Tannic acid facilitates expression of the polypyrimidine tract binding protein and alleviates deleterious inclusion of CHRNA1 exon P3A due to an hnRNP H-disrupting mutation in congenital myasthenic syndrome

Abstract

We recently reported that the intronic splice-site mutation IVS3-8G>A of CHRNA1 that encodes the muscle nicotinic acetylcholine receptor alpha subunit disrupts binding of a splicing repressor, hnRNP H. This, in turn, results in exclusive inclusion of the downstream exon P3A. The P3A(+) transcript encodes a non-functional alpha subunit that comprises 50% of the transcripts in normal human skeletal muscle, but its functional significance remains undetermined. In an effort to search for a potential therapy, we screened off-label effects of 960 bioactive chemical compounds and found that tannic acid ameliorates the aberrant splicing due to IVS3-8G>A but without altering the expression of hnRNP H. Therefore, we searched for another splicing trans-factor. We found that the polypyrimidine tract binding protein (PTB) binds close to the 3' end of CHRNA1 intron 3, that PTB induces skipping of exon P3A and that tannic acid increases the expression of PTB in a dose-dependent manner. Deletion assays of the PTB promoter region revealed that the tannic acid-responsive element is between positions -232 and -74 from the translation initiation site. These observations open the door to the discovery of novel therapies based on PTB overexpression and to detecting possible untoward effects of the overexpression.

Figure 1.

Chimeric CHRNA1 exon P3A minigene for screening of chemical compounds. CHRNA1 exon P3A and its flanking introns are inserted after position 834 (arrowhead) of the firefly luciferase cDNA. Skipping of exon P3A should generate an intact luciferase molecule, whereas inclusion of exon P3A should abrogate it. The hnRNP H-binding ‘TGGG’ motif is underlined.

Figure 2.

Tannic acid alleviates aberrant inclusion of CHRNA1 exon P3A due to IVS3-8G>A by facilitating the expression of PTBP1. (A) RT–PCR of wild-type and mutant CHRNA1 minigenes in HEK293T cells with increasing concentrations of tannic acid. (B) Real-time RT–PCR to quantify the ratio of P3A-skipped transcript arising from wild-type and mutant CHRNA1 minigenes in HEK293T cells with increasing concentrations of tannic acid. The ratios are represented by the mean and SD. For the mutant minigene, 100 µm tannic acid increases the ratio of the P3A(−)-transcript. (C) Real-time RT–PCR to demonstrate a dose-dependent increase of PTBP1 mRNA by tannic acid. The ratios are normalized to that in the absence of tannic acid. Means and SD are represented.

Figure 3.

PTB binds close to the 3′ splice site of exon P3A and induces skipping of exon P3A. (A) ³²P-labeled RNA probe is incubated with nuclear extracts of indicated cells, followed by UV cross-linking and SDS–PAGE. Autoradiographs demonstrate a ∼60-kDa molecule (arrowhead). (B) ³²P-labeled RNA probe is incubated with nuclear extracts of HEK293T cells followed by UV cross-linking and RNase digestion. The bound molecules are immunoprecipitated with anti-PTB (At-PTB) or control antibody (cont Ab), and separated on SDS–PAGE. Only anti-PTB antibody precipitates a ∼60-kDa molecule. (C) Western blotting with anti-PTB or anti-SRp55 antibody of affinity purified HEK293T nuclear extract. Affinity purification also demonstrates the binding of PTB to the wild-type (WT) intron 3, but not to the scrambled RNA probe (Scr). NUC represents a lane loaded with nascent nuclear extract. (D) Real-time RT–PCR analysis of HeLa cells transfected with the wild-type minigene along with the indicated siRNA or cDNA. The %P3A values are normalized to those of control values. Down-regulation of PTBP1 enhances, whereas overexpression of PTBP1 silences the recognition of exon P3A. Bars represent the mean and SD of three experiments.

Figure 4.

Scanning deletion analysis of the PTBP1 promoter region. (A) Luciferase promoter assays to demonstrate responses to 80 µm tannic acid. Relative luciferase activities are first calculated by dividing the firefly luciferase activity by the cotransfected Renilla luciferase activity. The calculated relative luciferase activity in the presence of 80 µm tannic acid is then divided by that in the absence of tannic acid to estimate the response. Means and SD (n = 4) are represented. The sizes of the PTBP1 promoter region before the translation initiation site are indicated. ‘232-Sp1’ and ‘232-AP2’ indicate a 232-nucleotide construct lacking two Sp1 elements [underlined in (B)] or three AP2 elements [indicated by dots in (B)], respectively. (B) PTBP1 promoter region from position –239 to +3. Arrowheads point to the 5′ ends of the 232-nucleotide and 74-nucleotide constructs in (A).