Phenotypic analysis of bovine chondrocytes cultured in 3D collagen sponges: effect of serum substitutes

Repair of damaged cartilage usually requires replacement tissue or substitute material. Tissue engineering is a promising means to produce replacement cartilage from autologous or allogeneic cell sources. Scaffolds provide a three-dimensional (3D) structure that is essential for chondrocyte function and synthesis of cartilage-specific matrix proteins (collagen type II, aggrecan) and sulfated proteoglycans. In this study, we assessed porous, 3D collagen sponges for in vitro engineering of cartilage in both standard and serum-free culture conditions. Bovine articular chondrocytes (bACs) cultured in 3D sponges accumulated and maintained cartilage matrix over 4 weeks, as assessed by quantitative measures of matrix content, synthesis, and gene expression. Chondrogenesis by bACs cultured with Nutridoma as a serum replacement was equivalent or better than control cultures in serum. In contrast, chondrogenesis in insulin-transferrin-selenium (ITS⁺³) serum replacement cultures was poor, apparently due to decreased cell survival. These data indicate that porous 3D collagen sponges maintain chondrocyte viability, shape, and synthetic activity by providing an environment favorable for high-density chondrogenesis. With quantitative assays for cartilage-specific gene expression and biochemical measures of chondrogenesis in these studies, we conclude that the collagen sponges have potential as a scaffold for cartilage tissue engineering.     

Chondrocyte transplantation for osteochondral defects with the use of suspension culture

Cartilage graft is considered to be useful in repairing chondral or osteochondral defects. One method of the cartilage graft is achieved by autologous chondrocyte transplantation following cell culture. However, chondrocytes change their phenotype during culture. We used costal chondrocytes cultured over agarose (suspension culture) as a source of graft materials. The suspension-cultured chondrocytes formed aggregate in culture. We first examined the expressions of cartilage-specific matrices of cultured chondrocytes after two weeks in culture. The chondrocytes cultured over agarose expressed more type II collagen mRNA than those cultured on plastic dishes did after two weeks in culture. Safranin O staining showed the presence of glycosaminoglycans in the chondrocyte culture over agarose, while glycosaminoglycans were not observed in the culture on plastic dishes. We then examined the changes of rat articular osteochondral defects after transplantation of suspension-cultured chondrocytes. The aggregate of suspension-cultured chondrocytes was easily picked up with forceps and transplanted in the osteochondral defects. The defects were filled with safranin O-stained hyaline cartilage tissue two weeks after chondrocyte transplantation. On the contrary, the fibrous materials, which were not stained with safranin O, were observed in the control defects. These results suggest that the suspension-cultured chondrocytes are useful for autologous cartilage grafts by preserving chondrocyte phenotype.     

Hyaline cartilage formation and enchondral ossification modeled with KUM5 and OP9 chondroblasts

What is it that defines a bone marrow‐derived chondrocyte? We attempted to identify marrow‐derived cells with chondrogenic nature and immortality without transformation, defining “immortality” simply as indefinite cell division. KUM5 mesenchymal cells, a marrow stromal cell line, generated hyaline cartilage in vivo and exhibited enchondral ossification at a later stage after implantation. Selection of KUM5 chondroblasts based on the activity of the chondrocyte‐specific cis‐regulatory element of the collagen α2(XI) gene resulted in enhancement of their chondrogenic nature. Gene chip analysis revealed that OP9 cells, another marrow stromal cell line, derived from macrophage colony‐stimulating factor‐deficient osteopetrotic mice and also known to be niche‐constituting cells for hematopoietic stem cells expressed chondrocyte‐specific or ‐associated genes such as type II collagen α1, Sox9, and cartilage oligomeric matrix protein at an extremely high level, as did KUM5 cells. After cultured OP9 micromasses exposed to TGF‐β3 and BMP2 were implanted in mice, they produced abundant metachromatic matrix with the toluidine blue stain and formed type II collagen‐positive hyaline cartilage within 2 weeks in vivo. Hierarchical clustering and principal component analysis based on microarray data of the expression of cell surface markers and cell‐type‐specific genes resulted in grouping of KUM5 and OP9 cells into the same subcategory of “chondroblast,” that is, a distinct cell type group. We here show that these two cell lines exhibit the unique characteristics of hyaline cartilage formation and enchondral ossification in vitro and in vivo. J. Cell. Biochem. 100: 1240–1254, 2007. © 2006 Wiley‐Liss, Inc.     

Is the repair of articular cartilage lesion by costal chondrocyte transplantation donor age-dependent? An experimental study in rabbits

The repair of chondral injuries is a very important problem and a subject of many experimental and clinical studies. Different techniques to induce articular cartilage repair are under investigation. In the present study, we have investigated whether the repair of articular cartilage folowing costal chondrocyte transplantation is donor age-dependent. Transplantation of costal chondrocytes from 4- and 24-week old donors, with artificially induced femoral cartilage lesion, was performed on fourteen 20-week-old New Zealand White male rabbits. In the control group, the lesion was left without chondrocyte transplantation. The evaluation of the cartilage repair was performed after 12 weeks of transplantation. We analyzed the macroscopic and histological appearance of the newly formed tissue. Immunohistochemistry was also performed using monoclonal antibodies against rabbit collagen type II. The newly formed tissue had a hyaline-like appearance in most of the lesions after chondrocyte transplantation. Positive immunohistochemical reaction for collagen II was also observed in both groups with transplanted chondrocytes. Cartilage from adult donors required longer isolation time and induced slightly poorer repair. However, hyaline-like cartilage was observed in most specimens from this group, in contrast to the control group, where fibrous connective tissue filled the lesions. Rabbit costal chondrocytes seem to be a potentially useful material for inducing articular cartilage repair and, even more important, they can also be derived from adult, sexually mature animals.     

Collagen XI chain misassembly in cartilage of the chondrodysplasia (cho) mouse

Molecular mechanisms controlling the assembly of cartilage-specific types II, IX and XI collagens into a heteropolymeric network of uniformly thin, unbanded fibrils are not well understood, but collagen XI has been implicated. The present study on cartilage from the homozygous chondrodysplasia (cho/cho) mouse adds support to this concept. In the absence of alpha1(XI) collagen chains, thick, banded collagen fibrils are formed in the extracellular matrix of cho/cho cartilage. A functional knock-out of the type XI collagen molecule has been assumed. We have re-examined this at the protein level to see if, rather than a complete knock-out, alternative type XI chain assemblies were formed. Mass spectrometry of purified triple-helical collagen from the rib cartilage of cho/cho mice identified alpha1(V) and alpha2(XI) chains. These chains were recovered in roughly equal amounts based on Coomassie Blue staining of SDS-PAGE gels, in addition to alpha1(II)/alpha3(XI) collagen chains. Using telopeptide-specific antibodies and Western blot analysis, it was further shown that type V/XI trimers were present in the matrix cross-linked to each other and to type II collagen molecules to form heteropolymers. Cartilage from heterozygous (cho/+) mice contained a mix of alpha1(V) and alpha1(XI) chains and a mix of thin and thick fibrils on transmission electron microscopy. In summary, the results imply that native type XI collagen molecules containing an alpha1(XI) chain are required to form uniformly thin fibrils and support a role for type XI collagen as the template for the characteristic type II collagen fibril network of developing cartilage.     

Stimulation by glucocorticoid of the synthesis of cartilage-matrix proteoglycans produced by rabbit costal chondrocytes in vitro

The effect of glucocorticoids on sulfated proteoglycan synthesis by rabbit costal chondrocyte cultures exposed to serum-free conditions has been examined. Low density cultures of rabbit costal chondrocytes were maintained on dishes coated with extracellular matrix produced by bovine corneal endothelial cells and exposed to a 9:1 mixture (v/v) of Dulbecco's modified Eagle's medium and Ham's F-12 medium supplemented with transferrin, high density lipoproteins, fibroblast growth factor, and insulin (Medium A). Chondrocytes maintained in the presence of Medium A supplemented with 10(-7) M hydrocortisone reorganized, at confluence, into a homogeneous cartilage-like tissue composed of round cells surrounded by a refractile matrix in which abundant thin collagen fibrils characteristic of type II collagen were observed. The cell ultrastructure and fibrils of the pericellular matrix were similar to those seen in vivo. In contrast, cells maintained in the presence of Medium A alone, once they reached confluence, formed a fibroblastic multilayer and produced thick collagen bundles. The level of 35SO4(2-) incorporated into large cartilage-specific proteoglycans in glucocorticoid-supplemented cultures was 33-fold higher than that of glucocorticoid-free cultures. The level of 35SO4(2-) incorporated into small ubiquitous proteoglycans was only 4-fold higher than that of glucocorticoid-free cultures. On the other hand, the level of [3H]glucosamine incorporated into hyaluronate in glucocorticoid-supplemented cultures was 4.5-fold lower than that of glucocorticoid-free cultures. Within 24 h of their addition to confluent cultures, hydrocortisone or dexamethasone markedly stimulated proteoglycan synthesis. This effect was not mimicked by androgens, estrogens, progesterone, or an inactive form of glucocorticoids such as deoxycorticosterone. This suggests that glucocorticoids have a direct and specific stimulatory effect on cartilage-specific proteoglycan synthesis and are essential for the maintenance of this synthesis in low density chondrocyte cultures.     

Development and phenotypic characterization of a high density in vitro model of auricular chondrocytes with applications in reconstructive plastic surgery

Cultivation of phenotypically stable auricular chondrocytes will have applications in autologous chondrocyte transplantation and reconstructive surgery of cartilage. Chondrocytes grown in monolayer culture rapidly dedifferentiate assuming a fibroblast-like morphology and lose their cartilage-specific pattern of gene expression. Three-dimensional high-density culture models mimic more closely the in vivo conditions of cartilage. Therefore, this study was undertaken to test whether the high-density cultures might serve as a suitable model system to acquire phenotypically and functionally differentiated auricular chondrocytes from porcine cartilage. Freshly isolated porcine auricular chondrocytes were cultured for 7 passages in monolayer culture. From each passage (passage 0 and 1-7) cells were introduced to high-density cultures and examined by transmission electron microscopy. Western blotting was used to analyse the expression of cartilage-specific markers, such as collagen type II and cartilage specific proteoglycan, fibronectin, cell adhesion and signal transduction receptor beta1-integrin, matrix metalloproteinases (MMP-9, MMP-13), cyclo-oxygenase (COX)-2 and the apoptosis commitment marker, activated caspase-3. When dedifferentiated auricular chondrocytes from monolayer passages 0-4 were cultured in high-density culture, they recovered their chondrocytic phenotype and formed cartilage nodules surrounded by fibroblast-like cells and synthesised collagen type II, proteoglycans, fibronectin and beta1-integrins. However, chondrocytes from monolayer passages 5-7 did not redifferentiate to chondrocytes even when transferred to high-density culture, and did not synthesize a chondrocyte-specific extracellular matrix. Instead, they produced increasing amounts of MMP-9, MMP-13, COX-2, activated caspase-3 and underwent apoptosis. Three-dimensional high-density cultures may therefore be used to obtain sufficient quantities of fully differentiated auricular chondrocytes for autologous chondrocyte transplantation and reconstructive plastic surgery.     

Development and characterization of a human articular cartilage‐derived chondrocyte cell line that retains chondrocyte phenotype

Chondrocytes, the only cell type present in articular cartilage, regulate tissue homeostasis by a fine balance of metabolism that includes both anabolic and catabolic activities. Therefore, the biology of chondrocytes is critical for understanding cartilage metabolism. One major limitation when studying primary chondrocytes in culture is their loss of phenotype. To overcome this hurdle, limited attempts have been made to develop human chondrocyte cell lines that retain the phenotype for use as a good surrogate model. In this study, we report a novel approach to the establishment and characterization of human articular cartilage‐derived chondrocyte cell lines. Adenoviral infection followed by culture of chondrocytes in 3‐dimensional matrix within 48 h post‐infection maintained the phenotype prior to clonal selection. Cells were then placed in culture either as monolayer, or in 3‐dimensional matrix of alginate or agarose. The clones were characterized by their basal gene expression profile of chondrocyte markers. Based on type II collagen expression, 21 clones were analyzed for gene expression following treatment with IL‐1 or BMP‐7 and compared to similarly stimulated primary chondrocytes. This resulted in selection of two clones that retained the chondrocyte phenotype as evidenced by expression of type II collagen and other extra‐cellular matrix molecules. In addition, one clone (AL‐4‐17) showed similar responses as primary chondrocytes when treated with IL‐1 or BMP‐7. In summary, this report provides a novel procedure to develop human articular cartilage‐derived chondrocyte cell lines, which preserve important characteristics of articular chondrocytes and represent a useful model to study chondrocyte biology. J. Cell. Physiol. 222: 695–702, 2010. © 2009 Wiley‐Liss, Inc.     

Human chondrocyte cell lines from articular cartilage of metatarsal phalangeal joints

Chondrocytes can be isolated from human adult cartilage from metatarsal phalangeal joints. After enzymatic digestion to isolate viable cells, confluent monolayers were obtained 2-4 weeks after the start of cell division. Chondrocytes cultures, initiated and maintained in HAM's F12 with bovine fetal serum without the addition of other growth factors, produced in vitro a matrix rich in collagen and proteoglycans. Although several studies reported phenotypic instability, our results showed that the cell retain for more than 5 months in culture their differentiated characteristics, including the ability to produce cartilage-specific molecules. Chondrocyte cell lines should be useful in studying the functions of these cells from normal and abnormal tissue and for pharmacological studies in vitro.     

Susceptibility of cartilage collagens type II, IX, X, and XI to human synovial collagenase and neutrophil elastase

The action of purified rheumatoid synovial collagenase and human neutrophil elastase on the cartilage collagen types II, IX, X and XI was examined. At 25 degrees C, collagenase attacked type II and type X (45-kDa pepsin-solubilized) collagens to produce specific products reflecting one and at least two cleavages respectively. At 35 degrees C, collagenase completely degraded the type II collagen molecule to small peptides whereas a large fragment of the type X molecule was resistant to further degradation. In contrast, collagen type IX (native, intact and pepsin-solubilized type M) and collagen type XI were resistant to collagenase attack at both 25 degrees C and 35 degrees C even in the presence of excess enzyme. Mixtures of type II collagen with equimolar amounts of either type IX or XI did not affect the rate at which the former was degraded by collagenase at 25 degrees C. Purified neutrophil elastase, shown to be functionally active against soluble type III collagen, had no effect on collagen type II at 25 degrees C or 35 degrees C. At 25 degrees C collagen types IX (pepsin-solubilized type M) and XI were also resistant to elastase, but at 35 degrees C both were susceptible to degradation with type IX being reduced to very small peptides. Collagen type X (45-kDa pepsin-solubilized) was susceptible to elastase attack at 25 degrees C and 35 degrees C as judged by the production of specific products that corresponded closely with those produced by collagenase. Although synovial collagenase failed to degrade collagen types IX and XI, all the cartilage collagen species examined were degraded at 35 degrees C by conditioned culture medium from IL1-activated human articular chondrocytes. Thus chondrocytes have the potential to catabolise each cartilage collagen species, but the specificity and number of the chondrocyte-derived collagenase(s) has yet to be resolved.     

Cell condensation in chondrogenic differentiation.

Reduction of intercellular spaces in the areas of prospective cartilage and bone formation (precartilage condensation) precedes chondrogenesis and may represent an important step in the process of cartilage differentiation during limb skeletogenesis. We have attempted to clarify the role of the microenvironment established during cell condensation, taking advantage of a tissue culture model system that allows condensation (i.e., increased cell density due to cell aggregation) and chondrogenic differentiation (i.e., synthesis of cartilage-specific extracellular matrix proteins, such as type II collagen and acquisition of a chondrocyte morphology) of chick embryo cartilage-derived undifferentiated cells. To prevent condensation cells were grown in carboxymethylcellulose and changes in the differentiation pathway were evaluated. In another series of experiments, we have separated single cells from the aggregated cells and analyzed their differentiation properties. Morphological analyses and the evaluation of type II collagen expression, at both the protein and the mRNA level, show that a reduced rate of cell clustering and cell to cell contact parallels a reduction of cell recruitment into the differentiation program. On the basis of our results, we suggest that the following cascade of events regulates the early stages of chondrocyte differentiation: (a) the acquisition of the ability to establish cell to cell contacts, (b) the formation of a permissive environment capable of activating the differentiation program, and (c) the expression of differentiation markers.     

Proteoglycan production by immortalized human chondrocyte cell lines cultured under conditions that promote expression of the differentiated phenotype

Large and small proteoglycans are essential components of articular cartilage. How to induce chondrocytes to repair damaged cartilage with normal ratios of matrix components after their loss due to degenerative joint disease has been a major research focus. We have developed immortalized human chondrocyte cell lines for examining the regulation of cartilage-specific matrix gene expression. However, the decreased synthesis and deposition of cartilage matrix associated with a rapid rate of proliferation has presented difficulties for further examination at the protein level. In these studies, proteoglycan synthesis was characterized in two chondrocyte cell lines, T/C-28a2 and tsT/AC62, derived, respectively, from juvenile costal and adult articular cartilage, under culture conditions that either promoted or decreased cell proliferation. Analysis of proteo[36S]glycans by Sepharose CL-4B chromatography and SDS-PAGE showed that the large proteoglycan aggrecan and the small, leucine-rich proteoglycans, decorin and biglycan, were produced under every culture condition studied. In monolayer cultures, a high initial cell density and conditions that promoted proliferation (presence of serum for T/C-28a2 cells or permissive temperature for the temperature-sensitive tsT/AC62 cells) favored cell survival and ratios of proteoglycans expected for differentiated chondrocytes. However, the tsT/AC62 cells produced more proteoglycans at the nonpermissive temperature. Culture of cells suspended in alginate resulted in a significant decrease in proteoglycan production in all culture conditions. While the tsT/AC62 cells continued to produce a larger amount of aggrecan than small proteoglycans, the T/C-28a2 cells lost the ability to produce significant amounts of aggrecan in alginate culture. In addition, our data indicate that immortalized chondrocytes may alter their ability to retain pericellular matrix under changing culture conditions, although the production of the individual matrix components does not change. These findings provide critical information that will assist in the development of a reproducible chondrocyte culture model for the study of regulation of proteoglycan biosynthesis in cartilage.     

IL-10 overexpression differentially affects cartilage matrix gene expression in response to TNF-α in human articular chondrocytes in vitro

Cartilage-specific extracellular matrix synthesis is the prerequisite for chondrocyte survival and cartilage function, but is affected by the pro-inflammatory cytokine TNF-α in arthritis. The aim of the present study was to characterize whether the immunoregulatory cytokine IL-10 might modulate cartilage matrix and cytokine expression in response to TNF-α. Primary human articular chondrocytes were treated with either recombinant IL-10, TNF-α or a combination of both (at 10 ng/mL each) or transduced with an adenoviral vector overexpressing human IL-10 and subsequently stimulated with 10 ng/ml TNF-α for 6 or 24 h. The effects of IL-10 on the cartilage-specific matrix proteins collagen type II, aggrecan, matrix-metalloproteinases (MMP)-3, -13 and pro-inflammatory cytokines were evaluated by real-time RT-PCR and immunohistochemistry. Transduced chondrocytes overexpressed high levels of IL-10 which significantly up-regulated collagen type II expression. TNF-α suppressed collagen type II and aggrecan, but increased MMP and cytokine expression in chondrocytes compared to the non-stimulated controls. The TNF-α mediated down-regulation of aggrecan expression was significantly antagonized by IL-10 overexpression, whereas the suppression of collagen type II was barely affected. The MMP-13 and IL-1β expression by TNF-α was slightly reduced by IL-10. These results suggest that IL-10 overexpression modulates some catabolic features of TNF-α in chondrocytes.     

An established rat cell line expressing chondrocyte properties

Chondrocytes express a well-characterized set of marker proteins making these cells useful for studies on differentiation and regulation of gene expression. Because of the inherent instability of primary rat chondrocytes in culture, and because several rat chondrocyte genes have been cloned and characterized (including the collagen II promoter and enhancer), a rat chondrocyte cell line would be especially useful. To obtain this line we infected primary fetal rat costal chondrocytes with a recombinant retrovirus (NIH/J-2) carrying the myc and raf oncogenes, which have been shown to have an "immortalizing" function. Following infection, a rapidly proliferating clonal line was isolated that maintained a stable phenotype through 45 passages (11/2 year in culture). This line, termed IRC, grows in suspension culture as multicellular aggregates and in monolayer culture as polygonal cells which accumulate an alcian blue-stainable matrix. IRC cells synthesize high levels of cartilage proteoglycan core protein, and link protein, but show reduced collagen II expression. In addition, the cells express virally derived myc mRNA and protein, but do not express v-raf. Retinoic acid, which is a known modulator of chondrocyte phenotype, down-regulates expression of chondrocyte marker proteins, while stimulating v-myc expression by IRC cells. These data suggest that v-myc expression by chondrocytes results in rapid cell division and maintenance of many aspects of the differentiated phenotype. These "immortalized" cells, however, remain responsive to agents such as retinoic acid which modulate cell phenotype. The potential exists for development of chondrocyte cell lines from diseased cartilage, as well as from human cartilage.