We achieved 37- and 43-fold mean sequence coverage of targeted exonic regions, with 73 and 77% of loci covered at 20-fold from tumour and matched blood samples, respectively (Supplementary Fig

We achieved 37- and 43-fold mean sequence coverage of targeted exonic regions, with 73 and 77% of loci covered at 20-fold from tumour and matched blood samples, respectively (Supplementary Fig.?S1). in combination in OSCC cells led to increased cell migration and clonal growth, resistance to Staurosporine-induced apoptosis and, in some cases, increased terminal differentiation. The OSCC lines thus represent a valuable resource for elucidating the impact of different mutations on tumour behaviour. (Caspase 8) and (an atypical cadherin) [6], which are not conventional oncogenes/tumour suppressor genes and can potentially have pleiotropic effects on tumour properties. For example, the association between mutation and overall survival in HNSCC differs according to the HPV status of the tumour [17], and loss of Caspase 8 not only has cell intrinsic effects [18] but can also trigger inflammation [19]. Furthermore, there is evidence for biological interactions between FAT1 and Caspase 8, with FAT1 acting as an antagonist of Caspase 8 in a synthetic lethal screen in cancer cell lines [20]. In this study, we set out to develop new OSCC lines, discover which mutations are tumour-acquired and determine whether they are indeed representative of mutational burden in primary tumours. We then used the lines to explore the impact of mutations in and on cell behaviour. Materials and methods Derivation of OSCC lines Anonymized biopsies of OSCC or normal oral mucosa were collected with appropriate Rabbit Polyclonal to CDH11 ethical approval (UK National Research Ethics Service (08/H0306/30). Cells were isolated and cultured on a feeder layer of J2 (22R)-Budesonide 3T3 cells in complete FAD medium as described previously [16]. Whole exome sequencing Genomic DNA was extracted from OSCC lines (passage 2C4) and patient-matched blood. Whole exome sequencing was performed by Beijing Genomics Institute (Hong Kong). Raw image files were processed by Illumina base calling Software 1.7 or base calling with default parameters, and the sequences of each individual were generated as 90?bp paired-end reads. High-quality reads were aligned against the NCBI human reference genome (hg19) using Burrows-Wheeler Aligner (v0.5.9) with default parameters. Picard (v1.54) was employed to mark duplicates and was followed by Genome Analysis Toolkit (v1.0.6076, GATK IndelRealigner) to improve alignment accuracy. Putative somatic single nucleotide variations (SNVs) were predicted by VarScan2.25 with the parameters as — min-coverage 5 –min-coverage-normal 5 –min-coverage-tumour 5 –min-var-freq 0.1 –min-freq-for-hom 0.75 –min-avg-qual 0 somatic-p-value 0.15. In order to obtain high confidence somatic SNVs, an in-house pipeline was applied. Somatic InDels were predicted by GATK SomaticInDelDetector with default parameters. A pipeline was developed to obtain high confidence somatic InDels; normal and tumour bam were reused to perform local realignment and germline indels were filtered for high confidence indels, with normal coverage and tumour coverage no less than 5. High confidence somatic single nucleotide variants and InDels were annotated using ANNOVAR. Functional impacts of missense mutations were predicted using SIFT, PolyPhen2, PhyloP, MutationTaster and LRT annotations. Prediction of driver genes and pathways The Oncodrive-fm method was applied, as (22R)-Budesonide previously (22R)-Budesonide published, to identify significantly mutant genes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways [21]. Pathway enrichment analysis was also performed to identify additional significantly mutated KEGG pathways. KEGG pathway analysis and clustering Whole exome sequencing data from The Cancer Gene Atlas (TCGA) HNSCC collection [6] were accessed from cBioPortal.org. KEGG pathway analysis was performed; Clog2(value of 0.05 was considered significant, unless otherwise noted. Results Whole exome sequencing of OSCC lines We derived multiple low passage polyclonal cell lines from primary oral squamous cell carcinoma biopsies by culture on a 3T3 J2 feeder layer in order to minimise selection for rapidly dividing cells [16]. Whole exome sequencing was performed on 16 lines, together with patient-matched blood. We achieved 37- and 43-fold mean sequence coverage of targeted exonic regions, with 73 and 77% of loci covered at 20-fold from tumour and matched blood samples, respectively (Supplementary Fig.?S1). Mutation rates varied.