Abstract

In humans and great apes, CHRNA1 encoding the muscle nicotinic acetylcholine receptor α subunit carries an inframe exon P3A, the inclusion of which yields a nonfunctional α subunit. In muscle, the P3A(−) and P3A(+) transcripts are generated in a 1:1 ratio but the functional significance and regulation of the alternative splicing remain elusive. An intronic mutation (IVS3-8G>A), identified in a patient with congenital myasthenic syndrome, disrupts an intronic splicing silencer (ISS) and results in exclusive inclusion of the downstream P3A exon. We found that the ISS-binding splicing trans-factor was heterogeneous nuclear ribonucleoprotein (hnRNP) H and the mutation attenuated the affinity of hnRNP for the ISS ∼100-fold. We next showed that direct placement of hnRNP H to the 3′ end of intron 3 silences, and siRNA-mediated downregulation of hnRNP H enhances recognition of exon P3A. Analysis of the human genome suggested that the hnRNPH-binding UGGG motif is overrepresented close to the 3′ ends of introns. Pursuing this clue, we showed that alternative exons of GRIP1, FAS, VPS13C and NRCAM are downregulated by hnRNP H. Our findings imply that the presence of the hnRNP H-binding motif close to the 3′ end of an intron is an essential but underestimated splicing regulator of the downstream exon.

INTRODUCTION

Congenital myasthenic syndromes (CMSs) are caused by genetic defects of presynaptic, synaptic or postsynaptic molecules at the motor endplate (EP) (1). CMSs are autosomal recessive disorders except for the slow channel syndrome, in which neuromuscular transmission is compromised by autosomal-dominant gain-of-function mutations of the acetylcholine receptor (AChR) subunit genes. Most CMS are postsynaptic and most of these are caused by mutations in AChR subunit genes. The adult-type muscle nicotinic AChR comprises four homologous α, β, δ and ε subunits with the stoichiometry of α2βδε, encoded by CHRNA1, CHRNB1, CHRND and CHNRE, respectively, whereas the fetal-type AChR harbors the γ, encoded by CHRNG, instead of the ε subunit. When the ε subunit is defective owing to a null or low-expressor mutation in CHRNE, persistent expression of the fetal-type γ-AChR at the EP partially rescues the phenotype (2–5). Null mutations appearing in both alleles of non-ε subunit genes are likely fatal due to the lack of a substituting subunit, and only few instances of low-expressor mutations in both alleles of CHRNA1 (6,7), CHRNB1 (8) and CHRND (9) have been documented.

In 1990, Beeson et al. (10) reported that CHRNA1 carries an extra 75 nt inframe exon, named P3A, between exons 3 and 4, and that CHRNA1 gives rise to P3A(+) and P3A(−) transcripts. The P3A(−) transcript encodes a functional α subunit that becomes incorporated into functional AChR, whereas the P3A(+) transcript encodes a nonfunctional α subunit not expressed on the cell surface, although exon P3A contains no stop codon (11). Exon P3A likely arises from the exonization of the retroposed mammalian interspersed repeat element (12). Exon P3A is alternatively spliced in humans, gorillas, chimpanzees and orangutans, but not in rhesus monkeys, gibbons, mandrills, marmosets, dogs and cats (12,13). In human skeletal muscle, the P3A(−) and P3A(+) transcripts are generated in a 1:1 ratio (14). The P3A(+) transcript is also expressed in the normal thymus gland and in nonneoplastic thymus glands of myasthenic patients, but is absent (15) or rarely expressed (16) in thymomas. The functional significance of the P3A(+) transcript in muscle or in the thymus gland has not been elucidated to date.

No fewer than 74% of human multi-exon genes are alternatively spliced, which provides for a diverse array of proteome from a limited number of genes (17). Alternative splicing is achieved by exonic or intronic splicing enhancers (ESEs, ISEs) and exonic or intronic splicing silencers (ESSs, ISSs) in combination with spatial and temporal expression of trans-acting splicing factors, such as serine/arginine-rich (SR) proteins and heterogeneous nuclear ribonucleoproteins (hnRNPs) (18,19). Exonic and intronic mutations or polymorphisms potentially affect splicing cis-elements of constitutive and alternative exons (1,20–31) as well as of cryptic exons (32,33). Splicing trans-factors have been identified for some splicing mutations or polymorphisms (27,31,33–36).

hnRNP H is a member of the hnRNP H family that comprises hnRNP H, H′, F and 2H9 encoded by HNRPH1, HNRPH2, HNRPF and HNRPH3, respectively. Each family molecule specifically recognizes poly(G) sequences (G-tracts) (37) and either enhances or silences pre-mRNA splicing of a variety of genes. For example, hnRNP H binds to an ISE in the src pre-mRNA and promotes neuron-specific inclusion of the N1 exon (38). Similarly, hnRNP H binds to an ISE downstream of a brain-specific exon in GRIN1 (39). In addition, hnRNP H activates an ESE in the immunodeficiency virus (40,41) and the PLP1 gene (42). On the other hand, hnRNP H binds to an ESS in an alternative exon 7 of the rat β-tropomyosin gene and induces skipping of exon 7 (43). Among the four splicing cis-elements of ESE, ISE, ESS and ISS, hnRNP H has never been reported to bind to an ISS.

Here we report that an intronic mutation (IVS3-8G>A) upstream of CHRNA1 exon P3A identified in a CMS patient disrupts an ISS and exclusively yields a P3A(+) transcript that encodes a nonfunctional α subunit. We show that a splicing trans-factor, hnRNP H, binds to an ISS, and that the mutation decreases the binding affinity ∼100-fold. In addition, we find that alternative splicing of GRIP1, FAS, VPS13C and NRCAM is downregulated by the hnRNP H-binding UGGG motif close to the 3′ end of an intron. An hnRNP H-binding motif close to the 3′ end of an intron is likely an essential but underestimated splicing cis-regulatory element in an unidentified subset of alternatively spliced genes.

RESULTS

Patient

An 18-year-old woman had severe myasthenic symptoms since birth. She has a decremental EMG response, no anti-AChR antibodies and respond partially to combined treatment to anticholinesterase medications and 3,4-diaminopyridine. Her parents are not consanguineous, and there are no similarly affected relatives. An intercostal biopsy was obtained at age 3.

EP studies

We previously reported some of the patient’s EP studies (44). In brief, the expression of acetylcholinesterase was well preserved but that of AChR was greatly attenuated at patient EPs. The number of AChRs per EP was reduced to ∼20% of normal (Table 1). On electron microscopy, the structural integrity of the junctional folds and nerve terminals was preserved but some postsynaptic regions were simpler than normal. Ultrastructural localization of AChR with peroxidase-labeled α-bungarotoxin (α-bgt) revealed marked decrease in the density and distribution of AChR on the junctional folds. The AChR index (defined as the length of the postsynaptic membrane reacting for AChR normalized for the length of the primary synaptic cleft) was reduced to 19% of normal in the patient (Fig. 1 and Table 1). The amplitude of the miniature EP potentials (MEPPs) was reduced to 23% of normal. The number of quanta released by nerve impulse was normal (Table 1). The analysis of the acetylcholine (ACh)-induced current noise suggested mild shortening of the channel-opening events. To summarize, the safety margin of neuromuscular transmission in the patient is mainly compromised by the AChR deficiency.

Figure 1.

Ultrastructural localization of EP AChR with peroxidase-labeled α-bgt at a patient (A) and a control (B) EP. The density and distribution of AChR at the patient EP is markedly attenuated but the postsynaptic junctional folds are well developed. (A) and (B): ×10 000.

Figure 1.

Ultrastructural localization of EP AChR with peroxidase-labeled α-bgt at a patient (A) and a control (B) EP. The density and distribution of AChR at the patient EP is markedly attenuated but the postsynaptic junctional folds are well developed. (A) and (B): ×10 000.

Table 1.

EP studies

 Patient Controls 
[125I]α-bgt binding sites/EP 1.0 × 106 12.82 ± 0.79 × 106 (13 adults); 4.7 × 106 (3 years old) 
AChR index 0.60 ± 0.18 (37) 3.01 ± 0.11 (85) 
EPP quantal contenta 25 ± 2.3 (14) 31 ± 1 (190) 
MEPP amplitude (mV)b 0.23 ± 0.090 (16) 1.00 ± 0.025 (165) 
 Patient Controls 
[125I]α-bgt binding sites/EP 1.0 × 106 12.82 ± 0.79 × 106 (13 adults); 4.7 × 106 (3 years old) 
AChR index 0.60 ± 0.18 (37) 3.01 ± 0.11 (85) 
EPP quantal contenta 25 ± 2.3 (14) 31 ± 1 (190) 
MEPP amplitude (mV)b 0.23 ± 0.090 (16) 1.00 ± 0.025 (165) 

Values represent mean ± SE. Numbers in parentheses indicate number of EPs except for α-bgt-binding sites where they indicate number of controls. T = 29 ± 0.5C° for EPP and MEPP recordings.

aQuantal content of EPP at 1 Hz stimulation corrected for resting membrane potential of −80 mV, nonlinear summation and non-Poisson release.

bCorrected for resting membrane potential of −80 mV and a mean muscle fiber diameter of 55 µm.

Mutation analysis

Direct sequencing of AChR α, β, δ and ε subunit genes revealed that the patient is heterozygous for two mutations. The first mutation was a G-to-A substitution at position –8 in intron 3 preceding exon P3A of the α subunit (IVS3-8G>A) (Fig. 2A and B). The second mutation was a C-to-T substitution at nucleotide 937 (c.937C>T) in exon 7, which predicts an arginine-to-tryptophan substitution at codon 313 in the long cytoplasmic loop linking the third and fourth transmembrane domains of the α subunit (p.R313W) (Fig. 2A and C). R313 is widely conserved across vertebrate species (Fig. 2C). As the AChR α subunit has a signal peptide of 20 amino acids, nucleotide and amino acid numbers traditionally start from position 1 of the mature peptide. Thus, c.937C>T and p.R313W correspond to c.997C>T and p.R333W, respectively, when positions are counted from the translation initiation site.

Figure 2.

Identified mutations in CHRNA1. (A) Genomic structure of CHRNA1 and identified mutations. Exons and introns are drawn to scale. Shaded areas represent untranslated regions. (B) Nucleotide sequence of and around exon P3A. Corresponding gene segments in great apes are aligned against the human gene. Discordant nucleotides are shown in italic letters. Intronic nucleotides are shown in lower case letters, and exonic nucleotides in upper case. Predicted amino acids are shown at the top. Arrowheads point to the branch points identified by lariat RT-PCR. The upstream branch point conforms to the human branch point consensus sequence of yUnAy (70), but the downstream one does not. (C) Alignment of part of the long cytoplasmic loop of human AChR subunits and α subunits of other species. R313 is conserved in all the human subunits except for δ, and in α subunits of all species. (D) Family analyses show that each mutation is heteroallelic in the patient and is recessive. The patient is indicated by an arrow and a shaded symbol. Half-shaded symbols represent asymptomatic heterozygous carriers.

Figure 2.

Identified mutations in CHRNA1. (A) Genomic structure of CHRNA1 and identified mutations. Exons and introns are drawn to scale. Shaded areas represent untranslated regions. (B) Nucleotide sequence of and around exon P3A. Corresponding gene segments in great apes are aligned against the human gene. Discordant nucleotides are shown in italic letters. Intronic nucleotides are shown in lower case letters, and exonic nucleotides in upper case. Predicted amino acids are shown at the top. Arrowheads point to the branch points identified by lariat RT-PCR. The upstream branch point conforms to the human branch point consensus sequence of yUnAy (70), but the downstream one does not. (C) Alignment of part of the long cytoplasmic loop of human AChR subunits and α subunits of other species. R313 is conserved in all the human subunits except for δ, and in α subunits of all species. (D) Family analyses show that each mutation is heteroallelic in the patient and is recessive. The patient is indicated by an arrow and a shaded symbol. Half-shaded symbols represent asymptomatic heterozygous carriers.

None of the two mutations was present in 200 normal alleles. We traced IVS3-8G>A by direct sequencing and p.R313W by AciI restriction analysis in family members and found that IVS3-8G>A was inherited from the asymptomatic father and p.R313W from the asymmetric mother, indicating that each mutation is heteroallelic and recessive (Fig. 2D). EP AChR deficiency is an autosomal recessive disorder, and haploinsufficiency of the AChR subunit genes usually exhibits no clinical symptoms (1).

p.R313W is a low expressor in HEK cells and p.R313W has mild fast-channel properties

Wild-type or mutant α subunit cDNA was expressed along with wild-type β, δ and ε subunit cDNAs in HEK cells. The total [125I]α-bgt binding to AChR expressed on the HEK cell surface was normalized to that measured for wild-type AChR. Compared with wild-type AChR, the expression of αR313W-AChR was markedly reduced (mean ± SD, 23.6 ± 7.3%, n = 6).

Single-channel patch-clamp recordings obtained from HEK cells expressing αR313W-AChR in the presence of low concentrations of ACh revealed that opening bursts of the mutant receptor were reduced to 69% of wild type (mean ± SE, 2.29 ± 0.13 ms, six patches versus 3.31 ± 0.12 ms, 21 patches; P < 0.001). The shortening of opening bursts in HEK cells was comparable with the previous observation that the mean duration of channel-opening events was reduced to 70% of normal at patient EPs (44). Thus, αR313W-AChR has mild fast-channel properties.

IVS3-8G>A markedly enhances recognition of exon P3A in muscle

We initially assumed that IVS3-8G>A was a rare polymorphism because the mutation did not affect any known cis-acting element in intron 3 including the polypyrimidine tract or the branch points that we determined with lariat RT-PCR (Fig. 2B). However, an exhaustive search for other mutations in the α subunit, including single allele analysis by the ‘conversion’ method (45), disclosed no other mutation in the patient.

Allele-specific RT-PCR of the patient muscle revealed that IVS3-8G>A markedly enhanced the incorporation of exon P3A into mature mRNA and prevented the expression of the functional P3A(−) transcript (Fig. 3). Real-time RT–PCR analysis of muscle mRNA showed that the ratio of the P3A(−) transcript to the total CHRNA1 transcripts [or %P3A(−); see Materials and Methods] of the allele harboring IVS3-8G>A was 1.8 ± 0.2% (mean ± SD, n = 3), whereas the %P3A(−) value of the other allele carrying p.R313W was 61.3 ± 6.1% (n = 3). The %P3A(−) values of alleles carrying p.R313W were similar to those of normal controls (48.4 ± 1.8%, n = 4). Skeletal muscle expressing CHRNA1 mRNA was not available from the asymptomatic father or the asymptomatic brother carrying IVS3-8G>A.

Figure 3.

IVS3-8G>A invariably results in recognition of exon P3A. (A) Four possible transcripts in the muscle of the patient. An allele carrying IVS3-8G>A potentially gives rise to P3A(−) (I) and P3A(+) (II) transcripts. Similarly, an allele harboring p.R313W potentially yields P3A(−) (III) and P3A(+) (IV) transcripts. Transcript I, however, does not exist in the muscle of the patient as shown in the following panels. (B) Transcripts from R313- and W313-alleles are specifically amplified by allele-specific RT-PCR. Nested RT-PCR spanning exons 3 to 4 reveals that an allele with R313 gives rise to a P3A(−) transcript [transcript II in (A)], whereas an allele with W313 yields both P3A(−) and P3A(+) transcripts [transcripts III and IV in (A)]. (C) P3A(−) and P3A(+) transcripts are specifically amplified by allele-specific RT-PCR. Nested RT-PCR products spanning p.R313W are digested by AciI, which cuts only the wild-type R313 fragment (lower band) and leaves the mutant W313 fragment undigested (upper band). The P3A(−) transcript exclusively arises from an allele with W313 [transcript III in (A)], whereas the P3A(+) transcript arises from both R313- and W313-alleles [transcripts II and IV in (A)].

Figure 3.

IVS3-8G>A invariably results in recognition of exon P3A. (A) Four possible transcripts in the muscle of the patient. An allele carrying IVS3-8G>A potentially gives rise to P3A(−) (I) and P3A(+) (II) transcripts. Similarly, an allele harboring p.R313W potentially yields P3A(−) (III) and P3A(+) (IV) transcripts. Transcript I, however, does not exist in the muscle of the patient as shown in the following panels. (B) Transcripts from R313- and W313-alleles are specifically amplified by allele-specific RT-PCR. Nested RT-PCR spanning exons 3 to 4 reveals that an allele with R313 gives rise to a P3A(−) transcript [transcript II in (A)], whereas an allele with W313 yields both P3A(−) and P3A(+) transcripts [transcripts III and IV in (A)]. (C) P3A(−) and P3A(+) transcripts are specifically amplified by allele-specific RT-PCR. Nested RT-PCR products spanning p.R313W are digested by AciI, which cuts only the wild-type R313 fragment (lower band) and leaves the mutant W313 fragment undigested (upper band). The P3A(−) transcript exclusively arises from an allele with W313 [transcript III in (A)], whereas the P3A(+) transcript arises from both R313- and W313-alleles [transcripts II and IV in (A)].

IVS3-8G>A disrupts an ISS

The mutation enhances the recognition of exon P3A either by generating an ISE or by abrogating an ISS. To distinguish between the two mechanisms, we constructed a pRBG4 minigene carrying AChR α exons 2, 3, P3A, 4 and their intervening introns (Fig. 4A). Next, we engineered naturally occurring and artificial mutations (Fig. 4B), transfected each minigene into COS cells and then measured the absolute copy numbers of the P3A(−) and P3A(+) transcripts by real-time RT–PCR. The patient mutation, as well as IVS3-8G>C and IVS3-8G>T, markedly enhanced the recognition of exon P3A (Fig. 4C), indicating that IVS3-8G is an essential nucleotide of an ISS that silences the recognition of exon P3A.

Figure 4.

Constructs of minigenes and splicing analyses. (A) Partial genomic structure of CHRNA1 and two minigene constructs. pRBG4 minigene carries four native exons of CHRNA1, whereas pSPL3 minigene harbors only CHRNA1 exon P3A flanked by proprietary 5′ and 3′ exons of pSPL3. A broken line in the pRBG4 minigene indicates the excluded 589 bp segment. (B) The patient’s mutation and two artificial mutations that have been introduced into the pRBG4 minigene construct. (C) Real-time RT–PCR analysis of COS cells transfected with the indicated pRBG4 constructs. (D) Real-time RT–PCR analysis of HeLa cells transfected with wild-type and mutant pSPL3 constructs. For (C) and (D), bars represent mean and SD of three experiments.

Figure 4.

Constructs of minigenes and splicing analyses. (A) Partial genomic structure of CHRNA1 and two minigene constructs. pRBG4 minigene carries four native exons of CHRNA1, whereas pSPL3 minigene harbors only CHRNA1 exon P3A flanked by proprietary 5′ and 3′ exons of pSPL3. A broken line in the pRBG4 minigene indicates the excluded 589 bp segment. (B) The patient’s mutation and two artificial mutations that have been introduced into the pRBG4 minigene construct. (C) Real-time RT–PCR analysis of COS cells transfected with the indicated pRBG4 constructs. (D) Real-time RT–PCR analysis of HeLa cells transfected with wild-type and mutant pSPL3 constructs. For (C) and (D), bars represent mean and SD of three experiments.

We also inserted exon P3A and its flanking introns into the modified exon-trapping vector pSPL3 (Fig. 4A) (46) and found that these constructs recapitulated the aberrant splicing in HeLa cells (Fig. 4D). We initially constructed pRBG4 minigenes and analyzed splicing cis-elements in COS cells. We later constructed pSPL3 minigenes and analyzed the splicing trans-factors in HeLa cells, because the experiment-to-experiment variability was less than in COS cells.

hnRNP H binds to the ISS element

In order to identify a splicing trans-factor that binds to the ISS sequence, we incubated a nuclear extract of HEK293T cells with an RNA probe carrying the 3′ end of CHRNA1 intron 3 with or without IVS3-8G>A and used a scramble sequence as a negative control. An ∼55 kDa protein was purified with the wild-type RNA but not with the scramble RNA (Fig. 5A). The molecular weight of the purified protein suggested hnRNP H and this was confirmed by western blotting (Fig. 5B). We next depleted hnRNP H from nuclear extract of HEK293 cells using anti-hnRNP F/H monoclonal antibody (mAb) conjugated to a protein G column (Fig. 5C). The depletion of hnRNP H also indicated that the ∼55 kDa protein was hnRNP H (Fig. 5D). In addition, the binding affinity of hnRNP H for the IVS3-8G>A RNA probe was decreased compared with that for the wild-type RNA probe (Fig. 5B). Biacore affinity measurements revealed that the apparent dissociation constants (Kd) estimated from the surface plasmon resonance profiles were 5.61 × 10−8 M for wild type and 5.08 × 10−6 M for IVS3-8G>A (Fig. 5E), indicating an ∼100-fold attenuation of affinity for hnRNP H.

Figure 5.

hnRNP H binds to an ISS upstream of exon P3A. (A) Syproruby staining of affinity-purified HEK293T nuclear extract shows a 55 kDa band (arrowhead) in wild-type intron 3 (WT) and faintly in the intronic mutant probe (IVS3-8G>A). Scr represents scrambled RNA probe. (B) Western blotting with anti-hnRNP F/H antibody shows that the purified molecule is hnRNP H. NUC represents lanes loaded with nascent nuclear extract. U2AF65, which binds to the polypyrimidine tract, is immunostained as a positive control, and SRp55 as a negative control. Note decreased affinity of the mutant RNA probe (IVS3-8G>A) for hnRNP H. (C) Western blotting of hnRNP H-depleted (ΔhnRNP H) and mock-depleted (Cont) nuclear extracts. PTBP1 is immunostained as a control. (D) Mock-depleted (Cont) and hnRNP H-depleted (ΔhnRNP H) nuclear extracts are purified with the indicated affinity beads and stained with Syproruby. Note that staining of a 55 kDa band (arrowhead) representing hnRNP H is decreased by depletion of hnRNP H. (E) Biacore surface plasmon resonance analysis to quantify the interaction between hnRNP H and an ISS. The wild type-RNA (WT) or intronic mutant-RNA (IVS3-8G>A) is immobilized to a sensor chip SA and the response signals are monitored on injection of the indicated concentrations of recombinant hnRNP H protein.

Figure 5.

hnRNP H binds to an ISS upstream of exon P3A. (A) Syproruby staining of affinity-purified HEK293T nuclear extract shows a 55 kDa band (arrowhead) in wild-type intron 3 (WT) and faintly in the intronic mutant probe (IVS3-8G>A). Scr represents scrambled RNA probe. (B) Western blotting with anti-hnRNP F/H antibody shows that the purified molecule is hnRNP H. NUC represents lanes loaded with nascent nuclear extract. U2AF65, which binds to the polypyrimidine tract, is immunostained as a positive control, and SRp55 as a negative control. Note decreased affinity of the mutant RNA probe (IVS3-8G>A) for hnRNP H. (C) Western blotting of hnRNP H-depleted (ΔhnRNP H) and mock-depleted (Cont) nuclear extracts. PTBP1 is immunostained as a control. (D) Mock-depleted (Cont) and hnRNP H-depleted (ΔhnRNP H) nuclear extracts are purified with the indicated affinity beads and stained with Syproruby. Note that staining of a 55 kDa band (arrowhead) representing hnRNP H is decreased by depletion of hnRNP H. (E) Biacore surface plasmon resonance analysis to quantify the interaction between hnRNP H and an ISS. The wild type-RNA (WT) or intronic mutant-RNA (IVS3-8G>A) is immobilized to a sensor chip SA and the response signals are monitored on injection of the indicated concentrations of recombinant hnRNP H protein.

hnRNP H promotes skipping of exon P3A

We next examined whether hnRNP H indeed plays an essential role in alternative splicing of exon P3. As there are 50 potential hnRNP H-binding sites with the GGG core sequence in our pSPL3 minigene, we inserted the MS2 RNA stem-loop structure, which is recognized with high affinity by the MS2 coat protein (47), in place of the hnRNP H-binding site of our interest of CHRNA1 intron 3 (Fig. 6A). The native hnRNP H-binding site spanning IVS3-8G was eliminated in this construct. When the MS2-substituted construct was co-transfected with a fusion cDNA encoding the MS2 coat protein and hnRNP H, the skipping of exon P3A was enhanced (Fig. 6B). The introduction of hnRNP H alone to the MS2-substituted construct marginally enhanced the skipping of exon P3A, suggesting that another but weak hnRNP H-responsive element remained in this construct.

Figure 6.

hnRNP H silences splicing of exon P3A. (A) Schematic presentation of the introduction of the MS2 RNA hairpin in place of the ISS, and a fusion protein carrying the MS2-coat protein (inverted U shape) and hnRNP H (oval). (B) Real-time RT–PCR analysis of HeLa cells transfected with the MS2-introduced pSPL3 minigene along with one-tenth amount of an empty vector (Cont), hnRNP H cDNA (H), MS2-hnRNP H fusion cDNA (MS2H) or MS2-EGFP-NLS fusion cDNA (MS2EGFP). The rightmost column represents results obtained with the wild-type (WT) minigene alone. Note that placement of hnRNP H to the native ISS site induces skipping of exon P3A, but placement of MS2-EGFP-NLS does not. Overexpression of hnRNP H may have a marginal effect. (C) Western blotting with anti-hnRNP F/H antibody reveals efficient downregulation of nuclear (NUC) and cytoplasmic (CYT) hnRNP H by siRNA. HeLa cells are transfected with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). (D) Real-time RT–PCR analysis of HeLa cells transfected with the wild-type pSPL3 minigene along with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). Downregulation of hnRNP H induces inclusion of exon P3A for the wild-type construct, but not for the mutant construct. For (B) and (D), bars represent mean and SD of three experiments.

Figure 6.

hnRNP H silences splicing of exon P3A. (A) Schematic presentation of the introduction of the MS2 RNA hairpin in place of the ISS, and a fusion protein carrying the MS2-coat protein (inverted U shape) and hnRNP H (oval). (B) Real-time RT–PCR analysis of HeLa cells transfected with the MS2-introduced pSPL3 minigene along with one-tenth amount of an empty vector (Cont), hnRNP H cDNA (H), MS2-hnRNP H fusion cDNA (MS2H) or MS2-EGFP-NLS fusion cDNA (MS2EGFP). The rightmost column represents results obtained with the wild-type (WT) minigene alone. Note that placement of hnRNP H to the native ISS site induces skipping of exon P3A, but placement of MS2-EGFP-NLS does not. Overexpression of hnRNP H may have a marginal effect. (C) Western blotting with anti-hnRNP F/H antibody reveals efficient downregulation of nuclear (NUC) and cytoplasmic (CYT) hnRNP H by siRNA. HeLa cells are transfected with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). (D) Real-time RT–PCR analysis of HeLa cells transfected with the wild-type pSPL3 minigene along with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). Downregulation of hnRNP H induces inclusion of exon P3A for the wild-type construct, but not for the mutant construct. For (B) and (D), bars represent mean and SD of three experiments.

We next knocked down the endogenous hnRNP H protein by siRNA. We observed efficient downregulation of hnRNP H in cells transfected with hnRNP H-siRNA compared with those with control siRNA (Fig. 6C). The downregulation of hnRNP H markedly reduced the ratio of the P3A(−) transcript in the wild-type construct, but had no effect in the IVS3-8G>A construct (Fig. 6D). These results indicate that the hnRNP H binds to the ISS element close the 3′ end of intron 3 and abrogates the formation of spliceosome for exon P3A, and that IVS3-8G>A hinders the effect of hnRNP H.

hnRNP H promotes skipping of alternative exons in four other genes

We next asked whether hnRNP H similarly works on other genes. To this end, we analyzed the entire human genome and looked for the ratio of an hnRNP H-recognition motif of UGGG (48) among AGGG, CGGG and UGGG at positions −4 to −60. We found the ratio of UGGG gradually increases toward the 3′ end of introns (Fig. 7A). However, the AGGG motif tends to be absent close to the 3′ end of introns because an AG dinucleotide potentially generates a cryptic 3′ splice site in this region (49). Similarly, the CGGG motif includes a CpG dinucleotide, which is infrequent throughout the human genome. Thus, the gradual increase of UGGG toward the 3′ end of introns may or may not represent preferential appearance of the hnRNP H-responsive splicing cis-element in this region.

Figure 7.

Effects of hnRNP H on splicing of other genes. (A) Ratios of AGGG, CGGG and UGGG appearing at positions −60 to −4 of 190 972 introns according to the NCBI human RefSeq annotations. (B) RT-PCR of HeLa cells transfected with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). Closed arrowheads point to exon-included products, whereas open arrowheads point to exon-skipped products. Downregulation of hnRNP H promotes exon inclusion in all the indicated genes, as in the case of CHRNA1 exon P3A. Experiments are repeated three times or more and representative results are shown.

Figure 7.

Effects of hnRNP H on splicing of other genes. (A) Ratios of AGGG, CGGG and UGGG appearing at positions −60 to −4 of 190 972 introns according to the NCBI human RefSeq annotations. (B) RT-PCR of HeLa cells transfected with nonspecific control siRNA (Cont) or specific siRNA for hnRNP H (H). Closed arrowheads point to exon-included products, whereas open arrowheads point to exon-skipped products. Downregulation of hnRNP H promotes exon inclusion in all the indicated genes, as in the case of CHRNA1 exon P3A. Experiments are repeated three times or more and representative results are shown.

To identify hnRNP H-regulated exons in other genes, we first chose 24 alternative introns, in which the UGGG motif is located within the last 12 nucleotides of an intron. We then analyzed effects of siRNA-mediated knocking down of endogenous hnRNP H in HeLa cells on splicing of these 24 exons (Table 2). We found that six exons were alternatively spliced in HeLa cells. The downregulation of hnRNP H resulted in the inclusion of the downstream exon in four of the six genes (Fig. 7B). Although CHRNA1 is not expressed in HeLa cells, we have shown that it is regulated by hnRNP H in skeletal muscle. To summarize, in five of 24 exons, the UGGG motif in the upstream intron is an hnRNP H-responsive ISS.

Table 2.

Twenty-four human genes carrying UGGG in the last 12 nucleotides of an intron immediately upstream of an alternatively spliced exon

Gene symbol Intron length (bp) Sequence at the 3′ end of intron Effect of downregulation of hnRNP H 
FAS 4505 CCUUUUUUCCUUGGGCAG Exon inclusiona 
FRMPD2 4171 UACUCUUCUUUUGGGCAG No effect on alternative splicing 
TACC2 864 CACACUCCCUGUGGGCAG NA 
USH1C 3395 CCUCUUUUCCCUGGGUAG NA 
MBNL1 7768 GCAAUGCAUGAUGGGCAG No effect on alternative splicing 
GRIP1 6581 AUCACUUCAUAUGGGCAG Exon inclusiona 
MADD 1726 GCAUUGCAUCUGGGGUAG NA 
FCN3 444 AUCCCUUCUCUGGGGCAG NA 
SETD5 943 UUUUCCUCCUUGGGAUAG NA 
SLC26A1 365 UCGCACACCCUGGGCCAG NA 
TMEM107 83 UCCAAAGGCCUGGGCCAG NA 
GGT1 543 CUUACCCCGUGGGUGCAG NA 
MOSPD3 207 CUCCUACCUUGGGACCAG NA 
KCNQ2 9050 CGGUUCCCGUGGGAGCAG NA 
GCOM1 3419 UAUCUUUGUGGGCUACAG NA 
VPS13C 5020 CUGCCAUGUGGGAUGCAG Exon inclusiona 
LDHD 291 CGGUGUCUUGGGUUCCAG NA 
ZFYVE27 1394 CCGCACCUUGGGAGGCAG NA 
ARHGEF10L 5080 GGCCGCUUGGGGCUCCAG ND 
BCR 2385 UCUCCCUUGGGGCUGCAG NA 
NRCAM 2189 AUAUUGAUGGGGAAAAAG Exon inclusiona 
CHRNA1 1310 UUUUCUGUGGGUGGACAG NDb 
KCNIP2 228 UGUGCACUGGGAAAGAAG NA 
AGC1 908 CUCCCCUGGGGGUUGCAG NA 
Gene symbol Intron length (bp) Sequence at the 3′ end of intron Effect of downregulation of hnRNP H 
FAS 4505 CCUUUUUUCCUUGGGCAG Exon inclusiona 
FRMPD2 4171 UACUCUUCUUUUGGGCAG No effect on alternative splicing 
TACC2 864 CACACUCCCUGUGGGCAG NA 
USH1C 3395 CCUCUUUUCCCUGGGUAG NA 
MBNL1 7768 GCAAUGCAUGAUGGGCAG No effect on alternative splicing 
GRIP1 6581 AUCACUUCAUAUGGGCAG Exon inclusiona 
MADD 1726 GCAUUGCAUCUGGGGUAG NA 
FCN3 444 AUCCCUUCUCUGGGGCAG NA 
SETD5 943 UUUUCCUCCUUGGGAUAG NA 
SLC26A1 365 UCGCACACCCUGGGCCAG NA 
TMEM107 83 UCCAAAGGCCUGGGCCAG NA 
GGT1 543 CUUACCCCGUGGGUGCAG NA 
MOSPD3 207 CUCCUACCUUGGGACCAG NA 
KCNQ2 9050 CGGUUCCCGUGGGAGCAG NA 
GCOM1 3419 UAUCUUUGUGGGCUACAG NA 
VPS13C 5020 CUGCCAUGUGGGAUGCAG Exon inclusiona 
LDHD 291 CGGUGUCUUGGGUUCCAG NA 
ZFYVE27 1394 CCGCACCUUGGGAGGCAG NA 
ARHGEF10L 5080 GGCCGCUUGGGGCUCCAG ND 
BCR 2385 UCUCCCUUGGGGCUGCAG NA 
NRCAM 2189 AUAUUGAUGGGGAAAAAG Exon inclusiona 
CHRNA1 1310 UUUUCUGUGGGUGGACAG NDb 
KCNIP2 228 UGUGCACUGGGAAAGAAG NA 
AGC1 908 CUCCCCUGGGGGUUGCAG NA 

NA, no alternative splicing is detected in HeLa cells.

ND, no expression is detected in HeLa cells.

aRT-PCR results are as shown in Figure 7B.

bAlthough not expressed in HeLa cells, CHRNA1 exon P3A is regulated by hnRNP H in skeletal muscle, as demonstrated in this paper.

DISCUSSION

Exon P3A recognition-enhancing mutation in CHRNA1

We here report a CMS patient with EP AChR deficiency, who carries a low-expressor missense mutation (p.R313W) on one allele and an exon P3A recognition-enhancing mutation (IVS3-8G>A) on the second allele (Fig. 2). We initially assumed that IVS3-8G>A is a rare polymorphism because it does not affect the branch point, the polypyrimidine tract or any other known cis-acting splicing element, and because it is located upstream of the nonfunctional exon P3A. This mutation, however, disrupts an ISS and exclusively yields a nonfunctional P3A(+) transcript in the patient muscle (Fig. 3).

Intronic mutations that activate a cryptic exon have been reported in two genes. In MTRR encoding the methionine synthase reductase, an intronic T-to-C mutation activates a 140 bp cryptic exon and yields a frameshifting transcript (32). In ATM encoding the ataxia telangiectasia mutated, an intronic four nucleotide deletion abolishes the interaction of an intron-splicing processing element with U1 snRNP and activates a 65 bp cryptic exon (33). CHRNA1 exon P3A is not a cryptic exon, because it is included in 50% of normal muscle transcripts (14). Intronic mutations in these genes all unexpectedly affect pre-mRNA splicing. Deep intronic regions and introns flanking nonfunctional exons are seldom scrutinized in mutation analysis, and even if a mutation is detected in this region, it could be ignored as a rare polymorphism. Our studies and the two previous reports underscore the importance of scrutinizing deep intronic regions and nonfunctional exons, especially when searching for second mutations in autosomal recessive disorders.

hnRNP H silences recognition of exon P3A

The RNA affinity-purification assays revealed that the ISS element in intron 3 of CHRNA1 is recognized by hnRNP H, and that IVS3-8G>A disrupts its binding by mutating the UGGG sequence to UGGA (Fig. 2B). As described in the Introduction, hnRNP H either enhances (38–42) or silences (43) the recognition of an alternative exon in pre-mRNA splicing. Although the exact mechanisms by which hnRNP H acts either as a splicing activator or as a repressor remain elusive, Kralovicova and Vorechovsky (50) demonstrate a position-dependent enhancement and the suppression of splicing by hnRNP H. They show that an hnRNP H-binding site upstream of the branch point of DQB1 intron 3 enhances the recognition of exon 4, whereas a site close to the 5′ end of DQB1 intron 3 silences it. Our current studies demonstrate that the hnRNP H-binding site close to the 3′ splice site silences the recognition of the downstream exon in CHRNA1, FAS, GRIP1, VPS13C and NRCAM (Fig. 7B). In contrast to the DQB1 gene, the hnRNP H-binding sites in these genes are immediately downstream of the polypyrimidine tract, where the essential splicing factor U2AF65 binds. Thus, hnRNP H likely competes with the binding of U2AF65 and silences the recognition of the downstream exon. The hnRNP H-binding UGGG motif downstream of the polypyrimidine tract is an essential splicing cis-element in a subset of alternatively spliced genes, and mutations affecting it have likely been underestimated.

Why do humans and great apes have CHRNA1 exon P3A?

Exon P3A is unique to humans and great apes (12,13) and is included in 50% of CHRNA1 transcripts in human skeletal muscle (14). The mechanisms regulating the alternative splicing of exon P3A have been totally unknown. One hypothesis is that the expression of hnRNP H is upregulated exclusively at subsynaptic nuclei and silences the recognition of exon P3A to achieve synapse-specific expression of the functional AChR α subunit, whereas hnRNP H is downregulated at the extrasynaptic nuclei to yield the nonfunctional α subunit harboring exon P3A. In mammals, AChR transcription is upregulated at the subsynaptic nuclei by a transcriptional cis-element N-box (51–53) and downregulated at the extrasynaptic nuclei by N-box (51,54) and an E-box (55,56). Alternative splicing of CHRNA1 exon P3A might have evolved to achieve additional regulation of synapse-specific expression of AChR. Further studies are required to elucidate the physiological significance of exon P3A.

MATERIALS AND METHODS

Muscle biopsies

Intercostal muscle specimen was obtained intact from origin to insertion from the patient and control subjects without muscle disease undergoing thoracic surgery. All human studies were in accord with the guidelines of the Institutional Review Board of the Mayo Clinic.

EP studies

AChR and acetylcholinesterase were detected in cryostat sections by two-color fluorescence (57). EPs were localized for electron microscopy and analyzed by established methods (58). Peroxidase-labeled α-bgt was used for the ultrastructural localization of AChR (59). The number of AChRs per EP was measured with [125I]α-bgt (44). MEPP and EPP recordings were performed, and estimates of the number of transmitter quanta released by nerve impulse were obtained as described elsewhere (44,60). Patch-clamp recordings from AChRs expressed in HEK cells were obtained in the cell-attached mode as described previously (61).

Mutation analysis

We directly sequenced AChR α, β, δ and ε subunit genes using genomic DNA isolated from muscle (2). Genomic DNA from single alleles was extracted from leucocytes using the ‘conversion’ method (45) by GMP Genetics. For the analysis of family members, we traced IVS3-8G>A by direct sequencing; for p.R313W (c.937C>T), we used restriction analysis of PCR products with AciI (New England Biolabs). p.R313W resulted in the loss of an AciI site. We screened 200 normal alleles of unrelated patients for the two identified mutations by allele-specific PCR. The allele-specific PCR primers for p.R313W were 5′-AGGACTCAGGACTTCCACATA-3′ and 5′-AGTGTCGATAAAAACCTTaCA-3′, where ‘a’ represents a deliberately introduced mismatched nucleotide to avoid misannealing of the mutant primer to the wild-type allele. The boldfaced A is complementary to the mutant nucleotide. The allele-specific primers for IVS3-8G>A were 5′-TCTTTCTCCTTTTCTGTtGA-3′ and 5′-GACGTGCTCACACTGTTTCA-3′, where ‘t’ represents an artificial mismatch and the boldfaced A is complementary to the mutant nucleotide.

Allele-specific RT-PCR

For allele-specific RT-PCR, mRNA was isolated from muscle using the Micro-FastTrack 2.0 mRNA isolation kit (Invitrogen). cDNA was synthesized using random hexamers and the Superscript II reverse transcriptase (Invitrogen) according to the manufacturer’s recommendations.

Transcripts arising from wild-type allele harboring R313 was amplified with 5′-AGTGTCGATAAAAACCTTaCG-3′, and that arising from the mutant allele harboring W313 was amplified using the primer, 5′-AGTGTCGATAAAAACCTTaCA-3′, where the lower case ‘a’ represents a mismatched nucleotide and the boldfaced G and A are complementary to the wild-type and mutant nucleotides, respectively. The opposite primer was 5′-GTCCTGGGCTCCGAACAT-3′ in exon 2, and the expected allele-specific RT-PCR product size was 966 bp. The amplicon was then used as a template for nested RT-PCR using primers, 5′-AACCAATGTGCGTCTGAAAC-3′ in exon 3 and 5′-TTTTTCACACCGCCATAGTC-3′ in exon 4, which spanned exon P3A. With these nested primers, a transcript lacking exon P3A would yield a 78 bp fragment, whereas a transcript harboring exon P3A would give rise to a 153 bp fragment.

Next, transcripts lacking and harboring exon P3A were specifically amplified using primers, 5′-AATGTGCGTCTGAAACAGCAA-3′ and 5′-AATGTGCGTCTGAAACAGGGT-3′, in which the boldfaced nucleotides anneal to the 5′ end of exons 4 and P3A, respectively. The opposite primer was 5′-TCTCTGCTCTGGTAGGTTCC-3′ in the 3′ untranslated region, and the expected allele-specific RT-PCR product size was 1209 bp. The RT-PCR amplicon was used as a template for nested RT-PCR using primers, 5′-TCACTGTCATCGTCATCAACA-3′ in exon 7 and 5′-GCCGCCGCATTGTTAGACTC-3′ at the boundary of exons 8 and 9 to amplify a 319 bp fragment which spanned exon 7 containing p.R313W. The nested RT-PCR product was then digested with AciI restriction enzyme (New England Biolabs). A PCR amplicon carrying wild-type R313 was expected to give rise to 247, 36, 29, 4 and 3 bp fragments, whereas an amplicon harboring mutant W313 was expected to yield 283, 29, 4 and 3 bp fragments.

Mapping of branch points of intron 3

We mapped the branch point of intron 3 by the lariat RT-PCR method (62). We synthesized cDNA from total RNA extracted from COS cells transfected with the wild-type CHRNA1 minigene using a CHRNA1-specific primer, 5′-TATCCCTGTTACCCATATTGA-3′ (IVS3+92 to IVS3+72) and Superscript II (Invitrogen). A fragment spanning the branch point was amplified by primers, 5′-ATGATGTTGCCTGCTTGAG-3′ (IVS3-65 to IVS3-46) and 5′-GCCCAGGTTTAATTTCAGTG-3′ (IVS3+61 to IVS3+42), cloned into pGEM-T (Promega). We sequenced four clones.

Construction and expression of wild-type and mutant AChRs

Sources of human α, β, δ and ε subunit cDNAs were as described previously (63–65). We amplified the α subunit cDNA containing P3A from control skeletal muscle by RT-PCR [pRBG4-α-P3A(+)]. All cDNAs were subcloned into the CMV-based expression vector pRBG4 (66) for expression in HEK cells. p.R313W was engineered into pRBG4 using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). The presence of the desired mutation and the absence of unwanted artifacts were confirmed by sequencing the entire inserts. HEK cells in a 10 cm2 dish were transfected with 5.4 µg of wild-type or mutant α subunit cDNA along with 2.7 µg each of wild-type β, δ and ε subunit cDNAs using 40.8 µl FuGENE-6 (Roche) according to the manufacturer’s recommendations. To correct for cell numbers, transfection efficiency and harvesting efficiency, we added 125 ng of the Renilla luciferase vector, phRL-TK (Promega), to each dish. The total number of [125I]α-bgt-binding sites expressed on the surface of transfected HEK cells was determined as described previously (2). Renilla luciferase activities were measured using the Renilla luciferase assay system (Promega) in a Turner Designs TD-20/20 Luminometer.

Construction of pRBG4 and pSPL3 minigenes for splicing analysis

We constructed a minigene spanning exons 2 to 4 of CHRNA1 to analyze pre-mRNA splicing in transfected COS cells (Fig. 4A). First, a 500 bp genomic fragment spanning nucleotide 18′ in exon 2 to IVS3+214 was amplified by PCR using Pfu DNA polymerase (Stratagene) from control genomic DNA. The 5′ ends of the forward and reverse primers carried XbaI and KpnI sites, respectively. Secondly, an 894 bp genomic fragment spanning IVS3-507 to nucleotide 36′ in exon 4 was similarly amplified by PCR. The 5′ ends of the forward and reverse primers carried KpnI and ClaI sites, respectively. The two PCR fragments were ligated using the KpnI site and then into the XbaI and ClaI sites of the CMV-based expression vector pRBG4 (66). Compared with wild-type CHRNA1, the minigene lacked a 589 bp segment in the middle of intron 3 (Fig. 4A).

Because nonsense-mediated mRNA decay potentially destroys a nonsense-containing transcript even for minigenes (67), we added the Kozak consensus sequence (68), 5′-CCACCATG-3′, at the 5′ end of the insert, to assure inframe translation of the minigene. The predicted open-reading frame was the same as that of CHRNA1. With this open-reading frame, a TAA stop codon should appear 11 codons downstream of the CHRNA1 insert.

We also constructed a minigene spanning exon P3A in the modified exon-trapping vector, pSPL3 (a discontinued product of Invitrogen), in which we introduced the CMV promoter in place of the SV40 promoter and also introduced NotI and PacI restriction sites in the intron (46). We amplified a 288 bp genomic fragment spanning IVS3-123 to IVS3A+90 by PCR. The PCR fragment was ligated into the NotI and PacI sites of the modified pSPL3 vector (Fig. 4A).

Naturally occurring and artificial mutations were engineered into the pRBG4 and pSPL3 minigenes using the QuikChange Site-Directed Mutagenesis Kit. The absence of artifacts was confirmed by sequencing the entire inserts.

Construction of hnRNP H expression vector with or without the MS2 coat protein and CHRNA1 minigene carrying the MS2 RNA sequence

The human HNRPH1 cDNA encoding hnRNP H (Open Biosystems) was subcloned into the CMV-based expression vector, pcDNA3.1/V5-His TOPO (Invitrogen). We obtained the MS2 coat protein cDNA by RT-PCR using MS2RNA (Roche) as a template. The 5′ ends of the forward and reverse primers carried XhoI and XbaI sites, respectively. The PCR fragment was ligated upstream of HNRPH1 cDNA described before to make a fusion cDNA.

We substituted the MS2 RNA sequence of ATGCACGATCACGGCATAA for the native GGGTGGA sequence from IVS3-10 to IVS3-4 of the pSPL3 minigene using the QuikChange Site-Directed Mutagenesis Kit. We thus eliminated a putative hnRNP H-binding sequence of GGG from the CHRNA1 minigene and inserted the MS2 RNA target sequence.

As a control, we constructed a fusion cDNA encoding EGFP, the MS2 coat protein, and three copies of the nuclear localization signal (NLS) of the simian virus 40 large T–antigen. First, we amplified the MS2 coat protein cDNA with a forward primer carrying an XhoI site at the 5′ end and with a reverse primer carrying three repeats of the NLS and a BamHI site at the 5′ end. The PCR fragment was ligated into the 3′ end of the EGFP cDNA in the pLEGFP-C1 vector (Clontech). Then, we amplified the EGFP-MS2-NLS fusion cDNA with PCR primers carrying HindIII and BamHI sites at the respective 5′ ends. We ligated the PCR fragment into the pCDNA3.1(+) vector (Invitrogen), transfected HEK293 cells and confirmed that the EGFP-MS2-NLS fusion protein was expressed in the nuclear fraction.

Transfection of minigene and isolation of total RNA

COS or HeLa cells were seeded into a six-well plate. After 24 h, cells were transfected with 1 µg of each minigene construct using 3 µl of the FuGENE 6 transfection reagent (Roche) according to the manufacturer’s recommendations. Fresh medium was added 16 h after transfection. Total RNA was extracted 40 h after transfection using the RNeasy Mini Kit (Qiagen) including the DNase I treatment according to the manufacturer’s instructions.

One-third of the isolated total RNA was used for cDNA synthesis using an oligo-dT primer and the Superscript II reverse transcriptase (Invitrogen). One-fortieth of the synthesized cDNA was used for each real-time RT–PCR in 20 µl.

Real-time RT–PCR for quantification of P3A(−) and P3A(+) transcripts

We estimated the absolute copy numbers of the P3A(−) and P3A(+) transcripts in muscle and in transfected cell lines with the LightCycler 1.2 real-time PCR instrument and the LightCycler FastStart DNA Master SYBR Green I Kit (Roche), or with the Mx3000P QPCR System and the Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen). We then calculated the ratio of P3A(−) transcript to total transcripts using the following equation:  

formula
.

We constructed and used five cDNA clones as standards. First, as a standard for the quantification of the P3A(−) transcript of the pRBG4 minigene transfected into COS cells, we PCR-amplified the P3A(−) transcript using cDNA from transfected COS cells with 5′-AATGTGCGTCTGAAACAGCAA-3′ spanning exons 3 and 4, and 5′-TGTGAAATTTGTGATGCTATTG-3′ in the polyadenylation signal site of pRBG4. We then cloned the amplicon into pGEM-T (Promega). We employed the same primer pair for the quantification of the P3A(−) transcript.

Secondly, as a standard for the quantification of the P3A(−) transcript of the pSPL3 minigene transfected into HeLa cells, we PCR-amplified the P3A(−) transcript using cDNA from transfected HeLa cells with 5′-TCTGAGTCACCTGGACAACC-3′ on the native 5′ exon of pSPL3 and 5′-GGGAGATCTCCAGGTTGCT-3′ spanning the native 5′ and 3′ exons of pSPL3. We employed the same primer pair for the quantification of the P3A(−) transcript.

Thirdly, as a standard for the quantification of the P3A(−) transcript in muscle, we used the pRBG4-α expression vector described earlier. For the quantification of the P3A(−) transcript, we used 5′-AATGTGCGTCTGAAACAGCAA-3′ spanning exons 3 and 4 and 5′-ACGTGATGTGGCCAGTGTACTG-3′ in exon 5.

Next, as a standard for the quantification of the P3A(+) transcript both in pRBG4-transfected COS cells and in muscle, we used the pRBG4-α-P3A(+) expression vector described earlier. For the quantification of the P3A(+) transcript, we used 5′-ACCACCGCCAGGTCGTGG-3′ in exon 2 and 5′-CTCATTCTGCAGATGAGAAAAC-3′ in exon P3A.

Finally, as a standard for the quantification of the P3A(+) transcript of the pSPL3 minigene transfected into HeLa cells, we PCR-amplified the P3A(+) transcript using cDNA from transfected HeLa cells with 5′-TCTGAGTCACCTGGACAACC-3′ on the native 5′ exon of pSPL3 and 5′-GGGAGATCTCCAGGTCTCA-3′ spanning the native 3′ exon of pSPL3 and exon P3A. We employed the same primer pair for the quantification of the P3A(−) transcript.

For each P3A(−)- or P3A(+)-specific primer pair described herein, we confirmed that no primer pair amplified the undesired P3A(+) or P3A(−) transcript. We also sequenced RT-PCR products to ensure that cryptic splice sites were not activated in any minigene constructs. After concentrations of the five standard cDNA clones were calibrated by amplifying a shared segment using the real-time PCR, we made a serial dilution of 1 × 108 to 1 × 103 copies of each plasmid to make a standard curve. We also confirmed that amplification efficiencies of the cDNA clones were similar to those of cDNA samples by simultaneously amplifying 1×, 10× and 100× diluted cDNA samples.

siRNAs to knockdown hnRNP H

For the downregulation of hnRNP H, we purchased the HP Validated siRNA (SI02654799) as well as AllStar Negative Control siRNA (1027281) from Qiagen. HeLa cells were plated 24 h before transfection on 12-well plates. The transfection reagent included 300 ng of the pSPL3 minigene, 50 pmol of siRNA and 1 µl of Lipofectamine 2000 in 100 µl DMEM.

RT-PCR spanning alternative exons of GRIP1, FAS, VPS13C and NRCAM

We isolated total RNA from HeLa cells transfected with control siRNA or hnRNP H siRNA and synthesized cDNA as described before. We employed the following primers for RT-PCR: GRIP1 sense primer, 5′-ACCAGCATGGAGTACTGTACAC-3′; GRIP1 antisense primer, 5′-CCTGCCCAGCCAATCCTA-3′; FAS sense, 5′-GGACCCTCCTACCTCTGGTTCTTAC-3′; FAS antisense, 5′-GCACTTGGTGTTGCTGGTGAG-3′; VPS13C sense, 5′-CCCTTCAAAAAGCAGCA GAA-3′; VPS13C antisense, 5′-CCCAGTGTGACACCAAATGA-3′; NRCAM sense, 5′-CCCTGATTCTCTTCCTGTGC-3′; NRCAM antisense, 5′-CCTTTGGCTTCACACTGGAT-3′.

Preparation of biotinylated RNA and RNA affinity purification assay

We synthesized biotinylated RNAs using the RiboMAX System (Promega) from a PCR-amplified fragment. The forward primer was 5′-TAATACGACTCACTATAGGGAGAC AGG-3′, where the italicized is T7 promoter and the boldfaced is for annealing to the reverse primer. The three reverse primers were: wild-type, 5′-CATGTCACCCTGTCCACCC ACAGAAAAGGAGCCTGTCTC-3′, where the boldfaced is for annealing to the forward primer; IVS3-8G>A, 5′-CATGTCACCCTGTCCATCCACAGAAAAGGAGCCTGTCTC-3′; scramble, 5′-GATGTCAGACTATCGACGTTCAGGATCGTCACCTGTCTC-3′. In each 20 µl reaction, 2 µg DNA template was transcribed by T7 polymerase in the presence of 7.5 mm UTP, 7.5 mm ATP, 7.5 mm GTP, 4.5 mm CTP and 3.0 mm Biotin-14-CTP (Invitrogen). Biotinylated RNAs (∼10 µg) were mixed with 10 µl of nuclear extract of HEK293 cells containing 1 mm DTT, 5 µg/μl yeast tRNA and RNase inhibitor (Toyobo) in the binding buffer (20 mm HEPES, pH7.8, 50 mm KCl, 3 mm MgCl2, 500 µm EDTA and 0.05% Triton X). After incubation on ice for 30 min, we added 25 µl of packed streptavidin-sepharose (GE Healthcare) and 25 µl of the binding buffer. The reaction mixture was shaken at 4°C for another 30 min and washed with the binding buffer four times. The bound proteins were resolved on a 10% SDS–polyacrylamide gel and the gel was stained with Syproruby (Invitrogen) according to the manufacturer’s instructions or analyzed by western blotting as described previously (69). The mouse mAb against hnRNP F/H (Santa Cruz Biotechnology, Inc.) is documented to recognize hnRNP H and F. Although recognition of hnRNP H′ is not documented by the manufacturer, the antibody likely recognizes hnRNP H′, because hnRNP H and H′ have only five mismatches among 449 amino acids.

Depletion of hnRNP H from nuclear extract

Depletion of hnRNP H from HEK293 cell nuclear extract was performed using Protein G HP spin trap (GE Healthcare) according to the manufacturer’s instructions. Anti-hnRNP F/H mAb (300 µl) or isotype-matched control antibody was conjugated to the Protein G column. The depletion of hnRNP H was confirmed by western blotting with anti-hnRNP F/H antibody.

Surface plasmon resonance analysis

We measured the binding affinities of hnRNP H for the wild-type and mutant CHRNA1 ISS motifs by surface plasmon resonance using Biacore 3000 (GE Healthcare). RNA probes were chemically synthesized with the 5′-biotin tag (Hokkaido System Science) and immobilized onto streptavidin-coated sensor chips (GE Healthcare) at a concentration of 200 resonance units by injecting 0.1 pmol/µl of the RNA probe in HBS-EP (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mm EDTA and 0.005% P20). The RNA probe sequences were as follows: wild-type, 5′-UUUCUCCUUUUCUGUGGGUGGACAGGGUGACAUGGUA-3′; IVS3-8G>A, 5′-UUUCUCCUUUUCUGUGGAUGGACAGGGUGACAUGGUA-3′.

Recombinant hnRNP H was expressed using the Bac-to-Bac baculovirus expression system (Invitrogen) according to the manufacturer’s instructions. The human hnRNP H cDNA in pcDNA3.1/V5-His TOPO was digested with BamHI and PmeI sites and ligated into the BamHI and StuI sites of the pFASTBAC1 vector (Invitrogen). Sf9 cells were infected with recombinant baculovirus, harvested after 48 h and lysed by sonication in PBS. His-tagged hnRNP H protein was purified using the His Trap HP column (GE Healthcare) according to the manufacturer’s instructions.

While the surface plasmon resonance signals were monitored, we injected recombinant hnRNP H over the chip at a flow rate of 20 µl/min for 3 min and then washed with HBS-EP.

In silico analysis of human genome

We extracted 190 972 introns from the entire human genome by mapping the ‘mRNA’ tags of the NCBI RefSeq human genome database build 36.3. We then looked for AGGG, CGGG and UGGG motifs close to the 3′ ends of the extracted introns. We developed Perl programs and performed the analysis using the PrimePower HPC2500/Solaris 9 supercomputer (Fujitsu).

FUNDING

This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan to A.M., M.I., T.M., and K.O., Grants-in-Aid from the Ministry of Health, Labor, and Welfare of Japan to K.O., the National Institutes of Health grant NS6277 to A.G.E. and by the Muscular Dystrophy Association research grant to A.G.E.

ACKNOWLEDGEMENTS

We are grateful to Jun Shinmi and Keiko Itano for technical assistance.

Conflict of Interest statement. None declared.

REFERENCES

1
Engel
A.G.
Ohno
K.
Sine
S.M.
Sleuthing molecular targets for neurological diseases at the neuromuscular junction
Nat. Rev. Neurosci.
 , 
2003
, vol. 
4
 (pg. 
339
-
352
)
2
Engel
A.G.
Ohno
K.
Bouzat
C.
Sine
S.M.
Griggs
R.C.
End-plate acetylcholine receptor deficiency due to nonsense mutations in the epsilon subunit
Ann. Neurol.
 , 
1996
, vol. 
40
 (pg. 
810
-
817
)
3
Ohno
K.
Quiram
P.A.
Milone
M.
Wang
H.L.
Harper
M.C.
Pruitt
J.N.
II
Brengman
J.M.
Pao
L.
Fischbeck
K.H.
Crawford
T.O.
, et al.  . 
Congenital myasthenic syndromes due to heteroallelic nonsense/missense mutations in the acetylcholine receptor epsilon subunit gene: identification and functional characterization of six new mutations
Hum. Mol. Genet.
 , 
1997
, vol. 
6
 (pg. 
753
-
766
)
4
Milone
M.
Wang
H.L.
Ohno
K.
Prince
R.
Fukudome
T.
Shen
X.M.
Brengman
J.M.
Griggs
R.C.
Sine
S.M.
Engel
A.G.
Mode switching kinetics produced by a naturally occurring mutation in the cytoplasmic loop of the human acetylcholine receptor epsilon subunit
Neuron
 , 
1998
, vol. 
20
 (pg. 
575
-
588
)
5
Croxen
R.
Young
C.
Slater
C.
Haslam
S.
Brydson
M.
Vincent
A.
Beeson
D.
End-plate gamma- and epsilon-subunit mRNA levels in AChR deficiency syndrome due to epsilon-subunit null mutations
Brain
 , 
2001
, vol. 
124
 (pg. 
1362
-
1372
)
6
Milone
M.
Shen
X.-M.
Ohno
K.
Harper
C.M.
Fukudome
T.
Stilling
G.
Brengman
J.M.
Engel
A.G.
Unusual congenital myasthenic syndrome with endplate AChR deficiency caused by alpha subunit mutations and a remitting-relapsing clinical course
Neurology
 , 
1999
, vol. 
51
 
Suppl. 2
pg. 
A185
  
[Abstract]
7
Maselli
R.A.
Chen
M.
Chen
T.-Y.
Kong
D.
Constance
B.
McDonald
C.
Fairclough
R.H.
Gomez
C.M.
Wollmann
R.L.
A heteroallelic nonsense/missense mutation in the acetylcholine receptor (AChR) alpha-subunit causing severe endplate AChR deficiency
Neurology
 , 
2002
, vol. 
58
 
Suppl. 3
pg. 
A230
  
[Abstract]
8
Quiram
P.A.
Ohno
K.
Milone
M.
Patterson
M.C.
Pruitt
N.J.
Brengman
J.M.
Sine
S.M.
Engel
A.G.
Mutation causing congenital myasthenia reveals acetylcholine receptor beta/delta subunit interaction essential for assembly
J. Clin. Invest.
 , 
1999
, vol. 
104
 (pg. 
1403
-
1410
)
9
Shen
X.M.
Ohno
K.
Fukudome
T.
Tsujino
A.
Brengman
J.M.
De Vivo
D.C.
Packer
R.J.
Engel
A.G.
Congenital myasthenic syndrome caused by low-expressor fast-channel AChR delta subunit mutation
Neurology
 , 
2002
, vol. 
59
 (pg. 
1881
-
1888
)
10
Beeson
D.
Morris
A.
Vincent
A.
Newsom-Davis
J.
The human muscle nicotinic acetylcholine receptor alpha-subunit exist as two isoforms: a novel exon
EMBO J.
 , 
1990
, vol. 
9
 (pg. 
2101
-
2106
)
11
Newland
C.F.
Beeson
D.
Vincent
A.
Newsom-Davis
J.
Functional and non-functional isoforms of the human muscle acetylcholine receptor
J. Physiol.
 , 
1995
, vol. 
489
 (pg. 
767
-
778
)
12
Krull
M.
Petrusma
M.
Makalowski
W.
Brosius
J.
Schmitz
J.
Functional persistence of exonized mammalian-wide interspersed repeat elements (MIRs)
Genome Res.
 , 
2007
, vol. 
17
 (pg. 
1139
-
1145
)
13
MacLennan
C.
Beeson
D.
Vincent
A.
Newsom-Davis
J.
Human nicotinic acetylcholine receptor alpha-subunit isoforms: origins and expression
Nucleic Acids Res.
 , 
1993
, vol. 
21
 (pg. 
5463
-
5467
)
14
Guyon
T.
Levasseur
P.
Truffault
F.
Cottin
C.
Gaud
C.
Berrih-Aknin
S.
Regulation of acetylcholine receptor alpha subunit variants in human myasthenia gravis. Quantification of steady-state levels of messenger RNA in muscle biopsy using the polymerase chain reaction
J. Clin. Invest.
 , 
1994
, vol. 
94
 (pg. 
16
-
24
)
15
Andreetta
F.
Baggi
F.
Antozzi
C.
Torchiana
E.
Bernasconi
P.
Simoncini
O.
Cornelio
F.
Mantegazza
R.
Acetylcholine receptor alpha-subunit isoforms are differentially expressed in thymuses from myasthenic patients
Am. J. Pathol.
 , 
1997
, vol. 
150
 (pg. 
341
-
348
)
16
Wilisch
A.
Gutsche
S.
Hoffacker
V.
Schultz
A.
Tzartos
S.
Nix
W.
Schalke
B.
Schneider
C.
Muller-Hermelink
H.K.
Marx
A.
Association of acetylcholine receptor alpha-subunit gene expression in mixed thymoma with myasthenia gravis
Neurology
 , 
1999
, vol. 
52
 (pg. 
1460
-
1466
)
17
Johnson
J.M.
Castle
J.
Garrett-Engele
P.
Kan
Z.
Loerch
P.M.
Armour
C.D.
Santos
R.
Schadt
E.E.
Stoughton
R.
Shoemaker
D.D.
Genome-wide survey of human alternative pre-mRNA splicing with exon junction microarrays
Science
 , 
2003
, vol. 
302
 (pg. 
2141
-
2144
)
18
Smith
C.W.
Valcarcel
J.
Alternative pre-mRNA splicing: the logic of combinatorial control
Trends Biochem. Sci.
 , 
2000
, vol. 
25
 (pg. 
381
-
388
)
19
Caceres
J.F.
Kornblihtt
A.R.
Alternative splicing: multiple control mechanisms and involvement in human disease
Trends Genet.
 , 
2002
, vol. 
18
 (pg. 
186
-
193
)
20
Ozsahin
H.
Arredondo-Vega
F.X.
Santisteban
I.
Fuhrer
H.
Tuchschmid
P.
Jochum
W.
Aguzzi
A.
Lederman
H.M.
Fleischman
A.
Winkelstein
J.A.
, et al.  . 
Adenosine deaminase deficiency in adults
Blood
 , 
1997
, vol. 
89
 (pg. 
2849
-
2855
)
21
Vuillaumier-Barrot
S.
Barnier
A.
Cuer
M.
Durand
G.
Grandchamp
B.
Seta
N.
Characterization of the 415G>A (E139K) PMM2 mutation in carbohydrate-deficient glycoprotein syndrome type Ia disrupting a splicing enhancer resulting in exon 5 skipping
Hum. Mutat.
 , 
1999
, vol. 
14
 (pg. 
543
-
544
)
22
D’Souza
I.
Poorkaj
P.
Hong
M.
Nochlin
D.
Lee
V.M.
Bird
T.D.
Schellenberg
G.D.
Missense and silent tau gene mutations cause frontotemporal dementia with parkinsonism-chromosome 17 type, by affecting multiple alternative RNA splicing regulatory elements
Proc. Natl Acad. Sci. USA
 , 
1999
, vol. 
96
 (pg. 
5598
-
5603
)
23
Cardozo
A.K.
De Meirleir
L.
Liebaers
I.
Lissens
W.
Analysis of exonic mutations leading to exon skipping in patients with pyruvate dehydrogenase E1 alpha deficiency
Pediatr. Res.
 , 
2000
, vol. 
48
 (pg. 
748
-
753
)
24
Ars
E.
Serra
E.
Garcia
J.
Kruyer
H.
Gaona
A.
Lazaro
C.
Estivill
X.
Mutations affecting mRNA splicing are the most common molecular defects in patients with neurofibromatosis type 1
Hum. Mol. Genet.
 , 
2000
, vol. 
9
 (pg. 
237
-
247
)
25
Lim
S.R.
Hertel
K.J.
Modulation of survival motor neuron pre-mRNA splicing by inhibition of alternative 3′ splice site pairing
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
45476
-
45483
)
26
Fackenthal
J.D.
Cartegni
L.
Krainer
A.R.
Olopade
O.I.
BRCA2 T2722R is a deleterious allele that causes exon skipping
Am. J. Hum. Genet.
 , 
2002
, vol. 
71
 (pg. 
625
-
631
)
27
Cartegni
L.
Krainer
A.R.
Disruption of an SF2/ASF-dependent exonic splicing enhancer in SMN2 causes spinal muscular atrophy in the absence of SMN1
Nat. Genet.
 , 
2002
, vol. 
30
 (pg. 
377
-
384
)
28
Moseley
C.T.
Mullis
P.E.
Prince
M.A.
Phillips
J.A.
III
An exon splice enhancer mutation causes autosomal dominant GH deficiency
J. Clin. Endocrinol. Metab.
 , 
2002
, vol. 
87
 (pg. 
847
-
852
)
29
Kimbell
L.M.
Ohno
K.
Engel
A.G.
Rotundo
R.L.
C-terminal and heparin-binding domains of collagenic tail subunit are both essential for anchoring acetylcholinesterase at the synapse
J. Biol. Chem.
 , 
2004
, vol. 
279
 (pg. 
10997
-
11005
)
30
Wang
G.S.
Cooper
T.A.
Splicing in disease: disruption of the splicing code and the decoding machinery
Nat. Rev. Genet.
 , 
2007
, vol. 
8
 (pg. 
749
-
761
)
31
Nielsen
K.B.
Sorensen
S.
Cartegni
L.
Corydon
T.J.
Doktor
T.K.
Schroeder
L.D.
Reinert
L.S.
Elpeleg
O.
Krainer
A.R.
Gregersen
N.
, et al.  . 
Seemingly neutral polymorphic variants may confer immunity to splicing-inactivating mutations: a synonymous SNP in exon 5 of MCAD protects from deleterious mutations in a flanking exonic splicing enhancer
Am. J. Hum. Genet.
 , 
2007
, vol. 
80
 (pg. 
416
-
432
)
32
Wilson
A.
Leclerc
D.
Rosenblatt
D.S.
Gravel
R.A.
Molecular basis for methionine synthase reductase deficiency in patients belonging to the cblE complementation group of disorders in folate/cobalamin metabolism
Hum. Mol. Genet.
 , 
1999
, vol. 
8
 (pg. 
2009
-
2016
)
33
Pagani
F.
Buratti
E.
Stuani
C.
Bendix
R.
Dork
T.
Baralle
F.E.
A new type of mutation causes a splicing defect in ATM
Nat. Genet.
 , 
2002
, vol. 
30
 (pg. 
426
-
429
)
34
Kashima
T.
Manley
J.L.
A negative element in SMN2 exon 7 inhibits splicing in spinal muscular atrophy
Nat. Genet.
 , 
2003
, vol. 
34
 (pg. 
460
-
463
)
35
Disset
A.
Bourgeois
C.F.
Benmalek
N.
Claustres
M.
Stevenin
J.
Tuffery-Giraud
S.
An exon skipping-associated nonsense mutation in the dystrophin gene uncovers a complex interplay between multiple antagonistic splicing elements
Hum. Mol. Genet.
 , 
2006
, vol. 
15
 (pg. 
999
-
1013
)
36
Wu
J.Y.
Kar
A.
Kuo
D.
Yu
B.
Havlioglu
N.
SRp54 (SFRS11), a regulator for tau exon 10 alternative splicing identified by an expression cloning strategy
Mol. Cell. Biol.
 , 
2006
, vol. 
26
 (pg. 
6739
-
6747
)
37
Caputi
M.
Zahler
A.M.
Determination of the RNA binding specificity of the heterogeneous nuclear ribonucleoprotein (hnRNP) H/H′/F/2H9 family
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
43850
-
43859
)
38
Chou
M.Y.
Rooke
N.
Turck
C.W.
Black
D.L.
hnRNP H is a component of a splicing enhancer complex that activates a c-src alternative exon in neuronal cells
Mol. Cell. Biol.
 , 
1999
, vol. 
19
 (pg. 
69
-
77
)
39
Han
K.
Yeo
G.
An
P.
Burge
C.B.
Grabowski
P.J.
A combinatorial code for splicing silencing: UAGG and GGGG motifs
PLoS Biol.
 , 
2005
, vol. 
3
 pg. 
e158
 
40
Zahler
A.M.
Damgaard
C.K.
Kjems
J.
Caputi
M.
SC35 and heterogeneous nuclear ribonucleoprotein A/B proteins bind to a juxtaposed exonic splicing enhancer/exonic splicing silencer element to regulate HIV-1 tat exon 2 splicing
J. Biol. Chem.
 , 
2004
, vol. 
279
 (pg. 
10077
-
10084
)
41
Caputi
M.
Zahler
A.M.
SR proteins and hnRNP H regulate the splicing of the HIV-1 tev-specific exon 6D
EMBO J.
 , 
2002
, vol. 
21
 (pg. 
845
-
855
)
42
Wang
E.
Dimova
N.
Cambi
F.
PLP/DM20 ratio is regulated by hnRNPH and F and a novel G-rich enhancer in oligodendrocytes
Nucleic Acids Res.
 , 
2007
, vol. 
35
 (pg. 
4164
-
4178
)
43
Chen
C.D.
Kobayashi
R.
Helfman
D.M.
Binding of hnRNP H to an exonic splicing silencer is involved in the regulation of alternative splicing of the rat beta-tropomyosin gene
Genes Dev.
 , 
1999
, vol. 
13
 (pg. 
593
-
606
)
44
Engel
A.G.
Nagel
A.
Walls
T.J.
Harper
C.M.
Waisburg
H.A.
Congenital myasthenic syndromes. I. Deficiency and short open-time of the acetylcholine receptor
Muscle Nerve
 , 
1993
, vol. 
16
 (pg. 
1284
-
1292
)
45
Yan
H.
Papadopoulos
N.
Marra
G.
Perrera
C.
Jiricny
J.
Boland
C.R.
Lynch
H.T.
Chadwick
R.B.
de la Chapelle
A.
Berg
K.
, et al.  . 
Conversion of diploidy to haploidy
Nature
 , 
2000
, vol. 
403
 (pg. 
723
-
724
)
46
Sahashi
K.
Masuda
A.
Matsuura
T.
Shinmi
J.
Zhang
Z.
Takeshima
Y.
Matsuo
M.
Sobue
G.
Ohno
K.
In vitro and in silico analysis reveals an efficient algorithm to predict the splicing consequences of mutations at the 5′ splice sites
Nucleic Acids Res.
 , 
2007
, vol. 
35
 (pg. 
5995
-
6003
)
47
Yi
R.
Bogerd
H.P.
Wiegand
H.L.
Cullen
B.R.
Both ran and importins have the ability to function as nuclear mRNA export factors
RNA
 , 
2002
, vol. 
8
 (pg. 
180
-
187
)
48
Jacquenet
S.
Mereau
A.
Bilodeau
P.S.
Damier
L.
Stoltzfus
C.M.
Branlant
C.
A second exon splicing silencer within human immunodeficiency virus type 1 tat exon 2 represses splicing of Tat mRNA and binds protein hnRNP H
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
40464
-
40475
)
49
Smith
C.W.
Porro
E.B.
Patton
J.G.
Nadal-Ginard
B.
Scanning from an independently specified branch point defines the 3′ splice site of mammalian introns
Nature
 , 
1989
, vol. 
342
 (pg. 
243
-
247
)
50
Kralovicova
J.
Vorechovsky
I.
Position-dependent repression and promotion of DQB1 intron 3 splicing by GGGG motifs
J. Immunol.
 , 
2006
, vol. 
176
 (pg. 
2381
-
2388
)
51
Koike
S.
Schaeffer
L.
Changeux
J.P.
Identification of a DNA element determining synaptic expression of the mouse acetylcholine receptor delta-subunit gene
Proc. Natl Acad. Sci. USA
 , 
1995
, vol. 
92
 (pg. 
10624
-
10628
)
52
Duclert
A.
Savatier
N.
Schaeffer
L.
Changeux
J.P.
Identification of an element crucial for the sub-synaptic expression of the acetylcholine receptor epsilon-subunit gene
J. Biol. Chem.
 , 
1996
, vol. 
271
 (pg. 
17433
-
17438
)
53
Ohno
K.
Anlar
B.
Engel
A.G.
Congenital myasthenic syndrome caused by a mutation in the Ets-binding site of the promoter region of the acetylcholine receptor epsilon subunit gene
Neuromuscul. Disord.
 , 
1999
, vol. 
9
 (pg. 
131
-
135
)
54
Sapru
M.K.
Florance
S.K.
Kirk
C.
Goldman
D.
Identification of a neuregulin and protein-tyrosine phosphatase response element in the nicotinic acetylcholine receptor epsilon subunit gene: regulatory role of an Rts transcription factor
Proc. Natl Acad. Sci. USA
 , 
1998
, vol. 
95
 (pg. 
1289
-
1294
)
55
Tang
J.
Jo
S.A.
Burden
S.J.
Separate pathways for synapse-specific and electrical activity-dependent gene expression in skeletal muscle
Development
 , 
1994
, vol. 
120
 (pg. 
1799
-
1804
)
56
Bessereau
J.L.
Stratford-Perricaudet
L.D.
Piette
J.
Le Poupon
C.
Changeux
J.P.
In vivo and in vitro analysis of electrical activity-dependent expression of muscle acetylcholine receptor genes using adenovirus
Proc. Natl Acad. Sci. USA
 , 
1994
, vol. 
91
 (pg. 
1304
-
1308
)
57
Hutchinson
D.O.
Walls
T.J.
Nakano
S.
Camp
S.
Taylor
P.
Harper
C.M.
Groover
R.V.
Peterson
H.A.
Jamieson
D.G.
Engel
A.G.
Congenital endplate acetylcholinesterase deficiency
Brain
 , 
1993
, vol. 
116
 (pg. 
633
-
653
)
58
Engel
A.G.
Engel
A.G.
Franzini-Armstrong
C.
Myology
 , 
1994
, vol. 
Vol. 2
 
New York
McGraw-Hill
(pg. 
1018
-
1045
)
59
Engel
A.G.
Lindstrom
J.M.
Lambert
E.H.
Lennon
V.A.
Ultrastructural localization of the acetylcholine receptor in myasthenia gravis and in its experimental autoimmune model
Neurology
 , 
1977
, vol. 
27
 (pg. 
307
-
315
)
60
Uchitel
O.
Engel
A.G.
Walls
T.J.
Nagel
A.
Atassi
M.Z.
Bril
V.
Congenital myasthenic syndromes. II. Syndrome attributed to abnormal interaction of acetylcholine with its receptor
Muscle Nerve
 , 
1993
, vol. 
16
 (pg. 
1293
-
1301
)
61
Shen
X.M.
Ohno
K.
Tsujino
A.
Brengman
J.M.
Gingold
M.
Sine
S.M.
Engel
A.G.
Mutation causing severe myasthenia reveals functional asymmetry of AChR signature cystine loops in agonist binding and gating
J. Clin. Invest.
 , 
2003
, vol. 
111
 (pg. 
497
-
505
)
62
Vogel
J.
Hess
W.R.
Borner
T.
Precise branch point mapping and quantification of splicing intermediates
Nucleic Acids Res.
 , 
1997
, vol. 
25
 (pg. 
2030
-
2031
)
63
Ohno
K.
Wang
H.L.
Milone
M.
Bren
N.
Brengman
J.M.
Nakano
S.
Quiram
P.
Pruitt
J.N.
Sine
S.M.
Engel
A.G.
Congenital myasthenic syndrome caused by decreased agonist binding affinity due to a mutation in the acetylcholine receptor epsilon subunit
Neuron
 , 
1996
, vol. 
17
 (pg. 
157
-
170
)
64
Schoepfer
R.
Luther
M.
Lindstrom
J.
The human medulloblastoma cell line TE671 expresses a muscle-like acetylcholine receptor. Cloning of the alpha-subunit cDNA
FEBS Lett.
 , 
1988
, vol. 
226
 (pg. 
235
-
240
)
65
Luther
M.A.
Schoepfer
R.
Whiting
P.
Casey
B.
Blatt
Y.
Montal
M.S.
Montal
M.
Linstrom
J.
A muscle acetylcholine receptor is expressed in the human cerebellar medulloblastoma cell line TE671
J. Neurosci.
 , 
1989
, vol. 
9
 (pg. 
1082
-
1096
)
66
Lee
B.S.
Gunn
R.B.
Kopito
R.R.
Functional differences among nonerythroid anion exchangers expressed in a transfected human cell line
J. Biol. Chem.
 , 
1991
, vol. 
266
 (pg. 
11448
-
11454
)
67
Ohno
K.
Milone
M.
Shen
X.-M.
Engel
A.G.
A frameshifting mutation in CHRNE unmasks skipping of the preceding exon
Hum. Mol. Genet.
 , 
2003
, vol. 
12
 (pg. 
3055
-
3066
)
68
Kozak
M.
An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs
Nucleic Acids Res.
 , 
1987
, vol. 
15
 (pg. 
8125
-
8148
)
69
Masuda
A.
Yoshikai
Y.
Kume
H.
Matsuguchi
T.
The interaction between GATA proteins and activator protein-1 promotes the transcription of IL-13 in mast cells
J. Immunol.
 , 
2004
, vol. 
173
 (pg. 
5564
-
5573
)
70
Gao
K.
Masuda
A.
Matsuura
T.
Ohno
K.
Human branch point consensus sequence is yUnAy
Nucleic Acids Res.
 , 
2008
, vol. 
36
 (pg. 
2257
-
2267
)