Genetic study of the NUS1 gene variants in Han Chinese patients with Parkinson’s disease

INTRODUCTION

Parkinson’s disease (PD) is the second most prevalent progressive neurodegenerative condition after Alzheimer’s disease^[1,2]. The globally pooled all-age prevalence of PD is 0.151%, which has been on the rise, especially over the past two decades, with the prevalence higher in males (0.154%) than females (0.149%), increasing with age^[3]. According to the Movement Disorder Society Clinical Diagnostic Criteria for PD (MDS-PD Criteria) in 2015, it centers on motor syndrome, including the cardinal manifestations of bradykinesia, rest tremor, and rigidity^[4]. This heterogeneous neurological disorder, which is pathologically characterized by aberrant deposition of α-synuclein and death of dopaminergic neurons in the substantia nigra, as well as nondopaminergic changes, presents various motor and non-motor symptoms^[5-7]. It is tied to a complex interplay of factors including genetics, aging, and environmental risk^[8,9], among which genetic factors are responsible for ~30% of PD cases, with 5%-10% following a monogenic inheritance pattern^[9,10-12].

Genome-wide association studies (GWAS) have provided new insights into the pathophysiology of PD, as over 90 independent significant risk variants have been revealed to be correlated with PD^[13,14], and next-generation sequencing approaches have further facilitated the identification of more PD-related gene variants^[15]. Beyond pathogenic variants in commonly implicated PD genes - such as alpha-synuclein (SNCA), leucine-rich repeat kinase 2 (LRRK2), parkin RBR E3 ubiquitin protein ligase (PRKN), PTEN-induced kinase 1 (PINK1), and parkinsonism-associated deglycase (PARK7, also known as DJ-1) - which typically exhibit with Mendelian inheritance patterns, several genetic variants with either susceptibility or protective roles have also been identified. These include variants in genes such as glucosylceramidase beta 1 (GBA1), microtubule-associated protein tau (MAPT), and DnaJ heat shock protein family (Hsp40) member C10 (DNAJC10), all of which are implicated in PD pathogenesis^[16-18].

The NUS1 dehydrodolichyl diphosphate synthase subunit gene (NUS1) was first identified as a candidate gene for PD through trio or quad whole-exome sequencing (WES), independent case-control analysis, and Drosophila study^[19]. Several subsequent studies have revealed rare or low-frequency NUS1 gene variants contributing to PD or have reported controversial conclusions^[20-25], indicating the veiled effects of the NUS1 gene variants in PD.

In this study, we aimed to screen the NUS1 gene variants in a PD-control cohort and evaluate the potential impacts on PD in the Han Chinese population, which may provide new insights into the NUS1-related genetic contribution to PD etiology.

METHODS

RESULTS

A total of three NUS1 (NM_138459.5) heterozygous missense variants, including c.127G>T (p.Ala43Ser, rs1327892878), c.487G>C (p.Asp163His, rs369403261), and c.537T>A (p.Asp179Glu, rs28362519), were revealed by WES and confirmed by Sanger sequencing in the cohort [Figure 1A]. All three variants were detected in sporadic PD cases, with one also identified in controls. The p.Ala43Ser variant was found in two PD patients with onset age over 50 years, and the p.Asp163His variant was discovered in a 73-year-old female patient suffering from bradykinesia, rigidity, and postural instability [Table 1]. The p.Asp179Glu variant was detected in 7 PD patients and 9 controls. Two variants p.Ala43Ser and p.Asp163His with a MAF < 0.001 or absent in East Asians according to 1000G, ExAC, and gnomAD were considered rare variants, whereas p.Asp179Glu with a MAF < 0.01 was categorized as a low-frequency variant [Supplementary Table 1].

Figure 1. The three identified NUS1 heterozygous missense variants, c.127G>T (p.Ala43Ser), c.487G>C (p.Asp163His), and c.537T>A (p.Asp179Glu). (A) The Sanger sequencing results of three missense variants, as well as a synonymous variant (c.126C>A, p.Arg42=, rs947563301) co-occurring with c.127G>T; (B) Sequence alignment of the NUS1-encoded NgBR protein in different species, with arrows pointing to the amino acids affected by the variants; (C) The locations of three variants in the pattern diagram and topological model of NgBR protein, mainly localized to the endoplasmic reticulum. NUS1: the NUS1 dehydrodolichyl diphosphate synthase subunit gene; TM: transmembrane domain; aa: amino acid; cis-PTase homology: cis-prenyltransferase homology domain; ER: endoplasmic reticulum; NgBR: neurite outgrowth inhibitor B receptor.

The variants, p.Ala43Ser and p.Asp163His, were predicted to be conflicting, in which over half of the pathogenicity prediction tools (4/7) predicted p.Asp163His as pathogenic, while all the tools predicted p.Asp179Glu as benign. Two missense variants, p.Ala43Ser and p.Asp163His, were predicted to reduce the protein stability by SAAFEC-SEQ or MUpro, while p.Asp179Glu was predicted to reduce the stability only by MUpro [Supplementary Table 1]. The amino acid residues, p.Ala43, p.Asp163, and p.Asp179, are highly conserved in the nine species from reptiles to mammals, except in lower organisms below zebrafish [Figure 1B].

No statistically significant difference was found in terms of gender or age between patients and controls (both P > 0.05). The Hardy-Weinberg equilibrium test indicated no deviation in the controls (all P > 0.05). Absent in controls, two potential PD-associated variants, p.Ala43Ser and p.Asp163His, had the genotype frequency of 0.39% and 0.20% in the case population, respectively. No statistically significant differences in genotypic distributions or allele frequencies between 512 PD patients and 516 controls (all P > 0.05) were revealed in three variants [Supplementary Table 2]. Further gene-based SKAT-O analysis showed no significant association between the identified rare variant and the PD phenotype (P = 0.1494 for p.Ala43Ser and P = 0.2893 for p.Asp163His), even combined (P = 0.1259 for two rare variants p.Ala43Ser and p.Asp163His, and P = 0.5784 for all three variants).

DISCUSSION

PD is a common complex neurodegenerative disease determined by monogenic variants or related to multiple factors^[36,37]. Since the watershed in 1997, when a pathogenic variant of the SNCA gene, known as the first PD-causative gene, was identified in autosomal dominant PD families^[38], a large number of genetic loci have been implicated in PD^[39]. To date, at least 26 loci and 22 genes have been definitively identified as responsible for monogenic PD^[39-41], although they account for less than 10% of cases^[42,43]. Additionally, variants in many susceptibility genes have been reported to increase the risk of PD development, recognized as common risk factors^[44-46]. In recent years, a research group has reported that the NUS1 gene variants are associated with PD mainly in Chinese^[19-21]. In contrast, other studies have suggested a lack of association between the NUS1 gene variants and PD in case-control cohorts of Chinese or European descent^[23-25].

The NUS1 gene, located on chromosome 6q22.1, contains 5 exons and encodes the transmembrane protein, neurite outgrowth inhibitor B (Nogo-B) receptor (NgBR), which is a subunit of cis-prenyltransferase (cis-PTase)^[47-49]. The NgBR includes two types of domains, the N-terminal transmembrane domain and the cis-PTase homology domain^[50]. It is mainly distributed in the endoplasmic reticulum (ER) and the cell membrane, with a relatively high expression in nerve cells and immune cells (http://biogps.org/). Its stable binding to dehydrodolichyl diphosphate synthase (DHDDS) stimulates the full activity of cis-PTase, acting in dolichol synthesis and protein glycosylation in ER^[51]. The N-terminal transmembrane domains, TM1 (transmembrane domain 1) and TM3 (transmembrane domain 3), determine the orientation of the C-terminal of NgBR, and the localization of the C-terminal is related to the NgBR function. When the C-terminal is oriented toward the ER lumen, as the minor proportion, it interacts with Niemann-Pick type C2 (NPC2) protein, stabilizes the nascent NPC2, and regulates intracellular cholesterol transport^[52,53]. When the C-terminal of NgBR is oriented toward the cytosol, as the major proportion, it interacts with and stabilizes DHDDS, participating in dolichol synthesis, and affecting the protein glycosylation^[50]. The NUS1 homozygous loss-of-function variant was identified to cause congenital disorder of glycosylation type Iaa (CDG1AA, OMIM 617082)^[54], while de novo heterozygous variants were shown to be associated with autosomal dominant intellectual developmental disorder-55 with seizures (MRD55, OMIM 617831)^[55]. Lower serum soluble level of Nogo-B, which may function via the Nogo-B/NgBR axis, in PD patients with poor motor function and worse disease progression, along with the observation of cholesterol accumulation and PD-related neurodegeneration features in Drosophila with NgBR ortholog loss, supported the potential role of NUS1 in the pathogenesis of PD^[52,56].

In this study, we identified three NUS1 heterozygous missense variants in our cohort, of which two variants, c.127G>T (p.Ala43Ser) and c.487G>C (p.Asp163His), were found only in patients, and c.537T>A (p.Asp179Glu) was found to have similar frequencies in PD patients and controls (7/512 vs. 9/516). The c.127G>T variant resulted in the substitution of non-polar hydrophobic alanine by polar hydrophilic serine (p.Ala43Ser). The other two variants, c.487G>C and c.537T>A, were predicted to affect amino acids in the cis-PTase homology domain, resulting in the conversion of acidic aspartic acid to basic histidine at position 163 (p.Asp163His), and the acidic residue substitution at position 179 (from aspartic acid to glutamic acid, p.Asp179Glu). The similar low frequency of p.Asp179Glu variant in our patients and controls, combined with the benign outcomes of pathogenicity predictive tools, suggested that it was not associated with PD, consistent with the findings and conclusions of previous studies^[23,24].

However, we suspected that p.Asp163His and p.Ala43Ser may pose a risk for PD in the population, which is supported by the following evidence: (i) Some de novo heterozygous variants in the NUS1 gene were disease-causing variants for MRD55^[55]. (ii) To date, no co-segregated disease-causing NUS1 variant has been found in a large family with PD^[19]. (iii) These two variants had low frequencies in the population and their bioinformatics predictions were conflicting, especially for the p.Ala43Ser variant. (iv) These two variants were identified in sporadic PD cases rather than familial cases. (v) The amino acids are highly conserved in multi-species. (vi) Particularly, the TM2 domain that the p.Ala43Ser affects may not be a determiner in the orientation of the C-terminal like TM1 and TM3 [Figure 1C]. Burden analysis of rare variants including p.Asp163His was reported to be deleterious and have an association with PD^[19-21]. Although statistical analysis showed no associations between the variants (p.Ala43Ser and p.Asp163His) and PD phenotype, combined with all the findings, we could not fully exclude a possible role in PD. Moreover, PD is a typical age-related neurological disease. These three patients were all over 50 years at age onset and had no positive family history, which suggested that the NUS1 gene variants may play a role in increasing susceptibility in PD at most, rather than cause PD in a monogenic inheritance pattern with low penetrance.

Furthermore, even though p.Ala43Ser and p.Asp163His may potentially contribute to PD susceptibility, the NUS1 variant(s) may affect less than 0.6% (3/512) of Han Chinese patients with PD. More studies with large-scale samples, as well as high-throughput sequencing approaches and valid bioinformatics tools, are required to verify the role of NUS1 gene variants in PD susceptibility. In conjunction with further screening of cases with neurological disorders, including CDG1AA and MRD55, elucidating the role of the NUS1 gene in neurological diseases, particularly neurodevelopmental conditions, will become touchable.

In summary, our study indicated that the NUS1 gene variants seem to not cause monogenic PD, which may exert a susceptibility to PD at most. The findings enhanced the understanding of the real impact of the NUS1 gene variants on monogenic PD by the limited samples in this study.