Rats
Ninety-six male F344/Nslc (Fisher) rats were obtained from Shimizu Laboratory Supplies (Kyoto, Japan). These Fisher rats, which were thirteen weeks of age and weighed 220 to 240 g, were used as recipients or donors. Two syngeneic transgenic male rats expressing green fluorescence protein (GFP) were obtained from SLC (Shizuoka, Japan). These GFP rats were thirteen weeks of age and were used as donors. Nine Fisher rats, which did not undergo an operation, were the source of bone marrow cells and mesenchymal stem cells as well as normal Achilles tendons. The remaining eighty-seven Fisher rats were used for Achilles tendon healing experiments, which involved biomechanical testing (seventy-two tendons), histological and immunohistochemical analysis (thirty-six tendons), histological analysis with fluorescence microscopy (twelve tendons), and reverse transcription-polymerase chain reaction (RT-PCR) (eighteen tendons). All rats were kept in single cages and fed a standard diet. The study protocol was approved by the Animal Experimentation Committee, Kansai Medical University (Osaka, Japan).
Preparation of Bone Marrow Cells and Mesenchymal Stem Cells
To prepare bone marrow cells from the Fisher and GFP rats, bone marrow was flushed from the femora and tibiae with use of a needle and syringe (3 mL; Nipro, Osaka, Japan) filled with Dulbecco modified Eagle medium (DMEM; Sigma-Aldrich, St. Louis, Missouri). A single cell suspension was generated by aspirating the bone marrow back through the syringe. After isolation, cell preparations were washed with DMEM to remove small pieces of bone and debris.
Mesenchymal stem cells were obtained from the femora and tibiae of Fisher rats, with use of a procedure modified from that of Chong et al.23. In brief, the bone marrow was mixed with DMEM containing 1% penicillin-streptomycin (Sigma-Aldrich) and 10% fetal bovine serum. Samples were washed and were centrifuged at 2000 rpm for five minutes. The resulting cell pellet was resuspended with medium and then plated onto the culture flasks. Cells were grown in 5% CO2 at 37°C for five days. Once confluent, the adherent cells were detached and serially subcultured. Second-passage cells were used for implantation. Cells isolated with this technique have a fibroblast-like appearance and have been shown to be capable of multipotent differentiation23.
Protocol
Rats were anesthetized with 5% Fluothane (halothane; Takeda Pharmaceutical Company Limited, Osaka, Japan) in an anesthetic chamber and then with 3% Fluothane in a mask. With use of aseptic technique, a horizontal skin incision was made directly over each Achilles tendon. A complete transverse incision was made with a surgical blade 7 mm from the calcaneal insertion of each Achilles tendon (Fig. 1, A). The skin only was sutured with 4-0 monofilament nylon, and both ankles were then placed in casts. Next, we injected 50 µL of iohexol (Omnipaque 240; Daiichi-Sankyo Company Limited, Tokyo, Japan) around the Achilles tendon to confirm that this volume was sufficient to diffuse completely around the tendon (Fig. 1, B). Fifty microliters of DMEM containing the donor bone marrow cells (3 × 106) or mesenchymal stem cells (3 × 106) were then injected around the Achilles tendon with use of a syringe (1 mL; Nipro) (Fig. 1, C). The rats were divided into three groups: (1) the BMC group (both Achilles tendons in the rat were injected with bone marrow cells), (2) the MSC group (both Achilles tendons in the rat were injected with mesenchymal stem cells), and (3) the non-treated group (both Achilles tendons in the rat were incised only). Both Achilles tendons in each rat were included in the same group (i.e., BMC, MSC, or untreated). One side of each rat was used for biomechanical evaluation, and the other side was used for histological evaluation. The sides were randomly selected.
Macroscopic Assessment
To examine the tendon healing process, the treated rats were killed and the surgical exposures were evaluated at fourteen and twenty-eight days after incision.
Biomechanical Testing
The Achilles tendon between the calcaneus and the musculotendinous junction was harvested at seven, fourteen, twenty-eight, and thirty-five days after incision. Six Achilles tendons in each group were used at each time point. Processing was performed as previously described30-32. The proximal and distal ends of the Achilles tendon were placed securely in serrated grips and were mounted onto a mechanical testing machine (Autograph AG-X; Shimadzu, Kyoto, Japan). The load cell capacity of this system was 100 N, and the Achilles tendon was pulled at a constant speed of 10 mm/min until failure. Data collected from Trapezium2, version 2.05, software (Shimadzu) were used to calculate the ultimate failure load of each Achilles tendon.
Tissue Preparation
The treated rats were killed, and the Achilles tendon between the calcaneus and the musculotendinous junction was harvested at seven or twenty-eight days after incision. The Achilles tendon was fixed for one week in 10% formalin. The specimens were embedded in paraffin, cut into 4-µm sections, and stained with hematoxylin and eosin (H & E) or Masson trichrome22,24,33.
Collagen fibers are considered to be important in the tendon healing process. Type-III collagen is the first collagen to be laid down during healing. Type-I collagen is responsible for the mechanical strength of the tendon34,35. To examine healing of the transversely incised Achilles tendon, we used immunohistochemistry to analyze the distribution of types-I and III collagen. In brief, sections were stained with use of anti-human type-I or III collagen goat polyclonal antibodies (SouthernBiotech, Birmingham, Alabama) at a dilution of 1:400, as reported previously36. After application of rabbit polyclonal anti-goat immunoglobulins (Dako, Tokyo, Japan) and washing with 0.05 M phosphate buffer (pH 7.6), sections were incubated in peroxidase-conjugated streptavidin (peroxidase labeled, number KO675; Dako) for ten minutes at room temperature with use of the LSAB (labeled streptavidin-biotin) method37. After washing, color was developed with use of 3-3'-diaminobenzidine (DAB). Mayer hematoxylin was used as a nuclear counterstain36,37. The glass slides were then examined with standard light microscopy, and photomicrographs were obtained at ×400 magnification.
Specimens injected with bone marrow cells from the GFP rats were fixed in isopentane, quick frozen in liquid nitrogen, and mounted for cryosectioning. The specimens were cut into 8 to 10-µm sections and mounted under comparable tension on glass slides. They were then examined with fluorescence microscopy for the presence of GFP-positive cells, and photomicrographs were obtained at ×400 magnification24.
Histopathological Analysis
Histopathological analysis of longitudinal sections of the Achilles tendon were analyzed with use of the semiquantitative Bonar histopathological scale38. Briefly, this scale consists of four features, each graded as 0, 1, 2, or 3: (1) tenocyte (spindle cell) morphology and proliferation, (2) the presence or absence of ground substance, (3) collagen bundle characteristics, and (4) vascularity. Each slide was examined by three histologists (N.O., T.K., and K.O.).
Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
We examined expression of TGF-ß, VEGF, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) at four days after incision. Levels of mRNA for all genes were determined with use of specific primers (Invitrogen Japan K.K., Tokyo, Japan). The sequences were: TGF-ß, 5'-TGTACGGCAGTGGCTGAAC-3' and 5'-ATTCATGTTGGACAACTGCTC-3'; VEGF, 5'-GTCTACCAGCGCAGCTATTG-3' and 5'-ACAGTGAACGCTCCAGGAT-3'; and GAPDH, 5'-GGGTGGTGCCAAAAGGGT-3' and 5'-GGAGTTGCTGTTGAAGTCACA-3'. TGF-ß mRNA was amplified with use of pairs of primers at 94°C for one minute, 60°C for one minute, and 72°C for one minute × 35 cycles, with a final extension at 72°C for ten minutes. VEGF mRNA was amplified at 94°C for one minute, 58°C for one minute, and 72°C for one minute × 40 cycles, with a final extension at 72°C for ten minutes. GAPDH mRNA was amplified at 94°C for one minute, 58°C for one minute, and 72°C for one minute × 30 cycles, with a final extension at 72°C for ten minutes. PCR products were electrophoresed on a 1% agarose gel (Invitrogen Japan), stained with ethidium bromide (0.5 g/mL), and visualized on an ultraviolet transilluminator (ATTO, Tokyo, Japan).
Statistical Analysis
One-way analysis of variance revealed significant intergroup differences in the ultimate failure load, the histopathological grade, and levels of mRNA among the BMC group, the MSC group, and the non-treated group (p < 0.05). Post hoc comparisons were performed with the unpaired Student t test to evaluate the difference between two groups. For multiple comparisons, a Bonferroni correction-adjusted level of significance was used at level 0.05/number of t test (p < 0.016). The Dunnett test was used to evaluate the difference in the ultimate failure load between each group and the normal Achilles tendon. Differences were considered significant when the p value was < 0.05. All data are expressed as mean and standard deviation.
Source of Funding
This study was supported by a grant from Strategic Research Base Development Program for Private Universities: matching fund subsidy from MEXT (Ministry of Education, Culture, Sports, Science and Technology), and the Brain Medical Research Center Project of the Ministry of Education. The funds were used to pay for all aspects of the study, including the animals, chemicals, and supplies.
Macroscopic Assessment
To examine the effects of bone-marrow-cell transplantation on the incised Achilles tendon, we evaluated the appearance of the tendons. The defect in the Achilles tendon was hardly detectable in the MSC group or the BMC group, and the appearances in these two groups were similar at fourteen days after incision. In contrast, the Achilles tendon defect was clearly detectable in the non-treated group at fourteen days after incision (Fig. 2). However, no macroscopic differences could be seen among the three groups at twenty-eight days after incision (data not shown).
Biomechanical Testing
The ultimate failure load in the BMC group was significantly greater than that in the non-treated group or the MSC group at seven days after incision (3.8 N vs. 0.9 N or 2.1 N, p < 0.016) and at fourteen days after incision (10.2 N vs. 6.1 N or 8.2 N, p < 0.016). At twenty-eight days after incision, the ultimate failure load in the BMC group was significantly greater than that in the non-treated group (33.8 N vs. 25.6 N, p < 0.01), and the tendons in the BMC group had reached almost normal strength (34.8 N). In the MSC group, the ultimate failure load was significantly greater than that in the non-treated group at seven days after incision (p < 0.01) and at fourteen days after incision (p < 0.016), but at twenty-eight days the difference was not significant although the ultimate failure load in the MSC group (29.3 N) was higher than that in the non-treated group (25.6 N). At thirty-five days after incision, the ultimate failure load in all groups did not differ significantly from that of normal Achilles tendon (Fig. 3). These findings indicate that injection of bone marrow cells around the Achilles tendon quickly improved biomechanical strength in the early stages of healing and for up to twenty-eight days after incision.
Histological Findings
At seven days after incision, the tenocyte scores in the non-treated group, the MSC group, and the BMC group were 2.8 ± 0.4 (mean and standard deviation), 2.2 ± 0.4, and 2.0 ± 0.6, respectively. Macroscopic examination showed more spindle cells in both the MSC group and the BMC group than in the non-treated group (Fig. 4, A, B, and C); however, there was no significant difference between the groups with regard to the tenocyte scores. At twenty-eight days after incision, the tenocyte scores in the non-treated group, the MSC group, and the BMC group were 2.2 ± 0.8, 1.0 ± 0.6, and 0.5 ± 0.5, respectively. In other words, there were significantly more mature spindle cells in both the MSC group and the BMC group than in the non-treated group (p < 0.016) (Fig. 4, D, E, and F).
At seven days after incision, the collagen scores in the non-treated group, the MSC group, and the BMC group were 2.8 ± 0.4, 2.3 ± 0.5, and 2.2 ± 0.4, respectively. Macroscopic findings indicated a less marked separation of fibers with complete loss of architecture in the MSC group and the BMC group than in the non-treated group. However, the collagen scores did not differ significantly among the three groups (Fig. 4, G, H, and I). At twenty-eight days after incision, the collagen scores in the non-treated group, the MSC group, and the BMC group were 2.2 ± 0.4, 1.0 ± 0.6, and 0.8 ± 0.8, respectively. This indicates that there was significantly more collagen arranged in tightly cohesive, well-demarcated bundles in the MSC group and the BMC group than in the non-treated group (p < 0.016) (Fig. 4, J, K, and L).
At seven days after incision, the vascularity scores in the non-treated group, the MSC group, and the BMC group were 1.3 ± 0.5, 2.5 ± 0.5, and 2.7 ± 0.5, respectively. Vascularity in the MSC group and the BMC group was significantly increased compared with that in the non-treated group (p < 0.016) (Fig. 4, A, B, and C). At twenty-eight days after incision, vascularity was hardly observable and the mean scores for vascularity did not differ significantly among the three groups (Fig. 4, D, E, and F).
Immunohistochemical Findings
Staining for type-III collagen showed that the spindle cells and the collagen fibers were more deeply stained in the MSC group and the BMC group than in the non-treated group at seven days after incision (Fig. 5, A, B, and C). However, the spindle cells and collagen fiber bundles were more weakly stained in all groups at twenty-eight days after incision than at seven days (Fig. 5, D, E, and F).
Staining for type-I collagen showed weak staining of spindle cells and collagen fibers in all groups at seven days after incision (Fig. 5, G, H, and I), but they were more strongly stained in all groups at twenty-eight days after incision than at seven days (Fig. 5, J, K, and L).
Histological Findings Under Fluorescence Microscopy
To examine the effects of donor-derived bone marrow cells injected around the incised Achilles tendon, we analyzed frozen sections from animals treated with bone marrow cells from GFP rats. Donor-derived GFP-positive cells were detectable around the incised region not only at seven days but also at twenty-eight days after incision.
Electrophoretic Analyses of RT-PCR Products
To investigate the growth factors produced around the incised Achilles tendon, we examined TGF-ß and VEGF expression with RT-PCR at four days after incision. Expression of TGF-ß in the BMC group was significantly increased compared with that in the MSC group or the non-treated group (1.6 vs. 1.3 or 0.6, p < 0.01) (Fig. 6, A). In addition, expression of VEGF in the BMC group was significantly increased compared with that in the MSC group or the non-treated group (1.7 vs. 1.1 or 0.9, p < 0.01) (Fig. 6, B).
When Achilles tendon rupture occurs, blood vessels are also ruptured and blood clots are formed around the injured tendon. These clots contain not only platelets but also several other cell types, which immediately release a variety of growth factors including insulin-like growth factor-1 (IGF-1), TGF-ß, VEGF, platelet-derived growth factor (PDGF), and basic fibroblast growth factor (bFGF)27,28. Human tendon healing is classified into five phases: the immediate postinjury phase, the inflammatory phase, the proliferation phase, the reparative phase, and the remodeling phase27. Healing depends on both intrinsic and extrinsic processes. Intrinsic healing occurs within the tendon itself as a result of the activity of intrinsic fibroblasts and an increased intratendinous blood supply from the synovium and the osseous insertion. Extrinsic healing involves factors outside the tendon such as ingrowth of extrinsic peripheral fibroblasts and extratendinous vascular invasion16,39. Fibroblasts are important in the tendon healing process. During the inflammatory phase, intrinsic fibroblasts proliferate and extrinsic fibroblasts migrate into the defect. These fibroblasts synthesize new extracellular matrix, consisting largely of collagens and glycosaminoglycans27,39.
Mesenchymal stem cells have the capacity to differentiate into several cell types including osteocytes, chondrocytes, myotubes, stromal cells, fibroblasts, and adipocytes40. Furthermore, mesenchymal stem cells produce growth factors and cytokines, and they also have immunomodulatory and anti-inflammatory effects26,41. Because of this, mesenchymal-stem-cell transplantation therapy has recently begun to be used to induce tissue healing, including tendon healing, tendon-to-bone healing, and cartilage repair, in animal models; the mesenchymal stem cells contribute to healing not only by direct differentiation but also by the production and release of paracrine factors such as growth factors and cytokines23,24,42. Hematopoietic stem cells have the capacity to differentiate into hematolymphoid cells such as lymphocytes, macrophages, granulocytes, eosinophils, erythroblasts, erythrocytes, and megakaryocytes. These cells secrete several growth factors29. Both hematopoietic stem cells and mesenchymal stem cells exist within the whole bone-marrow-cell fraction. Therefore, we hypothesized that, since both mesenchymal stem cells and hematopoietic stem cells are necessary for accelerating Achilles tendon healing, transplanting the whole bone-marrow-cell fraction into the incised tendon would be more effective than transplantation of mesenchymal stem cells alone. We found that transplanted GFP-positive cells could be detected around the incised region not only at seven days but also at twenty-eight days after incision. These findings suggest that the donor-derived cells were able to migrate and differentiate into fibroblasts, thereby accelerating the tendon healing process. In the MSC group, the ultimate failure load at seven and fourteen days after incision was significantly higher than that in the non-treated group, although the ultimate failure load at twenty-eight days after incision was not significantly greater than that in the non-treated group. In contrast, in the BMC group, the ultimate failure load at twenty-eight days after incision was significantly greater than that in the non-treated group (Fig. 3). These findings indicate that bone-marrow-cell transplantation therapy is more effective than transplantation of mesenchymal stem cells alone in achieving Achilles tendon healing in this rat model.
Several growth factors are contained within the clots formed around the injured tendon27,28,43. Growth factors such as IGF-1, TGF-ß, VEGF, PDGF, and bFGF play key roles during multiple stages of the tendon healing process16,27. In particular, TGF-ß and VEGF are important for inducing healing27. TGF-ß regulates various biological processes including cell proliferation, migration, differentiation, apoptosis, and extracellular matrix deposition27. In addition, TGF-ß modulates proteoglycan deposition and stimulates the production of collagens by fibroblasts44. VEGF is required for the formation of the initial vascular plexus in early granulation tissue development27. VEGF is also a powerful stimulator of angiogenesis and is involved in multiple stages of the tendon healing process, especially during the proliferative and remodeling phases27. Zhang et al.16 reported that the local injection of exogenous VEGF during healing can significantly improve tensile strength in the rat Achilles tendon because VEGF can induce the production of other growth factors and thus improve fibroblast proliferation. Notably, there have been reports that both TGF-ß and VEGF increase tendon healing and that they can enhance each other's production26,45,46. In our study, analysis with RT-PCR at four days after incision showed that expression of TGF-ß and VEGF in the BMC group and the MSC group was significantly increased compared with that in the non-treated group. Furthermore, expression of TGF-ß and VEGF in the BMC group was significantly greater than that in the MSC group (Fig. 6).
The Achilles tendon is made up of large amounts of type-I collagen together with small amounts of other collagens such as type III. Type-I collagen is a mature form of collagen and is mainly responsible for the mechanical strength of the tendon. On the other hand, type-III collagen, which is thinner and more extensible than type-I collagen, is not a major collagen in normal tendon. However, type-III collagen is important during the earliest stage of tendon healing because it can rapidly form cross-links and stabilize the precarious repair site34,35. In the present study, the spindle cells and the newly formed type-III-collagen-positive fibers were much more prevalent in the BMC group than in the MSC group or the non-treated group during the reparative phase of the tendon healing process (at seven days after incision; Fig. 5, A, B, and C). Furthermore, the spindle cells and type-I-collagen-positive fibers were much more detectable in the BMC group than in the MSC group or the non-treated group during the remodeling phase (at twenty-eight days after incision; Fig. 5, J, K, and L). These findings indicate that the Achilles tendon healed more rapidly in the BMC group than in the MSC group or the non-treated group.
On the basis of these findings, we postulate that a variety of growth factors, such as TGF-ß and VEGF, that are produced and released by the various cell types and platelets within the clots activate fibroblasts within or outside the tendon itself and induce the synthesis of types-I and III collagen. Whole bone marrow cells contain a variety of cell types in the same way as clots, and mesenchymal stem cells and hematopoietic stem cells in particular play important roles in organ regeneration. Therefore, we hypothesized that mesenchymal stem cells might be more activated when combined with hematopoietic stem cells and that this might further accelerate the healing process over and above that resulting from the transplantation of mesenchymal stem cells alone. In the present study, the levels of TGF-ß and VEGF were significantly increased in the BMC group compared with the levels in the MSC group. Therefore, we suggest that the proliferation of mesenchymal stem cells and their differentiation into fibroblasts might be accelerated by the increased levels of growth factors secreted from hematolymphoid cells, resulting in a more rapid increase in the levels of types-I and III collagen and, finally, more rapid recovery of the ultimate failure load in the BMC group.
In this model, mesenchymal-stem-cell transplantation therapy was effective in accelerating the regeneration of the Achilles tendon following rupture. However, it has a disadvantage in that mesenchymal stem cells must be harvested over a period of a few days from fresh whole bone marrow cells in order to obtain sufficient quantities of purified cultured mesenchymal stem cells for transplantation. In order to use cultured mesenchymal stem cells for human transplantation therapy in the future, it is necessary to develop a culture system that allows mesenchymal stem cells to be obtained safely and quickly. Currently, fresh whole bone marrow cells are obtained from the human ilium and injected systemically for the treatment of such disorders as myeloma, leukemia, and aplastic anemia. Therefore, our method of transplantation therapy using whole bone marrow cells for the treatment of Achilles tendon rupture may provide a more readily available alternative than the use of cultured mesenchymal stem cells for transplantation therapy. Because human Achilles tendon rupture differs from the rat Achilles tendon rupture model, it is unclear whether bone-marrow-cell transplantation therapy will be effective in humans.