Paget disease of bone: mapping of two loci at 5q35-qter and 5q31

Abstract

Paget disease of bone is characterized by focal increases of the bone-remodeling process. It is the second most common metabolic bone disease after osteoporosis. Genetic factors play a major role in the etiology of Paget disease of bone, and two loci have been mapped for the disorder: PDB1 and PDB2. The gene(s) causing the typical form of the disorder remains to be characterized. To decipher the molecular basis of Paget disease of bone, we performed genetic linkage analysis in 24 large French Canadian families (479 individuals) in which the disorder was segregating as an autosomal dominant trait. After exclusion of PDB2, a genomewide scan was performed on the three most informative family nuclei. LOD scores >1.0 were observed at seven locations. The 24 families were then used to detect strong evidence for linkage to chromosome 5q35-qter. Under heterogeneity, a maximum LOD score of 8.58 was obtained at D5S2073, at straight theta= .1. The same characteristic haplotype was carried by all patients in eight families, suggesting a founder effect. A recombination event in a key family confined the disease region within a 6-cM interval between D5S469 and the telomere. The 16 other families, with very low conditional probability of linkage to 5q35-qter, were further used, to map a second locus at 5q31. Under heterogeneity, a maximum LOD score of 3.70 was detected at D5S500 with straight theta=.00. Recombination events refined the 5q31 region within 12.2 cM, between D5S642 and D5S1972. These observations demonstrate the mapping of two novel loci for Paget disease of bone and provide further evidence for genetic heterogeneity of this highly prevalent disorder. It is proposed that the 5q35-qter and 5q31 loci be named "PDB3" and "PDB4," respectively.

Figure 1

Pedigree structure and haplotypes of the extended pagetic families DD and TH, showing linkage to chromosome 5q35-qter. Individuals selected for the genomewide scan are indicated by an asterisk (*). Circles and squares represent females and males, respectively; blackened symbols denote individuals affected by Paget disease of bone; empty white symbols denote individuals whose status is unknown or for whom the data were insufficient for phenotypic classification; an “N” within a white symbol denotes that the individual is normal; a question mark (?) within a white symbol denotes that the diagnosis is ambiguous; and a diagonal line through a symbol denotes that the individual is deceased. The number in parentheses below each symbol denotes the individual's age at the time of the study. “D5S” numbers for each marker genotyped are shown on the left of the third generation of each pedigree; refinement of the disease locus is shown on the left of these “D5S” numbers. The disease-associated haplotypes are denoted by the vertical black bars; within the haplotypes, maternal crossover is denoted by a slash (/), paternal crossover by a back slash (\).

Figure 2

Pedigree structure and haplotypes spanning the 5q31 region in extended family LF. Blackened symbols with a white lower-left corner denote individuals diagnosed with concomitant Paget disease of bone and benign osteopetrosis; tiled symbols denote individuals diagnosed with osteopetrosis only; arrows (→) indicate key recombinant events; all other symbols are as described in figure 1. A spontaneous mutation was detected in marker AFM263wg5 (D5S436) in individual LF-045 and is enclosed within parentheses.

Figure 3

Summary of genomewide scan for susceptibility loci for Paget disease of bone, in three French Canadian family nuclei. Two-point LOD scores are shown, and markers with LOD scores >1.0 are indicated. Gray vertical lines denote chromosome boundaries.

Figure 4

Multipoint parametric LOD scores at 5q35-qter in eight families showing linkage. The phenotypic model was modified according to penetrance values estimated on the basis of 101 individuals who carry the disease haplotype. Allele frequencies were estimated on the basis of all genotyped pedigrees, according to Boehnke's (1991) method. Genetic distances are from genetic maps reported elsewhere (Dib et al. 1996)—except for UT6507, which was mapped relative to other markers by use of Human Genome Project (National Center for Biotechnology Information) draft sequences.

Figure 5

Pedigree structure and haplotypes spanning the 5q31 region in extended family FT. Symbols are as described in figures 1 and 2.