Genomic profiling of hereditary epithelial ovarian cancer using tiling resolution array CGH

Heredity is a major risk factor for ovarian cancer, but many families escape detection. Refined diagnosis of ovarian cancers linked to the breast and ovarian cancer (HBOC) syndrome and the hereditary nonpolyposis colorectal cancer (HNPCC) syndrome would allow cancer prevention in high risk families. In order to delineate genetic profiles of hereditary ovarian cancer, we applied genome wide array comparative genomic hybridization to 24 sporadic tumors, 12 HBOC associated tumors (BRCA1 mutations) and eight HNPCC associated tumors (mismatch repair gene mutations). Unsupervised cluster analysis identified two distinctive clusters related to genetic complexity. Most sporadic and HBOC associated tumors had complex genetic profiles with multiple gains and losses with an average of 41% of the genome altered, whereas mismatch repair defective tumors had stable genetic profiles, with an average of 18% of the genome altered. Losses of 4q34, 13q12-q32 and 19p13 were overrepresented in the HBOC subset, gains of chromosomes 17 and 19 characterized the HNPCC tumors and gains of 20q11 were more common in the sporadic tumors. The genetic distinction between HBOC and HNPCC associated ovarian cancer suggests that genetic profiles can be applied for refined classification of hereditary cases and reflects tumor development along different genetic pathways. DNA was extracted from formalin fixed, paraffin embedded tumor tissue according to protocols from the UCSF Waldman Laboratory, San Francisco, CA, USA (http://cc.ucsf.edu/people/waldman/Protocols/paraffin.html), with an additional purification step using Phase Lock Gel tubes (Eppendorf AG, Hamburg, Germany). DNA quality was assessed using a Ready-To-Go RAPD analysis kit (GE Healthcare, Little Chalfont, UK) with primers 5’-AATCGGGCTG-3’ and 5’-GAACGGGTG-3’. PCR products were validated on a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Tiling 32k BAC microarrays, with contiguous genome wide coverage, were produced at the Microarray DNA Resource Centre, SCIBLU Genomics, Department of Oncology, Lund University, Sweden (http://www.lth.se/sciblu). Labeling and hybridization were performed in short, 2-8 μg tumor DNA and 2 µg reference DNA (Promega Corporation, Madison, WI, USA) were labeled with Cy3-dCTP and Cy5-dCTP, using BioPrime Array CGH Genomic Labeling System (Invitrogen Life Technologies, Carlsbad, CA, USA). Tumor and reference DNA were pooled and mixed with Human COT-1 DNA (Invitrogen). Hybridizations were conducted using the MAUI® Hybridization System (BioMicro systems Inc, Salt Lake City, UT, USA) and the slides were scanned in an Agilent Microarray Scanner (Agilent Technologies).

遗传是卵巢癌的主要危险因素，但诸多家族未被检出。对与乳腺与卵巢癌综合征（breast and ovarian cancer syndrome, HBOC）及遗传性非息肉病性结直肠癌综合征（hereditary nonpolyposis colorectal cancer syndrome, HNPCC）相关的卵巢癌开展精准诊断，可助力高危家族的癌症预防。
为阐明遗传性卵巢癌的遗传谱，我们对24例散发性肿瘤、12例HBOC相关肿瘤（伴BRCA1基因突变）以及8例HNPCC相关肿瘤（伴错配修复基因突变）实施了全基因组阵列比较基因组杂交（genome wide array comparative genomic hybridization）检测。无监督聚类分析识别出两个与遗传复杂度相关的独特聚类簇。多数散发性肿瘤与HBOC相关肿瘤呈现复杂的遗传谱，存在大量拷贝数扩增与缺失，平均有41%的基因组发生变异；而错配修复缺陷型肿瘤的遗传谱较为稳定，平均仅18%的基因组发生变异。
在HBOC相关肿瘤亚组中，4q34、13q12-q32及19p13区域的拷贝数缺失显著富集；HNPCC相关肿瘤以17号和19号染色体拷贝数扩增为特征；而20q11区域的拷贝数扩增在散发性肿瘤中更为常见。HBOC与HNPCC相关卵巢癌之间的遗传差异表明，遗传谱可用于遗传性卵巢癌病例的精准分型，同时也反映了肿瘤通过不同遗传通路发生发展的过程。
本研究按照美国加利福尼亚州旧金山加州大学旧金山分校瓦尔德曼实验室（UCSF Waldman Laboratory）的操作流程（http://cc.ucsf.edu/people/waldman/Protocols/paraffin.html），从福尔马林固定石蜡包埋肿瘤组织中提取DNA，并额外使用Phase Lock Gel管（德国汉堡艾本德公司，Eppendorf AG）完成纯化步骤。采用Ready-To-Go 随机扩增多态性DNA（RAPD）分析试剂盒（英国小查尔方特通用电气医疗集团，GE Healthcare），以引物5’-AATCGGGCTG-3’和5’-GAACGGGTG-3’对DNA质量进行评估。PCR产物通过生物分析仪（美国加利福尼亚州圣克拉拉安捷伦科技公司，Agilent Technologies）完成验证。
本研究使用的具备全基因组连续覆盖能力的32k细菌人工染色体（BAC）铺列微阵列，由瑞典隆德大学肿瘤学系SCIBLU基因组学中心微阵列DNA资源中心制备（http://www.lth.se/sciblu）。标记与杂交步骤简述如下：取2~8 μg肿瘤DNA与2 μg参考DNA（美国威斯康星州麦迪逊普洛麦格公司，Promega Corporation），采用BioPrime阵列CGH基因组标记系统（美国加利福尼亚州卡尔斯巴德英潍捷基生命科技公司，Invitrogen Life Technologies），分别以Cy3-dCTP与Cy5-dCTP进行标记。将标记后的肿瘤DNA与参考DNA混合，并加入人类COT-1 DNA（英潍捷基公司）。杂交反应采用MAUI®杂交系统（美国犹他州盐湖城BioMicro系统公司）进行，芯片扫描使用安捷伦微阵列扫描仪（安捷伦科技公司）完成。