Abstract Gene therapies are promising for diseases previously considered incurable. Adeno-associated virus serotype 9 (AAV9) demonstrates remarkable tropism for motor neurons (MNs) and represents an exciting candidate to target genetic causes of motor neuron diseases like amyotrophic lateral sclerosis (ALS). However, systemic delivery risks immunogenicity and off-target effects, therefore localised delivery to the CNS is advantageous. We assessed MN transduction in wild-type post-natal mice using AAV9-controlled, cytomegalovirus-promoter driven, enhanced GFP expression. Intra-cisterna magna (ICM) and intra-cerebroventricular (ICV) methods were compared. Four weeks post-delivery, GFP positivity in MN and astrocytes were quantified via immunohistochemical approaches and viral genome copy number determined by qPCR. All delivery methods achieved high MN transduction in lumbar spinal cord (> 68%). Unilateral ICV delivery provided the highest and most consistent levels (89 ± 3%), and minimal peripheral viral copies. ICV delivery resulted in higher astrocytic transduction, most notably in the cortex. Brainstem MN transduction was high with all methods (> 55%). We failed to find evidence of neuronal transduction in motor cortex. Viral genome copies trended higher in spinal cord and brainstem with ICV approaches, however further work is required to understand how bilateral repeated dose delivery leads to more profound increases. Whilst several routes of administration into cerebrospinal fluid exist, direct comparisons for targeting MNs in vivo remain limited. Overall, all methods of CNS-directed delivery result in high levels of motor neuron transduction in the lumbar spinal cord and brainstem, but not in motor cortex. Unilateral ICV appears to provide the best balance between consistent, high levels of transduction and low off-target effects. However, ICM might be the better option if seeking to avoid astrocytic transduction.
Mortimer et al. (Wed,) studied this question.