• Parkinson’s disease
• paraoxonase
• epidemiology

Parkinson’s disease (PD) is thought to be caused by a combination of genetic and environmental factors. Epidemiological studies have found associations of PD with pesticide exposure, or suspected pathways of pesticide exposure, such as rural residence and well water consumption. Many organophosphorous insecticides (for example, chlorpyrifos and diazinon) are bioactivated to potent cholinesterase inhibitors by the cytochromes P450, and the resulting toxic oxon forms are hydrolysed by paraoxonase (PON1). Genetic variants of detoxifying enzymes or pesticide metabolising enzymes, such as paraoxonase, may confer a predisposition to PD and thus are considered candidate genes for association studies.

Multiple polymorphisms have been identified in the PON1 gene. The coding region contains two common polymorphisms, at amino acid codons 55 and 192. The glutamine (Q) to arginine (R) substitution at codon 192 causes substrate dependent differences in the kinetics of hydrolysis: compared with the PON1 192Q isoform, PON1 192R has higher activity towards paraoxon and chlorpyrifos oxon, but lower activity towards diazoxon, soman, and sarin. The leucine (L) to methionine (M) substitution at codon 55 does not affect the catalytic efficiency of substrate hydrolysis by the enzyme, but the PON1 55M allele is correlated with decreased mRNA and protein levels, because of linkage disequilibrium with a single nucleotide polymorphism (SNP) at position -108 of the promoter region of the gene. Five SNPs have been identified in the promoter region, at positions -108, -126, -162, -832, and -909. The -108 SNP has been shown to have the greatest effect on arylesterase activity, accounting for about 22% of the total variance, followed by the polymorphisms at positions 192, 55, and -162, which account for about 5.7 %, 4.1%, and 1.1% of the total variance in arylesterase activity, respectively. Cell culture studies indicate an approximately twofold change in PON1 gene transcription attributable to the -108C/T and -162G/A SNPs, with the -108C and -162A providing more efficient transcription.

A significant association of the 192R allele with PD was found in a population of patients of Japanese ethnicity with comparatively low mean age of onset.¹ In contrast, no difference in genotypes distributions was found between PD cases and controls for the PON1 Q192R polymorphism in an Australian study² nor in another study on subjects of Russian ethnicity.³ In a more recent study by Akhmedova et al⁴ on the same Russian population, the M55 allele was found to be associated with PD. No associations for the amino acid codon 55 and 192 polymorphisms were found in a study from China.⁵

In this study, we examined associations of two promoter (G-162A and C-108T) and two coding region (M55L and Q192R) polymorphisms in PON1 with PD. Newly diagnosed idiopathic PD patients (n=150; 91 men and 59 women), aged 37 to 88 years, were identified by neurology and general medical practice clinics of the group health cooperative (GHC) from the Puget Sound area in western Washington State. Inclusion criteria for the cases were the presence of at least two of the four cardinal signs of PD: bradykinesia, resting tremor, cogwheel rigidity, and postural reflex impairment. Exclusion criteria included the use of certain drugs during the 12 months preceding symptom onset, history of multiple cerebrovascular events, or another explanation for parkinsonism symptoms.

Controls (n=244; 158 men and 86 women), aged 44 to 84 years, were identified from GHC enrollees without past histories of PD or other neurodegenerative disorders. Controls were matched to cases by birth decade, sex, and year of enrolment in GHC. All subjects were of non-Hispanic white ethnicity. Study subjects were volunteers who were informed of the purpose of the study. Study forms and procedures were approved by the Institutional Review Board committees on Human Subjects Research at the University of Washington and GHC Center for Health Studies.

A PCR/dye terminator cycle sequencing based assay was used to detect the -162 G/A and -108 C/T genetic variants within the paraoxonase 1 (PON1) gene. TaqMan Detection System based assays were developed to identify the PON1 55 T/A and PON1 192 A/G variants. Odds ratios and χ² tests were calculated using of SPSS software for Windows; α=0.05 was taken as the level of significance. Logistic regression models were used to calculate adjusted odds ratios and to test for statistical significance of interactions. Haplotypes were inferred using EH software.

Among controls, we observed the following allelic frequencies: -162 A = 0.23, G = 0.77; -108 T = 0.46, C = 0.54; 55M = 0.35, L = 0.65; 192R = 0.30, Q = 0.70. As shown in table 1, there were no significant differences in the genotype distributions of cases and controls for any of the four PON1 polymorphisms. The distribution of three marker haplotypes was not significantly different between cases and controls (χ²₍₇₎=4.78, p=0.69). There also was no difference in the distribution of four marker (including -162) haplotypes between cases and controls (χ²₍₁₅₎=9.98, p=0.82).

We also tested for interactions between PON1 genotypes and age (<60, =60), sex, and smoking. No conclusive evidence of interaction between any PON1 genotype and either smoking or sex was found. However, we did detect an interaction between PON1 192 genotype and age. Interestingly, among cases, PON1 192 QQ genotype frequency increased with age (χ²₍₄₎ for genotype distribution =3.9×10⁻⁵), whereas among controls, 192 QQ genotype frequency decreased with age (χ²₍₄₎ for genotype distribution =0.09). Mean age at diagnosis, however, did not differ appreciably between PON1 192 genotypes (mean (SD) age for Q/Q cases =68.9 (9.1), Q/R cases =65.7 (10.4), R/R cases =66.5 (8.3)).

In contrast with the reports of Kondo and Yamamato¹ and Akhmedova et al,⁴ our results do not indicate that PD is associated with specific PON1 genotypes. In addition to the coding region polymorphisms investigated previously, we assessed the role of two promoter mutations but found no evidence of association. These findings suggest that PON1 genotypes may not be predictive of PD, although there remains the possibility of interactions with pesticide exposures. Considerably larger studies will be required to investigate such interactions.