This page will trace the history of multiple ephiphyseal dysplasia in the context of it being recognised as a distinct clinical entity; the delineation of the ‘Ribbing’ and ‘Fairbank’ forms; identification of the genetic causes; refining of disease mechanisms and finally the identification of therapeutic targets.
The recognition of MED as a distinct clinical entity
The first half of the 19th Century saw MED described as a recognisable and ‘distinct clinical entity’. However, the initial clinical reports used a variety of different terminologies including:-
“Double coxa vara with other deformities” (Barrington-Ward 1912)
“Chondro-dystrophia Foetalis” (White 1924)
“Dwarf with stippled epiphyses” (Buxton 1930)
“Epiphysial dysostosis” (Jansen 1934)
“Hereditary deforming dyschondroplasia” (Gardiner-Hill 1937)
Finally, in his 1945 paper in the “Proceedings of the Royal Society of Medicine”, Thomas Fairbanks presented a series of 15 cases (including those previously described under different terminology) and stated that:-
“It is the purpose of this paper to describe the features of a condition which I believe should be regarded as a clinical entity, and as a developmental error resulting from unknown cause”
Initially, Fairbank’s had suggested the term “Epiphyseal Dysplasia, Generalisata” in 1935, but this was later changed to “Epiphyseal Dysplasia, Multiplex” and cases under this new title were first shown by Wiles and Yarrow at the 1938 meeting of the Royal Society of Medicine.
The term “Epiphyseal Dysplasia, Multiplex” continued to be used through the 1940s and 50s (Fairbanks 1947; Waugh 1952) until Edmund Shephard first used the now well-recognised term “Multiple Epiphyseal Dsyplasia” in his 1956 publication of the same name. Later papers by Herbert Barrie (1958) and Cyril Monty (1962) reinforced the use of this name.
Delineation of the ‘Ribbing’ and ‘Fairbank’ forms
Content to follow
Identification of the genetic causes
In many ways the 1958 seminal paper by Herbert Barrie, Cedric Carter and John Sutcliffe, at the Hospital of Sick Child in Great Ormond Street, was the first step in identifying the genetic causes of MED.
By 1958 Barrie and his colleagues knew the following:-
- some human disease could be attributed to “genetic causes” (Sir Archibold Garrod 1902).
- that “one gene encodes one protein” (Beadle and Tatum 1941).
- that “DNA was the molecule that mediates heredity” (Avery, MacLeod and McCarty 1944).
- that “DNA was the molecule that mediates heredity” (Hershey and Chase 1952).
- that “DNA was a double helix” (Watson and Crick 1953).
However, the concept of ‘genetic heterogeneity’ in human disease had not been considered and in 1958 no genetic cause to any human disease had yet been identified. It is therefore remarkable that on the basis of clinical presentation alone Barrie and colleagues concluded that:-
“This suggest that two genes at least are responsible – one for the relatively mild form seen in some cases, and another for the more severe form: but to prove this point conclusively it would be necessary to take more radiographs than is justified for a point of purely theoretical importance.”
The candidate gene approach (1988-1992)
The first attempts to identify the genetic basis of MED (and the phenotypically related pseudoachondroplasia; PSACH) relied on a candidate gene approach using restriction fragment length polymorphism’s (RFPLs) in small nuclear families. However, these studies failed to link PSACH or MED to any of the genes being tested.
Single family linkage analysis (1991-1993)
The ascertainment and collection of DNA from two multi-generation families with MED and mild PSACH/MED respectively was the turning point for pin-pointing the chromosomal localisation of a PSACH-MED gene to the pericentromeric region of chromosome 19 as reported in Briggs 1993 and Oehlmann 1994. The identification of these genetic markers then allowed a the gene in series of smaller PSACH families to be localised to the same region in Hecht 1993.
However, the story starts several years earlier when Dr David Rimoin of Cedars-Sinai Medical Centre had the opportunity to visit Hawaii in the early 1990s and was able to ascertain a large family with features of both mild pseudoachondroplasia and multiple epiphyseal dysplasia.
The phenotype of this family was characterised by a waddling gait, short limbs and early onset osteoarthritis. The radiographic presentation resembled pseudoachondroplasia in childhood and multiple epiphyseal dysplasia in adults.
Genetic linkage analysis performed in Dan Cohn’s lab at Cedars-Sinai Medical Centre in Los Angeles and in James Weber’s lab at the Marshfield Medical Centre in Wisconsin identified the first genetic marker linked to disease in this family.
This anonymous genetic marker was called D19S253 and it was a tetra-nucleotide tandem repeat located on chromosome 19 and provided the first evidence for the location of the gene that caused mild pseudoachondroplasia/multiple epiphyseal dysplasia in this family. Subsequent analysis with neighbouring genetic makers (D19S199, D19S222 and D19S49) confirmed the location of this gene to a 6.3cM interval in the pericentromeric region of chromosome 19 between 19p13.1 and 19q12.
The cover of the December issue of Genomics featured the image of a chondrocyte showing the typical lamella appearance of retained material in pseudoachondroplasia. This image was courtesy of Helen Gruber (CSMC) and its selection as the cover image was most likely due to the then Editor, Victor McKusick’s long-standing interest in skeletal dysplasias, and pseudoachondroplasia in particular.
This localisation of this mild PSACH/MED gene to chromosome 19 then allowed the testing of smaller families with typical forms of PSACH by Jacqui Hecht in Texas. This analysis confirmed a similar genetic linkage and the paper detailing this study was published back-to-back in Genomics.
In a completely independent study, Rob Knowlton and colleagues in Philadelphia demonstrated linkage of a large family with multiple epihpyseal dysplasia to the same region of chromosome 19.
Confirmation of genetic heterogeneity and identification of the EDM2 (COL9A2) locus
British Family A with MED
The availability of DNA from a large 3-generation British family with mild MED allowed the identification of the second genetic loci (EMD2). Interestingly, this family had originally been described by Barrie and colleagues in their seminal paper of 1955. Furthermore, the same family had been studied by Paul Wordsworth and colleagues in 1988 and used to exclude type II collagen as the causative gene.
Unfortunately, at that time it had not been possible to identify the precise mutation in COL9A2.
Dutch Family V with MED
As part of his Thesis work Jan van Mourik ascertained a large Dutch family with a mild form of MED.
Several studies have since confirmed COL9A2 at the EDM2 locus and demonstrated that all the MED-causeing mutations are clustered in the splice donor site of exon 2, which is a very remarkable grouping and points to the COL3 domain of type IX collagen as an important disease mechanism.
COL9A1 (EDM6) and COL9A3 (EDM3) become the next MED genes
In separate linkage studies in two unrelated multigenerational families Petteri Paasilta in Leena Ala-Kokko’s lab and Carsten Bonnemann in Louis Kunkel’s lab identified COL9A3 as the next genetic loci for autosomal dominant MED (EDM3).
Finally, Malwina Czarny-Ratajczak idenfitied COL9A1 as the final collagen IX geneitc loci for MED (EDM6).