Autosomal Recessive Juvenile Parkinsonism (AR-JP): Genetic Diagnosis

Autosomal recessive juvenile parkinsonism (AR-JP) is a familial levodoparesponsive parkinsonism resulting from Lewy body negative degeneration of nigral neurons in the zona compacta of the substantia nigra (1-4). The first proposal for a distinct clinical entity with recessively inherited parkinsonism was made in Japan and was termed "paralysis agitans with marked diurnal fluctuations of symptoms" (1). This syndrome was later designated as autosomal recessive form of juvenile parkinsonism (2). It was subsequently found to be linked to the 17-cM region on chromosome 6q25.2-27, and the locus was recently designated Park2 (3,5). Through the study of a patient who had homozygous microdeletion of the marker D6S305 (5), the responsible gene was identified by positional cloning and was designated parkin (6). Linkage and mutation analysis to date have shown that founders of mutations in this gene are multiple and widely distributed in the world (7-13). Abnormalities in this gene, which are specific for AR-JP, include homozygous exonic deletions, small deletions, and point mutations. The presence of homozygous exonic deletions strengthens the notion that nigral neurodegeneration in AR-JP is caused by loss of function of the parkin protein.


Introduction
Autosomal recessive juvenile parkinsonism (AR-JP) is a familial levodoparesponsive parkinsonism resulting from Lewy body negative degeneration of nigral neurons in the zona compacta of the substantia nigra (1-4). The first proposal for a distinct clinical entity with recessively inherited parkinsonism was made in Japan and was termed "paralysis agitans with marked diurnal fluctuations of symptoms" (1). This syndrome was later designated as autosomal recessive form of juvenile parkinsonism (2). It was subsequently found to be linked to the 17-cM region on chromosome 6q25.2-27, and the locus was recently designated Park2 (3,5). Through the study of a patient who had homozygous microdeletion of the marker D6S305 (5), the responsible gene was identified by positional cloning and was designated parkin (6). Linkage and mutation analysis to date have shown that founders of mutations in this gene are multiple and widely distributed in the world (7-13). Abnormalities in this gene, which are specific for AR-JP, include homozygous exonic deletions, small deletions, and point mutations. The presence of homozygous exonic deletions strengthens the notion that nigral neurodegeneration in AR-JP is caused by loss of function of the parkin protein.

Clinical and Pathologic Manifestations of AR-JP
The cardinal features of AR-JP are early-onset parkinsonism with a benign course and remarkable response to levodopa. The following clinical features are also important to support the diagnosis of AR-JP (Table 1): 1. Mild focal dystonia, which often manifests as unilateral foot dystonia-dorsiflexion of the big toe or pes equinovarus deformity. Dorsiflexion of the big toe can be easily observed when the patient sits on a high chair or walks with bare feet. In some cases, truncal dystonia is the first symptom. 2. Sleep benefit, which can be identified by asking patients whether their parkinsonian symptoms improve after naps, or whether their symptoms are much milder upon awakening in the morning compared with the evening. 3. Extremely slow progression of the disease and absence of dementia even in the terminal stages of the disease. 4. Rare occurrence of autonomic dysfunction such as constipation or neurogenic bladder. 5. Fine postural finger tremor. 6. Hyperactive deep tendon reflexes with a negative Babinski sign. 7. Dopa-induced dyskinesia, which soon follows the dramatic dopa responsiveness. 8. Wearing-off phenomenon, which is frequently encountered in a relatively early phase of the disease.

Family Interview
Family interviews typically reveal multiple affected individuals in one generation with no appearance of the disease in previous generations or offspring of pateints. Thus, if the patient has no siblings, AR-JP can manifest as a sporadic early-onset parkinsonism. Although 51% of AR-JP families (9/17) have consanguineous marriages (4), a sufficiently large proportion (49%) have no history of consanguinity despite exhaustive family interviews (4). Nevertheless, patients from these non-consanguineous families frequently have homozygous haplotypes (63%; see Subheading 1.2.4.), which indicates the presence of an ancient consanguineous loop. In such cases, the parents' families frequently originated from the same geographic area.

Structure and Expression of the parkin Gene
The parkin gene consists of 12 exons encoding 465 amino acids, with a molecular weight of 51,652 D (Fig. 1). The full-length cDNA, which has been isolated from human skeletal muscle and fetal brain cDNA libraries consists of 2860 bp with an open reading frame of 1395 bp. The N-terminal 76 amino acid residues show homology to ubiquitin (65% positive, 33% identical). The characteristic cysteine-rich motif (Cys-X2-Cys-X9-Cys-X1-His-X2-Cys-X4-Cys-X4-Cys-X2-Cys) is also found at the C-terminus of parkin. The parkin gene is ubiquitously transcribed. Northern blot analysis using full-length parkin cDNA as probe revealed a 4.5 kb mRNA in almost all tissues (6). In the brain, parkin mRNA is present in several regions, including cerebellum, substantia nigra, cerebral cortex, brainstem, putamen, caudate, hippocampus, amygdala, and thalamus. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis using leukocyte RNA revealed no full-length mRNA but a shorter transcript in which exons 3, 4, and 5 are spliced out. In the brain, the full-length transcript, as well as a small amount of mRNA with a spliced-out exon 5, has been detected by RT-PCR (14).

Analysis of Exon Deletions by Genomic PCR
A wide variety of deletion mutations in the parkin gene have been reported so far ( Table 2). If the patient is homozygous for the deletion, it is detectable by lack of a genomic PCR product using intron primers encompassing the deleted exons. However, if a patient is heterozygous for the deletions (compound heterozygote: see Note 1), only the exon whose deletion is shared by both chromosomes fails to be amplified. If no part of the deletion is shared between the two chromosomes, exon PCR cannot detect any deletion. For example, if an individual receives exon 3 deletion from the father and exon 4 deletion from the mother, exon PCR cannot detect any deletion. Southern blot analysis is not dependable for evaluation of such small changes in gene dosage. Accordingly, when the patient shows a heterozygous haplotype for mark-ers on the AR-JP locus, negative results from exon PCR do not necessarily mean that the patient has no deletion in the AR-JP gene.
To date, exon PCR has been effective for the detection of deletions in 57% of chromosome 6q-linked recessive juvenile parkinsonism (12 of 21 families) in Japan and in 25% (3 of 12 families) in Europe and North Africa (11). Thus, 25-57% of clinical AR-JP can be detected by exon PCR.

Exon Sequencing
When exon PCR shows no deletion, the next step is to sequence each exon and its boundaries. A wide variety of point mutations in the parkin gene have been reported so far ( Table 2).
Homozygous one-point mutations, small insertions or deletions at the same nucleotide site on both chromosomes could be detected. Alternatively, if the patient is a compound heterozygote (see Note 1), one-point heterozygous mutation at the same nucleotide position might also be detected. When a heterozygous point mutation is observed, the presence of a compound heterozygote with a deletion in the other chromosome is possible. When two-point mutations are observed at different sites, it is necessary to exclude the possible presence of two mutations residing on the same chromosome. This can be done by sequencing a carrier who has only one disease chromosome, which can be detected by haplotype analysis. If only one of these two-point mutations is observed in the carrier, the patient is a compound heterozygote. If both of the two-point mutations are present in the carrier, the patient may have two point mutations in one chromosome. In the latter case, it is still possible that the patient is a compound heterozygote with a deletion in one chromosome and two-point mutations in the other. When a new homozygous one-point mutation is identified, the possibility of polymorphic mutation should be assessed. Several polymorphic mutations in the parkin gene have been reported (Table 2) (13,15).

Haplotype Analysis
As mentioned above, haplotype analysis is mandatory to interpret correctly the results of exon deletions and point mutations (see Note 2). If an affected patient has a heterozygous haplotype for the parkin gene, he/she is expected to be a compound heterozygote, receiving different mutations from each parent. In AR-JP derived from a consanguineous marriage, the patient usually receives the identical mutation from both parents, and thus should be homozygous for polymorphic marker alleles located in and around the parkin gene. However, in AR-JP, mutations in the parkin gene are variable and widely distributed in the world. This multiple-founder effect increases the likelihood of the occurrence of the disease from nonconsanguineous marriages, resulting in compound heterozygotes.
It should be noted that although the normal carrier state (heterozygote) of a deletion cannot be detected by conventional exon PCR, it can be detected by haplotype analysis if the individual belongs to the same family as the affected proband and the parents have heterozygous haplotypes (Figs. 2-4). When only patients' samples are available, it is desirable to calculate allele frequencies of the markers in the general population from which affected families originate. If the frequencies of the marker alleles are rare, haplotype homozygosity alone is sufficient to indicate the true linkage of the haplotype to the disease (see Note 3).

Analysis of parkin mRNA and Protein
Absence or truncation of parkin transcripts can be detected by RT-PCR using tissue RNA samples. The presence of tissue-specific splicing of parkin transcripts should be taken into consideration. For example, full-length parkin transcript is absent in peripheral leukocytes (14). When a specific antibody is available, Western blot analysis using tissue samples can detect abnormalities of parkin translated products. Analysis of the parkin mRNA and protein has just begun, and further studies should become available in the near future. Such analyses would be helpful in the diagnosis of AR-JP when genomic studies are not informative.

Perspectives
Even when PCR-based studies of homozygous exonic deletions and point mutations are negative in a particular patient, the diagnosis of AR-JP cannot be excluded if haplotype analysis shows a heterozygous haplotype. Individual patients might be compound heterozygotes having two different exonic deletions that do not share a common segment. When a hetrozygous point mutation is present in a patient, a compound state with one deletion and one point mutation should be evaluated. Thus, without a sensitive method to detect small changes in gene dosage such as heterozygous deletions, the diagnosis of AR-JP should depend on the efforts to put together the results of PCR-based analysis of the mutation and haplotype studies of the pedigree.
The existence of multiple founder mutations in the parkin gene and the high proportion of nonconsanguinity in AR-JP pedigrees (49%) indicate a high frequency of compound heterozygotes and asymptomatic carriers of parkin mutations in the normal population, resulting in a potentially high prevalence of sporadic cases of AR-JP (4). The major obstacle for assessing the latter possibility is the difficulty in detecting deletion heterozygotes.
Recently, real-time PCR monitoring by fluorescent-energy transfer techniques such as TaqMan or LightCycler system have been introduced to detect such small differences in gene dosage (16). These technical improvements could enable the detection of deletion heterozygotes in the parkin gene. At the  . PCR products generate a complex of peaks, with several smaller peaks are located left of the highest peak, which represents shorter products generated by the skipping phenomenon of amplification. This phenomenon is often observed in amplification of short nucleotide repeats. The length difference between each of these skipping peaks is 2 bp, because D6S305 is a dinucleotide repeat polymorphic marker. Two alleles (226 and 234 bp) are seen in this family. Individuals 1 and 2 are parents who are first cousins. Individual 1 shows a single allele (234 bp). Individual 2 shows a single allele, which is different in size from that of individual 1 (226 bp). If individuals 1 and 2 are homozygotes, all offspring should show a heterozygous allotype (226/234 bp). However, individual 4 shows a single allele (234 bp), indicating that this person received a null (deleted) allele (shown as X in the family pedigree) from individual 2. This in turn suggests that individual 2 has heterozygous deletion of this marker (X/226 bp). On the other hand, individual 3, who has clinical AR-JP, shows no PCR product, indicating that she received two deleted alleles, one from each parent (X/X). The latter observation means that individual 1 also has a heterozygous deletion of the marker (X/234 bp). Individual 5 shows a heterozygous allotype (226/234 bp), indicating that he received no deleted allele from either parent. These findings taken together indicate that the responsible gene for AR-JP resides in close proximity to D6S305. same time, the full sequencing of the genomic region in and around the parkin gene is in progress, which will enable the detection of mutations in the noncoding region of this gene as well.
α-Synuclein aggregation is considered a major cause of Lewy body formation. Nonetheless, the cell death pathway triggered by α-synuclein aggregation is not clear. The unique feature of neuronal death in AR-JP, namely, absence of Lewy body formation, suggests the possibility that the downstream event in the cell death cascade triggered by α-synuclein aggregation might share the same biochemical pathway involving parkin (17). Genetic analysis of mutations in the parkin gene in AR-JP patients will eventually contribute to the elucidation of the functional role of parkin in the pathogenesis of Parkinson's disease.

MATERIALS 2.1. Exon Deletions
1. Chimeric primers with M13 universal and reverse primer sequences at their 5' ends are used for exon PCR as well as for exon sequencing (

1.
A recessive disease is caused by the presence of two mutations, each of which has occurred in the same gene residing on homologous chromosomes. A compound heterozygote is a patient who has two different mutations on each of homologous chromosomes. As each of the mutations is derived from a different ancestor of the disease mutation, a compound heterozygote has two different haplotypes (see Note 8), which originate from different ancestors of the mutation. 2. A haplotype is a set of alleles on one chromosome. Alleles are alternative forms of a gene or marker occupying the same locus on homologous chromosomes. As human cells have two copies of each chromosome (diploid cells), an individual always has a pair of alleles, one from each parent. Accordingly, an individual has two haplotypes. If alleles are very closely linked, haplotypes within a kindred are transmitted as units. However, when alleles are not closely located, recombination by crossing over occurs and haplotypes are changed. Homozygotes have the same alleles or haplotypes on both homologous chromosomes, whereas heterozygotes have different alleles or haplotypes. 3. In a consanguineous pedigree, each parent is usually a carrier of the same mutation, i.e., has a single identical mutation derived from a single person who first acquired the mutation in an earlier generation, i.e., the ancestor of the mutation. Accordingly, if a patient born from a consanguineous marriage has homozygous haplotypes for the markers that flank or reside in a certain gene, this is a strong indication that the patient has two identical mutations in the same gene (theory of homozygosity mapping) (18). The probability for homozygosity to show true linkage is heavily dependent on the rarity of the alleles or haplotypes showing homozygosity. This is based on the fact that if the frequency of the marker in the control population is rare, the chance for its heterozygosity in the general population as well as in the parents increases. The latter, in turn, increases the power of detection for the single identical allele to be transmitted from each parent to the affected person (homozygosity by descent). On the other hand, if the marker frequency is high in the general population, homozygosity by chance increases and, therefore, homozygosity in the patient by itself is not informative (homozygos-ity by state). Thus, to substantiate segregation of the haplotype with the disease, especially when only information about the patients' haplotypes is available, knowledge of the allele frequencies of the markers that constitute homozygous haplotypes are important. Analysis of 30-50 DNA samples obtained from normal persons is sufficient to determine allele frequencies of the markers. 4. PCR with primers without M13 universal sequences are also possible. In this case, the extracted DNA can be directly sequenced using internal primer sequences or can be subcloned into the TA-vector plasmid (TA-vector cloning kit, Invitrogen) without filling in the ends of the DNA fragment. The insert in the TAvector can be sequenced with universal primers (M13 and M13 reverse). Several clones should be assessed to exclude possible PCR-and cloning-based mutations. 5. If extra bands in PCR products are observed in the gel, cutting the band corresponding to the expected size, extraction, and purification of DNA with the Quiaquick Gel extraction kit (Qiagen) is recommended for further sequencing. 6. Single-strand sequencing using T7 polymerase is an alternative method for the sequencing. The major merit of single-strand sequencing with T7 polymerase is uniformity of signal intensity, allowing easy detection of heterozygous mutations. The sequencing kit (Autoread sequencing kit) can be purchased from Pharmacia (Uppsala, Sweden). The single-strand template is recovered from the PCR product by magnetic force. As one of the PCR primers is biotin-labeled, the addition of streptavidin-coated magnetic beads (Dynal) to the PCR product results in their binding to the biotin-labeled DNA strand. Accordingly, the biotin labeled strand is isolated by magnetic force. A sequencing sample is applied in four lanes (A, C, G, and T) of the sequencing gel and analyzed with a Pharmacia ALF2 fluorescence autosequence analyzer. Universal sequences are added to the 5' end of PCR primers and fluorescein isothiocyanate (FITC)-labeled universal primers are used for sequencing. FITC-labeled universal primers are included in the Autoread sequencing kit (5'-CGACGTTTAAAACGACGGCCAGT-3' for M13 primer and 5'-CAGGAGGCAGCTATGAC-3' for M13 reverse primer). Sequencing primers must be derived from the region located at least one nucleotide internal to the site of PCR primers. 7. Scale-out of the peak of the signal occurs when dilution of the sample is insufficient. 8. Two different types of size standards-internal and external-are used. As internal standards, two size markers encompassing the size of the PCR product are loaded in the same lane with the PCR product. For example, Sizers 200 and 300 are loaded with the product whose expected size is between 200 and 300 bp. The molecular size of the peak of the PCR product is determined by reading the retention times of respective peaks of internal standards flanking the PCR product. As external standards, only the sizer markers are loaded in the lane that is called as the reference lane. For each group of sample lanes (usually four to five lanes) with reference lanes on both sides, the standard curves of the reference lanes are calculated. The molecular size of the PCR sample is calculated by first using the external standard and then adjusting the resulting standard curves to the internal reference points.