Lewis rats (male, 4 weeks old) were used for this experiment. Rats were anesthetized with ketamine (Ketalar, Sankyo, Japan) (6 mg/100 g body weight) and medetomidine hydrochloride (Domitor, Orion, Finland) (0.04 mg/100 g body weight). The femoral bone marrow was flushed out with 10 ml of minimum essential medium (α-MEM, invitrogen, California) containing 15% fetal bovine serum (FBS) and antibiotics (penicillin 100 U/ml, streptomycin 100 μg/ml). Cells were seeded on a 100 mm dish and incubated in 5% CO₂ at 37 °C. For this study, passage-2 BMSCs were used [16].

Under intramuscular sedation, whole blood of Lewis rats (8-10 weeks old) was drawn from the heart into tubes containing anticoagulant. PRP was isolated by traditional two-step centrifugation. Centrifugation was carried out at 24 °C for 10 minutes at 600 g and the supernatant was transferred to another tube. Centrifugation of the supernatant, 2,840 g was conducted for 15 minutes at 24 °C. PRP and blood were counted automatically using a hematology analyzer (Celltac α, Nihon Kohden, Japan). Concentrations of platelets in PRP were adjusted as follows, 500 x 10⁴/μl, 250 x 10⁴/μl, 125 x 10⁴/μl, 62.5 x 10⁴/μl, 31.25 x 10⁴/μl, 15.625 x 10⁴/μl, and absence (control) for in vitro study. PRP was activated with 2% calcium chloride solution and bovine thrombin.

BMSCs were plated at a density of 1.0 x 10³ cells/150µl/ well in 96-well plates with 10 μl of PRP. The proliferation of BMSCs was measured on 1, 2, 3, 4 and 5 days using MTS assay kit (CellTiter 96 AQueous One Solution Cell Proliferation, Promega, Wisconsin). The assay was performed according to manufacturer instructions. This test was performed 5 times with 5 samples for each group.

BMSCs were plated at a density of 2.0 x 10⁴ cells/ml in 24-wel culture plates (1ml was added in each well) with 50 μl of PRP. On 2, 4, 6, 8, 10, 12 and 14 days, the medium was removed and cells were lysed with 200 μl of extraction reagent (CelLytic-M, Sigma-Aldrich, Missouri). ALP activity was measured using p -nitrophenyl phosphatase as a substrate (LabAssay ALP, Wako Pure Chemical Industries Japan). The assay was performed according to manufacturer instructions. This test was performed 4 times with 3 samples for each group.

BMSCs suspension and prepared collagen solution (Cellmatrix type I, Nitta Gelatin Inc. Japan) were mixed at 4 °C to achieve a mixture of cell density at 2 x 10⁶ cells/ml and a collagen concentration of 2.0 mg/ml. Five collagen mixtures consisting of different concentrations of PRP (200 μl) and BMSCs were prepared as follows, i) BMSCs/PRP(20) group of BMSCs and PRP (platelet 20 x 10⁴/μl), ii) BMSCs/PRP(100) group of BMSCs and PRP (platelet 100 x 10⁴/μl), iii) BMSCs/PRP(500) group of BMSCs and PRP (platelet 500 x 10⁴/μl), iv) BMSCs group, and v) PRP group (platelet 100 x 10⁴/μl). A two ml of collagen mixture (4 x 10⁶/cell) was poured onto one ml of collagen gel leyer without cells in each well and incubated for 15 minites in 5% CO₂ at 37 °C to prepare for collagen gels for implantation. To obtain a floating collagen gel, the attached collagen mixture was gently lifted off from the bottom of the 12-well plates and 2 ml α-MEM containing 15% FBS was added to the mixture. The collagen gel was then returned for incubation overnight and, having contracted. Then, two collagen gels with cells were implanted into each femur defect site.

A rat femoral segmental defect model was used in this study [16]. A 6-mm defect was made on the femoral diaphysis of recipient rats. Through an anterolateral approach, a high density polyethylene plate (4 x 4 x 23 mm) was attached to the lateral aspect of recipient femora (female, 8-10 weeks old). Six groups were prepared (Table 1). Two collagen gels described above were implanted in the femur defect site (Fig. 1). Our institution’s committee for animal experimentation approved all animal experiments (approval number 20045, Yamagata University Faculty of Medicine, February 20, 2009).

Fig. (1). Transplantation of bone marrow stromal cells (BMSCs) in femoral segmental defect model. To evaluate the survival of transplanted BMSCs by fluorescence in situ hybridization analysis, BMSCs harvested from male rats were transplanted to the defect of female rat.

Radiographic examination was performed using a soft X-ray apparatus (Softex; Softex Co, Japan) for 30 seconds exposure at 25 kV and 3 mA. Serial radiographs of rat femora were examined at 0, 2, 4, 6 and 8 weeks postoperatively. Levels of new bone formation were evaluated by soft X-ray. The total area to be analyzed corresponded to the entire area of the original surgical defect. Quantification of the bone fill percentage of newly formed bone was carried out using an image software package (ImageJ V1.41, NIH, Maryland).

At 2, 4, 6 and 8 weeks after implantation, harvested femora were fixed in 4% paraformaldehyde in phosphate buffered saline (pH 7.2) for 48 hours. Fixation femora were demineralized in 14% ethylenediaminetetraacetic acid (pH 7.2). Samples were embedded in paraffin, cut into 5 μm sections, and stained in hematoxylineosin was examined via light microscopy.

To evaluate the survival of transplanted BMSCs in newly formed bone, fluorescence in situ hybridization (FISH) analysis [17, 18] was performed on paraffin embedded histological sections of femur at 2, 4, 6 and 8 weeks after implantation. The rat Y-chromosome probe in plasmid was kindly provided by Dr. Barbara Hoebee (National Institute of Public Health and Environment Protection, Netherlands). The probe was labeled with digoxigenin by nick translation (DIG-Nick Translation Mix, Roche Diagnostics, Switzerland), and applied to the pretreated bone samples. Probe hybridization occurred over night at 37 °C using HYBrite (Vysis, Illinois). Hybridized slides were stained with rhodamin labeled anti-digoxigenin antibody, counterstained with 4', 6 -diamidino-2-phenylindole, dihydrochloride (DAPI). The survival rate of transplanted BMSCs was defined by Y-probe positive cell count in 100 non-overlapping nuclei in newly formed bone [16].

To evaluate the vessel immunohistochemistry analysis was performed on paraffin embedded histological sections of the femur at 2 and 4 weeks postoperatively. Histological sections were treated in protein kinase solution for immunostaining. Platelet endothelial cell adhesion molecule-1 (PECAM-1, rabbit anti-human IgG, Santa Cruz Biotechnology, Texas) was the primary antibody used. Staining was performed using Vector Elite ABC kits (Vector, California). The number of blood vessel cavities was measured using two fields per slide viewed at x 200 magnification. The fields examined were at the border of the existing bone and bone defect (Fig. 2). Three rats were used in each group and immunostaining was performed 2 times with each sample.

Fig. (2). Immunohistochemistry of vascular marker PECAM-1. To evaluate the vessel immunohistochemistry analysis was performed at 2 and 4 weeks after implantation. * The number of the blood vessel cavities was measured at the border of the existing bone and bone defect.

The data was performed by one-way analysis of variance (ANOVA) using the SPSS 2. We evaluated significant differences between two groups by fluorescence in situ hybridization using the Mann-Whitney U test. A p value of < 0.05 indicated statistical significance. All of the results were expressed as the mean ± standard deviation.

On days 3, 4 and 5, cellular proliferation was enhanced by the addition of PRP compared to control groups (Fig. 3). On day 5, the number of BMSCs with PRP (500 x 10⁴/μl) was approximately a 4-fold increase compared to the control groups. Increases in the number of cells were PRP concentration-dependent.

Fig. (3). Effect of platelet-rich plasma (PRP) on the proliferation of bone marrow stromal cells (BMSCs). The numbers of cells cultured with PRP with different concentrations of platelets were shown. (p < 0.05 * : versus control, † : versus 15.6 x 104/μl, ‡ : versus 31.2 x 104/μl, §: versus 62.5 x 104/μl, || : versus 125 x 104/μl, ¶ : versus 250 x 104/μl, ** : versus 500x 104/μl).

On days 2 and 4, with the addition of PRP, ALP activity decreased compared to control groups. However, on days 6, 8, 10 and 12, ALP activity with PRP (125 x 10⁴/μl, 62.5 x 10⁴/μl) increased compared to control groups. Thereafter, ALP activity decreased over time, but maintained a state that was higher than control (Fig. 4).

Fig. (4). Alkaline phosphatase (ALP) activity with platelet-rich plasma (PRP) of different platelet concentrations. ALP activity was increased with PRP (62.5 x 104/μl) on day 8.

BMSCs/PRP (100) group showed a significant increase in newly formed bone compared with BMSCs and PRP groups (p < 0.05) (Fig. 5). The bone fill percentage of newly formed bone in the BMSCs/PRP (100), BMSCs, PRP (100) and control groups were 46.9%, 10.9%, 12.5%, 18.5% at 8 weeks. The bone fill percentage of newly formed bone in the BMSCs/PRP (100) group was 46.9% and increased significantly compared with other groups (p < 0.05) (Fig. 6). In the BMSCs/PRP (100) group, complete bone union was not realized, but it suggested possible adequate bone formation.

Fig. (5). Radiographic findings. BMSCs/PRP (100) group showed significant increase in newly formed bone compared with BMSCs and PRP group at 8 weeks. (A-D) BMSCs/PRP (100) at 0, 2, 4 and 8 weeks postoperatively. (E-H) BMSCs group at 0, 2, 4 and 8 weeks. (I-L) PRP group at 0, 2, 4 and 8 weeks, (M-P) control (defect) group at 0, 2, 4 and 8 weeks.

Fig. (6). The bone fill percentage of newly formed bone. BMSCs/PRP (100) group showed significant increase in newly formed bone compared with other groups. * statistical significance, p < 0.05.

At 8 weeks in BMSCs/PRP (100) group, woven bone was partially observed, but lamellae bone was not adequately formed after transplant. In contrast, the defects in BMSCs and PRP (100) group demonstrated sparse woven bone at the proximal and the distal end and connective tissue was observed within the defect space. At 8 weeks in BMSCs / PRP (100) group, bone marrow was found in the newly formed bone area (Fig. 7).

Fig. (7). Histological findings (Hematoxylin-Eosin staining). (A) At 8 weeks in BMSCs/PRP (100) group, newly bone formed on the proximal fragment extended the length of the defect to the distal in histological section. (B) At 8 weeks in BMSCs/PRP (100) group, woven bone was partly observed. CB: original cortical bone, NB: newly formed bone, BM: bone marrow, WB: woven bone, LB: lamellae bone.

The detection efficiency was found to be 71 ± 4.8% in male positive normal controlof the rat without defect. This indicated the probe worked sufficientry. The Y-probe positive cells were found in the newly formed bone tissue. In BMSCs/PRP (100) and BMSCs group, the survival rate of cells with Y-probe positive were 19.7% and 17.4% in newly formed bone tissue, respectively at 8 weeks after surgery (Fig. 8).

Ingrowth was more significant in BMSCs/PRP (100) and PRP (100) group than in the control group at 2 and 4 weeks. However, a significant difference was not recognized between the two groups (Table 2, Fig. 9). In the BMSCs group the number of blood vessel cavities formed were comparable to those of the control group.

Fig. (8). FISH with Y-probe. The Y-probe positive cells were found in the newly formed bone tissue. The number of donor cells in BMSCs/PRP (100) group was similar to BMSCs group.

Fig. (9). Immunohistochemistry of vascular marker PECAM-1 at 4 weeks after implantation. The number of vascular vessels after implantation in BMSCs/PRP (100) group. Arrow: vascular vessel.