Strontium ion attenuates osteoarthritis through inhibiting senescence and enhancing autophagy in fibroblast-like synoviocytes

Osteoarthritis (OA) mainly occurs in the elderly population and seriously affects their quality of life (QOL). Strontium (Sr) ions have shown positive effects on bone tissue and are promising for OA treatment. However, the adequate treatment dosage and underlying mechanisms are unclear. This study investigated the effects and underlying mechanisms of different concentrations of Sr ions in a mouse model of OA induced by destabilization of the medial meniscus (DMM) surgery. DMM-induced OA mice received intra-articular injections of different concentrations of Sr ions, and a suitable concentration of Sr ions was found to improve OA. Furthermore, we investigated the mechanism by which Sr ions mediate senescence and autophagy in fibroblast-like synoviocytes (FLSs) in the synovial tissues of DMM-induced OA mice. OA mice treated with 10 µl of 5 mmol/L SrCl2 showed the greatest improvement in pain-related behavior and cartilage damage. In addition, in vivo and in vitro experiments revealed that Sr ions inhibit senescence and improve the autophagic function of FLSs. We also found that enhancement of the autophagic function of FLSs could effectively slow down senescence. Therefore, we show that Sr ions through the AMPK/mTOR/LC3B-II signal axis improve FLSs autophagy function and delay FLSs senescence, and furthermore, improve OA. These results suggest that senescence and autophagy function of FLSs may serve as promising targets for OA treatment, and that Sr ions may inhibit OA progression through these two targets.


Introduction
Osteoarthritis (OA) is a degenerative disease that damages the articular cartilage and entire joint tissue [1]. As an agerelated disease, its incidence rate increases with the aging population [2]. Many treatments have been applied to OA [3,4]. Early stage OA is mainly treated conservatively with oral medication or drug injections into the joint cavity to reduce joint damage, but the outcome is often unsatisfactory [5]. Most patients with OA eventually require surgery [6], which seriously affects their quality of life and imposes a significant economic burden. Therefore, nonsurgical Autophagy is an intracellular homeostatic mechanism that scavenges dysfunctional organelles and proteins [15]. Deficits in autophagy lead to cellular aging and are associated with age-related diseases [14]. In the FLS of OA patients, the expression of LC3B-I and LC3B-II (microtubule-associated protein 1 light chain 3), markers of autophagy, is reduced, suggesting decreased autophagy [16,17]. Autophagy is regulated by multiple signaling pathways, and mTOR is considered to be one of the most important proteins that affect autophagy [18,19]. Therefore, the regulation of autophagy may be a viable strategy for treating OA.
Strontium ions (Sr 2+ ) positively affect bone metabolism. Strontium ranelate (SR) has been used to treat postmenopausal osteoporosis (OP) [20] because it can enhance bone formation and inhibit bone resorption by osteoclasts. SR also slows OA progression [21]. In addition, Sr-doped biomaterials have shown promising results in the treatment of OA [22]. These findings suggest that Sr 2+ is an effective therapeutic agent for OA. However, the optimal dose and mechanism of action for OA remain unclear.
In this study, we investigated the mechanisms by which Sr 2+ inhibits OA progression. We induced OA in a mouse model by destabilization of the medical meniscus (DMM) surgery and measured the dose-response curve of Sr 2+ injected into the knee articular cavity. We found that intraarticular Sr ion treatment could slow the progression of OA, which is related to the inhibition of cellular senescence and enhancement of autophagy.

Experimental animals
Male C57/BL6 mice (10 weeks old) were obtained from the Animal Center of Air Force Military Medical University. All mice were kept on a 12 h light-on/12 h light-off cycle and had access to food and water ad libitum. All animal experimental procedures were approved by the Animal Research Committee of the Air Force Military Medical University, People's Republic of China.

OA mouse model and sr treatment
We performed destabilization of the medial meniscus (DMM) surgery on the left hindlimb knee joint following previously published protocols [11]. The mice were randomly assigned to five groups (n = 6 per group): osteoarthritis mice (OA), osteoarthritis mice with different treatments, OA + SR 0.1, OA + SR 1.0, OA + SR 2.0, and OA + SR 5.0. Two weeks after surgery, the mice received intra-articular injection of 10 µl saline (OA group) or SrCl 2 (0.1 mM for OA + SR 0.1, 1 mM for OA + SR 1.0, 2 mM for OA + SR 2.0, and 5 mM for OA + SR 5.0) twice per week for two consecutive weeks.

Catwalk assessment
On days 0, 14, 28, and 42 after SrCl 2 intra-articular injection, a catwalk analysis (CatWalk XTTM, Noldus) was performed to evaluate pain-related behavior in mice [23]. The mouse walked over a glass plate illuminated by a fluorescent tube light from below. A high-speed camera captured the footprint images when the soles of the mouse touched the glass plate for catwalk gait analysis. Prior to DMM surgery, the mouse was habituated to the catwalk gait analyzer daily for two weeks (15 min/day) to allow it to pass through the analyzer channel at a constant speed. The gait parameters (footprint area, stand, swing, intensity, and swing speed) were recorded at each time point. The ratio of the left hindlimb to the right hindlimb (LH/RH) was calculated.

Von Frey test
On days 0, 14, 28, and 42 after SrCl 2 injection, the von Frey test was performed to evaluate pain-related behavior, as described previously [24]. The paw withdrawal threshold was recorded as the minimum paw withdrawal value.

Isolation, culture and grouping of FLSs
Synovial tissues were isolated from normal and OA mice. FLSs were isolated from synovial tissues through enzymatic digestion using 2 mg/ml collagenase A and 0.1 mg/ ml DNase I (Roche). Dissociated cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Non-adherent cells were removed after 1 d of incubation. The cells were detached using trypsin-EDTA (Gibco) and split into new media when they reached approximately 90% confluence. FLSs from normal synovial tissues were classified as the Con group, and FLSs from OA mouse synovial tissues without SrCl 2 and with 5 mM SrCl 2 were classified as the OA and OA + SR groups, respectively.

Safranin O staining
Mouse knee joints were fixed in 4% paraformaldehyde in PBS for one week, dehydrated in ethanol, and embedded in paraffin after decalcification with ethylenediaminetetraacetic acid (EDTA). The tissue was then sliced into 5 μm sections and prepared for safranin O staining. The severity of knee OA was further quantified according to the Osteoarthritis Research Society International (OARSI) grades [25].

Quantitative real-time PCR
The FLSs were cultured for 5 days before collection. Total RNA was extracted using an RNA extraction kit (TRIzol reagent, Invitrogen Life Technologies) and reverse-transcribed into cDNA using the PrimeScript RT Master Mix (TaKaRa). Real-time PCR was performed on a Bio-Rad CFX96 PCR Manager System using the primers listed in Table S1. GAPDH was used as the housekeeping gene control.

Statistical analysis
We used Student's t-test for comparison between the two groups. One-way analysis of variance (ANOVA) was used to analyze the data from different groups, and the least significant difference (LSD) test was used for multiple comparisons. We computed Pearson's correlation coefficient between the OARSI score and each index. Data are presented as mean ± SD. The significance levels are given as *p < 0.05, **p < 0.01, and ***p < 0.001.

Sr 2+ alleviated pain-related mice behaviors
We used destabilization of the medial meniscus (DMM) surgery to model osteoarthritis (OA) in mice [26] and administered a series of intra-articular injections of different doses of SrCl 2 (twice per week for two weeks, starting two weeks after surgery). We used the Catwalk and von Frey tests to evaluate pain-related behavior in mice with OA. The CatWalk system is an automated video-based gait analysis system used to assess footfall and gait variation in rodents [27]. What it can record are static and dynamic parameters such as paw pressure, imprint area, stance phase duration, swing phase duration, stride length, and ice were received intra-articular injection different concentration Sr ionlimb coordination [28]. The CatWalk system has been used to investigate rodent models of arthritis, mainly in rats and mice [29,30]. CatWalk was assessed using the ratio of the left and right hindlimb measurements. Baseline interlimb coordination in mice, measured using the regularity index, is the ratio of left to right hindlimbs equal to 1 [31]. We found that mice receiving the highest dose of SrCl 2 (OA + SR 5.0 group) exhibited significant pain alleviation on days 14, 28, and 42, as indicated by the reduced swing ( Fig. 1 A), increased standing (Fig. 1B), increased footprint area (Fig. 1 C), increased pressure during contact between the foot and plate (intensity) (Fig. 1D), higher swing speed (Fig. 1E), and higher paw withdrawal threshold (Fig. 1 F) of the operated limb (p < 0.05). Mice in the OA + SR 1.0 and OA + SR 2.0 groups, which received lower doses of SrCl 2 , showed a partial alleviation of pain (p < 0.05), whereas mice receiving the lowest dose (OA + SR 0.1 group) did not show any pain alleviation (p > 0.05).

Sr 2+ inhibited the progression of OA
To examine the histological improvement of Sr 2+ in knee osteoarthritis, knee joints of OA mice with or without Sr 2+ treatment were collected 42 days after the injection. Safranin a dose-dependent manner, but not in the OA + SR 0.1 group (Fig. 2 A). In addition, we found a progressively increased thickness of osteophytes in the medial tibial plateau of OA mice, and OA + SR 1.0, OA + SR 2.0, and OA + SR 5.0, inhibited osteophyte formation (p < 0.001) in a dose-dependent manner; however, the OA + SR 0.1 group showed no difference compared to the OA group (p > 0.05) (Fig. 2B). Moreover, mice in the OA group showed increased thickening O staining showed that the articular surface structure of OA group was abnormal with an obvious loss of proteoglycan (Fig. 2 A), the features of early osteoarthritis are destruction of superficial zones of articular cartilage, which includes destruction of collagen and loss of proteoglycans, thereby leading to "mechanical softening" of the extracellular matrix [32]. Proteoglycan loss was significantly improved in the OA + SR 1.0, OA + SR 2.0, and OA + SR 5.0 groups in  All data are presented as mean ± SD, n = 6/group. *p < 0.05, **p < 0.01, ***p < 0.001 related to cell senescence, were significantly higher in the FLSs of OA mice than in those of normal mice (p < 0.05), whereas SrCl 2 treatment significantly reduced their expression in the OA model FLSs (p < 0.05) (Fig. 3 A-C). Double immunofluorescence labeling was used to stain p16 INK4a and vimentin in the synovial tissue. The FLSs of the OA group showed a serious senescence phenotype, and Sr ions weakened senescence of the FLSs (Fig. 3D). Quantitative immunofluorescence analysis revealed that the number of p16 INK4a positive cells in the OA group was the highest, followed by the OA + SR group, and the normal group was the lowest, with a statistical difference (p < 0.001) (Fig. 3E). The mRNA levels of SASP, including TNF-α, IL-6, IL-1β, MMP3, and MMP13 in FLSs were significantly higher in the OA group than in the normal group (p < 0.001). Sr 2+ significantly decreased the mRNA levels of SASP in FLSs (p < 0.001), although they were still higher than those in the normal group (p < 0.001) (Fig. 3 F).

Sr 2+ improves autophagy function of FLS
Autophagy plays a vital role in metabolic homeostasis, and LC3 is a key factor in this process [33]. Under normal conditions, LC3 expression at the baseline was stable. The of the subchondral bone in the proximal tibia, resulting in an increased subchondral bone score (based on the OARSI scoring system) on day42 after injection. Sr 2+ (OA + SR 2.0 and OA + SR 5.0) partially alleviated subchondral bone changes (p < 0.01). However, the OA + SR 0.1 groups did not alleviate the subchondral bone changes compared to the OA group (p > 0.05) (Fig. 2 C). We then quantified joint pathology using OARSI scoring (Fig. 2D). We found that the OA + SR 0.1 group and OA groups had the highest scores, indicating the worst pathological changes. In contrast, OA + SR 5.0 group had the lowest score (p < 0.001), and OA + SR 1.0(p < 0.01)/ OA + SR 2.0 (p < 0.001) groups had intermediate scores, suggesting less severe pathology (Fig. 2D).

Sr 2+ inhibits FLSs senescence
To further explore the mechanism of Sr 2+ in the remission of OA, we isolated FLSs from normal and OA mice and cultured them in media containing 5 mM SrCl 2 for three days. Quantitative real-time PCR and western blotting were performed to quantify the expression of cell senescencerelated mRNAs and protein. We found that the mRNA and protein expression levels of p16 INK4a and p21, which are  (Fig. 5 A, B). mTOR is a protein downstream of AMPK. Meanwhile, OA increased the level of phosphorylated mTOR (p-mTOR) (p < 0.001), and Sr ions reversed this effect (p < 0.001) (Fig. 5 A, B). Furthermore, the addition of compound C, an AMPK inhibitor, to the OA + SR group restored phosphorylated mTOR (p < 0.001) and lowered LC3B-II levels as well as the ratio of LC3B-II to LC3B-I (Fig. 5 C-E).
Correlation analysis revealed that the OARSI score was significantly correlated with the expression of LC3B and p16 INK4a in synovial tissue (Table S2). To verify the correlation between autophagy and senescence in FLSs from OA mice, these cells were treated with rapamycin or 3-MA to activate or inhibit autophagy, respectively. Rapamycin significantly reduced the expression of p16 INK4a (p < 0.05) and the mRNA levels of SASP factors (p < 0.05), including TNF-α, IL-6, IL-1β, Mmp3, and Mmp13 were all decreased significantly (Fig. 5 F-H). Conversely, 3-MA significantly increased the expression of p16 INK4a (p < 0.05) and mRNA levels of SASP factors (p < 0.05) (Fig. 5I-K).
ratio of LC3-II to LC3-I (LC3-II/LC3-I) is an important indicator of autophagic flux in which LC3-I is converted to LC3-II. The autophagic aptamer p62 links the ubiquitin pathway with the autophagy pathway and is degraded during this process [34]. We further examined the expression of autophagy-related markers LC3B-II and p62, in FLSs. Western blot analysis showed that the OA group had the lowest expression of LC3B-II and the highest expression of p62 (p < 0.001), increased expression of LC3B-II in FLS, and decreased expression of p62 in FLSs (p < 0.001) (Fig. 4 A, B). The ratio of LC3B-II to LC3B-I was significantly higher in the SR group than that in the OA and OA + SR group (p < 0.001). Immunofluorescence staining of LC3B in synovial tissue showed that the intensity of LC3B in the OA group was the lowest, whereas Sr ions increased the intensity of LC3B (p < 0.001) (Fig. 4 C, D).

Sr2 + improves the autophagic function of FLSs through the AMPK/mTOR/LC3B-II signaling axis and affects cellular senescence
We further studied the signaling axis in which Sr ions improved FLSs autophagy. Western blotting showed that the OA group had a reduced level of phosphorylated AMPK groups of mice. All data were presented as mean ± SD, n = 3/group. *p < 0.05, **p < 0.01, ***p < 0.001; red scale bar = 100 μm, yellow scale bar = 200 μm strontium ions can effectively promote cartilage differentiation of dedifferentiated fat cells [37]. Cheng demonstrated the most significant effect of Sr on osteogenic differentiation at a concentration of 3 mM without any obvious cytotoxicity [38]. Cai observed that Sr ions simultaneously enhanced the proliferation of human chondrocytes, upregulated the expression of cartilage-specific genes, and improved the secretion of glycosaminoglycans [39]. Li found that certain concentrations of tannic acid (TA)/Sr2+-coated components potently slowed OA and provided cartilage protection in a rat model [40]. Based on these relevant experimental results, and to identify the most appropriate concentration dose, we tested a range of Sr concentrations (0.1 ~ 5 mM) for joint cavity injection. In our experimental subgroup, the highest relative concentration (5 mM) yielded the best outcome in terms of pain alleviation and reduction in joint pathology.
FLSs form the lining of joints and play an important role in OA pathogenesis. They produce not only extracellular matrix and joint lubricants to protect the articular cartilage, but also inflammatory factors and proteases to destroy the articular cartilage, leading to the occurrence of OA [9,41]. Therefore, FLSs are increasingly considered good therapeutic targets for OA. Previous studies have shown that the FLSs of OA patients and DMM mice exhibit senescence and impaired autophagy [11]. In this study, we also found that

Discussion
Sr ions play an important role in maintaining human health. As an oral drug, strontium ranelate can effectively slow the occurrence and development of osteoporosis [20]. At the same time, studies have shown that strontium ranelate has a protective effect on OA [21,35]. This was mainly because Sr 2+ promotes osteogenic differentiation and inhibits osteoclast differentiation. Nevertheless, the optimal dose and mechanism of action of strontium ranelate for OA treatment remain unclear, limiting its clinical application. Recent studies have locally applied Sr ions. For example, Guo Shuo fabricated a sustained release system consisting of Sr 2+ -loaded sodium titanate nanorods and found that Sr 2+ could effectively promote bone repair at bone-defect sites [36]. Deng Cuijun also constructed bioactive ceramic scaffolds containing Sr 2+ to help rebuild the complex interface between cartilage and subchondral bone [22]. Thus, we hypothesized that injecting Sr 2+ into the joint cavity would reduce the occurrence and progression of OA. In this study, SrCl 2 was injected into the joint cavity to observe the effect of strontium ions on OA and synovitis as well as the optimal concentration and possible mechanism.
Okita cultured dedifferentiated fat cells with 1.5 mM strontium ions and found that this concentration of All data were presented as mean ± SD, n = 3/group. *p < 0.05, **p < 0.01, ***p < 0.001 the digestive system into the blood circulation and approximately 99% of the absorbed strontium is stored in the bone. Under normal physiological conditions, strontium in extracellular fluid is continuously exchanged with strontium in bone tissue [47,48]. Studies have shown that Sr ions have side effects on the heart. Therefore, an increasing number of researchers believe that local use of strontium ions is safer [49]. In the future, we need to explore strontium ion concentrations in more detail and clarify the different intracellular signaling pathways. After clarifying the therapeutic mechanism of strontium ions, we will continue to attempt to combine them with other drugs with different mechanisms of action to achieve better therapeutic effects. FLSs in the synovial tissue of OA mice were more senescent than those in normal mice, and the mRNA levels of SASP, including TNF-α, IL-6, IL-1β, Mmp3, and Mmp13, were significantly increased. These data indicate that an inflammatory microenvironment exists in the knee joints of OA mice, hindering repair of the damaged joint surface. We also found that senescence of FLSs in OA mice was correlated with damage to autophagic function.

Supplementary Information
Strontium ions have been proven to have multiple effects on bone regeneration and repair, such as promoting osteogenesis, inhibiting bone resorption, increasing internal vascularization of bone tissue, and promoting the polarization of macrophages to pro-repair type M2 [42]. However, the mechanism of action of strontium ions in OA has not been extensively investigated. Since strontium ranelate can regulate the autophagic function of bone marrow mesenchymal stem cells [38], we speculate that strontium ions may also regulate the autophagic function of FLSs. The results of this study show that strontium ions can enhance the autophagy function of FLSs in the synovium of OA mice through the AMPK/mTOR/LC3B-II signal axis. AMP-dependent protein kinase (AMPK) is involved in energy metabolism and autophagy. Notably, AMPK is a well-established regulator of autophagy that inhibiting mTOR [43]. In this study, based on the observation that compound C suppresses AMPK phosphorylation [44], we found that Sr ions can increase AMPK phosphorylation, thereby reducing downstream mTOR phosphorylation and ultimately improving the autophagic function of FLSs. This is consistent with a previous finding that treatment with the AMPK inhibitors C and Sr reduced autophagy and decreased osteogenic differentiation of cells [38]. Therefore, the AMPK/mTOR pathway might be a critical regulator of Sr-induced autophagy. The improvement of autophagy function significantly prevented the aging of FLSs and the consequent SASP.
In this study, we found that injecting different concentrations of strontium ions into the joint cavity of OA mice could effectively slow the OA process, and a dose of 5mM showed the best effect. The reason for this difference from other concentrations that have been reported may be related to the different models of the degree of injury [45,46]. Our data indicate that the decrease in autophagy in FLSs of OA mice contributes to their senescence, and strontium ions improve autophagy and delay senescence in FLSs through the AMPK/mTOR/LC3B-II signaling pathway. Although the aforementioned trials have shown that strontium ions have promising therapeutic effects for OA and their therapeutic mechanism has also been described, more research is needed before it is possible to apply strontium ions in the clinical treatment of OA.
Approximately 320 mg of strontium is present in normal adults, of which approximately 30% can be absorbed by