Systemic delivery of AAV-GCDH ameliorates HLD-induced phenotype in a glutaric aciduria type I mouse model

Abstract

Glutaric aciduria type 1 (GA1) is a rare inherited metabolic disorder caused by a deficiency of glutaryl-coenzyme A dehydrogenase (GCDH), with accumulation of neurotoxic metabolites, resulting in a complex movement disorder, irreversible brain damage, and premature death in untreated individuals. While early diagnosis and a lysine restricted diet can extend survival, they do not prevent neurological damage in approximately one-third of treated patients, and more effective therapies are required. Here we report the efficacy of adeno-associated virus 9 (AAV9)-mediated systemic delivery of human GCDH at preventing a high lysine diet (HLD)-induced phenotype in Gcdh ^-/- mice. Neonatal treatment with AAV-GCDH restores GCDH expression and enzyme activity in liver and striatum. This treatment protects the mice from HLD-aggressive phenotype with all mice surviving this exposure; in stark contrast, a lack of treatment on an HLD triggers very high accumulation of glutaric acid, 3-hydroxyglutaric acid, and glutarylcarnitine in tissues, with about 60% death due to brain accumulation of toxic lysine metabolites. AAV-GCDH significantly ameliorates the striatal neuropathology, minimizing neuronal dysfunction, gliosis, and alterations in myelination. Magnetic resonance imaging findings show protection against striatal injury. Altogether, these results provide preclinical evidence to support AAV-GCDH gene therapy for GA1.

Keywords: AAV; adeno-associated virus; gene replacement; glutaric aciduria; metabolic disease; viral vectors.

Graphical abstract

Figure 1

AAV9 vector expressing GCDH from the CAG promoter shows superior phenotype amelioration than that expressed from the PGK promoter (A) Schematic illustration of the AAV-P-GCDH and AAV-GCDH vectors. (B) Western blot analysis of GCDH in liver and brain extracts from 6-week-old Gcdh KO mice treated with 7.5 × 10¹² vg/kg of the indicated virus (n = 3). (C) C5DC content in liver and brain extracts from WT, KO, or KO-treated animals with AAV-P-GCDH or AAV-GCDH (n = 4 to 9). Data are expressed as the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Figure 2

Intravascular administration of AAV-GCDH in KO young adult mice improves metabolite accumulation in the liver and in the striatum after HLD exposure at 5 months after treatment AAV-GCDH was injected via tail vein at a dose (7.5 × 10¹² vg/kg) into 1-month-old mice; 1 week later, mice were placed onto an HLD regime, or kept on a standard diet (SD), for 3 weeks or 5 months. (A) Western blot analysis of GCDH in liver and striatum lysates at 1 month and 5 months after treatment. Quantification of GCDH from different individuals (n = 4–6). C5DC, GA, and 3-OH GA were measured in WT, KO, and KO-AAV-GCDH treated mice in the liver at 1 month (B) or 5 months (C) after therapy, and in the striatum at 1 month (D) or 5 months (E) after therapy. Data are expressed as the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. Statistical significance between WT-standard diet and WT-HLD were observed for C5DC in (B, D, and E) and for GA and 3-OH GA in (C). Statistical significance between KO-standard diet and KO-HLD were observed for GA in (C, D, E) and for 3-OH GA in (B, E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Figure 3

Intravascular administration of AAV-GCDH in KO-neonatal mice restores enzyme activity and prevents HLD-induced metabolite accumulation in the striatum at 1 month and 6 months after treatment AAV-GCDH was injected into the temporal vein (at 5 × 10¹³ vg/kg) at P1, and at weaning were fed a standard diet (SD) or an HLD for 4 days or 5 months. Readouts were performed at 1 month and 6 months after viral injection. (A) Western blot analysis of GCDH in liver and striatum lysates at 1 month and 6 months after treatment. Quantification of GCDH from different individuals is shown (n = 4–6). C5DC, GA, and 3-OH GA concentrations in liver are shown for 1 month (B) and 6 months (C) after therapy, and in the striatum at 1 month (D) or 6 months (E) after therapy. Data are expressed as the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. Statistical significance between WT-standard diet and WT-HLD were observed for C5DC in (E), and for GA and 3-OH GA in (C). Statistical significance between KO-standard diet and KO-HLD were observed for C5DC (B and D), GA (B–E), and 3-OHGA (B, D, E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Figure 4

Intravascular administration of AAV-GCDH in KO-neonatal mice prevents from HLD-induced GA accumulation in the striatum at 1 month after therapy AAV-GCDH were injected in the temporal vein at a dose 7.5 × 10¹² vg/kg (Low dose) at P1. At weanling were exposed to HLD for 4 days. Analysis were performed 1 month after viral injection. (A) Western blot analysis of GCDH in the striatum. Quantification of GCDH from different individuals (n = 6). (B) Analysis of GA concentration (n = 6). Results also show the GA concentration of AAV-GCDH treated mice at 5 × 10¹³ vg/kg (Figure 3D; high dose). Data are expressed as the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Figure 5

MRIs One month-old WT or KO mice were fed a standard diet (SD) or HLD regime for 4 days and either treated with AAV-GCDH or saline; mice were then used for the MRI studies. (A) Proton MR spectroscopy imaging was performed in mice striatum (n = 6-7). Estimated concentration of NAA+NAAG are indicated. Data are expressed as the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. ∗p < 0.05. (B) Representative brain images of MD maps obtained from DWI.

Figure 6

Neonatal AAV-GCDH ameliorates striatal injury Immunohistochemical analysis of striatum alterations in the indicated groups of mice. Hematoxylin and eosin staining showing progressive vacuolation in KO-HLD mice. Scale bar, 50 μm, and GFAP, glial marker; scale bar, 100 μm, and MBP, myelin marker. Quantification was performed on histology images with 1 section/mouse, digitalized with a ScanScope slide scanner, and analyzed with QuPath 0.3.0 software. Results are presented as percentage of vacuole area, GFAP-positive cells, or area of MBP staining. Data are expressed ase the means ± SEM. Significance was assessed using a two-tailed Mann-Whitney test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p > 0.001.

Figure 7

Neonatal AAV-GCDH administration to KO-HLD mice results in normal lifespan Follow-up of 150 days for mice survival in Gcdh KO mice treated (n = 25) or not (n = 33) with AAV-GCDH and exposed to HLD at weaning. p value was calculaed by log rank test (Mantle-Cox).