Macrophage-Derived Extracellular DNA Initiates Heterotopic Ossification

Heterotopic ossification (HO) severely affects people's lives; however, its pathological mechanism remains poorly understood. Although extracellular DNA (ecDNA) has been shown to play important roles in pathological calcification, its effects in HO development and progression remain unknown. The in vivo rat Achilles tendon injury model and in vitro collagen I calcification model were used to evaluate the effects of ecDNA in the ectopic calcifications and the main cell types involved in those pathological process. Histology, immunofluorescent staining, reverse transcriptase-polymerase chain reaction analysis and micro-computed tomography were used to identify the distribution of macrophage-derived ecDNA and elucidate their roles in HO. The results showed that the amount of ecDNA and ectopic calcification increased significantly and exhibited a strong correlation in the injured tendons of HO model compared with those of the controls, which was accompanied by a significantly increased number of M2 macrophages in the injured tendon. During in vitro co-culture experiments, M2 macrophages calcified the reconstituted type I collagen and ectopic bone collected from the injured tendons of HO rats, while those effects were inhibited by deoxyribonuclease. More importantly, deoxyribonuclease reversed the pathological calcification in the injured rat tendon HO model. The present study showed that ecDNA from M2 macrophages initiates pathological calcification in HO, and the elimination of ecDNA might be developed into a clinical strategy to prevent ectopic mineralization diseases. The use of deoxyribonuclease for the targeted degradation of ecDNA at affected tissue sites provides a potential solution to treat diseases associated with ectopic mineralization.


INTRODUCTION
Heterotopic ossification (HO) is a disease comprising abnormal bone formation in extraskeletal soft tissue [1], and its principal clinical signs include localized pain, swelling, joint stiffness, and limited range of motion [2].Currently, treatment strategies for HO are limited to local radiation therapy, the use of bisphosphonates, nonsteroidal anti-inflammatory drugs, and surgical resection in severe cases [3][4][5][6].However, these treatments have problems of low curative effect and high recurrence rate, which brings a huge economic burden to patients.Therefore, clarifying the underlying pathological mechanism of HO is important to investigate novel clinical strategies for HO therapy.
DNA is generally considered as the carrier of genetic information; however, recent studies showed that extracellular DNA (ecDNA), identified in microscopical calcifications derived from breast cancer [7], gallstones [8], and atherosclerotic plaques [9,10], was associated with the occurrence of the ectopic calcification of calcium phosphate.Furthermore, our recent study showed that ecDNA, because of its polyanionic property, is capable of stabilizing supersaturated calcium phosphate solution and functions as an initiator of collagen intrafibrillar mineralization, leading to the complete mineralization of collagen matrices [11].The mechanism of ecDNA in the mineralization of collagen is attributed to the relatively stable formation of amorphous liquid droplets triggered by attraction of DNA to the collagen fibrils via hydrogen bonding [11].Notably, tissue injury, one of the most common causative factors of various types of HO, leads to massive cell death and the release of DNA fragments into the extracellular matrix during HO progression [12].However, the major cell types that deliver ecDNA after tissue injury and the pathogenic roles of ecDNA on the development of HO are still unknown.
Tissue injury at the beginning of HO leads to a well-orchestrated inflammatory response, with the recruitment of macrophages, neutrophils, and lymphocytes [13].While it is difficult to attribute HO pathogenesis and progression to a specific type of inflammation, it is important to point out that macrophages serve as the most critical sentinels and regulators of the immune system during inflammation [14,15].Macrophages can differentiate into different phenotypes within different micro-environments, namely, the M1-like phenotype (pro-inflammatory) and M2-like phenotype (anti-inflammatory), based on cell surface markers [16].The study has reported the crucial roles of macrophages in HO formation, particularly in the initiation stage, because selective depletion of macrophages significantly prohibited ectopic bone formation in a rat Achilles tendon injury model [17].Furthermore, a recent study showed that macrophages were present and persisted throughout the ectopic bone formation process, suggesting the critical role played by macrophages in HO formation [13].Nucleic acids can be released from cells in extracellular vesicles (EVs) and macrophage-derived EVs promote the deposition of calcium orthophosphate and atherosclerotic plaque formation, suggesting that macrophages might serve as the origins of ecDNA within the injured tissue of HO [10].However, the phenotypic heterogeneity of the macrophage population and roles of macrophage-derived ecDNA during HO development remain unclear.
In this study, the in vivo rat Achilles tendon injury model was used to investigate the associations of ecDNA and ectopic calcification formation at the site of the injured tendon, and the phenotypic heterogeneity of macrophages was further tested during HO development.Furthermore, in vitro calcification assays were carried out by co-culturing macrophages with the reconstituted type I collagen or ectopic bone collected from the HO rats to evaluate the role of macrophages in ecDNA-induced ectopic calcification.Finally, deoxyribonuclease treatment was used in the in vitro and in vivo ectopic bone formation models to further confirm the effects of macrophage-derived ecDNA in the ectopic calcification during HO development.Our findings propose a novel paradigm to understand the functional impact of ecDNA in the progression of ossification in ectopic mineralization diseases.

Experimental Animals
Male Sprague-Dawley rats (8-weeks old, weighing 200-300 g) were purchased from the Laboratory Animal Research Centre of the Fourth Military Medical University, and used for all in vivo experiments in the present study.All surgical procedures used in those experiments were approved by the Institutional Animal Care and Use Committee of the Fourth Military Medical University (ethics approval number: 20220906).All animals were kept under specific-pathogen-free conditions and fed with sterilized food and distilled water during the study.

The Achilles Tenotomy Model and Group Designation
To examine the mechanism of HO osteogenesis, an Achilles tenotomy model was created in rats.After each rat anesthetized by intraperitoneal injection of sodium pentobarbital (40 mg kg −1 ), the heels of both hind legs were shaved and sterilized, then the skin and subcutaneous tissue were incised to expose the Achilles tendon.In the HO group, the Achilles tendon was transected along the midpoint and the ends of the tendon were clamped repeatedly about five times.In the sham control group, the Achilles tendon was not transected.Each operation was completed by the same experimenter within 10 min and all efforts were made to minimize suffering.None of the rats died during the entire experimental period and no differences were noticed in the ectopic bone formation between the right and left legs.
In the first run of the experiment, the rats were euthanized after 3, 6, and 9 weeks.The left and right Achilles tendons from rats were fixed using 4% paraformaldehyde for 24 h, decalcified using 10% ethylenediaminetetraacetic acid (EDTA; pH 7.3) for 2 weeks (the demineralization medium was changed every two days).Then, the left Achilles tendons were dehydrated in an ascending graded series of ethanol, embedded in paraffin, and processed into 6 μm-thick sections.These sections were used for hematoxylin-eosin (HE) staining (n = 3).The right Achilles tendons were dehydrated using 30% sucrose at 4 °C for 48 h.Next, the specimens were embedded in Tissue-Tek optimal cutting temperature (Sakura, Netherlands) and processed into 5 μm-thick sections using a cryogenic microtome for immunofluorescence staining (n = 3).In the second run of the experiment, the rats were also euthanized after 3, 6, and 9 weeks.Gene expression in the Achilles tendons was analyzed using quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR), and a single sample was obtained by pooling together every two out of six Achilles tendons samples (n = 3).In addition, we investigated the ectopic bone masses of the Achilles tendons for surface-associated ecDNA at 6 and 9 weeks after surgery (n = 3).In the third run of the experiment, the rats were euthanized after 9 weeks.The left Achilles tendons were used to investigate surface-associated ecDNA of ectopic bone masses cultured with M0 or M2 macrophages in osteogenic medium (n = 3).The right Achilles tendons were used to investigate further mineralization of ectopic bone masses in vitro (n = 3).In the fourth run of the experiment, we inhibited the occurrence of ectopic ossification by local injection of DNase-I in the third week and injected every three days as well as the rats were also euthanized after 6 weeks.The left Achilles tendons with the heel (calcaneus) and lower tibia were dissected from rats and fixed using 4% paraformaldehyde for 24 h, before being scanned and analyzed using micro-computed tomography (Micro-CT; Inveon, Siemens Preclinical, Knoxville, TN, USA) at high-resolution (n = 3).The right Achilles tendons were used for HE staining and immunofluorescence staining (n = 3).

Preparation of 2D Single-Layer Reconstituted Collagen
Collagen fibrils were self-assembled from rat tail tendon-derived collagen dissolved in acetic solution (5 mg mL −1 ) using the dialysis method.The collagen solution was dripped onto 12-well chambers and air dried at room temperature.

Hematoxylin-Eosin (HE) Staining
For HE staining, sections were deparaffinized and rehydrated in xylene and a declining graded series of ethanol, stained with hematoxylin for 7 min, rinsed in running tap water for 5 min, differentiated with 1% hydrochloric acid-ethanol for 5 s, rinsed in running tap water for 1 min, and stained eosin for 3 min.All sections were rinsed and then mounted using resinous medium.
M1 or M2 macrophages seeded on 12-well chamber slides (n = 6) were fixed using 4% paraformaldehyde and incubated with goat serum for 30 min at room temperature.The cell samples were exposed to the following primary antibodies at 4 ℃ for 12 h: anti-iNOS antibody (1/100, ab15323, Abcam) and anti-MRC1 antibody (1/100, ab64693, Abcam).After washing, secondary antibodies were added and incubated for 1 h in the dark.After washing, the cells were mounted with DAPI for CLSM.
The images were captured using CLSM and the integrated fluorescence intensity was analyzed using ImageJ software (NIH, Bethesda, MD, USA).For each sample, three fields of view were selected randomly and the average fluorescence intensity was calculated as the data using Image J software.

Quantitative Real-Time Reverse Transcription Polymerase Chain Reaction (qRT-PCR)
To investigate the influence of macrophages on the formation of HO, gene expression levels of Inos, Il1b (encoding interleukin-1 beta), Tnfa (encoding tumor necrosis factor alpha), Mrc1, Retnlb (encoding resistin-like beta, also known as FIZZ1) and Arg1 (encoding Arginase 1) of the Achilles tendon tissue in trauma areas were evaluated using qRT-PCR.Total RNA was extracted using the Tripure Isolation Reagent (11,667,165,001, Roche, San Francisco, CA, USA) according to the manufacturer's instructions.Gapdh (encoding glyceraldehyde-3-phosphate dehydrogenase) was used as the housekeeping gene.
The concentration and purity of the extracted RNA were determined by measuring the absorbance at 260 and 280 nm (BioTek, Winooski, VT, USA).cDNA was synthesized using a PrimeScript RT reagent kit (Takara Bio Inc., Shiga, Japan).The quantitative real-time polymerase chain reaction (qPCR) was performed using the cDNA as the template on a 7500 Real Time PCR System (Applied Biosystems, Foster City, CA, USA).Results obtained after calibration using the Gapdh expression level were calculated using the 2 −ΔΔCt method and presented as fold increases relative to the non-stimulated control (technical replicates n = 3 for each group) [18].Sense and antisense primers were designed based on the published cDNA sequences using Primer Express 5.0 (Thermo Fisher Scientific, Waltham, MA, USA; Supplementary Table 1).

Mineralization of Ectopic Bone Masses
The ectopic bone masses were fixed using 4% paraformaldehyde, rinsed with PBS, and prepared by ultrasonic cleaning and autoclaving.CaCl 2 •2H 2 O solution (1.6 mm) was mixed with an equal volume of K 2 HPO 4 solution (1 mm) to provide materials for mineralization.The pH of the final mineralization medium was adjusted to 7.0 and the prepared media were clear, without precipitation.The ectopic bone masses were cultured with M0 or M2 macrophages in osteogenic medium for 10 days and then incubated in the CaCl 2 •2H 2 O and K 2 HPO 4 medium for Alizarin red S staining for 14 days.

Micro-Computed Tomography (Micro-CT)
Briefly, samples were scanned at 80 kV and 500 μA.Two-dimensional slices with 39 μm isotropic resolution were generated with a three dimensional (3D) image reconstruct based on the scanned information using the Inveon Research Workplace software (Siemens Medical Solutions USA, Inc., Hoffman Estates, IL, USA).A region of interest was positioned in the injury site and the bone mineral density (BMD) and bone volume to total volume ratio (BV/TV) were measured using the Inveon Research Acquisition software.

Alizarin Red S Staining of Calcification
Mineralized nodules were analyzed after culturing the cells from different groups in medium for 5, 7, and 14 days.The cells were stained with Alizarin red S (40 mmol L −1 , pH 4.2) after fixation with 10% formaldehyde.After 15 min, excessive dye was removed by washing with distilled water.The images were captured under a Zeiss microscope (Thornwood, NY, USA) and the positive area was analyzed using ImageJ software.

Statistical Analysis
Statistical analyses were performed using the Graph-Pad Prism 5 package (GraphPad Software, La Jolla, CA, USA).All data are expressed as the mean ± standard deviation.Data acquired for each assay were evaluated for their normality and homoscedasticity assumptions before the use of parametric statistical methods.When these assumptions were not violated, the data were analyzed using a t-test (two groups) or one-way analysis of variance (ANOVA) (more than two groups).To ensure the validity of observations, all quantitative experiments were repeated at least three times.Statistical significance was preset at α = 0.05.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).

Increased ecDNA Deposition Associated with Ectopic Calcification During HO Progression
HE staining was used to observe the features of ectopic bone formation in the rat Achilles tendon injury model.At 3 weeks post-surgery, disorganized local cells and infiltrated inflammatory cells were observed within the injured tendon; however, ectopic bone could not be seen.Immature woven bone at 6 weeks post-surgery and a developed cancellous bone at 9 weeks post-surgery were clearly identified in the HO groups, but were not observed in the sham groups (Fig. 1a).Quantitative analysis of HE staining revealed that the BV/TV ratios in the 9-week HO groups were significantly higher than those in the other groups (p < 0.05, Fig. 1b).These results demonstrated that the Achilles tenotomy model successfully simulated the occurrence of HO.
Immunofluorescent staining of ectopic bone mass and the sections of injured tendon tissues from HO rats were then used to observe the distribution of ecDNA and their association with the ectopic bone formation.Stereomicroscopy observation of the ectopic bone masses prepared from the 6 and 9 week HO rats demonstrated a large amount of ecDNA attached onto the calcified tissues, and the amount of ecDNA was significantly higher in the 9-week HO rats than in the 6-week HO rats (p < 0.05, Fig. 1c, d).Likewise, as shown in Fig. 2a, the calcified regions (red) and ecDNA (arrows) showed little distribution in the sham control and 3-week HO groups, but gradually increased in the 6-week HO group, reaching their highest levels in the 9-week HO group.The increased ecDNA in the injured tendons from the 6-and 9-week HO groups was located near to the calcified regions.The areas of calcified regions and ecDNA both increased significantly in the 9-week HO groups compared with those in the other groups (p < 0.05, Fig. 2b, c), while those in 6-week HO groups were significantly higher than those in 3-week HO group and sham control group (p < 0.05).Analysis of the Pearson correlation coefficient showed a strong correlation between areas of ecDNA and calcified regions in the 6-week and 9-week HO groups (p < 0.05, r = 0.98, Fig. 2d).These data suggested that increased ecDNA deposition is associated with ectopic calcification during HO progression.

M2 Macrophages Infiltrated Around the Ectopic Bone During HO Progression
Identification of the association between ecDNA and ectopic calcification in vivo prompted us to explore the origin of the ecDNA and its effect on the mineralized tissue.Macrophages and neutrophils are the most critical sentinels and regulators to inflammation; therefore, we hypothesized that ecDNA from macrophages or neutrophils participates in HO development.Immunofluorescent staining and qRT-PCR were conducted to detect the infiltration of different types of macrophages and neutrophils in the process of HO.The results showed that CD68 + cells started to increase in large numbers in the 3-week HO group, reaching a maximum in the 9-week HO group (Fig. 3a), but we found no neutrophils expression in around the mineralized zone (Supplementary Fig. 1).In addition, the amount of CD68 + cells was significantly higher in the HO groups than in the sham controls (p < 0.05, Fig. 3b), and the amount of CD68 + cells in the 9-week HO group was significantly higher than that in the 3and 6-week groups (p < 0.05).So, we think that ecDNA from macrophages participates in HO development.To further determine the effects of M1 or M2 macrophages during HO development, we detected iNOS + cells (M1 macrophages) and MRC1 + cells (M2 macrophages) using immunofluorescent staining.The results showed that iNOS + cells were largely absent in the 3-, 6-and 9-week HO groups and sham controls groups, and did not exhibit any difference among the HO and control groups (p > 0.05).The trends of the number of MRC1 + cells among groups were identical to those of CD68 + cells, exhibiting a significant increase in the HO groups compared with those in the sham controls, and reaching a maximum in the 9-week HO group (Fig. 3b).In accordance with the immunofluorescence results for macrophages, the mRNA levels of M1-related genes (Inos, Il1b, and Tnfa) showed no significant difference among the HO groups and their sham controls (p > 0.05), while those of M2-related genes (Mrc1, Retnlb, and Arg1) in the HO groups increased significantly compared with those in the controls (all p < 0.05, Fig. 3c).These results suggested that M2 macrophages might be one of the origins of ecDNA and participate in the ectopic calcification in HO.

M2 Macrophage-Derived ecDNA Promotes Type I Collagen Mineralization
The effects of M2 macrophages-derived ecDNA in ectopic calcification were further investigated by coculturing M0 and M2 macrophages with reconstituted type I collagen to identify ectopically mineralized nodules in vitro.Immunofluorescence staining and qRT-PCR were conducted to confirm the polarization of naïve (M0) macrophages into M2 phenotypes (Supplementary Fig. 2).
The collagen mineralization induced by naïve (M0) macrophages or M2 macrophages in osteogenic medium was detected using Alizarin red staining (Fig. 4a and b).Collagen fibers co-cultured with M0 macrophages were not mineralized at any time point.Collagen fibers co-cultured with M2 macrophages were apparently mineralized at 5 days and the degree of mineralization increased with time.The degree of collagen calcification with M2 macrophages was significantly higher than that in the M0 macrophage group, and reached the highest level at 14 days compared with that at 5 or 7 days; the M0 macrophage group had no obvious mineralization (p < 0.05).Treatment of the M2 macrophages with DNase-I resulted in a significant reduction of the collagen mineralization at 5, 7, or 14 days (p < 0.05).(Fig. 4c).Taken together, our findings reveal a critical role for M2 macrophages in ecDNAinduced ectopic calcification.

M2 Macrophage-Derived ecDNA Adhered on Ectopic Bone Masses and Induced Further Mineralization ex vivo
To confirm the critical role of M2 macrophages in ecDNA-induced ectopic calcification, ectopic bone masses prepared from the 9-week HO rats were cultured with M0 or M2 macrophages in osteogenic medium for 10 days (Fig. 5a, b).Stereomicroscopy of the ectopic bone masses demonstrated that ecDNA on the surface of the ectopic bone was significantly increased when cultured with M2 macrophages, compared with that cultured with M0 macrophages (p < 0.05).After treatment of the M2 macrophages with DNase-I, ecDNA was significantly reduced (p < 0.05) (Fig. 5c).To clarify the effect of ecDNA on the mineralized tissue, the ectopic bone masses co-cultured with M0 or M2 macrophages were incubated in 1.6 mm CaCl 2 •2H 2 O and 1 mm K 2 HPO 4 medium for 14 days and stained using Alizarin red  ◂ Macrophage-Derived Extracellular DNA Initiates Heterotopic Ossification staining and ecDNA fluorescent staining (Fig. 6a).After incubation, few calcified nodules and little ecDNA were observed on the surface of ectopic bone masses cultured with M0 macrophages, while abundant calcified nodules and ecDNA were identified on the surface of ectopic bone masses cultured with M2 macrophages.DNase-I treatment reversed the calcified nodules formation in the M2 macrophage co-culture groups (Fig. 6b).The extent of calcification assessed in the M2 macrophage group was significantly higher compared with that of the ectopic bone mass cultured with M0 or M2 macrophages treated with DNase-I (p < 0.05, Fig. 6c).Pearson correlation coefficient analysis also demonstrated a strong correlation between ecDNA and calcified deposition (p < 0.05, r = 0.98, Fig. 6d).

Deoxyribonuclease Treatment Reversed the Ectopic Bone Formation in the Rat Achilles Tenotomy HO Model
DNase-I treatment was designed to investigate the role of ecDNA in the occurrence of HO. (Fig. 7a).Data from Micro-CT analysis showed that heterotopic bone was formed by 6 weeks after tenotomy injected with PBS, while the injection of DNase-I resulted in no heterotropic bone formation (Fig. 7b).H&E staining of the HO specimens revealed thick cartilage layers adjacent to cancellous bone and marrow in the HO group injected with PBS.By contrast, the area of HO bone marrow was significantly decreased after treatment with DNase-I.Alizarin red staining of the sections further confirmed that blocking the release of ecDNA in vivo using DNase-I effectively reversed pathological calcification in the progression of HO (Fig. 7c,  d).Statistical analysis also showed that the bone mineral density (BMD), BV/TV, and the fluorescent intensity of Alizarin red in the HO group treated with DNase-I were significantly lower than those in the group treated with PBS (Fig. 7e, f).Altogether, these data indicated that inhibition of ecDNA activity by DNase-I treatment effectively reversed HO in the Achilles tenotomy model.

DISCUSSION
Despite our understanding that extracellular nucleic acids mineralize collagen matrices and the discovery of ecDNA deposition in calcified tendon regions, little is known about the contribution of ecDNA to the occurrence of HO [11].In this study, we utilized the rat Achilles tendon injury model and showed that the amounts of ecDNA and ectopic calcification both increased during HO progression and exhibited a strong correlation, accompanied by increased M2 macrophage infiltration.The novel mechanism, that M2 macrophages-derived ecDNA is responsible for the ectopic bone formation, was validated by the in vitro calcification assay in which co-culturing macrophages reconstituted type I collagen or ectopic bone mass.Furthermore, degrading ecDNA using deoxyribonuclease reversed M2 macrophage-induced calcification in vitro and the ectopic bone formation in the injured rat tendon HO model.Taken together, we demonstrated that ecDNA derived from M2 macrophages initiates ectopic calcification and serves as a potential therapeutic target for HO.
Recent evidence identified ecDNA on the microscopical calcifications from atherosclerotic plaques, gallstones, and breast cancer, suggesting that free forms of ecDNA are involved in ectopic mineralization [7][8][9].In our results, calcified regions and ecDNA showed little or no distribution in the sham group and 3-week HO group, but gradually increased in the 6-week HO group, reaching their highest levels in the 9-week HO group.The amount of ecDNA and calcified deposition in the injured tendon tissues of 6-week and 9-week HO rats and in in vitro M2 macrophage-induced ectopic bone formation were strongly correlated, suggesting that ecDNA mediates the ectopic bone formation in HO.Inhibition of ecDNA activity by DNase-I treatment effectively reversed the ectopic bone formation in vivo and in vitro, further confirming the independent role of ecDNA in ectopic calcification.These results point to the potential of ecDNA as a mineralization initiator for inducing calcium phosphate deposition in HO.In the early stages of human atheroma, free DNA was also identified as a potential nidus for calcium phosphate precipitation and hydroxyapatite crystallization [19].The polyanionic property of ecDNA molecules mean that they are capable of stabilizing calcium and phosphate, promoting the relatively stable formation of amorphous liquid droplets, and inducing ectopic mineralization in different body tissues [20][21][22][23].Although HO is believed to develop through a process of endochondral ossification histologically, our results reveal that ecDNAinduced ectopic calcium and phosphate deposition also contributes to the onset and progression of HO.It has been reported that nucleic acids are not inert molecules, but trigger diverse biological reactions [24].They are involved in innate immune reactions related to bacterial killing by neutrophil extracellular traps, and serve as the damaging factor in cardiovascular pathologies, as well as thrombus and edema formation [25,26].Therefore, further studies are need to explore the other pathogenic effects ecDNA during HO progression.More importantly, our results supported the view that deoxyribonuclease is the powerhouse strategy in solving the issue of unwarranted ectopic collagen mineralization.From a future perspective, harnessing ecDNA is suggested as a promising therapeutic strategy for pathological calcification disease.
Upon injury, circulating monocytes from peripheral blood respond and infiltrate into the injury site, where they undergo the monocyte-to-macrophage transition and polarization, secreting a plethora of proinflammatory mediators, TNF-α, IL-1, and transforming growth factor β (TGF-β), creating an osteogenesis-promoting niche for multipotent cell accumulation and differentiation into mature bone cells [11].In our study, we demonstrated that M2 macrophages are responsible for the pathogenesis of ectopic mineralization through delivering ecDNA to the ossification site.Our results showed enrichment of M2 macrophages, but not M1 macrophages, in the injured areas of HO.M1 macrophages commonly initiate inflammatory responses immediately after various injuries [27,28]; therefore, in our results, the staining of iNOS + structures was also detected, but was largely absent at 3 weeks after HO.M2 polarization has been verified to take part in bone tissue repair [29][30][31].Olmsted-Davis et al. also demonstrated the essential role of M2 macrophages in regulating angiogenesis with the secretion of vascular endothelial growth factor (VEGF) during HO development [32].Consistent with these studies, our results illustrated that M2 macrophages are similarly involved in the occurrence of HO.Intrigued by these in vivo observations, we further utilized different types of macrophages that had been cultured with reconstituted type I collagen or ectopic bone mass from HO rats, which demonstrated that M2-derived ecDNA promoted collagen fibril mineralization and further mineralization of ectopic bone masses in CaP solution.Heterogeneous macrophages are regarded as the "butterflies" that drive a sequence of events and ultimately promote HO; however, interestingly, our findings have led to a new approach for M2 macrophages in the formation of extraskeletal bone [33,34].
While we have verified that ecDNA released by M2 macrophages is critical for inducing calcification in HO, the modes of ecDNA release from macrophages are unclear.ecDNA has been described to be released from dead cells or from cells in association with vesicular structures [35, Fig. 8 Tissue injury at the beginning of HO leads to an inflammatory response with the recruitment of macrophages.Exosomes, apoptosis, and necrosis of M2 macrophages all lead to the release of ecDNA into the ECM.The polyanionic properties of ecDNA molecules make them capable of stabilizing calcium and phosphate, promoting the relatively stable formation of amorphous liquid droplets and inducing the ectopic mineralization in HO. 36].Macrophages control bone physiology by secreting the EVs containing nucleic acid cargoes, and interestingly, EVs derived from M2 macrophages significantly enhanced bone regeneration [29].The death of macrophages was not apparent in the present in vivo and in vitro models of bone calcification; therefore, we inferred that ecDNA was mainly released from the macrophages through the secretion of EVs.Further studies are needed to determine the method of release of ecDNA from macrophages.
To sum up, the present study showed that ecDNA from M2 macrophages initiates pathological calcification in HO, and the elimination of ecDNA might be developed into a clinical strategy to prevent ectopic mineralization diseases.The use of deoxyribonuclease for the targeted degradation of ecDNA at affected tissue sites provides a potential solution to treat diseases associated with ectopic mineralization (Fig. 8).

Fig. 1
Fig. 1 Extracellular DNA and calcified deposition identified from the ectopic bone masses in vivo.a Representative images of hematoxylin & eosin (HE) staining of sections prepared from the rat Achilles tendon calcification (bar: 300 or 50 μm).b Quantitative analysis of bone volume to total volume ratio (BV/TV) according to the HE images.c Bright field (BF) and representative images of immunofluorescence staining prepared from the ectopic bone masses at 6 and 9 weeks after tenotomy (bar: 2 mm or 500 μm).d Quantitative analysis of the percentage of surface ecDNA per field of view.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.

Fig. 2
Fig. 2 Increased ecDNA deposition associated with ectopic calcification during HO progression.a Representative images of immunofluorescence staining (bar: 100 or 25 μm) of the rat Achilles tendon tissue showing spatial distribution of ecDNA within the mineralized tissue.Alizarin red staining, red; Nucleic acids, green; DAPI, blue.b Quantitative analysis of the fluorescence intensity of Alizarin red per field of view.c Quantitative analysis of fluorescence intensity of ecDNA per field of view.d Strength of the association between ecDNA and calcified distributions using Pearson's correlation coefficient.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.

Fig. 3 Fig. 4
Fig. 3 M2 macrophages infiltrated around the ectopic bone during HO progression.a Representative images of immunofluorescence staining of CD68, iNOS, and MRC1 of sections prepared from the rat Achilles tendon calcification (bar: 100 μm).b Quantification of the mean fluorescence intensity (MFI) of CD68, iNOS, and MRC1 expression per field of view.c qRT-PCR analysis of the gene expression of the macrophages from the sham and HO groups of different times.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.◂

Fig. 5
Fig. 5 M2 macrophage-derived ecDNA adhered on ectopic bone masses and induced further mineralization ex vivo.a Schematic diagram showing ectopic bone masses cultured with macrophages with the incorporation of DNase-I.b BF and representative images of immunofluorescence staining of nucleic acids prepared from the ectopic bone masses cultured with macrophages for 10 days (bar: 2 mm).c Quantitative analysis of the percentage of surface ecDNA per field of view.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.

Fig. 6
Fig. 6 Further mineralization of ectopic calcification directed by CaP solution in vitro.a Schematic diagram showing ectopic bone masses cultured with macrophages before incubation in CaP solution for 14 days.b Representative images of immunofluorescence staining, Alizarin red staining, and nucleic acids prepared from the ectopic bone masses cultured with CaP solution for 14 days.c Quantitative analysis of the percentage of surface ecDNA per field of view.d Strength of the association between ecDNA and calcified distributions using Pearson's correlation coefficient.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.

Fig. 7
Fig. 7 Deoxyribonuclease treatment reversed the ectopic bone formation in the rat Achilles tenotomy HO model.a Schematic diagram showing the inhibitory activity of ecDNA in the rat Achilles tendon injury model through the injection of DNase-I.b Representative images of 3D-reconstructed Micro-CT of the different groups (bar: 5 mm), c HE (bar: 100 μm) and d immunofluorescence staining and Alizarin red staining of the Achilles tendon injury in the 6-week sham and HO with DNase I injection groups (bar: 100 μm).e Quantitative analysis of bone volume/total volume (BV/TV) and bone mineral density (BMD) of the different groups.f Quantitative analysis of fluorescence intensity of Alizarin red per field of view.For all charts, groups labeled with different lowercase letters are significantly different (p < 0.05).Data represent the means ± standard deviations.CT, computed tomography.