Patient‐control association study of the Leucine‐Rich repeat kinase 2 (LRRK2) gene in South African Parkinson's disease patients

Abstract

The leucine-rich repeat kinase 2 (LRRK2) gene is of interest to Parkinson's disease (PD) as it has been implicated in both familial and sporadic forms of the disorder.1 PD-susceptibility alleles in LRRK2 appear to be ethnic-specific with G2385R,2 R1628P3 and A419V4 identified in Asian populations, whereas M1646T is found in Caucasians.4 A haplotype protecting against development of PD is present in Chinese (N551K-R1398H)5 and Caucasians (N551K-R1398H-K1423K).4 Further studies are necessary to investigate the contribution of LRRK2 to PD-susceptibility in various populations worldwide. To this end we investigated whether variants in LRRK2 were associated with PD in a South African patient series comprising 205 PD patients and 378 controls of different ethnicities: Caucasian, Mixed ancestry, Xhosa-speaking Black African and Indian/Asian (Supplementary Table 1). For the purposes of our study the Afrikaner Caucasian individuals, hereafter referred to as Afrikaner, were analyzed separately from the ‘non-Afrikaner’ Caucasians. All LRRK2 exonic variants, published or reported up to April 1, 20104 were genotyped using the MassArray iPLEX platform (Sequenom, San Diego, CA, USA). Statistical analyses were performed using R (www.r-project.org) and R package haplo.stats. Logistic regression was used to assess individual single nucleotide polymorphisms (SNPs) and haplotype associations with PD. With group sizes of 64 patients and 93 controls (similar to our Afrikaner group), a significance level of 5%, and assuming a control frequency of 5%, we had 80% power to detect an additive allelic odds ratio of 3.1. Of the 117 variants genotyped, 30 were polymorphic in at least one ethnic group. All variants were in Hardy-Weinberg equilibrium. In this exploratory analysis a number of novel associations with PD were found (Table 1; Supplementary Table 2), although an association with a variant common to all ethnic groups was not detected. The M1646T variant was not present in the Black African individuals. Furthermore, this variant was not associated with PD in any of the other ethnic groups; this may be related to small sample sizes or possibly due to differences in genetic substructure (Supplementary Fig. 1). The previously-identified protective haplotype (N551K-R1398H-K1423K) did not show a significant association with PD. However, of interest is the fact that greater diversity in the haplotype structure was observed in the Black African and Mixed ancestry individuals (five haplotypes) than the Caucasians (two haplotypes) (Supplementary Table 3) which is important for future association studies. Table 1 Case-control association results for the LRRK2 variants that showed evidence for association in at least one South African ethnic group. Previous mutation-screening studies on LRRK2 in African populations found that upwards of 30% of PD patients in North African Berber Arabs harbor the pathogenic G2019S mutation. In contrast, the present study found G2019S to be relatively uncommon in the South African population (4/205, 2%) reflecting the fact that extensive genetic diversity across different African populations exists.7 Taken together, our findings further support the idea that genetic risk factors in LRRK2 for PD are ethnic-specific. While it is acknowledged that the group sizes are small, this study is of interest as it is the first case-control association study on LRRK2 in a sub-Saharan African population. It would be important for this work to be duplicated in diverse populations to see how the results compare and contrast.