Mg-BGNs/DCECM Composite Scaffold for Cartilage Regeneration: A Preliminary In Vitro Study

Cartilage lesions can lead to progressive cartilage degeneration; moreover, they involve the subchondral bone, resulting in osteoarthritis (OA) onset and progression. Bioactive glasses, with the dual function of supporting both bone and cartilage regeneration, have become a promising biomaterial for cartilage/bone engineering applications. This is especially true for those containing therapeutic ions, which act as ion delivery systems and may further promote cartilage repair. In this study, we successfully fabricated Mg-containing bioactive glass nanospheres (Mg-BGNs) and constructed three different scaffolds, DCECM, Mg-BGNs-1/DCECM (1% Mg-BGNs), and Mg-BGNs-2/DCECM (10% Mg-BGNs) scaffold, by incorporating Mg-BGNs into decellularized cartilage extracellular matrix (DCECM). All three scaffolds showed favorable microarchitectural and ion controlled-release properties within the ideal range of pore size for tissue engineering applications. Furthermore, all scaffolds showed excellent biocompatibility and no signs of toxicity. Most importantly, the addition of Mg-BGNs to the DCECM scaffolds significantly promoted cell proliferation and enhanced chondrogenic differentiation induction of mesenchymal stem cells (MSCs) in pellet culture in a dose-dependent manner. Collectively, the multifunctional Mg-BGNs/DCECM composite scaffold not only demonstrated biocompatibility but also a significant chondrogenic response. Our study suggests that the Mg-BGNs/DCECM composite scaffold would be a promising tissue engineering tool for osteochondral lesions, with the ability to simultaneously stimulate articular cartilage and subchondral bone regeneration.


Introduction
It is well known that articular cartilage lesions can lead to progressive cartilage degeneration; moreover, these lesions involve the subchondral bone. This results in an early onset of osteoarthritis (OA) [1,2]. OA is a leading cause of disability among adults [3]. As articular cartilage lacks an innate self-repair ability with avascularity, it poses a significant challenge for cartilage regeneration [4]. Tissue engineering strategies, which involve a combination of cells and bioactive scaffolds, have become a promising approach for cartilage repair and regeneration [5]. Bioactive scaffolds play a significant role in tissue engineering [6]. The ideal scaffold biomaterials should be biocompatible, biodegradable, and have a porous architecture that can promote cell adhesion and proliferation. In particular, the scaffold should provide bioactive stimuli for target tissue regeneration and formation [7].
For cartilage tissue engineering, decellularized cartilage extracellular matrix (DCECM) is a promising scaffold biomaterial. This is due to its ability to provide a cartilage-specific extracellular matrix for cell-matrix interactions, such as collagen II, glycosaminoglycans (GAGs), and certain active growth factors [8,9]. It has been found that the cartilage extracellular matrix can promote chondrogenesis of mesenchymal stem cells (MSCs) and reduce the dedifferentiation phenomenon of chondrocytes during in vitro expansion [10,11].
The DCECM was fabricated using our previous differential centrifugation method [31,32]. The swine articular cartilage was washed with distilled water, cut into pieces (1 mm × 1 mm × 1 mm), and then treated with hydrogen peroxide (Sigma-Aldrich, St. Louis, MO, USA). After washing with distilled water, the minced cartilage was homogenized using a machine (Kinematica AG, Lucerne, Switzerland) at a low temperature (4 • C). The decellularized cartilage matrix was obtained using the differential centrifugation method with a high-speed, low-temperature centrifuge (Thermo, Osterode, Germany). The cartilage homogenate samples were first centrifuged at 2000 rpm for 10 min. Next, the sediment was removed and the supernatant was recentrifuged at 10,000 rpm for 30 min. The supernatant was then removed; the remaining sedimentary slurry was the decellularized cartilage extracellular matrix hydrogel.

Characterization of Mg-BGNs
The microstructure of Mg-BGNs was observed using scanning electron microscopy (SEM; Gemini 300, Zeiss, Jena, Germany) and transmission electron microscopy (TEM; TALOS F200 X, FEI, Waltham, MA, USA). The constituent elements of the Mg-BGNs were determined using TEM elemental mapping.

Ion Release of Mg-BGNs/DCECM Scaffolds
The release of Mg, Si, and Ca ions from the Mg-BGNs-1/DCECM and Mg-BGNs-2/DCECM scaffolds was assayed over 7 days using inductively coupled plasma-atomic emission spectroscopy (ICP-AES; Avio 500). Briefly, the Mg-BGNs-1/DCECM and Mg-BGNs-2/DCECM scaffolds (20 mg) were dispersed in 10 mL of phosphate-buffered saline (PBS) in a 37 • C incubator shaker. At the set time points of 3, 6, 12, 24, 48, 96, 120, and 168 h, the samples were collected for ion assessment. Rabbit BMSCs were isolated as described in our previous study, and this study was performed according to a protocol approved by the Institutional Animal Care and Use Committee (IACUC) of Renji Hospital affiliated to Shanghai Jiao Tong University Medical College (identification code: RJ2021-0525; date: 20 May 2021). The P0 cells were cultured in α-minimum essential medium (α-MEM; Gibco) with 10% fetal bovine serum (FBS; Gibco) at 37 • C in a 5% CO 2 incubator. The cells were diluted 1:3 at 90% confluence. Passage 3 was used for cytocompatibility studies.

Live/Dead Staining and Cell Viability Analysis
The cytotoxicity of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds was evaluated using a live/dead assay kit (Beyotime, Shanghai, China). After seeding BMSCs on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds and cultured for 1, 3, and 7 days, the cell/scaffold complex was harvested. It was subsequently washed with PBS (Gibco, Life Technologies, Grand Island, NY, USA) and incubated with calcein AM and propidium iodide (PI) for 30 min according to the manufacturer's instructions (Beyotime, Shanghai, China). After washing with PBS three times, the complex was observed using a confocal microscope (Leica SP8, Germany). Cell viability was calculated as follows: live cells/total cells × 100%. Five horizons per group were used for cell viability analysis, and the images were analyzed using Imaris software (ver. 7.4. software (Bitplane, Zurich, Switzerland).

Cell Morphology
The morphology of cells seeded on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds was observed using a SEM (Gemini 300, Zeiss). The cell/scaffold complex was harvested after culturing for 3 days; it was then fixed with 2.5% glutaraldehyde for 24 h. Subsequently, it was dehydrated with gradient alcohol and dried to a critical point (EM CPD300; Leica, Wetzlar, Germany) using CO 2 . The samples were then sputter-coated with gold. The microstructure of the cell scaffolds was observed using SEM (Gemini 300, Zeiss).

Chondrogenic Differentiation Induction in BMSC Pellets
We evaluated the induction of chondrogenic differentiation of DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds in BMSC pellets using a Transwell system. The BMSC pellets were formulated as described in our previous study with 5 × 10 5 BMSCs. The DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were transferred to the lower well of the Transwell plate (Corning Inc., Corning, NY, USA). The BMSCs were placed on the upper well of the Transwell plate (Corning, USA). Chondrogenic medium (Cyagen, Santa Clara, CA, USA) was added to the wells and changed every 3 days. After culturing in the well for 21 days, the pellets were harvested for analysis.

Histological and Immunohistochemical Analysis
After being cultured in the Transwell for 21 days, the pellets were harvested and photographed. The pellets were fixed in 4% paraformaldehyde overnight, embedded in paraffin, and prepared into 7 µm sections. These sections were stained with hematoxylineosin (HE), Alcian blue (AB), and immunohistochemical (IHC) for collagen II (primary antibody Col II 1:100, Abcam, Boston, MA, USA).

GAG/DNA Analysis
The GAG and DNA content of the pellets were determined according to our previous methods. We used the 1,9-dimethylmethylene blue (DMMB) assay method to measure GAG content using the Tissue GAG Total Content DMMB Colorimetry kit (Genmed Scientific Inc., Shanghai, China). DNA was extracted using the TIANamp Genomic DNA kit (TIANamp, Beijing, China) and quantified using the PicoGreen DNA assay kit (Invitrogen, Carlsbad, CA, USA).

RT-PCR
The expression of chondrogenic-related genes (SOX9), collagen II, aggrecan, and collagen I) in the pellets treated with the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were determined by real-time polymerase chain reaction (RT-PCR). The RT-PCR procedure was performed according to the general protocol. After extraction from the pellets with TRIzol (Invitrogen, Waltham, MA, USA), messenger ribonucleic acid (mRNA) was reverse transcribed into complementary DNA (cDNA) using a ReverTra Ace kit (Toyobo, Osaka, Japan), and then quantified using RT-PCR on a LightCycler 480 system (Roche Applied Science, Indianapolis, IN, USA). The primers used are listed in Table 1.

Statistical Analysis
Data are expressed as mean ± standard deviation (SD). Statistical analysis was performed using one-way analysis of variance (ANOVA) with SPSS Statistics version 22 (IBM, Armonk, NY, USA). A value of p < 0.05 was considered statistically significant.

Characterization of Mg-BGNs
The nanoscale morphology of the Mg-BGNs was observed using field-emission scanning electron microscopy (FESEM) ( Figure 1A) and field-emission transmission electron microscopy (FETEM) ( Figure 1B). The average diameter of the uniformly sized nanospheres was 108.2 ± 22.5 nm. The chemical composition of the Mg-BGNs was characterized by scanning transmission electron microscopy (STEM)-energy dispersive spectroscopy (EDS). As shown in Figure 1C-H, elemental maps revealed that Mg, Si, Ca, and O are homogeneously distributed within the Mg-BGNs.

Statistical Analysis
Data are expressed as mean ± standard deviation (SD). Statistical analysis was performed using one-way analysis of variance (ANOVA) with SPSS Statistics version 22 (IBM, Armonk, NY, USA). A value of p < 0.05 was considered statistically significant.

Characterization of Mg-BGNs
The nanoscale morphology of the Mg-BGNs was observed using field-emission scanning electron microscopy (FESEM) ( Figure 1A) and field-emission transmission electron microscopy (FETEM) ( Figure 1B). The average diameter of the uniformly sized nanospheres was 108.2 ± 22.5 nm. The chemical composition of the Mg-BGNs was characterized by scanning transmission electron microscopy (STEM)-energy dispersive spectroscopy (EDS). As shown in Figure 1C-H, elemental maps revealed that Mg, Si, Ca, and O are homogeneously distributed within the Mg-BGNs.

Morphology of DCECM and Mg-BGNs/DCECM Scaffolds
The microstructures of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds are shown in Figure 2. The SEM images clearly show that all three scaffolds exhibit a similar highly interconnected porous structure with a mean pore size of 80-200 µm. The high magnification images indicate that the Mg-BGNs-1/DCECM scaffold contains a small number of Mg-BGNs, whereas the Mg-BGNs-2/DCECM scaffold consists of abundant, uniform, and dispersed Mg-BGNs. The SEM results confirm that the Mg-BGNs tend to aggregate in the walls of the pores in the scaffolds.

Morphology of DCECM and Mg-BGNs/DCECM Scaffolds
The microstructures of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds are shown in Figure 2. The SEM images clearly show that all three scaffolds exhibit a similar highly interconnected porous structure with a mean pore size of 80-200 um. The high magnification images indicate that the Mg-BGNs-1/DCECM scaffold contains a small number of Mg-BGNs, whereas the Mg-BGNs-2/DCECM scaffold consists of abundant, uniform, and dispersed Mg-BGNs. The SEM results confirm that the Mg-BGNs tend to aggregate in the walls of the pores in the scaffolds.

Ionic Release of Mg-BGNs/DCECM Scaffolds
The ionic release behaviors of the Mg-BGNs-1/DCECM and Mg-BGNs-2/DCECM scaffolds were examined using ICP-AES. The ionic concentrations of Mg, Si, and Ca in the scaffold extracts were recorded for up to 7 days, as shown in Figure 4. The release rate was rather rapid at first (up to 3 days) for Mg ions. This decreased slightly by day 7 ( Figure  4A). The release rate of Si ions was relatively stable for up to 7 days ( Figure 4B). The release behavior of Ca ions was similar to that of Mg ions-rapid up to 3 days and then decreasing slightly by day 7( Figure 4C). Overall, the Mg, Si, and Ca ionic release rates of the Mg-BGNs-2/DCECM scaffold are more rapid than those of the Mg-BGNs-1/DCECM scaffold.

Ionic Release of Mg-BGNs/DCECM Scaffolds
The ionic release behaviors of the Mg-BGNs-1/DCECM and Mg-BGNs-2/DCECM scaffolds were examined using ICP-AES. The ionic concentrations of Mg, Si, and Ca in the scaffold extracts were recorded for up to 7 days, as shown in Figure 4. The release rate was rather rapid at first (up to 3 days) for Mg ions. This decreased slightly by day 7 ( Figure 4A). The release rate of Si ions was relatively stable for up to 7 days ( Figure 4B). The release behavior of Ca ions was similar to that of Mg ions-rapid up to 3 days and then decreasing slightly by day 7 ( Figure 4C). Overall, the Mg, Si, and Ca ionic release rates of the Mg-BGNs-2/DCECM scaffold are more rapid than those of the Mg-BGNs-1/DCECM scaffold.
was rather rapid at first (up to 3 days) for Mg ions. This decreased slightly by day 7 ( Figure  4A). The release rate of Si ions was relatively stable for up to 7 days ( Figure 4B). The release behavior of Ca ions was similar to that of Mg ions-rapid up to 3 days and then decreasing slightly by day 7( Figure 4C). Overall, the Mg, Si, and Ca ionic release rates of the Mg-BGNs-2/DCECM scaffold are more rapid than those of the Mg-BGNs-1/DCECM scaffold.

Cell Viability Analysis of Scaffolds
The cytocompatibility of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds was evaluated using a live/dead cell staining method. BMSCs were seeded on the three different scaffolds and cultured for 1, 3, and 7 days. Subsequently, live/dead staining was performed and observed using a confocal microscope. The results are shown in Figure 5A. From this, we know that most of the cells in the scaffolds were live (green) and few cells died (red); this indicates that all the three different scaffolds had no obvious cytotoxicity. Quantitative cell viability analysis also confirms that the viability of BMSCs in DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds after 1, 3, and 7 days was more than 95%, and there was no significant difference among the three different groups ( Figure 5B).

Cell Viability Analysis of Scaffolds
The cytocompatibility of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds was evaluated using a live/dead cell staining method. BMSCs were seeded on the three different scaffolds and cultured for 1, 3, and 7 days. Subsequently, live/dead staining was performed and observed using a confocal microscope. The results are shown in Figure 5A. From this, we know that most of the cells in the scaffolds were live (green) and few cells died (red); this indicates that all the three different scaffolds had no obvious cytotoxicity. Quantitative cell viability analysis also confirms that the viability of BMSCs in DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds after 1, 3, and 7 days was more than 95%, and there was no significant difference among the three different groups ( Figure 5B).

Cell Proliferation and Attachment
The cell proliferation effect of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were evaluated using a CCK-8 assay kit. As shown in Figure 6A, on day 1, there was no significant difference between the three different groups. However, the cell proliferation effect of the Mg-BGNs-2/DCECM group was higher than the DCECM

Cell Proliferation and Attachment
The cell proliferation effect of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were evaluated using a CCK-8 assay kit. As shown in Figure 6A, on day 1, there was no significant difference between the three different groups. However, the cell proliferation effect of the Mg-BGNs-2/DCECM group was higher than the DCECM group (p < 0.05) on day 3. On day 7, the Mg-BGNs-2/DCECM group's cell proliferation effect was higher than the Mg-BGNs-1/DCECM group, and this, in turn, was higher than the DCECM group (p < 0.05). The results indicate that Mg-BGNs-2/DCECM scaffolds can promote the proliferation of BMSCs compared to the Mg-BGNs-1/DCECM and DCECM groups. The BMSCs attached to the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were observed using SEM ( Figure 6B). It can be seen that BMSCs spread well on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds. Cell attachment and growth were also observed by F-actin staining. As shown in Figure 6C, the cytoskeletal morphology of BMSCs seeded on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds showed that all the BMSCs on the three scaffolds appeared to grow and spread well after culturing for 1, 3, and 7 days. They also exhibited elongated, multilayered morphologies with the extended spreading of actin filaments.  Figure 6B). It can be seen that BMSCs spread well on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds. Cell attachment and growth were also observed by F-actin staining. As shown in Figure  6C, the cytoskeletal morphology of BMSCs seeded on the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds showed that all the BMSCs on the three scaffolds appeared to grow and spread well after culturing for 1, 3, and 7 days. They also exhibited elongated, multilayered morphologies with the extended spreading of actin filaments.

Chondrogenic Differentiation Induction of Scaffolds
We evaluated the chondrogenic differentiation induction capacity of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds in BMSC pellet culture in vitro. The pellets were cultured in a chondrogenic conditioned medium for 21 days, as shown in Figure 7B. Figure 7C-E shows the histological observation of the pellets for HE, AB, and ICH of collagen II staining. The AB staining shows the greatest intensity in the pellets treated with Mg-BGNs-2/DCECM. The IHC collagen II of the Mg-BGNs-2/DCECM group was greater than that of the Mg-BGNs-1/DCECM group, which was also greater than that of the DCECM group.

Chondrogenic Differentiation Induction of Scaffolds
We evaluated the chondrogenic differentiation induction capacity of the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds in BMSC pellet culture in vitro. The pellets were cultured in a chondrogenic conditioned medium for 21 days, as shown in Figure 7B. Figure 7C-E shows the histological observation of the pellets for HE, AB, and ICH of collagen II staining. The AB staining shows the greatest intensity in the pellets treated with Mg-BGNs-2/DCECM. The IHC collagen II of the Mg-BGNs-2/DCECM group was greater than that of the Mg-BGNs-1/DCECM group, which was also greater than that of the DCECM group.  To evaluate the glycosaminoglycan (GAG) content of the pellets treated with the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds, we calculated GAG normalized for DNA content. GAG per DNA content of the Mg-BGNs-2/DCECM group was significantly higher than that of the Mg-BGNs-1/DCECM group (p < 0.05). This was also higher than that of the DCECM group (p < 0.05) ( Figure 7F).
The expression of chondrogenic differentiation-related genes (SOX 9, Col II, Aggrecan, and Col I) in the pellets treated with the DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds were evaluated by RT-PCR. The SOX 9 expression of the pellet treated with Mg-BGNs-2/DCECM is higher than that of the DCECM and Mg-BGNs-1/DCECM group (p < 0.05) ( Figure 7G). The Col II expression of the Mg-BGNs-2/DCECM group is higher than that of the DCECM and Mg-BGNs-1/DCECM groups (p < 0.01) ( Figure 7H). The aggrecan expression of the Mg-BGNs-2/DCECM group is higher than that of the Mg-BGNs-1/DCECM group, which is, in turn, higher than that of the DCECM group (p < 0.05) ( Figure 7I). There are no significant differences in the Col I expressions among the three groups ( Figure 7J). All the results indicate that the Mg-BGNs-2/DCECM has a better chondrogenic differentiation induction effect than DCECM and Mg-BGNs-1/DCECM.

Discussion
Osteochondral tissues are the continuous organization that involves both the articular cartilage and subchondral bone tissue, and both tissues play a significant role in OA onset and progression. The ideal treatment strategy not only promotes cartilage repair but also enhances subchondral bone regeneration. Tissue engineering scaffolds with the capacity to simultaneously stimulate cartilage and subchondral bone regeneration may be an ideal therapeutic strategy for osteochondral lesions. Bioactive glasses are promising biomaterials for the treatment of osteochondral injuries due to their dual functions in bone and cartilage regeneration. A number of studies have already confirmed that bioactive glasses can promote osteogenic differentiation of MSCs and are beneficial for bone tissue regeneration [33,34]. In recent years, some studies have found that some therapeutic ions, such as Mg and Si, have some beneficial effects on cartilage regeneration. Mg plays a key role in cellular energy metabolism and has a positive effect on chondrogenesis of MSCs through integrin-signaling proteins, TGFB1, octamer-binding protein (NONO), and others [35]. Furthermore, Mg can enhance chondrocyte growth and cartilage extracellular matrix synthesis (aggrecan and collagen II) [20]. Si can stimulate cartilage reconstruction by activating the HIF pathway and has a positive effect on OA therapy [36]. Considering that bioactive glasses are excellent carrier systems for ions, bioactive glasses containing therapeutic ions might be ideal tissue engineering biomaterials for osteochondral regeneration.
In this study, we fabricated sol-gel-derived Mg-BGNs. We then incorporated Mg-BGNs into a DCECM to construct a Mg-BGNs/DCECM composite scaffold. The overall aim of the present study was to investigate the effects of incorporating Mg-BGNs into DCECM scaffolds. These include the microarchitectural and physicochemical properties, cytocompatibility, and the ability of the scaffolds to enhance chondrogenic differentiation in BMSC pellet culture. The results demonstrate that Mg-containing bioactive glass nanospheres can be successfully fabricated by the sol-gel method. Furthermore, the three scaffolds (DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM scaffolds) constructed via the freeze-drying method were highly porous, had suitable pore size for tissue engineering, and sustainably released Mg, Si, and Ca ions. Moreover, all three scaffolds have excellent biocompatibility, showing no signs of toxicity. Most importantly, the addition of Mg-BGNs not only resulted in cell proliferation but also enhanced chondrogenic differentiation of MSCs in the pellet culture in a dose-dependent manner. Collectively, these results indicate that the Mg-BGNs/DCECM scaffold developed in this study shows potential for use in osteochondral lesion repair.
We proposed the development of a multifunctional composite scaffold for the treatment of osteochondral lesions, which requires a matrix that is not only capable of supporting cartilage regeneration but is also capable of contributing to bone regeneration.
Bioactive glass is an ideal biomaterial for bone regeneration. Various studies have already explored its osteogenic capabilities. Therefore, in this study, we focused on the induction of chondrogenic differentiation by Mg-BGNs. We found that the incorporation of Mg-BGNs into the DCECM scaffold successfully enhances the chondrogenic differentiation of MSCs in pellet culture. The Mg-BGNs-2/DCECM scaffold resulted in a significant increase in GAG secretion in comparison to the Mg-BGNs-1/DCECM, which also had a higher GAG content than the DCECM control group. The AB and Col II staining indicates that the pellet treated with Mg-BGNs-2/DCECM shows higher intensity than that of Mg-BGNs-1/DCECM and DCECM. Additionally, the RT-PCR results also show that the expression of the chondrogenic differentiation-related genes SOX 9, Col II, and Aggrecan in the Mg-BGNs-2/DCECM group are higher than those of the DCECM control group.
To the best of our knowledge, this is the first study to successfully demonstrate the incorporation of Mg-containing bioactive glass nanospheres into a natural DCECM scaffold. Before freeze-drying, we added Mg-BGNs to the DCECM hydrogels and made them homogeneously distributed throughout the hydrogels. We fabricated three different scaffolds: DCECM, Mg-BGNs-1/DCECM, and Mg-BGNs-2/DCECM. The SEM results indicate that the pore sizes of all three scaffolds are similar and within the range of 80-200 µm, which is the ideal range for tissue engineering applications according to a previous study [37]. The FTIR and XPS results confirm the successful incorporation of Mg-BGNs into the DCECM scaffold. Additionally, ICP-AES analysis demonstrates that Mg-BGNs/DCECM can control the release of Mg, Si, and Ca ions in a dose-dependent manner. Cell viability analysis indicates that all three scaffolds show excellent cytocompatibility and no signs of toxicity. Cytocompatibility analysis shows that all three scaffolds promote cell adhesion and growth. The cell proliferation analysis indicates that there is a significant increase in cell number on the scaffolds containing Mg-BGNs, which may be attributed to the ionic extracts from the Mg-BGNs/DCECM scaffold. In addition, it has been found that the combination of Mg-BGNs with DCECM could significantly promote chondrogenesis in comparison to the DCECM-only control. Thus, we can postulate that the enhanced cartilage regenerative capacity of the Mg-BGNs/DCECM scaffold can be attributed to the combination of Mg-containing bioactive glass nanospheres.
This study had some limitations. First, it only focused on the assessment of the feasibility of the application of Mg-BGNs/DMECM composite scaffolds for cartilage regeneration in vitro. The assessment of these novel materials in an in vivo study will be performed in our next study. Second, it must be highlighted that the incorporation of bioactive glass into DCECM scaffolds would alter the stiffness of the construction, and this may also influence the differentiation of MSCs. As this study only focused on the in vitro assessment, the biomechanical variation of the addition of BG into DCECM scaffolds would also be performed in our next study, together with the in vivo study. Third, the containing certain therapeutic ions (like magnesium) bioactive glass could favor hydroxyapatite (HA) formation, because magnesium and calcium could be incorporated into hydroxyapatite crystals, while HA deposition is unwanted in cartilage. However, it was unclear whether the containing certain therapeutic ions bioactive glass could cause heterotopic ossification in the cartilage layer. This is also a limitation of this study, and we will further investigate the HA deposition issue possibly produced by the bioactive glass in the following study. Furthermore, the composite scaffold used in this study was homogeneous architecture, while the osteochondral structure was heterogeneous. Finally, although many studies have confirmed that bioactive glasses are widely used for bone tissue regeneration, the osteogenic capacity of the Mg-BGNs/DCECM composite scaffold is still not very clear. We would like to evaluate its effect on osteogenesis in a future study.

Conclusions
In this study, Mg-containing bioactive glass nanospheres (Mg-BGNs) were successfully prepared using the sol-gel method and incorporated into the DCECM to construct a Mg-BGNs/DCECM composite scaffold. The Mg-BGNs/DCECM composite scaffold had a highly porous structure, suitable pore size for tissue engineering, good biocompatibility, promoted MSC proliferation and showed enhanced chondrogenic differentiation in comparison to the DCECM scaffold. Given the results of this study, we suggest that the Mg-BGNs/DCECM composite scaffold can simultaneously stimulate articular cartilage and subchondral bone regeneration, which satisfies the requirement for osteochondral lesions, and provides an alternative selection for cartilage/bone regeneration.