Advertisement

Activation of innate immunity by 14-3-3 ε, a new potential alarmin in osteoarthritis

Open ArchivePublished:March 12, 2020DOI:https://doi.org/10.1016/j.joca.2020.03.002

      Summary

      Objective

      The innate immune system plays a central role in osteoarthritis (OA). We identified 14-3-3ε as a novel mediator that guides chondrocytes toward an inflammatory phenotype. 14-3-3ε shares common characteristics with alarmins. These endogenous molecules, released into extracellular media, are increasingly incriminated in sustaining OA inflammation. Alarmins bind mainly to toll-like receptor 2 (TLR2) and TLR4 receptors and polarize macrophages in the synovium. We investigated the effects of 14-3-3ε in joint cells and tissues and its interactions with TLRs to define it as a new alarmin involved in OA.

      Design

      Chondrocyte, synoviocyte and macrophage cultures from murine or OA human samples were treated with 14-3-3ε. To inhibit TLR2/4 in chondrocytes, blocking antibodies were used. Moreover, chondrocytes and bone marrow macrophage (BMM) cultures from knockout (KO) TLRs mice were stimulated with 14-3-3ε. Gene expression and release of inflammatory mediators [interleukin 6 (IL-6), monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor alpha (TNFα)] were evaluated via reverse transcription quantitative polymerase chain reaction (RT-qPCR) and ELISA.

      Results

      In vitro, 14-3-3ε induced gene expression and release of IL6 and MCP1 in the treated cells. The inflammatory effects of 14-3-3ε were significantly reduced following TLRs inhibition or in TLRs KO chondrocytes and BMM.

      Conclusions

      14-3-3ε is able to induce an inflammatory phenotype in synoviocytes, macrophages and chondrocytes in addition to polarizing macrophages. These effects seem to involve TLR2 or TLR4 to trigger innate immunity. Our results designate 14-3-3ε as a novel alarmin in OA and as a new target either for therapeutic and/or prognostic purposes.

      Keywords

      Abbreviations

      BMMs
      bone marrow derived macrophages
      BSA
      bovine serum albumin
      cDNA
      complementary deoxyribonucleic acid
      DAMP
      damage-associated molecular pattern
      DMEM
      Dulbecco's modified Eagle's medium
      EDTA
      Ethylenediaminetetraacetic acid
      ELISA
      enzyme linked immunosorbent assay
      FBS
      fetal bovine serum
      FLS
      fibroblast-like synoviocyte
      Glu
      glutamine
      H&E
      hematoxylin and eosin
      HPRT
      hypoxanthine-guanine phosphoribosyltransferase
      IL
      interleukin
      KO
      knockout
      LPS
      lipopolysaccharides
      mAB
      monoclonal antibody
      MCP1
      monocyte chemoattractant protein-1
      MMPs
      matrix metalloproteinases
      mRNA
      messenger ribonucleic acid
      OA
      osteoarthritis
      OxPAPC
      oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine
      PBS
      phosphate buffered saline
      PMB
      pattern-recognition receptors
      PRRs
      pattern-recognition receptors
      PS
      penicillin/streptomycin
      RA
      rheumatoid arthritis
      RNA
      ribonucleic acid
      RT-PCR
      reverse transcription polymerase chain reaction
      TLR
      toll like receptor
      TNFα
      tumor necrosis factor alpha
      WT
      wild-type

      Introduction

      Osteoarthritis (OA) is a highly complex and the most prevalent joint disorder with a total of 242 million people affected worldwide
      Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013.
      . Clinical symptoms include severe pain, joint stiffness and reduced function, which seriously decrease quality of life
      • Felson D.T.
      Osteoarthritis of the knee.
      . Unfortunately, only a few drugs are weakly effective for treating symptoms, and no disease-modifying osteoarthritis drugs (DMOADs) are available to date. Initially considered cartilage driven, OA is a much more complex disease with inflammatory mediators released by cartilage, bone and synovium
      • Hügle T.
      • Geurts J.
      What drives osteoarthritis?—Synovial versus subchondral bone pathology.
      . Increasing evidence suggests that inflammation is present in OA and has raised the possibility that inflammation and the innate immune system could be active players in the development and progression of OA
      • Berenbaum F.
      Osteoarthritis as an inflammatory disease (osteoarthritis is not osteoarthrosis!).
      ,
      • Herrero-Beaumont G.
      • Pérez-Baos S.
      • Sánchez-Pernaute O.
      • Roman-Blas J.A.
      • Lamuedra A.
      • Largo R.
      Targeting chronic innate inflammatory pathways, the main road to prevention of osteoarthritis progression.
      . Innate immune responses involve the activation of resident leukocytes such as macrophages, production of inflammatory mediators (cytokines, chemokines, and lipid metabolites), recruitment of neutrophils and monocytes/macrophages, and aims to eliminate invading microorganisms and injured tissues
      • Chen G.Y.
      • Nuñez G.
      Sterile inflammation: sensing and reacting to damage.
      . Histopathological studies have confirmed that immune cell infiltration is extremely common in OA histological specimens
      • Furman B.D.
      • Kimmerling K.A.
      • Zura R.D.
      • et al.
      Articular ankle fracture results in increased synovitis, synovial macrophage infiltration, and synovial fluid concentrations of inflammatory cytokines and chemokines.
      . The innate immune system participates in inflammation triggered by host molecules or fragments collectively called danger-associated molecular patterns (DAMPS) or alarmins
      • Chen G.Y.
      • Nuñez G.
      Sterile inflammation: sensing and reacting to damage.
      ,
      • Oppenheim J.J.
      • Yang D.
      Alarmins: chemotactic activators of immune responses.
      ,
      • Nefla M.
      • Holzinger D.
      • Berenbaum F.
      • Jacques C.
      The danger from within: alarmins in arthritis.
      .
      The term “alarmin” was proposed by Oppenheim and co-workers in 2005 to classify proteins that are rapidly released during infection or tissue damage, activating immune cells after interaction with their specific receptors
      • Oppenheim J.J.
      • Yang D.
      Alarmins: chemotactic activators of immune responses.
      . Alarmins are now considered to be markers of destructive processes in joints
      • Liu-Bryan R.
      • Terkeltaub R.
      Emerging regulators of the inflammatory process in osteoarthritis.
      . Because of their fast release as a result of cell stressor nonprogrammed cell death, alarmins are among the first factors to be secreted and, as such, act as first responders to stimuli. In addition to their role in disease initiation, alarmins also amplify and sustain inflammatory processes and, thus, play a notable role in the pathogenesis of inflammatory conditions
      • Manfredi A.A.
      • Capobianco A.
      • Bianchi M.E.
      • Rovere-Querini P.
      Regulation of dendritic- and T-cell fate by injury-associated endogenous signals.
      ,
      • Tamaki Y.
      • Takakubo Y.
      • Hirayama T.
      • et al.
      Expression of toll-like receptors and their signaling pathways in rheumatoid synovitis.
      . A number of alarmins have been detected at high levels in OA tissues and synovial fluid, including HMGB1, UA, ATP, thymosin β4 and various S100 proteins, enhancing catabolic processes and inflammatory responses that contribute to disease progression
      • Nefla M.
      • Holzinger D.
      • Berenbaum F.
      • Jacques C.
      The danger from within: alarmins in arthritis.
      ,
      • Ke X.
      • Jin G.
      • Yang Y.
      • et al.
      Synovial fluid HMGB-1 levels are associated with osteoarthritis severity.
      • Wang L.
      • Zhang H.
      • Shao L.
      • et al.
      S100A12 levels in synovial fluid may reflect clinical severity in patients with primary knee osteoarthritis.
      • Sunahori K.
      • Yamamura M.
      • Yamana J.
      • et al.
      The S100A8/A9 heterodimer amplifies proinflammatory cytokine production by macrophages via activation of nuclear factor kappa B and p38 mitogen-activated protein kinase in rheumatoid arthritis.
      .
      Synovitis, the inflammation of the synovium, can occur in early stages of OA
      • Sellam J.
      • Berenbaum F.
      The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis.
      . OA synovitis directly contributes to several clinical signs and symptoms including joint swelling and effusion, and reflects the structural progression of the disease
      • Sellam J.
      • Berenbaum F.
      The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis.
      ,
      • Loeser R.F.
      • Goldring S.R.
      • Scanzello C.R.
      • Goldring M.B.
      Osteoarthritis: a disease of the joint as an organ.
      . Moreover, synovitis is significantly associated with OA severity
      • Raghu H.
      • Lepus C.M.
      • Wang Q.
      • et al.
      CCL2/CCR2, but not CCL5/CCR5, mediates monocyte recruitment, inflammation and cartilage destruction in osteoarthritis.
      . Macrophages are the main immune cell type in the healthy synovium and are likely the front-line cells that sense joint damage. These cells also contribute to OA progression in response to alarmins by producing matrix metalloproteinases (MMPs) and cytokines
      • Bondeson J.
      • Blom A.B.
      • Wainwright S.
      • Hughes C.
      • Caterson B.
      • van den Berg WB.
      The role of synovial macrophages and macrophage-produced mediators in driving inflammatory and destructive responses in osteoarthritis.
      . The main morphological characteristic of synovitis is macrophage accumulation in the intimal lining
      • Bondeson J.
      • Blom A.B.
      • Wainwright S.
      • Hughes C.
      • Caterson B.
      • van den Berg WB.
      The role of synovial macrophages and macrophage-produced mediators in driving inflammatory and destructive responses in osteoarthritis.
      . Macrophages are characterized by heterogeneity and plasticity in response to stimuli from their microenvironment, leading to a continuum of phenotypes where M1/M2 are the 2 extremes
      • Sica A.
      • Mantovani A.
      Macrophage plasticity and polarization: in vivo veritas.
      . M1 macrophages are activated by interferon-γ and lipopolysaccharide (LPS) or tumor necrosis factor alpha (TNF-α), resulting in the secretion of large amounts of proinflammatory cytokines and mediators such as TNF-α, interleukin (IL)-1 and IL-6
      • Lopa S.
      • Leijs M.J.C.
      • Moretti M.
      • Lubberts E.
      • van Osch G.J.V.M.
      • Bastiaansen-Jenniskens Y.M.
      Arthritic and non-arthritic synovial fluids modulate IL10 and IL1RA gene expression in differentially activated primary human monocytes.
      . M2 macrophages have been further divided into specific subtypes and possess anti-inflammatory activity
      • Shapouri-Moghaddam A.
      • Mohammadian S.
      • Vazini H.
      • et al.
      Macrophage plasticity, polarization, and function in health and disease.
      . Studies have shown that macrophages accumulate and become polarized (M1 or M2) in the synovium during OA development
      • Chen Y.
      • Jiang W.
      • Yong H.
      • He M.
      • Yang Y.
      • Deng Z.
      • Li Y.
      Macrophages in osteoarthritis: pathophysiology and therapeutics.
      . The classification of macrophages into M1/M2 subtype is reductive. Recently, the heterogeneity of macrophage phenotypes in OA patients has been studied and revealed a more complex classification
      • Wood M.J.
      • Leckenby A.
      • Reynolds G.
      • et al.
      Macrophage proliferation distinguishes 2 subgroups of knee osteoarthritis patients.
      . However, the role of macrophages, their polarization in OA development and the underlying mechanisms are still unknown.
      Activated synovial cells (synoviocytes or macrophages) secrete several degenerative enzymes and inflammatory mediators, as well as alarmins
      • Wang L.
      • Zhang H.
      • Shao L.
      • et al.
      S100A12 levels in synovial fluid may reflect clinical severity in patients with primary knee osteoarthritis.
      ,
      • Manferdini C.
      • Paolella F.
      • Gabusi E.
      • et al.
      Adipose stromal cells mediated switching of the pro-inflammatory profile of M1-like macrophages is facilitated by PGE2: in vitro evaluation.
      . These alarmins activate pattern-recognition receptors (PRRs) including toll-like receptors (TLRs) in the OA-affected cartilage and synovium
      • van Lent P.L.E.M.
      • Blom A.B.
      • Schelbergen R.F.P.
      • et al.
      Active involvement of alarmins S100A8 and S100A9 in the regulation of synovial activation and joint destruction during mouse and human osteoarthritis.
      ,
      • Liu-Bryan R.
      • Terkeltaub R.
      The growing array of innate inflammatory ignition switches in osteoarthritis.
      , which in turn amplifies inflammation and degeneration of cartilage. TLRs are transmembrane receptors that display binding affinity for a variety of DAMPs. According to their ligands, location and signaling pathways, ten different genes can be distinguished in humans
      • Liu-Bryan R.
      Synovium and the innate inflammatory network in osteoarthritis progression.
      . TLRs activate signaling pathways that result in the production of cytokines, chemokines, and various inducible molecules associated with the immune response. TLR2 and TLR4 are overexpressed in OA cartilage, and their presence correlates with histopathological damage
      • Bobacz K.
      • Sunk I.G.
      • Hofstaetter J.G.
      • et al.
      Toll-like receptors and chondrocytes: the lipopolysaccharide-induced decrease in cartilage matrix synthesis is dependent on the presence of toll-like receptor 4 and antagonized by bone morphogenetic protein 7.
      ,
      • Kim H.A.
      • Cho M.-L.
      • Choi H.Y.
      • et al.
      The catabolic pathway mediated by Toll-like receptors in human osteoarthritic chondrocytes.
      .
      Our team identified 14-3-3ε as a novel soluble mediator that is critical in the communication between subchondral bone and cartilage in OA
      • Priam S.
      • Bougault C.
      • Houard X.
      • et al.
      Identification of soluble 14-3-3∊ as a novel subchondral bone mediator involved in cartilage degradation in osteoarthritis: soluble 14-3-3∊ in bone-cartilage communication.
      . The 14-3-3 proteins, a family of seven isoforms (β, ε, γ, η, θ, σ, ξ), are involved in a wide range of vital regulatory processes by binding to more than 200 intracellular proteins
      • Sun S.
      • Wong E.W.P.
      • Li M.W.M.
      • Lee W.M.
      • Cheng C.Y.
      14-3-3 and its binding partners are regulators of protein–protein interactions during spermatogenesis.
      . Under normal conditions, these proteins reside intracellularly; however, some reports indicate its presence extracellularly. Externalization of 14-3-3 appears to be mediated by a non-classical pathway, similar to IL1-β, due to the absence of a signal peptide sequence
      • Nickel W.
      The mystery of nonclassical protein secretion.
      . 14-3-3 proteins are released into the extracellular space through an exosomal pathway
      • Chavez-Muñoz C.
      • Morse J.
      • Kilani R.
      • Ghahary A.
      Primary human keratinocytes externalize stratifin protein via exosomes.
      . Extracellular 14-3-3s are now thought to play an important role in the pathogenesis of certain inflammatory conditions. In OA, our team showed that 14-3-3ε, released by osteoblasts in response to mechanical stress, skews chondrocytes toward a pro-catabolic phenotype by strongly inducing the expression of MMPs in an CD13/APN-dependent manner
      • Priam S.
      • Bougault C.
      • Houard X.
      • et al.
      Identification of soluble 14-3-3∊ as a novel subchondral bone mediator involved in cartilage degradation in osteoarthritis: soluble 14-3-3∊ in bone-cartilage communication.
      ,
      • Nefla M.
      • Sudre L.
      • Denat G.
      • et al.
      The pro-inflammatory cytokine 14-3-3 is a ligand of CD13 in cartilage.
      . 14-3-3ε seems to share common characteristics with other alarmins (such as ATP and thymosin β4) in inducing degradation of the cartilage matrix
      • Nefla M.
      • Holzinger D.
      • Berenbaum F.
      • Jacques C.
      The danger from within: alarmins in arthritis.
      . Extracellular 14-3-3ε proteins could, therefore, be classified as alarmins that are derived from activated or damaged osteoblasts.
      Our aim is to investigate the effects of 14-3-3ε on the different cell types of the joint and its role in the activation of innate immunity by studying interactions with its potential receptors TLR2 and TLR4.

      Methods

      Materials

      All reagents were purchased from Sigma–Aldrich (Lyon, France), unless stated otherwise. Fetal bovine serum (FBS) was obtained from Invitrogen (Cergy-Pontoise, France). Liberase TM and complete protease inhibitor mixture were from Roche Diagnostics (Meylan, France). Recombinant human 14-3-3ε was from Enzo Life Sciences. Anti-TLR2 antibody and OXPAPc (oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine) were from InvivoGen (Toulouse, France). Anti-TLR4 antibody was from Santa Cruz Technology (Heidelberg, Germany).

      Collection of human OA synovium

      Human OA knee explants were obtained from patients undergoing total knee arthroplasty due to OA at Saint-Antoine Hospital (Paris) or at the Maussins clinic (Paris) (BioJOINT, a biobank of OA human knee, legal authorization: CPP Paris Ile de France V, CNIL reference: MMS/HGT/AR177404). Informed consent for the use of tissue and clinical data was obtained from each patient before surgery. Experiments with human samples were approved by a French Institutional Review Board (Comité de Protection des Personnes, Paris Ile de France V and Commission Nationale de Informatique et des Libertés).

      Mice

      Mice on a C57BL/6 J background, 8–12 weeks old, were used in all experiments. Wild-type (WT) mice were purchased from Janvier Laboratories. The animal housing facility was granted approval (C 75-12-01) by the French Administration. All experiments were conducted according to the European Communities Council Directive (2010/63/UE) and approved by the Regional Animal Care and Use Committee (Ile-de-France, Paris, no5; agreement number 00917.02 and 4625).
      Tlr2−/−, and Tlr4−/− mice were kind gifts from Professor Shizuo Akira (Osaka University, Japan) and with the collaboration of Professor F. Pene (Institut Cochin, France). All knockout (KO) were maintained in the specific pathogen-free (SPF) animal facility of the Cochin Institute.

      Cell cultures

      Detailed protocols are described in supplemental data.

      Synovial explants

      For explants, synovium from OA patients was cut into small pieces aseptically and incubate in 24-well plate before treatment.

      Primary culture of synovial fibroblasts

      Primary cultures of fibroblast like synoviocytes (FLSs) are obtained after enzymatic digestion (collagenase and DNase) of human synovial membrane samples (Biojoint biobank). The digestion solution was then placed in culture flasks during an overnight incubation. The digestion solution was removed, and adherent cells were washed 3 times with phosphate buffered saline (PBS) and cultured in FLS growth medium at 37°C in a humidified atmosphere (5% CO2). After reaching their confluence, FLS were counted and cultured in 12-well plates for stimulations
      • Eymard F.
      • Pigenet A.
      • Citadelle D.
      • et al.
      Induction of an inflammatory and prodegradative phenotype in autologous fibroblast-like synoviocytes by the infrapatellar fat pad from patients with knee osteoarthritis.
      .

      Primary culture of murine articular chondrocytes

      Mouse primary chondrocytes were isolated from the articular cartilage of 5 to 6-day-old C57Bl6 mice from Janvier (St. Berthevin, France). All experiments were performed according to protocols approved by the French and European ethics committees (Comité Régional d'Ethique en Expérimentation Animale N°3 de la région Ile de France). Each littermate among the mice was used for one experiment. After 1 week, the cells were incubated in fasting medium for 24 h before treatment
      • Gosset M.
      • Berenbaum F.
      • Thirion S.
      • Jacques C.
      Primary culture and phenotyping of murine chondrocytes.
      .
      In addition, experiments were also performed using TLR2−/− and TLR4−/− mice.

      Primary culture of murine bone marrow derived macrophages cells

      Bone marrow derived macrophages phagocytic precursor cells were isolated from femurs and tibiae of WT and TLR2−/− and TLR4−/− mice. These precursors were differentiated into adherent mature bone marrow derived macrophages (BMM) for 7 days in complete medium containing 10 ng/ml of macrophage colony stimulating factor (PeproTech, Neuilly-sur-Seine, France).

      Culture of the THP-1 cell line

      Human monocytic THP1 cells (American Type Culture Collection, Rockville, MD, USA) were kind gifts from Professor Rouis (Sorbonne University, France). Thp1 cells were cultivated in flasks and then differenciated into macrophages with a 24 h treatment with Phorbol 12-Myristate 13-Acetate (50 nM) PMA (Sigma, Saint-Louis, USA). After 72 h, the macrophages were starved before treatment.

      Treatment by recombinant 14-3-3ε

      All cell cultures were stimulated with recombinant 14-3-3ε at 1 μg/ml for 24 h. Supernatants and total mRNA collected after cell lysis were harvested and stored at −80°C.
      For blocking antibody and pharmacological experiments, murine articular chondrocytes were pretreated for 20 min with increasing concentrations (1 and 5 μg/ml) of a mouse TLR2 or TLR4 antibody or oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (OxPAPC) at 0.3, 3 and 30 μg/ml before the treatment by recombinant 14-3-3ε.

      RNA extraction and quantitative real-time polymerase chain reaction (RT-PCR)

      Total RNA was extracted from murine chondrocytes using the ReliaPrep RNA Cell Miniprep System kit (Promega, Madison, WI, USA) from human synovial fibroblasts, BMM and THP1 by Trizol chloroform. Concentrations were determined by spectrophotometry (Eppendorf, Le Pecq, France). Reverse transcription was performed with 500 ng of total RNA with the Omniscript RT kit (Qiagen). mRNA levels were quantified with the Light Cycler LC480 (Roche Diagnostics, Indianapolis, IN, USA). PCR amplification conditions are described in supplemental data. Product formation was detected at 72°C in the fluorescein isothiocyanate channel. The mRNA levels were normalized to those of murine hypoxanthine-guanine phosphoribosyltransferase (HPRT) or Human 18S. Specific primer sequences are presented in Table S1.

      Protein secretion quantification by ELISA

      Total mouse and human IL6, monocyte chemoattractant protein-1 (MCP1), TNFα and MMP-3 secretion were assayed in cell-free supernatants using an enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems and Abbexa) according to the manufacturer's instructions. Concentrations were analyzed in duplicate at serial dilutions and determined by comparison against a standard curve.

      Endotoxin tests

      Protocol of the experiments is described in supplemental data.

      Statistical analysis

      The choice of the number of experiments was established by power analysis tests and previous and published work from our laboratory. The small number of experiments used for each part of the work can be considered as a limitation. All data were showed as mean values ± s.e.m. In Fig. 5, the stimulated condition (14-3-3ε) was normalized to 1 in order to study the inhibition rates and data were showed as mean values with 95% confidence intervals (CI). Statistical analyses were performed with the Mann Whitney test to compare mean values between 2 groups or by the Wilcoxon test when analyses were based on patient paired-matched samples (Fig. 1, Fig. 2). One-way analysis of variance (ANOVA) and two-way ANOVA with the Bonferroni multiple comparisons post-test were used to compare mean values between more than 2 groups using GraphPad Prism software (GraphPad Software, San Diego, CA). P < 0.05 was considered statistically significant. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns: not significant.
      Fig. 1
      Fig. 1Stimulation of IL6 and MCP1 protein release by synovium explants from OA patients with 14-3-3ε. Synovium explants from OA patients were incubated with medium in the presence or absence of 14-3-3ε (1 μg/ml) for 24 h. A, B: Protein levels of IL6 (A) (n = 5) and MCP1 (B) (n = 5) released by human synovium explants treated with the control or 14-3-3ε were measured using the ELISA. Bars show the mean ± s.e.m. (Wilcoxon test) ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001, ns, not significant.
      Fig. 2
      Fig. 2Stimulation of IL6 and MCP1 expression and protein release in response to 14-3-3ε in primary cultures of human FLSs. Synovium explants from OA patients were digested and seeded in 6-well culture plates until confluence. The primary cultures were then stimulated with the control medium in the presence or absence of recombinant 14-3-3ε (10 μg/ml) for 24 h. A,B: Total RNA was extracted, and mRNA levels of IL6 (A) (n = 6) and MCP1 (B) (n = 6) were determined by qRT-qPCR. Protein levels of IL6 (C) (n = 6) and MCP1 (D) (n = 6) in cell supernatants were measured by ELISA. Bars show the mean ± s.e.m. (Wilcoxon test) ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns, not significant.
      Fig. 3
      Fig. 3Stimulation of IL6, MCP1 and CD38 mRNA expression and IL6, MCP1, and TNFα protein release in human macrophages (derived from the THP1 cell line). Stimulation of macrophages (THP1 cells) was performed using control medium or 14-3-3ε (1 μg/ml) treatment. A,B,C: Total RNA was extracted, and mRNA levels of IL6 (A) (n = 5), MCP1 (B) n = 5), and CD38 (C) (n = 5) were determined by qRT-PCR. D, E, F: Protein levels of IL6 (D) (n = 5), MCP1 (E) (n = 5) and TNFα (F) (n = 5) in cell supernatants were measured by ELISA. Bars show the mean ± s.e.m. (Mann Whitney test) ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns, not significant.
      Fig. 4
      Fig. 4Stimulation of IL6 and MCP1 protein release by BMMs from WT, TLR2 or TL4 KO mice in response to 14-3-3ε treatment. Stimulation of BMMs were performed using control medium or 14-3-3ε (1 μg/ml) treatment. Protein levels of IL6 (A) (n = 4) and MCP1 (B) (n = 4) in cell supernatants were measured by ELISA. Bars show the mean ± s.e.m. (Two way ANOVA with a Bonferroni post-test) ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns, not significant.
      Fig. 5
      Fig. 5Involvement of TLR2 and TLR4 in 14-3-3ε-induced MMP-3 (degradative enzyme) and IL-6 (pro-inflammatory cytokine) release by murine articular chondrocytes. Murine articular chondrocytes were treated with TLR2, TLR4 blocking antibodies or the pharmacologic inhibitor OxPAPC for 15 min and then stimulated with recombinant 14-3-3ε (1 μg/ml) for 24 h. (A, B, C) Total RNA was extracted, and mRNA levels of MMP-3 were determined by qRT-PCR to examine the inhibitory effects of anti-TLR4 (A) (n = 5), anti-TLR2 (B) (n = 5), and OxPAPC (C) (n = 5). (D, E, F) Protein levels of MMP-3 in cell supernatants were measured by ELISA to examine the inhibitor effects of anti-TLR4 (D) (n = 5), anti-TLR2 (E) (n = 5), and OxPAPC (F) (n = 5). 14-3-3ε-stimulated cells released an average of 350 ng/ml of MMP-3. (G, H, I) Total RNA was extracted, and mRNA levels of IL6 were determined by qRT-PCR to examine the inhibitory effects of anti-TLR4 (G) (n = 5), anti-TLR2 (H) (n = 5), and OxPAPC (I) (n = 5). (J, K, L) Protein levels of IL6 in cell supernatants were measured by ELISA to examine the inhibitory effects of anti-TLR4 (J) (n = 5), anti-TLR2 (K) (n = 5), and OxPAPC (L) (n = 5). 14-3-3ε-stimulated cells released an average of 2 ng/ml of MMP-3. Bars show the mean values with 95% confidence intervals (One-way ANOVA with a Bonferroni post-test) ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns, not significant.

      Results

      Stimulation of synovium explants from OA patients by 14-3-3ε elicits the release of pro-inflammatory factors

      To verify whether 14-3-3ε is able to induce synovium inflammation in humans, we used synovium explants from OA patients to mimic the pathophysiological environment as closely as possible. Stimulation of these human synovium explants with recombinant 14-3-3ε (1 μg/ml) induced the secretion of MCP1 and IL6 protein [Fig. 1(A) and (B)]. Indeed, mean difference of IL6 protein release showed a 5.7-fold increase between control and 14-3-3ε stimulation (0.6, 95% CI [0.3;0.9] vs 3.4, 95% CI [2.7;4.1] respectively) [Fig. 1(A)] and mean difference of MCP1 protein release showed a 9.5-fold increase between control and 14-3-3ε stimulation (4.8, 95% CI [0.9;8.6] vs 45.6, 95% CI [9.9;81.3] respectively) [Fig. 1(B)].

      14-3-3ε elicits a pro-inflammatory phenotype in FLSs

      To more precisely study the impact of 14-3-3ε on the two main cell types residing in the synovium, FLS and macrophages, we stimulated primary cultures of FLS from OA patients with 14-3-3ε recombinant protein (1 μg/ml) for 24 h. The levels of IL-6 and MCP1 in controls samples are below the detection threshold. We found that stimulated synoviocytes had increased mRNA expression and secretion of both IL6 and MCP1 [Fig. 2(A)–(D)]. Mean difference of IL6 mRNA expression showed an increase of 7.5 fold between control and 14-3-3ε stimulation (0.5, 95% CI [0;1.0] vs 3.7, 95% CI [2.1;5.3] respectively) and MCP1 mRNA expression showed an increase of 4.4 fold mean difference between control and 14-3-3ε (0.6, 95% CI [0.1;1.1] vs 2.6, 95% CI [0.9;4.3] respectively) [Fig. 2(A) and (B)]. Mean difference of protein secretion of IL6 and MCP1 showed a fold increase of 3.7 (5.5, 95% CI [−0.6;11.6] vs 20.54 95% CI [−0.7;41.7]) and 31.8 (1.7, 95% CI [1.0;2.4] vs 54.2, 95% CI [32.7;75.7]) between control and 14-3-3ε stimulation respectively [Fig. 2(C) and (D)].

      14-3-3ε skews macrophages toward a pro-inflammatory phenotype involving TLRs signaling

      Stimulation of human macrophages derived from the THP1 cell line with 14-3-3ε induced mRNA expression and secretion of both IL6 and MCP1 [Fig. 3(A), (B), (D), (E)]. Mean difference of mRNA expression levels of IL6 and MCP1 showed a fold increase of 7.2 (0.1, 95% CI [0;0.2] vs 0.7, 95% CI [0.5;0.8]) and 3.1 (0.2, 95% CI [0;0.4] vs 0.7, 95% CI [0.5;0.8]) between control and 14-3-3ε stimulation respectively. Similarly, the release of IL6, MCP1 and TNFα protein was increased in the supernatants of macrophages stimulated with 14-3-3ε, showing a fold mean difference of 17.8 (0.03, 95% CI [0;0.05] vs 6.2, 95% CI [1.9;10.4]), 18.9 (23.9, 95% CI [23;24.8] vs 453, 95% CI [379.8;525.6]) and 15.1 (0.04, 95% CI [0;0.07] vs 0.6, 95% CI [0.2;1.1]) between control and 14-3-3ε stimulation respectively [Fig. 3(D)–(F)]. Moreover, macrophages subjected to 14-3-3ε stimulation displayed increased mRNA expression of CD38, another marker associated with the M1 pro-inflammatory phenotype, with a 6.6-fold mean difference between control and 14-3-3ε stimulation (0.1, 95% CI [0;0.2] vs 0.9, 95% CI [0.7;1.1]) [Fig. 3(C)].
      To assess the implications of TLR2 and TLR4 in the cellular response to 14-3-3ε, we used primary cultures of bone marrow derived macrophages (BMMs) from TLR2 or TLR4 KO mice and stimulated them with recombinant 14-3-3ε. Untreated WT, TLR2 KO and TLR4 KO BMM showed no protein release of IL6 and MCP1, whereas IL-6 and MCP1 release were increased by WT BMM cells stimulated with 14-3-3ε. IL-6 protein release was significantly decreased in the supernatants of TLR2 and TLR4 KO BMM stimulated with 14-3-3ε compared to the WT BMM stimulated with 14-3-3ε (64% inhibition for TLR2, WT 2.5, 95% CI [0.9;4.1] vs TLR2 KO 0.9, 95% CI [0.4;1.3] and 84% inhibition for TLR4; TLR4 KO 0.4, 95% CI [0.2;0.6]) [Fig. 4(A)]. MCP1 showed the same tendency (22% inhibition for TLR2 WT 2.8, 95% CI [−2.0;7.7] vs TLR2 KO 2.2, 95% CI [−1.5;6.0] and 68% inhibition for TLR4; TLR4 KO 0.9, 95% CI [−2.1;3.8]) [Fig. 4(B)].

      14-3-3ε elicits a catabolic and inflammatory phenotype in murine articular chondrocytes involving TLR2 and TLR4

      Murine chondrocytes were sensitive to stimulation with 14-3-3ε recombinant protein and showed increased mRNA expression and secretion of pro-catabolic (MMP3) and pro-inflammatory (IL6) factors.
      To confirm the involvement of TLR2 and TLR4 in 14-3-3ε signaling, murine articular chondrocytes were pre-treated with specific TLR2, TLR4 blocking antibody or the pharmacologic inhibitor (OxPAPC) inhibiting both receptors followed by 14-3-3ε stimulation. MMP3 mRNA expression and protein secretion induced by 14-3-3ε were significantly and dose-dependently reduced by TLR4 blocking antibody treatment (MMP3 mRNA expression inhibition fold: 78% (95% CI [−0.1;0.6]) (1 μg/ml) and 77% (95% CI [−0.02;0.5]) (5 μg/ml) [Fig. 5(A)]; secretion: 69% (95% CI [−0.18;0.81]) (1 μg/ml) and 84% (95% CI [−0.09;0.42]) (5 μg/ml) [Fig. 5(D)]. Anti-TLR2 blocking antibody reproduced the same pattern (mRNA expression: 79% (95% CI [−0.01;0.43]) (1 μg/ml) and 83% (95% CI [0.04;0.30]) (5 μg/ml) [Fig. 5(B)]; secretion 73% (95% CI [−0.07;0.61]) (1 μg/ml) and 74% (95% CI [0.04;0.48]) (5 μg/ml) [Fig. 5(E)]. Furthermore, the pharmacologic inhibitor OxPAPC, inhibiting both receptors, markedly decreased MMP3 mRNA expression and protein release in a dose-dependent manner (mRNA expression: from 71% (95% CI [0.04;0.54]) (0.3 μg/ml) to 98% (95% CI [−0.01;0.04]) (30 μg/ml) [Fig. 5(C)]; secretion: from 64% (95% CI [0.14;0.58]) (0.3 μg/ml) to 96% (95% CI [−0.01;0.08]) (30 μg/ml) [Fig. 5(F)].
      Inhibition of TLR2 and/or TLR4 also impacted the mRNA and protein expression of the pro-inflammatory cytokine IL6 by murine chondrocytes. (IL6 mRNA expression: anti-TLR4 treatment: 73% (95% CI [−0.15;0.68]) (1 μg/ml) and 71% (95% CI [−0.10;0.69]) (5 μg/ml) [Fig. 5(G)]; secretion: 79% (95% CI [−0.12;0.55]) (1 μg/ml) and 81% (95% CI [−0.10;0.48]) (5 μg/ml) [Fig. 5(J)]; anti-TLR2 treatment: IL6 mRNA expression 53% (95% CI [−0.02;0.97]) (1 μg/ml) and 51% (95% CI [0.04;0.94]) (5 μg/ml) [Fig. 5(H)]; secretion 69% (95% CI [−0.03;0.64]) (1 μg/ml) and 78% (95% CI [0.01;0.43]) (5 μg/ml) [Fig. 5(K)]. Inhibiting both receptors simultaneously with OxPAPC resulted in decreases IL6 expression and secretion (IL6 mRNA expression: from 69% (95% CI [0.01;0.60]) (0.3 μg/ml) to 92% (95% CI [0;0.15]) (30 μg/ml) [Fig. 5(I)]; secretion: from 69% (95% CI [0.05;0.58]) (0.3 μg/ml) to 96% (95% CI [−0.02;0.09]) (30 μg/ml) [Fig. 5(L)]. Stimulations of chondrocytes with the different treatments showed high variability between cultures.
      To confirm the results obtained in KO BMM, we also stimulated TLR2 or TLR4 KO murine articular chondrocytes with 14-3-3ε. TLR2 and TLR4 KO murine chondrocytes exhibited decreased mRNA expression and protein release of IL6 but also mRNA expression of MMP3 and MMP13 compared to WT chondrocytes (Fig. 6). IL6 mRNA expression was significantly decreased in TLR2 and TLR4 KO chondrocytes treated with 14-3-3ε compared to the WT chondrocytes (75% for TLR2; WT 65, 95% CI [−53.7;183.8] vs TLR2 KO 16.5, 95% CI [3.6;29.4] and 97% for TLR4; TLR4 KO 1.8, 95% CI [1.0;2.5]) [Fig. 6(B)], and its release was greatly attenuated in TLR2 (77% for TLR2 (WT 21.8, 95% CI [−1.5;45.1] vs TLR2 KO 5.0, 95% CI [2.0;8.0]) and TLR4 (by 97% for TLR4, TLR4 KO 0.7, 95% CI [0.5;0.9]) KO chondrocytes in response to 14-3-3ε stimulation respectively [Fig. 6(A)]. Similarly, MMP3 and MMP13 mRNA expression levels were also decreased in TLR2 and TLR4 KO chondrocytes (For MMP3: 60% for TLR2, WT 988, 95% CI [−588;2566] vs TLR2 KO 394.1, 95%CI [242.3;545.8] and 94% for TLR4; TLR4 KO 61.4, 95% CI [46.9;75.9]; For MMP13: 22% for TLR2, WT 0.5, 95% CI [0;1] vs TLR2 KO 66.4, 95%CI [48.6;84.2] and 77% for TLR4; TLR4 KO 20.2, 95% CI [10.4;30.1]) [Fig. 6(C) and (D)]. Taken together, these results demonstrate the involvement of both TLR2 and TLR4 receptors in 14-3-3ε signaling.
      Fig. 6
      Fig. 6Stimulation of IL6 mRNA expression and Il6, MMP3 and MMP13 protein release by murine articular chondrocytes from WT, TLR2 or TL4 KO mice in response to 14-3-3ε treatment. Stimulation of murine chondrocytes was performed using control medium or 14-3-3ε (1 μg/ml) treatment. Total RNA was extracted, and mRNA levels of IL6 (B) (n = 4), MMP3 (C) (n = 4) and MMP13 (D) (n = 4) were determined by qRT-PCR. Protein levels of IL6 (A) (n = 4) in cell supernatants was measured by ELISA. Bars show the mean ± s.e.m. (Two way ANOVA post-test)∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ns, not significant.

      Discussion

      We have recently identified 14-3-3ε as a new soluble mediator involved in deleterious biochemical interactions between bone and cartilage
      • Priam S.
      • Bougault C.
      • Houard X.
      • et al.
      Identification of soluble 14-3-3∊ as a novel subchondral bone mediator involved in cartilage degradation in osteoarthritis: soluble 14-3-3∊ in bone-cartilage communication.
      . In the present study, we showed that 14-3-3ε is a new alarmin and can act particularly in the synovium during OA pathogenesis. Its role in the activation of innate immunity leading to synovitis could be due to interactions with its potential receptors, TLR2 and TLR4.
      It is now established that up to 50% of OA patients have synovitis, as demonstrated by magnetic resonance imaging, ultrasonography and arthroscopy
      • Bijlsma J.W.
      • Berenbaum F.
      • Lafeber F.P.
      Osteoarthritis: an update with relevance for clinical practice.
      . Based on these results and to study the involvement of 14-3-3ε in the synovium of the OA joint, we analyzed the effects of 14-3-3ε on whole synovial tissue explants from OA patients by measuring the release of pro-inflammatory factors. We found that IL6 and MCP1 release in culture media was significantly increased after 14-3-3ε stimulation compared with the control explants. In these complete synovial explants from OA patients, the pathophysiological environment was conserved. There was obvious heterogeneity in the degree of inflammation of the synovium between patient samples, which could be due to the different pathological mechanisms leading to OA among patients
      • Van Spil W.E.
      • Kubassova O.
      • Boesen M.
      • Bay-Jensen A.-C.
      • Mobasheri A.
      Osteoarthritis phenotypes and novel therapeutic targets.
      ,
      • Blom A.B.
      • van Lent P.L.
      • Libregts S.
      • et al.
      Crucial role of macrophages in matrix metalloproteinase–mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3.
      . Although fewer synovial macrophages are present in OA compared with RA, they are crucial for the production of proinflammatory cytokines such as IL6
      • Bondeson J.
      • Blom A.B.
      • Wainwright S.
      • Hughes C.
      • Caterson B.
      • van den Berg WB.
      The role of synovial macrophages and macrophage-produced mediators in driving inflammatory and destructive responses in osteoarthritis.
      . Previous studies have shown that selective depletion of synovial macrophages during experimental OA largely reduces cartilage damage and osteophyte formation, which are 2 major hallmarks of OA
      • Blom A.B.
      • van Lent P.L.
      • Libregts S.
      • et al.
      Crucial role of macrophages in matrix metalloproteinase–mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3.
      . Thus, we would like to separately analyze the role in synovitis of 2 main cell types residing in the synovium: FLSs and macrophages. We found that 14-3-3ε was able to increase the mRNA expression and protein secretion of IL6 (7.5 and 4.3-fold increase respectively) and MCP1 (4.4 and 32-fold increase) in FLS. These cells were able to respond to 14-3-3ε resulting in an inflammatory phenotype. Moreover, Thp1 cells, a human monocyte cell line, were used to address whether 14-3-3ε could polarize these cells toward a pro-inflammatory macrophage phenotype. In our current study, we showed that 14-3-3ε increased the mRNA expression and protein secretion of IL6 (7.2 and 17.8-fold increase respectively) and MCP1 (3.1 and 19-fold increase). In addition to the experiments on Thp1, we also performed stimulations of primary cultures of murine macrophages (BMMs) by 14-3-3ε and studied the expression of different markers. mRNA expressions of pro-inflammatory mediators such as iNOS, IL-1β and TNFα were increased whereas mRNA expression of anti-inflammatory marker (EGR2) was decreased in BMM (Fig. S1 Supplemental data). This protein subsequently appears to polarize macrophage toward the M1 phenotype. In the case of knee OA, a study analyzing M1 macrophages and M2 macrophages in synovial fluid in normal vs OA knees found a higher ratio of M1/M2 in OA vs normal knees, and the ratio was significantly correlated to the Kellgren–Lawrence grade
      • Liu B.
      • Zhang M.
      • Zhao J.
      • Zheng M.
      • Yang H.
      Imbalance of M1/M2 macrophages is linked to severity level of knee osteoarthritis.
      . These results suggest that macrophage polarization may indeed play a role in the control and even progression of OA disease. However, it should be noted that the classification of macrophages into M1/M2 subtype is reductive. This ability of macrophages to modify their phenotype in response to external signals gives rise to a broad spectrum of possibilities depending on their interactions
      • Wood M.J.
      • Leckenby A.
      • Reynolds G.
      • et al.
      Macrophage proliferation distinguishes 2 subgroups of knee osteoarthritis patients.
      . Our study indicates that 14-3-3ε can induce a proinflammatory environment in the OA synovium through stimulation of macrophages which possibly contributes to the joint destruction that occurs during OA. Similar results have been obtained previously for the alarmin S100A9 in the synovium
      • van den Bosch M.H.
      • Blom A.B.
      • Schelbergen R.F.
      • et al.
      Alarmin S100A9 induces proinflammatory and catabolic effects predominantly in the M1 macrophages of human osteoarthritic synovium.
      .
      Interestingly, high levels of many alarmins have been described in the synovial fluid of OA patients
      • Ke X.
      • Jin G.
      • Yang Y.
      • et al.
      Synovial fluid HMGB-1 levels are associated with osteoarthritis severity.
      . Numerous studies have shown that these DAMPs stimulate synovial cell proliferation, influence hypertrophic chondrocyte differentiation and induce inflammatory and pro-catabolic events in vitro, and they promote synovitis and cartilage degradation in vivo in murine OA models
      • Schelbergen R.F.P.
      • Blom A.B.
      • van den Bosch M.H.J.
      • et al.
      Alarmins S100A8 and S100A9 elicit a catabolic effect in human osteoarthritic chondrocytes that is dependent on Toll-like receptor 4.
      . We hypothesized that 14-3-3ε is a new alarmin involved in OA. Our results showed that this protein was able to lead to an inflammatory and catabolic response in joint similarly to other alarmins. Moreover, OA is mainly linked to activation of innate immunity by the binding of damage-associated molecular patterns (DAMPs) to so-called PRRs
      • Kawai T.
      • Akira S.
      The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors.
      . Of central importance in the PRR family are the TLRs
      • Kawai T.
      • Akira S.
      The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors.
      . TLR2 and TLR4 are overexpressed in OA cartilage, and their presence correlates with histopathological damage
      • Kim H.A.
      • Cho M.-L.
      • Choi H.Y.
      • et al.
      The catabolic pathway mediated by Toll-like receptors in human osteoarthritic chondrocytes.
      . Blockade of TLR signaling -as shown in TLR2/TLR4 and in MyD88 knockout mice-downregulates cartilage catabolic response in vitro, and it can protect animals from experimental OA
      • Nasi S.
      • Ea H.-K.
      • Chobaz V.
      • et al.
      Dispensable role of myeloid differentiation primary response gene 88 (MyD88) and MyD88-dependent toll-like receptors (TLRs) in a murine model of osteoarthritis.
      . Moreover, synovial fluid proteins from patients with OA activate macrophages via TLR2/TLR4 receptors, in turn translocating NF-κB to the nucleus
      • Liu-Bryan R.
      • Terkeltaub R.
      Emerging regulators of the inflammatory process in osteoarthritis.
      . Synovial fibroblasts are sensitive to both mechanical alterations and DAMPs due to the expression of different TLRs on the cell membrane, resulting in increased synthesis of pro-inflammatory mediators
      • Liu-Bryan R.
      • Terkeltaub R.
      The growing array of innate inflammatory ignition switches in osteoarthritis.
      . In particular, TLR-2 and TLR-4 are used by many alarmins
      • Park J.S.
      • Svetkauskaite D.
      • He Q.
      • et al.
      Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein.
      .
      To examine whether 14-3-3ε response was driven by TLR, we used TLR2 and TLR4 KO mice to study their potential involvement in macrophage and chondrocyte responses. Our results showed that these two receptors were involved in the inflammatory and catabolic phenotype after stimulation with 14-3-3ε, with TLR4 showing predominant involvement. To further investigate the involvement of TLR2 and TLR4, we used blocking antibodies against them and a pharmacological inhibitor (OXPAPC) that is able to inhibit both TLR2 and TLR4. Our results validated the results obtained in the KO mice and confirmed that 14-3-3ε could elicit a catabolic and inflammatory phenotype in murine articular chondrocytes and a pro-inflammatory phenotype in BMM macrophages. Interestingly, a TLR4 monoclonal antibody has recently been demonstrated to have an adequate safety profile, and a phase II clinical trial in patients with RA has been launched
      • Monnet E.
      • Lapeyre G.
      • van Poelgeest E.
      • et al.
      Evidence of NI-0101 pharmacological activity, an anti-TLR4 antibody, in a randomized phase I dose escalation study in healthy volunteers receiving LPS.
      .
      In the present study, we used recombinant 14-3-3ε protein produced in Escherichia coli, similarly to many commercially available recombinant proteins. Although this expression system has many advantages, including rapid expression, high yields, ease of culture and low cost
      • Demain A.L.
      • Vaishnav P.
      Production of recombinant proteins by microbes and higher organisms.
      , the proteins recovered may be contaminated with endotoxin, a highly complex LPS constitutive of the outer membrane of most gram-negative bacteria
      • Heumann D.
      • Roger T.
      Initial responses to endotoxins and Gram-negative bacteria.
      . LPS is recognized by a receptor complex composed of TLR4, CD14 and MD-2
      • Triantafilou M.
      • Triantafilou K.
      Lipopolysaccharide recognition: CD14, TLRs and the LPS-activation cluster.
      ,
      • Pålsson-McDermott E.M.
      • O'Neill L.A.J.
      Signal transduction by the lipopolysaccharide receptor, Toll-like receptor-4.
      . Consequently, using recombinant 14-3-3ε in this study, we wanted to be sure that the effects of 14-3-3ε on joint tissues were due to the protein itself and not to endotoxin contamination. Low levels of endotoxins measured by a LAL kit (14-3-3ε contained less than 0.15 ng/ml of LPS, data not shown) and no significant inhibition by the polymyxin B (PMB) on 14-3-3ε chondrocyte stimulation (Fig. S2 supplemental data) and 14-3-3ε macrophages stimulation (data not shown) confirmed the proper effect of 14-3-3ε recombinant protein. Moreover, in our previous study
      • Priam S.
      • Bougault C.
      • Houard X.
      • et al.
      Identification of soluble 14-3-3∊ as a novel subchondral bone mediator involved in cartilage degradation in osteoarthritis: soluble 14-3-3∊ in bone-cartilage communication.
      , we demonstrated that immunodepletion and blocking of 14-3-3ε in conditioned media of compressed osteoblasts inhibited its catabolic effect on chondrocytes confirming that 14-3-3ε itself is involved in the establishment of a procatabolic phenotype in chondrocytes.
      Thus, although we cannot rule out that a small component of the effects observed herein were due to endotoxin contamination, we are confident that the cellular responses resulted from the activity of 14-3-3ε protein itself.
      Taken together, our results designate 14-3-3ε as a novel alarmin for further exploration in OA for either therapeutic or prognostic purposes.

      Contributions

      • -
        Conception and design: MM, FB, CJ, FP
      • -
        Collection and assembly of data: MM, LS, AR, AP, MN, CR
      • -
        Analysis and interpretation of the data: MM, CJ, FB
      • -
        Drafting of the article: MM, CJ, FB,
      • -
        Critical revision of the article for important intellectual content: MM, CJ, FB, FP, GA-L,
      • -
        Final approval of the article: MM, LS, MN, FP, CR, AR, AP, GA-L, FB, CJ

      Conflict of interest

      No potential conflicts of interest relevant to this article were reported. All authors disclose any financial and personal relationships with other people or organizations that could potentially and inappropriately influence their work and conclusions.

      Role of the funding source

      This work was supported by grants from Institut National de la Santé et de la Recherche Médicale ( INSERM) , Sorbonne University, French Society of Rheumatology (Société Française de rhumatologie) and foundation Arthritis – Courtin. M.M. was supported by a doctoral fellowship from Ministère de l'Enseignement Supérieur et de la Recherche .

      Acknowledgments

      The authors thank the Department of Orthopaedic Surgery and Traumatology of Saint-Antoine Hospital for providing human OA tissues. The authors thank Dr. F. Pène (Institut Cochin, INSERM U1016, CNRS UMR8104, Paris, France; Université Paris Descartes, Sorbonne Paris Cité, Paris France) for his kind gift of TLR2 and TLR4 KO mice. The authors thank T. Ledent, L. Dinard, A. Guyomard, T. Coulais, and Q. Pointout (animal housing facility, INSERM, Saint-Antoine Research Center, Sorbonne University, Paris) for their excellent work.

      Appendix A. Supplementary data

      The following are the supplementary data to this article:

      References

      1. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013.
        Lancet. 2015; 386: 743-800https://doi.org/10.1016/S0140-6736(15)60692-4
        • Felson D.T.
        Osteoarthritis of the knee.
        N Engl J Med. 2006; 354: 841-848https://doi.org/10.1056/NEJMcp051726
        • Hügle T.
        • Geurts J.
        What drives osteoarthritis?—Synovial versus subchondral bone pathology.
        Rheumatology (Oxford). 2017; 56: 1461-1471https://doi.org/10.1093/rheumatology/kew389
        • Berenbaum F.
        Osteoarthritis as an inflammatory disease (osteoarthritis is not osteoarthrosis!).
        Osteoarthritis Cartilage. 2013; 21: 16-21https://doi.org/10.1016/j.joca.2012.11.012
        • Herrero-Beaumont G.
        • Pérez-Baos S.
        • Sánchez-Pernaute O.
        • Roman-Blas J.A.
        • Lamuedra A.
        • Largo R.
        Targeting chronic innate inflammatory pathways, the main road to prevention of osteoarthritis progression.
        Biochem Pharmacol. 2019; 165: 24-32https://doi.org/10.1016/j.bcp.2019.02.030
        • Chen G.Y.
        • Nuñez G.
        Sterile inflammation: sensing and reacting to damage.
        Nat Rev Immunol. 2010; 10: 826-837https://doi.org/10.1038/nri2873
        • Furman B.D.
        • Kimmerling K.A.
        • Zura R.D.
        • et al.
        Articular ankle fracture results in increased synovitis, synovial macrophage infiltration, and synovial fluid concentrations of inflammatory cytokines and chemokines.
        Arthritis Rheumatol. 2015; 67: 1234-1239https://doi.org/10.1002/art.39064
        • Oppenheim J.J.
        • Yang D.
        Alarmins: chemotactic activators of immune responses.
        Curr Opin Immunol. 2005; 17: 359-365https://doi.org/10.1016/j.coi.2005.06.002
        • Nefla M.
        • Holzinger D.
        • Berenbaum F.
        • Jacques C.
        The danger from within: alarmins in arthritis.
        Nat Rev Rheumatol. 2016; 12: 669-683https://doi.org/10.1038/nrrheum.2016.162
        • Liu-Bryan R.
        • Terkeltaub R.
        Emerging regulators of the inflammatory process in osteoarthritis.
        Nat Rev Rheumatol. 2015; 11: 35-44https://doi.org/10.1038/nrrheum.2014.162
        • Manfredi A.A.
        • Capobianco A.
        • Bianchi M.E.
        • Rovere-Querini P.
        Regulation of dendritic- and T-cell fate by injury-associated endogenous signals.
        Crit Rev Immunol. 2009; 29: 69-86
        • Tamaki Y.
        • Takakubo Y.
        • Hirayama T.
        • et al.
        Expression of toll-like receptors and their signaling pathways in rheumatoid synovitis.
        J Rheumatol. 2011; 38: 810-820https://doi.org/10.3899/jrheum.100732
        • Ke X.
        • Jin G.
        • Yang Y.
        • et al.
        Synovial fluid HMGB-1 levels are associated with osteoarthritis severity.
        Clin Lab. 2015; 61: 809-818
        • Wang L.
        • Zhang H.
        • Shao L.
        • et al.
        S100A12 levels in synovial fluid may reflect clinical severity in patients with primary knee osteoarthritis.
        Biomarkers. 2013; 18: 216-220https://doi.org/10.3109/1354750X.2013.766262
        • Sunahori K.
        • Yamamura M.
        • Yamana J.
        • et al.
        The S100A8/A9 heterodimer amplifies proinflammatory cytokine production by macrophages via activation of nuclear factor kappa B and p38 mitogen-activated protein kinase in rheumatoid arthritis.
        Arthritis Res Ther. 2006; 8: R69https://doi.org/10.1186/ar1939
        • Sellam J.
        • Berenbaum F.
        The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis.
        Nat Rev Rheumatol. 2010; 6: 625-635https://doi.org/10.1038/nrrheum.2010.159
        • Loeser R.F.
        • Goldring S.R.
        • Scanzello C.R.
        • Goldring M.B.
        Osteoarthritis: a disease of the joint as an organ.
        Arthritis Rheum. 2012; 64: 1697-1707https://doi.org/10.1002/art.34453
        • Raghu H.
        • Lepus C.M.
        • Wang Q.
        • et al.
        CCL2/CCR2, but not CCL5/CCR5, mediates monocyte recruitment, inflammation and cartilage destruction in osteoarthritis.
        Ann Rheum Dis. 2017; 76: 914-922https://doi.org/10.1136/annrheumdis-2016-210426
        • Bondeson J.
        • Blom A.B.
        • Wainwright S.
        • Hughes C.
        • Caterson B.
        • van den Berg WB.
        The role of synovial macrophages and macrophage-produced mediators in driving inflammatory and destructive responses in osteoarthritis.
        Arthritis Rheum. 2010; 62: 647-657https://doi.org/10.1002/art.27290
        • Sica A.
        • Mantovani A.
        Macrophage plasticity and polarization: in vivo veritas.
        J Clin Investig. 2012; 122: 787-795https://doi.org/10.1172/JCI59643
        • Lopa S.
        • Leijs M.J.C.
        • Moretti M.
        • Lubberts E.
        • van Osch G.J.V.M.
        • Bastiaansen-Jenniskens Y.M.
        Arthritic and non-arthritic synovial fluids modulate IL10 and IL1RA gene expression in differentially activated primary human monocytes.
        Osteoarthritis Cartilage. 2015; 23: 1853-1857https://doi.org/10.1016/j.joca.2015.06.003
        • Shapouri-Moghaddam A.
        • Mohammadian S.
        • Vazini H.
        • et al.
        Macrophage plasticity, polarization, and function in health and disease.
        J Cell Physiol. 2018; 233: 6425-6440https://doi.org/10.1002/jcp.26429
        • Chen Y.
        • Jiang W.
        • Yong H.
        • He M.
        • Yang Y.
        • Deng Z.
        • Li Y.
        Macrophages in osteoarthritis: pathophysiology and therapeutics.
        Am J Transl Res. 2020 Jan 15; 12: 261-268https://doi.org/10.1002/jcp.26429
        • Wood M.J.
        • Leckenby A.
        • Reynolds G.
        • et al.
        Macrophage proliferation distinguishes 2 subgroups of knee osteoarthritis patients.
        JCI Insight. 2019 Jan 24; 4 (pii: 125325)https://doi.org/10.1172/jci.insight.125325
        • Manferdini C.
        • Paolella F.
        • Gabusi E.
        • et al.
        Adipose stromal cells mediated switching of the pro-inflammatory profile of M1-like macrophages is facilitated by PGE2: in vitro evaluation.
        Osteoarthritis Cartilage. 2017; 25: 1161-1171https://doi.org/10.1016/j.joca.2017.01.011
        • van Lent P.L.E.M.
        • Blom A.B.
        • Schelbergen R.F.P.
        • 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: 1466-1476https://doi.org/10.1002/art.34315
        • Liu-Bryan R.
        • Terkeltaub R.
        The growing array of innate inflammatory ignition switches in osteoarthritis.
        Arthritis Rheum. 2012; 64: 2055-2058https://doi.org/10.1002/art.34492
        • Liu-Bryan R.
        Synovium and the innate inflammatory network in osteoarthritis progression.
        Curr Rheumatol Rep. 2013; 15: 323https://doi.org/10.1007/s11926-013-0323-5
        • Bobacz K.
        • Sunk I.G.
        • Hofstaetter J.G.
        • et al.
        Toll-like receptors and chondrocytes: the lipopolysaccharide-induced decrease in cartilage matrix synthesis is dependent on the presence of toll-like receptor 4 and antagonized by bone morphogenetic protein 7.
        Arthritis Rheum. 2007; 56: 1880-1893https://doi.org/10.1002/art.22637
        • Kim H.A.
        • Cho M.-L.
        • Choi H.Y.
        • et al.
        The catabolic pathway mediated by Toll-like receptors in human osteoarthritic chondrocytes.
        Arthritis Rheum. 2006; 54: 2152-2163https://doi.org/10.1002/art.21951
        • Priam S.
        • Bougault C.
        • Houard X.
        • et al.
        Identification of soluble 14-3-3∊ as a novel subchondral bone mediator involved in cartilage degradation in osteoarthritis: soluble 14-3-3∊ in bone-cartilage communication.
        Arthritis Rheum. 2013; 65: 1831-1842https://doi.org/10.1002/art.37951
        • Sun S.
        • Wong E.W.P.
        • Li M.W.M.
        • Lee W.M.
        • Cheng C.Y.
        14-3-3 and its binding partners are regulators of protein–protein interactions during spermatogenesis.
        J Endocrinol. 2009; 202: 327-336https://doi.org/10.1677/JOE-09-0041
        • Nickel W.
        The mystery of nonclassical protein secretion.
        Eur J Biochem. 2003; 270: 2109-2119https://doi.org/10.1046/j.1432-1033.2003.03577.x
        • Chavez-Muñoz C.
        • Morse J.
        • Kilani R.
        • Ghahary A.
        Primary human keratinocytes externalize stratifin protein via exosomes.
        J Cell Biochem. 2008; 104: 2165-2173https://doi.org/10.1002/jcb.21774
        • Nefla M.
        • Sudre L.
        • Denat G.
        • et al.
        The pro-inflammatory cytokine 14-3-3 is a ligand of CD13 in cartilage.
        J Cell Sci. 2015; 128: 3250-3262https://doi.org/10.1242/jcs.169573
        • Eymard F.
        • Pigenet A.
        • Citadelle D.
        • et al.
        Induction of an inflammatory and prodegradative phenotype in autologous fibroblast-like synoviocytes by the infrapatellar fat pad from patients with knee osteoarthritis.
        Arthritis Rheumatol. 2014 Aug; 66: 2165-2174https://doi.org/10.1002/art.38657
        • Gosset M.
        • Berenbaum F.
        • Thirion S.
        • Jacques C.
        Primary culture and phenotyping of murine chondrocytes.
        Nat Protoc. 2008; 3: 1253-1260https://doi.org/10.1038/nprot.2008.95
        • Bijlsma J.W.
        • Berenbaum F.
        • Lafeber F.P.
        Osteoarthritis: an update with relevance for clinical practice.
        Lancet. 2011; 377: 2115-2126https://doi.org/10.1016/S0140-6736(11)60243-2
        • Van Spil W.E.
        • Kubassova O.
        • Boesen M.
        • Bay-Jensen A.-C.
        • Mobasheri A.
        Osteoarthritis phenotypes and novel therapeutic targets.
        Biochem Pharmacol. 2019; 165: 41-48https://doi.org/10.1016/j.bcp.2019.02.037
        • Blom A.B.
        • van Lent P.L.
        • Libregts S.
        • et al.
        Crucial role of macrophages in matrix metalloproteinase–mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3.
        Arthritis Rheum. 2007; 56: 147-157https://doi.org/10.1002/art.22337
        • Liu B.
        • Zhang M.
        • Zhao J.
        • Zheng M.
        • Yang H.
        Imbalance of M1/M2 macrophages is linked to severity level of knee osteoarthritis.
        Exp Ther Med. 2018; 16: 5009-5014https://doi.org/10.3892/etm.2018.6852
        • van den Bosch M.H.
        • Blom A.B.
        • Schelbergen R.F.
        • et al.
        Alarmin S100A9 induces proinflammatory and catabolic effects predominantly in the M1 macrophages of human osteoarthritic synovium.
        J Rheumatol. 2016; 43: 1874-1884https://doi.org/10.3899/jrheum.160270
        • Schelbergen R.F.P.
        • Blom A.B.
        • van den Bosch M.H.J.
        • et al.
        Alarmins S100A8 and S100A9 elicit a catabolic effect in human osteoarthritic chondrocytes that is dependent on Toll-like receptor 4.
        Arthritis Rheum. 2012; 64: 1477-1487https://doi.org/10.1002/art.3349543
        • Kawai T.
        • Akira S.
        The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors.
        Nat Immunol. 2010; 11: 373-384https://doi.org/10.1038/ni.1863
        • Nasi S.
        • Ea H.-K.
        • Chobaz V.
        • et al.
        Dispensable role of myeloid differentiation primary response gene 88 (MyD88) and MyD88-dependent toll-like receptors (TLRs) in a murine model of osteoarthritis.
        Joint Bone Spine. 2014; 81: 320-324https://doi.org/10.1016/j.jbspin.2014.01.018
        • Park J.S.
        • Svetkauskaite D.
        • He Q.
        • et al.
        Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein.
        J Biol Chem. 2004; 279: 7370-7377https://doi.org/10.1074/jbc.M306793200
        • Monnet E.
        • Lapeyre G.
        • van Poelgeest E.
        • et al.
        Evidence of NI-0101 pharmacological activity, an anti-TLR4 antibody, in a randomized phase I dose escalation study in healthy volunteers receiving LPS.
        Clin Pharmacol Ther. 2017; 101: 200-208https://doi.org/10.1002/cpt.522
        • Demain A.L.
        • Vaishnav P.
        Production of recombinant proteins by microbes and higher organisms.
        Biotechnol Adv. 2009; 27: 297-306https://doi.org/10.1016/j.biotechadv.2009.01.008
        • Heumann D.
        • Roger T.
        Initial responses to endotoxins and Gram-negative bacteria.
        Clin Chim Acta. 2002; 323: 59-72https://doi.org/10.1016/S0009-8981(02)00180-8
        • Triantafilou M.
        • Triantafilou K.
        Lipopolysaccharide recognition: CD14, TLRs and the LPS-activation cluster.
        Trends Immunol. 2002; 23: 301-304https://doi.org/10.1016/S1471-4906(02)02233-0
        • Pålsson-McDermott E.M.
        • O'Neill L.A.J.
        Signal transduction by the lipopolysaccharide receptor, Toll-like receptor-4.
        Immunology. 2004; 113: 153-162https://doi.org/10.1111/j.1365-2567.2004.01976.x