Mesenchymal stem cell sheet transplantation combined with locally released simvastatin enhances bone formation in a rat tibia osteotomy model

Cytotherapy, 2013; 15: 44e56

Mesenchymal stem cell sheet transplantation combined with locally released simvastatin enhances bone formation in a rat tibia osteotomy model

YIYING QI1,*, TENGFEI ZHAO2,*, WEIQI YAN1, KAN XU1, ZHONGLI SHI1 & JIANWEI WANG1 1

Department of Orthopedic Surgery, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China, and 2Hangzhou Binjiang Hospital, The Second Affiliated Hospital (Binjiang Branch), School of Medicine, Zhejiang University, Hangzhou, China

Abstract Nonunion of fractured bones is a common clinical problem for orthopedic surgeons. This study aimed to investigate the effects of simvastatin locally applied from calcium sulfate (CS) combined with a mesenchymal stem cell (MSC) sheet on fracture healing. In vitro, the proliferation and differentiation of rat bone marrowederived MSCs stimulated by simvastatin were investigated. In vivo, an osteotomy model was made in rat tibia, and fractured tibias were treated with CS, CS/ simvastatin, CS/MSC sheet or simvastatin-loaded CS with MSC or untreated (control). Tibias were harvested at 2 or 8 weeks and underwent real-time quantitative polymerase chain reaction, x-ray, micro-CT and histological analysis. The expression levels of bone morphogenetic protein 2, alkaline phosphatase, osteocalcin, osteoprotegerin and vascular endothelial growth factor of simvastatin-induced MSCs increased with the concentrations of the simvastatin, significantly higher than those in the MSCs group. At 2 weeks, the CS/simvastatin/MSC sheet group showed significantly higher expressions of bone morphogenetic protein 2, alkaline phosphatase, osteocalcin, osteoprotegerin and vascular endothelial growth factor, with more callus formation around the fracture site compared with the other four groups. At 8 weeks, complete bone union was obtained in the CS/simvastatin/MSC sheet group. By contrast, newly regenerated bone tissue partially bridged the gap in the CS/simvastatin group and the CS/MSC sheet group; the control and CS group showed nonunion of the tibia. These results show that both simvastatin and the MSC sheet contributed to the formation of new bone and that the tibia fracture was completely healed by transplantation of the MSC sheet with locally applied simvastatin. Such MSC sheet with locally applied simvastatin might contribute to the treatment of fractures, bone delayed unions or nonunions in clinical practice. Key Words: fracture, mesenchymal stem cells, MSC sheet, nonunion, simvastatin

Introduction Delay and failure of bone union are common clinical problems for orthopedic surgeons. Approximately 5e10% of fractures may result in delayed union or nonunion (1,2). For many years, autogenous bone grafts or free vascularized bone grafts have been widely used for nonunion treatment (3,4). However, harvesting of the grafts from the iliac crest is associated with donor site morbidity and particularly associated with chronic pain. It has been shown that the process of fracture healing could be enhanced by the local application of bone morphogenetic protein 2 (BMP-2), which has the potential to act as an autogenous bone graft substitute (5,6). Simvastatin, a cholesterol-lowering

drug, can stimulate new bone formation by induction of BMP-2 (7). In addition, a previous study (8) has confirmed that simvastatin is a potential drug in the treatment of fracture healing. Local application of simvastatin appears to be reasonable because simvastatin is poorly distributed to bone, and <5% of an oral dose reaches the systemic circulation (9). Moreover, local application helps to prevent systemic side effects of drugs. Among various bone substitutes, calcium sulfate (CS) has excellent biocompatibility and osteoconductivity, and its potential as an effective carrier for local release of antibiotics and growth factors has been well demonstrated (10e14). We believe that CS can be an effective carrier for local delivery of simvastatin.

*These authors contributed equally to this work. Correspondence: Jianwei Wang, MD, Department of Orthopedic Surgery, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China. E-mail: [email protected]; or Weiqi Yan, Department of Orthopedic Surgery, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China. E-mail: [email protected] (Received 15 April 2012; accepted 23 July 2012) ISSN 1465-3249 Copyright Ó 2013, International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcyt.2012.10.006

MSC sheet transplantation with simvastatin for bone repair Mesenchymal stem cells (MSCs) are multipotent cells, and MSCs combined with different scaffolds were shown to improve bone regeneration (15e17). However, the number of implanted MSCs was limited because of the low surface-to-volume ratio of the scaffolds (18). The cell-sheet technique can provide highly efficient cell delivery (19,20). In addition, the cell sheet can be easily detached from the culture substrate, and the adhesion molecules on the cell surface and cell-cell interactions remain intact (21). A previous study (22) showed that such an MSC sheet could form bone tissue after transplantation into a subcutaneous site. Therefore, the present study aimed to investigate the effects of locally applied simvastatin from a CS scaffold with an MSC sheet on bone formation in a rat tibia osteotomy model. This study was designed to probe into the following contents: (i) the local release of simvastatin from CS scaffolds in vitro; (ii) the effects of simvastatin on the proliferation, osteogenic and angiogenic differentiation of MSCs in vitro; and (iii) the effects and possible mechanism of locally applied simvastatin from a CS scaffold combined with the MSC sheet on bone formation in a rat tibia osteotomy model. Methods Fabrication of simvastatin-releasing CS scaffolds Osteoset (Wright Medical, Arlington, TN, USA), a medical-grade CS powder, was used in this study. Simvastatin powder (Hisun Pharmaceutical Co, Ltd, Taizhou, China) was dissolved in 75% ethanol at concentrations of 200 mg/mL. For the preparation of simvastatin-releasing CS scaffolds, 20 mg of CS powder, 2.5 mL of simvastatin solution and 5 mL distilled water were mixed and stirred in a dish for 15 seconds. The mixture was then transferred to a circular mold of 2-mm diameter and 1.0-mm thickness to make cylinders for implantation. Finally, 0.5 mg simvastatin was applied in each scaffold.

45

Single CS scaffolds served as controls. In addition, a putative marrow cavity was drilled along the axis of the CS constructs for the insertion of needles during the implantation (Figure 1A). All samples were sterilized by g radiation (>25 kGy). In vitro simvastatin release from simvastatin-releasing CS scaffolds assay The simvastatin-releasing CS scaffolds were placed in 5 mL of phosphate-buffered saline (PBS) at 37
C, and the PBS was exchanged at 1, 2, 4, 6, 8, 11, 14 and 21 days, respectively. At every time point, the solution absorbance was measured at a wavelength of 238 nm by UV-visible spectrophotometer; the simvastatin concentration was determined from the standard curve prepared with various amounts of simvastatin. MSC culture and cell sheet preparation Rats were euthanized with CO2, and the femurs and tibias were then removed. The bones were washed in a-minimal essential medium (a-MEM) supplemented with 10% (v/v) penicillin/streptomycin (Gibco, Life Technologies, Grand Island, New York, USA). Both ends of the femurs and tibias were cut away from the epiphysis, and the bone marrow was flushed out of the bone with 10 mL of medium in a syringe. The cells were filtered through a 70-mm cell strainer and centrifuged at 300g for 5 min. The cell pellet was resuspended in 10 mL of a-MEM supplemented with 10% fetal bovine serum (FBS, Gibco) and plated in a culture plate. Cells were maintained at 37
C in a humidified atmosphere with 5% CO2, and the medium was changed every 2 days. When adherent cells reached 80e90% confluence, they were detached with 0.25% trypsin-EDTA (Gibco) and replated at 1:3 in regular growth medium to allow for continued passaging. MSCs of passage 3 with the same cell line were used in all tests. Cells performed a maximum of six doublings during expansion and culturing procedures.

Figure 1. (A) A putative marrow cavity was drilled along the axis of the CS constructs. (B) Cultured MSCs were lifted as a cell sheet with the use of a scraper. (C) The transverse osteotomy was made in the proximal one third of the tibia, and the periosteum at the fracture site was removed. (D) A cell sheet combined with simvastatin-loaded CS was transplanted onto the osteotomy site.

46

Y. Qi et al.

To create the cell sheet, the released cells (passage 3) were seeded at 4  104 cells/cm2 onto flasks cultured in a-MEM at 37
C in a humidified atmosphere with 5% CO2 for approximately 1 week, following the procedures of previous studies (14,15). The MSC sheet performed a maximum of 17 doublings during expansion and culturing procedures. The cells were rinsed with PBS (Gibco) twice, and the cells were then lifted as a cell sheet with the use of a scraper (Figure 1B). Proliferation of MSCs stimulated by simvastatin in vitro Simvastatin was dissolved in 75% ethanol at concentrations of 1 M and diluted into different concentrations with PBS for in vitro study. The MSCs were seeded in 96-well plates (4  103 cells/well) with different concentrations of simvastatin (0.01 mmol/L, 0.1 mmol/L, 0.5 mmol/L, 1 mmol/L, 2 mmol/L, 5 mmol/L and 10 mmol/L) for proliferation assay. After culturing for 1, 3 and 7 days, the 3-(4,5-dimehyl2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay was performed. This assay is based on the ability of mitochondrial dehydrogenases to oxidize thiazolyl blue (MTT), a tetrazolium salt 3-(4, 5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-terazolium bromide, to an insoluble blue formazan product. The optical density of the plates was then read on a microplate reader (Bio-Rad Model 550, Hercules, CA, USA), with the use of test and reference wavelengths of 570 nm. This test was repeated three times. The growth curves of MSCs cultured in different concentrations were drawn. Each study was repeated in triplicate.

simvastatin (0.01 mmol/L, 0.1 mmol/L and 1 mmol/L). After culturing for 3 and 7 days, the differentiation of MSCs was analyzed. Real-time quantitative polymerase chain reaction assays for cultured MSCs, in vitro fabricated MSC sheet and simvastatin-induced MSCs Total cellular RNA was extracted from cultured MSCs, MSC sheet and simvastatin-induced MSCs with the use of Trizol Reagent (Invitrogen, Carlsbad, CA). Complementary DNAs (cDNAs) were synthesized with the use of a commercial kit (SuperScript II Reverse Transcriptase, Invitrogen) according to the manufacturer’s recommendations. Each group involved four wells. Real-time polymerase chain reaction (PCR) quantification was performed with the SYBR Premix Ex Taq Kit (TaKaRa) on an iQTM5 multiplex real-time fluorescence quantitative PCR instrument (Bio-Rad). The gene expression levels of BMP-2, alkaline phosphatase (ALP), osteocalcin (OCN), osteoprotegerin (OPG) and vascular endothelial growth factor (VEGF) were determined, following the procedures described previously (21). Primers for target genes and the internal control gene are listed in Table I. 18S ribosomal RNA was used as an internal control to adjust the differences between samples. Thermal cycle conditions were 1 min at 95
C for activation of the Universal mixture AmpliTaq Gold Polymerase, followed by 45 cycles of 10 seconds at 95
C for denaturing and 25 seconds at 62
C for annealing and extension.

Animal models Differentiation of MSCs stimulated by simvastatin in vitro MSCs were seeded in 6-well plates (4  104 cells/ well) in a-MEM supplemented with 1% (v/v) penicillin/streptomycin and different concentrations of

An experimental rat tibia osteotomy model was made according to a previous study (21). The rats were anesthetized with an intraperitoneal injection of 5% ketamine hydrochloride (2 mL/kg body weight). A lateral incision (1.5 cm) was made on the proximal

Table I. Nucleotide primers used for real-time quantitative polymerase chain reaction. Genes 18S ribosomal RNA BMP-2 Osteocalcin VEGF OPG ALP

Oligonucleotide sequence (50 -30 )

Product size

Forward: GAATTCCCAGTAAGTGCGGGTCATA Reverse: CGAGGGCCTCACTAAACCATC Forward: ACAGCTGGTCTCAGGTAAGACCAC Reverse: CCATGGCAGTAAAAGGCATGATAGCCT Forward: CTCACTCTGCTGGCCCTGAC Reverse: CACCTTACTGCCCTCCTGCTTG Forward: GTCACCACCACACCACCATCGT Reverse: CTCCTCTCCCTTCATGTCAGGCT Forward: CCGAATTGGCTGAGTGTTCTGGT Reverse: CTTGCGAGCTGTGTCTCCGTTT Forward: ACCTGACTGACCCTTCCCTCT Reverse: CAATCCTGCCTCCTTCCACTAG

105bp 219bp 111bp 76bp 94bp 102bp

MSC sheet transplantation with simvastatin for bone repair site of the tibia, and the muscle was divided longitudinally to expose the tibia. The periosteum was removed as much as possible from the proximal to the distal site of the tibia. A transverse osteotomy from front to back was then made in the proximal one third of the tibia with the use of an oscillating mini saw; the gap was 1 mm, which was equal to the height of CS scaffolds (Figure 1C). A small incision was then made on the medial aspect of the knee, and the patella was deflected laterally to expose the tibiae nod, where a small hole was drilled. A 21-gauge needle was then inserted through the hole to the distal end of the tibia, resulting in loose fixation. Eighty osteotomy sites of 40 rats were implanted with CS alone (CS group, n ¼ 8) or simvastatinreleasing CS (CS/simvastatin group, n ¼ 8) or CS with an MSC sheet (CS/MSC sheet group, n ¼ 8) or simvastatin-releasing CS with an MSC sheet (CS/simvastatin/MSC sheet group, n ¼ 8) (Figure 1D). Untreated osteotomy sites served as controls (n ¼ 8). Unprotected weight bearing was allowed immediately after the operation. Real-time PCR analysis Three tibia specimens were used for each group to determine gene expressions of BMP-2, ALP, OCN, OPG and VEGF at 2 weeks after surgery, following the procedures described in previous studies (23,24). Three bone samples incorporating the individual fracture region were obtained from tibias of each rat. Primers for target genes and the internal control gene are listed in Table I. The extraction of total RNA from bone samples, synthesis of cDNA and real-time PCR quantification were consistent with the method described above. Radiographic examination On the day before killing, the rats were generally anesthetized (ketamine hydrochloride, same dosage as above) to obtain posterior-anterior radiographs of the fractured tibia. Radiographs were taken with a dual track molybdenum/rhodiumþ Mo target mammography machine (22 KV, 250 mAS, GE, Fairfield, CT, USA) at 2 and 8 weeks after surgery to evaluate callus formation and bridging bone formation at the fracture sites. Radiographs of each animal were assessed under a triple-blinded protocol by means of a scoring scale that was based on rebridgement of the cortices and acceleration of healing (Table II). Gray values of the fracture sites and healthy tibias were analyzed with medical image analyzing software (Image J 1.43u, NIH, Bethesda, MD, USA), and the scales were set between 0 and 255. In addition, the therapeutic results after 8 weeks were determined through 3D

47

reconstruction of microcomputed tomography (micro-CT) images (mCT 40, Scanco Medical, Bassersdorf, Switzerland). Histological observations Specimens were fixed in 10% paraformaldehyde, were decalcified with pH 7.4, 10% EDTA solution for 4 weeks and embedded in paraffin. Sagittal plane sections (7 mm) from each sample were prepared, stained with hematoxylin and eosin (H&E) and Safranin-O and examined under a light microscope (Olympus, Tokyo, Japan). Statistical analysis Results are expressed as means and standard deviations and were examined by one-way analysis of variance. Data analyses were performed with the use of SPSS software (version 15.0; SPSS, Inc, Chicago, IL, USA). A Turkey test was used for multiple comparisons, and the level of significance was set at P < 0.05. Results In vitro release behavior of simvastatin from simvastatinreleasing CS The absorption and calibration curves are shown in Figure 2A,B. The in vitro release pattern of simvastatin from simvastatin-releasing CS is shown in Figure 2C. On day 1, approximately 9.5% of the simvastatin was released and stable release was maintained. By day 21, approximately 95% of the loaded simvastatin was released. Effects of different concentrations of simvastatin on cell proliferation The effects of simvastatin at different concentrations on the proliferation of MSCs are shown in Figure 3A. Table II. Radiographic scoring scale based on rebridgement of the cortices and acceleration of healing. 0 1 2 3 4 5 6 7

No bridging, no callus formation No bridging, initiation of a small amount callus No bridging, obvious initial callus formation near fracture No bridging, marked callus formation near and around fracture site Rebridging of at least one of the cortices, marked callus formation near and around fracture site Rebridging of at least one of the cortices, marked and complete callus formation around fracture site Rebridging of both cortices, and/or some resolution of the callus Clear rebridging of both cortices and resolution of the callus

48

Y. Qi et al. 0.01 mmol/L, 0.1 mmol/L and 1 mmol/L could obviously upregulate the expression of BMP-2, ALP, OCN, OPG and VEGF of MSCs (Figure 3C,D). Moreover, the expression levels of BMP-2, ALP, OCN, OPG and VEGF increased with the concentrations of the simvastatin from 0.01 mmol/L, 0.1 mmol/L to 1 mmol/L, significantly higher than those in the MSCs group. However, the upregulation of BMP-2, ALP, OCN, OPG and VEGF genes was not obvious from 3 days to 7 days in all concentrations of simvastatin. Gene expression profiles in vivo Gene expression experiments were conducted to investigate whether recruitment and augmented fracture healing could be linked to BMP-2, ALP, OCN, OPG and VEGF. The CS/simvastatin/MSC sheet group showed significantly higher expression of BMP-2, ALP, OCN, OPG and VEGF at 2 weeks, compared with the other four groups (Figure 4A,B). The CS/simvastatin group showed levels of BMP-2, ALP, OPG and VEGF expression similar to that in the CS/MSC sheet group, significantly higher than those of CS group and control group. Radiographic analysis

Figure 2. (A, B) Absorption and calibration curves of simvastatin, respectively. (C) The in vitro release pattern of simvastatin from simvastatin-releasing CS.

MTT assays showed that simvastatin >1 mmol/L began to slow down the cell growth at 7 days. Compared with the control group, the cell proliferation was somewhat inhibited when the concentration of simvastatin in media was >1 mmol/L. Moreover, lower concentration of simvastatin (0.01 mmol/L, 0.1 mmol/L, 0.5 mmol/L) can promote the proliferation of MSCs.

Gene expressions of cultured MSCs, MSC sheet and simvastatin-induced MSCs The expression levels of BMP-2, ALP, OCN, OPG and VEGF of the MSC sheet were significantly higher than those of cultured MSCs (Figure 3B). The present data indicate that cell sheets have the potential for angiogenesis as well as osteogenesis. Simvastatin at

All rats ate a normal diet and behaved appropriately after surgery, and all survived until the scheduled date of euthanasia with no apparent complications, such as obvious infection or skin necrosis. Moreover, no serious side effects of simvastatin, such as toxicity, were seen in the present study. Radiographs taken at 2 weeks showed more callus formation around the fracture site in the CS/ simvastatin group, the CS/MSC sheet group and the CS/simvastatin/MSC sheet group than that in the control and CS group (Figure 5AeE). The cortical gap was present in all groups; however, the gap was shortest and obscure in the CS/simvastatin/MSC sheet group (Figure 5E). The amount of bridging bone formation increased with time in all groups. At 8 weeks after surgery, the control group and the CS group showed less callus formation at the fracture site and the cortical gap was still present, resulting in established nonunion (Figure 5F,G). In the CS/ simvastatin group and the CS/MSC sheet group, more bridging callus formation at the fracture site and the gap was obscure (Figure 5H,I). In the CS/ simvastatin/MSC sheet group, the cortical gap disappeared, indicating bone union, and presented enhanced consolidation of the fracture site (Figure 5J). The results of radiological scores and gray values are shown in Figure 5KeN. The CS/ simvastatin/MSC sheet group showed the highest

MSC sheet transplantation with simvastatin for bone repair

49

Figure 3. (A) The effect of simvastatin at different concentrations on the proliferation of MSCs. (B) The gene expression levels of BMP-2, ALP, OCN, OPG and VEGF between the MSC sheet and MSCs; the MSC sheet showed higher expression of BMP-2, ALP, OCN, OPG and VEGF than did MSCs. (C, D) Gene expression of BMP-2, ALP, OCN, OPG and VEGF of MSCs treated with 0.01 mmol/L, 0.1 mmol/L, 1 mmol/L simvastatin and control (without simvastatin) for 3 and 7 days, respectively. MSCs treated with simvastatin showed significantly higher expression of BMP-2, ALP, OCN, OPG and VEGF than those of MSCs alone; with the increase of simvastatin, the expression of BMP-2, ALP, OCN, OPG and VEGF of MSCs treated with simvastatin was upregulated. *P < 0.05.

scores and gray values at both 2 weeks and 8 weeks. However, at 2 weeks, the gray values of the CS/simvastatin/MSC sheet group were significantly less than

those of normal tibia. At 8 weeks, the gray values of the CS/simvastatin/MSC sheet group were close to those of normal tibia. The radiological scores and gray values in the CS/MSC sheet group and the CS/simvastatin group were significantly higher than those of the CS group and the control group. No significant differences in radiological scores and gray values were found between the CS/MSC sheet group and the CS/ simvastatin group, the control group and the CS group. At 8 weeks after implantation, the micro-CT image demonstrated that the fractures treated with CS or that were untreated had obvious fracture gaps (Figure 6A,B), and the gap in control group was much wider than that in the CS group (Figure 6F,G). Fractures treated with CS/simvastatin or with the CS/MSC sheet had significantly reduced cortical fracture gaps, and newly regenerated bone tissue partially bridged the gap (Figure 6C,D,H,I). The CS/simvastatin/MSC sheet group showed complete bone union; the gap disappeared, and cortical bone connected together (Figure 6E,J). Histological analysis

Figure 4. Gene expression of BMP-2, ALP, OCN, OPG and VEGF of fractured sites in different groups at 2 weeks after surgery. Sim, simvastatin. *Compared with control and CS group P < 0.05; #compared with other four groups P < 0.05.

At 2 weeks after implantation, the cortical fracture gaps in all groups are still obvious, especially in the CS group and the control group. The CS/simvastatin group, the CS/MSC sheet group and the CS/

50

Y. Qi et al.

Figure 5. Radiographic analysis, radiographic scores and gray values of fracture sites at 2 and 8 weeks after surgery in each group (A, F: control group; B, G: CS group; C, H: CS/simvastatin group; D, I: CS/MSC sheet group; E, J: CS/simvastatin/MSC sheet group; K, L: radiographic scores of each group at 2 and 8 weeks after surgery, respectively; M, N: Gray values of each group at 2 and 8 weeks after surgery, respectively; Sim: simvastatin). *Compared with control and CS group P < 0.05; #compared with other four groups P < 0.05.

simvastatin/MSC sheet group showed newly formed bone tissue around the fracture sites. A thick callus consisting of newly formed woven bone tissue was observed in the CS/simvastatin/MSC sheet group (Figure 7). At 8 weeks, there was no bridging bone formation at the fracture site in the CS group and the control group. The gaps were filled with chondrocytes and fibrous tissue, resulting in the nonunion of the tibia (Figure 8). In the CS/simvastatin group and the CS/MSC sheet group, the calluses on both distal and proximal sites were partially united. However, in the un-united gaps, the chondrogenic

areas and fibrous tissue still existed. In the CS/ simvastatin/MSC sheet group, the calluses on both distal and proximal sites were completely united, and the gap and the chondrogenic areas had disappeared, indicating bony union of the fracture. Discussion The present study demonstrated that locally applied simvastatin combined with an MSC sheet had a synergistic effect on healing of the fractured bones. The present findings suggest a beneficial effect of

MSC sheet transplantation with simvastatin for bone repair

51

Figure 6. Micro-CT images of the fracture sites in each group at 8 weeks after surgery (A, F: control group; B, G: CS group; C, H: CS/simvastatin group; D, I: CS/MSC sheet group; E, J: CS/simvastatin/MSC sheet group).

simvastatin, locally applied from CS during fracture healing, which was similar to the effect of the MSC sheet. In an in vitro study, the MTT test showed that simvastatin <1 mmol/L (including 1 mmol/L) had no obvious inhibition on the proliferation of rat MSCs. However, simvastatin >1 mmol/L began to slow down the MSC growth and obviously inhibited the MSC growth with increased concentrations of simvastatin. A previous study (25) suggested that the appropriate concentration of simvastatin was 0.5e1 mmol/L for in vitro culture and that MSCs could not proliferate in a medium containing >2.5 mmol/L of simvastatin. Kupcsik et al. (26) and Zhou et al. (27) also found that simvastatin >1 mmol/L had an inhibitory or cytotoxic effect on the growth of MSCs. On the basis of a cell proliferation test, the osteogenic differentiation of rat MSCs co-cultured with 0.01e1 mmol/L simvastatin was determined. When MSCs were treated with 0.01 mmol/L, 0.1 mmol/L and 1 mmol/L simvastatin, the expressions of BMP-2, ALP, OCN, OPG and VEGF increased. Similar results of simvastatin on osteoblastic differentiation of other cells were found in many previous experiments. For human bone marrowederived MSCs, the osteogenic differentiation treated with simvastatin was significantly enhanced from 0.01 mmol/L to 1

mmol/L (28). For MC3T3-E1 cells, significant osteogenic-inducing effects were observed at a concentration of 0.1 mmol/L and 0.01 mmol/L simvastatin, and the effect of 0.1 mmol/L simvastatin was more obvious than 0.01 mmol/L simvastatin (29). When human osteoblasts and the MG-63 cell line were treated with different concentrations of simvastatin (0.001 mmol/L, 0.01 mmol/L, 0.1 mmol/L, 1 mmol/L), the expression of osteoblast-related genes was upregulated (30). In another study, treatment of human osteoblasts with atorvastatin enhanced the expression of ALP, OCN and OPG with a maximum effect at 1 mmol/L (31). However, other studies reported contradictory results. Kupcsik et al. (26) demonstrated that lower concentrations of statins (1 mmol/L) failed to induce calcification of human MSCs. Sonobe et al. (32) reported that simvastatin at the concentration of 0.01 mmol/L did not enhance the osteogenic differentiation of rat MSCs. It is believed that this discrepancy may result from the different culture systems and target cells. Moreover, there was no obvious increase for the genes’ expression of BMP-2, ALP, OCN, OPG and VEGF from day 3 to day 7, especially for the high concentration of simvastatin (1 mmol/L). The mechanism by which simvastatin can induce osteogenic differentiation of rat MSCs is unknown. It is generally agreed that the

52

Y. Qi et al.

Figure 7. Histological observation of fracture sites in each group at 2 weeks after surgery (w: woven bone; f: fibrous tissue; Sim: simvastatin; A-C, F-H, K-M, P-R, U-X: H&E staining; D, E, I, J, N, O, S, T, Y, Z: Safranin-O staining).

MSC sheet transplantation with simvastatin for bone repair

53

Figure 8. Histological observation of fracture sites in each group at 8 weeks after surgery (w: woven bone; c: cortical bone; f: fibrous tissue; arrow: chondrocytes; Sim: simvastatin; A, B, D, E, G, H, J, K, M, N: H&E staining; C, F, I, L, O: Safranin-O staining).

54

Y. Qi et al.

bone-forming effects of the simvastatin are associated with an increased BMP-2 expression level (7). BMP-2 might also exerts its main role in the osteogenesis of the simvastatin-stimulated MSCs and promote the expression of ALP, OCN, OPG and VEGF. However, the receptors of signal pathways, which simvastatin acts through, might be desensitized with long periods and high concentrations of simvastatin stimulation. This might explain the lack of change in the expression of BMP-2 from day 3 to day 7, which thus affected the expressions of ALP, OCN, OPG and VEGF. When simvastatin is used as a bone anabolic factor in vivo, local application through the use of a biocompatible, stable and drug-delivering scaffold has a therapeutic advantage by preventing systemic side effects and avoiding injections to the fracture site. Previous studies have investigated the dose effect of locally applied simvastatin on bone regeneration. In the study of Wong and Rabie (33), 0.5 mg of simvastatin in combination with collagen matrix was implanted into calvarial defects of rabbits, and 308% more new bone was present in the collagen/ simvastatin group compared with the collagen-alone group at 14 days. Stein et al. (34) found that 0.5 mg simvastatin appeared to be the optimal dose for single local application, which produced the best bone growth/inflammation ratio. On the basis of these findings, in the present study, 0.5 mg of simvastatin was incorporated into the CS scaffolds to investigate its effects on fracture healing. Local administration of statins with some carrier systems has been shown to significantly promote bone formation and healing (8,35). The results of these studies are in accordance with the present findings. In this study, the increased area of calluses, the increased bone formation and enhanced consolidation of the fracture healing in CS/simvastatin group proved that simvastatin had an anabolic effect on bone formation. Moreover, no serious side effects of simvastatin were seen in the present study, in accordance with Mundy’s study (7). The exact mechanisms of simvastatin on accelerating fracture healing are unclear. Statins have been demonstrated to stimulate BMP-2 production by endogenous bone cells, thereby enhancing bone formation in areas where remodeling is taking place (7), as shown in the present study. The CS/simvastatin group showed higher expression of BMP-2 at 2 weeks after implantation compared with CS and the control group. This would contribute to increased differentiation of recruited osteogenic cells to osteoblasts and subsequently enhanced osteogenesis in the fracture site. In consequence, ALP, which is known to be a differentiation marker of mature osteoblasts, and OCN, a late osteoblast

marker, are more highly expressed compared with CS and the control group. OPG, which is a protein produced by osteoblasts and had suppressor effects on osteoclast differentiation, also showed higher expression. Previous in vitro studies showed that mevastatin (36) and lovastatin (37) inhibited osteoclast differentiation. Neovascularization at the fracture site is considered to be necessary to achieve bone formation and union for fracture healing. The present in vitro study showed that the MSC sheet more highly expressed an angiogenic factor, VEGF, as well as osteogenic markers (BMP-2, ALP, OCN, OPG), compared with those of MSCs. Therefore, the MSC sheet produced VEGF around the fracture site, and, subsequently, VEGF induced revascularization and new bone formation. In addition, the higher expressions of osteogenic genes, especially BMP-2 and mineralized matrix, also can enhance osteogenesis at the fracture site. This was further clarified by the gene expression, x-ray, micro-CT and histological observations of the CS/MSC sheet group in vivo. The CS/MSC sheet group showed not only higher expressions of BMP-2, ALP, OCN, OPG and VEGF but also better bony bridging, compared with the control and CS groups. In a previous study (21), MSC sheet transplantation was shown to enhance bone formation in a rat nonunion model. Another possible reason for the effects of simvastatin on bone regeneration may be the effect on enhancing gene expression of VEGF (38). Maeda et al. (39) reported that statins augmented the expression of VEGF in osteoblastic cells in vitro. VEGF has been shown to induce BMP-2 expression, indirectly stimulating osteoblast activity (40). Elevated expression of VEGF also may result in the osteogenic differentiation of rat MSCs after stimulation with simvastatin. BMP-2 has also been shown to stimulate the expression of VEGF and vice versa (40,41). Therefore, osteogenic differentiation of rat MSCs may be stimulated by simvastatin as a consequence of upregulated expression of BMP-2 and VEGF or through a secondary response as the result of the increased complementary molecule. Simvastatin combined with the MSC sheet had synergistic effect on fracture healing, as shown in the CS/simvastatin/MSC sheet group. The CS/simvastatin/MSC sheet group showed the highest expression of osteogenic genes at 2 weeks and complete bony bridging at 8 weeks observed from x-ray, micro CT and histology. We think that our technique of cell sheet transplantation combined with locally released simvastatin for the treatment of fracture and nonunion has roles in VEGF and BMP-2eproducing cell transplantation as well as mineralization matrix transplantation around a fracture site.

MSC sheet transplantation with simvastatin for bone repair MSCs can be easily obtained from the bone marrow of patients. After proliferation in vitro, autologous MSC sheets formed. CS has been widely used in clinical situations and shows good biocompatibility and osteoconductivity. In addition, the present study demonstrated that CS was an effective carrier for local delivery of simvastatin. Therefore, simvastatin-releasing CS combined with the power of the MSC sheet could be applied to the sites of fracture or bone delayed unions or bone nonunions in clinical settings. By the combined effects of simvastatin locally released from CS and angiogenesis of the MSC sheet, fracture healing might be greatly improved. Conclusions In conclusion, both simvastatin and the MSC sheet contributed to the formation of new bone. MSC sheet transplantation together with locally applied simvastatin enhanced bone formation at the fracture sites. Such an MSC sheet and locally applied simvastatin might have great potential for the treatment of fractures and bone nonunions in clinical settings. Acknowledgments The project was supported by the Science Technology Program of Zhejiang Province (2008C13025), the Natural Science Foundation of China (81071259, 30901531) and the Natural Science Grants of Zhejiang Province (Y2090283, Y2090440). Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. References 1. Einhorn TA. Enhancement of fracture healing. Instr Course Lect. 1996;45:401e16. 2. Rubin C, Bolander M, Ryaby JP, Hadjiargyrou M. The use of low-intensity ultrasound to accelerate the healing of fractures. J Bone Joint Surg Am. 2001;83-A:259e70. 3. Yajima H, Tamai S, Mizumoto S, Inada Y. Vascularized fibular grafts in the treatment of osteomyelitis and infected nonunion. Clin Orthop Relat Res. 1993;293:256e64. 4. Safoury Y. Free vascularized fibula for the treatment of traumatic bone defects and nonunion of the forearm bones. J Hand Surg Br. 2005;30:67e72. 5. Szpalski M, Gunzburg R. Recombinant human bone morphogenetic protein-2: a novel osteoinductive alternative to autogenous bone graft? Acta Orthop Belg. 2005;71:133e48. 6. Govender S, Csimma C, Genant HK, Valentin-Opran A, Amit Y, Arbel R, et al. Recombinant human bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, controlled, randomized study of four hundred and fifty patients. J Bone Joint Surg Am. 2002;84-A(12): 2123e34.

55

7. Mundy G, Garrett R, Harris S, Chan J, Chen D, Rossini G, et al. Stimulation of bone formation in vitro and in rodents by statins. Science. 1999;286:1946e9. 8. Pauly S, Luttosch F, Morawski M, Haas NP, Schmidmaier G, Wildemann B. Simvastatin locally applied from a biodegradable coating of osteosynthetic implants improves fracture healing comparable to BMP-2 application. Bone. 2009;45: 505e11. 9. Bellosta S, Ferri N, Bernini F, Paoletti R, Corsini A. Nonlipid-related effects of statins. Ann Med. 2000;32:164e76. 10. Turner TM, Urban RM, Gitelis S, Kuo KN, Andersson GB. Radiographic and histologic assessment of calcium sulfate in experimental animal models and clinical use as a resorbable bone-graft substitute, a bone-graft expander, and a method for local antibiotic delivery. One institution’s experience. J Bone Joint Surg Am. 2001;83-A(Suppl 2):8e18. 11. McKee MD, Wild LM, Schemitsch EH, Waddell JP. The use of an antibiotic-impregnated, osteoconductive, bioabsorbable bone substitute in the treatment of infected long bone defects: early results of a prospective trial. J Orthop Trauma. 2002;16: 622e7. 12. Mousset B, Benoit MA, Delloye C, Bouillet R, Gillard J. Biodegradable implants for potential use in bone infection: an in vitro study of antibiotic-loaded calcium sulphate. Int Orthop. 1995;19:157e61. 13. Dacquet V, Varlet A, Tandogan RN, Tahon MM, Fournier L, Jehl F, et al. Antibiotic-impregnated plaster of Paris beads: trials with teicoplanin. Clin Orthop Relat Res. 1992;282:241e9. 14. Benoit MA, Mousset B, Delloye C, Bouillet R, Gillard J. Antibiotic-loaded plaster of Paris implants coated with poly lactide-co-glycolide as a controlled release delivery system for the treatment of bone infections. Int Orthop. 1997;21:403e8. 15. Dumas A, Moreau MF, Gherardi RK, Basle MF, Chappard D. Bone grafts cultured with bone marrow stromal cells for the repair of critical bone defects: an experimental study in mice. J Biomed Mater Res A. 2009;90:1218e29. 16. Giannoni P, Mastrogiacomo M, Alini M, Pearce SG, Corsi A, Santolini F, et al. Regeneration of large bone defects in sheep using bone marrow stromal cells. J Tissue Eng Regen Med. 2008;2:253e62. 17. Kon E, Muraglia A, Corsi A, Bianco P, Marcacci M, Martin I, et al. Autologous bone marrow stromal cells loaded onto porous hydroxyapatite ceramic accelerate bone repair in critical-size defects of sheep long bones. J Biomed Mater Res. 2000;49:328e37. 18. Zhou J, Lin H, Fang T, Li X, Dai W, Uemura T, et al. The repair of large segmental bone defects in the rabbit with vascularized tissue engineered bone. Biomaterials. 2010;31: 1171e9. 19. Ouyang HW, Toh SL, Goh J, Tay TE, Moe K. Assembly of bone marrow stromal cell sheets with knitted poly (L-lactide) scaffold for engineering ligament analogs. J Biomed Mater Res B Appl Biomater. 2005;75:264e71. 20. Yang J, Yamato M, Kohno C, Nishimoto A, Sekine H, Fukai F, et al. Cell sheet engineering: recreating tissues without biodegradable scaffolds. Biomaterials. 2005;26:6415e22. 21. Nakamura A, Akahane M, Shigematsu H, Tadokoro M, Morita Y, Ohgushi H, et al. Cell sheet transplantation of cultured mesenchymal stem cells enhances bone formation in a rat nonunion model. Bone. 2010;46:418e24. 22. Akahane M, Nakamura A, Ohgushi H, Shigematsu H, Dohi Y, Takakura Y. Osteogenic matrix sheet-cell transplantation using osteoblastic cell sheet resulted in bone formation without scaffold at an ectopic site. J Tissue Eng Regen Med. 2008;2:196e201. 23. Nyan M, Miyahara T, Noritake K, Hao J, Rodriguez R, Kuroda S, et al. Molecular and tissue responses in the healing

56

24.

25.

26.

27.

28.

29.

30.

31.

32.

Y. Qi et al. of rat calvarial defects after local application of simvastatin combined with alpha tricalcium phosphate. J Biomed Mater Res B Appl Biomater. 2010;93:65e73. Ayukawa Y, Yasukawa E, Moriyama Y, Ogino Y, Wada H, Atsuta I, et al. Local application of statin promotes bone repair through the suppression of osteoclasts and the enhancement of osteoblasts at bone-healing sites in rats. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009;107:336e42. Wadagaki R, Mizuno D, Yamawaki-Ogata A, Satake M, Kaneko H, Hagiwara S, et al. Osteogenic induction of bone marrow-derived stromal cells on simvastatin-releasing, biodegradable, nano- to microscale fiber scaffolds. Ann Biomed Eng. 2011;39:1872e81. Kupcsik L, Meurya T, Flury M, Stoddart M, Alini M. Statin-induced calcification in human mesenchymal stem cells is cell death related. J Cell Mol Med. 2009;13:4465e73. Zhou Y, Ni Y, Liu Y, Zeng B, Xu Y, Ge W. The role of simvastatin in the osteogenesis of injectable tissue-engineered bone based on human adipose-derived stromal cells and platelet-rich plasma. Biomaterials. 2010;31:5325e35. Baek KH, Lee WY, Oh KW, Tae HJ, Lee JM, Lee EJ, et al. The effect of simvastatin on the proliferation and differentiation of human bone marrow stromal cells. J Korean Med Sci. 2005;20:438e44. Maeda T, Matsunuma A, Kawane T, Horiuchi N. Simvastatin promotes osteoblast differentiation and mineralization in MC3T3-E1 cells. Biochem Biophys Res Commun. 2001;280: 874e7. Ruiz-Gaspa S, Nogues X, Enjuanes A, Monllau JC, Blanch J, Carreras R, et al. Simvastatin and atorvastatin enhance gene expression of collagen type 1 and osteocalcin in primary human osteoblasts and MG-63 cultures. J Cell Biochem. 2007;101:1430e8. Viereck V, Grundker C, Blaschke S, Frosch KH, Schoppet M, Emons G, et al. Atorvastatin stimulates the production of osteoprotegerin by human osteoblasts. J Cell Biochem. 2005; 96:1244e53. Sonobe M, Hattori K, Tomita N, Yoshikawa T, Aoki H, Takakura Y, et al. Stimulatory effects of statins on bone marrow-derived mesenchymal stem cells: study of a new

33.

34.

35.

36.

37.

38. 39.

40.

41.

therapeutic agent for fracture. Biomed Mater Eng. 2005;15: 261e7. Wong RW, Rabie AB. Statin collagen grafts used to repair defects in the parietal bone of rabbits. Br J Oral Maxillofac Surg. 2003;41:244e8. Stein D, Lee Y, Schmid MJ, Killpack B, Genrich MA, Narayana N, et al. Local simvastatin effects on mandibular bone growth and inflammation. J Periodontol. 2005;76:1861e70. Lee Y, Schmid MJ, Marx DB, Beatty MW, Cullen DM, Collins ME, et al. The effect of local simvastatin delivery strategies on mandibular bone formation in vivo. Biomaterials. 2008;29:1940e9. Woo JT, Nakagawa H, Krecic AM, Nagai K, Hamilton AD, Sebti SM, et al. Inhibitory effects of mevastatin and a geranylgeranyl transferase I inhibitor (GGTI-2166) on mononuclear osteoclast formation induced by receptor activator of NF kappa B ligand (RANKL) or tumor necrosis factor-alpha (TNF-alpha). Biochem Pharmacol. 2005;69:87e95. Fisher JE, Rogers MJ, Halasy JM, Luckman SP, Hughes DE, Masarachia PJ, et al. Alendronate mechanism of action: geranylgeraniol, an intermediate in the mevalonate pathway, prevents inhibition of osteoclast formation, bone resorption, and kinase activation in vitro. Proc Natl Acad Sci U S A. 1999;96:133e8. Weis M, Heeschen C, Glassford AJ, Cooke JP. Statins have biphasic effects on angiogenesis. Circulation. 2002;105:739e45. Maeda T, Kawane T, Horiuchi N. Statins augment vascular endothelial growth factor expression in osteoblastic cells via inhibition of protein prenylation. Endocrinology. 2003;144: 681e92. Bouletreau PJ, Warren SM, Spector JA, Peled ZM, Gerrets RP, Greenwald JA, et al. Hypoxia and VEGF upregulate BMP-2 mRNA and protein expression in microvascular endothelial cells: implications for fracture healing. Plast Reconstr Surg. 2002;109:2384e97. Deckers MM, van Bezooijen RL, van der Horst G, Hoogendam J, van Der Bent C, Papapoulos SE, et al. Bone morphogenetic proteins stimulate angiogenesis through osteoblast-derived vascular endothelial growth factor A. Endocrinology. 2002;143:1545e53.