Galectin-3 in osteoarthritis: from a protective to a destructive role

Background

Osteoarthritic (OA) chondrocytes are able to re-express numerous genes normally activated in the growth plate, and more particularly in the hypertrophic zone. Among several genes, we are interested in studying galectin-3 (gal-3) since we have recently demonstrated that its expression was increased in OA cartilage [1]. gal-3 is a mammalian lectin, which interacts with β-galactoside residues and is involved in numerous functions such as adhesion, splicing activity, cell cycle regulation, as well as a receptor for advanced glycation end products (AGE receptor). These functions are related to the gal-3 cellular localization. Indeed, this protein may be found in the plasma membrane, in cytoplasm and in the nucleus.

Objective

In the present study, we investigated the role(s) of gal-3 using both the monoiodoacetate-induced OA model and in vitro experiments.

Methods

OA was induced by a single injection of iodoacetate (5 mg/ml, 2 μl) into each knee joint of 4-month-old mice (WT) or gal-3 null mice (KO). Mice were sacrified 7, 14 and 21 days after the single injection. Histologic evaluation was performed on sagittal sections of mouse knee joint. The severity of the OA lesions was graded on a scale of 0–14 in a blinded fashion, by two independent observers, using the histologic/histochemical scale of Mankin. Intracellular and extracellular roles of gal-3 were investigated in both human chondrocytes and in chondrogenic ATDC5 cells, a mouse cell line derived from the 129 strain.

Results

Intra-articular injection of monoiodoacetate, which induced osteoarthritis, upregulated the expression of gal-3 in WT mice 7 days post injection, reaching a statistical significance 14 days post injection (P < 0.05). The histologic grading score indicated that KO mice (control group) had a poorer quality of cartilage compared with WT mice (control group). Moreover, the induction of OA in KO mice showed a marked decreased of bone area, noticeable 7 days post injection (P < 0.05).

According to the results obtained, it seemed that gal-3 was important for the cartilage homeostasis. Colnot and colleagues have suggested that gal-3 could be implicated in chondrocyte survival [2]. Therefore, we treated OA chondrocytes with sodium nitroprusside (SNP), which is known to generate chondrocyte cell death. Our results showed that gal-3 was much further decreased than was Bcl2 in experiments performed under the same conditions [3]. Moreover, SNP decreased the gal-3 phosphorylation, which is a key process in the capacity of gal-3 to prevent cell death. Finally, ATDC5 cells transfected with a gal-3-expressing vector were more resistant to SNP-induced cell death compared with those transfected with the empty vector. On the other hand, Ohshima and colleagues found gal-3 in synovial fluid, particularly during inflammation [4]. Therefore, we investigated the potential role of exogenous gal-3 in chondrocyte cultures. Surprisingly, we found that exogenous gal-3 induced chondrocyte death.

One of the most fascinating phenomena is the regulation of gal-3 secretion. Indeed, several cells produced gal-3 but not all are able to secrete a great amount of it, chondrocytes belonging to the latter category. Conversely, gal-3 is secreted in a much greater quantity by inflammatory cells that could affect – at least locally (i.e. at the pannus level) – chondrocyte survival.