| BackgroundSkeleton development and homeostasis depend on the coordinated balance between bone formation by osteoblasts and bone resorption by osteoclasts. Osteoclasts originate from monocytes and mediate bone resorption, but undifferentiated mesenchymal stem cell(MSC)-derived osteoblasts play an essential role in skeleton development and bone formation. During osteoblast differentiation, three osteoblast-specific or highly expressed transcription factors, Runt-related transcription factor 2(Runx2/Cbfa1), Osterix(Osx, also called Sp7), and activating transcription factor 4(ATF4) function sequentially to regulate the expression of osteoblastic marker genes, including alkaline phosphatase(Alp) and osteocalcin(Ocn). It has been widely accepted that Runx2 and Osx are master osteogenic factors because Runx2- or Osx-deficiency mice fails to form mature osteoblasts and bone. Runx2 expresses in prehypertrophic chondrocytes and osteoblasts and plays multiple roles in chondrogenesis and osteogenesis. Osx, which acts downstream of Runx2, is specifically expressed in osteoblast lineage cells and essential for osteoblast differentiation and bone formation in embryonic, postnatal and adult mice.Artificial transcription factors are an effective technology in endogenous gene regulation. An artificial transcription factor consists of a DNA-binding domain, an effector domain and a nuclear localization signal(NLS). The DNA-binding domain mainly functions in recognizing DNA triplets with high specificity and affinity, the effector domain activating or suppressing the expression of target gene, and the NLS delivering the artificial transcription factor into nuclei. The Cys2His2-type artificial zinc finger protein(ZFP), which is comprised of several zinc finger motifs, is the most promising candidate for DNA-binding domain. Each zinc finger motif consists of approximately 30 amino acids that code for two β-strands and one ɑ-helix, whose amino acids at position-1, 2, 3, 6 interact with 3rd, 4th, 2nd, and 1st bases of overlapping 4-bp DNA targets to confer the specificity of artificial transcription factor.To overcome the error-prone and time-consuming procedures in designing ZFPs, Zinc Finger Tools which can be accessed at URL http:www.zincfingertools.org was created. By Zinc Finger Tools, we can find appropriate target sites of our target DNA sequence for gen e regulation. Then the amino acid sequence of ZFPs binding to the target sites can be generated using the database of experimentally characterized zinc finger domains, after which ZFPs can be fused to effector domain and NLS to form functional artificial transcription factors that activate or suppress gene expression.Micro RNAs(miRNAs) are a class of small non-coding RNAs which mediate the inhibition of translation and/or mRNA degradation by binding to the 3 ’untranslated region(UTR) of their target m RNAs. miRNAs are generated from several hundred-nucleotide long primary transcripts(pri-mi RNAs) that are processed by the enzymes Drosha and Dicer to ~22-nucleotide mature mi RNAs. Emerging evidence has indicated that mi RNAs play important roles in development, homeostasis and diseases. They are involved in a wide range of biological and pathological processes, including cell differentiation, proliferation and apoptosis.Dicer deficiency in chondrocytes, osteoblasts and osteoclasts suggested essential roles of mi RNAs in skeletal development and bone homeostasis. Several miRNAs have been shown to regulate the osteogenic differentiation of mesenchymal stem cells(MSCs) and osteoblasts. For example, mi R-133 inhibits BMP2-induced osteogenic differentiation of C2C12 cells by repressing runt-related transcription factor 2(Runx2). miR-143 and mi R-145 suppresses osteogenic differentiation of MC3T3-E1 cells by targeting Osterix. mi R-214 negatively regulates osteogenesis of MC3T3-E1 cells and bone formtion by inhibiting ATF4. miR-218 promotes commitment and differentiation of MC3T3-E1 and MSCs by down-regulating Wnt signaling inhibitors SOST, DKK2 and SFRP2. mi R-146 a inhibits osteogenic differentiation by targeting JMJD3 in human mesenchymal stem cells. However, mi RNAs that regulate osteoblast differentiation and bone formation still require further investigation.Heparin-binding epidermal growth factor-like growth factor(HBEGF), which is a member of the epidermal growth factor(EGF) family, can bind to EGF receptor(EGFR) with high affinity and activate EGFR signaling, including extracellular signal-regulated kinase(ERK), JNK and phosphatidylinositol 3-kinase(PI3K)/Akt. EGFR signaling is involved in numerous physiological and pathological processes, including osteogenic differentiation and bone formation. Osteoblast proliferation and differentiation are abnormal in EGFR-deficient mice, with impaired bone formation and skeleton structure. HBEGF-EGFR signaling inhibits osteogenic differentiation of MSCs, MC3T3-E1 and C2C12 cells.Methods1. Design of artificial transcription factors targeting Osx: The sequence of Osx promoter was obtained by literature search in Pubmed; Then the sequence of Osx promoter was submitted to the Zinc Finger Tools(http:www.zincfingertools.org), a online software for design of artificial transcription factor, after which three best zinc finger proteins(ZFPs) were selected by parameters such as highest score, no TSO, high specificity and distance to transcription start site; The amino acids sequence of zinc finger protein was translated to nucleotide sequence, which was then combined with nucleotide sequences of effector domain, nuclear locolization signal and HA tag to form the complete nucleotide seque nce of artificial transcription factor.2. Construction and expression of artificial transcription factor expression vector: The complete nucleotide sequence of artificial transcription factor was optimized and synthesized by Shanghai Sangon and then cloned into vector pc DNA3.1 to form artificial transcription factor expression vector pCMV-ZFP-VP64.3. Assay of activation of Osx promoter by artificial transcription factor and binding specificity of artificial transcription factor with Osx promoter: The expr ession vector p CMV-ZFP-VP64 and reporter vector p GL3-OsxP were co-transfected into MC3T3-E1 cells, then dual-luciferase reporter assay was performed to determine whether artificial transcription factor could activate Osx promoter. The artificial transcrip tion factor with highest luciferase was selected for further study. Its binding specificity to Osx promoter was evaluated by electrophoretic mobility shift assay(EMSA) and chromatin immunoprecipitation assay(ChIP).4. The impact of artificial transcription factor on osteogenic differentiation of MC3T3-E1 cells: After transfection of expression vector p CMV-ZFP-VP64 to MC3T3-E1 cells, cells were subjected to osteogenic induction for different days. Then real-time PCR for Alp and Ocn mRNA levels, Alp staining for Alp activity and Alizarin Red staining for calcium deposition were performed.5. The impact of artificial transcription factor on local bone mass in OVX mice: The nucleotide sequence of artificial transcription factor was cloned into lentivirus vect or p LOV.Ubic.EGFP to form vector p Ubic-ZFP-VP64. After transfection of lentivirus vector p Ubic-ZFP-VP64 to mesenchymal stem cells, Western blot was performed to determine whether ZFP-VP64 could be successfully expressed. 48 h after transfection, MC3T3-E1 cells were subjected to osteogenic induction, then real-time PCR for Alp and Ocn m RNA levels as well as Alp staining for Alp activity were performed to determine the effect of ZFP-VP64 on osteogenic differentiation of mesenchymal stem cells. Ovariectomized mice were used for evaluation of impact of ZFP-VP64 on local femur. 8 weeks and 16 weeks after ovariectomy(OVX), microCT scanning was performed to assess bone loss in mice. 8 weeks after OVX, the mesenchymal stem cells transfecting with p Ubic-ZFP-VP64 were injected into bone marrow cavity of left femur, then micro CT scanning was performed 8 weeks after injection to determine the impact of ZFP-VP64 on local bone of OVX mice. After micro CT scanning, the left femur was subject to immunohistochemistry(IHC) t o determine the expression of ZFP-VP64 and Osx at the surface of trabecular bone.6. Prediction of microRNAs(mi RNAs) targeting Osx: Bioinformatical analysis using Target Scan, mi Randa, and PicTar was performed to explore mouse Osx-targeting mi RNAs.7. The correlation between predicted miRNAs and bone mineral density(BMD) and osteogenic differentiation: The predicted mi RNAs level in left fumer of OVX mice was compared with that in left femur of Sham-operated mice to determine whether there was correlation between miRNAs and BMD. The mi RNAs which are correlated with BMD were selected for further study. Real-time PCR for Alp, Ocn and miRNAs levels during osteogenic differentiation of MC3T3-E1 and BMSCs were performed to determine whether there was correlation between selected mi RNAs and osteogenic differentiation.8. The impact of selected mi RNAs on osteogenic differentiation of MC3T3-E1 and mesenchymal stem cells: After transfection of agomir or antagomir of selected mi RNA s into MC3T3-E1 or mesenchymal stem cells, cells were subject to osteogenic differentiation. The impact of selected mi RNAs on osteogenic differentiation of MC3T3-E1 and mesenchymal stem cells were evaluated by changes of Alp and Ocn mRNA levels and Alp activities in MC3T3-E1 and BMSCs during osteogenic differentiation.9. Study on molecular mechanisms under the impact of selected mi RNAs on osteogenic differentiation of MC3T3-E1 cells: 48 h after cotransfection of mi RNAs with Osx 3’UTR into MC3T3-E1 cells, dual-luciferase reporter assay was performed to determine whether Osx 3’UTR is the target of selected mi RNAs. If yes, selected mi RNAs mediated osteogenic differentiation by downregulation of Osx; If not, other target would be predicted by Target Scan, miRanda and Pic Tar to identify the real mechanism under which the selected mi RNAs mediated osteoblast differentiation.Results1. Design of artificial transcription factor targeting Osx: The Osx promoter was submitted to online artificial transcription factor-designing software Zinc Finger Tools. According to following parameters: high score, no TSO, high specificity, and different distance to transcription start site, three target sites of Osx promoter located in-94/-77,-387/-370, and-841/-824 was selected, which could be bound by zinc finger proteins named ZFP1, ZFP2, and ZFP3, respectively. Then nucleotide suquence of zinc finger proteins would be combined with nucleotide sequences of effector domain VP64, nuclear localization signal, and HA tag to form the nucleotide sequence of artificial transcription factor ZFP1-VP64, ZFP2-VP64, and ZFP3-VP64, which was cloned into pcDNA3.1 to form artificial transcription factor expression factor p CMV-ZFP-VP64.2. Assay of activation of Osx promoter by artificial transcription factor and binding specificity of artificial transcription factor with Osx promoter: After cotransfection of artificial transcription factor expressing vector pCMV-ZFP1-VP64, pCMV-ZFP2-VP64, or p CMV-ZFP3-VP64 with Osx promoter reporter vector pGL3-OsxP into MC3T3-E1 cells, dual-luciferase reporter assay was performed. Results showed that all the three transcription factors ZFP1-VP64, ZFP2-VP64, and ZFP3-VP64 could increase the luciferase activity, among which ZFP2-VP64 was the most potent one with an increase of luciferase activit y to 6.9-fold compared with the empty vector. Meanwhile, ZFP2-VP64 increased the Osx promoter activity in a dose-dependent manner, indicating the specific activation of Osx promoter by ZFP2-VP64. Further, EMSA results showed that a shift protein-DNA complex band was detected with labeled probe(containing-387/-370 site) but not with mutated probe. Moreover, the specificity of the interaction was confirmed by competition of the shift protein-DNA complex band by cold probe in a dose-dependent manner. To further confirme the specific binding of ZFP2-VP64 and Osx promoter, ChIP was also performed. We designed a set of primers encompassing-465/-270 site in the Osx promoter, a region that included ZFP2-VP64 binding site and performed PCR amplification. Results showed a band of expected size(196 bp) when anti-HA tag antibody was used in immunoprecipitation, whereas no band was detected using control Ig G, suggesting direct binding of ZFP2-VP64 with-465/-270 site of Osx promoter.3. ZFP2-VP64 upregulated Osx endogenous expression and promoted osteogenic differentiation of MC3T3-E1 cells: After transfection of different dose of p CMV-ZFP2-VP64 vector into MC3T3-E1 cells, we observed an increase of Osx mRNA in a dose-dependent fashion, with the maximum of 6.8 fold by 1.2 μg p CMV-ZFP2-VP64. Western blot also showed that p CMV-ZFP2-VP64 upregulated Osx expression at protein level in a dose-dependent manner. All these data indicated that the Osx-promoter-specific ZFP2-VP64 could activate the endogenous Osx expression, as we expected. Further, we determined whether ZFP2-VP64-induced Osx upregulation could promote osteoblast differentiation. Western blot assay 72 h after transfection indicated that ZFP2-VP64 did not cause a significant change for the osteogenic transcription factor Runx2, which acts upstream of Osx. However, the transcription factor ATF4 acting downstream of Osx was increased by ZFP2-VP64. Alp and Ocn mRNA levels were upregulated by introduction of p CMV-ZFP2-VP64 into MC3T3-E1 cells after 3- and 7-day induction with osteogenic differentiation medium. Consistent with changes in Alp and Ocn mRNA levels, we found that the presence of ZFP2-VP64 enhanced Alp staining after 7-day and Alizarin Red staining after 21-day induction of MC3T3-E1 cells with differentiation medium. Thus, ZFP2-VP64-induced Osx upregulation could promote osteoblast differentiation.4. ZFP2-VP64 locally reduced ovariectomy(OVX)-induced bone loss in mice: ZFP2-VP64 was stably overexpressed in p Ubic-ZFP2-VP64-overexpressing BMSCs. Introduction of ZFP2-VP64 into BMSCs resulted in dramatically increased Alp and Ocn m RNA levels at the late stage of osteogenic induction. Moreover, ZFP2-VP64 markedly enhanced Alp activity at day 7 and day 21. These results indicate d that ZFP2-VP64 was sufficient to promote osteogenic differentiation of BMSCs in vitro. Mice with only OVX exhibited significant decreases in bone mineral density(BMD), trabecular bone volume(BV/TV), trabecular thickness(Tb.Th), trabecular number(Tb.N), cortical bone thickness(Ct.Th), cortical area(Ct.Ar), and an increase in trabecular separation(Tb.Sp) compared with sham group, indicating successful induction of osteopenia in mice at 8 weeks and 16 weeks after OVX. 8 weeks after transplantation of ZFP2-VP64-overexpressing BMSCs into left femur marrow cavity of OVX mice, trabecular but not cortical bone mass was increased compared with control groups, indicating that ZFP2-VP64 could locally reduce bone loss in OVX mice, at least in part. 8 weeks after transplantation, IHC analysis showed that there were more HA tag- and Osx-positive cells along the trabecular bone surface in ovariectomized mice transplanted with pUbic-ZFP2-VP64-overexpressing BMSCs than in ovariectomized mice transplanted with pUbic-empty-overexpressing BMSCs, suggesting osteoblast differentiation of transplanted pUbic-ZFP2-VP64-overexpressing BMSCs in OVX mice.5. Prediction of micro RNAs(miRNAs) targeting Osx: By TargetScan, mi Randa, and Pic Tar, mi R-96 and miR-383 were predicted to target Osx 3’UTR. mi R-96 was downregulated in femur of OVX mice compared with that of wildtype mice, whereas there was not significant difference for miR-383 level between femur of OVX mice and that of wildtype mice, which indicated that there was correlation between miR-96 but not mi R-383 and bone marrow density. Thus, mi R-96 was selected for further study.6. mi R-96 promoted osteogenic differentiation of MC3T3-E1 and mesenchymal stem cells: The expression levels of mi R-96 in MC3T3-E1 cells and primary osteoblasts were significantly higher than that in mesenchymal stem cells. Moreover, the mi R-96 level was gradually increased during osteogenic differentiation of MC3T3-E1 cells and mesenchymal stem cells. These results suggested positive correlation between miR-96 and osteogenic differentiation. After transfection of agomir-96 into MC3T3-E1 and mesenchymal stem cells, Alp and Ocn m RNA levels as well as Alp activity were increased, while introduction of antagomir-96 into these cells resulted in decreases of Alp and Ocn mRNA levels as well as Alp activity. These results indicated that mi R-96 could promote osteogenic differentiation of MC3T3-E1 cells and mesenchymal stem cells.7. mi R-96 suppressed HBEGF-EGFR signaling by targeting HBEGF in MC3T3-E1 cells: Dual-luciferase reporter assay confirmed that Osx 3’UTR was not a target of mi R-96. We further predicted another target HBEGF to be a target of mi R-96 by bioinformatic softwares Target Scan, miRanda, and PicTar and confirmed that HBEGF was a target of mi R-96 by dual-luciferase reporter asaay. mi R-96 inhibited HBEGF protein but not mRNA level, indicating that mi R-96 could mediate the inhibition of translation but not mRNA degradation of HBEGF by binding to HBEGF 3’UTR. Further, overexpression of mi R-96 suppressed the phosphorylation but not total levels of EGFR, ERK, and Akt. Conversely, the EGFR signaling was activated by inhibition of mi R-96. All these results suggested that EGFR signaling was inhibited by miR-96.8. HBEGF inhibited mi R-96-induced osteogenic differentiation in MC3T3-E1 cells: Knockdown of HBEGF by siRNA promoted mi R-96-induced osteogenic differentiation reflected by Alp and Ocn mRNA levels as well as Alp activity in MC3T3-E1 cells, while overexpression of HBEGF by HBEGF expression vector inhibited miR-96-induced osteogenic differentiation in the cells. All these results further confirmed that mi R-96 mediated the induction of osteogenic differentiation by targeting HBEGF.Conclusions1. In this study, we successfully designed an artificial transcription factor ZFP2-VP64 targeting Osx promoter by a online software Zinc Finger Tools. ZFP2-VP64 could specifically bind to-387/-370 site of Osx promoter and significantly activated the transcription of Osx gene. ZFP2-VP64-induced Osx upregulation could promote osteogenic differentiation of osteoblasts and mesenchymal stem cells and locally reduced ovariectomy-induced bone loss in mice. ZFP2-VP64 may have the potential to be developed to an anti-osteoporosis drug and deserves further study.2. In the present study, miR-96 was predicted to be an important microRNA which might play great roles in osteogenesis by targeting Osx 3’UTR. However, HBEGF, but not Osx, was a target of mi R-96. Further, mi R-96 could promote osteoblast differentiation by suppressing HBEGF-EGFR signaling. |
PDF Full Text Request