NRF2 polymorphism and susceptibility to ischemic stroke in a Chinese population

Introduction

Cerebral stroke is a cerebrovascular disease influenced by multiple factors. The blockage or rupture of blood vessels in the brain can cause a localized interruption of blood flow, triggering a series of pathological damage processes. In China, cerebral stroke is the leading cause of death and disability among adults, with the highest incidence rate worldwide (Wang et al., 2022b). Among the different types of strokes, ischemic stroke (IS), which results from intracranial vascular occlusion, is the most prevalent. The incidence of IS in China has increased from 117 cases per 100,000 individuals in 2005 to 145 cases per 100,000 individuals in 2019, with a prevalence rate of 1,700 cases per 100,000 individuals in the same year (Wang et al., 2022a). In Yunnan Province, the incidence of IS rose from 220 cases per 100,000 people in 1990 to 674 cases per 100,000 people in 2017. During the same period, the disability-adjusted life-years attributed to IS increased from 770.7 to 815.4 per 100,000 individuals (Liu et al., 2023). These trends indicate a growing disease burden in both Yunnan and China as a whole.

Several modifiable risk factors, including hypertension, smoking, diabetes, and hyperlipidemia, contribute to IS. However, genetic factors are also recognized as significant contributors to disease susceptibility. Single-nucleotide polymorphisms (SNPs) in the genome influence an individual’s inherent risk of developing IS (Boehme, Esenwa & Elkind, 2017). With advancements in genomic technologies, the role of genetic factors in IS onset and progression has gained increasing attention. Multiple studies have reported associations between genetic polymorphisms and IS susceptibility (Almeida, 2013).

Nuclear factor erythroid 2-related factor 2 (NRF2) is a key regulatory molecule that protects cells against oxidative stress. It plays a vital role in maintaining redox homeostasis and overall brain function. In response to oxidative stress, NRF2 becomes activated and regulates over 250 downstream target genes involved in oxidative homeostasis. These include heme oxygenase 1, NADPH quinone oxidoreductase 1, and catalase (Xu et al., 2020). A previous report suggests that excessive reactive oxygen species (ROS) in IS leads to NRF2 upregulation, and the activation of its downstream target genes helps counteract oxidative damage, underscoring the neuroprotective role of NRF2 in IS (Farina et al., 2021). Certain SNPs in NRF2 have been associated with disease susceptibility. For example, the 617 C > A SNP has been linked to an increased risk of oxidant-induced acute lung injury (Marzec et al., 2007). Additionally, patients with cholangiocarcinoma who carry the NRF2 rs6726395 GG genotype have a longer median survival time (344 ± 138 days, 95% confidence interval (CI) [73–615] days) compared to those with the AA/AG genotype (172 ± 37 days, 95% CI [100–244] days) (Khunluck et al., 2014). Recent studies have also demonstrated associations between NRF2 SNPs and susceptibility to alcoholic liver disease (ALD), chronic kidney disease (CKD), and cardiovascular disease (Adam et al., 2017; Jerotic et al., 2019; Nunes dos Santos et al., 2019). However, the relationship between NRF2 polymorphism and IS risk in the Chinese population remains unexplored.

In this study, we investigated the correlation between NRF2 SNPs and IS susceptibility in a Chinese population from Yunnan. We further analyzed the impact of these SNPs on NRF2 expression and examined their effects on ROS levels in the peripheral blood of patients with IS.

Materials & Methods

Results

Distribution analysis of rs35652124 across different regions and populations

Analyzing the distribution of rs35652124 across different regions and populations is valuable for understanding its influence on the incidence of IS in the Yunnan population. Globally, the frequencies of the rs35652124 T and C alleles are 62.44% and 37.56%, respectively. In the East Asian population, the T and C allele frequencies are 44.84% and 55.16%, respectively. The allele distribution in our study population aligns with the trends in East Asia. The distribution of this variant in other regions is shown in Fig. 1.

We also examined the distribution of the three rs35652124 genotypes across different populations. As shown in Fig. 2, genotype distribution varies significantly among populations. Table 7 indicates that the genotype distribution pattern in our study population is consistent with that of the Chinese Han population. However, significant differences were observed when compared with the Chinese Dai population in East Asia, the Finnish and British populations in Europe, the Colombian population in the Americas, and the Gambian and Yoruba populations in the Africa (p < 0.05).

Table 7:

Analysis of three genotypes of rs35652124 in different populations worldwide.

Region	Population	Frequency of genotyping (N, %)			χ2	p-value
		TT	TC	CC
East Asian	Yunnan	36 (25.5%)	54 (38.3%)	51 (36.2%)	Ref
	CDX	12 (12.9%)	49 (52.7%)	32 (34.4%)	7.042	0.030
	CHB	30 (29.1%)	47 (45.6%)	26 (25.2%)	3.310	0.191
	JPT	18 (17.3%)	44 (42.3%)	42 (40.4%)	2.357	0.308
South Asian	ITU	23 (22.5%)	52 (51.0%)	27 (26.5%)	4.134	0.127
	BEB	19 (22.1%)	37 (43.0%)	30 (34.9%)	0.583	0.747
European	FIN	42 (42.4%)	45 (45.5%)	12 (12.1%)	18.644	<0.001
	GBR	46 (50.5%)	37 (40.7%)	8 (8.8%)	26.174	<0.001
American	CLM	25 (26.6%)	56 (59.6%)	13 (13.8%)	15.815	<0.001
	PEL	24 (28.2%)	37 (43.5%)	24 (28.2%)	1.513	0.469
African	GWD	83 (73.5%)	27 (23.9%)	3 (2.7%)	67.969	<0.001
	YRI	68 (63.0%)	38 (35.2%)	2 (1.9%)	54.515	<0.001

DOI: 10.7717/peerj.19742/table-7

Notes:

BEB

Bengali in Bangladesh

CDX

Chinese Dai

CHB

Chinese Han in Beijing, China

CLM

Columbian

FIN

Finnish

GBR

British

GWD

Gambian

ITU

Indian Telugu in the UK

JPT

Japanese in Tokyo

PEL

Peruvian in Lima

Ref

reference

YRI

Yoruba

Analysis of NRF2 genetic variation and NRF2 levels in peripheral blood

As illustrated in Fig. 3A, NRF2 levels in the peripheral blood plasma of patients with IS (248.56 [191.63, 324.86] pg/mL) were significantly higher than those in the control group (173.34 [145.99, 214.82] pg/mL) (p < 0.05). To further investigate whether rs35652124 polymorphisms influence NRF2 expression in patients with IS, we compared NRF2 levels among different genotypes at this locus. As shown in Figs. 3B–3C, neither homozygous nor heterozygous genotypes of rs35652124 significantly affected the NRF2 levels in the peripheral blood of patients with IS. These findings suggest that while genetic variation at rs35652124 is associated with IS susceptibility, it does not significantly modulate NRF2 expression levels in patients with IS.

eQTL

An eQTL is a genetic variant that influences gene expression and plays a role in biological functions. We analyzed the tissue-specific expression of rs35652124 using the GTEx database. As shown in Fig. 4, rs35652124 affects NRF2 expression in the bladder, tibial nerve, thyroid, and several other tissues. However, no significant correlation was observed between rs35652124 and NRF2 expression in blood or brain tissues.

Analysis of NRF2 genetic variation and ROS levels in peripheral blood

ROS levels in peripheral blood plasma was measured using ELISA. As shown in Fig. 5A, patients with IS had significantly higher ROS levels (160.3 [141.23, 194.61] pg/mL) than the control group (141.68 [100.43, 185.19] pg/mL) (p < 0.05). However, subgroup analysis based on rs35652124 genotype revealed no significant association between rs35652124 polymorphic variations and ROS expression levels (Figs. 5B–5C).

Discussion

Several studies have demonstrated that NRF2 plays a crucial neuroprotective role in the development and progression of various diseases (Liu, Locascio & Doré, 2019). Endogenous activation of NRF2 occurs following cerebral ischemia and helps prevent brain injury. Loss of NRF2 in mice (NRF2^−/−) exacerbates the progression of IS-related brain injury, with NRF2-deficient mice exhibiting more severe cerebral infarctions on day 3 and poorer sensorimotor function on day 28 in a permanent ischemic model (Jerotic et al., 2019). The absence of NRF2 reduces the body’s and cells’ antioxidant capacity and downregulates the downstream protective proteins, impairing the response to ischemia. In addition, aggravation of IS triggers inflammatory activation, further worsening ischemic injury and creating a vicious cycle. Previous research has indicated that polymorphisms in NRF2 can influence its affinity, expression, and activity (Song et al., 2016). However, its relationship with IS development among East Asian population has not been previously reported. This study revealed that NRF2 polymorphism is associated with IS risk, with the TC genotype of rs35652124 in the co-dominant model potentially increasing IS susceptibility in a Chinese population from Yunnan.

Among conventional risk factors, hypertension is the most prevalent modifiable risk factor for IS, with up to 84% of patients with acute IS having hypertension (McManus & Liebeskind, 2016). Hypertension has also been identified as a key contributor to stroke-related mortality (Pistoia et al., 2016). Chronic hypertension remains a persistent long-term risk factor for acute IS. Elevated fasting blood glucose level indicate abnormal glucose metabolism and diabetes severity, increasing IS risk. Oxidative stress and mitochondrial dysfunction are critical mechanisms in hyperglycemia-induced ischemic brain injury. Under hyperglycemic conditions, excessive mitochondrial ROS production, impaired mitochondrial energy generation, and the release of apoptotic factors contribute to endothelial cell apoptosis and dysfunction, thereby exacerbating disease risk (Zhang, Yan & Shi, 2013). Furthermore, hyperlipidemia accelerates pathophysiological processes such as oxidative stress in endothelial cells, inflammation, and lipid peroxidation, all of which facilitate atherosclerosis development (Alkahtani, 2022). Hyperlipidemia also elevates IS risk by promoting conditions such as hypertension, insulin resistance, obesity, and cardiovascular and cerebrovascular diseases (Kloska et al., 2020). However, in this study, the observed variations in allele and genotype distribution of rs35652124 across different regions and populations may be attributed to genetic heterogeneity influenced by regional, environmental, ethnic, and lifestyle differences. Stratified analysis of rs35652124 revealed no significant association between its genotype and confounding factors such as age, sex, hypertension, diabetes, and lipid levels. These findings suggest that the association between rs35652124 and IS risk may not be significantly influenced by interactions with these factors.

Their pilot findings suggest that in children with ASD (autism spectrum disorder) the NEF2L2 rs35652124 polymorphism impacts adaptive responses that may potentially link to ASD severity, and genotype rs35652124 CC can be protective with respect to oxidative stress characteristic of the pathogenesis of autism (Porokhovnik et al., 2023). Sorour et al. (2021) reported that carriers of the T allele of rs35652124 were 3.329 times more likely to develop vitiligo than carriers of the C allele. In the PRIME study, rs35652124 was associated with aging-related phenotypes, including adverse drug reactions, clinical frailty scores, and multiple diseases (Scutt et al., 2020). Moreover, genetic variation at rs35652124 has been linked to the inflammatory characteristics of the liver and the development of cirrhosis in ALD (Nunes dos Santos et al., 2019). It is worth noting that our study found that the TC genotype in the co-dominant model may be a risk factor for IS, correlating with susceptibility to the disease. This heterozygous mutation has an impact on the occurrence of the disease, which may be caused by dose-dependent reversal. And the Epistasis caused by the possible interaction of homozygous mutations with variations of other genes makes the CC genotype not associated with the occurrence of IS.

Notably, a previous study using in vitro dual-luciferase reporter assay revealed that the rs35652124-C allele reduced NRF2 promoter activity and expression in human microvascular endothelial cells (Marczak et al., 2012). However, the rs35652124 T > C variant is located in the core sequence of the putative binding site of transcription factor-specific protein 1 (SP1). The T > C change may enhance SP1 binding to the NRF2 promoter region, thereby increasing its activity and upregulating NRF2 expression. To investigate the mechanism by which NRF2 rs35652124 influences susceptibility to IS, we examined whether this variant was associated with altered plasma NRF2 levels in patients with IS. Our results showed that NRF2 expression in peripheral blood was significantly higher in the IS group than in the control group. However, no significant differences in NRF2 content were observed among the three co-dominant genotypes of rs35652124. Further analysis using eQTL data confirmed that rs35652124 was not associated with plasma NRF2 levels. Notably, the potential effect of rs35652124 genotypes on intracellular NRF2 expression remains unverified due to the absence of cell-based experimental evidence in this study.

Previous studies have demonstrated that increased ROS levels in experimental IS models lead to endogenous NRF2 activation, with significantly elevated NRF2 expression in ischemic infarction areas (Farina et al., 2021). We, therefore, analyzed the relationship between rs35652124 and ROS levels in peripheral blood. The results indicated no significant differences in ROS content among the three co-dominant genotypes of rs35652124. In IS, elevated ROS levels triggers NRF2 activation. Since blood samples in this study were collected during the early stage of IS, any enhancement in antioxidant capacity due to NRF2 activation may not have been evident. This finding suggests that a single genetic variation at rs35652124 does not directly alter ROS levels in the peripheral blood of patients with early IS. However, ROS content in the peripheral blood of the IS group was significantly higher than in the control group, likely due to cerebral ischemia-induced ROS accumulation, which activates endogenous oxidative stress and contributes to pathological damage.

Recent studies have also reported that specific NRF2 haplotypes are associated with differences in promoter activity and the severity of chronic obstructive pulmonary disease (Hua et al., 2010). In our study, the CCTTGGC haplotype of NRF2 (sequence: rs13005431–rs4893819–rs6721961–rs35652124–rs6726395–rs2364723–rs2706110) was found to increase IS risk.

Unfortunately, the other six genetic loci analyzed in this study did not show significant associations with IS occurrence. According to a study by Nunes dos Santos et al. (2019), the rs4893819 locus in the NRF2 promoter region is not associated with ALD susceptibility and has no effect on NRF2 expression or liver inflammatory activity. In addition, a genome-wide association study by Maraganore et al. (2005) found no link between rs13005431 and susceptibility to Parkinson’s disease. In a study on ALD, rs6721961 had no specific correlation with ALD occurrence, and had no effect on liver inflammation in patients with ALD (Nunes dos Santos et al., 2019). The rs6726395 locus of NRF2 has been linked to susceptibility to lung function impairment caused by smoking (Masuko et al., 2011). Genetic variants of rs2364723 are significantly associated with a reduced risk of CKD and may interact with rs35652124 to influence CKD risk (Gómez-García et al., 2022). The TAMRISK study found that the TT genotype of rs2706110 increased the risk of cerebrovascular diseases and suggested that insufficient NRF2 expression may contribute to their development (Kunnas, Määttä & Nikkari, 2016).

Thus, while NRF2 polymorphisms have different effects on various diseases, further research is needed to determine their role in individual susceptibility to IS. A limitation of our study is that no functional analysis was conducted to elucidate the mechanism by which rs35652124 influences IS. Future studies should include dual-luciferase reporter assays to address this gap. Furthermore, the existence of dose–effect deficiency is explored through functional mechanism studies such as epigenetic regulation and protein interactions. Moreover, our study was limited to a single population. We plan to expand our research to multiregional and multiethnic populations to validate our findings and further explore genetic risk factors for IS.

Conclusions

The polymorphic locus rs35652124 in NRF2 was associated with susceptibility to IS in a Chinese population, with the TC genotype in the co-dominant model identified as a risk factor. However, the rs35652124 variant did not influence peripheral blood NRF2 levels or affect IS risk by altering ROS levels, suggesting that its role in IS susceptibility may not be mediated through oxidative stress. In addition, the CCTTGGC haplotype was identified as a potential risk factor for IS. While our findings highlight a genetic association between NRF2 polymorphisms and IS, further functional studies are needed to elucidate the underlying mechanisms.

Supplemental Information