Open reading frame correction using splice-switching antisense oligonucleotides for the treatment of cystic fibrosis

Abstract

CFTR gene mutations that result in the introduction of premature termination codons (PTCs) are common in cystic fibrosis (CF). This mutation type causes a severe form of the disease, likely because of low CFTR messenger RNA (mRNA) expression as a result of nonsense-mediated mRNA decay, as well as the production of a nonfunctional, truncated CFTR protein. Current therapeutics for CF, which target residual protein function, are less effective in patients with these types of mutations due in part to low CFTR protein levels. Splice-switching antisense oligonucleotides (ASOs), designed to induce skipping of exons in order to restore the mRNA open reading frame, have shown therapeutic promise preclinically and clinically for a number of diseases. We hypothesized that ASO-mediated skipping of CFTR exon 23 would recover CFTR activity associated with terminating mutations in the exon, including CFTR p.W1282X, the fifth most common mutation in CF. Here, we show that CFTR lacking the amino acids encoding exon 23 is partially functional and responsive to corrector and modulator drugs currently in clinical use. ASO-induced exon 23 skipping rescued CFTR expression and chloride current in primary human bronchial epithelial cells isolated from a homozygote CFTR-W1282X patient. These results support the use of ASOs in treating CF patients with CFTR class I mutations in exon 23 that result in unstable CFTR mRNA and truncations of the CFTR protein.

Keywords: antisense oligonucleotides; cystic fibrosis; human bronchial epithelial cells; nonsense-mediated decay; splicing.

Fig. 1.

CFTR-Δ23 function is comparable to CFTR-W1282X and is responsive to CF modulators. (A) Schematic of CFTR exon 23 in relation to its position within NBD2 and the position of the CFTR-W1282X mutation. Symmetric exons are colored green. (B) Schematic of the CFTR-W1282X and CFTR-Δ23 constructs transfected into FRT cells. (C) Average conductance traces from FRT cells stably transfected with CFTR-Δ23, CFTR-W1282X, or empty vector. Cells were pretreated with vehicle (DMSO, solid lines), C18 (dotted lines), or VX-445+VX-661 (dashed lines). The time of compound additions (Forskolin, VX-770, or Inh-172) is indicated. (D) Average AUC was quantified for the forskolin and VX-770 test periods for each construct. Error bars are ± SEM. Two-way ANOVA; Dunnett’s multiple comparison test to vehicle within groups, ****P < 0.0001. CFTR-Δ23: DMSO and VX-770, n = 11; C18, n = 6; VX-445+VX-661, n = 5. CFTR-W1282X: DMSO and VX-770, n = 9; C18, n = 5; VX-445+VX-661, n = 4. empty vector: DMSO and VX-770, n = 8; C18, n = 5; VX-445+VX-661, n = 3. (E) CFTR protein analysis using WES capillary electrophoresis, Bands C and B, isolated from FRT cells stably transfected with empty vector, CFTR-WT, CFTR-W1282X, or CFTR-Δ23 constructs treated with vehicle or C18. β-actin was used as a control for protein expression. (F) Quantification of total CFTR C band from E, normalized to β-actin. Error bars are ±SEM. Empty vector: n = 1, CFTR-WT: n = 4, CFTR-W1282X: n = 2, CFTR-Δ23: DMSO, n = 4; C18, n = 3.

Fig. 2.

ASOs induce CFTR-W1282X exon 23 skipping. (A) Schematic of the CFTR dysfunction caused by the W1282X mutation in exon 23. The c.3846 G > A (W1282X) mutation creates a PTC in exon 23, leading to degradation of the transcript via NMD and reducing translation of a semifunctional, truncated CFTR protein (indicated by red arrows). ASO induced exon 23 skipping, eliminates the PTC, restores CFTR mRNA stability, and increases CFTR expression (green arrows). (B) Diagram of ASO target sites (green lines) on human CFTR exon 23 pre-mRNA. The location of the CFTR-W1282X mutation is indicated by a red vertical line. (C) RT-PCR analysis of CFTR exon 23 splicing in a CFF16HBEge-W1282X cell line treated with the indicated ASO (10 μM). Exon 23 skipping or cryptic splice site activation was quantified (Δ23 or cryptic/[Δ23+cryptic+W1282X] × 100) and is indicated below each lane. (D) Sequence alignment of ASO-23A and ASO-23B to exon 23. (E) RT-PCR analysis of exon 23 splicing in CFF16HBEge-W1282X cells treated with ASO-23A, ASO-23B, ASO-23A + ASO-23B (ASO-23AB, 10 μM each), or ASO-C (20 μM). Exon 23 skipping or cryptic splice site activation (percent of total) was quantified and is indicated below each lane. (F) RT-PCR analysis of exon 23 splicing in CFF16HBEge-W1282X cells treated with ASO-23AB or ASO-C at indicated concentrations. Exon 23 skipping and cryptic splice site activation was quantified and is indicated below each lane. β-actin is a control for RNA expression in all experiments. Primer-annealing sites are shown above amplicon diagrams.

Fig. 3.

ASO-induced exon skipping increases membrane conductance of a CFTR-W1282X cell line. (A) Conductance (Gt) traces of a CFF16HBEge-W1282X clonal cell line selected for high resistance (CFF16HBEge-W1282X-SCC:3F2). Cells were transfected with vehicle (black line), ASO-C (gray line), or ASO-23AB (green lines) at indicated concentrations. Cells were pretreated with DMSO (solid line) or VX-445+VX-661 (dashed lines). (B) Average AUC of the conductance trace in A was quantified for the forskolin + VX-770 test period for each treatment group. Error bars are ±SEM. One-way ANOVA; Dunnett’s multiple comparison test to vehicle + DMSO, **P < 0.01, and ***P < 0.001. n = 3, except for 10 μM ASO-23AB in which n = 2. (C) RT-PCR analysis of exon 23 splicing in CFF16HBEge-W1282X-SCC:3F2 cells in A. β-actin is a control for RNA expression. (D) Quantification of exon 23 skipping (percent of total) in C. Primer-annealing sites are diagrammed. Error bars are ±SEM. One-way ANOVA; Dunnett’s multiple comparison test to vehicle, *P < 0.05, and ***P < 0.001. n = 3, except for 10 μM ASO-23AB in which n = 2. (E) The calculated AUC, shown in B, correlated with exon 23 skipping (percent), shown in D (simple linear regression).

Fig. 4.

ASO treatment induces exon 23 skipping, stabilizes CFTR mRNA, and rescues CFTR function in primary hBE cells isolated from a patient homozygous for CFTR-W1282X. (A) Equivalent current (Ieq) traces of primary hBE cells from a homozygous CFTR-W1282X CF donor. Cells were transfected with vehicle, ASO-C, or ASO-23AB (320 μM total) and pretreated with DMSO, C18, or VX-445+VX-661. (B) Average AUC of the current traces in A was quantified for the forskolin or forskolin + VX-770 test periods for each treatment group. Error bars are ±SEM. Two-way ANOVA; Dunnett’s multiple comparison test to DMSO within treatment groups, ####P < 0.01. Two-way ANOVA; Dunnett’s multiple comparison test to vehicle within treatment groups, ****P < 0.0001. n = 4, except C18 treatment in which n = 3. (C) RT-PCR analysis of exon 23 splicing in cells from A. β-actin is a control for RNA expression. (D) Quantification of exon 23 skipping in C. Error bars are ±SEM. One-way ANOVA; Dunnett’s multiple comparison test, ****P < 0.0001. n = 4. (E) RT-PCR analysis of CFTR mRNA expression (exons 11 to 14) from cells analyzed in A, compared to hBE cells from a non-CF donor. (F) RT-qPCR analyses of total CFTR mRNA (exon 11[WT]-12) from cells analyzed in A, compared to hBE cells from non-CF and F508del-CFTR donors. Error bars are ±SEM. One-way ANOVA; Tukey’s multiple comparison test, ****P < 0.0001. n = 4. (G) Capillary electrophoresis (WES) analysis of CFTR protein isolated from cells in A. Protein from a non-CF donor and a CF donor homozygous for CFTR-F508del is also shown. SNRPB2 is a loading control. (H) Quantification of the total CFTR (B + C Bands)/SNRP2 normalized to vehicle + DMSO shown in G. Error bars are ±SEM. One-way ANOVA; Dunnett’s multiple comparison test to vehicle + DMSO, *P < 0.05, and ****P < 0.0001. n = 3. Primer-annealing sites are shown above amplicon diagrams.

Fig. 5.

ASO treatment does not affect modulator activity in hBE cells isolated from a CF patient compound heterozygous for CFTR-W1282X and F508del. (A) Equivalent current (Ieq) traces of primary hBE cells isolated from a CF donor heterozygous for CFTR-W1282X and F508del. Cells were transfected with vehicle, ASO-C, or ASO-23AB (320 μM total). Cells were pretreated with DMSO, C18, or VX-445+VX-661. (B) Average AUC of the current traces in A was quantified for the forskolin + VX-770 test periods for each treatment group. Error bars are ±SEM. Two-way ANOVA; Dunnett’s multiple comparison test to DMSO within treatment groups, #P < 0.05, ####P < 0.01. Two-way ANOVA; Dunnett’s multiple comparison test to vehicle within treatment groups, ns = P > 0.05. n = 3. (C) RT-PCR analysis of exon 23 splicing in hBE CFTR-W1282X/F508del cells in A. β-actin is a control for RNA expression. (D) Quantification of exon 23 skipping in C. Error bars are ±SEM. One-way ANOVA; Dunnett’s multiple comparison test, **P < 0.01. n = 3. (E) RT-PCR analysis of non-F508del CFTR mRNA expression (exons 11WT-14) from cells analyzed in A, compared to hBE cells from a non-CF donor. (F) RT-qPCR analyses of total CFTR mRNA from non-F508del alleles (exons 11WT-12) in cells analyzed in A, compared to hBE cells from non-CF and F508del-CFTR donors. Error bars are ±SEM. One-way ANOVA; Tukey’s multiple comparison test, ****P < 0.0001. n = 4, one outlier identified and removed with ROUT outlier analysis (Q = 5%) in WT. (G) RT-PCR analysis of CFTR-F508del mRNA expression (exons 11F508del-14) from cells analyzed in A, compared to hBE cells from a non-CF donor. (H) RT-qPCR analyses of total CFTR mRNA from F508del alleles (exons 11WT-12) in cells analyzed in A, compared to hBE cells from non-CF and F508del-CFTR donors. Error bars are ±SEM. One-way ANOVA; Tukey’s multiple comparison test, ns = P > 0.05. n = 4. Primer-annealing sites are shown above amplicon diagrams; the dashed line indicates a mismatch.