Therapeutic potential and mechanism of kinetin as a treatment for the human splicing disease familial dysautonomia

Mutations that affect the splicing of pre-mRNA are a major cause of human disease. Familial dysautonomia (FD) is a recessive neurodegenerative disease caused by a T to C transition at base pair 6 of IKBKAP intron 20. This mutation results in variable tissue-specific skipping of exon 20. Previously, we reported that the plant cytokinin kinetin dramatically increases exon 20 inclusion in RNA isolated from cultured FD cells. The goal of the current study was to investigate the nature of the FD splicing defect and the mechanism by which kinetin improves exon inclusion, as such knowledge will facilitate the development of future therapeutics aimed at regulating mRNA splicing. In this study, we demonstrate that treatment of FD lymphoblast cell lines with kinetin increases IKBKAP mRNA and IKAP protein to normal levels. Using a series of minigene constructs, we show that deletion of a region at the end of IKBKAP exon 20 disrupts the ability of kinetin to improve exon inclusion, pinpointing a kinetin responsive sequence element. We next performed a screen of endogenously expressed genes with multiple isoforms resulting from exon skipping events and show that kinetin's ability to improve exon inclusion is not limited to IKBKAP. Lastly, we highlight the potential of kinetin for the treatment of other human splicing disorders by showing correction of a splicing defect in neurofibromatosis.     

Identification of the first non-Jewish mutation in familial Dysautonomia

Familial Dysautonomia is an autosomal recessive disease with a remarkably high carrier frequency in the Ashkenazi Jewish population. It has recently been estimated that as many as 1 in 27 Ashkenazi Jews is a carrier of FD. The FD gene has been identified as IKBKAP, and two disease-causing mutations have been identified. The most common mutation, which is present on 99.5% of all FD chromosomes, is an intronic splice site mutation that results in tissue-specific skipping of exon 20. The second mutation, R696P, is a missense mutation that has been identified in 4 unrelated patients heterozygous for the major splice mutation. Interestingly, despite the fact that FD is a recessive disease, normal mRNA and protein are expressed in patient cells. To date, the diagnosis of FD has been limited to individuals of Ashkenazi Jewish descent and identification of the gene has led to widespread diagnostic and carrier testing in this population. In this report, we describe the first non-Jewish IKBKAP mutation, a proline to leucine missense mutation in exon 26, P914L. This mutation is of particular significance because it was identified in a patient who lacks one of the cardinal diagnostic criteria for the disease-pure Ashkenazi Jewish ancestry. In light of this fact, the diagnostic criteria for FD must be expanded. Furthermore, in order to ensure carrier identification in all ethnicities, this mutation must now be considered when screening for FD.     

Splicing of phenylalanine hydroxylase (PAH) exon 11 is vulnerable: molecular pathology of mutations in PAH exon 11

In about 20-30% of phenylketonuria (PKU) patients, phenylalanine (Phe) levels can be controlled by cofactor 6R-tetrahydrobiopterin (BH(4)) administration. The phenylalanine hydroxylase (PAH) genotype has a predictive value concerning BH(4)-response and therefore a correct assessment of the mutation molecular pathology is important. Mutations that disturb the splicing of exons (e.g. interplay between splice site strength and regulatory sequences like exon splicing enhancers (ESEs)/exon splicing silencers (ESSs)) may cause different severity of PKU. In this study, we identified PAH exon 11 as a vulnerable exon and used patient derived lymphoblast cell lines and PAH minigenes to study the molecular defect that impacted pre-mRNA processing. We showed that the c.1144T>C and c.1066-3C>T mutations cause exon 11 skipping, while the c.1139C>T mutation is neutral or slightly beneficial. The c.1144T>C mutation resides in a putative splicing enhancer motif and binding by splicing factors SF2/ASF, SRp20 and SRp40 is disturbed. Additional mutations in potential splicing factor binding sites contributed to elucidate the pathogenesis of mutations in PAH exon 11. We suggest that PAH exon 11 is vulnerable due to a weak 3' splice site and that this makes exon 11 inclusion dependent on an ESE spanning position c.1144. Importantly, this implies that other mutations in exon 11 may affect splicing, since splicing is often determined by a fine balance between several positive and negative splicing regulatory elements distributed throughout the exon. Finally, we identified a pseudoexon in intron 11, which would have pathogenic consequences if activated by mutations or improved splicing conditions. Exonic mutations that disrupt splicing are unlikely to facilitate response to BH(4) and may lead to inconsistent genotype-phenotype correlations. Therefore, recognizing such mutations enhances our ability to predict the BH(4)-response.     

Mutations causing defective splicing in the human hprt gene.

Ten intron mutations and one exon mutation giving rise to defective splicing in the human gene for hypoxanthine phosphoribosyl transferase (hprt) in T-lymphocytes have been characterized. The splicing mutants were detected by PCR amplification of hprt cDNA and direct sequencing. Nine of the mutants showed skipping of whole exons or parts of exons in the cDNA, one mutant had an inclusion of an intron sequence into the cDNA, and one mutant showed both inclusion of an intron sequence and skipping of exons as well as a normal cDNA. Genomic PCR and direct sequencing of the splice sites involved showed one deletion of three base pairs and 10 different single base alterations to be responsible for these splice alterations. One mutation in the last base pair of exon 6 causing skipping of the entire exon 6 was found, whereas an identical mutation in the last base pair of exon 2 caused no aberrant splicing. It was also found that a deletion mutation in the pyrimidine rich stretch of the acceptor site of intron 7 caused skipping of the entire exon 8, whereas a base substitution in the last base of intron 7 caused exclusion of only the first 21 base pairs of exon 8 as a result of the activation of a cryptic acceptor site in exon 8. The results show that many different types of mutations at several different sites can cause splicing errors in the hprt gene and that the sequence differences between the splice sites influence the possible spectrum of mutations in each site.     

Extensive in silico analysis of NF1 splicing defects uncovers determinants for splicing outcome upon 5' splice-site disruption

We describe 94 pathogenic NF1 gene alterations in a cohort of 97 Austrian neurofibromatosis type 1 patients meeting the NIH criteria. All mutations were fully characterized at the genomic and mRNA levels. Over half of the patients carried novel mutations, and only a quarter carried recurrent minor-lesion mutations at 16 mutational warm spots. The remaining patients carried NF1 microdeletions (7%) and rare recurring mutations. Thirty-six of the mutations (38%) altered pre-mRNA splicing, and fall into five groups: exon skipping resulting from mutations at authentic splice sites (type I), cryptic exon inclusion caused by deep intronic mutations (type II), creation of de novo splice sites causing loss of exonic sequences (type III), activation of cryptic splice sites upon authentic splice-site disruption (type IV), and exonic sequence alterations causing exon skipping (type V). Extensive in silico analyses of 37 NF1 exons and surrounding intronic sequences suggested that the availability of a cryptic splice site combined with a strong natural upstream 3' splice site (3'ss)is the main determinant of cryptic splice-site activation upon 5' splice-site disruption. Furthermore, the exonic sequences downstream of exonic cryptic 5' splice sites (5'ss) resemble intronic more than exonic sequences with respect to exonic splicing enhancer and silencer density, helping to distinguish between exonic cryptic and pseudo 5'ss. This study provides valuable predictors for the splicing pathway used upon 5'ss mutation, and underscores the importance of using RNA-based techniques, together with methods to identify microdeletions and intragenic copy-number changes, for effective and reliable NF1 mutation detection.     

A deep intronic mutation in FGB creates a consensus exonic splicing enhancer motif that results in afibrinogenemia caused by aberrant mRNA splicing, which can be corrected in vitro with antisense oligonucleotide treatment

We previously described a novel homozygous point mutation (FGB c.115-600A>G) located deep within intron 1 of the fibrinogen beta gene (FGB), as a likely cause of afibrinogenemia. While this was the only mutation detected, its pathological mechanism was unclear. Here we show the mutation causes the inclusion of a 50-bp cryptic exon by creating a consensus heptad motif recognized by the spliceosome recruiting protein pre-mRNA splicing factor (SF2)/arginine/serine-rich alternative splicing factor (ASF) splicing factor 2/alternative splicing factor (SF2/ASF). Translation of the aberrant mRNA would result in truncation of the Bbeta chain, preventing fibrinogen synthesis. Selective introduction of a second mutation into the enhancer motif abolished the SF2/ASF binding motif and re-established normal pre-mRNA splicing. Subsequent introduction of antisense phosphorodiamidate morpholino oligonucleotides (PMOs) into transfected cells containing the mutant construct blocked the protein-RNA interaction and successfully restored normal splicing ( approximately 50% at 2 microM and approximately 90% at 10 microM). The molecular characterization of this case has revealed a unique disease mechanism, shown the importance of screening for deep intronic mutations, and provided evidence that antisense gene therapy is potentially practical for the treatment of diseases caused by this class of mutation.     

Molecular analyses of novel ASAH1 mutations causing Farber lipogranulomatosis: analyses of exonic splicing enhancer inactivating mutation

Farber lipogranulomatosis is a rare autosomal recessive lysosomal storage disorder caused by mutations in the ASAH1 gene. In the largest ever study, we identified and characterized ASAH1 mutations from 11 independent Farber disease (FD) families. A total of 13 different mutations were identified including 1 splice, 1 polypyrimidine tract (PPT) deletion and 11 missense mutations. Eleven mutations were exclusive to the Indian population. The IVS6+4A>G splice and IVS5‐16delTTTTC PPT deletion mutations resulted in skipping of exon 6 precluding thereby the region responsible for cleavage of enzyme precursor. A missense mutation (p.V198A) resulted in skipping of exon 8 due to inactivation of an exonic splicing enhancer (ESE) element. This is the first report of mutations affecting PPT and ESE in the ASAH1 gene resulting in FD.     

An exon skipping-associated nonsense mutation in the dystrophin gene uncovers a complex interplay between multiple antagonistic splicing elements

A nonsense mutation c.4250T>A (p.Leu1417X) in the dystrophin gene of a patient with an intermediate phenotype of muscular dystrophy induces partial in-frame skipping of exon 31. On the basis of UV cross-linking assays and pull-down analysis, we present evidence that the skipping of this exon is because of the creation of an exonic splicing silencer, which acts as a highly specific binding site (UAGACA) for a known repressor protein, hnRNP A1. Recombinant hnRNP A1 represses exon inclusion both in vitro and in vivo upon transient transfection of C2C12 cells with Duchenne muscular dystrophy (DMD) minigenes carrying the c.4250T>A mutation. Furthermore, we identified a downstream splicing enhancer in the central region of exon 31. This region functions as a Tra2beta-dependent exonic splicing enhancer (ESE) in vitro when inserted into a heterologous splicing reporter, and deletion of the ESE showed that incorporation of exon 31 depends on the Tra2beta-dependent enhancer both in the wild-type and mutant context. We conclude that dystrophin exon 31 contains juxtaposed sequence motifs that collaborate to regulate exon usage. This is the first elucidation of the molecular mechanism leading to exon skipping in the dystrophin gene and allowing the occurrence of a milder phenotype than the expected DMD phenotype. The knowledge of which cis-acting sequence within an exon is important for its definition will be essential for the alternative gene therapy approaches based on modulation of splicing to bypass DMD-causing mutations in the endogenous dystrophin gene.