiefer, F., J. Brumell, N. Al-Alawi, S. Latour, A. Cheng, A. Veillette, S. Grinstein, T. Pawson. 1998. The Syk protein tyrosine kinase is essential for Fc receptor signaling in macrophages and neutrophils. Mol. Cell Biol. 1998. 18:4209.
MacDonald, A. S., A. D. Straw, B. Bauman, E. J. Pearce. 2001. CD8– dendritic cell activation status plays an integral role in influencing Th2 response development. J. Immunol. 167:1982.
Borregaard, N., J. B. Cowland. 1997. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89:3503.
Bainton, D. F.. 1999. Distinct granule populations in human neutrophils and lysosomal organelles identified by electron microscopy. J. Immunol. Methods 232:153.
Gullberg, U., N. Bengtsson, E. Bulow, E. Garwicz, A. Lindmark, I. Olsson. 1999. Processing and targeting of granule proteins in human neutrophils. J. Immunol. Methods 232:201.
Beutler, B., A. Cerami. 1989. The biology of cachectin/TNF- primary mediator of the host response. Annu. Rev. Immunol. 7:625.
Black, R. A., C. T. Rauch, C. J. Kozlosky, J. J. Peschon, J. L. Slack, M. F. Wolfson, B. J. Castner, K. L. Stocking, P. Reddy, S. Srinivasan, et al 1997. A metalloproteinase disintegrin that releases tumor-necrosis factor- from cells. Nature 385:729.
Grohmann, U., M. L. Belladonna, R. Bianchi, C. Orabona, E. Ayroldi, M. C. Fioretti, P. Puccetti. 1998. IL-12 acts directly on DC to promote nuclear localization of NF-B and primes DC for IL-12 production. Immunity 9:315.
Aliberti, J., J. P. Viola, A. Vieria-de-Abreu, P. T. Bozza, A. Sher, J. Scharfstein. 2003. Cutting Edge: bradykinin induces IL-12 production by dendritic cells: a danger signal that drives Th1 polarization. J. Immunol. 170:485.
Serbina, N. V., T. P. Salazar-Mathar, C. A. Biron, W. A. Kuziel, E. A. Pamer. 2003. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19:59.(Soumaya Bennouna and Eric)