Identifying genetic etiology of ischemic stroke based on pleiotropy of obesity related genes: A sibling study

  • Kun WANG 1 ,
  • Huairong WANG 1 ,
  • Huan YU 1 ,
  • Ruotong YANG 1 ,
  • Liuyan ZHENG 1 ,
  • Jingxian WU 1 ,
  • Xueying QIN 1, 2 ,
  • Tao WU 1, 2 ,
  • Dafang CHEN 1, 2 ,
  • Yiqun WU , 1, 2, * ,
  • Yonghua HU 1, 2
Expand
  • 1. Department of Epidemiology and Biostatistics, Peking University School of Public Health, Beijing 100191, China
  • 2. Key Laboratory of Epidemiology of Major Diseases (Peking University), Ministry of Education, Beijing 100191, China
WU Yiqun, e-mail,

Received date: 2024-11-15

  Online published: 2025-06-13

Supported by

the National Natural Science Foundation of China(81703291)

Copyright

All rights reserved. Unauthorized reproduction is prohibited.

Abstract

Objective: To identify genetic etiology of ischemic stroke (IS) based on pleiotropy of obesity related genes. Methods: A discordant sib-pair study was designed based on the Fangshan family cohort in Beijing. Body mass index (BMI) polygenic risk score (PRS) was first constructed under different P values. Using the polygenic transmission disequilibrium test (pTDT), we then compared the actual BMI genetic risk of siblings with IS to their expected risk, to analyze whether higher BMI was over-transmitted to siblings with IS. The single nucleotide polymorphism (SNP) that comprised the PRS over-transmitted with IS and that corresponded to the highest heritability of IS were identified as a pleiotropy SNPs set between BMI and IS. This set was then utilized as a candidate set to identify and verify risk SNPs asso-ciated IS by transmission disequilibrium test. Finally, we identified independent genomic risk loci and mapped to genes, we then explored the biological function of the identified risk loci and genes by functional annotation and pathway enrichment. Results: A total of 541 participants were enrolled, with an average age of (58.4±8.1) years, including 326 discordant sib pairs of ischemic stroke. Compared with non-IS participants, IS participants with males, education level below junior high school, hypertension and hyperlipidemia accounted for a higher proportion (P < 0.05). For all the BMI PRS, we found that the actual genetic risk of BMI in siblings with IS was higher than their expectation, suggesting that genetic risk associated with high BMI was over-transmitted with IS. Compared with other SNP sets, the set (P < 5×10-4) corresponded to the best analytical statistics of pTDT and the highest heritability of IS and was identified as the pleiotropy SNP set between BMI and IS. Within this set, there were 45 SNPs having linkage and association with IS, which were located in 43 independent genomic risk loci and mapped to 40 genes. These genes were significantly enriched in the lipid metabolism pathway. The rs2232852 corrected by multiple tests was mapped to CYB5R1 and ADIPOR1, which were related to lipid metabolism and the ferroptosis pathway. Conclusion: Pleiotropy between BMI-related genes and IS was observed. Forty-five SNPs were found with linkage and association with IS in the pleiotropy gene set and mapped to 40 genes, which were functionally enriched in lipid metabolic pathways. The rs2232852 corrected by multiple tests during association analysis validation was mapped to CYB5R1 and ADIPOR1, which were related to lipid metabolism and the ferroptosis pathway, suggesting that lipid metabolism and ferroptosis played an important role in the development of IS.

Cite this article

Kun WANG , Huairong WANG , Huan YU , Ruotong YANG , Liuyan ZHENG , Jingxian WU , Xueying QIN , Tao WU , Dafang CHEN , Yiqun WU , Yonghua HU . Identifying genetic etiology of ischemic stroke based on pleiotropy of obesity related genes: A sibling study[J]. Journal of Peking University(Health Sciences), 2025 , 57(3) : 448 -455 . DOI: 10.19723/j.issn.1671-167X.2025.03.007

脑卒中是全球成人重要的致死和致残原因[1],缺血性脑卒中(ischemic stroke,IS)占所有脑卒中的60%~90%[2]。IS是由遗传和环境因素共同影响的复杂疾病[3-4],其遗传度约为30%~40%[5]。目前已识别与IS相关的遗传风险位点近百个[6-9],其中部分IS遗传风险位点与其危险因素同时存在关联[6, 10],即存在遗传多效性。遗传多效性在人类复杂疾病中普遍存在,指单个遗传变异与两个或以上性状同时存在关联[11]。利用遗传多效性开展遗传病因研究,已广泛应用在精神类、癌症、免疫和炎症等具有遗传相关的疾病中[12-15]。肥胖是IS重要的危险因素之一,体重指数(body mass index, BMI)作为肥胖的重要指标,与IS存在较强的遗传相关性。利用遗传变异位点在BMI和IS两个性状中效应的富集作用,能够提高分析遗传变异与IS关联关系的研究效率。目前大部分基于遗传多效性的关联分析采用一般人群设计,人群分层较难很好地控制[16]。家系资料中,研究对象的遗传背景均一,不仅能够提供更丰富的遗传信息,因人群分层出现假阳性关联的可能性也较低[17]。针对家系资料,Weiner等[18]根据经典的传递不平衡检验方法进行扩展,提出了多基因传递不平衡检验(polygenic transmission disequilibrium test, pTDT),用于探索精神分裂症相关基因与孤独谱系障碍的遗传关联。相较于既往IS遗传病因研究,遗传多效性分析方法可提高研究效率,本研究利用这一家系遗传多效性分析方法,采用表型不一致同胞对研究设计[19],探索BMI相关基因与IS的关联,然后在与BMI和IS存在多效性关联的基因集中,进一步识别IS的遗传风险位点,以期为理解IS遗传病因提供科学依据。

1 资料与方法

1.1 研究设计

采用表型不一致同胞对设计,通过比较患IS者实际遗传风险和其未患病同胞估计的期望遗传风险,探索BMI相关基因与IS的关联关系。

1.2 研究对象

研究资料来源于“北方农村地区居民常见慢性非传染性疾病家系队列研究”,项目现场位于北京市房山区的周口店镇、青龙湖镇等9个乡镇,研究资料于2013—2017年收集完成,该家系队列研究已获得北京大学生物医学伦理委员会批准(IRB00001052-13027)。
本研究选择家系队列中IS患者及其未患病同胞作为研究对象,其中IS同胞的纳入标准为:(1)年龄≥18岁;(2)经二级以上医院明确诊断为IS;(3)至少存在一个未患IS的全同胞;(4)基线调查资料完整;(5)基线外周血生物样本保存完好;(6)研究对象自愿参加研究并且签署知情同意书;(7)已完成全基因组芯片检测。排除标准为:(1)有重大疾病或严重慢性病者,或行动不便不能配合调查者;(2)同胞IS诊断不明确;(3)同胞不愿意参加。未患病同胞除未患IS外,纳入、排除标准和IS患者相似。
本研究纳入541例表型不一致同胞,来自226个两人及以上同胞对家系,将同一个家系内的一位IS患者和非患者组成一对表型不一致同胞,共组成了326对表型不一致同胞。本研究还将家系队列中未构成表型不一致同胞对的696例研究对象作为验证人群,采用病例对照设计,对在表型不一致同胞对中发现的与IS存在连锁和关联的位点进行关联分析验证。
采用标准调查问卷收集研究对象的一般人口学资料、行为生活习惯、疾病史、家族史和用药史等信息。体格检查、临床检测及血生化检测指标由经统一培训的专业人员进行。

1.3 IS诊断信息的收集

IS的诊断信息由现场问卷调查、医院病历资料核查、影像学检查等多种途径进行多重确认。现场问卷调查询问研究对象经二级及以上医院确诊IS的情况,由工作人员根据调查中获得的IS患者就诊信息到相应医院查阅患者病历,摘抄入院和出院诊断,并记录各种辅助检查结果(包括头颅CT和/或MRI检查、经颅多普勒超声检查、颈动脉超声检查、颈椎放射影像检查、心电图、彩色多普勒超声心动图检查等)。未患IS疾病的研究对象需经过非卒中状态确证问卷确认[20]

1.4 基因组信息的收集

采用盐析法提取外周血DNA,采用紫外分光光度法测定DNA浓度,采用Illumina ASA芯片测定基因型,共得到74万单核苷酸多态性位点(single nucleotide polymorphism, SNP)。
对基因分型信息进行质量控制,包括样本质量控制和SNP质量控制两个步骤。
样本质量控制:(1)修正与问卷调查性别不一致的样本;(2)剔除基因分型成功率 < 5%的样本;(3)标记杂合度偏离>3倍标准差的样本;(4)根据血缘一致性修正与问卷调查亲缘关系不一致的样本;(5)剔除孟德尔误差率>5%的样本;(6)剔除明显偏离祖先遗传背景的样本。
SNP质量控制:(1)剔除缺失率>5%的SNP;(2)剔除在IS病例和对照中缺失率不同(P < 1×10-5)的SNP;(3)剔除不满足哈迪-温伯格(Hardy-Weinberg)平衡(1×10-6)的SNP;(4)剔除最小等位基因频率 < 1%的SNP;(5)剔除孟德尔误差率>10%的SNP。共有50万个SNP通过质量控制。
采用Shapeit2软件[21]进行基因定相,采用IMPUTE 2软件[22]以千人基因组计划[23]三期数据为参考模板进行基因型填补。最终521万个SNP通过填补后质量控制,纳入后续分析。

1.5 BMI多基因评分的构建

采用聚类处理和阈值设定法,选择基因组500 kb范围内独立位点(r2≥0.1),利用2023年来自亚洲人群的GWAS研究[24]中报道的SNP与BMI之间关联效应值作为权重,构建研究对象的BMI多基因评分。选择通过全基因组显著性阈值(5×10-8)的SNP构建多基因风险评分(polygenic risk score, PRS),评估BMI相关遗传风险与IS的关联关系,分别选择P值小于5×10-8、5×10-7、5×10-6、5×10-5、5×10-4、5×10-3的位点构建多基因评分,对应的SNP集合作为BMI相关基因集合,之后进一步确定与BMI和IS存在遗传多效性的最佳遗传变异集合,以上过程在Plink 1.9[25]软件中完成。

1.6 统计学分析

1.6.1 统计描述

采用R 4.4.1统计软件,对研究对象的一般人口学特征、体格检查、临床检查和血生化检测指标进行描述性分析,连续性变量以均数±标准差表示,正态分布数据采用t检验进行组间差异比较,非正态分布数据采用Wilcoxon秩和检验进行组间差异比较;分类变量以频数(百分比)表示,采用卡方检验。

1.6.2 筛选与BMI和IS存在遗传多效性的SNP集合

采用pTDT分析BMI相关基因与IS的多效性关系,该方法的基本思想为每个个体均有相同的机会从亲代继承某一等位基因,若某等位基因与IS患病无关,那么患IS个体与其未患IS同胞继承亲代该等位基因的机会相同。针对BMI多基因风险,若其与IS患病无关,则患IS个体的多基因风险期望值应与未患IS同胞相同,患IS个体的实际多基因风险与其未患病同胞无差异。首先计算表型不一致同胞对中患IS者与不患IS者不同P值下BMI PRS的差值[偏差(deviation, DEV)][18],然后比较DEV与0是否存在差异(P < 0.1为差异有统计学意义),进而判断该多基因风险是否与BMI和IS存在多效性。如果P < 0.1,认为IS患病状态与BMI的多基因风险具有共同向下传递的倾向,说明BMI的PRS与IS相关,具有遗传多效性,以上分析通过GCTA[26] R 4.4.1软件完成。DEV的计算公式为:$ \mathrm{DEV}=\frac{P R S_{\mathrm{IS}}-P R S_{\mathrm{nolS}}}{S D\left(P R S_{\mathrm{NIS}}\right)}$PRSIS是指患IS者BMI PRS,PRSnoIS是指未患病同胞BMI PRS,SD(PRSNIS)是指样本中非IS患者BMI PRS的标准差。计算不同P值PRS对应的SNP对IS遗传度的贡献(hSNP2),通过线性回归评估PRS对IS表型的解释程度(R2)。选择pTDT检验的显著性水平最优、对IS的遗传度解释最大、回归拟合程度最佳的SNP集合作为候选分析集,进一步分析单个SNP与IS的关系。

1.6.3 单位点SNP与IS的关联分析与验证

同胞对传递不平衡检验是通过检测在同胞对中,特定遗传标记的等位基因从父母传递给表型不一致同胞是否存在不平衡的现象,用以判断相关基因和疾病是否存在连锁和关联。本研究在IS表型不一致同胞中,通过同胞对传递不平衡检验,分析BMI与IS的遗传多效性变异集合中的SNP与IS是否同时存在连锁和关联,分析的显著性阈值设置为P < 0.05。将识别的与IS存在连锁和关联的位点进一步在验证人群中通过Logistic回归进行关联分析验证,分析调整年龄、性别和前两个主成分,通过Bonferroni进行多重检验校正(显著性阈值设为0.05/纳入位点个数),以上分析通过Plink 1.9软件完成。

1.6.4 注释和富集

将识别的与IS存在连锁和关联的SNP位点进行独立基因座划分和基因定位,将与识别的SNP存在连锁不平衡(r2≥0.6)的位点所在位置,对连锁不平衡区块距离 < 250 kb者进行合并,划分为独立基因座,然后通过位置进行基因映射,将SNP映射到10 kb窗口内的蛋白编码基因上。根据“基因型-组织表达项目”(第8版)(Genotype-Tissue Expression Project version 8.0, GTEx Ⅴ8),基因在不同组织中的表达水平,对映射基因的差异表达进行分析,根据基因本体论(gene ontology, GO)通路注释和《京都基因与基因组百科全书》(Kyoto Encyclopedia of Genes and Genomes, KEGG)通路注释进行功能富集分析。以上通过Plink 1.9软件、遗传关联的功能映射和注释工具(functional mapping and annotation of genetic associations, FUMA)[27]完成。

2 结果

2.1 研究对象的一般特征

本研究共纳入541例研究对象,来自于226个两人及以上家系,其中IS患者为276例(51.0%)。IS患者的平均年龄、男性比例、初中以下教育程度比例、患高血压比例、患高血脂症的比例均高于非IS患者(P < 0.05,表 1)。研究对象平均BMI为(26.2±3.4) kg/m2,利用通过全基因组显著性阈值(5×10-8)的SNP构建PRS,IS患者BMI PRS的均值为(0.047±0.157)分,高于其非患病同胞的(0.032±0.166)分,但两者差异无统计学意义。
表1 研究对象的基本特征

Table 1 Basic characteristics of participants

Items Total (n=541) Participants with IS (n=276) Participants without IS (n=265) P value
Age/years, ${\bar x}$±s 58.4±8.1 60.1±7.4 56.6±8.5 < 0.001
Male, n (%) 310 (57.3) 171 (62.0) 139 (52.5) 0.032
Married, n (%) 468 (86.5) 236 (85.5) 232 (87.5) 0.872
Junior high school education or above, n (%) 290 (53.6) 136 (49.3) 154 (58.1) 0.048
Smoker, n (%) 287 (53.1) 147 (53.3) 140 (52.8) 0.957
Drinker, n (%) 243 (44.9) 131 (47.5) 112 (42.3) 0.227
Adequate exercise, n (%) 159 (29.4) 86 (31.2) 73 (27.5) 0.193
BMI/(kg/m2), ${\bar x}$±s 26.2±3.4 26.2±3.3 26.3±3.6 0.637
Diabetes, n (%) 172 (31.8) 94 (34.1) 78 (29.4) 0.288
Hypertension, n (%) 276 (51.0) 158 (57.2) 118 (44.5) 0.004
Hyperlipidemia, n (%) 179 (33.1) 114 (41.3) 65 (24.5) < 0.001
BMI-PRS, ${\bar x}$±s 0.040±0.162 0.047±0.157 0.032±0.166 0.284

IS, ischemic stroke; BMI, body mass index; BMI-PRS, BMI polygenic risk score which SNP with P < 5×10-8 were included to calculate PRS;SNP, single nucleotide polymorphism; PRS, polygenic risk score.

2.2 识别与BMI和IS存在遗传多效性的SNP集合

在326对IS表型不一致同胞对中,利用pTDT分析确定与BMI和IS存在遗传多效性的SNP集合。选择不同P值范围下的SNP集合构建PRS,IS患者的BMI实际遗传风险与其预期风险之间的差值(DEV)均大于0,提示与高BMI水平相关的遗传风险与IS患病状态存在共同传递的倾向。比较不同P值筛选的基因集度量的BMI PRS与IS关联分析结果,P值小于5×10-4构建的PRS得分pTDT检验的显著性水平最优,该集合中包含的3 031个独立SNP对IS的遗传度解释最大、回归拟合程度最优(表 2),选择该集合作为与BMI和IS存在遗传多效性的SNP集合,进行SNP位点分析。
表2 BMI多基因风险与IS的关联

Table 2 Association of BMI polygenic risk with IS

P SNPn DEV (mean) PpTDT hSNPn2 R2
5×10-7 399 0.081 0.126 1×10-6 6.99×10-4
5×10-6 644 0.088 0.095 1×10-6 6.91×10-4
5×10-5 1 244 0.052 0.242 1×10-6 2.06×10-4
5×10-4 3 031 0.080 0.072 2×10-6 1.20×10-3
5×10-3 9 172 0.045 0.314 2×10-6 3.89×10-4

P is the P value for SNP that comprise the PRS; SNPn is the number of SNP filtered under different P; DEV, differences of BMI PRS between siblings with IS and without IS; PpTDT is the significance for the pTDT test; hSNPn2, heritability of SNPn for IS; R2, R-square for liner regression of IS and BMI PRS. IS, ischemic stroke; BMI, body mass index; SNP, single nucleotide polymorphism; PRS, polygenic risk score; pTDT, polygenic transmission disequilibrium test; DEV, deviation.

2.3 分析与IS存在关联的SNP位点

对上述选择的3 031个SNP在表型不一致同胞对中进行传递不平衡检验,发现45个SNP与IS同时存在连锁与关联(P < 0.05,表 3),这些位点中有6个SNP在验证人群中仍显示与IS存在关联(P < 0.05,表 4),其中有1个SNP(rs2232852)通过Bonferroni校正的显著性阈值(P < 0.05/45),它与IS的关联强度OR值为1.53(95%CI:1.21~1.94,P=4.582×10-5)。
表3 与IS同时存在连锁与关联的SNP对应的基因座及其映射基因

Table 3 SNP that are linked and associated with IS corresponded to genomic risk loci and mapped gene

SNP Cytogenetic band Genomic risk loci (chromosome: start position-end position) Gene
rs118074101 1p13.3 1:110750031-110750031 KCNC4, SLC6A7
rs12093350 1q22 1: 155468732-155793969 ASH1L, MSTO1, YY1AP1, DAP3, GON4L
rs2232852 1q32.1 1:202924936-202938778 ADIPOR1, CYB5R1
rs34378983 1q32.3 1:214331416-214404556
rs75747776 2p24.1 2:20409638-20502198 SDC1, PUM2
rs12612799 2p16.1 2:56316415-56385145
rs147755783 2q24.1 2:158031410-158031410
rs1460669 2q24.3 2:164613829-164653060
rs141123347 2q24.3 2:169174780-169174780
rs17053757 3p21.1 3:54169266-54171774 CACNA2D3
rs10025110 4p12 4:44901237-45014151
rs59257388 4q21.22 4:83203517-83270007 HNRNPD
rs3811801 4q23 4:100244319-100336102 ADH1B, ADH1C, ADH7
rs76339045 4q32.3 4:167710607-167710607 SPOCK3
rs13177532 5p13.2 5:36728340-36783545
rs12656015 5q11.2 5:53089622-53160978
rs7723244 5q22.1 5:111261184-111443109 NREP
rs2043478 5q31.2 5:136498493-136498493 SPOCK1
rs181895 5q31.3 5:141769375-141822539
rs2062536 5q34 5:165795362-165795362
rs368399960 6p21.32 6:32300809-32331002 C6orf10
rs13202872 6p12.3 6:51189632-51264082
rs6947395 7q11.22 7:69403462-69406661 AUTS2
rs10268638 7q33 7:137100832-137129484 DKGI
rs6999964 8q24.22 8:132862920-132862920
rs10961656 9p22.3 9:14639666-14694602 ZDHHC21
rs12343952 9q33.1 9:122445599-122497319
rs74829026 10p15.3 10:2298746-2298746
rs117023276 10p12.31 10:21246955-21291331 NEBL
rs138468034 10q22.2 10:77321693-77366416 C10orf11
rs664706 10q23.31 10:89773567-89777068
rs192881652 11p15.1 11:18034532-18034532 TPH1, SERGEF
rs308754 11q14.2 11:88161676-88172516
rs1820460 12q23.3 12:107704705-107704705 BTBD11
rs4477562
rs2785821
13q14.3 13:54091759-54272104
rs11631335 15q15.1 15:40395604-40421006 BMF
rs9646281 16q12.1 16:52264631-52264631
rs75659809 16q23.1 16:77018698-77018698
rs11877418 18q11.2 18:20748733-20787099 CABLES1, TMEM241
rs77247480
rs56080693
18q12.1 18:27163063-27426754
rs148990504 18q21.31 18:54255607-54255607 TXNL1
rs769449 19q13.32 19:45410002-45428234 TOMM40, APOE, APOC1
rs117988645 22q11.21 22:20093967-20174270 DGCR8, TRMT2A, RANBP1, ZDHHC8, AC006547.14

IS, ischemic stroke; SNP, single nucleotide polymorphism.

表4 通过与IS关联分析验证的SNP位点

Table 4 SNP verified by association analysis

SNP Effect allele MAF OR(95%CI) P
rs2232852 C 0.387 1.53 (1.21-1.94) 4.582×10-5
rs1460669 T 0.415 1.37 (1.08-1.73) 0.010
rs12343952 T 0.247 1.42 (1.07-1.90) 0.017
rs75659809 T 0.009 0.25 (0.07-0.85) 0.026
rs141123347 G 0.013 0.32 (0.12-0.88) 0.027
rs11631335 G 0.236 0.77 (0.60-1.00) 0.049

IS, ischemic stroke; SNP, single nucleotide polymorphism; MAF, minor allele frequency.

2.4 与IS存在关联SNP的功能探索

45个与IS存在连锁和关联的SNP位于43个独立基因座,映射在40个基因上(表 3),其中7个基因在既往报道中与IS有关,分别为DAP3YY1AP1ADIPOR1TOMM40APOEAPOC1TPH1,通过多重检验校正仍与IS呈显著性关联的位点rs2232852映射到CYB5R1ADIPOR1基因编码区域。这些基因在尾状基底神经节、脑颈部脊髓、下丘脑、心房、血液和小肠中差异表达(P < 0.05),在脑颈部脊髓、小肠、血液、输卵管和胰腺中低表达(P < 0.05),无显著高表达的组织。这些基因在脂质代谢通路中显著富集(P < 0.05),富集通路包括极低密度脂蛋白颗粒清除、乳糜微粒残余清除、高密度脂蛋白颗粒组装和重塑、磷脂和胆固醇外排、脂蛋白代谢以及脂肪酸降解等通路。其中,关联验证通过多重检验校正的rs2232852位点映射于CYB5R1ADIPOR1基因,与脂质代谢、铁死亡通路有关。

3 讨论

本研究采用表型不一致同胞设计,在与BMI和IS存在遗传多效性的基因集中,分析识别了45个SNP与IS存在连锁和关联,其中6个SNP得到进一步关联分析验证,这些遗传位点映射在40个基因,它们富集于脂质代谢相关通路。
本研究发现,IS患者的BMI PRS高于其未患病同胞,高BMI水平相关的遗传风险与IS患病状态存在共同传递的倾向,BMI遗传风险与IS患病密切相关,这一结果提示BMI与IS之间可能存在共同的遗传基础。肥胖是IS重要的危险因素,BMI是肥胖的重要指标,较高的BMI水平与IS风险增加有关[28],但BMI与IS之间的表型关联是否反映了共同遗传基础,在既往研究中的证据有限。有研究显示,BMI与IS存在显著的遗传相关性,在欧洲裔人群中,二者的遗传相关系数为0.20,在东亚人群中为0.16[6],既往研究发现的部分IS相关风险位点和基因(如12q24、APOE)同时与BMI存在关联[6, 29-30]APOE基因是BMI和IS的遗传多效性基因。在本研究中,通过pTDT发现了与BMI和IS存在遗传多效性的遗传变异集合,为BMI和IS的遗传关联提供了证据。本研究发现的基因与脂质合成和代谢密切相关,在脂质代谢相关通路富集,高脂血症同时是肥胖和IS的重要危险因素[31-32],提示IS和BMI的遗传多效性基因可能通过脂质合成与代谢发挥作用,是BMI和IS遗传关联和共享遗传基础的潜在机制。此外,高BMI与全身炎症反应呈正相关,血浆白细胞数量、促炎蛋白表达和脂质参数的升高[33-34],可能会诱发针对大血管和小血管的炎症状态,对心血管系统产生有害影响[35-36],这也可能是BMI和IS遗传多效性和共同遗传基础的潜在机制。
本研究在与BMI和IS存在遗传多效性的基因集中发现有45个与IS存在连锁和关联的SNP,识别了40个与IS相关的基因,其中7个基因在既往报道中与IS存在关联关系:ADIPOR1 [37]TOMM40 [38]APOE[29]与IS风险相关;DAP3与吸烟人群IS风险有关[39]APOC1与早发IS风险相关[40]YY1AP1突变与格兰奇综合征(Grange syndrome)相关,该综合征表现之一为IS[41]TPH1可作为IS的潜在药物靶点[42],通过调节氧化还原反应保护神经系统。APOE基因在脂质代谢通路富集上发挥重要作用,既往报道其与阿尔茨海默病(Alzheimer disease,AD)、IS及其发病年龄、动脉粥样硬化、脂质代谢紊乱、BMI等表型相关。APOE蛋白是一种载脂蛋白,参与血浆脂蛋白的产生、转化和清除,通过参与脂蛋白介导的组织间脂质分布,在血浆和组织脂质稳态中起关键作用[43]。除了通过脂质水平影响IS外,APOE基因多态性影响IS的机制还可能与炎症有关。既往研究显示,APOE介导胶质细胞炎症,释放可溶性介质到大脑中的炎症细胞,产生高浓度的促炎细胞因子和低浓度的抗炎细胞因子,通过有害的神经炎症表型导致脑血管损伤[44-45]APOE基因参与的脂质代谢、细胞炎症也是BMI和IS遗传多效性的可能机制,为IS发生的潜在机制提供了一定的证据。
本研究发现的与IS存在连锁和关联的SNP中,在验证人群中通过多重检验校正仍与IS呈显著性关联的位点为rs2232852,映射在CYB5R1ADIPOR1基因编码区域。CYB5R1基因参与脂肪酸的去饱和及延长、胆固醇的生物合成以及红细胞中的高铁血红蛋白还原[43]CYB5R1基因与细胞色素P450还原酶一起促进铁死亡过程[46]ADIPOR1基因与脂代谢和糖类代谢有关[43],也参与了铁死亡的过程[37]。铁死亡是一种铁依赖性调节的细胞死亡形式[47],主要与铁代谢、氨基酸代谢和脂质代谢相关[48]。铁代谢紊乱与缺血性脑损伤密切相关,脂质过氧化作为铁死亡的重要特征之一,其标志物在IS患者脑组织中高表达[49]。铁死亡还与动脉粥样硬化、心肌梗死和缺血/再灌注等多种心脑血管疾病密切相关[48, 50]。铁死亡以及参与脂质合成和代谢可能在IS的发生发展中起到一定作用。
本研究仍存在一定的局限性。首先,研究样本量有限,无法进行IS亚型的亚组分析。作为探索性研究,为识别更多可能的IS风险基因,在发现IS连锁与关联SNP的过程中放宽了统计学检验阈值,并未进行Bonferroni校正。未来在扩大样本量的情况下,应考虑不同亚型IS的细致分析,并进行更为严格的验证分析。其次,IS与肥胖存在遗传多效性,基于BMI相关基因遗传多效性能够发现更多的IS遗传风险位点,但本研究仅纳入BMI一种肥胖相关指标进行分析,未来研究可以利用更多与肥胖相关的表型,例如腰臀比、腰围等指标开展类似研究。第三,本研究利用公开数据库对发现的与IS相关的基因进行功能分析,有待通过细胞水平实验或动物模型研究进一步进行相关功能验证,为基因功能提供更直接的证据。
本研究的优势有:(1)采用表型不一致同胞对设计,由于同胞对有着相似的遗传背景,因人群分层而产生的阳性结果能在研究设计过程中较好控制;(2)基于BMI相关基因遗传多效性探索IS的遗传病因,发现了BMI和IS的共享遗传基础和遗传多效性,并利用了遗传多效性基因同时对BMI和IS的效应富集作用而提高风险位点的识别能力。
综上所述,BMI相关基因和IS存在遗传多效性,在多效性基因集中发现了45个IS相关位点,映射到40个基因上,它们的功能富集于脂质代谢相关通路。通过关联验证分析多重检验校正的rs2232852位点对应CYB5R1ADIPOR1基因,这两个基因不仅与脂质代谢相关,也涉及铁死亡通路,这提示了脂质代谢和铁死亡可能是IS发生的潜在机制。本研究不仅仅为IS的遗传病因和可能的发病机制提供新的见解,也为其他复杂疾病的病因探索提供了思路——基于遗传多效性提高复杂疾病遗传病因的研究效率。

利益冲突  所有作者均声明不存在利益冲突。

作者贡献声明  武轶群、胡永华、秦雪英、王坤:提出研究思路;武轶群、秦雪英、王坤:设计研究方案;王坤、王淮蓉、杨若彤、于欢、郑柳燕、吴婧娴:收集、分析、整理数据;王坤、武轶群:撰写论文;王坤、王淮蓉、杨若彤、于欢、郑柳燕、吴婧娴、武轶群、秦雪英、胡永华、陈大方、吴涛:总体把关和审定论文。

1
Global, regional, and national burden of stroke and its risk factors, 1990-2019: A systematic analysis for the Global Burden of Disease Study 2019[J]. Lancet Neurol, 2021, 20(10): 795-820.

2
Campbell BCV , Khatri P . Stroke[J]. Lancet, 2020, 396 (10244): 129- 142.

DOI

3
Black M , Wang W , Wang W . Ischemic stroke: From next generation sequencing and GWAS to community genomics?[J]. OMICS, 2015, 19 (8): 451- 460.

DOI

4
Dichgans M . Genetics of ischaemic stroke[J]. Lancet Neurol, 2007, 6 (2): 149- 161.

DOI

5
Hu Y , Haessler JW , Manansala R , et al. Whole-genome sequencing association analyses of stroke and its subtypes in ancestrally diverse populations from trans-omics for precision medicine project[J]. Stroke, 2022, 53 (3): 875- 885.

DOI

6
Mishra A , Malik R , Hachiya T , et al. Stroke genetics informs drug discovery and risk prediction across ancestries[J]. Nature, 2022, 611 (7934): 115- 123.

DOI

7
Malik R , Chauhan G , Traylor M , et al. Multiancestry genome-wide association study of 520, 000 subjects identifies 32 loci associated with stroke and stroke subtypes[J]. Nat Genet, 2018, 50 (4): 524- 537.

DOI

8
Matarin M , Brown WM , Dena H , et al. Candidate gene polymorphisms for ischemic stroke[J]. Stroke, 2009, 40 (11): 3436- 3442.

DOI

9
Dichgans M , Markus HS . Genetic association studies in stroke: Methodological issues and proposed standard criteria[J]. Stroke, 2005, 36 (9): 2027- 2031.

DOI

10
Holdt LM , Teupser D . Recent studies of the human chromosome 9p21 locus, which is associated with atherosclerosis in human populations[J]. Arterioscler Thromb Vasc Biol, 2012, 32 (2): 196- 206.

DOI

11
Stearns FW . One hundred years of pleiotropy: A retrospective[J]. Genetics, 2010, 186 (3): 767- 773.

DOI

12
Romero C , Werme J , Jansen PR , et al. Exploring the genetic overlap between twelve psychiatric disorders[J]. Nat Genet, 2022, 54 (12): 1795- 1802.

DOI

13
Lu H , Qiao J , Shao Z , et al. A comprehensive gene-centric pleiotropic association analysis for 14 psychiatric disorders with GWAS summary statistics[J]. BMC Med, 2021, 19 (1): 314.

DOI

14
Gong W , Guo P , Li Y , et al. Role of the gut-brain axis in the shared genetic etiology between gastrointestinal tract diseases and psychiatric disorders: A genome-wide pleiotropic analysis[J]. JAMA Psychiatry, 2023, 80 (4): 360- 370.

DOI

15
Wang J , Luo GY , Tian T , et al. Shared genetic basis and causality between schizophrenia and inflammatory bowel disease: Evidence from a comprehensive genetic analysis[J]. Psychol Med, 2024, 54 (10): 2658- 2668.

DOI

16
Price AL , Patterson NJ , Plenge RM , et al. Principal components analysis corrects for stratification in genome-wide association studies[J]. Nat Genet, 2006, 38 (8): 904- 909.

DOI

17
Ott J , Kamatani Y , Lathrop M . Family-based designs for genome-wide association studies[J]. Nat Rev Genet, 2011, 12 (7): 465- 474.

DOI

18
Weiner DJ , Wigdor EM , Ripke S , et al. Polygenic transmission disequilibrium confirms that common and rare variation act additively to create risk for autism spectrum disorders[J]. Nat Genet, 2017, 49 (7): 978- 985.

DOI

19
Laird NM , Blacker D , Wilcox M . The sib transmission/disequilibrium test is a Mantel-Haenszel test[J]. Am J Hum Genet, 1998, 63 (6): 1915- 1916.

20
Meschia JF , Lojacono MA , Miller MJ , et al. Reliability of the questionnaire for verifying stroke-free status[J]. Cerebrovasc Dis, 2004, 17 (2/3): 218- 223.

21
O'Connell J , Gurdasani D , Delaneau O , et al. A general approach for haplotype phasing across the full spectrum of relatedness[J]. PLoS Genet, 2014, 10 (4): e1004234.

DOI

22
Howie BN , Donnelly P , Marchini J . A flexible and accurate genotype imputation method for the next generation of genome-wide association studies[J]. PLoS Genet, 2009, 5 (6): e1000529.

DOI

23
Auton A , Brooks LD , Durbin RM , et al. A global reference for human genetic variation[J]. Nature, 2015, 526 (7571): 68- 74.

DOI

24
Chen CY , Chen TT , Feng YCA , et al. Analysis across Taiwan Biobank, Biobank Japan, and UK Biobank identifies hundreds of novel loci for 36 quantitative traits[J]. Cell Genom, 2023, 3 (12): 100436.

DOI

25
Chang CC , Chow CC , Tellier LC , et al. Second-generation PLINK: Rising to the challenge of larger and richer datasets[J]. Gigascience, 2015, 4, 7.

26
Yang J , Lee SH , Goddard ME , et al. GCTA: A tool for genome-wide complex trait analysis[J]. Am J Hum Genet, 2011, 88 (1): 76- 82.

DOI

27
Watanabe K , Taskesen E , van Bochoven A , et al. Functional mapping and annotation of genetic associations with FUMA[J]. Nat Commun, 2017, 8 (1): 1826.

DOI

28
Kim MS , Kim WJ , Khera AV , et al. Association between adiposity and cardiovascular outcomes: An umbrella review and meta-analysis of observational and Mendelian randomization studies[J]. Eur Heart J, 2021, 42 (34): 3388- 3403.

DOI

29
Tan J , Li B , Cao J , et al. APOE gene polymorphism in ischemic stroke patients from Huizhou and its correlation with blood lipids and homocysteine[J]. J Stroke Cerebrovasc Dis, 2024, 33 (11): 107990.

DOI

30
Kulminski AM , Loika Y , Culminskaya I , et al. Independent associations of TOMM40 and APOE variants with body mass index[J]. Aging Cell, 2019, 18 (1): e12869.

DOI

31
Wang C , Du Z , Ye N , et al. Hyperlipidemia and hypertension have synergistic interaction on ischemic stroke: Insights from a general population survey in China[J]. BMC Cardiovasc Disord, 2022, 22 (1): 47.

DOI

32
Rao W , Su Y , Yang G , et al. Cross-sectional associations between body mass index and hyperlipidemia among adults in Northeastern China[J]. Int J Environ Res Public Health, 2016, 13 (5): 516.

DOI

33
Ryder E , Diez-Ewald M , Mosquera J , et al. Association of obesity with leukocyte count in obese individuals without metabolic syndrome[J]. Diabetes Metab Syndr, 2014, 8 (4): 197- 204.

DOI

34
Schmidt FM , Weschenfelder J , Sander C , et al. Inflammatory cytokines in general and central obesity and modulating effects of physical activity[J]. PLoS One, 2015, 10 (3): e0121971.

DOI

35
Fantuzzi G , Mazzone T . Adipose tissue and atherosclerosis: Exploring the connection[J]. Arterioscler Thromb Vasc Biol, 2007, 27 (5): 996- 1003.

DOI

36
Singer G , Granger DN . Inflammatory responses underlying the microvascular dysfunction associated with obesity and insulin resis-tance[J]. Microcirculation, 2007, 14 (4/5): 375- 387.

37
Liao W , Wen Y , Zeng C , et al. Integrative analyses and validation of ferroptosis-related genes and mechanisms associated with cerebrovascular and cardiovascular ischemic diseases[J]. BMC Genomics, 2023, 24 (1): 731.

DOI

38
Yamase Y , Horibe H , Ueyama C , et al. Association of TOMM40 and SLC22A4 polymorphisms with ischemic stroke[J]. Biomed Rep, 2015, 3 (4): 491- 498.

DOI

39
Cheng X , Ferino E , Hull H , et al. Smoking affects gene expression in blood of patients with ischemic stroke[J]. Ann Clin Transl Neurol, 2019, 6 (9): 1748- 1756.

DOI

40
Chang CT , Liao HY , Chang CM , et al. Oxidized ApoC1 on MALDI-TOF and glycated-ApoA1 band on gradient gel as potential diagnostic tools for atherosclerotic vascular disease[J]. Clin Chim Acta, 2013, 420, 69- 75.

DOI

41
Ciuffetelli Alamo IV , Kwartler CS , Regalado ER , et al. Grange syndrome due to homozygous YY1AP1 missense rare variants[J]. Am J Med Genet A, 2019, 179 (12): 2500- 2505.

DOI

42
Yin S , Pang A , Liu C , et al. Peptide OM-LV20 protects astrocytes against oxidative stress via the PAC1R/JNK/TPH1 axis[J]. J Biol Chem, 2022, 298 (10): 102429.

DOI

43
Stelzer G , Rosen N , Plaschkes I , et al. The GeneCards suite: From gene data mining to disease genome sequence analyses[J]. Curr Protoc Bioinformatics, 2016, 54, 1.30.1- 1.30.33.

44
Tai LM , Ghura S , Koster KP , et al. APOE-modulated Aβ-induced neuroinflammation in Alzheimer' s disease: Current landscape, novel data, and future perspective[J]. J Neurochem, 2015, 133 (4): 465- 488.

DOI

45
Tai L M , Thomas R , Marottoli FM , et al. The role of APOE in cerebrovascular dysfunction[J]. Acta Neuropathol, 2016, 131 (5): 709- 723.

DOI

46
Yan B , Ai Y , Sun Q , et al. Membrane damage during ferroptosis is caused by oxidation of phospholipids catalyzed by the oxidoreductases POR and CYB5R1[J]. Mol Cell, 2021, 81 (2): 355-369. e10.

DOI

47
Dixon SJ , Lemberg KM , Lamprecht MR , et al. Ferroptosis: An iron-dependent form of nonapoptotic cell death[J]. Cell, 2012, 149 (5): 1060- 1072.

DOI

48
Zhang Y , Xin L , Xiang M , et al. The molecular mechanisms of ferroptosis and its role in cardiovascular disease[J]. Biomed Pharmacother, 2022, 145, 112423.

DOI

49
Hu X , Bao Y , Li M , et al. The role of ferroptosis and its mechanism in ischemic stroke[J]. Exp Neurol, 2024, 372, 114630.

DOI

50
Guo J , Tuo QZ , Lei P . Iron, ferroptosis, and ischemic stroke[J]. J Neurochem, 2023, 165 (4): 487- 520.

DOI

Outlines

/