Sage Journals: Discover world-class research

Abstract

Neurogenic heterotopic ossifications (NHOs) are ectopic bones developing in periarticular muscles after severe central nervous system injuries such as spinal cord injuries (SCI) and traumatic brain injuries (TBI). The pathogenesis of NHO is still poorly understood, and the only effective therapy remains surgical resection. Using our unique mouse model of NHO development after SCI and muscle injury, we have previously shown that oncostatin M (OSM), a cytokine of the interleukin-6 (IL-6) superfamily, contributes to NHO pathogenesis. However, a possible role of IL-6 in NHO pathogenesis remains unexplored. Herein, we establish that IL-6 mRNA is significantly upregulated in muscles developing NHO after SCI and that IL-6 protein is significantly elevated in the blood of mice developing NHO. However, administration of anti-IL-6 and anti-IL-6 receptor α chain (IL6RA) neutralizing antibodies after SCI and muscle injury did not impact NHO development. Relevant to the human pathology, recombinant human IL-6 and/or soluble IL-6 receptor α chain (sIL6RA) did not increase the mineralization of human fibro-adipogenic progenitors (FAPs) isolated from muscles surrounding NHO biopsies or the expression of osteoblast genes. As we have previously shown that macrophages are critical for NHO development, we also examined an indirect role of IL-6 on FAPs via peripheral blood monocytes. Results showed that conditioned media from human CD14⁺ monocytes stimulated with either IL-6 and/or sIL6RA did not increase NHO mineralization or expression of osteoblast genes. Finally, we show that mRNA for IL6RA were expressed at low levels in mesenchymal progenitors isolated from mouse and human muscles compared with OSM receptor α chain and GP130 mRNA which were abundantly expressed by these cells. Overall, our findings suggest that the contribution of the IL-6 pathway to NHO pathogenesis is minimal and that treatment with IL6RA or IL-6 neutralizing antibodies is unlikely to have beneficial effect to stop the development of this pathology.

Introduction

Neurogenic heterotopic ossifications (NHOs) are heterotopic bones that develop in periarticular muscles after severe central nervous system (CNS) injuries such as spinal cord injury (SCI), traumatic brain injury (TBI), or stroke.^1–3 The periarticular location of NHO is the cause of their pathogenicity by reducing the range of motion to completely ankylose the joints, which impairs patients’ ability to sit, dress, and eat independently, thus increasing morbidity, incapacity, and delaying rehabilitation.^4,5 NHO can also cause nerve and blood vessel compression, further increasing patient morbidity.⁶ Due to our poor understanding of NHO pathogenesis,^1,7 the only effective therapy is surgical resection^8–10; however, NHO recurrence is observed in 6% of operated patients.⁶ There is, therefore, a clear unmet need for the identification of prophylactic treatments that target specific pathways responsible for NHO pathogenesis.

To address this need, we developed an animal model of SCI-induced NHO in genetically unmanipulated mice, where SCI is combined with a muscle injury via an intramuscular injection of cardiotoxin (CDTX), a well-accepted model of muscle injury/repair.^11–13 In this model, NHO exclusively develop in the CDTX-injured muscles of mice with SCI¹⁴ but not in sham-operated mice. Using this model, we have demonstrated that NHOs are derived from muscle mesenchymal fibro-adipogenic progenitors (FAPs) and not from myogenic stem cells (also called satellite cells [SC]).¹⁵ We also established that macrophages are key drivers of NHO^14,16 (lymphocytes and neutrophils have no role^17,18), whereby the SCI exacerbates macrophage-mediated inflammation,¹⁹ with increased production of oncostatin M (OSM)¹⁶ and interleukin-1 (IL-1),¹⁹ which both promote NHO via their receptors OSMR and IL1R1, respectively, and similar observations were noted in the human setting.^16,19

In a gene expression microarray analysis of whole mouse injured muscles, interleukin-6 (IL-6) was found overexpressed in CDTX-injured muscles from mice with SCI, which develop NHO, compared with control CDTX-injured muscles from mice without SCI, which do not develop NHO.¹⁹ IL-6 and OSM are two functionally related inflammatory cytokines that bind to two distinct heterodimeric high-affinity receptors which both comprise the GP130 transmembrane signaling chain as OSMR:GP130 for the OSM receptor and IL6R:GP130 for the IL-6 receptors. Both cytokines activate JAK kinases and STAT transcription factors, particularly STAT3.^20–23 Given that we have previously shown that JAK/STAT signaling is important for NHO development, it makes IL-6 or its receptor attractive potential targets to inhibit NHO development as several neutralizing antibodies against IL-6 or IL6RA have been approved for several chronic inflammatory conditions.²¹ Furthermore, it has been reported that neutralizing anti-IL6R antibody tocilizumab reduces neuroinflammation in several animal models of SCI.^24,25 This prompted us to investigate a possible role of IL-6 in NHO pathogenesis. Herein, we report that IL-6 has no role in NHO development in mice and no osteogenic effects in NHO-associated human muscle-derived FAPs.

Materials and Methods

Animals

C57BL/6 mice were sourced from the Animal Resource Center. Mice were housed at the Translational Research Institute, Biological Research Facility, under specific pathogen-free conditions. All mice were fed a standard diet chow (Specialty Feeds) with ad libitum water access and simulated diurnal cycle. All mouse procedures were approved by the Health Sciences Animal Ethics Committee of The University of Queensland (ethics number 2021AE000155 and 2021AE000537) and performed in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.

SCI-induced NHO mouse model

NHO mouse model was carried out as previously described in 6- to 7-week-old female mice to be able to express bladders twice daily in an ethical manner.^14,16,26 Under general anesthesia, mice received a spinal cord transection at T11–T13 and an intramuscular injection (i.m.) of cardiotoxin (CDTX) purified from the venom of Naja pallida (Latoxan) at 0.32 mg/kg in the hamstring muscles in a final volume of 50 μL. Mice then received a subcutaneous injection of ciprofloxacin (10 mg/kg) and buprenorphine (0.075 mg/kg) and recovered on a heat pad. As SCI causes complete paraplegia, mouse bladders were expressed manually twice daily throughout the experiments. Mice were given Bactrim (800 mg/L, Roche) in drinking water following surgical procedures as prophylaxis for bladder infections. All mice were randomly assigned into groups with a maximum of five mice per cage.

Animal treatments

Post-surgery, C57BL/6 mice were injected intraperitoneally every other day with 200 μg neutralizing rat anti-mouse anti-IL-6 receptor α chain (IL6RA) (Bio X Cell, catalogue #BE00047, clone 15A7) together with 200 μg neutralizing rat anti-mouse IL-6 (catalogue #BE0046, clone MP5-20F3), or 200 μg of matching rat IgG2b anti-keyhole limpet hemocyanin isotype control (catalogue #BE0090, clone LTF-2) together with 200 μg rat IgG1 anti-horseradish peroxidase isotype control (catalogue #BE088, clone HRPN) from immediately after surgery until day 6 post-surgery.

Mouse tissue collection

At pre-determined endpoints, mice were euthanized by CO₂ asphyxiation or anesthetized in an isoflurane chamber and blood collected by cardiac puncture. Hindlimbs were fixed and decalcified and processed as previously described.¹⁶ Blood samples were collected by terminal cardiac puncture in tubes with 4 mM ethylenediaminetetraacetic acid (EDTA) final and centrifuged as described.²⁷ Plasma samples were aliquoted and stored at −80°C.

Purification of mouse muscle progenitors and leukocytes for RNA extraction

Muscle leukocytes were isolated from C57BL/6 mouse hamstring muscles 4 days post-SCI+CDTX using a skeletal muscle dissociation kit (Miltenyi Biotech) as per manufacturer’s instructions. Total muscle leukocytes were sorted into multiple populations using a BD FACS Aria Fusion (BD Biosciences) using fluorescent monoclonal antibodies (Biolegend): CD45-BV785 (clone 30-F11), anti-TER119-FITC (clone TER119), CD45R/B220-FITC (clone RA3-6B2), CD3ε-FITC (clone 145-2C11), CD11b-BV510 (clone M 1/70), anti-F4/80-APC (clone BM8), anti-Ly6C-Pacific Blue (clone HK1.4), anti-Ly6G-PE (clone 1A8), and anti-CD48 APCCy7 (clone HM48-1) and cell viability with 7-aminoactinomycin D (7-AAD). Monocytes/macrophages were sorted as CD45⁺ Ter119⁻ B220⁻ CD3ε⁻ CD11b⁺ F4/80⁺ CD48⁺ Ly6G⁻ with various levels of Ly6C expression and granulocytes as CD45⁺ Ter119⁻ B220⁻ CD3ε⁻ CD11b⁺ F4/80⁻ CD48⁻ Ly6G⁺. Muscle progenitor cells (MPCs) were isolated from hamstring muscles of naïve C57BL/6 muscles and stained with the flowing fluorescent monoclonal antibodies: CD45-BV785 (clone 30-F11), anti-TER119-FITC (clone TER119), CD45R/B220-FITC (clone RA3-6B2), CD3ε-FITC (clone 145-2C11), CD11b-FITC (clone M 1/70), anti-Gr1-FITC (clone RB6-8C5), CD31-BV421 (clone 390), anti-Sca1-PECY7 (clone D7), CD34-e660 (clone RAM34), and anti-integrin α7-PE (clone R2F2) and cell viability with FVS700. Populations were sorted using a BD FACS Aria Fusion according to the following phenotypes: SC: CD45⁻ lineage-(Ter119, B220, CD3ε, CD11b, Gr1)⁻ CD31⁻ CD34⁺ Sca1⁻ integrin α7⁺; FAPs: CD45⁻ Lin⁻ CD31⁻ CD34⁺ Sca1⁺ integrin α7⁻ as previously described.¹⁵ All populations were sorted directly into 1 mL Trizol LS (ThermoFisher) and frozen until extraction.

Mouse RNA extraction and quantitative real-time polymerase chain reaction

For RNA isolation of muscle, frozen muscle samples were homogenized in Trizol (Life Technologies) as previously published.¹⁹ mRNA was isolated from sorted cells using chloroform separation from initial Trizol preparations followed by a GeneJET RNA cleanup and concentration micro kit (ThermoFisher). Reverse transcription was performed with all RNA samples using the SensiFAST™ cDNA Synthesis Kit (ThermoFisher) as per manufacturer’s instructions. mRNA expression was analyzed using a single-step reverse transcription quantitative real-time polymerase chain reaction (qRT-PCR) on a ViiA 7 Real-Time PCR System (Life Technologies) with PCR setting: 20 sec at 95°C and then 40 cycles of 95°C (1 sec) and 60°C (20 sec). Taqman system using TaqMan™ fast PCR Master Mix and TaqMan™ (Supplementary Table S1). Ct values were normalized by the expression of the house-keeping genes Rps20 (for whole muscle) or Hprt (for sorted cells) and presented as ratio to house-keeping gene.

Mouse blood collection for quantification of plasma cytokines

Blood was collected in tubes containing EDTA (4 mM final for 1 mL blood collected) and plasma was isolated by centrifugation (1,000 g for 10 min), plasma was subsequently transferred into new tubes and spun again (1,000 g for 10 min), and plasma stored at −80°C until use. Plasma cytokine concentrations were measured in triplicate using bead-based immunoassay (LEGENDplex, mouse inflammation panel, BioLegend Cat #740150) and analyzed on a CytoFLEX benchtop flow cytometer.

Micro-computerized tomography and NHO volume quantification

NHO volumes were measured using the Molecubes β-Cube and X-Cube µPET-CT system (Molecubes). 3D reconstitutions and NHO volumes were quantified using VivoQuant^TM 2021 (Invicro). To quantify NHO volumes, the region of interest (ROI) was drawn around the muscles containing NHO and these were then carefully checked from three dimensions to ensure adjacent long bones were not included in the ROI. Calcified NHO regions were defined as above the threshold of 450 Hounsfield units.

Human ethics statement and samples collection

Samples were obtained with the informed consent of the patients and under the approval of an independent ethics committee (BANKHO, Comité de Protection des Personnes approval no. 09025). Muscle surrounding NHOs were collected from surgical waste following their excision from patients with TBI or SCI at Raymond Poincaré Hospital (Garches, France).

Isolation of human FAPs

Muscle fragments were minced using scalpel and small scissors and incubated in 1.5 mg/mL pronase (Sigma-Aldrich) in α-MEM for 45 min in a 37°C water bath. After addition of alpha minimum essential medium (α-MEM) supplemented with 15% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S), the cell suspension was filtered through a 100 µm cell strainer followed by a 40 µm cell strainer (BD Falcon). Isolated MPCs were maintained 10 days in α-MEM supplemented with 15% FBS, 1% P/S, and 10 ng/mL basic fibroblast growth factor (R&D Systems). Human muscle progenitor cells (MPCs) were trypsinized and incubated for 30 min with biotinylated anti-human platelet-derived growth factor receptor alpha (PDGFRα) (BAF322, R&D Systems) goat polyclonal antibody and CD56-PE (clone B159, BD Pharmigen) monoclonal antibody in PBS 2% FBS, 2 mM EDTA or with control isotypes IgG1 PE (A07796, Beckman Coulter), and biotinylated goat IgG (BAF108, R&D Systems). Cells were washed and incubated for 30 min with streptavidin-APC/Cy7 and 7-actinomycin D viability dye (Sony). Cells were washed and filtered through a 30 µm cell strainer (Sysmex) and sorted using a FACSAria III SORP sorter (BD Biosciences). PDGFRα⁺ CD56⁻ FAPs were seeded at 3,000 cells per cm² in α-MEM supplemented with 10% FBS and 1% P/S.

In vitro osteogenic differentiation assay and mineralization quantification

Human FAPs were seeded at 3,000 cells per cm² in α-MEM 10% FBS and 1% P/S. After 3 days, culture medium was replaced by control medium (α-MEM 10% FBS 1% P/S) or osteogenic medium (α-MEM 10% FBS 1% P/S, 0.052 mg/mL dexamethasone, 12.8 µg/mL ascorbic acid, and 2.15 mg/mL β-glycerophosphate [Sigma-Aldrich]) for 12–14 days. Human recombinant IL-6 and IL-6Rα (PeproTech) were directly incubated with FAPs at the indicated concentrations. At the end of the differentiation process, cells were washed in PBS, fixed in 70% ethanol, washed in water, and stained in 20 g/L Alizarin red S (Sigma-Aldrich). Excess stain was removed during three washing steps. Alizarin Red S dye was extracted with 0.5N hydrochloric acid and 5% sodium dodecyl sulfate (SDS) and quantified by spectrophotometry at 405 nm.

Human cell RNA extraction and RT-qPCR

Total RNAs were extracted using QIAzol lysis reagent (Qiagen) and chloroform (Sigma-Aldrich) extraction. RNA precipitation was performed using isopropanol (Sigma-Aldrich) and GlycoBlue (Ambion) for 20 min at −20°C. Pellets were washed with 75% ethanol before RNAse free water resuspension. Total RNA concentration were evaluated by NanoDrop (ThermoFisher), pre-diluted to 20 ng/µL for mRNA, and stored at −80°C. Reverse transcriptions were performed with RT² First Strand kit (Qiagen). Post-RT cDNA were diluted to 1:80 before qRT-PCR amplification using Quantitect SYBR Green and Quantitect primers (Qiagen). Three reference genes were selected (FAPs: HPRT, RPLP0, and PPIA; CD14⁺ monocytes: ACTB, GAPDH, and PPIA) using Genorm (v3.4) and normalized by the geometric mean of 1.9^ΔCt to reference gene.

Human CD14⁺ monocytes isolation and preparation of conditioned media

Human CD14⁺ blood monocytes were isolated from whole blood donation buffy coat from the Centre de Transfusion Sanguine des Armées (Clamart, France). All donors gave their consent for the research use of residues not used in the preparation of blood products. Peripheral blood mononuclear cells (PBMCs) were recovered using Ficoll gradient (Pan-Biotech). CD14⁺ monocytes were then isolated following magnetic separation using CD14 microbeads (Miltenyi Biotec). CD14⁺ monocytes were seeded at 1.25 × 10⁶ cells per cm² and stimulated with 100 ng/mL lipopolysaccharides (LPS) (L6529, Sigma-Aldrich), 100 ng/mL of recombinant human IL-6, sIL6Rα (Peprotech), or the combination of both IL-6 and sIL6Rα. Conditioned media (CM) CM^IL-6, CM^sIL-6Rα, CM^{IL-6+sIL-6Rα}, or without stimulation (CM^∅) were collected, centrifuged for 5 min at 500 g, and used in a 1:10 ratio in cell culture medium during osteogenic differentiation assays.

Statistical analysis

Data is represented as mean ± standard deviation. Statistically significant differences were determined using two-sided Mann–Whitney test, one-way analysis of variance (ANOVA or two-way ANOVA) in PRISM 7 (GraphPad software, California, USA).

Results

IL-6 is over-induced in injured muscles after SCI

In our previous study, we found Il6 mRNA overexpressed in CDTX-injured muscles developing NHO 2 days post-surgery in a gene expression microarray from whole muscles.¹⁹ To validate these initial findings, we quantified Il6 mRNA by qRT-PCR on whole muscles from a distinct cohort of mice that underwent SCI or sham surgery, with or without CDTX-induced muscle injury (Fig. 1A). CDTX-mediated muscle injury significantly increased Il6 mRNA compared with uninjured muscles regardless of the presence or absence of SCI. However, Il6 mRNA was significantly more abundant in CDTX-injured muscles developing NHO from mice with SCI (Fig. 1A). At the protein level, IL-6 concentration in blood plasma was significantly increased in mice with SCI + muscle injury compared to mice with muscle injury alone at both 17 and 72 h post-surgery (Fig. 1B). As activated monocytes/macrophages are a major source of IL-6 (Supplementary Figure S1), monocyte subsets and granulocytes were sorted from CDTX-injured muscles from mice that underwent SCI 4 days post-injury, and Il6 mRNA were quantified by qRT-PCR. All monocyte subsets (Ly6C^high, Ly6C^mid, and Ly6C^low) expressed detectable amounts of Il6 mRNA, with lower levels in granulocytes (Fig. 1C).

FIG. 1.

Il6 and Il6ra mRNA expression during NHO development in mice. (A) Validation of Il6 expression in injured muscles by qRT‐PCR. C57BL/6 mice underwent SCI and CDTX muscle injury. Whole hamstrings were collected 2 days post-surgery. Il6 mRNA was quantified relative to Rps20 housekeeping gene. (B) Plasma IL-6 levels in C57BL/6 mice at 17 or 72 h post-CDTX and -SCI+CDTX. (C) Expression of Il6 mRNA transcripts in sorted muscle monocytes/macrophages CD45⁺ Ter119⁻ B220⁻ CD3ε⁻ CD11b⁺ F480 ⁺ Ly6C^high (Ly6Chi), Ly6C^mid (Ly6cmid) and Ly6C^-/low (Ly6Clo), CD45⁺ CD11b⁺ F4/80⁻ Ly6G⁺ granulocytes (Grans), and positive control sample (sorted BM CD45⁺ CD11b⁺ myeloid cells from granulocyte colony stimulating factor (GCSF)-treated mice). Il6 mRNA expression was quantified relative to relative to housekeeping gene Hprt. Each dot represents one separate mouse; mean ± SD. CDTX, cardiotoxin; IL-6, interleukin-6; IL6RA, IL-6 receptor α chain; NHO, neurogenic heterotopic ossification; qRT-PCR, quantitative real-time polymerase chain reaction; SCI, spinal cord injuries; SD, standard deviation.

Neutralizing antibodies for IL-6 and IL6RA does not reduce NHO development in mice

To test for a potential role for elevated IL-6 in NHO pathogenesis, two cohorts of mice underwent SCI and CDTX-mediated muscle injury and were administered a combination of neutralizing anti-mouse IL-6 together with anti-mouse IL6RA (200 μg/kg of each antibody) or with isotype-matched control monoclonal antibody, daily from immediately after surgery until 1 week post-surgery. These antibodies have been previously shown to be effective to block endogenous IL-6 effects in vivo in mice.^28,29 Surprisingly, in vivo neutralization of IL-6 and IL6RA had no effect of NHO development after SCI (Fig. 2)

FIG. 2.

Anti-IL6/IL6RA administration does not alter NHO development after SCI in mice. C57BL/6 female mice underwent a complete spinal cord transection between vertebrae T11 and T13 in conjunction with a muscle injury via intramuscular injection of CDTX in the right hamstring muscle (0.32 mg/kg). Mice were injected i.p. with either isotype controls (rat IgG2b + rat IgG1) or neutralizing rat anti-mouseIL6 (clone MP5-20F3) together with anti-mouse IL6RA (clone 14A7) (200 μg/mouse on day 0, 2, 4, and 6 post-surgery). (A–B) NHO volumes measured by μCT at 7 and 21 days post-surgery with representative images of NHO development at 7 days post-surgery. Each dot represents one mouse, mean ± SD; there was no statistical difference between the two groups by Mann–Whitney test. (C–E) Expression of (C) Il6ra, (D) Osmr, and (E) Il6st mRNA transcripts by qRT-PCR in muscle SC (CD45⁻ Lin⁻ CD31⁻ Sca1⁻ CD34⁺), FAPs (CD45⁻ Lin⁻ CD31⁻ Sca1⁺ CD34⁺), myoblasts (Myo) (CD45⁻ Lin⁻ CD31⁻ ), and endothelial cells (Endo) (CD45⁻ Lin⁻ CD31⁺) and bone marrow monocytes and granulocytes as controls. qRT-PCR mRNA expression relative to housekeeping gene Hprt. Each dot represents one mouse; mean ± SD. p-Values were calculated by one-way ANOVA. μCT, micro-computerized tomography; ANOVA, analysis of variance; CDTX, cardiotoxin; IL-6, interleukin-6; IL6RA, IL-6 receptor α chain; NHO, neurogenic heterotopic ossification; qRT-PCR, quantitative real-time polymerase chain reaction; SC, satellite cells; SCI, spinal cord injuries; SD, standard deviation.

To explain the lack of effect of IL-6 and IL6RA neutralization, we hypothesized that mouse muscle FAPs, which are the cells-of-origin of NHO in muscles,¹⁵ may only express low levels of IL6RA that would make them nonresponsive to IL-6. To test this, FAPs, SC, myoblasts, and endothelial cells were sorted from muscles of naïve mice and expression of Il6ra, Osmr, and Il6st (GP130) mRNA was quantified by qRT-PCR and compared to expression in bone marrow monocytes and granulocytes (Fig. 2C–E). Il6ra mRNA was very low in all muscle cell subsets (SC, FAPs, myoblasts, and endothelial cells) compared to bone marrow monocytes and granulocytes (Fig. 2C). Conversely, Osmr mRNA was expressed in all muscle cell types, whereas Osmr mRNA was undetectable in bone marrow monocytes and granulocytes (Fig. 2D), consistent with previous reports.²⁷ On the other hand, Il6st (GP130) mRNA was significantly higher in muscle FAPs and endothelial cells compared to bone marrow monocytes and granulocytes (Fig. 2E). Our findings are broadly consistent with the dataset in the BioGPS portal showing very low expression of Il6ra mRNA in osteoblasts and skeletal muscles compared to leukocytes and immune organs (Supplementary Figure S2), whereas expression of Osmr and Il6st mRNA is highest in mouse mesenchymal cells but undetectable (for Osmr) or very low (for Il6st) expression in blood and bone marrow leukocytes (Supplementary Figure S3 and Figure S4).

IL-6 does not induce osteogenic differentiation of NHO-associated human FAPs in vitro

We next isolated PDGRα⁺ CD56⁻ muscle-derived FAPs from muscle surrounding NHO from surgical resection biopsies from five different patients. We have previously shown that these cells are the cells-of-origin of NHO in humans.¹⁵ When FAPs were cultured in osteogenic conditions for 2 weeks with increasing doses of human IL-6, no additional calcium mineralization was observed (Fig. 3A). As IL-6 can signal to cells that do not express IL6RA at their surface via non-canonical signaling, in which IL-6 first complexes to shed soluble IL6RA (sIL6RA), FAPs were incubated in the presence of increasing concentrations of both human IL-6 plus sIL6RA or sIL6RA alone. Calcium mineralization was not enhanced either by any of these treatments (Fig. 3A).

FIG. 3.

IL-6 and and/or sIL6RA have no direct effect on human FAPs osteogenic differentiation. (A) Human FAPs (n = 5 biopsies/patients) were cultured for 2 weeks in OB supplemented with purified recombinant human IL-6, sIL6RA, or IL-6 + sIL6RA at indicated concentrations. Alizarin Red mineralization was quantified by spectrophotometry at 405 nm. Results were normalized for each patient/biopsy to the level of mineralization obtained in the absence or IL6 or sIL6R. (B–H) mRNA expression of osteogenic markers in muscle-derived FAPs from five different NHO biopsies/patients after 7 days of in vitro culture in OB supplemented with recombinant human IL-6, sIL6RA, or IL-6 + sIL6RA combination as indicated concentrations. mRNA expression of RUNX2, SOX9, BMP2, ALPL, COL1A1, COL1A2, and ATF4 were measured by qRT-PCR. Each dot represents a separate patient biopsy. Bars are mean ± SD. FAPs, fibro-adipogenic progenitors; IL-6, interleukin-6; LPS, lipopolysaccharides; OB, osteogenic medium; qRT-PCR, quantitative real-time polymerase chain reaction; sIL6RA, soluble IL-6 receptor α chain; SD, standard deviation.

To further document the lack of effect of IL-6 and sIL6RA on osteogenic differentiation of FAPs, RNA was extracted after 7 days of osteogenic culture to quantify the expression of mRNA encoding osteoblast differentiation markers RUNX2, SOX9, BMP2, ALPL, COL1A1, COL1A2, and ATF4 by qRT-PCR. None of these osteoblast differentiation markers was induced by IL-6 or sIL6RA alone or in combination (Fig. 3B–H). Overall this suggests that IL-6 has no direct role in inducing muscle-derived FAPs osteogenic differentiation.

We have recently shown that bacterial lipopolysaccharide (LPS) enhanced NHO development after SCI in mice and enhanced osteogenic differentiation of human FAPs in cultures.³⁰ The effect of LPS may be directly mediated via mouse or human muscle FAPs, as both cells express the LPS receptor toll-like receptor 4 (TLR4),³⁰ and indirectly mediated via monocytes/macrophages infiltrating the injured muscles, as these monocytes secrete pro-osteogenic NHO-inducing cytokines, such as OSM and IL-1 in response to LPS and other inflammatory stimuli.^16,19 To assess whether IL-6 may act indirectly via monocytes/macrophages which express IL6R mRNA (Fig. 4A), human CD14⁺ monocytes were purified from healthy adult donors and cultured with human IL-6, sIL6RA, or the combination of both to produce conditioned media (CM). CM from monocytes in response to IL-6 and/or sIL6RA, when added to FAPs osteogenic assays, had no effect on calcium mineralization or the expression of RUNX2, SOX9, BMP2, ALPL, COL1A1, COL1A2, and ATF4 mRNA (Fig. 5).

FIG. 4.

RNA expression of IL-6 family members in human FAPs from NHO biopsies and peripheral blood mononuclear cells. mRNA expression by qRT-PCR relative to housekeeping genes for (A) IL6R, (B) OSMR, and (C) IL6ST from human muscle FAPs and CD14⁺ peripheral blood mononuclear cells (Mono/mac). Each dot represents one donor. Bars represent mean ± SD. FAPs, fibro-adipogenic progenitors; NHO, neurogenic heterotopic ossification; qRT-PCR, quantitative real-time polymerase chain reaction; SD, standard deviation.

FIG. 5.

Indirect effect of IL-6 and and/or sIL6RA on human FAPs osteogenic differentiation via CD14⁺ human PBMCs. (A) Human FAPs (n = 5 biopsies/patients) were cultured for 14 days under OB supplemented with CM from CD14⁺ PBMCs from three healthy donors (A–C) stimulated with 100 ng/mL IL-6 (CM^IL-6), sIL6RA (CM^IL-6Rα) or IL-6 + sIL6Rα (CM^IL-6+IL-6Rα) (100 ng/mL each), or unstimulated (CM^Ø). Pooled CM from CD14⁺ PBMCs stimulated with100 ng/mL LPS (CM^LPS) was used as positive control. Alizarin Red mineralization was quantified by spectrophotometry at 405 nm. (B–H) RUNX2, SOX9, BMP2, ALPL, COL1A1, COL1A2, and ATF4 were measured by qRT-PCR. Each dot represents a separate patient biopsy. Bars are mean ±SD; p-Values were calculated by one-way ANOVA. ANOVA, analysis of variance; CM, conditioned medium; FAPs, fibro-adipogenic progenitors; IL-6, interleukin-6; LPS, lipopolysaccharides; OB, osteogenic medium; qRT-PCR, quantitative real-time polymerase chain reaction; sIL6RA, soluble IL-6 receptor α chain; SD, standard deviation.

Human FAPs express high levels of OSMR and IL6ST but low levels of IL6R

Finally, we measured IL6R, OSMR, and IL6ST mRNA expression in human FAPs from patients and in monocytes from healthy donors (Fig. 4). Similar to the mouse, OSMR and IL6ST mRNA were highly expressed by FAPs with undetectable to low levels, respectively, in monocytes. In contrast IL6R mRNA was highly expressed in monocytes with very low levels in FAPs.

Discussion

Our data suggest that unlike its related cytokine OSM,¹⁶ IL-6 has little role in the pathogenesis of NHO. Indeed, prophylactic treatment of mice that underwent SCI and muscle injury with neutralizing doses of function-blocking anti-mouse IL-6 and anti-IL6RA monoclonal antibodies had no effect on subsequent NHO development in vivo. Likewise, recombinant human IL-6 alone or in combination with sIL6RA did not increase osteogenic differentiation of human FAPs in vitro either directly or indirectly via monocytes. Altogether, these observations suggest that unlike OSM and IL-1, IL-6 is unlikely to have a significant role in NHO pathogenesis in humans or mice.

These results were unexpected because IL-6 expression is enhanced in injured muscles developing NHO after SCI, similar to OSM, IL-1α, and IL-1β which do exert a role in NHO pathogenesis.^16,19 Similar to OSM, IL-1α, and IL-1β, IL-6 is also produced by activated monocytes/macrophages. Furthermore, IL-6 through its interaction with IL6RA and GP130 activates JAK kinases and STAT transcription factors^20,21 similar to the interaction between OSM, OSMR, and GP130, and treatment with the JAK1/2 inhibitor ruxolitinib significantly reduces NHO development after SCI in mice.²⁶ The major difference between OSM and IL-6 is the expression pattern of their respective receptor α chains OSMR and IL6RA. Consistent with publicly available gene expression databases, we find that while mouse and human OSMR and IL6ST (GP130) mRNA are abundantly expressed in muscle mesenchymal progenitor cells, IL6RA is expressed at low levels. In theory, IL-6 should be able to signal in cells that do not express IL6RA as long as GP130 is expressed via the non-canonical pathway in which IL-6 binds to sIL6RA in solution to form a soluble complex that then binds to and causes dimerization of transmembrane GP130 and initiates signaling through JAK1, JAK2, and TYK2 kinase activation^20,31 and additional signaling pathways such as RAS, MAP kinase, and phosphoinositol-3 kinase / protein kinase B (AKT) pathways,^32,33 which are also activated by OSM.²³ However, because of the high stochiometric ratio of OSMR over IL6RA chains at the surface of muscle mesenchymal cells as suggested by mRNA expression data, most of GP130 chains would be preferentially engaged with OSMR and OSM rather than with IL6RA. Furthermore, the interaction of OSM with its receptor is unusual as it can directly bind to GP130 and then recruit the OSMR α chain,^34,35 whereas in the case of IL-6, it must first bind to transmembrane IL6RA or sIL6RA to be able to induce GP130 dimerization and signaling. Therefore, both the high stochiometric ratio of OSMR over IL6RA and the unique way OSM binds to GP130 and then OSMR may contribute to the potent effect of OSM and the lack of effect of IL-6 in inducing osteogenic differentiation of FAPs.

In conclusion, it appears that although IL-6 is strongly induced in the circulation and injured muscles after SCI (as is OSM), IL-6 is unlikely to play a major role in NHO pathogenesis. Therefore, while treatment with JAK kinase inhibitors may provide some benefits to reduce NHO development in victims of CNS injuries, IL-6 or ILRA neutralizing antibodies such as tocilizumab, which is approved for the treatment of various forms of arthritis and cytokine release syndrome, are unlikely to have any benefit to reduce NHO development.

Footnotes

Acknowledgments

The authors acknowledge the assistance of the Translational Research Institute flow cytometry and preclinical imaging facilities and the University of Queensland biological resources facility.

Authors’ Contributions

D.G., H.-W.T., S.B., J.-P.L., and K.A.A. conceived the study, designed the experiments, and interpreted the results. D.G., H.W.T., B.R., V.B., S.S., S.G.S., M.-E.G., and K.A.A. performed the experiments and acquired data. F.G. and M.S. procured human biopsies for the study. D.G., S.G., J.-P.L., and K.A.A. wrote and edited the article.

Funding Information

This work was funded by award W81XWH21105560 from the Congressionally Approved Spinal Cord Injury Research Program of the U.S. Department of Defense to J.P.L., K.A.A., H.-W.T., S.B., and F.G., and by the Mater Foundation.

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Abbreviations Used

References

Brady

, Shultz

, McDonald

, et al. Neurological heterotopic ossification: Current understanding and future directions. Bone, 2018; 109(2018):35–42; doi: 10.1016/j.bone.2017.05.015

Ohlmeier

, Suero

, Aach

, et al. Muscle localization of heterotopic ossification following spinal cord injury. Spine J, 2017; 17(10):1519–1522; doi: 10.1016/j.spinee.2017.04.021

Genet

, Jourdan

, Schnitzler

, et al. Troublesome heterotopic ossification after central nervous system damage: A survey of 570 surgeries. PLoS One, 2011; 6(1):e16632; doi: 10.1371/journal.pone.0016632

Garland

, Orwin

. Resection of heterotopic ossification in patients with spinal cord injuries. Clin Orthop Relat Res, 1989(242):169–176.

Vanden Bossche

, Vanderstraeten

. Heterotopic ossification: A review. J Rehabil Med, 2005; 37(3):129–136.

Almangour

, Schnitzler

, Salga

, et al. Recurrence of heterotopic ossification after removal in patients with traumatic brain injury: A systematic review. Ann Phys Rehabil Med, 2016; 59(4):263–269; doi: 10.1016/j.rehab.2016.03.009

Alexander

, Tseng

H-W

, Salga

, et al. When the nervous system turns skeletal muscles into bones: How to solve the conundrum of neurogenic heterotopic ossification. Curr Osteoporos Rep, 2020; 18(6):666–676; doi: 10.1007/s11914-020-00636-w

Genêt

, Minooee

, Jourdan

, et al. Troublesome heterotopic ossification and stroke: Features and risk factors. A case control study. Brain Inj, 2015; 29(7–8):866–871; doi: 10.3109/02699052.2015.1005133

Genet

, Chehensse

, Jourdan

, et al. Impact of the operative delay and the degree of neurologic sequelae on recurrence of excised heterotopic ossification in patients with traumatic brain injury. J Head Trauma Rehabil, 2012; 27(6):443–448; doi: 10.1097/HTR.0b013e31822b54ba

10.

Genet

, Marmorat

, Lautridou

, et al. Impact of late surgical intervention on heterotopic ossification of the hip after traumatic neurological injury. J Bone Joint Surg Br, 2009; 91(11):1493–1498; doi: 10.1302/0301-620X.91B11.22305

11.

Mounier

, Théret

, Arnold

, et al. AMPKα1 regulates macrophage skewing at the time of resolution of inflammation during skeletal muscle regeneration. Cell Metab, 2013; 18(2):251–264; doi: 10.1016/j.cmet.2013.06.017

12.

Arnold

, Henry

, Poron

, et al. Inflammatory monocytes recruited after skeletal muscle injury switch into antiinflammatory macrophages to support myogenesis. J Exp Med, 2007; 204(5):1057–1069.

13.

Lemos

, Babaeijandaghi

, Low

, et al. Nilotinib reduces muscle fibrosis in chronic muscle injury by promoting TNF-mediated apoptosis of fibro/adipogenic progenitors. Nat Med, 2015; 21(7):786–794; doi: 10.1038/nm.3869

14.

Genêt

, Kulina

, Vaquette

, et al. Neurological heterotopic ossification following spinal cord injury is triggered by macrophage-mediated inflammation in muscle. J Pathol, 2015; 236(2):229–240; doi: 10.1002/path.4519

15.

Tseng

H-W

, Girard

, Alexander

, et al. Spinal cord injury reprograms muscle fibroadipogenic progenitors to form heterotopic bones within muscles. Bone Res, 2022; 10(1):22; doi: 10.1038/s41413-022-00188-y

16.

Torossian

, Guerton

, Anginot

, et al. Macrophage-derived oncostatin M contributes to human and mouse neurogenic heterotopic ossifications. JCI Insight, 2017; 2(21):e96034; doi: 10.1172/jci.insight.96034

17.

Alexander

, Tseng

H-W

, Kulina

, et al. Lymphocytes are not required for neurogenic heterotopic ossification development after spinal cord injury. Neurotrauma Rep, 2022; 3(1):87–96; doi: 10.1089/neur.2021.0072

18.

Tseng

H-W

, Kulina

, Salga

, et al. Neurogenic heterotopic ossifications develop independently of granulocyte colony-stimulating factor and neutrophils. J Bone Miner Res, 2020; 35(11):2242–2251; doi: 10.1002/jbmr.4118

19.

Tseng

H-W

, Kulina

, Girard

, et al. Interleukin-1 is overexpressed in injured muscles following spinal cord injury and promotes neurogenic heterotopic ossification. J Bone Miner Res, 2022; 37(3):531–546; doi: 10.1002/jbmr.4482

20.

Heinrich

, Behrmann

, Muller-Newen

, et al. Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J, 1998; 334(Pt 2):297–314; doi: 10.1042/bj3340297

21.

Jones

, Jenkins

. Recent insights into targeting the IL-6 cytokine family in inflammatory diseases and cancer. Nat Rev Immunol, 2018; 18(12):773–789; doi: 10.1038/s41577-018-0066-7

22.

Sims

. The JAK1/STAT3/SOCS3 axis in bone development, physiology, and pathology. Exp Mol Med, 2020; 52(8):1185–1197; doi: 10.1038/s12276-020-0445-6

23.

Sims

, Lévesque

J-P

. Oncostatin M: Dual regulator of the skeletal and hematopoietic systems. Curr Osteoporos Rep, 2024; 22(1):80–95; doi: 10.1007/s11914-023-00837-z

24.

Karatas

, Erdi

, Kaya

, et al. Neuroprotective effects of tocilizumab on experimentally-induced spinal cord ischemia-reperfusion injury. World Neurosurg, 2018; 124:e208–e213; doi: 10.1016/j.wneu.2018.12.069

25.

Luo

, Yao

, Shi

, et al. Tocilizumab promotes repair of spinal cord injury by facilitating the restoration of tight junctions between vascular endothelial cells. Fluids Barriers CNS, 2023; 20(1):1; doi: 10.1186/s12987-022-00399-9

26.

Alexander

, Tseng

H-W

, Fleming

, et al. Inhibition of JAK1/2 tyrosine kinases reduces neurogenic heterotopic ossification after spinal cord injury. Front Immunol, 2019; 10:377; doi: 10.3389/fimmu.2019.00377

27.

Bisht

, McGirr

, Lee

S-Y

, et al. Oncostatin M regulates Hematopoietic Stem Cell (HSC) niches in the bone marrow to restrict HSC mobilization. Leukemia, 2022; 36(2):333–347; doi: 10.1038/s41375-021-01413-z

28.

Barber

, Andrade

, McBerry

, et al. Role of IL-6 in Mycobacterium avium–associated immune reconstitution inflammatory syndrome. J Immunol, 2014; 192(2):676–682; doi: 10.4049/jimmunol.1301004

29.

Tsukamoto

, Senju

, Matsumura

, et al. IL-6-mediated environmental conditioning of defective Th1 differentiation dampens antitumour immune responses in old age. Nat Commun, 2015; 6:6702; doi: 10.1038/ncomms7702

30.

Salga

, Samuel

, Tseng

, et al. Bacterial lipopolysaccharides exacerbate neurogenic heterotopic ossification development. J Bone Miner Res, 2023; 38(11):1700–1717; doi: 10.1002/jbmr.4905

31.

Reeh

, Rudolph

, Billing

, et al. Response to IL-6 trans- and IL-6 classic signalling is determined by the ratio of the IL-6 receptor α to gp130 expression: Fusing experimental insights and dynamic modelling. Cell Commun Signal, 2019; 17(1):46; doi: 10.1186/s12964-019-0356-0

32.

Guo

, Xu

, Lu

, et al. Interleukin-6 signaling pathway in targeted therapy for cancer. Cancer Treat Rev, 2012; 38(7):904–910; doi: 10.1016/j.ctrv.2012.04.007

33.

Hsu

, Shi

, Frost

, et al. Interleukin-6 activates phosphoinositol-3′ kinase in multiple myeloma tumor cells by signaling through RAS-dependent and, separately, through p85-dependent pathways. Oncogene, 2004; 23(19):3368–3375; doi: 10.1038/sj.onc.1207459

34.

Gearing

, Comeau

, Friend

, et al. The IL-6 signal transducer, gp130: An oncostatin M receptor and affinity converter for the LIF receptor. Science, 1992; 255(5050):1434–1437; doi: 10.1126/science.1542794

35.

Mosley

, De Imus

, Friend

, et al. Dual oncostatin M (OSM) receptors. Cloning and characterization of an alternative signaling subunit conferring OSM-specific receptor activation. J Biol Chem, 1996; 271(51):32635–32643; doi: 10.1074/jbc.271.51.32635

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.54 MB

0.57 MB

0.56 MB

0.57 MB

0.89 MB

Interleukin-6 Does Not Contribute to Neurogenic Heterotopic Ossification Pathogenesis

Abstract

Introduction

Materials and Methods

Animals

SCI-induced NHO mouse model

Animal treatments

Mouse tissue collection

Purification of mouse muscle progenitors and leukocytes for RNA extraction

Mouse RNA extraction and quantitative real-time polymerase chain reaction

Mouse blood collection for quantification of plasma cytokines

Micro-computerized tomography and NHO volume quantification

Human ethics statement and samples collection

Isolation of human FAPs

In vitro osteogenic differentiation assay and mineralization quantification

Human cell RNA extraction and RT-qPCR

Human CD14+ monocytes isolation and preparation of conditioned media

Statistical analysis

Results

IL-6 is over-induced in injured muscles after SCI

Neutralizing antibodies for IL-6 and IL6RA does not reduce NHO development in mice

IL-6 does not induce osteogenic differentiation of NHO-associated human FAPs in vitro

Human FAPs express high levels of OSMR and IL6ST but low levels of IL6R

Discussion

Footnotes

Acknowledgments

Authors’ Contributions

Funding Information

Author Disclosure Statement

Supplementary Material

Abbreviations Used

References

Supplementary Material

Human CD14⁺ monocytes isolation and preparation of conditioned media