Mechanobiology: Cartilage and Chondrocyte [3] 1586034367

Citation preview

Biomedical and

?

Health Research

Mechanobiology: Cartilage and Chondrocyte

Volume 3

Edited by J.-F. Stoltz

GE AND CHONDROCYTE

'

!'

■

>

Biomedical and Health Research Volume 61 Recently published in this series: Vol. 59.

M. Wolman and R. Manor (Eds.), Doctors’ Errors and Mistakes of Medicine: Must Health Care Deteriorate?

Vol. 58.

S. Holm and M. Jonas (Eds.), Engaging the World: The Use of Empirical Research in Bioethics and the Regulation of Biotechnology

Vol. 57. Vol. 56.

A. Nosikov and C. Gudex (Eds.), EUROHIS: Developing Common Instruments for Health Surveys P. Chauvin and the Europromed Working Group (Eds.), Prevention and Health Promotion for the Excluded and the Destitute in Europe

Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol.

55. 54. 53. 52. 51. 50. 49. 48. 47.

Vol. Vol. Vol. Vol.

46. 45. 44. 43.

J. Matsoukas and T. Mavromoustakos (Eds.), Drug Discovery and Design: Medical Aspects I. M. Shapiro, B.D. Boyan and H.C. Anderson (Eds.), The Growth Plate C. Huttin (Ed.), Patient Charges and Decision Making Behaviours of Consumers and Physicians J. -F. Stoltz (Ed.), Mechanobiology: Cartilage and Chondrocyte, Vol. 2 G. Lebeer (Ed.), Ethical Function in Hospital Ethics Committees R. Busse, M. Wismar and P.C. Berman (Eds.), The European Union and Health Services T. Reilly (Ed.), Musculoskeletal Disorders in Health-Related Occupations H. ten Have and R. Janssens (Eds.), Palliative Care in Europe - Concepts and Policies H. Aldskogius and J. Fraher (Eds.), Glial Interfaces in the Nervous System - Role in Repair and Plasticity I. Philp (Ed.), Family Care of Older People in Europe

Vol. 42. Vol. 41. Vol. 40. Vol. 39. Vol. Vol. Vol. Vol.

38. 37. 36. 35.

Vol. 34. Vol. 33. Vol. 32. Vol. Vol. Vol. Vol. Vol.

31. 30. 29. 28. 27.

L. Turski, D.D. Schoepp and E.A. Cavalheiro (Eds.), Excitatory Amino Acids: Ten Years Later R. Coppo and Dr. L. Peruzzi (Eds.), Moderately Proteinuric IgA Nephropathy in the Young B. Shaw, G. Semb, P. Nelson, V. Brattstrom, K. Moisted and B. Prahl-Andersen, The Eurocleft Project 1996-2000 J. -F. Stoltz (Ed.), Mechanobiology: Cartilage and Chondrocyte T.M. Gress (Ed.), Molecular Pathogenesis of Pancreatic Cancer H. Leino-Kilpi, M. Valimaki, M. Arndt, T. Dassen, M. Gasull, C. Lemonidou, P.A. Scott, G. Bansemir, E. Cabrera, H. Papaevangelou and J. Me Parland, Patient’s Autonomy, Privacy and Informed Consent ’ J.-M. Graf von der Schulenburg (Ed.), The Influence of Economic Evaluation Studies on Health Care Decision-Making N. Yoganandan and F.A. Pintar (Eds.), Frontiers in Whiplash Trauma J.M. Ntambi (Ed.), Adipocyte Biology and Hormone Signaling F. F. Pari, Estrogens, Estrogen Receptor and Breast Cancer M. Schlaud (Ed.), Comparison and Harmonisation of Denominator Data for Primary Health Care Research in Countries of the European Community G. J. Bellingan and G.J. Laurent (Eds.), Acute Lung Injury: From Inflammation to Repair H. H. Goebel, S.E. Mole and B.D. Lake (Eds.), The Neuronal Ceroid Lipofuscinoses (Batten Disease) B.J. Njio, B. Prahl-Andersen, G. ter Heege, A. Stenvik and R.S. Ireland (Eds.), Quality of Orthodontic Care - A Concept for Collaboration and Responsibilities B.J. Njio, A. Stenvik, R.S. Ireland and B. Prahl-Andersen (Eds.), EURO-QUAL J.-F. Stoltz, M. Singh and P. Riha, Hemorheology in Practice G. Pawelec (Ed.), EUCAMBIS: Immunology and Ageing in Europe A.M.N. Gardner and R.H. Fox, The Venous System in Health and Disease P.A. Frey and D.B. Northrop (Eds.), Enzymatic Mechanisms

ISSN 0929-6743

Mechanobiology: Cartilage and Chondrocyte Volume 3 Edited by

J.-F. Stoltz Hematologie-Hemorheologie, UMR CNRS 7563, Faculte de Medecine, Universite Henri Poincare, Vandcevre-les-Nancy, France

X. MTItmi INSTITUTES OF BEMJH mn LIBRARY ffe-.

). Using its definition, we also obtain F)Tr (i\

— = exp(£/Am)E/(er).

(9)

Therefore, the derivative takes the form

(10) A user-defined fibrillar element (Appendix A) was incorporated into the commercial finite element code ABAQUS, which was used to work with the existing porous element for the nonfibrillar matrix. Finite element solutions were obtained for tensile and unconfined compression testing. Analytical solu¬ tions were found (Appendix B) and used to test the finite element results when fluid pressurization was not considered.

184

L.P. Li and W. Herzog / The role of viscoelasticity of collagen fibers

3. Results For convenience of derivation of the closed form solutions used to test the finite element results, the effect of finite deformation was not included in Figs 1-3. Finite deformation theory was used for all other cases (Figs 4-9). Consider a 5-step tensile relaxation test. In each step, 2% strain was applied in the direction parallel to the articular surface at a rate of 0.15%/s, and then sufficient time was given to allow for full relaxation (Charlebois, McKee and Buschmann, J. Biomech. Eng. 126 (2004), 129-137). We have verified that the fluid has negligible impact on the mechanical response in tension using a fibril reinforced model (results not shown). Therefore, in our simulations, the tensile stress was determined analytically (in Appendix B) using the hereditary integral (1), once the elastic stress of the nonfibrillar matrix (Eme) had been added. The elastic properties (given in the caption of Fig. 1) were obtained using the data measured at equilibrium. The parameters for the relaxation function (2) (three exponentials were taken) were extracted by fitting the model to the experimental data (legend in Fig. 1). The same elastic and viscoelastic properties determined for the tensile test (for the horizontal di¬ rection) were then applied to the horizontal fibers in the analysis of unconfined compression (closed form solution given by (B.4)), and surprisingly, no viscoelastic response was visible when fluid pres¬ surization was absent (Fig. 2). The transient response shown in Fig. 2 was almost all produced by fluid flow/pressure. Increasing values of Poisson’s ratio were examined to determine if Poisson’s ratio could account for the loss of viscoelasticity in compression (Fig. 3a). The ratio of peak and equilibrium stresses did not increase significantly with increased compression rates (case iii. Fig. 3b), or with greater collagen viscoelastic relaxation (case ii, Fig. 3b), or with faster collagen viscoelastic dissipation (case iv, Fig. 3b). The impact of collagen viscoelasticity on the fluid pressurization was then examined for strain rates of 0.15 and 15%/s, respectively. In both cases, the viscoelastic dissipation in the radial fibers resulted

1.6

Experimental — Model g = 0.870, 7=10

a

g = 0.036, X2=100

CL

1.2

g = 0.273, X3=1000

C/5

0.8 c w

0.4

0.0 40000

L

L

80000

120000

160000

Time (s) Fig. 1. Simulation of a 5-step tensile relaxation test using the viscoelastic model. 2% tensile strain was applied at each step at a rate of 0.15%/s. The Young’s modulus of the nonfibrillar matrix was Em ~ 0.25 MPa. The fibrillar modulus, obtained by fitting the tensile stress at equilibrium, was El = (0.5 + 250er) MPa (the instantaneous modulus Gr(0)E^ = 2.179E/). The effect of fluid flow was not considered in this case, and thus the analytical solution could be obtained using Eqs (B.2) and (B.3).

L.P. Li and W. Herzog / The role of viscoelasticity of collagen fibers 80

Fluid Flow Considered .Total Compressive Axial Stress Fluid Flow Not Considered - Total Compressive Axial Stress -Radial Fibrillar Stress

60

2a

185

;

40

uH on 20

0 J—i—i—i—I—i—i—i—i_I

0

500

i

i

i

1000

i

I

1500

i

i

i

i

1

i

i

i

2000

i

I

2500

Time (s) Fig. 2. Compressive axial stress and radial fibrillar stress in unconfined compression predicted by the viscoelastic model. The transient response could be neglected if fluid pressurization in the tissue were absent. 2% compressive axial strain was applied at each step at a rate of 0.15%/s. The moduli were the same as those for Fig. 1. In addition, Poisson’s ratio of the nonfibrillar matrix was Um = 0.3. The permeability was k = 0.003 exp(10 x dilatation) mm4/Ns [27], and the void ratio was 3.5 when applicable. The radius and thickness of the disk were 1.5 and 1.0 mm, respectively.

Time (s)

Time (s)

a)

b)

Fig. 3. Compressive axial stress for the first step shown in Fig. 2 when fluid pressurization was not considered. The conditions were altered as follow: (a) vm was changed from 0.30 to 0.42 and 0.48; (b) conditions were altered in reference to case (i) as follow: (ii) *)= f &{NiZ dVo = 0, Jv„

where ur and uz are the vectors containing the nodal displacements. Supposing uz is an approximation for uz, we need to find Au2 so that

Flz(uz+Auz) = 0.

(A.3)

Therefore, letting dFz be equal to A FI results in

^(uz)Mi^-Flz(uz).

(A.4)

Here — Fz are elements of the residual vector, and kfj = dFz/duJz are elements of the Jacobian matrix, both of which must be evaluated in the user subroutine. If the fibrils are in tension in the axial direction,

l)

dal de, jNiz dVo,

Jv0 dez Qui

(A.5)

[ ^p-N^N^dVo.

V0 dez

Similarly, if the fibrils are in tension in the radial direction (and thus in the circumferential direction),

kr Ki]

dal Vo L dsr

N. N +&AMi iyi,r1 v?,r ' 0 9

dee

rz

dVn

(A.6)

and —FI defined in (A.2) are elements of the residual vector.

Appendix B. Viscoelastic stresses in uniaxial tensile and compression testing in the absence of fluid pressurization A multi-step relaxation test can be defined by the strain and strain rate as

F=l£k~l \ek

att = 0’ at t = tk,

._(cift

•

+ .... A

•

A A

-0 020

A -0 025 J A velocity # pressure

| A^locty ♦ pressure]

Fig. 6. (A) The predicted hydrostatic pressure (MPa) within the gel was greatest at the center and zero at the outer surface. (B,C,D) The predicted fluid flow varied non-linearly along each of the central axes of the gel and was greatest in the Z-direction due to the high pressure gradients across the thickness. All results correspond to the point of maximal peg displacement.

inhibited proteoglycan (PG) synthesis but not protein synthesis relative to the unloaded controls. The 20% displacement loading also inhibited PG synthesis to the same extent as 10% and did not affect protein synthesis, while the 5% displacement did not significantly decrease PG or protein synthesis. These results suggest that there may be a threshold amplitude of cyclic tension that inhibits chondrocyte production of proteoglycans. This response is also consistent with previous experiments where 10% dis¬ placement at 1.0 Hz decreased PG synthesis in chondrocytes and fibrochondrocytes [24], However, the significant decreases in protein synthesis seen in the previous study were not seen here. This inconsis¬ tency may be due to variability in cellular activity between animals or to a relatively small inhibition of protein synthesis, making any effect difficult to consistently detect at statistically significant levels. A small number of previous studies have investigated the effects of cyclic tension on chondrocyte ma¬ trix synthesis in monolayer culture. In general, higher magnitudes of tensile strain ( 17%) inhibited PG synthesis while smaller strains (~5%) increased synthesis [6,7,23]. Although, the local environment and cellular deformations in 3D and monolayer culture are significantly different, these results are generally consistent with the magnitude dependent inhibition of PG synthesis seen in the current study. The immunostaining and confocal microscopy images demonstrated that chondrocytes within the fib¬ rin gels maintained their phenotype in all loading groups. These cells stained positively for extracellu¬ lar type II collagen and aggrecan core protein indicating synthesis of the major articular cartilage ma¬ trix components. In addition, cells within the construct maintained rounded chondrocyte morphologies. Based on the spread morphology and lack of matrix staining, cells at the surface of the construct ap¬ peared to dedifferentiate similarly to chondrocytes in monolayer culture [25], In previous studies, cyclic

386

J.T. Connelly et al. / Influence of cyclic tension amplitude on chondrocyte matrix synthesis

tension induced a significant number of chondrocytes within the gel to take on a three-dimensional stel¬ late morphology [24]. However this phenomenon was not observed in the current study and may have been due to inter-animal variability. The results of the finite element analyses revealed important insights into the mechanical environ¬ ment of the fibrin construct during cyclic tensile loading. Within the central region (analyzed for matrix synthesis), there was a fairly wide range of strains. For example, the maximum principal strains at 10% displacement were approximately 1-5%. As a result, the chondrocytes within the central region most likely experienced a wide range of cellular deformations, which could have produced a variety of biosynthetic responses. Previous poroelastic simulations of the micro-environment in articular car¬ tilage predicted large variability in the local cellular strains even for uniform macroscopic strains due to mismatches in the cell-matrix mechanical properties and complexity of the extracellular matrix [8]. Such a non-uniform strain field therefore makes it difficult to isolate how specific amplitudes of dynamic tension affect matrix synthesis. This issue is further complicated by an overlap in the range of principal strains at different displace¬ ment magnitudes. As seen in Fig. 5, the maximum principal strains for the 10% and 20% loading groups overlapped between 1.4% and 5.4% strains, and the 5% and 10% loading groups overlapped between 0.6% and 3.0%. With a large number of elements in each group experiencing similar deformations, it is not surprising that no significant differences in matrix synthesis were detected between any of the loaded groups. Therefore, it is possible that cyclic tension inhibits PG synthesis in a dose dependent manner rather than as a threshold response, but an experimental system with a more uniform strain field is necessary to distinguish the two hypotheses. The poroelastic model of the fibrin gel also provided interesting information about the hydrostatic pressure and fluid flow distributions, which have been theorized as stimuli of chondrocyte matrix syn¬ thesis [3,11,14,18,22]. Due to the high permeability of the fibrin gel, the media easily flowed through the construct, and only minimal hydrostatic pressures developed (~—40 Pa). As a result, the total stress within the gel was primarily carried by the solid component, unlike articular cartilage tissue where the low permeability causes the total stress to include larger hydrostatic pressures [17]. It should be noted that in vivo fluid velocities in articular cartilage have also been estimated within the same order of mag¬ nitude as those calculated in this study [3,4], As seen with the calculated strain distributions, identifying the extent to which pressure, fluid flow or other aspects of the local environment regulate matrix synthe¬ sis is difficult due to the variability throughout the construct. The combination of loading experiments and finite element analysis allowed for detailed examination of how the amplitude of cyclic tension influences chondrocyte matrix synthesis, and several conclu¬ sions can be drawn from this study. (1) Dynamic tension consistently inhibits proteoglycan production. (2) This response depends on the magnitude of tension in either a threshold or dose dependent man¬ ner. (3) Mechanical parameters vary greatly within the central region of the fibrin construct, possibly inducing a variety of cellular responses. Clearly, future studies using an experimental system with a more uniform mechanical environment are necessary to isolate the effects of various amplitudes. In ad¬ dition, other loading parameters such as strain rate and duration of loading should be investigated to fully characterize the effects of oscillatory tension on chondrocyte matrix synthesis.

References [1] R.A. Ariens, H. Philippou, C. Nagaswami, J.W. Weisel, D.A. Lane and P.J. Grant, The factor XIII V34L polymorphism accelerates thrombin activation of factor XIII and affects cross-linked fibrin structure, Blood 96 (2000), 988-995.

J.T. Connelly et al. / Influence of cyclic tension amplitude on chondrocyte matrix synthesis

387

U] M.D. Buschmann, Y.A. Gluzband, A.J. Grodzinsky and E.B. Hunziker, Mechanical compression modulates matrix biosyn¬ thesis in chondrocyte/agarose culture, J. Cell Sci. 108 (Pt 4) (1995), 1497-1508. [3] M.D. Buschmann, Y.J. Kim, M. Wong, E. Frank, E.B. Hunziker and A.J. Grodzinsky, Stimulation of aggrecan synthesis in cartilage explants by cyclic loading is localized to regions of high interstitial fluid flow, Arch. Biochem. Biophys. 366 (1999), 1-7. [4] F. Eckstein, M. Tieschky, S. Faber, K.H. Englmeier and M. Reiser, Functional analysis of articular cartilage deformation, recovery, and fluid flow following dynamic exercise in vivo, Anat. Embryol. (Berl.) 200 (1999), 419-424. [5] R.W. Farndale, C.A. Sayers and A.J. Barrett, A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures, Connect. Tissue Res. 9 (1982), 247-248. [6] T. Fujisawa, T. Hattori, K. Takahashi, T. Kuboki, A. Yamashita and M. Takigawa, Cyclic mechanical stress induces extra¬ cellular matrix degradation in cultured chondrocytes via gene expression of matrix metalloproteinases and interleukin-1 J. Biochem. (Tokyo) 125 (1999), 966-975. [7] K. Fukuda, S. Asada, F. Kumano, M. Saitoh, K. Otani and S. Tanaka, Cyclic tensile stretch on bovine articular chondro¬ cytes inhibits protein kinase C activity, J. Lab. Clin. Med. 130 (1997), 209-215. [8] F. Gudak and V.C. Mow, The mechanical environment of the chondrocyte: a biphasic finite element model of cell-matrix interactions in articular cartilage, J. Biomech. 33 (2000), 1663-1673. [9] C.J. Hunter, S.M. Imler, P. Malaviya, R.M. Nerem and M.E. Fevenston, Mechanical compression alters gene expression and extracellular matrix synthesis by chondrocytes cultured in collagen I gels, Biomaterials 23 (2002), 1249-1259. [ 10] C.J. Huntei and M.E. Fevenston, The influence of repair tissue maturation on the response to oscillatory compression in a cartilage defect repair model. Biorheology 39 (2002), 79-88. [11] Y.J. Kim, F.J. Bonassar and A.J. Grodzinsky, The role of cartilage streaming potential, fluid flow and pressure in the stimulation of chondrocyte biosynthesis during dynamic compression, J. Biomech. 28 (1995), 1055-1066. [12] Y.J. Kim, R.F. Sah, J.Y. Doong and A.J. Grodzinsky, Fluorometric assay of DNA in cartilage explants using Hoechst 33258, Anal. Biochem. 174 (1988), 168-176. [13] Y.J. Kim, R.F. Sah, A.J. Grodzinsky, A.H. Plaas and J.D. Sandy, Mechanical regulation of cartilage biosynthetic behavior: physical stimuli, Arch. Biochem. Biophys. 311 (1994), 1-12. [14] P. Malaviya and R.M. Nerem, Fluid-induced shear stress stimulates chondrocyte proliferation partially mediated via TGFbetal, Tissue Eng. 8 (2002), 581-590. [15] R.F. Mauck, M.A. Soltz, C.C. Wang, D.D. Wong, P.H. Chao, W.B. Valhmu, C.T. Hung and G.A. Ateshian, Functional tissue engineering of articular cartilage through dynamic loading of chondrocyte-seeded agarose gels, J. Biomech Em 122 (2000), 252-260. [16] K. Messner and J. Gao, The menisci of the knee joint. Anatomical and functional characteristics, and a rationale for clinical treatment, J. Anat. 193 (Pt 2) (1998), 161-178. [17] N. Mukherjee and J.S. Wayne, Load sharing between solid and fluid phases in articular cartilage: II - Comparison of experimental results and u-p finite element predictions, J. Biomech. Eng. 120 (1998), 620-624. [ 18] J.J. Parkkinen, J. Ikonen, M.J. Lammi, J. Laakkonen, M. Tammi and H.J. Helminen, Effects of cyclic hydrostatic pressure on proteoglycan synthesis in cultured chondrocytes and articular cartilage explants, Arch. Biochem. Biophys. 300 (1993), 458^165. [19] J.J. Parkkinen, M.J. Lammi, H.J. Helminen and M. Tammi, Local stimulation of proteoglycan synthesis in articular carti¬ lage explants by dynamic compression in vitro, J. Orthop. Res. 10 (1992), 610-620. [20] R.L. Sah, Y.J. Kim, J.Y. Doong, A.J. Grodzinsky, A.H. Plaas and J.D. Sandy, Biosynthetic response of cartilage explants to dynamic compression, J. Orthop. Res. 7 (1989), 619-636. [21] D.H. Sierra, A.W. Eberhardt and J.E. Lemons, Failure characteristics of multiple-component fibrin-based adhesives, J. Biomed. Mater. Res. 59 (2002), 1-11. [22] R.L. Smith, B.S. Donlon, M.K. Gupta, M. Mohtai, P. Das, D.R. Carter, J. Cooke, G. Gibbons, N. Hutchinson and D.J. Schurman, Effects of fluid-induced shear on articular chondrocyte morphology and metabolism in vitro J Orthop Res 13(1995), 824-831. [23] T. Toyoda, S. Saito, S. Inokuchi and Y. Yabe, The effects of tensile load on the metabolism of cultured chondrocytes Clin Orthop. (1999), 221-228. [24] E.J. Vanderploeg, S.M. Imler, K.R. Brodkin, A.J. Garcia and M.E. Levenston, Oscillatory tension differentially modulates matrix metabolism and cytoskeletal organization in chondrocytes and fibrochondrocytes, J. Biomech. (in press). [25] K. von der Mark, V. Gauss, H. von der Mark and P. Muller, Relationship between cell shape and type of collagen synthe¬ sised as chondrocytes lose their cartilage phenotype in culture, Nature 267 (1977), 531-532.

Biorheology 41 (2004) 389-399 IOS Press

389

Adipose-derived adult stem cells for cartilage tissue engineering Farshid Guilak3’*, Hani A. Awada, Beverley Fermora, Holly A. Leddya and Jeffrey M. Gimbleb a Departments of Surgery and Biomedical Engineering, Duke University Medical Center, Durham, NC 27710, USA b Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA 70808, USA Abstract. Tissue engineering is a promising therapeutic approach that uses combinations of implanted cells, biomaterial scaf¬ folds, and biologically active molecules to repair or regenerate damaged or diseased tissues. Many diverse and increasingly complex approaches are being developed to repair articular cartilage, with the underlying premise that cells introduced exoge¬ nously play a necessary role in the success of engineered tissue replacements. A major consideration that remains in this field is the identification and characterization of appropriate sources of cells for tissue-engineered repair of cartilage. In particular, there has been significant emphasis on the use of undifferentiated progenitor cells, or “stem” cells that can be expanded in culture and differentiated into a variety of different cell types. Recent studies have identified the presence of an abundant source of stem cells in subcutaneous adipose tissue. These cells, termed adipose-derived adult stem (ADAS) cells, show character¬ istics of multipotent adult stem cells, similar to those of bone marrow derived mesenchymal stem cells (MSCs), and under appropriate culture conditions, synthesize cartilage-specific matrix proteins that are assembled in a cartilaginous extracellular matrix. The growth and chondrogenic differentiation of ADAS cells is strongly influenced by factors in the biochemical as well as biophysical environment of the cells. Furthermore, there is strong evidence that the interaction between the cells, the extracellular biomaterial substrate, and growth factors regulate ADAS cell differentiation and tissue growth. Overall, ADAS cells show significant promise for the development of functional tissue replacements for various tissues of the musculoskeletal system. Keywords: Articular cartilage, chondrocyte, pre-adipocyte, stromal cell, osteoarthritis, collagen, proteoglycan

1. Introduction Under normal physiologic circumstances, articular cartilage may function for decades as a nearly fric¬ tionless articulating surface in diarthrodial joints, while exposed to loads of several times body weight. This remarkable function is attributed to the unique structure and composition that determine the me¬ chanical properties of the cartilage extracellular matrix [51]. The cartilage extracellular matrix is main¬ tained by the metabolic activity of a sparse population of cells (chondrocytes) embedded within the tissue. Due to its lack of vascularity and the low metabolic activity of chondrocytes, articular cartilage exhibits a limited capacity for intrinsic repair. Isolated chondral or osteochondral lesions may be a sig¬ nificant source of pain and loss of function, and will rarely, if ever, heal spontaneously. Even minor lesions or injuries may lead to progressive damage and joint degeneration [34,35]. *Address for correspondence: Farshid Guilak, PhD, Orthopaedic Research Laboratories, Department of Surgery, Division of Orthopaedic Surgery, 375 MSRB, Box 3093, Duke University Medical Center, Durham, North Carolina 27710, USA. Tel.: + 1919 684 2521; Fax: +1 919 681 8490; E-mail: [email protected].

0006-355X704/$ 17.00 © 2004 - IOS Press and the authors. All rights reserved

390

F. Guilak et al. /Adipose-derived, adult stem cells

One of the potential explanations for the poor repair response of articular cartilage is the lack of a blood supply or a source of undifferentiated cells that can promote repair. To overcome these biological limitations, many surgeons have used drilling, abrasion, and microfracture of the subchondral bone to induce bleeding in a repair site in the cartilage [2,8,22,24,35]. This approach generally promotes the formation of a fibrocartilaginous repair tissue, suggesting that the cells responsible for cartilage repair do not differentiate into a true chondrocytic lineage. While the repair tissue is in some cases satisfactory and can decrease pain and morbidity in the short term, fibrocartilaginous repair tissue differs in its mechanical properties in comparison with native articular cartilage [28] and therefore may not function effectively as a long term replacement for normal tissue [55]. Other techniques for cartilage repair have included the transplantation of allograft or autograft cartilage in an effort to restore tissue function. One of the main challenges with these techniques is promoting the functional integration between the implanted graft and the host cartilage. Furthermore, the long-term efficacy of such methods remains unproven, and there is now evidence that significant morbidity may be associated with the donor site of autograft cartilage in the joint [43].

2. Tissue engineering of articular cartilage To address the clinical need for the repair or regeneration of articular cartilage, significant attention has turned to tissue engineering approaches [3,5,13,15,20,21,23,26,33,47-49,59,65,68]. Despite rapid and exciting advances in the field, few clinical applications have been developed for cartilage tissue en¬ gineering, and tissue engineers continue to face significant challenges in repairing or replacing tissues that serve a predominantly biomechanical function. At this writing, one cell-based therapy is available clinically for cartilage repair. The Carticel™ procedure (Genzyme Biosurgery, Cambridge, MA, USA) involves the isolation and amplification of autologous chondrocytes from articular cartilage of an unaf¬ fected region of the joint, followed by surgical implantation of cells into the cartilage defect [11]. A flap of autologous periosteal tissue is then used to cover the defect. Clinical outcomes of this procedure are good to excellent [10,50], although animal studies have not shown long-term success [9]. Furthermore, it is now apparent that the harvest procedure required to provide autologous cells may initiate joint degen¬ eration [43]. Therefore there still exists a continuing and unmet need for a readily available and abundant source of chondrocyte progenitor cells for cartilage tissue engineering.

3. Adult stem cells in tissue engineering The adult stem cell is defined as an “undifferentiated (unspecialized) cell that is found in a differen¬ tiated (specialized) tissue; it can renew itself and become specialized to yield all of the specialized cell types of the tissue from which it originated” [39]. In other words, the adult stem cell retains the capacity for self-renewal as well as the potential for differentiation into one or more specialized cell types. It is now apparent that many adult tissues harbor cells with the potential to differentiate into multiple cell types. These cells have been described most commonly as mesenchymal stem cells (or MSCs), but also as multipotential adult stem cells, human marrow stromal cells, or mesenchymal progenitors [14,16,17, 37,52,58,61-63,67,72]. Due to their accessibility, adult stem cells have been used extensively in a variety of tissue engineering applications. Numerous sources of cells with multipotent or pluripotent differentiation capabilities have been found in various musculoskeletal tissues, including bone marrow [14,61,63], adipose tissue [20,29,

F. Guilak et al. / Adipose-derived adult stem cells

391

74,75], trabecular bone [58], periosteum [53,59], synovial tissue [18,57], muscle [36,44,73], and several other tissues. These cells have formed the basis for tissue-engineered repair of different musculoskele¬ tal tissues [19,23,65]. However, despite several common characteristics among these cells, significant differences exist in their proliferation and differentiation capabilities, their expression of various cell surface markers, their abundance and ease of harvest, and therefore, their potential utility in tissue engi¬ neering applications. Specific to articular cartilage, a number of tissue engineering approaches have explored embryonic or adult stem cells as a source of progenitor cells. For example, murine embryonic stem cells stimulated with bone morphogenetic protein 4 (BMP-4) have been shown to differentiate along the mesodermal lin¬ eages [40,54]. Subpopulations of the totipotent embryonic stem cells expressing the VEGF receptor (flk1) and/or PDGFR-alpha differentiated into the chondrogenic lineage when cultured in 3D pellets and ex¬ posed to TGF-/33. The chondrogenic lineage was determined by mRNA expression of cartilage-specific genes (proteoglycans and type II collagen) and synthesis of type II collagen protein. The chondrogenic differentiation was synergistically enhanced in the presence of platelet-derived growth factor (PDGF) and the subsequent culture with BMP-4 [54], The primary source of adult stem cells for cartilage tissue engineering has been from the bone marrow. Bone marrow-derived MSCs undergo chondrogenesis in vitro when maintained in pellet culture in the presence of TGF-/3, based on the expression and synthesis of proteoglycans and collagen type II [7,38,46,61,72]. Other tissues have also been shown as sources of chondrocyte progenitors, including muscle [1], periosteum [60], synovium [18,57], skin/foreskin [56], and trabecular bone [58,66],

4. Adipose tissue as a source of adult stem cells Adipose tissue harbors a population of multipotent progenitor cells that can be induced to differentiate along multiple mesodermal and ectodermal lineages under controlled in vitro culture conditions (Fig. 1) Digest and separate stromal cells from fat tissue & other cells