Abstract
目的
通过大数据筛选与肺腺癌预后相关的关键基因并探讨其临床价值和潜在机制。
方法
基于基因表达综合数据库（GEO）中获得的GSE18842，GSE27262以及GSE33532基因表达谱进行数据分析；生物信息学方法筛选肿瘤组织和正常肺组织的差异表达基因，对其进行京都基因与基因组百科全书（KEGG）和基因本体论（GO）富集分析后进行蛋白质-蛋白质相互作用网络（PPI）、模组、表达差异和预后分析和筛选。35例非小细胞肺癌标本和35例配对的癌旁正常组织，共70例组织标本分为肿瘤组和正常组对MAD2L1和TTK的表达进行了免疫组化验证。
结果
共有256个基因的表达谱数据有统计学差异（P < 0.05），包括66个上调基因，190个下调基因。进行功能分析后筛选出32个上调基因。32个基因中的29与肺腺癌预后显著相关。相较与正常肺组织，所有29个基因均在肺腺癌组织中高表达并主要富集在细胞周期通路。其中7个关键基因与纺锤体组装检查点（SAC）复合体紧密相关，负责调控细胞G2/M期行为。我们选择了SAC相关基因TTK和MAD2L1，在肺腺癌患者组织标本中观察到了TTK和MAD2L1相较与癌旁正常肺组织的过表达。
结论
以TTK和MAD2L1为代表的7个SAC复合体相关基因在肺腺癌患者中存在过表达，且其过表达与预后相关。
Keywords: 生物信息学分析, 基因芯片, 非小细胞肺癌, 肺腺癌, 差异表达基因
Abstract
Objective
To screen the key genes related to the prognosis of lung adenocarcinoma through big data analysis and explore their clinical value and potential mechanism.
Methods
We analyzed GSE18842, GSE27262, and GSE33532 gene expression profile data obtained from the Gene Expression Omnibus (GEO). Bioinformatics methods were used to screen the differentially expressed genes in lung adenocarcinoma tissues and KEGG and GO enrichment analysis was performed, followed by PPI interaction network analysis, module analysis, differential expression analysis, and prognosis analysis. The expressions of MAD2L1 and TTK by immunohistochemistry were verified in 35 non-small cell lung cancer specimens and paired adjacent tissues.
Results
We identified a total of 256 genes that showed significant differential expressions in lung adenocarcinoma, including 66 up-regulated and 190 down-regulated genes. Thirty-two up-regulated core genes were screened by functional analysis, and among them 29 were shown to significantly correlate with a poor prognosis of patients with lung adenocarcinoma. All the 29 genes were highly expressed in lung adenocarcinoma tissues compared with normal lung tissues and were mainly enriched in cell cycle pathways. Seven of these key genes were closely related to the spindle assembly checkpoint (SAC) complex and responsible for regulating cell behavior in G2/M phase. We selected SAC-related proteins TTK and MAD2L1 to test their expressions in clinical tumor samples, and detected their overexpression in lung adenocarcinoma tissues as compared with the adjacent tissues.
Conclusion
Seven SAC complex-related genes, including TTK and MAD2L1, are overexpressed in lung adenocarcinoma tissues with close correlation with the prognosis of the patients.
Keywords: bioinformatical analysis, microarray, non-small cell lung cancer, lung adenocarcinomas, differentially expressed genes
肺癌是全世界发病率和死亡率最高的癌症[1]。其中非小细胞肺癌（NSCLC）约占80%。肺腺癌和肺鳞癌是NSCLC中最常见的两种组织学类型。仍有超过半数的患者无法从狭义的靶向治疗中获益[2-3]。
基于肺癌的研究发现对纺锤体组装检查点（SAC）相关蛋白KIAA0101和YL-9的抑制能引起SAC复合体失活进而介导抗癌作用[4-5]。TTK作为SAC复合体的核心组分，负责维持SAC复合体的正常功能[6]，其抑制剂也被认为是肿瘤治疗的新手段[7]。尽管以TTK抑制剂为代表的SAC复合体抑制剂有着良好的前景[7-8]，但由于其耐药性，一般建议多种SAC抑制剂联用或与其他药物联用，不建议单独使用[8-9]。鉴于SAC复合体的组分较为复杂，包括CCNB1、CCNB2、TTK、MAD2L1在内的数十种组分共同参与其调节，厘清SAC复合体相关基因在肿瘤中的表达特征，对肿瘤精准治疗和相关药物在肿瘤中的应用有潜在的意义，而目前没有相关报道。
因此，本研究通过利用3个基因芯片表达谱数据集（GSE18842、GSE27262和GSE33532）进行生物信息学分析，最终筛选7个以TTK和MAD2L1为代表的SAC复合体相关基因在肺腺癌中过表达，有可能成为肺腺癌靶向治疗新的研究发现和潜在靶点。
1. 资料和方法
1.1. 基因芯片数据
NSCLC标本以及正常组织标本的基因表达数据来自GEO数据库[10]的GSE18842、GSE27262和GSE33532数据集。上述3个基因芯片数据均基于GPL570平台（[HG-U133Plus2]Affymetrix Human Genome U133 Plus 2.0 Array）。数据总共包含91个病人的241个标本。
1.2. 差异表达基因的数据分析
肿瘤组织相较于正常组织的差异表达基因数据通过GEO2R工具[11]进行第一步分析。分组分析后按照|logFC| > 2以及adjust P值< 0.05为标准得出各个数据集的差异表达基因（DEGs）并计算3个基因集共同的差异表达基因。logFC > 2的被认为是上调基因，logFC < -2的被认为是下调基因。
1.3. 基因富集分析
使用R语言程序包ClusterProfiler[12]分别对差异表达基因进行生物学过程（BP），分子功能（MF），细胞组分（CC）和京都基因与基因组百科全书（KEGG）信号通路的富集以及可视化。使用Benjamini & Hochberg法（BH法）进行P值矫正。
1.4. 蛋白质-蛋白质相互作用网络（PPI）构建和分析
PPI使用基因相互作用检索工具（STRING）[13]生成并初步分析。将构建PPI网络要求的最低相互作用得分设置为0.4（满分为1）。通过Cytoscape软件分析和标记PPI数据。使用分子复合物模组分析工具（MCODE）插件对PPI网络进行聚类，计算各个节点的信息得出Score值和功能模块。相关参数设置为：Degree Cutoff= 2，Max. Depth=100，K-Core=2，Node Score Cutoff=0.2。
1.5. MCODE基因的生存分析
使用生存分析工具Kaplan Meier-plotter[14]进行生存分析。选择中位数作为基因表达量的Cutoff值，有多个探针的单个基因选择JetSet最佳探针。计算后的P 值、风险比（HR）和95%置信区间绘于图上。
1.6. 基因表达量分析
使用基因表达谱分析工具（GEPIA）[15]基于TCGA数据库计算肺腺癌肿瘤组织和正常组织的基因表达数据，肿瘤和正常组织数量绘于图上。
1.7. 免疫组化数据库分析
使用人类蛋白质图谱（HPA）数据库[16]验证肺腺癌患者的肿瘤组织和正常组织中CCNB1、CCNB2、MAD2L1、CDC20、AURKA和TTK的蛋白表达。
1.8. 非小细胞肺癌患者组织标本
两张35对70例肺小细胞肺癌组织和匹配的癌旁正常组织的组织芯片来自武汉Servicebio公司组织库2013~2015年手术切除标本。其中男性24例，女性11例，年龄58~73岁，平均年龄65.2岁；鳞癌15例，腺癌16例、腺鳞癌和其他类型4例；AJCC第七版临床分期：Ⅰ期19例，Ⅱ期4例，Ⅲ期12例。有淋巴结转移16例，无淋巴结转移19例，均无远处转移；所有患者均知情同意。芯片中的肿瘤组织由直径1.5 mm的芯针从术后肿瘤标本最有组织学代表性的部分采集，经10%中性甲醛溶液固定、常规脱水、石蜡包埋后制作为4 μm厚度的组织芯片。手术和标本获取途径和过程均符合伦理要求。
1.9. 免疫组化实验
脱蜡、水化和抗原修复等处理完毕的两张组织芯片分别使用PBS清洗，并使用牛血清蛋白封闭30 min。分别用MAD2L1抗体和TTK抗体4 ℃孵育过夜，清洗后二抗室温孵育1 h，DAB显色后苏木紫复染。脱水、透明、封片、镜检等处理后使用3D HISTECH的Pannoramic MIDI扫描仪扫描成像。每个点位均随机取4~5个视野进行观察，在病理科专家指导下分三人分别按照染色强度和所占肺腺癌细胞/癌旁肺上皮细胞比例进行判定和统计。
1.10. 统计学方法
研究使用SPSS 19.0软件进行统计分析，两组间数据的比较使用t检验，P < 0.05被认为差异具有统计学意义。免疫组化的评分标准为H-SCORE=∑(pi×i)=（弱阳性染色细胞占比×1）+（中等阳性染色细胞占比×2）+（强阳性染色细胞占比×3）其中pi代表阳性细胞数量占所有细胞数量的百分比，i代表染色强度。
2. 结果
2.1. 差异表达基因的获取
本项研究中使用了151例非小细胞肺癌组织以及90例正常肺组织的基因芯片数据，通过GEO2R分析工具，我们从GSE18842、GSE27262和GSE33532中按照|logFC| > 2以及adjust P < 0.05为标准分别获取了1060、671和795个差异表达基因。使用获得的数据绘制韦恩图，对三个数据集中得出的差异表达基因取交集，得出了256个共同差异表达基因，其中上调基因66个，下调基因190个（图 1）。
1.
来自三个基因数据集（GSE18842、GSE27262和GSE33532）的差异基因绘制的韦恩图（<a href="http://bioinformatics.psb.ugent.be/webtools/Venn/" target="_blank">http://bioinformatics.psb.ugent.be/webtools/Venn/</a>）
Venn diagrams (<a href="http://bioinformatics.psb.ugent.be/webtools/Venn/" target="_blank">http://bioinformatics.psb.ugent.be/webtools/Venn/</a>) of 256 common differentially expressed genes (DEGs) identified in the 3 datasets (GSE18842, GSE27262, and GSE33532). <bold>A</bold>: 190 DEGs are down-regulated in the 3 datasets (logFC > 0); <bold>B</bold>: 66 DEGs are up-regulated in the 3 datasets (logFC < 0).
2.2. 差异表达基因的GO和KEGG富集分析
使用ClusterProfiler对所有256个差异表达基因进行富集分析后得出：生物学过程方面，上调的差异表达基因主要富集于有丝分裂核分裂、细胞分化、细胞器分裂、有丝分裂核分裂调节、参与有丝分裂的微管细胞骨架和核分裂调控等过程。下调基因主要富集于细胞-基质粘附、血管形成、循环系统中的血管过程、细胞-基质粘附调控、变形细胞迁移和细胞外结构组织等过程；分子功能方面，上调基因主要富集于金属内肽酶活性、细胞外基质结构成分、赋予抗张强度的细胞外基质结构成分、金属肽酶活性、糖胺聚糖结合和组蛋白激酶活性等。下调基因主要富集于糖胺聚糖结合、细胞外基质结构成分、硫化合物结合、肝素结合、跨膜受体蛋白丝氨酸/苏氨酸激酶结合和离子通道结合等；细胞组分方面，上调基因主要富集于纺垂体、浓缩染色体、着丝粒区、着丝粒、染色体、着丝粒区，浓缩染色体着丝粒和纺锤极。下调基因主要富集于收缩纤维部分、应力纤维、收缩肌动蛋白丝束、收缩纤维、肌动蛋白丝束和放线菌素（图 2，表 1）。
2.
66个上调DEGs和190个下调DEGs的GO富集分析直方图（前20个结果）
Histogram of the top 20 results of GO enrichment analysis of 66 up-regulated DEGs (A-C) and 190 down-regulated DEGs (D-F).
1.
差异表达基因的GO富集分析
Gene ontology analysis of the DEGs
| Expression | Category | ID | Term | n | P |
| Up-Regulated | BP | GO:0140014 | Mitotic nuclear division | 17 | < 0.01 |
| BP | GO:0000280 | Nuclear division | 19 | < 0.01 | |
| BP | GO:0048285 | Organelle fission | 19 | < 0.01 | |
| BP | GO:0007088 | Regulation of mitotic nuclear division | 12 | < 0.01 | |
| BP | GO:1902850 | Microtubule cytoskeleton organization involved in mitosis | 11 | < 0.01 | |
| BP | GO:0051783 | Regulation of nuclear division | 12 | < 0.01 | |
| CC | GO:0005819 | Spindle | 14 | < 0.01 | |
| CC | GO:0000779 | Condensed chromosome, centromeric region | 10 | < 0.01 | |
| CC | GO:0000776 | Kinetochore | 10 | < 0.01 | |
| CC | GO:0000775 | Chromosome, centromeric region | 11 | < 0.01 | |
| CC | GO:0000777 | Condensed chromosome kinetochore | 9 | < 0.01 | |
| CC | GO:0000922 | Spindle pole | 10 | < 0.01 | |
| MF | GO:0004222 | Metalloendopeptidase activity | 5 | < 0.01 | |
| MF | GO:0005201 | Extracellular matrix structural constituent | 5 | < 0.01 | |
| MF | GO:0030020 | Extracellular matrix structural constituent conferring tensile strength | 3 | < 0.01 | |
| MF | GO:0008237 | Metallopeptidase activity | 5 | < 0.01 | |
| MF | GO:0005539 | Glycosaminoglycan binding | 5 | < 0.01 | |
| MF | GO:0035173 | Histone kinase activity | 2 | < 0.01 | |
| Down-Regulated | BP | GO:0031589 | Cell-substrate adhesion | 18 | < 0.01 |
| BP | GO:0001570 | Vasculogenesis | 9 | < 0.01 | |
| BP | GO:0003018 | Vascular process in circulatory system | 12 | < 0.01 | |
| BP | GO:0010810 | Regulation of cell-substrate adhesion | 13 | < 0.01 | |
| BP | GO:0001667 | Ameboidal-type cell migration | 17 | < 0.01 | |
| BP | GO:0043062 | Extracellular structure organization | 16 | < 0.01 | |
| CC | GO:0044449 | Contractile fiber part | 10 | < 0.01 | |
| CC | GO:0001725 | Stress fiber | 6 | < 0.01 | |
| CC | GO:0097517 | Contractile actin filament bundle | 6 | < 0.01 | |
| CC | GO:0043292 | Contractile fiber | 10 | < 0.01 | |
| CC | GO:0032432 | Actin filament bundle | 6 | < 0.01 | |
| CC | GO:0042641 | Actomyosin | 6 | < 0.01 | |
| MF | GO:0005539 | Glycosaminoglycan binding | 11 | < 0.01 | |
| MF | GO:0005201 | Extracellular matrix structural constituent | 9 | < 0.01 | |
| MF | GO:1901681 | Sulfur compound binding | 10 | < 0.01 | |
| MF | GO:0008201 | Heparin binding | 8 | < 0.01 | |
| MF | GO:0070696 | Transmembrane receptor protein serine/threonine kinase binding | 3 | < 0.01 | |
| MF | GO:0044325 | Ion channel binding | 6 | < 0.01 |
对差异表达基因进行KEGG富集分析的结果得出，调基因主要富集于细胞周期、卵母细胞减数分裂、孕酮介导的卵母细胞成熟、p53信号通路和ECM-受体相互作用（图 3A），下调基因主要富集于疟疾（图 3B，表 2）。
3.
66个上调基因和190个下调基因的KEGG富集气泡图
Bubble plot of KEGG enrichment analysis results for 66 up-regulated DEGs (A) and 190 down-regulated DEGs (B).
2.
KEGG富集结果
KEGG pathway analysis of DEGs
| DEG | Pathway ID | Name | Count | GeneRatio | P.adjust |
| Up-regulated | |||||
| hsa04110 | Cell cycle | 6 | 18% | 0.00040954 | |
| hsa04114 | Oocyte meiosis | 6 | 18% | 0.00040954 | |
| hsa04914 | Progesterone-mediated oocyte maturation | 5 | 15% | 0.001106453 | |
| hsa04115 | p53 signaling pathway | 4 | 12% | 0.00353443 | |
| hsa04512 | ECM-receptor interaction | 4 | 9% | 0.006116403 | |
| Down-regulated | |||||
| hsa05144 | Malaria | 5 | 6% | 0.022398395 | |
| hsa04514 | Cell adhesion molecules (CAMs) | 7 | 9% | 0.056375369 | |
2.3. 蛋白质-蛋白质相互作用网络（PPI）以及模组分析
将256个差异表达基因导入STRING在线工具后，203个基因被识别，得出包含203个基因节点，862根连线的PPI网络。其中58个上调，145个下调（图 4A）。将PPI网络数据进行整理后导入Cytotype进行MCODE聚类和模组分析，取计算后Score值最大的模块作为最终的功能模块，得出32个在PPI网络中占据最主要地位的差异表达基因（MCODE基因），32个MCODE基因均为上调基因（图 4B）。
4.
差异表达基因使用STRING生成的PPI网络及其MCODE分析结果
PPI network analysis of DEGs using STRING online tool and the MCODE plugin. A: PPI network containing 203 DEGs, each node representing a protein, and each line representing an interaction between proteins. Blue node represents a down-regulated gene, and a red one an up-regulated gene. B: Module analysis by Cytoscape's MCODE plugin (degree cutoff=2, node score cutoff=0.2, k-core=2, and max. Depth=100).
2.4. MCODE基因的生存分析以及TCGA差异表达分析
将32个MCODE基因使用Kaplan Meier plotter基于EGA、TCGA和GEO数据库的所有866名患者进行生存分析后发现，32个基因中的29个在肺腺癌患者中有着显著更差的总生存期（图 5，表 3）。全部32个基因对于肺鳞癌患者的预后都无统计学意义。这可能是由于前一步的MCODE模组分析提取MCODE基因的原理是基于蛋白质互相作用，从而导致MCODE基因之间有着很强的协同效应，进而从非小细胞肺癌（包括肺鳞癌以及肺腺癌）的差异表达基因中筛选出了与肺腺癌最为相关的基因。接着，我们使用GEPIA工具基于TCGA数据库验证了上述29个差异表达基因在肺腺癌组织和正常肺组织中的表达情况，结果表明，相较于正常肺组织，包括TTK和MAD2L1在内的所有29个核心基因全部在肺腺癌组织中上调，且上调具有统计学意义（P < 0.05，图 6）。
5.
使用Kaplan Meier plotter对32个DEGs的表达和肺腺癌患者总生存期的关系进行分析后具有统计学差异（P < 0.05）的代表基因
Prognostic analysis of 32 core genes using Kaplan Meier plotter online tool for survival analysis. 29 of the 32 core genes are associated with a shorter survival of patients with lung adenocarcinoma (P < 0.05). A, B: Representative genes TTK and MAD2L1.
3.
32个基因的生存分析结果
Correlation of the 32 candidate genes with the patients' survival outcomes
| Category | Genes |
| Genes with significantly worse survival (P < 0.05) | RRM2 UBE2T CENPU MELK KIAA0101 BIRC5 CENPF TTK ZWINT NUF2 BUB1 KIF20A DLGAP5 UBE2C NEK2 CCNB1 CCNB2 ASPM AURKA TPX2 CDKN3 GTSE1 KIF4A TYMS MAD2L1 CDC20 ANLN TOP2A CEP55 |
| Genes without significantly worse survival (P>0.05) | HMMR CDCA7 KIF11 |
6.
29个基因在肺腺癌组织和正常肺组织中的表达量对比分析后具有统计学差异的代表基因
Expression of 29 core genes in lung adenocarcinoma and normal lung tissues. 29 prognostic-related genes were analyzed using the GEPIA online tool, and all 29 genes were significantly higher in lung adenocarcinoma (P < 0.05). (A-B) Representative genes TTK and MAD2L1.
2.5. 29个核心基因的二次KEGG富集分析
为了进一步探究29个关键基因可能参与调节的信号通路，我们使用ClusterProfiler重新对得出的29个基因进行KEGG富集分析。结果表明，5个基因（CCNB1、CCNB2、MAD2L1、BUB1和CDC20）同时富集在卵母细胞减数分裂以及细胞周期通路中，1个基因（AURKA）单独富集在卵母细胞减数分裂通路中，1个基因（TTK）单独富集在细胞周期通路中（表 4）。有趣的是，这七个基因均为SAC复合体相关基因。同时近期有过卵母细胞减数分裂通路在非生殖细胞肿瘤中发挥调控作用的报道，包括结肠癌、膀胱癌和宫颈癌[17-19]。
4.
29个基因的KEGG二次富集分析
KEGG pathway enrichment re-analysis of 29 selected genes
| Pathway ID | Name | Count | GeneRatio | P.adjust | Genes |
| hsa04110 | Cell cycle | 6 | 43% | 4.83E-07 | CCNB1 MAD2L1 CCNB2 BUB1 AURKACDC20 |
| hsa04114 | Oocyte meiosis | 6 | 43% | 4.83E-07 | CCNB1 MAD2L1 CCNB2 BUB1 TTK CDC20 |
| hsa04914 | Progesterone-mediated oocyte maturation | 5 | 36% | 3.81E-06 | CCNB1 MAD2L1 CCNB2 BUB1 AURKA |
| hsa04115 | p53 signaling pathway | 4 | 29% | 3.33E-05 | CCNB1 CCNB2 RRM2 GTSE1 |
2.6. 验证组织样本中关键DEGs的表达
为了进一步确认我们使用生物信息学方法筛选的预后相关的关键DEG在肿瘤组织中的表达，我们使用免疫组化（IHC）验证了其在人肺腺癌组织中的表达情况。首先，我们使用了The Human Protein Atlas（HPA）数据库验证全部7个基因在肺腺癌患者组织标本中的表达。除去BUB1基因的表达情况未被数据库收录外，其他全部6个基因在肿瘤组织和正常组织中的表达情况如图所示（图 7A）。结果表明，6个关键差异表达基因（CCNB1、CCNB2、MAD2L1、CDC20、AURKA和TTK）均在肿瘤标本中呈现过表达趋势。为了进一步验证关键差异基因在肿瘤患者中的高表达趋势，我们挑选了两个具有代表性的基因（MAD2L1和TTK），用免疫组化的方法在肺腺癌患者的肿瘤组织标本和癌旁正常组织上进行了体外实验，我们观察到，相较于癌旁组织，肿瘤组织的细胞核和细胞质的MAD2L1（图 7B）和TTK（图 7C）都出现了过表达的趋势，这与我们的假设相符。
7.
使用免疫组化验证MAD2L1和TTK在肺腺癌肿瘤组织和癌旁组织中的表达情况
Validation of MAD2L1 and TTK expression in human lung adenocarcinoma samples and normal lung tissues using immunohistochemistry. A: Representative images of key genes expression in the human lung adenocarcinoma samples and normal lung tissues using the HPA database. B: Representative images of IHC analysis stained with anti-MAD2L1antibodies (brown) on adjacent tissues (top) and lung adenocarcinoma tissues (bottom). C: Representative images of IHC analysis stained with anti-TTK antibodies (brown) on adjacent tissues (top) and lung adenocarcinoma tissues (bottom).
3. 讨论
本研究通过利用3个基因芯片表达谱数据集（GSE18842、GSE27262和GSE33532）进行生物信息学分析，筛选出29个与肺腺癌预后相关且在肺癌组织中显著高表达的基因。最终确定了7个以TTK和MAD2L1为代表的SAC复合体相关基因在肺腺癌中过表达。上述发现提示，针对肺腺癌患者，靶向SAC复合体相关分子将可能成为肺腺癌靶向治疗新的研究发现和潜在靶点。
细胞周期是哺乳动物生长发育必不可缺的关键步骤，细胞周期的异常和紊乱以及相关蛋白的变化是人类癌症的一个标志[20]。其中，细胞周期检查点的异常被认为是导致多种肿瘤的重要原因。细胞周期检查点负责在细胞周期的关键节点上检查基因组异常，启动基因修复或凋亡信号，确保细胞能够正常进入细胞周期的下一阶段。细胞周期检查点相关蛋白也被认为是抗肿瘤治疗的关键靶点[21]。本研究筛选出的关键基因集中在细胞周期通路，且尤其与G2/M期检查点和纺锤体组装检查点（SAC）密切相关。这不但说明了肺腺癌中普遍存在细胞周期信号通路尤其是细胞周期检查点的异常，也为这些基因进一步的临床转化和药物干预提供了坚实的基础。
临床中常规化疗药物紫杉醇类和长春碱类药物共属抗微管类药物（AMCDs），是使用最为广泛的抗肿瘤药物，其抗肿瘤作用和耐药机制都和SAC的功能密切相关[22-23]。而TTK（又称Mps1）是SAC的重要组成部分，对于维持正常的细胞周期有着关键作用[6]。TTK在包括胰腺癌、胶质母细胞癌、乳腺癌和甲状腺癌的多种肿瘤中过表达[24-25]。近期，一项基于三阴乳腺癌的研究发现[7]，TTK激酶抑制剂BOS172722可导致SAC功能失调，加速染色体错配的异常肿瘤细胞分裂，最终导致凋亡。并且BOS172722和抗微管类药物连用能够产生强大的协同作用，在体内实验中观察到了更小的肿瘤和更长的生存期，目前基于TTK抑制剂联合治疗的I期临床实验正在进行中（NCT03328494）。我们的研究也发现相较于正常组织或癌旁组织，TTK在肺腺癌过表达。紫杉醇类药物在非小细胞肺癌中有着广泛的应用，是肺癌的一线化疗药。因此，未来基于TTK抑制剂和AMCDs的联合治疗模式或许可以为肺腺癌患者带来新的希望。同时，另一个TTK抑制剂CFI-402257介导的SAC抑制能够导致肿瘤细胞的基因组不稳定，诱发新的抗原表位，增强PD-1免疫治疗的疗效[8]，相应的I期临床实验也仍在进行中（NCT02792465）。目前，PD-1/ PD-L1信号轴的免疫检查点抑制剂在非小细胞肺癌中已经取得了突破性的应用和发展[26]，免疫治疗和TTK抑制剂的联合应用值得期待。综上所述，我们研究筛选出的关键基因TTK可能成为未来肺腺癌的关键治疗靶点，基于TTK抑制剂的联合模式治疗肺腺癌可能成为一种新的治疗模式，也需要更充分的临床及转化研究来进一步证实。
此外，同属细胞周期蛋白家族的G2期有丝分裂特异性周期蛋白B1（CCNB1）和G2期有丝分裂特异性周期蛋白B2（CCNB2）也同样是SAC复合体相关蛋白，分别与微管和高尔基复合体共定位，负责结合CDK1调控细胞在G2/M期的正常行为。二者都被观察到在多种肿瘤中过表达，且被认为与多种肿瘤的预后相关[27-29]。对CCNB1或CCNB2的敲除也能够对多种肿瘤产生抑制作用[30-32]。目前，以此为靶点的大多数尝试都针对和细胞周期蛋白B结合的CDK1亚基。我们注意到一项基于结直肠癌和胰腺癌的研究指出CDK1是KRAS突变的合成致死靶点，CDK1的抑制会通过合成致死效应导致KRAS突变的肿瘤细胞死亡[33]。而KRAS突变是非小细胞肺癌最常见的驱动基因突变之一，且会导致“难治性”肺癌。CDK1抑制剂在细胞周期蛋白B过表达的KRAS突变肺腺癌患者上的应用也许会为该类患者的治疗带来曙光。
类似的，CDC20和BUB1也是SAC复合体的重要组成部分，在乳腺癌、胰腺癌和结肠癌中都观察到了CDC20与预后相关的过表达[34]，对CDC20的抑制能够导致前列腺癌细胞对多西他赛的敏感性增强[35]。而BUB1的抑制剂BAY1816032也在乳腺癌中有着杀伤肿瘤细胞和抗微管类药物增敏的作用[36]。本研究发现的SAC相关基因在肺腺癌中呈现出的“联合”过表达状态暗示肺腺癌或许是SAC相关靶点治疗的理想瘤种。
综上所述，本研究通过对肺腺癌相关基因芯片大数据的深入挖掘，运用一系列生物信息学分析手段，最终得出了7个关键基因在肺腺癌中高表达，并且验证了这些基因直接与患者预后显著相关。同时也发现了这些关键基因参与细胞周期调控的主要功能。上述发现可能为肺癌靶向治疗提供新的生物标志物，同时也为肺癌联合治疗提供新的理论依据。
Biography
刘铸，硕士，E-mail: awypass@163.com
Funding Statement
国家自然科学基金（81872399，81672756）；广东省自然科学基金（2017A030311023，2018030310285）
Supported by National Natural Science Foundation of China (81872399, 81672756)
Contributor Information
刘 铸 (Zhu LIU), Email: awypass@163.com.
吴 德华 (Dehua WU), Email: 18602062748@163.com.
董 忠谊 (Zhongyi DONG), Email: dongzy1317@foxmail.com.
References
- 1.Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394–424. doi: 10.3322/caac.21492. [Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2018, 68(6): 394-424.] [DOI] [PubMed] [Google Scholar]
- 2.Zhang YL, Yuan JQ, Wang KF, et al. The prevalence of EGFR mutation in patients with non-small cell lung cancer: a systematic review and meta-analysis. Oncotarget. 2016;7(48):78985–93. doi: 10.18632/oncotarget.12587. [Zhang YL, Yuan JQ, Wang KF, et al. The prevalence of EGFR mutation in patients with non-small cell lung cancer: a systematic review and meta-analysis[J]. Oncotarget, 2016, 7(48): 78985-93.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhou XY, Shou JW, Sheng J, et al. Molecular and clinical analysis of Chinese patients with anaplastic lymphoma kinase (ALK)- rearranged non-small cell lung cancer. Cancer Sci. 2019;110(10):3382–90. doi: 10.1111/cas.14177. [Zhou XY, Shou JW, Sheng J, et al. Molecular and clinical analysis of Chinese patients with anaplastic lymphoma kinase (ALK)- rearranged non-small cell lung cancer[J]. Cancer Sci, 2019, 110 (10): 3382-90.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chang YC, Tseng YL, Leu WJ, et al. Discovery of novel agents on spindle assembly checkpoint to sensitize vinorelbine-induced mitotic cell death against human non-small cell lung cancers. Int J Mol Sci. 2020;21(16):E5608. doi: 10.3390/ijms21165608. [Chang YC, Tseng YL, Leu WJ, et al. Discovery of novel agents on spindle assembly checkpoint to sensitize vinorelbine-induced mitotic cell death against human non-small cell lung cancers[J]. Int J Mol Sci, 2020, 21(16): E5608.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lei H, Wang K, Jiang T, et al. KIAA0101 and UbcH10 interact to regulate non-small cell lung cancer cell proliferation by disrupting the function of the spindle assembly checkpoint[J], 2020.
- 6.Pachis ST, Kops GJPL. Leader of the SAC: molecular mechanisms of Mps1/TTK regulation in mitosis. Open Biol. 2018;8(8):180109. doi: 10.1098/rsob.180109. [Pachis ST, Kops GJPL. Leader of the SAC: molecular mechanisms of Mps1/TTK regulation in mitosis[J]. Open Biol, 2018, 8(8): 180109.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Anderhub SJ, Mak GW, Gurden MD, et al. High proliferation rate and a compromised spindle assembly checkpoint confers sensitivity to the MPS1 inhibitor BOS172722 in triple-negative breast cancers. Mol Cancer Ther. 2019;18(10):1696–707. doi: 10.1158/1535-7163.MCT-18-1203. [Anderhub SJ, Mak GW, Gurden MD, et al. High proliferation rate and a compromised spindle assembly checkpoint confers sensitivity to the MPS1 inhibitor BOS172722 in triple-negative breast cancers [J]. Mol Cancer Ther, 2019, 18(10): 1696-707.] [DOI] [PubMed] [Google Scholar]
- 8.Mason JM, Wei X, Fletcher GC, et al. Functional characterization of CFI-402257, a potent and selective Mps1/TTK kinase inhibitor, for the treatment of cancer. Proc Natl Acad Sci USA. 2017;114(12):3127–32. doi: 10.1073/pnas.1700234114. [Mason JM, Wei X, Fletcher GC, et al. Functional characterization of CFI-402257, a potent and selective Mps1/TTK kinase inhibitor, for the treatment of cancer[J]. Proc Natl Acad Sci USA, 2017, 114(12): 3127-32.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Koch A, Maia A, Janssen A, et al. Molecular basis underlying resistance to Mps1/TTK inhibitors. Oncogene. 2016;35(19):2518–28. doi: 10.1038/onc.2015.319. [Koch A, Maia A, Janssen A, et al. Molecular basis underlying resistance to Mps1/TTK inhibitors[J]. Oncogene, 2016, 35(19): 2518-28.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Clough E, Barrett T: The gene expression omnibus database, Statistical genomics: Springer, 2016: 93-110.
- 11.Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional genomics data sets: update. http://pubmedcentralcanada.ca/pmcc/articles/PMC3013736/ Nucleic Acids Res. 2013;41(Database issue):D991–5. doi: 10.1093/nar/gks1193. [Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional genomics data sets: update[J]. Nucleic Acids Res, 2013, 41(Database issue): D991-5.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yu G. clusterProfiler: universal enrichment tool for functional and comparative study. http://www.biorxiv.org/node/78933. BioRxiv. 2018:256784. [Yu G. clusterProfiler: universal enrichment tool for functional and comparative study[J]. BioRxiv, 2018: 256784.] [Google Scholar]
- 13.Szklarczyk D, Gable AL, Lyon D, et al. STRING v11: proteinprotein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607–D613. doi: 10.1093/nar/gky1131. [Szklarczyk D, Gable AL, Lyon D, et al. STRING v11: proteinprotein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets[J]. Nucleic Acids Res, 2019, 47(D1): D607-D613.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nagy Á, Lánczky A, Menyhárt O, et al. Validation of miRNA prognostic power in hepatocellular carcinoma using expression data of independent datasets. Sci Rep. 2018;8(1):9227. doi: 10.1038/s41598-018-27521-y. [Nagy Á, Lánczky A, Menyhárt O, et al. Validation of miRNA prognostic power in hepatocellular carcinoma using expression data of independent datasets[J]. Sci Rep, 2018, 8(1): 9227.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tang ZF, Li CW, Kang BX, et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017;45(W1):W98–W102. doi: 10.1093/nar/gkx247. [Tang ZF, Li CW, Kang BX, et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses[J]. Nucleic Acids Res, 2017, 45(W1): W98-W102.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Thul PJ, Lindskog C. The human protein atlas: a spatial map of the human proteome. Protein Sci. 2018;27(1):233–44. doi: 10.1002/pro.3307. [Thul PJ, Lindskog C. The human protein atlas: a spatial map of the human proteome[J]. Protein Sci, 2018, 27(1): 233-44.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Qi YW, Qi HW, Liu ZY, et al. Bioinformatics analysis of key genes and pathways in colorectal cancer. J Comput Biol. 2019;26(4):364–75. doi: 10.1089/cmb.2018.0237. [Qi YW, Qi HW, Liu ZY, et al. Bioinformatics analysis of key genes and pathways in colorectal cancer[J]. J Comput Biol, 2019, 26(4): 364-75.] [DOI] [PubMed] [Google Scholar]
- 18.Wu X, Peng L, Zhang YQ, et al. Identification of key genes and pathways in cervical cancer by bioinformatics analysis. Int J Med Sci. 2019;16(6):800–12. doi: 10.7150/ijms.34172. [Wu X, Peng L, Zhang YQ, et al. Identification of key genes and pathways in cervical cancer by bioinformatics analysis[J]. Int J Med Sci, 2019, 16(6): 800-12.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li S, Liu XP, Liu TZ, et al. Identification of biomarkers correlated with the TNM staging and overall survival of patients with bladder cancer. Front Physiol. 2017;8:947. doi: 10.3389/fphys.2017.00947. [Li S, Liu XP, Liu TZ, et al. Identification of biomarkers correlated with the TNM staging and overall survival of patients with bladder cancer[J]. Front Physiol, 2017, 8: 947.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Leal-Esteban L C, Fajas L. Cell cycle regulators in cancer cell metabolism. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease. 2020;1866(5):165715. doi: 10.1016/j.bbadis.2020.165715. [Leal-Esteban L C, Fajas L. Cell cycle regulators in cancer cell metabolism[J]. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease, 2020, 1866(5): 165715.] [DOI] [PubMed] [Google Scholar]
- 21.Wenzel ES, Singh ATK. Cell-cycle checkpoints and aneuploidy on the path to cancer. http://europepmc.org/abstract/MED/29275292. In Vivo. 2018;32(1):1–5. doi: 10.21873/invivo.11197. [Wenzel ES, Singh ATK. Cell-cycle checkpoints and aneuploidy on the path to cancer[J]. In Vivo, 2018, 32(1): 1-5.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Visconti R, Grieco D. Fighting tubulin-targeting anticancer drug toxicity and resistance. Endocr Relat Cancer. 2017;24(9):T107–T117. doi: 10.1530/ERC-17-0120. [Visconti R, Grieco D. Fighting tubulin-targeting anticancer drug toxicity and resistance[J]. Endocr Relat Cancer, 2017, 24(9): T107-T117.] [DOI] [PubMed] [Google Scholar]
- 23.Raab M, Kobayashi NF, Becker S, et al. Boosting the apoptotic response of high-grade serous ovarian cancers with CCNE1 amplification to paclitaxel in vitro by targeting APC/C and the pro-survival protein MCL-1. Int J Cancer. 2020;146(4):1086–98. doi: 10.1002/ijc.32559. [Raab M, Kobayashi NF, Becker S, et al. Boosting the apoptotic response of high-grade serous ovarian cancers with CCNE1 amplification to paclitaxel in vitro by targeting APC/C and the pro-survival protein MCL-1[J]. Int J Cancer, 2020, 146(4): 1086-98.] [DOI] [PubMed] [Google Scholar]
- 24.Xie Y, Wang AQ, Lin JZ, et al. Mps1/TTK: a novel target and biomarker for cancer. J Drug Target. 2017;25(2):112–8. doi: 10.1080/1061186X.2016.1258568. [Xie Y, Wang AQ, Lin JZ, et al. Mps1/TTK: a novel target and biomarker for cancer[J]. J Drug Target, 2017, 25(2): 112-8.] [DOI] [PubMed] [Google Scholar]
- 25.Simon Serrano S, Sime W, Abassi Y, et al. Inhibition of mitotic kinase Mps1 promotes cell death in neuroblastoma. Sci Rep. 2020;10(1):11997. doi: 10.1038/s41598-020-68829-y. [Simon Serrano S, Sime W, Abassi Y, et al. Inhibition of mitotic kinase Mps1 promotes cell death in neuroblastoma[J]. Sci Rep, 2020, 10(1): 11997.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE- 042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393(10183):1819–30. doi: 10.1016/S0140-6736(18)32409-7. [Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE- 042): a randomised, open-label, controlled, phase 3 trial[J]. Lancet, 2019, 393(10183): 1819-30.] [DOI] [PubMed] [Google Scholar]
- 27.Chai N, Xie HH, Yin JP, et al. FOXM1 promotes proliferation in human hepatocellular carcinoma cells by transcriptional activation of CCNB1. Biochem Biophys Res Commun. 2018;500(4):924–9. doi: 10.1016/j.bbrc.2018.04.201. [Chai N, Xie HH, Yin JP, et al. FOXM1 promotes proliferation in human hepatocellular carcinoma cells by transcriptional activation of CCNB1[J]. Biochem Biophys Res Commun, 2018, 500(4): 924-9.] [DOI] [PubMed] [Google Scholar]
- 28.Chen ZD, Chen JH, Huang X, et al. Identification of potential key genes for hepatitis B virus-associated hepatocellular carcinoma by bioinformatics analysis. J Comput Biol. 2019;26(5):485–94. doi: 10.1089/cmb.2018.0244. [Chen ZD, Chen JH, Huang X, et al. Identification of potential key genes for hepatitis B virus-associated hepatocellular carcinoma by bioinformatics analysis[J]. J Comput Biol, 2019, 26(5): 485-94.] [DOI] [PubMed] [Google Scholar]
- 29.Li R, Jiang XM, Zhang YG, et al. Cyclin B2 overexpression in human hepatocellular carcinoma is associated with poor prognosis. Arch Med Res. 2019;50(1):10–7. doi: 10.1016/j.arcmed.2019.03.003. [Li R, Jiang XM, Zhang YG, et al. Cyclin B2 overexpression in human hepatocellular carcinoma is associated with poor prognosis [J]. Arch Med Res, 2019, 50(1): 10-7.] [DOI] [PubMed] [Google Scholar]
- 30.Zhang H, Zhang X, Li X, et al. Effect of CCNB1 silencing on cell cycle, senescence, and apoptosis through the p53 signaling pathway in pancreatic cancer. http://onlinelibrary.wiley.com/doi/10.1002/jcp.26816. J Cell Physiol. 2018;234(1):619–31. doi: 10.1002/jcp.26816. [Zhang H, Zhang X, Li X, et al. Effect of CCNB1 silencing on cell cycle, senescence, and apoptosis through the p53 signaling pathway in pancreatic cancer[J]. J Cell Physiol, 2018, 234(1): 619-31.] [DOI] [PubMed] [Google Scholar]
- 31.Lei CY, Wang W, Zhu YT, et al. The decrease of cyclin B2 expression inhibits invasion and metastasis of bladder cancer. Urol Oncol. 2016;34(5):237.e1–10. doi: 10.1016/j.urolonc.2015.11.011. [Lei CY, Wang W, Zhu YT, et al. The decrease of cyclin B2 expression inhibits invasion and metastasis of bladder cancer[J]. Urol Oncol, 2016, 34(5): 237.e1-10.] [DOI] [PubMed] [Google Scholar]
- 32.Li HX, Tian XF, Wang PQ, et al. MicroRNA-582-3p negatively regulates cell proliferation and cell cycle progression in acute myeloid leukemia by targeting cyclin B2. Cell Mol Biol Lett. 2019;24:66. doi: 10.1186/s11658-019-0184-7. [Li HX, Tian XF, Wang PQ, et al. MicroRNA-582-3p negatively regulates cell proliferation and cell cycle progression in acute myeloid leukemia by targeting cyclin B2[J]. Cell Mol Biol Lett, 2019, 24: 66.] [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
- 33.Costa-Cabral S, Brough R, Konde A, et al. CDK1 is a synthetic lethal target for KRAS mutant tumours. PLoS One. 2016;11(2):e0149099. doi: 10.1371/journal.pone.0149099. [Costa-Cabral S, Brough R, Konde A, et al. CDK1 is a synthetic lethal target for KRAS mutant tumours[J]. PLoS One, 2016, 11(2): e0149099.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gayyed MF, El-Maqsoud NM, Tawfiek ER, et al. A comprehensive analysis of CDC20 overexpression in common malignant tumors from multiple organs: its correlation with tumor grade and stage. Tumour Biol. 2016;37(1):749–62. doi: 10.1007/s13277-015-3808-1. [Gayyed MF, El-Maqsoud NM, Tawfiek ER, et al. A comprehensive analysis of CDC20 overexpression in common malignant tumors from multiple organs: its correlation with tumor grade and stage[J]. Tumour Biol, 2016, 37(1): 749-62.] [DOI] [PubMed] [Google Scholar]
- 35.Li K, Mao YH, Lu L, et al. Silencing of CDC20 suppresses metastatic castration-resistant prostate cancer growth and enhances chemosensitivity to docetaxel. Int J Oncol. 2016;49(4):1679–85. doi: 10.3892/ijo.2016.3671. [Li K, Mao YH, Lu L, et al. Silencing of CDC20 suppresses metastatic castration-resistant prostate cancer growth and enhances chemosensitivity to docetaxel[J]. Int J Oncol, 2016, 49(4): 1679-85.] [DOI] [PubMed] [Google Scholar]
- 36.Siemeister G, Mengel A, Fernández-Montalván AE, et al. Inhibition of BUB1 kinase by BAY1816032 sensitizes tumor cells toward taxanes, ATR, and PARP inhibitors in vitro and in vivo. Clin Cancer Res. 2019;25(4):1404–14. doi: 10.1158/1078-0432.CCR-18-0628. [Siemeister G, Mengel A, Fernández-Montalván AE, et al. Inhibition of BUB1 kinase by BAY1816032 sensitizes tumor cells toward taxanes, ATR, and PARP inhibitors in vitro and in vivo[J]. Clin Cancer Res, 2019, 25(4): 1404-14.] [DOI] [PubMed] [Google Scholar]