Neonatal systemic gene therapy restores cardiorespiratory function in a rat model of Pompe disease

Abstract

Absence of functional acid-α-glucosidase (GAA) leads to early-onset Pompe disease with cardiorespiratory and neuromuscular failure. A novel Pompe rat model (Gaa ^-/-) was used to test the hypothesis that neonatal gene therapy with adeno-associated virus serotype 9 (AAV9) restores cardiorespiratory neuromuscular function across the lifespan. Temporal vein administration of AAV9-DES-GAA or sham (saline) injection was done on post-natal day 1; rats were studied at 6-12 months old. Whole-body plethysmography showed that reduced inspiratory tidal volumes in Gaa ^-/- rats were corrected by AAV-GAA treatment. Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI) revealed that AAV-GAA treatment normalized diaphragm muscle glycogen as well as glycans. Neurophysiological recordings of phrenic nerve output and immunohistochemical evaluation of the cervical spinal cord indicated a neurologic benefit of AAV-GAA treatment. In vivo magnetic resonance imaging demonstrated that impaired cardiac volumes in Gaa ^-/- rats were corrected by AAV-GAA treatment. Biochemical assays showed that AAV treatment increased GAA activity in the heart, diaphragm, quadriceps and spinal cord. We conclude that neonatal AAV9-DES-GAA therapy drives sustained, functional GAA expression and improved cardiorespiratory function in the Gaa ^-/- rat model of Pompe disease.

Figure 1.. AAV-GAA treatment normalizes body weight and breathing.

Data obtained during baseline room air breathing i.e., “eupnea”. A: Example of breathing patterns measured using whole body plethysmography. B: Body weight is normalized after AAV treatment. C: Tidal volume (VT, ml/br) is restored after AAV treatment. D: Minute ventilation ($\dot{\mathrm{V}}\mathrm{E}$, ml/min) is restored after AAV treatment. Respiratory rate (E), metabolic rate (as estimated via ${\dot{\mathrm{V}}\mathrm{CO}}_2$; F) and the ratio of $\dot{\mathrm{V}}\mathrm{E}$ to ${\dot{\mathrm{V}}\mathrm{CO}}_2$ (G) all showed a strong trend to be impacted by AAV treatment. Statistical test: 2-way RM ANOVA. The treatment effect P-value is reported on each plot. *, p<0.05 vs. Pompe + AAV-GAA. S-D: Sprague-Dawley

Figure 2.. Spatial metabolomic profile of the diaphragm muscle at age 12 months indicates normalization of glycogen after AAV-GAA.

A: Examples of diaphragm tissue evaluated using MALDI. The heat map shows the gradient of glycogen in diaphragm (represented by chain length +7, 1175m/z). B: Normalization of diaphragm glycogen after neonatal AAV-GAA treatment. C-D: Unsupervised clustering heatmap analysis for the glycome (C) and metabolome/lipidome (D). the treatment group is indicated by the top row, and the relative expression of each molecule is indicated by the color on the heat map. Relative abundance (RA) plots show glucose (E), 3-Phosphoglyceric acid (3PG) (F), arachidonic acid (G), glycerophosphorylethanolamine (GPE) (H), docosahexaenoic acid (I), and stearic acid (J). Statistical tests: B: 2-way RM ANOVA; treatment effect P-value is reported on plot. E-J: 1-way ANOVA; treatment effect P-value is reported on each plot. *, p<0.05 vs. Pompe+AAV-GAA; S-D: Sprague-Dawley. Color scheme for treatment groups is the same on all panels.

Figure 3.. Impact of AAV-GAA treatment on diaphragm myofibers.

A: Example photomicrographs from each group. B: Type I and IIb/x myofiber size was normalized in Pompe rats following the AAV-GAA treatment, with values comparable to that observed in Sprague-Dawley (S-D) rats. C: AAV-GAA treatment increased the overall number of diaphragm Type IIb/x fibers (treatment, P<0.001). Statistical test: 1-way ANOVA; P-value is reported on each plot. *, p<0.05 vs. Pompe+AAV-GAA

Figure 4.. Phrenic nerve recordings.

A: Examples of recordings of inspiratory bursting recorded in the phrenic nerve of anesthetized rats during baseline conditions. B: Inspiratory burst amplitude (v). Pompe rats treated with AAV-GAA showed a strong trend for increased burst amplitude (P=0.060). C: Respiratory rate (bursts per minute) was similar between saline and AAV-GAA treated rats (P=0.806). D: Heart rate (beats per min) was greater in AAV-GAA vs. saline treated rats (P=0.036). Statistical test: 2-way RM ANOVA. The treatment effect P-value is reported on each plot. S-D: Sprague-Dawley

Figure 5.. Representative photomicrographs of spinal cord tissue.

Mid-cervical (C4-5) spinal cord sections were stained with NeuN (neurons) IV-58 (Glycogen), GFAP, and Iba1, and evaluated using fluorescence microscopy. The images demonstrate the expected marked increase in neuronal glycogen in Pompe+Saline rats, and a reduction in glycogen after AAV-GAA treatment. A: Low power images showing glycogen staining in spinal grey matter. B-D: Higher power images showing staining for neurons (NeuN) and glycogen (B), GFAP and glycogen (C), and Iba1 and glycogen (D). S-D: Sprague-Dawley

Figure 6.. GAA activity and glycogen content.

A: Assays done at age 6 months, neonatal AAV-GAA treatment increased GAA activity and reduced glycogen in heart and diaphragm. B: Assays done at age 12 month; neonatal AAV-GAA treatment increased GAA activity in heart, diaphragm, quadriceps and spinal cord. Statistical tests: A: 1-way ANOVA. *, different than Pompe+AAV-GAA. B: t-test. S-D: Sprague-Dawley

Figure 7.. Impact of AAV-GAA on the heart.

Representative MRI images are shown in panel A. B: Ex vivo assessment of the heart weight to body weight ratio (HW:BW) shows a reduction in size after AAV-GAA. C: Cardiac output (CO) was increased after AAV-GAA treatment. D-E: Stroke volume (SV) and ejection fraction (EF) are variable in saline treated Pompe rats but are more consistent after AAV-GAA treatment. F: End-systolic and end-diastolic volumes are normalized following the AAV-GAA treatment. G: Volume index is increased after AAV-GAA. *, p<0.05 vs. other two groups; ^, p<0.05 vs. Pompe+AAV-GAA. S-D: Sprague-Dawley

Figure 8.. Impact of AAV-GAA on ECG.

A: R-R interval, B: PR interval, C: P-wave amplitude, D: R-wave amplitude. The R-R interval and R-wave amplitude were normalized after AAV-GAA therapy. Statistical test: 1-way ANOVA. The treatment effect P-value is reported on each plot. *, p<0.05 vs. other two groups; ^, p<0.05 vs. Pompe+AAV-GAA; #, p<0.05 vs. Sprague-Dawley (SD)