BoneKEy-Osteovision | Commentary

Of viruses, flies and bumps: Hereditary multiple exostoses



DOI:10.1138/2001037

Hereditary multiple exostoses is a disorder of the growth plate in which cartilaginous excrescences grow laterally from the growth plate and subsequently ossify. In addition to these deformities, the affected limb may be shortened and bowed. Hereditary multiple exostoses is the commonest of inherited bone tumors, with a prevalence of about 1/50,000-1/100,000. The disorder is dominantly inherited and genetically heterogeneous: about 70% of multiple exostoses families show linkage to markers on chromosome 8 (Type 1, OMIM 133700), other loci are on chromosome 11 (Type 2, OMIM 133701) and chromosome 19 (Type 3, OMIM 600209). Knowledge about hereditary multiple exostoses has taken two strides forwards in the past six years. A few years ago, triangulation between three entirely different lines of work identified two responsible genes, disclosed their function, and pointed to a possible pathogenetic mechanism for the disorder. The second stride has been taken by a series of papers in the past year; it may be a step in another direction, however ().

Genes for hereditary multiple exostoses were found by positional cloning in the regions of 8q24 () and 11p12 (). The two genes encoded related proteins of unknown function called exostocins. A research group interested in cell surface proteoglycans happened upon the function of the exostocins about two years later. Using complementation to restore the infectivity of herpes simplex virus to a heparan sulfate-deficient cell line, they cloned an enzyme necessary for synthesis of heparan sulfate and found the cDNA to be identical to the product of the EXT1 gene (). Subsequent studies have confirmed that both exostocin 1 and its homologue exostocin 2 have glycosyltransferase activities and can add D-glucuronic acid and N-acetylglucosamine molecules to a substrate (). A complex between exostocins 1 and 2 is necessary for full heparan sulfate polymerase activity; this presumably accounts for the phenotypic identity of mutations at the two loci ().

The third intersecting line of work involved identification of the locus for a Drosophila mutation called tout velu, roughly translated as “hairy all over” with a phenotype reminiscent of the Hedgehog phenotype (). Long-range signaling by Hedgehog proteins requires both cleavage of the protein and addition of cholesterol to its N-terminal fragment, which is the long range morphogen (). Hedgehog signaling normally activates gene expression up to 10 cell diameters away from Hedgehog producing cells in the wing disc of the fly, but long-range signaling was abolished by the tout velu mutation, even though Hedgehog could signal to adjacent cells. Other experiments were interpreted as showing that tout velu does not affect secretion of the Hedgehog protein but rather its ability to diffuse and to bind to target cells. Molecular cloning of tout velu disclosed that the gene is the fly ortholog of the human EXT1 gene, and thus a putative heparan sulfate polymerase. The tout velu result indicates that in addition to the poorly understood function of cholesterol in Hedgehog transport, cell surface heparan sulfate proteoglycans are also somehow involved.

How do the EXT genes function in vertebrate development? Ablation of the EXT1 gene in the mouse is lethal at embryonic day 8.5. Homozygous embryos fail to gastrulate and have little organized mesoderm or extraembyonic tissue (). Staining for Indian hedgehog, a member of the vertebrate hedgehog family, is severely reduced in visceral endoderm of EXT1(-/-) embryos, or by treatment with heparitinase. Heterozygous mice are grossly normal except for 10% shortening of their limb bones, but they do not develop exostoses.

How then do we account for the origin of exostoses in humans with EXT mutations? It is likely that hereditary multiple exostoses are benign neoplasms that arise because of loss of heterozygosity (LOH) at the EXT loci in chondrocytes, although only a few lesions have been analyzed for LOH (). A cell-autonomous defect would lead to proliferation of affected cells, which must retain their ability to participate in endochondral bone formation. How would abnormal production of heparan sulfate proteoglycans be linked to proliferation? A gain of function in a cartilage-specific growth factor response would account for the restriction of tumors to the growth plate in the face of widespread expression of the EXT genes. We know, for example, that Indian hedgehog (Ihh) is a critical chondrocyte growth factor in the growth plate, and that association of Ihh with cells is impaired in the EXT1(-/-) mouse (). The Indian hedgehog receptor Patched is not ordinarily expressed in Ihh-expressing cells, but could be in the mutant.

The early collapse of the developmental sequence in the EXT(-/-) mouse is much more severe than the Ihh(-/-) phenotype, however, suggesting that signaling pathways other than Hedgehog are affected by the absence of exostocin 1 in the mouse (in contrast to the fly), and could be in cartilage development as well. Cell surface heparan sulfate proteoglycans are necessary for transport and/or binding of proteins in the Wnt, fibroblast growth factor and Hedgehog families (), all of which have roles in the development of cartilage. To complicate matters further, two recent papers report a marked rearrangement of the actin cytoskeleton in EXT chondrocytes, both in vivo and in cell culture (). This could be the consequence either of altered cell surface heparan sulfate proteoglycans or of altered binding of a growth factor which signals for rearrangement of the cytoskeleton. In either case, the highly characteristic cytological abnormality may be a vital clue to the nature of the chondrocyte abnormality in hereditary multiple exostoses.


Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.