Introduction

Parkinson’s disease (PD) is a common progressive neurodegenerative disorder, affecting 1–2% of people older than 65 years [1]. The pathogenesis of PD has not been fully elucidated, but genetic factors play an important role in PD [2, 3]. Leucine-rich repeat kinase 2 (LRRK2) variants are the most common genetic causes of familial late-onset and sporadic PD [4]. The frequency of LRRK2 mutation carriers in PD patients is greatly variable in different populations. LRRK2 G2019S is typically found in North Africa and South European populations, whereas it is rare in Asians [5]. LRRK2 G2385R and R1628P are two common Asian-specific variants with twofold increased risk for sporadic PD in Asian population [6,7,8]. Besides, LRRK2 mutation carriers are reported to show incomplete penetrance and a highly variability of age at onset (AAO) of motor symptoms [9, 10]. Therefore, other genetic or environmental modifiers are probably regulating the AAO in LRRK2-related PD patients [11, 12].

A recent genome-wide linkage and association study (GWAS) identified rs2421947 (C/G) in dynamin 3 (DNM3) gene-modified AAO of LRRK2 G2019S carriers in Tunisian Arab-Berber population, and the median AAO of GG carriers of rs2421947 was 12.5 years earlier than CC carriers [13]. Afterwards, another study reported no relation between rs2421947 and AAO of LRRK2 G2019S carriers in Spain population [14]. A recent research showed that rs2421947 was not associated with AAO of LRRK2-PD in Asian population [15]. Nevertheless, these findings have not been confirmed in other Asian populations.

Variants in the α-synuclein (SNCA) gene have been associated with an increased risk of PD [14]. Besides, rs356219 (A/G) in SNCA has also been reported to regulate the AAO of LRRK2-related PD in European populations [14, 16]. Cyclin G-associated kinase (GAK) gene has been reported to modify the alpha-synuclein expression and toxicity, and GAK rs1564282 was associated with an increased PD risk with an interaction with SNCA rs356219 [17, 18]. In this study, we aimed to investigate whether rs2421947, rs356219, and rs1564282 modulate the AAO of LRRK2-related PD patients in Chinese population.

Methods

Study participants

The study enrolled a large cohort of Han Chinese population from the first affiliated hospital of Zhengzhou University, including 732 sporadic PD patients and 1992 healthy controls. All the patients were diagnosed independently by two neurologists based on the United Kingdom Parkinson’s Disease Society Brain Bank [19]. Each patient was assessed by detailed neuropsychological evaluations, and each control was checked by two experienced neurologists to exclude the abnormal neurological examination. Age of PD onset was defined as the age of appearance of first symptoms that were self-reported by each patient. The study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University. Written informed consent was obtained from each participant.

Genetic analysis

Genomic DNA was extracted from peripheral blood according to the standard protocols. The genotypes of LRRK2 G2385R and R1628P, DNM3 rs2421947, and DNM3 rs2421947 were performed by Sanger sequencing. The genotyping of DNM3 rs2421947 was performed using 5′-TCCTGCTGAACGACTAAGGT-3′ as the forward primer and 5′-CTCTCAGTCACGTTTTGCTACA-3′ as the reverse primer. The genotyping of GAK rs1564282 was performed using 5′-TTCCCTCTTGTGGAACTGCT-3′ as the forward primer and 5′-GGTGGATACAGGGCTGTCAGT-3′ as the reverse primer. The genotyping of LRRK2 G2385R and R1628P variants, and SNCA rs356219 were also performed using the primers reported [20, 21].

Statistical analysis

Hardy–Weinberg equilibrium for each variant among cases and controls, and Linkage disequilibrium analysis was analyzed by SNPstat software [22]. Differences in allele and genotype distributions were analyzed using chi-square test to calculate the frequency significance and the odds ratio (OR). Age at examination, age at onset, and disease duration were assessed using the two-tailed Student’s t test. A two-tailed P value < 0.05 was considered statistically significant. The statistical analysis was performed using SPSS version 20.0 (IBM, Armonk, NY, USA).

Results

Demographic characteristics and LRRK2 variant frequencies in the PD cases and healthy controls are summarized in Table 1. 121 of the 732 PD patients carried at least one LRRK2 variant (82 patients carried G2385R, 46 patients carried 1628P, and 8 patients carried the both variants). 145 of the 1992 healthy controls carried at least one LRRK2 variant (80 controls carried G2385R, 68 controls carried R1628P, and 3 controls carried the both variants). Each variant was in Hardy–Weinberg equilibrium in the control subjects (P value cut-off = 0.01). There was no linkage disequilibrium between SNCA rs356219,DNM3 rs2421947, and GAK rs1564282.

Table 1 Demographic characteristics of the participants
Full size table

Association between rs2421947/rs356219/rs1564282 and LRRK2-related Parkinson’s disease are analyzed in Table 2. The SNCA rs356219-G allele was significantly different between LRRK2-related PD cases and healthy controls with LRRK2 carriers (OR 1.50, 95% CI 1.08–2.01, P = 0.016). The individuals with AG + GG genotypes have an increased risk compared to those with AA genotype (OR 2.27, 95% CI 1.25–4.12, P = 0.006). Nonetheless, no similar significance was observed in the DNM3 rs2421947, and GAK rs1564282.

Table 2 Association between rs2421947/rs356219/rs1524282 and LRRK2-related Parkinson’s disease
Full size table

The clinical characteristics of LRRK2-related PD patients between different genotypes are also analyzed in Table 3. The LRRK2-related PD patients with SNCA rs356219 AG + GG genotypes were associated with earlier age at onset compared with AA genotype (mean AAO of AG + GG: 58 years old; mean AAO of AA: 62 years old), but PD onset symptoms among the AG + GG and AA genotypes of SNCA rs356219 were not significantly different. No similar significance was observed in the DNM3 rs2421947 and GAK rs1564282.

Table 3 Clinical characteristics of LRRK2-related Parkinson’s disease between different genotypes of DNM3 rs2421947 and SNCA rs356219
Full size table

Discussion

LRRK2 mutations are associated with autosomal dominant PD with incomplete penetrance, and the penetrance varies among different variants. The heterogeneity has led to the hypothesis that some modifications influence the penetrance of LRRK2 [9,10,11]. In this study, we found the mean AAO of GG + GA genotype of SNCA rs356219 was 4 years earlier than AA genotype in LRRK2-related PD patients.

Recently, DNM3 rs2421947 has been identified associated with the AAO of LRRK2 G2019S PD in the Arab-Berber population [13]. In addition, there seemed to be a higher level of expression of dynamin 3 protein in GG carriers than that in CC carriers, and the localization of Dnm3 was affected significantly in the LRRK2 G2019S knock-in mice [13]. However, our study showed no association of rs2421947 and AAO in LRRK2-related PD patients. Though the association of DNM3 with PD onset was confirmed in the meta-population with HR = 1.61, the France population showed no association with HR = 0.71 and the Norway population showed slightly association with HR = 1.17 [13]. Besides, a Spanish study also reported no association of rs2421947 and LRRK2 G2019S PD onset [14]. Our results were consistent with the analysis in the France and Spanish populations.

These findings are conflicting due to several factors. One possible explanation is that the molecule pathogenesis may be different between LRRK2 G2019S and G2385R/R1628P variants. G2019S variant is located in the kinase domain and increases the kinase activity, while G2385R is resided in the WD40 domain, and decreases the kinase activity and increases GTPase activity [23]. It was reported that G2385R was associated with a greater risk of apoptosis and cell death [6]. R1628P is resided in the COR domain, and it can increase indirectly kinase activities by enhancing the cyclin-dependent kinase 5(Cdk5) phosphorylation of LRRK2 and leading to neuronal death [24]. It may result in different interaction effects with DNM3. Another reason is that single variant is subtle, and additional genetic modifiers and/or environment modifiers could also contribute to the penetrance of LRRK2 [11].

SNCA is a well-established causative gene for PD, and α-synuclein has been recognized as a critical protein in the pathogenesis of PD [25,26,27]. Previous studies have demonstrated that rs356219 in the 3'UTR of SNCA increased the risk of sporadic PD [16]. It influenced the AAO of LRRK2-related PD in European population, with an earlier 7–11 years mean AAO for the GG genotype. In our study, we found a similar result that the mean AAO of GG + AG carriers of SNCA rs356219 was 4 years earlier than AA carriers (62 years for AA, and 58 years for GG + AG). These studies showed that rs356219 in SNCA could be a penetrance modifier for LRRK2-related PD, alone or in combination with additional risk factors. Previous studies have demonstrated that rs356219 in SNCA was related with increased α-synuclein expression in blood, cerebellum, and substantia nigra of PD patients [28, 29]. Further functional study is essential to elucidate its pathophysiologic role in LRRK2 carries. Though GAK rs1564282 was reported to be associated with an increased PD risk, and interact with SNCA, our study showed no evidence of GAK on AAO of LRRK2-related PD.

Although the precise physiological mechanism of LRRK2 underling PD remains largely unclear, many studies have showed that LRRK2 was associated with vesicle trafficking, protein degradation, cytoskeletal maintenance, and neurite morphology [30,31,32]. The Asian-specific LRRK2 variants, G2385R and R1628P, were associated with increased risk of PD. Under the condition of certain situation, such as oxidative stress, the LRRK2 variants were more toxic and associated with a higher rate of apoptosis [6, 33]. The potential genetic modifiers of AAO of LRRK2-related PD may provide insight into the disease therapeutics by delaying the PD onset. However, there are limitations in this study. Potential gene–gene and gene–environment interactions were not evaluated, and further functional studies are essential to elucidate the roles of these variants in LRRK2 carries.

In conclusion, our study suggested that SNCA rs356219 but not DNM3 rs2421947 or GAK rs1564282 modulated the AAO of LRRK2-related PD in Han Chinese population. Given the limited number size of this study, further studies in larger scale populations will be required to validate our findings.