Effective Treatment of Colq-Deficient Mice with Adeno-Associated Virus Type Rh74-Mediated Gene Therapy

Mutations in human COLQ, which encodes the collagen-like tail subunit (ColQ) of asymmetrical acetylcholinesterase (AChE), cause congenital myasthenic syndrome (CMS) with deficiency of end plate AChE. A valuable animal model of COLQ-CMS is the Colq-deficient (Colq^-/-) mouse, which lacks asymmetrical AChE in skeletal and cardiac muscles. Mutant Colq^-/- mice fail to thrive, and many die before reaching maturity. With the aim of developing a treatment for COLQ-CMS, Colq^-/- mice were injected at postnatal day 26-29 with three doses of an adeno-associated virus type rh74 carrying full-length human COLQ (AAVrh74-COLQ): 5 × 10¹³ viral genomes per kilogram (vg/kg) (intravenously [IV]), 1 × 10¹⁴ vg/kg (IV), and 2 × 10¹⁴ vg/kg (1 × 10¹⁴ vg/kg IV + 1 × 10¹⁴ vg/kg intraperitoneally). Motor performance was evaluated using rotarod, grip strength, and wire hang tests weekly for 12 weeks. Voluntary ambulation and repetitive nerve stimulation (RNS) were assessed once before euthanasia. Protein and RNA expression of COLQ was measured via immunohistochemistry (IHC) and reverse transcriptase quantitative PCR (RT-qPCR), respectively. Mice treated with AAVrh74-COLQ at 1 × 10¹⁴ and 2 × 10¹⁴ vg/kg doses showed 100% survival and no adverse side effects. Mice injected with 2 × 10¹⁴ vg/kg showed almost full recovery and similar scores to wild type that were significantly higher than vehicle-injected mutants for grip strength (p value <0.0001), rotarod (p value <0.0001), and RNS (p value <0.0001). Similar improvements were observed in mice injected with 1 × 10¹⁴ vg/kg, although the recovery of grip strength was incomplete. Mice injected with 5 × 10¹³ vg/kg showed incomplete recovery. IHC demonstrated full recovery of protein expression in 1 × 10¹⁴ and 2 × 10¹⁴ vg/kg mice, and RT-qPCR unambiguously demonstrated that the source of the ColQ was human COLQ. In summary, a single treatment of AAVrh74-COLQ (1 × 10¹⁴ to 2 × 10¹⁴ vg/kg) was effective and safe for Colq^-/-mice, which reproduce many of the clinical features of the human COLQ-CMS phenotype. Thus, these results support a similar therapy for patients affected with COLQ-CMS.