Summary Statement

Extended expansion of cells derived from equine articular cartilage reveal maintenance of chondrogenic potency and no evidence of senescence up to 100 population doublings. The data suggests the reclassification of these cells from progenitor cells to stem cells.

Osteoarthritis (OA) is the most common form of joint disease leading to disability and dependence. In severe cases of knee OA, the joint is deemed irrecoverable and total knee replacements are indicated. Tissue engineering is a possible solution for this pathology and previous work from our laboratory has demonstrated that it is possible to isolate and expand chondroprogenitor cells in vitro from healthy knee-joint articular cartilage. Work presented here describes the detection and isolation of chondroprogenitor cells derived from osteoarthritic cartilage following total knee replacement in patients with severe OA, suggesting a pool of viable cells from this degenerate region which has been previously deemed non-recoverable.

Human articular cartilage was excised from tibial plateaux (TP's) obtained from total knee replacements following the diagnoses of severe OA. Cells were isolated by a sequential pronase and collagenase digestion and subject to a fibronectin adhesion assay. Cells were expanded in monolayer in supplemented growth medium. Clonal 3D pellet cultures were established in chondrogenic and osteogenic differentiation media. Adipogenic cultures were also established in monolayer cultures. Histological procedures, immunohistochemistry and molecular biology were undertaken in order to determine the extent of differentiation. In addition, osteochondral plugs were excised from the TP's and wax embedded for further histological and immunohistochemical analysis.

Clonal cell lines obtained from osteoarthritic knee-joint cartilage using the fibronectin adhesion assay were isolated and successfully cultured to a maximum of 60 population doublings whilst still demonstrating a chondrogenic capacity. Three-D pellet cultures after 21 days of chondrogenic induction produced smooth and iridescent pellets which stained positively for toluidine blue and safranin O. Positive labelling for collagen type II and aggrecan were also observed. Following osteogenic induction; evidence of mineralisation was indicated by the von Kossa stain. Adipogenic induction revealed a positive result. Osteochondral plugs demonstrated sporadic positive labelling in the surface region for putative stem cell marker Stro-1.

Chondroprogenitor cells isolated from osteoarthritic display a strong chondrogenic phenotype, and have the ability to be induced into different lineages. These findings suggest the presence of a pool of viable chondroprogenitors from osteoarthritic tissue which was otherwise deemed irrecoverable.

Unique progenitor cells have been identified recently and successfully cultured in vitro from human articular cartilage. These cells are able to maintain chondrogenic potential upon extensive expansion. In this study, we have developed a sheep, ex-vivo model of cartilage damage and repair, using these progenitor cells. This study addresses the question can such a model be used to determine factors required for progenitor cell proliferation, differentiation and integration of matrix onto bone. The hypothesis was that sheep allogenic cartilage derived progenitor cells could regenerate artificially damaged sheep articular cartilage in an osteochondral culture model. Progenitor cells were derived from ovine articular cartilage using a differential adhesion assay to fibronectin and expanded clonally. These clonal cells were marked with lentiviral vectors derived from the Human Immunodeficiency Virus-1. When a self-inactivating lentiviral vector encoding a ubiquitous phosphoglycerate kinase promoter, driving a Green Fluorescent Protein (GFP) reporter gene, was used to transduce these cells, up to 80% of these progenitor cells expressed GFP. Normal sheep medial femoral condyles containing about 2mm thick sub-condral bone were obtained and 4mm circular defects created on the cartilage surface using a biopsy punch. Condyles were cultured for two weeks in vitro with GFP labelled progenitor cells within a fibrin glue scaffold (Tisseel Lyo) and matrix production (collagen) as determined by spatially offset Raman spectroscopy and immunohistochemistry was demonstrated. Progenitor cells were able to proliferate and differentiate into collagen producing cells. Such an ex-vivo model system is an effective tool for the analysis of cartilage repair from various sources of stem cells. These ex-vivo experiments and variations on defect type, size, titration of scaffold and progenitor cell numbers requirements can further be used as a basis for screening prior to in vivo experiments.

Hyaline cartilage defects are a significant clinical problem for which a plethora of cartilage repair techniques are used. One such technique is cartilage replacement therapy using autologous chondrocyte or mesenchymal stem cell (MSC) implantation (ACI). Mesenchymal stem cells are increasingly being used for these types of repair technique because they are relatively easy to obtain and can be expanded to generate millions of cells. However, implanted MSCs can terminally differentiate and produce osteogenic tissue which is highly undesirable, also, MSCs generally only produce fibrocartilage which does not make biomechanically resilient repair tissue, an attribute that is crucial in high weight-bearing areas. Tissue-specific adult stem cells would be ideal candidates to fill the void, and as we have shown previously in animal model systems [Dowthwaite et al, 2004, J Cell Sci 117;889], they can be expanded to generate hundreds of millions of cells, produce hyaline cartilage and they have a restricted differential potential. Articular chondroprogenitors do not readily terminally differentiate down the osteogenic lineage.

At present, research focused on isolating tissue-specific stem cells from articular cartilage has met with modest success. Our results demonstrate that using differential adhesion it is possible to easily isolate articular cartilage progenitor populations from human hyaline cartilage and that these cells can be subsequently expanded in vitro to a high population doubling whilst maintaining a normal karyotype. Articular cartilage progenitors maintain telomerase activity and telomere length that are a characteristic of progenitor/stem cells and differentiate to produce hyaline cartilage.

In conclusion, we propose the identification and characterisation of a novel articular cartilage progenitor population, resident in human cartilage, which will greatly benefit future cell-based cartilage repair therapies.

One reason why NICE (National Institute for Clinical Excellence) does not support operations by the NHS to heal hyaline cartilage lesions using a patients own cells is because there is no clear evidence to show that these operations are beneficial and cost-effective in the long term. Specifically, NICE identified a deficiency of high quality cartilage being produced in repaired joints. The presence of high quality cartilage is linked to long-lasting and functional repair of cartilage. The benchmark for quality, NICE stipulate, is repair cartilage that is stiff and strong and looks similar to the normal tissue surrounding it, i.e. mature hyaline articular cartilage.

Biopsy material from autologous cartilage implantation surgical procedures has the appearance of immature articular cartilage and is frequently a mixture of hyaline and fibrocartilage. Osteoarthritic cartilage, in its early stages, also exhibits characteristics of immature articular cartilage in that it expresses proteins found in embryonic and foetal developmental stages, and is highly cellular as evidenced through the presence of chondrocyte clusters. Therefore, an ability to modulate the phenotype and the structure of the extracellular matrix of articular cartilage could positively affect the course of repair and regeneration of articular cartilage lesions. In order to do this, the biochemical stimuli that induce the transition of an essentially unstructured amorphous cartilage mass (immature articular cartilage) to one that is highly structured and ordered, and biomechanically adapted to its particular function (mature articular cartilage) has to be identified.

We show for the first time, that fibroblast growth factor-2 and transforming growth factor beta-1 induce precocious maturation of immature articular cartilage. Our data demonstrates that it is possible to significantly enhance maturation of cartilage tissue using growth factor stimulation; consequently this may have applications in transplantation therapy, or through phenotypic modulation of osteoarthritic chondrocytes in diseased cartilage in order to stimulate growth and maturation of repair tissue.

A novel scoring system for the grading of osteoarthritis has been developed.

Scoring systems for the measurement of Osteoarthritis (OA) are essential for the understanding of the osteoarthritic process. OA is a mutifactorial degenerative joint disease affecting not only hyaline cartilage but also the surrounding tissues and particularly the subchondral bone. It as questionable as to why the articular cartilage remains the sole component used for histopathological assessment. The intimate relationship between the subchondral bone and overlying cartilage provide major difficulty in their independent measurement.

A new scoring system has been developed to incorporate the subchondral bone into the assessment process and relating it to the structure of the overlying hyaline cartilage, which together permit a more accurate description of the degree of degenerate change.

The new scoring system was developed from the analysis of 26 operative specimens from tibial plateau (TP) from patients who underwent total knee replacement (TKR). Multiple osteochondral plugs were taken from weight-bearing regions of the whole TP. The specimens were fixed and decalcified before being sectioned and stained with Masson's trichrome.

Using a standard imaging system (Photoshop) the areas of bone and hyaline cartilage were identified and measured. Further parameters 1) cartilage thickness 2) tidemark integrity, 3) surface integrity 4) cartilage morphology were measured using a numeric measurement scale.

The scoring system indicated a relationship between the area of subchondral bone and the hyaline cartilage degeneration. The overall sum of scores was also successful in distinguishing between the milder and more severe samples of OA. More comprehensive inter and intra observer variability needs to be tested in order validate the system. Quantifying changes to the subchondral bone may also serve beneficial to clinicians, as it is possible that monitoring these changes clinically could lead to early identification of OA.

Hyaline cartilage defects are a significant clinical problem for which a plethora of cartilage repair techniques are used. One such technique is cartilage replacement therapy using autologous chondrocyte or mesenchymal stem cell (MSC) implantation (ACI). Mesenchymal stem cells are increasingly being used for these types of repair technique because they are relatively easy to obtain and can be expanded to generate millions of cells. However, implanted MSCs can terminally differentiate and produce osteogenic tissue which is highly undesirable, also, MSCs generally only produce fibrocartilage which does not make biomechanically resilient repair tissue, an attribute that is crucial in high weight-bearing areas. Tissue-specific adult stem cells would be ideal candidates to fill the void, and as we have shown previously in animal model systems [Dowthwaite et al, 2004, J Cell Sci 117;889], they can be expanded to generate hundreds of millions of cells, produce hyaline cartilage and they have a restricted differential potential. Articular chondroprogenitors do not readily terminally differentiate down the osteogenic lineage.

At present, research focused on isolating tissue-specific stem cells from articular cartilage has met with modest success. Our results demonstrate that using differential adhesion it is possible to easily isolate articular cartilage progenitor populations from human hyaline cartilage and that these cells can be subsequently expanded in vitro to a high population doubling whilst maintaining a normal karyotype. Articular cartilage progenitors maintain telomerase activity and telomere length that are a characteristic of progenitor/stem cells and differentiate to produce hyaline cartilage.

In conclusion, we propose the identification and characterisation of a novel articular cartilage progenitor population, resident in human cartilage, which will greatly benefit future cell-based cartilage repair therapies.

The three-dimensional architecture of bovine articular cartilage collagen and its relationship to split lines has been studied with scanning electron microscopy. In the middle and superficial zones, collagen was organised in a layered or leaf-like manner. The orientation was vertical in the intermediate zone, curving to become horizontal and parallel to the articular surface in the superficial zone. Each leaf consisted of a fine network of collagen fibrils. Adjacent leaves merged or were closely linked by bridging fibrils and were arranged according to the split-line pattern. The surface layer (lamina splendens) was morphologically distinct. Although ordered, the overall collagen structure was different in each plane (anisotropic) a property described in previous morphological and biophysical studies. As all components of the articular cartilage matrix interact closely, the three-dimensional organisation of collagen is important when considering cartilage function and the processes of cartilage growth, injury and repair.