Parkin Deletions in a Family with Adult-Onset, Tremor-Dominant Parkinsonism:

Expanding the PhenotypeChristine Klein, MD,*† Peter P. Pramstaller, MD,‡ Bernhard Kis, MD,†‡ Curtis C. Page, BA,*Martin Kann, BS,* Joanne Leung, BS,* Heather Woodward, BS,* Claudio C. Castellan, MD,§

Monika Scherer, MD,‡ Peter Vieregge, MD,† Xandra O. Breakefield, PhD,* Patricia L. Kramer, PhD,¶and Laurie J. Ozelius, PhD*\

A gene for autosomal recessive parkinsonism, PARK2 (parkin), has recently been identified on chromosome 6q andshown to be mutated in Japanese and European families, mostly with early-onset parkinsonism. Here we present a largepedigree from South Tyrol (a region of northern Italy) with adult-onset, clinically typical tremor-dominant parkinsonismof apparently autosomal dominant inheritance. Haplotype analysis excluded linkage to the chromosome 2p, 4p, and 4qregions that harbor genes associated with autosomal dominant parkinsonism, but implicated the parkin locus on chro-mosome 6q. Compound heterozygous deletions in the parkin gene (one large and one truncating) were identified in 4affected male siblings. The patients were clinically indistinguishable from most patients with idiopathic Parkinson’sdisease. None of them displayed any of the clinical hallmarks described in patients with previously reported parkinmutations, including diurnal fluctuations, benefit from sleep, foot dystonia, hyperreflexia, and early susceptibility tolevodopa-induced dyskinesias. Two affected female individuals carried one (truncating) of the two deletions in a het-erozygous state with an apparently normal allele. We conclude that the phenotypic spectrum associated with mutationsin the parkin gene is broader than previously reported, suggesting that this gene may be important in the etiology of themore frequent late-onset typical Parkinson’s disease.

Klein C, Pramstaller PP, Kis B, Page CC, Kann M, Leung J, Woodward H, Castellan CC, Scherer M, Vieregge P,Breakefield XO, Kramer PL, Ozelius LJ. Parkin deletions in a family with adult-onset tremor-dominant

parkinsonism: expanding the phenotype. Ann Neurol 2000;48:65–71

A genetic contribution to the etiology of at least somecases of parkinsonism is now well established1 and ap-pears to be particularly important when the disease be-gins at or before age 50 years.2 In families with auto-somal dominantly inherited parkinsonism, one geneand two chromosomal regions have recently been im-plicated.3–7 Mutations in the a-synuclein gene onchromosome 4q (PARK1) have been found in familieswith parkinsonism of Italian/Greek3 and German ori-gin,4 and two linked loci for autosomal dominant par-kinsonism have been mapped to chromosome 2p(PARK3)5 and to chromosome 4p.6 Furthermore, amissense mutation in the ubiquitin carboxy-terminalhydrolase L1 (UCH-L1) gene on chromosome 4p ap-pears to be associated with parkinsonism in 1 family.7

Mutations in another gene for parkinsonism on chro-

mosome 6q, parkin (PARK2), have been detected infamilies with autosomal recessive juvenile parkinson-ism (ARJP).8 Finally, another form of familial parkin-sonism, grouped under the term frontotemporal de-mentia and parkinsonism, is linked to chromosome 17(FTDP-17), with mutations identified in the taugene.9,10 On closer inspection, many of the publishedparkinsonian kindreds show unusual features, both onclinical and pathological grounds, and are thereforenot clinically typical Parkinson’s disease. Herein, wedescribe the clinical and genetic characteristics of alarge pedigree from South Tyrol (a region of northernItaly) with adult-onset, clinically typical parkinsonismof pseudo-dominant inheritance, including 4 siblingswith compound heterozygous deletions in the parkingene.

From the *Molecular Neurogenetics Unit, Massachusetts GeneralHospital, and Departments of Neurology and Genetics and Neuro-science Program, Harvard Medical School, Boston, MA; †Depart-ment of Neurology, Medical University of Lubeck, Lubeck, Ger-many; Departments of ‡Neurology and §Genetics, RegionalGeneral Hospital, Bolzano-Bozen, Italy; ¶Department of Neurology,Oregon Health Science University, Portland, OR; and \MolecularGenetics, Albert Einstein College of Medicine, Bronx, NY.

Received Jan 7, 2000, and in revised form Feb 28. Accepted forpublication Feb 29, 2000.

Address correspondence to Dr Ozelius, Molecular Genetics,AECOM, 1300 Morris Park Avenue, Bronx, NY 10461.

Copyright © 2000 by the American Neurological Association 65

Patients and MethodsPatientsAfter giving informed consent, 31 members of family LA(Fig 1) underwent a standardized neurological examinationby a movement disorders specialist (P.P.P.). Clinical infor-mation on 18 deceased patients or individuals was takenfrom hospital records or by history. Diagnosis of parkinson-ism was established according to the UK Parkinson’s DiseaseSociety Brain Bank criteria.11 Patients were defined as clini-cally definite if they had at least two of three cardinal motorsigns (resting tremor, rigidity, bradykinesia) and a consistentresponse to levodopa, or as possibly affected if only one ofthe major signs of the classic triad was present. The Hoehn-Yahr scale was used to rate on-phase parkinsonian symp-toms.12 Secondary parkinsonism and other Parkinson plussyndromes were excluded by history, absence of additionalneurological signs, normal laboratory tests (including full

blood count; urine analysis; serum electrolytes; liver, renal, andthyroid function tests; vitamin B12 and folate levels; VDRL;copper, ceruloplasmin, and acanthocytes), and neuroimaging(computed tomography [CT] or magnetic resonance imaging[MRI] brain scan). Neither positron emission tomographic(PET) scans or autopsy studies have been performed yet.

DNA AnalysisTesting for the Ala53Thr and the Ala30Pro mutations in thea-synuclein gene and the Ile93Met mutation in the UCH-L1gene was performed using published primers for polymerasechain reaction (PCR) amplification followed by a restrictiondigest, as described3,4 or by direct DNA sequencing of thePCR product.7 Genotyping was carried out according tostandard protocols (Research Genetics; http://www.resgen.com) with the polymorphic DNA markers D2S2320,D2S380, D2S327, D2S2114 (2p); D4S2366, D4S2397,

Fig 1. Pedigree of Family LA with genotypes at polymorphic DNA markers D6S1003–21.9 cM–D6S1550–0.0 cM–D6S305–0.0cM–D6S411–0.0 cM–D6S1579–7.0 cM–D6S1277 (marker distances according to Marshfield [http://www.marshmed.org/genetics]). Haplotypes were assigned with the assistance of SIMWALK2. Markers in parentheses were inferred on the basis of off-spring data; markers surrounded by question marks could not be phased with certainty. Disease chromosome 1 from the paternalbranch is indicated by a 1 below the haplotype; disease chromosome 2 from the maternal side is indicated by a 2. The position ofthe deletion on disease chromosome 1 is indicated by a x. Sequence analysis confirmed that individual 550 carried the single basepair deletion in disease chromosome 2. Cases 405 and 511 potentially carry disease chromosome 1 based on their haplotypes, but noRNA was available to confirm this at the molecular level. Obligate recombination events at the markers D6S1003 and D6S1277place the putatively linked region in a 28.9-cM area of chromosome 6q. The hemizygous marker D6S305 is located in intron 7 ofparkin; D6S411 is at the same genetic map position as D6S305.

66 Annals of Neurology Vol 48 No 1 July 2000

D4S1627, D4S2367 (4p); D4S2361, D4S2371, D4S2623(4q); and D6S1003, D6S1550, D6S305, D6S411,D6S1579, D6S1599, D6S1277 (6q).

Furthermore, five additional highly polymorphic DNAmarkers (average heterozygosity, 0.83) were run to verify ap-propriate inheritance (D9S2161, ASS, D9S63, D19S246,and D19S178). PCR products were analyzed on a LI-COR(Lincoln, NE) automated sequencer or detected by silverstaining or autoradiography. Single-strand conformationalpolymorphism (SSCP) analysis was performed on all 12 ex-ons of parkin, as well as flanking intron sequences, usingpublished primers.8 An additional pair of exon 3 primers wasused for SSCP analysis (Exon3Fnew 59-aga aac gcg gcg ggaggc tg-39; Exon3Rnew 59-act gag ctg ctg agg tcc ac-39).

Linkage AnalysisWe conducted linkage analysis using FASTLINK13 andVITESSE14 with the markers listed above on chromosomes2p, 4p, and 4q. For these markers, we assumed autosomaldominant inheritance. Because of the late onset and complexnature of this disease, only definitely affected individualswere considered “affected,” and all other family memberswere considered “unknown” with regard to disease status.Haplotypes for the chromosome 6q markers were con-structed with assistance from SIMWALK2.15

Sequence AnalysisDideoxy cycle sequencing was performed with the Perkin-Elmer AmpliCycle Sequencing kit (Perkin-Elmer, FosterCity, CA) using published primers8 labeled with a33P-dATP(2,000 Ci/mmol; NEN, Boston, MA). Direct cycle sequenc-ing (step 1: 95°C 2 minutes; step 2: 95°C 1 minute; step 3:60°C 1 minute; step 4: 72°C 1 minute; cycle steps 2–4325; step 5: 4°C 5 minutes) was performed after enzymaticclean-up with exonuclease I and shrimp alkaline phosphatase(USB, Cleveland, OH) for 15 minutes at 37°C and 15 min-utes at 85°C.

RT-PCR AnalysisLymphoblasts were separated from a fresh whole blood sam-ple from the index patient (Patient 509) by centrifugationthrough Histopaque (Sigma, St Louis, MO) at 400 3 g for30 minutes. The isolated lymphoblasts were washed once incold phosphate-buffered saline, and total cell RNA was pu-rified using Trizol reagent (Gibco, Gaithersburg, MD) ac-cording to the manufacturer’s specifications. First-strandcDNA was synthesized from this RNA. The following prim-ers were used to amplify parkin cDNA spanning exons 1 to9: forward primer in exon 1, GATTACCCAGGAGAC-CGCTG; reverse primer in exon 9, GGTACGCTTCTTTA-CATTCC (Fig 2c).

ResultsClinical DataFamily LA originated from a small village in the West-ern Alps of South Tyrol in Italy. There was no knownparental consanguinity in ancestors traced back until1657 and no circumstantial evidence that the diseasemight have been environmentally acquired.

Eight patients (5 men, 3 women) were definitely af-fected; 3 women were possibly affected (see Fig 1).Mean age at onset of the definite cases was 57.3 615.0 years (range, 31–76 years), and that of the possi-ble cases was 60.0 6 5.0 years (range, 55–65 years).The mode of transmission was consistent with autoso-mal dominant inheritance with high penetrance, affect-ing individuals in two successive generations with 50%of offspring of an affected individual being definitelyaffected. Tremor was the presenting symptom in all ofthe definite cases. Similarly, tremor dominated as thefirst symptom in the possibly affected cases. Mean dis-ease duration in the 8 definitely affected patients was16.3 6 12.8 years. All treated patients responded wellto L-dopa therapy, with 4 patients experiencinglevodopa-related dyskinesias after a mean interval of7.5 6 2.3 years. Clinical features are summarized inthe Table. CT (Cases 502, 505, 509, 528, 534, 550)and MRI brain scan (Case 503) were normal. Depres-sion was present in 1 patient (Case 528) who was noton antiparkinsonian medication.

Linkage Analysis of Loci for Parkinsonism andMutations in the parkin GeneNone of the affected family members carried theAla53Thr or the Ala30Pro mutation in the a-synucleingene, nor did they have the Ile93Met mutation in theUCH-L1 gene. Furthermore, multipoint linkage analy-sis excluded the candidate regions on chromosomes 2p,4p, and 4q (LOD scores were below 22.0 across theentire region; data not shown). However, haplotypeanalysis in the region of the parkin gene on chromo-some 6q did show evidence of linkage. Further, 4 af-fected siblings (Cases 502, 503, 505, 509) and some oftheir offspring (Cases 604, 609, 616, 640) appeared tobe hemizygous for alleles at markers D6S305 (locatedin intron 7 of parkin) and D6S411. This was indicativeof a deletion in the region containing parkin for one ofthe alleles on chromosome 6q coming from the de-ceased father, Case 401 (disease chromosome 1) (seeFig 1). Nonpaternity was not an issue as correct trans-mission of parental alleles was observed at all othermarkers tested on chromosomes 2p, 4p, 4q (data notshown) and 6q.

In addition, all 4 of these affected siblings with theapparent deletion in parkin also shared a common hap-lotype for the marker alleles at D6S1550 to D6S1579(5-6-4-1), defined as disease chromosome 2 (see Fig 1),suggesting the possibility of a second mutation in thisparkin allele. SSCP analysis for the 12 exons of parkin,including flanking intron sequences, revealed that the 4affected siblings, as well as all other living carriers ofthis haplotype (see Fig 1), showed a band shift in exon9 (see Fig 2a). Sequencing revealed that this corre-sponded to a single base pair deletion at position 1072,inherited from the deceased mother (Case 411) (see Fig

Klein et al: Clinical and Genetic Study in Parkinson’s Disease 67

2b). This deletion causes a frameshift that results in theaddition of 111 novel amino acids to the parkin pro-tein (starting at amino acid 324) and then proteintruncation in exon 12 of parkin (see Fig 2c). The fouraffected male siblings (Cases 502, 503, 505, 509) areall compound heterozygotes for these two parkin dele-tions. The possibly affected sibling (Case 501) carriedonly the parkin deletion from her mother (Case 411),as did an affected maternal cousin (Case 550), thusconfirming that the disease chromosome 2 comes fromthe maternal side. Direct sequencing of all exons in theother 2 maternal affected cousins (Cases 528 and 534)identified no parkin mutations.

One endpoint of the large deletion on disease chro-mosome 1 was defined by the presence of “double se-quence” in affected individuals for exon 9 (see Fig 2b),indicating that exon 9 was intact. To define the other

end of this large deletion, we sequenced PCR productsgenerated by RT-PCR of RNA from the index patientwith compound heterozygous deletions in parkin, usingexonic primers located in exons 1 and 9 (see Fig 2c).This showed that only the cDNA derived from diseasechromosome 2 could be amplified (see Fig 2b; absenceof “double sequence”), confirming that the other end-point of the large deletion extends upstream of theprimer in exon 1. Therefore, the large deletion on dis-ease chromosome 1 encompasses exon 1 through in-tron 7 (containing marker D6S305) but does not ex-ceed intron 8 (see Fig 2c).

DiscussionAnalysis of Family LA reveals an interesting combina-tion of clinical and molecular findings, including com-pound heterozygous mutations in parkin in 4 of 8 af-

Fig 2. (a) Mutation found in exon 9 ofparkin in affected individuals and carriersof disease chromosome 2. Single-strand con-formational polymorphism of exon 9 ofparkin showing a band shift in the affectedindividual compared with a normal con-trol. (b) Sequence of exon 9 of parkin in anormal control (arrow indicates the C atposition 1071 that is deleted in affectedindividuals), an affected individual (arrowshows beginning of “double sequence”), andof the RT-PCR product of exon 9 of anaffected individual, showing the sequence ofdisease chromosome 2 only (arrow points tothe deleted C). The reverse strand sequenceis shown. (c) Schematic representation ofthe parkin cDNA and disease chromosomes1 and 2. The position of exonic primersused for RT-PCR experiments is indicated.

68 Annals of Neurology Vol 48 No 1 July 2000

fected individuals (Cases 502, 503, 505, 509) with aphenotype resembling idiopathic Parkinson’s diseasewith apparently autosomal dominant inheritance. Twopatients (Cases 411, 550) and a possibly affected case(Case 501) were heterozygous with a parkin mutationon one allele and an apparently normal other allele,and 2 other affected individuals (Cases 528, 534) didnot carry any detectable parkin mutations by sequenceanalysis. This confirms and further extends recent find-ings16 that parkin mutations can occur in patients witha phenotype that is indistinguishable from that of id-iopathic Parkinson’s disease and that these mutationscan occur as compound heterozygous mutations.

Family LA is characterized by the development ofclinically typical, tremor-dominant, levodopa-responsiveparkinsonism with slow progression. Average age at dis-ease onset of all 8 affected individuals was in the mid-fifties, with mean onset of 48.0 6 13.5 years in the 4affected siblings with compound heterozygous deletionsin parkin. This is only slightly lower than the mean ageof onset in sporadic idiopathic Parkinson’s disease17 butdiffers from the overall young onset age of early-onsetparkinsonism, which is considered to be below 40 years(,20 years in AR-JP).8,18,19 To date, only 4 patientswith disease onset in their fifties, from 4 families display-ing autosomal recessive parkinsonism, have been re-ported to have mutations in the parkin gene.16,20 Age ofonset in 1 of the affected siblings from family LA (Case502) was 64 years, which is the latest age of onset re-ported in a case with proven parkin mutations thus far(Table). None of the patients in family LA, regardless ofthe genetic status, displayed any of the hallmarks de-scribed in families with previously reported parkin

mutations, including diurnal fluctuations, sleep benefit,foot dystonia, hyperreflexia, and early susceptibility tolevodopa-induced dyskinesias.8,16,18,21 Similarly, amongthe reported parkinsonian kindreds with autosomaldominant inheritance,22–27 only very few show a pheno-type entirely typical for idiopathic Parkinson’s disease.5,6

To date, only a single autopsy report on a provenparkin-related case has been reported and no Lewybodies were found.28 However, literature review of thepostmortem analyses of AR-JP patients reveals thatclinically typical cases have been published bothwith29,30 and without Lewy bodies.31–33 Unfortu-nately, in the current study, postmortem analysis tocheck for the typical pathological hallmarks of Parkin-son’s disease, most importantly the presence of Lewybodies, was not performed in the deceased individual(Case 505). Given the clinical phenotype of FamilyLA, one would intuitively expect Lewy bodies to bepresent. However, this and the general role of Lewybodies in parkin-related parkinsonism remain to beclarified.

AR-JP is inherited in an autosomal recessive fashionand frequently occurs in the offspring of consanguine-ous marriages.8 There was no evidence of consanguin-ity in our family, and before the genetic analysis themode of inheritance appeared dominant. However,parkinsonism in the 4 affected siblings (Cases 502,503, 505, 509) with compound heterozygous deletionsin PARK2 is clearly consistent with an autosomal re-cessive mode of inheritance. Three of the 5 remainingindividuals with definite (Cases 411, 550) or possibleparkinsonism (Case 501) carried the truncating dele-tion (disease chromosome 2) in a heterozygous state

Table. Clinical Features of Affected Family Members

Case No.Sex/Age(yr) Life Spana

Age atOnset(yr)

DiseaseDuration(yr)

FirstSign

Cardinal PDSigns

PD TherapyHoehn-YahrScale

Source ofInformationRT B R PI Response

TreatmentComplication

Definitely affected502 M/71 b. 1927 64 7 Tremor 1 1 1 Ldb — II PE/G/V503 M/68 b. 1930 49 19 Tremor 1 1 1 1 Ldb MF, Dys IV PE/G/V505 M 1933–1998 48 16 Tremor 1 1 1 1 Ld 1 DAb Dys III PE/G/V509 M/63 b. 1935 31 32 Tremor 1 1 1 1 Ldb Dys III PE/G/V528 F/71 b. 1927 55 16 Tremor 1 1 Dab — II PE/G/V534 M/78 b. 1920 75 3 Tremor 1 1 1 1 Ld 1 DA 1 Seb — II PE/G/V550 F/77 b. 1921 76 1 Tremor 1 1 1 Ldb — II PE/G/V411 F 1896–1992 60 36 Tremor 1 1 1 1 Ldb Not reported III MR

Possibly affected523 F/64 b. 1934 61 3 Tremor 1 No therapy — PE/G/V501 F/72 b. 1926 65 7 Tremor 1 1 No therapy — PE/G/V307 F 1855–1950 55 35 Tremor 1 No therapy — H

aCases showing only year of birth (b.) are still alive.bResponsive to therapy.

RT 5 resting tremor; B 5 bradykinesia; R 5 rigidity; PI 5 postural instability; Ld 5 L-dopa; DA 5 dopamine agonist; Se 5 selegiline; Dys 5dyskinesia; MF 5 motor fluctuations; PE 5 personally examined; G 5 genotyped; V 5 videotape; MR 5 medical records; H 5 by history.

Klein et al: Clinical and Genetic Study in Parkinson’s Disease 69

with an apparently normal allele. Interestingly, all 3 in-dividuals were women and had particularly late onsetof the disease, 60 and 76 years in the definite cases and65 years in the possible case. One could speculate thatsome parkin mutations might act in a dominant nega-tive fashion and increase susceptibility for parkinson-ism, possibly in association with additional risk factors(ie, environmental insult or genetic/endogenous fac-tors). However, no risk factors were indicated in these3 individuals, and the study by Abbas and colleagues16

did not support a dominant negative effect of truncat-ing mutations among parents of patients. However,they did report on 3 families in which they could onlyidentify a single mutant allele.16 Follow-up studies onthe 44 descendants of Cases 501 to 507 and 509 (notall of whom are indicated in Fig 1), who have notreached the age of disease onset yet (age at last exam-ination was between 9 and 44 years in the 26 descen-dants that have been examined) but most of whomcarry either disease chromosome 1 or 2 may serve tofurther elucidate whether some parkin mutations canact in a dominant negative fashion.

Another explanation for the parkinsonism seen inthe maternal branch is the possibility that a second par-kin mutation is located in a region of the gene that wasnot sequenced (eg, the promoter region or an intron)or may be present as an exon deletion or duplicationthat might be missed by PCR exon amplification andsequencing. Finally, affected status of these individualscould also be due to involvement of a different genefor parkinsonism on the maternal side. The origin ofparkinsonism in this branch of the family requires fur-ther study, but is reminiscent of a family with parkin-sonism displaying genetic heterogeneity describedpreviously.34

Point mutations in the parkin gene have been de-scribed to be more common than deletions in AR-JP.16

However, Family LA shows a novel single basepair de-letion on one allele (2) and a larger deletion, compris-ing several exons, on the other (1).

At this point, no clear-cut genotype-phenotype cor-relations have been observed16,35 between mutations inparkin and disease onset, or symptoms. Interestingly,the four affected siblings in the present family whocarry identical compound heterozygous deletions inparkin and lived in the same environment throughouttheir lives, display marked intrafamilial phenotypicvariability (see Table). On the other hand, these sameindividuals (Cases 502, 503, 505, 509) with the com-pound heterozygous deletions phenotypically resembledCases 528 and 534 without identified parkin muta-tions. Given the relatively small number of reportedparkin mutations, the almost exclusive screening of fa-milial cases with autosomal recessive inheritance andjuvenile onset, and the lack of information on modifyingfactors such as exposure to environmental toxicants, ad-

ditional studies are warranted to further clarify the phe-notypic spectrum associated with parkin mutations.

Although our data give rise to several open ques-tions, including the origin of parkinsonism in the ma-ternal branch of the family and the issue of presence orabsence of Lewy bodies, our findings provide an im-portant piece of information to neurologists and genet-icists interested in Parkinson’s disease. Taken together,compound heterozygous deletions in parkin can be as-sociated with a phenotype consistent with classic idio-pathic Parkinson’s disease and may occur in familieswith apparently dominant inheritance. Parkinsonismdue to the exact same parkin mutation may result indifferent ages of onset, while genetically different indi-viduals may be phenotypically indistinguishable.

This work was supported by the Neuroepidemiology Project SouthTyrol/NEPT (P.P.P., M.S., C.C., B.K.), the Dystonia Medical Re-search Foundation (L.O., P.L.K., X.O.B.), NINDS grant NS38372(X.O.B.), the Deutsche Forschungsgemeinschaft (C.K.), and theBlowitz-Ridgeway Foundation (X.O.B.).

We are grateful to the patients and family members who partici-pated in this study. We thank Drs L. Golbe and N. Quinn forcritical evaluation of the videotapes.

References1. Duvoisin RC. Role of genetics in the cause of Parkinson’s dis-

ease. Mov Disord 1998;13(Suppl 1):7–122. Tanner CM, Ottmann R, Goldman SM, et al. Parkinson’s dis-

ease in twins: an etiologic study. JAMA 1999;281:341–3463. Polymeropoulos MH, Lavedan C, Leroy E, et al. Mutation in

the a-synuclein gene identified in families with Parkinson’s dis-ease. Science 1997;276:2045–2047

4. Kruger R, Kuhn W, Muller T, Woitalla D, et al. Ala30Promutation in the gene encoding a-synuclein in Parkinson’s dis-ease. Nat Genet 1998;18:106–108

5. Gasser T, Muller-Myhsok B, Wszolek ZK, et al. A susceptibil-ity locus for Parkinson’s disease maps to chromosome 2p13.Nat Genet 1998;18:262–265

6. Farrer M, Gwinn-Hardy K, Muenter M, et al. A chromosome4p haplotype segregating with Parkinson’s disease and posturaltremor. Hum Mol Genet 1999;8:81–85

7. Leroy E, Boyer R, Auburger G, et al. The ubiquitin pathway inParkinson’s disease. Nature 1998;395:451–452

8. Kitada T, Asakawa S, Hattori N, et al. Mutations in the parkingene cause autosomal recessive juvenile parkinsonism. Nature1998;392:605–608

9. Hutton M, Lendon CL, Rizzu P, et al. Association of missenseand 59-splice-site mutations in tau with the inherited dementiaFTDP-17. Nature 1998;393:702–705

10. Spillantini MG, Murrell JR, Goedert M, et al. Mutation in thetau gene in familial multiple system tauopathy with preseniledementia. Proc Natl Acad Sci USA 1998;95:7737–7741

11. Gibb WRG, Lees AJ. The relevance of the Lewy body to thepathogenesis of idiopathic Parkinson’s disease. J Neurol Neuro-surg Psychiatry 1988;51:745–752

12. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, andmortality. Neurology 1967;17:427–442

13. Schaffer A, Gupta S, Shriram K, Cottingham R Jr. Avoidingrecomputation in linkage analysis. Hum Hered 1994;44:225–237

70 Annals of Neurology Vol 48 No 1 July 2000

14. O’Connell JR, Weeks DE. The Vitesse algorithm for rapid, ex-act multilocus linkage analysis via genotype set-recording andfuzzy inheritance. Nat Genet 1995;11:402–408

15. Sobel E, Lange K. Descent graphs in pedigree analysis: appli-cations to haplotyping, location scores, and marker-sharing sta-tistics. Am J Hum Genet 1996;58:1323–1337

16. Abbas N, Lucking CB, Ricard S, et al. A wide variety of mu-tations in the parkin gene are responsible for autosomal reces-sive parkinsonism in Europe. French Parkinson’s Disease Ge-netics Study Group and the European Consortium on GeneticSusceptibility in Parkinson’s Disease. Hum Mol Genet 1999;8:567–574

17. Hoehn M. The natural history of Parkinson’s disease in thepre-levodopa and post-levodopa eras. Neurol Clin 1992;10:331–339

18. Yamamura Y, Sobue I, Ando K, et al. Paralysis agitans of earlyonset with marked diurnal fluctuation of symptoms. Neurology1973;23:239–244

19. Jones AC, Yamamura Y, Almasy L, et al. Autosomal recessivejuvenile Parkinsonism maps to 6q25.2-q27 in four ethnicgroups: detailed genetic mapping of the linked region. Am JHum Genet 1998;63:80–87

20. Tassin J, Durr A, de Broucker T, et al. Chromosome 6-linkedautosomal recessive early-onset parkinsonism: linkage in Euro-pean and Algerian families, extension of the clinical spectrum,and evidence of a small homozygous deletion in one family.The French Parkinson’s Disease Genetics Study Group, TheEuropean Consortium on Genetic Susceptibility in Parkinson’sDisease. Am J Hum Genet 1998;63:88–94

21. Ishikawa A, Tsuji S. Clinical analysis of 17 patients in 12 Jap-anese families with autosomal recessive type juvenile parkinson-ism. Neurology 1996;47:160–166

22. Golbe LI, Di Iorio G, Bonavita V, et al. A large kindred withautosomal dominant Parkinson’s disease. Ann Neurol 1990;27:276–282

23. Markopoulou K, Wszolek ZK, Pfeiffer RF. A Greek-Americankindred with autosomal dominant levodopa-responsive parkin-sonism and anticipation. Ann Neurol 1995;38:373–378

24. Waters CH, Miller CA. Autosomal dominant Lewy body par-

kinsonism in a four-generation family. Ann Neurol 1994;35:59–64

25. Wszolek ZK, Cordes M, Calne DB, et al. Hereditarer MorbusParkinson: Bericht uber drei Familien mit autosomal-dominantem Erbgang. Nervenarzt 1993;64:331–335

26. Wszolek ZK, Pfeiffer B, Fulgham JR, et al. Western Nebraskafamily (family D) with autosomal dominant parkinsonism.Neurology 1995;45:502–505

27. Muenter MD, Forno LS, Hornykiewicz O, et al. Hereditaryform of parkinsonism-dementia. Ann Neurol 1998;43:768–781

28. Mori H, Kondo T, Yokochi M, et al. Pathologic and biochem-ical studies of juvenile parkinsonism linked to chromosome 6q.Neurology 1998;51:890–892

29. Yokochi M, Narabayashi H, Iizuka R, Nagatsu T. Juvenileparkinsonism: some clinical, pharmacological, and neuropatho-logical aspects. Adv Neurol 1984;40:407–413

30. Sugita Y, Yokochi M, Ida A, et al. A 43-year old woman with18 years history of parkinsonism. No To Shinkei 1996;48:963–971

31. Rajput AH, Uitti RJ, Sudhakar S, Rozdilsky B. Parkinsonismand neurofibrillary tangle pathology in pigmented nuclei. AnnNeurol 1989;25:602–606

32. Takahashi H, Ohama E, Suzuki S, et al. Familial juvenileparkinsonism: clinical and pathologic study in a family. Neu-rology 1994;44:4371–4441

33. Takubo H, Kondo T, Mori H, et al. A 62-year-old man withfamilial parkinsonism with the onset at 24 years of the age. NoTo Shinkei 1996;48:587–597

34. Leroy E, Anastasopoulos D, Konitsiotis S, et al. Deletions inthe Parkin gene and genetic heterogeneity in a Greek familywith early onset Parkinson’s disease. Hum Genet 1998;103:424–427

35. Lucking CB, Abbas N, Durr A, et al. Homozygous deletions inparkin gene in European and North African families with au-tosomal recessive juvenile parkinsonism. The European Consor-tium on Genetic Susceptibility in Parkinson’s Disease and theFrench Parkinson’s Disease Genetics Study Group. Lancet1998;352:1355–1356

Klein et al: Clinical and Genetic Study in Parkinson’s Disease 71