ECM Decorated Electrospun Nanofiber for Improving Bone Tissue Regeneration

Optimization of nanofiber surface properties can lead to enhanced tissue regeneration outcomes in the context of bone tissue engineering. Herein, we developed a facile strategy to decorate elctrospun nanofibers using extracellular matrix (ECM) in order to improve their performance for bone tissue engineering. Electrospun PLLA nanofibers (PLLA NF) were seeded with MC3T3-E1 cells and allowed to grow for two weeks in order to harvest a layer of ECM on nanofiber surface. After decellularization, we found that ECM was successfully preserved on nanofiber surface while maintaining the nanostructure of electrospun fibers. ECM decorated on PLLA NF is biologically active, as evidenced by its ability to enhance mouse bone marrow stromal cells (mBMSCs) adhesion, support cell proliferation and promote early stage osteogenic differentiation of mBMSCs. Compared to PLLA NF without ECM, mBMSCs grown on ECM/PLLA NF exhibited a healthier morphology, faster proliferation profile, and more robust osteogenic differentiation. Therefore, our study suggests that ECM decoration on electrospun nanofibers could serve as an efficient approach to improving their performance for bone tissue engineering.


Introduction
Large-sized bone defects cannot heal by themselves, due to the limited regenerative capability of bone tissue; therefore, a variety of bone regeneration strategies have been developed in the past 30 years to address this critical clinical problem [1][2][3]. Among these approaches, bone tissue engineering, which uses a combination of scaffolds, cells, and growth factors to regenerate new bone, still holds the promise of fully solving this important matter [4,5]. As one of three key components for bone tissue engineering, scaffolds play a pivotal role in successfully regenerating bone. Researchers have developed various approaches to fabricating scaffolds for bone regeneration, such as salt-leaching, freeze-drying, gas foaming, phase separation, and sintering, etc. [6][7][8][9][10][11]. For instance, Ma et al. developed a series of scaffolds with controllable pore size, porosity and other parameters using a phase separation method [6]. Bone tissue is generally considered to be a nanomaterial with nanoscale collagen fibers and hydroxyapatite organized into a hybrid structure [12,13]. However, most of these methods are capable of making scaffolds with micrometer features and fail to mimic the natural structure of bone.
Electrospinning is a facile and versatile method for fabricating fibrous scaffolds from a rich variety of materials [14][15][16]. Electrospun scaffolds possess very high surface-to-volume ratio, pore sizes ranging from several to tens of micrometers, and adjustable porosities of up to more

Electrospinning
To prepare a PLLA nanofibrous scaffold, PLLA (M n = 60.145 kDa, MWD = 1.64) pellets were dissolved in dichloromethane (DCM)/N,N-dimethylformamide (DMF) (50/50 vol %) solvent system at 12 wt %. The PLLA solution was fed into a plastic syringe connected to a stainless steel capillary tube at a feeding rate of 2.0 mL/h, controlled by a micro infusion pump (WZ-50C2, Zhejiang University Medical Instrument Co., Hangzhou, China). A positive DC high-voltage power supply (DW-P303-1AC, Tianjin Dongwen High-Voltage Power Supply Factory, Tianjin, China) was connected to the capillary tube to generate a high electric field of 2.0 Kv/cm between the capillary tube and the grounded collector, extending for a distance of 10 cm. A flat plate wrapped in aluminum foil was used as the collector to spin scaffolds with a random fibrous structure. The electrospun scaffolds were dried under vacuum at room temperature for 72 h.
The morphology of the electrospun scaffolds was observed using field emission scanning electron microscopy (Zeiss DSM 982, FESEM, Oberkochen, Germany) at 3 keV.

Cell Culture
MC3T3-E1 cells were cultured in alpha minimum essential medium (α-MEM) supplemented with 10% fetal bovine serum (FBS) (Corning, New York, NY, USA) and 1% pen-strep (Corning, New York, NY, USA). Cells were grown in a humidified atmosphere of 5% CO 2 at 37 • C. The culture medium was changed every other day. An osteogenic medium, consisting of α-MEM plus 10 mM β-glycerol phosphate and 50 mg/mL-ascorbic acid (Sigma, St. Louis, MO, USA), was used for osteogenic differentiation of the cells.
Mouse bone marrow stromal cells (mBMSCs) were derived from 6-8-week-old mice. The study was conducted in accordance with relevant national legislation on the use of animals for research and the protocol was approved by the Ethics Committee of Zhejiang University Laboratory Animal Center. Briefly, mice were euthanized using CO 2 asphyxiation. Femurs and tibias were dissected from surrounding tissues. Bone marrow was harvested by flushing bones with α-MEM medium containing 10% FBS using a 25-G needle. The cells were filtered through a 70 µm cell strainer and plated at a density of 1 × 10 8 cells in 100 mm culture dish. Half of the culture medium was replaced by fresh medium to remove the non-adherent cells at day 4. The cells were allowed to grow for 7-8 days in α-MEM containing 10% FBS and then passaged. mBMSCs were used at passage 2 for all the experiments performed in this study.

ECM Decoration on Electrospun Nanofibers
PLLA nanofibrous (PLLA NF) was cut into 8.0 mm circular samples using a biopsy punch with a diameter of 8.0 mm. All the samples were then sterilized using ethylene oxide gas before cell culture. Sterile PLLA NF samples were glued to the bottom of a 24-well plate using pharma-grade Vaseline to prevent floating and soaked in cell culture medium for 12 h. After that, MC3T3-E1 cells were seeded on PLLA NF at 5 × 10 4 cells/cm 2 . MC3T3-E1 cells were allowed to grow on PLLA NF for 2 weeks with osteogenic medium described previously. The culture was terminated after 2 weeks by washing the samples with PBS once and all the samples were fixed in 4% paraformaldehyde and buffered with PBS for 30 min. After that, all samples were collected in conical tubes and frozen in a −20 • C freezer overnight and then thawed in a 37 • C water bath. To remove all the cellular components of MC3T3-E1 cells, the cells were treated with 1% SDS (Life Technology, Waltham, MA, USA) in deionized water for 30 min ( Figure S1). The decellularized samples were washed with PBS extensively to remove any detergent residue and cellular components. The obtained ECM decorated PLLA NF (ECM/PLLG NF) were kept in PBS at 4 • C for further use.

Cell Attachment and Morphology
mBMSCs were seeded on both PLLA NF and ECM/PLLA NF in a 24-well plate at a seeding density of 2 × 10 4 cells/cm 2 in 1.0 mL medium (n = 10). The cells were allowed to adhere to the test materials for 2 h. After that, all the cell culture medium was slowly aspirated, and all the samples were washed gently with PBS twice. Half of the samples were then fixed in 2.5% glutaraldehyde for 1 h. Then they were incubated in 0.1 mol/L sodium cacodylate buffer for 1 h. The specimens with fixed cells were dehydrated in graded ethanol series and critical point dried. All specimens were then sputter coated with gold-palladium. The cell morphology on different substrates was observed using FESEM (JEOL 6335F, Akishima, Japan) at an acceleration voltage of 2 keV. The other half samples were then washed by PBS three times to remove the unattached and loosely bound cells from material surfaces. CellTiter-Blue ® assay was used to measure the density of cells left on the samples. 0.5 mL fresh medium containing 10% CellTiter-Blue ® dye (Promega, Madison, WI, USA) was added to each well and incubated for 2 h. Aliquots of 200 µL from each well were transferred into a 96-well plate and read at (λ ex = 570 nm/λ em = 590 nm) using a fluorescence microplate reader (Bio-tek MQX, Winooski, VT, USA). The cell number on each sample was calculated based on a calibration curve prepared with predetermined cell numbers. In another set of experiments, mBMSCs were seeded on both PLLA NF and ECM/PLLA NF in a 24-well plate at a seeding density of 2 × 10 4 cells/cm 2 in 1.0 mL medium (n = 5). The cells were allowed to grow on the specimens for 7 days. After that, cells were fixed by 4% paraformaldehyde in PBS for 20 min at room temperature. Cells were permeabilized in 0.5% Triton X-100 in PBS for 15 min. To visualize both actin and the nucleus, tetramethylrhodamine isothiocyanate (TRITC)-conjugated phalloidin and 0.5 µg/mL 4 ,6-diamidino-2-phenylindole dihydro-chloride (DAPI) were added to the cells and incubated for 30 min. After extensive washing with PBS, the cells were imaged using a fluorescent microscope (Zeiss Axiovert 200M, Oberkochen, Germany) with filters appropriate for TRITC and DAPI.

Cell Metabolism
In another set of experiments, mBMSCs were seeded onto the scaffolds (PLLA NF and ECM/PLLA NF) at a density of 2 × 10 4 cells/cm 2 in 1.0 mL of medium (n = 8). At each predetermined time points (day 3, 7 and 14), each scaffold was carefully transferred into another 24-well. 0.5 mL 10% CellTilter Blue in α-MEM was added into each well and incubated at 37 • C for 2 h. Aliquots of 200 µL from each well were transferred into a 96-well plate and read at (λ ex = 570 nm/λ em = 590 nm) using a fluorescence microplate reader (Bio-tek MQX, Winooski, VT, USA). The cell number on each sample was calculated based on a calibration curve prepared with predetermined cell numbers.
Finally, the cell morphology on different substrates was examined using field emission scanning electron microscopy (FESEM, LEO/Zeiss DSM 982, Oberkochen, Germany). To assess the cell distribution and extracellular matrix deposition, mBMSCs were seeded and cultured on PLLA NF and ECM/PLLA NF for 14 days. After that, all the specimens were prepared as described earlier in this section and then subjected to FESEM observations.

ALP Activity
The osteogenic differentiation of mBMSCs was determined by measuring their alkaline phosphatase (ALP) activity. Briefly, after 3, 7 and 14 days of culture, all media were removed, and each specimen was washed with PBS twice. 200 µL lysis buffer containing 0.5% Triton X-100 was added into each well and incubated for 20 min. The cell lysis was then centrifuged at 3000× g at 4 • C for 5 min. Aliquots of supernatants were subjected to a total protein assay using a BCA assay kit (Thermo Scientific, Waltham, MA, USA). The ALP activity was measured by colorimetric assay with reagent mixture composed of 5 mM p-nitro-phenol phosphate disodium (p-NPP), 1 mM MgCl 2 , and 0.15 M 2-amino-2-methyl-1-propanol (AMP) (Sigma, St. Louis, MO, USA) together with an equal volume amount of nitrophenyl phosphate (10 mM). The optical density of the solution was measured at 405 nm using a microplate reader (Biotek, MQX200, Winooski, VT, USA). The ALP activity was expressed as total activity per microgram protein per sample.

Characterization of Mineralized Tissue Formation
Formation of mineralized tissue was assessed by Alizarin Red-S (ARS) staining at 14 days of osteogenic culture. The cultured specimens were washed with PBS and fixed in 10% (v/v) formaldehyde at RT for 30 min. The cells were then washed twice with excess dH2O prior to addition of 1 mL of 40 mM ARS (pH 4.1) per well for 30 min. After aspiration of the unincorporated ARS, the wells were washed four times with 4 mL dH2O while shaking for 10 min. The specimens were then imaged using a Cannon scanner (Tokyo, Japan). To quantify the amount of calcium deposited on each type of specimen, the mineral component was dissolved in 1.0 M HCl and the calcium concentration was measured as follows: 5 µL of sample solution was mixed with 195 µL of a 0.4 mM Arsenazo III prepared in 0.02 M Tris buffer (pH = 7.40). The amount of calcium quantitatively detected by measuring the absorbance of the 615 nm band on a microplate reader (Biotek, MQX200, Winooski, VT, USA) and compared it to a set of standards with known calcium concentrations.

Mineralization
To test the apatite formation capability on both PLLA NF and ECM/PLLA NF, the nanofibrous mats were incubated in modified simulated body fluid (mSBF) at 37 • C 7 days with continuous rotation. The mSBFs were prepared by adding the following reagents into deionized water in the order shown: 141 mM NaCl, 4.0 mM KCl, 0.5 mM MgSO 4 , 1.0 mM MgCl 2 , 4.2 mM NaHCO 3 , 20.0 mM HEPES, 5.0 mM CaCl 2 , and 2.0 mM KH 2 PO 4 . The pH of mSBF was then adjusted to 6.80 using HCl/NaOH. In particular, each nanofibrous mat was incubated in 50 mL mSBF in a conical tube to form the mineral coatings. The mSBFs were refreshed daily throughout the entire coating process in order to maintain consistent ion concentrations for mineral coating growth. After 7 days of incubation, the scaffolds were rinsed with deionized water and lyophilized. After that, the specimens were gently washed with de-ionized water and air dried. The formation of the apatite coating on the surface of titanium disks was then observed using FESEM (JEOL 6335F Akishima, Japan) at 5 kV.

Statistical Analysis
All quantitative data are given as mean ± standard deviation. The numerical results obtained in this study were subjected to a two-tailed Student's t-test. The significance of the results was evaluated at a significance level of p < 0.05 based on the p value of comparison.

Generation and Characterization of ECM Decorated PLLA NF
Similar to decellularized tissues, cell-derived ECM provides a natural scaffold for cell attachment, proliferation and differentiation as it is a bioactive materials containing a series of fibrillary protein, matrix molecules, and embedded growth factors [36]. Therefore, the utility of cell-derived ECM for enhancing the performance of electrospun PLLA nanofibers is confirmed to be a feasible approach. Specifically, we used MC3T3-E1 cell line to generate ECM on PLLA NF surface, as its ECM contains osteogenic and osteoinductive factors accumulated during its differentiation.

Mineralization
To test the apatite formation capability on both PLLA NF and ECM/PLLA NF, the nanofibrous mats were incubated in modified simulated body fluid (mSBF) at 37 °C 7 days with continuous rotation. The mSBFs were prepared by adding the following reagents into deionized water in the order shown: 141 mM NaCl, 4.0 mM KCl, 0.5 mM MgSO4, 1.0 mM MgCl2, 4.2 mM NaHCO3, 20.0 mM HEPES, 5.0 mM CaCl2, and 2.0 mM KH2PO4. The pH of mSBF was then adjusted to 6.80 using HCl/NaOH. In particular, each nanofibrous mat was incubated in 50 mL mSBF in a conical tube to form the mineral coatings. The mSBFs were refreshed daily throughout the entire coating process in order to maintain consistent ion concentrations for mineral coating growth. After 7 days of incubation, the scaffolds were rinsed with deionized water and lyophilized. After that, the specimens were gently washed with de-ionized water and air dried. The formation of the apatite coating on the surface of titanium disks was then observed using FESEM (JEOL 6335F Akishima, Japan) at 5 kV.

Statistical Analysis
All quantitative data are given as mean ± standard deviation. The numerical results obtained in this study were subjected to a two-tailed Student's t-test. The significance of the results was evaluated at a significance level of p < 0.05 based on the p value of comparison.

Generation and Characterization of ECM Decorated PLLA NF
Similar to decellularized tissues, cell-derived ECM provides a natural scaffold for cell attachment, proliferation and differentiation as it is a bioactive materials containing a series of fibrillary protein, matrix molecules, and embedded growth factors [36]. Therefore, the utility of cellderived ECM for enhancing the performance of electrospun PLLA nanofibers is confirmed to be a feasible approach. Specifically, we used MC3T3-E1 cell line to generate ECM on PLLA NF surface, as its ECM contains osteogenic and osteoinductive factors accumulated during its differentiation.  In order to obtain ECM decoration on PLLA NF surface, we employed a cell-based strategy to generate ECM coating on the nanofibrous PLLA surface. As shown in Figure 1A, nanofibers with uniform size were formed after electrospinnnig. Although fiber size variation can still be observed when zoomed in (Figure 1B,C,G,H), the average diameter of the nanofibers was around 350 nm based on image-based statistics ( Figure 1I). After decorating with ECM, it was observed that a layer of ECM uniformly covered on the nanofiber while still preserving the nanoscale morphology of the electrospun fibers. Compared to nanofibers without ECM decoration, it was noticed that the average size of the nanofibers was slightly increased, which is probably due to the swelling of PLLA during long cell culture times.

Cell Attachment and Spread
Although polyester-based nanofibers fabricated by electrospinning have been widely used in tissue engineering and regenerative medicine [15], the cell attachment behaviors of these materials are still considered to be suboptimal due to their high hydrophobicity. We hypothesized that ECM decoration on electrospun nanofibers could significantly improve cell attachment by providing abundant adhesive components such as fibronectin, vitronectin, etc. on the materials' surface [37,38]. This hypothesis was verified by showing a high cell attachment ratio on ECM/PLLA NF compared to PLLA NF ( Figure 2G). The percentage of cells attached to ECM/PLLA NF was 30% more than for PLLA NF. The morphology of mBMSCs after being seeded on nanofibers is shown in Figure 2. Cells on PLLA NF exhibited a round morphology with limited filopodia extension ( Figure 2A). The side view of individual cell showed that the cell only physically attached to the nanofiber, and no further interactions between the nanofibers and the cell were observed ( Figure 2B,C). In contrast, rBMSCs grown on ECM/PLLA NF demonstrated a flat polygonal shape on the material surface, and most of them exhibited multipolar extensions. As shown in Figure 2E, cells on ECM/PLLA NF tend to stretch their filopodia along the nanofiber direction, indicating that the ECM coating on these fibers had remarkably enhanced their interactions with cell anchoring points. This feature of the ECM decorated nanofibers was clearly visible in high-magnification SEM images, which clearly showed that the cells spread following the orientation of the nanofibers. Therefore, ECM deposited by MC3T3-E1 cells can effectively improve cell attachment on electrospun nanofibers. In order to obtain ECM decoration on PLLA NF surface, we employed a cell-based strategy to generate ECM coating on the nanofibrous PLLA surface. As shown in Figure 1A, nanofibers with uniform size were formed after electrospinnnig. Although fiber size variation can still be observed when zoomed in (Figure 1B,C,G,H), the average diameter of the nanofibers was around 350 nm based on image-based statistics ( Figure 1I). After decorating with ECM, it was observed that a layer of ECM uniformly covered on the nanofiber while still preserving the nanoscale morphology of the electrospun fibers. Compared to nanofibers without ECM decoration, it was noticed that the average size of the nanofibers was slightly increased, which is probably due to the swelling of PLLA during long cell culture times.

Cell Attachment and Spread
Although polyester-based nanofibers fabricated by electrospinning have been widely used in tissue engineering and regenerative medicine [15], the cell attachment behaviors of these materials are still considered to be suboptimal due to their high hydrophobicity. We hypothesized that ECM decoration on electrospun nanofibers could significantly improve cell attachment by providing abundant adhesive components such as fibronectin, vitronectin, etc. on the materials' surface [37,38]. This hypothesis was verified by showing a high cell attachment ratio on ECM/PLLA NF compared to PLLA NF ( Figure 2G). The percentage of cells attached to ECM/PLLA NF was 30% more than for PLLA NF. The morphology of mBMSCs after being seeded on nanofibers is shown in Figure 2. Cells on PLLA NF exhibited a round morphology with limited filopodia extension (Figure 2A). The side view of individual cell showed that the cell only physically attached to the nanofiber, and no further interactions between the nanofibers and the cell were observed ( Figure 2B,C). In contrast, rBMSCs grown on ECM/PLLA NF demonstrated a flat polygonal shape on the material surface, and most of them exhibited multipolar extensions. As shown in Figure 2E, cells on ECM/PLLA NF tend to stretch their filopodia along the nanofiber direction, indicating that the ECM coating on these fibers had remarkably enhanced their interactions with cell anchoring points. This feature of the ECM decorated nanofibers was clearly visible in high-magnification SEM images, which clearly showed that the cells spread following the orientation of the nanofibers. Therefore, ECM deposited by MC3T3-E1 cells can effectively improve cell attachment on electrospun nanofibers.

Cell Proliferation and Differentiation
To determine the effect of ECM decoration on mBMSCs viability, cell numbers on different treated PLLA nanofibers were recorded at day 3, 7 and 14. As shown in Figure 3A, cell numbers on both tested groups increased significantly as culture time extended. At day 3, the cell numbers on

Cell Proliferation and Differentiation
To determine the effect of ECM decoration on mBMSCs viability, cell numbers on different treated PLLA nanofibers were recorded at day 3, 7 and 14. As shown in Figure 3A, cell numbers on both tested groups increased significantly as culture time extended. At day 3, the cell numbers on ECM/PLLA NF were slightly higher than those on PLLA NF, which might be due to the faster cell settlement on ECM-decorated nanofibers. The difference in cell numbers between ECM/PLLA NF and PLLA NF had become even more significant by day 7. At day 14, the cell numbers for ECM treated group were 70% higher than those for PLLA NF, indicating that the presence of ECM not only improved cell attachment at the initial stage, but also provided proliferative signal, stimulating the cells to grow faster. ECM/PLLA NF were slightly higher than those on PLLA NF, which might be due to the faster cell settlement on ECM-decorated nanofibers. The difference in cell numbers between ECM/PLLA NF and PLLA NF had become even more significant by day 7. At day 14, the cell numbers for ECM treated group were 70% higher than those for PLLA NF, indicating that the presence of ECM not only improved cell attachment at the initial stage, but also provided proliferative signal, stimulating the cells to grow faster. ALP activity was evaluated as an early marker of osteogenic differentiation of mBMSCs cells. Although ALP activity in both groups was very low at day 3, it was noticed that ALP on ECM/PLLA NF was still slightly higher than PLLA NF group ( Figure 3B). ALP activity on both tested groups increased gradually between 3 and 7 days, and there is a noticeable difference between ECM/PLLA NF and PLLA NF, suggesting that ECM presence on nanofiber surface could increase ALP activity, which may be further assigned to mineral deposition during MC3T3-E1 differentiation. A dramatic increase in ALP activity was observed during the second week of culture due to the switch of culture medium from basal medium to osteogenic medium. Although the ALP activity on PLLA NF surfaces increased by over 3 folds, the ALP activity on the ECM-treated surfaces were still much higher than the untreated group, which indicates that the ECM decoration approach used here is more favorable for osteogenic differentiation. Similar results have also been observed by other investigators in the context of bone tissue engineering. For instance, Sutha et al. showed in their study that embryonic stem cell-derived ECM possessed osteoinductive factors, which promoted ectopic bone formation in vivo [39]. Since MC3T3-E1 cell is an osteoblastic cell line which can differentiate into osteoblast, osteogenic factors can be secreted during their differentiation [40]. Thus, the enhanced osteogenic differentiation of mBMSCs on ECM/PLLA NF might be attributed to the osteogenic signals contained in MC3T3-E1 derived ECM.  Figure 4 shows the mBMSC cells growing on the two types of surface after 14 days of culture; however, cell coverage on ECM/PLLA NF appears to be more uniform compared to PLLA NF. Similar results were also observed in Figure 5 using SEM. Both surfaces were covered with a layer of cells. In particular, cells on PLLA NF exhibited a flatter morphology and tended to fully spread out. However, there are still small gaps between cells, and nanofibers were still exposed without cell layers covering them. Meanwhile, mBMSCs on ECM/PLLA NF grew tightly to each other and tended to form cell colonies ( Figure 5D). Moreover, these cells demonstrated a spindle-shaped morphology, which is more similar to the morphology of osteoblasts. At a high magnification, although cells on PLLA NF showed a healthy morphology, the surface was only covered a single layer of cells. In comparison, ECM/PLLA NF surface was heavily covered by multiple layers of mBMSCs, which displayed numerous interactions between different cells ( Figure 5B,E). Furthermore, we also found that new ALP activity was evaluated as an early marker of osteogenic differentiation of mBMSCs cells. Although ALP activity in both groups was very low at day 3, it was noticed that ALP on ECM/PLLA NF was still slightly higher than PLLA NF group ( Figure 3B). ALP activity on both tested groups increased gradually between 3 and 7 days, and there is a noticeable difference between ECM/PLLA NF and PLLA NF, suggesting that ECM presence on nanofiber surface could increase ALP activity, which may be further assigned to mineral deposition during MC3T3-E1 differentiation. A dramatic increase in ALP activity was observed during the second week of culture due to the switch of culture medium from basal medium to osteogenic medium. Although the ALP activity on PLLA NF surfaces increased by over 3 folds, the ALP activity on the ECM-treated surfaces were still much higher than the untreated group, which indicates that the ECM decoration approach used here is more favorable for osteogenic differentiation. Similar results have also been observed by other investigators in the context of bone tissue engineering. For instance, Sutha et al. showed in their study that embryonic stem cell-derived ECM possessed osteoinductive factors, which promoted ectopic bone formation in vivo [39]. Since MC3T3-E1 cell is an osteoblastic cell line which can differentiate into osteoblast, osteogenic factors can be secreted during their differentiation [40]. Thus, the enhanced osteogenic differentiation of mBMSCs on ECM/PLLA NF might be attributed to the osteogenic signals contained in MC3T3-E1 derived ECM. Figure 4 shows the mBMSC cells growing on the two types of surface after 14 days of culture; however, cell coverage on ECM/PLLA NF appears to be more uniform compared to PLLA NF. Similar results were also observed in Figure 5 using SEM. Both surfaces were covered with a layer of cells. In particular, cells on PLLA NF exhibited a flatter morphology and tended to fully spread out. However, there are still small gaps between cells, and nanofibers were still exposed without cell layers covering them. Meanwhile, mBMSCs on ECM/PLLA NF grew tightly to each other and tended to form cell colonies ( Figure 5D). Moreover, these cells demonstrated a spindle-shaped morphology, which is more similar to the morphology of osteoblasts. At a high magnification, although cells on PLLA NF showed a healthy morphology, the surface was only covered a single layer of cells. In comparison, ECM/PLLA NF surface was heavily covered by multiple layers of mBMSCs, which displayed numerous interactions between different cells ( Figure 5B,E). Furthermore, we also found that new ECM deposition on ECM/PLLA NF was more abundant than PLLA NF. Mineralized nodules can be clearly seen on ECM/PLLA NF, whereas they are sparse on PLLA NF ( Figure 5C,F), which implies that decoration of ECM secreted by MC3T3-E1 cells enhanced newly deposited ECM by BMSC cells and osteoblastic cells. ECM deposition on ECM/PLLA NF was more abundant than PLLA NF. Mineralized nodules can be clearly seen on ECM/PLLA NF, whereas they are sparse on PLLA NF ( Figure 5C,F), which implies that decoration of ECM secreted by MC3T3-E1 cells enhanced newly deposited ECM by BMSC cells and osteoblastic cells.  We also studied the mineral deposition in terms of calcium on both surfaces after two weeks. It was found that both groups were mineralized, showing a layer of calcium mineral on the surface ( Figure 6). Nevertheless, it was found that calcium deposition on ECM/PLLA NF was about 12.59 mg/sample, which is much higher than that of PLLA NF (8.72 mg/sample); thus, decoration of MC3T3-E1 derived ECM substantially increased calcium deposition during mBMSC differentiation.   We also studied the mineral deposition in terms of calcium on both surfaces after two weeks. It was found that both groups were mineralized, showing a layer of calcium mineral on the surface ( Figure 6). Nevertheless, it was found that calcium deposition on ECM/PLLA NF was about 12.59 mg/sample, which is much higher than that of PLLA NF (8.72 mg/sample); thus, decoration of MC3T3-E1 derived ECM substantially increased calcium deposition during mBMSC differentiation. It is shown that mBMSCs grown on ECM/PLLA NF developed multilayer structure and ECM deposition on this surface is more robust. Red arrow: mineralized nodule formation after culture.

Cell Morphology and Mineral Deposition
We also studied the mineral deposition in terms of calcium on both surfaces after two weeks. It was found that both groups were mineralized, showing a layer of calcium mineral on the surface ( Figure 6). Nevertheless, it was found that calcium deposition on ECM/PLLA NF was about 12.59 mg/sample, which is much higher than that of PLLA NF (8.72 mg/sample); thus, decoration of MC3T3-E1 derived ECM substantially increased calcium deposition during mBMSC differentiation.

Biomineralization
Biomineralization capability is closely related to the bone-forming potential of biomaterials in vivo; therefore, it is commonly used to characterize the bioactivity of biomaterials [41]. To investigate the effect of ECM decoration on PLLA nanofiber biomineralization, both PLLA NF and ECM/PLLA NF were soaked in mSBF to evaluate their mineral-forming capability. As shown in Figure 7A, scattered mineral clusters were formed on PLLA NF after soaking in mSBF for 24 h. When zoomed in, it can be noticed that these minerals mainly grew along the fiber directions, instead of uniformly coating the surface ( Figure 7A,B). On the other hand, mineral growth on ECM/PLLA NF showed a much faster trend. It was found that mineral coating had homogeneously formed along the nanofibers, which increased the diameter of the nanofibers from several-hundred nanometers to about 1 µm ( Figure 7E,F). The localization of mineral coating on nanofibers indicates that mineral nucleation mainly occurred on the nanofibers, which may be due to the presence of MC3T3-E1derived ECM. When we extended the biomineralization time to 48 h, it was found that a thick mineral coating covered the whole nanofiber surface, causing PLLA NF to lose its nanoscale features ( Figure  7C,D). However, this situation did not occur with ECM/PLLA NF. Instead, the mineral coating continued to grow along the nanofiber direction, and preserved its nanoscale morphology ( Figure  7G,H). The results from biomineralization indicate that ECM decoration can facilitate mineral deposition on PLLA nanofiber surface, which is beneficial for bone regeneration.

Biomineralization
Biomineralization capability is closely related to the bone-forming potential of biomaterials in vivo; therefore, it is commonly used to characterize the bioactivity of biomaterials [41]. To investigate the effect of ECM decoration on PLLA nanofiber biomineralization, both PLLA NF and ECM/PLLA NF were soaked in mSBF to evaluate their mineral-forming capability. As shown in Figure 7A, scattered mineral clusters were formed on PLLA NF after soaking in mSBF for 24 h. When zoomed in, it can be noticed that these minerals mainly grew along the fiber directions, instead of uniformly coating the surface ( Figure 7A,B). On the other hand, mineral growth on ECM/PLLA NF showed a much faster trend. It was found that mineral coating had homogeneously formed along the nanofibers, which increased the diameter of the nanofibers from several-hundred nanometers to about 1 µm ( Figure 7E,F). The localization of mineral coating on nanofibers indicates that mineral nucleation mainly occurred on the nanofibers, which may be due to the presence of MC3T3-E1-derived ECM. When we extended the biomineralization time to 48 h, it was found that a thick mineral coating covered the whole nanofiber surface, causing PLLA NF to lose its nanoscale features ( Figure 7C,D). However, this situation did not occur with ECM/PLLA NF. Instead, the mineral coating continued to grow along the nanofiber direction, and preserved its nanoscale morphology ( Figure 7G,H). The results from biomineralization indicate that ECM decoration can facilitate mineral deposition on PLLA nanofiber surface, which is beneficial for bone regeneration. inserted images were Alizarin Red S staining of calcium. Calcium deposition were significantly higher on ECM/PLLA NF than on PLLA NF (p < 0.05) after 14 days. * indicates significant differences.

Biomineralization
Biomineralization capability is closely related to the bone-forming potential of biomaterials in vivo; therefore, it is commonly used to characterize the bioactivity of biomaterials [41]. To investigate the effect of ECM decoration on PLLA nanofiber biomineralization, both PLLA NF and ECM/PLLA NF were soaked in mSBF to evaluate their mineral-forming capability. As shown in Figure 7A, scattered mineral clusters were formed on PLLA NF after soaking in mSBF for 24 h. When zoomed in, it can be noticed that these minerals mainly grew along the fiber directions, instead of uniformly coating the surface ( Figure 7A,B). On the other hand, mineral growth on ECM/PLLA NF showed a much faster trend. It was found that mineral coating had homogeneously formed along the nanofibers, which increased the diameter of the nanofibers from several-hundred nanometers to about 1 µm ( Figure 7E,F). The localization of mineral coating on nanofibers indicates that mineral nucleation mainly occurred on the nanofibers, which may be due to the presence of MC3T3-E1derived ECM. When we extended the biomineralization time to 48 h, it was found that a thick mineral coating covered the whole nanofiber surface, causing PLLA NF to lose its nanoscale features ( Figure  7C,D). However, this situation did not occur with ECM/PLLA NF. Instead, the mineral coating continued to grow along the nanofiber direction, and preserved its nanoscale morphology ( Figure  7G,H). The results from biomineralization indicate that ECM decoration can facilitate mineral deposition on PLLA nanofiber surface, which is beneficial for bone regeneration.

Conclusions
In conclusion, electrospun PLLA nanofibers were successfully decorated by ECM derived from osteoblastic cells by decellularization of MC3T3-E1 cells grown for two weeks. The results demonstrated that application of ECM on PLLA nanofibers enhanced mouse bone marrow stromal cell (mBMSC) adhesion, supported cell proliferation and promoted osteogenic differentiation of mBMSCs. Therefore, this study provided an alternative platform for directly translating the regenerative potential of ECM into clinical therapeutics and motivating the development of cell-derived ECM materials for broad regenerative medicine applications.