Ribosomal Frameshifting Selectively Modulates the Assembly, Function, and Pharmacological Rescue of a Misfolded CFTR Variant

The cotranslational misfolding of the cystic fibrosis transmembrane conductance regulator chloride channel (CFTR) plays a central role in the molecular basis of cystic fibrosis (CF). The misfolding of the most common CF variant (ΔF508) remodels both the translational regulation and quality control of CFTR. Nevertheless, it is unclear how the misassembly of the nascent polypeptide may directly influence the activity of the translation machinery. In this work, we identify a structural motif within the CFTR transcript that stimulates efficient −1 ribosomal frameshifting and triggers the premature termination of translation. Though this motif does not appear to impact the interactome of wild-type CFTR, silent mutations that disrupt this RNA structure alter the association of nascent ΔF508 CFTR with numerous translation and quality control proteins. Moreover, disrupting this RNA structure enhances the functional gating of the ΔF508 CFTR channel at the plasma membrane and its pharmacological rescue by the CFTR modulators contained in the CF drug Trikafta. The effects of the RNA structure on ΔF508 CFTR appear to be attenuated in the absence of the ER membrane protein complex (EMC), which was previously found to modulate ribosome collisions during “preemptive quality control” of a misfolded CFTR homolog. Together, our results reveal that ribosomal frameshifting selectively modulates the assembly, function, and pharmacological rescue of a misfolded CFTR variant. These findings suggest interactions between the nascent chain, quality control machinery, and ribosome may dynamically modulate ribosomal frameshifting in order to tune the processivity of translation in response to cotranslational misfolding.


Slip-Site 7 G G G G A A A A A A G G A A G A A U U C U A U U C U C A A U C C A A U C A A C U C U A U A C G A
...674 764...

Slip-Site 6 A A C A A A
. Potential Slip Sites in CFTR.The relative positions of various canonical and near-canonical X 1 XXY 4 YYZ 7 sites within the CFTR trasncript along with downstream secondary structure predictions are shown.(A) A schematic depicts the nucleotide position of the protein domain boundaries and potential slip-sites within the CFTR transcript.(B) Efficient frameshifting typically requires the formation of stable RNA secondary structure 5-8 basepairs downstream of the slippery sequence.We therefore utilized the Vienna RNAfold program to predict the mRNA secondary structure within the 75 nucleotides that lie 5 basepairs downstream of each putative slippery site.Cartoons depict the most stable predicted secondary structure at each site in relation to the slippery sequence (blue) for each potential slip-site not characterized in this work.

A B
Carmody & Roushar et al.In conjunction with the results in Fig. 6 showing that CFTR-specific modulators accelerate quenching, these results demonstrate that the observed cellular quenching reaction is rate-limited by CFTR conductance.
Figure S10 ΔF508+Inh  WT CFTR-mediated hYFP quenching is slower in EMC6 knockout cells relative to parental HEK293T cells.As a result, the effect of the ΔF508 mutation appear less pronounced.Moreover, the RF mut modification has little, if any, impact on ΔF508 CFTR-mediated hYFP quenching in EMC6 knockout cellsthe p-value from a two-sample t-test indicated the difference between the half-lives derived from cells expressing these variants are not statistically significant.
Carmody & Roushar et al. Figure S11 Carmody & Roushar et al.A GAPDH loading control is included for reference.Though trends were consistent, quantitative variations in total CFTR expression levels in the presence of correctors were observed across western blots for different biological replicates.Trends were more precise when measured by flow cytometry due to the fact that this analysis is insensitive to variations in transfection efficiency.Nevertheless, C/B ratios from replicate experiements were quite consistent.For this reason, we chose to report expres-sion measurements from flow cytometry and maturation efficiencies from western blots in Fig. 6.
Figure S2.Reverse Transcriptase PCR of CFTR Dual Luciferase Reporters.CFTR frameshift remporters were transiently expressed in HEK293T cells prior to the extraction of the total cellular mRNA and the amplification of each reporter cDNA with reverse transcriptase.The products of these reactions are shown on an image of a representative 1% agarose gel.Reactions produced a clean, strong band in the expected molecular weight range (~3 kb) in all cases, which indicates the reporter constructs generate the expected transcript without any significant alternative splicing.

Figure S3 .Figure S4 .Figure S5 .RFmutFigure S6 .
Figure S3.Ribosomal frameshifting reporter measurements in HEK293T Cells.The efficiency of ribosomal frameshifting during the translation of a structured region within the CFTR transcript was assessed in HEK293T cells using a bicistronic dual luciferase reporter system.(A, B) Dot plots indicate the raw RLuc and FLuc luminescence intensities within the lysates of HEK293T cells transiently expressing these constructs, respectively.Results were taken from three biological replicates with 4 transfections per sample each.Central hashes represent the average intensities and whiskers reflect their standard deviation.(C) A bar graph depicts the average -1 ribosomal frameshifting efficiencies for each construct in HEK293T cells.Error bars reflect the standard deviation to serve as a measure of precision.

Figure S7 .Figure S9 .
Figure S7.Relative abundance of CFTR transcripts in transiently transfected CFBE 41 ocells.Quantitative PCR (qPCR) was used to compare the abundance of the CFTR transcript in CFBE41o-cells transiently expressing WT (gray), WT RF mut (white), ∆F508 (red), and ∆F508 RF mut (blue).Cells were harvested two days after transfection prior to the extraction of cellular RNA.Cellular mRNA were then reverse-transcribed into DNA and real-time PCR was used to track TaqMan probe amplification from the cDNA for CFTR and an endogenous control transcript (ACTB).∆Ct values for each variant were used to quantify the abundance of each transcript relative to that of WT (RQ), which are plotted in the bar graph above.Points represent individual values from three biological replicates.Bars report the average values and the whiskers show the 90th percentile value.A mock sample that was transfected without the expression vector is shown for reference.The results reveal that transient transfections generate a high degree of variability in transfection efficiency, which is likely to obscure any deviations in transcript stability.

Figure S10 .
Figure S10.Impact of CFTR Inh-172 on the Observable hYFP Quenching Kinetics.The imact of the CFTR-specific inhibitor 172 on the kinetics of halide-sensitive yellow fluorescent protein (hYFP) quenching upon activation of stably expressed CFTRs in HEK293T cells was characterized by flow cytometry.A) HEK293T cells stably expressing WT CFTR were stimulated with 25 µM forskolin (0.06% DMSO Vehicle) to activate CFTR prior to measurement of the change in cellular hYFP: mKate intensity ratio measurements over time in the presence (red) and absence (gray) of 10 µM Inh-172 by flow cytometry.Single cell intensity ratios are plotted against the time and the fits for a single-exponential decay function are shown for reference.The fitted quenching half-life value increases 89% in the presence of inhibitor.A) HEK293T cells stably expressing ∆F508 CFTR were stimulated with 25 µM forskolin (0.06% DMSO Vehicle) to activate CFTR prior to measurement of the change in cellular hYFP: mKate intensity ratio measurements over time in the presence (red) and absence (gray) of 10 µM Inh-172 by flow cytometry.Cellular intensity ratios are plotted against the time and the fits for a single-exponential decay function are shown for reference.The fitted quenching half-life value increases 45% in the presence of inhibitor.In conjunction with the results in Fig.6showing that CFTR-specific modulators accelerate quenching, these results demonstrate that the observed cellular quenching reaction is rate-limited by CFTR conductance.
Figure S11.CFTR Function in EMC6 Knockout Cells.The functional conductance of stably expressed CFTR variants was compared in the context of an EMC6 knockout HEK293T cell line by measuring the time-dependent quenching of a halide-sensitive yellow fluorescent protein (hYFP).A) HEK293T cells stably expressing WT (white), WT RF mut (gray), ΔF508 (red), and ΔF508 RF mut (blue) were stimulated with 25 µM forskolin (0.06% DMSO Vehicle) to activate CFTR prior to measurement of the change in cellular hYFP: mKate intensity ratio measurements over time by flow cytometry.Cellular intensity ratios are plotted against the time and the global fits of the decay are shown for reference.B) A bar graph depicts the globally fit half-life for hYFP quenching for each variant.Values represent the average fitted values for each variant normalized relative to the WT value (n = 3).WT CFTR-mediated hYFP quenching is slower in EMC6 knockout cells relative to parental HEK293T cells.As a result, the effect of the ΔF508 mutation appear less pronounced.Moreover, the RF mut modification has little, if any, impact on ΔF508 CFTR-mediated hYFP quenching in EMC6 knockout cellsthe p-value from a two-sample t-test indicated the difference between the half-lives derived from cells expressing these variants are not statistically significant.

Figure S12 .
Figure S12.Impact of ribosomal frameshifting on the pharmacological rescue of CFTR expression in HEK293T cells.Western blotting was used to compare the expression and matu-ration of transiently expressed CFTR variants in the presence of CFTR modulators in HEK293T cells.A representative western blot depicting the relative abundance of the mature (band C) and immature (band B) CFTR glycoforms of each indicated variant in HEK293T cells treated for 16 hours with DMSO (Vehicle) or 3uM VX-661 + 3uM VX-445 is shown.A GAPDH loading control is included for reference.Though trends were consistent, quantitative variations in total CFTR expression levels in the presence of correctors were observed across western blots for different biological replicates.Trends were more precise when measured by flow cytometry due to the fact that this analysis is insensitive to variations in transfection efficiency.Nevertheless, C/B ratios from replicate experiements were quite consistent.For this reason, we chose to report expres-sion measurements from flow cytometry and maturation efficiencies from western blots in Fig.6.