Figure 2. CRISPR/Cas9 manipulations enable rapid and highly efficient in vivo deletions of Clstn3 in Purkinje cells.

(A) Schematic of the sgRNA design strategy. Both sgRNAs target the positive strand of DNA, with sg66 targeting exon2, and sg21 targeting exon3. (B) Schematic of the AAV-DJ expression cassette in which sgRNAs and tdTomato (tdT) synthesis are driven by U6 and CAG promoters, respectively. Control mice were infected with AAVs that lacked sgRNAs but were otherwise identical. (C) Experimental strategy for CRISPR-mediated acute Clstn3 deletions in the cerebellum. AAVs expressing the sgRNAs and tdTomato were stereotactically injected into the cerebellum of constitutively expressing Cas9 mice at P21, and mice were analyzed after P50. (D) Quantitative RT-PCR shows that the CRISPR-mediated Clstn3 deletion severely suppresses Clstn3 mRNA levels in the total cerebellum. Relative gene expression levels were first normalized to GAPDH using threshold cycle (CT) values, and then normalized to control. (E) Immunoblotting analyses confirm that the CRISPR-mediated deletion greatly suppresses Clstn3 protein levels in the overall cerebellum (left, representative immunoblot; right, summary graph of quantifications using fluorescently labeled secondary antibodies). (F & G) Representative images of a sagittal cerebellar section from a mouse that was stereotactically infected with AAVs as described in C (F, overview of the cerebellum; G, cerebellar cortex; red = AAV-encoded tdTomato; blue, DAPI). Note that AAVs infect all Purkinje cells but few granule cells or inhibitory neurons. Data in panels D and E are means ± SEM. Statistical analyses were performed using double-tailed unpaired t-test for D and E (***p < 0.001). Numbers of animals for each experiment are indicated in the graphs.

Figure 2—figure supplement 1. Predicted genome editing patterns by the sgRNAs used for the Clstn3 KO in the current study and analysis of potential off-target effects of the sgRNAs using genomic sequencing of targeted Clstn3 KO cerebellum.

(A & B) Primer design strategies for RT-qPCR detecting Clstn3 KO efficiency (A) and off-target effects (B). (C & D) Predicted genome editing effects for sg66 (A) and sg21 (B). (E & F) The three top-ranked potential off-target sites for sgRNA66 as identified by sequence predictions from the design website (https://zlab.bio/guide-design-resources) (C), and their analysis by genomic sequencing of Clstn3 KO cerebellum, demonstrating that CRISPR with sgRNA66 does not detectably mutant these three sites (D). Arrow indicates potential cutting positions. (G & H) Same as C and D, but for sgRNA21.

Figure 2—figure supplement 2. The more distal cortex of the cerebellum is also infected by the AAVs that were stereotactically injected, but the expression levels are lower like due to a lower virus titer in areas more distant to the injection site.

(A & B) Overview of the infected cerebellum (A) and zoomed-in images from more distal areas of the cerebellar cortex (B). These images complement those seen in Figure 2F to illustrate the weak but clearly detectable tdTomato expression even in the distal cerebellar cortex. Note that the tdTomato expression directly correlates with the viral load because tdTomato is encoded by the viruses directly and not produced by a Cre-dependent recombination event (see Figure 2B).