Abstract Purpose: Articular cartilage, is of great importance for physiological mobility. The structure and the function of this tissue, which is characterized by a poor self-repair capacity, are frequently disrupted or damaged upon physical trauma or in degenerative osteoarthritis (OA). Given that musculoskeletal disorders are the leading cause of poor performance or early retirement of sports and race horses, treating horses for these disorders is relevant. The proposed therapeutic approach, which was first developed for human, has the potential of being considered as a pre-clinical step for human medicine because the horse is recognized as an excellent model for the study of articular cartilage disorders in humans. This study aims to improve the Autologous Chondrocytes Implantation (ACI) technique by using Mesenchymal Stem cells (MSCs) from bone marrow (easy to collect in horse) as the cell source. Thus, MSCs were, first, characterized before being cultured with chondrogenic conditions in order to find the combination that best enhances and stabilizes the characteristics of the chondrocyte phenotype, in order to perform clinical trials in horse. Methods: MSCs from equine bone marrow were isolated and expanded in monolayer cultures until 4 passages. These cells were characterized by analyzing their proliferative potential, their pluripotency (with microenvironmental stimuli) and their capacity to express MSCs specific phenotypic markers defining the MSCs (flow cytometry). In parallel, MSCs differentiation in chondrocytes was accomplished via a combinatory approach based on the association of 3D-culture in type I collagen sponges, low oxygen tension with chondrogenic factors (BMP-2, TGF-?1) and RNA interference (siRNA to down-regulate type I collagen and HtrA1 protein expression). Finally, an extensive analysis at gene and protein levels corresponding to differentiated, dedifferentiated and hypertrophic chondrocyte phenotypes was performed. Results: Our results show a very high proliferation potential of MSCs isolated from equine bone marrow. Furthermore, the isolated cells satisfy the various criteria of stem cells definition (pluripotency, expression of surface markers). In addition, siRNAs targeting equine Col1a1 and Htra1 have been functionally validated. Finally, we show that the BMP-2 and TGF-?1 combination strongly induces the differentiation of MSCs in chondrocytes. Thus, the combined use of specific culture conditions defined within the laboratory, with specific growth factors and siRNAs association leads to the in vitro synthesis of a hyaline type neo-cartilage by chondrocytes which present an optimal phenotypic index, comparable to the one established with induction of specific differentiation of human MSCs in chondrocytes, and close to the one of differentiated/healthy chondrocytes. Conclusions: These data represent a first step in the development of equine clinical trials, which are planed - to better understand the reaction of cartilage tissue after a few weeks of intra-articular implantation and - to make the proof of concept in a large animal model. This approach will allow us to explore the criteria necessary to begin to consider the development of cell therapy for cartilage repair in humans.