Aberrant expression of CPSF1 promotes head and neck squamous cell carcinoma via regulating alternative splicing.
Akihiro SakaiMizuo AndoTakahito FukusumiShuling RenChao LiuJesse QualliotineSunny HaftSayed SadatYuki SaitoTheresa W GuoGuorong XuRoman SasikKathleen M FischJ Silvio GutkindElana J FertigAlfredo A MolinoloJoseph A CalifanoPublished in: PloS one (2020)
Alternative mRNA splicing increases protein diversity, and alternative splicing events (ASEs) drive oncogenesis in multiple tumor types. However, the driving alterations that underlie the broad dysregulation of ASEs are incompletely defined. Using head and neck squamous cell carcinoma (HNSCC) as a model, we hypothesized that the genomic alteration of genes associated with the spliceosome may broadly induce ASEs across a broad range of target genes, driving an oncogenic phenotype. We identified 319 spliceosome genes and employed a discovery pipeline to identify 13 candidate spliceosome genes altered in HNSCC using The Cancer Genome Atlas (TCGA) HNSCC data. Phenotypic screens identified amplified and overexpressed CPSF1 as a target gene alteration that was validated in proliferation, colony formation, and apoptosis assays in cell line and xenograft systems as well as in primary HNSCC. We employed knockdown and overexpression assays followed by identification of ASEs regulated by CPSF1 overexpression to identify changes in ASEs, and the expression of these ASEs was validated using RNA from cell line models. Alterations in expression of spliceosome genes, including CPSF1, may contribute to HNSCC by mediating aberrant ASE expression.
Keyphrases
- genome wide
- poor prognosis
- bioinformatics analysis
- genome wide identification
- binding protein
- high throughput
- transcription factor
- dna methylation
- copy number
- long non coding rna
- oxidative stress
- squamous cell carcinoma
- machine learning
- gene expression
- genome wide analysis
- papillary thyroid
- artificial intelligence
- deep learning
- amino acid
- squamous cell
- nucleic acid