In the context of skeletal tissue repair, tissue regeneration therapy is the local application of autologous (host-derived) cells to promote reconstruction of tissue defects caused by trauma, disease or surgical procedures. The objective of the tissue regeneration therapy approach is to deliver high densities of repair- competent cells (or cells that can become competent when influenced by the local environment) to the defect site in a format that optimizes both initial wound mechanics and eventual neotissue production. For soft tissue repair, it is likely that an implant vehicle(s), will be required to 1) transport and constrain the autologous cells in the defect site and 2) provide initial mechanical stability to the surgical site. In an optimal system, it is likely that the vehicle will slowly biodegrade at a rate comparable to the production of neotissue and development of strength in the reparative tissue (1) .

The tissue regeneration therapy approach contrasts significantly with more passive approaches to wound repair in which no attempt is made to deliver or to recruit reparative cells to the defect site. For example, in the case of anterior cruciate ligament (ACL) repair with synthetic (presumably "inert") polymer

grafts, the healing process depends entirely on local cellular responses to initiate and control the incorporation of a permanent implant (2) .

Recently, more active devices have been tested using matrix scaffolds designed to deliver and/or to direct cellular processes. These have included, for example, tendon or ACL repair (3-7) , meniscus repair (8-11) and articular cartilage repair (12-15) . Alternatively, the use of locally delivered peptide factors, intended to stimulate recruitment of reparative cells and their attachment and/or differentiation, have also been investigated (16-19) .

In perhaps the best documented tendon repair experiments to date, Silver, Dunn and their colleagues have described extensive investigations of the performance of collagen fiber prostheses for Achilles tendon (3-5) and anterior cruciate ligament (ACL) (6,7) repair in rabbits. They report that at 52 weeks postimplantation in the Achilles tendon defect, the reconstructed tendon (prosthesis + repair tissue) was about 66% as strong as the normal tissue for all implants tested, including an autologous tendon graft and glutaraldehyde- or carbodiimide-crosslinked collagen fiber composites (5) . Both the autologous implants and the carbodiimide- crosslinked prostheses were observed to biodegrade rapidly, then regain strength rapidly as new tissue was produced. Glutaraldehyde cross-linked prostheses biodegraded much more slowly in the Achilles tendon model and became surrounded by a thick capsule that eventually stopped the degradation process. While the neotendon developed in these studies was similar to normal, it was not identical. For example, the crimp angle of the neotendon collagen was similar to normal tendon in all implants, but the length of the neotendon crimp was less than about 30% of normal for the collagen prosthetic devices. In addition, the moduli of the neotendons

formed from the more rapidly degrading implants (autologous tendon and carbodiimide-crosslinked collagen fibers) were significantly lower than for normal tendon. Finally, the neotendon observed did not assemble with the fascicle microarchitecture of normal tendon. These researchers conclude that the rate of degradation of the r -^sthesis, and the consequent transfer of load to the n _c w tissue, may be as important as the initial prosthesis tensile strength in determining the ultimate properties of the repair tissue (5) . A similar generation of neoligament was observed in the ACL implants after 20 weeks, although the recovery of strength of the tissue may be somewhat slower in the avascular synovial environment (7) .

Based on this evidence, it is clear that at least in the healthy animal, repair-competent cells can be recruited from the tissues surrounding defects in tendons and ligaments, and that these cells will initiate the production of neotissue. It remains unclear from these investigations to what extent the recruited cells represented differentiated phenotypes (e.g., tendon fibroblasts) , as opposed to undifferentiated pluripotent stem cells, or whether increased numbers of such cells would enhance the rate of synthesis or the microarchitecture and mechanical properties of the neotissue produced.

Many cell-mediated processes related to the production of skeletal tissue depend on the number of cells involved, both in the rate and magnitude of the effect. For example, in the in vi tro production of connective tissue, the rate of collagen gel contraction by fibroblasts embedded in the gel is dependent on the number of cells present in the culture (20) . A similar gel-contracting activity has also been correlated with cell density-dependent secretion of a contraction- promoting factor by endothelial cells (21) . In addition,

the extent of fibroblast orientation in cultures grown on collagen gels is directly related to the initial cell density (22) . This cell orientation effect has been correlated with the observation of "organizing centers" in the culture, the number of which has been suggested to be a direct indicator of morphogenetic capacity at the molecular and cellular levels (23) .

Cell density-dependent differentiation was clearly demonstrated in the culture of chick limb bud cells (24) . When cultured at very low density (10 ⁶ cells/35mm dish) , these cells do not exhibit chondrogenic or osteogenic properties. At "intermediate" cell culture densities (2 x 10 ⁶ cells/35mm dish) , the cells exhibit the maximum frequency of osteogenesis, while at still higher density (5 x 10 ⁶ cells) the maximum frequency of chondrocyte phenotypes is observed.

In each instance cited above, the number of cells initially present strongly influences the nature of cell-mediated processes involved in skeletal tissue formation and the rate at which these developmental and physiological processes occur. Therefore, in the reparative processes of skeletal tissues, Caplan and coworkers have hypothesized that some minimum threshold of cell number may be required at the repair site before formation of "normal" neotissue can occur (25) . Furthermore, in many cases, this minimum threshold may exceed the number of recruitable reparative cells, including less committed cells that can differentiate to repair competent phenotypes; therefore, the extent to which the reparative process can occur may be limited by this single parameter.

Preliminary investigations of the tissue regeneration therapy approach have recently been conducted in a tendon repair model in the Achilles tendon of the rabbit (25) . There were three components to this

model: the defect, the cells and the vehicle to deliver the cells to the defect site. The delivery vehicle in this model must restrain the cells at the defect site, stabilize the tissue mechanics, then slowly biodegrade as new tissue is produced.

The present invention relates to an implant for repair of a tissue defect, which implant comprises a physiologically compatible load-bearing member having means for securing under tension tissue adjacent to the defect to be repaired, means for supporting a tissue reparative cell mass in the defect and a tissue reparative cell mass supported thereby. In its simplest form, the invention involves the production of an appropriate polymeric material containing the cells around a fibrous, degradable fixation device which is then employed to secure the cells in the desired anatomic location. This approach is a general surgical method for delivering and securing autologous cells to soft tissue defects, including tendon, ligament, meniscus or muscle, in which the cell delivery device must be fastened at one or both ends to soft tissue interfaces.

While one preferred material for the gel matrix employed in the specific example above was composed of purified Type I collagen fibrils, other materials that can likewise be used include, for example, 1) cell- contracted collagen gels containing other components of the extracellular matrix, such as proteoglycans or glycosaminoglycans, glycoproteins (fibronectin, laminin, etc.), other collagens (e.g., Types II or IV), elastin and/or bioactive peptide growth factors or cytokines,- 2) other biopolymers such as fibrin; 3) synthetic biodegradable fibers made from such polymers as polylactic or polyglycolic acids, polycaprolactones, or polyamino acids, or their copolymers, which could be cast or wound into or onto the suture; or 4) composite structures such as collagen/polylactic acid structures.

In addition to simple single-filament sutures, multifilament devices produced by braiding, weaving or knitting biodegradable fibrous materials including sutures or collagen fibers or the like can also be used. Cells could in general be attached to such devices by cell- mediated processes such as contraction of collagen gels, or by non-cell-mediated physical or chemical processes such as gel-casting gelatin, or a winding of cell-containing fibrous or membranous structures around the device. Such implantation devices could have one or more needles attached at each end of the device to facilitate fixation to soft or hard tissues, and could be of a variety of geometries including planar, cylindrical or tubular construction, depending upon the specific tissue to be repaired, the mode of fixation of the implant and/or the method used to attach the cell-containing biomatrix combination to the implantation device.

The present invention relates to a device and method for implantation of any type of cells that will effect tissue repair. Although the invention is not limited to any particular cell type, a particularly preferred embodiment includes human mesenchymal stem cells (MSCs) , or any of their committed or differentiated progeny. The cells are preferably obtained from the animal for which the implant is intended, and can preferably be culture expanded prior to implant. The animal is preferably a human.

In a specific embodiment of this invention, methods have been demonstrated for culturing MSCs or tendon fibroblasts onto double-needle Dexon sutures by causing the cells to contract collagen gels around the central region of the sutures. The autologous cell/collagen gel/suture composite device can be directly implanted between the free ends of full-thickness tendon defects, such as for repair of the human Achilles tendon, ligament such as for repair of the anterior cruciate ligament, or cartilage such as for

repair of articular cartilage, meniscus, or the disc of the temporomandibular joint.

In the embodiment shown in Figure 1, implant 1 comprises a strand of suture material 2 and a gel matrix 4 containing reparative cells 6 and which has been contracted around central portion 8 of suture 2. Suture 2 has free ends 10 and 12 which are used to rejoin the tissue T adjacent the defect. As shown free ends 10 and 12 have been sewn into the body of the tissue thereby not only holding the ends of the tendon in place but also holding gel matrix 4 in position in the defect.

Figure 2 shows a mold assembly 15 which can be used to form an implant of the invention. Mold assembly 15 includes mold 16 in which the cell-containing gel matrix is formed around suture 2 which is shown here with needles 3 and 5 at the ends thereof. Tension wire 18 which holds suture 2 under tension in mold 16 and incubation dish 20 in which the matrix preparation is incubated to set the gel. A specific embodiment of this is described in the example below.

Example 1

A mold assembly was used to prepare an implant for repair of a tissue defect in accordance with the invention. Small, glass cylinders, 5 mm x 27 mm, which had had their ends fused shut, were cut longitudinally through the center to form glass, canoe-shaped molds. Stiff surgical wires were bent to form small, bow-shaped tension wires with ends shaped to set just 2 mm deep into the glass molds. The glass mold was placed into a 100 mm culture dish with a suture spanning the tension wire situated in the center of the mold in preparation for the gel suspension to be poured.

Autologous mesenchymal stem cells (4 x 10 ^s cells) were suspended in 0.5 ml of 2 X DMEM-LG and mixed thoroughly to create a single-cell suspension. Then 0.5 ml sterilized type I collagen solution (Pancogene S™, Gattefosse SA, Lyon, France; 3 mg/ml; dialyzed into O.OOIM HCl was added to the cell suspension and pipetted up and down to form a homogenous suspension of cells in the gel. This gel suspension was immediately poured into the prepared glass mold in the culture dish. The lid was placed over the dish and it was put into the incubator at 37° C for 15-20 minutes to set the gel. After gelation was complete, the dish was flooded with medium without serum until the glass mold was covered and put back into the incubator for 4-6 hours . Contraction of the gel by the cells occurred to the extent that the gel was detached from the walls of the mold and decreased in diameter and length by about 10%. If the cells are cultured in this apparatus for approximately 20 hours, the gel contracts to approximately 60% of its original radial dimension. At the 4 hour time point, the gel was firmly attached to the central suture, such that the suture and tension spring could be lifted out of the medium, the tension spring removed, and the gel implanted in the surgical defect as described.

Tissue repair devices prepared by this procedure were implanted in rabbit Achilles tendon defect model either with or without a Vicryl sheath. Histological observations from these implants at 1, 3 and 8 weeks indicate that neotendon tissues are formed as early as 1-3 weeks by this procedure. These early neotendon tissues are morphologically similar to tissues produced from tendon cell or MSC implantation in the Vicryl sheath repair model at later timepoints.

Cited Literature

1. Goodship AE and Cooke P. Bicompatibility of tendon and ligament prostheses . Critical Reviews in Biocompatibility 1986.-2 (4) :303-334.

2. Bonnarens F.O. and Drez, D. , Jr. Biomechanics of artificial ligaments and associated problems. In: Jackson DW, Drez Jr. D, Eds. The anterior cruciate deficient knee: New concepts in ligament Repair. St. Louis: C.V. Mosby Co., 1987;239-253.

3. Goldstein JD, Tria AJ, Zawadshy JP, Kato YP, Christiansen D, Silver FH. Development of a Reconstituted Collagen Tendon Prosthesis: A preliminary study. J Bone Jt SuriQ 1989;71A(8) .1183-1191.

4. Hsu SYC, Cheng JCY, Chong YW, Leung PC. Glutaraldehyde-treated bioprosthetic substitute for rabbit Achilles tendon. Biomaterials 1989;10:258-264.

5. Kato YP, Dunn MG, Zawadsky JP, Tria AJ, Silver FH. Regeneration of Achilles tendon with a collagen tendon Prosthesis: Results of a one-year implantation study. J Bone Jt Surq 1991;73A:561-574.

6. Kato YP, Dunn MG, Tria AJ, Zawadsky JP, Silver FH. Preliminary assessment of a collagen fiber ACL prosthesis. Proceedings of the 17th Annual Meeting of the Society for Biomaterials, Abstract 265, Scottsdale Arizona 1991.

7. Dunn MG, Tria AJ, Kato YP, Bechler JR, Ochner RS, Zawadsky JP and Silver FH. Anterior cruciate ligament reconstruction using a composite collagenous prosthesis. A biomechanical and histologic study in rabbits. Am. J. Sports Med. 1991;20:507-515.

8. Klompmaker J, Jansen HWB, Veth RPH, de Groot JH, Nijenhuis AJ, Pennings AJ. Porous polymer implant for

repair of meniscal lesions : A preliminary study in dogs . Biomaterials 1991;12:810-816.

9. Henning CE, Lynch MA, Yearout KM, Vequist SW, Stallbaumer RJ, Decker KA. Arthroscopic meniscal repair using an exogenous fibrin clot. Clin Ortho 1990;252:64-72.

10. Wood DJ, Minns RJ, Strover A. Replacement of the rabbit medial meniscus with a polyester-carbon fibre bioprosthesis. Biomaterials 1990; 11: 13-16.

11. Stone KR, Rodkey WG, Webber RJ, McKinney L, Steadman JR. Collagen-based prostheses for meniscal regeneration. Clin Ortho 1990;252:129-135.

12. Grande DA, Pitman MI, Peterson L, Menche 0, Klein M. The repair of experimentally produced defects in rabbit articular cartilage by autologous chondrocyte transplantation. J Ortho Res 1989;7:208-218.

13. Grande DA. Technique for healing lesions in cartilage. U.S. Patent Number 4,846,835, July 11, 1989.

14. von Schroeder HP, Kwan M, Amiel D, Coutts RD. The use of polylactic acid matrix and periosteal grafts for the reconstruction of rabbit knee articular defects. J Biomed Mat Res 1991;25:329-339.

15. Wakitani S, Kimura T, Hirooka A, et al. Repair of rabbit articular surfaces with allograft chondrocytes embedded in collagen gel. J Bone Joint Surg 1989;71-B:74- 80.

16. Wang EA, Rosen V, D'Alessandro JS, et al. Recombinant human bone morphogenetic protein induces bone formation. Biochem 1990;87:220-224.

17. Syftestad GT, Lucas PA, Ohgushi H, Caplan Al. Chondrogenesis as an in vi tro response to bioactive factors extracted from adult bone and nonskeletal tissues. In: Thomhill T, SennA, eds. Development and diseases of cartilage and bone matrix. , UCLA Symposium Volume, New York: Alan Liss, Inc., 1987;187-199.

18. Syftestad GT, Lucas PA, Caplan Al. The in vi tro chondrogenic response of limb bud mesenchyme to a water- soluble fraction prepared from demineralized bone matrix. Differentiation 1985;29:230-237.

19. Lucas PA, Syftestad GT, Caplan Al. A water-soluble fraction from adult bone stimulates the differentiation of cartilage in explants of embryonic muscle. Differentiation 1988;37:47-52.

20. Bell E, Ivarsson B, Merrill C. Production of a tissue¬ like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vi tro. Proc Natl Acad Sci USA 1979;76 (3) :1274-1278.

21. Guidry C, Hohn S, Hook M. Endothelial cells secrete a factor that promotes fibroblast contraction of hydrated collagen gels. J Cell Biol 1990;110:519-528.

22. Klebe RJ, Caldwell H, Milam S. Cells transmit spatial information by orienting collagen fibers. Matrix 1989;9:451-458.

23. Klebe RJ, Overfelt TM, Magnuson VL, Steffensen B, Chen D, Zardeneta G. Quantitative assay for morphogenesis indicates the role of extracellular matrix components and G proteins. Proc Natl Acad Sci USA 1991;88:9588-9592.

24. Caplan Al. Effects of the nicotinamide-sensitive teratogen 3-acetylpyridine on chick limb cells in culture. EXP Cell Res 1970;62:341-355.

25. Caplan Al, Fink DJ, Goto T. , Linton AE, Young RG, Wakitani S, Goldberg VM, Haynesworth SE. Mesenchymal Stem Cells and Tissue Repair In: Jackson DW et al. , eds. The Anterior Cruciate Ligament: Current and Future Concepts, New York: Raven Press, Ltd., 1993; 405-417.