Neurodegenerative diseases such as Parkinson’s disease (PD) and multiple system atrophy (MSA) imposed a severe burden on our aging society. More and more people would suffer from neurodegenerative disorders with increasing life expectancy (Dorsey et al., 2007). Although their causes were not yet fully understood, new genetic insights of PD and some genome-wide association studies (GWAS) have strengthened the evidence that PD has a considerable genetic component (Puschmann, 2013; Chang et al., 2017; Nalls et al., 2019; Blauwendraat et al., 2020). And MSA pedigrees with both autosomal dominant and autosomal recessive inheritance patterns have been reported as well (Wullner et al., 2004; Hara et al., 2007; Itoh et al., 2014).

To date, mechanisms underlying PD and MSA pathogenesis remain not fully understood, but recent studies have shown that neuroinflammation might play an important role in their pathogenesis (Jellinger, 2014; De Virgilio et al., 2016; Gelders et al., 2018). The association between inflammation and neurodegeneration has long been observed in PD (McGeer and McGeer, 2004; Phani et al., 2012; Gelders et al., 2018). In addition, recent studies have shown that inflammation contributions to the abnormal accumulation of α-synuclein, which targets dopamine neurons in the midbrain, forming Lewy bodies in PD (Ohama and Ikuta, 1976; Schain and Kreisl, 2017; Bras et al., 2020). In MSA, which is different from PD, such aggregates usually can be found in oligodendrocytes. However, it has been reported that neuroinflammation also closely links to oligodendroglial α-syncleinopathy (Hoffmann et al., 2019). Therefore, inflammation-related genes may play a part in the pathogenesis of such diseases.

WNT3 (Wnt Family Member 3) is within a protein-protein interaction network with PD risk genes, such as LRRK2, SNCA and MAPT. It might regulate the gene expression of these PD-related genes in multiple immune cells contributing to PD risk (Witoelar et al., 2017). HLA-DRB5 (Major Histocompatibility Complex, Class II, DR Beta 5), whose expression products play a central role in the immune system by presenting peptides derived from extracellular proteins, was found to have a possible pathogenic effect Alzheimer’s disease (AD) risk (Karch and Goate, 2015). IL1R2 (Interleukin 1 Receptor Type 2) is another inflammation-related gene and encodes a cytokine receptor. Polymorphisms, rs2074404 in WNT3 and rs17425622 in HLA-DRB5, have been reported to be associated with PD (Liu et al., 2011; Witoelar et al., 2017). However, such association was not investigated in patients from Chinese. And the previous results of the association between IL1R2 rs34043159 and PD in the Chinese population were contradictory (Chen et al., 2018; Li et al., 2018; Gao et al., 2020; Hu et al., 2020; Fang et al., 2021).

Hence, in this study, we sought to investigate whether these three polymorphisms in inflammation-associated genes could modulate the risk of developing PD and MSA in a Southwest Chinese population.

The current study explored the association between polymorphisms in inflammation-related genes and the risk of developing PD and MSA. We replicated the association of HLA-DRB5 rs17425622 and IL1R2 rs34043159 with PD, and first analyzed it with MSA. In addition, we firstly studied the possible modification effects of WNT3 rs2074404 on PD and MSA. The results indicated the WNT3 rs2074404 could modify the risk for PD, especially for LOPD rather than MSA or EOPD, but not in HLA-DRB5 rs17425622 and IL1R2 rs34043159.

WNT3, encode secreted signaling protein, that possesses significant functions during neural development, such as neural stem cells (NSC) proliferation, axonal guidance, and neuronal survival (Gonzalez-Fernandez et al., 2014; Duan et al., 2018). A previous study indicated WNT3 is a candidate risk gene of PD (Liu et al., 2011), and a protein-protein interaction network of WNT3 with known causative PD genes like LRRK2 was also perceived (Witoelar et al., 2017). The locus rs2074404 of WNT3 was previously found to be a CD (Coeliac disease) associated polymorphism (Dubois et al., 2010; Franke et al., 2010; Amundsen et al., 2014). It was also shown that gut inflammation is a main pathological feature occurring in both PD and CD and this inflammation may contribute to α-synuclein aggregation (Dzamko et al., 2017; Drobny et al., 2021). Additionally, some SNPs in genes responsible for binding bacterial metabolites and intestinal homeostasis were associated with PD (Gorecki et al., 2020). All above suggested CD have a link with neurodegenerative diseases, in some respects of genetics and intestinal flora (Voros et al., 2013; Cao et al., 2018; Spielman et al., 2018). Our study found the minor G allele of rs2074404 had a 1.182-bold increased risk of PD, which further supported there may be shared pathways and similar pathogenesis between CD and PD. Analyzing rs2074404 using two genetic models (dominant and recessive) showed that rs2074404 was significantly associated with PD in the recessive mode and with LOPD, suggesting that the G allele was a risk allele for PD.Nevertheless, it didn’t reach significance in the EOPD subgroup. One study indicated that there might be differences in inflammatory responses between EOPD and LOPD, and peripheral immune disorders were more severe in LOPD patients (Jiang et al., 2019). It was also possible that the long-term inflammatory stimulation due to rs2074404 variants increased the risk of LOPD instead of EOPD. Mechanistically, rs2074404, an intron variant mapped in WNT3, is not located on the NRRE predicted by NUBIScan, however, we found an NRRE component, ER6 (everted repeats separated by 6 base pairs), is not far upstream from it (Supplementary Figure S1), and the different allele of rs2074404 might affect the ER6 element binding and further influence WNT3 expression. Furthermore, we found that WNT3 rs2074404 has a three-dimensional interacting with rs916888 that also has a similar link to WNT3 rs415430 (Lu et al., 2017) (Supplementary Table S4), which has been reported to be associated with PD in an Ashkenazi Jewish population (Liu et al., 2011). However, whether rs2074404 affects the expression of WNT3, or the fact and potential mechanisms that rs2074404 or rs415430 in WNT3 were associated with the increased PD risk need more studies. Moreover, this locus makes no difference on MSA onset. Onset age might be one reason, because WNT3 rs2074404 increased the risk of development for PD, mainly in LOPD, but not in EOPD. For MSA, the mean onset age was younger than that in LOPD, which makes MSA suffer from a relative shorter-term inflammatory stimulation mediated by WNT3 rs2074404. Even now, the role of neuroinflammation in MSA cannot be ignored. As our previous study has shown, rs3135500 variant in the NOD2 gene whose products act a role in the immune response might increase the risk for MSA (Cao et al., 2018). However, there were no relevant studies on whether WNT3 rs2074404 could contribute to the development of MSA. In this study, it was an exploratory study, the smaller sample size of MSA patients might be another reason which contributing the negative results. It is worth point out that our study may be the first one to directly report that rs2074404 is a polymorphism for PD risk with specific data, Therefore, our findings need to be confirmed in more studies with larger sample-size.

IL1R2 encodes the receptor of IL1 (interleukin-1), a critical inflammation cytokine. The association between IL1 and dopaminergic dysfunction has been reported (Williams-Gray et al., 2016). And this locus has been found to be associated with PD and AD among the Chinese population in some researches (Li et al., 2018; Chen K. et al., 2020; Fang et al., 2021), but in some were not (Chen et al., 2018; Hu et al., 2020). In this current study, rs34043159 was also found to have no effect on both PD and MSA. And then, after a comprehensive searching, we pooled five previous studies including this locus in Chinese and the current analysis (Figure 1; Table 5) together which came out with a nonsignificant result similar to ours, whether performed by random-effects or fixed-effect model. However, using the sample from Chinese (Figure 1, 7,626 allele in patients and 9,322 allele in controls), the statistical power for the rs34043159 polymorphism was 0.677 if the MAF in controls was 38.3% and the OR of developing PD in patients relative to controls 1.08 (Chen et al., 2018; Li et al., 2018; Gao et al., 2020; Hu et al., 2020; Fang et al., 2021), indicated the lack of statistical power to adequately assess the small impact of this risk variant. Further investigations with a larger sample or from experimental evidence are needed to confirm if there is a link between it and PD. In addition, origin-specific effects of the risk locus or potential environmental factors may contribute to these inconsistent results. For MSA, this is the first study to explore rs34043159 on MSA and found out that it is not associated with MSA, whether other loci of IL1R2 region may make a difference on MSA onset need more evidence.

For HLA-DRB5 rs17425622, it is a risk locus for UC (ulcerative colitis) (Anderson et al., 2011). Braak et al. (2006) suggested that the pathological process in PD originates in the gastrointestinal tract, and then a hypothesis called “gut-brain axis theory” has been proposed, which consists of bidirectional communication between the central nervous system and the enteric nervous system (Carabotti et al., 2015). Indeed, as mentioned above, some researches showed that chronically inflamed gut may trigger α-syn deposition, increase the permeability of the blood-brain barrier leading to neuroinflammation (Lema Tome et al., 2013; Pellegrini et al., 2018). However, whether the SNPs on inflammatory bowel disease genes would increase the risk of PD is worth further exploration. The association of rs17425622 with PD was first explored in a large-scale meta-analysis of genome-wide association, which showed the T allele of rs17425622 could reduce the risk of PD (OR: 0.79, p = 8.07e−5) in the USA population (Nalls et al., 2014). In addition, a recent study for mapping the HLA locus in PD among Europeans also validated associations for the SNPs rs17425622 (OR: 0.92, 95% CI (0.90, 0.95)) (Yu et al., 2021). However, our preliminary exploration of this locus in the Chinese PD and MSA population did not yield meaningful results. In this study, the statistical power for the rs17425622 polymorphism was 0.857 if the MAF in controls was 21.7% and the OR of developing PD in patients relative to controls 0.79 (Nalls et al., 2014), which indicated the sample size involved in our study was enough. The genetic heterogeneity of different people might be an essential reason for the divergent findings.

The strengths of our study are we included a relatively large and ethnically homogeneous cohort, and WNT3 rs2074404 related to PD in the Chinese population was tested for the first time. However, this study has some weaknesses or limitations. First, this genetic study did not take the gene-gene or gene-environment interaction into consideration. Second, our study was a single-center study. Thus, more multi-centers and larger studies are expected.

We found the allele G of WNT3 rs2074404 was a risk allele for PD and particularly, for the LOPD subgroup. However, we did not detect the significant association of IL1R2 rs34043159, HLA-DRB5 rs17425622 with PD and MSA in a Chinese population. More studies for the WNT3 signaling pathway would be beneficial in developing new therapeutic strategies.