SNP-array profiling of 21 novel primary and metastatic colorectal cancer cell lines [Illumina HumanExome-12 v1.0 BeadChip]

Background: In vitro models are an essential tool towards understanding the molecular characteristics of colorectal cancer (CRC) and the testing of therapies for CRC. To this end we established 21 novel CRC cell lines of which six were derived from liver metastases. Extensive genetic, genomic, transcriptomic and methylomic profiling was performed in order to characterize these new cell lines and all data is made publically available. Additionally, sensitivity of oxaliplatin was tested as a measure for chemotherapy resistance. Results: DNA copy-number alterations (CNA) were compared between primary and metastasis derived cell lines. In concordance with previous studies copy-number gain of chr20, and loss of chr8p were found highly specific for liver metastases. Previously reported BRAF-mutation associated DNA methylation profiles could be validated on the genome-wide DNA methylation profiles of these cell lines. 47.6% of the loci previously reported to associate with BRAF mutation status were reproduced in this dataset. When examining the gene expression profiles in conjunction with these DNA methylation results, we identified 20 genes of which the gene expression correlated with the DNA methylation status, including MEIS1, LRAT and STC2. These genes have previously been reported to be subject to transcriptional regulation through DNA hypermethylation, validating our approach. Conclusions: By combining mutation profiles with CNA and gene expression profiles we constructed an overview of the alterations in the major CRC-related signalling pathways. The mutation profiles, along with the genome, transcriptome and methylome data of these cell lines will be made publically available . This combined dataset puts these cell lines among the best characterized CRC cell lines, allowing researchers to select appropriate cell line models for their particular experiment, making optimal use of these novel cell lines as in vitro model for CRC. SNP-array analysis of 21 novel CRC cell lines; 16 with Illumina HumanExome-12 v1.2 BeadChip and 5 with Illumina HumanExome-12 v1.0 BeadChip.

背景：体外模型（in vitro models）是解析结直肠癌（colorectal cancer, CRC）分子特征、开展CRC治疗药物测试的核心工具。为此，我们构建了21株新型CRC细胞系，其中6株源自肝转移灶。为全面表征这批新细胞系，我们对其开展了全面的遗传、基因组、转录组（transcriptome）与甲基化组（methylome）谱型分析，所有数据均已公开可用。此外，我们还以奥沙利铂（oxaliplatin）敏感性作为化疗耐药性的评价指标进行了检测。
结果：我们比较了原发灶来源与转移灶来源细胞系之间的DNA拷贝数变异（copy-number alterations, CNA）特征。与既往研究结果一致，20号染色体（chr20）拷贝数扩增与8号染色体短臂（chr8p）拷贝数缺失被证实与肝转移灶具有高度特异性关联。既往报道的与BRAF突变（BRAF mutation）相关的DNA甲基化谱特征，可在本批细胞系的全基因组DNA甲基化谱中得到验证：本数据集复现了47.6%既往报道的与BRAF突变状态相关的基因座。结合基因表达谱与上述DNA甲基化结果进行分析，我们鉴定出20个基因，其基因表达水平与DNA甲基化状态显著相关，包括MEIS1、LRAT及STC2。上述基因此前已被报道可通过DNA高甲基化介导转录调控，验证了本研究分析方法的可靠性。
结论：通过整合突变谱、CNA数据与基因表达谱，我们梳理了结直肠癌相关主要信号通路的异常改变情况。本批细胞系的突变谱、基因组、转录组及甲基化组数据均将公开可用。本整合数据集使这批细胞系成为目前表征最全面的CRC细胞系之一，可帮助研究人员为特定实验选择合适的细胞系模型，从而更高效地利用这些新型细胞系作为CRC体外研究模型。本研究针对21株新型CRC细胞系开展了单核苷酸多态性芯片（SNP-array）分析，其中16株使用Illumina HumanExome-12 v1.2 微珠芯片，5株使用Illumina HumanExome-12 v1.0 微珠芯片。