Tissue engineering bones take great advantages in massive bone defect repairing; under the induction of growth factors, seed cells differentiate into osteoblasts, and the scaffold materials gradually degrade and are replaced with neogenetic bones, which simulates the actual pathophysiological process of bone regeneration. However, mechanism research is required and further developed to instruct elements selection and optimization. In the present study, we prepared vascular endothelial growth factor/bone morphogenetic protein-2- nanohydroxyapatite/collagen (VEGF/ BMP-2- nHAC/ PLGAs) scaffolds and inoculated mouse MC3T3-E1 preosteoblasts to detect osteogenic indexes and activation of related signaling pathways. The hypothesis is to create a three-dimensional environment that simulates bone defect repairing, and p38 mitogen-activated kinase (p38) inhibitor was applied and osterix shRNA was transferred into mouse MC3T3-E1 preosteoblasts to further investigate the molecular mechanism of crosstalk between BMP-2 and VEGF. Our results demonstrated the following: (1) BMP-2 and VEGF were sustainably released from PLGAs microspheres. (2) nHAC/PLGAs scaffold occupied a three-dimensional porous structure and has excellent physical properties. (3) MC3T3-E1 cells proliferated and differentiated well in the scaffold. (4) Osteogenic differentiation related factors expression of VEGF/BMP-2 loaded scaffold was obviously higher than that of other groups; p38 inhibitor SB203580 decreased the nucleus/cytoplasm ratio of osterix expression. To conclude, the active artificial bone we prepared could provide a favorable growth space for MC3T3-E1 cells, and osteogenesis and maturation reinforced by simultaneous VEGF and BMP-2 treatment may be mainly through the activation of the p38 MAPK pathway to promote nuclear translocation of osterix protein.
Bone defect repair is a complex process involving several cell types as osteoblasts, endothelial cells, and so on, besides various growth factors [
Configuration and crystal size of nanohydroxyapatite (nHA) are similar to the natural bone that holds the advantage of nice biocompatibility, osteogenic activity, and osteoinduction. But slow degradation that impairs new bone formation restricts its application. The degradation rate, biocompatibility, and plasticity of collagen are all excellent, while its mechanical strength is poor [
MC3T3-E1 cells, a mouse preosteoblast cell line that has been widely used as a good model for studying osteogenic differentiation in vitro, especially under the actin of ECM signaling pathways [
PLGA microspheres were fabricated by a modified water- oil- water (w/o/w) double emulsion solvent evaporation method: The primary organic phase was prepared by dissolving PLGA (Sigma) in dichloromethane (100 g/L). VEGF or BMP-2 (PeproTech) was dissolved in bovine serum albumin (BSA, Sigma) solution (w
Tails of rats were soaked in 75% ethanol for 10 min and then washed with 0.9% sodium chloride solution three times. After skin was peeled, tendons were extracted, soaked in 75% ethanol for 15 min, and then washed with deionized water repeatedly. The tendons were trimmed using scissors into lmm3 tissue mass, soaked in 0.05M Tris / HCL solution (pH 7.5), and then stored at 4°C overnight. Tris / HCL was replaced with 0.05 M acetic acid solution (pH 4.76, 2 L acetic acid for 5 g tendon) and placed at 4°C for 4 days. The solution was centrifuged (4°C, 320 g, 10 min); the obtained supernatant was filtered through a 50 um mesh to acquire crude collagen solution. 0.14mol / L NaOH solution was added to the above solution (V
Rat tail collagen was dissolved in deionized water to form collagen solution (20mg/ml). nHAP powder (Nanjing Emperor Nano Material Co., Ltd) was homodispersed in collagen solution (w
The surface and internal morphology of the scaffold were observed and photographed using scanning electron microscopy (SEM; Hitachi) after the samples were broken and sputter-coated with gold. Then SEM images were analyzed by ImageJ software to measure pore size and microsphere particle diameter. The scaffold was immersed into hexane (primary volume was V1) to be impregnated, the total volume of hexane and scaffold was V2; the residual hexane volume was V3 after removing hexane-impregnated scaffolds. The porosity of the scaffold was calculated as (V1-V3)/(V2-V3) × 100. Water absorption rate (W) of the scaffolds was measured according to the following equation (m1 indicates dry weight of samples; m2 indicates wet weight of samples saturated with water in air): W=(m2-m1)/m1 × 100%. The compressive strength and elastic modulus of the scaffold were acquired by calculating the stress-strain curve measured using universal mechanical properties testing instrument (ZWICKZ005).
20 mg growth factor loaded microspheres were immersed in 2 ml PBS and then oscillated in 37°C constant temperature water bath for 28 days. 1 ml supernatant was aspirated and replaced with fresh phosphate buffer saline (PBS, pH 7.4, Hyclone) every 24 h. The amount of released BMP-2 or VEGF in the supernatant was detected, respectively, using ELISA assay kit (R&D) according to the instructions. After being washed, diluted standards or samples were added into each well, followed by being incubated with diluted detected antibody at room temperature for 2 hours. Then wells were washed again 6 times and filled with diluted Streptavidin-HRP to incubate at room temperature for 45 minutes, followed by being incubated with substrate solution away from light for 30 minutes at room temperature. After adding stop solution, the optical density of wells can be determined using microplate reader under double wavelength of 450 and 570/ 630 nm within 30 minutes. Concentration of samples can be calculated according to a formula derived from concentration and optical density of standards, then accumulative release curves were plotted to describe the release behavior of BMP-2/VEGF.
MC3T3-E1 cells (purchased from National Infrastructure of cell line resource) were inoculated into the scaffold and incubated with common culture medium (Dulbecco’s Modified Eagle’s Medium-high glucose plus 10% fetal bovine serum) for 1, 3, and 5 days, then samples were fixed with 2.5% glutaraldehyde overnight and then dehydrated with alcohol (30%, 50%, 70%, 80%, 90%, 95%, 100%, and 100%, each for 10 min) to be observed under SEM.
MC3T3-E1 cells were inoculated into the scaffold at a density of 1 × 108/ml and cultured in common medium with or without BMP-2 (1
MC3T3-E1 cells | scaffold | BMP-2 | VEGF | |
---|---|---|---|---|
A | + | _ | _ | _ |
B | + | + | + | _ |
C | + | + | _ | + |
D | + | + | + | + |
At the 3rd day, adherent cells were separated from scaffold using trypsin and resuspended with common culture medium. Cell suspension and 1/10 volume of cell counting kit- 8 (CCK-8) solution (Beyotime) were added to a 96-well plate and incubated at 37°C for 2 h. Then absorbance was detected using microplate reader under 450 nm wavelength to reveal proliferation of MC3T3-E1 cells.
At 7th, 14th, and 21th days, adherent cells were separated from scaffold by 0.05% Triton X-100 and ultrasonic treatment (150W, 250s) in ice bath. Then the supernatant was collected after centrifugation (4°C, 12000 g, 15 min), 20
At the 21th day, adherent cells were separated from scaffold using trypsin, then inoculated, and cultured in culture plate for 21 days. Cells were fixed and then stained using 0.2% alizarin red solution (Solarbio, pH 8.3). Formation of mineralized nodules was observed under microscope (Olympus).
At the 7th, 14th, and 21th days, cell-scaffold complex was trimmed, then immersed in RNAiso Plus (TaKaRa), and triturated evenly. The supernatant was collected after being centrifuged (4°C, 12000 g, 15 min), mixed with chloroform (V mouse GAPDH (Forward: 5′- CTTTGTCAAGCTCATTTCCTGG - 3′; Reverse: 5′- TCTTGCTCAGTGTCCTTGC - 3′) mouse Dlx-5 (Forward: 5′- AGCTACCTGGAGAACTCGG - 3′; Reverse: 5′- CCCAAAACTGAGCAAGAGAAAG - 3′) mouse osterix (Forward: 5′- CCTCTCCCTTCTCCCTCTC -3′; Reverse: 5′- CTGGAGCCATAGTGAGCTTC -3′) mouse col1 (Forward: 5′- CATAAAGGGTCATCGTGGCT - 3′; Reverse: 5′- TTGAGTCCGTCTTTGCCAG - 3′)
At the 7th, 14th, and 21th days, cell-scaffold complex was trimmed, then immersed in lysate (RIPA lysis buffer with 1 mM PMSF, Beyotime), and ultrasonically homogenized on ice to extract total protein. Concentration of protein samples was measured using BCA protein assay kit. After being mixed with loading buffer, protein samples were denatured in 95°C water bath for 10 min. Protein bands (30
p38 inhibitor SB203580 and Akt inhibitor LY294002 (20
Data were analyzed using SPSS package 20.0. Comparisons among the groups were analyzed with independent sample
nHAC/PLGAs scaffold displayed a cylindrical shape with a diameter of 0.8cm and height of 0.8cm, being milky white and rough and having a foamy surface (Figure
Physical parameters of nHAC/PLGAs scaffold(
Pore size ( | Diameter of PLGAs ( | Porosity (%) | Water absorption rate (%) | Elastic modulus (MPa) | Compressive strength (MPa) |
---|---|---|---|---|---|
176 ± 93 | 9.95 ± 1.93 | 79.46 ± 7.78 | 561.51 ± 19.59 | 0.72 ± 0.17 | 2.85 ± 0.49 |
(a) General appearance and surface topography under SEM, (b) ×100, (c) ×900 of nHAC/PLGAs scaffold, and (d) releasing characteristics of BMP-2 and VEGF from PLGA microspheres.
Burst release was observed during the first 7 days; the released amount of BMP-2 and VEGF was 60.20% and 51.98%, respectively. A steady release was observed between the 8th and 17th days; the released amount of BMP-2 and VEGF was 29.53% and 28.82%, respectively. Only 6.85% of BMP-2 and 9.37% of VEGF were released during the remaining 11 days (Figure
At the first day under SEM observation, a small number of MC3T3-E1 cells were observed; they displayed ovary or irregular morphology and loosely stuck on the scaffold surface (Figure
MC3T3-E1 cells growing in nHAC/PLGAs scaffold for 1 day (a) and 5 days (b).
Compared with the control group, the proliferative activity of VEGF group was higher (P<0.01), but that of BMP-2 group was similar (P>0.05). The proliferative activity of BMP-2/VEGF group was higher (P<0.01) than that of BMP-2 but lower (P<0.01) than that of the VEGF group (Table
Proliferative activity (OD value) and ALP activity (DEA enzyme activity,
Time (days) | Control group | VEGF group | BMP-2 group | BMP-2/VEGF group | |
---|---|---|---|---|---|
Proliferative activity | 3 | 0.8507 ± 0.0317 | 1.3987 ± 0.1304 | 0.7385 ± 0.1160## | 1.1096 ± 0.0632 |
| |||||
ALP activity | 7 | 4.2738 ± 0.8482 | 4.6884 ± 0.0979## | 29.8862 ± 1.4436 | 38.0940 ± 2.1503 |
14 | 4.3097 ± 0.8896 | 4.8018 ± 1.0833## | 47.5750 ± 3.0548 | 59.1847 ± 1.8423 | |
21 | 4.0477 ± 1.1751 | 4.9057 ± 1.1357## | 54.8533 ± 2.9016 | 78.0085 ± 6.8198 | |
| |||||
Relative expression of col 1 mRNA | 7 | 1.0927 ± 0.0758 | 1.1907 ± 0.0939## | 1.3614 ± 0.2367 | 1.6776 ± 0.2768 |
14 | 1.1592 ± 0.1075 | 1.1095 ± 0.4211## | 1.5429 ± 0.1926 | 1.9635 ± 0.3183 | |
21 | 0.9377 ± 0.0719 | 0.8569 ± 0.0447## | 2.5563 ± 0.3960 | 6.6649 ± 0.5074 | |
| |||||
Relative expression of col 1 protein | 7 | 0.0935 ± 0.0269 | 0.1288 ± 0.0201 | 0.1551 ± 0.0235 | 0.1908 ± 0.0840 |
14 | 0.0968 ± 0.0043 | 0.0963 ± 0.0277## | 0.3205 ± 0.0238 | 0.6859 ± 0.0372 | |
21 | 0.1182 ± 0.0278 | 0.1310 ± 0.0254## | 0.4019 ± 0.0099 | 0.7932 ± 0.0635 | |
| |||||
Relative expression of RUNX2 protein | 7 | 0.0163 ± 0.0017 | 0.0258 ± 0.0082## | 0.0968 ± 0.0057 | 0.0762 ± 0.0158 |
14 | 0.0210 ± 0.0026 | 0.0294 ± 0.0073## | 0.1357 ± 0.0246 | 0.1401 ± 0.0260 | |
21 | 0.0175 ± 0.0046 | 0.0292 ± 0.0064## | 0.1930 ± 0.0832 | 0.1500 ± 0.0386 | |
| |||||
Relative expression of Dlx-5 mRNA | 7 | 0.9742 ± 0.0921 | 1.0566 ± 0.0504## | 2.0851 ± 0.2293 | 2.1265 ± 0.4104 |
14 | 1.0388 ± 0.0515 | 0.9120 ± 0.3276## | 2.4911 ± 0.2984 | 3.6416 ± 0.7365 | |
21 | 0.9527 ± 0.0644 | 0.8845 ± 0.0413## | 3.3096 ± 0.3557 | 4.5190 ± 1.3563 | |
| |||||
Relative expression of Dlx-5 protein | 7 | 0.0216 ± 0.0114 | 0.0348 ± 0.0012## | 0.1017 ± 0.0254 | 0.1161 ± 0.0230 |
14 | 0.0253 ± 0.0064 | 0.0218 ± 0.0036## | 0.0865 ± 0.0239 | 0.1506 ± 0.0289 | |
21 | 0.0291 ± 0.0029 | 0.0447 ± 0.0109## | 0.2042 ± 0.0234 | 0.4216 ± 0.0362 | |
| |||||
Relative expression of osterix mRNA | 7 | 0.8809 ± 0.1061 | 0.7559 ± 0.0399# | 1.2075 ± 0.1732 | 1.3938 ± 0.4592 |
14 | 1.0333 ± 0.0924 | 0.8624 ± 0.2732## | 1.7024 ± 0.2401 | 2.1167 ± 0.4100 | |
21 | 0.8365 ± 0.4278 | 0.6243 ± 0.0860## | 2.3860 ± 0.2616 | 6.9909 ± 1.6734 | |
| |||||
Relative expression of osterix protein | 7 | 0.0171 ± 0.0024 | 0.0153 ± 0.0045## | 0.0400 ± 0.0048 | 0.0814 ± 0.0182 |
14 | 0.0166 ± 0.0042 | 0.0209 ± 0.0028## | 0.0947 ± 0.0147 | 0.5140 ± 0.0287 | |
21 | 0.0260 ± 0.0057 | 0.0243 ± 0.0036## | 0.2001 ± 0.0269 | 0.4832 ± 0.0376 |
Values are mean ± sd.
Proliferative activity (a) and ALP activity (b) of MC3T3-E1 cells in nHAC/PLGAs scaffold (con: control group; V: VEGF group; B: BMP-2 group; B+V: BMP-2/VEGF group) (
At the 7th day, compared with the control group, the ALP activity of BMP-2 group was significantly higher (P<0.01), but that of VEGF group was similar (P>0.05). ALP activity of BMP-2/VEGF group was significantly higher (P<0.01) than that of the BMP-2 and VEGF groups. At the 14th and 21th days, the ALP activity of each group displayed similar trends to that at the 7th day (Table
At the 21th day, cells in the control group displayed elongated morphology and were randomly arranged; they almost cannot be stained by alizarin red. Cells in the other three groups were arranged in a turbo form, with orange colored calcium nodules deposited in the extracellular matrix. The calcium nodules in the VEGF group were small in number and volume; they were scattered and lightly colored. The number and volume of calcium nodules in the BMP-2 group were bigger than those of the VEGF group, with several calcium nodules clustered together. Calcium nodules were filled with culture plate of BMP-2/VEGF group, with most calcium nodules clustered together and deeply colored (Figures
At the 7th day, compared with the control group, the relative expression of col1 mRNA and protein in BMP-2 group was higher (P<0.05), but that of VEGF group was similar (P>0.05). Col1 mRNA (P<0.01) and protein (P>0.05) expression of BMP-2/VEGF group were significantly or slightly higher than those of the BMP-2 and VEGF groups. At the 14th and 21th days, relative expression of col1 mRNA and protein of each group displayed similar trends to that at the 7th day, except for their expression in BMP-2/VEGF group which was significantly higher than that of BMP-2 and VEGF groups (P<0.01) (Table
Relative expression of col1 mRNA (a), col1 protein (b), and western blot images of RUNX2, Dlx-5, osterix, col1, and
At the 7th day, compared with the control group, the relative expression of RUNX-2 protein in BMP-2 group was significantly higher (P<0.01), but that of the VEGF group was similar (P>0.05). RUNX-2 protein expression of BMP-2/VEGF group was similar (P>0.05) to that of BMP-2 group and significantly higher (P<0.01) than that of VEGF group. At the 14th and 21th days, the relative expression of RUNX-2 protein of each group displayed similar trends to that at the 7th day (Table
Relative expression of RUNX2 protein (a), Dlx-5 mRNA (b), Dlx-5 protein (c), osterix mRNA (d), and osterix protein (e) of MC3T3-E1 cells in nHAC/PLGAs scaffold (
At the 7th day, compared with the control group, the relative expression of Dlx-5 mRNA and protein in BMP-2 group was significantly higher (P<0.01), but that of VEGF group was similar (P>0.05). Dlx-5 mRNA and protein expression of BMP-2/VEGF group were similar (P>0.05) to that of BMP-2 group and significantly higher (P<0.01) than that of VEGF group. At 14th and 21th days, relative expression of Dlx-5 mRNA and protein of each group displayed similar trends to that of the 7th day, except for their expression of BMP-2/VEGF group that was higher (P<0.05 or P<0.01) than that of BMP-2 group (Table
At 7th day, compared with control group, relative expression of osterix mRNA and protein in BMP-2 group was significantly higher (P<0.01), but that of VEGF group was similar (P>0.05). Osterix mRNA and protein expression of BMP-2/VEGF group were higher than that of VEGF group and slightly higher than that of BMP-2. At 14th and 21th days, relative expression of osterix mRNA and protein of each group displayed similar trends to that at the 7th day, except for their expression in BMP-2/VEGF group that was significantly higher than that of BMP-2 and VEGF group (P<0.01) (Table
pSmad expression was almost not detected in control group and VEGF group, while it was obviously detected in BMP-2 and BMP-2/VEGF group, but there was no significant difference between them. pAkt expression was almost not detected in control group and BMP-2 group, while it was obviously detected in VEGF and BMP-2/VEGF group, and its expression in BMP-2/VEGF group was significantly higher than that of VEGF group. p-p38 expression was detected in VEGF, BMP-2,and BMP-2/VEGF group; its expression in BMP-2 group was significantly higher than that in VEGF group, and its expression in BMP-2/VEGF group was significantly higher than that in VEGF and BMP-2 group. There was a little difference among the expressions of signal pathway proteins at different stages of osteogenesis; thus, the interaction between VEGF and BMP-2 enhanced the activation of PI3K/Akt and p38 MAPK pathways (Figure
(a) Expression of osteogenic related signaling pathway proteins of MC3T3-E1 cells in nHAC/PLGAs scaffold. (b) Expression of col1 protein of MC3T3-E1 cells treated with LY294002 and SB203580. (c) Expression of Dlx-5 and osterix protein of MC3T3-E1 cells treated with SB203580. (d) Expression of osterix protein in nucleus and cytoplasm of MC3T3-E1 cells in BMP-2/VEGF group (con: control group; V: VEGF group; B: BMP-2 group; B+V: BMP-2/VEGF group; nu: nucleus, cy: cytoplasm).
At the 21th day, under treatment with LY294002, no calcium nodules were detected in control group and VEGF group, only a few lightly-colored calcium nodules scattered in BMP-2 and BMP-2/VEGF group, and col1 protein expression in all groups changed a little (Figures
Calcium nodules detection of MC3T3-E1 cells treated with (a) LY294002, (b) SB203580, and (c) osterix shRNA in nHAC/PLGAs scaffold (1: control group; 2: VEGF group; 3: BMP-2 group; 4: BMP-2/VEGF group).
Osterix knockdown by lentivirus carrying osterix shRNA can be stable in our experimental period (Figures
Expression of GFP protein in MC3T3-E1 cells treated with osterix shRNA unloaded ((a) 3rd day after transfection) or loaded ((b) 3rd day and (c) 24th day after transfection) lentivirus (1. Images in fluorescent state; 2. Images in ordinary state). (d) Osterix protein expression of MC3T3-E1 cells treated with osterix shRNA loaded or unloaded lentivirus at the 3rd day and 24th day after transfection. (e) Expression of col1 protein in MC3T3-E1 cells treated with osterix shRNA loaded or unloaded lentivirus at 24th day after transfection (con: control, Mock: unloaded lentivirus; shRNA: loaded lentivirus).
Collagen, which constitutes a skeleton of extracellular matrix and participates in cellular vital activities, possesses strong toughness, high tensile strength, and low immunogenicity; thus, it has been considered as an ideal material in tissue engineering. As the parent phase of the scaffold, collagen interweaves into a three-dimensional porous network structure [
The PLGA microspheres we prepared take advantage of concentrating loaded BMP-2/ VEGF inside the shell, which can achieve sustainable release by dissolution degradation. These PLGA microspheres and nHA dispersed in the collagen; the hydrophilic surrounding environment and polar reaction as well as hydrogen bond among these molecules further help maintain the stability of loaded growth factors [
Researchers have reported that 3D culture can provide cells with a similar microenvironment as that in vivo and avoid contact inhibition and spontaneous senescence [
Our results show that BMP-2 and VEGF sustainably released nHAC/ PLGAs scaffold possessed excellent physical properties and biocompatibility to bear pressure from the surrounding tissues and provide space for seed cells to attach and grow. In the 3D microenvironment provided by the scaffold, VEGF alone significantly promoted MC3T3-E1 cells proliferation, BMP-2 alone significantly promoted MC3T3-E1 cells osteogenic differentiation in nHAC/ PLGAs scaffold, and osteoinductive effect of simultaneous VEGF and BMP-2 treatment further enhanced, which may be mainly through increasing nuclear translocation of osterix protein by activation of the p38 MAPK pathway.
Data in the present article has been displayed as figures and tables above.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
The present article enjoys supports from Key Laboratory for Immunology and Dermatology of Health’s Ministry, The First Hospital of China Medical University. This work was financially supported by National Natural Science Foundation of China (grant no. 51272286), Liaoning Medical Peak Construction Project (2010014), and Natural Science Foundation of Liaoning province (20102296).