Testing and validation of an in vitro 3D human chondrocyte model to replace animal use in mechanobiology research

Lead Research Organisation: Cardiff University
Department Name: School of Biosciences


During everyday movement e.g. walking, running, climbing stairs, our knee joints are exposed to mechanical forces arising from body weight. Articular cartilage, which lines the ends of our bones, functions to absorb and dissipate the forces experienced by our joints; the cells - called chondrocytes can sense these alterations in force regulating the production of cartilage to support its function. As we age or when our joints are exposed to trauma i.e. sports injury, cartilage can begin to degrade and joint health is negatively impacted.

How do we know so much about this phenomenon? Over the decades, we have relied on the use of animal models to study joint health and to mimic tissue ageing and degeneration. Historically, large animal models including horses and dogs were used, however small animal models, specifically rodents are used for many of these investigative studies today. The rodent models aim to mimic mechanical forces experienced in the joint and by manipulating these forces can determine what happens to the tissue during ageing and disease onset; both surgical and non-surgical approaches are used to alter weight-bearing in the knee, or conversely 'off-loading' where the rear end of a rodent is suspended to prevent weight bearing on the hind legs. Although these models provide information on how the chondrocytes sense and respond to changes in the forces applied, the procedures are considered moderate in severity by the Home Office. Furthermore, fundamental differences between rodents and humans in their anatomy and biomechanics likely contributes to the low success rate for research translation to the clinic. Yet, it has been conservatively estimated that typically 3,708 mice and 486 rats are used per annum worldwide in such experiments.

Why can't we use in vitro models to study these responses? Several in vitro alternatives have been developed to obviate the need for in vivo animal research in line with the 3Rs ethos of refinement, reduction and replacement. Unfortunately, these models fall short in replicating the unique features of articular cartilage and are incapable of forming the correct composition and structural features which give the tissue its highly specialised biomechanical function. Importantly, these models fail to support the extensive communication that exists between chondrocytes and the tissue it resides in which are imperative to how the cells can sense and respond accordingly.

Is there a suitable non-animal technology alternative? We have developed a novel three-dimensional model system which utilises human chondro-progenitors i.e. the precursor cell type to mature chondrocytes that have actively produced a highly organised tissue that develops into a cartilage-like tissue with the correct molecular composition to support mechanical function. However, this model has not previously been used to investigate mechanical responses and forms the basis of this proposal. We aim to determine whether this model responds to physiological and non-physiological forces in a similar way to in vivo animal models to validate it as a replacement system. We will assess how the cells in this human in vitro model respond to load by mapping the forces experienced by cells through the depth of the tissue followed by analysing changes at the gene level. We will then be able to compare the responses to those detected in two in vivo loading models using our previously acquired data to enable validation and provide evidence of utility of this non-animal technology. Once validated, we will widely publicise the model, invite interested users to our laboratory to learn how to establish the model and overall calculate that we can realistically reduce experimental animal use by at least 40% in this field. Use of this human cell derived model could also provide long-term translational impact in facilitating the identification and screening of new targets to prevent cartilage catabolism and preserve joint health.

Technical Summary

Articular cartilage dissipates mechanical forces across the joint surface which is imperative for healthy ageing of the joint. Mechanistic insight of 'mechano-signalling' has been largely obtained from in vivo studies involving rodents, zebrafish, dogs, sheep and horses. These studies, typically moderate in severity, are now predominantly performed on rodents utilising an average of 3,708 mice and 486 rats per annum worldwide. Surgical or non-surgical destabilisation of the knee joint is commonly used to induce abnormal loading profiles. However, these in vivo models do not accurately reflect human cartilage, either in anatomy or biomechanics, and likely contributes to the low success rate for clinical translation. In vitro models are not attractive as they do not recapitulate the phenotypic and biomechanical features of native cartilage. To address this shortfall and promote animal replacement, we have developed a new proprietary human 3D in vitro tissue model which utilises immortalised human chondroprogenitor cells that display cartilage-like properties including depth-dependent zonal stratification, matrix organisation and biomechanical functionality. These immortalised cells provide an accessible source for reproducibly generating stratified cartilage. This proposal will assess model applicability for mechanobiology studies by modelling the mechanical strain experienced by the cells through the construct depth. RNA sequencing will be performed on cell populations exposed to differential strain magnitudes and mechanically regulated mRNAs and miRNAs identified. Validation of transcriptomic data against our existing in vivo loading data will be critically important to ensue confidence in end users that the proposed model is a suitable replacement for animal studies. This 3D human chondrocyte model can provide a more relevant screening platform to support early pre-clinical studies to deliver translational impact in identifying new targets to maintain joint health.


10 25 50