New endemic familial parkinsonism in south Moravia, Czech Republic and its genetical background

Abstract An increased prevalence of familial neurodegenerative parkinsonism or cognitive deterioration was recently found in a small region of southeastern Moravia. The aim of the study was to assess the genetic background of this familial disease. Variants in the ADH1C, EIF4G1, FBXO7, GBA + GBAP1, GIGYF2, HTRA2, LRRK2, MAPT, PRKN, DJ-1, PINK1, PLA2G6, SNCA, UCHL1, VPS35 genes were examined in 12 clinically positive probands of the pedigree in which familial atypical neurodegenerative parkinsonism was identified in previous epidemiological studies. Libraries were sequenced by massive parallel sequencing (MPS) on the Personal Genome Machine (PGM; Ion Torrent). Data were analyzed using Torrent Suite and IonReporter software. All variants were then verified and confirmed by Sanger sequencing. We identified 31 rare heterozygous variants: 11 missense variants, 3 synonymous variants, 8 variants in the UTR region, and 9 intronic variants. Six variants (rs1801334, rs33995883, rs35507033, rs781737269, rs779760087, and rs63750072) were evaluated as pathogenic by at least one in-silico predictor. No single “founder” pathogenic variant associated with parkinsonism has been found in any of the probands from researched pedigree. It may rather be assumed that the familial occurrence of this disease is caused by the combined influence of several “small-effect” genetic variants that accumulate in the population with long-lasting inbreeding behavior.


Introduction
The results of epidemiological studies indicate that the prevalence of Parkinson disease (PD) and neurodegenerative parkinsonism in the general population is around 1.6% in Europe and approximately 1.5% in Asia and the Americas. [1][2][3][4][5][6][7][8] In the population over 65 years of age, the prevalence of these diseases ranges from 1.1% to 2.2% [1][2][3][4][5][6][7][8][9][10][11] and even higher prevalences have been repeatedly found in small, isolated populations. [3,4,6,9] We recently described such a phenomenon in 10 villages of a remote, small, rural, and isolated Moravian region of Czech Republic called "Hornacko" [Upper Lands]. The last census, conducted in 2011, determined the presence of 8664 inhabitants, 2927 of which were over 50 years of age. "Hornacko" people have their own specific customs and traditions, which are reflected in the folk art, crafts, architecture, costumes, songs, and dances; also, the local dialect is very different from that of the surrounding areas ( Fig. 1). This area is traditionally isolated from the neighboring Moravian regions and the people only rarely migrate permanently out of here. [12,13] In 2011 to 2015, we conducted an epidemiological study in this region which has found a significantly increased prevalence of neurodegenerative parkinsonism. The overall prevalence in the population over 50 years of age was 2.8% (95% confidence interval (CI), 2.2-3.4); the prevalence in the population between 50 and 64 years of age was 1.9% (95% CI, 1.2-2.5) and it was 4.06% (95% CI, 2.9-5.1) in the population over 65 years of age. [14,15] Three large pedigrees with a familial occurrence of parkinsonism were found in which clinical geneticists identified an autosomal-dominant inheritance pattern. [16] Two of these pedigrees eventually compiled one large family tree spanning several generations from 1840 to the present. [17] In this large pedigree, 12 probands with apparent clinical signs of parkinsonism were identified (Fig. 2). These probands were referred to a tertiary movement disorders center for detailed clinical and laboratory examinations including DNA sampling in order to obtain information about the clinical, biological, and genetic characteristics of this endemic and familial disorder. The village from which these families originate (Javorník nad Velicǩou, PCN CZ-69674, elevation 412 m, population 720) is very specific even in this region: the population belongs almost entirely to the Lutheran denomination, unlike all neighboring villages, which are entirely Roman-Catholic. Marriages between the inhabitants of Javorník and residents of other villages were and still are quite exceptional, so the local population is practically isolated from a genetic point of view. [18]

Methods
The study was approved by the ethics committee of Palack y University Medical School, and all patients and probands gave their informed consent.
All 12 probands who were evaluated as positive for parkinsonian symptoms were examined by a movement disorders specialist. The United Kingdom Parkinson's Disease society Brain Bank clinical diagnostic criteria clinical diagnostic criteria were used for the diagnosis of parkinsonism; this was diagnosed when at least 2 of 4 Class I symptoms (resting tremor, rigidity, bradykinesia, and impaired postural reflexes) were present in a subject not receiving antiparkinsonian medication. Cognitive functions were assessed using the Mini Mental State Examination (MMSE) tool; cognitive deterioration was diagnosed when the MMSE score was 24. All positive probands were then examined in detail to exclude secondary forms of parkinsonism. The examination protocol included detailed neurological, psychological, and psychiatric examinations, examination of cerebrospinal fluid, magnetic resonance imaging of the brain; neurophysiological examinations were also performed (electroencephalographic recordings, electromyography, and assessments of visual, auditory, somatosensory, and motor evoked potentials). [16,17] The clinical details of all probands are presented in Table 1.
DNA was isolated from peripheral blood by a DNA silica-gel membrane based isolation kit (QIAamp DNA Mini Kit -Qiagene; Strasse 1, 40724 Hilden, Germany). Each DNA sample was

Discussion
Some past studies have suggested that people living in rural areas could be more exposed to putative environmental, mainly agricultural, influences and would be more likely to develop parkinsonism than those living in urban areas. [19] However, the area surveyed in our study is surrounded by a much larger rural area, in which the agricultural and environmental conditions are quite similar; to date, no signs of higher parkinsonism incidence in this (or any other) part of Central Europe have been reported. A genetic origin of familial parkinsonism thus seems to be the more probable explanation of presence in this population.
The molecular-genetic DNA analysis performed in 12 clinically positive probands in our pedigree has found several new variants in the loci associated with familial parkinsonism in the past:  The missense substitution was evaluated as pathogenic using prediction programs (in SIFT pathogenic; in PolyPhen-2 probably pathogenic). It leads to an exchange of acidic for uncharged polar amino acid. Rare occurrence and also a higher degree of evolutionary conservation of the site has been reported. Moura et al [20] identified the variant in 9 (6.8%) of his patients; it was evaluated as a nonpathogenic polymorphism. Lücking et al [21] reported a statistically significant difference between patients and controls: the variant was found in 10/194 patients and in 12/125 controls. In another recent study, this variant was found in similar proportions in both patients and controls. [22] 4.2. NM_198578.3: c.6241A>G LRRK2(PARK8) The substitution is located in an evolutionarily conserved region, which could, according to Polyphen-2, have a pathogenic character. However, SIFT evaluated this variant as benign. Its benign role is supported by the recent finding of Benitez et al, [23] who found this variant in 17 of 478 PD patients and 15 of 337 controls. In another study, Foo et al [24] found the variant in none of the patients and in 1 control (0.25%).

NM_198578.3: c.4541G>A LRRK2 (PARK8)
This missense variant in the coding region leads to the exchange of alkaline amino acid for polar uncharged in the phylogenetically conserved region. According to SIFT, it was evaluated as benign; according to PolyPhen-2, it was evaluated as probably pathogenic. Because of its rare occurrence in the human population and the high phylogenetic conservation of the region, it presumably influences the protein function. On the other hand, Nichols et al [25] described this variant as probably nonpathogenic. Toft et al [26] found the variant in 1 large kindred but they did not confirm its full segregation with parkinsonism. An analysis of the case-control series did not confirm an increased risk of parkinsonism with this variant, and no association with PD has been reported.

NM_198578.3: c.1898C>T LRRK2 (PARK8)
This is a missense variant in a coding conserved region, and it leads to the exchange of uncharged polar for nonpolar amino acid. It was evaluated as pathogenic (SIFT and PolyPhen-2). A pathogenic character can thus be expected, but no association with PD has been reported so far.

NM_001123066.3: c.689A>G MAPT (TAU)
This is a missense variant which leads to the substitution of uncharged polar amino acid for alkaline. The site is evolutionarily conserved. The variant was predicted as pathogenic. Jin et al [27] studied variants of MAPT associated with Alzheimer disease in the Spanish population. The variant was identified in 18 patients (10% probands) and 8 controls (5.8%). An Table 2 Genetic variants identified in probands with clinical signs of parkinsonism. association between the variant and the level of tau protein was described in the cerebrospinal fluid. [28] The variant has not yet been functionally analyzed. The detailed study of the presence of mutual variants in the researched pedigree indicates that the genetic background of this endemic neurodegenerative disease has a clearly heterogenous character: 31 different variants were found in 12 genes. When we tracked the variants within the pedigree, we found a dominant trait in only 3 branches of the family and only for some variants. In patients 3 and 11, we found similar variants of the EIF4G1 gene. In contrast, LRRK2 variants were present in the maternal proband only. In patients 6 and 12, we found similar variants of ADH1C, LRRK2, and PRKN; MAPT variants were present in the maternal proband only. In patients 2 and 10, similar variants of PRKN, PLA2G6, VPS35, and MAPT were found. Variants of EIF4G1 and LRRK2 were present in the maternal proband only, while the variants of PINK1, DJ-1, and GIGYF2 were found only in her daughter. In other pedigree branches, we did not find any similarity in the variant presence in consanguineous relatives. Though many other genes could potentially be associated with PD; we focused on the most common candidate ones.

Conclusion
Our study has provided an evidence that the recently described endemic neurodegenerative parkinsonism, which can manifest in different phenotypes (see Table 1), has also different genetic background and apparently heterogenous traits. In this moleculargenetic study of the core pedigree, we have not found one "founder" pathogenic variant associated with parkinsonism in all or the most PD patients of this pedigree. Therefore, it could rather be assumed that the familial occurrence of this disease is caused by the combined influence of several "small-effect" genetic variants that accumulate in the population with long-lasting inbreeding behavior. Whether there is any link to the VPS35 positive familial parkinsonism, which has been discovered in the relatively near region of Upper Austria by Zimprich et al [29] and Struhal et al [30] remains open for further research. For the planned functional analyses, we have so far collected only one brain specimen from the research pedigreee. [31] Therefore, our future study will also utilize proper cell lines with combinations of targeted mutagenesis to assess the effects of particular rare variants.