Improved therapeutic approach for spinal muscular atrophy via ubiquitination-resistant survival motor neuron variant

Background Zolgensma is a gene-replacement therapy that has led to a promising treatment for spinal muscular atrophy (SMA). However, clinical trials of Zolgensma have raised two major concerns: insufficient therapeutic effects and adverse events. In a recent clinical trial, 30% of patients failed to achieve motor milestones despite pre-symptomatic treatment. In addition, more than 20% of patients showed hepatotoxicity due to excessive virus dosage, even after the administration of an immunosuppressant. Here, we aimed to test whether a ubiquitination-resistant variant of survival motor neuron (SMN), SMNK186R, has improved therapeutic effects for SMA compared with wild-type SMN (SMNWT). Methods A severe SMA mouse model, SMA type 1.5 (Smn(-/-); SMN2+/+; SMN triangle 7(+/-)) mice, was used to compare the differences in therapeutic efficacy between AAV9-SMNWT and AAV9-SMNK186R. All animals were injected within Postnatal Day (P) 1 through a facial vein or cerebral ventricle. Results AAV9-SMNK186R-treated mice showed increased lifespan, body weight, motor neuron number, muscle weight and functional improvement in motor functions as compared with AAV9-SMNWT-treated mice. Lifespan increased by more than 10-fold in AAV9-SMNK186R-treated mice (144.8 +/- 26.11 days) as compared with AAV9-SMNWT-treated mice (26.8 +/- 1.41 days). AAV9-SMNK186R-treated mice showed an ascending weight pattern, unlike AAV9-SMNWT-treated mice, which only gained weight until P20 up to 5 g on average. Several motor function tests showed the improved therapeutic efficacy of SMNK186R. In the negative geotaxis test, AAV9-SMNK186R-treated mice turned their bodies in an upward direction successfully, unlike AAV9-SMNWT-treated mice, which failed to turn upwards from around P23. Hind limb clasping phenotype was rarely observed in AAV9-SMNK186R-treated mice, unlike AAV9-SMNWT-treated mice that showed clasping phenotype for more than 20 out of 30 s. At this point, the number of motor neurons (1.5-fold) and the size of myofibers (2.1-fold) were significantly increased in AAV9-SMNK186R-treated mice compared with AAV9-SMNWT-treated mice without prominent neurotoxicity. AAV9-SMNK186R had fewer liver defects compared with AAV9-SMNWT, as judged by increased proliferation of hepatocytes (P < 0.0001) and insulin-like growth factor-1 production (P < 0.0001). Especially, low-dose AAV9-SMNK186R (nine-fold) also reduced clasping time compared with SMNWT. Conclusions SMNK186R will provide improved therapeutic efficacy in patients with severe SMA with insufficient therapeutic efficacy. Low-dose treatment of SMA patients with AAV9-SMNK186R can reduce the adverse events of Zolgensma. Collectively, SMNK186R has value as a new treatment for SMA that improves treatment effectiveness and reduces adverse events simultaneously.