CRISPR/Cas13d targeting suppresses repeat-associated non-AUG translation of C9orf72 hexanucleotide repeat RNA

Abstract

A hexanucleotide GGGGCC repeat expansion in the non-coding region of the C9orf72 gene is the most common genetic mutation identified in patients with amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). The resulting repeat RNA and dipeptide repeat proteins from non-conventional repeat translation have been recognized as important markers associated with the diseases. CRISPR/Cas13d, a powerful RNA-targeting tool, has faced challenges in effectively targeting RNA with stable secondary structures. Here we report that CRISPR/Cas13d can be optimized to specifically target GGGGCC repeat RNA. Our results demonstrate that the CRISPR/Cas13d system can be harnessed to significantly diminish the translation of poly-dipeptides originating from the GGGGCC repeat RNA. This efficacy has been validated in various cell types, including induced pluripotent stem cells and differentiated motor neurons originating from C9orf72-ALS patients, as well as in C9orf72 repeat transgenic mice. These findings demonstrate the application of CRISPR/Cas13d in targeting RNA with intricate higher-order structures and suggest a potential therapeutic approach for ALS and FTD.

Keywords: Gene therapy; Genetics; Molecular biology; Neurodegeneration; Neuroscience.

Figure 1. The one-vector CRISPR/Cas13d system and the test for its knockdown efficiency.

(A) Schematic of the one-vector CRISPR/Cas13b/d system construct (top) and the EGFP reporter construct (bottom). (B) Schematic of type VI-d (left) and VI-b (right) crRNA structures with the target RNA. The crRNAs carry a direct repeat sequence (blue) to facilitate the binding with their corresponding Cas13 enzyme, and a spacer sequence (red) specific for the target RNA, r(GGGGCC)n (purple). (C and D) The knockdown efficiency test in HEK293 cells via cotransfection of the CRISPR/Cas13d vector (C), or the CRISPR/Cas13b vector (D), and the reporter construct. Immunoblotting of EGFP showed the knockdown efficiency for different guide RNAs (gRNAs). Data are presented as means ± SD of 3 independent experiments and were analyzed with ordinary 1-way ANOVA with Dunnett’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2. Guided Cas13d specifically decreased translation associated with GGGGCC repeat–containing transcripts.

(A) Schematic of the inducible luciferase-based C9orf72 RAN translation reporter system in HeLa Flp-In cells. (B) The reporter cells stably expressing Cas13d and gRNA S24 and S30 showed lower signals of NanoLuc and firefly luciferase signal from the (GGGGCC)70-containing reporter transcripts. The significant reduction of the NanoLuc luciferase signal relative to the firefly luciferase signal demonstrates that the repeat-associated RAN translation is inhibited by Cas13d-mediated S24 or S30 treatment. (C) The control cells harboring the reporter without the G4C2 repeat showed no effect of the Cas13d gRNAs on the translation of the reporters. (D) Immunoblot analysis of C9orf72 protein showed that the C9orf72 protein level was unaffected in HeLa RAN translation reporter cell lines stably expressing Cas13d-NT30, Cas13d-S24, and Cs13d-S30. (E) Transient cotransfection of CRISPR/Cas13d constructs with either the GA-frame, GP-frame, or GR-frame or the No-G4C2-repeat control construct in HEK293 cells showed that Cas13d-S24 and Cas13d-S30 significantly reduced both NanoLuc and firefly luciferase signals from the GA-, GP-, and GR-frame but not the negative No-G4C2-repeat control reporter compared with the non-targeting control Cas13d-NT30. (F) Immunoblot analysis of C9orf72 protein showed that the C9orf72 protein level was unaffected in HEK293 cells cotransfected with CRISPR/Cas13d and either the GA-frame, GP-frame, or GR-frame or the No-G4C2-repeat control construct. Data are presented as means ± SD of 3 or 4 biological replicates as indicated by the number of dots in each graph, and were analyzed with ordinary 1-way ANOVA with Dunnett’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3. Guided Cas13d suppressed RAN translation in cells derived from patients carrying the C9orf72 hexanucleotide repeat expansion.

(A) Schematic of RAN translation product detection in human iPSCs stably expressing Cas13d and gRNA via lentivirus transduction. (B) ELISA quantification in multiple C9-ALS patient iPSC cell lines showed significant reduction of poly-GP and poly-GA levels by Cas13d-S24 and CRISPR/S30 compared with the non-targeting control Cas13d-NT30. (C) Quantification of relative RNA levels of Cas13d in the C9-ALS patient iPSC cell lines showed variable Cas13d levels among lines, while in each line there were no significant differences among the S24, S30, and non-targeting NT30 groups. (D and E) Linear regression and correlation analyses showed a strong positive correlation between Cas13d expression level and poly-GP (D) and poly-GA (E) knockdown efficiency among C9-ALS patient iPSC lines. Pearson’s correlation coefficients and 2-tailed P value were computed. (F) Schematic of poly-GP and poly-GA detection in iMNs derived from human C9-ALS patient iPSCs. (G) ELISA quantification in iMN lines derived from multiple iPSC cell lines showed significant reduction in poly-GP and poly-GA levels by Cas13d-S24 and CRISPR/S30 compared with the non-targeting control Cas13d-NT30. (H) Quantification of relative RNA levels of Cas13d in the C9-ALS patient iMN lines showed variable Cas13d levels among lines, while in each line there were no significant differences among the S24, S30, and non-targeting NT30 groups. (I and J) Linear regression and correlation analyses showed a strong positive correlation between Cas13d expression level and poly-GP (I) and poly-GA (J) knockdown efficiency among C9-ALS patient iMN lines. Pearson’s correlation coefficients and 2-tailed P value were computed. Data are presented as means ± SD of 2–4 biological replicates as indicated by the number of dots in each graph, and were analyzed with ordinary 1-way ANOVA with Dunnett’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 4. The CRISPR/Cas13d system degraded targetable short but not long GGGGCC repeat RNA.

(A) Purified Cas13d showed a nearly 100% degradation efficiency of the target RNA r(NT24) with gRNA NT24 but not the other gRNAs, confirming the specificity and high cleavage activity of the Cas13d system. (B) The target RNA r(GGGGCC)2 was too short to be degraded by Cas13d. (C–E) The purified Cas13d showed partial degradation of the target RNAs r(GGGGCC)5 (C), r(GGGGCC)8 (D), and r(GGGGCC)12 (E) with gRNA S24 or S30 but not the other gRNAs, indicating a compromised cleavage activity of Cas13d targeting GGGGCC repeat RNAs and a trend of decreased cleavage efficiency with increased repeat lengths. (F) Cas13d was unable to degrade r(GGGGCC)28, demonstrating the limited activity of Cas13d to target or cleave longer GGGGCC repeat RNAs. The cleavage assay was performed in a buffer containing 0.3 μM of gRNA, 0.6 μM of Cas13d protein, and 40 ng/μL of target RNA. Data are presented as means ± SD of 3 independent experiments and were analyzed with unpaired 2-tailed Student’s t test (A) and ordinary 1-way ANOVA with Dunnett’s multiple-comparison test (B–F). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.001.

Figure 5. The CRISPR/Cas13d system suppressed RAN translation in C9orf72 repeat transgenic mice.

(A) Schematic of poly-GP and poly-GA detection in mice treated with AAV9 expressing Cas13d and gRNA. (B and C) Quantification showed decreased levels of poly-GP (B) or poly-GA (C) in C9-500 BAC mice but not in the WT mice, when the mice were treated with AAV9 expressing Cas13d-S24 or Cas13d-S30, compared with those treated with control AAV9 expressing the non-targeting Cas13d-NT30. The number of dots in each group indicates the number of mice in the corresponding group. Data are presented as means ± SD and were analyzed with unpaired 1-tailed Student’s t test. *P < 0.05, ****P < 0.0001.