Toll-like receptor TLR2 and TLR9 ligation triggers neutrophil activation ...

RHEUMATOLOGY

Rheumatology 2013;52:1183–1189 doi:10.1093/rheumatology/kes415 Advance Access publication 12 February 2013

Original article Toll-like receptor TLR2 and TLR9 ligation triggers neutrophil activation in granulomatosis with polyangiitis Julia U. Holle1, Mirca Windmo¨ller1, Christina Lange1, Wolfgang L. Gross1, Karen Herlyn1 and Elena Csernok1 Abstract Objective. The aim of the study was to characterize the expression of TLR2, TLR4 and TLR9 in PMNs of patients with granulomatosis with polyangiitis (GPA) and to elucidate the role of these receptors in GPA with respect to neutrophil activation. Methods. The expression of TLR2, TLR4 and TLR9 was determined on ex vivo PMNs in whole blood samples of GPA patients (n = 35) and healthy controls (HCs) (n = 24). Isolated PMNs were stimulated in vitro with TLR agonists and assessed for degranulation, membrane proteinase 3 (mPR3) expression, soluble L-selectin shedding and cytokine production (IL-8) in five GPA patients and five HCs. The priming effects of TLR2 and TLR9 ligation were assessed by measurement of serine protease activity after stimulation with PR3-ANCA.

Conclusion. Expression of TLR2, TLR4 and TLR9 in PMNs and the TLR-induced activation of PMNs was comparable in GPA and HC. mPR3 upregulation by TLR2 and TLR9 stimulation and the priming effect of TLR ligands on PMNs may have a potential implication for triggering disease activity during infection in GPA. Key words: granulomatosis with polyangiitis, ANCA-associated vasculitis, innate immunity, PMN, Toll-like receptor.

Introduction Granulomatosis with polyangiitis (GPA) is an autoimmune disease of unknown aetiology that is characterized by granuloma formation and ANCA-associated small vessel vasculitis [1, 2]. PMNs are thought to play a key role in the 1 Department of Rheumatology and Immunology, University Hospital Schleswig-Holstein, Campus Lu¨beck and Klinikum Bad Bramstedt, Bad Bramstedt, Germany.

Submitted 9 March 2012; revised version accepted 7 December 2012. Correspondence to: Julia U. Holle, Department of Rheumatology and Immunology, University Hospital Schleswig-Holstein, Campus Luebeck and Klinikum Bad Bramstedt, Oskar-Alexander-Straße 26, 24576 Bad Bramstedt, Germany. E-mail: [email protected]

mediation of small vessel vasculitis as they undergo a respiratory burst, degranulation, adhere to and migrate through endothelial cell layers upon interaction with ANCA [3, 4]. Importantly, the interaction of ANCA with PMNs occurs via engagement of ANCA with its target antigen, proteinase 3 (PR3), and Fc-y receptors on the plasma membrane of PMNs [5]. The level of membrane expression of PR3 (mPR3) correlates with disease activity and relapse in generalized GPA [6–8]. Furthermore, infections trigger local disease activity in the ear, nose and throat (ENT) tract and relapses that have been linked to Staphylococcus aureus infections and staphylococcal toxic shock syndrome toxin-1 [9–11]. Antimicrobial therapy with trimethoprim/sulfomethoxazole is efficient in

! The Author 2013. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: [email protected]

BASIC SCIENCE

Results. There were no significant differences in the ex vivo expression of TLRs on PMNs in HCs and GPA patients. Stimulation of TLR4 and TLR9 induced MPO release, stimulation with TLR2, TLR4 and TLR9 ligands elicited IL-8 production and stimulation of TLR2 and TLR9 led to an upregulation in mPR3 expression on PMNs with no significant differences between GPA and HC after 1 or 24 h stimulation. Priming of PMNs with TLR2 and TLR9 ligands induced degranulation after subsequent stimulation with PR3-ANCA, which was comparable to priming with TNF-a.

Julia U. Holle et al.

reducing relapse rates in GPA [12, 13], which further supports the role of infectious agents in triggering disease activity. In antimicrobial host defence, Toll-like receptors (TLRs) are crucial in the sensing of invading pathogens and in the initiation of first defence mechanisms. Apart from TLR3, all TLRs are expressed on PMNs. TLR4 is regarded as the major LPS receptor, whereas PMN activation by components of gram-positive bacteria such as bacterial lipopetides, lipoteichoic acids and peptidoglycans are mediated via TLR2 (in combination with TLR1 and TLR6) and by bacterial DNA motifs (CpG motifs) via TLR9. Both TLR2 and TLR4 are increased in septic states and TLR2 is upregulated after exposure to various bacterial stimuli [14–16]. In particular, TLR2- and TLR4-mediated effects have been studied in PMNs [17], which are implicated in regulation of neutrophil activation, migration and apoptosis [16]. Whereas the effect of TLR activation on mPR3 expression is unknown, TLR2 and TLR4 activation is known to induce ROS production, L-selectin shedding, IL-8 release and the inhibition of neutrophil apoptosis [17, 18]. In AAV, only one study has addressed the expression of TLRs in peripheral blood leukocytes and found an increased expression of TLRs by monocytes and NK cells, but not by granulocytes [19]. Furthermore, no difference between nasal S. aureus carriers and non-carriers was found [19]. Regarding peripheral blood cell activation upon TLR stimulation, only cytokine production has been assessed after stimulation of whole blood cultures with TLR ligands showing no differences between GPA and HC [19]. The aim of the study was to determine ex vivo TLR expression in PMNs in a large GPA cohort and healthy individuals and to elucidate the role of TLRs on neutrophil activation with respect to degranulation (assessed by MPO and serine protease release), membrane proteinase 3 (mPR3) expression, L-selectin shedding and cytokine release (IL-8) after in vitro stimulation with TLR agonists.

Material and methods Patients All GPA patients had to fulfil the definition according to the Chapel Hill conference and the criteria of the ACR [1, 2] to be selected for the study. Disease stages were applied according to the current European League Against Rheumatism (EULAR) definitions [20]. Disease activity was assessed using the BVAS, version 3 [21]. All patients gave written consent before taking part. Ethics approval was obtained from the University of Lu¨beck (ethics approval for subproject 2 of Clinical Research Unit KFO 170, No. 07-035). For assessing TLR expression, 35 GPA patients and 24 healthy individuals were studied as controls (HCs). For stimulation experiments, five GPA patients and five HC were analysed. Patients and HCs had no clinical evidence of infection at the time of testing. Furthermore, eight GPA patients with infections were assessed for TLR expression and compared with eight patients without infections and controls. The study was carried out in accordance with the Declaration of Helsinki.

1184

Preparation of whole blood and neutrophils Five millilitres of EDTA-anticoagulated whole blood was added to 45 ml of PBS and centrifuged at 233 g for 10 min at 20 C. The cell pellet was resuspended in 50 ml of PBS. This was repeated twice. The cell pellet was then adjusted to 5 ml in PBS and ready for monoclonal antibody staining. PMNs were isolated as described earlier [22]. For analysis of TLR and mPR3 expression levels in PMNs, whole blood samples were used.

Staining of neutrophils and flow cytometry analysis One hundred microlitres of washed whole blood and isolated neutrophils at a concentration of 2  106/ml were stained with PE-conjugated mouse monoclonal antibody IgG2a antibody against TLR2 and TLR4 or with an irrelevant control antibody (Coulter Immunotech, Krefeld, Germany) for 30 min at 4 C according to the manufacturer’s protocol. For intracellular staining of TLR9, cells were first stained with a FITC-conjugated anti-CD66 monoclonal antibody, fixed and permeabilized and then incubated with a PE-conjugated rat monoclonal IgG2a antibody against TLR9. Stained cells were analysed by FACSCalibur followed by analysis using CellQuest software (Becton Dickinson). The median fluorescence intensity (MFI) and MFI ratio of control and specific antibody MFI values were recorded for each sample. A minimum of 10 000 events were collected per sample and analysed using a flow cytometer.

Stimulation of neutrophils by TLR ligands Isolated PMNs (2  106) of five GPA patients and five HCs were stimulated with TLR ligands of TLR2 (PAM3CSK4, 100 ng/ml; Invivo Gen, San Diego, CA, USA), TLR4 (LPS, 100 ng/ml; Sigma, Taufkirchen, Germany) and TLR9 (CpG DNA, 50 mg/ml; Invivo Gen, San Diego, CA, USA) for 1 and 24 h, respectively. For the detection of TLR2 and TLR9, PMNs were pretreated with GM-CSF, which is required as an additional stimulant to induce TLR2 and TLR9 (50 ng/ml; Immunotools, Friesoythe, Germany) for 1.5 h, which are otherwise not detectable [23]. TLR ligands were used at concentrations described earlier to stimulate PMNs [23]. The viability of PMNs after 24 h was assessed using trypan blue staining. More than 60% of cells were viable after 24 h.

Analysis of mPR3 expression PMNs were assessed for mPR3 expression by flow cytometry before and after stimulation with TLR ligands for 1 h, as described earlier [22]. Briefly, cells were stained with 2 mg/ml of FITC-conjugated anti-PR3 murine monoclonal antibody (WGM2) or with an irrelevant FITC-conjugated IgG1 control antibody (Coulter Immunotech, Krefeld, Germany) according to the manufacturer’s protocol.

Degranulation assay to assess a possible priming effect of TLR ligands For each assay, 2  106 purified neutrophils were primed with TNF-a (2 ng/ml; Cell Systems, Berlin, Germany) and TLR2 and nine ligands (CpG DNA, 50 mg/ml and

www.rheumatology.oxfordjournals.org

TLR-induced activation in PMNs of GPA patients

lipoprotein PAM3CSK4, 100 ng/ml) for 60 min at 37 C, followed by stimulation with monoclonal antibodies anti-PR3 clone WGM2 33 mg/ml for 60 min at 37 C. Serine protease (elastase, proteinase 3 and cathepsin G) activity was assessed by the cleavage of MeO-Suc-Ala-ProVal-pNA (Sigma, Taufkirchen, Germany) compared with a standard curve. The effects of PR3-ANCA were performed on three donors.

Results Patients Thirty-five GPA patients (median age 61.5 years, range 21–80 years, 23 males and 12 females) and 24 HCs (median age 57.5 years, range 29–81 years, 13 males and 11 females) were analysed for TLR expression in unstimulated PMNs. Of the GPA patients, 9 had inactive disease (BVAS = 0), 15 had active disease (BVAS = 1–9, median 5) and 10 were highly active (BVAS = >9, 14). For stimulation experiments, five GPA (median age 51 years, range 20–59 years, 2 males and 3 females) and five HCs (median age 43 years, range 24–58 years, 3 males and 2 females) were analysed. The median BVAS of the GPA patients analysed in stimulation experiments was 4 (range 0–10).

Analysis of MPO release, IL-8 secretion and L-selectin shedding The supernatants of the stimulated PMNs were tested for the release of MPO (MPO Enzyme Immunoassay Test Kit, BioCheck), production of IL-8 (Quantikine Human CXCL8/ IL-8 ELISA, R&D Systems) and L-selectin shedding (Human sL-selectin/CD62L ELISA, R&D Systems) according to the manufacturer’s protocols.

Ex vivo TLR expression in PMNs Statistical analysis

There were no significant differences in the membrane expression of TLR2, TLR4 and intracellular expression of TLR9 between all GPA patients and HCs (P > 0.05, data not shown). When active GPA (BVAS 1–9) and highly

Statistics were performed using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA, USA) using one-tailed Mann–Whitney U test (Fig. 1, Tables 1 and 2).

FIG. 1 Membrane expression of proteinase 3 on PMNs.

PMNs stained with FITC-conjugated anti-PR3 murine monoclonal antibody (WGM2) were analysed by flow cytometry after priming with GM-CSF (1.5 h) and stimulation with the ligand of TLR9 (1 h). Representative histograms of a monomodal (A) and a bimodal (B) distribution of proteinase 3 surface expression.

TABLE 1 Expression of mPR3 in unstimulated PMNs and after stimulation with TLR2, TLR4 and TLR9 ligand mPR3 expression after 1 h [MFI ratio, mean (S.D.)] Stimulant

HC

GPA

P (HC vs GPA)

Unstimulated TLR2 ligand, P (unstimulated vs stimulated)

1.7 (0.2) 2.8 (0.4) 0.008 2.3 (0.9) 0.3 3.8 (1.2) 0.008

2.0 (0.6) 4.4 (1.8) 0.032 2.8 (0.9) 0.17 4.2 (1.6) 0.008

0.55 0.22 – 0.55 – 0.22 –

TLR4 ligand, P (unstimulated vs stimulated) TLR9 ligand, P (unstimulated vs stimulated)

www.rheumatology.oxfordjournals.org

1185

1186

0.55 0.31 – 1.0 – 0.55 – 6.1 (2.8) 0.1 (1.7) 0.008 5.6 (2.7) 0.69 0.1 (1.5) 0.008 0.69 0.15 – 0.15 – 0.31 – TLR9 ligand, P (unstimulated vs stimulated)

TLR4 ligand, P (unstimulated vs stimulated)

51.2 (25.1) 66.8 (28.8) 0.55 59.0 (22.2) 0.42 131.2 (34.9) 0.008 Unstimulated TLR2 ligand, P (unstimulated vs stimulated)

48.8 (14.3) 73.2 (37.2) 0.55 60.0 (13.5) 0.42 154.8 (65.8) 0.008

1.0 1.0 – 1.0 – 0.42 –

33.2 (71.5) 236.8 (104.9) 0.016 204.8 (84.4) 0.016 404.9 (114.8) 0.008

61.2 (42.1) 362.9 (34.3) 0.008 131.6 (61.8) 0.016 339.8 (159.2) 0.016

4.4 (3.0) 0.5 (1.1) 0.008 7.8 (3.8) 0.84 0.4 (0.4) 0.008

P (HC vs GPA) GPA HC P (HC vs GPA) GPA HC Stimulant

GPA

P (HC vs GPA)

HC

IL-8 secretion after 1 h [pg/ml, mean (S.D.)] MPO release after 1 h [ng/ml, mean (S.D.)]

TABLE 2 MPO release, IL-8 secretion and L-selectin shedding after incubation of PMNs with TLR ligands

L-Selectin

shedding after 24 h [ng/ml, mean (S.D.)]

Julia U. Holle et al.

active GPA (BVAS > 9) were compared with controls, there was no significant difference in the expression pattern of TLR2, TLR4 and TLR9 (P > 0.05, data not shown). Eight GPA patients with infections [six males and two females, median age 66 years (range 27–78 years), median BVAS 0.5 (range 0–99)] were assessed for TLR expression and compared with eight age- and gender-matched GPA patients with comparable disease activity but without infection [six males and two females, median age 64 years (range 40–80 years), median BVAS 0.5 (range 0–10)] and controls without evidence of infection [seven males and one female, median age 67 (range 46–73 years)]. There was a trend towards increased TLR2 and TLR4 expression of GPA patients with infection compared with patients with no infection and controls, but this was not significant (data not shown). Intracellular TLR9 expression was significantly higher in GPA patients with infection (TLR9 MFI ratio 1.22 ± 1.22) compared with GPA patients without infection (TLR9 MFI ratio 0.91 ± 0.07, P = 0.036) and controls (TLR9 MFI ratio 0.97 ± 0.27, P = 0.093). There was no difference in mPR3 expression in unstimulated ex vivo PMNs of HCs and GPA patients (Table 1).

MPO release, IL-8 production, L-selectin shedding and mPR3 expression upon PMN stimulation with TLR ligands Stimulation of TLR9 induced MPO release, and stimulation with TLR2, TLR4 and TLR9 ligands elicited IL-8 production. L-selectin shedding was downregulated after stimulation of PMNs with TLR2 and TLR9 ligand (Table 2). Furthermore, stimulation of TLR2 and TLR9 upregulated mPR3 expression (Table 1 and Fig. 1). There were no significant differences in MPO release, IL-8 production, L-selectin shedding and mPR3 expression of unstimulated PMNs in HCs versus GPA patients and stimulated PMNs in HCs versus GPA patients (Tables 1 and 2). As GM-CSF was used in combination with TLR ligand stimulation, we assessed the effect of GM-CSF on mPR3 expression in the absence and presence of TLR9 ligand. Representative FACS plots show that GM-CSF induces mPR3 expression, which is further enhanced by TLR9 ligand (Fig. 1). Table 2 displays the results for MPO release, IL-8 secretion and L-selectin shedding. For MPO release and IL-8 secretion, data are presented after 1 h incubation with TLR ligands. Incubation for 24 h resulted in greater MPO release and IL-8 secretion with no differences between HCs and GPA patients (data not shown). With respect to L-selectin shedding, data on the 24 h stimulation are displayed. Incubation with TLR ligands for 1 h only resulted in no significant shedding (data not shown).

Priming experiments to assess priming effects of TLR ligation In order to assess whether TLR2 and TLR9 ligands can act as a priming agent, degranulation (serine protease activity) of PMNs of three healthy donors was assessed after incubation with TLR2 ligand, TLR9 ligand or TNF-a and subsequent treatment with PR3-ANCA. Pretreatment

www.rheumatology.oxfordjournals.org

TLR-induced activation in PMNs of GPA patients

with TLR ligands and subsequent stimulation with PR3-ANCA led to a significant increase in serine protease activity compared with PMNs that were pretreated with TLR ligands but not stimulated with PR3-ANCA. This priming effect was comparable to priming with TNF-a (Fig. 2).

FIG. 2 TLR agonist-mediated effects on degranulation.

Serine protease activity of PMNs of healthy donors (n = 3) after stimulation with ligands of TLR2 and TLR9 for 1 h and a subsequent stimulation with PR3-ANCA for 1 h. Values shown are means ± S.D. *P = 0.05.

Discussion GPA activity is known to be triggered by infections, i.e. activity in the ENT tract by local S. aureus infection. The aim of this study was to assess basic antimicrobial host defence mechanisms in GPA with regard to TLR activation. We found that in individuals without evidence of infection, the ex vivo expression of extracellular TLR2, TLR4 and intracellular TLR9 are comparable in PMNs of GPA and controls. Furthermore, there were no differences in expression in active GPA compared with inactive GPA and controls. These findings confirm data from Tadema et al. [19]. In contrast, in individuals with GPA and clinical evidence of infection, we found an increased expression of TLR9 compared with individuals with GPA and no infection. We found a trend towards increased TLR2 and TLR4 expression in infections, which was not significant, probably due to the small number of patients we assessed with infections or a consequence of rapid downregulation of TLR4. The findings are in line with previous studies demonstrating an increased expression of TLRs in infections (i.e. sepsis) [15, 18]. Furthermore, the results suggest that TLR upregulation is regulated in a complex manner and associated with an upregulation in the context of infection—as expected—but not in the context of disease activity in GPA.

FIG. 3 Role of TLR activation in GPA.

TLRs become upregulated during infection. Stimulation of TLR2 and TLR9 enhances upregulation membrane PR3 (mPR3) expression on PMNs. TLR2 and TLR9 ligands function as priming agents and include degranulation upon stimulation with PR3-ANCA.

www.rheumatology.oxfordjournals.org

1187

Julia U. Holle et al.

Moreover, we confirmed previous reports [17, 19] that stimulation of TLR2 and TLR4 induces MPO release and IL-8 production and demonstrated that the same is valid for TLR9 stimulation. We found no difference between GPA and controls, which implies that the regulation of these innate immune functions is not differentially regulated in controls and GPA. We cannot rule out that GM-CSF pretreatment that was done in stimulation experiments with TLR2 and TLR9 ligands may have had an influence on our results with respect to increased MPO release, IL-8 secretion, L-selectin shedding and mPR3 expression, yet pretreatment with GM-CSF is usually performed to allow TLR2 and TLR9 detection [16, 24]. We provide evidence that TLR2 and TLR9 ligands act as priming agents in a similar fashion as TNF-a to induce degranulation of PMNs. This underlines a possible role of infectious agents triggering ANCA-mediated PMN activation to induce disease activity in AAV. Furthermore, we demonstrate that stimulation of TLR2 and TLR9 enhances upregulation of mPR3 expression of PMNs. Even if this occurred in controls to the same extent as in GPA, this finding provides an explanation for the triggering of GPA activity by infections. First, the presence of mPR3 on the plasma membrane is the prerequisite for interaction with PR3-ANCA to further activate PMNs, and evidence suggests that ANCA-induced neutrophil activation is stronger (with respect to superoxide burst and degranulation) in mPR3high neutrophils compared with mPR3low neutrophils [25]. Second, an upregulation of PR3 on the plasma membrane of PMNs may provide an increased amount of antigen to antigen-presenting cells (i.e. dendritic cells) to facilitate an initialization of an adaptive immune response (Fig. 3). Intriguingly, TLR2 and TLR9 have recently been shown to have a T cell polarizing function in a mouse model of ANCA-associated vasculitis [26]. Summers et al. [26] found that glomerulonephritis developed only after co-administration of MPO with either of the TLR ligands. Our findings together with the results of Summers et al. support the evidence that infections may be linked via TLR ligation to autoimmunity against PR3- and MPO-ANCA. In contrast to previous studies [17] showing increased shedding of L-selectin after stimulation of TLR2 and TLR4, the shedding of L-selectin was downregulated by stimulation of TLR2 and TLR9 in both GPA patients and controls in our study. L-Selectin mediates the initial interaction (rolling) of PMNs with endothelial cells and serves as a marker of PMN activation. The observed differences may be explained by methodological differences, as in previous studies, and shedding was assessed by measuring the expression of L-selectin on PMNs by flow cytometry, whereas in our study L-selectin was detected in the supernatant of PMNs by ELISA. To summarize, there were no significant differences in the expression of TLR2, TLR4 and TLR9 and in the TLR-induced activation of circulating PMNs between controls and GPA patients as assessed by degranulation, IL-8 secretion, L-selectin shedding and mPR3 expression. The findings that (1) mPR3 expression is enhanced by TLR2

1188

and TLR 9 stimulation and (2) TLR2 and TLR9 ligands act as priming agents may be of particular relevance in GPA, as this may play a role in triggering disease activity in the presence of infection. Rheumatology key messages TLR activation enhances upregulation of mPR3 expression on PMNs in healthy controls and GPA patients. . In AAV, some TLR ligands may act as priming agents to induce ANCA-mediated neutrophil activation. .

Acknowledgements We thank Monika Backes and Evelyn Grage-Griebenow for technical assistance. Funding: This work was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG) to E.C. and J.U.H (subproject 2 within Clinical Research Unit KFO170). Disclosure statement: The authors have declared no conflicts of interest.

References 1 Fries JF, Hunder GG, Bloch DA et al. The American College of Rheumatology 1990 criteria for the classification of vasculitis. Summary. Arthritis Rheum 1990;33:1135–6. 2 Jennette JC, Falk RJ, Andrassy K et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187–92. 3 Falk RJ, Terrell RS, Charles LA et al. Anti-neutrophil cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Natl Acad Sci USA 1990;87:4115–9. 4 Calderwood JM, Williams JM, Morgan MD et al. ANCA induces beta2 integrin and CXC chemokine-dependent neutrophil-endothelial cell interactions that mimic those of highly cytokine-activated endothelium. J Leukoc Biol 2005;77:33–43. 5 Porges AJ, Redecha PB, Kimberly WT et al. Anti-neutrophil cytoplasmic antibodies engage and activate human neutrophils via Fc gamma RIIa. J Immunol 1994;153:1271–80. 6 Witko-Sarsat V, Lesavre P, Lopez S et al. A large subset of neutrophils expressing membrane proteinase 3 is a risk factor for vasculitis and rheumatoid arthritis. J Am Soc Nephrol 1999;10:1224–33. 7 Muller Kobold AC, Kallenberg CGM et al. Leukocyte membrane expression of proteinase 3 correlates with disease activity in patients with Wegener’s granulomatosis. Br J Rheumatol 1998;37:901–7. 8 Rarok AA, Stegeman CA, Limburg PC et al. Neutrophil membrane expression of proteinase 3 (PR3) is related to relapse in PR3-ANCA-associated vasculitis. J Am Soc Nephrol 2002;13:2232–8.

www.rheumatology.oxfordjournals.org

TLR-induced activation in PMNs of GPA patients

9 Stegeman CA, Tervaert JW, Sluiter WJ et al. Association of chronic nasal carriage of Staphylococcus aureus and higher relapse rates in Wegener granulomatosis. Ann Intern Med 1994;120:12–7.

18 Sabroe I, Dower SK, Whyte MKB. The role of Toll-like receptors in the regulation of neutrophil migration, activation and apoptosis. Clin Infect Dis 2005;41: S421–6.

10 Popa ER, Stegeman CA, Abdulahad WH et al. Staphylococcal toxic-shock-syndrome-toxin-1 as risk factor for disease relapse in Wegener’s granulomatosis. Rheumatology 2007;46:1029–33.

19 Tadema H, Abdulahad WH, Stegeman CA et al. Increased expression of Toll-like receptors by monocytes and natural killer cells in ANCA-associated vasculitis. PLoS One 2011;6:e24315.

11 Laudien M, Gadola SD, Podschun R et al. Nasal carriage of Staphylococcus aureus and endonasal activity in Wegener’s granulomatosis as compared to rheumatoid arthritis and chronic rhinosinusitis with nasal polyps. Clin Exp Rheumatol 2010;28(1 suppl 57):51–5.

20 Hellmich B, Flossmann O, Gross WL et al. EULAR recommendations for conducting clinical studies and/or clinical trials in systemic vasculitis: focus on anti-neutrophil cytoplasm antibody-associated vasculitis. Ann Rheum Dis 2007;66:605–17.

12 Stegeman CA, Tervaert JW, de Jong PE et al. Trimethoprimsulfamethoxazole (co-trimoxazole) for the prevention of relapses of Wegener’s granulomatosis. Dutch Co-Trimoxazole Wegener Study Group. N Engl J Med 1996;335:16–20.

21 Mukhtyar C, Lee R, Brown D et al. Modification and validation of the Birmingham Vasculitis Activity Score (version 3). Ann Rheum Dis 2009;68:1827–32.

13 Reinhold-Keller E, De Groot K, Rudert H et al. Response to trimethoprim/sulfamethoxazole in Wegener’s granulomatosis depends on the phase of disease. QJM 1996;89:15–23. 14 Kobayashi SD, Braughton KR, Whitney AR et al. Bacterial pathogens modulate an apoptosis differentiation program in human neutrophils. Proc Natl Acad Sci USA 2003;100: 1281–6. 15 Ha¨rter L, Mica L, Stocker R et al. Increased expression of Toll-like receptor-2 and -4 on leukocytes from patients with sepsis. Shock 2004;22:403–9.

22 Csernok E, Ernst M, Schmitt W et al. Activated neutrophils express proteinase 3 on their plasma membrane in vitro and in vivo. Clin Exp Immunol 1994;95:244–50. 23 Hayashi F, Means TK, Luser AD. Toll-like receptors stimulate human neutrophil function. Blood 2003;201: 2660–9. 24 Hellmich B, Csernok E, Trabandt A et al. Granulocyte-macrophage colony-stimulating factor (GM-CSF) but not granulocyte colony-stimulating factor (G-CSF) induces plasma membrane expression of proteinase 3 (PR3) on neutrophils in vitro. Clin Exp Immunol 2000;120:392–8.

16 Hayashi F, Means TK, Luster AD. Toll-like receptors stimulate human neutrophil function. Blood 2003;102:2660–9.

25 Schreiber A, Luft FC, Kettritz R. Membrane proteinase 3 expression and ANCA-induced neutrophil activation. Kidney Int 2004;65:2172–83.

17 Sabroe I, Prince LR, Jones EC et al. Selective roles for Toll-like receptor (TLR) 2 and TLR4 in the regulation of neutrophil activation and life span. J Immunol 2003;170: 5268–75.

26 Summers SA, Steinmetz OM, Gan PY et al. TLR2 induces TH17 myeloperoxidas autoimmunity while Toll-like receptor 9 drives Th1 autoimmunity. Arthritis Rheum 2011; 63:1124–35.

www.rheumatology.oxfordjournals.org

1189