Figure S1. Viral transduction of the AAV9-SynI vector in the mouse. Western blot analyses using an anti-Flag-tag antibody showed a widespread distribution of the AAV9-SynI vector in the central nervous system, including the spinal cord, cortex, cerebellum and olfactory bulb of AR2 mice intravenously injected with AAV9-Flag-hADAR2 (AAV). Saline; AR2 mice treated with saline.
Figure S2. Expression of AAV-delivered ADAR2 in Neuro2a cells and wild-type mouse brains. RT-PCR (A) and Western blot analyses (B) of ADAR2 demonstrated that hADAR2 was expressed in Neuro2a cells infected with AAV9-hADAR2 or AAV9-hADAR2E396A. (C) The editing efficiency at the CYFIP2 mRNA K/E site was significantly higher in Neuro2a cells infected with AAV9-hADAR2 compared with those infected with AAV9-GFP- or AAV9-hADAR2E396A (mean ± s.e.m.; n = 6 for each group). **p < 0.01 (Mann–Whitney U-test). D. Editing efficiencies at the Q/R site of GluA2 pre-mRNA and the K/E site of CYFIP2 mRNA in the brains of wild-type mice injected with AAV9-hADAR2E396A in the cerebral cortex. Editing efficiencies were significantly decreased at the injection site, but not in the remote ipsilateral (2.0–2.5 mm posterior to the injection site) or contralateral hemisphere. *p < 0.05, ***p < 0.0001 (Mann–Whitney U-test,). The editing efficiencies at these sites were not changed in the mouse brain area injected with AAV9-GFP. All error bars represent the s.e.m., n = 4–12.
Figure S3. Expression of AAV9-Flag-hADAR2 in the AR2 mouse spinal cord after intravenous injection. Immunohistochemistry for Flag (green) and ChAT (red) in the spinal cords of an AR2 mouse intravenously injected with AAV9-Flag-hADAR2. The boxed areas represent the area for high power magnification image of Fig 1D. Scale bar, 50 mm. (B and C) Magnified view of the dotted line areas in (A). Both ChAT-positive anterior horn cells (AHCs) (B) and ChAT-negative dorsal horn cells (DHCs) (C) exhibit Flag immunoreactivity. Scale bars, 20 mm. (D) Numbers of AHCs classified by immunoreactivity to Flag and ADAR2 in AR2 mice intravenously injected with AAR9-Flag-hADAR2 or saline. AR2 mice (n = 4 for each group) were examined at 33–36 weeks of age, ***p < 0.001 (Student's t-test). All error bars represent the s.e.m.
Figure S4. GFAP and MAC2 immunoreactivity in the spinal cords of wild-type mice injected with AAV9-GFP. Axial section of the spinal cord of wild type mice (21 weeks of age) 11 weeks after the intravenous injection with AAV9-GFP (7.2 × 1011 vg/body). There was little or virtually no increase of GFAP and MAC2 immunoreactivity in the spinal cord. There was no GFAP-positive astrocytes or MAC2-positive microglia around the GFP-positive neurons (bottom). Scale bars, 50 mm (upper panels) and 20 mm (bottom panels).
Figure S5. GFAP and MAC2 immunoreactivity in the spinal cord of AAV-injected AR2 mice. In contrast to the intense GFAP and MAC2 immunoreactivity in the spinal anterior horns (AH) of the saline-treated AR2 mice (Saline), there was virtually no increase of the immunoreactivity in the AH of AR2 mice intravenously injected with AAV9-Flag-hADAR2 (AAV). Scale bars, 50 mm (upper panels) and 20 mm (bottom panels).
Figure S6. Rotarod performance of AAV-injected AR2 mice. Rotarod performance was significantly better in AR2 mice intravenously injected with AAV9-Flag-hADAR2 at presymptomatic stages (n = 11; AAV) than in saline-treated AR2 mice (n = 13; Saline) (Student's t-test against Saline group, *p < 0.05, ***p < 0.0001). All error bars represent the s.e.m. (B) Injection of the AAV vector into AR2 mice that already exhibited declined rotarod performance prevented the progression of motor dysfunction (n = 5; AAV; Student's t-test against Saline group, **p < 0.01, ***p < 0.0001). All error bars represent the s.e.m.
Figure S7. Magnified view of the ventral roots of the fifth lumber segment (L5). Each panel is the ventral root of either saline-injected (Saline, n = 3) or AAV9-Flag-hADAR2-injected AR2 mice (AAV, n = 3). Arrowheads indicate axonal degeneration. Scale bar: 10 mm.
Figure S8. Effects on the expression of ADAR2 in the brain and spinal cord. The relative abundance of hADAR2 mRNA was significantly higher in the brains of AR2 mice injected with AAV9-Flag-hADAR2 (AAV) (n = 3) than in those of saline-injected AR2 mice (Saline) (n = 3), *p < 0.05 (Mann–Whitney U-test). All error bars represent the s.e.m. (B and C) Immunoblot analysis of brain (B) and spinal cord extracts (C) demonstrated that the expression of ADAR2 protein did not significantly differ between AAV and Saline groups.
Figure S9. Rescue of TDP-43 in motor neurons. Immunohistochemistry for TDP-43 (green) and Flag (red) in the right half of the spinal cord of AR2 mice intravenously injected with AAV9-Flag-hADAR2. The boxed areas represent the area for high power magnification image of Fig 5A. Scale bar, 50 mm. (B) There were AHCs devoid of TDP-43 immunoreactivity (arrowheads) and those exhibiting numerous TDP-43-positive cytoplasmic aggregates without nuclear TDP-43 immunoreactivity (double arrowheads) in the saline-treated AR2 mice (Saline). Flag-positive AHCs exhibited intense nuclear and faint cytoplasmic TDP-43 immunoreactivity (arrows) in AAV9-Flag-ADAR2-injected AR2 mice (AAV). Scale bar is 20 mm.
Figure S10. The signal intensity of TDP-43-positive and -negative AHCs. Immunohistochemistry for TDP-43 (green) and Flag (red) in the spinal cord. AHCs were recognized as large (≥20 μm diameter) and TO-PRO-3-positive (blue) cells in the AH. The panels represent the enlarged image of Fig 5A. The profile images showed fluorescence intensity of TDP-43 along the yellow line. In Fig 5C, we designated AHCs as TDP-43-positive when the signal intensity was more than threefold higher (>60) than the background (average ± s.e.m., 8.80 ± 0.12; max 20).
Table S1. Primer sequences.
Table S2. TaqMan probes used for quantitative polymerase chain reaction.