In Vivo Mechanical Condition Plays an Important Role for Appearance of Cartilage Tissue in ES Cell Transplanted Joint Masaaki Nakajima,1,2 Shigeyuki Wakitani,3 Yasuji Harada,1 Akira Tanigami,4 Naohide Tomita1 1
International Innovation Center, Kyoto University, Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501, Japan
2
Department of Physical Therapy, School of Health Sciences, Kibi International, University, 8 Iga-machi, Takahashi, Okayama 716-8508, Japan
3
Department of Orthopaedic Surgery, School of Medicine, Osaka City University, 1-5-17, Asahi-cho, Abeno-ku, Oosaka 545-8586, Japan
4
Fujii Memorial Research Institute, Otsuka Pharmaceutical Co., Ltd., 1-11-1, Karasaki, Ohtsu, Shiga 520-0106, Japan
Received 26 October 2006; accepted 8 May 2007 Published online 3 August 2007 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jor.20462 ABSTRACT: The objective of this study was to evaluate the effects of the mechanical environment on the formation of cartilage tissue in transplanted embryonic stem (ES) cells. Full-thickness osteochondral defects were created on the patella groove of SD rats, and ES cells (CCE ES cells obtained from 129/Sv/Ev mice and Green ES FM260 ES cells obtained from 129SV [D3] - Tg [NCAGEGFP] CZ—001–FM260Osb mice) were transplanted into the defects embedded in collagen gel. The animals were randomly divided into either the joint-free group (JF group) or the joint-immobilized group (JI group) for 3 weeks after a week postoperatively. The results showed that cartilage-like tissue formed in the defects of the JF group whereas large teratomatous masses developed in the defects of the JI group. Some parts of the cartilage-like tissue and the teratomatous masses were positively stained with immunostain for GFP when the Green ES FM260 ES cells were transplanted. It is suggested that the environment plays an important role for ES cells in the process of repairing cartilage tissue in vivo. ß 2007 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 26:10–17, 2008
Keywords: transplanted joint; regeneration; ES cells; chondrogenic lineage; mechanical condition
INTRODUCTION Articular cartilage has previously been thought to have a limited capacity for spontaneous repair. However, several reports have shown successful repair of defects in articular cartilage using cell transplantation and so on.1–6 Drawbacks of this method are that only a limited number of cells can be obtained from the donor site, and that the cultured cells only have a limited proliferative ability. Embryonic stem (ES) cells have the capacity to differentiate into any of the three germ layer cells: gut epithelium (endoderm), smooth muscle (mesoderm), or neural epithelium (ectoderm).7,8 The regulation of this differentiation is not yet clear. Generally, when transplanted, ES cells form terato-
Correspondence to: Naohide Tomita (Telephone: þ81-75753-9201; Fax: þ81-75-753-9201.; E-mail:
[email protected]) ß 2007 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
10
JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2008
mas containing derivatives of all three embryonic germ layers. It was reported that ES cells transplanted into the spinal cord differentiated into neural cells in vivo.9,10 These findings suggest that the interaction between ES cells and the surrounding tissues plays an important role for tissue formation. Wakitani and our group have shown that when ES cells (CCE ES cells) are injected into mouse knee joints, they form teratomas,11 whereas when ES cells are transplanted into osteochondral defects, they form cartilage-like tissue.12 It has also been reported that the neochondrogenesis in the healing of full-thickness articular cartilage defects seems to occur through differentiation of the mesenchymal stem cells in the subchondral tissues to chondrocytes as a result of the stimulation provided by the joint motion.13 As in the case of the mesenchymal stem cells, mechanical stimulation may affect neochondrogenesis in the healing of full-thickness articular cartilage defects using ES cell transplantation. The objective of this study was to evaluate the effect of mechanical factors on the formation of cartilage tissue in
11
IN VIVO MECHANICAL CONDITION AND CARTILAGE TISSUE IN ES CELLS
transplanted ES cell joints by using tail-suspension and joint immobilizing techniques.
MATERIALS AND METHODS Animals and Experimental Protocol Eighteen 12-week-old male SD rats (Charles River Japan Inc., Yokohama, Japan) were used in this study. All animals had an operation for creating defects and implantation on the patella groove bilaterally. A fullthickness osteochondral defect was created on the patellar groove bilaterally. The sizes of the defects were 2 mm in diameter and 2 mm in depth. Ten rats received collagen gel with CCE ES cells transplanted into the right knees and collagen gel without cells into the left knees, while the other eight rats received collagen gel with Green ES FM260 ES cells in the right knees and collagen gel without cells in the left knees. The animals were randomly divided into either joint-immobilized group (JI group) or joint-free group (JF group), as shown in Table 1. Knee joints were immobilized and a tailsuspension procedure was performed for the JI group. Only the tail-suspension procedure was performed for the JF group where knee movement was allowed. The difference in mechanical environment between the JI and JF groups was thought to be the presence or absence of cyclic loading applied by patella movement. These tail-suspension and joint-immobilization procedures were performed 1 week postoperatively to make the cell graft survival equal (Fig. 1A). The animals were sacrificed 4 weeks after the surgery, and the knee joints were histologically evaluated (Fig. 1A).
to their germ line and then production of chimera mice was confirmed. Green ES FM260 ES cells were used to determine whether the repair tissue in the defect derived from the transplanted cells. Both types of cells were separately cultured by us on feeder cells (R. J.) inactivated by mitomycin C. The ES cells were embedded into a collagen gel (type-I collagen obtained from porcine Achilles tendon; Nitta Gelatine, Osaka, Japan) at 48C at a cell density of 107 cells/mL, and were gelated in 15 mL aliquots at 378C. Surgery The rats were anesthetized using an intraperitoneal injection of ketamine (10 mg/100 g; Sankyo Co. Ltd., Tokyo, Japan) and xylazine (1 mg/100 g; Bayer Co. Ltd., Tokyo, Japan). The knee joints were opened via the parapatellar-medial approach and the patella was retracted. An osteochondral defect (2 mm in diameter and 2 mm in depth) was created on the patellar groove of each femur with a hand drill (Fig. 1B) and the gel was inserted into the defect (Fig. 1C). The collagen gels were implanted into the defect within 60 min after cell seeding. The cell seeded collagen gels were refrigerated at 48C until just before they were embedded. In both cases (with or without ES cells), the gel volume was 15 mL. The calculated volume of the defect was 6.28 mL. When the gel was placed in the hole, some of the gel protruded from the defect. However, by using gauze to remove the exuded water that extruded from the gel, the gel was confined to the defect hole. The rectus femoris and patella were then carefully replaced over the defect and the wound was sutured with nylon thread. The recipient animals received daily subcutaneous injections of cyclosporin (14 mg/ kg; Novartis Pharma AG, Basel, Switzerland).
Preparation of ES Cells CCE ES cells obtained from 129/Sv/Ev mice were kindly provided by Dr. Elizabeth J. Robertson (Department of Molecular and Cellular Biology, Harvard University). Green ES FM260 ES cells obtained from 129SV (D3) - Tg (NCAG-EGFP) CZ - 001 - FM260Osb mice were kindly provided by Professor Masaru Okabe (Genome Information Research Center, Research Institute for Microbial Diseases, Osaka University). Green ES FM260 ES cells are GFP (green fluorescent protein) gene transfected ES cells. GFP can function as a protein tag. Green ES FM260 ES cells were electroporated with pCAGGSN-EGFP DNA linearized by HindIII. The ES cell lines were transmitted
Tail-Suspension Procedure and Limb-Immobilization Procedure In order to control the mechanical environment within the knee joint, the hind limbs of all the rats were elevated using a tail-suspension procedure.14 With this technique both the hind limbs become non-weight bearing whilst the forelimbs are weight bearing, and can be used for locomotion and access to food and water (Fig. 1D). In the JI group, a Kirschner wire (0.7 mm in diameter) was inserted percutaneously through the middle part of the tibia into the femur in order to immobilize the knee joint (Fig. 1E).15 The joints were immobilized at an
Table 1. Experimental Group and Loading Condition No Group
Side
Joint free group (JF group)
Right Left
ES-cellsþcollagen gel Collagen gel
Joint immobilized group (JI group)
Right Left
ES-cellsþcollagen gel Collagen gel
DOI 10.1002/jor
Transplantation
Loading Condition
CCE
FM260
Loading Loading
5 5
4 4
Immobilized Immobilized
5 5
4 4
JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2008
12
NAKAJIMA ET AL.
Figure 1. Experimental design. (A) Experimental procedure. After a lapse of 1 week postoperatively, tail-suspension and joint-immobilization procedures were performed. (B) The defect depth was constantly maintained at 2.0 mm by hand drill with a polyester guide. (C) The skin and knee joint capsule were incised at the medial side of the patella and patella tendon. The patella was retracted outside of the patella groove and created the defect. The site of transition between the patella and patella tendon was used to create the defect when the knee joints were held at a 908 flexion angle. (D) The rats were elevated off their hind limbs by tail suspension. A sling was attached to the tail with adhesive tape and was put on the guide rail through a clip. The length of the sling was adjusted so that the limbs did not touch the ground. The clip rotated freely and moved on the guide rail freely. Therefore, the rats could walk on the forelimb freely inside cage. (E) In the jointimmobilized group, a Kirschner wire was inserted from the tibial tuberosity to the diaphysis of the femur at a knee joint angle of 908 to keep it from cutting into the knee joint capsule. (F) The patella did not cover over the defect hole under a knee angle at 908 (1). The defect hole was totally covered by the patella at maximum-flexed knee angle (2). In the joint-free group, the rats frequently moved their hind limbs, as well as forelimbs, by associated movement when they locomoted with their forelimbs.
JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2008
DOI 10.1002/jor
IN VIVO MECHANICAL CONDITION AND CARTILAGE TISSUE IN ES CELLS
angle of 908, where the defects were not totally covered by the patella (Fig. 1F). The end of the wire was placed subcutaneously in order to prevent infection.
Histological Evaluation The rats were sacrificed 4 weeks after the surgery. The distal femur was fixed using 10% buffered formalin solution, and the tissues decalcified and sectioned. Hematoxylin-eosin, toluidine blue, or safranin O and fast green, type-I and type-II collagen immunohistochemical stainings and GFP immunohistochemical staining were performed. Each histological view was graded using the histological scale shown in Table 2,5 where the histological features were divided into five categories: cell morphol-
Table 2. Histological Grading Scale for the Defects of Cartilage Category Cell morphology Hyaline cartilage Mostly hyaline cartilage Mostly fibro cartilage Mostly noncartilage Noncartilage only Matrix-staining (metachromasia) Normal (compared with host adjacent cartilage) Slightly reduced Markedly reduced No metachromatic stain Surface regularity Smooth (concave;>3/4, convex;2/3b 1/3–2/3b
