HnRNP A1 controls a splicing regulatory circuit promoting mesenchymal-to-epithelial transition

Epithelial-to-mesenchymal transition (EMT) is an embryonic program used by cancer cells to acquire invasive capabilities becoming metastatic. ΔRon, a constitutively active isoform of the Ron tyrosine kinase receptor, arises from skipping of Ron exon 11 and provided the first example of an alternative splicing variant causatively linked to the activation of tumor EMT. Splicing of exon 11 is controlled by two adjacent regulatory elements, a silencer and an enhancer of splicing located in exon 12. The alternative splicing factor and oncoprotein SRSF1 directly binds to the enhancer, induces the production of ΔRon and activates EMT leading to cell locomotion. Interestingly, we now find an important role for hnRNP A1 in controlling the activity of the Ron silencer. HnRNP A1 is able to antagonize the binding of SRSF1 and prevent exon skipping. Notably, hnRNP A1, by inhibiting the production of ΔRon, activates the reversal program, namely the mesenchymal-to-epithelial transition, which instead occurs at the final metastasis sites. Also, hnRNP A1 affects Ron splicing by regulating the expression level of hnRNP A2/B1, which similarly to SRSF1 can promote ΔRon production. These results shed light on how splicing regulation contributes to the tumor progression and provide potential targets to develop anticancer therapies.     

Autoregulation of Fox protein expression to produce dominant negative splicing factors

The Fox proteins are a family of regulators that control the alternative splicing of many exons in neurons, muscle, and other tissues. Each of the three mammalian paralogs, Fox-1 (A2BP1), Fox-2 (RBM9), and Fox-3 (HRNBP3), produces proteins with a single RNA-binding domain (RRM) flanked by N- and C-terminal domains that are highly diversified through the use of alternative promoters and alternative splicing patterns. These genes also express protein isoforms lacking the second half of the RRM (FoxΔRRM), due to the skipping of a highly conserved 93-nt exon. Fox binding elements overlap the splice sites of these exons in Fox-1 and Fox-2, and the Fox proteins themselves inhibit exon inclusion. Unlike other cases of splicing autoregulation by RNA-binding proteins, skipping the RRM exon creates an in-frame deletion in the mRNA to produce a stable protein. These FoxΔRRM isoforms expressed from cDNA exhibit highly reduced binding to RNA in vivo. However, we show that they can act as repressors of Fox-dependent splicing, presumably by competing with full-length Fox isoforms for interaction with other splicing factors. Interestingly, the Drosophila Fox homolog contains a nearly identical exon in its RRM domain that also has flanking Fox-binding sites. Thus, rather than autoregulation of splicing controlling the abundance of the regulator, the Fox proteins use a highly conserved mechanism of splicing autoregulation to control production of a dominant negative isoform.     

Splicing factor SRSF1 negatively regulates alternative splicing of MDM2 under damage

Genotoxic stress induces alternative splicing of the oncogene MDM2 generating MDM2-ALT1, an isoform attributed with tumorigenic properties. However, the mechanisms underlying this event remain unclear. Here we explore MDM2 splicing regulation by utilizing a novel minigene that mimics endogenous MDM2 splicing in response to UV and cisplatinum-induced DNA damage. We report that exon 11 is necessary and sufficient for the damage-specific alternative splicing of the MDM2 minigene and that the splicing factor SRSF1 binds exon 11 at evolutionarily conserved sites. Interestingly, mutations disrupting this interaction proved sufficient to abolish the stress-induced alternative splicing of the MDM2 minigene. Furthermore, SRSF1 overexpression promoted exclusion of exon 11, while its siRNA-mediated knockdown prevented the stress-induced alternative splicing of endogenous MDM2. Additionally, we observed elevated SRSF1 levels under stress and in tumors correlating with the expression of MDM2-ALT1. Notably, we demonstrate that MDM2-ALT1 splicing can be blocked by targeting SRSF1 sites on exon 11 using antisense oligonucleotides. These results present conclusive evidence supporting a negative role for SRSF1 in MDM2 alternative splicing. Importantly, we define for the first time, a clear-cut mechanism for the regulation of damage-induced MDM2 splicing and present potential strategies for manipulating MDM2 expression via splicing modulation.     

RNA binding protein RALY promotes Protein Arginine Methyltransferase 1 alternatively spliced isoform v2 relative expression and metastatic potential in breast cancer cells

Aberrant expression of Protein Arginine Methyltransferases (PRMTs) has been observed in several cancer types, including breast cancer. We previously reported that the PRMT1v2 isoform, which is generated through inclusion of alternative exon 2, is overexpressed in breast cancer cells and promotes their invasiveness. However, the precise mechanism by which expression of this isoform is controlled and how it is dysregulated in breast cancer remains unknown. Using a custom RNA interference-based screen, we identified several RNA binding proteins (RBP) which, when knocked down, altered the relative abundance of the alternatively spliced PRMT1v2 isoform. Amongst the top hits were SNW Domain containing 1 (SNW1) and RBP-associated with lethal yellow mutation (RALY), which both associated with the PRMT1 pre-mRNA and upon depletion caused an increase or decrease in the relative abundance of PRMT1v2 isoform mRNA and protein. Most importantly, a significant decrease in invasion was observed upon RALY knockdown in aggressive breast cancer cells, consistent with targeting PRMT1v2 directly, and this effect was rescued by the exogenous re-expression of PRMT1v2. We show that SNW1 expression is decreased, while RALY expression is increased in breast cancer cells and tumours, which correlates with decreased patient survival. This work revealed crucial insight into the mechanisms regulating the expression of the PRMT1 alternatively spliced isoform v2 and its dysregulation in breast cancer. It also provides proof-of-concept support for the development of therapeutic strategies where regulators of PRMT1 exon 2 alternative splicing are targeted as an approach to selectively reduce PRMT1v2 levels and metastasis in breast cancer.     

Primary structure and developmental acidic to basic transition of 13 alternatively spliced mouse fast skeletal muscle troponin T isoforms

Large samples of original cDNAs encoding neonatal and adult mouse fast skeletal muscle troponin T (fTnT) have been isolated and characterized. The results demonstrate expression relationships of 8 alternatively spliced exons of the fTnT gene and reveal the primary structure of as many as 13 fTnT isoforms that diverge into acidic and basic classes due to differential mRNA splicing in the N-terminal variable region. In the C-terminal variable region encoded by the mutually exclusive exons 16 and 17, the splicing pathway and structure of exon 16 appears to be adult fTnT-specific, suggesting an adaptation to the functional demands of mature fast skeletal muscle. The cloned cDNAs were expressed in E. coli as standards to identify a high M_r to low M_r, acidic to basic fTnT isoform transition in postnatal developing skeletal muscles. Different from the developmental cardiac TnT switch generated by alternative splicing of a single exon, the fTnT isoform transition is an additive effect of alternative splicing of multiple N-terminal-coding exons, especially exons 4, 8 and fetal that are expressed at higher frequencies in the neonatal than in the adult muscle. The developmental fTnT isoform primary structure transition in both N- and C-terminal variable regions suggest a physiological importance of the apparently complex TnT isoform expression.     

A nonsense exon in the Tpm1 gene is silenced by hnRNP H and F

As well as generating protein isoform diversity, in some cases alternative splicing generates RNAs that harbor premature termination codons and that are subject to nonsense-mediated decay (NMD). We previously identified an apparent pseudo-exon in the rat α-tropomyosin (Tpm1) gene as a probable genuine alternatively spliced exon that causes NMD when spliced into Tpm1 RNA. Here, we report the analysis of cis-acting splicing regulatory elements within this “nonsense exon.” Guided by the data set of predicted splicing enhancer and silencer elements compiled by Zhang and Chasin, we made a series of mutations through the nonsense exon and found that like authentic exons it is densely packed with enhancer and silencer elements. Strikingly, 11 of 13 tested mutations behaved as predicted computationally. In particular, we found that a G-rich silencer at the 5′ end, which is crucial for skipping of the nonsense exon, functions by binding hnRNP-H and F.     

Cloning and identification of two novel splice variants of human PD-L2

T cell activation is dependent upon signals deliveredthrough the antigen-specific T cell receptors and costimula-tory molecules [1,2]. The B7 family of costimulatory mol-ecules provides signals that are critical for both stimula-ting and inhibiting T cell…     

The neurofibromatosis type I pre-mRNA is a novel target of CELF protein-mediated splicing regulation

The CUG-BP and ETR-3 like factors (CELF) are a family of six highly conserved RNA-binding proteins that preferentially bind to UG-rich sequences. One of the key functions of these proteins is to mediate alternative splicing in a number of tissues, including brain, heart and muscle. To fully understand the function of CELF proteins, it is important to identify downstream targets of CELF proteins. In this communication, we report that neurofibromatosis type I (NF1) exon 23a is a novel target of CELF protein-mediated splicing regulation in neuron-like cells. NF1 regulates Ras signaling, and the isoform that excludes exon 23a shows 10 times greater ability to down-regulate Ras signaling than the isoform that includes exon 23a. Five of the six CELF proteins strongly suppress the inclusion of NF1 exon 23a. Over-expression or siRNA knockdown of these proteins in cell transfection experiments altered the levels of NF1 exon 23a inclusion. In vitro binding and splicing analyses demonstrate that CELF proteins block splicing through interfering with binding of U2AF⁶⁵. These studies, combined with our previous investigations demonstrating a role for Hu proteins and TIA-1/TIAR in controlling NF1 exon 23a inclusion, highlight the complex nature of regulation of this important alternative splicing event.     

Transforming growth factor-beta1 regulates fibronectin isoform expression and splicing factor SRp40 expression during ATDC5 chondrogenic maturation

Fibronectin (FN) isoform expression is altered during chondrocyte commitment and maturation, with cartilage favoring expression of FN isoforms that includes the type II repeat extra domain B (EDB) but excludes extra domain A (EDA). We and others have hypothesized that the regulated splicing of FN mRNAs is necessary for the progression of chondrogenesis. To test this, we treated the pre-chondrogenic cell line ATDC5 with transforming growth factor-beta1, which has been shown to modulate expression of the EDA and EDB exons, as well as the late markers of chondrocyte maturation; it also slightly accelerates the early acquisition of a sulfated proteoglycan matrix without affecting cell proliferation. When chondrocytes are treated with TGF-beta1, the EDA exon is preferentially excluded at all times whereas the EDB exon is relatively depleted at early times. This regulated alternative splicing of FN correlates with the regulation of alternative splicing of SRp40, a splicing factor facilitating inclusion of the EDA exon. To determine if overexpression of the SRp40 isoforms altered FN and FN EDA organization, cDNAs encoding these isoforms were overexpressed in ATDC5 cells. Overexpression of the long-form of SRp40 yielded an FN organization similar to TGF-beta1 treatment; whereas overexpression of the short form of SRp40 (which facilitates EDA inclusion) increased formation of long-thick FN fibrils. Therefore, we conclude that the effects of TGF-beta1 on FN splicing during chondrogenesis may be largely dependent on its effect on SRp40 isoform expression.