SEARCH

SEARCH BY CITATION

Keywords:

  • EPC-based therapy;
  • fracture healing;
  • segmental bone defect;
  • neovascularization;
  • osteogenesis

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The objective of this study was to evaluate the effects of local endothelial progenitor cell (EPC) therapy on bone regeneration in a rat model. A segmental bone defect (5 mm) was created in the femur and fixed with a mini-plate. There were two groups: EPC-treated (N = 28) and control (N = 28). Seven animals were sacrificed from each group at 1, 2, 3, and 10 weeks postoperatively. Healing of the defect was evaluated with radiographic, histological, and quantitative micro-computed tomography (micro-CT) scans. Radiographically, mean scores of the EPC and control groups were, respectively, 1.16–0.61 (p < 0.05) at 1 week, 2.53–1.54 (p < 0.05) at 2 weeks, and 4.58–2.35 at 3 weeks (p < 0.05). At 10 weeks, all the animals in the EPC-treated group had complete union (7/7), but in the control group none achieved union (0/7). Histological evaluation revealed that specimens from EPC-treated animals had abundant new bone and vessel formation compared to that in controls. Micro-CT assessment of the samples from the animals sacrificed at 10 weeks (N = 14) showed significantly improved parameters of bone volume (36.58–10.57, p = 0.000), bone volume density (0.26–0.17, p = 0.000), model index −2.22–2.79, p = 0.000), trabecular number (1.28–0.91, p = 0.063), trabecular thickness (0.21–0.15, p = 0.001), trabecular spacing (0.63–1.07, p = 0.022), bone surface (353.75–152.08, p = 0.000), and bone surface to bone volume ratio (9.54–14.24, p = 0.004) for the EPC group compared to control, respectively. In conclusion, local EPC therapy significantly enhanced bone regeneration in a segmental defect model in rat femur diaphysis. © 2010 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 28:1007–1014, 2010

Bone healing constitutes a unique regenerative process which requires coordinated coupling between osteogenesis and angiogenesis. Although regeneration of the bone can be completed without any scarring, a significant percentage of fractures fail to heal adequately.1, 2 Segmental bone defects after severe trauma, infection, and surgical removal of tumors remain a major clinical problem to be addressed. Vascular in-growth at the fracture site has a cardinal role in the healing process and regeneration of the bone postfracture.3 This observation has provided the rationale for the investigations showing the positive effects of vascular endothelial growth factor (VEGF) and erythropoietin (EPO) on fracture healing.4, 5

Endothelial progenitor cells (EPCs), first described by Asahara et al.6 in 1997, are bone marrow-derived cells with the ability to differentiate into endothelial cells and to participate in the establishment of neovasculature mainly by indirect, paracrine mechanisms. EPCs have been shown to express various endothelial surface markers such as CD34, VEGFR2, and CD133, and to home to sites of ischemia. These findings constituted the basis for the use of ex vivo expanded EPCs as a novel therapeutic option to augment neovascularization in animal models of myocardial infarction and limb ischemia.7 Studies in myocardial infarction models demonstrated that infusion of ex vivo expanded EPCs significantly improved blood flow and cardiac function, and reduced left ventricular scarring.8 Likewise, ex vivo expanded EPCs improved the neovascularization in rat models with hind limb ischemia.9 It was described previously that EPCs mobilized by signals from the bone regeneration sites may contribute to neovascularization and, thus, new bone formation in fracture healing.10 However, effects of local treatment with ex vivo expanded EPCs on healing of a segmental bone defect have not been reported to date.

The purpose of this investigation was to evaluate the effects of the local use of ex vivo expanded EPCs on the stimulation of angiogenesis and the promotion of bone healing at a fracture site in a rat femur osteotomy model.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Experimental Design

Fifty-six adult male Fisher-344 rats (250–300 g) (Charles River, Quebec, Canada) were randomly assigned to two groups of equal size: the EPC group (N = 28) and the control group (N = 28). All handling and treatment procedures were approved by St. Michael's Hospital Institutional Animal Care and Use Committee. Rat bone marrow-derived EPCs were isolated and cultured for 7–10 days in endothelial cell growth medium. A segmental bone defect of 5 mm was created in the rat femur diaphysis, and stabilized with a mini-plate and screws (Synthes, Mississauga, Canada). The treatment group received 1 × 106 EPCs within gel-foam locally at the area of the bone defect, and control animals received only saline gel-foam with no cells. Seven animals were sacrificed from each group at 1, 2, 3, and 10 weeks postoperatively. Assessment of bone healing was performed by radiographs, histology, and quantitative micro-computed tomography (micro-CT).

Isolation, Culturing, and Characterization of EPCs

Femur and tibia bone marrow from syngenic Fisher-344 rats (250–300 g) were flushed. Wash-out solution was centrifuged by using Ficoll density-gradient to isolate the EPCs from the buffy layer similar to previously described methods.11 Collected cells were plated on fibronectin-coated tissue culture flasks and maintained in endothelial basal medium (EBM-2, Clonetics, San Diego, CA) supplemented with EGM-2-MV SingleQuots which contain VEGF, R3-insulin-like growth factor-1 (R3-IGF-1), human epidermal growth factor-B (hEGF-B), hydrocortisone, gentamycin-amphotericin (GA-1000), 5% fetal bovine serum (FBS), human fibroblast growth factor (hFGF-B), and ascorbic acid. After 4 days in culture, nonadherent cells were removed by suction of the media and washing with phosphate-buffered saline (PBS), new media (EGM-2-MV) was added, and the culture was continued for 7–10 days.

Cultured cells were initially characterized under phase contrast microscopy for the morphologic assessment of the attached cells within the tissue culture flask. A cobblestone pattern which is typical for confluent endothelial cells was detected in all the flasks (Fig. 1A).12 The phenotypes of cultured cells were further characterized according to their ability for uptake of Dil-labeled acetylated-LDL to confirm their endothelial lineage. We incubated the adherent cells with 10 µg/mL Dil-Ac-LDL (Biomedical Technologies, Stoughton, MA) at 37°C for 4 h. After the incubation, cells were fixed with 3% formaldehyde for 20 min and viewed under a fluorescent microscope. The entire cell populations on each slide were effectively stained positive with Dil-Ac-LDL (Fig. 1B).13

thumbnail image

Figure 1. (A) Cobblestone morphology of cultured EPCs at day 10 (original magnification, ×20). (B) View of Dil-Ac-LDL-labeled EPCs under fluorescent microscope (original magnification, ×40).

Download figure to PowerPoint

Surgical Procedure

Thirty minutes before every surgery, pieces of sterile gel-foam (each having cubic dimensions 5 × 5 × 5 mm) were aseptically impregnated with either 0.3 mL PBS (control), or with 0.3 mL media containing 1 × 106 cells (EPC) depending on the study group of the animal. Anesthesia was induced with 5% Isoflurane inhalation and maintained with 2% Isoflurane inhalation throughout the procedure. The right thigh area was shaved, scrubbed with Betadine (povidone-iodine, 10% solution), and draped. Under aseptic conditions, the entire length of the right femur was exposed through a lateral approach with a minimum damage to muscles. The mid-diaphysis of the femur was marked by measuring equal distances from both the proximal and the distal ends of the bone. By having this point as the reference, two mid-diaphyseal osteotomies were created with an oscillating saw to form a 5-mm segmental bone defect. Fixation of the femurs was performed with a mini-plate (1.5 mm) with five holes. Two cortical screws (1.5 mm) were inserted on each side of the osteotomy, and the middle hole of the plate was left empty as it was at the level of the osteotomy gap in all the procedures. After inserting the gel-foam piece into the osteotomy gap, the muscles were sutured from deeper to the superficial layers, and fascia and skin were closed with absorbable sutures. Animals with intraoperative complications such as split fractures or nonstandard osteotomies were excluded from the study before being assigned to either group. Figure 2 schematizes the surgical procedure.

thumbnail image

Figure 2. (A) Collected EPCs (1 × 106) are suspended in 0.3 mL media and added into a gel-foam piece. (B) After the removal of a 5-mm bone piece from the mid-diaphysis of the right femur and plate fixation, gel-foam containing the cells is implanted in to the gap. (C) Enhanced bone regeneration is expected.

Download figure to PowerPoint

The rats were allowed unrestricted movements and walking in their cages after they recovered from the anesthesia, and they were observed daily for activity and wound-healing.

Radiographic Assessment

Anteroposterior radiographs of the operated femur were made on the day of sacrifice for the animals in the 1-, 2-, and 3-week groups. For the animals sacrificed at 10 weeks postoperatively, anteroposterior radiographs were made every 14 days. However, only the radiographs that were made on the day of sacrifice (10 weeks) were used for radiographic assessment of this group.

For the animals sacrificed at 1, 2, and 3 weeks, assessment of bone healing was performed by scoring the radiographs from 0 to 5 according to percentage of the bone defect filled with new bone and density of the callus formation (Table 1).14 The evaluation was done by two independent and blinded readers, each on two occasions. The mean radiographic score for each limb was then calculated.

Table 1. Radiographic Analysisa
ScoreVolumeDensity
  • a

    Scoring was performed based on the percentage of bone filling and the density of callus at the osteotomy gap.

00N/A
11%–20%Low
221%–40%Intermediate
341%–60%Intermediate
461%–80%Dense
581%–100%Dense

Evaluation of the radiographs of the animals that were sacrificed at 10 weeks was performed according to the presence or absence of bridging callus formation between both ends of the osteotomy gap with complete bone filling of the defect. This evaluation was also done by two independent and blinded readers, each on two occasions.

Specimen Harvest

The animals were sacrificed at the predetermined time points after induction of general anesthesia with 5% Isoflurane inhalation. The operated femur was harvested with a portion of surrounding soft tissue and fixed in 4% paraformaldehyde solution for 24 h. After this step, plates and screws were removed from the bone. The specimens that were taken from the animals in the 1-, 2-, and 3-week groups (N = 42) were put through histologic processing sequentially. The femurs from the animals that were sacrificed at 10 weeks (N = 14) were stored in 70% ethanol until quantitative micro-CT was done.

Quantitative Micro-CT

Micro-CT scans of each femur in the 10-week group (N = 14) were conducted with a micro-tomography system (MicroCT40, Scanco Medical, Basserdorf, Switzerland). All trimmed femur samples were placed in a poly-ethyl-imid (PEI) holder with 70% ethanol and scanned at 70 kVp and 114 µA. The specimens were scanned in high-resolution mode, and scanning time for each specimen was approximately 1.5 h.

The final three-dimensional images were composed of 1,000–1,100 axial-cut slices, each with 6 µm in thickness. After scanning and reconstruction, a region of interest (ROI) including the osteotomy site was determined, and quantification of bone morphometry was performed.15

Histological Procedure

The fixed histology specimens from the animals in the 1-, 2-, and 3-week groups (N = 42) were cut through the soft tissue margins both proximal and distal to the new bone, then completely through the osteotomy lines, in order not to lose any newly formed bone tissue. The tissue samples were processed without decalcification, and were dehydrated and embedded in paraffin.

The specimens from the 10-week group animals (N = 14) were decalcified within EDTA solution following micro-CT scanning and were cut 5 mm beyond the osteotomy lines on both sides. The samples then were dehydrated and embedded in paraffin.

The samples within paraffin blocks were longitudinally cut into 5-µm sections, and prepared slides were stained with hematoxylin and eosin (H&E). After staining, the slides were evaluated qualitatively at 20× magnification for comparison of the EPC group with the controls, according to the amount of new bone and vessel formation.16

Statistical Analysis

SPSS Base 16.0 software for Windows (© 2007 SPSS Inc., Chicago, IL) was used to calculate mean and SE for each group. Charts were used to display the data distribution and to visually assess the differences between study groups. The significance of differences in mean values between groups was statistically tested by ANOVA. A value of p < 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

There were no superficial or deep infections of the surgical site during the postoperative period in any of the animals. All the animals had returned to their usual activities with nonrestricted weight bearing in their cages within 72 h postoperatively. No signs of adverse reactions were observed following transplantation of the cells in the treatment group due to the syngenic nature of the animals.

Radiographic Analysis

The radiographs of 28 animals from the EPC group and 28 animals from the control group were analyzed. Comparison between subgroups including seven rats from each group was done according to the postoperative week that the animals were sacrificed. At 1 week, new bone formation in the osteotomy gap was unremarkable in either group. However X-rays from the EPC group animals had more radiographic signs of early callus formation compared to controls (Fig. 3A, B). The scores tended to be higher in the EPC group than the control group with mean scores of 1.2–0.6 (p < 0.05), respectively.

thumbnail image

Figure 3. Plain radiographs from different animals at different time points: (A, C, E, G) from the EPC group, and (B, D, F, H) from control group animals at 1, 2, 3, and 10 weeks, respectively. Arrowheads indicating the osteotomy gap.

Download figure to PowerPoint

At 2 weeks, callus formation was evident especially in the radiographs from the EPC group animals (Fig. 3C, D). The difference in mean radiographic scores between the EPC group and the control group had increased compared to the first week. EPC-treated animals had a mean score of 2.5, and the control group had a mean score of 1.5 (p < 0.05).

At 3 weeks, radiographic signs of bone healing were significant in both groups. Evaluation of EPC group radiographs revealed noticeably more callus formation compared to controls (Fig. 3E, F). In the EPC group, out of seven animals, five had healed with bridging callus formation, whereas in the control group, none of the animals had bridging callus. The difference between mean radiographic scores increased further compared to 2 weeks. The mean score of the EPC and the control groups were 4.6 and 2.4, respectively (p < 0.05). Table 2 and Figure 4 show the mean radiographic scores at 1, 2, and 3 weeks.

Table 2. Mean Radiography Scores with SE and p-Values
ScoresMean Value ± SE for Each Groupp-Value
EPCControl
  1. EPC, endothelial progenitor cell.

Week 11.2 ± 0.20.6 ± 0.10.041
Week 22.5 ± 0.31.5 ± 0.20.023
Week 34.6 ± 0.12.4 ± 0.30.000
thumbnail image

Figure 4. Bar graph showing the mean radiography scores at 1, 2, and 3 weeks for the EPC (blue columns) and control (green columns) groups (ANOVA, p < 0.05 at every time period).

Download figure to PowerPoint

At 10 weeks, radiographic evaluation demonstrated a major difference between groups. All seven animals in the EPC group had complete union with 100% dense callus filling of the osteotomy gap, whereas in the control group, none of the animals had union or bridging callus formation (Fig. 3G, H).

Micro-CT Analysis

Assessment of micro-CT scans of the samples from the EPC (N = 7) and control (N = 7) animals, that were sacrificed at 10 weeks postoperatively, quantified the considerable differences between groups in new bone formation at the osteotomy site (Fig. 5). The following new bone structural parameters were calculated and statistically analyzed from the region of interest at the osteotomy site: bone volume (BV, mm3), bone volume density (BV/TV, %), model index (SMI), trabecular number (Tb.N, mm−1), trabecular thickness (Tb.Th, mm), trabecular spacing (Tb.Sp, mm), bone surface (BS, mm), and bone surface to bone volume ratio (BS/BV, mm−1). Table 3 includes the mean values with SE, and p-values. Figure 6 shows the graphic representation of each of the calculated parameters. The micro-CT bone structural parameters showed significant differences between treatment and control groups.

thumbnail image

Figure 5. Micro-CT images showing superior bone healing with filling of the entire defect with new bone in the EPC-treated group (A), compared to insufficient bone formation in the control group (B).

Download figure to PowerPoint

Table 3. Quantitative Micro-CT: Mean Parameter Values with SE and p-Values
ParameterMean Value ± SE for Each Groupp-Value
EPCControl
  1. EPC, endothelial progenitor cell; BV, bone volume; BV/TV, bone volume density; SMI, model index; Tb.N, trabecular number; Tb.Th, trabecular thickness; Tb.Sp, trabecular spacing; BS, bone surface; BS/BV, bone surface to bone volume ratio.

BV (mm3)36.58 ± 2.3410.57 ± 1.200.000
BV/TV (%)0.26 ± 0.180.17 ± 0.010.000
SMI−2.22 ± 0.742.79 ± 0.660.000
Tb.N (1/mm)1.28 ± 0.120.91 ± 0.130.063
Tb.Th (mm)0.21 ± 0.010.15 ± 0.010.001
Tb.Sp (mm)0.63 ± 0.891.07 ± 0.140.022
BS (mm)353.75 ± 31.77152.08 ± 26.140.000
BS/BV (1/mm)9.54 ± 0.3314.24 ± 1.280.004
thumbnail image

Figure 6. Histograms for new bone structural parameters showing significantly increased bone formation in the EPC-treated animals: (A) bone volume, (B) bone volume density, (C) model index, (D) trabecular number, (E) trabecular thickness, (F) trabecular spacing, (G) bone surface, and (H) bone surface to bone volume ratio (ANOVA, p < 0.05 for each parameter).

Download figure to PowerPoint

Histologic Analysis

At 1 week, new bone tissue and vessel formation were unremarkable in either group histologically. Remnants of gel-foam were abundant at the osteotomy gap in both groups. However, slides from the EPC group revealed considerably more cellular elements at the osteotomy site compared to controls (Fig. 7A, B). At 2 weeks, new bone tissue formation became evident particularly in the EPC group. The amount of woven bone tissue and vessel formation at the osteotomy gap was markedly increased in the EPC group compared to control group (Fig. 7C, D). At 3 weeks, the EPC group animals showed more organized filling of the osteotomy defect. There was significant trabecular bone tissue formation within the gap in the EPC group compared to control group, where scarce bone tissue and gel-foam remnants still filled the majority of the defect (Fig. 7E, F).

thumbnail image

Figure 7. Representative histologic specimens from the EPC and control groups at 1, 2, 3, and 10 weeks. Specimens at 1 week show increased cellular elements in the EPC slide (A) compared to control (B) in which remnants of gel-foam seem more abundant (H&E; original magnification, ×20). Specimens at 2 weeks demonstrate more woven bone and blood vessel formation in the EPC (C) compared to control group (D) (H&E; original magnification, ×20). Specimens at 3 weeks show more trabecular bone formation in the EPC (E) compared to control group (F) (H&E; original magnification, ×20). Histological pictures of the osteotomy gap and the adjacent bone at 10 weeks demonstrate that the EPC group (G) had remarkably more newly formed trabecular bone and bridging of the osteotomy gap, whereas the control group (H) had predominantly fibrotic tissue filling of the osteotomy gap with patchy areas of ossification (H&E; original magnification, ×2).

Download figure to PowerPoint

At 10 weeks, assessment of the EPC group animals revealed abundant callus formation and ossification across the entire gap, and that the osteotomy defects were filled with newly formed trabecular and cortical bone. Conversely, the control group animals showed patchy filling of the defect with newly ossified areas and fibrotic tissue (Fig. 7G, H).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Endothelial progenitor cells have been shown by many researchers to be effective in cell-based therapies to improve vascularization for a variety of therapeutic applications.17, 18 To our knowledge, this is the first study reporting the effects of local EPC therapy on healing of a segmental bone defect. Our results demonstrated that local use of EPCs in a defect model in a rat femur significantly improved callus formation and healing of the osteotomy gap.

In our study design, we primarily aimed to evaluate the effects of local use of EPCs at a segmental bone defect, and compare the difference that EPC therapy could make versus a control group. Therefore, the size of the osteotomy gap (5 mm) we created, although smaller than those reported in similar studies, was not considered as a limitation.19 In addition, we subdivided both the EPC and the control groups into 1-, 2-, 3-, and 10-week groups in order to be able to monitor the effects of EPC therapy both during early and later stages of fracture healing, and the differences between the two groups at these time points postfracture. Our analysis revealed the remarkable effects of EPCs in enhancing bone regeneration even at earlier time points, such as 2 and 3 weeks after the surgery. Findings at 10 weeks after the surgery also showed an impressive difference between the two groups. Complete fracture union was observed in all the animals in the EPC group, whereas in the control group, none of the animals had bridging callus formation. These results encourage us to speculate that local EPC therapy may become an effective cell-based treatment modality to promote bone healing in the near future.

An important question concerning postfracture local EPC therapy relates to the possible mechanisms that contribute to enhanced bone regeneration. Previous reports have demonstrated EPCs' potential to differentiate into endothelial cells and participate in neovascularization.20–22 Moreover, it was reported that there is a high plasticity among cells of the myeloid lineage and, under specific growth conditions, these cells may be possibly differentiating into cells of another lineage with distinct functional properties.23 Supporting this, Matsumoto et al.24 recently highlighted multilineage differentiation of CD34-positive cells into endothelial cells and also osteoblasts to develop a favorable environment for fracture healing via vasculogenesis and osteogenesis. In our study, we did not correlate the amount of new vessel formation at the osteotomy gap, particularly during the early phase of fracture healing, with the significant differences in the amount of new bone formation during the same postoperative time periods. However, we comment that transplanted EPCs also had osteogenic effects along with vasculogenesis in the treatment group. Further investigation is required to determine the relative proportions of locally transplanted EPCs that differentiate into endothelial cells for neovascularization or into osteoblasts for osteogenesis, and promote the tremendous bone regeneration potential we observed in our study. It would also raise the possibility that EPCs may be acting by paracrine mechanisms releasing chemokines and growth factors that may enhance bone regeneration.

Several animal studies have previously suggested that transplantation of culture expanded bone marrow-derived mesenchymal stem cells (MSCs) has value in bone-healing applications.25 We did not compare EPC therapy with other local cell-based therapies such as MSCs, and this may appear to be a potential limitation of this study that warrants discussion. However, since there were no previous studies in the literature reporting the effects of local EPC therapy on bone regeneration in a segmental defect model, our primary objective was to demonstrate the value of EPCs as a new cell-based treatment modality to promote bone healing. Hence, future studies focusing on the local use of EPCs postfracture, and comparing EPCs with other cell-based treatment modalities, will be important to fully realize the potential of these cells.

Another point which requires discussion is the use of gel-foam as a scaffold to transfer the EPCs to the osteotomy site. Gel-foam provides a mechanical matrix that facilitates clotting and offers a framework for deposition of the cellular elements of blood.26 Given the importance of hematoma formation postfracture, we speculate that gel-foam did not have any inhibitory effect on fracture healing; on the contrary, it possibly provided a superior environment at the osteotomy site. In our study design, all the animals in both groups were treated with gel-foam pieces of same size to minimize any additional influence on bone regeneration at the fracture site.

We transferred the EPCs within gel-foam and 0.3 ml media including growth factors, as opposed to the control group where gel-foam was implanted with 0.3 ml saline. Our aim was to keep the cells viable from the time of collection until transplantation to the animals, and this may appear as a limitation of our study. In a separate experiment to determine the effect of the media and growth factors, we compared a group of animals treated with gel-foam and 0.3 ml media including growth factors with control animals where gel-foam was implanted with 0.3 ml saline. There was no difference between these two groups with regard to bone healing during a period of 10 weeks postoperatively. Hence, we state that the superior healing effect we observed in the group treated with EPCs should be attributed solely to the use of EPCs.

In conclusion, local use of ex-vivo expanded endothelial progenitor cells significantly enhanced bone healing in a segmental defect model in rat femur diaphysis, compared with that in control animals. On the basis of these results, local use of EPCs should be further investigated as a promising cell-based therapy to promote bone regeneration at a fracture site.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

This work was funded by the Osteosynthesis and Trauma Care (OTC) Foundation, and the Physicians' Services Incorporated Foundation (PSIF).

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES