Targeted pseudouridylation: An approach for suppressing nonsense mutations in disease genes

Abstract

Nonsense mutations create premature termination codons (PTCs), activating the nonsense-mediated mRNA decay (NMD) pathway to degrade most PTC-containing mRNAs. The undegraded mRNA is translated, but translation terminates at the PTC, leading to no production of the full-length protein. This work presents targeted PTC pseudouridylation, an approach for nonsense suppression in human cells. Specifically, an artificial box H/ACA guide RNA designed to target the mRNA PTC can suppress both NMD and premature translation termination in various sequence contexts. Targeted pseudouridylation exhibits a level of suppression comparable with that of aminoglycoside antibiotic treatments. When targeted pseudouridylation is combined with antibiotic treatment, a much higher level of suppression is observed. Transfection of a disease model cell line (carrying a chromosomal PTC) with a designer guide RNA gene targeting the PTC also leads to nonsense suppression. Thus, targeted pseudouridylation is an RNA-directed gene-specific approach that suppresses NMD and concurrently promotes PTC readthrough.

Keywords: PTC read-through; RNA modification; box H/ACA RNA; nonsense mutations; nonsense suppression; nonsense-mediated mRNA decay; premature termination codon; pseudouridine; targeted pseudouridylation.

Figure 1.

Nonsense suppression by targeted pseudouridylation. (A) Box H/ACA RNA-guided RNA pseudouridylation is diagramed on the left. Box H/ACA gRNA (black line) and the four common core proteins, Nhp2, Gar1, Nop10, and Cbf5 (Dyskerin/NAP57 in mammals), are shown, and they form the box H/ACA RNP complex. During pseudouridylation, the guide sequences of box H/ACA gRNA form base-pairing interactions with the substrate RNAs (red lines), specifying the target uridine to be pseudouridylated (Ψs, indicated by arrows). Nonsense suppression by targeted pseudouridylation is schematized on the right. Double-stranded DNA (with an open reading frame) is shown. A PTC created by a nonsense mutation and the normal stop codon are indicated. After transcription and processing, mRNA (red) is produced. Normally, the presence of the PTC triggers NMD, degrading the PTC-containing mRNA. A small fraction of PTC-containing mRNA escapes from NMD and is translated, but translation stops at the PTC. In the presence of a designer gRNA (with its guide sequence designed to target the uridine of the PTC), the target uridine is site-specifically converted into Ψ, leading to NMD suppression and PTC-readthrough (full-length proteins are restored). (B) Two gRNA expression plasmids are diagrammed. One contained an independent gRNA gene and the other contained a gRNA gene within the intron of a hybrid β-globin gene whose 5’ end sequence is altered (LINK). Both were under the control of the CMV promoter. (C) After transfection of cells with the independent gRNA plasmid (lane 1) or with the intronic gRNA plasmid (lane 2), total RNA was recovered and Northern analysis was carried out using a DNA oligonucleotide complementary to a specific sequence of the designer gRNA (upper panel). The mature gRNA is indicated. As a control, the RNA sample was also probed with anti-U6 DNA oligonucleotide (lower panel). (D) Pre-mRNA splicing and gRNA production was monitored by RT-PCR at different time points during cell culture. The β-globin pre-mRNA with an altered 5’ end sequence (the LINK sequence, see B) and the two pairs of primers (R1-F1 and R2-F2; except for R1, all the other primers are specific to the exogenously transfected β-globin gene) are shown schematically. At indicated time points post-transfection with the intron-encoded designer gRNA plasmid (see B), cells were collected, RNA recovered, and RT-PCR performed. Primer pairs R1-F1 and R2-F2 were used to detect pre-mRNA/mRNA and gRNA, respectively. ACTB mRNA was also measured by another pair of primers as a loading control. The asterisks indicate two bands whose identities are currently unknown. (E) The N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X) pre-mRNA and mRNA are shown. The PTC (Q39X) and its flanking sequences are also shown. The three exons (E1, E2, and E3), two introns (black lines), and the FLAG tag are indicated. (F) After co-transfection of HEK293T cells with a plasmid containing an N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X) gene (see E) and a plasmid encoding the gRNA (intron-encoded; see B and D) specific (lanes 3 and 4) or non-specific (lanes 1 and 2) for targeting the PTC, total RNA was recovered and pseudouridylation assay (CMC-modification followed by primer-extension) performed using a primer complementary to a short sequence of β-Thalassemia downstream of the PTC (upper panel). As a control, primer-extension was also performed with a primer complementary to U6 (lower panel). The image was cropped for alignment purpose (indicated by the line).

Figure 2.

Nonsense suppression of the β-Thalassemia gene by targeted pseudouridylation. (A) HEK293T cells were either transfected with a plasmid encoding N-terminal FLAG-tagged wild-type β-globin (lane 1) or co-transfected with a plasmid containing a N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X) gene (see Fig 1E) and a plasmid encoding gRNA specific (lane 3) or non-specific (lane 2) for targeting the PTC. Cell lysates were prepared and anti-FLAG IP and western blotting conducted (Lower panel). Total RNA was also isolated from the lysates for RT-PCR analyses, measuring the levels of β-Thalassemia (β-globin) mRNA (Upper panel) and gRNA (middle panel). MUP served as a loading control for RT-PCR and Tubulin served as a loading control for western blot analysis. The image was cropped for alignment purpose. (B) Quantification of the levels of mRNA and full-length protein observed in (A). After normalization against the control mRNA (MUP), the relative β-Thalassemia mRNA level was calculated. Likewise, after normalization against the control protein (Tubulin), the relative restored full-length protein level was also calculated. All quantifications were based on three independent experiments (error bars indicate S.D.-and (*) represents P < 0.05 calculated by Student's t test.). (C) To compare targeted pseudouridylation with G418 treatment, HEK293T cells were transfected with a plasmid encoding N-terminal FLAG-tagged wild-type β-globin (lane 1) or co-transfected with a plasmid containing an N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X/PTC39) gene and a plasmid encoding a non-PTC-specific gRNA (lane 2) or a PTC-specific gRNA (lane 3). In parallel, HEK293T cells were transfected with the plasmid containing the N-terminal tagged β-Thalassemia (β-globin Q39X) gene (see Fig 1E) and then incubated in a medium containing G418 (125 μg/mL) (lane 4). In another parallel experiment, HEK293T cells were co-transfected with the plasmid containing the N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X/PTC39) gene and the plasmid encoding the non-PTC-specific gRNA and then incubated in a medium containing G418 (125 μg/mL) (lane 5). Western analysis (lower panel) and RT-PCR were conducted (upper panel) as in (A). The image was cropped for alignment purpose. (D) Quantitative analysis of the levels of mRNA and full-length protein observed in (C). Calculation/analysis was exactly as in (B).

Figure 3.

The persistent effect of targeted pseudouridylation on nonsense suppression. (A) HEK293T cells were transfected with a plasmid encoding N-terminal FLAG-tagged wild-type β-globin (lane 1) or co-transfected with a plasmid containing an N-terminal FLAG-tagged β-Thalassemia (β-globin Q39X/PTC39) gene and a plasmid encoding a non-PTC-specific gRNA (lanes 1, 3, 5 and 7) or a PTC-specific gRNA (lanes 2, 4, 6 and 8). At different time points/days post transfection (day 3, day 8, day 11 and day16, indicated above each lane), cell lysates were made (using equal number of cells), and anti-FLAG IP and western were performed as in Fig 2A (lower panel). Total RNA was also prepared for RT-PCR analysis (Upper panel) as in Fig 2A. ACTB mRNA and Tubulin served as controls for RT-PCR and western, respectively. The image was cropped for alignment purpose. (B) Quantitative analysis of the levels of mRNA, gRNAs and full-length protein observed in (A). Calculation/analysis was exactly as in Fig 2B.

Figure 4.

Nonsense suppression by targeted pseudouridylation in different sequence contexts. (A) β-Thalassemia (β-globin Q39X) mRNA is diagrammed. The N-terminal FLAG tag, exons 1, 2, 3 (E1, E2 and E3), and the normal stop codon are indicated. Q39X/PTC39 and its flanking sequences (Blue) were substituted with the CFTR (G542X) or IDUA (W392X) PTC and flanking sequences (Red). (B) HEK293T cells were transfected with a plasmid encoding N-terminal FLAG-tagged wild-type β-globin (lane 1) or a plasmid containing the CFTR PTC-substituted β-globin gene (lane 2), or co-transfected with the plasmid containing the CFTR PTC-substituted β-globin gene and a plasmid encoding a non-PTC-specific gRNA (lane 3) or a PTC-specific gRNA (lane 4). Cell lysates were made, and the levels of mRNA and full-length protein were measured, as in Fig 2A. The image was cropped for alignment purpose. (C) Quantitative analysis of the levels of mRNA and full-length protein observed in (B). Calculation/analysis was exactly as in Fig 2B. (D) Experiments were performed exactly as in (B), except that a plasmid containing the IDUA PTC-substituted (instead of CFTR PTC-substituted) β-globin gene was used. Co-transfection with this plasmid and a plasmid encoding a PTC-specific gRNA is shown in lane 3. The image was cropped for alignment purpose. (E) Quantitative analysis of the levels of mRNA and full-length protein observed in (D). Calculation/analysis was exactly as in Fig 2B. (F) HEK293T cells were transfected with a plasmid encoding CFTR N-terminal Binding Domain (NBD1, C-terminal FLAG-tagged and intronless) or co-transfected with a plasmid containing a PTC (G542X)-containing NBD1 gene (C-terminal FLAG-tagged and intronless) and a plasmid encoding a non-PTC-specific gRNA (lane 2) or a PTC-specific gRNA (lane 3). Cell lysates were made, and the levels of mRNA and full-length protein were measured, as in Fig 2A. (G) Quantitative analysis of the levels of mRNA and the full-length CFTR NBD1 protein observed in (F). Calculation/analysis was exactly as in Fig 2B. (H) Experiments were performed exactly as in (F), except that a plasmid containing an intronless, C-terminal FLAG-tagged, full-length wild-type or PTC (W392X)-containing IDUA gene (instead of the CFTR NBD1 gene) was used. The image was cropped for alignment purpose. (I) Quantitative analysis of the levels of mRNA and the full-length IDUA protein observed in (H). Calculation/analysis was exactly as in Fig 2B.

Figure 5.

Nonsense suppression by targeted pseudouridylation in a disease model cell lines. (A) The 16HBE14o- human bronchial epithelial cell line, carrying a wild-type CFTR gene (lane 1) or a chromosomal PTC (G542X)-containing CFTR gene (lanes 2 and 3), was transfected with no plasmid (lane 1) or a plasmid encoding a PTC-specific (lane 2) or non-PTC-specific (lane 3) gRNA. Cells were harvested and cell lysates prepared. Total RNA was recovered and RT-PCR performed to measure the level of CFTR mRNA (wild-type or PTC mutant) (Upper panel). The lysates were also used for western analysis to measure the full-length CFTR protein (using anti-CFTR antibody) (Lower panel). ACTB mRNA and Tubulin served as a control for RT-PCR and western, respectively. Two forms of CFTR full-length protein (Band B and Band C) are indicated. (B) Quantitative analysis of the levels of CFTR mRNA and full-length CFTR protein observed in (A). Calculation/analysis was exactly as in Fig 2B. (C) The mouse embryonic fibroblasts carrying an endogenous PTC-containing IDUA gene (W392X) was also used for the nonsense suppression assay. Cells were transfected with a plasmid containing a non-PTC-specific gRNA gene (lane 2), a plasmid containing a PTC-specific gRNA gene (lane 3), or treated with G418 (250 μg/mL) (lane 4). As a control, wild-type mouse embryonic fibroblasts were also analyzed in parallel (lane 1). RT-PCR (upper panel) and western analysis (Lower panel) were performed as in Fig 2A except that different primers and anti-IDUA antibody were used. The image was cropped for alignment purpose. (D) Quantitative analysis of the levels of IDUA mRNA and full-length IDUA protein observed in (C). Calculation/analysis was exactly as in Fig 2B.

Figure 6.

Nonsense suppression by an mCherry gene carrying an artificial intron that encodes a gRNA. (A) An artificial gRNA intron was inserted into the mCherry gene. HEK293T cells were transfected with the wild-type β-globin gene (without gRNA) (lane 1), or co-transfected with the mutant β-globin (with CFTR PTC, G542X) gene and an mCherry gene containing a non-PTC specific gRNA intron (lane 2) or a PTC-specific gRNA intron (lane 3). For direct comparison, cells were also co-transfected with the mutant β-globin (with CFTR PTC, G542X) gene and the β-globin gene carrying a PTC-specific gRNA in its first intron (used in all above experiments) (lane 4). RT-PCR and western analysis were performed as in Fig 2A. (B) Quantitative analysis of the levels of β-globin mRNA (with CFTR PTC), gRNA, and full-length β-globin protein (with G542X) observed in (A). Calculation/analysis was exactly as in Fig 2B. (C) mCherry fluorescence was detected in the cells transfected with the mCherry gene carrying an artificial (non-PTC-specific or PTC-specific) gRNA intron (first two rows), but not in the cells transfected with the β-globin gene carrying a gRNA intron (the bottom row).

Figure 7.

IDUA protein functional assay. (A) The IDUA gene encodes α-L-iduronidase, an enzyme that hydrolyzes 4MU-iduronide, releasing fluorescent 4MU (schematized). (B) Cell culture and cell transfection were conducted as in Fig 4H. Cell lysates, prepared from HEK293T cells transfected with a plasmid carrying a PTC (W392X)-containing IDUA gene (Sample #1) or co-transfected with this plasmid and a plasmid encoding a non-PTC-specific gRNA (Sample #2) or a PTC-specific gRNA (Sample #3), were used for the IDUA functional assay. The standard curve was also carried out and shown (the left graph). The IDUA activity value of Sample #3 was within the linear range of standard curve. The graph shown on the right is an excerpt (of the graph shown on the left) focusing on the three experimental samples. (C) The IDUA functional assay was also carried out using cell lysates prepared from transfected mouse embryonic fibroblasts carrying an endogenous PTC-containing IDUA gene (W392X) (see Fig 5C). Sample #1 was from cells transfected with a plasmid containing a non-PTC-specific gRNA gene; Sample #2 was from cells transfected with a plasmid containing a PTC-specific gRNA gene; Sample #3 was from cells treated with G418 (250 μg/mL).