Association of AMPK Pathway-Related Gene Polymorphisms with Symptomatic Intracranial Atherosclerotic Stenosis in a Chinese Han Population

Abstract

Background and Objective:

Recently, AMP-activated protein kinase (AMPK) signaling was confirmed to be intimately associated with atherosclerosis. Evidence indicates that genetic susceptibility plays an important role in the etiology of symptomatic intracranial atherosclerotic stenosis (sICAS), however few genes have been pinpointed being etiologically associated. This study investigated possible links between single nucleotide polymorphisms (SNPs) of AMPK-related genes and sICAS in Han Chinese subjects.

Methods:

Target gene sequencing was carried out in 400 sICAS Han Chinese patients and 1007 healthy controls for 11 AMPK pathway-related genes. Chi-squared testing and multiple logistic regression in dominant, recessive, and additive models were used to evaluate the association between SNPs and risk of sICAS. Bonferroni corrections were performed with a p < (0.05/44 = 0.0011) as statistically significant. Further subgroup data analyses was conducted using chi-squared or t-tests.

Results:

There were 44 common variants of 11 candidate genes distributed differently between sICAS patients and healthy controls, among which the INSR rs78312382 SNP remained significant even after a Bonferroni correction. Logistic regression analysis showed that rs78312382 was significantly associated with the risk of sICAS in both dominant and additive models (p_Bonferroni = 7.874e⁻⁵ and 0.000506, respectively), with the A allele being much more prevalent in the sICAS group (p = 0.000404).

Conclusions:

Variants of the INSR rs7831282 locus may play an important role in the development of sICAS among the Han Chinese with the A allele being a risk factor and a potential biomarker for this illness.

Introduction

Symptomatic intracranial atherosclerotic stenosis (sICAS) is a main cause of ischemic stroke, especially in Asian populations (Koo, 2015) and most relevant to incidence, disability, mortality, and recurrence of large artery atherosclerotic ischemic stroke, the most common ischemic stroke subtype (Wang et al., 2014; Pan et al., 2017). Many studies have proposed that sICAS is a complex disease caused by environmental factors, such as hypertension, diabetes mellitus, age, smoking, and metabolic syndrome, as well as genetic factors, such as single nucleotide variants in ring finger protein 213 and adiponectin gene (ADIPOQ) (Cui et al., 2017; Liao et al., 2017, 2019). However, few candidate genes have been localized and the concrete pathogenesis of sICAS is still largely unknown.

AMP-activated protein kinase (AMPK) is expressed almost universally in eukaryotes as a heterotrimeric protein comprising catalytic (α) and regulatory (β and γ) subunits. Each subunit has multiple phosphorylation sites associated with anabolic or catabolic regulation through gene transcription or further protein phosphorylation (Hardie, 2014; Wang et al., 2018). In addition to its major role in energy homeostasis, AMPK signaling has also been found to be involved in metabolic diseases (Carling, 2007), especially atherosclerosis (AS) (Ma et al., 2017).

Wang et al. (2017) found reduction of atheromatous macrophages in apolipoprotein E-deficient mice, an ideal AS model, after activation of AMPK signaling, indicating its potential protective role in AS. AMPK signaling was also found to be involved in multiple pathophysiological processes, such as vascular smooth muscle cell proliferation, endothelial autophagy, proinflammatory reactions, and endoplasmic reticulum stress in the endothelium, which are all implicated in the occurrence and development of AS (Xiong et al., 2014; Hwang et al., 2017; Jojima et al., 2017). In addition, this kind of anti-AS effect on AMPK signaling is believed to be relevant to the enhanced antiatherogenic effect of high-density lipoprotein (Ma et al., 2017). Since atherogenesis is the most common pathological change in ICAS, AMPK signaling is hypothesized to play an important role in ICAS as well.

Genes reported to be involved in AMPK signaling include ADIPOQ, protein kinase AMP-activated catalytic subunit α 1 (PRKAA1), sirtuin 1 (SIRT1), insulin (INS), INS receptor (INSR), peroxisome proliferator-activated receptor γ (PPARG), hepatocyte nuclear factor 4 α (HNF4A), interleukin 6 (IL-6), leptin (LEP), LEP receptor (LEPR), and resistin (RETN). Among them, some have been shown to be closely associated with AS and/or stroke.

Adiponectin has been confirmed to be important in antiatherogenic, anti-inflammatory, antioxidant, vasoprotective, and insulin-sensitizing processes (Katsiki et al., 2017), and rs2241767 and rs182052 single nucleotide polymorphisms (SNPs) of ADIPOQ have both been associated with a higher risk of sICAS in a Han Chinese population (Cui et al., 2017). PRKAA1 deficiency has been shown to inhibit glycolysis, compromise endothelial cell proliferation, and promote AS lesions in hyperlipidemic mice (Yang et al., 2018), whereas deletion of SIRT1 can accelerate vascular aging and AS through increasing oxidative stress and inflammation, promoting foam cell formation, and inducing autophagy (Kitada et al., 2016).

Defective INSR may lead to insulin resistance, which could induce apoptosis or necrosis of foam cells and promote atherogenesis through activating endoplasmic reticulum stress pathways in macrophages of plaques. Moreover, exogenous insulin infusion can alleviate AS in mildly diabetic apolipoprotein E-deficient mice by inhibiting inflammation, lowering plasma lipid levels, and improving endothelial function (Park et al., 2018). Therefore, INSR has been acknowledged as an independent risk factor for cardiovascular disease (Han et al., 2006). Furthermore, some proinflammatory cytokines, such as IL-6, LEP, and RETN, can mediate insulin resistance and development of atherogenesis through stimulating inflammation and neoangiogenesis (Beltowski, 2006; Kantorova et al., 2011; Huang et al., 2017).

However, few SNPs among AMPK-related genes have been found to be associated with sICAS, and whether the above genes are directly linked to sICAS remains unknown. Based on the mentioned evidence, this study investigated whether AMPK-related genes and/or gene variants contribute to the development and occurrence of sICAS in Han Chinese people. The current findings will be useful for understanding pathological mechanisms of sICAS as well as provide theoretical evidence for potential biomarkers of sICAS susceptibility and early prevention.

Methods

Study population

The study subjects were continuously recruited from Xiangya Hospital of Central South University (Changsha, Hunan, China) from July 2015 to December 2017. A total of 400 patients with clinically confirmed transient ischemic attacks or acute ischemic stroke due to ICAS were enrolled. Diagnostic criteria for acute ischemic stroke and transient ischemic attacks were consistent with that utilized in previous studies (Duan et al., 2014; Long et al., 2017).

Clinical information, including gender, age, and risk factors for ischemic stroke (e.g., diabetes mellitus, dyslipidemia, hypertension, alcoholism, smoking, medical history, and family history), was assessed. Hypertension was defined as blood pressure at least 140/90 mmHg or current use of antihypertensive medication. Diabetes mellitus was defined as casual glucose ≥200 mg/dL, fasting plasma glucose ≥126 mg/dL, or use of antidiabetic agents (Zhan et al., 2012). Dyslipidemia was defined as serum triglycerides ≥1.7 mM, serum total cholesterol ≥5.2 mM, serum low-density lipoproteins cholesterol ≥3.4 mM, or serum high-density lipoproteins <1.0 mM according to Chinese guidelines for the prevention and treatment of dyslipidemia in adults (Joint Committee for Developing Chinese Guidelines on Prevention and Treatment of Dyslipidemia in Adults, 2016). Smoking and alcoholism status were evaluated based on self-report at registration.

Peripheral blood samples were obtained in the morning after overnight fasting for 12 h and underwent laboratory testing for fasting blood glucose, triglycerides, total cholesterol, and high- and low-density lipoprotein levels. Allelic frequencies of controls were obtained from the GeneSky in-house database (1007 samples). This study was approved by the Ethics Committee of Central South University, and all patients signed informed consent before inclusion.

Diagnosis of ICAS

ICAS was defined as ≥50% diameter stenosis at proximal and/or terminal portions of the anterior circulation or basilar artery without abnormal vascular networks in the region of the basal ganglia assessed by computed tomography, magnetic resonance, or digital subtraction angiography (Yu et al., 2017). Angiography imaging findings were diagnosed by two trained neurologists. Those patients who had extracranial arterial stenosis or other causes of ischemic stroke, such as cardioaortic embolism, vasculitis, or arterial dissection, were excluded.

Targeted sequencing and variant filtration

Genomic DNA was extracted from peripheral blood using a Blood Gen Midi Kit (ComWin Biotech Co., Ltd., Beijing, China) according to the manufacturer's instructions. The quality and quantity of DNA samples were assessed using Nanodrop 2000. A custom capture array was designed to capture target regions containing exonic and regulatory regions (i.e., introns, splice sites, 3′- and 5′-untranslated regions, and promotors) of 11 genes related to AMPK signaling, including PRKAA1, PPARG, IL-6, LEP, LEPR, ADIPOQ, HNF4A, INS, INSR, SIRT1, and RETN. Target gene DNA was captured following the previous described method (Wei et al., 2011). Target sequencing was performed on an Illumina Hiseq high-throughput sequencing platform using 150-bp double-end sequencing mode.

Raw FastQ data were aligned to the human genome (hg19, University of California-Santa Cruz Genome Browser) using the Burrows-Wheeler aligner. Variant calling was performed with VarScan and Haplotype-Caller from the Genome Analysis Toolkit. After sequencing, variants were filtered on the following terms: (a) common in-reference databases (allele frequency was >0.05 in 1000 Genome Project Han Chinese population); (b) the recall rates of single nucleotide variant typing of case and control samples were >0.8. Phen-2 and SIFT were used to predict the effect of functional variants.

Statistical analysis

Data were analyzed with SPSS 21.0 and presented as means ± standard deviations or number of subjects (percentage). Genotype distribution and allele frequencies between study groups, which were analyzed by chi-squared test, were in accordance with Hardy-Weinberg equilibrium. Three genetic model analyses (dominant, recessive, and additive models) were used to further evaluate associations between SNPs and risk of sICAS. A p < (0.05/44 = 0.0011) was accepted as statistically significant after Bonferroni correction. Subgroup analysis by history of diabetes was performed to evaluate its association with candidate SNPs in these sICAS cases. Differences between allele frequencies and clinical characteristics of allele subgroups between allele carriers and noncarriers were compared using the chi-squared or Student's t-tests, with a p < 0.05 being considered significant.

Results

Clinical characteristics of sICAS patients

Altogether, 400 sICAS patients were recruited. Patient characteristics and laboratory test results are summarized in Table 1. Data from 1007 healthy controls from the GeneSky in-house database are used for comparison.

Table 1.

Characteristics of Symptomatic Intracranial Atherosclerotic Stenosis Patients and Controls

Variables	sICAS (n = 400)
Female, n (%)	150 (37.5%)
Male, n (%)	250 (62.5%)
Age (years)
Mean ± SD (Range)	59.4 ± 11.5 (26-87)
Medical history
Hypertension, n (%)	305 (76.3%)
Type 2 diabetes, n (%)	125 (31.3%)
Hyperlipemia, n (%)	265 (66.3%)
Smoking, n (%)	165 (41.3%)
Alcohol drinking, n (%)	118 (29.5%)
History of stroke, n (%)	69 (17.3%)
Laboratory values
FBG (mM)	6.24
TG (mM)	1.92
TC (mM)	4.55
HDL (mM)	1.15
LDL (mM)	2.84

Control individuals were obtained from GeneSky in-house database, clinical characteristics were not available.

FBG, fasting blood glucose; HDL, high-density lipoproteins; LDL, low-density lipoproteins; sICAS, symptomatic intracranial atherosclerotic stenosis; TC, total cholesterol; TG, triglycerides.

Identification of common variants in sICAS patients

Forty-four common variants of the 11 candidate genes were identified in 400 subjects, including 17 exonic, 15 intronic, 6 3′-untranslated region, 5 splice site, and 1 noncoding RNA exonic variants. Among them, four exonic variants were nonsynonymous (three in LEPR and one in PPARG), whereas one in INS (c.*9C>T) and one in PPARG (c.1431C>T) were predicted to be deleterious. More details for all variants are given in Table 2.

Table 2.

Common Variants Identified in Symptomatic Intracranial Atherosclerotic Stenosis Patients

Gene	Position	Gene region	Function	Variants^a		SNP ID	Freq_Alt (1000 g)^b	SIFT^c	POLYPhenV2^d
Gene	Position	Gene region	Function	cDNA	Amino acid	SNP ID	Freq_Alt (1000 g)^b	SIFT^c	POLYPhenV2^d
ADIPOQ	Chr3: 186570892	Exonic	Synonymous	c.45T>G	p.G15G	rs2241766	0.279	—	—
ADIPOQ-AS1	Chr3: 186572642	ncRNA_exonic	—	—	—	rs1501298	0.984	—	—
HNF4A	Chr20: 43034693	Splicing	—	c.41-5C>T	—	rs745975	0.168	—	—
HNF4A	Chr20: 43034660	Intronic	—	—	—	rs736824	0.554	—	—
IL6	Chr7: 22768249	Intronic	—	—		rs2066992	0.308	—	—
INS	Chr11: 2181073	3′UTR	—	c.9C>T	—	rs3842752	0.126	D	—
	Chr11: 2181060	3′UTR	—	c.22A>C	—	rs3842753	0.649	T	—
	Chr11: 2182224	Splicing	—	—	—	rs689	0.649	—	—
INSR	Chr19: 7267939	Intronic	—	—	—	rs3745544	0.051	—	—
	Chr19: 7117426	Splicing	—	c.3759-5C>T	—	rs78312382	0.051	—	—
	Chr19: 7163065	Exonic	Synonymous	c.2007C>T	p.F669F	rs2962	0.063	—	—
	Chr19: 7150491	Intronic	—	—	—	rs41509747	0.070	—	—
	Chr19: 7141775	Exonic	Synonymous	c.2559C>T	p.N853N	rs2229431	0.086	—	—
	Chr19: 7152703	Intronic	—	—	—	rs3745548	0.136	—	—
	Chr19: 7184518	Exonic	Synonymous	c.783C>T	p.D261D	rs891087	0.138	—	—
	Chr19: 7163214	Splicing	—	c.1862-4A>G	—	rs2245649	0.159	—	—
	Chr19: 7163154	Exonic	Synonymous	c.1918C>T	p.L640L	rs2963	0.159	—	—
	Chr19: 7163140	Exonic	Synonymous	c.1932A>C	p.P644P	rs2245655	0.160	—	—
	Chr19: 7166138	Intronic	—	—	—	rs3815902	0.204	—	—
	Chr19: 7184651-7184651	Splicing	—	c.653-3->TC	—	rs3835070	0.233	—	—
	Chr19: 7166376	Exonic	Synonymous	c.1650G>A	p.A550A	rs2059806	0.244	—	—
	Chr19: 7167951	Intronic	—	—	—	rs2860177	0.247	—	—
	Chr19: 7166388	Exonic	Synonymous	c.1638C>T	p.D546D	rs2229429	0.256	—	—
	Chr19: 7163230	Intronic				rs2245648	0.271	—	—
	Chr19: 7125297	Exonic	Synonymous	c.3219C>T	p.H1073H	rs1799817	0.292	—	—
	Chr19: 7166109	Intronic	—	—	—	rs2059807	0.614	—	—
	Chr19: 7116963	3′UTR	—	c.104A>G	—	rs1051690	0.877	—	—
LEPR	Chr1: 66081664	Intronic	—	—	—	rs13306520	0.079	—	—
	Chr1: 66099013	3′UTR	—	c.161C>A	—	rs17415296	0.097	—	—
	Chr1: 66075952	Exonic	Nonsynonymous	c.1968G>C	p.K656N	rs1805094	0.142	T	P
	Chr1: 66102755-66102755	3′UTR	—	c.57_58insTTACT	—	rs75054066	0.236	—	—
	Chr1: 66067109	Exonic	Synonymous	c.1029T>C	p.S343S	rs1805134	0.253	—	—
	Chr1: 66036441	Exonic	Nonsynonymous	c.326A>G	p.K109R	rs1137100	0.320	T	D
	Chr1: 66102257	Exonic	Synonymous	c.3057G>A	p.P1019P	rs1805096	0.529	—	—
	Chr1: 66088701	Intronic	—	—	—	rs12067936(#)	0.563	—	—
	Chr1: 66058513	Exonic	Nonsynonymous	c.668A>G	p.Q223R	rs1137101	0.584	T	B
PPARG	Chr3: 12393125	Exonic	Nonsynonymous	c.34C>G	p.P12A	rs1801282	0.070	D	B
PPARG	Chr3: 12475557	Exonic	Synonymous	c.1431C>T	p.H477H	rs3856806	0.126	—	—
PRKAA1	Chr5: 40777531-40777531	Intronic	—	—	—	rs144157386	0.360	—	—
RETN	Chr19: 7735297	3′UTR	—	c.62G>A	—	rs3745368	0.090	—	—
	Chr19: 7734814	Intronic	—	—	—	rs34788323	0.095	—	—
	Chr19: 7734369	Intronic	—	—	—	rs3219177	0.165	—	—
SIRT1	Chr10: 69666598	Exonic	Synonymous	c.994T>C	p.L332L	rs2273773	0.127	—	—
SIRT1	Chr10: 69651125	Intronic	—	—	—	rs7896005	0.324	—	—

Nucleotide numbering is based on the DNA reference sequence NM_004797, NM_001287184, NM_001185098, NM_001079817, NM_001003679, NM_002303, NM_015869, NM_001193374, and NM_012238.

Frequency of Alt allele in 1000 Genome Project.

SIFT score prediction: D, damaging; T, tolerated.

POLYPHEN score prediction: B, benign; P, possibly damaging; D, probably damaging; —, not available.

Distribution of common variants in sICAS and control subjects

Genotypic and allelic frequency distributions of the 44 SNPs are given in Supplementary Tables S1 and S2. Genotype distributions of the 44 SNPs appeared compatible with Hardy-Weinberg equilibrium in control subjects (data not shown). Chi-squared tests revealed significant differences in genotype or/and allele frequency of eight SNPs in four genes between controls and sICAS patients (Table 3). However, only the locus rs78312382 in INSR was differently distributed between the two groups after Bonferroni correction, with the frequency of allele A being 0.219 in the sICAS group and 0.141 in the control (p_Bonferroni = 0.000404, odds ratio = 1.705, 95% confidence interval = 1.383-2.102) and a GG/GA/AA genotype distribution of 0.593/0.373/0.032 in sICAS patients and 0.740/0.237/0.022 in controls (p = 5.007e⁻⁷). There were more GA and AA carriers in the sICAS group than in the control.

Table 3.

Genotype or Allele Frequency of Common Variants in Symptomatic Intracranial Atherosclerotic Stenosis Patients and Controls

Gene	SNP ID	Allele			OR (95% CI)	p ^a	p _Bonferroni ^b	Genotype^c			p ^a
Gene	SNP ID	Ref/Alt	Case Ref/Alt	ControlRef/Alt	OR (95% CI)	p ^a	p _Bonferroni ^b	Wt/Het/Hom	Case	GENESKY	p ^a
ADIPOQ-AS1	rs1501298	G/A	0/800	32/1728	—	0.000124	0.103	GG/GA/AA	0/0/400	0/32/848	0.000112
HNF4A	rs745975	C/T	664/136	1612/402	0.821 (0.662-1.018)	0.07165	1.000	CC/CT/TT	270/124/6	656/300/51	0.00928
INSR	rs3745544	G/A	624/176	1640/374	1.237 (1.011-1.513)	0.0385	1.000	GG/GA/AA	241/142/17	671/298/38	0.0757
	rs78312382	G/A	623/175	1724/284	1.705 (1.383-2.102)	4.895e⁻⁷	0.000404	GG/GA/AA	237/149/13	743/238/23	5.007e⁻⁷
	rs3835070	-/GA	759/23	1887/107	0.534 (0.337-0.845)	0.00652	1.000	-/-GA/GAGA	368/23/0	890/107/0	0.00527
	rs2229429	G/A	763/37	1867/147	0.616 (0.425-0.891)	0.00964	1.000	GG/GA/AA	364/35/1	863/141/3	0.0265
	rs1799817	G/A	436/364	1214/796	1.273 (1.079-1.502)	0.00416	1.000	GG/GA/AA	117/202/81	376/462/167	0.0121
LEPR	rs17415296	C/A	640/24	1727/111	0.583 (0.371-0.915)	0.0177	>0.0011	CC/CA/AA	309/22/1	812/103/4	0.0539

The bold values are statistically significant.

p values were calculated using a chi-square analysis, threshold of statistical significance was set at p < 0.05.

p_Bonferroni < 0.05/44 = 0.0011 was considered statistically significant.

Genotype presented as wild type/heterozygous/homozygous.

Alt, alternate allele(s); CI, confidence interval; OR, odds ratio; Ref, reference allele(s).

Further logistic regression analysis under the genetic models (additive, dominant, and recessive) showed that INSR rs3745544, rs78312382, rs3835070, rs2229429, and rs1799817 and LEPR rs17415296 were significantly associated with sICAS in dominant and additive models, whereas HNF4A rs745975 correlated with sICAS in the recessive model. However, only rs78312382 still exhibited significance in both dominant and additive models after Bonferroni correction (p_Bonferroni = 7.874e⁻⁵ and 0.000506, respectively, Table 4).

Table 4.

Logistic Regression Analysis for Common Variants of Genes Involved in AMP-Activated Protein Kinase Pathway

	SNP ID	Dominant			Recessive			Additive
	SNP ID	p ^a	OR (95% CI)	p _Bonferroni ^b	p^a	OR (95% CI)	p _Bonferroni ^b	p ^a	OR (95% CI)	p _Bonferroni ^b
HNF4A	rs745975	0.400	0.8999 (0.7035-1.151)	1.000	0.00401	0.2855 (0.1215-0.6706)	1.000	0.0757	0.8251 (0.6673-1.02)	1.000
INSR	rs3745544	0.0239	1.318 (1.037-1.674)	1.000	0.677	1.132 (0.6312-2.03)	1.000	0.0394	1.236 (1.01-1.511)	1.000
	rs78312382	9.545e⁻⁸	1.946 (1.524-2.485)	7.874e⁻⁵	0.303	1.436 (0.7204-2.865)	1.000	6.128e⁻⁷	1.722 (1.391-2.132)	0.000506
	rs3835070	0.00599	0.5199 (0.326-0.8289)	1.000	NA	NA	NA	0.00599	0.5199 (0.326-0.8289)	1.000
	rs2229429	0.00779	0.5927 (0.4032-0.8713)	1.000	0.879	0.8388 (0.08699-8.087)	1.000	0.00944	0.609 (0.4188-0.8856)	1.000
	rs1799817	0.00391	1.446 (1.126-1.857)	1.000	0.171	1.274 (0.9486-1.711)	1.000	0.00475	1.267 (1.075-1.492)	1.000
LEPR	rs17415296	0.0170	0.5649 (0.3534-0.903)	1.000	0.741	0.6911 (0.07697-6.205)	1.000	0.0204	0.5885 (0.3759-0.9213)	1.000

The bold values are statistically significant.

p values were calculated using a logistic analysis in dominant model, recessive model, and additive model, respectively. Threshold of statistical significance was p < 0.05.

p_Bonferroni < 0.05/44 = 0.0011 was considered statistically significant.

Subgroup analysis based on history of diabetes

As INSR mediates the insulin function of target cells, and defective INSR may lead to insulin resistance and eventually type 2 diabetes, the potential relationship between INSR rs78312382 and diabetes in sICAS cases was examined. The results revealed no difference in genotype or allele distribution of rs78312382 in patients with or without diabetes, nor was there a significant difference after further analysis in any of the genetic models (Table 5).

Table 5.

Subgroup Analysis Based on History of Diabetes Mellitus

Gene	SNP ID	Genetic statistical analysis			OR (95% CI)	p ^a
INSR	rs78312382	Chi-square analysis
		Allele			1.076 (0.792-1.462)	0.688
			Diabetes	Nondiabetes
		G	193	430
		A	57	118
		Genotype			—	0.208
		GG	70	167
		GA	53	96
		AA	2	11
		Logistic analysis
		Dominant models			1.226 (0.799-1.882)	0.351
		Recessive models			0.389 (0.085-1.781)	0.224
		Additive models			1.084 (0.744-1.580)	0.674

Threshold of statistical significance was set at p < 0.05.

Clinical characteristics of patients with and without allele A of rs78312382

The SNP rs78312382 was detected in 399 of the 400 sICAS patients. Among them, 162 carried allele A. Further analysis revealed there were no statistical differences between allele A carriers and noncarriers among the risk factors of sICAS (age, drinking, smoking, history of stroke, hypertension, diabetes, and hyperlipemia) or laboratory test results (fasting blood glucose, triglycerides, total cholesterol, and high- and low-density lipoproteins) (Table 6).

Table 6.

Clinical Characteristics of Symptomatic Intracranial Atherosclerotic Stenosis Patients with and without A Allele of rs78312382

Characteristics	Patients without A allele of rs78312382 (n = 237)	Patients with A allele of rs78312382 (n = 162)	p ^a
Female, n (%)	85 (35.9%)	64 (39.5%)	0.460
Age (years)
Mean ± SD	58.8 ± 11.6	60.3 ± 11.1	0.199
Medical history
Hypertension, n (%)	176 (74.3%)	128 (79.0%)	0.274
Type 2 diabetes, n (%)	70 (29.5%)	55 (33.9%)	0.350
Hyperlipemia, n (%)	155 (65.4%)	110 (67.9%)	0.604
History of stroke, n (%)	41 (17.2%)	28 (17.2%)	0.996
Smoking, n (%)	99 (41.8%)	65 (40.1%)	0.742
Alcohol drinking, n (%)	66 (27.8%)	52 (31.1%)	0.361
Laboratory values
FBG (mM)	6.07	6.50	0.089
TG (mM)	2.01	1.78	0.113
TC (mM)	4.50	4.61	0.405
HDL (mM)	1.15	1.15	0.988
LDL (mM)	2.80	2.90	0.317

Continuous variable was analyzed through t-test, and categorical variable was analyzed by chi-square test. Threshold of statistical significance was set at p < 0.05.

Discussion

Numerous studies have reported the importance of AMPK in the occurrence and development of AS (Xiong et al., 2014; Hwang et al., 2017; Jojima et al., 2017; Ma et al., 2017; Wang et al., 2017). Since atherogenesis has been identified as the most common etiology of sICAS in different ethnic populations, this study investigated whether AMPK also participates in the pathophysiological process of sICAS through sequencing exonic and regulatory regions of genes and SNPs involved in AMPK signaling. Overall, the present results revealed 44 common variants in a population of Han Chinese subjects.

INSR, located at 19p13.2, plays a pivotal role in insulin signaling pathways, like Ras/mitogen-activated protein kinase and phosphatidylinositol-3-kinase-Akt/protein kinase B, which play an important role in cell growth and differentiation and insulin metabolic function (Taniguchi et al., 2006). INSR mutations are the most common cause of insulin resistance, which is found in atherogenesis.

To date, there have been no clinical reports regarding the relationships between INSR variants and sICAS. The results of this study indicated, for the first time, significant differences in allele and genotype distribution of INSR rs78312382 between sICAS patients and healthy controls, suggesting the presence of rs78312382 in the splicing region of INSR is closely related to sICAS in Han Chinese individuals. In addition, the A allele of this SNP was more frequently detected in sICAS patients than in healthy controls, whereas the GA/AA genotype was associated with significant risk of sICAS under the dominant model. These results indicate that allele A of rs78312382 may be a genetic risk factor for sICAS.

However, the underlying mechanism of this association remains unclear. Downregulation of INSR signaling has been reported to promote endoplasmic reticulum stress, inhibit sarco/endoplasmic reticulum calcium adenosine triphosphatase, lead to calcium overload, and then cause necrosis of foam cells (Ozcan and Tabas, 2016). Furthermore, a recent study highlighted that overexpression of INSR promotes platelet-derived growth factor-stimulated smooth muscle cell proliferation and migration, thereby accelerating atherogenesis (Bi et al., 2016).

Other than the correlation with AS already mentioned, insulin resistance is also a hallmark of type 2 diabetes mellitus, and an association between INSR variants and diabetes has been identified previously (Defronzo, 2010; Bodhini et al., 2012). In particular, the T allele of rs1799817 was found to be higher in controls than in patients suffering from diabetes in south India, illustrating that the T allele of INSR rs1799817 plays a protective role against diabetes (Bodhini et al., 2012).

Since diabetes is a non-negligible risk factor for sICAS and considering the role of INSR SNPs in diabetes (Huang et al., 2016), the correlation between INSR SNPs and sICAS is presumed to depend on diabetes in this middle phenotype. However, the results of this study showed there was no significant difference in allele A frequency in sICAS patients with and without diabetes. Similarly, the rate of diabetes showed no significant difference between allele A rs78312382 carriers and noncarriers in this study. Therefore, the association between rs78312382 and risk of sICAS could not be explained by the intermediating effect of diabetes.

In addition, SNP rs78312382 located in the splice region of INSR alone would not be functional in the regulation of the sICAS phenotype. Nevertheless, it might be in near-complete or complete linkage disequilibrium with an as of yet unidentified functional variant. Hence, further studies focused on this gene and its function are needed.

Apart from rs78312382, differences in genotype and/or allele frequency of HNF4A rs745975, ADIPOQ-AS1 rs1501298, LEPR rs17415296, and INSR rs3745544, rs3835070, rs2229429, and rs1799817 were also detected for the first time between sICAS patients and healthy controls.

Since HNF4A rs745975 has been shown to be associated with metabolic syndrome (Stephanie-May et al., 2009; Marcil Valérie et al., 2015), the role of rs745975 in AS is thought to depend on its contribution to metabolic disorder. However, this speculation needs further investigation. Accumulating evidence also supports the correlation between ADIPOQ or LEP SNPs and AS in diverse human populations (Turgut et al., 2016; Cui et al., 2017). Although no association has been detected between rs1501298/rs17415296 and AS previously, this may be due to differences in inclusion criteria and/or ethnical genetic heterogeneity.

Furthermore, logistic regression analysis results herein revealed a link between these SNPs (except rs1501298) and sICAS in the three genetic models. Although these links were no longer significant after Bonferroni correction, this kind of association with genetic susceptibility to sICAS should not be overlooked and requires further verification in a larger sample size.

There are limitations to this study. First, data for healthy control individuals were obtained from the GeneSky in-house database, and their clinical characteristic information was not available. Therefore, the influence of environmental risk factors could not be completely eliminated. Second, this was a single-center study with a relatively limited sample size and ethnic group.

In summary, this study detected 44 single nucleotide variants of genes involved in AMPK signaling, and common variants in 8 loci of 4 genes were possibly related to sICAS. However, only INSR rs78312382 remained significantly associated with sICAS after Bonferroni correction. Moreover, this is the first study to investigate the association between allele A of INSR rs78312382 and risk of sICAS. These results provide an important basis for uncovering and understanding the genetic background of sICAS, although such relationships require further exploration in other ethnic populations in future.

Footnotes

Acknowledgments

The authors thank all of the patients for their participation in the project.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This project was supported by the National Natural Science Foundation of China (Grant Nos. 81671166, 81571151, 81601140, and 81641039) and the National Science and Technology Basic Resources Survey Project of China (Grant No. 2018FY100900).

Supplementary Material

References

Beltowski

(2006) Leptin and atherosclerosis. Atherosclerosis, 189:47-60.

, Ding

, He

, et al. (2016) miR-503 inhibits platelet-derived growth factor-induced human aortic vascular smooth muscle cell proliferation and migration through targeting the insulin receptor. Biomed Pharmacother, 84:1711-1716.

Bodhini

, Sandhiya

, Ghosh

, et al. (2012) Association of His1085His INSR gene polymorphism with type 2 diabetes in South Indians. Diabetes Technol Ther, 14:696-700.

Carling

(2007) The role of the AMP-activated protein kinase in the regulation of energy homeostasis. Novartis Foundation Symposium, 286:72-81.

Cui

, Zhou

, Li

, et al. (2017) Association of ADIPOQ single nucleotide polymorphisms with the risk of intracranial atherosclerosis. Int J Neurosci, 127:427-432.

Defronzo

(2010) Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the missing links. The Claude Bernard Lecture 2009. Diabetologia, 53:1270-1287.

Duan

, Zhan

, Song

, et al. (2014) Detection of platelet microRNA expression in patients with diabetes mellitus with or without ischemic stroke. J Diabetes Complications, 28:705-710.

Han

, Liang

, Devries-Seimon

, et al. (2006) Macrophage insulin receptor deficiency increases ER stress-induced apoptosis and necrotic core formation in advanced atherosclerotic lesions. Cell Metab, 3: 257-266.

Hardie

(2014) AMPK—sensing energy while talking to other signaling pathways. Cell Metab, 20:939-952.

10.

Huang

, Li

, Chen

, et al. (2016) Family history of diabetes and the risk of subclinical atherosclerosis. Diabetes Metab, 42:170-177.

11.

Huang

, Li

, Chen

, et al. (2017) The association of circulating MiR-29b and interleukin-6 with subclinical atherosclerosis. Cell Physiol Biochem, 44:1537-1544.

12.

Hwang

, Jung

, Choi

, et al. (2017) Knockdown of sestrin2 increases pro-inflammatory reactions and ER stress in the endothelium via an AMPK dependent mechanism. Biochim Biophys Acta Mol Basis Dis, 1863:1436-1444.

13.

Joint Committee for Developing Chinese guidelines on Prevention and Treatment of Dyslipidemia in Adults. (2016) Chinese guideline for the management of dyslipidemia in adults. Zhonghua Xin Xue Guan Bing Za Zhi, 44:833-853.

14.

Jojima

, Uchida

, Akimoto

, et al. (2017) Liraglutide, a GLP-1 receptor agonist, inhibits vascular smooth muscle cell proliferation by enhancing AMP-activated protein kinase and cell cycle regulation, and delays atherosclerosis in ApoE deficient mice. Atherosclerosis, 261:44-51.

15.

Kantorova

, Chomova

, Kurca

, et al. (2011) Leptin, adiponectin and ghrelin, new potential mediators of ischemic stroke. Neuro Endocrinol Letters, 32:716-721.

16.

Katsiki

, Mantzoros

, Mikhailidis

(2017) Adiponectin, lipids and atherosclerosis. Curr Opin Lipidol, 28:347-354.

17.

Kitada

, Ogura

, Koya

(2016) The protective role of Sirt1 in vascular tissue: its relationship to vascular aging and atherosclerosis. Aging, 8:2290-2307.

18.

Koo

(2015) The latest information on intracranial atherosclerosis: diagnosis and treatment. Interv Neurol, 4:48-50.

19.

Liao

, Deng

, Dai

, et al. (2017) Rare variants of RNF213 and moyamoya/non-moyamoya intracranial artery stenosis/occlusion disease risk: a meta-analysis and systematic review. Environ Health Prev Med, 22:75.

20.

Liao

, Zhang

, Li

, et al. (2019) Rare RNF213 variants and the risk of intracranial artery stenosis/occlusion disease in Chinese population: a case-control study. BMC Med Genet, 20:55.

21.

Long

, Zhan

, Yuan

, et al. (2017) The expression of microRNA-223 and FAM5C in cerebral infarction patients with diabetes mellitus. Cardiovasc Toxicol, 17:42-48.

22.

, Wang

, Yang

, et al. (2017) AMPK activation enhances the anti-atherogenic effects of high-density lipoproteins in apoE−/− mice. J Lipid Res, 58:1536-1547.

23.

Marcil

Valérie

, Devendra

, Seidman

, et al. (2015) Hepatocyte nuclear factor 4 alpha polymorphisms and the metabolic syndrome in French-Canadian Youth. PLoS One 10:. e0117238.

24.

Ozcan

, Tabas

(2016) Calcium signalling and ER stress in insulin resistance and atherosclerosis. J Intern Med, 280:457-464.

25.

Pan

, Meng

, Jing

, et al. (2017) Association of multiple infarctions and ICAS with outcomes of minor stroke and TIA. Neurology, 88:1081-1088.

26.

Park

, Li

, Evcimen

, et al. (2018) Exogenous insulin infusion can decrease atherosclerosis in diabetic rodents by improving lipids, inflammation, and endothelial function. Arterioscler Thromb Vasc Biol, 38:92-101.

27.

Stephanie-May

, John

, Tuomo

, et al. (2009) Interaction between HNF4A polymorphisms and physical activity in relation to type 2 diabetes-related traits: results from the Quebec Family Study. Diabetes Res Clin Pract, 84:211-218.

28.

Taniguchi

, Emanuelli

, Kahn

(2006) Critical nodes in signalling pathways: insights into insulin action. Nat Rev Mol Cell Biol, 7:85-96.

29.

Turgut

, Topsakal

, Ata

, et al. (2016) Leptin receptor gene polymorphism may affect subclinical atherosclerosis in patients with acromegaly. Avicenna J Med Biotechnol, 8:145-150.

30.

Wang

, Ma

, Zhao

, et al. (2017) AMPK activation reduces the number of atheromata macrophages in ApoE deficient mice. Atherosclerosis, 258:97-107.

31.

Wang

, Liu

, Zhai

, et al. (2018) AMPK-mediated regulation of lipid metabolism by phosphorylation. Biol Pharm Bull, 41:985-993.

32.

Wang

, Zhao

, Liu

, et al. (2014) Prevalence and outcomes of symptomatic intracranial large artery stenoses and occlusions in China: the Chinese Intracranial Atherosclerosis (CICAS) Study. Stroke, 45:663-669.

33.

Wei

, Ju

, Yi

, et al. (2011) Identification of sequence variants in genetic disease-causing genes using targeted next-generation sequencing. PLoS One, 6:e29500.

34.

Xiong

, Yepuri

, Forbiteh

, et al. (2014) ARG2 impairs endothelial autophagy through regulation of MTOR and PRKAA/AMPK signaling in advanced atherosclerosis. Autophagy, 10:2223-2238.

35.

Yang

, Xu

, Ma

, et al. (2018) PRKAA1/AMPKα1-driven glycolysis in endothelial cells exposed to disturbed flow protects against atherosclerosis. Nat Commun, 9:4667.

36.

, Lu

, Li

, et al. (2017) Correlation of plasma vascular endothelial growth factor and endostatin levels with symptomatic intra- and extracranial atherosclerotic stenosis in a Chinese Han population. J Stroke Cerebrovasc Dis, 26:1061-1070.

37.

Zhan

, Yuan

, Wang

, et al. (2012) Association of lipopolysaccharide-binding protein gene polymorphisms with cerebral infarction in a Chinese population. J Thromb Thrombolysis, 34:260-268.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB