Two SNPs (rs1801282 and -1279G/A) of PPARγ are associated with Parkinson's disease in a northern Chinese population

Background: This study aimed to assess the association between PPARγ gene polymorphism and susceptibility to Parkinson’s disease (PD) in a northern Chinese population. Methods : We conducted a case-control study which including 391 outpatients with PD and 391 healthy matched individuals. All subject genotypes on PPARγ gene in rs3856806, rs1801282, -1279G/A were determined by polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) analysis. Results : Our results showed participants with AG and AG+AA (dorminant model) genotypes of -1279G/A had a higher genetic risk of PD compared to those with GG ( p = 0.024, OR = 1.781, 95% CI = 1.073-2.956; p = 0.024,OR = 1.768, 95% CI = 1.078-2.898). A allele of the -1279G/A polymorphism was presumably correlated with increased risk of PD ( p = 0.037,OR = 1.639, 95% CI = 1.027-2.616) and male PD ( p =0.032, OR = 1.998, 95% CI = 1.051-3.798) as well as early-onset Parkinson’s disease (EOPD)( p =0.019, OR = 2.667, 95% CI = 1.263-5.629). Stratification analysis by age for rs1801282 indicated a significant genotype difference between EOPD and controls( p =0.005) as well as late-onset Parkinson’s disease(LOPD) and controls( p =0.008). G allele frequency of rs1801282 in EOPD subjects was significantly higher than it in controls( p = 0.006, OR =3.093, 95% CI = 1.446-6.615) and LOPD ( p = 0.009, OR =2.899, 95% CI = 1.344-6.253). Conclusions : The study showed that in a northern Chinese population, the A allele

neurons from injury [2,5]. The pathogenesis of PD remains unclear; however, the interaction between genetics and environment is considered to be involved in PD development [6]. Further, related studies have indicated an association between the susceptibility genes and PD development [7][8][9].
The chromosome site for the PPARγ gene is 3p-25 [10] and many genetic variations of this gene have been found. Three gene loci sited in different functional region of PPARγ, with rs1801282 located in exon B of PPARγ which related with decreased PPARγ activity [11], and rs3856806 located in exon 6 [11] as well as -1279G/A located in the promoter area. Among these mutations, previous studies have revealed that the rs1801282 polymorphism might be related to PD or AD susceptibility [12][13][14][15]. One study found that type2 diabetes mellitus was associated with subsequent PD [16], and rs3856806 polymorphism was probabely related with type2 diabetes in a Chinese Han population [17]. Moreover, -1279G/A within PPARγ gene as a novel polymorphism that has never been studied in patients with PD. We aimed to assess association between PPARγ genetic polymorphisms (rs1801282, rs3856806, and − 1279G/A) and PD in a cohort of northern Chinese subjects. Additionally, we conducted stratification analyses by sex and age as well as haplotype analysis for the first time to better understand the association between PPARγ and PD risk .

Study subjects
In this case-control study, we enrolled 391 patients with PD (185 females and 206 males; age 62.42±9.32 years) who met the criteria of the UK PD Brain Bank and 391 gender, age, and ethnicity matched healthy individuals (178 females and 213 males; age: 61.08±10.16 years) without neurologic or psychiatric disorders, T2DM, or cardiopathy. All subjects in the PD group were enrolled from the Neurology clinic department of the Affiliated Hospital of Qingdao University while the controls were randomly recruited from the Health Examination center. As shown in Table 1, there was no significant between-group difference in the age (p = 0.056) and sex (p = 0.616). Polymorphism (PCR-RFLP) analysis. We used three restriction enzymes and three pairs of primers as previously described [18][19][20].  Table 2 Primary information on genotyping assays for three SNPs of PPARγ gene

Results
The characteristics of all the study subjects are summarized in Table 1. The Hardy-Weinberg equilibrium analysis of the three SNPs were > 0.05, which indicated there was no sample bias. Table 3 shows the results of the between-group analysis of dominant, and recessive genetic models as well as genotype and allele frequency for rs3856806 and − 1279G/A. We just detected two genotypes in all samples for rs1801282, so only genotype and allele frequency about rs1801282 are presented in the Table 3. Regarding rs1801282 polymorphisms, subgroup analysis in terms of age indicated statistical differences between the early-onset PD (EOPD) (age of diagnosis < 50) and healthy group (p = 0.005).
Compared to the healthy group, the EOPD subgroup had a higher G allele frequency (OR = 3.093, 95% CI = 1.446-6.615, p = 0.006 in Table 4). Further, significant difference was found between the EOPD and late-onset PD (LOPD) (age of diagnosis > 50) (p = 0.008). Compared with the LOPD subgroup, the EOPD subgroup had a higher frequency of the G allele of rs1801282 (p = 0.009, OR = 2.899, 95% CI = 1.344-6.253). Furthermore, no significant difference was found among other subgroups in the allele frequencies and genotype distribution (p > 0.05) in Table 4.  Table 3 and Table 4).
According to Table 5

Discussion
We found that the allele A of -1279G/A was significantly associated with PD risk in a northern Chinese population. Subgroup analysis showed that both − 1279G/A and rs1801282 were associated with EOPD and both A allele of -1279G/A polymorphism and G allele of rs1801282 polymorphism within PPARγ gene might be a risk factor for EOPD in a northern Chinese population. In addition, our finding showed that A allele of -1279G/A may increase the risk of male PD and AG + AA genotype may increase the susceptibility of PD in the dominant model. However, there was no mutual effect between the rs3856806 polymorphism and PD in the dominant and recessive model.
PPARγ has been reported to be associated with the development of numerous degenerative diseases.
Additionally, an association of the rs1801282 genotype with early-onset AD has been reported in the Finnish population [13]. In the absence of the APOEε4 subgroup with PPARγ Pro12Ala polymorphism (rs1801282), the onset age of AD patients with the Pro/Ala genotype was found 4.6 years earlier than Pro/Pro genotype carriers in northeast China [15]. Another study on Italians reported that the risk of developing AD in Ala carriers was twice as high as that of octogenarian controls [21]. Interestingly, analysis after gender stratification showed that the G allele of rs1801282 played a protective role in females but had an opposite effect in male Caucasians from the UK [22]. Tanner CM et al demonstrated the effect of genetic factors was more obvious in the EOPD subgroup than it in the LOPD subgroup [23]. In a certain sense, above studies seems to tell us rs1801282 is more likely a age-related mutation after so many different racial analysis.
A previous association study on PPARγ gene polymorphism in Japanese patients with Parkinson's disease with dementia (PDD) (n = 171) and controls (n = 136) reported no significant between-group differences in the genotypic frequencies of the SNPs (rs3856806 and rs1801282) or the haplotype analysis for the 2 PPARγ-SNPs [24]. Yang investigated the PPARγ gene polymorphism (rs3856806 and rs1801282) in a Southern Chinese cohort comprising patients with PD (n = 206) and controls (n = 210) and reported no significant between-group difference in the genotype distribution [12]. Our results are clearly consistent with those of the aforementioned studies; however, we performed further subgroup analysis based on age and sex using a larger sample size from a different geographical area. This is the first study to provide evidence that the A allele of -1279G/A SNP may contribute to increasing PD risk in northern Chinese population. Our findings might contribute toward developing a novel therapy and a feasible means of predicting the risk for PD.
To date, the pathophysiology of PD remains elusive. A review reported that the most probable nosogenesis involves lesions of the ubiquitin-proteasome system (UPS), oxidative stress anomaly, and mitochondrial deficiency [25]. UPS dysfunction might result in excess or misfolded proteins in the brain, in turn resulting in PD development. Rosiglitazone, a PPARγ agonist, not only reduces mHtt aggregates containing ubiquitin and heat shock factor1 (HSF1) but also enhance the function of the UPS, HSF1, and heat shock protein27/70 (HSP27/70) in N2A cells [26]. Moreover, in human neuroblastoma SH-SY5Y cells, rosiglitazone improves the expression of SOD, catalase, Bcl-2, and Bax, which might attribute to the prevention of mitochondrial impairment induced by the 1-methyl-4phenylpyridinium ion (MPP+) [27]. In addition, ciglitazone, a PPARγ agonist, plays a role in protecting neurons from oxidative stress by regulating mitochondrial fusion and fission in hippocampal neurons [28]. Overall, PPARγ play a protective role in the development of PD.
To our knowledge, the rs1801282 G allele variation shows decreased transcriptional activity of PPARγ in vitro [29], suggesting potential association with PD. -1279G/A is located in the promoter region of PPARγ and the polymorphism of the promoter region tends to affect gene expression and contribute to the occurrence of disease [30]. The ENCODE project and bioinfomatics approaches identified the − 1279G/A(rs7647481) as the cis-regulatory vabriant which exerts the effect of regulating transranscriptional activity, and ultimately contributes to different sensitivity of insulin in primary adipose cells [31]. In addition, Yin Yang 1(YY1) has been confirmed as an allele-specific transcription factor for − 1279G/A (rs7647481) A allele, which may affects the level of PPARγ [31]. Taken together, the − 1279G/A polymorphism is a cis-regulatory vabrian [31], which may play a role in decreasing transcriptional activity of PPARγ, leading to the low expression level of PPARγ which confers to susceptibility for the development of PD. In fact, the precise molecular mechanism is more complicated than we expect, and hence, further research is required to elucidate the underlying mechanism between variants and disease risk.
Our study had several limitations, such as limited sample size, race, and mutational sites. Therefore, there is a need for more studies with more ethnicities and larger cohorts to assess the association of PPARγ polymorphism with PD susceptibility.

Conclusion
This is the first study to demonstrate that the A allele of -1279G/A and the G allele of rs1801282

Acknowledgments
We appreciate all subjects for participating this study and we would like to acknowledge the investigators for their helpful comments regarding this paper.

Funding
This work was supported by grants from Natural Science Foundation of China(81571225). The study design and the collection of the PD data were supported by the funding.

Availability of data and materials
The data used and analyzed in the current study are available from the corresponding author on