The Study Of Structural Mutations And Functional Verification Of Target Genes Based On Whole Genome Sequencing In Esophageal Squamous Cell Carcinoma

Cancer genomes harbor various somatic forms of genetic alterations spanning from nucleotide-level alterations(including point mutations and small insertions/deletions) to large chromosomal events(including structural variations and copy-number changes), some of which may be the driver mutations for development of ESCC. SVs can promote the development of tumor by a number of ways. In the past study, SVs can damage the functions of tumor suppressor genes directly, and increase the copy numbers of key oncogenes, or form new oncogenetic fusion genes. Therefore, it is a powerful way to identify the SVs regions on tumor chromosomes and the genes in mutated regions.ESCC is the predominant form of esophageal cancer in China. Recently, we and other teams have studied the mutations in ESCC genome, and identified some driver genes with point mutations. Which include some famed genes in ESCC, such as TP53, PIK3 CA, NOTCH1 and CDKN2 A, and new driver genes correlation with ESCC, such as ZNF750, AJUBA, FAT1, et al. But there is no a report about the SVs in genome of ESCC. Objective:Analysis genome structure variations(including structural rearrangement and copy number variation) and describe the structure mutation spectrum and mechanism spectrum in esophageal squamous cell carcinoma. To detect the frequency of complex structure mutation including, kataegis, gene-fusion and BFB cycle in ESCC. And identification of the target genes derived from structural variations in esophageal squamous cell carcinoma. To identify chromothripsis the chromosomal regions with copy number variations, the target genes in which regions, and new driver genes for development of ESCC. Study the expression level of the new driver genes in ESCC and the ability of impacting of ESCC cell proliferation, invasion and migration. To preliminarily explore the role mechanism of the new driver gene in ESCC. Methods:1. Download the WGS data of 31 ESCC from EGA, and analyze the genome structure mutations by corresponding bioinformatics methods. Describe the structure mutation spectrum and mechanism spectrum in esophageal squamous cell carcinoma. Detect the frequency and the target genes derived from complex structure mutation, including chromothripsis, kataegis, gene-fusion and BFB cycle in ESCC.2. FISH demonstrated the amplified CCND1, FGFR1 and LETM2 derived from chromothripsis and BFB cycle;FISH and PCR sanger sequencing demonstrate the fusion events of TRAPPC9-CLVS1 and EIF3E-RAD51 B.3. Downregulate the endogenous expression levels of FGFR1 and LETM2 in esophageal squamous cancer cells by RNAi, test the changes of cell proliferation, invasion and migration by MTT and transwell assay.4. Combined analysis of 31 WGS data and 123 CGH data of ESCC by Patchwork and GISTIC, identify the gain and lost regions on chromosomes and the target genes in which, and further to identify the new driver gene involved in the development of ESCC.5. Detect the CDCA7 expression levels in ESCC tissues by RT-PCR and IHC based on tissue microarray. Downregulate the endogenous expression levels of CDCA7 in ECA109 cell by RNAi, test the changes of cell proliferation, invasion, migration and apoptosis by MTT, transwell and flow cytometry.6.we performed RNA-seq of CDCA7 knockdown cells and cells transfected with Plv sh RNA-puro vector to screen the differentially expressed genes, select the genes related to cell proliferation and apoptosis, described the CDCA7 target pathway in proliferation and apoptosis in ESCC and validated the pathway levels by western blot. Results:1. A total of 5,204 SVs were identified from the 31 ESCC genomes with an average of 168 SVs per tumor. Five categories of SVs were observed, including deletions, tandem duplications, inversions, insertions, and chromosomal translocations. Deletions and chromosomal translocations were the main types in ESCC, make up 35% and 42% respectively.2. NHEJ and alt-EJ were the dominant mechanisms cause SVs in ESCC.3. Across 5,204 SVs, we found that 3,376 SVs occurred in the region of genes and were predicted to directly disrupt sequence of gene. SVs resulted in CDKN2 A deleted recurrently and NOTCH1 inactived in ESCC.4. We observed chromothripsis involved in chromosome 8 of ESCC-16 T. we found a high-level focal amplification of 8p12, FGFR1 and LETM2 locus. FISH analysis demonstrated DM-derived the amplification of FGFR1 and LETM2. LETM2 expressed highly in tumor tissue than in normal tissue. FGFR1 downregulation prevented cell proliferation and migration/invasion. Downregulated LETM2 prevented cell proliferation but has no effect on cell migration/invasion as monitored by MTT or in vitro cell migration and invasion assays.5. We screened gene fusion events across 31 ESCC genomes and identified a total of 173 in-frame fusion genes and 231 out-frame fusion genes affected by SVs. We identified two inframe fusions, TRAPPC9-CLVS1 and EIF3E-RAD51 B, which have the potential to drive carcinogenesis. Then we validated the EIF3E-RAD51 B and TRAPPC9-CLVS1 fusion transcripts via FISH and PCR sanger sequencing.6. We found locally rearranged variations concentrated in chromosome 3 of ESCC-14 T and somatic mutations clustered in the region of 16.9 Mb to 17.5 Mb, and most substitutions within the hypermutated region being characterized by C>T transitions in Tp Cp X trinucleotides, indicates a potential tumorigenic mechanism of kataegis in ESCC development.7. BFB plays an important role in oncogene amplification in ESCC tumors. We observed BFB involving 21 ESCCs. Of these 21 ESCCs, 9 showed evidence of BFB on chromosome 11 led to a focal amplification of CCND1. And the other regions amplified by BFB cycles harbor oncogenes such as EGFR, ERBB2, MMPs, and MYC. FISH showed unbalanced amplified signals indicating that the CCND1 amplification was created by BFB cycles in ESCC. Additionally, we also found inter-chromosomal SVs enriched in CCND1 locus on chromosome 11. Our results demonstrated that at least two mutational mechanisms, focal amplification via BFB cycles and inter-chromosomal translocations, result in CCND1 amplification in ESCC.8. In CNAs analysis, 19 of 31 ESCCs that could be used to determine absolute copy number. Consistently, frequent arm-level changes were observed in ESCC, including frequent copy-number gains of 3q, 5p, 7p, 8q, 12 p, 17 p, 20 p, and 20 q and universal deletions affecting 3p, 4p, 4q, 5q, 10 p, 13 q, and 21 q. Meanwhile, 70% of loss of heterozygosity was copy neutral loss of heterozygosity(CN-LOH) in ESCC. Especially we observed frequent CN-LOH on 13 q and 17 p. And WGD occurred in 13 out of 19 ESCC genomes.9. We applied GISTIC to copy-number profiling from a combination of 31 WGS and 123 CGH data. This analysis yielded 11 amplification peaks and 13 deletion peaks, including cancer genes EGFR, CDK6, AKT1, MYC, CCND1, CDKN2 A, and others.10. Specifically, we identified a focal amplified region corresponding to CDCA7 in 5 out of 31 ESCC genomes. Moreover, we observed that most of individuals with ESCC tumors showed statistically higher expression level of CDCA7 compared with that of normal tissues as determined by real-time PCR and immunohistochemistry analyses(p<0.001).11. CDCA7 knockdown significantly inhibited cell growth and promoted cell apoptosis(p<0.05) but had no differential effect on cell migration and invasion in ESCC cells, indicating that CDCA7 might involve cell proliferation and apoptosis but not metastasis in ESCC.12. By performing RNA-seq of CDCA7 knockdown cells and cells transfected with p LVsh RNA-puro vector, we observed a positive and highly significant enrichment of the expression of cell proliferation or apoptosis-associated target genes, including TRAIL-R, CASP10, IL1R1, CASP7, BCL2 and CASP9. CDCA7 knockdown led to the decrease of phospho-ERK1/2 by WB, suggests that CDCA7 might regulate cell proliferation via FGF21-ERK1/2 MAPK pathway rather than other pathways in ESCC tumorigenesis. Conclusion:1. There are a large number of structural mutations in 31 ESCC genomes. The basic types include deletion, terminal replication, inversion, insertion and chromosomal translocation. Deletion and chromosomal translocation are the main types.2. Complex genomic rearrangements, such as chromothripsis, kataegis and BFB, are an integral part of mutation mechanisms contributing to ESCC development. Which result in the amplification of oncogene FGFR1, LETM2 and CCND1, and derive the potential tumorigenic fusion genes, TRAPPC9-CLVS1 and EIF3E-RAD51 B.3. CN-LOH and WGD are common in the genome of ESCC, may be contribute to the development of ESCC.4. CDCA7, which promoting the proliferation and inhibiting the apoptosis of ESCC cells, is one of the amplified genes in ESCC. And may be a new driver gene revolved in the occurrence and development of ESCC.