Systemic gene therapy for methylmalonic acidemia using the novel adeno-associated viral vector 44.9

Abstract

Methylmalonic acidemia (MMA) is a severe and potentially lethal autosomal recessive inborn error of metabolism most frequently caused by mutations in the methylmalonyl-CoA mutase (MMUT) gene. Proof-of-concept adeno-associated virus (AAV) gene therapy studies using mouse models of MMA have demonstrated promise for this therapeutic approach but translation to the clinic could be limited by preexisting capsid immunity and vector potency. Here we explore the efficacy of a novel clade E capsid, 44.9, as a serotype for systemic AAV gene therapy for MMA. An anti-AAV44.9 neutralizing antibody (NAb) survey in adult volunteers (n = 19) and a large cohort of MMA patients (n = 48) revealed a seroprevalence rate of ∼26% and 13%, respectively. The efficacy of AAV44.9 gene delivery was examined in two murine models of MMA, representing neonatal lethal and juvenile phenotypes of MMA. Systemic delivery of the AAV44.9-Mmut vector prevented lethality and lowered disease-related metabolites in MMA mice. Tissue biodistribution and transgene expression studies in treated MMA mice showed that AAV44.9 was efficient at transducing the liver and heart. In summary, we establish that AAV44.9 exhibits a low prevalence of preexisting NAb in humans, is highly efficacious in the treatment of clinically severe MMA mouse models and is therefore a promising vector for clinical translation.

Keywords: AAV 44.9; MMA; liver directed gene therapy; methylmalonic acidemia; neutralizing antibodies; systemic gene therapy.

Graphical abstract

Figure 1

Neutralizing antibodies (NAb) against AAV capsids in healthy volunteers and MMA patients (A) Seroreactivity in healthy volunteers against AAV2 and AAV44.9 at NAb titers ≥1:16. (B) Seroreactivity in MMA patients with MMUT deficiency against AAV9 and AAV44.9 at NAb titers ≥1:16. NA, not available; NT, not transplanted; T, transplanted. p < 0.05 one-way ANOVA with post hoc Tukey test compared with AAV NAb in healthy volunteers.

Figure 2

AAV44.9 vectors efficiently transduce liver and heart in mice following systemic delivery (A) A representative image of a wild-type young adult Balb/c mouse treated by tail vein injection with 1 × 10¹² vg/mouse AAV44.9 expressing a firefly luciferase reporter gene. Luciferase expression was assessed in live mice 2 months later using the IVIS Lumina platform to quantify luminescence, which is depicted with a color bar scale in the inset of the panel. (B) Luminescence of three dissected and isolated livers and hearts, respectively, are displayed, each image is complemented by a color bar scale. (C) Luciferase transduction quantification from organs in (B). Luminescence is expressed as Average Radiance n = 3. ∗p < 0.05. (D and E) Transduction efficiency and specificity of AAV44.9 Cre-recombinase in ROSA^mT/mG (Tomato) transgenic mice as assessed by fluorescence expression of GFP. Mice received 1 × 10¹² vg/mouse by tail vein administration. One month after treatment, the liver (D) and heart (E) were collected, and analyzed by confocal microscopy. Panels, from left to right, show labeled cells with DAPI (blue), mTomato (red), mGFP (green), and finally mTomato/mGFP (merged), in their respective channels. Near complete conversion to green fluorescence is appreciated in both tissues following Cre expression. The untreated control organs maintain the membrane-localized Tomato fluorescence.

Figure 3

Survival of Mmut^−/− mice, vector biodistribution, and MMUT protein expression after treatment with AAV44.9-CBA-Mmut Mmut^−/− mice received an intrahepatic injection of 2 × 10¹¹ vg of AAV44.9-CBA-Mmut per pup at birth, and were sacrificed on day of life 60. (A) AAV44.9-CBA-Mmut treated Mmut^−/− mice had a significant increase in survival in comparison with untreated Mmut^−/− mice (∗∗∗p < 0.005, Log rank (Mantel-Cox) test). (B) Vector biodistribution of AAV44.9-CBA-Mmut in Mmut^−/− treated mice at day 60, n = 4 for all tissues studied (∗∗p < 0.01, two-way ANOVA). (C) MMUT hepatic protein expression in AAV44.9-CBA-Mmut treated Mmut^−/− mice (n = 3) determined by western blot analysis at day 60. Beta-actin served as a loading control. (D) Quantification of hepatic MMUT protein expression in (C) normalized to beta-actin. NS, not significant (one-way ANOVA with Kruskal-Wallis test); WT, wild type. Error bars = $\pm$ SEM.

Figure 4

Improvement in disease-related metabolites and growth in MMA mice after AAV44.9-CBA-Mmut treatment Mmut^−/−MCK-Mmut⁺ received 7 × 10¹² vg/kg of AAV44.9-CBA-Mmut at 1 month of life by retro-orbital delivery. (A) Methylmalonic acid levels were measured in the plasma and compared with pretreatment levels. (∗∗p < 0.004, one-way ANOVA with Kruskal-Wallis test). (B) Methylcitrate levels were measured in the plasma and compared with pretreatment levels. (∗p < 0.04, one-way ANOVA with Kruskal-Wallis test). (C) Growth in mice before and after AAV44.9-CBA-Mmut treatment in female Mmut^−/−MCK-Mmut⁺ relative to untreated Mmut^−/−MCK-Mmut⁺ and wild-type (WT) mice. (D) Growth in mice before and after AAV44.9-CBA-Mmut treatment in male Mmut^−/−MCK-Mmut⁺ relative to untreated Mmut^−/−MCK-Mmut⁺ and WT mice (∗p < 0.05, ∗∗p < 0.005, two-way ANOVA). Error bars = $\pm$ SEM.