Copper-binding proteins and exonic splicing enhancers and silencers

Eukaryotic DNA codes not only for proteins but contains a wealth of information required for accurate splicing of messenger RNA precursors and inclusion of constitutively or alternatively spliced exons in mature transcripts. This "auxiliary" splicing code has been characterized as exonic splicing enhancers and silencers (ESE and ESS). The exact interplay between protein and splicing codes is, however, poorly understood. Here, we show that exons encoding copper-coordinating amino acids in human cuproproteins lack ESEs and/or have an excess of ESSs, yet RNA sequencing and expressed sequence tags data show that they are more efficiently included in mature transcripts by the splicing machinery than average exons. Their largely constitutive inclusion in messenger RNA is facilitated by stronger splice sites, including polypyrimidine tracts, consistent with an important role of the surrounding intron architecture in ensuring high expression of metal-binding residues during evolution. ESE/ESS profiles of codons and entire exons that code for copper-coordinating residues were very similar to those encoding residues that coordinate zinc but markedly different from those that coordinate calcium. Together, these results reveal how the traditional and auxiliary splicing motifs responded to constraints of metal coordination in proteins.

Exonic splicing code and coordination of divalent metals in proteins

Exonic sequences contain both protein-coding and RNA splicing information but the interplay of the protein and splicing code is complex and poorly understood. Here, we have studied traditional and auxiliary splicing codes of human exons that encode residues coordinating two essential divalent metals at the opposite ends of the Irving-Williams series, a universal order of relative stabilities of metal-organic complexes. We show that exons encoding Zn2+-coordinating amino acids are supported much less by the auxiliary splicing motifs than exons coordinating Ca2+. The handicap of the former is compensated by stronger splice sites and uridine-richer polypyrimidine tracts, except for position -3 relative to 3' splice junctions. However, both Ca2+ and Zn2+ exons exhibit close-to-constitutive splicing in multiple tissues, consistent with their critical importance for metalloprotein function and a relatively small fraction of expendable, alternatively spliced exons. These results indicate that constraints imposed by metal coordination spheres on RNA splicing have been efficiently overcome by the plasticity of exon-intron architecture to ensure adequate metalloprotein expression.

Transposon clusters as substrates for aberrant splice-site activation

Transposed elements (TEs) have dramatically shaped evolution of the exon-intron structure and significantly contributed to morbidity, but how recent TE invasions into older TEs cooperate in generating new coding sequences is poorly understood. Employing an updated repository of new exon-intron boundaries induced by pathogenic mutations, termed DBASS, here we identify novel TE clusters that facilitated exon selection. To explore the extent to which such TE exons maintain RNA secondary structure of their progenitors, we carried out structural studies with a composite exon that was derived from a long terminal repeat (LTR78) and AluJ and was activated by a C > T mutation optimizing the 5ʹ splice site. Using a combination of SHAPE, DMS and enzymatic probing, we show that the disease-causing mutation disrupted a conserved AluJ stem that evolved from helix 3.3 (or 5b) of 7SL RNA, liberating a primordial GC 5ʹ splice site from the paired conformation for interactions with the spliceosome. The mutation also reduced flexibility of conserved residues in adjacent exon-derived loops of the central Alu hairpin, revealing a cross-talk between traditional and auxilliary splicing motifs that evolved from opposite termini of 7SL RNA and were approximated by Watson-Crick base-pairing already in organisms without spliceosomal introns. We also identify existing Alu exons activated by the same RNA rearrangement. Collectively, these results provide valuable TE exon models for studying formation and kinetics of pre-mRNA building blocks required for splice-site selection and will be useful for fine-tuning auxilliary splicing motifs and exon and intron size constraints that govern aberrant splice-site activation.

Cancer-Associated Substitutions in RNA Recognition Motifs of PUF60 and U2AF65 Reveal Residues Required for Correct Folding and 3' Splice-Site Selection

U2AF65 (U2AF2) and PUF60 (PUF60) are splicing factors important for recruitment of the U2 small nuclear ribonucleoprotein to lariat branch points and selection of 3′ splice sites (3′ss). Both proteins preferentially bind uridine-rich sequences upstream of 3′ss via their RNA recognition motifs (RRMs). Here, we examined 36 RRM substitutions reported in cancer patients to identify variants that alter 3′ss selection, RNA binding and protein properties. Employing PUF60- and U2AF65-dependent 3′ss previously identified by RNA-seq of depleted cells, we found that 43% (10/23) and 15% (2/13) of independent RRM mutations in U2AF65 and PUF60, respectively, conferred splicing defects. At least three RRM mutations increased skipping of internal U2AF2 (~9%, 2/23) or PUF60 (~8%, 1/13) exons, indicating that cancer-associated RRM mutations can have both cis- and trans-acting effects on splicing. We also report residues required for correct folding/stability of each protein and map functional RRM substitutions on to existing high-resolution structures of U2AF65 and PUF60. These results identify new RRM residues critical for 3′ss selection and provide relatively simple tools to detect clonal RRM mutations that enhance the mRNA isoform diversity.

Thioflavin T Monitoring of Guanine Quadruplex Formation in the rs689-Dependent INS Intron 1

The human proinsulin gene (INS) contains a thymine-to-adenine variant (rs689) located in the 3′ splice site (3′ ss) recognition motif of the first intron. The adenine at rs689 is strongly associated with type 1 diabetes. By weakening the polypyrimidine tract, the adenine allele reduces the efficiency of intron 1 splicing, which can be ameliorated by antisense oligonucleotides blocking a splicing silencer located upstream of the 3′ ss. The silencer is surrounded by guanine-rich tracts that may form guanine quadruplexes (G4s) and modulate the accessibility of the silencer. Here, we employed thioflavin T (ThT) to monitor G4 formation in synthetic DNAs and RNAs derived from INS intron 1. We show that the antisense target is surrounded by ThT-positive segments in each direction, with oligoribonucleotides exhibiting consistently higher fluorescence than their DNA counterparts. The signal was reduced for ThT-positive oligonucleotides that were extended into the silencer, indicating that flanking G4s have a potential to mask target accessibility. Real-time monitoring of ThT fluorescence during INS transcription in vitro revealed a negative correlation with ex vivo splicing activities of corresponding INS constructs. Together, these results provide a better characterization of antisense targets in INS primary transcripts for restorative strategies designed to improve the INS splicing defect associated with type 1 diabetes.

Heterogeneity in the extent of linkage disequilibrium among exonic, intronic, non-coding RNA and intergenic chromosome regions

Whole-genome sequence data enable construction of high-resolution linkage disequilibrium (LD) maps revealing the LD structure of functional elements within genic and subgenic sequences. The Malecot-Morton model defines LD map distances in linkage disequilibrium units (LDUs), analogous to the centimorgan scale of linkage maps. For whole-genome sequence-derived LD maps, we introduce the ratio of corresponding map lengths kilobases/LDU to describe the extent of LD within genome components. The extent of LD is highly variable across the genome ranging from ~38 kb for intergenic sequences to ~858 kb for centromeric regions. LD is ~16% more extensive in genic, compared with intergenic sequences, reflecting relatively increased selection and/or reduced recombination in genes. The LD profile across 18,268 autosomal genes reveals reduced extent of LD, consistent with elevated recombination, in exonic regions near the 5' end of genes but more extensive LD, compared with intronic sequences, across more centrally located exons. Genes classified as essential and genes linked to Mendelian phenotypes show more extensive LD compared with genes associated with complex traits, perhaps reflecting differences in selective pressure. Significant differences between exonic, intronic and intergenic components demonstrate that fine-scale LD structure provides important insights into genome function, which cannot be revealed by LD analysis of much lower resolution array-based genotyping and conventional linkage maps.

Stoichiometries of U2AF35, U2AF65 and U2 snRNP reveal new early spliceosome assembly pathways

The selection of 3΄ splice sites (3΄ss) is an essential early step in mammalian RNA splicing reactions, but the processes involved are unknown. We have used single molecule methods to test whether the major components implicated in selection, the proteins U2AF35 and U2AF65 and the U2 snRNP, are able to recognize alternative candidate sites or are restricted to one pre-specified site. In the presence of adenosine triphosphate (ATP), all three components bind in a 1:1 stoichiometry with a 3΄ss. Pre-mRNA molecules with two alternative 3΄ss can be bound concurrently by two molecules of U2AF or two U2 snRNPs, so none of the components are restricted. However, concurrent occupancy inhibits splicing. Stoichiometric binding requires conditions consistent with coalescence of the 5΄ and 3΄ sites in a complex (I, initial), but if this cannot form the components show unrestricted and stochastic association. In the absence of ATP, when complex E forms, U2 snRNP association is unrestricted. However, if protein dephosphorylation is prevented, an I-like complex forms with stoichiometric association of U2 snRNPs and the U2 snRNA is base-paired to the pre-mRNA. Complex I differs from complex A in that the formation of complex A is associated with the loss of U2AF65 and 35.

A birth of bipartite exon by intragenic deletion

BACKGROUND

Disease-causing mutations that activate transposon-derived exons without creating a new splice-site consensus have been reported rarely, but they provided unique insights into our understanding of structural motifs required for inclusion of intronic sequences in mature transcripts.

METHODS

We employ a combination of experimental and computational techniques to characterize the first de novo bipartite exon activation in genetic disease.

RESULTS

The exon originated from two separate introns as a result of an in-frame COL4A5 deletion associated with a typical Alport syndrome. The deletion encompassed exons 38 through 41 and activated a cryptic 3' and 5' splice site that were derived from intron 37 and intron 41, respectively. The deletion breakpoint was in the middle of the new exon, with considerable complementarity between the two exonic parts, potentially bringing the cryptic 3' and 5' splice site into proximity. The 3' splice site, polypyrimidine tract and the branch site of the new exon were derived from an inactive, 5' truncated LINE-1 retrotransposon. This ancient LINE-1 copy sustained a series of mutations that created the highly conserved AG dinucleotide at the 3' splice site early in primate development. The exon was fully included in mature transcripts and introduced a stop codon in the shortened COL4A5 mRNA, illustrating pitfalls of inferring disease severity from DNA mutation alone.

CONCLUSION

These results expand the repertoire of mutational mechanisms that alter RNA processing in genetic disease and illustrate the extraordinary versatility of transposed elements in shaping the new exon-intron structure and the phenotypic variability.

Antisense Oligonucleotides Modulating Activation of a Nonsense-Mediated RNA Decay Switch Exon in the ATM Gene

ATM (ataxia-telangiectasia, mutated) is an important cancer susceptibility gene that encodes a key apical kinase in the DNA damage response pathway. ATM mutations in the germ line result in ataxia-telangiectasia (A-T), a rare genetic syndrome associated with hypersensitivity to double-strand DNA breaks and predisposition to lymphoid malignancies. ATM expression is limited by a tightly regulated nonsense-mediated RNA decay (NMD) switch exon (termed NSE) located in intron 28. In this study, we identify antisense oligonucleotides that modulate NSE inclusion in mature transcripts by systematically targeting the entire 3.1-kb-long intron. Their identification was assisted by a segmental deletion analysis of transposed elements, revealing NSE repression upon removal of a distant antisense Alu and NSE activation upon elimination of a long terminal repeat transposon MER51A. Efficient NSE repression was achieved by delivering optimized splice-switching oligonucleotides to embryonic and lymphoblastoid cells using chitosan-based nanoparticles. Together, these results provide a basis for possible sequence-specific radiosensitization of cancer cells, highlight the power of intronic antisense oligonucleotides to modify gene expression, and demonstrate transposon-mediated regulation of NSEs.

Alternative splicing of U2AF1 reveals a shared repression mechanism for duplicated exons

The auxiliary factor of U2 small nuclear ribonucleoprotein (U2AF) facilitates branch point (BP) recognition and formation of lariat introns. The gene for the 35-kD subunit of U2AF gives rise to two protein isoforms (termed U2AF35a and U2AF35b) that are encoded by alternatively spliced exons 3 and Ab, respectively. The splicing recognition sequences of exon 3 are less favorable than exon Ab, yet U2AF35a expression is higher than U2AF35b across tissues. We show that U2AF35b repression is facilitated by weak, closely spaced BPs next to a long polypyrimidine tract of exon Ab. Each BP lacked canonical uridines at position -2 relative to the BP adenines, with efficient U2 base-pairing interactions predicted only for shifted registers reminiscent of programmed ribosomal frameshifting. The BP cluster was compensated by interactions involving unpaired cytosines in an upstream, EvoFold-predicted stem loop (termed ESL) that binds FUBP1/2. Exon Ab inclusion correlated with predicted free energies of mutant ESLs, suggesting that the ESL operates as a conserved rheostat between long inverted repeats upstream of each exon. The isoform-specific U2AF35 expression was U2AF65-dependent, required interactions between the U2AF-homology motif (UHM) and the α6 helix of U2AF35, and was fine-tuned by exon Ab/3 variants. Finally, we identify tandem homologous exons regulated by U2AF and show that their preferential responses to U2AF65-related proteins and SRSF3 are associated with unpaired pre-mRNA segments upstream of U2AF-repressed 3′ss. These results provide new insights into tissue-specific subfunctionalization of duplicated exons in vertebrate evolution and expand the repertoire of exon repression mechanisms that control alternative splicing.

Exon-centric regulation of ATM expression is population-dependent and amenable to antisense modification by pseudoexon targeting

ATM is an important cancer susceptibility gene that encodes a critical apical kinase of the DNA damage response (DDR) pathway. We show that a key nonsense-mediated RNA decay switch exon (NSE) in ATM is repressed by U2AF, PUF60 and hnRNPA1. The NSE activation was haplotype-specific and was most promoted by cytosine at rs609621 in the NSE 3' splice-site (3'ss), which is predominant in high cancer risk populations. NSE levels were deregulated in leukemias and were influenced by the identity of U2AF35 residue 34. We also identify splice-switching oligonucleotides (SSOs) that exploit competition of adjacent pseudoexons to modulate NSE levels. The U2AF-regulated exon usage in the ATM signalling pathway was centred on the MRN/ATM-CHEK2-CDC25-cdc2/cyclin-B axis and preferentially involved transcripts implicated in cancer-associated gene fusions and chromosomal translocations. These results reveal important links between 3'ss control and ATM-dependent responses to double-strand DNA breaks, demonstrate functional plasticity of intronic variants and illustrate versatility of intronic SSOs that target pseudo-3'ss to modify gene expression.

X-linked Alport syndrome associated with a synonymous p.Gly292Gly mutation alters the splicing donor site of the type IV collagen alpha chain 5 gene

BACKGROUND

X-linked Alport syndrome (XLAS) is a progressive hereditary nephropathy caused by mutations in the type IV collagen alpha chain 5 gene (COL4A5). Although many COL4A5 mutations have previously been identified, pathogenic synonymous mutations have not yet been described.

METHODS

A family with XLAS underwent mutational analyses of COL4A5 by PCR and direct sequencing, as well as transcript analysis of potential splice site mutations. In silico analysis was also conducted to predict the disruption of splicing factor binding sites. Immunohistochemistry (IHC) of kidney biopsies was used to detect α2 and α5 chain expression.

RESULTS

We identified a hemizygous point mutation, c.876A>T, in exon 15 of COL4A5 in the proband and his brother, which is predicted to result in a synonymous amino acid change, p.(Gly292Gly). Transcript analysis showed that this mutation potentially altered splicing because it disrupted the splicing factor binding site. The kidney biopsy of the proband showed lamellation of the glomerular basement membrane (GBM), while IHC revealed negative α5(IV) staining in the GBM and Bowman's capsule, which is typical of XLAS.

CONCLUSIONS

This is the first report of a synonymous COL4A5 substitution being responsible for XLAS. Our findings suggest that transcript analysis should be conducted for the future correct assessment of silent mutations.

The role of short RNA loops in recognition of a single-hairpin exon derived from a mammalian-wide interspersed repeat

Splice-site selection is controlled by secondary structure through sequestration or approximation of splicing signals in primary transcripts but the exact role of even the simplest and most prevalent structural motifs in exon recognition remains poorly understood. Here we took advantage of a single-hairpin exon that was activated in a mammalian-wide interspersed repeat (MIR) by a mutation stabilizing a terminal triloop, with splice sites positioned close to each other in a lower stem of the hairpin. We first show that the MIR exon inclusion in mRNA correlated inversely with hairpin stabilities. Employing a systematic manipulation of unpaired regions without altering splice-site configuration, we demonstrate a high correlation between exon inclusion of terminal tri- and tetraloop mutants and matching tri-/tetramers in splicing silencers/enhancers. Loop-specific exon inclusion levels and enhancer/silencer associations were preserved across primate cell lines, in 4 hybrid transcripts and also in the context of a distinct stem, but only if its loop-closing base pairs were shared with the MIR hairpin. Unlike terminal loops, splicing activities of internal loop mutants were predicted by their intramolecular Watson-Crick interactions with the antiparallel strand of the MIR hairpin rather than by frequencies of corresponding trinucleotides in splicing silencers/enhancers. We also show that splicing outcome of oligonucleotides targeting the MIR exon depend on the identity of the triloop adjacent to their antisense target. Finally, we identify proteins regulating MIR exon recognition and reveal a distinct requirement of adjacent exons for C-terminal extensions of Tra2α and Tra2β RNA recognition motifs.

Identification of U2AF(35)-dependent exons by RNA-Seq reveals a link between 3' splice-site organization and activity of U2AF-related proteins

The auxiliary factor of U2 small nuclear RNA (U2AF) is a heterodimer consisting of 65- and 35-kD proteins that bind the polypyrimidine tract (PPT) and AG dinucleotides at the 3′ splice site (3′ss). The gene encoding U2AF35 (U2AF1) is alternatively spliced, giving rise to two isoforms U2AF35a and U2AF35b. Here, we knocked down U2AF35 and each isoform and characterized transcriptomes of HEK293 cells with varying U2AF35/U2AF65 and U2AF35a/b ratios. Depletion of both isoforms preferentially modified alternative RNA processing events without widespread failure to recognize 3′ss or constitutive exons. Over a third of differentially used exons were terminal, resulting largely from the use of known alternative polyadenylation (APA) sites. Intronic APA sites activated in depleted cultures were mostly proximal whereas tandem 3′UTR APA was biased toward distal sites. Exons upregulated in depleted cells were preceded by longer AG exclusion zones and PPTs than downregulated or control exons and were largely activated by PUF60 and repressed by CAPERα. The U2AF(35) repression and activation was associated with a significant interchange in the average probabilities to form single-stranded RNA in the optimal PPT and branch site locations and sequences further upstream. Although most differentially used exons were responsive to both U2AF subunits and their inclusion correlated with U2AF levels, a small number of transcripts exhibited distinct responses to U2AF35a and U2AF35b, supporting the existence of isoform-specific interactions. These results provide new insights into function of U2AF and U2AF35 in alternative RNA processing.

X-linked Alport syndrome caused by splicing mutations in COL4A5

BACKGROUND AND OBJECTIVES

X-linked Alport syndrome is caused by mutations in the COL4A5 gene. Although many COL4A5 mutations have been detected, the mutation detection rate has been unsatisfactory. Some men with X-linked Alport syndrome show a relatively mild phenotype, but molecular basis investigations have rarely been conducted to clarify the underlying mechanism.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

In total, 152 patients with X-linked Alport syndrome who were suspected of having Alport syndrome through clinical and pathologic investigations and referred to the hospital for mutational analysis between January of 2006 and January of 2013 were genetically diagnosed. Among those patients, 22 patients had suspected splice site mutations. Transcripts are routinely examined when suspected splice site mutations for abnormal transcripts are detected; 11 of them showed expected exon skipping, but others showed aberrant splicing patterns. The mutation detection strategy had two steps: (1) genomic DNA analysis using PCR and direct sequencing and (2) mRNA analysis using RT-PCR to detect RNA processing abnormalities.

RESULTS

Six splicing consensus site mutations resulting in aberrant splicing patterns, one exonic mutation leading to exon skipping, and four deep intronic mutations producing cryptic splice site activation were identified. Interestingly, one case produced a cryptic splice site with a single nucleotide substitution in the deep intron that led to intronic exonization containing a stop codon; however, the patient showed a clearly milder phenotype for X-linked Alport syndrome in men with a truncating mutation. mRNA extracted from the kidney showed both normal and abnormal transcripts, with the normal transcript resulting in the milder phenotype. This novel mechanism leads to mild clinical characteristics.

CONCLUSIONS

This report highlights the importance of analyzing transcripts to enhance the mutation detection rate and provides insight into genotype-phenotype correlations. This approach can clarify the cause of atypically mild phenotypes in X-linked Alport syndrome.

Natural history of genetically proven autosomal recessive Alport syndrome

BACKGROUND

Autosomal recessive Alport syndrome (ARAS) is a rare hereditary disease caused by homozygous or compound heterozygous mutations in either the COL4A3 or COL4A4 genes. Failure to diagnose ARAS cases is common, even if detailed clinical and pathological examinations are carried out. As the mutation detection rate for ARAS is unsatisfactory, we sought to develop more reliable diagnostic methods and provide a better description of the clinicopathological characteristics of this disorder.

METHODS

A retrospective analysis of 30 genetically diagnosed patients with ARAS in 24 pedigrees was conducted. The mutation detection strategy comprised three steps: (1) genomic DNA analysis using polymerase chain reaction (PCR) and direct sequencing; (2) mRNA analysis using reverse transcription (RT)-PCR to detect RNA processing abnormalities; (3) semi-quantitative PCR using capillary electrophoresis to detect large heterozygous deletions.

RESULTS

Using the three-step analysis, we identified homozygous or compound heterozygous mutations in all patients. Interestingly, 20% of our ARAS patients showed normal expression of α5 in kidney tissue. The median age of developing end-stage renal disease was 21 years.

CONCLUSIONS

The strategy described in this study improves the diagnosis for ARAS families. Although immunohistochemical analysis of α5 can provide diagnostic information, normal distribution does not exclude the diagnosis of ARAS.

Optimal antisense target reducing INS intron 1 retention is adjacent to a parallel G quadruplex

Splice-switching oligonucleotides (SSOs) have been widely used to inhibit exon usage but antisense strategies that promote removal of entire introns to increase splicing-mediated gene expression have not been developed. Here we show reduction of INS intron 1 retention by SSOs that bind transcripts derived from a human haplotype expressing low levels of proinsulin. This haplotype is tagged by a polypyrimidine tract variant rs689 that decreases the efficiency of intron 1 splicing and increases the relative abundance of mRNAs with extended 5' untranslated region (5' UTR), which curtails translation. Co-expression of haplotype-specific reporter constructs with SSOs bound to splicing regulatory motifs and decoy splice sites in primary transcripts revealed a motif that significantly reduced intron 1-containing mRNAs. Using an antisense microwalk at a single nucleotide resolution, the optimal target was mapped to a splicing silencer containing two pseudoacceptor sites sandwiched between predicted RNA guanine (G) quadruplex structures. Circular dichroism spectroscopy and nuclear magnetic resonance of synthetic G-rich oligoribonucleotide tracts derived from this region showed formation of a stable parallel 2-quartet G-quadruplex on the 3' side of the antisense retention target and an equilibrium between quadruplexes and stable hairpin-loop structures bound by optimal SSOs. This region interacts with heterogeneous nuclear ribonucleoproteins F and H that may interfere with conformational transitions involving the antisense target. The SSO-assisted promotion of weak intron removal from the 5' UTR through competing noncanonical and canonical RNA structures may facilitate development of novel strategies to enhance gene expression.

Alport syndrome caused by a COL4A5 deletion and exonization of an adjacent AluY

Mutation-induced activation of splice sites in intronic repetitive sequences has contributed significantly to the evolution of exon–intron structure and genetic disease. Such events have been associated with mutations within transposable elements, most frequently in mutation hot-spots of Alus. Here, we report a case of Alu exonization resulting from a 367-nt genomic COL4A5 deletion that did not encompass any recognizable transposed element, leading to the Alport syndrome. The deletion brought to proximity the 5′ splice site of COL4A5 exon 33 and a cryptic 3′ splice site in an antisense AluY copy in intron 32. The fusion exon was depleted of purines and purine-rich splicing enhancers, but had low levels of intramolecular secondary structure, was flanked by short introns and had strong 5′ and Alu-derived 3′ splice sites, apparently compensating poor composition and context of the new exon. This case demonstrates that Alu splice sites can be activated by outlying deletions, highlighting Alu versatility in shaping the exon–intron organization and expanding the spectrum of mutational mechanisms that introduce repetitive sequences in mRNAs.

Compensatory signals associated with the activation of human GC 5' splice sites

GC 5′ splice sites (5′ss) are present in ∼1% of human introns, but factors promoting their efficient selection are poorly understood. Here, we describe a case of X-linked agammaglobulinemia resulting from a GC 5′ss activated by a mutation in BTK intron 3. This GC 5′ss was intrinsically weak, yet it was selected in >90% primary transcripts in the presence of a strong and intact natural GT counterpart. We show that efficient selection of this GC 5′ss required a high density of GAA/CAA-containing splicing enhancers in the exonized segment and was promoted by SR proteins 9G8, Tra2β and SC35. The GC 5′ss was efficiently inhibited by splice-switching oligonucleotides targeting either the GC 5′ss itself or the enhancer. Comprehensive analysis of natural GC-AG introns and previously reported pathogenic GC 5′ss showed that their efficient activation was facilitated by higher densities of splicing enhancers and lower densities of silencers than their GT 5′ss equivalents. Removal of the GC-AG introns was promoted to a minor extent by the splice-site strength of adjacent exons and inhibited by flanking Alu repeats, with the first downstream Alus located on average at a longer distance from the GC 5′ss than other transposable elements. These results provide new insights into the splicing code that governs selection of noncanonical splice sites.

Cryptic splice sites and split genes

We describe a new program called cryptic splice finder (CSF) that can reliably identify cryptic splice sites (css), so providing a useful tool to help investigate splicing mutations in genetic disease. We report that many css are not entirely dormant and are often already active at low levels in normal genes prior to their enhancement in genetic disease. We also report a fascinating correlation between the positions of css and introns, whereby css within the exons of one species frequently match the exact position of introns in equivalent genes from another species. These results strongly indicate that many introns were inserted into css during evolution and they also imply that the splicing information that lies outside some introns can be independently recognized by the splicing machinery and was in place prior to intron insertion. This indicates that non-intronic splicing information had a key role in shaping the split structure of eukaryote genes.

Prediction of single-nucleotide substitutions that result in exon skipping: identification of a splicing silencer in BRCA1 exon 6

Missense, nonsense, and translationally silent mutations can inactivate genes by altering the inclusion of mutant exons in mRNA, but their overall frequency among disease-causing exonic substitutions is unknown. Here, we have tested missense and silent mutations deposited in the BRCA1 mutation databases of unclassified variants for their effects on exon inclusion. Analysis of 21 BRCA1 variants using minigene assays revealed a single exon-skipping mutation c.231G>T. Comprehensive mutagenesis of an adjacent 12-nt segment showed that this silent mutation resulted in a higher level of exon skipping than the 35 other single-nucleotide substitutions. Exon inclusion levels of mutant constructs correlated significantly with predicted splicing enhancers/silencers, prompting the development of two online utilities freely available at http://www.dbass.org.uk. EX-SKIP quickly estimates which allele is more susceptible to exon skipping, whereas HOT-SKIP examines all possible mutations at each exon position and identifies candidate exon-skipping positions/substitutions. We demonstrate that the distribution of exon-skipping and disease-associated substitutions previously identified in coding regions was biased toward top-ranking HOT-SKIP mutations. Finally, we show that proteins 9G8, SC35, SF2/ASF, Tra2, and hnRNP A1 were associated with significant alterations of BRCA1 exon 6 inclusion in the mRNA. Together, these results facilitate prediction of exonic substitutions that reduce exon inclusion in mature transcripts.

DBASS3 and DBASS5: databases of aberrant 3'- and 5'-splice sites

DBASS3 and DBASS5 provide comprehensive repositories of new exon boundaries that were induced by pathogenic mutations in human disease genes. Aberrant 5'- and 3'-splice sites were activated either by mutations in the consensus sequences of natural exon-intron junctions (cryptic sites) or elsewhere ('de novo' sites). DBASS3 and DBASS5 currently contain approximately 900 records of cryptic and de novo 3'- and 5'-splice sites that were produced by over a thousand different mutations in approximately 360 genes. DBASS3 and DBASS5 data can be searched by disease phenotype, gene, mutation, location of aberrant splice sites in introns and exons and their distance from authentic counterparts, by bibliographic references and by the splice-site strength estimated with several prediction algorithms. The user can also retrieve reference sequences of both aberrant and authentic splice sites with the underlying mutation. These data will facilitate identification of introns or exons frequently involved in aberrant splicing, mutation analysis of human disease genes and study of germline or somatic mutations that impair RNA processing. Finally, this resource will be useful for fine-tuning splice-site prediction algorithms, better definition of auxiliary splicing signals and design of new reporter assays. DBASS3 and DBASS5 are freely available at http://www.dbass.org.uk/.

Transposable elements in disease-associated cryptic exons

Transposable elements (TEs) make up a half of the human genome, but the extent of their contribution to cryptic exon activation that results in genetic disease is unknown. Here, a comprehensive survey of 78 mutation-induced cryptic exons previously identified in 51 disease genes revealed the presence of TEs in 40 cases (51%). Most TE-containing exons were derived from short interspersed nuclear elements (SINEs), with Alus and mammalian interspersed repeats (MIRs) covering >18 and >16% of the exonized sequences, respectively. The majority of SINE-derived cryptic exons had splice sites at the same positions of the Alu/MIR consensus as existing SINE exons and their inclusion in the mRNA was facilitated by phylogenetically conserved changes that improved both traditional and auxiliary splicing signals, thus marking intronic TEs amenable for pathogenic exonization. The overrepresentation of MIRs among TE exons is likely to result from their high average exon inclusion levels, which reflect their strong splice sites, a lack of splicing silencers and a high density of enhancers, particularly (G)AA(G) motifs. These elements were markedly depleted in antisense Alu exons, had the most prominent position on the exon-intron gradient scale and are proposed to promote exon definition through enhanced tertiary RNA interactions involving unpaired (di)adenosines. The identification of common mechanisms by which the most dynamic parts of the genome contribute both to new exon creation and genetic disease will facilitate detection of intronic mutations and the development of computational tools that predict TE hot-spots of cryptic exon activation.

Disease-causing mutations improving the branch site and polypyrimidine tract: pseudoexon activation of LINE-2 and antisense Alu lacking the poly(T)-tail

Cryptic exons or pseudoexons are typically activated by point mutations that create GT or AG dinucleotides of new 5' or 3' splice sites in introns, often in repetitive elements. Here we describe two cases of tetrahydrobiopterin deficiency caused by mutations improving the branch point sequence and polypyrimidine tracts of repeat-containing pseudoexons in the PTS gene. In the first case, we demonstrate a novel pathway of antisense Alu exonization, resulting from an intronic deletion that removed the poly(T)-tail of antisense AluSq. The deletion brought a favorable branch point sequence within proximity of the pseudoexon 3' splice site and removed an upstream AG dinucleotide required for the 3' splice site repression on normal alleles. New Alu exons can thus arise in the absence of poly(T)-tails that facilitated inclusion of most transposed elements in mRNAs by serving as polypyrimidine tracts, highlighting extraordinary flexibility of Alu repeats in shaping intron-exon structure. In the other case, a PTS pseudoexon was activated by an A>T substitution 9 nt upstream of its 3' splice site in a LINE-2 sequence, providing the first example of a disease-causing exonization of the most ancient interspersed repeat. These observations expand the spectrum of mutational mechanisms that introduce repetitive sequences in mature transcripts and illustrate the importance of intronic mutations in alternative splicing and phenotypic variability of hereditary disorders.

Ab initio prediction of mutation-induced cryptic splice-site activation and exon skipping

Mutations that affect splicing of precursor messenger RNAs play a major role in the development of hereditary diseases. Most splicing mutations have been found to eliminate GT or AG dinucleotides that define the 5' and 3' ends of introns, leading to exon skipping or cryptic splice-site activation. Although accurate description of the mis-spliced transcripts is critical for predicting phenotypic consequences of these alterations, their exact nature in affected individuals cannot often be determined experimentally. Using a comprehensive collection of exons that sustained cryptic splice-site activation or were skipped as a result of splice-site mutations, we have developed a multivariate logistic discrimination procedure that distinguishes the two aberrant splicing outcomes from DNA sequences. The new algorithm was validated using an independent sample of exons and implemented as a free online utility termed CRYP-SKIP (http://www.dbass.org.uk/cryp-skip/). The web application takes up one or more mutated alleles, each consisting of one exon and flanking intronic sequences, and provides a list of important predictor variables and their values, the overall probability of activating cryptic splice vs exon skipping, and the location and intrinsic strength of predicted cryptic splice sites in the input sequence. These results will facilitate phenotypic prediction of splicing mutations and provide further insights into splicing enhancer and silencer elements and their relative importance for splice-site selection in vivo.

Position-dependent repression and promotion of DQB1 intron 3 splicing by GGGG motifs

Alternative splicing of HLA-DQB1 exon 4 is allele-dependent and results in variable expression of soluble DQbeta. We have recently shown that differential inclusion of this exon in mature transcripts is largely due to intron 3 variants in the branch point sequence (BPS) and polypyrimidine tract. To identify additional regulatory cis-elements that contribute to haplotype-specific splicing of DQB1, we systematically examined the effect of guanosine (G) repeats on intron 3 removal. We found that the GGG or GGGG repeats generally improved splicing of DQB1 intron 3, except for those that were adjacent to the 5' splice site where they had the opposite effect. The most prominent splicing enhancement was conferred by GGGG motifs arranged in tandem upstream of the BPS. Replacement of a G-rich segment just 5' of the BPS with a series of random sequences markedly repressed splicing, whereas substitutions of a segment further upstream that lacked the G-rich elements and had the same size did not result in comparable splicing inhibition. Systematic mutagenesis of both suprabranch guanosine quadruplets (G(4)) revealed a key role of central G residues in splicing enhancement, whereas cytosines in these positions had the most prominent repressive effects. Together, these results show a significant role of tandem G(4)NG(4) structures in splicing of both complete and truncated DQB1 intron 3, support position dependency of G repeats in splicing promotion and inhibition, and identify positively and negatively acting sequences that contribute to the haplotype-specific DQB1 expression.

Variants in the human insulin gene that affect pre-mRNA splicing: is -23HphI a functional single nucleotide polymorphism at IDDM2?

Predisposition to type 1 diabetes and juvenile obesity is influenced by the susceptibility locus IDDM2 that includes the insulin gene (INS). Although the risk conferred by IDDM2 has been attributed to a minisatellite upstream of INS, intragenic variants have not been ruled out. We examined whether INS polymorphisms affect pre-mRNA splicing and proinsulin secretion using minigene reporter assays. We show that IVS1-6A/T (-23HphI+/-) is a key INS variant that influences alternative splicing of intron 1 through differential recognition of its 3' splice site. The A allele resulted in an increased production of mature transcripts with a long 5' leader in several cell lines, and the extended mRNAs generated more proinsulin in culture supernatants than natural transcripts. The longer mRNAs were significantly overrepresented among beta-cell-expressed sequenced tags containing the A allele as compared with those with T alleles. In addition, we show that a rare insertion/deletion polymorphism IVS1+5insTTGC (IVS-69), which is exclusively present in Africans, activated a downstream cryptic 5' splice site, extending the 5' leader by 30 bp. These results indicate that -23HphI and IVS-69 are the most important INS variants affecting pre-mRNA splicing and suggest that -23HphI+/- is a common functional single nucleotide polymorphism at IDDM2.

Identification of splicing silencers and enhancers in sense Alus: a role for pseudoacceptors in splice site repression

Auxiliary splicing signals in introns play an important role in splice site selection, but these elements are poorly understood. We show that a subset of serine/arginine (SR)-rich proteins activate a cryptic 3' splice site in a sense Alu repeat located in intron 4 of the human LST1 gene. Utilization of this cryptic splice site is controlled by juxtaposed Alu-derived splicing silencers and enhancers between closely linked short tandem repeats TNFd and TNFe. Systematic mutagenesis of these elements showed that AG dinucleotides that were not preceded by purine residues were critical for repressing exon inclusion of a chimeric splicing reporter. Since the splice acceptor-like sequences are present in excess in exonic splicing silencers, these signals may contribute to inhibition of a large number of pseudosites in primate genomes.

Branch site haplotypes that control alternative splicing

We show that the allele-dependent expression of transcripts encoding soluble HLA-DQbeta chains is determined by branchpoint sequence (BPS) haplotypes in DQB1 intron 3. BPS RNAs associated with low inclusion of the transmembrane exon in mature transcripts showed impaired binding to splicing factor 1 (SF1), indicating that alternative splicing of DQB1 is controlled by differential BPS recognition early during spliceosome assembly. We also demonstrate that naturally occurring human BPS point mutations that alter splicing and lead to recognizable phenotypes cluster in BP and in position -2 relative to BP, implicating impaired SF1-BPS interactions in disease-associated BPS substitutions. Coding DNA variants produced smaller fluctuations of exon inclusion levels than random exonic substitutions, consistent with a selection against coding mutations that alter their own exonization. Finally, proximal splicing in this multi-allelic reporter system was promoted by at least seven SR proteins and repressed by hnRNPs F, H and I, supporting an extensive antagonism of factors balancing the splice site selection. These results provide the molecular basis for the haplotype-specific expression of soluble DQbeta, improve prediction of intronic point mutations and indicate how extraordinary, selection-driven DNA variability in HLA affects pre-mRNA splicing.

Deletion and reduced expression of the Fanconi anemia FANCA gene in sporadic acute myeloid leukemia

Fanconi anemia (FA) is an autosomal recessive chromosomal instability disorder caused by mutations in one of seven known genes (FANCA,C,D2,E,F,G and BRCA2). Mutations in the FANCA gene are the most prevalent, accounting for two-thirds of FA cases. Affected individuals have greatly increased risks of acute myeloid leukemia (AML). This raises the question as to whether inherited or acquired mutations in FA genes might be involved in the development of sporadic AML. Quantitative fluorescent PCR was used to screen archival DNA from sporadic AML cases for FANCA deletions, which account for 40% of FANCA mutations in FA homozygotes. Four heterozygous deletions were found in 101 samples screened, which is 35-fold higher than the expected population frequency for germline FANCA deletions (P<0.0001). Sequencing FANCA in the AML samples with FANCA deletions did not detect mutations in the second allele and there was no evidence of epigenetic silencing by hypermethylation. However, real-time quantitative PCR analysis in these samples showed reduced expression of FANCA compared to nondeleted AML samples and to controls. These findings suggest that gene deletions and reduced expression of FANCA may be involved in the promotion of genetic instability in a subset of cases of sporadic AML.

Fine-scale mapping at IGAD1 and genome-wide genetic linkage analysis implicate HLA-DQ/DR as a major susceptibility locus in selective IgA deficiency and common variable immunodeficiency

Selective IgA deficiency (IgAD) and common variable immunodeficiency (CVID) are the most common primary immunodeficiencies in humans. A high degree of familial clustering, marked differences in the population prevalence among ethnic groups, association of IgAD and CVID in families, and a predominant inheritance pattern in multiple-case pedigrees have suggested a strong, shared genetic predisposition. Previous genetic linkage, case-control, and family-based association studies mapped an IgAD/CVID susceptibility locus, designated IGAD1, to the MHC, but its precise location within the MHC has been controversial. We have analyzed a sample of 101 multiple- and 110 single-case families using 36 markers at the IGAD1 candidate region and mapped homozygous stretches across the MHC shared by affected family members. Haplotype analysis, linkage disequilibrium, and homozygosity mapping indicated that HLA-DQ/DR is the major IGAD1 locus, strongly suggesting the autoimmune pathogenesis of IgAD/CVID. This is supported by the highest excess of allelic sharing at 6p in the genome-wide linkage analysis of 101 IgAD/CVID families using 383 marker loci, by previously reported restrictions of the T cell repertoires in CVID, the presence of autoantibodies, impaired T cell activation, and a dysregulation of a number of genes in the targeted immune system. IgAD/CVID may thus provide a useful model for the study of pathogenesis and novel therapeutic strategies in autoimmune diseases.

The HLA-DRA*0102 allele: correct nucleotide sequence and associated HLA haplotypes

Here we correct the nucleotide sequence of a single known variant of the HLA-DRA gene. We show that the coding regions of the HLA-DRA*0101 and HLA-DRA*0102 alleles do not differ at two codons as reported previously, but only in codon 217. Using nucleotide sequencing and DNA samples from individuals homozygous in the major histocompatibility complex, we found that the variant, leucine 217-encoding HLA-DRA*0102 allele was present on the haplotypes HLA-B*0801, DRB1*03011, DQB1*0201 (ancestral haplotype AH8.1), HLA-B*07021, DRB1*15011, DQB1*0602 (AH7.1), HLA-B*1501, DRB1*15011, DQB1*0602, HLA-B*1501, DRB1*1402, DQB1*03011 and HLA-A3, B*07021, DRB1*1301, DQB1*0603. The HLA-DRA*0101 allele coding for valine 217 was observed on the haplotypes HLA-B*5701, DRB1*0701, DQB1*03032 (AH57.1), HLA-DRB1*04011, DQB1*0302, HLA-DRB1*0701, DQB1*0202, and HLA-DRB1*0101, DQB1*05011.

Coding haplotype analysis supports HCR as the putative susceptibility gene for psoriasis at the MHC PSORS1 locus

PSORS1, near HLA-C, is the major genetic determinant of psoriasis. We present genetic and structural evidence suggesting a major role for the HCR gene at the PSORS1 locus. Genotyping of 419 families from six populations revealed that coding single-nucleotide polymorphisms of HCR formed a conserved allele HCR*WWCC that associated highly significantly with psoriasis and with the HLA-Cw6 allele in all populations. Because of strong linkage disequilibrium between HLA-Cw6 and HCR*WWCC, the two genes could not be genetically distinguished by this sample size. However, the variant HCR allele was predicted to differ in secondary structure from the wild-type protein. HCR protein expression in lesional psoriatic skin differed considerably from that observed in normal skin. These results provide strong evidence for the HCR*WWCC allele as a major genetic determinant for psoriasis, probably by a mechanism impacting on keratinocyte proliferation.

Does 77C-->G in PTPRC modify autoimmune disorders linked to the major histocompatibility locus?

A 77G allele of the gene encoding CD45, also known as the protein tyrosine phosphatase receptor-type C gene (PTPRC), has been associated with multiple sclerosis (MS). Here we determine allele frequencies in large numbers of MS patients, primary immunodeficiencies linked to major histocompatibility complex (MHC) locus and over 1,000 controls to assess whether aberrant splicing of PTPRC caused by the 77C-->G polymorphism results in increased susceptibility to these diseases. Our results show no difference in the frequency of the 77G allele in patients and controls and thus do not support a causative role for the polymorphism in the development of disorders with a strong autoimmune component in etiology.

Short tandem repeat (STR) haplotypes in HLA: an integrated 50-kb STR/linkage disequilibrium/gene map between the RING3 and HLA-B genes and identification of STR haplotype diversification in the class III region

We present a dense STR/linkage disequilibrium(LD)/gene map between the RING3 and HLA-B loci, reference allelic sizes on the most prevalent HLA haplotypes and their allelic frequencies in pedigree founders. This resource will facilitate LD, evolution and gene mapping studies, including comparisons of HLA and STR haplotypes and identification of HLA recombinants. The map was constructed by testing novel and previously reported STRs using a panel of 885 individuals in 211 families and 60 DNA samples from cell lines and bone marrow donors homozygous in the HLA-A, -B and -DR loci selected from over 15 000 entries into the registry of Swedish bone marrow donors. We have also analysed the variability of STR alleles/haplotypes on the most prevalent HLA haplotypes to identify STRs useful for fine mapping of disease genes in the region previously implicated in susceptibility to many disorders. The analysis of 40 HLA-A*01, B*0801, DRB1*03011, DQB1*0201 haplotypes in homozygous donors showed a surprising stability in 23 STRs between the class II recombination hot spot and HLA-B, with the average of 1.9% (16/838) variant alleles. However, 40% variant alleles were found at the D6S2670 locus in intron 19 of the tenascin-X gene both in the families and homozygous donors. The nucleotide sequence analysis of this STR showed a complex polymorphism consisting of tetra- (CTTT)(8-18) and penta-nucleotide (CTTTT)(1-2) repeats, separated by an intervening non-polymorphic sequence of 42 bp. The HLA-A1, B*0801, DRB1*03011, DQB1*0201 haplotypes had five (CTTT)(14-18)/(CTTTT)(2) variants with a predominant (CTTT)(16) allele, implicating the tetranucleotide component as the source of this ancestral haplotype diversification, which may be due to the location of D6S2670 in the region of the highest GC content in the human MHC.

Allelic imbalance on chromosomes 13 and 17 and mutation analysis of BRCA1 and BRCA2 genes in monozygotic twins concordant for breast cancer

To study genetic changes associated with the development of breast cancer and the extent of its hereditary predisposition, paraffin-embedded tissue samples were obtained from monozygotic twin pairs concordant for breast cancer through the linked Swedish Twin and Cancer Registries. DNA samples extracted from the matched tumour and normal tissues of nine twin pairs were analysed for allelic imbalance using a series of microsatellite markers on chromosomes 13 and 17, containing loci with known tumour suppressor genes. Multiple losses of constitutional heterozygosity (LOH), consistent with a loss of large genomic region, the whole chromosome or chromosome arm, was found in at least three pairs of twins. One double mitotic crossover was identified in one tumour sample in a pair concordant for LOH at multiple loci on both chromosomes. Recombination breakpoints were mapped to regions delineated by D13S218 and D13S263, and D13S155 and D13S279, respectively. In general, no genetic effect of losing the same allele within a twin pair was found. However, for one marker at chromosome 13 (D13S328, between the BRCA2 and the RB-1 loci) and two markers on chromosome 17 (D17S786, distal to the p53 locus, and D17S855, an intragenic BRCA1 marker) the proportion of twin pairs with the same LOH was significantly higher than expected. These regions may reflect hereditary genomic changes in our sample set. In addition, tumour DNA samples from a subset of 12 twin pairs were analysed for BRCA1 and BRCA2 mutations using exon-by-exon single-strand conformation polymorphism analysis. Two unclassified BRCA2 variants, with a putative pathogenic effect, were identified, but no pathogenic alterations were found in the BRCA1 gene.

Low frequency of E-cadherin alterations in familial breast cancer

In order to explore the possible role of E-cadherin in familial cancer, 19 familial breast cancer patients, whose tumours demonstrated loss of heterozygosity (LOH) at the E-cadherin locus, were screened for germline mutations. No pathogenic germline alterations were detected in these individuals. However, a somatic mutation was found (49-2A-->C) in one of the tumours. This tumour showed a pattern of both ductal and lobular histology. Another 10 families with cases of breast, gastric and colon cancer were also screened for germline mutations, and no mutations were found. A missense mutation in exon 12 of E-cadherin (1774G-->A; Ala592Thr) was previously found in one family with diffuse gastric cancer, and colon and breast cancer. An allelic association study was performed to determine whether the Ala592Thr alteration predisposes to breast cancer. In total, we studied 484 familial breast cancer patients, 614 sporadic breast cancer patients and 497 control individuals. The frequencies of this alteration were similar in these groups. However, a correlation between the Ala592Thr alteration and ductal comedo-type tumour was seen. These results, together with previously reported studies, indicate that germline mutations and, more commonly, somatic mutations in E-cadherin may have an influence on the behaviour of the tumours, rather than predispose to breast cancer.