Depth-dependent confined compression modulus of full-thickness bovine articular cartilage.

The objective of this study was to determine the equilibrium confined compression modulus of bovine articular cartilage as it varies with depth from the articular surface. Osteochondral samples were compressed by 8, 16, 24, and 32% of the cartilage thickness and allowed to equilibrate. Intratissue displacement within the cartilage was measured with use of fluorescently labeled chondrocyte nuclei as intrinsic, fiducial markers. Axial strain was then calculated in nine sequential 125 microns thick cartilage layers comprising the superficial 1,125 microns and in a 250 microns thick layer of cartilage adjacent to the cartilage-bone interface. Adjacent osteochondral cores were also tested in confined compression to determine the equilibrium stresses required to achieve the same levels of compression. Stress-strain data for each layer of each sample were fit to a finite deformation stress-strain relation to determine the equilibrium confined compression modulus in each tissue layer. The compressive modulus increased significantly with depth from the articular surface and ranged from 0.079 +/- 0.039 MPa in the superficial layer to 1.14 +/- 0.44 MPa in the ninth layer. The deepest layer 250 microns thick, had a modulus of 2.10 +/- 2.69 MPa. These moduli were markedly different from the apparent "homogeneous" modulus for full-thickness cartilage (0.38 +/- 0.12 MPa) and ranged from 21 to 560% of that value. The relatively low moduli and the compression-induced stiffening of the superficial layers suggest that these layers greatly affect the biomechanical behavior of cartilage, such as during confined compression testing. The delineation of the depth-dependent modulus provides a basis for detailed study of the relationship between the composition, structure, and function of cartilage in such processes as aging, repair, and degeneration.

Chondrocyte transplantation to articular cartilage explants in vitro.

The transplantation of chondrocytes has shown promise for augmenting the repair of defects in articular cartilage. This in vitro study examined the efficiency of the transplantation of bovine chondrocytes onto articular cartilage disks and the ability of the transplanted chondrocytes to subsequently synthesize and deposit proteoglycan. The radiolabeling of chondrocyte cultures with [3H]thymidine, followed by 4 days of chase incubation, resulted in the incorporation of 98% of the radiolabel into DNA (as assessed by susceptibility to DNase). At the end of the culture period, the [3H]DNA was stable, with a half-life of radioactivity loss into the medium of 73 days. With use of radiolabeled chondrocytes for quantitation, the efficiency of transplantation onto a cartilage substrate was 93 +/- 4% for seeding densities of as much as 650,000 cells per cm2 and a seeding duration of 1 hour. These findings were confirmed both by tracking cells stained with 5-chlormethylfluorescein diacetate and by quantitating DNA. During the 16 hours after seeding onto a cartilage substrate (in which the endogenous cells had been lysed by lyophilization), the transplanted cells synthesized sulfated proteoglycan in direct proportion to the number of cells seeded. Most (83%) of the newly synthesized proteoglycan was released into the medium rather than retained within the layer of transplanted cells and the recipient cartilage substrate. Comparative studies with lyophilized-rehydrated or live cartilage as the recipient substrate indicated a similar efficiency of chondrocyte seeding and proteoglycan synthesis by the seeded chondrocytes. The transplanted cells retained the chondrocyte phenotype, as judged by a high proportion of the [35S]macromolecules being in the form of aggrecan that was capable of aggregating with hyaluronan and link protein, as well as by immunostaining within and around the transplanted cells for type-II, but not type-I, collagen. These results indicate that the number of chondrocytes transplanted onto a cut cartilage surface greatly affects the level of matrix synthesis; this in turn may affect repair.

Effect of seeding duration on the strength of chondrocyte adhesion to articular cartilage.

Chondrocyte adhesion to cartilage may play an important role in the repair of articular defects by maintaining cells in positions where their biosynthetic products can contribute to the repair process. The objective of this in vitro study was to determine the effect of the duration of seeding time on the ability of chondrocytes to resist detachment from cartilage when subjected to mechanical perturbation (fluid-induced shear stress). Suspensions of adult bovine articular chondrocytes were prepared from primary, high-density monolayer cultures and infused into a parallel-plate shear-flow chamber where they settled onto 50-microm-thick sections of bovine articular cartilage at a density of approximately 20,000 cells/cm2. The chondrocytes were seeded and allowed to attach to the cartilage surface for specific durations (5-40 minutes) in medium including 10% serum at 22 degrees C, after which the cells were exposed to fluid flow-induced shear stresses (6-90 Pa). The fraction of detached cells at each shear stress was calculated from microscopic images. Shear stress was applied for 1 minute because this length of time was sufficient to induce steady-state cell detachment. Increasing the duration of cell seeding led to a more firm attachment of chondrocytes to cartilage. After 9 minutes of seeding, 50% cell detachment was induced by gravitational force alone. After 40 minutes of seeding, 50% detachment required 26 Pa of shear stress. Extrapolation of the data to account for the effect of repeated applications of cell suspensions to an individual cartilage substrate indicated that for a freshly prepared cartilage section, 50% detachment was induced by gravity after 25 minutes of seeding and by 2.3 Pa of shear stress after 40 minutes of seeding. The increase in resistance to shear stress-induced cell detachment with increasing seeding duration suggests that it may be beneficial to allow chondrocytes to stabilize in the absence of applied load for some time after chondrocyte transplantation for cartilage repair in vivo.

Integrative repair of articular cartilage in vitro: adhesive strength of the interface region.

The objective of this study was to quantify the strength of the repair tissue that forms at the interface between pairs of cartilage explants maintained in apposition in an in vitro culture system. Articular cartilage explants were harvested from calves and from adult bovine animals, dissected into uniform blocks, and incubated in pairs within a chamber that maintained a 4 x 5 mm area of tissue overlap. Following 1-3 weeks of incubation, integrative repair was assessed by testing samples in a tensile single-lap configuration to estimate adhesive strength. After incubation in medium containing 20% fetal bovine serum, the adhesive strength between pairs of calf cartilage blocks and pairs of adult bovine cartilage blocks increased at a rate of 7.0 and 10.5 kPa/week, respectively. This repair process appeared to be dependent on viable cells, since lyophilization of adult bovine cartilage before incubation completely inhibited the development of an interface with a measurable adhesive strength. The repair process was dependent on serum components in the medium. Incubation of sample pairs for 3 weeks in medium supplemented with 20% fetal bovine serum resulted in a relatively high proteoglycan content as well as a relatively high adhesive strength (34 kPa), whereas incubation in basal medium with or without 0.1% bovine serum albumin resulted in a 54-70% lower proteoglycan content and a 65-88% lower adhesive strength. Samples incubated for 3 weeks with serum also had a 20% higher DNA content than samples maintained in basal medium. Histological analysis indicated some cell division at the free surfaces of the explant and also occasional cells within the interface region between explants.

Depth- and strain-dependent mechanical and electromechanical properties of full-thickness bovine articular cartilage in confined compression.

Compression tests have often been performed to assess the biomechanical properties of full-thickness articular cartilage. We tested whether the apparent homogeneous strain-dependent properties, deduced from such tests, reflect both strain- and depth-dependent material properties. Full-thickness bovine articular cartilage was tested by oscillatory confined compression superimposed on a static offset up to 45%. and the data fit to estimate modulus, permeability, and electrokinetic coefficient assuming homogeneity. Additional tests on partial-thickness cartilage were then performed to assess depth- and strain-dependent properties in an inhomogeneous model, assuming three discrete layers (i = 1 starting from the articular surface, to i = 3 up to the subchondral bone). Estimates of the zero-strain equilibrium confined compression modulus (H(A0)), the zero-strain permeability (kp0) and deformation dependence constant (M), and the deformation-dependent electrokinetic coefficient (ke) differed among individual layers of cartilage and full-thickness cartilage. HiA0 increased from layer 1 to 3 (0.27 to 0.71 MPa), and bracketed the apparent homogeneous value (0.47 MPa). ki(p0) decreased from layer 1 to 3 (4.6 x 10(-15) to 0.50 x 10(-15) m2/Pa s) and was less than the homogeneous value (7.3 x 10(-15) m2/Pa s), while Mi increased from layer 1 to 3 (5.5 to 7.4) and became similar to the homogeneous value (8.4). The amplitude of ki(e) increased markedly with compressive strain, as did the homogeneous value: at low strain, it was lowest near the articular surface and increased to a peak in the middle-deep region. These results help to interpret the biomechanical assessment of full-thickness articular cartilage.

Video microscopy to quantitate the inhomogeneous equilibrium strain within articular cartilage during confined compression.

The objectives of this study were to develop a method to quantitate the displacement and strain fields within articular cartilage during equilibrium confined compression, and to use the method to determine the variation of the equilibrium confined compression modulus with depth from the articular surface in bovine cartilage. The method made use of fluorescently labeled chondrocyte nuclei as intrinsic fiducial markers. Articular cartilage was harvested from the patellofemoral groove of adult bovines and trimmed to rectangular blocks 5 mm long, 0.76 mm wide, and 500 microns deep with the articular surface intact. Test specimens were stained with the DNA binding dye Hoechst 33258, placed in a custom confined compression chamber, and viewed with an epifluorescence microscope equipped for video image acquisition. Image processing was used to localize fluorescing chondrocyte nuclei in uncompressed and compressed (approximately 17%) specimens, allowing determination of the intra-tissue displacement profile. Strain was determined as the slope of linear regression fits of the displacement data in four sequential 125-microns-thick layers. Equilibrium strains varied 6.1-fold from the articular surface through 500 microns of cartilage depth, with the greatest compressive strain in the superficial 125-microns layer and the least compressive strain in the two deepest 125-microns layers. Thus, the four successive 125-microns layers have moduli that are 0.44 (superficial), 1.07, 2.39, and 2.67 (deep) times the apparent modulus for a 500-microns thick cartilage sample assumed to be homogeneous.

Gelatin-based resorbable sponge as a carrier matrix for human mesenchymal stem cells in cartilage regeneration therapy.

Adult mesenchymal stem cells (MSCs), found in the bone marrow, have the potential to differentiate into multiple connective tissue types, including cartilage. In this study, we examined the potential of a porous gelatin sponge, Gelfoam, for use as a delivery vehicle for MSCs in cartilage regeneration therapy. Adult human MSCs (hMSCs) were seeded throughout the gelatin sponge after a 2-h incubation period. When cultured for 21 days in vitro in a defined medium supplemented with 10 ng/mL of TGF-beta 3, hMSC/Gelfoam constructs produced a cartilage-like extracellular matrix containing sulfated glycosaminoglycans (s-GAGs) and type-II collagen, as evident upon histologic evaluation. Constructs loaded with a cell suspension of 12 x 10(6) cells/mL produced an extracellular matrix containing 21 microg of s-GAG/microg of DNA after 21 days of culture. This production was more efficient than constructs loaded at higher or lower cell densities, indicating that the initial seeding density influences the ability of cells to produce extracellular matrix. When implanted in an osteochondral defect in the rabbit femoral condyle, Gelfoam cylinders were observed to be very biocompatible, with no evidence of immune response or lymphocytic infiltration at the site. Based on these observations we conclude that Gelfoam resorbable gelatin sponge is a promising candidate as a carrier matrix for MSC-based cartilage regeneration therapies.

Quantitation and localization of cartilage degeneration following the induction of osteoarthritis in the rabbit knee.

OBJECTIVE: To develop and apply a new video imaging technique to quantify and localize Indian ink staining of cartilage of the rabbit femorotibial joint after the induction of osteoarthritis by unilateral transection of the anterior cruciate ligament (ACLT). METHODS: Nine weeks after surgery, femora and tibiae from 11 ACLT and contralateral control knees were harvested and positioned to obtain calibrated gray-scale images of the ink-painted articular cartilage surfaces that are opposed with the knee in 90 degrees flexion. Images were processed so that areas of normal cartilage gave a relatively high reflectance score, whereas ink-stained fibrillated cartilage and exposed bone gave low scores. RESULTS: Comparison of the medial and lateral femoral condyles and tibial plateaus (MFC, LFC, MTP, LTP) of control and ACLT knees showed that the area of the MTP not covered by the meniscus had a significantly lower reflectance score (P < 0.001) than other areas. ACLT led to an 11% decrease (P < 0.001) in the overall reflectance score. The reflectance score decreased as a traditional morphological grading of degeneration increased. ACLT-induced degeneration had a predilection for the posteromedial aspects of the joint, and to a lesser extent, the anterolateral aspects. In the tibial plateaus, ACLT caused significant degeneration in the covered, but not the uncovered, areas. Image scores of opposing cartilage surfaces (i.e., MFC vs MTP and LFC vs LTP) were significantly (R = 0.56-0.70, P < 0.001) correlated in ACLT and control knees. DISCUSSION: Identification and characterization of cartilage areas prone to degeneration may be particularly useful for further analysis of biochemical and biomechanical mechanisms in osteoarthritis, as well as the efficacy of therapeutic interventions.