Elimination of Toxic Microsatellite Repeat Expansion RNA by RNA-Targeting Cas9

Abstract

Microsatellite repeat expansions in DNA produce pathogenic RNA species that cause dominantly inherited diseases such as myotonic dystrophy type 1 and 2 (DM1/2), Huntington's disease, and C9orf72-linked amyotrophic lateral sclerosis (C9-ALS). Means to target these repetitive RNAs are required for diagnostic and therapeutic purposes. Here, we describe the development of a programmable CRISPR system capable of specifically visualizing and eliminating these toxic RNAs. We observe specific targeting and efficient elimination of microsatellite repeat expansion RNAs both when exogenously expressed and in patient cells. Importantly, RNA-targeting Cas9 (RCas9) reverses hallmark features of disease including elimination of RNA foci among all conditions studied (DM1, DM2, C9-ALS, polyglutamine diseases), reduction of polyglutamine protein products, relocalization of repeat-bound proteins to resemble healthy controls, and efficient reversal of DM1-associated splicing abnormalities in patient myotubes. Finally, we report a truncated RCas9 system compatible with adeno-associated viral packaging. This effort highlights the potential of RCas9 for human therapeutics.

Keywords: ALS; CRISPR; Huntington’s disease; RNA-targeting Cas9; adeno-associated virus; gene therapy; microsatellite repeat expansion; minimal Cas9; myotonic dystrophy.

Figure 1. Visualizing Microsatellite Repeat Expansion RNAs with CRISPR/Cas9

(A) Schematic description of recognition and imaging of microsatellite repeat expansion RNA with low levels of RNA-targeting Cas9 (RCas9) fused to EGFP. (B) RNA-FISH and RCas9 (EGFP) imaging in COS-M6 cells transfected with (CTG)¹⁰⁵, (CCTG)³⁰⁰, or (CAG)⁸⁰. Each pair of rows features sgRNAs targeting the specified repeat (+) or a non-targeting sgRNA (NT). Scale bars are 10 μm. See also Figure S1 and Tables S1 and S2.

Figure 2. Degradation of Microsatellite Repeat Expansion RNA with RNA-Targeting Cas9

(A) Schematic description of elimination of microsatellite repeat expansion RNA with RNA-targeting Cas9 (RCas9) fused to EGFP or PIN domain. (B) CUG RNA foci measured by RNA-FISH in COS-M6 cells transfected with (CTG)¹⁰⁵, either non-targeting sgRNA (NT), CUG-targeting sgRNA (+), or no sgRNA (−), and with (+) or without (−) HA-tagged PIN-dCas9. Scale bars in (B)–(E) are 10 μm. (C) CUG RNA foci measured by RNA-FISH in COS-M6 cells transfected with (CTG)¹⁰⁵ and PIN-dCas9, with either non-targeting sgRNA (NT) or CUG-targeting sgRNA (+), and with (+) or without (−) cognate PAMmer. (D) CCUG RNA foci measured by RNA-FISH in COS-M6 cells transfected with (CCTG)³⁰⁰ and PIN-dCas9 or dCas9, with either non-targeting sgRNA (NT) or CCUG-targeting sgRNA (+). (E) GGGGCC RNA foci measured by RNA-FISH in COS-M6 cells transfected with (GGGGCC)¹²⁰ and PIN-dCas9 or dCas9, with either non-targeting sgRNA (NT) or GGGGCC-targeting sgRNA (+). (F) Quantification of RNA-FISH signal in COS-M6 cells transfected with various MREs and PIN-dCas9 or dCas9 with MRE-targeting (+) or non-targeting (NT) sgRNA. Cells containing at least 1 RNA focus are considered positive for MRE RNA. Measurements are normalized to the condition with the MRE-targeting sgRNA and MRE RNA but lacking dCas9. Error bars denote SDs determined from 3 biological replicates enumerating 100 transfected cells each. (G) RNA dot blot of (CUG)^exp levels in COS-M6 cells transfected with (CTG)¹⁰⁵, CTG-targeting or non-targeting (NT) sgRNA, and various forms of Cas9 (PIN-dCas9, dCas9-GFP, and wtCas9). U6 snRNA served as a loading control in (G)–(I). (H) RNA dot blot of (CCUG)^exp levels in COS-M6 cells transfected with (CCTG)¹⁰⁵, CCTG-targeting or non-targeting (NT) sgRNA, and various forms of Cas9 (PIN-dCas9, dCas9-GFP, and wtCas9). (I) RNA dot blot of (GGGGCC)^exp levels in COS-M6 cells transfected with (GGGGCC)¹⁰⁵, GGGGCC-targeting or non-targeting (NT) sgRNA, and various forms of Cas9 (PIN-dCas9, dCas9-GFP, and wtCas9). See also Figures S2 and S3 and Tables S1 and S2.

Figure 3. RCas9 Binds and Targets CUG Repeat Expansion RNA for Degradation

(A) Gel shift assay of in vitro transcribed radiolabeled (CUG)¹² RNA and increasing doses of COS-M6 cellular extract (0, 1.25, 2.5, 5, 10, 20, or 40 μg of total cellular protein) containing dCas9-GFP and sgRNAs (1.5 μg transfected plasmid per 1.5 million cells) targeting this repetitive RNA (CUG sgRNA) or a non-targeting control (NT-sgRNA) with 40 μg cell extract. The rightmost lane includes a GFP antibody (40 μg cell extract). (B) Immunoprecipitation and dot blot quantitation of radiolabeled in vitro transcribed (CUG)⁵⁴ after incubation with extract from cells expressing dCas9-GFP and sgRNAs targeting this repetitive RNA (CUG) or a non-targeting control (NT). (C) In vitro cleavage assay of radiolabeled (CUG)¹² RNA combined with increasing concentrations of extract (10, 20, 40 μg) from COS-M6 cells expressing dCas9-GFP or PIN-dCas9 and a CUG-targeting sgRNA or nontargeting (NT) sgRNA, as indicated. (D) RNA dot blot assay of CUG^exp RNA levels in COS-M6 cells transfected with plasmids encoding (CTG)¹⁰⁵, and the indicated amounts of dCas9-GFP (top) or PIN-dCas9 (middle). U6 snRNA served as a loading control. Bottom: densitometric quantification. (E) Scheme describing the relative activity of dCas9-GFP and PIN-dCas9 in the context of CUG microsatellite repeat expansion (MRE) RNA elimination as observed in (D). (F) Scheme describing a tetracycline-inducible expression system for (CTG)⁹⁶⁰ repeat expansion in DMPK to assess whether RCas9-mediated repeat expansion RNA can be eliminated in the absence of transcription. TRE is the tetracycline responsive element. (G) Detection of CUG^exp RNA foci by RNA-FISH in COS-M6 cells transfected and treated with doxycycline according to (E). Scale bars are 20 μm. (H) Quantification of CUG^exp-FISH signal in COS-M6 cells transfected and treated with doxycycline according to (E). Cells containing at least 1 RNA focus were considered positive for CUG repeat RNA. Measurements were normalized to the total number of GFP-positive cells. Error bars denote SDs determined from 3 biological replicates enumerating 100 GFP-positive cells each. (I) Quantification of CUG^exp containing DMPK transcript levels in COS-M6 cells transfected and treated with doxycycline according to (E) using RT-qPCR. DMPK mRNA levels were normalized to GAPDH and expressed as fold change over no-doxycycline condition. Error bars denote SDs determined from 3 biological replicates. (J) RNA blot analysis quantifying CUG^exp containing DMPK transcript levels in COS-M6 cells transfected with the tetracycline-inducible expression system for (CTG)⁹⁶⁰ and treated with doxycycline according to (E). The legend from (I) also applies to this panel. See also Figure S2 and Tables S1 and S2.

Figure 4. Truncated Forms of Cas9 Maintain the Ability to Eliminate Repeat Expansion RNAs

(A) The domain structure of full-length (FL) Cas9 and three truncation constructs lacking the HNH domain (ΔHNH), the REC2 and HNH domain (ΔHNH, ΔREC2), or composed of the REC-lobe only (REC-only). (B) RNA dot blot assay of (CUG)^exp levels in COS-M6 cells transfected with (CTG)¹⁰⁵, CUG-targeting (CUG) or non-targeting (NT) sgRNA, and full-length, ΔHNH, or REC-only dCas9 fused to a PIN domain. (C) RNA dot blot assay of (CUG)^exp levels in COS-M6 cells transfected with (CTG)¹⁰⁵, CUG-targeting (CUG) or non-targeting (NT) sgRNA, and ΔHNH, ΔREC2 dCas9 truncation with and without a fused PIN RNA endonuclease domain. (D) RNA dot blot assay of (CCUG)^exp levels in COS-M6 cells transfected with (CCTG)¹⁰⁵, CCUG-targeting or non-targeting (NT) sgRNA, and ΔHNH fused to PIN compared to other forms of Cas9 (PIN-dCas9, dCas9, and wtCas9). (E) RNA dot blot assay of (CAG)^exp levels in COS-M6 cells transfected with (CAG)¹⁰⁵, CAG-targeting (CAG) or non-targeting (NT) sgRNA, and ΔHNH fused to PIN compared to other forms of Cas9 (PIN-dCas9, dCas9, and wtCas9). See also Table S2.

Figure 5. Degradation of CUG^exp RNA in DM1 Patient Myoblasts and CCUG^exp RNA in DM2 Patient Fibroblasts

(A) RNA-FISH with (CAG)¹⁰ probes recognizing (CUG)^exp RNA and HA immunofluorescence in control and DM1 primary myoblasts transduced with no sgRNA (−), non-targeting sgRNA (NT), or CUG-targeting sgRNA (CUG) with (+) or without (−) PIN-dCas9-HA. Scale bars in (A) and (B) are 20 μm. (B) RNA-FISH with (CAGG)¹⁰ probes recognizing (CCUG)^exp RNA and HA immunofluorescence in control and DM2 primary fibroblasts transduced with non-targeting sgRNA (NT), or CCUG-targeting (CCUG) sgRNA with PIN-dCas9-HA. (C) Quantification of RNA foci in patient-derived myoblast cells transduced with non-targeting (NT) or MRE-targeting sgRNAs (+) and PIN-dCas9. Cells containing at least one RNA focus are considered positive for MRE RNA. Error bars denote SDs determined from 3 biological replicates counting 100 patient cells each. (D) Dot blot assay to detect (CUG)^exp in control and DM1 primary myoblasts transduced with non-targeting sgRNA (NT) or CUG-targeting sgRNA (CUG) and PIN-dCas9. (E) RNA dot blot assay of (CUG)^exp RNA levels in DM1 patient myoblasts transduced with CTG-targeting sgRNA or template strand (CAG)-targeting sgRNAs, andCas9-GFP. (F) RIP-PCR for PIN-dCas9 in DM1 patient cells. The fold enrichment for each primer pair (DMPK, TCF4, AR, and intergenic) was calculated for the CUG-targeting sgRNA relative to the non-targeting sgRNA. See also Figure S2 and Tables S1 and S2.

Figure 6. MRE Elimination Reverses Downstream Molecular Phenotypes Associated with DM1 and HD

(A) RNA-FISH and EGFP fluorescence in COS-M6 cells transfected with plasmid (CTG)¹⁰⁵, MBNL1-EGFP, and non-targeting or CTG-targeting sgRNA. Scale bars in (A) and (B) are 20 μm. (B) RNA-FISH and MBNL1 immunofluorescence in primary DM1 patient myoblasts transduced with PIN-dCas9 and a non-targeting (NT) sgRNA (top) or CUG-targeting sgRNA (bottom). (C) Quantification of MBNL1 nuclear foci after RCas9 treatment with either non-targeting (NT) or CUG-targeting (CUG) sgRNA in primary DM1 patient myoblasts. Control myoblasts from an unaffected individual are shown for comparison. Error bars denote SD determined from three biological replicates of 100 transfected cells each. (D) RT-PCR to evaluate alternative splicing in primary DM1 patient myoblasts transduced with PIN-dCas9 and non-targeting (NT) or CUG-targeting sgRNA. (E) Western blot for polyglutamine protein (polyQ) from extracts COS-M6 cells transfected with (+) or without (−) (CAG)⁸⁰ plasmid, and either with (+) or without (−) CAG-targeting sgRNA or non-targeting sgRNA (NT) and the indicated dCas9 constructs. See also Figures S2 and S6 and Tables S1, S2, and S5.

Figure 7. CUG-Targeting RCas9 Corrects Transcriptome-wide DM1-Related Alternative Splicing Defects

(A) Hierarchical clustered heatmap of exon inclusion indices of top 350 differential AS events in control and DM1 myoblasts. (B) Hierarchical clustered heatmap of exon inclusion indices of top 350 differential AS events in control and DM1 myotubes. (C) Scatterplot showing exon inclusion or splicing indices (SI) of top 350 differential AS events of control and DM1 myotubes transduced with non-targeting sgRNA (NT) and PIN-dCas9. (D) Scatterplot showing exon inclusion or splicing indices (SI) of top 350 differential AS events of control and DM1 myotubes transduced with CUG-targeting sgRNA (CUG) and PIN-dCas9. (E) Pie charts showing complete and partial/no reversal of top 350 DM1-related AS events after treatment with CUG-sgRNA and Cas9-PIN in myoblasts (left) and myotubes (right). (F) RNA-seq data represented on UCSC genome-browser tracks of MBNL2 exon 6 (left) and FGFR1 exon 3 (right) showing DM1-related mis-splicing events in myoblasts or myotubes from unaffected individuals (ctrl) or DM1 patients (DM1), transduced with lentivirus expressing dCas9-PIN and either non-targeting (NT) or CUG-targeting sgRNA (CUG), or GFP only, as indicated. (G) RT-PCR of candidate AS events uncovered by RNA-seq analysis. (H) Expression levels of the indicated muscle differentiation markers evaluated by RNA-seq in myotubes differentiated from DM1 patient or healthy control myoblasts treated with non-targeting or CUG-targeting sgRNAs. (I) Schematic description of the therapeutic mechanism of the RCas9 system in the context of myotonic dystrophy type I. See also Figures S2, S4, and S5 and Tables S2–S5.