Research ArticleNeurodegenerative Disease

An miRNA-mediated therapy for SCA6 blocks IRES-driven translation of the CACNA1A second cistron

See allHide authors and affiliations

Science Translational Medicine  13 Jul 2016:
Vol. 8, Issue 347, pp. 347ra94
DOI: 10.1126/scitranslmed.aaf5660
  • Fig. 1. Somatic gene transfer of CACNA1A IRES–driven α1ACTSCA6 causes Purkinje cell degeneration in mice.

    (A) Representative immunofluorescence images of AAV9-injected mouse brain and cerebellum from 4-week-old mice stained with FLAG-specific antibodies and fluorescent secondary antibody. Scale bar, 5 mm. (B) Relative α1ACT-FLAG immunofluorescence intensities in AAV9-injected mouse hippocampus, cerebral cortex, and cerebellum (n = 6). Relative α1ACT-FLAG immunofluorescence is expressed as signal intensities per unit area quantified by the National Institutes of Health (NIH) ImageJ software. n.s., not significant. (C) Relative α1ACT mRNA expression in the cerebellum of AAV9-injected mice at 4 weeks of age (n = 4). (D) Western blot and densitometric analyses showing α1ACT-Q11-FLAG and α1ACT-Q33-FLAG expression in the cerebellum of AAV9-injected mice at 4 weeks of age (n = 4). (E) Representative immunofluorescence images of AAV9-injected mouse cerebellum from 4-week-old mice stained with calbindin-specific antibodies and fluorescent secondary antibody. Scale bar, 100 μm. (F to H) Molecular layer thickness (F), density of dendritic tree (G), and Purkinje cell count of AAV9-injected mouse cerebellum from 4-week-old mice (n = 6) (H). GFP, AAV9-GFP mice; Q11, AAV9-α1ACT-Q11 mice; Q33, AAV9-α1ACT-Q33 mice. All data represent means ± SEM. **P < 0.01. Student’s t test in (B) to (D). One-way analysis of variance (ANOVA) in (F) to (H).

  • Fig. 2. CACNA1A IRES–driven α1ACTSCA6 causes an early-onset ataxia and motor deficits in mice.

    (A to E) Clinical features of AAV9-injected mice. Body weight (A), rotarod test (B), and open-field assay (C) at 4 weeks of age. Stride length (D) and stride frequencies (E) assessed by DigiGait analysis at 4 weeks of age (n = 12, 6 male and 6 female mice per group). GFP, AAV9-GFP mice; Q11, AAV9-α1ACT-Q11 mice; Q33, AAV9-α1ACT-Q33 mice. All data represent means ± SEM. *P < 0.05, **P < 0.01. Two-way ANOVA in (A) and (B). One-way ANOVA in (C) to (E).

  • Fig. 3. miR-3191-5p inhibits the CACNA1A IRES–driven translation of α1ACT while sparing α1A and CACNA1A mRNA expression in HEK293 cells.

    (A) Schematic representation of a bicistronic control reporter vector and a bicistronic CACNA1A IRES reporter vector. SV40, simian virus 40; polyA, polyadenylation signal sequence. (B) Relative dual-luciferase CACNA1A IRES reporter activities in HEK293 cells treated with miR-711, miR-3191-5p, miR-4786-3p, or an miRNA negative control (NC) (n = 6). The ratio of firefly luciferase to Renilla luciferase activities of a bicistronic CACNA1A IRES reporter vector was normalized to that of a bicistronic control reporter vector. (C and D) Western blot and densitometric analyses showing α1A-FLAG and α1ACT-FLAG expression in HEK293 cells treated with three miRNAs or NC (n = 6). (C)CACNA1A-Q11. (D) CACNA1A-Q33. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. (E and F) Relative CACNA1A-Q11 (E) and CACNA1A-Q33 (F) mRNA expression in HEK293 cells treated with three miRNAs or NC (n = 6). (G) Relative endogenous CACNA1A mRNA expression in HEK293 cells treated with miR-3191-5p or NC (n = 6). All data represent means ± SEM. **P < 0.01, ***P < 0.001. Student’s t test in (B) to (G).

  • Fig. 4. Ago4 is required for miR-3191-5p–mediated inhibition of CACNA1A IRES–driven α1ACT translation.

    (A) Relative dual-luciferase CACNA1A IRES reporter activities in HEK293 cells treated with si-Ago1, si-Ago2, si-Ago3, si-Ago4, or a scrambled control siRNA (si-C) in the presence of miR-3191-5p or an miRNA NC (n = 6). The ratio of firefly luciferase to Renilla luciferase activities of a bicistronic CACNA1A IRES reporter vector was normalized to that of a bicistronic control reporter vector. (B) Western blot and densitometric analyses showing α1A-Q33-FLAG and α1ACT-Q33-FLAG expression in HEK293 cells treated with four siRNAs or si-C in the presence of miR-3191-5p or NC (n = 6). (C and D) Amount of miR-3191-5p (C) and CACNA1A-Q33 (D) mRNA binding to Ago1 to Ago4 that was immunoprecipitated using antibodies against Ago1 to Ago4 or a control immunoglobulin G (IgG) (n = 6). (E) Relative amount of endogenous miR-3191-5p in HEK293 cells treated with si-Ago4 or si-C (n = 6). All data represent means ± SEM. **P < 0.01, ***P < 0.001. Student’s t test in (A) to (E).

  • Fig. 5. miR-3191-5p inhibits the translational initiation of CACNA1A IRES–driven α1ACT by eIF4AII and eIF4GII.

    (A to D) Relative dual-luciferase CACNA1A IRES reporter activities in HEK293 cells treated by overexpression of either eIF4AI (AI) (A), eIF4AII (AII) (B), eIF4GI (GI) (C), eIF4GII (GII) (D), or a control vector (C) in the presence of miR-3191-5p or an miRNA NC (n = 6). The ratio of firefly luciferase to Renilla luciferase activities of a bicistronic CACNA1A IRES reporter vector was normalized to that of a bicistronic control reporter vector. (E and F) Western blot and densitometric analyses showing α1A-Q33-FLAG and α1ACT-Q33-FLAG expression in HEK293 cells treated by overexpression of either AII (E), GII (F), or C in the presence of miR-3191-5p or NC (n = 6). (G and H) Relative CACNA1A-Q33 mRNA expression in HEK293 cells treated by overexpression of either AII (G), GII (H), or C in the presence of miR-3191-5p or NC (n = 6). (I and J) Amount of IRES-α1ACT-Q33 mRNA binding to AII (I), GII (J), or a control IgG in the presence of vectors expressing miR-3191-5p (miR-3191) or an miRNA control (miR-C) that was immunoprecipitated using antibodies against AII, GII, or a control IgG (n = 6). (K) Immunoprecipitation (IP) from HEK293 cells transfected with AII- and GII-expressing vectors using an antibody against AII, GII, or a control IgG. Recovered proteins were analyzed using Western blot. Input, 10%. (L) Relative dual-luciferase CACNA1A IRES reporter activities in HEK293 cells treated with si-AII, si-GII, or a scrambled si-C (n = 6). The ratio of firefly luciferase to Renilla luciferase activities of a bicistronic CACNA1A IRES reporter vector was normalized to that of a bicistronic control reporter vector. All data represent means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. Student’s t test in (A) to (F), (I), (J), and (L). One-way ANOVA in (G) and (H).

  • Fig. 6. AAV9-mediated therapeutic delivery of miR-3191-5p blocks CACNA1A IRES–driven α1ACT translation in mice.

    (A and B) Representative immunofluorescence images of AAV9-injected mouse brain and cerebellum from 4-week-old mice stained with GFP-specific (A) and FLAG-specific (B) antibodies and fluorescent secondary antibody. Scale bar, 5 mm. (C and D) Relative GFP (C) and α1ACT-FLAG (D) immunofluorescence intensities in AAV9-injected mouse hippocampus, cerebral cortex, and cerebellum (n = 6). Relative GFP and relative FLAG immunofluorescence are expressed as signal intensities per unit area quantified by NIH ImageJ software. (E) Relative α1ACT mRNA expression in the cerebellum of AAV9-injected mice at 4 weeks of age (n = 4). (F) Western blot and densitometric analyses showing α1ACT-Q33-FLAG expression in the cerebellum of AAV9-injected mice at 4 weeks of age (n = 4). (G and H) Amount of miR-3191-5p (G) and α1ACT-Q33 (H) mRNA binding to Ago4 that was immunoprecipitated using an antibody against Ago4 or a control IgG in the cerebellum of AAV9–Q33–miR-3191-5p mice at 4 weeks of age (n = 6). GFP, AAV9-GFP mice; mock, AAV9–Q33–miR-mock mice; 3191, AAV9–Q33–miR-3191-5p mice. All data represent means ± SEM. **P < 0.01. One-way ANOVA in (C). Student’s t test in (D) to (H).

  • Fig. 7. miR-3191-5p prevents Purkinje cell degeneration caused by CACNA1A IRES–driven α1ACTSCA6 in mice.

    (A) Representative immunofluorescence images of AAV9-injected mouse cerebellum from 4-week-old mice stained with calbindin-specific antibodies and fluorescent secondary antibody. Scale bar, 100 μm. (B to D) Molecular layer thickness (B), density of dendritic tree (C), and Purkinje cell counts (D) in AAV9-injected mouse cerebellum from 4-week-old mice (n = 6). GFP, AAV9-GFP mice; mock, AAV9–Q33–miR-mock mice; 3191, AAV9–Q33–miR-3191-5p mice. All data represent means ± SEM. **P < 0.01. Student’s t test in (B) to (D).

  • Fig. 8. AAV9–miR-3191-5p prevents the ataxia and motor deficits caused by CACNA1A IRES–driven α1ACTSCA6 in mice.

    (A to E) Clinical features of AAV9-injected mice. Body weight (A), rotarod test (B), and open-field assay (C) at 4 weeks of age. Stride length (D) and stride frequencies (E) assessed by DigiGait analysis at 4 weeks of age (n = 12, 6 male and 6 female mice per group). GFP, AAV9-GFP mice; mock, AAV9–Q33–miR-mock mice; 3191, AAV9–Q33–miR-3191-5p mice. All data represent means ± SEM. *P < 0.05, **P < 0.01. Two-way ANOVA in (A) and (B). Student’s t test in (C) to (E).

Supplementary Materials

  • www.sciencetranslationalmedicine.org/cgi/content/full/8/347/347ra94/DC1

    Fig. S1. Schematic representation of AAV9 vectors and AAV9 transduction efficiency into Purkinje cells.

    Fig. S2. Histopathological examination of the AAV9-injected mouse cerebral cortex and hippocampus.

    Fig. S3. The long-term follow-up of AAV9-injected mouse behavioral phenotypes and CACNA1A IRES–driven α1ACT expression in mice.

    Fig. S4. Stem-loop structure of CACNA1A IRES and the predicted binding sites of miR-711, miR-3191-5p, and miR-4786-3p.

    Fig. S5. Effects of miR-3191-5p on mutated CACNA1A IRES templates.

    Fig. S6. Effects of eIF4AII and eIF4GII on mutated CACNA1A IRES templates in the presence or absence of miR-3191-5p.

    Fig. S7. Cotransduction of AAV9-α1ACT-Q33 with either AAV9–miR-mock or AAV9–miR-3191-5p.

    Fig. S8. The long-term follow-up of therapeutic effect of miR-3191-5p on mouse behavioral phenotypes and CACNA1A IRES–driven α1ACTSCA6 expression in mice.

    Fig. S9. Potential mouse mRNAs targeted by human miR-3191-5p.

    Fig. S10. Histopathological examination of the brain, cerebellum, heart, lung, liver, and kidney of wild-type mice injected with AAV9–miR-3191-5p.

    Table S1. Source data.

    Movie S1. DigiGait video of an 8-week-old AAV9-GFP mouse on treadmill.

    Movie S2. DigiGait video of an 8-week-old AAV9-α1ACT-Q11 mouse on treadmill.

    Movie S3. DigiGait video of an 8-week-old AAV9-α1ACT-Q33 mouse on treadmill.

    Movie S4. DigiGait video of an 8-week-old AAV9–Q33–miR-mock mouse on treadmill.

    Movie S5. DigiGait video of an 8-week-old AAV9–Q33–miR-3191-5p mouse on treadmill.

  • Supplementary Material for:

    An miRNA-mediated therapy for SCA6 blocks IRES-driven translation of the CACNA1A second cistron

    Yu Miyazaki, Xiaofei Du, Shin-ichi Muramatsu, Christopher M. Gomez*

    *Corresponding author. E-mail: cgomez{at}neurology.bsd.uchicago.edu

    Published 13 July 2016, Sci. Transl. Med. 8, 347ra94 (2016)
    DOI: 10.1126/scitranslmed.aaf5660

    This PDF file includes:

    • Fig. S1. Schematic representation of AAV9 vectors and AAV9 transduction efficiency into Purkinje cells.
    • Fig. S2. Histopathological examination of the AAV9-injected mouse cerebral cortex and hippocampus.
    • Fig. S3. The long-term follow-up of AAV9-injected mouse behavioral phenotypes and CACNA1A IRES–driven α1ACT expression in mice.
    • Fig. S4. Stem-loop structure of CACNA1A IRES and the predicted binding sites of miR-711, miR-3191-5p, and miR-4786-3p.
    • Fig. S5. Effects of miR-3191-5p on mutated CACNA1A IRES templates.
    • Fig. S6. Effects of eIF4AII and eIF4GII on mutated CACNA1A IRES templates in the presence or absence of miR-3191-5p.
    • Fig. S7. Cotransduction of AAV9-α1ACT-Q33 with either AAV9–miR-mock or AAV9–miR-3191-5p.
    • Fig. S8. The long-term follow-up of therapeutic effect of miR-3191-5p on mouse behavioral phenotypes and CACNA1A IRES–driven α1ACTSCA6 expression in mice.
    • Fig. S9. Potential mouse mRNAs targeted by human miR-3191-5p.
    • Fig. S10. Histopathological examination of the brain, cerebellum, heart, lung, liver, and kidney of wild-type mice injected with AAV–miR-3191-5p.
    • Legend for table S1
    • Legends for movies S1 to S5

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Table S1 (Microsoft Excel format). Source data.
    • Movie S1 (.mov format). DigiGait video of an 8-week-old AAV9-GFP mouse on treadmill.
    • Movie S2 (.mov format). DigiGait video of an 8-week-old AAV9-α1ACT-Q11 mouse on treadmill.
    • Movie S3 (.mov format). DigiGait video of an 8-week-old AAV9-α1ACT-Q33 mouse on treadmill.
    • Movie S4 (.mov format). DigiGait video of an 8-week-old AAV9–Q33–miRmock mouse on treadmill.
    • Movie S5 (.mov format). DigiGait video of an 8-week-old AAV9–Q33–miR- 3191-5p mouse on treadmill.

Stay Connected to Science Translational Medicine

Navigate This Article