The authors state that they have no conflicts of interest.
Bone Healing and Migration of Cord Blood—Derived Stem Cells Into a Critical Size Femoral Defect After Xenotransplantation†
Version of Record online: 23 APR 2007
Copyright © 2007 ASBMR
Journal of Bone and Mineral Research
Volume 22, Issue 8, pages 1224–1233, August 2007
How to Cite
Jäger, M., Degistirici, Ö., Knipper, A., Fischer, J., Sager, M. and Krauspe, R. (2007), Bone Healing and Migration of Cord Blood—Derived Stem Cells Into a Critical Size Femoral Defect After Xenotransplantation. J Bone Miner Res, 22: 1224–1233. doi: 10.1359/jbmr.070414
- Issue online: 4 DEC 2009
- Version of Record online: 23 APR 2007
- Manuscript Accepted: 17 APR 2007
- Manuscript Revised: 5 MAR 2007
- Manuscript Received: 27 NOV 2006
- stem cells;
- cord blood;
- bone healing;
Stem cell and tissue engineering—based therapies have become a promising option to heal bony defects in the future. Human cord blood—derived mesenchymal stem cells were seeded onto a collagen/tricalcium phosphate scaffold and xenotransplanted into critical size femoral defects of 46 nude rats. We found a survival of human cells within the scaffold and surrounding bone/bone marrow up to 4 wk after transplantation and an increased bone healing rate compared with controls without stem cells. This study supports the application of cord blood stem cells for bone regeneration.
Introduction: The treatment of critical size bone defects is still a challenging problem in orthopedics. In this study, the survival, migration, and bone healing promoting potency of cord blood—derived stem cells were elucidated after xenotransplantation into a critical size femoral defect in athymic nude rats.
Materials and Methods: Unrestricted somatic stem cells (USSCs) isolated from human cord blood were tested toward their mesenchymal in vitro potency and cultivated onto a collagen I/III and β-tricalcium phosphate (β-TCP) scaffold. The biomaterial-USSC composite was transplanted into a 4-mm femoral defect of 40 nude rats and stabilized by an external fixator. Twelve animals without USSCs served as controls. Cell survival, migration, and bone formation were evaluated by blood samples, X-rays, and histological and immunocytochemical analysis of different organs within a maximal postoperative follow-up of 10 wk.
Results: Of the 52 nude rats, 46 animals were evaluated (drop-out rate: 11.5%). Human-derived stem cells showed an engraftment within the scaffold and adjacent femur up to 4 wk after xenotransplantation. With further time, the human cells were destroyed by the host organism. We found a significant increase in bone formation in the study group compared with controls. USSC transplantation did not significantly influence blood count or body weight in athymic nude rats. Whereas the collagen I/III scaffold was almost resorbed 10 wk after transplantation, there were still significant amounts of TCP present in transplantation sites at this time.
Conclusions: Human cord blood—derived stem cells showed significant engraftment in bone marrow, survived within a collagen-TCP scaffold up to 4 wk, and increased local bone formation in a nude rat's femoral defect.
The treatment of critical size musculoskeletal defects is still a challenging problem in orthopedics. Besides conventional surgical procedures, the application of growth factors, tissue engineering techniques, and cell-based therapies has become a promising option to support or substitute surgery over the last few years. However, to date, the side effects of immunosuppressive agents do not allow for allogenic stem cell transplantation to treat local musculoskeletal defects.
Cord blood (CB)-derived stem cells have been in clinical use to treat hematological malignancies for years.(1) As shown by several in vitro studies, there is also evidence that human cord blood contains mesenchymal progenitor cells that are capable of differentiating into osteogenic, adipogenic, or chondrogenic lines under defined stimuli.(2,3) Moreover, some authors have reported a lower immunogenic reaction in patients after unrelated or non—HLA-matched allogenic cord blood transplantation compared with bone marrow—derived cells.(4,5) Because of both biological aspects, cord blood may be an attractive source for stem cell—based therapy of musculoskeletal defects and tissue engineering in the future.
Several different types of cord blood—derived mesenchymal stem cells with an osteogenic differentiation potential have been published in the literature.(6–9) One well-defined and characterized CD45-negative subpopulation of human cord blood—derived somatic stem cells is described as the unrestricted somatic stem cell (USSC).(3,10–12) Besides their mesenchymal multipotency, these cells are characterized by an adherent growth, a capability to differentiate both into the hematopoietic and neurogenic lines, a lower degree of telomere shortening in vitro compared with other somatic stem cells, a phenotypic stability, and a persisting multipotency property up to higher passages. Because of its immature nature compared with adult stem cells, some experimental results also indicate that USSCs allow for an engraftment in different animal models. In addition, USSCs produce functionally significant amounts of hematopoiesis-supporting cytokines and are superior to bone marrow—derived mesenchymal stem cells (MSCs) in expansion of CD34+ cells from CB as shown by Kögler et al.(11) Moreover, it was shown that USSCs are not restricted to humans but can also be found in other species.(13)
In this study, the osteogenic in vivo potential and engraftment of CB-derived USSCs after xenotransplantation were investigated.
MATERIALS AND METHODS
After isolation of CB-derived mononuclear from healthy newborns by density gradient centrifugation as described by Kögler et al.,(3) cells were cultivated (5000 cells/cm2) and expanded in DMEM-low glucose media (1 g/liter), supplemented with 30% FCS and 100 nM dexamethasone in 5 vol.%CO2 at 37°C. Colony-forming units were observed within 5 days in vitro. After a confluent monolayer formed, cells were passaged supported by 0.05% trypsin/0.02% EDTA as described previously.(3) According to the Declaration of Helsinki in its present form, the parents of all donors had agreed with written consent before delivery. For the xenotransplantation, we used USSCs isolated from the cord blood samples of seven different donors and five different passages (2 × P4, 2 × P5, 5 × P6, 2 × P7, and 1 × P8, the maximal cultivation time was 8 wk). USSCs were tested in vitro toward their mesenchymal multipotency (osteo-, chondro-, adipogenic) before xenotransplantation using lineage-specific stimulation (data not shown).
A porcine collagen I/III membrane (ACI Maix; Matricel), which has been in clinical use for autologous chondrocyte transplantation for years and has been tested for its osteogenic in vitro potential on human bone marrow cells previously, was used as a scaffold for USSCs.(14) In addition, USSCs were also cultivated onto β-TCP granules [β-Ca3(PO4)2, Ca/P = 1.50, 90% interconnected pore space, particle size: 100 nm, pore size: 1–1000 μm; Vitoss, Fa. Orthovita], which had shown good cytocompatibility properties in vitro.(15) Twelve hours before xenotransplantation, USSCs were detached from culture flasks supported by 0.05% trypsin/0.02% EDTA and resuspended. A total suspension volume of each 100 μl (3 × 107 cells/ml) was pipetted onto a 0.5 × 0.5-cm collagen I/III membrane and also onto 10 mg β-TCP. After an initial incubation period of 2 h at 37°C to allow for cellular adherence, the culture dishes were filled with DMEM-low glucose containing 30% FCS and 10−7 M dexamethasone followed by an overnight incubation at 37°C. As controlled by light microscopy, the human cells adhered to the scaffolds (TCP, collagen sponge) and were able to migrate into the superficial layers of both biocarriers. After in vitro cultivation and before transplantation, the majority of human cells were located at the biomaterial surface and in its superficial layers.
Immediately before surgery, the collagen USSC and TCP USSC transplants were combined by wrapping the collagen membrane around the solid TCP granule and fixed by a 3.0 Ethibond suture under sterile conditions.
Animals and xenotransplantation
All animal experiments were performed in accordance with the local ethical committee (Project G67/2001). For xenotransplantation experiments, we used 52 adult athymic rnu nude rats (Crl:RNU-Rats; Charles River Laboratories) older than 12 wk. Animals were anesthetized by 0.0315 mg Fentanyl/100 mg body weight (Hypnorm; Janssen Pharmaceutica, Beerse, Belgium) intraperitoneally and 0.5 mg diazepam (Valium; Hoffmann-La Roche, Wyhlen, Germany) intramuscularly. Before surgery, the right hind limb was shaved and scrubbed with antiseptic fluid. Using a lateral approach, a critical size femoral defect (CSD) of 4.0 mm size was created on nude rats on the right hind limb and fixed by external fixator as previously described.(16,17) A femoral osteotomy was performed using a high-speed cutter (Moto-Flex, Dremel, Germany) with a 3.2-mm burr head. To reduce local heat and prevent soft tissue from bony debris contamination, the osteotomy was carried out under constant irrigation with 0.9% NaCl solution. The USSC-biomaterial composite was implanted into the osseous defect zone, and the wound was closed.
The animals were kept in type III cages (800 cm2), with 55% humidity at 22 ± 2°C, 12-h illumination (320 lux). The animals were fed ssmiffR/M-Zucht ad libitum. To diminish the infection risk, the drinking water was supplemented with antibiotics: 5 days before surgery, nude rats received a combination of sulfadimidine (50 mg/kg body mass), sulfathiazole (50 mg/kg body mass), and trimethoprim (20 mg/kg body mass). If signs of infection were noted, such as increased white blood count or reduction of body weight, drinking water was supplemented with 0.3 g amoxycillin/1000 ml water. All animals were kept in an incubator (Uni Protect, DIN 40050; Ehret, Emmendingen, Germany; positive pressure of 150 hPa) under S1 conditions. Table 1 shows the number and distribution of animals for the evaluation of bone healing, migration, and survival analysis of human USSCs after xenotransplantation.
Blood samples of 20 μl venous blood under anesthesia and weight of the nude rats were taken twice a week. White (WBC) and red blood count (RBC), hemoglobin (HGB), hematocrit (HCT), platelets (PLT), mean corpuscular volume (MCV), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDV), mean platelet volume (MPV), and differential blood count (lymphocytes, monocytes, granulocytes) served as parameters to detect any systemic influences after xenotransplantation using an abc Animal Blood Counter (ABX Diagnostics, Montpellier, France). Exclusion parameters were a reduction of >20% of total body weight persisting >7 days and signs of deep infection.
To evaluate new bone formation, lateral X-rays of the nude rat's hind limb were performed directly after surgery and 1, 2, 3, 4, 8, and 10 weeks after surgery (settings: 73 kV, 32 ms, Optimus; Philips/Dental film, Kodak ultra-speed DF-49; Eastman Kodak, Rochester, NY, USA). The transplant area was evaluated for bone formation macroscopically and histologically at these times and classified as “solid fusion,” “bone formation within the defect zone,” or “no bone formation within the defect zone.”
Cell survival and migration analysis
Corresponding to the kinetics points used to evaluate bone formation, we analyzed different organs for engraftment of human cells. Immediately after death, fresh tissue specimens of different organs (spleen [apex], liver [right lobe], kidney, heart [left ventricle], lung [apex], thyroid gland, stomach, small intestinal bowel, pancreas, transplant/biomaterial) were taken, frozen in liquid nitrogen, and stored at −80°C. After external fixator and soft tissue removal, the right femur was preserved in formaldehyde and decalcified in EDTA-containing solution over 6 wk (12.11 g TRIS Base, 800 ml aqua dest., 100 g EDTA, 50 ml 1 M NaOH, and 75 g PVP K25, at pH 7.4 and room temperature). All organs were examined macro- and microscopically for signs of infection or tumors. Based on macroscopical examination, biomaterial resorption was valued semiquantitatively by the following score: no resorption, 1; partial resorption (biomaterial structure/geometry intact), 0.75; only traces of the biomaterial visible, 0.25; complete resorption (macroscopically no signs of biomaterial visible), 0. The number of human cells was evaluated by immunohistology. Because the discrimination of the single signals was too poor to receive reliable quantitative results by laser scanning microscopy, the number of human cells was counted optically (see below).
Histology and immunohistology
For immunocytochemical evaluation, we used 10-μm frozen tissue sections (Kryostat; Zeiss), which were fixed onto poly-l-lysine—covered slides. Specimens were fixed in 4% formaldehyde at 4°C for 5 min and rinsed in aqua dest./PBS. The tissue samples were incubated with a monoclonal mouse anti-human nuclei antibody (Chemicon) in PBSTA for 12 h at 4°C to identify the human cells within the different nude rat's organs. This antibody was used because a HLA class I expression was negative in USSCs before transplantation using immunofluorescent microscopy (Figs. 1A and 1B). After rinsing in PBS again, specimens were incubated for 1 h with a second antibody system [avidin-biotin-complex: biotinylated anti-mouse IgG (H+L)/biotinylated anti-rabbit IgG (H+L); Vector Laboratories]. After this, a 3,3-diaminobenzidine reaction was performed for optical visualization (incubation with Streptavidin, Alexa Fluor 635 conjugate, concentration 1:100 in PBSTA; Molecular Probes). The samples were rinsed in PBS and DAPI stained (Vectashield: 1.5 μg DAPI/ml; Vector Laboratories), and the slices were sealed with Depex (Serva).
Formaldehyde-fixed tissue specimens were also stained with H&E using incubation in Mayer's Hämalaun solution (5 min), rinsed in aqua dest. and alcoholic HCl solution, incubated in 1% Eosin-G-solution (2 min), rinsed in aqua dest., and dehydrated in graded alcohols (70%/96%/99%). Finally, the probes were incubated in xylol for 10 min and covered with Depex.
All slides were analyzed using fluorescent and episcopic light microscopy (Axiovert 200; Zeiss) in combination with a computer-supported imaging picture analysis system (Axiovision; Zeiss). A total area of 350 × 270 μm = 945.500 μm2 was evaluated toward human nuclear antigen (HNA)-positive cells and quantified. In addition, the corresponding tissues of an athymic nude that did not undergo surgery served as a negative control for histological and immunohistochemical evaluation.
Table 1 gives an overview of the distribution of the different groups and kinetic points.
The Student's t-test for independent statistical groups and the F-test were used for statistical analysis. p < 0.01 was rated highly statistically significant and p < 0.05 was statistically significant, whereas p > 0.05 showed no significance. The average values (X) and SD served as descriptive parameters.
Because of perioperative complications (two deep wound infections, two significant diarrhea with a reduction of body weight >50 g, and two implant failures), 46 of the original 52 animals were evaluated at follow-up (drop-out rate: 11.5%). A total of 4140 microscopic display images were evaluated toward antigen-positive cells (45 nude rats + 1 animal without surgery as control, 10 organs, 9 images per organ). We found neither macro- nor microscopic signs of tumor formation in the analyzed tissues and organs. Corresponding to a good physical condition postoperatively, the body weight of the nude rats was not influenced negatively by the xenotransplantation and was comparable with controls (Fig. 2).
There were no statistically significant differences between the study and the control groups in RBC, WBC, the number of granulocytes, monocytes, or lymphocytes, MCH, MVC, MPV, or RDW. Table 2 shows the p values and correlations for the different blood count parameters. From day 21, there was a slight increase in WBC in both groups. MCHC showed a significant increase in controls compared with the study group, whereas the MCV in study and control groups decreased after surgery (preoperative mean values: XUSSC = 50.5 ± 3.7 [SD] versus Xcontrol = 49.8 ± 1.7, postoperative values: XUSSC = 46.0 ± 0.7 versus X control = 45.25 ± 0.5, correlation after surgery: 0.94). There were no significant changes in MCH and MPV during follow-up. In the study group, MCHC decreased temporarily from day 21 to day 42 (XUSSC/21 = 40.9 ± 6.7 g/dl, versus XUSSC/42 = 33.9 ± 7.0 g/dl) but increased afterward to XUSSC/70 = 38.9 ± 4.3 g/dl at final follow-up. In contrast, we found a temporary increase in MCHC between days 14 and 28 in controls (Xcontrol/14 = 38.2 ± 3.2 g/dl to Xcontrol/28 = 43.9 ± 5.4 g/dl). The RDW showed a slight but nonsignificant increase postoperatively compared with preoperative values for both groups.
Compared with controls, the rate of new bone formation was significantly increased within the transplantation area after USSCs transplantation as shown by X-rays, macroscopic inspection, and histology (Figs. 3A—3H). No bone healing of the defect was found in controls during the follow-up corresponding to a critical size defect. One week after surgery, there were no signs of bone healing, and besides a demarcation of the collagen I/III carrier, there were no signs of implant resorption in the USSC+ group or in controls. Two weeks after USSC transplantation, one of five nude rats showed interfragmentary bone formation on X-rays and macroscopically; another animal showed the same 3 wk postoperatively (Fig. 3C). Four weeks after surgery, three of five nude rats showed new bone formation, but only one of those had macroscopic healing of the defect corresponding to bony bridging of the critical size gap. Eight weeks after xenotransplantation, another nude rat of five showed bony healing, whereas we found only mild bone formation within the transplant in two rats. A solid fusion in the study group was found in 4 of 10 animals 10 wk after xenotransplantation, whereas an additional 5 animals showed significant bone formation. One nude rat showed no signs of bone formation after USSC transplantation at follow-up. Moreover, we found no correlation of bone healing in regard to the cell passage. There were two bone defect healings each for USSC passages 4 and 6 and one each in passages 5 and 7 (four different donors). There were also differences in kinetics of collagen I/III and TCP resorption between the study and control groups as shown in Figs. 4A and 4B. The progressive bone healing after USSC transplantation was also confirmed by histological evaluation of the former transplant area, showing a remodeling of the biomaterial and increasing deposition of calcified matrix during follow-up (Figs. 3D—3G). In only a few locations within the transplantation area did we detect cells with morphological characteristics of chondrocytes. These cells contacted directly to the newly formed bone and could indicate an intermediate cartilage stage before osteoblastic differentiation.
Two weeks after USSC transplantation, we found a diffuse cellular infiltration within the collagen I/III mesh. After 3 wk, some cells built palisade-like formations along the collagen fibers. These cells were characterized by a cuboid shape with a mean diameter of 25 μm and a central nucleus. Other parts of the implant showed an infiltration with cells with histological characteristics of lymphocytes (small round cells, round nucleus). At later follow-up, we also found signs of calcification of the scaffold. However, 10 wk after USSC transplantation, the collagen I/III carrier was completely resorbed except for a few loose fibers (average diameter, 25 μm) in the study group, whereas a loose fibrous network of the collagen biomaterial was still remaining in control nude rats (Figs. 4C and 4D). In contrast, we found no significant differences between the resorption kinetics for TCP between the study and controls. In both groups, TCP showed an incomplete resorption within the 10-wk follow-up period. In some areas, we found TCP- and collagen scaffold—associated multinuclear cells corresponding to a local osteoclast activation.
Migration and survival of human cells
One week after xenotransplantation, we detected HNA+ cells within the scaffold but in higher numbers in the fragment ends (bone and bone marrow) of the femur next to the transplantation area. Two weeks after transplantation, the number of HNA+ cells increased in both the collagen scaffold and the femur (number of cells/945,500 μm2: femurweek 1 179.3 versus femurweek 2 191.3; biomaterialweek 1 35.8 versus biomaterialweek 2 59.3). At the femur, some HNA+ cells showed colony-forming units (CFUs) that reached a maximum size of about 500 μm in diameter. After the second week after transplantation, there was a significant reduction in HNA+ cells in the femur and scaffold (number of cells/945,500 μm2: femurweek 3: 134.0 versus femurweek 4: 118.2, biomaterialweek 3: 41.7 versus biomaterialweek 4: 21.0). From week 8 after surgery, we did not find HNA+ cells in the transplantation area or in the surrounding femur. There were no signs of cell migration in other organs as evaluated by immunofluorescent microscopy (Figs. 5A—5C). However, corresponding to a decrease in the number of HNA+ cells in the transplant and femur, we observed a continuous increase of cells that contained HNA+ material in the spleen, corresponding to a delayed immunogenic reaction of the recipient (number of cells/945,500 μm2: 10.5 after 1 wk, 18.8 after 2 wk, and a maximum of 57.0 cells after 3 wk after surgery). Within the spleen, there was an association of these cells to the periarteriolar lymphoid sheaths (PALS) and also to the reactive outer zone of the follicles at weeks 3 and 4 (Figs. 5A—5C, 6A, and 6B). After the peak level at week 3 after USSC transplantation, a significant decrease of these cells followed (number of cells/945,500 μm2: 35.6 at 4 wk, 15.4 at 8 wk, and at least 12.5 cells at 10 wk).
In this study, we showed that USSCs can survive in an athymic nude rat after xenotransplantation up to 4 wk. Although it was shown previously that USSCs express only small amounts of HLA,(3) which is a strong stimulative antigen for an immunogenic response, human USSCs were killed with further engraftment by the host organism. One explanation of this phenomenon could be the increased expression of human-specific antigens after USSC transplantation. As shown for embryonic and other stem cell types, a change of the local microenvironment and a nonphysiological niche may increase cellular aging and also induce a progressive differentiation including a change of antigen pattern (e.g., HLA-I).(18–21) In addition, it is evident that there is still some degree of immunocompetence present in athymic rats that might result in an unspecific or even specific xenogenic reaction against human cells.
It was shown by Drapper et al.(18) and LeBlanc et al.(22) that retinoic acid and INF-γ potently induce HLA-I and HLA-II expression. Because USSCs were not supplemented by INF or retinoic acid in vitro and did not express HLA-I as shown by immunocytochemical stainings before xenotransplantation, these or other cytokines may have induced the expression of human-specific antigens with further engraftment of transplanted USSCs in vivo. The progressive loss of hypoimmunogenic properties of USSCs and an increasing production of soluble immunomodulatory factors, including interleukin-10, TGF-β, prostaglandin E2, and others as described for adult MSCs,(23) also would be in correspondence to a delayed immunogenic defense of the host organism as shown by an increasing number of phagocytosed human cells in the spleen. However, other authors have reported that IFN-γ also shows immunosuppressive effects on T and to a lesser degree to B cells by inducing indoleamine 2,3-dioxygenase synthesis in MSCs.(24,25) Because of a lack of T cells in athymic rats, it is unclear if indoleamine 2,3-dioxygenase-related immunosuppressive effects played a role. It is also possible that the majority of the immunogenic response and destruction of human USSCs is based on an interaction between B cells, phagocytes, and natural killer cells (NKC). Erices et al.(26) were some of the first authors to show the hypoimmunogenity of human CB—derived MSCs after xenotransplantation. They injected 5 × 105 human CB MSCs into the tail lateral vein of unconditioned nude mice and detected human DNA in the recipient's bone marrow and other tissues such as cardiac muscle, teeth, and spleen 5 mo after transplantation. Although Erices et al.(26) emphasized a strong migration and surviving potential of xenotransplanted human CB MSCs, three rodents may be too small for general conclusions. Moreover, the detection of β-globulin encoding human DNA in different host organs does not prove the cell survival of human cells within the maximal follow-up, because human DNA could also derive from dead human cells that may be phagocytosed by the host's immunogenic cells. However, Erices et al. showed by recultivation of engrafted human CB MSCs taken from nude mice bone marrow after xenotransplantation that some human cells survived and were able to differentiate into osteoblasts under lineage-specific stimulation.
The positive influence of MSCs on the engraftment potential of different progenitor cell types and its immunosuppressive effects were shown in several studies.(27–29) Maitra et al.(30) reported that 8 of 10 mice co-transplanted with umbilical cord blood cells and MSCs showed persistent engraftment. In another study, in't Anker et al.(31) co-transplanted hematopoietic (CD34+) cord blood—derived stem cells together with fetal mesenchymal (CD45−) stem cells into irradiated NOD/SCID mice and showed supportive effects to engraftment and survival by MSCs. Here, primary MSCs but not culture-expanded MSCs were detected in recipient mice, suggesting that the primary cells were able to home and that this capacity was lost after expansion. Therefore, another reason why we found an engraftment of USSCs only within the scaffold and the bone marrow/femur close to the transplantation area may be also the use of in vitro expanded USSCs (passages 4–7). However, the mechanism behind the enhanced engraftment of MSCs in vivo is poorly understood.
Grinnemo et al.(32) transplanted bone marrow—derived MSCs into ischemic rat myocardium and found a high engraftment rate corresponding to an absence of xenoreactivity compared with immunocompetent SpragueDawley rats. The good engraftment of human MSCs after xenotransplantation corresponds with data by other authors.(33–37) There is also evidence that, in autologous tissue engineering concepts, transplanted mesenchymal cells die within further time after transplantation.(38)
Because the body weight of the animals in our study was not significant affected by USSC transplantation and there were no significant differences between study groups and controls in blood count, we can assume that there was not a severe inflammatory or immunogenic systemic response after USSC xenotransplantation in nude rats.
In this study, we showed significantly increased bone formation after USSC xenotransplantation into a critical-sized femoral defect. This corresponds to the in vitro data of Chang et al.(6) who showed that CB-derived MSCs have a stronger osteogenic potential than bone marrow—derived MSCs. However there is a lack of data about the osteogenic in vivo potential of CB-derived stem cells in the literature. In contrast, several in vivo studies indicate the high bone regenerative potential of bone marrow—derived stem cells.(39–43)
In this study, we showed significantly increased bone formation after USSC xenotransplantation into a critical size femoral defect. A treatment with unseeded scaffolds as a control arm neither showed solid bone fusion nor callus formation within the transplantation area. Interestingly, immunohistochemistry with human nuclear antigen revealed the presence of human stem cells only detectable until 4 weeks after implantation. These data might indicate that it is not the human stem cells that actively participate in repair tissue formation. Our results argue more for human stem cells as the communication medium to initiate a cell signaling cascade, which results in the recruitment of host-derived stem cells. Similar data have been presented by Eyckmans and Luyten(44) Long-term results within our study at 10 wk after implantation showed 4 of 10 defects had solid bone fusion and an additional 5 of 10 defects had clear callus formation in the center of the transplantation area.
In this study, we did not use any cytokine stimulation or transfected progenitor cells as described by Peterson et al.(45) to promote the bone healing potency of the transplanted cells. However, we did not show that human osteoblasts or other human cells derived from USSCs were responsible for the in vivo bone regeneration after xenotransplantation. It would be also possible that cytokines released by USSCs or a local inflammatory response at the transplantation site may have stimulated the rat's cells toward an osteogenic differentiation. Based on our data, we can assume that human USSCs mainly provided growth factors to the site, which enabled rat skeletal cells start forming bone.
Considering our own experiences and recent papers from the literature that indicate the good cytocompatibility and osteogenic promoting potency in MSCs, we used a collagen I/III mesh and highly porous β-TCP as a scaffold for USSC transplantation.(14,15,46) In addition, several other authors suggest co-administration of growth factors (e.g., BMP-2 or BMP-9) to increase the osteogenic potential of transplanted MSCs.(47–52) Dayoub et al.(48) infected the musculature of nude rats with bone marrow—derived hMSCs transduced with recombinant adenovirus (ad-BMP-9) and found ectopic bone formation up to 84 days in contrast to noninfected controls. Corresponding to our results, none of the recipients showed any clinical evidence of systemic toxicity. In another study, Peterson et al.(45) used BMP-2—infected progenitor cells isolated from adipose tissue and transplanted them into a critical size femoral defect in a nude rat model. Within 8 wk postoperatively, they showed bone healing in 11 of 12 animals. Similar to our study, they used a collagen/ceramic carrier as cellular scaffold.
In conclusion, human CB-derived stem cells promote local bone regeneration after xenotransplantation into a critical-sized femoral defect in an athymic nude rat. There is evidence that these cells survive within the transplantation area rat for up to 4 wk postoperatively.
The authors thank S Lensing-Höhn, H Denck (Orthopaedic Research Lab, Duesseldorf), I Schrey (Animal Research Lab, Duesseldorf), and P Wernet and G Kögler (Institute of Transplantation Diagnostic and Cell Therapy, Duesseldorf), as well as Matricel GmbH/Herzogenrath for support. Parts of this study were funded by the Wirtschaftsministerium NRW, TPW, Germany.
- 472004 Cranial bone defect healing is accelerated by mesenchymal stem cells induced by coadministration of bone morphogenetic protein-2 and basic fibroblast growth factor. Wound Repair Regen 12: 252–259., , , ,