The emerging role of splicing factors in cancer

Abstract

Recent progress in global sequence and microarray data analysis has revealed the increasing complexity of the human transcriptome. Alternative splicing generates a huge diversity of transcript variants and disruption of splicing regulatory networks is emerging as an important contributor to various diseases, including cancer. Current efforts to establish the dynamic repertoire of transcripts that are generated in health and disease are showing that many cancer-associated alternative-splicing events occur in the absence of mutations in the affected genes. A growing body of evidence reveals changes in splicing-factor expression that correlate with cancer development, progression and response to therapy. Here, we discuss how recent links between cancer and altered expression of proteins implicated in splicing regulation are bringing the splicing machinery to the fore as a potential target for anticancer treatment.

Figure 1

Alternative splicing of FGFR2. (A) FGF signalling is mediated by four FGFR tyrosine kinases known as FGFR1–4, which have crucial roles in morphogenesis, development, angiogenesis and wound healing. FGFRs are composed of an extracellular ligand-binding portion consisting of three immunoglobulin-like domains (D1, D2 and D3), a single transmembrane (TM) helix and a cytoplasmic portion that contains protein tyrosine kinase activity. Ligand binding and specificity reside in D2, D3 and the linker that connects them. Alternative splicing in the carboxy-terminal half of D3 is an important determinant of FGF–FGFR binding specificity. (B) The transcripts (pre-mRNA) encoding FGFR2 contain two alternative exons: IIIb and IIIc. Inclusion of exon IIIb instead of exon IIIc introduces an amino-acid sequence in the second half of D3 that is less likely to form the hydrophobic core required for efficient interaction with FGF2 (Plotnikov et al, 2000). Inclusion of exon IIIb occurs predominantly in epithelial cells, whereas inclusion of exon IIIc is exclusively detected in cells of mesenchymal origin. A switch from exon IIIb to exon IIIc inclusion accompanies the progression of androgen-sensitive, well-differentiated prostate carcinomas to androgen-insensitive, poorly differentiated tumours (Carstens et al, 1997; Yan et al, 1993). (C) In mesenchymal cells, exon IIIb silencing depends on the combined effect of weak splice sites, an exonic splicing silencer that binds hnRNP A1, and two flanking intronic splicing silencers that bind to PTB. The binding of hnRNP A1 and PTB to the splicing silencers inhibits the recruitment of U1 and U2 snRNPs to the weak splice sites of exon IIIb. In epithelial cells, exon IIIb silencing is countered by binding of TIA-1 to an intronic activating sequence located downstream of exon IIIb. TIA-1 binding promotes the recruitment of U1 snRNP to the weak 5′ splice site of exon IIIb. In addition, FOX-2 binds to intronic and exonic (U)GCAUG sequence elements, and contributes to both exon IIIb activation and exon IIIc repression. FOX-2 proteins are differentially expressed in IIIb⁺ cells in comparison to IIIc⁺ cells, and overexpression of FOX-2 is sufficient to induce a switch in splice choice from IIIc to IIIb (Baraniak et al, 2006 and references therein). FGF fibroblast growth factor; FGFR2, fibroblast growth factor receptor 2; FOX-2 forkhead box-2; hnRNP heterogeneous nuclear ribonucleoparticle; HSPG, heparan sulphate proteoglycan; PTB, polypyrimidine-tract binding protein; snRNP, small-nuclear ribonucleoprotein particle; TIA-1, cytotoxic granule-associated RNA binding protein.

From left: Maria Carmo-Fonseca, Sandra Martins & Ana Rita Grosso