摘要:目的" 利用生物信息學(xué)方法探究A型急性主動(dòng)脈夾層(AAAD)的病理機(jī)制、關(guān)鍵靶點(diǎn)以及診斷生物標(biāo)志物。方法" GEO數(shù)據(jù)庫(kù)下載GSE52093數(shù)據(jù)集,基于差異分析和WGCNA算法,構(gòu)建基因共表達(dá)模塊,對(duì)關(guān)鍵模塊基因進(jìn)行功能富集分析及PPI網(wǎng)絡(luò)分析,探究關(guān)鍵模塊基因參與AAAD的分子機(jī)制,并篩選出與AAAD密切相關(guān)的Hub基因。結(jié)果" 從GSE52093數(shù)據(jù)集中共鑒定出218個(gè)差異表達(dá)基因,其中上調(diào)基因135個(gè),下調(diào)基因82個(gè);另基于差異表達(dá)基因識(shí)別出4個(gè)基因模塊,包括Turquoise模塊(95個(gè)基因)、Blue模塊(91個(gè)基因)、Yellow模塊(20個(gè)基因)、Grey模塊(11個(gè)基因),其中Turquoise模塊與AAAD相關(guān)性最顯著。關(guān)鍵基因模塊在多種生物途徑中富集,包括肌動(dòng)蛋白細(xì)胞骨架的調(diào)節(jié)(regulation of actin cytoskeleton)、鈣信號(hào)通路(calcium signaling pathway)、黏著斑途徑(focal adhesion)、血管平滑肌收縮(vascular smooth muscle contraction)、心肌細(xì)胞中的腎上腺素能信號(hào)轉(zhuǎn)導(dǎo)(adrenergic signaling in cardiomyocytes)、Wnt信號(hào)通路(Wnt signaling pathway)等。從PPI網(wǎng)絡(luò)中篩選出10個(gè)與AAAD密切相關(guān)的Hub基因,分別為ACTC1、ACTN4、NEXN、MYOZ2、CSRP1、PPP1R12B、RYR2、MYL9、ACTG2、CASQ2,這些Hub基因在AAAD中均表達(dá)下調(diào)。結(jié)論" 本研究利用生物信息學(xué)方法篩選出與AAAD密切相關(guān)的生物途徑和Hub基因,將為AAAD病理機(jī)制研究以及診斷生物標(biāo)物和治療靶點(diǎn)的發(fā)現(xiàn)提供基礎(chǔ)。
關(guān)鍵詞:A型急性主動(dòng)脈夾層;樞紐基因;加權(quán)基因共表達(dá)網(wǎng)絡(luò)分析;生物信息學(xué)分析
中圖分類(lèi)號(hào):R543.1" " " " " " " " " " " " " " " " "文獻(xiàn)標(biāo)識(shí)碼:A" " " " " " " " " " " " " " " " "DOI:10.3969/j.issn.1006-1959.2023.07.002
文章編號(hào):1006-1959(2023)07-0009-08
Identification of Hub Genes and Molecular Mechanisms of Type A Acute Aortic Dissection Based
on Weighted Gene Co-expression Network Analysis
GONG Dao-xing
(Department of Vascular Surgery,Changsha First Hospital,Changsha 410000,Hunan,China)
Abstract:Objective" To explore the pathological mechanism, key targets and diagnostic biomarkers of type A acute aortic dissection (AAAD) by bioinformatics methods.Methods" The GSE52093 dataset was downloaded from the GEO database. Based on the difference analysis and WGCNA algorithm, the gene co-expression module was constructed. The key module genes were subjected to functional enrichment analysis and PPI network analysis to explore the molecular mechanism of key module genes participating in AAAD, and Hub genes closely related to AAAD were screened.Results" A total of 218 differentially expressed genes were identified from the GSE52093 dataset, including 135 up-regulated genes and 82 down-regulated genes. In addition, four gene modules were identified based on differentially expressed genes, including Turquoise module (95 genes), Blue module (91 genes), Yellow module (20 genes) and Grey module (11 genes), among which Turquoise module had the most significant correlation with AAAD. Key gene modules were enriched in multiple biological pathways, including regulation of actin cytoskeleton, calcium signaling pathway, focal adhesion pathway, vascular smooth muscle contraction, adrenergic signaling in cardiomyocytes, Wnt signaling pathway and so on. Ten Hub genes closely related to AAAD were screened from the PPI network, namely ACTC1, ACTN4, NEXN, MYOZ2, CSRP1, PPP1R12 B, RYR2, MYL9, ACTG2 and CASQ2, which were down-regulated in AAAD.Conclusion" In this study, bioinformatics methods were used to screen out the biological pathways and Hub genes closely related to AAAD, which will provide a basis for the study of the pathological mechanism of AAAD and the discovery of diagnostic biomarkers and therapeutic targets.
Key words:Type A acute aortic dissection;Hub genes;Weighted gene co-expression network analysis;Bioinformatics analysis
急性主動(dòng)脈夾層(acute aortic dissection,AAD)即主動(dòng)脈內(nèi)膜損傷破裂,血液進(jìn)入中膜層,于主動(dòng)脈兩層之間形成新的含血腔隙,是一種病情危重、發(fā)展迅速、病死率高的以胸背部劇烈疼痛為主要表現(xiàn)的主動(dòng)脈疾病[1]。根據(jù)Stanford分型,急性主動(dòng)脈夾層分為A型與B型兩類(lèi)。A型急性主動(dòng)脈夾層(type A acute aortic dissection,AAAD)主要累及主動(dòng)脈弓、降主動(dòng)脈及腹主動(dòng)脈,大約占AAD的2/3[2]。由于主動(dòng)脈瓣反流、心力衰竭和中風(fēng)等嚴(yán)重并發(fā)癥,AAD死亡率較高[3],而AAAD術(shù)后死亡率高達(dá)9.3%~28.6%。未接受任何治療的AAAD患者3個(gè)月死亡率約為90%,24 h內(nèi)的自然死亡率達(dá)30%,48 h內(nèi)的自然死亡率可達(dá)50%[4]。然而,由于AAAD的癥狀與肺栓塞、急性冠狀動(dòng)脈綜合征和心肌梗死等疾病相似,經(jīng)常會(huì)導(dǎo)致診斷不及時(shí),影響患者生命健康[5]。因此,早期診斷對(duì)于患者獲得及時(shí)治療,改善預(yù)后至關(guān)重要。但目前臨床上缺乏有效且方便的診斷生物標(biāo)志物,故識(shí)別新的潛在生物標(biāo)志物對(duì)于AAAD的治療尤為重要。近年來(lái)越來(lái)越多的研究認(rèn)為[6-9],AAAD的發(fā)生和基因突變、主動(dòng)脈結(jié)構(gòu)異常、主動(dòng)脈中層的退行性變、慢性炎癥等有關(guān),但AAAD的病理機(jī)制仍不清楚。研究AAAD的分子機(jī)制,對(duì)于認(rèn)識(shí)疾病的發(fā)生、發(fā)展和轉(zhuǎn)歸以及治療靶點(diǎn)的發(fā)現(xiàn)和靶向藥物的研發(fā)有著重要意義。隨著基因表達(dá)譜的發(fā)展,基于生物信息學(xué)方法,通過(guò)篩選病理和生理?xiàng)l件下差異表達(dá)基因,是尋找疾病的生物標(biāo)志物以及病理機(jī)制和關(guān)鍵治療靶點(diǎn)的重要手段[10]。目前,通量測(cè)序技術(shù)已在AAAD的研究中廣泛使用,因此整合先進(jìn)生物信息學(xué)方法是探究AAAD分子機(jī)制,尋找診斷生物標(biāo)物和關(guān)鍵靶點(diǎn)的十分方便有效的途徑。本研究采用加權(quán)基因共表達(dá)網(wǎng)絡(luò)分析(WGCNA)對(duì)AAAD相關(guān)的表達(dá)譜芯片進(jìn)行生物信息學(xué)分析,以期為AAAD病理機(jī)制研究以及診斷生物標(biāo)物和治療靶點(diǎn)的發(fā)現(xiàn)提供基礎(chǔ)。
1資料與方法
1.1數(shù)據(jù)來(lái)源" 從GEO數(shù)據(jù)庫(kù)中下載AAD相關(guān)的基因表達(dá)譜數(shù)據(jù)集(GSE52093),GSE52093基于GPL10558(Illumina HumanHT-12 V4.0表達(dá)芯片),包含7個(gè)AAAD組織標(biāo)本和5個(gè)相鄰正常組織標(biāo)本。
1.2基因差異表達(dá)分析" 使用R軟件(版本3.6.3)BiocManager軟件包,分別對(duì)GSE52093數(shù)據(jù)集進(jìn)行背景校正、標(biāo)準(zhǔn)歸一化處理和表達(dá)式值計(jì)算。以|log(FC)|≥1.0,Plt;0.05為閾值分析差異表達(dá)基因,并繪制火山圖進(jìn)行可視化。
1.3WGCNA" R軟件包構(gòu)建所有差異基因的WGCNA分析,具體流程包括:首先,選擇網(wǎng)絡(luò)建設(shè)的軟閾值。軟閾值是鄰接矩陣成為0到1之間的連續(xù)值,使構(gòu)建的網(wǎng)絡(luò)符合冪律分布,更接近真實(shí)的生物網(wǎng)絡(luò)狀態(tài);其次,使用blockwiseModules函數(shù)構(gòu)建無(wú)標(biāo)度網(wǎng)絡(luò),然后進(jìn)行模塊劃分分析,以確定基因共表達(dá)模塊,該模塊可以對(duì)具有相似表達(dá)模式的基因進(jìn)行分組。這些模塊通過(guò)使用動(dòng)態(tài)樹(shù)切割算法將聚類(lèi)樹(shù)切割成分支來(lái)定義,并分配不同的顏色進(jìn)行可視化;然后,計(jì)算每個(gè)基因模塊的特征值(ME);接著,計(jì)算每個(gè)模塊中ME與臨床特征之間的相關(guān)性;最后,進(jìn)一步計(jì)算模塊基因的基因顯著性。計(jì)算并篩選GSE52093數(shù)據(jù)集WGCNA分析下的關(guān)鍵模塊基因,用于后續(xù)分析。
1.4功能富集分析" 為了確定GSE52093數(shù)據(jù)集WGCNA分析下關(guān)鍵模塊基因的生物功能,使用DAVID數(shù)據(jù)庫(kù)(https://david.ncifcrf.gov/)進(jìn)行功能富集分析。
1.5差異基因的相互作用" 使用String數(shù)據(jù)庫(kù)(http://string-db.org),對(duì)關(guān)鍵模塊基因進(jìn)行蛋白質(zhì)-蛋白質(zhì)相互作用(protein-protein interaction,PPI)分析。然后,通過(guò)CytoHubba插件篩選PPI網(wǎng)絡(luò)中的關(guān)鍵基因,以最大相關(guān)標(biāo)準(zhǔn)(maximal clique centrality,MCC)算法,從PPI網(wǎng)絡(luò)中篩選出得分最高的10個(gè)基因作為本次研究的Hub基因。
2結(jié)果
2.1差異基因表達(dá)分析" 對(duì)GSE52093數(shù)據(jù)集進(jìn)行標(biāo)準(zhǔn)歸一化處理,從GSE52093數(shù)據(jù)集中共鑒定出218個(gè)差異表達(dá)基因,其中上調(diào)基因135個(gè),下調(diào)基因82個(gè),見(jiàn)圖1。
2.2 WGCNA網(wǎng)絡(luò)的構(gòu)建" 分別對(duì)GSE52093數(shù)據(jù)集中差異基因進(jìn)行WGCNA分析。設(shè)置WGCNA網(wǎng)絡(luò)的軟閾值為28,根據(jù)模塊相似度大于0.25標(biāo)準(zhǔn)合并相似模塊?;贕SE52093數(shù)據(jù)集差異表達(dá)基因識(shí)別出4個(gè)基因模塊,包括Turquoise模塊(95個(gè)基因)、Blue模塊(91個(gè)基因)、Yellow模塊(20個(gè)基因)、Grey模塊(11個(gè)基因),見(jiàn)圖2?;赑earson相關(guān)性分析方法,通過(guò)計(jì)算各個(gè)基因模塊與樣本表型之間的關(guān)系,繪制出基因模塊與樣本表型熱圖,結(jié)果顯示Turquoise模塊與AAAD相關(guān)性最顯著,見(jiàn)圖3。
2.3功能富集分析" 利用DAVID數(shù)據(jù)庫(kù)對(duì)WGCNA分析下的95個(gè)Turquoise模塊基因進(jìn)行GO注釋和KEGG功能分析,其中GO分析結(jié)果顯示,關(guān)鍵模塊基于涉及201種生物過(guò)程(biological process,BP)、33種細(xì)胞成分(cellular component,CC)、31種分子功能(molecular function,MF)3大類(lèi)。BP涉及基于肌動(dòng)蛋白絲的過(guò)程(actin filament-based process)、肌肉結(jié)構(gòu)發(fā)育(muscle structure development)、肌動(dòng)蛋白細(xì)胞骨架組織(actin cytoskeleton organization)、離子傳輸?shù)恼{(diào)節(jié)(regulation of ion transport)、金屬離子遷移的調(diào)節(jié)(regulation of metal ion transport)、肌細(xì)胞分化(muscle cell differentiation)等過(guò)程。CC包括肌節(jié)(sarcomere)、肌原纖維(myofibril)、收縮纖維(contractile fiber)、肌動(dòng)蛋白細(xì)胞骨架(actin cytoskeleton)、細(xì)胞-基底連接(cell-substrate junction)、細(xì)胞間連接(cell-cell junction)等。在MF中,共同的關(guān)鍵模塊基因主要富集于激酶結(jié)合(kinase binding)、肌動(dòng)蛋白結(jié)合(actin binding)、激酶活性(kinase activity)、蛋白激酶結(jié)合(protein kinase binding)、轉(zhuǎn)錄因子結(jié)合(transcription factor binding)、鈣調(diào)素結(jié)合(calmodulin binding)、蛋白質(zhì)絲氨酸/蘇氨酸/酪氨酸激酶活性(protein serine/threonine/tyrosine kinase activity)等分子功能。KEGG功能富集結(jié)果顯示,關(guān)鍵模塊基因富集于53條信號(hào)通路中,包括肌動(dòng)蛋白細(xì)胞骨架的調(diào)節(jié)(regulation of actin cytoskeleton)、鈣信號(hào)通路(calcium signaling pathway)、糖尿病心肌?。╠iabetic cardiomyopathy)、黏著斑途徑(focal adhesion)、血管平滑肌收縮(vascular smooth muscle contraction)、心肌細(xì)胞中的腎上腺素能信號(hào)轉(zhuǎn)導(dǎo)(adrenergic signaling in cardiomyocytes)、Wnt信號(hào)通路(Wnt signaling pathway)等。KEGG及GO分析各類(lèi)別下前20條目見(jiàn)圖4。
2.4 Hub基因的PPI網(wǎng)絡(luò)構(gòu)建及Hub基因的篩選" 使用String軟件對(duì)共同模塊基因進(jìn)行PPI網(wǎng)絡(luò)分析?;贛CC算法,通過(guò)CytoHubba plugin從PPI網(wǎng)絡(luò)中篩選出相關(guān)度最大的前10個(gè)Hub基因,分別為ACTC1、ACTN4、NEXN、MYOZ2、CSRP1、PPP1R12B、RYR2、MYL9、ACTG2、CASQ2。與正常對(duì)照組相比,這些Hub基于在AAAD中均下調(diào),其低表達(dá)可能與AAAD進(jìn)程密切相關(guān),見(jiàn)圖5、表1。
3討論
AAAD是一種嚴(yán)重的主動(dòng)脈疾病,表現(xiàn)為內(nèi)膜撕裂穿透升主動(dòng)脈中膜,其發(fā)病機(jī)制復(fù)雜,與多種基因的相互調(diào)控有關(guān)。同時(shí),由于缺乏快速、有效的診斷和治療方法,AAAD在臨床上具有很高的死亡率。因此,探究AAAD分子機(jī)制,尋找行之有效的診斷生物標(biāo)物具有十分重要的意義。
本研究基于生物信息學(xué)方法,首先對(duì)GSE52093數(shù)據(jù)集進(jìn)行差異分析和WGCNA分析,識(shí)別出與AAAD相關(guān)性最佳的關(guān)鍵模塊基因。對(duì)關(guān)鍵模塊基因進(jìn)行功能富集分析,結(jié)果顯示這些模塊基因主要富集在肌動(dòng)蛋白絲的過(guò)程、肌動(dòng)蛋白細(xì)胞骨架組織、離子傳輸?shù)恼{(diào)節(jié)等生物過(guò)程,肌動(dòng)蛋白細(xì)胞骨架的調(diào)節(jié)、鈣信號(hào)通路、黏著斑途徑、血管平滑肌收縮、Wnt信號(hào)通路等信號(hào)途徑,該結(jié)果揭示了導(dǎo)致AAAD可能的生物途徑。
血管平滑肌細(xì)胞(VSMCs)是主動(dòng)脈中膜的重要成員,具有收縮和合成2種細(xì)胞表型[11]。正常情況下,VSMCs多為合成表型,合成細(xì)胞外基質(zhì)(包括膠原、彈性蛋白和蛋白聚糖等)維持主動(dòng)脈正常彈性和張力。在AAAD中,VSMCs向收縮表型轉(zhuǎn)換,細(xì)胞凋亡作用增強(qiáng),增殖和遷移能力減弱[12,13]。研究表明[14,15],收縮型VSMCs可通過(guò)肌動(dòng)蛋白絲形成過(guò)程、肌動(dòng)蛋白細(xì)胞骨架組織形成及調(diào)節(jié)、黏著斑途徑等生物途徑介導(dǎo)AAAD的發(fā)生或發(fā)展。鈣信號(hào)通路在維持血管平滑肌正常生理功能中發(fā)揮著重要作用,當(dāng)VSMCs中Ca2+濃度升高,將引起血管收縮,反之則導(dǎo)致血管舒張[16,17]。研究發(fā)現(xiàn)[18],鈣信號(hào)通路受到過(guò)度抑制、VSMCs對(duì)Ca2+敏感性減弱,可導(dǎo)致血管收縮能力下降,引起主動(dòng)脈中膜退化,使血管變得脆弱。
此外,本研究通過(guò)對(duì)模塊基因進(jìn)行PPI網(wǎng)絡(luò)分析,結(jié)合MCC計(jì)算方法,候選出與AAAD相關(guān)的10個(gè)Hub基因,包括ACTC1、ACTN4、NEXN、MYOZ2、CSRP1、PPP1R12B、RYR2、MYL9、ACTG2、CASQ2。與正常對(duì)照組相比,這些Hub基于在AAAD中均顯著下調(diào)。已有研究報(bào)道[19],NEXN、ACTG2、RYR2等基因在心血管疾病中發(fā)揮著重要作用。Wu L等[20]研究發(fā)現(xiàn),NEXN可通過(guò)促進(jìn)VSMCs增殖和遷移,改善主動(dòng)脈損傷。Wang W等[21]研究發(fā)現(xiàn),ACTG2可抑制VSMCs表型轉(zhuǎn)換,改善主動(dòng)脈夾層。但目前相關(guān)研究仍非常缺乏,以上基因與AAAD的關(guān)系機(jī)制仍需進(jìn)一步探索。
綜上所述,本研究利用生物信息學(xué)方法篩選出與AAAD密切相關(guān)的生物途徑和Hub基因,將為AAAD病理機(jī)制研究以及診斷生物標(biāo)物和治療靶點(diǎn)的發(fā)現(xiàn)提供基礎(chǔ)。
參考文獻(xiàn):
[1]Gawinecka J,Sch?nrath F,von Eckardstein A.Acute aortic dissection: pathogenesis,risk factors and diagnosis[J].Swiss Med Wkly,2017,147:w14489.
[2]Jassar AS,Sundt TM.How should we manage type A aortic dissection?[J].Gen Thorac Cardiovasc Surg,2019,67(1):137-145.
[3]Duan W,Wang W,Xia L,et al.Clinical profiles and outcomes of acute type A aortic dissection and intramural hematoma in the current era:lessons from the first registry of aortic dissection in China[J].Chin Med J (Engl),2021,134(8):927-934.
[4]王洪巖.急性主動(dòng)脈夾層患者的入院血鈉和血糖對(duì)其短期預(yù)后的影響[D].唐山:華北理工大學(xué),2021.
[5]Gao H,Sun X,Liu Y,et al.Analysis of Hub Genes and the Mechanism of Immune Infiltration in Stanford Type a Aortic Dissection[J].Frontiers in Cardiovascular Medicine,2021,8:680065.
[6]Yamamoto T,Endo D,Shimada A,et al.Surgical treatment of acute aortic dissection in a patient with SLE and prior antiphospholipid syndrome on therapy for over 30 years:a case report[J].BMC Cardiovascular Disorders,2022,22(1):216.
[7]Zheng J,Guo J,Huang L,et al.Genetic diagnosis of acute aortic dissection in South China Han population using next-generation sequencing[J].International Journal of Legal Medicine,2018,132(5):1273-1280.
[8]Cifani N,Proietta M,Tritapepe L,et al.Stanford-A acute aortic dissection,inflammation,and metalloproteinases:a review[J].Annals of Medicine,2015,47(6):441-446.
[9]Lei Y,Jiang Z,Chen J,et al.Type A aortic dissection in pregnant patients with fibrillin-1 gene mutations: Two case reports and a literature review[J].Exp Ther Med,2018,16(6):4407-4414.
[10]Zou H,Qiu B,Lai S,et al.Role of ferroptosis-related genes in Stanford type a aortic dissection and identification of key genes:new insights from bioinformatic analysis[J].Bioengineered,2021,12(2):9976-9990.
[11]Owens G,Kumar M,Wamhoff B.Molecular regulation of vascular smooth muscle cell differentiation in development and disease[J].Physiological Reviews,2004,84(3):767-801.
[12]Wang W,Wang T,Wang Y,et al.Integration of Gene Expression Profile Data to Verify Hub Genes of Patients with Stanford A Aortic Dissection[J].BioMed Research International,2019,2019:3629751.
[13]Zhou X,Cheng J,Chen Z,et al.Role of c-Abl in Ang II-induced aortic dissection formation:Potential regulatory efficacy on phenotypic transformation and apoptosis of VSMCs[J].Life Sciences,2020,256:117882.
[14]Huang B,Niu Y,Chen Z,et al.Integrin α9 is involved in the pathopoiesis of acute aortic dissection via mediating phenotype switch of vascular smooth muscle cell[J].Biochemical and Biophysical Research Communications,2020,533(3):519-525.
[15]Li Y,Gao S,Han Y,et al.Variants of Focal Adhesion Scaffold Genes Cause Thoracic Aortic Aneurysm[J].Circulation Research,2021,128(1):8-23.
[16]陳琳,程玲霞,楊帆,等.血管平滑肌細(xì)胞自噬對(duì)小鼠主動(dòng)脈夾層形成的影響[J].中山大學(xué)學(xué)報(bào)(醫(yī)學(xué)科學(xué)版),2021,42(2):226-234.
[17]Martinsen A,Dessy C,Morel N.Regulation of calcium channels in smooth muscle:new insights into the role of myosin light chain kinase[J].Channels (Austin),2014,8(5):402-413.
[18]Somlyo A,Somlyo A.Ca2+ sensitivity of smooth muscle and nonmuscle myosin II:modulated by G proteins,kinases,and myosin phosphatase[J].Physiological Reviews,2003,83(4):1325-1358.
[19]Fan G,Kaβmann M,Hashad A,et al.Differential targeting and signalling of voltage-gated T-type Ca 3.2 and L-type Ca 1.2 channels to ryanodine receptors in mesenteric arteries[J].The Journal of Physiology,2018,596(20):4863-4877.
[20]Wu L,Li Y,Zhang D,et al.LncRNA NEXN-AS1 attenuates proliferation and migration of vascular smooth muscle cells through sponging miR-33a/b[J].RSC Advances,2019,9(48):27856-27864.
[21]Wang W,Liu Q,Wang Y,et al.LINC01278 Sponges miR-500b-5p to Regulate the Expression of ACTG2 to Control Phenotypic Switching in Human Vascular Smooth Muscle Cells During Aortic Dissection[J].Journal of the American Heart Association,2021,10(9):e018062.
收稿日期:2022-06-09;修回日期:2022-08-22
編輯/杜帆
作者簡(jiǎn)歷:龔道星(1990.4-),男,湖南華容縣人,碩士,住院醫(yī)師,主要從事胸腹主動(dòng)脈瘤、下肢動(dòng)脈缺血、糖尿病足、下肢靜脈曲張的研究