Skeletal muscle effects of antisense oligonucleotides targeting glycogen synthase 1 in a mouse model of Pompe disease

Abstract

Pompe disease (PD) is a progressive myopathy caused by the aberrant accumulation of glycogen in skeletal and cardiac muscle resulting from the deficiency of the enzyme acid alpha-glucosidase (GAA). Administration of recombinant human GAA as enzyme replacement therapy (ERT) works well in alleviating the cardiac manifestations of PD but loses sustained benefit in ameliorating the skeletal muscle pathology. The limited efficacy of ERT in skeletal muscle is partially attributable to its inability to curb the accumulation of new glycogen produced by the muscle enzyme glycogen synthase 1 (GYS1). Substrate reduction therapies aimed at knocking down GYS1 expression represent a promising avenue to improve Pompe myopathy. However, finding specific inhibitors for GYS1 is challenging given the presence of the highly homologous GYS2 in the liver. Antisense oligonucleotides (ASOs) are chemically modified oligomers that hybridise to their complementary target RNA to induce their degradation with exquisite specificity. In the present study, we show that ASO-mediated Gys1 knockdown in the Gaa^-/- mouse model of PD led to a robust reduction in glycogen accumulation in skeletal muscle. In addition, combining Gys1 ASO with ERT slightly further reduced glycogen content in muscle, eliminated autophagic buildup and lysosomal dysfunction, and improved motor function in Gaa^-/- mice. Our results provide a strong foundation for validation of the use of Gys1 ASO, alone or in combination with ERT, as a therapy for PD. We propose that early administration of Gys1 ASO in combination with ERT may be the key to preventative treatment options in PD. KEY POINTS: Antisense oligonucleotide (ASO) treatment in a mouse model of Pompe disease achieves robust knockdown of glycogen synthase (GYS1). ASO treatment reduces glycogen content in skeletal muscle. Combination of ASO and enzyme replacement therapy (ERT) further improves motor performance compared to ASO alone in a mouse model of Pompe disease.

Keywords: Enzyme replacement therapy (ERT); Gaa‐/‐ mouse model; Pompe disease; antisense oligonucleotides (ASOs); glycogen synthase 1 (GYS1); skeletal muscle.

FIGURE 1

Gaa^{− / –} mice show progressive accumulation of glycogen in skeletal muscle followed by deterioration of muscle function. (A) Glycogen content in quadriceps from Gaa^−/− mice at 1, 2, and 3 months (mo) of age compared to 1‐month‐old wild‐type (WT) C57BL/6 mice. n = 5 per group, as indicated in parenthesis on the x‐axis. (B) Graph illustrates the time that the mice spent on an accelerating rotarod before falling, to assess muscle strength as the animals age from 2 to 14 months. The mean decline in motor function is represented by the blue line within the shaded grey area, which indicates the SEM range. The number of mice in each measurement group is indicated at the bottom of the graph (n). Statistical analysis in A and B were performed using one‐way ANOVA in GraphPad Prism software. *p < .05, **p < .01, ***p < .005, ****p < .001.

FIGURE 2

Early treatment of Gaa^{− / −} mice with Gys1 ASOs lowers target mRNA and protein expression, leading to reduced glycogen accumulation in skeletal muscle. (A) Schematic illustration of the early treatment paradigm with Gys1 ASOs in Gaa^−/− mice from 1 to 4.5 months of age. WT: wild type. (B) qPCR analysis of Gys1 mRNA in quadriceps muscle. (C) Representative images of Western blot analysis of GYS1 protein in quadriceps, diaphragm, and heart (n = 5–7 per group). (D) Glycogen content measured using a biochemical assay in quadriceps muscle, diaphragm, and heart from Gaa^−/− mice dosed with either PBS, control ASO (Ctrl ASO), or two different Gys1 ASOs. The number of samples in each group is listed in parentheses on the x‐axis. (E) Periodic acid‐Schiff (PAS) staining of histological sections of quadriceps muscle. Scale bars: 50 µm. (F) PAS median intensities of quadriceps muscle fibres in controls and treated groups. Statistical analysis was performed using one‐way ANOVA in GraphPad Prism software. ***p < .005, ****p < .0001.

FIGURE 3

Therapeutic treatment with Gys1 ASO in Gaa^{− / –} mice partially reverses glycogen accumulation in skeletal muscle and preserves muscle strength. (A) Schematic illustration of the reversal treatment paradigm with Gys1 ASO#2 in Gaa^−/− mice from 3 to 6.5 months of age. WT: wild type. (B) Graph illustrates the time that mice spent on an accelerating rotarod before falling, to assess muscle strength. (C) qPCR analysis of Gys1 mRNA in quadriceps muscle. The number of samples in each group is listed in parentheses on the x‐axis. (D) Representative images of Western blot analysis of GYS1 protein in quadriceps, diaphragm, and heart (n = 5 per group). GAPDH was used as loading control. (E) Glycogen content measured using a biochemical assay in quadriceps muscle, diaphragm and heart from Gaa^−/− mice dosed with either PBS, a control ASO (Ctrl ASO), or Gys1 ASO. The number of samples in each group is listed in parentheses on the x‐axis. (F) Representative images from PAS staining of histological sections of quadriceps muscle. Scale bars: 50 µm. (G) PAS median intensities of quadriceps muscle fibres in controls and treated groups. Statistical analysis was performed using unpaired t‐test (Figure 3B) or one‐way ANOVA in GraphPad Prism software. *p < .05, ***p < .005, ****p < .001, ns: not significant.

FIGURE 4

Combination treatment with Gys1 ASO and ERT reduces glycogen accumulation and improves motor function in Gaa^{− / −} mice. (A) Schematic illustration of the study where 4‐month‐old Gaa^−/− mice were dosed with Gys1 ASO#2 and ERT in combination. WT: wild type. (B) Graph shows the results of a rotarod test to assess muscle strength. The time that the mice spent on a rotating rod was recorded monthly over the 3 months of treatment and expressed as per cent change versus baseline measurements (collected before the start of dosing) for each group; n = 5 per group. (C) Per cent change of rotarod data at the final time point. Statistical analysis in B and C were performed by two‐way ANOVA using multiple comparisons in GraphPad Prism software. ****p < .001 between the Gaa^−/− mice that received control ASO and Gys1 ASO#2 or Gys1 ASO#2+ERT. (D) qPCR analysis of Gys1 mRNA in quadriceps muscle. The number of samples in each group is listed in parentheses on the x‐axis. (E) Representative images of Western blot analysis of GYS1 protein in quadriceps muscle (n = 5), diaphragm (n = 3), and heart (n = 5 per group). GAPDH was used as loading control. (F) Glycogen content measured using a biochemical assay in quadriceps muscle, diaphragm and heart from wild‐type or Gaa^−/− mice dosed with either PBS, Gys1 ASO#2, Gys1 ASO#2+ERT, or ERT alone (n = 5 per group). (G) Representative images from PAS staining of histological sections of quadriceps muscle. Scale bars: 50 µm. (H) PAS median intensities of quadriceps muscle fibres in controls and treated groups, data are the mean ± SD from 5 images each group. Statistical analysis in D, F and H were performed using one‐way ANOVA. *p < .05, **p < .01, ***p < .005, ns: not significant.

FIGURE 5

Normalisation of autophagic flux in skeletal muscle of Pompe mice after treatment with Gys1 ASO, alone or in combination with ERT. (A) Western blot analysis of LC3‐II, LC3‐II/LC3‐I ratio, and p62 proteins in quadriceps muscle and heart from mice in cohort 3. GAPDH was used as loading control. WT: wild type. (B) Western blot analysis of LAMP1 expression. (C) Representative images of LAMP1 immunostaining in histological sections of quadriceps muscle from Gaa^−/− mice treated with Gys1 ASO#2, alone or in combination with ERT, scale bars: 100 µm. (D) Quantification of LAMP1‐positive cells as percentage of the total cell count in quadriceps muscle sections. LAMP1: lysosomal‐associated membrane protein 1. All experiments were conducted with n = 3–5 samples per group. The graphs depict the quantification of Western blots from three to five independent experiments. Statistical analysis was performed using one‐way ANOVA. *p < .05, **p < .01, ***p < .005, ns: not significant.

FIGURE 6

Model of pharmacological intervention in Pompe disease combining GYS1 ASO‐based SRT and ERT. Proposed adjunct therapeutic approach for Pompe disease based on the reduction of the synthesis of new glycogen via GYS1 ASO‐mediated SRT, combined with the ERT‐mediated clearance of preexisting glycogen that has accumulated in the lysosome because of GAA inactivation. Such combinatorial strategy could potentially result in greater benefit for Pompe patients, especially in skeletal muscle, where ERT alone is not remarkably effective. SRT: substrate reduction therapy; ERT: enzyme replacement therapy; GYS1: glycogen synthase 1; ASO: antisense oligonucleotide; GAA: glucosidase, alpha acid; rhGAA: recombinant human GAA.