Chemical Inhibition of Pre-mRNA Splicing in Living Saccharomyces cerevisiae.

The spliceosome mediates precursor mRNA splicing in eukaryotes, including the model organism Saccharomyces cerevisiae (yeast). Despite decades of study, no chemical inhibitors of yeast splicing in vivo are available. We have developed a system to efficiently inhibit splicing and block proliferation in living yeast cells using compounds that target the human spliceosome protein SF3B1. Potent inhibition is observed in yeast expressing a chimeric protein containing portions of human SF3B1. However, only a single point mutation in the yeast homolog of SF3B1 is needed for selective inhibition of splicing by pladienolide B, herboxidiene, or meayamycin in liquid culture. Mutations that enable inhibition also improve splicing of branch sites containing mismatches between the intron and small nuclear RNA-suggesting a link between inhibitor sensitivity and usage of weak branch sites in humans. This approach provides powerful new tools for manipulating splicing in live yeast and studies of spliceosome inhibitors.

Antisense targeting of 3' end elements involved in DUX4 mRNA processing is an efficient therapeutic strategy for facioscapulohumeral dystrophy: a new gene-silencing approach.

Defects in mRNA 3'end formation have been described to alter transcription termination, transport of the mRNA from the nucleus to the cytoplasm, stability of the mRNA and translation efficiency. Therefore, inhibition of polyadenylation may lead to gene silencing. Here, we choose facioscapulohumeral dystrophy (FSHD) as a model to determine whether or not targeting key 3' end elements involved in mRNA processing using antisense oligonucleotide drugs can be used as a strategy for gene silencing within a potentially therapeutic context. FSHD is a gain-of-function disease characterized by the aberrant expression of the Double homeobox 4 (DUX4) transcription factor leading to altered pathogenic deregulation of multiple genes in muscles. Here, we demonstrate that targeting either the mRNA polyadenylation signal and/or cleavage site is an efficient strategy to down-regulate DUX4 expression and to decrease the abnormally high-pathological expression of genes downstream of DUX4. We conclude that targeting key functional 3' end elements involved in pre-mRNA to mRNA maturation with antisense drugs can lead to efficient gene silencing and is thus a potentially effective therapeutic strategy for at least FSHD. Moreover, polyadenylation is a crucial step in the maturation of almost all eukaryotic mRNAs, and thus all mRNAs are virtually eligible for this antisense-mediated knockdown strategy.

Pro-apoptotic effects of splice-switching oligonucleotides targeting Bcl-x pre-mRNA in human glioma cell lines.

Alternative splicing is a near-ubiquitous phenomenon with important roles in human diseases, including cancers. Splice-switching oligonucleotides (SSOs) have emerged as a class of antisense therapeutics that modulate alternative splicing by hybridizing to the pre-mRNA splice site. The Bcl-x gene is alternatively spliced to express anti­apoptotic Bcl-xL and pro-apoptotic Bcl-xS. Bcl-xL expression is upregulated in many cancers and is considered a general mechanism by which cancer cells evade apoptosis. By redirecting Bcl-x pre-mRNA splicing from Bcl-xL to Bcl-xS, SSO exerted pro-apoptotic and chemosensitizing effects in various cancer cell lines. In this study, we investigated the effects of SSO targeting Bcl-x pre-mRNA in human glioma cell lines. First, we performed reverse transcription-polymerase chain reaction (RT-PCR) and western blotting to determine the mRNA and protein expression levels of Bcl-xL in glioma cell lines (U87 and U251) and a normal human astrocyte cell line (HA1800). Then, the Bcl-x SSO was designed to bind to the downstream 5' alternative splice site of exon 2 in Bcl-x pre-mRNA and was modified using 2'-O-methoxyethyl-phosphorothioate. An oligonucleotide targeting aberrantly spliced human ß-globin intron was used as a negative control. The SSOs were delivered with a cationic lipid into glioma and astrocyte cell lines. The antitumor effects of the SSOs were assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assays and flow cytometry, and the switch in production from Bcl-xL to Bcl-xS was analyzed by RT-PCR and western blotting. Bcl-xL mRNA and protein were highly expressed in both glioma cell lines. The Bcl-x SSO modified Bcl-x pre-mRNA splicing and had pro-apoptotic effects on the glioma cell lines. By contrast, the lipid alone and the control SSO did not affect Bcl-xL expression or induce apoptosis. Our study demonstrated the antitumor activity of an SSO that targets Bcl-x pre-mRNA splicing in glioma cell lines. Bcl-x SSO may be a potential strategy for treating gliomas.

Hepatotoxicity of high affinity gapmer antisense oligonucleotides is mediated by RNase H1 dependent promiscuous reduction of very long pre-mRNA transcripts.

High affinity antisense oligonucleotides (ASOs) containing bicylic modifications (BNA) such as locked nucleic acid (LNA) designed to induce target RNA cleavage have been shown to have enhanced potency along with a higher propensity to cause hepatotoxicity. In order to understand the mechanism of this hepatotoxicity, transcriptional profiles were collected from the livers of mice treated with a panel of highly efficacious hepatotoxic or non-hepatotoxic LNA ASOs. We observed highly selective transcript knockdown in mice treated with non-hepatotoxic LNA ASOs, while the levels of many unintended transcripts were reduced in mice treated with hepatotoxic LNA ASOs. This transcriptional signature was concurrent with on-target RNA reduction and preceded transaminitis. Remarkably, the mRNA transcripts commonly reduced by toxic LNA ASOs were generally not strongly associated with any particular biological process, cellular component or functional group. However, they tended to have much longer pre-mRNA transcripts. We also demonstrate that the off-target RNA knockdown and hepatotoxicity is attenuated by RNase H1 knockdown, and that this effect can be generalized to high affinity modifications beyond LNA. This suggests that for a certain set of ASOs containing high affinity modifications such as LNA, hepatotoxicity can occur as a result of unintended off-target RNase H1 dependent RNA degradation.

Cytoplasmic viral RNA-dependent RNA polymerase disrupts the intracellular splicing machinery by entering the nucleus and interfering with Prp8.

The primary role of cytoplasmic viral RNA-dependent RNA polymerase (RdRp) is viral genome replication in the cellular cytoplasm. However, picornaviral RdRp denoted 3D polymerase (3D(pol)) also enters the host nucleus, where its function remains unclear. In this study, we describe a novel mechanism of viral attack in which 3D(pol) enters the nucleus through the nuclear localization signal (NLS) and targets the pre-mRNA processing factor 8 (Prp8) to block pre-mRNA splicing and mRNA synthesis. The fingers domain of 3D(pol) associates with the C-terminal region of Prp8, which contains the Jab1/MPN domain, and interferes in the second catalytic step, resulting in the accumulation of the lariat form of the splicing intermediate. Endogenous pre-mRNAs trapped by the Prp8-3D(pol) complex in enterovirus-infected cells were identified and classed into groups associated with cell growth, proliferation, and differentiation. Our results suggest that picornaviral RdRp disrupts pre-mRNA splicing processes, that differs from viral protease shutting off cellular transcription and translation which contributes to the pathogenesis of viral infection.

Inhibition of U4 snRNA in human cells causes the stable retention of polyadenylated pre-mRNA in the nucleus.

Most human pre-mRNAs contain introns that are removed by splicing. Such a complex process needs strict control and regulation in order to prevent the expression of aberrant or unprocessed transcripts. To analyse the fate of pre-mRNAs that cannot be spliced, we inhibited splicing using an anti-sense morpholino (AMO) against U4 snRNA. As a consequence, splicing of several selected transcripts was strongly inhibited. This was accompanied by the formation of enlarged nuclear speckles containing polyadenylated RNA, splicing factors and the nuclear poly(A) binding protein. Consistently, more polyadenylated pre-mRNA could be isolated from nucleoplasmic as well as chromatin-associated RNA fractions following U4 inhibition. Further analysis demonstrated that accumulated pre-mRNAs were stable in the nucleus and that nuclear RNA degradation factors did not re-localise to nuclear speckles following splicing inhibition. The accumulation of pre-mRNA and the formation of enlarged speckles were sensitive to depletion of the 3' end processing factor, CPSF73, suggesting a requirement for poly(A) site processing in this mechanism. Finally, we provide evidence that the pre-mRNAs produced following U4 snRNA inhibition remain competent for splicing, perhaps providing a biological explanation for their stability. These data further characterise processes ensuring the nuclear retention of pre-mRNA that cannot be spliced and suggest that, in some cases, unspliced transcripts can complete splicing sometime after their initial synthesis.

RAM function is dependent on Kapß2-mediated nuclear entry.

Eukaryotic gene expression is dependent on the modification of the first transcribed nucleotide of pre-mRNA by the addition of the 7-methylguanosine cap. The cap protects transcripts from exonucleases and recruits complexes which mediate transcription elongation, processing and translation initiation. The cap is synthesized by a series of reactions which link 7-methylguanosine to the first transcribed nucleotide via a 5' to 5' triphosphate bridge. In mammals, cap synthesis is catalysed by the sequential action of RNGTT (RNA guanylyltransferase and 5'-phosphatase) and RNMT (RNA guanine-7 methyltransferase), enzymes recruited to RNA pol II (polymerase II) during the early stages of transcription. We recently discovered that the mammalian cap methyltransferase is a heterodimer consisting of RNMT and the RNMT-activating subunit RAM (RNMT-activating mini-protein). RAM activates and stabilizes RNMT and thus is critical for cellular cap methylation and cell viability. In the present study we report that RNMT interacts with the N-terminal 45 amino acids of RAM, a domain necessary and sufficient for maximal RNMT activation. In contrast, smaller components of this RAM domain are sufficient to stabilize RNMT. RAM functions in the nucleus and we report that nuclear import of RAM is dependent on PY nuclear localization signals and Kapß2 (karyopherin ß2) nuclear transport protein.

A high-throughput splicing assay identifies new classes of inhibitors of human and yeast spliceosomes.

The spliceosome is the macromolecular machine responsible for pre-mRNA splicing, an essential step in eukaryotic gene expression. During splicing, myriad subunits join and leave the spliceosome as it works on the pre-mRNA substrate. Strikingly, there are very few small molecules known to interact with the spliceosome. Splicing inhibitors are needed to capture transient spliceosome conformations and probe important functional components. Such compounds may also have chemotherapeutic applications, as links between splicing and cancer are increasingly uncovered. To identify new splicing inhibitors, we developed a high-throughput assay for in vitro splicing using a reverse transcription followed by quantitative PCR readout. In a pilot screen of 3080 compounds, we identified three small molecules that inhibit splicing in HeLa extract by interfering with different stages of human spliceosome assembly. Two of the compounds similarly affect spliceosomes in yeast extracts, suggesting selective targeting of conserved components. By examining related molecules, we identified chemical features required for the activity of two of the splicing inhibitors. In addition to verifying our assay procedure and paving the way to larger screens, these studies establish new compounds as chemical probes for investigating the splicing machinery.

Clinical trials using antisense oligonucleotides in duchenne muscular dystrophy.

Duchenne muscular dystrophy (DMD) is a severe muscle wasting disorder caused by mutations in the DMD gene, affecting 1 in 3500 newborn males. Complete loss of muscle dystrophin protein causes progressive muscle weakness and heart and respiratory failure, leading to premature death. Antisense oligonucleotides (AONs) that bind to complementary sequences of the dystrophin pre-mRNA to induce skipping of the targeted exon by modulating pre-mRNA splicing are promising therapeutic agents for DMD. Such AONs can restore the open reading frame of the DMD gene and produce internally deleted, yet partially functional dystrophin protein isoforms in skeletal muscle. Within the last few years, clinical trials using AONs have made considerable progress demonstrating the restoration of functional dystrophin protein and acceptable safety profiles following both local and systemic delivery in DMD patients. However, improvement of AON delivery and efficacy, along with the development of multiple AONs to treat as many DMD patients as possible needs to be addressed for this approach to fulfill its potential. Here, we review the recent progress made in clinical trials using AONs to treat DMD and discuss the current challenges to the development of AON-based therapy for DMD.

Establishment of a monitoring system to detect inhibition of mRNA processing.

A number of proteins complete mRNA processing in the nucleus, thus, inhibitor of mRNA processing is worth finding to analyze the mechanism of mRNA maturation in detail. Here, we established a monitoring system for mRNA processing using a test compound, spliceostatin A (SSA), which inhibits mRNA splicing. This system should serve to facilitate the discovery of novel compounds from natural resources that inhibit mRNA processing.

A LIN28-dependent structural change in pre-let-7g directly inhibits dicer processing.

Several recent studies have provided evidence that LIN28, a cytoplasmic RNA-binding protein, inhibits the biogenesis of members of the let-7 microRNA family at the Dicer step in both mammals and Caenorhabditis elegans. However, the precise mechanism of inhibition is still poorly understood. Here we report on an in vitro study, which combined RNase footprinting, gel shift binding assays, and processing assays, to investigate the molecular basis and function of the interaction between the native let-7g precursor (pre-let-7g) and LIN28. We have mapped the structure of pre-let-7g and identified some regions of the terminal loop of pre-let-7g that physically interact with LIN28. We have also identified a conformational change upon LIN28 binding that results in the unwinding of an otherwise double-stranded region at the Dicer processing site of pre-let-7g. Furthermore, we showed that a mutant pre-let-7g that displays an open upper stem inhibited pre-let-7g Dicer processing to the same extent as LIN28. The data support a mechanism by which LIN28 can directly inhibit let-7g biogenesis at the Dicer processing step.

Five-step process for screening antisense compounds for efficacy: gene target IL-12Rb2.

Antisense technologies are widely used for the inhibition of gene expression. Although traditionally the AUG start codon of the open reading frame is targeted to disrupt ribosome assembly and initiation, an emerging approach is targeting sequences to disrupt pre-mRNA splicing. The primary advantage to using this approach is a positive read-out for an antisense effect through detection of a novel splice product, but additional benefit can be found in generating a novel splice product with altered functional properties. The antisense compounds used here are phosphorodiamidate morpholino oligomers conjugated to an arginine-rich cell penetrating peptide (P-PMO). We describe a five-step process for selecting the best candidate antisense compound for altering IL-12Rb2 expression including (1) detecting mRNA splice products by RT-PCR, (2) measuring protein expression, (3) evaluating protein function, (4) checking cellular viability, and (5) validating efficacy of the final candidate compound. The significance of targeting exons composed of a number of base pairs divisible by 3 is also discussed. The five steps described here for selecting the best candidate P-PMO to alter IL-12Rb2 expression should be applied for designing and screening antisense compounds for other gene targets.

Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43.

We used cross-linking and immunoprecipitation coupled with high-throughput sequencing to identify binding sites in 6,304 genes as the brain RNA targets for TDP-43, an RNA binding protein that, when mutated, causes amyotrophic lateral sclerosis. Massively parallel sequencing and splicing-sensitive junction arrays revealed that levels of 601 mRNAs were changed (including Fus (Tls), progranulin and other transcripts encoding neurodegenerative disease-associated proteins) and 965 altered splicing events were detected (including in sortilin, the receptor for progranulin) following depletion of TDP-43 from mouse adult brain with antisense oligonucleotides. RNAs whose levels were most depleted by reduction in TDP-43 were derived from genes with very long introns and that encode proteins involved in synaptic activity. Lastly, we found that TDP-43 autoregulates its synthesis, in part by directly binding and enhancing splicing of an intron in the 3' untranslated region of its own transcript, thereby triggering nonsense-mediated RNA degradation.

Identification of SAP155 as the target of GEX1A (Herboxidiene), an antitumor natural product.

GEX1A is a microbial product with antitumor activity. HeLa cells cultured with GEX1A accumulated p27(Kip) and its C-terminally truncated form p27*. GEX1A inhibited the pre-mRNA splicing of p27, producing p27* from the unspliced mRNA containing the first intron. p27* lacked the site required for E3 ligase-mediated proteolysis of p27, leading to its accumulation in GEX1A-treated cells. The accumulated p27* was able to bind to and inhibit the cyclin E-Cdk2 complex that causes E3 ligase-mediated degradation of p27, which probably triggers the accumulation of p27. By using a series of photoaffinity-labeling derivatives of GEX1A, we found that GEX1A targeted SAP155 protein, a subunit of SF3b responsible for pre-mRNA splicing. The linker length between the GEX1A pharmacophore and the photoreactive group was critical for detection of the GEX1A-binding protein. GEX1A serves as a novel splicing inhibitor that specifically impairs the SF3b function by binding to SAP155.

Inhibition of dicing of guanosine-rich shRNAs by quadruplex-binding compounds.

RNA interference is triggered by small hairpin precursors that are processed by the endonuclease dicer to yield active species such as siRNAs and miRNAs. To regulate the RNAi-mediated suppression of gene expression, we imagined a strategy that relies on the sequence-specific inhibition of shRNA precursor processing by immediate RNA-small molecule interactions. Here, we present a first step in this direction by augmenting shRNAs with guanosine-rich sequences that are prone to fold into four-stranded structures. The addition of small molecules that selectively bind to such quadruplex sequences should allow for the specific inhibition of dicing of shRNAs that contain suitable G-rich elements. In an attempt to find compounds that protect against dicer processing, we have examined the effects of quadruplex-binding compounds on the dicer processing of shRNAs containing G-quadruplexes. Although a variety of small molecules that are known to bind to quadruplexes inhibited in vitro dicing of shRNAs, only two substance classes, namely certain porphyrazines and bisquinolinium compounds, showed selective inhibition of G-rich shRNAs compared to control sequences lacking guanine-rich elements. The G-rich shRNAs displayed a potent knockdown of gene expression in mammalian cell culture, but the effect was not influenced by addition of the respective quadruplex-binding compounds.

Mechanism of action of a new antitumor ribonucleoside, 1-(3-C-ethynyl-beta-D-ribo-pentofuranosyl)cytosine (ECyd, TAS-106), differs from that of 5-fluorouracil.

1-(3-C-ethynyl-beta-D-ribo-pentofuranosyl)cytosine (ECyd, TAS-106), is a new antitumor cytidine analogue, inhibiting RNA synthesis. In this study we investigated the cellular growth inhibition, intracellular metabolism, cell cycle phase specificity, and RNA synthesis of TAS-106 compared with those of 5-fluorouracil (5-FU), known to possess both DNA- (inhibition of thymidylate synthase activity) and RNA-synthesis-inhibiting activity (inhibition of RNA function). The IC50 values of TAS-106 and 5-FU ranged from 0.0173 to 3.11 microM, and from 6.80 to >1,000 microM, respectively, in a panel of 10 human tumor cells, indicating that TAS-106 possesses greater cytotoxicity than 5-FU. Using excess thymidine-synchronized cells, TAS-106 and 5-FU appeared to exert their cytotoxic effects independently of the cell cycle. The intracellular metabolism and the effect on pre-rRNA processing of TAS-106 differed from those of 5-FU. More than 50% of 5-FU incorporated into the cells was in the unchanged form, while 5-FU incorporated into RNA was approximately 20%. On the other hand, TAS-106 was incorporated in a time-dependent manner into the cells and rapidly converted to its mono-, di- and tri-phosphate form, however, the amount incorporated into RNA fraction was very small. 5-FU incorporated into RNA was confirmed to impair the normal processing of ribosomal RNA (formation of 34/32S RNA from 45S RNA), however, TAS-106 did not affect pre-rRNA processing and may be involved in the inhibition of the synthesis of ribosomal RNA. We concluded that intracellular accumulation and retention of the active metabolite of TAS-106, 3'-ethynylcytidine 5'-triphosphate (ECTP), may contribute to its potent cytotoxicity. The unique mechanism of antitumor activity and intensive cellular metabolism of TAS-106 could contribute to cancer chemotherapy through the pathways different from those of 5-FU or other antitumor nucleosides.

Novel targeting of cyclooxygenase-2 (COX-2) pre-mRNA using antisense morpholino oligonucleotides directed to the 3' acceptor and 5' donor splice sites of exon 4: suppression of COX-2 activity in human amnion-derived WISH and myometrial cells.

Increased expression of cyclooxygenase-2 (COX-2) has been implicated in the onset of both term and preterm labor. In this context, both selective and nonselective COX-2 inhibitors have been used in clinical trials to determine their efficacy in delaying preterm labor. However, recent evidence indicates that these tocolytics may have potentially adverse fetal and maternal side effects. Therefore, the development of more specific and nontoxic agents to inhibit COX-2 needs to be considered. We have evaluated whether antisense morpholino oligonucleotides have therapeutic potential in inhibiting COX-2 by specifically targeting both the 3' and 5' acceptor and donor sites of exon 4 of COX-2's pre-mRNA sequence. Confocal microscopy on "live" cells illustrated high levels of penetrance of antisense morpholino oligonucleotides using the Endo-Porter formula (Gene-Tools, LLC, Philomath, OR), with delivery efficiencies of 82 and 78%, respectively, in amnion-derived WISH and myometrial cells. Substantial inhibition by the morpholino oligonucleotides of COX-2 expression, induced by lipopolysaccharide administration, was observed at both the mRNA and protein levels. Loss of enzymic activity of COX-2 was confirmed using a sensitive COX enzyme activity assay, which reflects the rate of conversion of arachidonic acid to prostaglandin H2. Our results indicate that antisense morpholino oligonucleotides significantly inhibit expression and activity of this enzyme in in vitro cultures of amnion-WISH and myometrial cells. The potential thus exists that a similar approach can be mimicked in vivo to produce a highly specific and nontoxic strategy to inhibit COX-2 activity with its subsequent effects on the better management of preterm labor and other inflammatory conditions.

Global and local dynamics of the U1A polyadenylation inhibition element (PIE) RNA and PIE RNA-U1A complexes.

The structure and dynamics of the polyadenylation inhibition element (PIE) RNA, free and bound to the U1A protein, have been examined using time-resolved FRET and 2-aminopurine (2AP) fluorescence. This regulatory RNA, located at the 3' end of the U1A pre-mRNA, adopts a U-shaped structure, with binding sites for a single U1A protein at each bend (box 1 and box 2). The distance between the termini of the arms of the RNA is sensitive to its three-dimensional structure. Using Cy3/Cy5 FRET efficiency to monitor binding of Mg(2+), we show that the PIE RNA binds two Mg(2+) ions, which results in a restriction of its distance distribution of conformations. Local RNA structure probing using 2AP fluorescence shows that the structure of box 2 changes in response to Mg(2+) binding, thus tentatively locating the ion binding sites. Steady-state FRET data show that the distance R between the termini of the PIE RNA stems decreases from 66 A in the free RNA, to 58 A when N-terminal RNA binding domains (RBD1) of U1A are bound, and to 53 A when U1A proteins bind. However, anisotropy measurements indicate that both Cy3 and Cy5 stack on the ends of the RNA. To examine the consequences of the restricted motion of the fluorophores, FRET data are analyzed using two different models of motion and then compared to analogous data from the Cy3/fluorescein FRET pair. We conclude that the error introduced into distance calculations by stacking of the dyes is within the error of our measurements. Distance distributions of the RNA structures show that the intramolecular distance between the arms of the PIE RNA varies on the time scale of the fluorescence measurements; the mean distance is dependent on protein binding, but the breadth of the distributions indicates that the RNA retains structural heterogeneity.