Pycnodysostosis

A number sign (#) is used with this entry because pycnodysostosis is caused by homozygous or compound heterozygous mutation in the cathepsin K gene (CTSK; 601105) on chromosome 1q21.

Clinical Features

The features of pycnodysostosis are deformity of the skull (including wide sutures), maxilla and phalanges (acroosteolysis), osteosclerosis, and fragility of bone. The disorder was first described and named by Maroteaux and Lamy (1962). Andren et al. (1962) simultaneously and independently delineated this syndrome. They found 11 patients reported under various designations and added the cases of monozygotic twins. Some of these cases have probably been diagnosed as osteopetrosis (see OPTB1, 259700), e.g., the case described by Seigman and Kilby (1950) in a black female whose parents were first or second cousins. Kajii et al. (1966) described a Japanese case in the daughter of a first-cousin marriage. From Portugal, Meneses de Almeida (1972) reported 7 cases in 4 families, of whom 3 had consanguineous parents.

Kozlowski and Yu (1972) described a child who had hematologic features, hepatosplenomegaly and anemia, similar to those of osteopetrosis.

The report by Mills and Johnston (1988) of 2 Scottish brothers, born in 1932 and 1944, described the late changes of the disorder, which included irregular resorption of the middle phalanges as well as the terminal phalanges.

In a 24-year-old man with typical features of pycnodysostosis, Figueiredo et al. (1989) found large porencephalic cysts. A brother also had pycnodysostosis. The parents were not known to be related but were born in the same small village in Madeira.

Edelson et al. (1992) examined 14 cases of pycnodysostosis in a small Arab village with 3,000 inhabitants. They pictured 4 affected sibs, including fraternal male twins. An extensive pedigree was presented. The features that they pointed out included stress fractures of the tibia and femur, spondylolysis of L4 and L5, healed hangman's fracture of C2 (fracture of the pedicle), and a double row of teeth resulting from persistence of deciduous teeth. Also see craniostenosis (123100) and acroosteolysis with osteoporosis and changes in skull and mandible (102500).

Biochemical Features

Soliman et al. (1996) reported defective growth hormone secretion in response to provocation and low insulin-like growth factor-1 (147440) concentration in 5 out of 6 patients with pycnodysostosis. Physiologic replacement with growth hormone increased IGF1 concentration and improved linear growth in these children. The IGF1 generation time ruled out significant resistance to growth hormone. Growth hormone treatment was used in 2 children. The normal TSH, free thyroxine, and 8-hour cortisol concentrations ruled out any significant abnormality of the hypothalamic-pituitary, thyroid, and adrenal axes in these patients. The normal sexual development, fertility, and serum gonadotropin and testosterone concentrations in the 2 affected adult males were evidence against any abnormality of the hypothalamic-pituitary-gonadal axis.

Inheritance

Sedano et al. (1968) found parental consanguinity in about 30% of reported cases, indicating autosomal recessive inheritance.

Mapping

In the inbred Arab kindred reported by Edelson et al. (1992), Gelb et al. (1995) demonstrated by linkage analysis that the pycnodysostosis locus is located on 1q21. Polymeropoulos et al. (1995) found the same linkage in an inbred Mexican kindred. In both cases, homozygosity mapping was used initially. In the Arab kindred, Gelb et al. (1995) found that 13 of 16 affected individuals were homozygous for the D1S305 allele, which had previously been assigned to the pericentromeric region of chromosome 1. Using markers flanking the centromere of chromosome 1, they localized the PKND locus to a region of about 4 cM between D1S442 and D1S305. D1S442 had previously been assigned to 1q21 by fluorescence in situ hybridization. They pointed to the interleukin-6 receptor (IL6R; 147880) and myeloid cell leukemia-1 (MCL1; 159552) as plausible candidate genes. IL6R induces the formation of osteoclasts and is highly expressed in osteoclasts from bone of patients with Paget disease (see 167250) and osteoarthritis. For the initial screening in the Mexican kindred, Polymeropoulos et al. (1995) used a pooling strategy in which DNA from affected individuals was pooled and genotyped. Using a total of 363 genetic markers, they compared the allelic ladders produced with those from a pool of unaffected heterozygous parents genotyped with the same markers. They noticed a reduction in the complexity of the number of alleles observed in the affected pool for markers D1S1595 and D1S534. Genetic linkage analysis was then performed in the family, confirming linkage for marker D1S1595 with a lod score of 4.11 at theta = 0.05, and for D1S534 with a lod score of 2.05 at theta = 0.08. Haplotype analysis in affected individuals placed the PKND gene in a 6-cM interval between markers D1S514 and D1S305. Polymeropoulos et al. (1995) suggested that macrophage colony-stimulating factor (CSF1; 120420) might be a candidate gene, but CSF1 maps to the proximal short arm of chromosome 1; indeed, they could demonstrate no mutations in the gene by SSCP analysis. They also suggested that one of the calcium-binding protein genes that are clustered at 1q21 may be the site of the mutation.

Molecular Genetics

Because cathepsin K (601105), a cysteine protease gene that is highly expressed in osteoclasts, maps to the same region as pycnodysostosis, Gelb et al. (1996) searched for mutations in the cathepsin K gene. They identified nonsense, missense, and stop codon mutations in patients (601105.0001-601105.0004). Transient expression of cDNA containing the stop codon mutation resulted in mRNA, but no immunologically detectable protein was present. The findings suggested to the authors that cathepsin K is a major protease in bone resorption, providing a possible rationale for the treatment of disorders such as osteoporosis and certain forms of arthritis. Bone resorption, a process mediated by osteoclasts, is characterized by the solubilization of inorganic mineral and subsequent proteolytic degradation of organic matrix, primarily type I collagen. In pycnodysostosis, osteoclast numbers are normal as are their ruffled borders and clear zones, but the region of demineralized bone surrounding individual osteoclasts is increased. Ultrastructural examination of these osteoclasts revealed large, abnormal cytoplasmic vacuoles containing bone collagen fibrils. These findings suggested that pycnodysostosis osteoclasts function normally in demineralizing bone, but do not adequately degrade the organic matrix. Cathepsin S (116845), which also maps to 1q, was ruled out as a candidate gene for pycnodysostosis (Gelb et al., 1996).

Gelb et al. (1998) identified paternal uniparental disomy for chromosome 1 as the molecular basis of pycnodysostosis in a patient who had normal birth weight and height, had normal psychomotor development at age 7 years, and had only the usual features of pycnodysostosis. The patient represented the first case of paternal uniparental disomy of chromosome 1 and provided conclusive evidence that paternally derived genes on human chromosome 1 are not imprinted. The missense mutation in this case, inherited only from the father, was ala277 to val (601105.0004).

In affected individuals from 8 unrelated families with pycnodysostosis, Hou et al. (1999) identified homozygosity for 8 different mutations in the cathepsin K gene.

Animal Model

For information on animal models of pycnodysostosis, see 601105.

History

Maroteaux and Lamy (1965) suggested that Toulouse-Lautrec (1864-1901) had pycnodysostosis. Features consistent with the disorder were dwarfing, parental consanguinity, bone fracture with relatively mild trauma, and probably large fontanels, prompting him to wear a hat much of the time. Maroteaux (1993) gave a charming and well-illustrated account of the case of Toulouse-Lautrec. Frey (1994) and Frey (1995) presented evidence suggesting that pycnodysostosis was not the diagnosis in the case of Toulouse-Lautrec. See the rebuttal by Maroteaux (1995) and the response to the rebuttal by Frey (1995).