The role of NOX2-derived reactive oxygen species in collagenase-induced osteoarthritis

Open ArchivePublished:September 05, 2018DOI:https://doi.org/10.1016/j.joca.2018.08.014

      Summary

      Objective

      Synovitis in collagenase-induced osteoarthritis (CiOA) is driven by locally released S100A8/A9 proteins and enhances joint destruction. S100A8/A9 can induce reactive oxygen species (ROS) release by phagocytes in OA synovium via neutrophil cytosolic factor-1 (Ncf1)-regulated NOX2 activation. In the present study we investigated whether NOX2-derived ROS affect joint pathology during CiOA.

      Methods

      CiOA was induced in knee joints of wild type (WT) and Ncf1-deficient (Ncf1**) mice. Synovial gene expression of NOX2-subunits was measured with quantitative real-time polymerase chain reaction (qRT-PCR). Joint pathology was assessed using histology and immunohistochemistry for aggrecan neo-epitope VDIPEN. Levels of inflammatory proteins were measured with Luminex or ELISA. Phagocytes in synovium, blood, bone marrow (BM) and spleen were analyzed with flow cytometry. ROS release by phagocytes was measured with a ROS detection kit.

      Results

      CiOA induction in knee joints of WT mice caused significantly increased synovial gene expression of NOX2 subunits. On day 7 of CiOA, cartilage damage and MMP activity, as measured by VDIPEN, were comparable between WT and Ncf1** mice. Synovial thickening, synovial S100A8/A9 levels and percentages of synovial macrophages, polymorphonuclear cells (PMNs), and monocytes were not different, as were levels of inflammatory mediators in serum and phagocyte percentages in blood, BM and spleen. On day 42 of CiOA, synovitis, cartilage damage, and osteophyte formation in Ncf1** mice were unaltered when compared to WT mice. ROS detection confirmed that Ncf1** PMNs lack functional NOX2, but in vitro macrophages showed ROS production, suggesting activation of compensatory mechanisms.

      Conclusions

      Absence of Ncf1-mediated ROS production does not alter joint pathology in CiOA.

      Keywords

      Introduction

      Osteoarthritis (OA) is a highly prevalent and disabling joint disease, characterized by cartilage damage, ectopic bone formation and fibrosis. Synovitis is thought to aggravate OA pathology, and the majority of OA patients show at least low grade synovial inflammation
      • Guermazi A.
      • Hayashi D.
      • Roemer F.W.
      • Zhu Y.
      • Niu J.
      • Crema M.D.
      • et al.
      Synovitis in knee osteoarthritis assessed by contrast-enhanced magnetic resonance imaging (MRI) is associated with radiographic tibiofemoral osteoarthritis and MRI-detected widespread cartilage damage: the MOST study.
      . This is reflected by hyperplasia and activation of resident macrophages, resulting in the release of inflammatory mediators such as IL-1β, TNFα, S100A8/A9, and KC
      • Rahmati M.
      • Mobasheri A.
      • Mozafari M.
      Inflammatory mediators in osteoarthritis: a critical review of the state-of-the-art, current prospects, and future challenges.
      . However, the role of classical cytokines like IL-1β and TNFα as a driving force in OA is disputed
      • van Dalen S.C.
      • Blom A.B.
      • Sloetjes A.W.
      • Helsen M.M.
      • Roth J.
      • Vogl T.
      • et al.
      Interleukin-1 is not involved in synovial inflammation and cartilage destruction in collagenase-induced osteoarthritis.
      • Magnano M.D.
      • Chakravarty E.F.
      • Broudy C.
      • Chung L.
      • Kelman A.
      • Hillygus J.
      • et al.
      A pilot study of tumor necrosis factor inhibition in erosive/inflammatory osteoarthritis of the hands.
      . The alarmins S100A8/A9 are toll-like receptor (TLR)4 ligands that are produced in large amounts for prolonged periods by the OA synovium. S100A8/A9 are crucial mediators of synovitis, cartilage destruction and osteophyte formation in both the collagenase-induced OA (CiOA) model and early symptomatic OA patients
      • van Lent P.L.
      • Blom A.B.
      • Schelbergen R.F.
      • Sloetjes A.
      • Lafeber F.P.
      • Lems W.F.
      • et al.
      Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
      • Schelbergen R.F.
      • de Munter W.
      • van den Bosch M.H.
      • Lafeber F.P.
      • Sloetjes A.
      • Vogl T.
      • et al.
      Alarmins S100A8/S100A9 aggravate osteophyte formation in experimental osteoarthritis and predict osteophyte progression in early human symptomatic osteoarthritis.
      . Furthermore, synovial macrophages have been shown to be crucial for the development of cartilage degeneration and osteophyte formation in the CiOA model
      • Blom A.B.
      • van Lent P.L.
      • Libregts S.
      • Holthuysen A.E.
      • van der Kraan P.M.
      • van Rooijen N.
      • et al.
      Crucial role of macrophages in matrix metalloproteinase-mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3.
      • Blom A.B.
      • van Lent P.L.
      • Holthuysen A.E.
      • van der Kraan P.M.
      • Roth J.
      • van Rooijen N.
      • et al.
      Synovial lining macrophages mediate osteophyte formation during experimental osteoarthritis.
      .
      S100A8/A9 can induce reactive oxygen species (ROS) production by activating nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) oxidases
      • Benedyk M.
      • Sopalla C.
      • Nacken W.
      • Bode G.
      • Melkonyan H.
      • Banfi B.
      • et al.
      HaCaT keratinocytes overexpressing the S100 proteins S100A8 and S100A9 show increased NADPH oxidase and NF-kappaB activities.
      . This family of outward-facing NOX complexes consists of NOX1–5 and dual oxidases DUOX1 and -2. They contain membrane-bound and cytosolic subunits which together transfer electrons across cellular membranes. Their differential tissue distribution and activation mechanisms enable the induction of a plethora of processes such as cellular signaling, gene expression, senescence, and cell death
      • Bedard K.
      • Krause K.H.
      The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
      . Under inflammatory conditions, NOX2 assembly and subsequent ROS production is strongly enhanced
      • Chenevier-Gobeaux C.
      • Lemarechal H.
      • Bonnefont-Rousselot D.
      • Poiraudeau S.
      • Ekindjian O.G.
      • Borderie D.
      Superoxide production and NADPH oxidase expression in human rheumatoid synovial cells: regulation by interleukin-1beta and tumour necrosis factor-alpha.
      • Rada B.
      • Leto T.L.
      Oxidative innate immune defenses by Nox/Duox family NADPH oxidases.
      .
      Although ROS are important for host defense, prolonged elevated levels promote tissue damage
      • Bedard K.
      • Krause K.H.
      The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
      . This may be a mechanism by which S100A8/A9 cause OA pathology. S100A8/A9 are particularly produced and released by monocytes and activated macrophages, which express NOX1, NOX2, and NOX4, as well as by polymorphonuclear cells (PMNs) which exclusively express NOX213. There are several indications that S100A8/A9 enhance NOX2 activity, which is initiated by phosphorylation of regulating subunit neutrophil cytosolic factor-1 (Ncf1). This results in recruitment of the other NOX2 cytosolic subunits Ncf2 and -4 and the GTPase Rac1, followed by an activating conformation change in the membrane-bound subunits Cytochrome b-245 light and heavy chain (encoded by genes Cyba and Cybb respectively)
      • Bedard K.
      • Krause K.H.
      The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
      . S100A8/A9 have been described to interact with Ncf2 thereby increasing its affinity for Cytochrome b-245 heavy chain
      • Doussiere J.
      • Bouzidi F.
      • Vignais P.V.
      The S100A8/A9 protein as a partner for the cytosolic factors of NADPH oxidase activation in neutrophils.
      • Berthier S.
      • Paclet M.H.
      • Lerouge S.
      • Roux F.
      • Vergnaud S.
      • Coleman A.W.
      • et al.
      Changing the conformation state of cytochrome b558 initiates NADPH oxidase activation: MRP8/MRP14 regulation.
      . Moreover, NOX2 can be activated by p38 MAPK-dependent translocation of S100A8/A9 towards the plasma membrane
      • Schenten V.
      • Brechard S.
      • Plancon S.
      • Melchior C.
      • Frippiat J.P.
      • Tschirhart E.J.
      iPLA2, a novel determinant in Ca2+- and phosphorylation-dependent S100A8/A9 regulated NOX2 activity.
      .
      ROS are highly reactive compounds with an extremely short half-life, so it is difficult to accurately measure levels in synovial fluid. NOX2-derived ROS production is increased in synoviocytes from OA patients and in the synovium during knee OA and is associated with increased levels of the NALP3 inflammasome
      • Chenevier-Gobeaux C.
      • Lemarechal H.
      • Bonnefont-Rousselot D.
      • Poiraudeau S.
      • Ekindjian O.G.
      • Borderie D.
      Superoxide production and NADPH oxidase expression in human rheumatoid synovial cells: regulation by interleukin-1beta and tumour necrosis factor-alpha.
      • Chenevier-Gobeaux C.
      • Simonneau C.
      • Therond P.
      • Bonnefont-Rousselot D.
      • Poiraudeau S.
      • Ekindjian O.G.
      • et al.
      Implication of cytosolic phospholipase A2 (cPLA2) in the regulation of human synoviocyte NADPH oxidase (Nox2) activity.
      • Clavijo-Cornejo D.
      • Martinez-Flores K.
      • Silva-Luna K.
      • Martinez-Nava G.A.
      • Fernandez-Torres J.
      • Zamudio-Cuevas Y.
      • et al.
      The overexpression of NALP3 inflammasome in knee osteoarthritis is associated with synovial membrane prolidase and NADPH oxidase 2.
      . Additionally, increased expression of the antioxidant superoxide dismutase (SOD) in OA synovium and serum suggests ROS involvement in OA pathology
      • Fitowska A.
      • Ostalowska A.
      • Dobrakowski M.
      • Hapeta B.
      • Kasperczyk A.
      • Stoltny T.
      • et al.
      Protein metabolism in the synovial membrane in the hip osteoarthritis.
      • Fernandez-Moreno M.
      • Soto-Hermida A.
      • Pertega S.
      • Oreiro N.
      • Fernandez-Lopez C.
      • Rego-Perez I.
      • et al.
      Mitochondrial DNA (mtDNA) haplogroups and serum levels of anti-oxidant enzymes in patients with osteoarthritis.
      . Moreover, ROS are associated with chondrocyte senescence and cartilage breakdown in an immune complex-mediated arthritis, which signifies a role for ROS in synovitis-mediated pathology
      • van Lent P.L.
      • Nabbe K.C.
      • Blom A.B.
      • Sloetjes A.
      • Holthuysen A.E.
      • Kolls J.
      • et al.
      NADPH-oxidase-driven oxygen radical production determines chondrocyte death and partly regulates metalloproteinase-mediated cartilage matrix degradation during interferon-gamma-stimulated immune complex arthritis.
      • Henrotin Y.E.
      • Bruckner P.
      • Pujol J.P.
      The role of reactive oxygen species in homeostasis and degradation of cartilage.
      . Nevertheless, despite these studies indicating that ROS production aggravates cartilage degeneration, pilot studies showed no effects on cartilage damage after induction of CiOA in Ncf1** mice. These mice lack NOX2-derived ROS due to a spontaneous mutation in the Ncf1 subunit and thus can be used to study the role of ROS in vivo
      • Huang C.K.
      • Zhan L.
      • Hannigan M.O.
      • Ai Y.
      • Leto T.L.
      P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+.
      .
      Therefore, in the present study we investigated the role of NOX2-derived ROS in the inflammatory CiOA model, which is characterized by cartilage degeneration and ectopic bone formation, in more detail and hypothesized that absence of ROS production by NOX2 would not affect OA pathology. Both local and systemic inflammation as well as cartilage degeneration and ectopic bone formation were assessed in the early and late stage of the disease.

      Methods

       Animals

      Male and female C57Bl/6 wild type (WT) mice were obtained from Janvier and Ncf1** mice in a C57Bl/6 background (B6(Cg)-Ncf1m1J/J) were obtained from The Jackson Laboratory. Ncf1** mice have a spontaneous point mutation leading to aberrant splicing and absence of intact Ncf1 protein
      • Huang C.K.
      • Zhan L.
      • Hannigan M.O.
      • Ai Y.
      • Leto T.L.
      P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+.
      • Sareila O.
      • Jaakkola N.
      • Olofsson P.
      • Kelkka T.
      • Holmdahl R.
      Identification of a region in p47phox/NCF1 crucial for phagocytic NADPH oxidase (NOX2) activation.
      . We used twelve-week-old mice housed in filter-top cages with corncob bedding under pathogen-free conditions and a 12 h light–dark cycle with unlimited access to food and water. Female mice were randomly allocated to their cages, whereas male mice were housed in groups how they arrived, to avoid fighting. All animal studies were performed according to the Dutch law and were approved by the local Animal Experimentation Committee (RU-DEC 2014-193).

       Induction of CiOA

      Animals were anesthetized using 2.5% isoflurane in O2 and CiOA was induced by two intra-articular injections of 1U collagenase type VII from Clostridium histolyticum (Sigma–Aldrich) into knee joints of WT and Ncf1** mice on day 0 and 2
      • van der Kraan P.M.
      • Vitters E.L.
      • van Beuningen H.M.
      • van de Putte L.B.
      • van den Berg W.B.
      Degenerative knee joint lesions in mice after a single intra-articular collagenase injection. A new model of osteoarthritis.
      . Animals were euthanized by cervical dislocation. Day 42 was taken as end point of the disease. Contralateral saline-injected knee joints were used as controls. Serum samples were collected on day 7 and 42 via retro-orbital punctures with non-heparinized 20 μl end-to-end capillaries (Vitrex).

       S100A8 injections

      Recombinant human S100A8 was expressed and purified in house as described previously
      • Fassl S.K.
      • Austermann J.
      • Papantonopoulou O.
      • Riemenschneider M.
      • Xue J.
      • Bertheloot D.
      • et al.
      Transcriptome assessment reveals a dominant role for TLR4 in the activation of human monocytes by the alarmin MRP8.
      . Recombinant S100A8 (5 μg) in a total volume of 6 μl was injected intra-articularly into knee joints of naive mice. 5 μg of bovine serum albumin (BSA; Sigma Aldrich, catalog #A3803) in a volume of 6 μl was injected as control into the contralateral knee joints.

       Assessment of joint swelling

      Joint swelling in WT and Ncf1** mice was determined using consecutive measurements in the same mouse on day 3 and 7 of CiOA, based on the accumulation of 99mTc in the affected joint as previously described
      • Kruijsen M.W.
      • van den Berg W.B.
      • van de Putte L.B.
      • van den Broek W.J.
      Detection and quantification of experimental joint inflammation in mice by measurement of 99mTc-pertechnetate uptake.
      . Briefly, after subcutaneous injection of 0.75 MBq 99mTc in 200 μl saline, external gamma radiation in the knee joints was measured. Swelling was calculated as ratio of gamma counts in the right (inflamed) knee joint to gamma counts in the left (control) knee joint.

       Preparation of RNA and quantitative real-time polymerase chain reaction (qRT-PCR)

      After sacrificing mice on day 7, 21 and 42 after induction of CiOA in WT mice, and on day 1 and 3 after injection of S100A8, mRNA was isolated from synovial specimens as described previously
      • Van Meurs J.B.
      • Van Lent P.L.
      • Joosten L.A.
      • Van der Kraan P.M.
      • Van den Berg W.B.
      Quantification of mRNA levels in joint capsule and articular cartilage of the murine knee joint by RT-PCR: kinetics of stromelysin and IL-1 mRNA levels during arthritis.
      . mRNA levels of NOX subunits were detected using a StepOnePlus qRT-PCR system (Applied Biosystems) using SYBRgreen master mix (Applied Biosystems) and specific primers (Biolegio). Table I shows the primer sequences. Relative quantification of the qRT-PCR signal was performed by correcting the Ct value of the gene of interest for GAPDH content (−ΔCt).
      Table IPrimers used for qRT-PCR
      GeneForward primer (5′ → 3′)Reverse primer (5′ → 3′)
      GAPDHGGCAAATTCAACGGCACAGTTAGTGGGGTCTCGCTCCTG
      CybaGAGGCACCATCAAGCAACCACACCCTCACTCGGCTTCTTT
      CybbACCCTCCTATGACTTGGAAATGGCGAACCAACCTCTCACAAAGGT
      Ncf1TGGACTGAGACTCTACAGCAAAGGCGCTCACTGCCTCCTCTCAT
      Ncf2GCTTCGGAACATGGTGTCTAAGACGACGCCGGTAGCTCAGTT
      Ncf4CCAACTGGCTACGATGCTACTTTCTCTGGAACTCACGCCTCATG

       Histological analysis of OA progression

      At day 7 and day 42 after induction of CiOA in WT and Ncf1** mice, animals were sacrificed to obtain histology of OA and saline-injected contralateral control joints. Hereto, total knee joints were fixed in 4% formalin, decalcified in 5% formic acid buffered in PBS and embedded in paraffin. Coronal sections (7 μm) were stained with hematoxylin/eosin (HE) or safranin O/fast green (SafO) to score pathology in a blinded manner in five sections per joint. Synovial inflammation in the patellar region was arbitrarily scored, using a scale of 0 (no inflammation) to 3 (substantial inflammation). Cartilage destruction was determined using a modified OARSI scoring system
      • Glasson S.S.
      • Chambers M.G.
      • Van Den Berg W.B.
      • Little C.B.
      The OARSI histopathology initiative – recommendations for histological assessments of osteoarthritis in the mouse.
      . Briefly, the OA score is the assessment of the grade (OA depth progression into cartilage; 0–6) multiplied by the stage (percentage damaged cartilage of the total surface irrespective of the underlying grade; 0–5). Ectopic bone surface was measured using an image analysis system (Leica Application Suite, Leica Microsystems). For cartilage destruction, 5 sections per knee were analyzed. For synovial inflammation and ectopic bone surface, 3 sections per knee were analyzed.

       Immunohistochemistry

      Immunostaining on knee joints was conducted to visualize the presence of aggrecan neo-epitopes NITEGE and VDIPEN, which are the result of aggrecan cleavage by ADAMTSs and MMPs, respectively. Specific primary antibodies directed against these cleavage sites were a kind gift from John Mort (McGill University, Montreal, Canada) and were used as described before
      • Blaney Davidson E.N.
      • Vitters E.L.
      • van Lent P.L.
      • van de Loo F.A.
      • van den Berg W.B.
      • van der Kraan P.M.
      Elevated extracellular matrix production and degradation upon bone morphogenetic protein-2 (BMP-2) stimulation point toward a role for BMP-2 in cartilage repair and remodeling.
      . Staining was scored using the Leica Application Suite (Leica Microsystems) for three sections per joint. No positive staining was found in sections stained for NITEGE. The cartilage surface area that stained positive for VDIPEN was expressed as relative staining of total area that was measured with WT joints set at an average of 1.

       Measurement of local and systemic protein levels during CiOA

      In order to isolate synovial specimens, knee joints from WT and Ncf1** mice, sacrificed on day 7 and 42 after induction of CiOA, were stripped from their skin and the tissue between the patella and tibiofemoral joint were isolated. Subsequently, punch biopsies were taken from the patellar area and biopsies were incubated in culture medium containing 0.1% BSA for 1 h at RT to obtain synovial washouts. S100A8/A9 protein levels were quantified by a sandwich enzyme-linked immunosorbent assay (ELISA)
      • Vogl T.
      • Tenbrock K.
      • Ludwig S.
      • Leukert N.
      • Ehrhardt C.
      • van Zoelen M.A.
      • et al.
      Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock.
      . Protein levels of other cytokines were determined with Luminex multianalyte technology, using the Bio-Plex™ 200 System (Bio-Rad).

       Determination of inflammatory cell populations using flow cytometry

      Relative numbers of phagocytes in synovium, bone marrow (BM), spleen, and blood from WT and Ncf1** mice were determined as described previously
      • Cremers N.A.J.
      • van den Bosch M.H.J.
      • van Dalen S.
      • Di Ceglie I.
      • Ascone G.
      • van de Loo F.
      • et al.
      S100A8/A9 increases the mobilization of pro-inflammatory Ly6Chigh monocytes to the synovium during experimental osteoarthritis.
      . Synovial tissue was digested in 0.1 mg/ml DNAse I (Roche), 1.0 mg/ml Collagenase D (Roche), and 2.4 mg/ml Dispase II (Sigma) at 37°C for 40 min. Complete femurs were crushed with mortar and pestle. Spleens were mashed through 70 μm cell strainers (Falcon). After erythrocyte lysis (155 mM NH4Cl/12 mM KHCO3/0.1 mM EDTA, pH 7.3) and intensive lavation with PBS, cells were counted and stained. Single, viable myeloid cells were positively selected for CD11b, followed by Ly6G and Ly6C to distinguish PMNs and monocytes, respectively. Monocytes were plotted for F4/80 and Ly6C to select macrophages, Ly6Chigh, and Ly6Clow monocytes, depending on the tissue.

       In vitro ROS detection

      Conditionally immortalized ER-Hoxb8 were generated from BM as previously described
      • Wang G.G.
      • Calvo K.R.
      • Pasillas M.P.
      • Sykes D.B.
      • Hacker H.
      • Kamps M.P.
      Quantitative production of macrophages or neutrophils ex vivo using conditional Hoxb8.
      . WT and Ncf1** ER-Hoxb8 cells were differentiated towards PMNs for 4 days with 2% supernatant from SCF-producing CHO cell line. Bone-marrow derived stromal cells were differentiated towards macrophages for 6 days with 15 ng/ml M-CSF (R&D Systems). Differentiated PMNs and macrophages were stimulated with 4 μM phorbol 12-myristate 13-acetate (PMA; LC Laboratories) or 1 μg/mL recombinant mouse S100A8 and NOX-derived ROS production was inhibited using 8 μM diphenyleneiodonium (DPI; Sigma–Aldrich). After 2 h (PMNs) or 24 h (macrophages) incubation at 37°C extracellular ROS release was detected with an Amplex Red Hydrogen Peroxide/Peroxidase Assay kit (Invitrogen) using a Clariostar Monochromator Microplate Reader (BMG Labtech).

       Statistical analysis

      Group sizes were calculated to be able to detect differences between groups with a power of 80% and a level of significance of 0.05. We based our sample size calculation on the hypothesis that there would be no difference in the severity of pathology between WT and Ncf1** mice. To this end, we decided that a difference in pathology smaller than 20% would be considered as not relevant. Using the standard formula and with an expected variation of 20%, this resulted in a total number of required animals per group of 16 for the main readout parameters cartilage degeneration, synovial inflammation and ectopic bone formation in this study. Statistical analyses were performed using Graphpad Prism version 7.03. 95% Confidence Intervals (CIs) for the relative differences between the two groups (WT and Ncf1) were calculated using the Confidence Interval Calculator on www.gigacalculator.com. The purpose of all statistical tests used was to determine whether differences between the genotypes WT and Ncf1** were significant and biologically relevant. Differences between groups for non-continuous outcome parameters, e.g., arbitrary scores and histological OA scores, were calculated using non-parametric tests. Differences between groups were tested using a Mann–Whitney U test (synovial inflammation), one-way analysis of variance (ANOVA) followed by a Bonferroni or Dunn's Multiple Comparison post-test for all relevant comparisons (cartilage damage, gene expression, protein levels, ectopic bone size, relative number of cells, ROS production, 99mTc accumulation) or two-way ANOVA (S100A8/A9 levels). Bonferroni correction for multiplicity was used to test for significant differences between the primary variable (genotype) and multiple secondary variables (time points, tibiofemoral cartilage areas, cytokines, cell types and conditions). Families of secondary variables are defined based on the aspect of pathology they represent. Complete families are always represented in not more than one figure. P-values lower than 0.05 were considered significant. Results are expressed as mean values ± 95% CIs.

      Results

       Synovial gene expression of NOX2 subunits is enhanced during CiOA

      To substantiate the involvement of ROS in CiOA, we analyzed a previously described microarray performed on the synovium of murine WT knee joints with CiOA
      • Remst D.F.
      • Blom A.B.
      • Vitters E.L.
      • Bank R.A.
      • van den Berg W.B.
      • Blaney Davidson E.N.
      • et al.
      Gene expression analysis of murine and human osteoarthritis synovium reveals elevation of transforming growth factor beta-responsive genes in osteoarthritis-related fibrosis.
      . Functional annotation clustering analysis showed a significant enrichment of the gene clusters ‘superoxide metabolic process’, ‘superoxide-generating NADPH oxidase activity’, ‘superoxide anion generation’ and ‘NADPH oxidase complex’ on day 7 (P < 0.05), suggesting synovial ROS production in this OA model. More detailed analysis of the NOX2 complex showed a significant up-regulation of Cyba, Cybb, Ncf1, Ncf2, and Ncf4 (15.0×, 13.2×, 15.9×, 13.4×, 13.2× increase respectively) on day 7 after induction of CiOA. Some of these genes were still significantly increased on day 21 but had returned to basal levels on day 42 [Fig. 1(A)–(E)]. The course of NOX2 subunit gene expression coincided with the severity of synovitis
      • van Lent P.L.
      • Blom A.B.
      • Schelbergen R.F.
      • Sloetjes A.
      • Lafeber F.P.
      • Lems W.F.
      • et al.
      Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
      . Interestingly, intraarticular injection of S100A8 resulted in increased expression of Ncf1, the critical regulatory subunit of NOX2, which reached significance on day 3 (4.1 fold increase compared to BSA-injected controls) [Fig. 1(F)]. This indicates that S100A8 could enhance Ncf1 expression in CiOA synovium which may aggravate OA pathology.
      Fig. 1
      Fig. 1Synovial gene expression of NOX2 subunits is increased during the course of CiOA. A–E) Synovial gene expression of the subunits of the NOX2 complex (Cyba, Cybb, Ncf1, Ncf2, and Ncf4) in CiOA and saline-injected contralateral joints was assessed using qRT-PCR on day 7, 21 and 42 after induction of the model. All subunits showed a significantly increased expression on day 7 that was still present on day 21 for Cyba, Ncf1 and Ncf2, but expression levels of all genes reduced to basal levels on day 42 (n = 4 joints per group). F) Shortly after injection of S100A8 into knee joints of naïve mice, elevated Ncf1 gene expression levels were found in synovium which reached significance on day 3. Control samples were collected from BSA-injected contralateral joints (n = 6 joints for day 1, n = 5 joints for day 3). Symbols represent individual knee joints. Gene expression levels are presented as -ΔCt compared to GAPDH. Closed symbols represent control joints, open symbols represent CiOA (AE) or S100A8-injected (F) knee joints. Bars show mean values ± 95% CIs.

       No difference in cartilage damage in early CiOA between WT and Ncf1** mice

      Next, we investigated whether NOX-2 derived ROS were involved in the development of cartilage degeneration after induction of CiOA. Because tibiofemoral dislocations are not a clinically relevant phenomenon in humans and give artificially high OA pathology in the CiOA model, we left them out of our analyses. On day 7 after induction, cartilage damage was relatively mild as compared to contralateral control joints [Fig. 2(A)]. Moreover, we observed no significant difference in cartilage degeneration between Ncf1** and WT joints (P > 0.999 for all areas and the total mean). Moreover, we investigated whether Ncf1 deficiency affect MMP activity in cartilage by assessing the aggrecan neoepitope VDIPEN. However, no significant differences in the presence of this neoepitope could be observed between WT and Ncf1** mice (relative score of 1.0 and 1.1 respectively, P > 0.999) [Fig. 2(B)].
      Fig. 2
      Fig. 2Ncf1 deficiency does not affect cartilage breakdown in knee joints on day 7 of CiOA. A) Cartilage damage on day 7 of CiOA was comparable between WT and Ncf1** mice, both on the individual areas of the tibiofemoral joint and on the total mean scores. Symbols represent the average score of 5 sections per joint. B) MMP activity in early cartilage damage was studied by assessing the relative staining of aggrecan neo-epitope VDIPEN in Ncf1** knee joints compared to WT knee joints. Arrows indicate positive cells. Isotype control stained sections showed no positive staining. No differences were found between WT and Ncf1** mice. Images shown are representative for the treatment groups (original magnification ×400). T = tibia, F = femur, M = meniscus. Symbols represent individual mice (n = 18 joints per group). Closed circles represent WT knee joints, open circles represent Ncf1** knee joints. Red crosses in A represent dislocated joints, which were excluded from analysis because tibiofemoral dislocations are not a clinically relevant phenomenon in humans and give artificially high OA pathology (n = 7 for WT mice, n = 3 for Ncf1** mice). Bars show mean values ± 95% CIs.

       Ncf1** mice show no difference in inflammation in early CiOA

      Because the development of CiOA is strongly dependent on synovitis, we studied the synovial activation in more detail. Joint swelling was significantly increased in OA as compared to contralateral control joints in both WT and Ncf1** mice (R/L ratio 1.6 and 1.7 on day 3, 1.3 and 1.4 on day 7, P = 0.02 and P < 0.0001 respectively), but no significant difference was found between both genotypes (P > 0.999) [Fig. 3(A)]. Moreover, whereas synovitis was increased compared to control joints, no significant differences between WT and Ncf1** mice were found (P = 0.3) [Fig. 3(B)]. Although protein levels of IL-1β, KC, MCP-1, and IL-10 in washouts were below the detection limit, S100A8/A9 levels in osteoarthritic joints were significantly increased compared to control joints but no differences between WT and Ncf1** joints were found (P = 0.5) [Fig. 3(C)]. Finally, to study the effects of Ncf1 deficiency on cell influx, relative numbers of different phagocyte subpopulations in the synovium were assessed with flow cytometry. However, no differences in the numbers of macrophages, PMNs, and both Ly6Chigh and Ly6low monocytes could be observed between WT and Ncf1** mice (P > 0.999 for all cell types) [Fig. 3(D)].
      Fig. 3
      Fig. 3Synovial inflammation on day 7 of CiOA is not affected by Ncf1 deficiency. A) The ratio of 99mTc uptake in CiOA vs saline-injected control knee joints was determined as a measure for swelling, but no differences were detected between WT and Ncf1** mice. Connecting lines represent the course of the mean R/L ratio (n = 8 mice per group). B) Synovial inflammation was clearly present in CiOA knee joints, but arbitrary scoring (0–3) showed no significant differences between WT and Ncf1** mice (n = 11 joints for WT and n = 15 joints for Ncf1**). Symbols represent the average of 3 sections per joint. C) Local protein levels of S100A8/A9 were increased in synovial washouts of CiOA joints when compared to control joints, but levels were comparable between WT and Ncf1** mice (n = 12 mice per group). D) Subsets of phagocytes present in synovium on day 7 of CiOA were assessed with flow cytometry, but the percentages of macrophages, PMNs and Ly6Chigh and Ly6Clow monocytes were comparable between WT and Ncf1** mice (n = 4 mice per group). Images shown are representative for the treatment groups (original magnification ×100). F = femur, S = synovium. Symbols represent individual knee joints. Closed circles represent WT knee joints, open circles represent Ncf1** knee joints. Bars show mean values ± 95% CIs.

       Ncf1 deficiency does not alter systemic inflammation in early CiOA

      Induction of CiOA results in elevated S100A8/A9 serum levels and shifts in immune cell populations in BM
      • Cremers N.A.J.
      • van den Bosch M.H.J.
      • van Dalen S.
      • Di Ceglie I.
      • Ascone G.
      • van de Loo F.
      • et al.
      S100A8/A9 increases the mobilization of pro-inflammatory Ly6Chigh monocytes to the synovium during experimental osteoarthritis.
      . Because the absence of NOX2-derived ROS could affect systemic inflammation, we determined cytokine levels in serum. We observed comparable levels of S100A8/A9, IL-1β, KC, MCP-1, and IL-10 in WT and Ncf1** mice (P > 0.9 for all mediators) [Fig. 4(A)]. Moreover, as expected, no macrophages and Ly6Clow monocytes were detected in BM, whereas relative numbers of PMNs and Ly6Chigh monocytes were not significantly different between WT and Ncf1** mice. In addition, relative numbers of macrophages, PMNs, Ly6Chigh and Ly6Clow monocytes were comparable between WT and Ncf1** mice in spleen and blood (P > 0.999 for both cell types) [Fig. 4(B)–(D)].
      Fig. 4
      Fig. 4Ncf1 deficiency does not alter systemic inflammation during early CiOA. A) To study systemic inflammation on day 7 after induction of CiOA in Ncf1** mice, concentrations of S100A8/A9, IL-1β, KC, MCP-1, and IL-10 in serum were measured. Levels were not different in Ncf1** mice when compared to WT mice (n = 18 mice per group). Additionally, proportions of different phagocyte subsets in BM (B), spleen (C), and blood (D) on day 7 of CiOA were assessed with flow cytometry (n = 4 mice per group). No significant differences were found in relative numbers of macrophages, PMNs, or Ly6Chigh and Ly6Clow monocytes in the analyzed compartments. Symbols represent individual samples. Closed circles represent WT knee joints, open circles represent Ncf1** knee joints. Bars show mean values ± 95% CIs.

       End stage CiOA pathology is not different between WT and Ncf1** mice

      Next we studied whether a prolonged lack of NOX2-derived ROS alters CiOA pathology in later stages of CiOA. In agreement with our findings at 7 days after induction of CiOA, we did not observe significant differences in cartilage degeneration between WT and Ncf1** mice at day 42 (P > 0.999 for all areas and the total mean) [Fig. 5(A)]. These findings were substantiated in male mice, in which no significant cartilage damage was found (P > 0.999 for all areas and the total mean on day 7 and 42 of CiOA) (Supplemental Figure 1). Moreover, no significant differences in synovial inflammation (P = 0.5) [Fig. 5(B)] and serum protein levels of IL-1β, KC, MCP-1, and IL-10 (P > 0.999) were found, although serum levels of S100A8/A9 were significantly higher in Ncf1** mice [Fig. 5(C)]. No positive staining of the neo-epitope VDIPEN reflecting MMP-activity was detected on day 42 (data not shown). Finally, the size of ectopic bone formation was scored at 5 sites in the knee joints. No differences were found between WT and Ncf1** mice at 4 out of the 5 sites (P > 0.999), but the size of ectopic bone structures in the femoral side of the medial collateral ligament was significantly lower in Ncf1** mice (44% reduction compared to WT) [Fig. 5(D)].
      Fig. 5
      Fig. 5Ncf1 deficiency does not alter joint pathology in end stage CiOA. Joint pathology in WT and Ncf1** mice on day 42 after induction of CiOA was studied. A) Cartilage damage on day 42 was comparable between WT and Ncf1** mice, both on the individual areas of the tibiofemoral joint as well as on the total mean score. Symbols represent the average score of 5 sections per joint. B) Synovial inflammation in the patellar region on day 42 was comparable between WT and Ncf1** mice. Symbols represent the average of 3 sections per joint. C) Serum levels of inflammatory mediators were comparable between WT and Ncf1** mice for S100A8/A9, IL-1β, KC, MCP-1, and IL-10.D) Ectopic bone formation was abundantly present in end stage CiOA knee joints (indicated by arrows), but no differences were found in the average size on different locations in the joint, except for the femoral side of the medial collateral ligament where the ectopic bone size was significantly lower in Ncf1** mice. Symbols represent the average of 3 sections per joint. MCL = medial collateral ligament, Ent. = femoral enthesophyte in the MCL. Images shown are representative for the treatment groups (original magnification ×400 (A), 200 (B), 50 (D)). S = synovium, T = tibia, F = femur, M = meniscus. Symbols represent individual joints or mice (n = 15 for WT, n = 18 for Ncf1**). Closed circles represent WT knee joints, open circles represent Ncf1** knee joints. Bars show mean values ± 95% CIs.

       Lack of ROS production by Ncf1** PMNs may be compensated by macrophages

      Our results suggest that NOX2-derived ROS are not involved in CiOA pathology, so we aimed to find a clue that may explain this finding. Therefore, ROS production was investigated in PMNs and macrophages, which are the main expressers of NOX2. ROS production by WT PMNs was strongly enhanced after PMA stimulation, but not S100A8 stimulation, and was completely inhibited after addition of the pan-NOX inhibitor DPI. In contrast to WT cells, ROS release by Ncf1** PMNs was completely absent despite stimulation with PMA and S100A8 [Fig. 6(A)]. Interestingly, Ncf1** macrophages still produced considerable amounts of ROS after S100A8 and PMA stimulation [Fig. 6(B)]. The S100A8 and PMA-induced ROS production was strongly inhibited in both Ncf1** and WT cells after addition of DPI. This suggests that in Ncf1 deficient macrophages compensatory mechanisms are activated that rescue the ROS production.
      Fig. 6
      Fig. 6NOX-mediated ROS production is negligible in Ncf1** PMNs but active in Ncf1** macrophages. To investigate ROS release by phagocytes in more detail, PMNs and macrophages were stimulated in vitro with S100A8 or PMA to induce ROS production or inhibited with the pan-NOX inhibitor DPI. A) WT PMNs showed evident ROS production after PMA stimulation, whereas this effect was absent in Ncf1** PMNs. The PMA-induced signal was inhibited by DPI. S100A8 did not induce ROS production in both WT and Ncf1** PMNs. B) ROS release by Ncf1** macrophages was comparable to WT macrophages. S100A8-induced ROS production was comparable between WT and Ncf1** macrophages. PMA-induced ROS production was evident in both genotypes, although significantly lower in Ncf1** macrophages. The signal was inhibited by DPI in all conditions. Symbols represent individual samples (n = 3 per group). Closed circles represent WT cells, open circles represent Ncf1** cells. Bars show mean values ± 95% CIs.

      Discussion

      Although the complex process of joint destruction in OA remains elusive, it is well-recognized that S100A8/A9 levels are elevated in OA joints
      • van Lent P.L.
      • Blom A.B.
      • Schelbergen R.F.
      • Sloetjes A.
      • Lafeber F.P.
      • Lems W.F.
      • et al.
      Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
      • Schelbergen R.F.
      • de Munter W.
      • van den Bosch M.H.
      • Lafeber F.P.
      • Sloetjes A.
      • Vogl T.
      • et al.
      Alarmins S100A8/S100A9 aggravate osteophyte formation in experimental osteoarthritis and predict osteophyte progression in early human symptomatic osteoarthritis.
      . These damage-associated molecular patterns can enhance ROS production by NOX complexes
      • Benedyk M.
      • Sopalla C.
      • Nacken W.
      • Bode G.
      • Melkonyan H.
      • Banfi B.
      • et al.
      HaCaT keratinocytes overexpressing the S100 proteins S100A8 and S100A9 show increased NADPH oxidase and NF-kappaB activities.
      • Doussiere J.
      • Bouzidi F.
      • Vignais P.V.
      The S100A8/A9 protein as a partner for the cytosolic factors of NADPH oxidase activation in neutrophils.
      • Berthier S.
      • Paclet M.H.
      • Lerouge S.
      • Roux F.
      • Vergnaud S.
      • Coleman A.W.
      • et al.
      Changing the conformation state of cytochrome b558 initiates NADPH oxidase activation: MRP8/MRP14 regulation.
      , suggesting that during OA exaggerated levels of oxygen radicals may be present in the affected joint
      • Rahmati M.
      • Mobasheri A.
      • Mozafari M.
      Inflammatory mediators in osteoarthritis: a critical review of the state-of-the-art, current prospects, and future challenges.
      . Prolonged elevated levels of ROS lead to oxidative stress that damages cartilage
      • Bedard K.
      • Krause K.H.
      The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
      , and this may be a mechanism by which elevated S100A8/A9 levels result in joint destruction.
      Intracellular ROS are by-products of oxygen metabolism and act as short-lived messengers in a variety of signaling pathways such as gene expression, cell differentiation, and intracellular pH homeostasis
      • Shadel G.S.
      • Horvath T.L.
      Mitochondrial ROS signaling in organismal homeostasis.
      . In chondrocytes, intracellular ROS regulate extracellular matrix synthesis, but elevated ROS production can drive chondrocytes towards a hypertrophic state
      • Morita K.
      • Miyamoto T.
      • Fujita N.
      • Kubota Y.
      • Ito K.
      • Takubo K.
      • et al.
      Reactive oxygen species induce chondrocyte hypertrophy in endochondral ossification.
      . Additionally, both high and low intracellular ROS concentrations can result in elevated MMP levels, leading to OA-like cartilage degradation
      • Henrotin Y.E.
      • Bruckner P.
      • Pujol J.P.
      The role of reactive oxygen species in homeostasis and degradation of cartilage.
      • Reed K.N.
      • Wilson G.
      • Pearsall A.
      • Grishko V.I.
      The role of mitochondrial reactive oxygen species in cartilage matrix destruction.
      • Del Carlo M.
      • Schwartz D.
      • Erickson E.A.
      • Loeser R.F.
      Endogenous production of reactive oxygen species is required for stimulation of human articular chondrocyte matrix metalloproteinase production by fibronectin fragments.
      . Due to this dual role of ROS, their levels are controlled by antioxidants such as SOD which are constitutively expressed
      • Zelko I.N.
      • Mariani T.J.
      • Folz R.J.
      Superoxide dismutase multigene family: a comparison of the CuZn-SOD (SOD1), Mn-SOD (SOD2), and EC-SOD (SOD3) gene structures, evolution, and expression.
      .
      Extracellular ROS are predominantly produced by phagocytic immune cells such as monocytes, macrophages, and PMNs. This local, temporary oxidative burst damages DNA, proteins, and lipids, and is used to impair invading microorganisms and pathogen-associated particles
      • Bedard K.
      • Krause K.H.
      The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
      . Extracellular levels of antioxidant enzymes are limited under physiological conditions, but increase during inflammation
      • Chaiswing L.
      • Oberley T.D.
      Extracellular/microenvironmental redox state.
      . When ROS levels are high for prolonged periods and exceed the capacity of antioxidants, oxidative stress leads to tissue damage. Excessive ROS trigger pro-apoptotic signaling through activation of stress-activated protein kinases, which include JNK and p38 MAPK
      • Benhar M.
      • Dalyot I.
      • Engelberg D.
      • Levitzki A.
      Enhanced ROS production in oncogenically transformed cells potentiates c-Jun N-terminal kinase and p38 mitogen-activated protein kinase activation and sensitization to genotoxic stress.
      . Despite this, we found no differences in synovial cellularity and numbers of apoptotic chondrocytes (data not shown) between WT and Ncf1** mice during CiOA.
      Previous studies have shown that particularly NOX2 is crucial in joint inflammation during experimental models of rheumatoid arthritis
      • Gelderman K.A.
      • Hultqvist M.
      • Pizzolla A.
      • Zhao M.
      • Nandakumar K.S.
      • Mattsson R.
      • et al.
      Macrophages suppress T cell responses and arthritis development in mice by producing reactive oxygen species.
      • van de Loo F.A.
      • Bennink M.B.
      • Arntz O.J.
      • Smeets R.L.
      • Lubberts E.
      • Joosten L.A.
      • et al.
      Deficiency of NADPH oxidase components p47phox and gp91phox caused granulomatous synovitis and increased connective tissue destruction in experimental arthritis models.
      . NOX2-derived ROS enhance expression of cartilage-degrading MMPs
      • van Lent P.L.
      • Nabbe K.C.
      • Blom A.B.
      • Sloetjes A.
      • Holthuysen A.E.
      • Kolls J.
      • et al.
      NADPH-oxidase-driven oxygen radical production determines chondrocyte death and partly regulates metalloproteinase-mediated cartilage matrix degradation during interferon-gamma-stimulated immune complex arthritis.
      , although the opposite has also been described
      • Kassim S.Y.
      • Fu X.
      • Liles W.C.
      • Shapiro S.D.
      • Parks W.C.
      • Heinecke J.W.
      NADPH oxidase restrains the matrix metalloproteinase activity of macrophages.
      . This suggests that NOX2-derived ROS could be one of the effector mechanisms of tissue destruction in OA. Nevertheless, we found no differences in MMP activity between Ncf1** and WT mice.
      The central role of the TLR4 ligand S100A8/A9 in the CiOA model was confirmed in S100A9−/− mice, which were largely protected against synovial thickening, cartilage damage and osteophyte formation. In agreement with this, we have shown that S100A8/A9 serum levels were higher in early symptomatic OA patients that showed progression of cartilage degeneration and osteophyte size as compared to non-progressors
      • van Lent P.L.
      • Blom A.B.
      • Schelbergen R.F.
      • Sloetjes A.
      • Lafeber F.P.
      • Lems W.F.
      • et al.
      Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
      • Schelbergen R.F.
      • de Munter W.
      • van den Bosch M.H.
      • Lafeber F.P.
      • Sloetjes A.
      • Vogl T.
      • et al.
      Alarmins S100A8/S100A9 aggravate osteophyte formation in experimental osteoarthritis and predict osteophyte progression in early human symptomatic osteoarthritis.
      . In contrast, no significant differences in these pathologies were observed in the destabilization of the medial meniscus (DMM) experimental OA model, where inflammation is scant. Here, we found increased gene expression of various NOX2 subunits in CiOA and that S100A8 enhanced gene expression of the NOX2-regulating subunit Ncf1 which also was biologically relevant, except for Ncf4 (data not shown). This suggests that the high levels of S100A8/A9 might mediate ROS production by NOX2 during CiOA, which may also explain the reduced joint destruction in S100A9−/− mice. However, we found no reduced inflammation, cartilage destruction or ectopic bone formation during the early and late stage of CiOA in Ncf1** mice. Relative differences were lower than 20% and therefore considered biologically irrelevant for most comparisons between WT and Ncf1** mice (data not shown). The 95% CIs mostly consist of the range in which differences are considered biologically irrelevant, although in some cases they do not completely exclude biologically relevant differences. This however, makes it highly unlikely that Ncf1 plays a defining role in CiOA development. Although S100A8 induced gene expression of various NOX complex components, induced NOX-dependent ROS production in our in vitro experiments and S100A8/A9 is generally accepted as an inducer of ROS, we cannot irrefutably conclude from our data whether the in vivo damaging effects of S100A8/A9 in CiOA run via the NOX-mediated production of ROS.
      NOX2 is expressed by both PMNs and macrophages. The increased expression of NOX2 can probably be attributed to increased numbers of macrophages after induction of CiOA, which are crucial for the development of pathology. In contrast, PMNs are only present during the first couple of days after induction of the model. Next to NOX2, macrophages can also assemble NOX1 and NOX4 complexes, which can give additional ROS production, whereas NOX2 is the only NOX complex present in PMNs. This might give clues why we observed that Ncf1** macrophages but not PMNs showed production of ROS upon stimulation by PMA and S100A8.
      Although NOX1 produces lower amounts of ROS than NOX2, NOX1 exhibits a significant constitutive activity
      • Rada B.
      • Leto T.L.
      Oxidative innate immune defenses by Nox/Duox family NADPH oxidases.
      . Interestingly, NOX1 is TLR4 dependent and may thus be stimulated by the high levels of S100A8/A9 that are present in OA
      • Xu Q.
      • Choksi S.
      • Qu J.
      • Jang J.
      • Choe M.
      • Banfi B.
      • et al.
      NADPH oxidases are essential for macrophage differentiation.
      • Kim J.S.
      • Yeo S.
      • Shin D.G.
      • Bae Y.S.
      • Lee J.J.
      • Chin B.R.
      • et al.
      Glycogen synthase kinase 3beta and beta-catenin pathway is involved in toll-like receptor 4-mediated NADPH oxidase 1 expression in macrophages.
      . Overexpression of S100A8/A9 in keratinocytes indeed resulted in enhanced gene expression of several subunits of the NOX1 complex
      • Benedyk M.
      • Sopalla C.
      • Nacken W.
      • Bode G.
      • Melkonyan H.
      • Banfi B.
      • et al.
      HaCaT keratinocytes overexpressing the S100 proteins S100A8 and S100A9 show increased NADPH oxidase and NF-kappaB activities.
      . Because we found that Ncf1** macrophages showed ROS production, one possibility might be that these cells compensate for the loss of NOX2 by increasing their expression of NOX1. Indeed, we observed a significantly increased NOX1 gene expression in the Ncf1** macrophages when compared to WT macrophages (2.9 fold increase, P < 0.05, data not shown).
      Additionally, the NOX1 complex shares many structural and functional similarities with the NOX2 complex. Like NOX2, the NOX1 complex consists of the same cytochrome b-245 light chain and Rac1. The cytosolic proteins Ncf1, Ncf2, and Ncf4, however, are substituted by NOXA1 and NOXO1, which likewise orchestrate the assembly and activity of NOX112. Due to the extensive homology NOXA1 and NOXO1 can replace Ncf1 and Ncf2, leading to functional NOX2 complexes, albeit with a lower yield
      • Geiszt M.
      • Lekstrom K.
      • Witta J.
      • Leto T.L.
      Proteins homologous to p47phox and p67phox support superoxide production by NAD(P)H oxidase 1 in colon epithelial cells.
      . This might explain how macrophages compensate for the loss of NOX2-derived ROS in Ncf1** mice and why we found similar joint destruction in WT and Ncf1** mice.
      Lastly, S100A8/A9 can initiate NOX2 oxidase activation independent of Ncf1 through a calcium-dependent specific conformation change in the membrane-bound subunits
      • Berthier S.
      • Paclet M.H.
      • Lerouge S.
      • Roux F.
      • Vergnaud S.
      • Coleman A.W.
      • et al.
      Changing the conformation state of cytochrome b558 initiates NADPH oxidase activation: MRP8/MRP14 regulation.
      • Berthier S.
      • Nguyen M.V.
      • Baillet A.
      • Hograindleur M.A.
      • Paclet M.H.
      • Polack B.
      • et al.
      Molecular interface of S100A8 with cytochrome b558 and NADPH oxidase activation.
      . This was shown in vitro in a cell-free system containing purified cytochrome b-245 and Ncf2, where addition of S100A8/A9 resulted in increased oxidase activation
      • Berthier S.
      • Paclet M.H.
      • Lerouge S.
      • Roux F.
      • Vergnaud S.
      • Coleman A.W.
      • et al.
      Changing the conformation state of cytochrome b558 initiates NADPH oxidase activation: MRP8/MRP14 regulation.
      . This may further explain why we found no differences in CiOA pathology between WT and Ncf1** mice. However, additional studies are necessary to debunk which mechanism is responsible for the restoration of the ROS production in Ncf1** macrophages.
      Our findings are inconsistent with studies in mice lacking NOX2-derived ROS reporting protection against lung emphysema and diaphragm contractile dysfunction in chronic heart failure
      • Trocme C.
      • Deffert C.
      • Cachat J.
      • Donati Y.
      • Tissot C.
      • Papacatzis S.
      • et al.
      Macrophage-specific NOX2 contributes to the development of lung emphysema through modulation of SIRT1/MMP-9 pathways.
      • Ahn B.
      • Beharry A.W.
      • Frye G.S.
      • Judge A.R.
      • Ferreira L.F.
      NAD(P)H oxidase subunit p47phox is elevated, and p47phox knockout prevents diaphragm contractile dysfunction in heart failure.
      . Interestingly, the opposite has also been described. Systemic lupus erythematosus
      • Olsson L.M.
      • Johansson A.C.
      • Gullstrand B.
      • Jonsen A.
      • Saevarsdottir S.
      • Ronnblom L.
      • et al.
      A single nucleotide polymorphism in the NCF1 gene leading to reduced oxidative burst is associated with systemic lupus erythematosus.
      and joint inflammation in several models of experimental arthritis (RA) were more severe in mice with a mutation in NOX2. However, in rheumatoid arthritis, high numbers of PMNs are present throughout the course of the disease, whereas PMNs are only present during the first couple of days after induction of CiOA. As PMNs have no other NADPH oxidases which could compensate for the loss of functional NOX2, a mutation may have a larger impact on progression of diseases where PMNs are prominent participants. In line with our findings that in the macrophage-driven CiOA model pathology was not affected in the Ncf1** mice, no inhibition of atherosclerosis that is highly dependent on macrophages was found in a model of hypercholesterolemia induced in NOX2-deficient mice
      • Kirk E.A.
      • Dinauer M.C.
      • Rosen H.
      • Chait A.
      • Heinecke J.W.
      • LeBoeuf R.C.
      Impaired superoxide production due to a deficiency in phagocyte NADPH oxidase fails to inhibit atherosclerosis in mice.
      .
      Here we report that Ncf1** mice, described to lack functional NOX2 complexes, show similar CiOA pathology as WT mice, even though gene expression of NOX2 subunits was significantly increased during early CiOA. This is likely mediated by compensatory mechanisms in macrophages that rescue the production of ROS in this cell type.

      Author contributions

      Conception and design of study: SvD, MK, FvdL, AB, PvdK, PvL, MvdB.
      Acquisition of data: SvD, NK, BW, MH, AS, NC, JR, TV, AB, MvdB.
      Analysis and interpretation of data: SvD, NK, NC, AB, PvL, MvdB.
      Drafting the article: SvD, NK, AB, PvL, MvdB.
      Revising the article critically: BW, MH, AS, NC, MK, FvdL, JR, TV, PvdK.
      Final approval of the submitted manuscript: SvD, NK, BW, MH, AS, NC, MK, FvdL, JR, TV, AB, PvdK, PvL, MvdB.

      Conflict of interest statement

      None of the authors had financial or personal relationships with people or organizations that could inappropriately influence the bias of the presented work.

      Role of the funding source

      The funding sources had no role in study design, in collection, analysis or interpretation of data, or in writing the manuscript and decision to submit the manuscript.

      Ethics approval

      All animal studies were according to the Dutch law and were approved by the local Animal Experimentation Committee (RU-DEC 2013-215).

      Acknowledgments

      The authors would like to thank Dr Peter Klaren (Dept. of Animal Physiology, Radboud University, Nijmegen) and Jordache Ramjith (Dept. of Health Evidence, Radboud university medical center) for their excellent statistical advise. This study was financially supported by the Dutch Arthritis Foundation (Reumafonds) grant number 12-2-405 and 16-1-402 .

      Appendix A. Supplementary data

      Figure thumbnail figs1

      References

        • Guermazi A.
        • Hayashi D.
        • Roemer F.W.
        • Zhu Y.
        • Niu J.
        • Crema M.D.
        • et al.
        Synovitis in knee osteoarthritis assessed by contrast-enhanced magnetic resonance imaging (MRI) is associated with radiographic tibiofemoral osteoarthritis and MRI-detected widespread cartilage damage: the MOST study.
        J Rheumatol. 2014; 41: 501-508
        • Rahmati M.
        • Mobasheri A.
        • Mozafari M.
        Inflammatory mediators in osteoarthritis: a critical review of the state-of-the-art, current prospects, and future challenges.
        Bone. 2016; 85: 81-90
        • van Dalen S.C.
        • Blom A.B.
        • Sloetjes A.W.
        • Helsen M.M.
        • Roth J.
        • Vogl T.
        • et al.
        Interleukin-1 is not involved in synovial inflammation and cartilage destruction in collagenase-induced osteoarthritis.
        Osteoarthritis Cartilage. 2017; 25: 385-396
        • Magnano M.D.
        • Chakravarty E.F.
        • Broudy C.
        • Chung L.
        • Kelman A.
        • Hillygus J.
        • et al.
        A pilot study of tumor necrosis factor inhibition in erosive/inflammatory osteoarthritis of the hands.
        J Rheumatol. 2007; 34: 1323-1327
        • van Lent P.L.
        • Blom A.B.
        • Schelbergen R.F.
        • Sloetjes A.
        • Lafeber F.P.
        • Lems W.F.
        • et al.
        Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
        Arthritis Rheum. 2012; 64: 1466-1476
        • Schelbergen R.F.
        • de Munter W.
        • van den Bosch M.H.
        • Lafeber F.P.
        • Sloetjes A.
        • Vogl T.
        • et al.
        Alarmins S100A8/S100A9 aggravate osteophyte formation in experimental osteoarthritis and predict osteophyte progression in early human symptomatic osteoarthritis.
        Ann Rheum Dis. 2016; 75: 218-225
        • Blom A.B.
        • van Lent P.L.
        • Libregts S.
        • Holthuysen A.E.
        • van der Kraan P.M.
        • van Rooijen N.
        • et al.
        Crucial role of macrophages in matrix metalloproteinase-mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3.
        Arthritis Rheum. 2007; 56: 147-157
        • Blom A.B.
        • van Lent P.L.
        • Holthuysen A.E.
        • van der Kraan P.M.
        • Roth J.
        • van Rooijen N.
        • et al.
        Synovial lining macrophages mediate osteophyte formation during experimental osteoarthritis.
        Osteoarthritis Cartilage. 2004; 12: 627-635
        • Benedyk M.
        • Sopalla C.
        • Nacken W.
        • Bode G.
        • Melkonyan H.
        • Banfi B.
        • et al.
        HaCaT keratinocytes overexpressing the S100 proteins S100A8 and S100A9 show increased NADPH oxidase and NF-kappaB activities.
        J Investig Dermatol. 2007; 127: 2001-2011
        • Bedard K.
        • Krause K.H.
        The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
        Physiol Rev. 2007; 87: 245-313
        • Chenevier-Gobeaux C.
        • Lemarechal H.
        • Bonnefont-Rousselot D.
        • Poiraudeau S.
        • Ekindjian O.G.
        • Borderie D.
        Superoxide production and NADPH oxidase expression in human rheumatoid synovial cells: regulation by interleukin-1beta and tumour necrosis factor-alpha.
        Inflamm Res. 2006; 55: 483-490
        • Rada B.
        • Leto T.L.
        Oxidative innate immune defenses by Nox/Duox family NADPH oxidases.
        Contrib Microbiol. 2008; 15: 164-187
        • Xu Q.
        • Choksi S.
        • Qu J.
        • Jang J.
        • Choe M.
        • Banfi B.
        • et al.
        NADPH oxidases are essential for macrophage differentiation.
        J Biol Chem. 2016; 291: 20030-20041
        • Doussiere J.
        • Bouzidi F.
        • Vignais P.V.
        The S100A8/A9 protein as a partner for the cytosolic factors of NADPH oxidase activation in neutrophils.
        Eur J Biochem. 2002; 269: 3246-3255
        • Berthier S.
        • Paclet M.H.
        • Lerouge S.
        • Roux F.
        • Vergnaud S.
        • Coleman A.W.
        • et al.
        Changing the conformation state of cytochrome b558 initiates NADPH oxidase activation: MRP8/MRP14 regulation.
        J Biol Chem. 2003; 278: 25499-25508
        • Schenten V.
        • Brechard S.
        • Plancon S.
        • Melchior C.
        • Frippiat J.P.
        • Tschirhart E.J.
        iPLA2, a novel determinant in Ca2+- and phosphorylation-dependent S100A8/A9 regulated NOX2 activity.
        Biochim Biophys Acta. 2010; 1803: 840-847
        • Chenevier-Gobeaux C.
        • Simonneau C.
        • Therond P.
        • Bonnefont-Rousselot D.
        • Poiraudeau S.
        • Ekindjian O.G.
        • et al.
        Implication of cytosolic phospholipase A2 (cPLA2) in the regulation of human synoviocyte NADPH oxidase (Nox2) activity.
        Life Sci. 2007; 81: 1050-1058
        • Clavijo-Cornejo D.
        • Martinez-Flores K.
        • Silva-Luna K.
        • Martinez-Nava G.A.
        • Fernandez-Torres J.
        • Zamudio-Cuevas Y.
        • et al.
        The overexpression of NALP3 inflammasome in knee osteoarthritis is associated with synovial membrane prolidase and NADPH oxidase 2.
        Oxid Med Cell Longev. 2016; 2016: 1472567
        • Fitowska A.
        • Ostalowska A.
        • Dobrakowski M.
        • Hapeta B.
        • Kasperczyk A.
        • Stoltny T.
        • et al.
        Protein metabolism in the synovial membrane in the hip osteoarthritis.
        Pol Orthop Traumatol. 2012; 77: 21-26
        • Fernandez-Moreno M.
        • Soto-Hermida A.
        • Pertega S.
        • Oreiro N.
        • Fernandez-Lopez C.
        • Rego-Perez I.
        • et al.
        Mitochondrial DNA (mtDNA) haplogroups and serum levels of anti-oxidant enzymes in patients with osteoarthritis.
        BMC Musculoskelet Disord. 2011; 12: 264
        • van Lent P.L.
        • Nabbe K.C.
        • Blom A.B.
        • Sloetjes A.
        • Holthuysen A.E.
        • Kolls J.
        • et al.
        NADPH-oxidase-driven oxygen radical production determines chondrocyte death and partly regulates metalloproteinase-mediated cartilage matrix degradation during interferon-gamma-stimulated immune complex arthritis.
        Arthritis Res Ther. 2005; 7: R885-R895
        • Henrotin Y.E.
        • Bruckner P.
        • Pujol J.P.
        The role of reactive oxygen species in homeostasis and degradation of cartilage.
        Osteoarthritis Cartilage. 2003; 11: 747-755
        • Huang C.K.
        • Zhan L.
        • Hannigan M.O.
        • Ai Y.
        • Leto T.L.
        P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+.
        J Leukoc Biol. 2000; 67: 210-215
        • Sareila O.
        • Jaakkola N.
        • Olofsson P.
        • Kelkka T.
        • Holmdahl R.
        Identification of a region in p47phox/NCF1 crucial for phagocytic NADPH oxidase (NOX2) activation.
        J Leukoc Biol. 2013; 93: 427-435
        • van der Kraan P.M.
        • Vitters E.L.
        • van Beuningen H.M.
        • van de Putte L.B.
        • van den Berg W.B.
        Degenerative knee joint lesions in mice after a single intra-articular collagenase injection. A new model of osteoarthritis.
        J Exp Pathol (Oxford). 1990; 71: 19-31
        • Fassl S.K.
        • Austermann J.
        • Papantonopoulou O.
        • Riemenschneider M.
        • Xue J.
        • Bertheloot D.
        • et al.
        Transcriptome assessment reveals a dominant role for TLR4 in the activation of human monocytes by the alarmin MRP8.
        J Immunol. 2015; 194: 575-583
        • Kruijsen M.W.
        • van den Berg W.B.
        • van de Putte L.B.
        • van den Broek W.J.
        Detection and quantification of experimental joint inflammation in mice by measurement of 99mTc-pertechnetate uptake.
        Agents Actions. 1981; 11: 640-642
        • Van Meurs J.B.
        • Van Lent P.L.
        • Joosten L.A.
        • Van der Kraan P.M.
        • Van den Berg W.B.
        Quantification of mRNA levels in joint capsule and articular cartilage of the murine knee joint by RT-PCR: kinetics of stromelysin and IL-1 mRNA levels during arthritis.
        Rheumatol Int. 1997; 16: 197-205
        • Glasson S.S.
        • Chambers M.G.
        • Van Den Berg W.B.
        • Little C.B.
        The OARSI histopathology initiative – recommendations for histological assessments of osteoarthritis in the mouse.
        Osteoarthritis Cartilage. 2010; 18: S17-S23
        • Blaney Davidson E.N.
        • Vitters E.L.
        • van Lent P.L.
        • van de Loo F.A.
        • van den Berg W.B.
        • van der Kraan P.M.
        Elevated extracellular matrix production and degradation upon bone morphogenetic protein-2 (BMP-2) stimulation point toward a role for BMP-2 in cartilage repair and remodeling.
        Arthritis Res Ther. 2007; 9: R102
        • Vogl T.
        • Tenbrock K.
        • Ludwig S.
        • Leukert N.
        • Ehrhardt C.
        • van Zoelen M.A.
        • et al.
        Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock.
        Nat Med. 2007; 13: 1042-1049
        • Cremers N.A.J.
        • van den Bosch M.H.J.
        • van Dalen S.
        • Di Ceglie I.
        • Ascone G.
        • van de Loo F.
        • et al.
        S100A8/A9 increases the mobilization of pro-inflammatory Ly6Chigh monocytes to the synovium during experimental osteoarthritis.
        Arthritis Res Ther. 2017; 19: 217
        • Wang G.G.
        • Calvo K.R.
        • Pasillas M.P.
        • Sykes D.B.
        • Hacker H.
        • Kamps M.P.
        Quantitative production of macrophages or neutrophils ex vivo using conditional Hoxb8.
        Nat Methods. 2006; 3: 287-293
        • Remst D.F.
        • Blom A.B.
        • Vitters E.L.
        • Bank R.A.
        • van den Berg W.B.
        • Blaney Davidson E.N.
        • et al.
        Gene expression analysis of murine and human osteoarthritis synovium reveals elevation of transforming growth factor beta-responsive genes in osteoarthritis-related fibrosis.
        Arthritis Rheumatol. 2014; 66: 647-656
        • Shadel G.S.
        • Horvath T.L.
        Mitochondrial ROS signaling in organismal homeostasis.
        Cell. 2015; 163: 560-569
        • Morita K.
        • Miyamoto T.
        • Fujita N.
        • Kubota Y.
        • Ito K.
        • Takubo K.
        • et al.
        Reactive oxygen species induce chondrocyte hypertrophy in endochondral ossification.
        J Exp Med. 2007; 204: 1613-1623
        • Reed K.N.
        • Wilson G.
        • Pearsall A.
        • Grishko V.I.
        The role of mitochondrial reactive oxygen species in cartilage matrix destruction.
        Mol Cell Biochem. 2014; 397: 195-201
        • Del Carlo M.
        • Schwartz D.
        • Erickson E.A.
        • Loeser R.F.
        Endogenous production of reactive oxygen species is required for stimulation of human articular chondrocyte matrix metalloproteinase production by fibronectin fragments.
        Free Radic Biol Med. 2007; 42: 1350-1358
        • Zelko I.N.
        • Mariani T.J.
        • Folz R.J.
        Superoxide dismutase multigene family: a comparison of the CuZn-SOD (SOD1), Mn-SOD (SOD2), and EC-SOD (SOD3) gene structures, evolution, and expression.
        Free Radic Biol Med. 2002; 33: 337-349
        • Benhar M.
        • Dalyot I.
        • Engelberg D.
        • Levitzki A.
        Enhanced ROS production in oncogenically transformed cells potentiates c-Jun N-terminal kinase and p38 mitogen-activated protein kinase activation and sensitization to genotoxic stress.
        Mol Cell Biol. 2001; 21: 6913-6926
        • Chaiswing L.
        • Oberley T.D.
        Extracellular/microenvironmental redox state.
        Antioxid Redox Signal. 2010; 13: 449-465
        • Gelderman K.A.
        • Hultqvist M.
        • Pizzolla A.
        • Zhao M.
        • Nandakumar K.S.
        • Mattsson R.
        • et al.
        Macrophages suppress T cell responses and arthritis development in mice by producing reactive oxygen species.
        J Clin Investig. 2007; 117: 3020-3028
        • van de Loo F.A.
        • Bennink M.B.
        • Arntz O.J.
        • Smeets R.L.
        • Lubberts E.
        • Joosten L.A.
        • et al.
        Deficiency of NADPH oxidase components p47phox and gp91phox caused granulomatous synovitis and increased connective tissue destruction in experimental arthritis models.
        Am J Pathol. 2003; 163: 1525-1537
        • Kassim S.Y.
        • Fu X.
        • Liles W.C.
        • Shapiro S.D.
        • Parks W.C.
        • Heinecke J.W.
        NADPH oxidase restrains the matrix metalloproteinase activity of macrophages.
        J Biol Chem. 2005; 280: 30201-30205
        • Kim J.S.
        • Yeo S.
        • Shin D.G.
        • Bae Y.S.
        • Lee J.J.
        • Chin B.R.
        • et al.
        Glycogen synthase kinase 3beta and beta-catenin pathway is involved in toll-like receptor 4-mediated NADPH oxidase 1 expression in macrophages.
        FEBS J. 2010; 277: 2830-2837
        • Geiszt M.
        • Lekstrom K.
        • Witta J.
        • Leto T.L.
        Proteins homologous to p47phox and p67phox support superoxide production by NAD(P)H oxidase 1 in colon epithelial cells.
        J Biol Chem. 2003; 278: 20006-20012
        • Berthier S.
        • Nguyen M.V.
        • Baillet A.
        • Hograindleur M.A.
        • Paclet M.H.
        • Polack B.
        • et al.
        Molecular interface of S100A8 with cytochrome b558 and NADPH oxidase activation.
        PLoS One. 2012; 7e40277
        • Trocme C.
        • Deffert C.
        • Cachat J.
        • Donati Y.
        • Tissot C.
        • Papacatzis S.
        • et al.
        Macrophage-specific NOX2 contributes to the development of lung emphysema through modulation of SIRT1/MMP-9 pathways.
        J Pathol. 2015; 235: 65-78
        • Ahn B.
        • Beharry A.W.
        • Frye G.S.
        • Judge A.R.
        • Ferreira L.F.
        NAD(P)H oxidase subunit p47phox is elevated, and p47phox knockout prevents diaphragm contractile dysfunction in heart failure.
        Am J Physiol Lung Cell Mol Physiol. 2015; 309: L497-L505
        • Olsson L.M.
        • Johansson A.C.
        • Gullstrand B.
        • Jonsen A.
        • Saevarsdottir S.
        • Ronnblom L.
        • et al.
        A single nucleotide polymorphism in the NCF1 gene leading to reduced oxidative burst is associated with systemic lupus erythematosus.
        Ann Rheum Dis. 2017; 76: 1607-1613
        • Kirk E.A.
        • Dinauer M.C.
        • Rosen H.
        • Chait A.
        • Heinecke J.W.
        • LeBoeuf R.C.
        Impaired superoxide production due to a deficiency in phagocyte NADPH oxidase fails to inhibit atherosclerosis in mice.
        Arterioscler Thromb Vasc Biol. 2000; 20: 1529-1535