Fig 2: AAV9-UBA1 gene therapy corrects UPS perturbations and increases FL-SMN mRNA and protein levels.(A–D) Western blot analysis of ß-catenin, polyubiquitin, and monoubiquitin protein levels in uninjected spinal muscular atrophy (SMA) and adeno-associated virus serotype 9–ubiquitin-like modifier activating enzyme 1 (AAV9-UBA1) SMA mouse hearts at P7 (n = 3 mice per treatment group; unpaired 2-tailed Student’s t test). Lanes were run on the same gel but were noncontiguous. (E–G) Western blot analysis of UBA1 and survival motor neuron (SMN) protein levels in uninjected control, uninjected SMA, AAV9-UBA1–treated control, and AAV9-UBA1–treated SMA P7 hearts (n = 3 mice per group; 1-way ANOVA with Tukey’s post-hoc test). (H) PCR products following mouse, human, and UBA1 viral cDNA amplification using primers that detect full-length SMN (FL-SMN) (top row) or ?7-SMN (bottom row). (I) Significant increase in FL-SMN, but not ?7-SMN, mRNA expression in the hearts of AAV9-UBA1–treated mice at P7, as detected by qRT-PCR quantification using the primers shown in H (n = 3 mice per treatment group; unpaired 2-tailed Student’s t test). ns (not significant) P > 0.05, *P = 0.05, **P = 0.01, ***P = 0.005, ****P = 0.001.

Fig 3: AAV9-UBA1 gene therapy delivers safe systemic increases in UBA1 expression.(A and B) Adeno-associated virus serotype 9–ubiquitin-like modifier activating enzyme 1 (AAV9-UBA1) delivered at P1 significantly increased UBA1 protein levels in spinal cord, gastrocnemius muscle, heart, liver, lung, and kidney in P7 control mice (n = 3 mice per group; unpaired 2-tailed Student’s t test). Lanes were run on the same gel but were noncontiguous. (C) PCR products following mouse, human, and UBA1 viral cDNA amplification using primers detecting both mouse Uba1 and human UBA1 cDNA (top row), unique to mouse Uba1 cDNA (middle row), and unique to human UBA1 cDNA (bottom row). (D) Significant increase in AAV-expressed human UBA1 mRNA, but not endogenous mouse Uba1 mRNA, in the hearts of P7 AAV9-UBA1–treated mice (n = 3 mice per group; unpaired, 2-tailed Student’s t test). (E) Representative Nissl-stained spinal cords (ventral horn) from uninjected control and AAV9-UBA1–treated control mice at P9 (scale bar: 250 µm). (F) No significant change in the number of motor neurons in the spinal cords of AAV9-UBA1–treated control P9 mice (n = 4 mice per group; unpaired, 2-tailed Student’s t test). (G) Representative confocal micrographs of neuromuscular junctions (NMJs) in the external oblique from uninjected control and AAV9-UBA1–treated control mice at P9 (axonal inputs in green; postsynaptic endplates in red; scale bar: 25 µm). (H) No significant change in axonal inputs at the NMJ in the external oblique muscle of AAV9-UBA1–treated control mice at P9 (n = 4 mice per group; unpaired, 2-tailed Student’s t test). (I) Uninjected control and AAV9-UBA1–treated control mice at P180 (scale bar: 2 cm). (J–L) No change in AAV9-UBA1–treated controls from P1 to P180 with respect to (J) weight (at P180 n = 4 mice per group), (K) survival (Kaplan-Meier survival analysis), or (L) righting reflex test performance (from P1 to P12–P14; unpaired, 2-tailed Student’s t test at each time point). ns (not significant) P > 0.05, *P = 0.05, **P = 0.01, ***P = 0.005, ****P = 0.001. As part of a different comparative analysis, uninjected littermate control data concerning motor neuron cell body counts (F) and NMJ axonal input (H) is also shown in Figure 5, B and D, respectively. Weight, survival and righting times (J–L) for uninjected controls are also shown in Figure 4, A–C.

Fig 4: Rescue of neuromuscular and systemic pathology in SMA mice treated with AAV9-UBA1.(A) Nissl-stained spinal cords (ventral horn) from uninjected control, uninjected spinal muscular atrophy (SMA), and adeno-associated virus serotype 9–ubiquitin-like modifier activating enzyme 1–treated (AAV9-UBA1–treated) SMA mice at P9 (scale bar: 250 µm). (B) Significant rescue of motor neurons in AAV9-UBA1–treated SMA mice at P9 (n = 4 mice per group). (C) Confocal micrographs of neuromuscular junctions (NMJs) in the external oblique from uninjected control, uninjected SMA, and AAV9-UBA1–treated SMA mice at P9 (axonal inputs in green; postsynaptic endplates in red; scale bar: 25 µm). (D) Significant improvement in NMJ pathology in the external oblique muscle of AAV9-UBA1–treated SMA mice at P9 (n = 4 mice per group). (E) Bright-field micrographs of individual teased muscle fibers from the external oblique of uninjected control, uninjected SMA, and AAV9-UBA1–treated SMA mice at P9 (scale bar: 100 µm). (F) Significant rescue of muscle fiber diameters in AAV9-UBA1–treated SMA mice at P9 (n = 4 mice per group). One-way ANOVA with Tukey’s post-hoc test for all analyses. *P = 0.05, **P = 0.01, ***P = 0.005, ****P = 0.001. As part of a different comparative analysis, uninjected littermate control data motor neuron cell body counts (B) and NMJ axonal input (D) is also shown in Figure 2, F and H, respectively.

Fig 5: Restoration of UBA1 in SMA mice reverses GARS dysregulation and rescues sensory neuron cell fate phenotypes. (A and B) Representative fluorescent western blot (A) and quantification (B) of SMN, UBA1 and GARS in dorsal root ganglia from late-symptomatic SMA mice and SMA mice injected with AAV9-UBA1 (SMA+AAV9-UBA1). a-Tubulin (a-Tub): loading control. n = 3 mice per condition. (C) Spinal column sections from lumbar segments 1 and 2 of late-symptomatic SMA and SMA+AAV9-UBA1 mice labelled with UBA1a (green), SMI32 (magenta) and DAPI. (D) Spinal column sections from lumbar segments 1 and 2 of late-symptomatic SMA and SMA+AAV9-UBA1 mice labelled with NF200 (magenta) and peripherin (green). (C and D) Dorsal root ganglia (DRG) are outlined. (E and F) Quantification of the percentage of NF200-positive (NF200+) (E) and peripherin-positive (peripherin+) (F) sensory neurons. Data from control mice is shown as reference. n = 3 mice per condition, n = 4 DRGs per mouse. (G and H) Quantification of the area of NF200+ (G) and peripherin+ (H) sensory neurons. n = 3 mice per condition, n = 2 DRGs per mouse (seven NF200+ and seven peripherin+ neurons per DRG were analysed); ns = not significant. *P < 0.05, **P < 0.01. See also Supplementary Figs 10 and 11.