Cell Transplantation

Embryonic stem (ES) cells are considered to be a potential tool for repairing articular cartilage defects, but so far it has been impossible to cause these cells to differentiate into chondrocytes exclusively, either in vivo or in vitro. To explore a potential new cell source of cell transplantation for articular cartilage defects, we transplanted ES cells into articular cartilage defects in immunosuppressed rats. ES cells (AB2.2 or CCE cells) were transplanted into articular cartilage defects in the patellar groove of immunosuppressed rats treated with cyclosporine. The cells were histologically observed until 8 weeks after transplantation. To determine whether the repair tissue in the defect in the AB2.2-transplanted group was derived from the transplanted cells, the neomycin-resistant gene, which had been transfected into AB2.2 cells but does not exist in rat cells, was used for detection. The cells produced cartilage, resulting in repair of the defects from 4 weeks until 8 weeks after the transplantation without forming any teratomas. The neomycin-resistant gene was detected in every sample, demonstrating that the repair tissue in the AB2.2-transplanted group was derived from the transplanted AB2.2 cells. The environment of osteochondral defects is chondrogenic for ES cells. ES cells may thus be a potential tool for repairing articular cartilage defects.

To clarify whether the mesenchymal cells derived from human placenta were available for bone regeneration, we investigated the effects of osteogenic induction on mesenchymal cells of fetal and maternal parts of the placenta. The osteogenic-induced mineralization in both types of cells was measured by von Kossa staining, and the calcium concentration and the expression of osteogenic markers were assayed by RT-PCR. In the mesenchymal cells of both parts, osteopontin, osteocalcin, alkaline phosphatase, and collagen type I, which are osteogenic markers, were expressed. Moreover, the mesenchymal cells of the fetal part of the placenta were mineralized for 3 weeks, but those of the maternal part were not. These results showed that the mesenchymal cells derived from human placenta had an osteogenic phenotype and that only the mesenchymal cells of the fetal part were capable of being used as a cell source for bone reconstitution.

Translational research involves application of basic scientific discoveries into clinically germane findings and, simultaneously, the generation of scientific questions based on clinical observations. At first, as basic research we investigated tissue-engineered bone regeneration using mesenchymal stem cells (MSCs) and platelet-rich plasma (PRP) in a dog mandible model. We also confirmed the correlation between osseointegration in dental implants and the injectable bone. Bone defects made with a trephine bar were implanted with graft materials as follows: PRP, dog MSCs (dMSCs) and PRP, autogenous particulate cancellous bone and marrow (PCBM), and control (defect only). Two months later, dental implants were installed. According to the histological and histomorphometric observations at 2 months after implants, the amount of bone–implant contact at the bone–implant interface was significantly different between the PRP, PCBM, dMSCs/PRP, native bone, and control groups. Significant differences were also found between the dMSCs/PRP, native bone, and control groups in bone density. These findings indicate that the use of a mixture of dMSCs/PRP will provide good results in implant treatment compared with that achieved by autogenous PCBM. We then applied this injectable tissue-engineered bone to onlay plasty in the posterior maxilla or mandible in three human patients. Injectable tissue-engineered bone was grafted and, simultaneously, 2–3 threaded titanium implants were inserted into the defect area. The results of this investigation indicated that injectable tissue-engineered bone used for the plasty area with simultaneous implant placement provided stable and predictable results in terms of implant success. We regenerated bone with minimal invasiveness and good plasticity, which could provide a clinical alternative to autogenous bone grafts. This might be a good case of translational research from basic research to clinical application.

Osteogenesis occurs in porous hydroxyapatite (HA) when porous HA blocks combined with marrow mesenchymal cells are grafted in vivo. In vitro bone formation occurs in HA pores when HA combined with marrow cells is cultured in osteogenic medium containing dexamethasone. This cultured bone/HA construct possesses higher osteogenic ability when it is grafted in vivo. In the present study, we compared the osteogenic potential of a cultured bone/HA construct with that of a marrow mesenchymal cell/HA composite. Marrow cells were obtained from the femoral bone shaft of 7-week-old, male Fischer 344 rats and were cultured in T-75 flasks. Cells were concentrated, then frozen and stored in liquid nitrogen for 6 months. The cryopreserved cells were then thawed and prepared for subculture in porous HA (5 × 5 × 5 mm, Interpore 500) and for implantation with porous HA. After 2 weeks of subculture, three cultured bone/HA constructs were separately implanted in the right side of the back of each syngeneic 7-week-old male Fischer rat, and three thawed cell/HA composites (without subculture) were separately implanted in the left side. These implants were harvested at 2 or 4 weeks postimplantation, and prepared for histological, biochemical, and genetic analysis. Alkaline phosphatase activity and osteocalcin content of cultured bone/HA constructs were much higher than those of the cell/HA composites at 2 and 4 weeks postimplantation. Histological examination and gene expression data agreed with these findings. The culture technique discussed herein should facilitate the development of biosynthetic bone implants with higher osteogenic capacity.

We developed fully opened interconnected porous calcium hydroxyapatite ceramics having two different pore sizes. One has pores with an average size of 150 μm in diameter, an average 40-μm interconnecting pore diameter, and 75% porosity (HA150). The other has pores with an average size of 300 μm in diameter, an average 60–100-μm interconnecting pore diameter, and 75% porosity (HA300). Because of its smaller pore diameter, HA150 has greater mechanical strength than that of HA300. These ceramics were combined with rat marrow mesenchymal cells and cultured for 2 weeks in the presence of dexamethasone. The cultured ceramics were then implanted into subcutaneous sites in syngeneic rats and harvested 2–8 weeks after implantation. All the implants showed bone formation inside the pore areas as evidenced by decalcified histological sections and microcomputed tomography images, which enabled three-dimensional analysis of the newly formed bone and calculation of the bone volume in the implants. The bone volume increased over time. At 8 weeks after implantation, extensive bone volume was detected not only in the surface pore areas but also in the center pore areas of the implants. A high degree of alkaline phosphatase activity with a peak at 2 weeks and a high level of osteocalcin with a gradual increase over time were detected in the implants. The levels of these biochemical parameters were higher in HA150 than in HA300. The results indicate that a combination of HA150 and mesenchymal cells could be used as an excellent bone graft substitute because of its mechanical properties and capability of inducing bone formation.

In this article we describe the expansion of in vitro osteogenic capability of human osteoblasts (HOS cells) after sorting by fluorescence-activated cell sorting (FACS) with the osteoblastic marker of human bone alkaline phosphatase (AP) monoclonal antibody. After culturing for 7 days, the HOS cells were incubated with fluorescein isothiocyanate (FITC)-labeled AP monoclonal antibody. The antibody recognized the cells with high AP activity (high AP cells), which were about 76% of the total cells. After the HOS cells were sorted, the high AP cells could be recovered, and almost all of them reacted strongly with the AP antibody. Therefore, we were able to condense the high AP cells about 1.3 times. We further cultured the sorted cells as well as the unsorted control cells. After the initial seeding, the culturing periods for both groups of cells were 20 days. At the end of this period, we measured AP activity per DNA and osteocalcin contents. In contrast to the low condensation ratio of the high AP cells in the sorted fraction, the AP activity and osteocalcin contents were about nine times and four times greater than those of the unsorted cells, respectively. These results demonstrated that using the sorting technique to isolate the high AP cells might be a useful method for applications in bone tissue engineering.

Previously, we found that hepatocyte growth factor receptor (c-Met)-and alpha-fetoprotein (AFP)-expressing cells were present in adult rat bone marrow, and that these cells also expressed hematopoietic stem cell markers, such as CD34, Thy-1, and c-Kit. When bone marrow cells were cultured in a hepatocyte growth medium (HGM) with HGF and EGF, colonies composed of polygonal cells resembling mature hepatocytes appeared by 2 weeks and grew very slowly because of overgrowth of stromal cells. At days 34–41, 2-mm2 sheets of hepatocyte-like cells were cut out of their colonies by scratching with an injection needle under observation with a phase contrast microscope, transferred into wells of 24-well plates, and cultured in the HGM medium in the presence or absence of HGF and EGF. When cells reached confluence, cells were detached with trypsin and EDTA and transferred step by step into bigger culture vessels. Thus, hepatocyte-like cells were expanded 1000-fold during less than 4 months. These cells were immunocytochemically stained for albumin and also for AFP and the hematopoietic stem cell markers described above, showing characteristics of oval cells. By RT-PCR, we detected mRNAs of tryptophan-2,3-dioxygenase and tyrosine aminotransferase, markers of hepatocytes at a terminal differentiation stage. The present culture system may be useful for supply of hepatocyte resources for cell transplantation therapy.

Fulminant hepatic failure (FHF) is still associated with high mortality despite recent advances in medical management. There is need of an effective and safe bioartificial liver (BAL) support to help keep patients with FHF alive until an organ becomes available for transplantation or the native liver recovers. The aim of this study was to establish highly functional liver cells by means of transfecting hepatocyte nuclear factor (HNF)-4 gene for the development of BAL. We constructed adenovirus vector carrying rat HNF-4 cDNA, and transfected to hepatoma-derived cell lines, HepG2 and HuH-7, to enforce expression of the exogenous HNF-4 gene. We analyzed expression of HNF-4, HNF-1, and liver-specific genes in cells infected by the adenovirus vector expressing HNF-4. Adenovirus-mediated HNF-4 gene transfer resulted in increases in expressions of HNF-4, HNF-1, and liver-specific genes such as apolipoproteins, α1-antitrypsin (α1-AT), phosphoenolpyruvate carboxy-kinase, cytochrome P450 families, and glutamine synthetase in transfected hepatoma cells. Cells overexpressing HNF-4 removed ammonia from medium supplemented with NH₄Cl to a greater extent than control cells. These findings demonstrated that transfected cell lines restored differentiated gene expressions and liver-specific function by the overproduction of HNF-4. HNF-4-overexpressing hepatocyte cell lines are useful for bioreactor of BAL systems.

Mesenchymal stem cells (MSCs) possess the capacity for site-specific differentiation of cell types in response to cues provided by different organs. This phenomenon suggests that MSCs participate in cutaneous wound regeneration. However, there are no prior reports on the influence of the local application of MSCs on cutaneous wound regeneration. To examine the effects of MSCs on wound regeneration, we cultured bone marrow cells of the femur of rats and treated the plastic adherent cells with a differentiation medium to induce differentiation. After treatment, we found that the bone marrow-derived plastic adherent cells possessed myogenesis, chondrogenesis, and adipogenesis capabilities, indicating that these cells are MSCs. The bone marrow-derived plastic adherent cells were injected intradermally into the skin of rats, and linear full-thickness incisional wounds were made immediately through the injected area. At 14 days after operation, wounds transplanted with bone marrow-derived plastic adherent cells had healed with very fine scars. Collagen architecture was thick and appeared to be similar to normal dermis. Histomorphologic scale analysis demonstrated significant differences between the control and the wounds transplanted with bone marrow-derived plastic adherent cells. These results indicate that transplanted MSCs can respond quite normally to wound healing and regenerate dermal structure.

For research in regenerative medicine, not only the study of cellular pluripotency but also knowledge of the reorganization of tissue structure is crucial. However, the latter will probably be more difficult to acquire. When small fragments of kidney (approx. 1 × 1 mm) were implanted in the liver of syngeneic LEW rats, the tissue survived at least 2 weeks with retention of normal structure including glomeruli and tubules. In contrast, no kidney structure survived when transplanted to subcutaneous sites, omentum, or spleen. Molecules involved in renal tubular function, such as megalin and glut2 transporter protein, were detectable in the implanted tissue by immunohistochemistry, suggesting that the cells were biologically active. Survival of cortex, medulla, and calyx tissues was then compared. All three components were still detectable 8 weeks after transplantation but cortex and medulla were replaced by granuloma at 6 months. Only calyx tissue survived for up to 12 months after transplantation. There was no marked difference in tissue survival, either when the recipient liver was partially resected or when infantile donor kidney was implanted instead of adult kidney. The present method opens new avenues in the development of regenerative medicine (i.e., tissue transplantation) as an intermediate modus between organ transplantation and cell transplantation.

Neuroepithelial stem cells (NESCs) have emerged as a possible donor material aimed at neural transplantation for the repair of damaged neural circuitry, particularly because of their propensity to differentiate into neurons. We previously ascertained in vitro that NESCs derived from rat early embryos could be amplified in culture containing basic fibroblast growth factors (bFGF), and that neurospheres grown for 7 days in the culture had a strong tendency to differentiate into neurons. In this report, we analyze immunohistochemically the biological nature of bFGF-responsive neurospheres derived from NESCs. We first succeeded in amplifying the number of NESCs from the mesencephalic neural plate of embryonic day 10 Wistar rats with the addition of bFGF. Grown neurospheres were labeled with bromodeoxyuridine (BrdU) in vitro and were stereotactically transplanted into the right striatum of the normal adult Wistar rat. Two weeks after transplantation, a viable graft in the host brain was observed. While many BrdU/Hu double positive cells were seen in the graft, and a few BrdU/nestin double positive cells were also seen, no BrdU/GFAP double positive cells could be identified. These results suggested that bFGF-responsive neurospheres derived from NESCs demonstrated a propensity to differentiate into neurons in the adult brain environment. Furthermore, following in vitro amplification of the original stem cell number with bFGF, the grown neurospheres preserved their propensity to differentiate vigorously into neurons. NESCs are thus suggested as a feasible candidate for intracerebral grafting donor materials aimed at reconstruction of damaged neural circuits.

We investigated efficacy and safety of implantation of autologous bone marrow mononuclear cells plus platelets, including endothelial progenitor cells (EPCs), for recovering refractory chronic peripheral arterial disease (PAD) using visual and quantitative analyses by ^99mTc-tetrofosmin (TF) perfusion scintigraphy, and also investigated various quantitative assessments objectively. We performed 12 consecutive cases and 19 limbs and hands with severe chronic PAD that were almost Fontaine class IV (11/12 cases, about 92%) in this trial. This treatment was very effective in relieving severe pain of PAD, especially for Buerger's disease. We used a visual analog scale (VAS) for measurement of pain level. The maximum pain level before implantation was 66.5 ± 5.0 mm, and it decreased to 12.1 ± 2.2 mm after implantation (p < 0.001). Rest pain in legs and fingers was resolved in 11 cases (11/12 cases, 92%). All patients could measure pain-free walking time on a treadmill, which improved remarkably (140 ± 53 s before implantation vs. 451 ± 74 s after implantation, p = 0.034). Resting ankle brachial pressure index (ABI) in legs implanted with bone marrow mononuclear cells was also improved (0.65 ± 0.08 before implantation vs. 0.73 ± 0.07 after implantation, p = 0.055). According to ^99mTc-TF perfusion scintigraphy, the proximal area (region from knee to ankle) was 1.32 ± 0.10 before implantation versus 1.56 ± 0.11 after implantation (p = 0.007). ^99mTc-TF perfusion scintigraphy in the distal area (region from ankle to end of toes, or from wrist to end of fingers) was 0.79 ± 0.06 before implantation versus 0.83 ± 0.06 after implantation (p = 0.29). Ischemic legs and hands that were injected showed increased perfusion blood flow. ^99mTc-TF perfusion scintigraphy was effective to estimate visual and quantitative analysis of collateral vessels in neovascularization. We were successful with this new treatment for the most severe, chronic PAD that was not curable by any of the current treatments. Thus, this therapeutic angiogenesis could be a new strategy for saving severe ischemic limbs and hands.

In this study, vascular-like tubular tissues called biotubes, consisting of autologous tissues, were prepared using in vivo tissue engineering. Their mechanical properties were evaluated for application as a small-caliber artificial vascular prosthesis. The biotubes were prepared by embedding six kinds of polymeric rods [poly(ethylene) (PE), poly(fluoroacetate) (PFA), poly(methyl methacrylate) (PMMA), segmented poly(urethane) (PU), poly(vinyl chloride) (PVC), and silicone (Si)] as a mold in six subcutaneous pouches in the dorsal skin of New Zealand White rabbits. For rods apart from PFA, biotubes were constructed after 1 month of implantation by encapsulation around the polymeric implants. The wall thickness of the biotubes ranged from about 50 to 200 μm depending on the implant material and were in the order PFA < PVC < PMMA < PU < PE. As for PE, PMMA, and PVC, the thickness increased after 3 months of implantation and ranged from 1.5-to 2-fold. None of the biotubes were ruptured when a hydrostatic pressure was gradually applied to their lumen up to 200 mmHg. The relationship between the intraluminal pressure and the external diameter, which was highly reproducible, showed a “J”-shaped curve similar to the native artery. The tissue mostly consisted of collagen-rich extracellular matrices and fibroblasts. Generally, the tissue was relatively firm and inelastic for Si and soft for PMMA. For PMMA, PE, and PVC the stiffness parameter (β value; one of the indexes for compliance) of the biotubes obtained was similar to those of the human coronary, femoral, and carotid arteries, respectively. Biotubes, which possess the ability for wide adjustments in their matrices, mechanics, shape, and luminal surface design, can be applied for use as small-caliber blood vessels and are an ideal implant because they avoid immunological rejection.

The effect of cross-linking of a hydroxyapatite/collagen (HA/Col) nanocomposite, in which HA nanocrystals and collagen fibers are aligned like natural bone by a self-organization mechanism between HA and collagen in vitro, on mechanical properties was examined. The influence of degree of cross-linking, as well as rhBMP-2 preadsorption to the composite on the substitution pattern and rate with bone, was examined. In Experiment 1, anterior fusion was carried out at the C3–C4 vertebrae on 10 dogs and they were implanted as follows: without cross-linking and without adsorbed rhBMP-2 (three dogs), with cross-linking and without adsorbed rhBMP-2 (three dogs), without cross-linking and with adsorbed rhBMP-2 (two dogs), and with cross-linking and adsorbed rhBMP-2 (two dogs). Implants were removed from each dog for histology determinations after 12, 16, and 24 weeks in the non-rhBMP-treated groups, and after 16 and 24 weeks in the rhBMP-treated groups. In Experiment 2, the HA/Col composites with cross-linking and both with and without rhBMP-2 pretreatment were implanted into a bone defect of 20 mm made in the central part of tibiae in dogs (N = 3 in each group). As a control, bone defects of 20 mm remained without implantation (N = 3). The dogs were allowed to walk using an Ilizarov extra skeletal fixator. The implants were removed after 12, 16, and 24 weeks from one dog in each group. The cross-linking of the HA/Col composite was effective in controlling both the mechanical strength and bioresorbability. A “self-organization process” on the HA/Col implant surface resulted in the formation of bone remodeling units in and around the implant. Radiographic and histological findings suggest that a combined treatment of cross-linking of the HA/Col composite with preadsorption of rhBMP-2 molecules may be a very suitable replacement of existing ceramic systems in the anterior fusion of the cervical spine, as well as inlay grafting of bone defects in weight-bearing sites.

Three-dimensional reconstructed organoids in vitro are valuable for not only regenerative medicine but also drug development. However, the manipulation of conventional three-dimensional cultures is not simple. We describe a nylon membrane ring-embedded or a pressed silk sheet-embedded scaffold made of collagen “vitrigel” that can facilitate three-dimensional cultures for reconstructing an epithelial-mesenchymal model or a hard connective tissue model, respectively. Here we define vitrigel as a gel in a stable state produced by rehydration after the vitrification of a traditional hydrogel. The collagen vitrigel was successfully prepared in three steps involving a gelation process in which a cold and clear neutral salt solution containing type I collagen formed an opaque and soft gel by incubation at 37°C, a vitrification process in which the gel becomes a rigid material like glass after sufficient drying out, and finally a rehydration process to convert the vitrified material into a thin and transparent gel membrane with enhanced gel strength. The framework-embedded collagen vitrigel scaffold that can be easily reversed by forceps was prepared by inserting a nylon ring or a silk sheet in the collagen solution prior to the gelation. The scaffold enabled culturing anchorage-dependent cells on both surfaces of the collagen vitrigel by the manipulation of two-dimensional cultures and consequently resulted in reconstructing a three-dimensional organoid. An intestinal epithelial-mesenchymal model was reconstructed by coculturing fibroblasts on the opposite side of monolayered Caco-2 cells on the nylon ring-embedded collagen vitrigel. Also, fibroblasts seeded on both surfaces of the silk sheet-embedded collagen vitrigel proliferated well and formed multilayers and some cells invaded into the vitrigel framed by the network of numerous strong silk filaments, suggesting a reconstruction of a hard connective tissue model. These data demonstrate that the collagen vitrigel is a valuable scaffold for tissue engineering.