Hu Qian Wan Decoction Promotes Tendon-bone Healing of the Rotator Cuff in Rats by Inducing the Release of Exosomes Containing Interleukin-1 Receptor Antagonists

Background: Hu Qian Wan decoction (HQWD) is a traditional Chinese medicine used to treat many orthopedic diseases, including osteoporosis and osteoarthritis. This study aims to explore the mechanism by which HQWD promotes tendon-bone healing in the rotator cuff. Methods: Tendon-bone healing of the rotator cuff, serum exosome expression, and changes in interleukin-1 receptor antagonist (IL-1RA) in exosomes were analyzed after intragastric administration of HQWD in the rotator cuff reconstruction model of rats. The effects of Hu Qian Wan (HQW) on exosomes, IL-1RA, and Nod-like receptor protein 3 (NLRP3) secretion were veried in rat bone marrow mesenchymal stem cells (BMSCs) and macrophage cells, and the effects of HQW on inammasomes were observed. Results: HQWD intragastric administration increased proteoglycan and collagen (cid:0) (Col (cid:0) ) expression in the tendon-bone interface, improved the tendon-bone interface area, and augmented the maximum breaking load and stiffness of the rotator cuff. HQWD intragastric administration also increased rat exosome and IL-1RA expression in vivo. HQW promoted exosome and IL-1RA expression and suppressed the activation of NLRP3 inammasomes in rat BMSCs and macrophage cells in vitro. Conclusions: HQWD inhibited inammasome-related inammation by inducing exosome and IL-1RA secretion in rats, thus, promoting tendon-bone healing of the rotator cuff.

Among the factors affecting tendon-bone healing is osteoporosis (OP), a cause that cannot be ignored.
Studies have shown that OP is an independent factor affecting tendon-bone healing in rotator cuff injuries, and a signi cant negative correlation exists between bone mineral density in patients and the success rate of arthroscopic surgical repair [17]. In the reconstructive animal model of rotator cuff injury, the intervention of sclerostin monoclonal antibody, which is used to treat OP, enhances tendon-bone healing in the animal model [18]. Therefore, we hypothesized that it is feasible to search for drugs among the osteoporosis medications to promote tendon-bone healing.
Traditional Chinese medicine is widely used in the treatment of OP, and the effect is remarkable [19]. HQWD is a classic traditional Chinese herbal formula composed of Phellodendron chinense Schneid, Chinemys reevesii, Anemarrhena asphodeloides Bge, Rehmannia glutinosa Libosch, Citrus reticulata Blanco, Paeonia lacti ora Pall, Cynomorium songaricum Rupr, Canis familiaris L, and Zingiber oj-jicinale Rosc. For decades, HQWD has been widely used in the treatment of OP and osteoarthritis, and it has achieved favorable clinical e cacy [20]. However, the mechanism of HQWD is unclear. In addition, HQWD can promote the proliferation and differentiation of bone marrow mesenchymal stem cells (BMSCs) [21]. Therefore, exploring the mechanism and the potential clinical indications of HQWD is crucial. Based on previous studies, we speculated that there is a potential role between HQWD and exosomes, and that HQWD can regulate tendon-bone healing through exosomes. This study aimed to explore the changes in the expression of exosomes induced by HQWD and the mechanism of HQWD in promoting tendon-bone healing.

HQWD preparation
HQWD was provided by the Department of Pharmacy in the A liated Hospital of Nanjing University of Chinese Medicine. HQWD is composed of crude herbs, including Phellodendron chinense Schneid, Chinemys reevesii, Anemarrhena asphodeloides Bge, Rehmannia glutinosa Libosch, Citrus reticulata Blanco, Paeonia lacti ora Pall, Cynomorium songaricum Rupr, Canis familiaris L, and Zingiber oj-jicinale Rosc (dosages shown in Table 1). The medicinal herb mixture was extracted in boiling water, and the aqueous extracts were vacuum-dried at 60 °C to obtain a powder. The powder was then stored at -20 °C. The content of crude herbs was determined and used for experiments by dissolution in pure water at desired concentrations (0.5 mg/ml). The ngerprint of HQWD was determined by high-performance liquid chromatography (Fig. S1, Supporting Information).

Animal experiments
All the experimental procedures were performed with the approval of the Experimental Research Institute of Nanjing University of Chinese Medicine and followed the guidelines of the Institutional Animal Care and Use Committee. A total of 72 rats (four-week-old Sprague-Dawley (SD) rats, male, 70-100 g, purchased from Bejing Charles River Company) were used for in vivo experiments. Rats were housed in an airconditioned room at 25°C with a 12 h light/dark cycle. Nine rats were used in BMSCs extraction and 63 rats were used in rotator cuff reconstruction model building. Rotator cuff reconstruction model included two groups: control group and HQWD group, and each group included 18 rats. Eighteen rats were used in tissue section analysis (Control group, 9;HQWD group,9). Eighteen rats were used in biomechanical test analysis (Control group, 9;HQWD group,9). Twenty-seven rats were used in interleukin-1β (IL-1β) and IL-1RA experiments (Control group, 9; IL-1β group, 9; IL-1RA group, 9). The rats were sacri ced after anesthesia of iso urane, and tissues were obtained. RCT and reconstruction model building: After successful anesthesia with iso urane, we incised the shoulder joint laterally and pulled the trapezius muscle away. The insertion of the supraspinatus to the humeral head was exposed. The supraspinatus insertion was resected. Approximately 2 mm of the distal tendon of the supraspinatus was cut off. Then, the skin was sutured. Four weeks later, a 3-0 nonabsorbable Propathene line was used to x the supraspinatus tendon to the humeral head through bone tunnels. The subjects were randomly divided into two groups: equal amounts of either normal saline (NS) or HQWD. The intragastric administration of the treatments to the rats was performed after rotator cuff reconstruction.
The interleukin-1β (IL-1β) and IL-1RA drugs were administered intravenously into the rat tail at 1 and 10 μg/kg, respectively, once a day after rotator cuff reconstruction.

Extraction and culture of rat BMSCs
BMSCs inside their femurs and tibiae were obtained after anesthesia, according to previous literature [22]. Isolated BMSCs were incubated with standard media comprising DMEM (Gibco) supplemented with 10% FBS and 1% double-antibiotics (streptomycin + penicillin; Gibco). After 3-5 days of incubation (at 80% con uence), cells were re-plated in 10 cm Petri dishes and maintained at 37 ℃ in a humidi ed atmosphere of 5% CO 2 and 95% air.

Extraction of exosomes
Blood was collected from rat hearts after administering anesthesia, and the serum was extracted after centrifugation. Supercentrifugal extraction was adopted, and the cell debris was removed from the serum with 300× g centrifugation for 10 min, followed by 29,500× g centrifugation for 20 min. The supernatant was ltered through a 0.22 µm disposable sterile lter and centrifuged at 120,000× g for 90 min. The resulting supernatant was discarded. The precipitate was dissolved in 100 μL of PBS and stored at −80°C after packaging.
After BMSCs had grown to about 80% fusion, the culture medium was replaced by exosome-depleted FBS-containing medium (EXO-FBS-250A-1; System Biosciences, Mountain View, CA, USA). The cells were then cultured for a further 48 h. The medium was collected and centrifuged at 4 °C at 300 × g for 10 min and at 2000 × g for 10 min. Supercentrifugal extraction was adopted, and the cell debris was removed from the serum with 300 × g centrifugation for 10 min, followed by 29,500 × g centrifugation for 20 min. The supernatant was ltered through a 0.22× µm disposable sterile lter and centrifuged at 120,000 × g for 90 min. The resulting supernatant was discarded. The precipitate was dissolved in 100 μL of PBS and stored at −80 °C after packaging. The protein content of the BMSC exosomes was measured by the bicinchoninic acid assay (BCA; Thermo Fisher, Waltham, MA, USA). A microplate reader (ELx800, BioTek, USA) was used to measure the absorbance at a wavelength of 562 nm.
Grouping of drug and experimental animals HQWD was provided by the Department of Pharmacy in the A liated Hospital of Nanjing University of Chinese Medicine. The drug composition was as follows: Phellodendron chinense Schneid, Chinemys reevesii, Anemarrhena asphodeloides Bge, Rehmannia glutinosa Libosch, Citrus reticulata Blanco, Paeonia lacti ora Pall, Cynomorium songaricum Rupr, Canis familiaris L, and Zingiber oj-jicinale Rosc ( Table 1). The water decoction of HQWD contained 0.5 mg/ml of crude drug. Rats were randomly assigned to the control group or the HQWD group after reconstructing the supraspinatus muscle. Normal saline was administered to the control group intragastrically. The HQWD group received HQWD decoction 5 g/kg/day intragastrically.

Immunohistochemistry
For the immunohistochemistry assays, the para n sections were de-waxed, dehydrated, and incubated overnight at 4 °C with anti-Col , anti-NLRP3, and anti-IL-1RA (diluted 1:100; Abcam). After the primary antibody was removed, the secondary antibody (diluted 1:100, Thermo Fisher) was added, and the sections were incubated for 1 h at room temperature. The stained cells were developed with diaminobenzidine and counterstained with hematoxylin.

Safranin and fast green staining
After the shoulder joint was xed and decalci ed, the shoulder was embedded in para n, sliced at 5 μm in the coronal position, and hydrated after dewaxing. For fast green staining, the slices were placed in the fast green dye liquid for 5-10 min. Excess dye was washed with water until the cartilage became colorless. The slices were then slightly soaked in the differentiation uid and then washed with tap water. For staining with safranin, the slices were placed in safranin dye liquid for 15-30 s and then quickly dehydrated with three cylinders of anhydrous ethanol. For the transparent seal, sections were cleared with xylene for 5 min and then sealed with a neutral gum seal. The cartilage stained red or orange with a green background.

Western blot analysis
Cells and tissues were placed on ice immediately following treatment and washed with ice-cold Hank's balanced salt solution. All the wash buffers and nal resuspension buffer included a 1× protease inhibitor cocktail (Pierce, Rockford, IL, USA), NaF (5 mM), and Na3VO4 (200 mM). The protein concentration of the lysate was measured using the BCA protein assay kit (Thermo Fischer). Nuclear or total cell proteins were resolved on 8% to 12% SDS-PAGE and transferred by electroblotting to nitrocellulose membranes (Bio-Rad, Hercules, CA, USA). The membranes were blocked in 5% bovine serum albumin reagent (Beyotime, Jiangsu, China) and incubated overnight at 4 °C with primary antibody dilution buffer (the dilution followed the speci cation; Abcam) and then incubated with horseradish peroxidase-conjugated antirabbit IgG (1:5000) for 2 h. Afterward, the membranes were developed using the enhanced chemiluminescence substrate LumiGLO (Millipore, Bedford, MA, USA) and exposed to X-ray lm. The bands were analyzed with Gel-Pro Analyzer 4.0 (Bio-Rad, Hercules, CA, USA).

Biomechanical test
Biomechanical tests were performed 4 weeks after the surgical reconstruction. Mutilation was performed at the lower end of the humerus. The supraspinatus and its insertion were carefully retained; the other muscles, except for the supraspinatus, were removed. The supraspinatus-humerus structure was obtained. The specimen was immediately placed in 4% paraformaldehyde solution, and the shoulder joint tissue was rmly xed on an INSTRON biomechanical tester (Instron, Boston, MA, USA). The instrument was loaded, and a displacement velocity of 5 mm/min was applied to test the maximum tensile load of the supraspinatus tissue. The loading load when the supraspinatus tendon was broken was observed and recorded in detail as the maximum breaking load (N). The linear slope of the load-displacement curve was taken as the stiffness.
Enzyme-linked immunosorbent assay IL-1RA was determined by enzyme-linked immunosorbent assay (ELISA) using a commercially available ELISA set (Sagon Biotech, Shanghai, USA). ELISA was performed according to the manufacturer's instructions. All samples and standards were measured in duplicate.
Statistical analysis SPSS 19.0 statistical software (SPSS, Inc., Chicago, IL, USA) was used for all statistical analyses. The measurements are presented as the mean ± standard deviation. The data were analyzed using a one-way analysis of variance followed by Bonferroni's post-hoc test for multiple comparisons (p < 0.05 was considered to indicate a statistically signi cant difference).

HQWD improved the maximum breaking load and stiffness of the rotator cuff in rats
Four weeks after the supraspinatus reconstruction, the maximum breaking load and stiffness of the supraspinatus-humerus construction were examined (Fig. 1a). The breaking load and stiffness in the HQWD group were higher than in the control group (Fig. 1b, c), indicating that HQWD improved the tendon-bone healing of the rotator cuff in rats.
HQWD promoted the tendon-bone interface healing of the rotator cuff in rats Based on the differences in breaking load and stiffness, we speculated that HQWD could promote tendon-bone healing of the rotator cuff in rats. The key to tendon-bone healing lies in the tendon-bone interface. The tendon-bone interface area and proteoglycans are crucial indicators of tendon-bone healing. Safranin/fast green staining showed that in the HQWD group, the tendon-bone interface area was larger, and the expression of proteoglycans was stronger than in the control group (Fig. 2a). These results were statistically signi cant (Fig. 2b). Therefore, HQWD promoted the growth and healing of the tendon-bone interface.
Tendons are composed mainly of collagen, 95% of which is Col . Immunohistochemistry (IHC) showed that HQWD increased the expression level of Col , con rming that the tendon-bone healing in the HQWD group was superior to that in the control group (Fig. 2c, d, e, and f). Western blot (WB) analysis was performed on the 3 mm tendinous portion of the distal supraspinatus and showed that HQWD promoted the expression of Col 1 (Fig. 2e, f). These results showed that HQWD promoted the tendon-bone healing of the rotator cuff in rats.

Effects of HQWD on the secretion of exosomes and IL-1RA in rats in vivo
Since exosomes have a strong regulatory effect on tissue metabolism, we then observed the effect of HQWD on exosomes in rats. Brie y, 48 h after the rats underwent intragastric administration of HQWD, serum exosomes were extracted and con rmed by electron microscopy and WB (Fig. 3a, b). The expression of the surface markers of exosomes, such as CD63, CD9, and CD81, increased, indicating that HQWD promoted the secretion of exosomes in rats in vivo (Fig. 3b, c). Based on the immune and metabolic regulation of exosomes on cells and the inhibitory effect of in ammation on tendon-bone healing, we speculated that exosomes carry proteins that regulate in ammatory factors. IL-1RA, which regulates IL-1, was expressed in the exosomes of the HQWD group. WB con rmed increased expression of IL-1RA in exosomes in the HQWD group compared with the control group (Fig. 3d, e). IL-1RA was also detected through IHC in the shoulder joint at 4 weeks after reconstruction. The expression of IL-1RA in the tendon-bone interface of the HQWD group was increased compared with that of the control group; the difference was statistically signi cant ( Fig. 3f, g).

HQW induced BMSCs of rats to release exosomes carrying IL-1RA
Based on the close correlation between BMSCs and tendon-bone healing, we speculated that HQW would induce BMSCs to release exosomes. When HQW was added to the medium of BMSCs, IL-1RA and exosomes were increased in the culture supernatant of BMSCs (Fig. 4a, b and c), and IL-1RA in the exosomes was increased as determined by WB (Fig. 4d, e), con rming that HQW promoted BMSCs to secrete exosomes containing IL-1RA.
IL-1β and IL-1RA regulated the tendon-bone healing of the rotator cuff in rats HQWD increased the secretion of exosomes and IL-1RA and promoted tendon-bone healing in the rotator cuff in rats. To verify the effects of IL-1β and IL-1RA on tendon-bone healing, we applied IL-1β and IL-1RA in rats after rotator cuff reconstruction. Consequently, IL-1β reduced the rotator cuff maximum breaking load and stiffness, restrained the tendon-bone healing, and decreased the expression of Col (Fig. 5a, b, c, and d). Conversely, IL-1RA improved the maximum breaking load and stiffness of the rotator cuff and promoted tendon-bone healing and the expression of Col (Fig. 5a, b, c, and d).

HQW inhibited the activation of the NLRP3 in ammasome
The NLRP3 in ammasome consists of NLRP3, ASC, and caspase-1. IL-1RA inhibits the formation of the interleukin-1 receptor-induced NLRP3 in ammasome, and NLRP3 in ammasome activation promotes the release of IL-1β. We speculated that HQW could inhibit the activation of the NLRP3 in ammasome, and thus, inhibit IL-1β expression. It was found in macrophages that HQW could antagonize the expression of lipopolysaccharide (LPS)-induced NLRP3 in ammasome component proteins and downstream effector proteins (ASC, pro-caspase-1, caspase-1, pro-IL-1β, and IL-1β), which con rmed that HQW can inhibit the activation of the NLRP3 in ammasome (Fig. 6a, b). We observed the effect of HQWD on the expression of NLRP3 in the rotator cuff reconstruction site in rats and con rmed that HQWD inhibited the expression of NLRP3 (Fig. 6c, d).

Discussion
This study aimed to nd a treatment for rotator cuff tears using the traditional Chinese medicines, and to explore the relationship and internal mechanism between HQWD and tendon-bone healing. Our study has con rmed that HQWD can induce the release of exosomes containing IL-1RA and inhibit the expression of NLRP3 in ammasomes in rats, thus, promoting tendon-bone healing of the rotator cuff. Meanwhile, HQW stimulated BMSCs to secrete exosomes carrying IL-1RA, and inhibited the activation of NLRP3 in ammasomes in macrophages. Notably, our work demonstrated that HQWD acted by activating IL-1RA to improve tendon-bone healing. As a result, tendon-bone interface regeneration and functional recovery in the rotator cuff were improved.
According to the reconstruction model of the rat rotator cuff, HQWD improved the supraspinatus maximum breaking load and stiffness, thereby promoting tendon-bone healing. Safranin/fast green staining veri ed the increased tendon-bone interface area and proteoglycan expression. Furthermore, HQWD promoted the growth and healing of the supraspinatus tendon. In addition, 95% of the tendon collagen was Col , which is thick and closely spaced and is the main component of tendon resistance [23]. Therefore, we investigated the expression of Col . IHC and WB veri ed the increased expression of Col in the tendon-bone interface of the rotator cuff in the HQWD group. Thus, HQWD promoted tendonbone healing.
Based on the activity of exosomes to regulate the immune and metabolic effects of various cells, we analyzed whether or not HQWD could promote tendon-bone healing through the release of exosomes in rats. Indeed, HQWD increased the level of exosomes in rats. Exosomes may carry substances that promote tendon-bone healing. IL-1β is a crucial cause of tendon degeneration and tear, and it is an important negative regulator of tendon-bone healing [24]. IL-1β also induces the generation of matrix metalloproteinases (MMPs), the main cause for the degeneration in cartilage proteoglycan and tendon collagen [25]. Therefore, regulation of IL-1β is crucial for tendon-bone healing. IL-1/IL-1R in organisms maintains a dynamic state of balance. IL-1RA can competitively bind to IL-1R, leading to the inactivation of IL-1R and the restrained effect of IL-1β [26]. IL-1RA restrains the release of MMPs and reduces the degradation of proteoglycan and collagen, thereby retarding the destruction of bone and cartilage [27]. Therefore, improving the expression of IL-1RA could promote tendon-bone healing. Hence, we hypothesized that HQWD induced IL-1RA secretion and promoted tendon-bone healing of the rotator cuff. WB detection of rat exosomes veri ed that IL-1RA expression was increased in the HQWD group and was detected in the shoulder joint, thereby verifying that the expression of IL-1RA in the tendon-bone interface of the HQWD group was increased compared with that of the control group. Furthermore, the high expression of IL-1RA in the tendon-bone interface of the rotator cuff indicated that the area of the tendonbone interface would be large and regulated.
BMSCs can differentiate into various types of cells, including tendon cells and bone cells, and they can promote tendon-bone healing of the rotator cuff [16,28]. HQWD promotes the proliferation and differentiation of BMSCs (Wang et al.,2016).. Therefore, we hypothesized that exosomes in our research originated from BMSCs. Our study veri ed that HQWD promoted the secretion of exosomes and IL-1RA from BMSCs. In order to verify the role of IL-1RA in promoting tendon-bone healing, we applied IL-1β and IL-1RA in rats and found that IL-1β inhibited tendon-bone healing, while IL-1RA promoted tendon-bone healing. IL-1RA has been found to inhibit the activation of NLRP3 in ammasomes, a class of polymeric protein complexes that play an important regulatory role in in ammation and immune response [29]. Upon activation of NLRP3, a key component that mediates the formation of in ammasomes, the adaptor protein ASC is induced to bind pro-caspase-1 to form an in ammatory complex that activates caspase-1.
ACS also shears and promotes the maturation and secretion of IL-1β and IL-18, thus, promoting the in ammatory response [30]. In recent years, NLRP3 has been found to be involved in the regulation and inhibition of tendon-bone healing and bone injury repair [31,32]. Therefore, we speculated that exosomes coated with IL-1RA could inhibit IL-1R/NLRP3-mediated in ammasome formation and promote tendonbone healing by acting on macrophages. We veri ed the inhibitory effect of HQWD on the NLRP3 in ammasome in macrophages and at the tendon-bone interface of the rotator cuff after reconstruction in rats and con rmed that HQWD can inhibit the activation of the NLRP3 in ammasome.
To date, no effective prevention or treatment has been developed for re-tears after rotator cuff reconstruction. Studies on the treatment of rotator cuff tears by traditional Chinese medicine and on the role of exosomes and IL-1RA in this condition are limited. The present study is the rst to reveal that HQWD promotes tendon-bone healing of the rotator cuff. This nding is of positive signi cance for treating rotator cuff tears with traditional Chinese medicine. We preliminarily discussed the mechanism of promoting tendon-bone healing of the rotator cuff by inducing exosomes and the secretion of IL-1RA in response to HQWD. We also con rmed for the rst time that exosomes promote tendon-bone healing of the rotator cuff through IL-1RA, demonstrating the crucial role of exosomes in tendon-bone healing. A limitation of this study is that the mechanism by which HQWD induces the expression of exosomes and IL-1RA was not fully elucidated. In addition, exosomes are extremely active and possess various functions, so whether the inhibition of NLRP3 in ammasome activation by IL-1RA is the most vital factor in promoting tendon-bone healing needs further study, as does the mechanism of tendon-bone healing induced by exosomes and IL-1RA. Nevertheless, the present study provides a new concept for exploring the effects of traditional Chinese medicine on exosomes and IL-1RA in tendon-bone healing of the rotator cuff.

Conclusions
By inducing the secretion of exosomes containing IL-1RA, HQWD inhibits NLRP3 in ammasome-induced in ammation and promotes tendon-bone healing of the rotator cuff.