Cell and Tissue Research

, Volume 316, Issue 2, pp 141–153

Custom-shaping system for bone regeneration by seeding marrow stromal cells onto a web-like biodegradable hybrid sheet

  • Kohei Tsuchiya
  • Taisuke Mori
  • Guoping Chen
  • Takashi Ushida
  • Tetsuya Tateishi
  • Takeo Matsuno
  • Michiie Sakamoto
  • Akihiro Umezawa
Regular Article

DOI: 10.1007/s00441-004-0862-1

Cite this article as:
Tsuchiya, K., Mori, T., Chen, G. et al. Cell Tissue Res (2004) 316: 141. doi:10.1007/s00441-004-0862-1


New bone for the repair or the restoration of the function of traumatized, damaged, or lost bone is a major clinical need, and bone tissue engineering has been heralded as an alternative strategy for regenerating bone. A novel web-like structured biodegradable hybrid sheet has been developed for bone tissue engineering by preparing knitted poly(DL-lactic-co-glycolic acid) sheets (PLGA sheets) with collagen microsponges in their openings. The PLGA skeleton facilitates the formation of the hybrid sheets into desired shapes, and the collagen microsponges in the pores of the PLGA sheet promote cell adhesion and uniform cell distribution throughout the sheet. A large number of osteoblasts established from marrow stroma adhere to the scaffolds and generate the desired-shaped bone in combination with these novel sheets. These results indicate that the web-like structured novel sheet shows promise for use as a tool for custom-shaped bone regeneration in basic research on osteogenesis and for the development of therapeutic applications.


Bone regeneration Tissue engineering Scaffold Marrow stroma Polymer KUSA-A1 cells 


New bone for the replacement or restoration of the function of traumatized, damaged, or lost bone is a major clinical and socioeconomic need. Bone formation strategies, although attractive, have yet to yield functional and mechanically competent bone. Autografts (bone obtained from another site in the same subject of the same species) are currently the gold standard for bone repair and substitution, but the use of autografts has several serious disadvantages, such as additional expense and trauma to the patient, the possibility of donor-site morbidity, and limited availability (Glowacki and Mulliken 1985; Bauer and Muschler 2000). Because of these problems, bone tissue engineering has been heralded as an alternative strategy for the regeneration of bone (Langer and Vacanti 1993; Crane et al. 1995; Boyan et al. 1999).

Bone has a highly organized structure composed of a calcified connective tissue matrix formed by the proliferation and differentiation of osteoprogenitors into mature osteoblasts (Maniatopoulos et al. 1988; Pitaru et al. 1993). The osteoblasts belong to the stromal fibroblastic system of the bone marrow, which contains other stromal cells, such as chondrocytes and myoblasts (Friedenstein 1976; Owen and Friedenstein 1988; Haynesworth et al. 1992). We have shown that mouse stromal cells are able to differentiate into cardiomyocytes (Makino et al. 1999; Gojo et al. 2003), endothelial cells, neuronal cells (Kohyama et al. 2001), and adipocytes (Umezawa et al. 1991). We have previously established a murine osteoblast cell line, KUSA-A1, and shown that clonal stromal cells can generate bone in vivo (Umezawa et al. 1992). Marrow stromal cells are expected to serve as a good source for cell therapy, in addition to embryonic stem cells and fetal cells. Although these precursor cells have been reported to be stem cells, it remains unknown as to whether they are homogeneous or whether they constitute subpopulations of cells committed to various lineages of differentiation (Owen and Friedenstein 1988). The acquisition of a large number of osteoblast precursors as a cell source and the control of differentiation are essential to the success of the production of tissue-engineered bone for clinical application (Minuth et al. 1998).

Temporary three-dimensional scaffolds play an important role in the manipulation of the functions of osteoblasts (Chicurel et al. 1998) and in the guidance of the formation of new bones into the desired shapes (Ishaug et al. 1997; Ishaug-Riley et al. 1998), in the bone tissue-engineering approach. These scaffolds should be biocompatible, osteoconductive, biodegradable, highly porous with a large surface-to-volume ratio, mechanically strong, and malleable into the desired shapes. Synthetic polymers, such as poly(lactic acid), poly(glycolic acid), and poly(DL-lactic-co-glycolic acid), which is abbreviated here as PLGA, are easily processed into the desired shapes and are mechanically strong (Langer and Vacanti 1993; Ishaug et al. 1997; Ishaug-Riley et al. 1998; Mikos et al. 1998). Moreover, their degradation time can be manipulated by controlling their crystallinity, molecular weight, and the ratio of lactic acid to glycolic acid co-polymer (Thomson et al. 1995). Collagen is the primary component of extracellular bone matrix and has been demonstrated to produce good osteoconductivity (Aronow et al. 1990). Synthetic polymers, on the other hand, lack cell recognition signals, and their hydrophobic properties hinder the uniform seeding of cells in three dimensions. However, since collagen scaffolds are mechanically weak, these materials have been hybridized to combine their advantages and provide excellent three-dimensional porous biomaterials for bone tissue engineering.

We have developed collagen-hybridized PLGA sponge and have reported good biocompatibility both for cartilage tissue engineering with mature bovine chondrocytes (Sato et al. 2001; Chen et al. 2003) and for bone tissue engineering with osteoblasts isolated from marrow cells (Ochi et al. 2003). The sponges organize satisfactory cartilage and bone tissue when the cells fill the pores of the scaffold. The uniform distribution of the cells throughout the scaffolds is imperative for the development of homogeneous tissue, but special seeding techniques, such as stir flask culture or perfusion bioreactor culture, are required to produce an even distribution reliably (Vunjak-Novakovic et al. 1998; Freed et al. 1999). Since simple static seeding methods tend to produce an uneven distribution with large patches of cells on the surface, we have hybridized collagen microsponges with PLGA sheets. The sheets have a collagen fiber network in the openings of the PLGA fiber sheets. The findings that the sheets trap cells, that they can be laminated or rolled to control their shape for tissue engineering, and that they also have the capacity to supply minerals by depositing apatite particulates on the surface of collagen microsponges indicate their great advantages. In this study, we have shown that, when the novel web-like structured sheets are used as a scaffold for bone tissue engineering, an even cell distribution and the control of its shape can be achieved.

Materials and methods

Scaffold fabrication

The hybrid sheet was prepared by allowing collagen microsponges to form in the openings of PLGA knitted sheets as previously described (Chen et al. 2003). Briefly, as shown in Fig. 1A, a knitted Vicryl sheet made of polylactin 910 (a 90:10 co-polymer of glycolic acid and lactic acid) was immersed in a type I bovine collagen acidic solution (pH 3.2, 0.5% by weight, Koken, Tokyo, Japan) and frozen at −80°C for 12 h. It was then freeze-dried under vacuum (0.2 Torr) for 24 h to allow the formation of collagen microsponges. The collagen microsponges were further cross-linked by treatment with glutaraldehyde vapor saturated with a 25% aqueous glutaraldehyde solution at 37°C for 4 h. After the cross-linking, the sponge was treated with a 0.1 M aqueous glycine solution to block unreacted aldehyde groups. After being washed with deionized water and freeze-dried, the collagen-hybridized PLGA (PLGA/COL) sheet was complete. The sheets were sterilized with ethylene oxide for cell culture.
Fig. 1A–C

Bone formation in collagen hybrid PLGA sponge. Macroscopic appearance of the collagen hybrid PLGA sponge (A). Complete bone formation of an in vivo 4-week construct based on KUSA-A1 cells and collagen hybrid PLGA sponge (B). Some constructs showed uneven bone distribution. Living cells did not completely fill the pore cavities of the sponge (C). ×5

Cell culture

KUSA-A1 cells were cultured as described previously (Umezawa et al. 1992; Kohyama et al. 2001; Ochi et al. 2003).

Cell seeding of a PLGA/COL sheet

A PLGA/COL sheet was placed into a 100-mm culture dish (Falcon) and covered with a silicone rubber framework. A 2-ml volume of KUSA-A1 cell suspension at a density of 5×106/ml was dropped onto the PLGA/COL sheet (area: 4 cm2). After cultivation for 6 h, the sheet was turned over and re-seeded with the same number of cells on the reverse side. For comparison, a KUSA-A1 suspension (1×107 cells/ml) was injected into collagen hybrid PLGA sponges, and the injected sponges were incubated at 37°C for more than 30 min. The sponges were then transplanted into the subcutaneous tissue of C3H mice as previously described (Ochi et al. 2003).

Scanning electron microscopy

PLGA sheets and PLGA/COL sheets were examined by scanning electron microscopy (SEM). They were cut into small pieces with scissors and coated with gold by means of a sputter coater (Sanyu Denshi, Tokyo, Japan; gas pressure: 50 mtorr, current: 5 mA, coating time: 180 s). The samples were examined with a JSM-6400Fs scanning electron microscope (JEOL, Tokyo, Japan) operated at a voltage of 3 kV.

Transmission electron microscopy

Samples cultured in vitro for 1 day and 2 weeks were examined by transmission electron microscopy (TEM). They were fixed in 2.5% glutaraldehyde, postfixed in 1% osmium tetroxide, dehydrated, and embedded in resin. Ultrathin sections (70–90 nm) were cut and stained with 2% uranyl acetate and Reynold’s lead citrate before being examined with a JEM-1200 EX microscope (JOEL) at 80 kV.

In vivo assay

All animals received humane care in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources and published by the US National Institutes of Health (NIH Publication no. 86–23, revised 1985). The operation protocols were accepted by the Laboratory Animal Care and Use Committee of the National Research Institute for Child and Health Development (Tokyo) and by Keio University School of Medicine.

Laminated sheet implantation onto calvarial defects

Surgery was performed under anesthesia with Nembutal (50 mg/kg, i.p.). A midline skin incision approximately 1 cm long was made on the dorsal surface of the cranium, and the periosteum was removed. A 4.3-mm-diameter full-thickness circular defect was created in the skull with a trephine bar (Hasegawa Medical, Tokyo, Japan) attached to an electric handpiece, with minimal penetration of the dura. The defect was covered with three sheets of cell-loaded scaffolds cut to fit the shape of the defect. The scalp was then closed with 6-0 nylon sutures, and the animals were given access to food and allowed to behave ad libitum. Ten defects were left untreated, 10 defects were treated with PLGA/COL sheets alone as controls, and 10 defects were filled with PLGA/COL sheets seeded with KUSA-A1 cells.

Custom-shaped bone formation in mice

With the aim of producing long bone, cylinder-like bone was formed by rolling KUSA-A1-seeded sheets around a silicone rod 3 mm in diameter. The ends of the sheets were hemmed with 4-0 Vicryl dissolvable stitches. The rolled sheets were transplanted into subcutaneous tissue for 4 weeks, flat or after being knotted. Tissue-engineered phalanges were formed in a similar manner. KUSA-A1-seeded sheets were wrapped around a silicone rubber block trimmed in advance to the shape of the distal phalanx and transplanted into subcutaneous tissue. After cultivation in syngeneic C3H/He mice, NOD/SCID mice, and NOD/SCID/IL2-receptor γ knock-out immunodeficient mice (NOG), the specimens were extracted and examined histologically.

Histological and Immunohistochemical staining

Calvaria, femurs, and subcutaneous specimens were dissected at various times after implantation and fixed and decalcified for 1 week in 10% EDTA (pH 8.0) solution. After dehydration in ascending concentrations of ethanol and xylene, the transplants were embedded in paraffin and sectioned. The paraffin sections were then deparaffinized, hydrated, and either were stained with hematoxylin and eosin or were immunohistochemically stained with anti-human Factor VIII mAb (DAKO, Carpenteria, Calif.) to detect angiogenesis.


Bone distribution of sponges

We previously reported that KUSA-A1 cells generated cuboidal bone when used in combination with PLGA/COL hybrid sponge. In those experiments, the KUSA-A1 cells were distributed evenly into sponges and had generated cuboidal bone in the subcutaneous tissue at 8 weeks (Fig. 1A, B); however, some of them exhibited uneven bone formation. The bone was generated mainly at the periphery of the sponge, like an eggshell (Fig. 1C).

Collagen-hybridized PLGA sheet

We developed the sheet-style scaffold to prevent uneven bone formation (Fig. 2A–C). The synthetic biodegradable polymer, PLGA, which was 200 μm thick, was hybridized with collagen microsponges. SEM analysis revealed that the collagen microsponges overlaid the interstices of the fabricated web-like PLGA sheets (Fig. 2D, E). This structure was expected to entrap a large number of cells.
Fig. 2A–E

Collagen hybrid PLGA sheet. The thin 200 µm sheets are easy to handle (A, B). SEM micrographs of a PLGA sheet without collagen (C) and a hybrid sheet with collagen (D). Higher magnification of the hybrid sheet (E)

Cells uniformly trapped in sheets with a web-like pattern

A cell suspension was simply dropped onto the sheets (Fig. 3A–C), and the collagen microsponges had filled with precipitated cells 30 min after seeding. The cells had adhered completely at 3 h (Fig. 3D–F), whereas the non-hybridized PLGA sheets did not trap cells efficiently. The cells continued to spread and generate matrix over the sheets, so that at 2 weeks, the sheets were completely covered by extracellular matrix, and the thickness of sheet had increased (Fig. 3G, H). After in vitro culture for 1 day, SEM examination revealed the preference of cells for components of the scaffold (Fig. 3I–K). A large number of cells had adhered to the collagen microsponge portion rather than to the PLGA fibers. Since synthetic polymers lack cell recognition signals, and since their hydrophobic properties hinder cell adhesion, the hybridization of collagen microsponges was advantageous in achieving cell retention.
Fig. 3A–K

Process of cell seeding. A–C Cell seeding of the sheet. The sheet was framed in silicone rubber (A), and the cell suspension was simply dropped onto the sheet (B). The silicone rubber was removed 6 h after seeding, and the sheet was cultured in growth medium (C). D–F Phase-contrast micrographs of a cell-seeded sheet. The openings of the PLGA/COL sheet (D) were filled with cells at 30 min (E) and were completely covered with abundant extracellular matrix at 1 week (F). G, H Toluidine-blue stained cross sections of the cell-seeded sheet at 1 day (G) and 2 weeks (H). The thickness of the sheet had increased (arrowheads PLGA fibers, arrows collagen fibers). I–K SEM micrographs of a PLGA/COL sheet after cell seeding. A portion of PLGA fiber (J) and collagen microsponge (K) are shown. D–F ×200, G, H ×400, I ×50, J, K ×200

The PLGA/COL sheets trapped significantly more cells than non-hybridized control PLGA sheets (Fig. 4A, B). The numbers of cells that had adhered to the sheets at 6 h after seeding increased with the initial cell seeding concentration (Fig. 4C).
Fig. 4A–C

Effect of collagen microsponges on cell adhesion. Microscopic appearance of a PLGA sheet (A) and PLGA/COL (B) sheet 1 day after seeding the same number of cells in suspension (1×107 cells). The seeding rates (C) indicate the differences in the number of adhering cells between the two sheets at the initial cell seeding concentrations. A, B ×100

KUSA-A1 cells produce abundant collagen fibrils on the scaffolds

TEM revealed that the cells were tightly adherent to the surface of the PLGA fibers covered with collagen microsponges 24 h after seeding. The cells were spindle-shaped and in close contact to each other. The cell nuclei had become large and bright at 2 weeks, and the nucleoli had also increased in size. The abundant collagen fibrils bridged the intercellular spaces and connected to the synthetic fiber surface (Fig. 5A). The cells contained extensive dilated rough endoplasmic reticulum in their cytoplasm (Fig. 5B), thereby demonstrating the high affinity between the cells and collagen and the active protein-synthesizing capacity of the cells on the scaffold.
Fig. 5A, B

TEM analysis of a PLGA/COL sheet with cells. Ultrastructural analysis was performed 2 weeks after cell seeding. Note the abundant collagen fibrils produced from the cells (A arrows) and the abundant rough endoplasmic reticulum (rough ER) at 2 weeks (B arrows)

Implantation of laminated collagen hybrid PLGA sheets with KUSA-A1 cells for calvarial defects

We investigated whether calvarial defects were restored by the implantation of laminated collagen hybrid PLGA sheets with KUSA-A1 cells. No bridging of the cranial defects with new bone was observed at 4 weeks or at 8 weeks in mice, either in the sham-operated group or in the sheet alone group (n=10). The healing response in the control groups consisted of only a thin layer of connective tissue spanning the defects. Because of the scaffold remnants, the connective tissue was thicker in the defects treated with sheets alone than in the sham-operated group. By contrast, implantation of the hybrid PLGA sheet with KUSA-A1 cells over the calvarial defects resulted in closure of the defects with newly synthesized bone within 4 weeks in 10 out of 10 trials (Fig. 6). To determine the contribution of the implanted cells to osteogenesis, we labeled the implanted cells with β-galactosidase. The percentage of implanted cells was 96% in the generated bone, implying that most of the osteoblasts were derived from the donor cells. The thickness of the generated bone was variable and seemed to be related to the numbers of sheets. The scaffolds persisted at 4 weeks but had been completely absorbed at 8 weeks. The synthesized bone contained many vascular channels, and vascular endothelial cells were positive when immunohistochemically stained with Factor VIII antibody at 4 weeks. The newly formed bone contained a more prominent marrow space at 8 weeks. Bone healing was monitored radiographically at designated times after surgery (Fig. 7); the results showed a gradual increase in newly formed bone in the calvarial defects, and complete healing at 8 weeks.
Fig. 6A–F

Cranial defects implanted with hybrid sheets containing osteoblasts. No bone formation in the defect is seen at 4 weeks (4W) or 8 weeks (8W) after a sham operation (A, D) or after implantation of PLGA/COL sheets alone (B, E). Bone formation can be observed following hematoxylin and eosin staining of KUSA-A1-loaded sheets 4 weeks (C) and 8 weeks (F) after the sheets were grafted into a cranial defect. Insets: High-power views. A–F ×20, Insets ×200

Fig. 7A–F

Radiographs of calvaria. A A 4.3-mm defect (arrowheads) was created on the left side of skull. The X-ray density of the KUSA-A1-seeded sheet gradually increased (B 0 week, C 1 week, D 2 weeks, E 4 weeks, F 8 weeks). A ×2, B–F ×6

Custom-shaped osteogenesis with PLGA/COL sheets

Since cylinder-shaped scaffolds are more suitable for segmental long bone defects, the cell-seeded sheets were rolled up around a hollow silicone rod (3 mm in diameter) immediately before implantation into subcutaneous tissue. After 4 weeks of cultivation in vivo, cylinder-shaped bone had formed (Fig. 8). Its outer circumference was covered with a fibrous layer, and ubiquitous vascular channels were connected to it. We then attempted to produce bone of more complex shapes (Fig. 9). The rolled sheets were knotted and transplanted into subcutaneous tissue, and uniquely shaped knots of bone were created. The cell-seeded sheets wrapped around silicone rubber blocks conformed well to the shape of the distal phalanx when removed from the mouse after 4 weeks.
Fig. 8A–H

Cylinder-shaped bone formation. A Protocol. B Solid bone formation around the silicone core after 4 weeks of subcutaneous tissue culture. Macroscopic views (C, D) and radiographs (E axial view, F longitudinal view) of a cylinder of bone after removal of the silicone core. G Histological view of a cross section. H Higher magnification of G. B ×1.4, C ×1, D ×1.4, E, F ×1, G ×6, H ×40

Fig. 9A–H

Generation of a bone knot and phalanx-like bone. A–D Bone knot. Bone formed in the predesigned knotted PLGA/COL sheet. A, C Macroscopic view of a bone knot. B, D Radiographs of the bone knot in mice. E–H Phalanx-like bone. E, G Macroscopic view of the bone generated (with thumb of K.T. for reference). F, H Radiographs of the phalanx-like bone in mice. A–D ×1.3, E–H ×1.1


The shape of the sheets used in this study was not planned in advance. However, the sheets could easily be molded into desired shapes, such as a cylinder, knot, and phalanx. Thickness and shape could be adjusted by simply laminating, lapping, curling, or rolling the sheet, implying that almost any desired shape of bone can be produced with the sheets.

Up-regulation of cell adhesiveness by hybridization of collagen microsponge to hydrophobic sheets

Cells find it difficult to adhere to hydrophobic scaffolds, but hybridization of such scaffolds with collagen microsponges increases their cell adhesiveness. We have previously reported a novel three-dimensional porous scaffold hybridizing synthetic PLGA (Ochi et al. 2003) and naturally derived collagen for cartilage tissue engineering (Chen et al. 2000, 2001a). Osteoprogenitor cells can be readily processed into three-dimensional porous structures with desired pore morphological features that fit the defect, prior to surgery. However, cell seeding becomes more difficult as the thickness of the scaffold increases. Cell density and glycosaminoglycan content have been found to decrease with thickness in cartilage tissue engineering (Freed et al. 1994). Introduction of a novel PLGA-hybrid sheet with a thin planar sheet structure greatly improves cell seeding and even cell distribution. Since cell adhesion to the collagen microsponge in the opening of the PLGA is specific and rapid, this adhesion may result in transmission of “adhesive” signals that are important for the maintenance and/or commitment of marrow-derived stromal cells, such as osteoblasts and chondrocytes. This novel hybrid sheet shows a high degree of cell adhesion capacity and a more even cell distribution than we had expected.

Validity of the PLGA/COL hybrid sheet as a scaffold for bone tissue engineering

Biomaterials are essential for bone tissue engineering, and whether permanent or biodegradable, naturally occurring or synthetic, they must be biocompatible and ideally should be osteoinductive and osteoconductive (Urist 1965; Thomson et al. 1995). We have employed a PLGA/COL hybrid, one of the most promising scaffolds for osteogenesis. High-molecular weight molecules, such as poly(lactic acid) and poly(glycolic acid) (Laurencin et al. 1996; Gopferich et al. 1999), and inorganic materials, such as ceramics (Ohgushi et al. 1996), hydroxyapatite (Freed et al. 1999), and β-tricalcium phosphate (TCP), have been evaluated as scaffolds for osteogenesis. Hydroxyapatite and TCP efficiently conduct osteoprogenitor cells into themselves and become highly integrated with adjoining native bone (Osborn and Newesely 1980; Ohgushi et al. 1990), i.e., they have excellent osteoconductivity and osseointegration properties. However, they do not induce osteogenic differentiation, i.e., they lack osteoinductivity. Moreover, problems arise associated with their slow biodegradability and their association with inflammation because of immunologic reactions.

Further modification of PLGA/COL sheets for bone tissue engineering

To circumvent these limitations, natural or synthetic materials and composite scaffolds based on poly(lactic acid), poly(glycolic acid), and their co-polymer, PLGA, have been developed to increase biodegradability (Ishaug et al. 1997; Ishaug-Riley et al. 1998) and decrease immunological reactions (Mikos et al. 1998). These synthetic polymers are mechanically stronger (Boyan et al. 1999) than naturally derived polymers, such as collagen, and the scaffold can be used either alone, in combination with osteoinductive growth factors, or with osteoconductive inorganic materials (Chen et al. 2001b; Kikuchi et al. 2002). Growth factors, such as bone morphogenetic proteins (Lane et al. 1999; Oldham et al. 2000; Peter et al. 2000) and vascular endothelial growth factor (Murphy et al. 2000; Tabata et al. 2000), can be incorporated into these synthetic polymers, and small hydroxyapatite particles can also be coated onto the polymers. Thus, the PLGA/COL hybrid sheet can be endowed with osteoinductivity and osteoconductivity to shorten the osteogenesis period after implantation and to obtain mechanical strength with plasticity. However, even after the addition of hydroxyapatite, the mechanical strength of the scaffold may be insufficient to maintain its original shape when the scaffold is used to treat long bone defects, and greater strength may be required to resist excessive mechanical overload or to support body weight. There is also concern that the acidic milieu associated with PLGA degradation may be toxic to cells and may induce inflammation (Bostman 1991; Wake et al. 1998), but we have found no evidence of such adverse reactions in our study.

Synthetic polymers are currently used for a number of orthopedic devices, including suture anchors and interference screws. Collagraft (Zimmer, Warsaw, Ind.), a composite of porous calcium phosphate granules and bovine-derived fibrillar collagen for bone regeneration (Cornell et al. 1991), was approved by the US Food and Drug Administration (FDA) in 1993 (Naughton 2002). Our hybrid sheet also consists of matrices that have been approved by the FDA. Some tissue-engineered skin replacement products are on the market, and the technology of tissue engineering for sheet materials is well-established. The scaffold used in this study retains large numbers of osteoprogenitors or osteoblasts derived from bone marrow, and because of its flexibility, assembles them into tissue of the desired shape in mice. The hybrid sheet is expected to become a useful scaffold for bone tissue engineering, and by taking advantage of its unique sheet form, it may be applicable elsewhere, such as in the regeneration of skin, blood vessels, ligaments, and periosteum, in addition to bone tissue engineering.


We sincerely thank Y. Takeda and S. Matsumoto for support throughout the work, and N. Hida, T. Inomata, Y. Hashimoto, and Y. Nakamura for providing expert technical assistance.

Copyright information

© Springer-Verlag 2004

Authors and Affiliations

  • Kohei Tsuchiya
    • 1
    • 2
    • 3
    • 4
  • Taisuke Mori
    • 1
    • 2
  • Guoping Chen
    • 4
    • 5
  • Takashi Ushida
    • 5
    • 6
  • Tetsuya Tateishi
    • 4
  • Takeo Matsuno
    • 3
  • Michiie Sakamoto
    • 2
  • Akihiro Umezawa
    • 1
  1. 1.Department of Reproductive Biology and PathologyNational Research Institute for Child and Health DevelopmentTokyoJapan
  2. 2.Department of PathologyKeio University School of MedicineTokyoJapan
  3. 3.Department of Orthopedic SurgeryAsahikawa Medical CollegeHokkaidoJapan
  4. 4.Biomaterials CenterNational Institute for Materials ScienceIbarakiJapan
  5. 5.Tissue Engineering Research CenterNational Institute of Advanced Industrial Science and TechnologyHyogoJapan
  6. 6.Center for Disease Biology and Integrative Medicine, School of MedicineUniversity of TokyoTokyoJapan

Personalised recommendations