New Bone Formation by Murine Osteoprogenitor Cells Cultured on Corticocancellous Allograft Bone

Transcription

1 New Bone Formation by Murine Osteoprogenitor Cells Cultured on Corticocancellous Allograft Bone Ehren R. Nelson, 1 Zhinong Huang, 1 Ting Ma, 1 Derek Lindsey, 2 Christopher Jacobs, 2 Robert L. Smith, 1 Stuart B. Goodman 1 1 Department of Orthopaedic Surgery #R153, Stanford University Medical Center, 300 Pasteur Drive, Stanford, California , 2 Bone and Joint RR&D Center, The Palo Alto Veterans Administration Health Care Center, Palo Alto, California Received 2 August 2007; accepted 21 February 2008 Published online 3 June 2008 in Wiley InterScience ( DOI /jor ABSTRACT: The gold standard for bone grafting in orthopedics is autograft, however autograft has a limited supply and is associated with significant morbidity at the harvest site. One alternative, allograft bone, provides an osteoconductive scaffold, is in less limited supply, and it does not require a harvest from the patient. However, allograft lacks both osteogenic cells and osteoinductive proteins that make autograft bone so advantageous. This study provides a model to investigate strategies for augmentation of corticocancellous allograft bone discs with bone marrow-derived osteoprogenitor cells (OPCs) plus exogenous growth factors in vitro. In this model, allograft bone discs were created by cutting 1-mm thick slices from the distal femur and proximal tibia of euthanized mice. The allografts were sterilized and scanned by microcomputed tomography (mct) to provide the pre-culture graft volume and trabecular characteristics. The discs were then seeded with OPCs harvested from murine bone marrow. The seeded grafts were placed in organ culture until harvest, after which they were re-scanned by mct and the data compared to the corresponding pre-culture data. In addition, bone morphogenetic protein-7 (BMP-7, also know as osteogenic protein-1 or OP-1), basic fibroblast growth factor (bfgf), and OP-1 combined with bfgf were added on a daily basis to the cultures. After final mct scanning, all grafts were sectioned and evaluated histologically after hematoxylin and eosin (H&E) staining. mct scans of cultured allografts with cells at 3, 5, and 9 weeks showed a time-dependent, statistically significant increase in bone volume. The trabecular thickness (Tb.Th.) of grafts, from both groups that were augmented with OP-1, showed a statistically significant increase in trabecular thickness of allografts with OPCs. These data suggest that bone marrow-derived OPCs adhere to, and produce, new bone on corticocancellous allograft in vitro. When exogenous OP-1 is added to this model, an increase in the production of bone onto the corticocancellous allograft bone disc is seen. This model allows for the investigation of the effects of multiple growth factors, and other interventions, on OPCs seeded onto allograft bone in vitro. ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 26: , 2008 Keywords: osteoprogenitor cells; allograft bone; growth factors; BMP-7; FGF Over 450,000 bone graft procedures are performed each year in the United States. 1. Bone grafts are commonly used for the treatment of defects after major bone trauma, infection, bone tumor resection, and hip and knee revision surgery. Autogenous bone grafting is the gold standard for treating bone defects. Autograft serves as an osteoconductive scaffold that contains both osteoinductive growth factors and cells of the osteogenic lineage. Additionally, there is no risk of an immunogenic host response to the graft. However, there is a limited supply of autogenous graft material and the harvest requires a second incision site that lengthens the operation and increases the risk of blood loss, infection, hematoma formation, and other complications. 2,3 Chronic pain at the harvest site has also been noted to be a major source of morbidity in up to 24% of these patients. 4 One alternative to autogenous bone grafting is the use of allogeneic bone, which accounts for over 200,000 bone graft procedures in the USA every year. 5 Allograft bone serves as an osteoconductive scaffold that supports bone formation, and because its supply is less limited, allows reconstruction of large bone defects. However, allograft bone lacks osteogenic precursor cells and only has a fraction of the osteoinductive capacity of autograft bone. 6 As a result, the incorporation of allograft bone proceeds more slowly than autograft bone. 7 Correspondence to: Stuart B. Goodman (T: ; F: ; goodbone@stanford.edu) ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. Synthetic scaffolds could potentially be used to deliver both osteogenic cells and growth factors to large bone defects. Although some scaffolding systems show promise, an appropriate, inert, completely resorbable scaffold has not yet found its way into clinical use. Pluripotent stem cells, such as mesenchymal cells retrieved from bone marrow, can be stimulated along the osteoblastic lineage when cultured in an environment containing appropriate nutrients and factors. 8 Several growth factors have been shown to be positive regulators of bone production in vitro and in vivo, including osteogenic protein -1 (OP-1) also known as bone morphogenic protein-7 (BMP-7), and basic fibroblast growth factor (bfgf) also known as FGF-2. 9 The current investigation introduces a novel model for the study of cultured osteoprogenitor cells (OPCs) on cortico-cancellous bone allografts to enhance the efficacy of allograft bone incorporation. Development of a successful allograft model would provide a basis for evaluating seeded allograft bone as a delivery system for osteoprogenitor cells. Potentially, this could facilitate accelerated, more efficacious graft incorporation. Such a model would also facilitate new investigations into the mechanisms of osteoprogenitor cell growth and differentiation on allograft bone. MATERIALS AND METHODS Bone Disc Slices Institutional guidelines for the care and use of laboratory animals were observed in all aspects of this project. Ten-weekold C57 male mice were killed with CO 2 gas, and the femora and tibiae excised and cleaned of soft tissue. Each bone was cut using a Buehler Isomet Plus Precison Saw (Lake Bluff, IL) into 1660

2 OPCS CULTURED ON CORTICOCANCELLOUS ALLOGRAFT BONE mm slices made from the metaphyses of the distal femora and proximal tibiae. Slices were taken from the metaphysis to obtain bone discs with trabecular bone in the marrow space. Each slice was then quantitatively evaluated at a resolution of 10.5 mm using a Scanco Medical micro-ct (Southeastern, PA). The mct scans allow characterization of the trabecular bone within the medullary space of the bone disc. The data collected includes total volume of the graft (TV), bone volume (BV), and bone volume/graft volume (BV/GV). In addition, the trabecular number (Tb.N.), trabecular thickness (Tb.Th.), and trabecular spacing (Tb.Sp.) were also evaluated. Data was obtained by using drawn contours on the 2D axial CT images to either include the cortical bone for whole graft data (TV, BV, BV/GV) or to exclude the cortex to obtain data pertaining to the trabecular bone alone (Tb.Th., Tb.N., Tb.Sp.). All data from each allograft bone disc was saved for later comparisons with post-harvest scans after various treatments. After scanning, each graft was placed in a labeled 0.5-ml centrifuge tube with sterile saline and stored at 808C. Before seeding with cells, the grafts were defatted and sterilized by placing the bone discs in a 1:1 solution of methanol:chloroform for 12 h, and then washed three times in 100% methanol, three times in sterile water, and three times in sterile saline. Osteoprogenitor Cell Harvest Bone marrow was collected from the femora of C57 male mice, aged weeks old. 10,11 The mice were sacrificed with CO 2 gas, shaved, and sterilized in 70% ethanol for 5 min and then 90% ethanol for 5 min. While maintaining sterile technique, the femora were surgically removed. Using a syringe and 25-gauge needle, the bone marrow was flushed by injecting 4 ml of culture medium through the marrow cavity into a 15-ml centrifuge tube. The cells were washed three times, resuspended, and counted with a hemocytometer. Cells were then plated in a 75-cm 2 culture flasks at a concentration of cells per flask in a-eagle s minimum essential medium (EMEM) with 10% FBS and penicillin/streptomycin antibiotics. Cells were allowed to expand for 5 days before seeding onto the allograft bone. Medium was changed at 3 and 5 days to remove nonadherent cells. Seeding of Bone Marrow-Derived OPCs onto Bone Discs Each allograft bone disc was placed into one well of a 96-well plate. After the initial 5-day cell expansion period in a-emem, the cells were lifted with 0.25% Trypsin/EDTA, pooled, spun down, and resuspended in augmented osteogenic medium (a-emem supplemented with 10% FBS, 100 mg/ml penicillin, 100 mg/ml streptomycin, 50 mg/ml L-ascorbic acid, and 0.01 M b-glycerophosphate, 10 nm dexamethasone) at a concentration of cells/ml; cells in 300 mlof osteogenic medium was added onto each bone disc. The plate was incubated for 72 h. After the 72-h seeding period, each graft was removed and placed into one well of a new 96-well plate containing 300 ml of fresh osteogenic medium. Each plate was placed in the incubator and the medium changed every 4 days, and the spent medium placed in a 0.5-ml tube and stored at 808C. Four seeded bone discs were harvested at one of three time points: 3 weeks, 5 weeks, and 9 weeks. After harvest, each sample was scanned by m CT, followed by histological processing. The second part of this study used the same allograft preparation and cell seeding procedure, with the addition of exogenous growth factors. Grafts were placed into one of five groups: (1) grafts with no osteoprogenitor cells added; (2) grafts with osteoprogenitor cells added; (3) grafts with cells þ OP-1 added; (4) grafts with cells þ bfgf added; and (5) grafts with cells þ OP-1 and bfgf. The OP-1 and bfgf were added daily at concentrations of 100 ng/day and 50 ng/day, respectively, as in our previous studies. 12 The OP-1, provided by Stryker Corporation, Kalamazoo, Michigan was resuspended in HCl, according to the manufacturer. All allograft bone samples were subsequently harvested at 4 weeks, after which they were scanned by mct, and processed for histology. Evaluation After harvest, the seeded bone discs were fixed for 30 min in 4% paraformaldehyde. The grafts were then placed in PBS and re-scanned by mct. Evaluation of the m CT data from the bone discs included the TV, BV, BV/GV, Tb.N., Tb.Th., and Tb.Sp. Data from each graft was then compared to the data from the original, pre-seed scan of that particular graft. This allowed an accurate and quantitative evaluation of any changes in bone volume and trabecular characteristics of the allograft bone discs. A paired t-test was used to evaluate the statistical differences between the control (pre-culture scans) and the seeded bone discs. After mct scanning, the grafts were stored in formalin, decalcified, paraffin-embedded, sectioned with a microtome, and stained with hematoxylin and eosin. The sections were examined morphologically for the presence of new cells and bone under a magnification of 40, 100, and 200. RESULTS Time Course Data mct scanning of grafts after in vitro culture showed increased new bone deposition on the graft with the addition of OPCs. Figure 1 shows the change in BV after treatment as the percent increase over control preculture scans. A significant progressive, time-dependent increase in bone volume was found at all time points. There was also a significant increase in BV/GV (Figure 2) at 5 and 9 weeks, and an expected, reciprocal decrease in trabecular spacing at these time points (not shown). In addition, the trabecular thickness (Tb.Th.) was significantly increased at 5 and 9 weeks (p < 0.05, Figure 3). The trabecular number (Tb.N.) showed no significant differences between weeks 3, 5, and 9. Figure 1. Percent increase in bone volume over time.

3 1662 NELSON ET AL. Figure 2. Percent increase in bone volume divided by graft volume over time. Addition of Growth Factors The addition of OP-1 to the seeded grafts significantly increased the osteogenic activity of the cells. The addition of OP-1 alone was the only group to show a statistically significant increase in all measured endpoints including total BV, BV/GV, and Tb.Th (7.71%). When combined with bfgf, OP-1 showed a statistically significant increase in the Tb.Th. of the grafts (7.58%) (Fig. 4), but failed to show an increase in BV/GV. Treatment with bfgf alone showed a significant increase in the BV/GV (9.6%), but with no significant difference in Tb.Th. at 4 weeks (Fig. 4). As in the earlier part of this study, grafts with cells alone showed a significant increase in the bone BV/GV (22.6%) (Fig. 5). Results from mct scanning of the group of grafts cultured with cells alone showed no significant change in the trabecular thickness (Tb.Th.) at 4 weeks (Fig. 4). Although the treated groups showed some increase in activity over the untreated controls, consistent increases in bone production in all treated groups were seen only in the groups treated with OP-1 alone. Histology Histological sections were evaluated qualitatively for graft cellularity and evidence of new bone formation. Control sections, i.e., grafts with no cells added, showed an intact trabecular structure with no cells or new bone present (Fig. 6). Grafts treated with cells and Figure 4. Changes in trabecular thickness with the addition of growth factors. OP-1 showed large numbers of cells surrounded by hematoxylin-stained extracellular matrix (ECM). Clusters of cells surrounded by ECM, often formed new woven bone bridges connecting existing trabeculae and other cell clusters. Grafts treated with bfgf demonstrated high cellularity, however the cells were surrounded by eosinophilic ECM with less staining for new bone and very few bridging connections compared to the OP-1-treated samples. Grafts treated with OP-1 and bfgf had a similar appearance to the OP-1-treated slides, but with less hematoxylin-stained new bone. DISCUSSION Morsellized cancellous allograft bone is commonly used in the reconstruction of bone defects, however incorporation may be delayed, restricting ambulation, prolonging protected weight-bearing, and compromising function. Experimental models to test the efficacy of specific interventions to hasten the incorporation of allograft bone would be valuable. The current model showed that bone marrow-derived OPCs will adhere to, and produce new bone on a corticocancellous allograft in vitro when cultured in osteogenic medium. At 5- and 9-week harvests, all grafts showed an increase in both bone volume and trabecular thickness when cultured with OPCs in osteogenic medium. The trabecular number demonstrated no Figure 3. Percent increase in trabecular thickness over time. Figure 5. Changes in bone volume divided by graft volume with the addition of growth factors.

4 OPCS CULTURED ON CORTICOCANCELLOUS ALLOGRAFT BONE 1663 Figure 6. Photomicrographs of the different treatment groups with allograft bone. Decalcified hematoxylin and eosin stained sections of murine allograft bone osteoprogenitor cells the growth factors OP-1 (osetogenic protein-1, bone morphogenetic protein-7) and FGF-2 (basic fibroblast growth factor, fibroblast growth factor-2). (A) Allograft with no cells; (B) allograft with osteoprogenitor cells only; (C) allograft with osteoprogenitor cells þ OP-1; (D) allograft with osteoprogenitor cells þ bfgf; (E) allograft with osteoprogenitor cells þ OP-1 and bfgf. (A) Shows the control group with the existing allograft trabeculae with no cells present. (B) Shows the presence of many cells in the interstices of the allograft trabecuale with minimal bridging bone as seen in the OP-1 treated group (C). (D) Shows the bfgf treated group, which demonstrates high cellularity and minimal new bone formation, similar to (B). (E), OP-1 with bfgf, shows cell clusters with some new bone formation, but less than in the OP-1 group. Original magnification: left, 100; right, 200. significant differences over time. This was expected, as the seeded cells were not predicted to increase the number of trabecular spicules in the static, in vitro environment in this model. These data show that the trabecular network provided by the corticocancellous allograft serves as a conductive surface for new bone deposition by bone marrow-derived OPCs. The addition of growth factors had a variable effect on net bone formation in this model, possibly due to the amounts or mode of delivery, the timing of delivery, or other in vitro conditions. It is well established that the addition of growth factors to biological systems often requires super-physiological levels to achieve the desired result. Further investigations examining various dosages and time of delivery of exogenous growth factors are therefore indicated. In addition, the use of bone marrow harvest for OPCs provides osteogenic precursors, but also provides a mixed population of cells that include other cell types that may alter the cellular responses of OPCs. The addition of OP-1 has been previously shown to increase osteoblastic bone production However, in the present model, when bone marrow-derived MSCs were seeded onto allograft bone in vitro, OP-1 increased trabecular thickness of the allografts, but failed to show an increase in total BV/GV. This may be due to increased matrix production, leading to an increase in trabecular thickness, or stimulation of osteoclastmediated bone resorption leading to an overall decrease in total bone volume. Monocyte/macrophages are found in the bone marrow aspirates used in the OPC harvest, providing the potential for osteoclast-like cellular activity. Additionally, only one concentration of OP-1 was added daily to the culture medium. Other studies have used doses between ng/ml, showing a concentrationdependent increase in osteogenic activity. 15 Previous studies have shown that bfgf can induce bone formation in vitro 9 and in vivo, 16 while others have shown that bfgf has an inhibitory effect on osteoblastic differentiation. 14,17 Using the present model, bfgf did not increase bone production, and furthermore, it decreased the stimulatory effects of OP-1 when the two growth factors were added simultaneously. Given bfgf s possible role as an inhibitor of osteoblastic differentiation, future studies to determine the optimal temporal relationships for the addition of growth factors such as bfgf and OP-1 using this model are necessary. The role of mct in this study was useful in two respects. First, it was effective in determining the allograft bone volume and trabecular characteristics before in vitro culture. Second, it allowed the evaluation of small changes in bone content and trabecular structure when seeded and cultured with MSCs. Although the use of mct in this project was ideal in some regards, it does have important limitations. Since mct can only detect radio-dense material, it is possible that new, uncalcified osteoid matrix produced by osteoblasts was overlooked. Thus, we consider the total bone values as low estimates of the new bone formed. Qualitative evaluation of histological sections was performed to support the quantitative data obtained by mct. These sections showed clear differences in the amount of bone formation between treatment groups, which appeared to correlate with the data obtained by mct.

5 1664 NELSON ET AL. This study demonstrates that addition of murine bone marrow-derived MSCs are capable of producing new bone on murine allogeneic corticocancellous bone discs. The model is unique and can facilitate an understanding of the biological mechanisms by which osteoprogenitors undergo proliferation, differentiation, and maturation to form new bone. These data also suggest that the addition exogenous growth factors to this cell-seeded allograft model can increase the activity of the OPCs and bone production onto a corticocancellous scaffold. Such an augmented allograft bone disc has the potential to provide the three key properties of autograft bone: osteoconductive scaffold, osteogenic OPCs, and osteoinductive growth factors, without the limitations and morbidities associated with the use of autografts. The techniques used in this study have the potential to accelerate allograft incorporation in the treatment of bone defects in humans. This in vitro model is versatile, simple, cost-effective, and translational. However, it must be remembered that this is an in vitro model that does not simulate all the mechanical, biological, and other stimuli normally subjected to bone cells and allograft. The model is also limited by the number of allograft bone discs with similar cancellous structure that can be harvested from each mouse (four in total: one distal femoral disc and one proximal tibial disc from each limb), the number of progenitor cells that can be harvested from the long bones of the mice, and any subsequent biologics that need to be added at specific time points. For the above experiments, approximately 10 mice were required for the allograft discs including all controls, and 8 mice were required for cell harvesting. The cell cultures are somewhat labor intensive, but an added advantage is the fact that the culture supernatants can also be analyzed for cellular byproducts. Thus, the model can serve as a reasonably high throughput method to evaluate the dose- and time-related effects of multiple growth factors and other molecules on osteogenesis of bone marrow-derived OPCs cultured with allograft bone. Furthermore, cells and materials from knockouts and other genetically altered mice will facilitate molecular studies on the mechanisms of bone formation on allograft bone, allowing for the evaluation of these variables and the effects of multiple growth factors and small molecules on OPCs seeded onto allograft bone in vitro. ACKNOWLEDGMENTS This study was supported by funding from the Musculoskeletal Transplant Foundation, The Stanford University Medical Scholars Program, NIH AR45989, the Bone and Joint Center of the Palo Alto Veterans Administration Health Care Center, and the Robert and Mary Ellenburg Chair in Surgery at Stanford University. REFERENCES 1. Betz RR Limitations of autograft and allograft: new synthetic solutions. Orthopedics 25(Suppl 5):s561 s Arrington ED, Smith WJ, Chambers HG, et al Complications of iliac crest bone graft harvesting. Clin Orthop Relat Res 329: Summers BN, Eisenstein SM Donor site pain from the ilium. A. complication of lumbar spine fusion. J Bone Joint Surg [Br] 71: Skaggs DL, Samuelson MA, Hale JM, et al Complications of posterior iliac crest bone grafting in spine surgery in children. Spine 25: Tomford WW, Mankin HJ Bone banking. Update on methods and materials. Orthop Clin North Am 30: Gazdag AR, Lane JM, Glaser D, et al Alternatives to autogenous bone graft: efficacy and indications. J Am Acad Orthop Surg 3: Ehrler DM, Vaccaro AR The use of allograft bone in lumbar spine surgery. Clin Orthop Relat Res 371: Jaiswal N, Haynesworth SE, Caplan AI, et al Osteogenic differentiation of purified, culture-expanded human mesenchymal stem cells in vitro. J Cell Biochem 64: Hanada K, Dennis JE, Caplan AI Stimulatory effects of basic fibroblast growth factor and bone morphogenetic protein-2 on osteogenic differentiation of rat bone marrowderived mesenchymal stem cells. J Bone Miner Res 12: Meirelles Lda S, Nardi NB Murine marrow-derived mesenchymal stem cell: isolation, in vitro expansion, and characterization. Br J Haematol 123: Sun S, Guo Z, Xiao X, et al Isolation of mouse marrow mesenchymal progenitors by a novel and reliable method. Stem Cells 21: Goodman SB, Song Y, Yoo JY, et al Local infusion of FGF-2 enhances bone ingrowth in rabbit chambers in the presence of polyethylene particles. J Biomed Mater Res A 65: Davis J, Tucci M, Franklin L, et al The effects of growth factors on the production of osteopontin and osteocalcin. Biomed Sci Instrum 42: Chaudhary LR, Hofmeister AM, Hruska KA Differential growth factor control of bone formation through osteoprogenitor differentiation. Bone 34: Yeh LC, Tsai AD, Lee JC Osteogenic protein-1 (OP-1, BMP-7) induces osteoblastic cell differentiation of the pluripotent mesenchymal cell line C2C12. J Cell Biochem 87: Park MS, Kim SS, Cho SW, et al Enhancement of the osteogenic efficacy of osteoblast transplantation by the sustained delivery of basic fibroblast growth factor. J Biomed Mater Res B Appl Biomater 79: Quarto N, Longaker MT FGF-2 inhibits osteogenesis in mouse adipose tissue-derived stromal cells and sustains their proliferative and osteogenic potential state. Tissue Eng 12: