Knee osteoarthritis, lumbar-disc degeneration and developmental dysplasia of the hip - an emerging genetic overlap

Genetics highlights relationships between biological systems, and as the number of defined osteoarthritis susceptibility alleles increases, there is the natural tendency to assess whether the alleles influence other musculoskeletal phenotypes. That has proven to be the case for the GDF5 polymorphism rs143383, a risk factor for knee osteoarthritis and several other common conditions, including lumbar-disc degeneration and developmental dysplasia of the hip. Another interesting example has recently emerged in the repeat polymorphism of the asporin gene, ASPN, which is also associated with these three phenotypes. Such discoveries increase our understanding of shared disease etiology but also emphasize the complexity of common genetic risk.

Pleiotropy refers to the capacity of a DNA mutation to have an eff ect on many biological systems and is a consequence of the fact that, during life, gene products are typically expressed, as so-called spatial and temporal eff ects, in a variety of cell types at a number of diff erent times. Pleiotropy of common alleles is an expected feature of human diseases that have a major genetic compo nent and manifests as an allele associated with several diff erent phenotypes. Th e musculoskeletal system is no exception and in the previous issue of Arthritis Research & Th erapy Shi and colleagues [1] brought another example to our attention. In their study of Han Chinese, the investigators reported an association of the D-repeat polymorphism of the asporin gene, ASPN, with developmental dysplasia of the hip (DDH). Th is polymorphism had originally shown an association with knee osteoarthritis (OA) and subsequently with lumbardisc degeneration (LDD), also in Asian cohorts [2,3].
Asporin is an extracellular matrix (ECM) macro molecule belonging to the small leucine-rich proteo glycan (SLRP) protein family. SLRP family members are able to bind other structural components of the ECM, such as collagen, as well as growth factors that tempor arily reside in the ECM, such as members of the trans forming growth factor-beta (TGF-β) superfamily. Asporin is present in a number of tissues, including cartilage, and the D-repeat polymorphism of ASPN resides within exon 2 of the gene. It gets its name from the fact that it is a triplet repeat coding for a polymorphic stretch of residues of aspartic acid (D is the one-letter code for this particular amino acid). Depending on the ethnic group, the repeat has up to 12 alleles encoding 8 to 19 D residues.
In the original OA association study, it was reported that the allele coding for 14 D-repeats (D14) was a risk allele for knee OA [2]. Th is was also the case for LDD [3] and is the case in the recent DDH study [1]. Furthermore, the D13 allele was found to be protective for knee OA, something also observed for DDH. OA and LDD can be viewed as having similar etiological routes, both of which involve degeneration of cartilage, albeit of diff erent types; that of OA is hyaline and that of LDD is fi brocartilage. Th ey are also diseases associated with aging. However, DDH is not associated with aging but is instead congenital, nor is it a degenerative disease but instead involves the displace ment of the proximal femur from the acetabulum.
What is particularly intriguing, therefore, is how the same alleles at the ASPN D-repeat contribute to phenotypes as diff erent as OA/LDD and DDH. Functional studies have suggested that asporin attenuates TGF-β signaling and that the extent of the attenuation is infl uenced by the number of D-repeats [2,4,5]. Perhaps,

Abstract
Genetics highlights relationships between biological systems, and as the number of defi ned osteoarthritis susceptibility alleles increases, there is the natural tendency to assess whether the alleles infl uence other musculoskeletal phenotypes. That has proven to be the case for the GDF5 polymorphism rs143383, a risk factor for knee osteoarthritis and several other common conditions, including lumbar-disc degeneration and developmental dysplasia of the hip. Another interesting example has recently emerged in the repeat polymorphism of the asporin gene, ASPN, which is also associated with these three phenotypes. Such discoveries increase our understanding of shared disease etiology but also emphasize the complexity of common genetic risk. therefore, the D-repeat genotype, with its ability to alter the length of the run of aspartic acids, combined with expected diff erences in the expression levels of the various members of the TGF-β superfamily during develop ment and in mature tissues, infl uences which skeletal sites are aff ected and which phenotypes result.
Th e reality will obviously be far more complex than this. For example, being infl uenced by genotypes at other loci and by non-genetic factors, the association of the D-repeat with OA, LDD, and DDH is far from penetrant. Th is is apparent when one recalls that, although the D-repeat can have an impact on knee OA in Asians, it does not contribute to knee OA in Europeans: genetic or non-genetic factors must account for this ethnic diff erence [6]. Furthermore, the D-repeat is associated with an aberrant hip shape that results in DDH in the young but is not associated with an increased risk of hip OA in older patients; in the latter, the eff ect on hip OA is presumably a temporal one.
Another example in the literature of pleiotropy of an OA locus is the single-nucleotide polymorphism rs143383, which is located in the 5' untranslated region of GDF5. Th is gene codes for the TGF-β superfamily member growth and diff erentiation factor 5. Like the ASPN D-repeat, rs143383 was fi rst reported to be associated with increased risk of large-joint OA, principally of the knee [7]. rs143383 was then shown to be associated with other joint-related disease phenotypes, including DDH and LDD [8][9][10]. Th ere are, therefore, striking similarities between the D-repeat and rs143383 associations; the obvious similarity is that when it comes to lower-limb OA, both variants are particularly relevant to knee disease, but when it comes to early development, the hip is the main repository of the genetic eff ect. We shall have to wait for the discovery of additional susceptibility alleles to assess whether this shared genetic risk between OA, LDD, and DDH is more widespread or whether it is restricted to members of the TGF-β signaling pathway. One thing that we can be certain of is that late-onset musculoskeletal diseases can inform our understanding of developmental conditions, and vice versa.