Identification of long non-coding RNAs expressed in knee and hip osteoarthritic cartilage.

OBJECTIVE: Long intergenic non-coding RNAs (lincRNAs) are emerging as key regulators in gene expression; however, little is known about the lincRNA expression changes that occur in osteoarthritis (OA). Here we aimed to define a transcriptome of lncRNAs in OA cartilage, specifically comparing the lincRNA transcriptome of knee and hip cartilage. METHOD: RNA-seq was performed on nucleic acid extracted from hip cartilage from patients undergoing joint replacement surgery because of either OA (n = 10) or because of a neck of femur fracture (NOF; n = 6). After transcript alignment, counts were performed using Salmon and differential expression for ENSEMBL lincRNAs determined using DESeq2. Hip RNA-seq lincRNA expression was compared to a knee dataset (ArrayExpress; E-MTAB-4304). ChIP-seq data from ENCODE was used to determine whether lincRNAs were associated with promoters (plncRNA) or unidirectional enhancer-like regulatory elements (elncRNAs). RESULTS: Our analysis of the hip transcriptome identified 1692 expressed Transcripts Per Million (TPM ≥1) Ensembl lincRNAs, of which 198 were significantly (FDR ≤0.05) differentially expressed in OA vs normal (NOF) cartilage. Similar analysis of knee cartilage transcriptome identified 648 Emsembl lincRNAs with 93 significantly (FDR ≤0.05) differentially expressed in intact vs damaged cartilage. In total, 1834 lincRNAs were expressed in both hip and knee cartilage, with a highly significant correlation in expression between the two cartilages. CONCLUSION: This is the first study to use RNA-seq to map and compare the lincRNA transcriptomes of hip and knee cartilage. We propose that lincRNAs expressed selectively in cartilage, or showing differential expression in OA, will play a role in cartilage homoeostasis.

Gene expression changes in damaged osteoarthritic cartilage identify a signature of non-chondrogenic and mechanical responses.

OBJECTIVES: Joint degeneration in osteoarthritis (OA) is characterised by damage and loss of articular cartilage. The pattern of loss is consistent with damage occurring only where the mechanical loading is high. We have investigated using RNA-sequencing (RNA-seq) and systems analyses the changes that occur in damaged OA cartilage by comparing it with intact cartilage from the same joint. METHODS: Cartilage was obtained from eight OA patients undergoing total knee replacement. RNA was extracted from cartilage on the damaged distal medial condyle (DMC) and the intact posterior lateral condyle (PLC). RNA-seq was performed to identify differentially expressed genes (DEGs) and systems analyses applied to identify dysregulated pathways. RESULTS: In the damaged OA cartilage, there was decreased expression of chondrogenic genes SOX9, SOX6, COL11A2, COL9A1/2/3, ACAN and HAPLN1; increases in non-chondrogenic genes COL1A1, COMP and FN1; an altered pattern of secreted proteinase expression; but no expression of major inflammatory cytokines. Systems analyses by PhenomeExpress revealed significant sub-networks of DEGs including mitotic cell cycle, Wnt signalling, apoptosis and matrix organisation that were influenced by a core of altered transcription factors (TFs), FOSL1, AHR, E2F1 and FOXM1. CONCLUSIONS: Gene expression changes in damaged cartilage suggested a signature non-chondrogenic response of altered matrix protein and secreted proteinase expression. There was evidence of a damage response in this late OA cartilage, which surprisingly showed features detected experimentally in the early response of cartilage to mechanical overload. PhenomeExpress analysis identified a hub of DEGs linked by a core of four differentially regulated TFs.

Proteoglycan biosynthesis in chondrocytes: protein A-gold localization of proteoglycan protein core and chondroitin sulfate within Golgi subcompartments.

The intracellular pathway of cartilage proteoglycan biosynthesis was investigated in isolated chondrocytes using a protein A-gold electron microscopy immunolocalization procedure. Proteoglycans contain a protein core to which chondroitin sulfate and keratan sulfate chains and oligosaccharides are added in posttranslational processing. Specific antibodies have been used in this study to determine separately the distribution of the protein core and chondroitin sulfate components. In normal chondrocytes, proteoglycan protein core was readily localized only in smooth-membraned vesicles which co-labeled with ricin, indicating them to be galactose-rich medial/trans-Golgi cisternae, whereas there was only a low level of labeling in the rough endoplasmic reticulum. Chondroitin sulfate was also localized in medial/trans-Golgi cisternae of control chondrocytes but was not detected in other cellular compartments. In cells treated with monensin (up to 1.0 microM), which strongly inhibits proteoglycan secretion (Burditt, L.J., A. Ratcliffe, P. R. Fryer, and T. Hardingham, 1985, Biochim. Biophys. Acta., 844:247-255), there was greatly increased intracellular localization of proteoglycan protein core in both ricin-positive vesicles, and in ricin-negative vesicles (derived from cis-Golgi stacks) and in the distended rough endoplasmic reticulum. Chondroitin sulfate also increased in abundance after monensin treatment, but continued to be localized only in ricin-positive vesicles. The results suggested that the synthesis of chondroitin sulfate on proteoglycan only occurs in medial/trans-Golgi cisternae as a late event in proteoglycan biosynthesis. This also suggests that glycosaminoglycan synthesis on proteoglycans takes place in a compartment in common with events in the biosynthesis of both O-linked and N-linked oligosaccharides on other secretory glycoproteins.

The intracellular localisation of proteoglycans and their accumulation in chondrocytes treated with monensin.

Pig laryngeal chondrocytes incubated in the presence of monensin showed inhibition of [35S]sulphate incorporation and decreased secretion of proteoglycan into the culture medium, but no large decrease in protein synthesis. This lead to the intracellular accumulation of proteoglycan protein core, which was detected in immunoprecipitates of cell extracts. Using the same antiserum protein core was localised by electron microscopy with protein A-coated gold. In control chondrocytes, it was detected only in elements of the Golgi and in secretory vesicles, but following monensin treatment labelling was more intense in the Golgi and extended into the distended cisternae of the rough endoplasmic reticulum. The results suggest that monensin blocks proteoglycan protein core translocation between different elements of the Golgi and that this occurs prior to the major site of chondroitin sulphate synthesis on proteoglycan.

The influence of link protein stabilization on the viscometric properties of proteoglycan aggregate solutions.

The dynamic, steady-shear and transient shear flow properties of precisely prepared link-stable (s0 136, 66% aggregate) and link-free (s0 93, 59% aggregate) proteoglycan aggregate solutions at concentrations ranging from 10 to 50 mg/ml were determined using a cone-on-plate viscometer in a mechanical spectrometer. All proteoglycan solutions tested possessed: (1) linear viscoelastic properties - as measured by the dynamic complex modulus under small amplitude steady oscillatory conditions (1 less than or equal to omega less than or equal to 100 rad/s) - and (2) nonlinear shear-rate dependent apparent viscosities and primary normal stress difference under steady shearing conditions (0.25 less than or equal to gamma less than or equal to 250 s-1). Our transient flow data show that all proteoglycan aggregate solutions exhibited transient stress overshoot effects in shear stress and normal stress. From these steady and transient flow data, we conclude that link protein stabilized aggregates have significant effects on their dynamic and steady-shear properties as well as transient flow properties. The transient stress overshoot data provide a measure of the energy per unit volume of fluid required to overcome the proteoglycan networks in solution from a resting state. Thus we found that link-stable aggregates form much stronger networks than link-free aggregates. This is corroborated by the fact that link-stable aggregates form more elastic (lower than delta) and stiffer (higher [G*]) networks than link-free aggregates. The complete spectrum of viscometric flow data is entirely compatible with the proposed role of link protein in adding structural stability to the proteoglycan-hyaluronate bond. In cartilage, the enhanced strength of the networks formed by link-stable aggregates may play an important role in determining the material properties of the tissue and thereby contribute to the functional capacity of cartilage in diarthrodial joints.

Immunoglobulin fold and tandem repeat structures in proteoglycan N-terminal domains and link protein.

Detailed primary sequence and secondary structure analyses are reported for the hyaluronate binding region (G1 domain) and link protein of proteoglycan aggregates. These are based on six full or partial sequences from the chicken, pig, human, rat and bovine proteins. Determinations of a full pig and a partial human link protein sequence are reported in the Appendix. Five sequences at the N terminus in both proteins were compared with the structures of 11 variable immunoglobulin (Ig) fold domains for which crystal structures are available. Despite only modest sequence homology, a clear alignment could be proposed. Analysis of this shows that the equivalents of the first and second hypervariable segments are now significantly longer, and both proteins have N-terminal extensions that are up to 23 residues in length. Secondary structure predictions showed that these sequences could be identified with available crystal structures for the variable Ig fold. However the hydrophobic residues involved in interactions between the light and heavy chains in Igs are replaced by hydrophilic charged groups in both proteins. These results imply that both proteins are members of the Ig superfamily, but exhibit structural differences distinct from other members of this superfamily for which crystal structures are known. The proteoglycan tandem repeat (PTR) is a repeat of 99 residues that is found twice in the amino acid sequence of link protein and the proteoglycan G1 domain adjacent to the Ig fold, and also twice in the proteoglycan G2 domain. A total of 16 PTRs was available for analysis. Compositional analyses show that these are positively charged if these originate from link protein, and negatively charged if from the G1 or G2 domains. The 16 Robson secondary structure predictions for the PTRs were averaged to improve the statistics of the prediction, and checked by comparison with Chou-Fasman calculations. A strong alpha-helix prediction was found at residues 13 to 25, and several beta-strands were predicted. The overall content is 18% alpha-helix and 28% beta-sheet, with 44% of the remaining sequence being predicted as turns. These analyses show that both the proteoglycan G1 domain and link protein are constructed from two distinct globular components, which may provide the two functional roles of these proteins in proteoglycan aggregation.

The distribution of aggregating proteoglycans in articular cartilage: comparison of quantitative immunoelectron microscopy with radioimmunoassay and biochemical analysis.

Electron microscopic immunolocalization and radioimmunoassay have been used to determine the variation with depth of the hyaluronate-binding region of proteoglycan in articular cartilage. The cartilage was cut into serial sections from the articular surface to the bony margin, the proteoglycans were extracted from each section and determined by radioimmunoassay using antibodies raised against proteoglycan binding region. Proteoglycans were found to be most abundant in the middle zone and least abundant near the articular surface. Biochemical analysis for hexuronate in the same extracts showed a distribution of proteoglycan in agreement with these and other published results. The binding region antiserum was used for electron microscopic immunolocalization of proteoglycan with ultrathin sections of cartilage embedded in Lowicryl K4M resin. After digestion of the sections with chondroitinase ABC, the proteoglycans were localized using the antiserum and protein A-coated gold particles as immunolabel. The density of labeling was quantified using a Magiscan image analysis system. Throughout the depth of the cartilage matrix labeling was higher in the pericellular regions compared to the intercellular regions, and variation of the amount of immunolabel with depth was found to show a good correlation with the results from radioimmunoassay. Intracellular labeling of proteoglycans was mainly found over the Golgi region and in membrane-bound (secretory) vesicles.

Viscoelastic properties of proteoglycan solutions with varying proportions present as aggregates.

Monomer and aggregated proteoglycans were prepared from pig laryngeal cartilage. Vascoelastic flow properties, comprising linear complex dynamic shear modulus, nonlinear steady-state shear-rate dependent viscosity, and primary normal stress difference, were measured in proteoglycan solutions containing varying proportions of aggregate (0-80%) and at different concentrations (10-50 mg/ml). Results were analyzed using the simple Oldroyd four-parameter nonlinear rate-type rheological equation. All solution properties were strongly dependent on proteoglycan concentration and on the proportion of aggregates present. Aggregation was found to have a great effect on the zero shear-rate viscosity at 50 mg/ml, which increased fivefold from 0-100% aggregate. The results showed that network formation in proteoglycan solutions increased with concentration from 10-50 mg/ml and also increased with aggregation. All proteoglycan solutions showed shear thinning, which was most marked with aggregated proteoglycan at high concentration (50 mg/ml), where the viscosity decreased tenfold from the zero shear-rate limit to the infinite shear-rate limit. The intermolecular interactions in the network were therefore increasingly disrupted by increasing shear rate, but repeated measurements showed that these were reversible changes and that testing did not induce disaggregation or degradation of proteoglycan. These rheological properties show that aggregation is likely to immobilize proteoglycan at high concentration within cartilage and to contribute to the material properties of the porous solid matrix of articular cartilage that are important for its load-bearing function.

Increased release of matrix components from articular cartilage in experimental canine osteoarthritis.

The release rates of specific components of the proteoglycan aggregates (G1 domain, the chondroitin sulfate and keratan sulfate containing portion of the protein core, and link protein) of the articular cartilage of mature beagles were studied at early stages of canine experimental osteoarthritis (OA), generated by transection of the anterior cruciate ligament. Analysis of cartilage explants and synovial fluids indicates that at early stages of experimental OA, there is increased release of the proteoglycan aggregates of the articular cartilage. This involves a release from the tissue of the components of the proteoglycan that are specifically involved with aggregation together with the glycosaminoglycans of the proteoglycan. These components were detected at elevated levels in the media of explants of cartilage from the operated joint, and in the synovial fluids of the operated joints.

The sulphation pattern in chondroitin sulphate chains investigated by chondroitinase ABC and ACII digestion and reactivity with monoclonal antibodies.

We have used progressive chondroitinase digestion of pig aggrecan in conjunction with ELISA assays and disaccharide analysis to derive information about the pattern of 4- and 6-sulphation in chondroitin sulphate chains. Digestion with chondroitinase ABC resulted in the release of mainly disaccharides from the nonreducing terminal of chondroitin sulphate chains but there was also the release of some tetra- and hexa-saccharides which were degraded to disaccharides with more extensive digestion. Chondroitinase ACII, in contrast, released only disaccharides. Analysis of the disaccharide composition of the intact and digested products at different stages of digestion showed that there was a slight increase in 6-sulphate content of the chains as they were shortened. Reaction of the partially digested proteoglycans with monoclonal antibodies 3-B-3 and 3-D-5 which recognise chains terminating in 6- or 4-sulphated disaccharides, respectively, showed major differences between chondroitinase ABC and ACII products. The results suggested that chondroitinase ABC preferentially cleaved next to 4-sulphated, rather than 6-sulphated disaccharides and this resulted in some oligosaccharides as well as disaccharide being released. Chondroitinase ACII also cleaved an additional disaccharide next to the linkage to protein of chondroitin sulphate, which was not removed by chondroitinase ABC and this disaccharide was mainly nonsulphated.

The structure of aggrecan and its turnover in cartilage.

Aggrecan is the major proteoglycan in cartilage. It has a multidomain structure with 3 globular and 2 extended segments. It forms large aggregates by binding to hyaluronan via the G1 domain, and link protein stabilizes the aggrecan-hyaluronan bond. The extended interglobular domain joining G1 and G2 domains is the main site of proteolytic attack in aggrecan turnover. One site of cleavage is reported to predominate, but the enzyme responsible for this cleavage has not been identified. A metalloproteinase, neutrophil collagenase, has been shown to cleave at this "aggrecanase" site in vitro; however, it has yet to be shown if metalloproteinases are responsible for this activity in cartilage.

[Regulation of cartilage matrix synthesis by chondrocytes]. / Régulation de la synthèse de la matrice cartilagineuse par les chondrocytes.

Articular cartilage has important load bearing properties. These depend on the integrity of its matrix which is formed by a dense network of collagen fibres (mainly type II) and a high concentration of proteglycan (mainly aggrecan). The matrix is maintained by the chondrocytes, which control the production and turnover of matrix components and are greatly affected by cytokines (such as IL-1 alpha,beta and TNF alpha) and growth factors (such as IGF-1 alpha, beta TGF beta). They have strongly antagonistic effects. IL-1 alpha, beta and TNF alpha inhibit proteoglycan synthesis and at slightly higher concentration they enhance the rates of matrix degradation. In contrast the growth factor IGF-1 stimulates proteoglycan biosynthesis and reduces matrix proteinase action. TGB beta has less direct anabolic effect on matrix biosynthesis in normal cartilage, but does not induce an anabolic response in isolated chondrocytes or in explants following IL-1 treatment. We have also investigated the action of IL-1 in inflammatory arthritis in vivo by treatment with IL-1 receptor antagonist, the natural IL-1 inhibitor. In antigen-induced arthritis the effects of the injection of rh IL-1 ra was measured over 3 days. There was little effect on the induction of joint swelling, cellular infiltration into synovium or cartilage proteoglycan depletion, but when given over a similar time in more chronic arthritis, 14 days after induction it had a major effect on suppressing synovial fibrosis although it still did not affect parameters of joint inflammation. The results suggested that IL-1 was not the major factor inducing inflammation in the joint, but was responsible for the excessive collagen deposition in the synovium in this experimental model of arthritis.

Fat pad-derived mesenchymal stem cells as a potential source for cell-based adipose tissue repair strategies.

BACKGROUND: Mesenchymal stem cells are able to undergo adipogenic differentiation and present a possible alternative cell source for regeneration and replacement of adipose tissue. The human infrapatellar fat pad is a promising source of mesenchymal stem cells with many source advantages over from bone marrow. It is important to determine whether a potential mesenchymal stem-cell exhibits tri-lineage differentiation potential and is able to maintain its proliferation potential and cell-surface characterization on expansion in tissue culture. We have previously shown that mesenchymal stem cells derived from the fat pad can undergo chondrogenic and osteogenic differentiation, and we characterized these cells at early passage. In the study described here, proliferation potential and characterization of fat pad-derived mesenchymal stem cells were assessed at higher passages, and cells were allowed to undergo adipogenic differentiation. MATERIALS AND METHODS: Infrapatellar fat pad tissue was obtained from six patients undergoing total knee replacement. Cells isolated were expanded to passage 18 and proliferation rates were measured. Passage 10 and 18 cells were characterized for cell-surface epitopes using a range of markers. Passage 2 cells were allowed to undergo differentiation in adipogenic medium. RESULTS: The cells maintained their population doubling rates up to passage 18. Cells at passage 10 and passage 18 had cell-surface epitope expression similar to other mesenchymal stem cells previously described. By staining it was revealed that they highly expressed CD13, CD29, CD44, CD90 and CD105, and did not express CD34 or CD56, they were also negative for LNGFR and STRO1. 3G5 positive cells were noted in cells from both passages. These fat pad-derived cells had adipogenic differentiation when assessed using gene expression for peroxisome proliferator-activated receptor γ2 and lipoprotein lipase, and oil red O staining. DISCUSSION: These results indicate that the cells maintained their proliferation rate, and continued expressing mesenchymal stem-cell markers and pericyte marker 3G5 at late passages. These results also show that the cells were capable of adipogenic differentiation and thus could be a promising source for regeneration and replacement of adipose tissue in reconstructive surgery.

The epitope characterisation and the osteogenic differentiation potential of human fat pad-derived stem cells is maintained with ageing in later life.

Some clinical settings are deficient in osteogenic progenitors, e.g. atrophic nonunited fractures, large bone defects, and regions of scarring and osteonecrosis. These benefit from the additional use of bone marrow-derived mesenchymal stem cells, but these cells exhibit an age-related decline in lifespan, proliferation and osteogenic potential. Therapeutic approaches for the repair of bone could be optimised by the identification of a stem cell source that does not show age-related changes. Fat pad-derived stem cells are capable of osteogenesis, but a detailed study of the effect of ageing on their epitope profile and osteogenic potential has so far not been performed. Fat pad-derived cells were isolated from 2 groups of 5 patients with a mean age of 57 years (S.D. 3 years) and 86 years (S.D. 3 years). The proliferation, epitope profile and osteogenic differentiation potential of cells from the 2 groups were compared. Cells isolated from the fat pad of both groups showed similar proliferation rates and exhibited a cell surface epitope profile similar but not identical to that of bone marrow-derived stem cells. The cells from both groups cultured in osteogenic medium exhibited osteogenesis as shown by a significant upregulation of alkaline phosphatase and osteocalcin genes, and significantly greater alkaline phosphatase enzyme activity compared to cells cultured in the control medium. The cells cultured in the osteogenic medium also showed greater calcium phosphate deposition on alizarin red staining. There was no significant difference between the osteogenic potential of the two age groups for any of the parameters studied. The fat pad is a consistent and homogenous source of stem cells that exhibits osteogenic differentiation potential with no evidence of any decline with ageing in later life. This has many potential therapeutic tissue engineering applications for the repair of bone defects in an increasingly ageing population.