Parkinson’s disease (PD) is the second most common neurodegenerative disease, affecting 1.14 and 1.06% of people over 60 years old worldwide and in China, respectively (Pringsheim et al., 2014; Cui et al., 2020). Only a minority of PD are known with genetic causes and familial hereditary, such as mutations in SNCA, LRRK2, and PARKIN. In contrast, most PD cases are sporadic with unclear etiology. Complicated interactions between genetic susceptibilities and environmental exposures are considered to be responsible for the pathogenesis of sporadic PD (Kalia and Lang, 2015).

The chemical 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine (MPTP) was discovered to induce parkinsonism in humans in the 1980s (Langston et al., 1984). MPTP was later found to be structurally similar to some known pesticides such as paraquat, raising the question whether pesticide exposures were associated with PD (Wirdefeldt et al., 2011). Multiple kinds and classes of pesticides have since been identified to link to PD development. For example, exposure to herbicide paraquat, insecticide rotenone or fungicide maneb leads to toxic effects similar to those of MPTP in animal models (Islam et al., 2021). Benomyl, another fungicide, inhibits aldehyde dehydrogenase activity, and results in accumulation of 3,4-dihydroxyphenylacetaldehyde (DOPAL), a dopamine metabolite highly toxic to dopaminergic neurons (Fitzmaurice et al., 2013). Organophosphate pesticides are also shown to be associated with increased risk of PD, probably due to the inhibition of acetylcholinesterase activity and the induction of oxidative stress (Sanchez-Santed et al., 2016). In addition, pesticide exposure exhibits the strongest association with PD among 11 environmental risk factors identified in a meta-analysis (Noyce et al., 2012).

Pesticides are metabolized in vivo by phase I enzymes such as cytochromes P450 (CYPs) and esterases into reactive metabolites. Phase II enzymes such as glutathione S-transferases (GSTs) then detoxify the reactive metabolites through catalytic conjugation with charged species such as reduced glutathione (Omiecinski et al., 2011). Genetic liabilities of these enzymes to diseases have been widely studied such as in context of diabetes, cancer and cardiovascular disorders (Gao et al., 2017; Elfaki et al., 2018). It has also been investigated whether the enzymes are genetically associated with PD. However, some of the results remain inconsistent and inconclusive, such as for CYP2D6, and GSTM1 (Ahmadi et al., 2000; Chan et al., 2003; Elbaz et al., 2004; Perez-Pastene et al., 2007).

Since the same class of xenobiotic metabolizing enzymes may functionally compensate each other or overlap, we hypothesize that the genetic effects of each class modulate PD risk stronger in a synergistic way than individually. We selected 14 polymorphic loci out of 13 genes which encode metabolizing enzymes of three classes, GSTs, esterases, and CYPs (Supplementary Table 1). Amongst, GSTM1 and GSTT1 are two deletion polymorphisms, and the remaining are single nucleotide polymorphisms (SNPs). These polymorphisms are either functional in modulating protein expression or enzyme activity or are reportedly associated with PD risk (Supplementary Table 1). Their genetic effects on PD were herein investigated individually and synergistically in a large Chinese cohort.

Xenobiotic metabolizing enzymes are key to the retention and deposition of environmental substances in human bodies. The exposures, in particular pesticide, have been well-recognized to aggravate PD risk. The association between pesticide and PD may be more prominent in eastern China due to the higher pesticide residues (Yu et al., 2020). By analyzing 13 functional or potential PD-associated polymorphisms in genes which encode three classes of metabolizing enzymes, including GSTs, CYPs, and esterases, we demonstrate that none of the polymorphisms are independently associated with PD risk. These observations are in line with results of a previous genome-wide association study (GWAS) in East Asian (Foo et al., 2020). On the other hand, there are synergistic effects on PD in polymorphisms of CYPs and GSTs, that is, the protection conferred, respectively, by simultaneous presence of the rs12441817/CC (CYP1A1) and rs2070676/GG + GC (CYP2E1) genotypes, as well as of the GSTM1-present, GSTT1-null, rs156697/AG + GG (GSTO2) and rs1695/AA (GSTP1) genotypes.

CYPs are the class of the most important phase I enzymes responsible for metabolism of various xenobiotic compounds, including drugs and pesticides (Elfaki et al., 2018). Genetic polymorphisms in CYPs are associated with altered enzymatic activities and disturbed expression levels. The changes in activity or expression may lead to inter-individual variations in xenobiotic metabolisms, and are implicated in various disease susceptibilities (Dardiotis et al., 2013; Elfaki et al., 2018). In context of PD, the C allele of rs12441817 (CYP1A1) was shown to be associated with reduced risk of PD (P = 5.22 × 10^–5, OR = 0.87, 95% CI = 0.81–0.93) in a European-descent based GWAS (Nalls et al., 2014). However, this association is not confirmed by the current study and the Asian GWAS (Foo et al., 2020). We are not sure whether it is ethnicity specific. In addition, rs2070676 (CYP2E1) showed an association with PD in a Swedish population (Shahabi et al., 2009), rs1048943 (CYP1A1) was associated with PD in a Japanese population (Takakubo et al., 1996), and rs762551 (CYP1A2) was reported to have a marginal association with PD risk in the United States and Denmark (Palacios et al., 2010; Chuang et al., 2016). Nonetheless, these associations are not independently observed in the current study, the Asian GWAS (Foo et al., 2020), the European GWAS (Nalls et al., 2014), and other case-control studies (Chan et al., 2002; Facheris et al., 2008; Siokas et al., 2021). As noted in Supplementary Table 1, the rs1048943 is a missense variant of CYP1A1 (T > C, Ile462Val) and functionally regulates the CYP1A1 enzyme activity (Shah et al., 2009). The CYP1A2 activity is also influenced by its rs762551 polymorphism (Denden et al., 2016). The rs4244285 is the most studied polymorphism of CYP2C19, and can alter the CYP2C19 mRNA reading frame leading to a truncated non-functional protein (Scott et al., 2012). As discussed above, these polymorphisms are not, at least not consistently, shown to be associated with PD, suggesting that bluntness of each enzyme per se may not be sufficient to affect PD susceptibility. On the other hand, our results demonstrate that simultaneous presence of the rs12441817/CC (CYP1A1) and rs2070676/GG + GC (CYP2E1) genotypes confers protection against PD risk, suggesting a synergistic genetic effect from these two enzymes. Coincidentally, the rs12441817/C and the rs2070676/G indeed served as the protective allele in the two positive studies (Shahabi et al., 2009; Nalls et al., 2014), which to some extent supports our finding. These results together indicate that a synergistic play more likely leads to a modulation of PD risk than individually.

Esterases, including paraoxonase 1 (PON1), paraoxonase 2 (PON2), and butyrylcholinesterase (BCHE), are responsible for catalyzing ester hydrolysis during Phase I reactions, and are critical in organophosphate metabolism (Satoh et al., 2002). Two of the most studied polymorphisms of PON1 are rs854560 (A > T, Leu55Met) and rs662 (T > C, Gln192Arg). The rs854560/A and rs662/C alleles result in higher PON1 enzyme activity (Rejeb et al., 2013). The rs854560 was excluded from the current study due to a low minor allele frequency in East Asians (T allele at 0.034; from gnomAD).¹ The rs12026 of PON2 (G > C, Ala148Gly) and rs1803274 of BCHE (G > A, Ala539Thr) are linked with lower enzyme activity at presence of the C and A allele, respectively (Rejeb et al., 2013; Habieb et al., 2021). Nonetheless, neither individual nor synergistic effect on PD risk is detected in these variants, suggesting that esterases may not be important in modulating the environmental and genetic interplay toward PD risk. Besides, there are two previous reports for rs662 and PD with results showing positive association (Kondo and Yamamoto, 1998) and no association (Wang and Liu, 2000).

The current study detects no individual association between the GST polymorphisms and PD risk. The rs4925 (GSTO1), rs156697 (GSTO2), and rs1695 (GSTP1) have not been reported in Chinese PD cohorts till now. The GSTM1 and GSTT1 deletion polymorphisms have been previously reported in relatively small Chinese populations, also showing no association with PD (Tao et al., 1998; Fong et al., 2006). The lack of PD association in these variants has been additionally observed in other ethnicities, including Caucasians, Australians and Japanese (Menegon et al., 1998; Ahmadi et al., 2000; Harada et al., 2001; Whitbread et al., 2004). Nonetheless, reports of significant association with PD also exist, such as for the GSTM1 variant in Chileans (Perez-Pastene et al., 2007), the GSTT1 variant in a UK population (De Palma et al., 1998), and the rs4925 in Americans (Wahner et al., 2007). Regardless of the individual performance, the present study for the first time reports a protection against PD synergistically conferred by the genotypes of GSTM1-present, GSTT1-null, rs156697/AG + GG (GSTO2) and rs1695/AA (GSTP1). The GSTM1-present and the rs1695/A allele are linked with elevated enzyme activities (Johansson et al., 1998; Hayes and Strange, 2000). In contrast, the GSTT1-null and the rs156697/G lead to reduced enzyme activities or protein levels (Hayes and Strange, 2000; Allen et al., 2012). Thus, the combination of these four genotypes appears to be mediocre in terms of collective enzyme activity. Interestingly, there is a similar case showing a synergistic effect from the alleles of GSTM1-null, GSTT1-present, GSTA1/T (low activity), and GSTP1/G (low activity) on promoting risk for clear cell renal cell carcinoma (Coric et al., 2016). In certain cases, high GST activity may lead to increased toxicity, such that GSTT1 dehalogenates and transforms halogenated organic compounds to toxic products (Landi, 2000). In addition, the expression of GSTs may be regulated by sex hormones (Coecke et al., 2000), which may partially explain our observed female-specific synergistic effect.

In conclusion, the current study demonstrates that polymorphisms of CYPs and GSTs, but not those of esterases, have synergistic impact on PD susceptibility. The variants are not associated with PD individually. Our findings suggest that more attention should be paid to the synergistic interplay for the genetic variations of xenobiotic metabolizing enzymes to better understand their roles in PD risk.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://datadryad.org/stash, https://datadryad.org/stash/share/bcA9u9G-dGMdGLw_TnpEtY9qPeJg7aXdRA7u35_tgxs.

The studies involving human participants were reviewed and approved by the Ethics Committee of the Second Affiliated Hospital and Yuying Children’s hospital, Wenzhou Medical University. The patients/participants provided their written informed consent to participate in this study.

H-HF designed the study. H-HF, K-YW, H-DY, M-JG, X-HY, and H-JD conducted the experiments. H-HF and J-HZ analyzed the data. B-QL and XZ contributed samples. H-HF and J-HZ wrote the manuscript. XZ and J-HZ supervised the study. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.