TY - JOUR
AU - Mastbergen, Simon, C
AB - Abstract Objectives The crosstalk between the immune and nervous system in the regulation of OA pain is increasingly becoming evident. GM-CSF signals in both systems and might be a new treatment target to control OA pain. Anti GM-CSF treatment has analgesic effects in OA without affecting synovitis scores, suggesting that treatment effects are not caused by local anti-inflammatory effects. We aimed to evaluate whether expression of GM-CSF and its receptor GM-CSFrα in synovial tissue is linked to synovial inflammation and/or knee pain in knee OA patients. Methods Cartilage and synovial tissue of knee OA patients (n = 20) was collected during total knee replacement. Cartilage damage was evaluated by histology and ex vivo matrix proteoglycan turnover. Synovial inflammation was evaluated by histology and ex vivo synovial production of TNF-α, (PGE2) and nitric oxide (NO). Numbers of synovial tissue cells expressing GM-CSF or GM-CSFrα were determined by immunohistochemistry. Pain was evaluated using WOMAC questionnaire and visual analogue scale (VAS) knee pain. Results Collected cartilage and synovial tissue had a typical OA phenotype with enhanced cartilage damage and synovial inflammation. GM-CSF and GM-CSFrα expressing cells in the synovial sublining correlated negatively with knee pain. Cartilage damage and synovial inflammation did not correlate with knee pain. Conclusion Unanticipated, the negative correlation between synovial tissue cells expressing GM-CSF(r) and OA knee pain suggests that local presence of these molecules does not promote pain, and that the role of GM-CSFr in OA pain is unrelated to local inflammation. Trial registration ToetsingOnline.nl NL18274.101.07. knee osteoarthritis, pain, GM-CSF, synovium Rheumatology key messages Synovial GM-CSF(r) expression is negatively correlated with clinical osteoarthritis pain. Mechanisms of action of GM-CSF in osteoarthritis pain are likely not related to local inflammation. Previously found inhibition of osteoarthritis pain by GM-CSF(r) might be due to systemic mechanisms. Introduction Increasing evidence suggests a bidirectional crosstalk between the immune system and nervous system as a regulator of OA pain, broadening the search for possible new treatment targets in OA pain. In a recently published review, Conaghan and colleagues focused on inflammation and peripheral nociception and suggested GM-CSF as one of the possible targets [1]. GM-CSF is a haematopoietic growth factor [2], but also has an important role in maintaining autoimmunity (e.g. after chemotherapy) and in inducing inflammation by activating various cell types [3]. GM-CSF also has direct effects on neurons, increasing mechanical hypersensitivity, as well as indirect effects via cyclooxygenase or chemokine (C-C motif) ligand 17 (CCL17)- and eicosanoid-dependent pathways [4]. GM-CSF is expressed in synovial membranes of RA as well as OA patients [5]. Several clinical trials with drugs targeting GM-CSF or its receptor (GM-CSFr) show beneficial effects in the treatment of RA. Mavrilimumab, a fully human monoclonal antibody (mAb) targeting GM-CSFr improved the Disease Activity Score 28 (DAS28) score after 12 weeks, and resulted in a greater percentage of ACR 20 responders at week 24 in a phase IIb randomized placebo-controlled trial in patients with moderate-severe RA [6]. Two human mAbs blocking GM-CSF, MOR103 and namilumab, also had beneficial effects and were well tolerated in RA [7, 8]. In OA, GM-CSF plays a role in disease and pain progression as well. In GM-CSF-deficient mice the development of joint damage and OA pain in a collagenase-induced instability mouse model of OA was significantly lower compared with wild-type mice [9]. In a phase IIa study in patients with inflammatory hand OA, anti-GM-CSF mAb induced larger reductions in hand pain, compared with placebo, although little differences were found in numbers of tender and swollen joints or MRI inflammation-related endpoints [10]. The discrepancy between improvement in clinical pain scores and unaltered synovitis supports the idea that OA pain is not just a result of local inflammation but that other mechanisms of action, possible systemic or related to the nervous system, play a role as well [1]. To investigate whether GM-CSF and its receptor are related to knee pain in knee OA patients and, if so, the role of synovial inflammation therein, we evaluate the correlation between the number of cells expressing GM-CSF or GM-CSFrα in the synovial tissue, inflammation, and clinical knee pain in OA patients. Methods Patient selection This study is a post hoc evaluation of data of the control group of a study on the disease modifying effects of celecoxib in end stage OA, registered under number NL18274.101.07 [11]. Twenty patients form the non-treated trial arm that did not receive medication in the year before total knee replacement (TKR) were included for the present evaluation and additional analyses. The trial was performed in accordance with the Declaration of Helsinki and approved by the Dutch Central Committee on research Involving Human Subjects (CCMO). All patients signed informed consent before inclusion. Clinical outcome parameters for pain Pain was evaluated using the WOMAC questionnaire pain subscale (0 = maximum pain, 100 = no pain) and the visual analogue scale (VAS) for pain (mean of VAS night, VAS morning and VAS general; 0 = no pain, 100 = maximum pain). Both were collected twice (6 weeks before surgery and just before surgery) and the average value of both time points was used. Immunohistochemistry of GM-CSF(r) At TKR surgery, synovial tissue samples were collected and kept in phosphate buffered saline for transport. Two samples of each donor were used for immunohistochemistry. Samples were embedded in Tissue Tek (Sakura; Torrance, CA, USA) and stored at −80°C. For the present study, 6-µm slices were cut with a cryotome and incubated with primary antibodies for GM-CSF (LS-C41956; LSBio, Seattle, WA, USA) or GM-CSFrα (CD116, alpha chain of GM-CSF r; LS-C40980, LSBio) for 1 h at room temperature. Subsequently, sections of both stainings were incubated with a secondary antibody (BrightVision Poly-AP-Anti Ms/Rb/Ra) for 30 min. Sections were counterstained with Alkaline Phosphate substrate and scored microscopically (magnification 40×) by one observer. The number of positive cells per mm2 synovial sublining was determined for GM-CSF as well as GM-CSFrα. Fig. 1 shows an example of the immunohistochemistry staining. Fig. 1 Open in new tabDownload slide Expression of GM-CSF and GM-CSFr in synovial sublining Expression of GM-CSF (A and B) and GM-CSFr (C and D) in synovial sublining. Magnification 4× (A and C) and 10× (B and D). Arrows indicate cells positive for target. Fig. 1 Open in new tabDownload slide Expression of GM-CSF and GM-CSFr in synovial sublining Expression of GM-CSF (A and B) and GM-CSFr (C and D) in synovial sublining. Magnification 4× (A and C) and 10× (B and D). Arrows indicate cells positive for target. Evaluation of synovial inflammation Two representative synovial tissue samples of each donor were embedded in paraffin, cut in 5-µm sections and stained with Haematoxylin-Eosin to evaluate synovial inflammation using modified Goldenberg and Cohen score (range 0–10). Two additional samples of each patient were cultured for three days in DMEM supplemented with glutamine (2 mM), penicillin (100 IU/ml), streptomycin sulphate (100 µg/ml), ascorbic acid (0.85 mM) and 10% human male AB+ serum to determine ex vivo produced inflammatory mediators. Supernatants were collected and centrifuged (1300 g). To determine PGE2EIA was used (514010-96w; Caymann Chemical, Ann Arbor, MI, USA). TNF-α was determined using ELISA (CHC1753; Invitrogen/Thermofisher; Carlsbad, CA, USA). Nitric oxide (NO) was determined using the standard Griess reaction. Detailed procedures published previously [11]. Evaluation of structural cartilage damage Cartilage samples were collected simultaneously with the synovial tissue. Two representative samples of each donor were embedded in paraffin, cut in 5-µm sections and stained with Safranin-O-fast-green to evaluate microscopic cartilage damage using modified Mankin score (range 0–11). Eight cartilage samples were cultured ex vivo at 37°C and proteoglycan (PG) synthesis and release were determined as measures of chondrocyte activity as described before [12]. In short, after one h of pre-culture, 20µCI Na235SO4 in DMEM was added to each sample. After 4 h of labelling, samples were washed 3× 45 min in medium and subsequently cultured for another 3 days in order to determine ex vivo release. Next, samples were digested with papain for 2 h at 65°C. GAGs were precipitated by addition of cetylpyridium chloride. 35 SO42--labelled GAGs were measured by liquid scintillation on a TRI-CARB 2800TR. Values were normalized to specific activity of the pulse medium, pulse time and wet weight of the explants and expressed as nmoles of sulphate incorporated per h per gram wet weight of the cartilage (nmol/h/g). For the total release of PGs, Alcian Blue staining of the medium was quantified photometrically with chondroitin sulphate as a reference. The total amount of GAGs released (blue staining) is expressed as a percentage of the original tissue content (% GAG release). Statistical analysis WOMAC scores were converted to 100-WOMAC. By doing that, for WOMAC as well as for VAS pain a higher value on the x-axis reflects more pain. All statistical tests were performed using IBM SPSS Statistics version 25.0.0.2. Shapiro-Wilk tests were used to see whether parameters were normally distributed. In case not normally distributed, a log transformation was performed (NO and PG release). Pearson correlation coefficient was determined to evaluate correlations between all parameters. For PGE2 log transformation did not result in a normal distribution, therefore a Spearman correlation coefficient was determined instead. P-values <0.05 were considered statistically significant. Results Patient characteristics Patients represented a typical OA population, comparable to the original cohort [11]; age 69.3 (5.7) years, 13/20 female (65%), BMI 26.7 (4.7) and Kellgren & Lawrence (K&L) grade ≥2 (Supplementary Table S1, available at Rheumatology online). GM-CSF and GM-CSFrα expressing cells in synovial sublining correlated with clinical OA pain and inflammation Correlations between GM-CSF and GM-CSFrα expressing cells in synovial sublining and clinical pain are shown in Fig. 2. Number of GM-CSF expressing cells in the synovial sublining was negatively correlated with WOMAC knee pain; r = −0.578, P <0.01, and VAS knee pain; r = −0.494, P = 0.027. For GM-CSFrα expressing cells in the synovial sublining, negative correlations were found as well; r = −0.481, P = 0.032 for WOMAC pain and r = −0.387, P = 0.091 for VAS pain. Ex vivo production of the inflammatory synovial mediators TNF-α, PGE2 and NO was negatively correlated with GM-CSF expressing cells, but not GM-CSFrα expressing cells (Supplementary Table S2, available at Rheumatology online). Fig. 2 Open in new tabDownload slide Correlation between GM-CSF and GM-CSFr expression and WOMAC and VAS knee pain Negative correlations were found between number of cells expressing GM-CSF or GM-CSFr in the synovial sublining and clinical knee pain measured by WOMAC and VAS pain scores (mean of VAS night, VAS morning and VAS general). Data is represented as individual values (dots) and a correlation (continuous line). VAS: visual analogue scale Fig. 2 Open in new tabDownload slide Correlation between GM-CSF and GM-CSFr expression and WOMAC and VAS knee pain Negative correlations were found between number of cells expressing GM-CSF or GM-CSFr in the synovial sublining and clinical knee pain measured by WOMAC and VAS pain scores (mean of VAS night, VAS morning and VAS general). Data is represented as individual values (dots) and a correlation (continuous line). VAS: visual analogue scale No relations between number of cells expressing GM-CSF or its receptor and cartilage damage or chondrocyte activity were detected (Supplementary Table S2, available at Rheumatology online). There was no statistically significant correlation between inflammation and clinical pain or cartilage damage and chondrocyte activity (Supplementary Table S3, available at Rheumatology online). Discussion In this study we identified that the number of cells expressing GM-CSF and GM-CSFrα in the sublining of the synovial tissue was negatively correlated with OA pain. It is known that in human [13] and equine [14] OA joints, nerve fibre density is decreased in superficial layers of synovial tissue, closer to the synovial surface. Therefore, in this study, only the deeper layer (sublining) of the synovial tissue was examined. Negative correlations between synovial GM-CSF and GM-CSFrα expression and WOMAC and VAS knee pain were found. The correlations between GM-CSF and pain might be skewed by one data point. Sensitivity analysis leaving out this value resulted in negative correlation with P-values of 0.293 and 0.210 for VAS and WOMAC pain, respectively. This would mean a tendency towards the negative correlation (or even absence of a correlation) between GM-CSF and pain, but not a positive correlation. We therefore considered our general conclusion valid. Taken the limited number of patients available into account, it was chosen to include all values of all subjects in the analyses. The negative correlation suggests that local GM-CSF at the synovium likely does not promote pain in advanced OA. Intriguingly, systemic neutralization of GM-CSF with neutralizing antibodies reduces pain and disease progression in an experimental model of OA [9] and inflammatory hand OA in humans [10], indicating that GM-CSF promotes pain. Possibly the stage of OA affects whether GM-CSF contributes to OA pain locally and/or systemically. Structural changes (e.g. cartilage degeneration and/or osteophyte formation) are more prominent in advanced OA, while in early OA more features of synovial inflammation can be present [15]. Otherwise, differences in tissue-dependent contributions of GM-CSF may be at the core of these differences in the contribution of GM-CSF in OA pain. The absence of a correlation between inflammatory markers and OA pain suggests that local synovitis is not the only mechanism contributing to OA pain. The lack of a strong correlation between radiographic damage and knee pain [16] also suggests involvement of other pain mechanisms than that caused directly by damage of the joint. Indeed, intrathecally administered mAb against GM-CSF reduced pain in a mouse model for neuropathic pain, indicating that GM-CSF in the central nervous system contributes to pain [17]. Neuropathic pain mechanisms and the central nervous system are also involved in OA pain [18, 19] indicating OA pain is not only the result of sensitization/activation of peripheral sensory neurons by damage/inflammation. For example, significant relationships were found between peripheral sensitization of local nociceptors (knee) and central sensitization of dorsal horn neurons (leg, arm) detected by lower pressure pain thresholds, and pain intensity in OA patients. Moreover, OA patients have enhanced evoked temporal summation and reduced descending inhibition pointing to aberrant central pain processing [18]. Future research needs to demonstrate whether the analgesic effects of peripheral anti GM-CSF treatments are caused through an effect at the central nervous system, locally at the joint or otherwise. In our study population, central sensitization most likely plays a dominant role in pain as patients had advanced knee OA. This might explain the absence of a positive correlation between local factors including GM-CSF(r) and knee pain parameters found in this study. In the EULAR recommendations for pain management in inflammatory arthritis and OA, a biopsychosocial model of pain is recommended [20]. It is important to realize that OA pain never only encompasses biological factors, whether local or systemic, but also psychological and social factors. OA is a very heterogeneous disease, consisting of multiple OA phenotypes (subtypes) with distinct underlying pathobiological and pain mechanisms and structural and functional consequences. The correlations between GM-CSF and GM-CSFrα and local and systemic pain parameters most likely vary between these OA phenotypes. Future research should focus on selecting OA patients with specific OA phenotypes, to see whether the role of GM-CSF differs between those groups and evaluate if one of these phenotypes might be more susceptible for beneficial effects of treatment targeting GM-CSF or its receptor. In conclusion, the negative correlation between local GM-CSF and GM-CSFrα expression and knee OA pain suggests that these molecules are not to be locally involved in OA knee pain development and supports a more systemic role of GM-CSF in OA pain, possibly related to the nervous system. Acknowledgements F.P.J.G.L., S.C.M. and N.E. were supported by the Dutch Arthritis Society (ReumaNederland) project LLP-9. We thank Yvonne. E. Karens for her help with the immunohistochemistry. All authors contributed to the set-up of the study, gathering data and scientifically contributed to writing the manuscript. Funding: This work was supported by the Dutch Arthritis Society (ReumaNederland) project 10–412. Disclosure statement: The authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Conaghan PG , Cook AD, Hamilton JA, Tak PP. Therapeutic options for targeting inflammatory osteoarthritis pain . Nat Rev Rheumatol 2019 ; 15 : 355 – 63 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Francisco-Cruz A , Aguilar-Santelises M, Ramos-Espinosa O et al. Granulocyte-macrophage colony-stimulating factor: not just another haematopoietic growth factor . Med Oncol 2014 ; 31 : 774 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Hamilton JA. Colony-stimulating factors in inflammation and autoimmunity . Nat Rev Immunol 2008 ; 8 : 533 – 44 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Cook AD , Christensen AD, Tewari D, McMahon SB, Hamilton JA. Immune cytokines and their receptors in inflammatory pain . Trends Immunol 2018 ; 39 : 240 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Farahat MN , Yanni G, Poston R, Panayi GS. Cytokine expression in synovial membranes of patients with rheumatoid arthritis and osteoarthritis . Ann Rheum Dis 1993 ; 52 : 870 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Burmester GR , McInnes IB, Kremer J et al. A randomised phase IIb study of mavrilimumab, a novel GM-CSF receptor alpha monoclonal antibody, in the treatment of rheumatoid arthritis . Ann Rheum Dis 2017 ; 76 : 1020 – 30 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Behrens F , Tak PP, Ostergaard M et al. MOR103, a human monoclonal antibody to granulocyte-macrophage colony-stimulating factor, in the treatment of patients with moderate rheumatoid arthritis: results of a phase Ib/IIa randomised, double-blind, placebo-controlled, dose-escalation trial . Ann Rheum Dis 2015 ; 74 : 1058 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Taylor PC , Saurigny D, Vencovsky J et al. Efficacy and safety of namilumab, a human monoclonal antibody against granulocyte-macrophage colony-stimulating factor (GM-CSF) ligand in patients with rheumatoid arthritis (RA) with either an inadequate response to background methotrexate therapy or an inadequate response or intolerance to an anti-TNF (tumour necrosis factor) biologic therapy: a randomized, controlled trial . Arthritis Res Ther 2019 ; 21 : 101 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Cook AD , Pobjoy J, Steidl S et al. Granulocyte-macrophage colony-stimulating factor is a key mediator in experimental osteoarthritis pain and disease development . Arthritis Res Ther 2012 ; 14 : R199 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Schett G , Bainbridge C, Berkowitz M et al. A phase IIa study of anti-GM-CSF antibody GSK3196165 in subjects with inflammatory hand osteoarthritis. Abstract 1365 . Arthritis Rheumatol 2018 ; 70(Suppl. 9 ): 1494 . Google Scholar OpenURL Placeholder Text WorldCat 11 Van Helvoort EM , Coeleveld K, de Boer TN et al. Lack of a clear disease modifying activity of celecoxib treatment of end-stage knee osteoarthritis: a randomized observer blinded clinical trial . JSM Bone Joint Dis 2017 ; 1 : 1006 . Google Scholar OpenURL Placeholder Text WorldCat 12 Mastbergen SC , Marijnissen AC, Vianen ME et al. The canine ‘groove’ model of osteoarthritis is more than simply the expression of surgically applied damage . Osteoarthritis Cartilage 2006 ; 14 : 39 – 46 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Eitner A , Pester J, Nietzsche S, Hofmann GO, Schaible HG. The innervation of synovium of human osteoarthritic joints in comparison with normal rat and sheep synovium . Osteoarthritis Cartilage 2013 ; 21 : 1383 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Pujol R , Girard CA, Richard H et al. Synovial nerve fiber density decreases with naturally-occurring osteoarthritis in horses . Osteoarthritis Cartilage 2018 ; 26 : 1379 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Benito MJ , Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B. Synovial tissue inflammation in early and late osteoarthritis . Ann Rheum Dis 2005 ; 64 : 1263 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Bedson J , Croft PR. The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature . BMC Musculoskelet Disord 2008 ; 9 : 116 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Nicol LSC , Thornton P, Hatcher JP et al. Central inhibition of granulocyte-macrophage colony-stimulating factor is analgesic in experimental neuropathic pain . Pain 2018 ; 159 : 550 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Arendt-Nielsen L , Nie H, Laursen MB et al. Sensitization in patients with painful knee osteoarthritis . Pain 2010 ; 149 : 573 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Fu K , Robbins SR, McDougall JJ. Osteoarthritis: the genesis of pain . Rheumatology 2018 ; 57 : iv43 – iv50 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Geenen R , Overman CL, Christensen R et al. EULAR recommendations for the health professional’s approach to pain management in inflammatory arthritis and osteoarthritis . Ann Rheum Dis 2018 ; 77 : 797 – 807 . Google Scholar PubMed OpenURL Placeholder Text WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Expression of granulocyte macrophage-colony stimulating factor and its receptor in the synovium of osteoarthritis patients is negatively correlated with pain
JF - Rheumatology
DO - 10.1093/rheumatology/keaa199
DA - 2020-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/expression-of-granulocyte-macrophage-colony-stimulating-factor-and-its-A80ly6d3YG
SP - 3452
EP - 3457
VL - 59
IS - 11
DP - DeepDyve
ER -