Introduction
Bone defects and non-union caused by various diseases, such as trauma and tumors, are common orthopedic disorders in the clinic. Treatment is limited by the finite nature of autogenous bone materials; however, with the rise of tissue engineering, this problem may be solved using new methods. Current research (1–3) shows that bone morphogenetic protein (BMP) belongs to the TGF-β superfamily, which plays a notable role in inducing bone differentiation. Among its family members, BMP2 is one of the strongest factors and the only growth factor that is capable of singly inducing bone formation. HIF1 expression was increased in low oxygen environments, and it could be combined with the hypoxia response element of target genes, regulate gene transcription of vascular endothelial growth factor (VEGF), and play a crucial role in tissue angiogenesis (4–7). The current study will build a recombinant adenovirus vector containing a BMP2 and three mutant HIF1α (HIF1αmu) (Ad-BMP2-IRES-HIF1αmu), to be transfected into rabbit bone marrow mesenchymal stem cells (MSCs), and observe the mRNA and protein expression of the two genes in the MSCs which will provide support for gene therapy of clinical bone defects and bone non-union.
Materials and methods
Primers
Primer sequences were designed according to the sequence information of the target gene human BMP2 and the HIF1αmu gene and the enzyme digestion site information of the pIRES2-EGFP vector. NheI and BamHI sites were respectively added at the 5′ end and the 3′ end of BMP2 primer, and a 6xHIS tag was added at the C-cleavage site of the BMP2 gene. BstXI and XbaI sites were respectively added at the 5′ end and 3′ end of the HIF1αmu primer, and a 3xFlag tag was added at the C-cleavage site. The study was approved by the ethics committee of Liaoning Medical University, Liaoning, China.
Vector construction
The HIF1αmu fragment was amplified by PCR using pShuttle-CMV-HIF1αmu-IRES-hrGFP-1 plasmid as the template. The target fragment was recovered and double enzyme digested with BstXI and XbaI. The vector pIRES2-EGFP was double enzyme digested with BstXI and XbaI and recovered by agarose gel electrophoresis. The BMP2 fragment was amplified by PCR with pShuttle-CMV-BMP2 plasmid as the template. The target fragment was double enzyme digested with NheI and BamHI and recovered by agarose gel electrophoresis. The correct pIRES2-HIF11αmu-flag vector was double enzyme digested with NheI and BamHI and the larger fragment was recovered. The previously mentioned enzyme was digested and the recovered target gene was ligated with the vector fragment for 2 h at 22°C, and then transformed into competent E. coli DH5α.
Screening and identifying of positive clones
The positive clones identified by colony PCR were transferred into Kan-LB and agitated overnight at 37°C, and then the plasmid was extracted the next day. The positive clones screened by colony PCR and enzyme digestion were sequenced. The confirmed clones were stored.
Recombination of adenovirus expression vector
BMP2-IRES-HIF1αmu recombinant fragment was amplified with BMP2-pIRES2-HIF1αmu vector as the template and then recovered. The target fragment BMP2-IRES-HIF1αmu was recombined to the pDONR221 vector using the BP recombinant system of Invitrogen (Carlsbad, CA, USA). The target sequence BMP2-IRES-HIF1αmu was recombined to adenovirus vector pAd-BMP2-IRES-HIF1αmu using the LP recombinant system of Invitrogen. The recombinant plasmids were sequenced, and stored at −20°C until further use.
Titration of virus liquid
The target gene was transferred into HEK293 cells in the exponential phase and the transfer process was observed under a fluorescence microscope. An end-point dilution assay was used to determine the titer of virus liquid.
Recombinant adenovirus transfection of rabbit MSCs
Group A (experimental group): transfection with Ad-BMP2-IRES-HIF1αmu; group B (positive control group 1): transfection with Ad-HIF1αmu-IRES-hrGFP-1; group C (positive control group 2): transfection with Ad-BMP2-IRES-hrGFP-1; group D (negative control group): transfection with Ad-IRES-hrGFP-1; group E (blank group): without transfection virus.
Rabbit MSCs within three generations were digested by trypsin and mixed, then transferred into 6-well plates at a density of 5×105/well, with routine culture for 24 h in a cell culture box. The supernatant was discarded following firm adherence.
Multiplicity of infection (MOI) = (pfu/ml value) × (virus liquid volume)/(cell number for transfection). Virus liquids at MOI of 50, 100, 150 and 200 were respectively transferred into MSCs, leaving two wells as the negative controls. Following incubation for 1 h at 37°C, the cells were routinely cultured for 48–72 h with complete medium and then observed under the inversion fluorescence microscope.
Every virus liquid in each group was respectively transferred into rabbit MSCs at its best MOI value. The largest MOI not causing a marked cytopathic effect was considered as the best MOI. In this experiment, the best MOI=100. Fresh serum medium (2 ml) was respectively added to each flask of cells, and cells were cultured for 3 h in a constant temperature cell culture box, then an additional 2 ml medium was added to each flask of cells, with continued culture for 24–72 h. The transfer effect was observed under the inversion fluorescence microscope.
RT-PCR
Total RNA was extracted according to a kit that contained 2 μl PrimeScript 1 Step Enzyme Mix and 25 μl 2X 1 Step Buffer (Dye Plus). Then 2 μl upstream primer and 2 μl downstream primer, 1 μl total RNA and 18 μl RNase-free ddH2O were added to make up the 50 μl RT-PCR reaction solution. The RT-PCR reaction was as follows: a cycle of 50°C for 30 min; a cycle of 94°C for 2 min; 30 cycles of 94°C for 30 sec, 60°C for 30 sec and 72°C for 1 min. After the reaction, 5–8 μl reaction solutions underwent 1% agarose gel electrophoresis for 40 min. The electrophoresis results were observed and the OD value of strips was detected using a gel imaging system. This was repeated for three times, and the relative OD values were separately calculated.
Western blot analysis
Cellular total protein in each group was extracted using cellular protein lysis buffer, and the protein concentration in each group was detected by BCA; 5% spacer gel and 8% separation gel were prepared for polyacrylamide gel electrophoresis at 60 V for 30 min, and then 150 V for 1 h. Following electrophoresis, the separation gel was washed three times and then placed in a transfer box at 100 mA for 30 min. The film was then incubated with the primary antibodies (1:1500) overnight at 4°C. The film was then washed and then incubated with the secondary antibodies and developer solution for 30 min at room temperature, avoiding light. The film was then washed with eluent three times to terminate the chromogenic reaction. The OD value of target strips and internal reference on the film were analyzed using a gel imaging system. The experiment was repeated three times, and the relative OD values were calculated.
Statistical analysis
The experimental data were statistically analyzed using SPSS17.0 software. Measurement data were expressed as the means ± standard deviation (±s) and analyzed by one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference.
Results
Recombinant identification
A 9789 bp recombinant plasmid strip, a 1191 bp BMP2 strip and a 2481 bp HIF1α strip were observed after agarose gel electrophoresis, and the zymogram analysis of recombinant plasmid was the same as expected. The BMP2 fragment and the HIF1α fragment in the recombinant plasmid completely coincided with the BMP2 (NM001200) CDS area and the HIF1α (NM001530) CDS area in Genebank, respectively, with insert fragments in the correct direction (Figs. 1 and 2A and B).
Adv-BMP-2-IRES-HIF1α<sup>mu</sup> identification
There were two strips at 3000 and 10000 bp as shown by agarose gel electrophoresis following enzyme digestion, which indicated that the shuttle plasmid had been successfully ligated to the adenovirus genome and Adv-BMP-2-IRES-HIF1αmu had been successfully constructed. The target fragment BMP2 was detected by RT-PCR and agarose gel electrophoresis, and there was a strip at 1191 bp, which is the same as expected. The target fragment HIF1α was detected by RT-PCR and agarose gel electrophoresis, and there was a strip at 2481 bp, which was the same as expected (Fig. 2C).
Transfer and titration of virus liquid
Under inverted fluorescence microscopy, the majority of cells appeared green and cell fragments had become detached and showed a cytopathic effect. The titer of virus liquid was 3.16×108 according to the end-point dilution assay (Fig. 3A and B).
Cell culture
After the original cultivation of rabbit bone marrow stromal cells for one week, there was cell adherence to the wall when the culture solution was changed. Cells were passaged when they reached 80% confluence. The third generation had a higher purity and a good adhesion (Fig. 3C and D).
RT-PCR
There was no marked difference in the HIF1α mRNA expression level between group A and B (P>0.05), with the same result among groups C, D and E (P>0.05). The HIF1α mRNA expression level in group A and B was markedly higher than that in groups C, D and E, with a significant difference (P<0.01). There was a significant difference in BMP2 mRNA expression level between group A and C (P>0.05), but no marked difference among groups B, D and E (P>0.05). The BMP2 mRNA expression level in group A and C were markedly higher than that in groups B, D and E, with a significant difference (P<0.01) (Fig. 4A and B, Table I).
Western blot analysis
The HIF1α protein expression level in group A and B was significantly higher than that in the other three groups, with a significant difference (P<0.01). However, there was no significant difference among groups C, D and E (P>0.1). BMP2 protein expression level in group A and C was significantly higher than that in the other three groups, with a significant difference (P<0.01). However, there was no significant difference among groups B, D and E (P>0.1). There was also a significant difference in the BMP2 protein expression level between group A and C (P<0.05) (Fig. 4C and D, Table II).
Discussion
Previous studies reported that the incidence rate of delayed union after fracture and non-union was approximately 5–10%, and treatment of this remains a big problem in the field of orthopedics (8–10). In the BMP family, BMP2 is the main signaling protein regulating bone formation (11–13). BMP2 could induce differentiation of pluripotent mesenchymal progenitor cells into osteoblast cells, promoting bone formation of osteoblast cells. There were also studies confirming that BMP2 was capable of inducing pluripotent mesenchymal progenitor cells to differentiate into osteoblast cells in vitro or in vivo(14–15). During the process of bone defect and fracture healing, BMP2 promoted the osteogenetic effect at the fracture site, but following osteoblast induction, the lack of blood supply at the fracture site and difficult angiogenesis affected the repair process of the fracture.
It has been demonstrated that the activating transcription factor HIF1α found in recent years is capable of activating low oxygen metabolism and regulating the transcription of angiogenesis-promoting genes. At the same time it also participates in the regulation of VEGF and SDF21 expression (16). Therefore, HIF1α has been considered as one of the most promising genes in the clinical application of promoting angiogenesis. However, HIF1α loses its transcription activity due to its easy degradation under usual oxygen conditions (17–19). Liu et al(20) successfully constructed recombinant adenovirus-mediated mutations of hypoxia inducible factor expression vector (HIF1αmu), which could be expressed under usual oxygen conditions and rapidly accumulate in cells and promote angiogenesis. The double gene co-expression of the adenovirus vector with BMP2 and HIF1αmu constructed in this experiment was expected to form a cooperative effect due to the expression of both genes (7), accelerating the repair process of fractures and bone defects. The past use of a single gene vector had a poor treatment effect, and conjunction with many genes and vectors often increased dosages of adenovirus and virus infection efficiency (20). The advantage of two genes with a single vector constructed in this experiment was the reduction of the dosage of adenovirus required during bone defect treatment.
According to the results of RT-PCR and western blot analysis, the expression level of HIF1α mRNA was significantly increased following transfection into MSCs in group B. The expression level of BMP2 mRNA was significantly increased following transfection into MSCs in group C. The expression level of BMP2 mRNA and HIF1α mRNA were both not increased following transfection into MSCs in group D and E, demonstrating that it was not adenovirus expression vector itself or hrGFP-1 that caused the increased expression level of BMP2 and HIF1α mRNA. The expression levels of HIF1α mRNA and BMP2 mRNA were significantly increased at the same time after transfection into MSCs in group A, and the expression of BMP2 was higher than that in group C. There was no significant difference in HIF1α expression between group A and B, indicating that mutant HIF1α was likely to promote the osteogenetic function of BMP2, which would provide a new direction for angiogenic treatment of bone defect diseases. This point was consistent with the experimental results we had expected.
The angiogenesis process in the osteogenesis system is an extremely complex physical process. Importing mutations of the HIF1α gene was an abnormal way to normalize tissue, and promoted the formation of a new vascular network, but with an unknown effect on other factors themselves participating in the process of angiogenesis. The effect of mutant genes on HIF1α itself will also require further research.