Genetic and clinical study of PARK7 in Japanese Parkinson's disease

Background Biallelic variants in PARK7, which encodes protein-nucleic acid deglycase DJ-1, can cause early-onset Parkinson's disease (PD). Although many patients with PARK7 variants have been identified from European and Middle Eastern ethnic groups, there have been no reports in the Japanese population. Objectives To determine the prevalence and clinical features of patients with PD harboring PARK7 variants in Japan. Methods We performed a molecular genetic analysis of PD patients with PARK7 variants identified using comprehensive panel sequencing, to explore the details of variants. Moreover, clinical neurological features were investigated, including neuroimaging analyses. This study followed STROBE guidelines. Results Four patients with biallelic rare variants of PARK7 were identified in the cohort. All four patients presented with levodopa-responsive parkinsonism, with an age at onset in the early 30s. Furthermore, two of the four patients had psychiatric complications. Dopamine transporter imaging revealed nigrostriatal pathway dysfunction. Conclusions To our knowledge, this is the first report of Japanese patients with PARK7 variants. We identified a relatively low frequency of PARK7 variants in patients in Japan. As opposed to typical patients with sporadic PD, the identified patients developed the disease in their 30s and presented with a variety of non-motor symptoms and complications. Further studies are needed to identify the clinical features related to PARK7 variants in Japanese patients with PD, and to analyze the pathophysiology of how the variants identified in the present study might affect DJ-1 function.

identify the clinical features related to PARK7 variants in Japanese patients with PD, and to analyze the pathophysiology of how the variants identified in the present study might affect DJ-1 function.

Introduction
Parkinson's disease (PD) is the second most common neurodegenerative disease after Alzheimer's disease, affecting more than 1 % of individuals over 60 years of age [1,2].The most prominent clinical features of PD are motor symptoms such as bradykinesia, rigidity, and resting tremor, which have an excellent response to levodopa therapy.Although the motor symptoms of PD are caused by the degeneration of dopaminergic neurons in the substantia nigra, patients also present with various non-motor symptoms that are caused by neurodegeneration of the non-dopaminergic system.The etiology of PD remains unclear and there are currently no curative therapies; however, it is anticipated that the identification of disease-related genes from patients with familial PD will contribute to breakthroughs in the study of PD.More than 30 different causative genes for familial PD have been reported, including dominant types of PD caused by variants in synuclein alpha (SNCA) and leucine rich repeat kinase 2 (LRRK2), and recessive types caused by variants in parkin RBR E3 ubiquitin protein ligase (PRKN), PTEN induced kinase 1 (PINK1), and Parkinsonism associated deglycase (PARK7, or DJ-1) [2].
DJ-1 was initially discovered as an oncogene; however, in 2003, Bonifati and colleagues identified two biallelic variants from two separate consanguineous families with early-onset PD (MIM#606324), thus identifying DJ-1 as a recessive causative gene for familial PD [3,4].PARK7 is located on chromosome 1p36 and consists of seven exons encoding 189 amino acids.The gene product, DJ-1, has a molecular weight of 22 kDa, is highly conserved across species, and is widely expressed in various tissues, especially the brain and testis [5].Since the original report of PARK7 as a PD-related gene, many validation studies have been conducted in diverse populations; however, the prevalence of pathogenic PARK7 variants in PD patients appears to be relatively low [6].To better understand the function of DJ-1 and the detailed genetic and clinical features associated with PARK7 biallelic variants, it is essential to study the etiology of DJ-1 and early-onset PD by accumulating as many patients as possible.Here, we investigated the characteristics of PARK7 variants in Japanese patients with PD, and explored the associated clinical features.We identified four patients from independent families with biallelic PARK7 variants in our Japanese PD cohort, and analyzed their clinical courses.Our findings provide a further understanding of the impact of PARK7 variants on clinical features of PD.

Participants
This study was approved by the ethics committee of the Juntendo University Faculty of Medicine (approval number M08-0477).All participants gave informed written consent before participation, and all patients met the standard diagnostic criteria of PD [1,7].Medical records were used to obtain clinical data such as clinical course; symptoms; and the results of neuroimaging such as brain magnetic resonance imaging (MRI), 123 I-metaiodobenzylguanidine (MIBG) myocardial scintigraphy, dopamine transporter single-photon emission computed tomography (DAT-SPECT), 99m Tc-ethyl cysteinate dimer SPECT ( 99m Tc-ECD-SPECT), and N-isopropyl-p-123 I-iodoamphetamine SPECT ( 123 I-IMP-SPECT).A total of 1716 Japanese patients with PD participated in the study.Patients in whom rare variants in PD-related genes other than PARK7 were identified before the start of this study were excluded.The cohort comprised 1241 patients with PD who underwent panel sequencing (described in section 2.2) from June 2016 to March 2023, and 475 patients who did not undergo panel sequencing as an additional analysis and who were under 50 years of age at onset.Patients considered to have developed PD due to PARK7 variants were subjected to neurological analyses.A study flow chart is shown in Fig. 1.

Mutation screening
We extracted genomic DNA from the peripheral blood of patients using the QIAamp DNA Blood Maxi kit (Qiagen, Venlo, Netherlands).Patients were screened for genes related to familial PD or dementia (panel ID: IAD103177_182) using high-throughput next-generation sequencing with an Ion Torrent System (Thermo Fisher Scientific, Waltham, MA, USA).The methods of our targeted gene panel screening have been described previously [8].Sanger sequencing was performed to validate non-synonymous variants detected by panel sequencing.Multiplex Ligation-Dependent Probe Amplification (MLPA) was used for copy number variant analysis using the SALSA MLPA® P051 Parkinson probe mix (MRC Holland, Amsterdam, Netherlands).Raw data for gene dosage quantification were collected on a SeqStudio sequencer (Thermo Fisher Scientific) and analyzed with Coffalyser software (MRC Holland).Probe ratios below 0.7 or above 1.3 were defined by the Coffalyser software as indicating a heterozygous deletion or a heterozygous duplication, respectively.Haplotype analysis was performed by fragment analysis and the genotyping of microsatellites located in flanking PARK7 using fluorescently labeled primers on a SeqStudio sequencer, and by the genotyping of single nucleotide variants using Sanger sequencing.Haplotype phases were determined by comparison with haplotypes from the parents of Patient D-II-2.The primer sequence information (for the Sanger sequencing and haplotype analysis) is available upon request.

Quantitative reverse transcription polymerase chain reaction (qPCR)
qPCR was performed to quantify PARK7 mRNA.Total RNA was collected from peripheral blood using the PAXgene RNA system (BD, Franklin Lakes, NJ, USA) according to the manufacturer's protocol.cDNA was synthesized from total RNA using the ReverTra Ace qPCR RT Master Mix (TaKaRa Bio, Tsu, Japan).qPCR was performed on the QuantStudio 7 Flex Real-Time PCR System (Thermo Fisher Scientific) using PowerUp SYBR Green Master Mix (Thermo Fisher Scientific) and PARK7-specific qPCR primers.Primer3Plus was used for primer design [9], and at least one primer was designed to be located at the exon-exon junction and amplify cDNA only.Two targets (exons 2-3 and exon 7) were set in PARK7.The qPCR experiments were performed in triplicate and repeated three times.Relative expression levels were calculated using ACTB as the reference, and were compared statistically by one-way analysis of variance and Tukey's post hoc test using JMP17 (JMP, Cary, NC, USA).The qPCR primer sequence information is available upon request.

Bioinformatic analysis
For this bioinformatic analysis, NM_007262 was used as the reference sequence.The pathogenicity of each missense variant was estimated using Rare Exome Variant Ensemble Learner (REVEL) [10].Furthermore, the Single Nucleotide Polymorphism Database (dbSNP) [11] was used to search for the allele frequencies of previously reported variants.

Age demographics of the participants
Of the 1716 patients with PD who were enrolled in the present study, 688 had familial PD (with the same phenotype identified in the same family) and 1028 had sporadic PD.The detailed age characteristics of the participants are described in Table 1.The mean age at onset was 14.4 years earlier in sporadic PD than in familial PD; this difference may have been caused by a sample selection bias because most of the early-onset PD cases who were analyzed additionally using Sanger sequencing had sporadic PD (Fig. 1).

Frequency of PARK7 variants in Japanese patients with PD
In our cohort, we identified four patients with biallelic rare variants of PARK7.The frequencies were 0.23 % of the total cohort, 0.29 % (2/688) of patients with familial PD, and 0.19 % (2/1028) of patients with sporadic PD.The prevalence of variants in patients with PD with an age at onset <50 years was 0.34 % (4/1168).No PARK7 variants were identified in patients with PD with an age at onset ≥50 years.PD, Parkinson's disease; SD, standard deviation; y, years.
M. Ishiguro et al.

Identified biallelic PARK7 variants in PD patients
Two rare variants and one exonic deletion were identified in four patients with PD from independent families: a homozygous c.218C > T (p.P73L) variant in one patient (A-III-1), a homozygous c.242dup (p.N81Kfs*4) variant in two patients (B-II-4 and C-II-1), and a compound heterozygous c.242dup (p.N81Kfs*4) variant and exon 6 deletion in one patient (D-II-2) (Fig. 2A-C).Further exploration of the exonic deletion breakpoint revealed that 4.9 kbp of hg38 chr1:7977016_7981953 was heterozygously deleted in Patient D-II-2 (Fig. 2D).When we compared the 300 bases forward of the deletion with the 300 bases at the deletion site, there was 86 % homology.Moreover, c.218C > T (p.P73L) was registered as rs367584305 and c.242dup (p.N81Kfs*4) was registered as rs1640403698; both were rare variants, especially rs1640403698, which has been detected in the Japanese population only.No other putative pathogenic variants in PD-related genes were identified via panel sequencing in these patients.In silico prediction by REVEL showed that c.218C > T (p.P73L) had a high REVEL score, of 0.837.In addition, c.242dup (p.N81Kfs*4) was predicted to produce a stop codon after four amino acids and cause nonsense-mediated mRNA decay.Indeed, the analysis of PARK7 mRNA expression in Patients C-II-1 and D-II-2, from whom RNA was able to be obtained, showed a prominent and significant decrease compared with controls (without PARK7 variants) (Fig. 3).Furthermore, c.242dup (p.N81Kfs*4) was identified in three of the four patients with PARK7 variants in our cohort.Patients C-II-1 and D-II-2 may have shared a common founder with a maximum range of approximately 1.1 Mbp (including PARK7), whereas B-II-4 had less useful and inconclusive genetic marker information (Table 2).We also detected a c.310G > A (p.A104T) variant in one PD patient; however, although c.310G > A (p.A104T) has been reported as a PD-associated variant, its pathogenicity remains unknown because it was a monoallelic variant (data not shown) [12].

Clinical findings of patients with biallelic PARK7 variants
Patient A-III-1 was a 48-year-old man who had experienced panic attacks since his teens.At the age of 35 years, he presented gait disturbance and dysphonia and was diagnosed with PD.Aged 40 years, wearing-off and psychiatric symptoms such as hallucinations, auditory hallucinations, and delusions appeared.His psychiatric symptoms improved with the discontinuation of dopamine agonists.During the off-state, he experienced dysarthria with stuttering and difficulty walking.At the age of 43 years, he frequently overdosed on levodopa, and was diagnosed with dopamine dysregulation syndrome.When his levodopa medications increased, he had increased agitation and irritability.From 45 years of age, he has been repeatedly hospitalized for long periods in mental hospitals to control his agitation.At the age of 48 years, his activities of daily living classify him as Hoehn and Yahr stage III.He has not experienced any cognitive decline.His younger sister had intellectual disability from birth without any movement disorders.Interviews with the patient's parents revealed that the patient's grandmother may also have had PD.Furthermore, given that the family's relatives lived in a historically closed area, the possibility of an accidental consanguineous marriage cannot be ruled out.In terms of imaging, the patient's brain MRI was normal.DAT-SPECT showed a reduction of dopamine transporter uptake (Fig. 4A).His heart to mediastinum (H/M) ratio did not decrease in MIBG myocardial scintigraphy (Fig. 4B).99m Tc-ECD-SPECT showed lower uptake in the right posterior lobe (Table 3 and Fig. 4D).
Patient B-II-4 was a 69-year-old woman.She noticed gait disturbance, resting tremor, and bradykinesia at the age of 32 years.Her older sister developed PD in her 30s and passed away aged 45 years, and her unaffected parents had a consanguineous marriage.The patient was diagnosed with PD, and levodopa ameliorated her parkinsonism.She developed hallucinations and delusions in her 40s.Dystonia of both lower extremities appeared at 47 years of age.At the age of 58 years, she reported sensory disturbances with hyporeflexia of all limbs; a nerve conduction study revealed axonal neuropathy.She has been bedridden since the age of 60 years, without any cognitive decline.In terms of imaging, the patient's brain MRI was normal, and DAT-SPECT showed a reduction of dopamine transporter uptake.Her H/M ratio did not decrease in MIBG myocardial scintigraphy. 123I-IMP-SPECT showed lower uptake in the frontal lobe (Table 3).
Patient C-II-1 was a 36-year-old man who developed tremor at the age of 32 years.He had no family history, and demonstrated good levodopa responsiveness.He presented with cervical dystonia, which improved with oral levodopa and dopamine agonists.Aged 36 years, he was taking 250 mg levodopa and 9 mg rotigotine and was classified as Hoehn and Yahr stage II.His brain MRI was normal (Fig. 4C).DAT-SPECT showed a reduction of dopamine transporter uptake (Fig. 4A).His H/M ratio was decreased in MIBG myocardial scintigraphy (Table 3 and Fig. 4B).
Patient D-II-2 was a 39-year-old man.He developed a resting tremor in his right hand at the age of 32 years.Aged 34 years, he had a shuffling gait.He then developed spasticity, and at the age of 37 years, he was suspected to have hereditary spastic paraplegia.His spasticity worsened, and he required assistance to walk.Myoclonus of the upper extremities was observed.In a levodopa challenge test, he showed improvements in masked face and stiffness of the upper extremities, but his improvement rate was 23 % in part III of the Movement Disorder Society-Unified Parkinson's Disease Rating Scale.He had cognitive decline, with a Mini-Mental Scale Examination score of 14 points.In imaging findings, his brain MRI was normal.DAT-SPECT showed a reduction of dopamine transporter uptake.His H/M ratio did not decrease in MIBG myocardial scintigraphy. 123I-IMP-SPECT showed lower uptake in the frontal and temporal lobes (Table 3).

Discussion
In the present study, we screened a large Japanese PD population for patients with PARK7 variants; we identified four patients with early-onset PD who had pathogenic biallelic variants.Of the Japanese PD patients with an age of onset <50 years, 0.34 % had biallelic PARK7 variants; this frequency is consistent with previous reports [6].However, it should be noted that our cohort excluded PD cases with previously identified pathological variants in PD-related genes other than PARK7, so the actual frequency was expected to be even lower.Notably, in three of the four PD patients, we detected a frameshift variant that resulted from the same type of single nucleotide insertion.Haplotype analysis suggested that at least two of the families with this variant shared a common founder.Moreover, database analysis revealed that this variant has been identified in the Japanese population only, suggesting that it may be a Japanese-specific variant.In addition, a single nucleotide substitution variant that was detected in the present study has been reported at very low allele frequencies in various ethnic groups.Furthermore, the encoded amino acids are highly conserved across species.The  identified missense variants were determined as pathogenic by REVEL, suggesting that these biallelic variants are causative for PD development.In addition, one patient had a heterozygous deletion in exon 6, and a comparison of the 300 bases forward of the deletion with the 300 bases at the deletion site showed 86 % homology.This site is therefore considered likely to be susceptible to recombination.By contrast, c.310G > A (p.A104T), which has already been reported as pathogenic for PD, was detected heterozygously in one patient with PD and its pathogenicity is unclear [12].
Most previously reported patients with PARK7 variants had PD onset in their 20s-30s, with rare cases of disease onset in infants and teenagers [13][14][15][16][17].Although the reported clinical features have been variable, patients generally present with good levodopa responsiveness, dystonia, and dyskinesia [13].In the present study, all patients shared a common onset of parkinsonism in their 30s.However, two patients (A-III-1 and D-II-2) had psychiatric symptoms, whereas two others (B-II-4 and C-II-1) with the same type of biallelic variants had dystonia.A range of complications are often reported in patients with PARK7 variants, including dysphagia, dysarthria, peripheral neuropathy [16,18,19], pyramidal tract disorders, dysphonia, and muscle atrophy; however, it may be that the type of complications that develop depend on the variant of each patient [15,[18][19][20].Cases with p.E163K and p.Q45X variants are reported to be complicated with amyotrophic lateral sclerosis [21,22].Interestingly, Patient B-II-4 had axonal peripheral neuropathy, and Patient D-II-2 had severe spasticity mimicking hereditary spastic paraplegia.In addition to amyotrophic lateral sclerosis complications, pyramidal tract signs were reportedly positive in three studies that mentioned pyramidal tract signs [14][15][16]18,19,23].Together, these findings suggest that PARK7 variants might also be associated with motor neuron degeneration.In all of our cases, brain MRI showed no abnormalities.Similarly, most previously reported patients with biallelic PARK7 variants reported have shown no abnormalities, although a few cases showed cortical or cerebellar atrophy, suggesting that PARK7 variants do not substantially affect brain atrophy [16,18].Striatal dopamine transporter binding was low in all three examined patients, three of our four patients showed normal H/M ratio values.In our previous studies, H/M ratios were heterogeneous in patients with familial PD-associated genes [8,[24][25][26].PARK7 variants may therefore lead to a loss of DJ-1 function, striatal neuronal loss, and typical symptoms of PD (i.e., with similar motor symptoms, age of onset, and responsiveness to levodopa therapy as typical PD).However, PARK7 variants may also cause various non-motor symptoms, psychiatric symptoms, and neuromuscular diseases with different variants and disease durations.
The monomeric DJ-1 protein contains seven β-strands and nine α-helices in a three-layered structure, with α-helices on either side of the parallel β-sheet [5].Previous studies have shown that DJ-1 forms dimers, which are made by interactions of the β-sheet β3 and the α-helices α1, α8, and α9.The H-bond between C-terminal residues Pro 184 and His126 is essential for dimer formation [27].In the case of Patient D-II-2, the deletion of exon 6 suggests that the C-terminal structure of the DJ-1 protein remains intact, because exon 7 is expected to remain in-frame.However, the deletion of His126 encoded in exon 6 is expected to prevent dimer formation.Chen and colleagues have also reported that His126 is an amino acid for the protease activity of DJ-1 [28].Together, these findings suggest that this amino acid is crucial for multiple functions of DJ-1, including dimer formation and protease activity.For the p.P73L variant, Pro 73 usually exists in close proximity to β4 in the β-sheet structure in the DJ-1 molecule.The amino acid substitution from proline to leucine is therefore expected to reduce the structural stability of DJ-1.The p.N81Kfs*4 variant led to significantly lower mRNA expression in the current study, suggesting that transcripts are rapidly degraded by nonsense-mediated mRNA decay, and that there is little or null DJ-1 protein.
The present study includes some limitations.First, the clinical data were evaluated as a cross-sectional study, without concern for time differences in each patient.Second, we did not perform any functional analyses using cell culture or model animals.Finally, because there is a high proportion of samples from cases of familial PD in our DNA bank, the prevalence reported in the present study might differ slightly from that of a general population of PD patients.

Conclusions
In conclusion, patients with PARK7 variants were rare in the Japanese population.Patients presented in their 30s and showed various clinical symptoms alongside levodopa-responsive parkinsonism.Further studies are needed to elucidate the mechanisms by which PARK7 variants may lead to PD.

Fig. 1 .
Fig. 1.Study flow chart.In the study, 1716 patients were analyzed using panel sequencing (seq) or Sanger seq.Pathological variants of PARK7 were identified in four patients.

Fig. 2 .
Fig. 2. Mutation analysis of PARK7.A: Family tree of PD patients with biallelic variants of PARK7.B: Electropherograms of pathological single nucleotide variations detected in this study.C: Results of the MLPA of Family D. D: Sequencing electropherogram of the deletion breakpoint from Patient D-II-2.

Table 1
Age characteristics of individuals.

Table 2 Haplotype analysis of PARK7 rare variant carriers.
Estimated common haplotypes are indicated by gray backgrounds.M.Ishiguro et al.

Table 3
Clinical features of patients with biallelic PARK7 variants.relationship with Takeda Pharmaceutical Company Limited that includes: speaking and lecture fees.Masahiko Tomiyama reports a relationship with Kyowa Kirin Co Ltd that includes: speaking and lecture fees.Masahiko Tomiyama reports a relationship with Ono Pharmaceutical Co Ltd that includes: speaking and lecture fees.Masahiko Tomiyama reports a relationship with Eisai Co Ltd that includes: speaking and lecture fees.Yuanzhe Li reports a relationship with Eisai Co Ltd that includes: employment.Yuanzhe Li reports a relationship with Nihon Medi-Physics Co Ltd that includes: employment.Nobutaka Hattori reports a relationship with Sumitomo Pharma Co Ltd that includes: board membership, consulting or advisory, paid expert testimony, and speaking and lecture fees.Nobutaka Hattori reports a relationship with Takeda Pharmaceutical Company Limited that includes: board membership, consulting or advisory, paid expert testimony, and speaking and lecture fees.Nobutaka Hattori reports a relationship with Kyowa Kirin Co Ltd that includes: board membership, consulting or advisory, paid expert testimony, and speaking and lecture fees.Nobutaka Hattori reports a relationship with Teijin Pharma Limited that includes: board membership, consulting or advisory, paid expert testimony, and speaking and lecture fees.Nobutaka Hattori reports a relationship with PARKINSON Laboratories Co. Ltd that includes: consulting or advisory and equity or stocks.Nobutaka Hattori reports a relationship with Ono Pharmaceutical Co Ltd that includes: board membership, paid expert testimony, and speaking and lecture fees.Nobutaka Hattori reports a relationship with Riken Center for Brain Science that includes: employment.Nobutaka Hattori reports a relationship with Abbvie GK that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with Otsuka Pharmaceutical Co Ltd that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with Novartis Pharma Kabushiki Kaisha that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with Eisai Co Ltd that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with DAIICHI SANKYO COMPANY, LIMITED that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with FP Pharmaceutical Corporation that includes: speaking and lecture fees.Nobutaka Hattori reports a relationship with The Michael J Fox Foundation that includes: funding grants.Nobutaka Hattori reports a relationship with International Parkinson and Movement Disorder Society that includes: funding grants.If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.