Summary Vascular endothelial growth factor A (VEGFA; hereafter referred to as VEGF) is a key regulator of physiological and pathological angiogenesis. Two families of VEGF isoforms are generated by alternate splice-site selection in the terminal exon. Proximal splice-site selection (PSS) in exon 8 results in pro-angiogenic VEGFxxx isoforms (xxx is the number of amino acids), whereas distal splice-site selection (DSS) results in anti-angiogenic VEGFxxxb isoforms. To investigate control of PSS and DSS, we investigated the regulation of isoform expression by extracellular growth factor administration and intracellular splicing factors. In primary epithelial cells VEGFxxxb formed the majority of VEGF isoforms (74%). IGF1, and TNFα treatment favoured PSS (increasing VEGFxxx) whereas TGFβ1 favoured DSS, increasing VEGFxxxb levels. TGFβ1 induced DSS selection was prevented by inhibition of p38 MAPK and the Clk/sty (CDC-like kinase, CLK1) splicing factor kinase family, but not ERK1/2. Clk phosphorylates SR protein splicing factors ASF/SF2, SRp40 and SRp55. To determine whether SR splicing factors alter VEGF splicing, they were overexpressed in epithelial cells, and VEGF isoform production assessed. ASF/SF2, and SRp40 both favoured PSS, whereas SRp55 upregulated VEGFxxxb (DSS) isoforms relative to VEGFxxx. SRp55 knockdown reduced expression of VEGF165b. Moreover, SRp55 bound to a 35 nucleotide region of the 3′UTR immediately downstream of the stop codon in exon 8b. These results identify regulation of splicing by growth and splice factors as a key event in determining the relative pro- versus anti-angiogenic expression of VEGF isoforms, and suggest that p38 MAPK-Clk/sty kinases are responsible for the TGFβ1-induced DSS selection, and identify SRp55 as a key regulatory splice factor.

Summary Angiogenesis is regulated by the balance of pro-angiogenic VEGF165 and anti-angiogenic VEGF165b splice isoforms. Mutations in WT1, the Wilms’ tumour suppressor gene, suppress VEGF165b and cause abnormal gonadogenesis, renal failure and Wilms’ tumours. In WT1 mutant cells, reduced VEGF165b was due to lack of WT1 mediated transcriptional repression of the splicing factor kinase SRPK1. WT1 bound to the SRPK1 promoter, and repressed expression through a specific WT1 binding-site. In WT1 mutant cells SRPK1-mediated hyperphosphorylation of the oncogenic RNA binding protein SRSF1 regulated splicing of VEGF, and rendered WT1 mutant cells pro-angiogenic. Altered VEGF splicing was reversed by wildtype WT1, knockdown of SRSF1 or SRPK1 and inhibition of SRPK1, which prevented in vitro and in vivo angiogenesis and associated tumour growth.

Vascular endothelial growth factor (VEGF) is produced either as a pro-angiogenic or anti-angiogenic protein depending upon splice site choice in the terminal, eighth exon. Proximal splice site selection (PSS) in exon 8 generates pro-angiogenic isoforms such as VEGF165, and distal splice site selection (DSS) results in anti-angiogenic isoforms such as VEGF165b. Cellular decisions on splice site selection depend upon the activity of RNA-binding splice factors, such as ASF/SF2, which have previously been shown to regulate VEGF splice site choice. To determine the mechanism by which the pro-angiogenic splice site choice is mediated, we investigated the effect of inhibition of ASF/SF2 phosphorylation by SR protein kinases (SRPK1/2) on splice site choice in epithelial cells and in in vivo angiogenesis models. Epithelial cells treated with insulin-like growth factor-1 (IGF-1) increased PSS and produced more VEGF165 and less VEGF165b. This down-regulation of DSS and increased PSS was blocked by protein kinase C inhibition and SRPK1/2 inhibition. IGF-1 treatment resulted in nuclear localization of ASF/SF2, which was blocked by SPRK1/2 inhibition. Pull-down assay and RNA immunoprecipitation using VEGF mRNA sequences identified an 11-nucleotide sequence required for ASF/SF2 binding. Injection of an SRPK1/2 inhibitor reduced angiogenesis in a mouse model of retinal neovascularization, suggesting that regulation of alternative splicing could be a potential therapeutic strategy in angiogenic pathologies.

The identity of the RNA binding proteins (RBPs) that govern cancer stem cell remains poorly characterized. The MSI2 RBP is a central regulator of translation of cancer stem cell programs. Through proteomics analysis of the MSI2 interacting RBP network and functional shRNA screening, we identified 24 genes required for in vivo leukemia and SYNCRIP was the most differentially required gene between normal and myeloid leukemia cells. SYNCRIP depletion increased apoptosis and differentiation while delaying leukemogenesis. Gene expression profiling of SYNCRIP depleted cells demonstrated a loss of the MLL and HOXA9 leukemia stem cell gene associated program. SYNCRIP and MSI2 interact indirectly though shared mRNA targets. SYNCRIP maintains HOXA9 translation and MSI2 or HOXA9 overexpression rescued the effects of SYNCRIP depletion. We validated SYNCRIP as a novel RBP that controls the myeloid leukemia stem cell program and propose that targeting these functional complexes might provide a novel therapeutic strategy in leukemia.

The bone marrow microenvironment influences malignant hematopoiesis but how it promotes leukemogenesis has not been elucidated. Additionally, the role of the bone marrow stroma in regulating clinical responses to DNA methyltransferase inhibitors (DNMTi) is also poorly understood. In this study, we conducted a DNA methylome analysis of bone marrow-derived stromal cells from myelodysplastic syndrome (MDS) patients and observed widespread aberrant cytosine hypermethylation occurring preferentially outside CpG islands. Stroma derived from 5-azacytidine-treated patients lacked aberrant methylation and DNMTi treatment of primary MDS stroma enhanced its ability to support erythroid differentiation. An integrative expression analysis revealed that the WNT pathway antagonist FRZB was aberrantly hypermethylated and underexpressed in MDS stroma. This result was confirmed in an independent set of sorted, primary MDS-derived mesenchymal cells. We documented a WNT/β-catenin activation signature in CD34+ cells from advanced cases of MDS where it associated with adverse prognosis. Constitutive activation of β-catenin in hematopoietic cells yielded lethal myeloid disease in a NUP98-HOXD13 mouse model of MDS, confirming its role in disease progression. Our results define novel epigenetic changes in the bone marrow microenvironment which lead to β-catenin activation and disease progression of MDS.

The angiogenic capability of colorectal carcinomas (CRC), and their susceptibility to anti-angiogenic therapy, is determined by expression of vascular endothelial growth factor (VEGF) isoforms. The intracellular protein T-cell Intracellular Antigen (TIA-1) alters post-transcriptional RNA processing and binds VEGF-A mRNA. We therefore tested the hypothesis that TIA-1 could regulate VEGF-A isoform expression in colorectal cancers. TIA-1 and VEGF-A isoform expression was measured in colorectal cancers and cell lines. We discovered that an endogenous splice variant of TIA-1 encoding a truncated protein, short TIA-1 (sTIA-1) was expressed in CRC tissues and invasive K-Ras mutant colon cancer cells and tissues but not in adenoma cell lines. sTIA-1 was more highly expressed in CRC than in normal tissues and increased with tumour stage. Knockdown of sTIA-1 or over-expression of full length TIA-1 (flTIA-1) induced expression of the anti-angiogenic VEGF isoform VEGF-A165b. Whereas flTIA-1 selectively bound VEGF-A165 mRNA and increased translation of VEGF-A165b, sTIA-1 prevented this binding. In nude mice, xenografted colon cancer cells over-expressing flTIA-1 formed smaller, less vascular tumours than those expressing sTIA-1, but flTIA-1 expression inhibited the effect of anti-VEGF antibodies. These results indicate that alternative splicing of an RNA binding protein can regulate isoform specific expression of VEGF providing an added layer of complexity to the angiogenic profile of colorectal cancer and their resistance to anti-angiogenic therapy.

RNA-binding proteins are altered in large-scale, genome-wide cancer studies, although it is unclear how these proteins control tumor progression. Using unbiased gene expression and immunohistochemical analysis of 802 stage I lung adenocarcinoma (LUAD) patients, we show that increased adenosine deaminase acting on double-stranded RNA (ADAR) expression correlates with tumor recurrence. Knockdown of ADAR in LUAD cells reduces mesenchymal properties, cellular migration, and invasion. Analysis of gene expression patterns following the loss of ADAR identifies enrichment in cell migration pathways, most notably focal adhesion kinase (FAK). We show that ADAR posttranscriptionally increases the FAK oncogene through physical interaction with its RNA binding domain and editing a specific intronic site, resulting in stabilization and increase of FAK transcript. Moreover, pharmacological inhibition of FAK blocks ADAR-induced increase of cell invasion in LUAD cells, suggesting a potential therapeutic application for ADAR high-expressing LUAD. Collectively, we identify ADAR as an important regulator of LUAD progression through its ability to stabilize FAK.

Myelodysplastic syndromes (MDS) are driven by complex genetic and epigenetic alterations. The MSI2 RNA-binding protein has been demonstrated to have a role in acute myeloid leukaemia and stem cell function, but its role in MDS is unknown. Here, we demonstrate that elevated MSI2 expression correlates with poor survival in MDS. Conditional deletion of Msi2 in a mouse model of MDS results in a rapid loss of MDS haematopoietic stem and progenitor cells (HSPCs) and reverses the clinical features of MDS. Inversely, inducible overexpression of MSI2 drives myeloid disease progression. The MDS HSPCs remain dependent on MSI2 expression after disease initiation. Furthermore, MSI2 expression expands and maintains a more activated (G1) MDS HSPC. Gene expression profiling of HSPCs from the MSI2 MDS mice identifies a signature that correlates with poor survival in MDS patients. Overall, we identify a role for MSI2 in MDS representing a therapeutic target in this disease.