Edited by: Ali Mobasheri, University of Oulu, Finland
Reviewed by: Martijn H. Van Den Bosch, Radboud University Nijmegen Medical Centre, Netherlands; Iris Maria Ribitsch, University of Veterinary Medicine Vienna, Austria; Florien Jenner, University of Veterinary Medicine Vienna, Austria; Sheila Laverty, Université de Montréal, Canada
This article was submitted to Comparative and Clinical Medicine, a section of the journal Frontiers in Veterinary Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Synovitis is a major component of osteoarthritis and is driven primarily by macrophages. Synovial macrophages are crucial for joint homeostasis (M2-like phenotype), but induce inflammation (M1-like) when regulatory functions become overwhelmed. Macrophage phenotypes in synovium from osteoarthritic and healthy joints are poorly characterized; however, comparative knowledge of their phenotypes during health and disease is paramount for developing targeted treatments. This study compared patterns of macrophage activation in healthy and osteoarthritic equine synovium and correlated histology with cytokine/chemokine profiles in synovial fluid. Synovial histology and immunohistochemistry for M1-like (CD86), M2-like (CD206, IL-10), and pan macrophage (CD14) markers were performed on biopsies from 29 healthy and 26 osteoarthritic equine joints. Synovial fluid cytokines (MCP-1, IL-10, PGE2, IL-1β, IL-6, TNF-α, IL-1ra) and growth factors (GM-CSF, SDF-1α+β, IGF-1, and FGF-2) were quantified. Macrophage phenotypes were not as clearly defined
Osteoarthritis (OA) is a leading cause of lameness and morbidity and presents significant treatment challenges in horses and people (
Upon defined stimulation
Specific information regarding macrophage phenotypes in joint disease is limited to
The objective of this study was to compare the expression of macrophage markers in the synovium of healthy equine carpal and metacarpophalangeal (MCP) joints and those with naturally occurring OA. The well-defined equine model for the study of OA (
Synovial fluid and synovial membrane biopsies were collected from 26 osteoarthritic joints (16 MCP joints and 10 radiocarpal/middle carpal joints) of horses undergoing arthroscopy or following euthanasia at the Hagyard Equine Medical Institute (Lexington, KY) or the Virginia-Maryland College of Veterinary Medicine (Blacksburg, VA). Written informed consent was received from owners prior to inclusion of horses in the study. Control samples from healthy joints (15 MCP and 14 carpal joints) were collected at the same hospitals from horses without history and evidence of lameness referable to the harvested joints and with grossly healthy articular surfaces at euthanasia. All procedures were performed under IACUC approval. Both healthy and OA samples were harvested from 13 horses, OA samples only from 8, and healthy samples only from 7 horses. The mean age of control horses used to harvest healthy samples (7.4 years) was similar to those with OA (6 years), and comparable to horses used for both purposes (8 years). Synovial inflammation was assessed by gross pathology, synovial membrane histology, and synovial fluid cytology and immunoassay quantification of concentrations of pro- and anti-inflammatory cytokines and growth factors. Synovial macrophage phenotype activation
A total of 29 horses (11 females and 18 castrated males), 3–15 years old (skeletally mature, but not aged) were recruited and lameness exams performed, including response to joint manipulation, joint flexion, gait analysis at the trot, and radiography. OA joints were from horses exhibiting Grade 1-3 (out of 5) lameness (
Synovial fluid (2 mL) was aseptically collected and aliquoted (EDTA and Protein LoBind microfuge tubes, Eppendorf®, Westbury, CT). Anticoagulant-free synovial fluid was immediately centrifuged at 12,000 × g for 10 min at 4°C and the supernatant stored at −20°C for cytokine and growth factor quantification. Two synovial membrane biopsies were obtained from each OA joint adjacent to the major cartilage alterations (
Synovial fluid cytology was processed for total nucleated cell count (TNCC) by hemocytometer and total protein (TP) by refractometer. Differential cell counts were performed following Romanowski stain (Microscopy Hemacolor®, Merck, Germany). Concentrations of pro- (IL-1β, IL-6, GM-CSF, TNF-α) and anti-inflammatory cytokines (IL-10, IL-1ra), chemokines (MCP-1, SDF-1), growth factors (IGF-1, FGF-2), and PGE2 in synovial fluid were quantified. Thawed samples (200 μL) were hyaluronidase-digested (10 μL of 100 IU hyaluronidase/mL acetate buffer; Worthington Biochemical Corporation, Lakewood, NJ) for 30 min at 37°C, centrifuged at 12,000 × g for 10 min at 4°C, and the supernatant recovered. Based on previous experience and interfering factors in cytokine detection in synovial fluid (
PGE2 was quantified by ELISA (R&D Systems, Minneapolis, MN). Hyaluronidase-digested samples were solid-phase extracted (500 μL synovial fluid in 490 μL 100% ethanol and 10 μL glacial acetic acid incubated at 23°C for 5 min), centrifuged at 2,500 × g for 8 min at room temperature, and the supernatant collected. Remaining analytes were quantified by bead-based multiplex assay (MILLIPLEX MAP Equine Cytokine/Chemokine Multiplex Assay with manufacturer modification to include IGF-1, SDF-1, and IL-1ra; Luminex 200 plate reader Millipore Sigma, Burlington, MA).
Fixed synovial membrane biopsies were paraffin-embedded, sectioned at 5 μm, and H&E-stained. Synovitis was scored based on the OARSI histopathology guide and included cell infiltration, vascularity, hyperplasia, edema, and fibrosis (
Data analysis was performed using SAS version 9.4 (SAS Institute, Inc, Cary, NC). Effects of different joints sampled (carpi vs. MCP) and effects of disease (healthy vs. OA) on outcomes were assessed using linear General Estimating Equations (GEE) in an incomplete block design. Each of the linear models specified joint, disease, and the interaction between joint and disease as fixed effects. Correlation between observations within horse (the blocking factor) were modeled by specifying a compound symmetry covariance matrix. The interaction between joint and disease was further analyzed (sliced) to extract comparisons between disease conditions within joint. Scatterplots and analysis of covariance models were used to determine associations between synovial fluid cytology, synovial membrane histology, and synovial membrane immunohistochemistry parameters with joint condition (healthy vs. OA). For the analysis of covariance models, immunohistochemistry parameters (macrophage markers) were specified as covariates (one parameter at a time) while disease was the design effect. Statistical significance was set to
To determine whether synovial fluid macrophage counts differed between healthy and OA joints, standard synovial fluid cytology analysis was performed (
Synovial fluid cytology from healthy and OA equine metacarpophalangeal and carpal joints (median, 95% Confidence Interval).
Metacarpophalangeal joints | 2.1 (1.5–2.4) | 91 (24–256) | 65 (55–73) | 28 (24–43) | 0 (0–3) | |
2.7 (1.1–3.9) | 68 (21–607) | 68 (49–79) | 27 (4–44) | 0 (0–3) | ||
Carpal joints | 2.4 (1.6–2.8) | 24 (19–221) | 58 (50–67) | 33 (28–48) | 2 (0–3) | |
3.1 (1.7–3.8) | 124 (14–204) | 61 (46–77) | 31 (16–39) | 2 (0–19) | ||
Overall | 2.1 (1.9–2.4) | 91 (24–156) | 64 (55–71) | 30 (25–43) | 1 (0–3) | |
2.7 (1.8–3.6) | 110 (36–173) | 65 (54–73) | 29 (18–39) | 0 (0–3) | ||
No significant differences were detected between samples from control and OA joints.
TNCC, Total Nucleated Cell Count. P-values < 0.05 highlighted in bold.
To assess the secretory response of synovial lining macrophages, concentrations of pro- (IL-1β, IL-6, GM-CSF, TNF-α, PGE2) and anti-inflammatory cytokines (IL-10, IL-1ra), chemokines (MCP-1, SDF-1) and growth factors (IGF-1, FGF-2) were quantified in synovial fluid (
Cytokine, chemokine, and growth factor concentrations in synovial fluid of healthy and OA equine joints (median, 95% Confidence Interval).
11.5 pg/mL | 0.3 pg/mL | 15.5 pg/mL | 2.3 pg/mL | 0.02 pg/mL | 9 pg/mL | 20.5 pg/mL | 23.2 pg/mL | 39 pg/mL | 1.5 pg/mL | ||
Metacarpo- phalangeal joints | U | U | |||||||||
– | – | – | – | – | – | ||||||
Carpal joints | |||||||||||
– | – | – | – | – | – | ||||||
Overall | |||||||||||
– | – | – | – | – | – |
Overall scores for histological assessment of the synovium for intimal hyperplasia (
Individual and composite histological parameters for H&E-stained equine synovial membrane (median, 95% Confidence Interval).
Metacarpo-phalangeal joints | 2 (1–2) | 2 (1–3) | 1 (0–1) | 1 (0–2) | 2 (2–3) | 7 (4–11) | |
2 (1–3) | 3 (1–4) | 1 (0–3) | 1.5 (1–3) | 2 (1–3) | 9.5 (6–14) | ||
Carpal joints | 2 (1–2) | 2 (1–2) | 0.5 (0–1) | 1 (1–2) | 2 (1–3) | 8.5 (5–9) | |
2 (1–3) | 2 (0–3) | 1 (1–2) | 2 (0–3) | 3 (1–3) | 9.5 (7–12) | ||
Overall | 2 (1–2) | 2 (1–2) | 1 |
1 (1–2) | 2 (1–3) | 8 (6–9) | |
2 (1–3) | 2.5 (0–4) | 1 |
2 (1–3) | 2.5 (2–3) | 9.5 (7–12) | ||
Synovial Hyperplasia was significantly higher in OA carpi, while Subintimal Edema was significantly higher in OA metacarpophalangeal joints.
Compared to OA joints with no or minimal signs of gross inflammation
The distribution of immunostaining for macrophage markers across the synovial lining differed between healthy and OA joints. In healthy joints, staining was largely limited to the base of synovial villi, while in OA joints the tips of villi were also frequently stained (
Composite immunohistochemical scores of macrophage markers in healthy and OA synovial membrane (median, 95% Confidence Interval).
Metacarpo-phalangeal joints | 4 (0–5) | 4 (0–6) | 4 (4–6) | 5 (4–6) | |
5 (0–7) | 5 (0–7) | 5 (0–6) | 6 (4–6) | ||
Carpal joints | 5 (4–6) | 6 (4–6) | 4 (0–6) | 6 (5–6) | |
5 (0–6) | 6 (5–8) | 5.5 (4–7) | 6 (5–7) | ||
Overall | 5 (0–5) | 5 (4–6) | 5 (0–6) | 5 (5–6) | |
6 (4–6) | 6 (5–7) | 5 (4–6) | 6 (5–6) | ||
Osteoarthritic metacarpophalangeal joints exhibited increased expression of all markers, while only CD206 expression was higher in OA carpi. P-values < 0.05 highlighted in bold.
Representative immunohistochemistry sections from healthy and OA equine synovial membrane at low (top 2 rows; scale bar = 100 μm) and high magnification (bottom 2 rows; scale bar = 50 μm) from the same histological section and demonstrating the median staining scores for macrophage markers (CD14, CD86 [M1], CD206 and, IL-10 [M2]). Staining for all markers was most intense on the vascular endothelium (black arrows). In healthy joints, staining was largely limited to the base of synovial villi (white arrowheads), while in OA joints the tips of villi were also frequently stained (black arrowheads). In healthy joints, staining for macrophage markers at the tips of villi was subtle and primarily located at the synovial lining around cell nuclei. In contrast, staining in OA joints was more diffusely distributed in the synovial lining around cell nuclei.
In the subset of OA joints with gross signs of synovitis, staining for CD86 was more markedly intense than remaining OA joints. A similar, but less consistent pattern was observed for CD14, IL-10, and CD206. For 4 horses, we were able to compare OA joints with gross signs of synovitis to the healthy contralateral joints of the same individual (
Sets of representative immunohistochemistry sections from healthy and grossly inflamed OA equine synovial membrane from the same horse (2 different horses; scale bar = 150 μm) demonstrating increased staining intensity and distribution for all selected markers in OA joints, denoting more consistently marked increases for CD86 staining.
This is the first study in any species comparing macrophage phenotypes in the synovium from healthy joints to those with naturally occurring OA. Markers widely used to define M1- (CD86) and M2-like (CD206 and IL-10) macrophages were similarly expressed in both groups. Expression for all markers varied with degree of synovial inflammation. While their expression was mildly increased in OA joints with low-grade inflammation (majority), it was markedly increased in grossly inflamed OA joints, with CD86 most highly expressed.
Under sustained inflammatory conditions, macrophages have lower expression of pro-resolving molecules such as IL-10 (
The concept of macrophage activation as either inflammatory (M1) or regulatory (M2) originated from monocyte-derived macrophages treated
Similar to CD86, the mannose receptor (CD206) has a pivotal function in host defenses during inflammation, clearance of debris, wound healing and remodeling, and resolution of inflammation. CD206 is also constitutively expressed in mature mononuclear phagocytes and the intensity of its expression is proportionate to demands for anabolic cytokine secretion, efferocytosis, and sensing of damage-associated molecular patterns (
Like CD86 and CD206, expression of IL-10 in the synovial membrane in our study was directly associated with the degree of synovial inflammation. After injury, macrophage activation leads to increased expression of IL-1, IL-6, and TNF-α, which is followed by proportional increases in expression of IL-10 as a compensatory, negative feedback (
The lower synovial fluid concentrations of IL-10 in OA metacarpophalangeal joints suggests that mechanisms compensating for tissue damage may be impaired or overwhelmed in OA joints. An
In response to injury, resident synovial macrophages form a protective immunological barrier in the synovial lining, similar to the hyperplastic synovium, secluding intra-articular structures. Exchange of solutes and cells from the sub-synovial to intra-articular space is restricted and could explain higher IL-10 staining in the synovium from OA joints with lower synovial fluid IL-10 concentrations than healthy joints (
Increased overall synovial fluid MCP-1 concentrations in OA, concomitant with clinical signs of joint inflammation, is consistent with the literature (
Traditionally, IL-1β and TNF-α have been considered the main drivers of disease processes in OA (
Although differences between normal and OA joints were observed for both carpal and metacarpophalangeal joints, differences were more often identified in metacarpophalangeal joints. One potential reason for this observation is that, due to a more distal location and higher range of motion, metacarpophalangeal joints are more exposed to higher mechanical loads and stress, and thus the response to trauma and tissue microdamage may be more marked. In fact, metacarpophalangeal joints are the most commonly affected site of injury in many equine disciplines (
Even though our experimental design was aimed at minimizing variability, synovial histological parameters can vary with joint and site within the joint, and could have contributed to a degree of variability among samples, preventing statistical inference. While assessing the expression of macrophage markers in synovial fluid cells using flow cytometry would have contributed to our findings in the synovial membrane, recent reports are in agreement with the pattern of expression identified in our study (
Combined with previously reported studies, our results suggest that synovial macrophages are strictly neither M1 nor M2, but represent a hybrid state of activation that overall displays a regulatory response and that ultimately targets resolution of the inflammatory process (
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The animal study was reviewed and approved by Virginia Tech Institutional Animal Care and Use Committee. Written informed consent was obtained from the owners for the participation of their animals in this study.
BM, DR, SW, and LD contributed substantially to study conception and design. BM and DR collected samples. BM was primarily responsible for data acquisition, analysis, and interpretation. KG, AO, and YN assisted BM with data collection and assembly. SB supervised the synovial fluid cytology performed by BM and KG. SW performed statistical analysis and consulted on its interpretation. BM and LD were responsible for manuscript preparation. All authors reviewed the final manuscript. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors thank the surgeons and surgery interns at Hagyard Equine Medical Institute for their invaluable contributions in collecting samples, and Dr. Renee Nodine from Horseshoe Valley Equine Center in helping recruit horses for the study.
metacarpophalangeal joints
classically activated/pro-inflammatory
suppressive/healing
macrophage chemoattractant protein 1
osteoarthritis/osteoarthritic
prostaglandin E2
stromal cell–derived factor 1
total nucleated cell count
total protein
LPS co-receptor along with Toll-like receptor 4— proposed mature macrophage marker
/T cell costimulatory receptor (B7.2)—proposed M1 marker
Mannose receptor 1 (MRC1)—proposed M2 marker.