Discussion

Introduction:
Joint disorders rank among the primary causes for wastage in performance horses and the articular issue that is by far most troublesome is the cartilage, which has an extremely low healing capacity. This insufficient repair capacity is strongly related to the extremely long turnover time of hundreds of years of collagen type II that forms the backbone of the extracellular matrix of articular cartilage (Maroudas 1980). These turnover times apply to mature individuals, but obviously not to young, growing individuals in which there is an active process of continuous remodelling. It lies therefore at hand to suppose that, if cartilage is to be conditioned (like muscular tissue or bone), this can only be achieved at a young age.

Physiological development of articular cartilage:
Due to anatomical and geometric particularities, joints are not loaded evenly over their surface. These variable loading conditions can only adequately be met by cartilage that possesses different mechanical properties in different sites and hence features different biochemical and ultrastructural characteristics, i.e. shows topographical heterogeneity. There are 2 possible origins of the topographical heterogeneity: genetic predetermination or formation in the early juvenile period. Work by Brama and co- authors showed that biochemical composition was homogeneous across the joint of the newborn foal, which situation they called the 'blank joint' (Brama et al. 2002). Topographical heterogeneity formed quickly in the early juvenile period and most was already present at age 5 months. Later, the underlying maturation process of articular cartilage in early life was indirectly confirmed using equine-specific cDNA micro-arrays, showing a distinct difference between neonatal and mature individuals in the expression of genes encoding matrix proteins and matrix modifying enzymes (Mienaltowski et al. 2008).
The changes during early life occur in a structural sense as well. The arcade structure of the collagen network (Benninghoff 1925), is typical for mature cartilage, but not yet present in the newborn.

Influencing the normal maturation process:
The transition from a blank, neonatal, joint to the mature joint featuring topographical heterogeneity, also called the process of functional adaptation, is driven by biomechanical loading, as shown by the so-called 'EXOC' study (van Weeren and Barneveld 1999). In this study 3 groups of foals were exercised differently from age one week onwards. One group was kept in box stalls for 24 h per day; one in identical box stalls, but additionally subjected to short bouts of exercise; the third group had free pasture exercise. The exercise regimens were maintained until weaning at 5 months of age when 24 foals were subjected to euthanasia and the remaining 19 subjected to a moderate exercise regimen, followed by euthanasia at 11 months. Topographical heterogeneity developed in the pastured group and in the boxed/sprinted group alike, but failed to develop in the boxed foals. This was direct proof of the steering effect of biomechanical loading. The most interesting observation was that, where proteoglycans and other factors such as subchondral bone density became normal after the common training programme from 5 - 11 months, collagen parameters remained abnormal. Therefore, for collagen there seems to be a limited window in time when the process of functional adaptation can take place.
In the EXOC study the pasture group came out best, raising the question whether the combination of pasture exercise and superimposed additional exercise would result in even better conditioning of the cartilage. In the GEXA study (Rogers et al. 2008) foals were raised divided into 2 groups. One group had free pasture exercise; the other was also subjected to additional exercise, increasing workload by a moderate, but significant, 30%.
The differences in exercise load in the GEXA study were much less extreme than in the earlier EXOC study, but a detailed study of contiguous 100 micrometre thick slices taken from the proximal articular surface of the proximal phalanx down to the tidemark at differently loaded sites showed exercise-related changes, which were interpreted as a precocious cessation of collagen remodelling at this site due to a load-induced speeding up of the normal maturation process (van Weeren et al. 2008).

Conclusion:
Articular cartilage is responsive to biomechanical loading and responsiveness is highest in the youngest individuals. The characteristic topographical heterogeneity of the extracellular matrix of mature cartilage is formed through physical exercise. Where the proteoglycan component of cartilage may still respond to exercise in (young) adult horses, no adaptive response of collagen has ever been demonstrated in this age group, making the early juvenile period the only opportunity to manipulate the collagen network. Lack of exercise seems deleterious, but no definitive answer can be given yet about the possible beneficial effect of additional exercise. For this, long-term epidemiological studies are necessary.