Systemic Delivery of AAVB1-GAA Clears Glycogen and Prolongs Survival in a Mouse Model of Pompe Disease

Abstract

Pompe disease is an autosomal recessive glycogen storage disorder caused by deficiency of the lysosomal enzyme acid alpha-glucosidase (GAA). GAA deficiency results in systemic lysosomal glycogen accumulation and cellular disruption in muscle and the central nervous system (CNS). Adeno-associated virus (AAV) gene therapy is ideal for Pompe disease, since a single systemic injection may correct both muscle and CNS pathologies. Using the Pompe mouse (B6;129-Gaa^Tm1Rabn/J), this study sought to explore if AAVB1, a newly engineered vector with a high affinity for muscle and CNS, reduces systemic weakness and improves survival in adult mice. Three-month-old Gaa^-/- animals were injected with either AAVB1 or AAV9 vectors expressing GAA and tissues were harvested 6 months later. Both AAV vectors prolonged survival. AAVB1-treated animals had a robust weight gain compared to the AAV9-treated group. Vector genome levels, GAA enzyme activity, and histological analysis indicated that both vectors transduced the heart efficiently, leading to glycogen clearance, and transduced the diaphragm and CNS at comparable levels. AAVB1-treated mice had higher GAA activity and greater glycogen clearance in the tongue. Finally, AAVB1-treated animals showed improved respiratory function comparable to wild-type animals. In conclusion, AAVB1-GAA offers a promising therapeutic option for the treatment of muscle and CNS in Pompe disease.

Keywords: AAV9; AAVB1; Pompe disease; glycogen; respiratory.

Figure 1.

Improved survival and behavior in Gaa^–/– animals treated with AAVB1 and AAV9 gene therapy. (A) Kaplan–Meier curve representing the percentage of animals surviving to the experimental endpoint at 180 days post injection. All wild type (WT) animals and 91% of both AAV9- and AAVB1-treated Gaa^–/– mice survived to 180 days, whereas only 50% of phosphate-buffered saline (PBS)-treated Gaa^–/– mice survived for the duration of the study. *p < 0.05 compared to PBS-treated Gaa^–/– animals. (B) Body weights (g) after injection until experimental endpoint. *p < 0.05 compared to 30 days. ^#p < 0.05 compared to WT. (C) Four-limb grip strength test measured in centiNewtons (cN) at 30, 90, and 180 days. *p < 0.05. (D) Inverted screen test time at 30, 90, and 180 days. *p < 0.05. Data are represented as mean ± standard error of the mean (SEM).

Figure 2.

Biodistribution of vector genomes in AAV9- and AAVB1-treated Gaa^–/– animals. Vector genomes (vg) contents in (A) respiratory related tissues—the lung, trachea, tongue, and diaphragm, (B) the liver, gastrocnemius muscle, and heart, and (C) the medulla and cervical, thoracic, and lumbar spinal cord 180 days after AAVB1 or AAV9 injection. AAVB1-injected animals had higher vg content in the tongue and in the thoracic cord compared to AAV9. Data are represented as mean ± SEM. *p < 0.05.

Figure 3.

Robust acid alpha-glucosidase (GAA) expression was noted by immunohistochemistry in both AAV9 and AAVB1 groups. Tissues were harvested 180 days following systemic AAV9 and AAVB1 injections, and the presence of GAA was detected by immunohistochemistry (brown). (A) GAA expression was seen in the heart, diaphragm, tongue, and gastrocnemius muscle. (B) In the medulla and spinal cord, motor neurons in both AAV9- and AAVB1-injected animals stained positive for GAA. Scale bar = 100 μm. Color images available online at www.liebertpub.com/hum

Figure 4.

GAA enzymatic activity evident 6 months following AAV injection. (A) GAA activity in both AAV9- and AAVB1-injected Gaa^–/– mice was robust in the heart, with levels well above WT levels. (B–D) GAA activity was not significantly different between AAV9 and AAVB1 animals in the diaphragm (B), tongue (C), and gastrocnemius muscle (D). (E) Lung GAA activity was significantly higher in the AAVB1-treated group compared to the AAV9 group. (F) Liver GAA activity levels were significantly lower in both AAV groups compared to WT. Data are represented as mean ± SEM. N.D., not detected. *p < 0.05.

Figure 5.

Systemic administration of AAV9 and AAVB1 reduced glycogen storage in Gaa^–/– mice. Periodic acid–Schiff staining of 1.5 μm sections of (A) the heart, (B) the diaphragm, (C) the tongue, and (D) the gastrocnemius muscle of Gaa^–/– animals injected with PBS, AAV9, or AAVB1 and WT animals. Dark staining corresponds to glycogen accumulation in cells. Note the robust accumulation of glycogen in the heart, diaphragm, tongue, and gastrocnemius muscle of the PBS-treated Gaa^–/– mice. In contrast, the cardiac muscle was almost completely cleared in animals treated with either AAV9 or AAVB1. The Gaa^−/− animals treated with AAVB1 had almost complete clearance of glycogen in the tongue, unlike the AAV9 group where considerable amounts of glycogen remained. The diaphragm and gastrocnemius muscle had partial clearance, with complete clearance seen in many muscle fibers. Scale bar = 100 μm. Biochemical assessment of glycogen accumulation in (E) the heart, (F) the diaphragm, (G) the tongue, and (H) the gastrocnemius muscle of Gaa^–/– animals injected with PBS, AAV9, or AAVB1 and WT animals. Data are represented as mean ± SEM. *p < 0.05.

Figure 6.

Respiratory phenotype was partially corrected by AAVB1 gene therapy in Gaa^–/– mice. Spontaneous breathing in Gaa^–/– animals 180 days following injections with PBS, AAV9, or AAVB1, or WT mice treated with PBS. These measurements were taken during a respiratory challenge with hypercapnia, and are presented as percent above baseline breathing. (A) No significant differences in breathing frequency were observed between groups. (B–D) AAVB1-treated Gaa^–/– animals responded to hypercapnic challenge with similar tidal volume (B), minute ventilation (C), and peak inspiratory flow rate (D) compared to WT animals. (E) No differences were observed in peak expiratory flow between groups. (F) AAVB1-treated mice had significantly lower inspiratory time compared to PBS-treated Gaa^–/– mice. (H) Airway and tissue resistance was measured in ventilated and sedated animals in response to incremental doses of methacholine. Data are represented as mean ± SEM. ^#p < 0.05 compared to WT. (I) PAS staining of plastic embedded tracheal sections of AAV9- and AAVB1-treated animals, respectively, revealed persistent glycogen accumulation in smooth muscle fibers. Scale bar = 100 μm. Color images available online at www.liebertpub.com/hum