Osteochondritis Dissecans (OCD)-Derived Chondrocytes Display Increased Senescence, Oxidative Stress, Chaperone-Mediated Autophagy and, in Co-Culture with Adipose-Derived Stem Cells (ASCs), Enhanced Expression of MMP-13

Osteochondritis dissecans (OCD) in equids, especially in sport horses, has become a growing issue as it contributes to the occurrence of lameness. Thus the aim of this study was to investigate the cytophysiological properties of OCD chondrocytes including expression of chondrogenic genes, apoptosis, mitochondria dynamics and autophagy. Horse chondrocytes were isolated from healthy (HE) and OCD cartilages. Properties of cells were evaluated using multiple assays e.g., polymerase chain reaction (PCR), immunofluorescence, Western blot. OCD chondrocytes were characterized by increased apoptosis and senescence. Expression of chondrogenic genes (vimentin, aggrecan) was decreased while mRNA levels of matrix metalloproteinase 13 significantly upregulated in comparison to HE cells. Moreover, OCD cells displayed increased mitochondrial fusion while fission events were diminished. Interestingly, chaperone mediated autophagy was triggered in those cells and it predominated over macroautophagy. Furthermore, co-culture of LPS-treated chondrocytes with adipose-derived stem cells (ASC) decreased p62/sequestosome 1 (SQSTM) and increases MMP-13 expression in OCD cells. Our results suggest that OCD affected horse chondrocytes are characterized by senescent phenotype due to endoplasmic reticulum stress and mitochondria dynamics deterioration. Expression of chondrogenic markers is decreased in those cells while expression of chaperone mediated autophagy (CMA)-related genes increased. Increased malfunctioning of cells leads to loss of their functionality and capacity to maintain tissue homeostasis.


Introduction
Osteochondrosis (OC) is a disease of the cartilage which negatively affects joint homeostasis. OC is found not only in humans but also in other mammals like dogs and horses. It is considered a serious condition which contributes to reduced mobility, poor performance and early retirement of sport horses causing adversefinancial consequences. OC is usually accompanied by osteochondritis dissecans (OCD), which occurs due to deterioration of endochondral ossification in the epiphyseal or metaphyseal growth the metabolic functions of chondrocytes. For example it was shown that mitochondrial fission 1 protein (Fis1) is significantly down-regulated in human OA chondrocytes compared to healthy cells and it induced accumulation and inhibition of lysosomes [18]. Deteriorated, dysfunctional organelles can be removed from cells via a specific form of autophagy called mitophagy. It was shown that activation of autophagy protects chondrocytes against mitochondrial dysfunction most likely by removal of damaged organelles and thus preventing oxidative stress [10]. In order to remove dysfunctional mitochondria, Parkin (E3 ubiquitin ligase) is selectively recruited to their membrane allowing them to be engulfed by autophagosomes. It was shown that OA chondrocytes are characterized by increased Parkin expression which protects them against reactive oxygen species (ROS) [19]. However, no data regarding mitochondria dynamics in OCD chondrocytes exists.
Another critical factor for chondrocyte deterioration is ER stress [20] Accumulation of misfolded proteins in the ER lumen triggers the unfolded protein response (UPR), which activates C/EBP homologous protein (CHOP) and caspases inducing chondrocyte death [21,22]. Recent data showed that ER stress inducer, tunicamycin, triggered CHOP and MMP-13 expression while decreasing collagen type II and aggrecan mRNA levels [23]. On the other hand, it was shown that targeting ER stress may become effective in the OA treatment [24]. Involvement of ER stress in the pathogenesis of OA [25] has been well studied, however the cross-talk of ER stress and OCD development is still missing.
The aim of this study was to answer the investigate the cytophysiological properties of equine OCD chondrocytes at molecular level. For that reason we investigated viability, chondrogenic and hypertrophic genes expression, senescence, apoptosis, ER stress, mitochondrial dynamics and autophagy in OCD-derived chondrocytes. It is absolutely necessary to establish the molecular properties of those cells while taking into consideration the testing of new drugs against OCD and their translation into human cartilage regeneration. As it has been performed in-depth in humans, the pathophysiological events in equine chondrocytes are less clear.

Experimental Section
All reagents used in this experiment were purchased from Sigma-Aldrich (Poland), unless indicated otherwise.

Animals and Collection of Cartilage Samples
The cartilage samples were removed during arthroscopic surgery as a part of the standard curettage of the subchondral bone after removing the OC-fragment, and further medical waste were collected for that study, explaining why agreement of the local ethical committee was not necessary. The cartilage was obtained from 7 horses, aged between 9-11 years, of both sexes, of body weight between 570-630 kg, and fed with commercial feed. The horses were used in a jumping discipline of low intensity. In all horses radiological evidence of fragmentation were visible. Arthroscopy was undertaken on the basis of one or more of the following features: synovial effusion, lameness localised to the affected joint. Before arthroscopic operation, the joint were swollen and inflamed. The chondrocytes were isolated from cartilages pieces that floated in the fetlock joint. Immediately after removal, fragments were inserted into the transport medium and delivered to the Experimental Biology Department at Wroclaw Environmental and Life Science University. Healthy cartilage was derived post mortem from 6 horses, aged between 9-12 years of both sexes, body weight between 550-645 kg, fed with commercial feed. Cartilage were harvest from the fetlock joint.

Adipose-Derived Stem Cells (ASC) and Chondrocytes' Isolation and Culture
Cartilage from healthy and OCD horses were placed in a Hank's balanced salt solution (HBSS) and washed vigorously. Next, the specimens were minced into small pieces and transferred to 1 mg/mL clostridial collagenase at 37 • C for 8 h. Then, the solution was filtered using 70 µm strainers following centrifugation at 1200 rpm for 8 min. After discarding the supernatant, the cell pellet was washed three times with HBSS. Cells were suspended in Dulbecco s modified Eagle s medium (DMEM) Hams-F12 with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin/amphotericin (P/S/A) and cultured at 37 • C, 95% air with 5% CO2.
Adipose-derived stem cells (ASC) were isolated from healthy horses in accordance to procedure described previously [26]. Ethics approval for the adipose tissue harvesting was obtained from Local Research Ethics Committee in Wroclaw (approval no. 84/2018). Briefly, subcutaneous adipose tissue (from the tail base of horses) was harvest and placed in HBSS supplemented with 1% P/S/A solution. Specimens were cut into small pieces with surgical scissors and transferred to collagenase type I (1 mg/mL) following incubation for 40 min at 37 • C. Then samples were centrifuged at 1200× g for 10 min at 23 • C and re-suspended in culture media (DMEM) with low glucose supplemented with 10% of FBS and 1% P/S/A solution.
After reaching 80%-90% of confluence, both ASC and chondrocytes were passaged using a trypsin solution (TrypLE Express, Life Technologies). Media were changed every 2-3 days.

Evaluation of Viability and Proliferation Rate
In order to evaluate viability, chondrocytes were seeded onto 24-well plates at the density of 4 × 10 4 cells per well. After 24 h, cells were subjected for analysis using resazurin-based assay (TOX-8) in accordance with the manufacturer s protocol. Briefly, culture media were collected and substituted with medium containing 10% of resazurin dye. Following 2 h incubation in a CO 2 incubator, 100 µ of medium from each well was transferred onto a 96-well plate. Absorbance was measured spectrophotometrically using a 96-well microplate reader (Epoch, BioTek, Winooski, VT, USA). Reduction of the dye was evaluated at a wavelength of 600 nm and a reference wavelength of 690 nm.
In order to investigate proliferation rate of cells, we performed the assay based on the BrdU incorporation into the newly synthesized DNA using the BrdU Cell Proliferation enzyme-linked immunosorbent assay (ELISA) Kit (Abcam, Cambridge, UK) following the manufacturer's protocol. Briefly, cells were seeded onto a 96-well plate at a concentration of 1 × 10 4 of cells per well. After the attachment of cells, BrdU reagent was added in a volume of 20 µL. Then cells were incubated for 24 h at 37 • C in CO 2 incubator. Next, cells were fixed and DNA was denaturated with supplied reagents. Incorporated BrdU was detected with anti-BrdU monoclonal antibody and horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody. Reaction was developed using chromogenic substrate tetra-methylbenzidine (TMB). Then, absorbance at 450 nm wavelength was measured.

Evaluation of Cellular Morphology
The morphology of chondrocytes was evaluated with a scanning electron microscope (Zeiss EVO LS15, Oberkochen, Germany). Prior experimental cells were washed with HBSS and fixed with 4% paraformaldehyde (PFA) for 45 min in room temperature. Next, cells were dehydrated in a graded ethanol series (concentration from 50% to 100%) and sputtered with gold (ScanCoat 6, Edwards, Oxford, UK), transferred to microscope chamber and observed using a SE1 detector, at 10 kV of filament s tension.
Proteoglycans were stained using Safranin O. Briefly, cells were fixed with 4% PFA and stained with Safranin O solution for 30 min at room temperature. To visualize senescence-associated β-galactosidase, we applied Senescence Cells Histochemical Staining Kit following the manufacturer s protocol. Photographs were taken using an inverted bright field and epifluorescent microscope (Zeiss, Axio Observer A.1, Oberkochen, Germany).
In order to visualize the mitochondrial net, chondrocytes were incubated in medium containing MitoRed dye (1:1000) for 30 min. After washing with HBSS, cells were fixed in 4% PFA and nuclei were counterstained with DAPI. Mitochondria depolarization was observed using JC-1 fluorescent probe (Thermo Scientific, Waltham, MA, USA) staining. 5 µM JC-1 was added to cells following 30 min incubation at 37 • C. Next, cells were washed with HBSS and fixed with 4% PFA. Additionally, viable and dead cells were visualized using the Cellstain Double Staining Kit following manufacturer' instruction. Briefly, live cells were stained with Calcein A.M while apoptotic with propidium iodide. Specimens were observed under epi-fluorescent microscope (Zeiss, Axio Observer A.1, Oberkochen, Germany).

Measurement of Superoxide Dismutase (SOD) Activity and Nitric Oxide (NO) Concentration
In order to perform the assays, culture media were collected after 24 h of culture. Superoxide dismutase (SOD) activity was evaluated using SOD assay kit. It determines SOD activity in an indirect assay method based on xanthine oxidase and a novel color reagent. Water-soluble formazan dye undergoes reduction with a superoxide anion and the rate of the reduction is linearly related to the xanthine oxidase (XO) activity, and is inhibited by SOD. Thus SOD activity is calculated as inhibition rate in%. NO levels were measured using commercially available Griess reagent kit (Life Technologies, Carlsbad, CA, USA)-colorimetric assay for nitrites, and nitrates that have been reduced to nitrites. Procedures were carried out following the manufacturer's recommendations.

Chondrocytes-ASC Co-Culture
ASC was cultured in a transwell insert at a density of 70 × 10 3 cells per well while chondrocytes were propagated on the bottom at the density of 50 × 10 4 cells per well. Both cells population were cultured in DMEM Hams-F12 with 10% FBS and 1% of PSA. Prior to co-culture, 1µg/mL of lipopolysaccharide (LPS) was added to treat the chondrocytes for 2 h. Then, ASC were added and indirectly co-cultured with chondrocytes for 24 h. After that, cells were subjected for reverse transcription polymerase chain reaction (RT-PCR) analysis.

Western Blotting
In order to perform Western blotting, cells were detached from culture dishes and homogenized in radioimmunoprecipitation assay buffer (RIPA) buffer supplemented with protease inhibitor cocktail. The lysates were centrifuged at 4 • C for 20 min (14,000g) and supernatants were transferred to new tubes. Thirty micrograms of protein were used for each sample. SDS-PAGE was performed at 100 V for 90 min. Proteins were transferred onto nitrocellulose membrane (Bio-Rad) using a transfer apparatus at 100 V for 1 h at 4 • C. Next, membranes were washed with Tris/NaCl/Tween buffer (TBST) and blocked overnight at 4 • C with 5% non-fat milk in TBST. Finally membranes were incubated overnight with primary antibody for mitochondrial fission factor MFF (Biorbyt orb325479) at a dilution of 1:500. Following washing, solution of appropriate secondary antibody conjugated with horseradish peroxidase (HRP) was applied. After 2 h incubation, the membrane was washed again with TBST and incubated with Luminata Forte substrate (Merck) and visualized with ChemidocMP (Biorad).

Statistics
The results were evaluated based on measurements obtained in subsequent repetitions. All experiments were performed at least in triplicate. Assessment of the normality of data was performed with the Shapiro-Wilk test. Differences between groups were determined using the non-parametric t-test. Statistical analysis was performed with GraphPad Prism 5 software (La Jolla, San Diego, CA, USA). Differences were considered statistically significant for p < 0.05.

Metabolic Activity and Morphology
Metabolic activity was established with TOX-8 reagent [ Figure 1A]. Interestingly, there were no differences in metabolic activity of cells at day 1, 4 and 7 of culture. Safranin staining confirmed the accumulation of proteoglycans in both, healthy (HE) and OCD cells [ Figure 1B]. Scanning electron microscopy (SEM) [ Figure 1C] and f-actin [ Figure 1D] revealed that HE chondrocytes were characterized by polygonal morphology while OCD cells displayed more fibroblastic shape.

Expression of Chondrogenic and Hypertrophic Genes
Expression of aggrecan (ACAN), which maintains chondrocytes phenotype by interactions with glycosaminoglycans was significantly decreased in OCD cells [ Figure 2  Metabolic activity was assessed using Alamar Blue assay after day 1 (d1), 2 (d2) and 7 (d7) of culture (A). Chondrocytes were stained with safranin and observed under phase contrast microscope (B). Furthermore, cell were observed in scanning electron microscopy (SEM) (C). F-actin was visualized using phallodin and captured with confocal microscope (D). Scale bars: brightfield-250 µm, confocal: 50 µm. Results expressed as mean ± standard deviation (S.D.).

Apoptosis and Senescence
HE chondrocytes were characterized by 80% confluence while OCD displayed slight loss of calcein fluorescence intensity [ Figure 3A]. What is more those cells were characterized by increased apoptosis, as in indicated by propidium iodide staining [ Figure 3B] and senescence-increased number of senescent-positive (blue) cells [ Figure 3C]. Using the MUSE Cell Analyzer and Caspase 3/7 activation kit, percentage of live, dead, apoptotic, apoptotic/dead cells in cultures was assessed [ Figure 3D]. Results indicated that the number of live cells was decreased while the number of apoptotic and dead cells increased in OCD groups which corresponds with calcein and propidium iodide staining. Expression of pro-apoptotic genes p53 [ Figure 3E] and p21 [ Figure 3F] was significantly increased in OCD chondrocytes. Interestingly, no differences were found in the expression of cas-3 (caspase 3) [ Figure 3G] while mRNA levels of cas-9 (caspase 9) [ Figure 3H] was increased in OA cells.

Expression of Endoplasmic Reticulum (ER) Stress-Related Genes
Expression of genes which participate in ER stress response including CHOP [ Figure 4A], PERK [ Figure 4B], eIF2-alpha [ Figure 4C] and BIP [ Figure 4D] was tested. mRNA levels of those genes were significantly upregulated in OCD chondrocytes in comparison to normal chondrocytes.

Assessment of Mitochondria Condition
Mitochondria of OCD cells were characterized by decreased fluorescence intensity which may indicate diminished metabolic activity of those organelles [ Figure 5

Assessment of Mitochondria Condition
Mitochondria of OCD cells were characterized by decreased fluorescence intensity which may indicate diminished metabolic activity of those organelles [ Figure 5A]. Furthermore, to investigate the mitochondrial membrane potential (MMP), cells were stained with JC-1 dye and observed under epifluorescent microscope. Number of mitochondria stained with red/orange was downregulated in OCD chondrocytes in comparison to healthy cells [ Figure 5B]. On the other hand the number of mitochondria emitting green fluorescence (with decreased MMP) was greater in OCD cells than in HE [ Figure 5C]. MMP was further evaluated using the MUSE Cell Analyser [ Figure 5D]. The data obtained revealed that increased number of dead cells in OCD group. In order to evaluate antioxidative defence of isolated cells, activity of SOD was tested [ Figure 5E]. OCD chondrocytes displayed decreased SOD activity in comparison to healthy cells. Amount of extracellular NO was comparable between groups [ Figure 5F].  . Furthermore, to investigate mitochondria condition, cells were stained with JC-1. Mitochondrial depolarization is indicated by a decrease in the red and increase in the green fluorescence. Polarized mitochondria are marked by orange-red fluorescence (B) while depolarization is indicated with green fluorescence (C). Quantitative data regarding the mitochondria condition was obtained using MitoPotential assay and MUSE Cell Analyzer (D). In order to evaluate antioxidative status of chondrocytes activity of superoxide dismutase (SOD) was established (E). Furthermore, amount of NO in culture media was evaluated (F). Scale bars: epifluorescent −250 µm. Results expressed as mean ± S.D. * p < 0.05, *** p < 0.001.

Mitochondrial Dynamics and Autophagy
Expression of genes involved in the process of mitochondrial fusion, including MNF (mitofusin 1) [ Figure 6A, MIEF-1 (mitochondrial elongation factor 1) [ Figure 6B] and MIEF2 (mitochondrial elongation factor 2) [ Figure 6C] was investigated. Interestingly expression of those genes was significantly upregulated in OCD chondrocytes. Contrary, expression of protein related to mitochondrial fission, FIS [ Figure 6D] was significantly downregulated in those cells. Furthermore, expression of genes involved in the progression of mitophagy, PINK (PTEN-induced kinase 1) [ Figure 6E] and PARKIN (parkin ligase) [ Figure 6F] was downregulated in those cells. Amount of MFF protein was investigated with western blot [ Figure 6G] however no significant difference between HE and OCD band thickness was observed. MitoPotential assay and MUSE Cell Analyzer (D). In order to evaluate antioxidative status of chondrocytes activity of superoxide dismutase (SOD) was established (E). Furthermore, amount of NO in culture media was evaluated (F). Scale bars: epifluorescent −250 μm. Results expressed as mean ± S.D. * p < 0.05, *** p < 0.001.

Mitochondrial Dynamics and Autophagy
Expression of genes involved in the process of mitochondrial fusion, including MNF (mitofusin 1) [ Figure 6

Evaluation of Autophagy
Immunofluorescence staining revealed increased accumulation of LAMP-2 in OCD chondrocytes [ Figure 7A]. Furthermore, expression of autophagy related genes was investigated with RT-PCR. Interestingly LC-3 mRNA levels were diminished in OCD cells [ Figure 7B] while no differences were noted in BECLIN expression [ Figure 7C]. What is more, expression of LAMP-2 [ Figure 7D] and HSC70 [ Figure 7E] was significantly upregulated in OCD chondrocytes which indicates that not macroautophagy but CMA predominates in those cells.  Results expressed as mean ± S.D. ** p < 0.01, *** p < 0.001.

Chondrocytes-ASC Co-Culture
The catabolism of chondrocytes was triggered with LPS treatment for 2 h. Then ASC were added to chondrocytes and co-cultured for 24 h. After that period cells were subjected for RT-PCR analysis. Interestingly, mitochondrial fission was diminished in OCD cells after co-culture with ASC as decreased expression of FIS was observed [ Figure 8A]. What is more, ASC enhanced the expression of MMP-13 in OCD cells [ Figure 8B]. Furthermore, during co-culture with ASC autophagy diminishes in OCD chondrocytes as decreased SQSTM expression was decreased in those cells [ Figure 8C].

Chondrocytes-ASC Co-Culture
The catabolism of chondrocytes was triggered with LPS treatment for 2 h. Then ASC were added to chondrocytes and co-cultured for 24 h. After that period cells were subjected for RT-PCR analysis. Interestingly, mitochondrial fission was diminished in OCD cells after co-culture with ASC as decreased expression of FIS was observed [ Figure 8(A)]. What is more, ASC enhanced the expression of MMP-13 in OCD cells [ Figure 8(B)]. Furthermore, during co-culture with ASC autophagy diminishes in OCD chondrocytes as decreased SQSTM expression was decreased in those cells [ Figure 8(C)].

Discussion
In this study we have shown for the first time the comprehensive analysis of healthy and OCD chondrocytes isolated from horses. We have analysed apoptosis, expression of ECM proteins, oxidative stress, autophagy and mitochondrial dynamics in order to reveal the major causes of chondrocytes deterioration during OCD. It is especially important to unravel and compare the characteristics of chondrocytes from humans and equids as horse have become a model to study OA in humans. Pathogenesis of those two diseases may be distinct, although deterioration of cellular functionality seems to be comparable which indicates that chondrocytes respond for OA and OCD in similar manner. However, an in-depth analysis of the molecular characteristics of equine

Discussion
In this study we have shown for the first time the comprehensive analysis of healthy and OCD chondrocytes isolated from horses. We have analysed apoptosis, expression of ECM proteins, oxidative stress, autophagy and mitochondrial dynamics in order to reveal the major causes of chondrocytes deterioration during OCD. It is especially important to unravel and compare the characteristics of chondrocytes from humans and equids as horse have become a model to study OA in humans. Pathogenesis of those two diseases may be distinct, although deterioration of cellular functionality seems to be comparable which indicates that chondrocytes respond for OA and OCD in similar manner. However, an in-depth analysis of the molecular characteristics of equine chondrocytes is currently missing. In our study we revealed that OCD chondrocytes share a set of common characteristics to previously studied OA tissue. Although some molecular changes are common for those cells, they consequences are distinct.
It was shown that both OA and OCD chondrocytes are characterized by increased senescence and degeneration [29,30]. Similarly, in our study we have observed increased expression of pro-apoptotic genes, increased number of apoptotic cells, and accumulation of senescence-associated β-galactosidase. It is worth noting that senescent chondrocytes lose their ability to replace their extracellular matrix. This suggests the hypothesis that senescence of cells within joint tissues may play a pathological role in the causation of OCD. In their study Xu et al. [31] injected either senescent or non-senescent cells into the knee joint area of mice and revealed that transplanting senescent cells caused pathological features suggestive of OA which highlights the importance of senescent cells in development of musculoskeletal disorders. In study performed by Kim et al. [32], it was shown that chondrocyte apoptosis from the OA horse' cartilages were significantly higher than from healthy cartilages. Although the precise molecular mechanism of senescence in chondrocytes is still not elucidated, it is thought to be associated with vascular endothelial growth factor (VEGF), p53, p21 and p16 [33]. In our study we have observed increased expression of p21, p53, BAX and caspase-9 in OCD chondrocytes. Activation of p53 is triggered by DNA damage and telomere shortening in order to inhibit cell cycle progression. Up-regulation of 53 activates p21 which leads to cellular senescence. Dai et al. [34] revealed that stress-induced senescence of chondrocytes is accompanied by enhanced expression of both p21 and p53. It was showed that chondrocytes expressing p53 possesses similar to OA chondrocytes morphology and downregulation of p53 expression can prevent apoptosis and senescence in those cells [35]. On the other hand, increased expression of caspase-3 and caspase-9 was noted, however, in human OA chondrocytes [36]. In this study, we revealed that OCD chondrocytes, similar to OA cells, highly express pro-apoptotic genes. Increased apoptosis and senescence in OCD chondrocytes and the interplay between these processes may be crucial in the pathogenesis and progression of that disease. Similar results were obtained by Semevolos et al. [30] who observed increased apoptosis of osteochondral junction chondrocytes from OC horses. Other mechanisms of cell death may be failing in OCD, thus apoptosis is triggered in order to induce cell death and ossification. It is supported by the fact that the OCD hallmark feature is delayed ossification of epiphyseal cartilage which in consequence leads to pain and decreased mobility [30]. Thus it is tempting to speculate that apoptosis and senescence of chondrocytes are the major cause of OCD development and progression. That thesis is supported by the fact that thymosin-β4-protein involved in inhibition of actin polymerization and regulation of apoptosis is increased in experimentally-induced osteochondrosis [37]. It highlights the importance of chondrocyte apoptosis during OCD.
Previous studies have shown that apoptosis can be considered as a marker for chondrocytes hypertrophy [38]. Hypertrophic chondrocytes are characterized by the expression of specific proteins such as RUNX-2, type X collagen, MMP-13 and osteocalcin. Increased expression of MMP-13 in OCD cells correlates with multiple other reports that indicated on its upregulation during pathological changes of cartilage including OA [39,40]. Similar results were obtained by Laverty et al. [41], who showed MMP-13 elevated activity in equine OC cartilage. What is more, it was shown that MMP-13 is activated by RUNX-2 overexpression by mitogen-activated protein kinase (MAPK) pathways and direct regulation of MMP-13 gene transcription [42,43]. As RUNX-2 plays a pivotal role in chondrocytes differentiation and matrix degeneration it was shown to be overexpressed in both, OC and OA cartilage [44,45]. Recently, RUNX-3 has been identified as a new RUNX2 target, which cooperatively regulates hypertrophic differentiation of ATDC5 chondrocytes [46]. This study also demonstrated that similar to those in humans, equine OA chondrocytes are characterized by hypertrophic phenotype as increased expression of MMP-13, RUNX-2 and RUNX-3 was observed. On the other hand, the expression of collagen type II, COMP and DCN was augmented in those cells which supports those observed in human OA degeneration of ECM [47].
An alternative explanation of OCD chondrocytes senescence is based on the damage caused by oxidative stress. Oxidative stress occurs when levels of nitric oxide (NO) and ROS exceeds the antioxidant capacities of cells. Many studies have shown that NO and ROS may be directly involved in the pathogenesis of musculoskeletal disorders [48,49] but no data regarding its involvement in OCD exist. To clarify the role of oxidative stress and mitochondria in OCD chondrocytes we looked for the presence of oxidative damage in those cells. The results obtained clearly indicated for the first time that oxidative stress affects OCD horses' chondrocytes contributing to increased apoptosis, ER stress and autophagy. Our findings are consistent with a large body of data showing that ROS are a major factor, however, in the development of OA [50,51]. We have shown that horse OCD chondrocytes are characterize by enhanced ROS accumulation and NO synthesis while SOD activity in those cells is significantly diminished. Thus, excessive ROS accumulation seems to be the mechanism underlaying chondrocyte deterioration in both, OA and OCD. Increased NO production from horse chondrocytes was indirectly evaluated by immunohistochemistry by Kim et al. who observed increased NO accumulation in cartilage from OA horses [32]. On the other hand, Fu et al. [52] have shown that decreased SOD activity was SIRT-3 dependent and that its expression was downregulated in human and mice OA cartilage. As a result of the overload with damaged and misfolded proteins in deteriorated chondrocytes, the activation of ER stress pathways occurs. Excessive ER stress leads to the activation of CHOP which triggers apoptosis. It was reported that ER stress-induced apoptosis and enhanced CHOP expression contributed to chondrocyte apoptosis along with OA progression [53]. This is in good agreement with our data as we observed increased expression of ER stress mediators such as CHOP, PERK and BIP in OCD horse chondrocytes. It corresponds with increased apoptosis in those cells as CHOP leads to decreased expression of Bcl-2 while enhancing the expression of BAX and caspases [54]. Moreover, similar to our results, it was shown that CHOP induces the expression of MMP-13 while decreasing the amount of collagen type II and aggrecan [55].
Numerous studies have shown that stress-triggered senescence is manifested by mitochondria damage and that mitochondria deterioration plays a key role in the production of ROS. Mitochondria serve as key organelles for energy production and protects cells from oxidative damage. However, the loss of mitochondria functionality has been observed during aging and many chronic diseases [56,57]. Thus, mitochondrial damage results not only in the loss of energy production but also exposes cells to oxidative damage. In the presented study we have observed decreased mitochondrial membrane potential in OCD chondrocytes and decreased expression of mitophagy promoting PINK. Mitophagy leads to the removal of depolarized and damaged mitochondria but in OCD chondrocytes that mechanism seems to be inhibited. Ansari et al. [19] suggested that mitophagy is a crucial mechanism which prevents ROS production and enhances the survival, however, of OA chondrocytes and that loss of mitophagy directly contributes to development of the disease. Accumulation of damaged mitochondria leads to ROS overload and, in consequence, to chondrocyte apoptosis. Compared with healthy cartilage, deteriorated chondrocytes fail to generate energy due to alternation in mitochondrial dynamics. In our study we have revealed that mitochondria from OCD chondrocytes are characterized by increased fusion which allows for the exchange of content including DNA, and metabolites between neighboring organelles, including damaged or senescent ones, in consequence promoting their survival [58]. Thus we speculate, that in the state of decreased mitophagy, mitochondrial fusion serves as a compensatory mechanism which mitigates the effects of mitochondria damage through the exchange of proteins and lipids with other organelles.
Autophagy is a process that has gained increased attention in OA development as it was shown to be triggered in hypertrophic chondrocytes [59]. However, autophagy has two faces and can serve as pro-apoptotic or a protective mechanism which is lost during OA [10]. However, the exact role of autophagy in OCD development since now has not been studied. Chang et al. have shown that the induction of autophagy prevented the accumulation of subdiploid cells in young chondrocytes, while it induced cell death by autophagy in OA chondrocytes [60]. Moreover, they did not observe differences in the expression of LC-3 and Beclin between healthy and OA chondrocytes. This is in good agreement with our data as we did not observe macroautophagy flux in OCD cells. However, we observed up-regulation of genes involved with the chaperone mediated autophagy which removes damaged proteins by directing them into lysosomes. Thus we speculate that that CMA may play a protective role in OCD chondrocytes. We speculate that CMA activates before macroautophagy as a last chance rescue mechanism as it was shown that advanced OA contributes to "chondropoptosis" in different layers of the articular cartilage [61].
Moreover, we have observed that LPS treated OCD cells while co-cultured with ASC displayed decreased expression of SQSTM and FIS while increased expression of MMP-13. This is an interesting phenomenon which indicates that during co-culture with ASC, mitochondrial fission and their removal through mitophagy may be diminished. On the contrary, Jiang et al. [62] have demonstrated that ASCs activate autophagy and inhibit the expression of MMPs in chondrocytes treated with IL-1β or LPS. Interestingly, we have observed significantly increased expression of MM-13 in LPS treated OCD chondrocytes. This may indicate that stem cells should not be applied to tremendously inflamed joints as they may exert the opposite effects. However, this phenomenon should be further investigated.

Conclusions
The presented study provides insights into the biology of horse OCD chondrocytes and the involvement of oxidative stress, mitochondria damage, and autophagy in the pathogenesis of disease from the point of chondrocytes' senescence. Our findings show on the decreased antioxidative capacities and increased NO levels in OCD chondrocytes and resulting mitochondria damage followed by cellular dysfunction. Furthermore, our results indicate that deterioration of those cells is also correlated with ER stress. We postulate that in order to deal with the accumulation of misfolded proteins and their aggregates, OCD chondrocytes activate chaperone-mediated autophagy as a rescue mechanism to remove harmful biomolecules and are damaged by ROS organelles. New efforts to prevent the development and progression of OCD in horses may include strategies aimed at the efficient protection and improvement of mitochondrial activity by reducing ROS levels and chondrocyte apoptosis would be a valid target to modulate cartilage degeneration. OCD chondrocytes share similar characteristics with OA chondrocytes, which indicates that the molecular basis of these diseases' pathophysiology are common.
Author Contributions: K.K., M.A.N., M.R., K.M. contributed to the conception and design of the study, and analysis and interpretation of the data. K.K. and K.M. contributed to the acquisition of data and the writing of the manuscript, M.Z. contributed to language editing and revision process.
Funding: Study was supported from the project aimed at maintenance and development of research potential entitled "Application of experimental biology in regenerative medicine" ID:B010/0055/18