The expression and function of Nod-like receptors in neutrophils

Authors


Professor L. O. Cardell, Division of ENT Diseases, Department of Clinical Sciences, Intervention and Technology, Karolinska Institutet, Karolinska University Hospital, S-141 86 Stockholm, Sweden. Email: lars-olaf.cardell@ki.se
Senior author: Anna-Karin Ekman, email: anna-karin.ekman@ki.se

Summary

Neutrophils make up an essential part of the innate immune system, and are involved both in the initial responses to pathogens, and in orchestrating later immune responses. Neutrophils recognize pathogens through pattern-recognition receptors (PRRs), which are activated by microbial motifs. The Nod-like receptors (nucleotide-binding domain leucine-rich repeat containing family; NLRs) constitute a recently discovered group of PRRs whose role in the neutrophil immune responses is not yet characterized. The present study aimed to investigate the expression and function of NLRs in neutrophils. Neutrophils were isolated from human peripheral blood, and the presence of nucleotide-binding oligomerization domain 1 (NOD1), NOD2 and NACHT-LRR-PYD-containing protein 3 (NLRP3) was evaluated with flow cytometry and immunohistochemistry. The expression of NOD1, NOD2 and NLRP3 messenger RNA was determined using real-time reverse transcription–polymerase chain reaction. Changes in neutrophil cytokine secretion, phenotype and migration following agonist-induced activation were studied using enzyme-linked immunosorbent assay, flow cytometry and a chemotaxis assay, respectively. No expression of NOD1 was found in isolated neutrophils and stimulation with the NOD1 ligand γ-d-glutamyl-meso-diaminopimelic acid induced no signs of activity. In contrast, a marked expression of NOD2 and NLRP3 was found. NOD2 activation with MurNAc-l-Ala-d-isoGln (MDP) resulted in interleukin-8 secretion, CD62 ligand down-regulation, CD11b up-regulation and increased migration towards an inflammatory stimulus. NLRP3 activation with alum caused interleukin-1β secretion and facilitated migration. Altogether, this suggests that NLRs may be a previously unknown pathway for neutrophil activation.

Introduction

Neutrophils constitute an essential part of our immunological defence. In response to invading microbes they release cytokines and chemokines, which attract other cells to the site of infection.1 The role of pattern-recognition receptors (PRRs) in this process is the focus of growing interest. Toll-like receptors (TLRs) are the best characterized PRRs. Recently, another group of PRRs, the Nod-like receptors (nucleotide-binding domain leucine-rich repeat containing; NLRs), has emerged.2,3 NLRs are found in the cell cytosol and so far 23 members of the NLR family have been described in humans.4 Among these, nucleotide-binding oligomerization domain 1 (NOD1), NOD2 and NACHT-LRR-PYD-containing protein 3 (NLRP3) are the best characterized. NOD1 is activated by the peptidoglycan fragment γ-d-glutamyl-meso-diaminopimelic acid (iE-DAP), and NOD2 recognizes the peptidoglycan fragment MurNAc-L-Ala-D-isoGln (MDP).5,6 The iE-DAP is commonest in Gram-negative bacteria whereas MDP is found in both Gram-negative and Gram-positive bacteria.4 NLRP3 is activated by danger signals such as extracellular ATP released upon cell damage or non-apoptotic cell death.7 Recently, NLRP3 has also been found to mediate effects in response to alum (aluminium hydroxide) adjuvant.8,9 Activation of the NOD1 and NOD2 receptors induces an inflammatory response via nuclear factor-κB-related pathways.10 NLRP3 is a constituent of the multiprotein complex known as the inflammasome, the activation of which results in increased interleukin-1β (IL-1β) production via caspase-1.11,12 NLRs have been implicated in several immunological diseases such as Crohn’s disease and asthma,13,14 but the knowledge of their role in neutrophils is still limited. The present study was designed to characterize the expression and function of NLRs in neutrophils.

Materials and methods

Materials

Polymorphprep™ was obtained from NycoMed (Axis-Shield PoC AS, Oslo, Norway); iE-DAP, γ-d-Glu-Lys (iE-Lys), MDP and MDP-negative control were purchased from Invivogen (San Diego, CA). Alum gel was acquired from Sigma-Aldrich (St Louis, MO). Phycoerythrin-conjugated CD62 ligand [CD62L-PE; mouse immunoglobulin G1 (IgG1), clone DREG56] and fluorescein isothiocyanate-conjugated (-FITC) CD11b (mouse IgG1, clone Bear1) were obtained from Immunotech (Marseille, France). RPMI-1640 medium supplemented with 0·3 g/l l-glutamine came from PAA Laboratories (Pasching, Austria), and penicillin/streptomycin were obtained from Invitrogen (Carlsbad, CA). Rabbit polyclonal anti-CARD4 antibodies, used for detection of NOD1, were purchased from Abcam (Cambridge, UK), as were mouse monoclonal anti-CARD15 (mouse IgG1 clone 2D9) and mouse monoclonal anti-NLRP3 (mouse IgG1, clone nalpy3-b), used to detect NOD2 and NLRP3 protein, respectively.

Neutrophil isolation

Venous peripheral blood was collected from healthy volunteers in Heparin test tubes (BD Vacutainer™ 367675). All participating subjects had given their informed consent, and the study was approved by the Regional Ethics Committee of Karolinska Institutet. The blood was layered on a Polymorphprep™ cushion and cells were isolated according to the manufacturer’s protocol. Briefly, neutrophils were isolated on the basis of density, washed once in 0·5 N RPMI-1640 to restore osmolarity, and then washed once more in RPMI-1640. The cell pellet was resuspended in RPMI-1640 and the cells were dyed with Gentiana violet then counted in a Bürker chamber. The neutrophils were subsequently diluted in complete medium consisting of RPMI-1640 supplemented with 0·3 g/l l-glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin. When the neutrophils were to be used for immunohistochemistry, they were isolated on a cushion of Ficoll-Paque™ plus (GE Healthcare Bio-sciences AB, Uppsala, Sweden), followed by lysis of remaining erythrocytes using ammonium chloride erythrocyte lysis solution (0·8% NH4Cl, 10 mm KHCO3, 0·1 mm ethylenediaminetetraacetic acid).

NLR protein expression

To examine the intracellular protein expression of NOD1, NOD2 and NLRP3, Intraprep™ permeabilization reagent kit (Immunotech) was used according to the manufacturer’s protocol. Briefly, freshly isolated cells were fixed and permeabilized. Thereafter, NLR antibodies and appropriate controls were added and samples were incubated for 15 min at room temperature in the dark. For detection of NOD1, rabbit polyclonal anti-CARD4 antibodies were used. These were fluorochrome-conjugated using polyclonal goat anti-rabbit IgG-H&L-FITC (Abcam). Isotype controls of rabbit IgG-FITC (Serotec, Oxford, UK) were used to rule out background fluorescence.

To detect NOD2 and NLRP3 expression, mouse monoclonal anti-CARD15 and mouse monoclonal anti-NLRP3 were conjugated using an Alexafluor 488 mouse IgG1 labelling kit (Molecular Probes, Eugene, OR). Isotype mouse IgG1 antibody (Dako, Glostrup, Denmark), conjugated with Alexa488, was used as control.

NOD1, NOD2 and NLRP3 protein expression in cells was analysed on a Coulter Epics XL-MCL (Beckman Coulter, Marseilles, France), collecting 10 000 events and gating neutrophils on the basis of forward-scatter and side-scatter properties. The data were analysed using Expo32 Analysis software (Beckman Coulter).

Immunohistochemistry

For immunohistochemistry, isolated neutrophils were fixed in a solution containing equal amounts of 10% neutral-buffered formalin (Applichem, VWR, Stockholm, Sweden) and 95% ethanol, along with 1% eosin, and were embedded in paraffin. The sections were used for immunohistochemistry to determine the expression of NOD1, NOD2 and NLRP3 protein. The sections were deparaffinated by immersion in xylene, and were rehydrated in pure ethanol and 95% ethanol. The antigens were retrieved by heating the slides with Target retrieval solution (Dako), and the cells were subsequently permeabilized using Triton-X (1% in Tris-buffered saline) for 20 min. Antibody staining was carried out in a moisture chamber using an EnVision™ kit (Dako) for mouse or rabbit 3,3′-diaminobenzidine (DAB) -staining. Briefly, the slides were incubated with a peroxidase blocker, followed by washing and the addition of primary antibody. The primary antibodies used were rabbit polyclonal anti-CARD4 for detection of NOD1, mouse monoclonal anti-CARD15 for detection of NOD2, and mouse monoclonal anti-NLRP3 for detection of NLRP3 expression. Universal negative control for mouse primary antibodies (Dako) was used as negative control for the NOD2 and NLRP3 antibodies, and antibody diluent (Dako) was used to rule out background staining. The slides were exposed to the primary antibodies for 2 hr. Detection of binding was subsequently carried out through the addition of a marked polymer, followed by washing and the addition of substrate chromogen. The slides were dehydrated in 95% ethanol, pure ethanol and xylene. Lastly, the slides were mounted with Pertex (Histolab, Gothenburg, Sweden) and were allowed to dry over night before analysis.

Real-time reverse transcription–polymerase chain reaction

Cellular RNA was extracted using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the supplier’s protocol, including an optional DNAse (Qiagen) treatment. Briefly, cells were lysed, DNA was removed through DNAse digestion, and RNA was eluted in RNAse-free water. Total RNA quantity and quality were assessed by spectrophotometer (260/280 nm). Reverse transcription to cDNA was carried out using Omniscript™ reverse transcriptase kit (Qiagen) with oligo-dT primer in a final volume of 20 μl using the Mastercycle personal polymerase chain reaction (PCR) machine (Eppendorf AG, Hamburg, Germany) at 37° for 1 hr.

Quantitative real-time PCR assays were performed using the SmartCyclerII system (Cepheid, Sunnyvale, CA). For detection of NOD1 and NLRP3, PCR was performed using QuantiTect™ SYBR®Green PCR kit (Qiagen) in a final volume of 25 μl with primers obtained from DNA-Technology (Aarhus, Denmark), designed using Primer express (Applied Biosystems, Foster City, CA) (Table 1). For examining messenger RNA (mRNA) levels of NOD2, Taqman® probes (Applied Biosystems) for NOD2 and β-actin were used in a final volume of 25 μl. The probe sequence for the NOD2 probe has not been made official.

Table 1.   Primer sequence. The primer sequence for nucleotide-binding oligomerization domain 1 (NOD1), NACHT-LRR-PYD-containing protein 3 (NLRP3) and β-actin
NOD1forward5′-GTG GAC AAC TTG CTG AAG AAT GAC-3′
reverse5′-CTG TAC CAG GTC CAG AAT TTT GC-3′
NLRP3forward5′-GCG ATC AAC AGG AGA GAC CTT TA-3′
reverse5′-GCT GTC TTC CTG GCA TAT CACA-3′
β-actinforward5′-CCA ACC GC GAGA AGA TG-3′
reverse5′-ACG GCC AGA GGC GTA CAG-3′

The reactions for NOD1 and NLRP3 were incubated at 95° for 15 min, and then processed for six cycles at 94° for 30 seconds followed by 66° for 60 seconds, with a stepping 2° decrease in temperature until reaching a cycle consisting of 94° for 30 seconds followed by 56° for 60 seconds. Subsequently, 40 cycles comprising 94° for 30 seconds and 55° for 60 seconds were performed. Specific PCR products were analysed by running melt curves. The reaction for NOD2 was incubated at 95° for 10 min, and then processed for 45 cycles at a temperature of 95° for 15 seconds, followed by 60° for 60 seconds.

Gene expression was assessed using the comparative cycle threshold (CT) method.15–17 The relative amount of mRNA was determined by subtracting the CT value for β-actin from the CT value of the samples. The amount of mRNA is expressed in relation to 100 000 mRNA molecules of β-actin (100 000 × 2−ΔCT).

Functional assay of NLRs

The neutrophils were cultured at a density of 4 × 106 cells/ml in complete medium. Cells were treated with 1 or 10 μg/ml NOD1 agonist iE-DAP, negative control iE-Lys, 1 or 10 μg/ml NOD2 agonist MDP, the negative control for MDP, alum in the concentrations of 10 and 100 μg/ml, or with vehicle as untreated cells. The cells were kept for 4 or 20 hr at 37° in a humidified air atmosphere with 5% CO2. Following culture, cells were harvested and used for flow cytometry. Cell culture supernatants were frozen and kept for subsequent analysis of IL-8 or IL-1β levels using IL-8 (sensitivity level 3·5 pg/ml) and IL-1β (sensitivity level 1 pg/ml) enzyme-linked immunosorbent assay (R&D systems, Minneapolis, MN).

Flow cytometry of CD11b and CD62L expression

Cells (1 × 106) were washed in phosphate-buffered saline (PBS; NaCl 0·15 m; KCl 0·0027 m; Na2HPO4 0·01 m; NaH2PO4 0·002 m) and were used for flow cytometry. The cells were suspended in 100 μl PBS, then CD11b-FITC and CD62L-PE were added and the samples were incubated for 20 min at room temperature in the dark. Following incubation, any remaining erythrocytes were lysed by addition of 2 ml High-yield lyse (Invitrogen). A total of 10 000 events were subsequently collected and gated based on forward-scatter and side-scatter properties. Fluorescence measurements were performed on a Coulter Epics XL flow cytometer (Beckman Coulter), and data were analysed using Expo32 ADC analysis software (Beckman Coulter).

Chemotactic assay

A total of 500 000 cells/well were cultured for 4 hr in a 48-well plate in 500 μl RPMI-1640 containing streptomycin and penicillin. Following culture, the cells were harvested, resuspended in fresh medium, and added to a Boyden microchamber cell culture insert (8μm pore size; BD Bioscience, Erembodegem, Belgium) in duplicates. The microchambers were placed in 12-well plates where each well contained 1·5 ml RPMI-1640 and IL-8 (10 ng/ml). The neutrophils were allowed to migrate for 3 hr at 37° in a 5% CO2 atmosphere. The cells that had migrated to the lower wells were collected and counted in a Bürker chamber where six random fields were counted.18 The chemotactic index was subsequently calculated as a ratio of the number of migrated NLR-stimulated neutrophils against the number of migrated untreated neutrophils.19

Statistical analysis

Statistical analysis was performed using the software GraphPad Prism 5 (GraphPad Software, San Diego, CA). All data are expressed as mean ± SEM, and n equals the number of subjects. To determine statistical significance, one-way repeated measurements analysis of variance with Bonferroni selected pairs post-test or Dunnett’s post-test was used. A P-value < 0·05 was considered significant.

Results

Flow cytometry was used to characterize the expression of NOD1, NOD2 and NLRP3. Analysis of the intracellular protein expression compared with isotype control antibody demonstrated a marked intracellular expression of NOD2 protein and a moderate expression of NLRP3. No NOD1 staining was seen (Fig. 1). These findings were confirmed with immunohistochemistry. Strong staining was observed in slides treated with NOD2 antibody (Fig. 2b). An intermediate expression of NLRP3 protein was also found (Fig. 2c). No NOD1 staining was observed (Fig. 2a).

Figure 1.

 Nod-like receptor (NLR) protein expression examined by flow cytometry. Freshly isolated neutrophils were stained for the expression of nucleotide-binding oligomerization domain 1 (NOD1) (a), NOD2 (b) and NACHT-LRR-PYD-containing protein 3 (NLRP3) (c) protein, and were analysed by flow cytometry. The receptor protein expression (dark grey histograms) is compared with isotype controls (transparent histograms) and, where applicable, with the secondary antibody used for conjugation of the NOD1 antibody (i.e. fluorescein isothiocyanate-conjugated goat anti-rabbit; light grey histograms). Expression of NOD2 and NLRP3 protein is clearly demonstrated. The data shown are representative of three independent experiments.

Figure 2.

 Nod-like receptor (NLR) protein expression examined by immunohistochemistry. Neutrophils (∼ 95–96% purity) were embedded in paraffin, sectioned and used for immunohistochemistry with DAB-staining of nucleotide-binding oligomerization domain 1 (NOD1), NOD2 and NACHT-LRR-PYD-containing protein 3 (NLRP3) protein. Brown staining indicates protein expression. No NOD1 protein could be visualized (antibody dilution 1 : 40) (a). Strong staining for NOD2 antibody (antibody dilution 1 : 20) (b), and a slightly weaker but still marked expression of NLRP3 protein at the dilution of 1 : 20 can be seen (c). Antibody diluent (d) and mouse isotype control mouse immunoglobulin G (e) show no staining.

Examination of the mRNA expression for the three NLRs using real-time reverse transcription (RT-) PCR showed no NOD1 mRNA (0·03 ± 0·03 mRNA in relation to 100 000 β-actin; = 4), high levels of NOD2 mRNA (9 285 ± 1 773 mRNA in relation to 100 000 mRNA β-actin; = 4) and moderate levels of NLRP3 mRNA (0·38 ± 0·15 mRNA in relation to 100 000 mRNA β-actin; = 4; Fig. 3). As a result of the different methods of measuring the mRNA (probe versus primer) the expression levels of NOD2 cannot be directly compared to the levels of NOD1 or NLRP3 mRNA.

Figure 3.

 Nod-like receptor (NLR) messenger RNA (mRNA) expression. Neutrophils were isolated from human peripheral blood and used for real-time reverse transcription–polymerase chain reaction investigation of the expression of nucleotide-binding oligomerization domain 1 (NOD1), NOD2 and NACHT-LRR-PYD-containing protein 3 (NLRP3) mRNA. Expression of NOD2 mRNA and NLRP3 mRNA was detected. The mRNA expression is shown in relation to 100 000 mRNA β-actin, and data are expressed as mean ± SEM; = 4.

To investigate NLR function, isolated neutrophils were treated with receptor ligands. The secretion of IL-8 and IL-1β was subsequently analysed. Interleukin-8 is a pro-inflammatory cytokine, secreted by activated neutrophils, whereas IL-1β is known to be released upon NLRP3 stimulation.

Stimulation for 4 hr with the NOD1 agonist iE-DAP did not affect the levels of IL-8 (1 μg/ml: 55·0 ± 21·6 pg/ml; 10 μg/ml: 60·6 ± 24·3 pg/ml; = 6; Fig. 4a) in comparison with negative control (1 μg/ml: 56·3 ± 22·0 pg/ml; 10 μg/ml: 45·4 ± 17·2 pg/ml) and untreated cells (51·3 ± 21·1 pg/ml). Treatment with the NOD2 agonist MDP (1 μg/ml) for 4 hr caused a small but not significant increase in IL-8 compared with negative control and untreated cells (1 μg/ml: 159·6 ± 62·8 pg/ml; negative control: 54·2 ± 18·2 pg/ml; untreated cells: 51·3 ± 21·1 pg/ml; = 6). The use of a 10 times higher concentration of MDP (10 μg/ml) resulted in a marked increase of IL-8 (192·1 ± 73·6 pg/ml) in comparison with negative control (82·3 ± 40·4 pg/ml; = 6; < 0·05; Fig. 4b) and untreated cells (51·3 ± 21·1 pg/ml; = 6; < 0·01). Treatment with alum did not affect the IL-8 secretion, regardless of concentration (10 μg/ml: 57·8 ± 10·5 pg/ml; 100 μg/ml: 70·1 ± 15·9 pg/ml untreated cells: 52·9 ± 11·7 pg/ml; = 6; Fig. 4c).

Figure 4.

 Interleukin-8 (IL-8) secretion by neutrophils stimulated with Nod-like receptor (NLR) ligands. Isolated neutrophils were treated with NLR ligands for 4 hr (a, b, c) or 20 hr (d, e, f). The levels of secreted IL-8 were subsequently measured by enzyme-linked immunsorbent assay. The cells were treated with nucleotide-binding oligomerization domain 1 (NOD1) ligand iE-DAP (a, d) at the concentrations of 1 μg/ml (NOD1 1 μg/ml) or 10 μg/ml (NOD1 10 μg/ml) with negative control iE-Lys (1 μg/ml: NOD1 neg 1 μg/ml; 10 μg/ml: NOD1 neg 10 μg/ml); NOD2 ligand MDP (b, e) at the concentrations of 1 μg/ml (NOD2 1 μg/ml) or 10 μg/ml (NOD2 10 μg/ml), with negative control for MDP(1 μg/ml: NOD2 neg 1 μg/ml; 10 μg/ml: NOD2 neg 10 μg/ml); or NACHT-LRR-PYD-containing protein 3 (NLRP3) ligand alum (c, f) at the concentration of 10 μg/ml (NLRP3 10 μg/ml) or 100 μg/ml (NLRP3 100 μg/ml). Treatment with 10 μg/ml MDP caused secretion of IL-8, which increased further after 20 hr. Data are expressed as mean ± SEM; = 6; *< 0·05, **< 0·01.

Twenty hours of treatment with iE-DAP did not affect the IL-8 secretion in neutrophils, neither compared with the levels at 4 hr (4 hr 1 μg/ml: 42·4 ± 13·8 pg/ml; 4 hr 10 μg/ml: 45·1 ± 11·2 pg/ml; 20 hr 1 μg/ml: 102·5 ± 22·5 pg/ml; 20 hr 10 μg/ml: 101·8 ± 25·7 pg/ml; = 5–6; Fig. 4d), nor in comparison with untreated cells (20 hr untreated cells: 124·1 ± 33·0 pg/ml). The IL-8 secretion observed after 4 hr of treatment with MDP increased even further after 20 hr of treatment, both for 1 μg/ml MDP (4 hr 1 μg/ml: 81·3 ± 23·2 pg/ml; 20 hr 1 μg/ml: 217·0 ± 51·4 pg/ml; = 6; < 0·05 versus MDP 4 hr 1 μg/ml; < 0·01 versus untreated cells 20 hr), and for 10 μg/ml (4 hr: 100·1 ± 28·0 pg/ml; 20 hr: 227·9 ± 60·0 pg/ml; = 6; < 0·05 versus MDP 4 hr 10 μg/ml; < 0·01; Fig. 4e). Twenty hours of treatment with alum did not cause any significant increase in IL-8 (4 hr 10 μg/ml: 57·8 ± 10·5 pg/ml; 4 hr 100 μg/ml: 70·1 ± 16·0 pg/ml; 20 hr 10 μg/ml: 94·9 ± 10·7 pg/ml; 20 hr 100 μg/ml: 179·1 ± 51·4 pg/ml; = 5; Fig. 4f).

Neither treatment with iE-DAP nor MDP gave rise to any secretion of IL-1β, regardless of concentration (all samples were below the level of detection; data not shown). In contrast, treatment with alum (10 μg/ml or 100 μg/ml) gave rise to an increase in the levels of IL-1β after 4 hr (alum 10 μg/ml: 6·8 ± 4·5 pg/ml; = 5; < 0·01 versus untreated cells; alum 100 μg/ml: 8·4 ± 3·2 pg/ml; untreated cells: 2·3 ± 2·2 pg/ml; = 5; < 0·05; Fig. 5) but not after 20 hr (alum 10 μg/ml: 1·9 ± 1·0 pg/ml; alum 100 μg/ml: 2·2 ± 1·1 pg/ml; untreated cells: 2·7 ± 1·3; = 5; data not shown).

Figure 5.

 Interleukin-1β (IL-1β) secretion by neutrophils stimulated with alum. Neutrophils were treated for 4 hr with NACHT-LRR-PYD-containing protein 3 (NLRP3) ligand alum at 10 μg/ml (NLRP3 10 μg/ml), 100 μg/ml (NLRP3 100 μg/ml) or were left untreated (untreated). The levels of IL-1β were subsequently measured. Neutrophils treated with alum caused an increase in IL-1β after 4 hr. Data are expressed as mean ± SEM; = 4; *< 0·05, **< 0·01.

Treatment with the NOD1 ligand did not alter the expression of the cell surface markers CD62L/l-selectin [mean fluorescence intensity (MFI) untreated cells: 22·6 ± 2·4; 1 μg/ml: 21·8 ± 3·0; 10 μg/ml: 21·1 ± 3·2; negative control 1 μg/ml: 23·7 ± 2·9; negative control 10 μg/ml: 23·7 ± 2·2; = 5; data not shown], or CD11b (MFI untreated cells: 23·5 ± 1·6; 1 μg/ml: 26·3 ± 2·3; 10 μg/ml: 27·3 ± 2·4; negative control 1 μg/ml: 25·5 ± 2·5; negative control 10 μg/ml: 26·0 ± 2·1; = 5; data not shown).

In contrast, MDP down-regulated the CD62L expression both at 1 μg/ml (MFI 1 μg/ml: 17·4 ± 2·6) and at 10 μg/ml (MFI 10 μg/ml: 16·3 ± 1·6) in comparison with negative controls (MFI negative control 1 μg/ml: 22·0 ± 2·9; < 0·05; negative control 10 μg/ml: 21·6 ± 2·8; = 5; < 0·01 and untreated cells (MFI 22·6 ± 2·4; < 0·05 and < 0·01 for each concentration, respectively; = 5; Fig. 6a,c). The NOD2 ligand also caused an up-regulation of CD11b compared with untreated cells (MFI untreated cells: 23·5 ± 1·6) at a concentration of 1 μg/ml (MFI 1 μg/ml: 32·6 ± 2·9; negative control 1 μg/ml:26·8 ± 3·1; < 0·01) and at a concentration of 10 μg/ml (MFI 10 μg/ml: 33·8 ± 3·9; negative control 10 μg/ml: 26·8 ± 2·9; = 5; < 0·001 versus untreated cells, < 0·05 versus negative control; Fig. 6b,d).

Figure 6.

 The expression of CD62L (CD62 ligand) and CD11b in neutrophils stimulated with Nod-like receptor (NLR) ligands. Neutrophils were treated for 4 hr with nucleotide-binding oligomerization domain 2 (NOD2) ligand MDP (1 μg/ml: NOD2 1 μg/ml; 10 μg/ml: NOD2 10 μg/ml), negative control for MDP (1 μg/ml: NOD2 neg 1 μg/ml; 10 μg/ml: NOD2 neg 10 μg/ml), or were left untreated (untreated). The expression of CD62L (a) and CD11b (b) was investigated by measuring mean fluorescence intensity (MFI). Data are expressed as mean ± SEM; = 5; *< 0·05, **< 0·01, ***< 0·001. The histograms show the shift in expression of CD62L (c) and CD11b (d) on cells treated with MDP 10 μg/ml (dark grey histograms), negative controls (light grey histograms) or untreated cells (transparent histograms). MDP treatment caused a down-regulation of CD62L, and an up-regulation of CD11b. The histogram data shown are representative of five independent experiments.

Treatment of isolated neutrophils with alum, did not markedly alter CD62L expression (MFI untreated cells: 27·9 ± 3·8; alum 10 μg/ml: 28·02 ± 3·7; alum 100 μg/ml: 28·1 ± 3·7; = 5; data not shown). Neither did alum treatment affect the expression of CD11b (MFI untreated cells 18·7 ± 3·2; alum 10 μg/ml: 19·9 ± 3·3; alum 100 μg/ml: 19·3 ± 3·4; = 5; data not shown).

The altered expression of CD62L and CD11b found in isolated neutrophils by Nod2 activation prompted us to investigate if NLR stimulation also affected the migratory properties of the cells. Treatment with 10 μg/ml MDP facilitated neutrophil migration induced by IL-8, both in comparison with untreated cells (untreated cells: 1·0 ± 0·0; 10 μg/ml: 2·1 ± 0·3; = 4; < 0·05) and negative control (negative control 10 μg/ml: 0·9 ± 0·2; = 4; < 0·01). A similar result was obtained after alum treatment (alum 100 μg/ml: 1·7 ± 0·2; = 4; < 0·01). Treatment with NOD1 ligand did not affect the chemotactic ability (10 μg/ml: 1·3 ± 0·2; negative control 10 μg/ml: 1·0 ± 0·1; = 4; Fig. 7).

Figure 7.

 Migratory responses by neutrophils stimulated with Nod-like receptor (NLR) ligands. Isolated neutrophils were treated for 4 hr with nucleotide-binding oligomerization domain 1 (NOD1) ligand (10 μg/ml; NOD1 10 μg/ml), NOD2 ligand (10 μg/ml; NOD2 10 μg/ml), with NACHT-LRR-PYD-containing protein 3 (NLRP3) ligand (100 μg/ml; NLRP3 100 μg/ml) or were left untreated (untreated). Following treatment, the cells were allowed to migrate for 3 hr towards interleukin-8 (IL-8; 10 ng/ml). Subsequently, the migrated cells were collected, and counted in a Bürker chamber. An increased migration following treatment with NOD2 and NLRP3 ligand was observed. Data are expressed as mean ± SEM; = 4; *< 0·05.

Discussion

The expression of NOD2 and NLRP3 protein was demonstrated in neutrophils using flow cytometry and immunohistochemistry. Corresponding mRNA was observed using real-time RT-PCR. Neither protein nor mRNA was detected for NOD1. The functional responses of NLR stimulation were investigated by treating isolated neutrophils with NOD1 ligand iE-DAP, NOD2 ligand MDP, or NLRP3 activator alum. The secretion of IL-8 and IL-1β, and the expression of the cell surface markers CD62L and CD11b were used as markers of neutrophil activation.1,20 In accordance with the protein expression data, treatment of isolated neutrophils with NOD1 ligand did not alter the secretion of IL-8 or IL-1β, nor the expression of CD11b or CD62L. NOD2 activation induced a time-dependent release of IL-8, an up-regulation of CD11b, and a corresponding down-regulation of CD62L. A chemotaxis assay was employed to study migration, and neutrophils treated with NOD2 or NLRP3 ligand displayed an increased migratory response to chemotactic stimuli.

Functional sets of NLRs have been demonstrated in a variety of cell types, including dendritic cells and monocytes.21 It has been reported that NLR activation may cause an inflammatory response which attracts neutrophils to the site of inflammation,22 but the presence of NLRs in neutrophils and the neutrophil responses upon their activation have not previously been demonstrated. In epithelial and antigen-presenting cells, NOD1 and NOD2 stimulation has been shown to cause cellular activation followed by release of cytokines, such as IL-8 and tumour necrosis factor-α.23 In analogy, we here demonstrate that NOD2 and NLRP3 stimulation can provoke a cytokine response in neutrophils. The altered expression of the adhesion molecules CD11b and CD62L, in combination with a facilitation of migratory properties, provide further evidence of neutrophil activation. The presently observed alum-induced secretion of IL-1β is well in line with previous findings in other cell types.9 The release of specific cytokines in response to stimulation of a particular NLR is believed to reflect the alternative signalling pathways used by the receptors. While NOD1 and NOD2 effects are mediated through nuclear factor-κB and the mitogen-activated protein kinase pathways, NLRP3 makes use of caspase-1.10

The responses of the NLRs are in many aspects similar to those of the TLRs. Both families are PRRs and recognize microbial patterns. Several TLRs have been described in neutrophils24,25 and their activation gives rise to effects similar to those we observed with NLR ligands, i.e. cytokine secretion, changed expression of the adhesion molecules CD62L and CD11b, and altered migration.24,26 The TLRs and NLRs exhibit different cellular localizations: NLRs are present in the cytosol, whereas the TLRs are found on the cell surface or endosomally.27,28 This difference in distribution may allow for PRR-mediated detection of bacteria regardless of bacterial compartmentalization.

It has been demonstrated that NLRs supplement TLRs when the TLR signalling is diminished.29 A synergistic effect between TLRs and NLRs has also been suggested. Hence, it appears that TLRs and NLRs might interact to potentiate the immune response.30–32 The detection of NLRs in neutrophils therefore suggests that these cells might have the ability to detect a wider range of microbes than they would by the use of TLRs alone. Furthermore, the possibility of TLR and NLR interactions also in neutrophils should not be ruled out.

As the neutrophils make up the first line of immune defence against foreign pathogens, they are likely to be exposed to microbial components.33 The immediate role of neutrophils in the host defence against infection and their early encounter with microbe antigen make these cells and their PRRs interesting targets for therapeutic manipulation. The present findings of functional NOD2 and NLRP3 in neutrophils might reflect a previously unknown pathway for activation of these cells.

Acknowledgements

The authors would like to thank Anna Karin Bastos, Josefine P Riikonen, Ann Reutherborg and Eva Thylander for help with the collection of samples, and Ingegerd Larsson for help with the real-time RT-PCR assay. This work was supported by grants from the Swedish Medical Research Council and the Swedish Heart and Lung Foundation.

Disclosures

There is no conflict of interest.

Ancillary