Skeletal development and osteoarthritis.

The limited ability of articular cartilage to repair after traumatic insult or during arthritic degeneration is widely attributed to the fact that the tissue lacks both innervation and vascularisation. However, comparative studies of articular cartilage repair in load bearing and non-load bearing areas of the joint, in addition to some in vitro studies suggest more strongly that adult articular cartilage has, intrinsically, a reasonable repair potential ifnot ability. ' The possible reasons for this repair limitation will be discussed below, but it is not always the case and, indeed, there are many reports in the developmental literature of substantial repair and/or regenerative capacity of cartilage. An excellent example is the newt limb which, after amputation, not only regenerates the articular cartilage, but also the entire cohort of lost skeletal elements and in the correct spatial pattern. Whilst mammals do not possess such a regenerative potential, during embryology, there can be a considerable degree of regulation which allows for replacement of lost parts.2 At what point in development is the repair capacity of mammalian cartilage restricted and for what reasons? Repair mechanisms apart, we may also ask what can we learn about degenerative joint disease by studying the developing constituent tissues of a joint? Many of the processes which occur during both pathology and repair, seem to mimic some of those processes which occurred during embryonic development. In the following pages I will consider some of the features which appear common to both osteoarthritic cartilage and developing cartilage. Such features include cell proliferation, elevated matrix synthesis and morphogenesis; taken together, these processes constitute growth and in osteoarthritis this is manifest as osteophyte formation. Before addressing these aspects, it is worth considering the primary components of the development of a cartilaginous element of the appendicular skeleton. Most work on the early development of the appendicular skeleton has been carried out on the embryonic chick. The chick is easily obtained, is cheap and readily accessible through holes cut in the shell. The incubation period of 21 days also facilitates experimental manipulation. In contrast, uterine development makes experimental manipulation of mammalian embryos very difficult. In addition, because the developing skeleton of the chick limb has been used widely as a model for pattern formation, there is a comprehensive literature on most aspects of early skeletogenesis. Whilst the relative lack of data on mammalian species is inconvenient, it is not restrictive since there is a tendancy for the more fundamental processes in embryology to be conserved across vertebrate classes. Consequently, most of the processes and concepts mentioned below will have been obtained and formulated through studies on the chick embryo, but to our knowledge, most will hold for the mammalian counterpart. The developing limb grows out from the flank of the embryo as an ectodermal outpushing containing mesenchyme which is derived from the somatopleure. At the tip of the early limb, a specialised thickening of the ectoderm develops which is known as the apical ectodermal ridge (AER).3 This ectodermal specialisation is essential for continued outgrowth through inductive influences and the underlying mesoderm, in turn, maintains the AER. Thus limb outgrowth is dependent on reciprocal ectodermal/mesodermal interactions3 which are required throughout the specification of the entire limb skeleton.4 5 For some time, the limb mesenchyme appears as a homogeneous population (histologically) fairly evenly distributed throughout the limb. About 12 hours before overt chondrogenesis (at 4 days of incubation), the presumptive cartilage cells condense to form prechondrogenic condensations which lie in the positions of the future skeletal elements and appear in a proximo-distal sequence. Consequently, the first to form in the leg are the condensations of the femur, tibia/fibula. The nature and significance of the prechondrogenic condensation has been the subject of some interest. Considerable cell/cell contact takes place within the condensation6 with gap junctions subsequently being identified7 and recent work has shown that the cells are able to transfer the dye lucifer yellow.8 Whilst some authors stress the importance of the interactions which occur within the prechondrogenic condensation it is, nevertheless, true that single mesenchymal cells isolated from the very early limb before condensation, can become chondrogenic when maintained under appropriate in vitro conditions.9 Consequently, the condensation process does not appear to be a prerequisite for chondrogenic differentiation, but may well facilitate the process by maintaining the cells in a rounded configuration which favours chondrogenesis.' It is more likely that condensation is essential for accurate morphogenesis. Before the synthesis of type II collagen (at 4-5 days incubation), the prechondrogenic condensation is strongly immunoreactive for type I collagen," fibronectin'2 and PG-M.'3 Department of Anatomy, University ofWales College of Cardiff, Cardiff, United Kingdom CW Archer