• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Dec 28, 2004; 101(52): 18087–18092.
Published online Dec 15, 2004. doi:  10.1073/pnas.0404504102
PMCID: PMC535800
Medical Sciences

Stimulation of proteoglycan synthesis by glucuronosyltransferase-I gene delivery: A strategy to promote cartilage repair


Osteoarthritis is a degenerative joint disease characterized by a progressive loss of articular cartilage components, mainly proteoglycans (PGs), leading to destruction of the tissue. We investigate a therapeutic strategy based on stimulation of PG synthesis by gene transfer of the glycosaminoglycan (GAG)-synthesizing enzyme, β1,3-glucuronosyltransferase-I (GlcAT-I) to promote cartilage repair. We previously reported that IL-1β down-regulated the expression and activity of GlcAT-I in primary rat chondrocytes. Here, by using antisense oligonucleotides, we demonstrate that GlcAT-I inhibition impaired PG synthesis and deposition in articular cartilage explants, emphasizing the crucial role of this enzyme in PG anabolism. Thus, primary chondrocytes and cartilage explants were engineered by lipid-mediated gene delivery to efficiently overexpress a human GlcAT-I cDNA. Interestingly, GlcAT-I overexpression significantly enhanced GAG synthesis and deposition as evidenced by 35S-sulfate incorporation, histology, estimation of GAG content, and fluorophore-assisted carbohydrate electrophoresis analysis. Metabolic labeling and Western blot analyses further suggested that GlcAT-I expression led to an increase in the abundance rather than in the length of GAG chains. Importantly, GlcAT-I delivery was able to overcome IL-1β-induced PG depletion and maintain the anabolic activity of chondrocytes. Moreover, GlcAT-I also restored PG synthesis to a normal level in cartilage explants previously depleted from endogenous PGs by IL-1β-treatment. In concert, our investigations strongly indicated that GlcAT-I was able to control and reverse articular cartilage defects in terms of PG anabolism and GAG content associated with IL-1β. This study provides a basis for a gene therapy approach to promote cartilage repair in degenerative joint diseases.

Keywords: osteoarthritis, gene transfer, chondrocyte, glycosyltransferase, glycosaminoglycan

Osteoarthritis (OA) is a progressive degenerative joint disease characterized by loss of articular cartilage components, mainly proteoglycans (PGs), leading to tissue destruction and hypocellularity, eventually resulting in loss of joint function (1, 2). A hallmark of OA is the marked increase in proinflammatory cytokines, such as IL-1β, that inhibit synthesis of PGs and collagen and enhance their degradation, perturbing the normal homeostasis of cartilage extracellular matrix (3). Stimulation of PG synthesis and/or inhibition of PG degradation, therefore, are of central importance for OA treatment (4). Various therapeutic strategies have been developed to antagonize the activity of proinflammatory cytokines (5). These strategies include administration of anticytokine antibodies (6), soluble cytokine receptor proteins (7, 8), ex vivo gene transfer, retrovirus expression of IL-1 receptor antagonists (9, 10), and adenovirus-mediated overexpression of IL-4 and IL-10 (11). Although several investigators have reported the benefits of agents that interfere with catabolic processes in OA treatment, relatively few studies have been conducted to stimulate anabolic activity in an attempt to enhance cartilage repair (12). In this regard, local administration of transforming growth factor (TGF)-β1 stimulated PG synthesis (13). However, adenovirus-mediated overexpression of TGF-β1 in rabbit synovium failed to stimulate cartilage matrix synthesis and even produced cartilage degradation, suggesting that TGF-β1 gene transfer may not be suitable for cartilage repair (14).

PGs consist of core proteins with covalently attached glycosaminoglycan (GAG) side chains. Because of their highly polyanionic nature, PGs interact with different proteins, cytokines, growth factors, and extracellular matrix proteins (15). Aggrecan, the major PG in articular cartilage contains chondroitin sulfate (CS) chains as the predominant GAGs. GAG biosynthesis is initiated by the formation of a GlcA-β1,3-Gal-β1,3-Gal-β1,4-Xyl tetrasaccharide primer, which is linked to serine residues of a core protein. The GAGs are built up on this linkage tetrasaccharide region by the alternate addition of N-acetylhexosamine and glucuronic acid residues (16). Recent identification of glycosyltransferases involved in GAG chain synthesis has highlighted the key role of galactose-β1,3-glucuronosyltransferase-I (GlcAT-I), which catalyses the transfer of a glucuronyl moiety from UDP glucuronic acid, onto the nonreducing end of the second galactose of the PG trisaccharide primer Gal-β1,3-Gal-β1,4-Xyl-β1-O-Ser (17). GlcAT-I has received much attention because it plays a central role at a branching point common to various GAG chains, and it has been suggested to be rate-limiting in GAG synthesis in Chinese hamster ovary cells (18). We have been deeply involved in structural and functional studies of human GlcAT-I (1921). We recently reported that IL-1β down-regulates GlcAT-I expression and activity and suggested that this may contribute to the reduced biosynthesis of GAGs observed in chondrocytes after IL-1β treatment (22).

Here, we show evidence that antisense inhibition of GlcAT-I expression in cartilage explants produced a dramatic decrease in matrix PG synthesis, emphasizing the essential role of this enzyme in GAG assembly. We found that GlcAT-I delivered by a nonviral gene transfer was able to enhance PG synthesis and overcome cartilage PG depletion produced by IL-1β. Furthermore, we brought evidence that GlcAT-I promoted the synthesis of CS GAG chains of the cartilage matrix. Taken together, these data indicate that therapy mediated through GlcAT-I gene delivery holds promise as a treatment for OA.

Materials and Methods

Construction of GlcAT-I Expression Vector, Rat Chondrocyte Cultures, GlcAT-I Transfection, IL-1β Treatment, and PG Synthesis. A human GlcAT-I cDNA was cloned (21) and ligated into the unique NheI-KpnI sites of mammalian expression vector pShuttle (Clontech) to generate pShuttle–GlcAT-I. Primary chondrocytes were isolated from rat articular cartilage (femoral head caps) and maintained in DMEM-F12 mix medium in six-well plates at 37°C in a humidified atmosphere supplemented with 5% CO2, as described in ref. 22. Cells were then transfected at 70% confluence by using Exgen 500 (Euromedex, Souffelweyersheim, France) as follows. One group was mock-transfected with an empty plasmid (control), a second group was mock-transfected and treated with 10 ng/ml IL-1β (Sigma) for 48 h, the third group was transfected with GlcAT-I expression vector (5 μg of plasmid DNA), and the last group was simultaneously transfected with GlcAT-I expression vector and treated with IL-1β. Cells were labeled with 5 μCi/ml (1 Ci = 37 GBq) 35S-sulfate for the last 6 h of the experimental period. PGs were extracted from cell layer and culture medium (23) with 4 M guanidine hydrochloride and precipitated by using cetylpyridinium chloride, and the PG synthesis rate was measured by liquid scintillation counting.

Cartilage Explant Culture and Treatments. Articular cartilage was aseptically excised from rat femoral head caps and maintained in DMEM-F12 mix medium at 37°C in a humidified atmosphere supplemented with 5% CO2. After 72 h, explants were serum-starved for 24 h before treatments. One group was mock-transfected with an empty plasmid and exposed to 10 ng/ml IL-1β for 48, 72, or 96 h. The second group was transfected with GlcAT-I expression vector, and the last group was simultaneously transfected with GlcAT-I expression vector and exposed to IL-1β for 48, 72, or 96 h. The control group was transfected with an empty vector. All explants were pulsed with 5 μCi/ml 35S-sulfate for the last 6 h before PG analysis.

A second set of experiments was designed to test whether GlcAT-I was able to promote recovery of PG synthesis after IL-1β-induced PG depletion. Explants were first exposed to IL-1β for 48, 72, or 96 h and then transfected with GlcAT-I-expression vector or empty plasmid and maintained for a further period of 6 days in culture medium before being pulsed with 5 μCi/ml 35S-sulfate for 6 h before PG analysis as described above.

Antisense Treatment of Cartilage Explants. Phosphorothioate oligonucleotides (20-mer) in antisense and sense orientation to rat GlcAT-I mRNA were synthesized (MWG-Biotech, Ebersberg, Germany): antisense, 5′-TCATGGCCGCGCCGCCGCCC-3′, complementary to nucleotides –16 to +4 (ending 1 bp down-stream to the initiator methionine codon); sense, 5′-GGGCGGCGGCGCGGCCATGA-3′, complementary to antisense served as a control oligonucleotide. Transfection was performed with Exgen 500 reagent by using 50 μM antisense or sense oligonucleotide in serum-free medium. Twenty-four hours after transfection, the medium was replaced by a serum-containing medium (i.e., oligonucleotide-free medium), and incubation continued for various periods before 35S-sulfate labeling and immunohistochemistry.

Histology and Immunohistochemistry. Paraformaldehyde-fixed and paraffin-embedded cartilage sections were either stained with toluidine blue or used for immunohistochemistry. Immunohistochemical detection of GlcAT-I protein and GAGs in sections was performed by using the Novostain Super ABC kit (Novo-Castra, Newcastle, U.K.) according to the manufacturer's protocol. The primary antibodies used were anti-GlcAT-I antibody (21), mouse monoclonal anti-chondroitin-6-sulfate 3B3 (Sigma), or mouse monoclonal anti-keratan sulfate 5D4 (ICN). The antigen–antibody complex was detected with the avidin-biotinylated horseradish peroxidase-conjugated antibody.

Assay of PG Content. Papain-digested cartilage was quantified for sulfated GAG content by 1,9-dimethylmethylene blue assay (24) with purified shark CS C (Sigma) as a standard. Total GAGs were measured before and after chondroitinase ABC treatment, and the CS level was calculated as total GAGs minus chondroitinase ABC digested GAGs.

Western Blot Analysis. The protein content of cell lysates from cultured chondrocytes was determined by using the Bradford method (25). Proteins (30 μg per lane) were separated by 12% SDS/PAGE, transferred onto an Immobilon (Millipore) membrane, and then probed with anti-GlcAT-I (21) antibody. For PG analysis, explants were extracted with 4 M guanidine hydrochloride and separated on 0.6% (wt/vol) composite agarose/1.2% (wt/vol) polyacrylamide gels (26). PGs were transferred to an Immobilon membrane; digested with chondroitinase ABC in 50 mM sodium acetate, pH 7.4, for 3 h at room temperature (27); then probed with a 1:1,000 (vol/vol) dilution of anti-chondroitin-4-sulfate, 2B6 antibodies. The blot was stripped and reprobed with a 1:1,000 (vol/vol) dilution of anti-chondroitin-6-sulfate and a 1:1,500 (vol/vol) dilution of anti-keratan-sulfate (anti-KS) antibodies, respectively. The blot was developed by using streptavidin-biotinylated horseradish peroxidase-conjugated antibody and detected with an enhanced chemiluminescence kit (ECL, Amersham Pharmacia).

CS Chain Analyses. CS chain disaccharide composition was performed on proteinase K-digested explants by using the fluorophore-assisted carbohydrate electrophoresis (FACE) method as described by Calabro et al. (28). Electrophoresis was carried out at 4°C by using miniature vertical slab gels of 29% acrylamide and 1% bisacrylamide at 5 W of constant power per gel. Fluorescent bands were immediately scanned in a UV-light box (Gel Doc System, Bio-Rad) and quantitated with Bio-Rad quantity one software. CS disaccharide standards were from Dextra Laboratories (Reading, U.K.).

CS chain size was determined as described by Funderburgh et al. (29). Briefly, chondroitin/dermatan sulfate PGs were separated from total 35S-labeled PGs by using fractional alcohol precipitation with 50% (vol/vol) ethanol. The precipitated PGs were resuspended in acetate buffer, pH 7.0 (0.1 M ammonium acetate containing 0.0005% phenol red), and hydrolyzed with 20 μg/ml proteinase K twice for 30 min at 45°C. The protein-free 35S-GAG chains were subjected to 15% SDS/PAGE, dried on filter paper, and subjected to autoradiography.

Biosynthetic Radiolabeling and Core Protein and Statistical Analyses. Explants were labeled with 10 μCi/ml [35S]methionine for 6 h before PG extraction by guanidine hydrochloride as above. Equal aliquots of extracted PGs were dialyzed and then digested with chondroitinase ABC (5 units/ml) and keratanase (5 units/ml) for 16 h at 37°C. Rates of protein synthesis were measured as incorporation of [35S]methionine into ethanol-insoluble protein by liquid scintillation counting. Results were expressed as means ± SEM of three to four independent samples in each group. One-way ANOVA with a Newman–Keuls post hoc test was used for statistics and was performed with prism 4.00 (GraphPad, San Diego).


Antisense Inhibition of GlcAT-I Expression Impairs PG Synthesis and Deposition in Cartilage Explants. Immunohistochemical analysis of explant sections showed a dramatic reduction in GlcAT-I protein expression in antisense-transfected cartilage 6 days after transfection compared with sense-transfected or control cartilage (Fig. 1A, compare a with b and c). Similar observations were made after 8 and 10 days after transfection (data not shown). Analysis of cartilage PG synthesis by 35S-sulfate incorporation at 6 and 8 days after transfection indicated a reduction of ≈47% and 58%, respectively, in antisense-treated cartilage (Fig. 1B). Furthermore, PG synthesis remained significantly inhibited (45%) in antisense-transfected explants 10 days after transfection (Fig. 1B). However, the rate of inhibition was less compared with that observed 8 days after transfection, suggesting a recovery from antisense inhibition.

Fig. 1.
Effects of antisense oligonucleotide inhibition of GlcAT-I protein expression in articular cartilage explants. (A) Immunohistochemical analysis of GlcAT-I expression. Antisense-transfected (a), sense-transfected (b), and control (c) sections were analyzed ...

Histological analysis of PG deposition in cartilage sections by using toluidine blue staining confirmed a significant loss of PG content in antisense-transfected explants (Fig. 1C a and b). These results were supported by immunohistochemistry, which showed a significant reduction in C-6-S epitope in cartilage extracellular matrix upon antisense oligonucleotide transfection (Fig. 1C c and d).

Gene Delivery of GlcAT-I in Chondrocytes and Cartilage Explants Stimulates PG Synthesis. The ability of pShuttle–GlcAT-I to drive the expression of the protein in chondrocytes was assessed by immunoblotting. The results showed a high expression 24 h after transfection and a further increase 48 h after transfection (Fig. 2A). Furthermore, analysis of PG synthesis 48 h after transfection showed a 30% increase of 35S-sulfate incorporation in GlcAT-I transfected chondrocytes compared with empty vector-transfected cells (Fig. 2B), indicating that overexpression of GlcAT-I significantly enhanced PG synthesis in chondrocytes.

Fig. 2.
Effects of overexpression of GlcAT-I in rat articular cultured chondrocytes. (A) Immunoblot of GlcAT-I protein expression in pShuttle–GlcAT-I-transfected chondrocytes 24 and 48 h after transfection. (B) PG synthesis in chondrocytes transfected ...

Next, we examined whether pShuttle–GlcAT-I vector was able to drive the expression of GlcAT-I in cartilage explants. Immunohistochemical analysis 6 days after transfection showed a strong intracellular staining of GlcAT-I protein in chondrocytes of transfected cartilage compared with empty vector-transfected explants (Fig. 3A a and b). Furthermore, RT-PCR and immunohistochemical analyses indicated that high level expression of GlcAT-I transcript and protein was maintained for >10 days after transfection (data not shown). Analysis of PG synthesis by 35S-sulfate incorporation 6 days after transfection demonstrated a 2.4-fold increase in synthesis rate in GlcAT-I-transfected cartilage compared with mock-transfected samples (Fig. 3B). These results were confirmed by toluidine blue staining, which revealed an increase in PG deposition in transfected explants (Fig. 3A c and d). Altogether, the data indicated that gene delivery-mediated GlcAT-I expression was effective in stimulating PG matrix synthesis in cultured chondrocytes and cartilage explants.

Fig. 3.
Effects of GlcAT-I gene delivery in rat articular cartilage explants. (A) Representative photomicrographs of GlcAT-I immunoreactivity from mock-transfected (a) or pShuttle–GlcAT-I-transfected (b) cartilage. Representative photomicrographs of toluidine ...

GlcAT-I Is Able to Overcome IL-1β-Induced PG Depletion in Chondrocytes and Cartilage Explants. Here, we examined whether gene delivery-mediated increase in GlcAT-I expression was able to overcome IL-1β-induced PG depletion in chondrocytes. IL-1β treatment reduced (28%) 35S-sulfate incorporation in empty vector-transfected chondrocytes. In contrast, no significant loss of PG synthesis was observed in GlcAT-I-transfected cells after IL-1β treatment (data not shown).

As for chondrocytes, cartilage explants were transfected with pShuttle–GlcAT-I vector and then exposed to IL-1β for different periods of time. PG synthesis was measured 48, 72, and 96 h after IL-1β treatment. A dose-dependent decrease in PG synthesis rate, up to 70% of control samples, was observed in nontransfected (Fig. 4A) or mock-transfected explants (data not shown). By contrast, no decrease in PG synthesis was found in GlcAT-I-transfected explants after IL-1β treatment (Fig. 4A). These findings demonstrate that GlcAT-I was able to overcome the loss of PGs induced by IL-1β in chondrocytes and cartilage explants.

Fig. 4.
Protective effects of GlcAT-I against IL-1β-induced loss of PG synthesis in cartilage. (A) Explants were transfected with pShuttle–GlcAT-I vector and exposed to IL-1β for 48, 72, or 96 h, and PG synthesis was measured. Data are ...

GlcAT-I Enhances Recovery of PG Synthesis After IL-1β Treatment. Explants were treated for different time intervals with IL-1β, then transfected or not with GlcAT-I expression vector and maintained in fresh medium for 6 days before PG analysis. We observed that PG synthesis in cartilage treated by IL-1β for 48, 72, and 96 h remained inhibited by 34%, 39%, and 53%, respectively, 6 days after the end of the treatments (Fig. 4B). Similar results were obtained with mock-transfected explants (data not shown). These findings indicated that normal cartilage recovery in terms of PG synthesis was low. Interestingly, in GlcAT-I-transfected cartilage, PG synthesis rate was restored to normal levels 6 days after each treatment period with IL-1β. These results demonstrated that GlcAT-I rescued cartilage PG synthesis after IL-1β-induced depletion (Fig. 4B).

GlcAT-I Increases the Cartilage GAG Content but Not the Size of the Chains. To determine whether GlcAT-I increases not only GAG synthesis but also GAG accumulation, the amount of sulfated GAGs was estimated. A 1.4-fold increase in GAG content was observed in GlcAT-I-transfected cartilage compared with the mock-transfected group (Fig. 5A). In contrast, IL-1β reduced the GAG content by 37% from the control value. However, no significant reduction of GAG content was observed in IL-1β-treated GlcAT-I-transfected cartilage, indicating that GlcAT-I was able to overcome IL-1β-induced GAG depletion in cartilage (Fig. 5A). Furthermore, we showed that chondroitinase ABC treatment of total GAGs from either the control or IL-1β-treated samples led to almost total digestion of GAGs, indicating that the predominant GAGs in rat cartilage were CS GAGs (Fig. 5A). The same results were obtained for both GlcAT-I-transfected and GlcAT-I-transfected/IL-1β-treated explants, suggesting that overexpression of GlcAT-I in rat cartilage led to stimulation of CS production.

Fig. 5.
Effect of GlcAT-I expression on GAG content and chain size. Explants were transfected with pShuttle–GlcAT-I vector or empty vector and exposed or not to IL-1β for 96 h. (A) Papain-digested cartilage was quantified for sulfated GAG content ...

To study whether GlcAT-I stimulation of GAG synthesis may result in modified chain size, 35S-sulfate-labeled GAGs were subjected to SDS/PAGE after proteolytic digestion of the core proteins. The results shown in Fig. 5B indicate that GAG chains were of similar size in all groups, although it is readily apparent that the amounts of GAG synthesized were higher in GlcAT-I compared with control. Furthermore, analysis of the rate of synthesis of PG core proteins indicated that there was no significant difference between empty vector- and GlcAT-I-transfected cartilage (data not shown).

GlcAT-I Expression Augments the Amount of Initiated CS Chains. Immunohistochemical analyses of sections from GlcAT-I-transfected explants disclosed a remarkable increase in C-6-S epitope compared with empty vector-transfected cartilage (Fig. 6A a and b). In contrast, C-6-S expression was greatly reduced in IL-1β-treated mock-transfected explants (Fig. 6Ac). Interestingly, in GlcAT-I-transfected explants, no significant reduction in C-6-S expression was produced by IL-1β treatment (Fig. 6Ad), indicating that GlcAT-I was able to overcome the loss of CS chains induced by this cytokine. In contrast, no significant differences were found in KS expression, whose synthesis does not require GlcAT-I enzyme, between control and GlcAT-I-transfected cartilage (data not shown).

Fig. 6.
Immunohistochemical and immunoblot analyses of the impact of GlcAT-I overexpression on cartilage GAGs. Explants were transfected and IL-1β-treated as described for Fig. 5. (A) Representative photomicrographs of C-6-S expression in control (a), ...

Immunoblot analysis of large PGs separated on composite gels (Fig. 6B) and quantitative assessment by densitometry (Fig. 6C) indicated a significant increase in the expression of C-4-S (34%) and C-6-S (38%) epitopes in GlcAT-I-transfected cartilage. As expected, IL-1β treatment produced a significant loss in C-4-S and C-6-S epitopes (18% and 15%, respectively). Interestingly, GlcAT-I transfection was able to contain CS decline produced by IL-1β in cartilage. In contrast, no effect of GlcAT-I expression was detected on KS epitope, confirming the specificity of GlcAT-I in biosynthesis of hexuronic acid-containing GAG chains (Fig. 6 B and C).

Analysis of the fine structure and disaccharide composition of hyaluronan (HA) and CS of rat cartilage was carried out by FACE. As shown in Fig. 7, a similar pattern of sulfated and unsulfated disacharides was observed in control and GlcAT-I transfected explants. This pattern indicated that 4-sulfated disaccharides were predominant. In contrast, 6-sulfated disaccharides were much less abundant. Quantitative analysis of the data indicated that GlcAT-I-transfected cartilage showed elevated levels of unsulfated (23%) and 4-sulfated CS disaccharides (29%). IL-1β caused a significant reduction of both unsulfated (25%) and sulfated disaccharides (30%) that was counteracted upon GlcAT-I expression. In contrast to CS, FACE analyses showed no significant difference in HA content between control and GlcAT-I-transfected cartilage, whereas treatment with IL-1β resulted in a reduction in HA in cartilage.

Fig. 7.
FACE analyses of HA and CS disaccharides produced from proteinase K-digested cartilage by hyaluronidase and chondroitinase ABC digestion. Explants were transfected and IL-1β-treated as described for Fig. 5. Disaccharides were identified by coelectrophoresis ...


Repair of damaged articular cartilage in OA is a clinical challenge. Because cartilage is an avascular and aneural tissue, normal mechanisms of tissue repair through recruitment of cells to the site of tissue destruction is not feasible. Current symptomatic treatments of OA by antiinflammatory drugs do not alter the progression of the disease (30). PG depletion induced by IL-1β, a principal mediator in OA, is a major factor in the onset and progression of joint destruction. Therefore, factors that accelerate PG production and deposition could safeguard against progression of OA. In an attempt to increase the anabolic activity of chondrocytes in OA cartilage, different studies reported the use of growth factors and bone morphogenetic proteins (12). Although these agents have shown some beneficial effects, they generally have shortcomings that limit their clinical application (14, 31, 32).

In this study, we developed a gene transfer strategy aimed to specifically stimulate GAG synthesis in an attempt to promote cartilage repair. GlcAT-I enzyme, which is responsible for the completion of the GAG–protein linkage tetrasaccharide sequence of PGs, was selected as a candidate gene with regard to its crucial role in GAG synthesis, as shown here and elsewhere (18, 33). We demonstrated that inhibition of GlcAT-I protein expression by antisense strategy resulted in a strong reduction of cartilage GAG synthesis and deposition, clearly indicating that GlcAT-I is an essential enzyme for matrix assembly. Likewise, loss of GlcAT-I activity in Chinese hamster ovary cell mutant led to a major defect in GAG synthesis (18). These observations paralleled studies in Caenorhabditis elegans, which showed that homozygous L4 worms derived from heterozygous hermaphrodites bearing sqv-8 gene (GlcAT-I equivalent) exhibited a dramatic reduction in CS-containing PGs (33).

However, it has also been suggested that GlcAT-I enzyme may be rate-limiting in GAG synthesis. This hypothesis was supported by Bai et al. (18), who showed that transfection of GlcAT-I-deficient Chinese hamster ovary cells, with GlcAT-I cDNA augmented the level of GAG synthesis by 2-fold compared with that seen in wild-type cells. Here, we showed that overexpression of GlcAT-I in chondrocytes and cartilage led to a significant stimulation of PG synthesis and deposition, indicating that the rate of GAG synthesis is, to some extent, related to the level of GlcAT-I expression.

Analysis of GAG chains by metabolic labeling showed that the chains produced by GlcAT-I-transfected cartilage are of similar size but more abundant than those made by mock-transfected cartilage. In agreement, FACE analysis indicated a higher amount of CS disaccharides. Furthermore, immunohistochemical and immunoblot analyses showed an increased number of CS stubs epitopes in GlcAT-I-transfected cartilage. These observations suggest that stimulation of GAG synthesis by GlcAT-I expression is due to an augmentation in the number of initiated CS chains and not to a lengthening of their size. In contrast, treatment with IL-1β led to a reduction in the abundance of GAG chains, as suggested by metabolic labeling and confirmed by FACE analysis, which showed a significant reduction in the amount of CS disaccharides. Furthermore, immunohistochemistry and Western blotting suggested that the cytokine produced a reduction in the number of GAG chains.

Our data indicate that GlcAT-I overexpression could efficiently counteract the inhibitory effects of IL-1β and maintain the anabolic activity of chondrocytes and cartilage explants in terms of PG synthesis and accumulation, as shown by 35S incorporation, histology, estimations of GAG content, and FACE analysis. The importance of GlcAT-I in the initiation of GAG chains synthesis, together with the inhibitory effect of IL-1β on its expression (18, 22), may account for the ability of this enzyme to overcome the deleterious effect of IL-1β when its activity was restored by gene delivery.

Our next goal was to investigate whether GlcAT-I gene delivery may have a beneficial effect in the latter stages of OA. Our findings demonstrated that GlcAT-I was able to restore PGs to normal levels in cartilage previously depleted from endogenous PGs by IL-1β. In concert, our investigations strongly indicated that GlcAT-I was capable of controlling and reversing the articular cartilage defect, in terms of PG anabolism, associated with IL-1β.

In OA, the loss of CS GAGs results in a reduction of global negative charge leading to impairment of the physicochemical and biological properties of PGs, such as tissue hydration, elasticity, and growth factor interaction. In this respect, a significant increase in CS synthesis gained upon GlcAT-I gene delivery may help to restore the structural and functional integrity of OA cartilage. It is also possible that increased production of CS GAGs may promote interactions with growth factors or other cytokines that may assist matrix repair by the chondrocytes and/or inhibit the degradation process. Interestingly, the protective effect of exogenous CS in the acute degradation of articular cartilage was reported in ref. 34.

In this study, we demonstrate that cultured chondrocytes and cartilage explants could be engineered by a nonviral gene delivery strategy to efficiently boost PG synthesis. This approach opens promising perspectives to enhance repair of cartilage defects by ex vivo GlcAT-I transfection of chondrocytes or cartilage grafts before transplantation. Furthermore, we show that overexpression of GlcAT-I was not only able to overcome the inhibitory effects of IL-1β on PG synthesis and deposition but also could stimulate the replenishment of GAG content in depleted cartilage. Altogether, this study supplies evidence for a potential cartilage protective effect of gene therapy by targeting the biosynthetic pathway of PGs. Besides overexpression of factors capable of antagonizing IL-1β or other mediators of matrix degradation, the introduction of a transgene capable of stimulating PG synthesis may provide a promising strategy to maintain and repair cartilage.


This work was supported by Fonds National pour la Science (Actions Concertées Incitatives), the IT2B program, the Programme National de Recherches sur les Maladies Ostéo-Articulaire from the Institut National de la Santé et de la Recherche Médicale, the Ligue Régionale contre le Cancer, the Contrat de Programme de Recherche Clinique, and the Programme Hospitalier de Recherche Clinique Centre Hospitalier Universitaire. Nancy. N. Venkatesan is recipient of a fellowship from the Centre National de la Recherche Scientifique and from the Ministère de la Recherche.


This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: CS, chondroitin sulfate; FACE, fluorophore-assisted carbohydrate electrophoresis; GlcAT-I, β1,3-glucuronosyltransferase-I; GAG, glycosaminoglycan; KS, keratan-sulfate; PG, proteoglycan; OA, osteoarthritis; HA, hyaluronan.


1. Felson, D. T. & Zhang, Y. (1998) Arthritis Rheum. 41, 1343–1355. [PubMed]
2. Hashimoto, S., Ochs, R. L., Rosen, F., Quach, J., McCabe, G., Solan, J., Seegmiller, J. E., Terkeltaub, R. & Lotz, M. (1998) Proc. Natl. Acad. Sci. USA 95, 3094–3099. [PMC free article] [PubMed]
3. Fernandes, J. C., Martel-Pelletier, J. & Pelletier, J. P. (2002) Biorheology 39, 237–246. [PubMed]
4. Evans, C. H. & Robbins, P. D. (1999) Rheum. Dis. Clin. N. Am. 25, 333–344. [PubMed]
5. Smolen, J. S. & Steiner, G. (2003) Nat. Rev. Drug Discovery 2, 473–488. [PubMed]
6. Feldmann, M. & Maini, R. N. (2003) Nat. Med. 9, 1245–1250. [PubMed]
7. Moreland, L. W., Margolies, G., Heck, L. W., Jr., Saway, A., Blosch, C., Hanna, R. & Koopman, W. J. (1996) J. Rheumatol. 23, 1849–1855. [PubMed]
8. Ghivizzani, S. C., Lechman, E. R., Kang, R., Tio, C., Kolls, J., Evans, C. H. & Robbins, P. D. (1998) Proc. Natl. Acad. Sci. USA 95, 4613–4618. [PMC free article] [PubMed]
9. Pelletier, J. P., Caron, J. P., Evans, C., Robbins, P. D., Georgescu, H. I., Jovanovic, D., Fernandes, J. C. & Martel-Pelletier, J. (1997) Arthritis Rheum. 40, 1012–1019. [PubMed]
10. Makarov, S. S., Olsen, J. C., Johnston, W. N., Anderle, S. K., Brown, R. R., Baldwin, A. S., Jr., Haskill, J. S. & Schwab, J. H. (1996) Proc. Natl. Acad. Sci. USA 93, 402–416. [PMC free article] [PubMed]
11. van de Loo, F. A. & van den Berg, W. B. (2002) Rheum. Dis. Clin. N. Am. 28, 127–149. [PubMed]
12. van den Berg, W. B., van der Kraan, P. M., Scharstuhl, A. & van Beuningen, H. M. (2001) Clin. Orthop. 391, S244–S250. [PubMed]
13. van Beuningen, H. M., van der Kraan, P. M., Arntz, O. J. & van den Berg, W. B. (1994) Lab. Invest. 71, 279–290. [PubMed]
14. Mi, Z., Ghivizzani, S. C., Lechman, E., Glorioso, J. C., Evans, C. H. & Robbins, P. D. (2003) Arthritis Res. Ther. 5, R132–R139. [PMC free article] [PubMed]
15. Iozzo, R. V. (1998) Annu. Rev. Biochem. 67, 609–652. [PubMed]
16. Prydz, K. & Dalen, K. T. (2000) J. Cell Sci. 113, 193–205. [PubMed]
17. Kitagawa, H., Ujikawa, M. & Sugahara, K. (1996) J. Biol. Chem. 271, 6583–6585. [PubMed]
18. Bai, X., Wei, G., Sinha, A. & Esko, J. D. (1999) J. Biol. Chem. 274, 13017–13024. [PubMed]
19. Gulberti, S., Fournel-Gigleux, S., Mulliert, G., Aubry, A., Netter, P., Magdalou, J. & Ouzzine, M. (2003) J. Biol. Chem. 278, 32219–32226. [PubMed]
20. Ouzzine, M., Gulberti, S., Levoin, N., Netter, P., Magdalou, J. & Fournel-Gigleux, S. (2002) J. Biol. Chem. 277, 25439–25445. [PubMed]
21. Ouzzine, M., Gulberti, S., Netter, P., Magdalou, J. & Fournel-Gigleux, S. (2000) J. Biol. Chem. 275, 28254–28260. [PubMed]
22. Gouze, J.-N., Bordji, K., Gulberti, S., Terlain, B., Netter, P., Magdalou, J., Fournel-Gigleux, S. & Ouzzine, M. (2001) Arthritis Rheum. 44, 351–360. [PubMed]
23. Larsson, S. E. & Kuettner, K. E. (1974) Calcif. Tissue Res. 14, 49–58. [PubMed]
24. Farndale, R. W., Buttle, D. J. & Barrett, A. J. (1986) Biochim. Biophys. Acta 883, 173–177. [PubMed]
25. Bradford, M. (1976) Anal. Biochem. 76, 248–254. [PubMed]
26. Heinegard, D., Sommarin, Y., Hedbom, E., Wieslander, J. & Larsson, B. (1985) Anal. Biochem. 151, 41–48. [PubMed]
27. Carney, S. L., Billingham, M. E., Caterson, B., Ratcliffe, A., Bayliss, M. T., Hardingham, T. E. & Muir, H. (1992) Matrix 12, 137–147. [PubMed]
28. Calabro, A., Hascall, V. C. & Midura, R. J. (2000) Glycobiology 10, 283–293. [PubMed]
29. Funderburgh, J. L., Mann, M. N. & Funderburgh, M. L. (2003) J. Biol. Chem. 278, 45629–45637. [PMC free article] [PubMed]
30. Hochberg, M. C., Altman, R. D., Brandt, K. D., Clark, B. M., Dieppe, P. A., Griffin, M. R., Moskowitz, R. W. & Schnitzer, T. J. (1995) Arthritis Rheum. 38, 1535–1540. [PubMed]
31. Loeser, R. F., Todd, M. D. & Seely, B. L. (2003) J. Rheumatol. 30, 1565–1570. [PubMed]
32. Glansbeek, H. L., van Beuningen, H. M., Vitters, E. L., Morris, E. A., van der Kraan, P. M. & van den Berg, W. B. (1997) Arthritis Rheum. 40, 1020–1028. [PubMed]
33. Bulik, D. A., Wei, G., Toyoda, H., Kinoshita-Toyoda, A., Waldrip, W. R., Esko, J. D., Robbins, P. W. & Selleck, S. B. (2000) Proc. Natl. Acad. Sci. USA 97, 10838–10843. [PMC free article] [PubMed]
34. Uebelhart, D., Thonar, E. J., Zhang, J. & Williams, J. M. (1998) Osteoarthritis Cart. 6, Suppl. A, 6–13. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...