Early-onset Parkinson's disease (PD) has been associated with different mutations in the Parkin gene (PARK2). To study distribution and type of Parkin mutations, we carried out a comprehensive literature review that demonstrated two prominent types of mutations among 379 unrelated mutation carriers: exon rearrangements involving exon 3, 4, or both, and alterations in exons 2 and 7, suggesting mutational hot spots or founders. To elucidate the origin of 14 recurrent Parkin mutations in our samples, we carried out a detailed haplotype analysis at the PARK2 locus. Thirty-eight mutation-positive individuals, available family members, and 62 mutation-negative individuals were genotyped. We determined allele frequencies and linkage disequilibrium (LD) to evaluate the significance of shared haplotypes. We observed no LD between markers at PARK2. Our data support a common founder for the most frequent Parkin point mutation (924C>T; exon 7) and indicate a mutational hot spot as cause of a common small deletion (255/256delA; exon 2). Furthermore, the most frequent Parkin exon deletion (Ex4del) arose independently in 2 of our subjects. However, it also occurred as the result of a founder mutation in 2 cases that shared identical deletion break points. This study provides evidence for both mutational hot spots and founder mutations as a source of recurrent mutations in Parkin, regardless of the mutation type.

We report the clinical, genetic, and neuropathological findings of a seven generation-spanning pedigree with 196 individuals, 25 of whom had levodopa-responsive parkinsonism. Genetic analyses indicated Parkin mutations in 77 subjects. Among the 25 patients, 5 carried compound heterozygous mutations and met criteria for definite Parkinson's disease (PD) according to UK PD Society Brain Bank guidelines; 8 subjects carried only a heterozygous Parkin mutation. The mutational status of five deceased patients was unknown, and seven PD patients had no Parkin mutation. Survival analyses showed a significant difference in the age-at-onset distribution between patients with compound heterozygous mutations and the groups of heterozygous carriers and subjects without detectable Parkin mutations. Autopsy of a 73-year-old patient, who carried two mutant Parkin alleles (delExon7 + del1072T), showed PD-type cell loss, reactive gliosis, and alpha-synuclein-positive Lewy bodies in the substantia nigra and locus ceruleus. Surviving neurons were reactive with antibodies to the N terminus of Parkin but not the In-Between-RING ("IBR") domain, which had been deleted by both mutations. This large Parkin pedigree represents a unique opportunity to prospectively study the role of heterozygous Parkin mutations as a PD risk factor, to identify additional contributors to the expression of late-onset PD in heterozygous carriers, and to reexamine the role of Parkin in inclusion formation.

Early-onset parkinsonism (EOP) may be associated with different mutations in the parkin gene, including exon deletions and duplications. To test for gene dosage alterations, we developed a new method of quantitative duplex PCR using the fluorescence resonance energy transfer technique on the LightCycler (Roche Diagnostics). In 21 patients with EOP, three mutations (a single base pair substitution in exon 3 and small deletions in exon 9) were detected by conventional mutational screening (single-strand conformation polymorphism and sequence analysis), while alterations of gene dosage were found in seven patients. We identified heterozygous and compound heterozygous deletions of exons 2, 3, 5 and 7. The latter was also found in the homozygous state. In addition, two heterozygous duplications of exon 4 were observed. Remarkably, two patients carried more than two parkin mutations. This is the first study systematically screening all 12 exons of parkin by real-time, kinetic quantification and clearly shows that mutational analysis of the parkin gene should include gene dosage studies. Furthermore, our method of quantitative PCR is easily applicable to any other gene to be screened for deletions or duplications of whole exons.

Recent studies have suggested an association between restless legs syndrome (RLS) and Parkinson's disease (PD). We present a large multigenerational family and a smaller family with RLS. A Parkin mutation was found in 10 of 20 patients from both families with idiopathic RLS but was not considered causative. The clinical phenotype did not differ between RLS patients with and without a Parkin mutation. Inheritance of RLS was consistent with autosomal dominant transmission, and linkage analysis excluded all three known loci for RLS.

No abstract