Overexpression of Galgt2 reduces dystrophic pathology in the skeletal muscles of alpha sarcoglycan-deficient mice

Abstract

Recent studies have shown that a number of genes that are not mutated in various forms of muscular dystrophy may serve as surrogates to protect skeletal myofibers from injury. One such gene is Galgt2, which is also called cytotoxic T cell GalNAc transferase in mice. In this study, we show that Galgt2 overexpression reduces the development of dystrophic pathology in the skeletal muscles of mice lacking alpha sarcoglycan (Sgca), a mouse model for limb girdle muscular dystrophy 2D. Galgt2 transgenic Sgca(-/-) mice showed reduced levels of myofiber damage, as evidenced by i) normal levels of serum creatine kinase activity, ii) a lack of Evans blue dye uptake into myofibers, iii) normal levels of mouse locomotor activity, and iv) near normal percentages of myofibers with centrally located nuclei. In addition, the overexpression of Galgt2 in the early postnatal period using an adeno-associated virus gene therapy vector protected Sgca(-/-) myofibers from damage, as observed using histopathology measurements. Galgt2 transgenic Sgca(-/-) mice also had increased levels of glycosylation of alpha dystroglycan with the CT carbohydrate, but showed no up-regulation of beta, gamma, delta, or epsilon sarcoglycan. These data, coupled with results from our previous studies, show that Galgt2 has therapeutic effects in three distinct forms of muscular dystrophy and may, therefore, have a broad spectrum of therapeutic potential for the treatment of various myopathies.

Figures

Figure 1

Creation of Galgt2 transgenic α sarcoglycan-deficient (Sgca−/−) mice. A: Galgt2 transgenic (CT) mice and non-transgenic littermates that were either heterozygous (Sgca+/−) or homozygous (Sgca−/−) for a deletion of α sarcoglycan were stained with a monoclonal antibody to α sarcoglycan or with CT2, an antibody that recognizes the CT carbohydrate. Control secondary antibody for CT2 (IgM only) and α sarcoglycan (IgG only) are also shown. Scale bar = 100 μm. B: 40 μg of whole muscle SDS protein lysate was blotted for Galgt2 protein and for α sarcoglycan protein. Actin was used as a control for protein loading and transfer. Asterisk indicates native molecular weight for the protein of interest. C: Sgca+/−CT and Sgca−/−CT muscles show similar increases in Galgt2 mRNA, relative to Sgca+/− and Sgca−/−, respectively, as measured by quantitative RT-PCR. D: Galgt2 mRNA was reduced in Sgca−/− muscles relative to Sgca+/−. Gas, gastrocnemius; Quad, quadriceps; TA, tibialis anterior; Tri, triceps; Dia, diaphragm. Errors are SD for 3 to 4 animals per condition (in C and D).

Figure 2

Characterization of muscle pathology in Galgt2 transgenic Sgca+/− and Sgca−/− mice. A: H&E staining of the gastrocnemius muscle of Galgt2 transgenic (CT) Sgca+/− and Sgca−/− mice and non-transgenic littermates. Scale bar =100 μm. B: Myofiber diameters were measured in the diaphragm (Dia), gastrocnemius (Gas), quadriceps (Quad), tibialis anterior (TA) and triceps (Tri) muscle of Galgt2 transgenic (CT) Sgca+/− and Sgca−/− mice and their non-transgenic, age-matched, littermates. CT muscles were reduced in size, regardless of Sgca genotype (P < 0.01 for all). C: Percentage of myofibers with central nuclei was greatly increased in Sgca−/− muscles (P < 0.001 for all versus Sgca+/−). Galgt2 transgene expression significantly lowered central nuclei in Sgca−/− muscles relative to Sgca−/− (P < 0.001 for all). Errors are SD from n = 3 to 4 animals per genotype (in B and C).

Figure 3

Evans blue dye uptake is reduced in exercised Galgt2 transgenic Sgca−/−muscles. Galgt2 transgenic (CT) Sgca+/− and Sgca−/− mice were compared with their non-transgenic littermates for uptake of Evans blue dye. A: Dye uptake is increased in the quadriceps muscles in Sgca−/− animals and is not elevated in Sgca+/−, Sgca+/−CT, and Sgca−/−CT animals. Scale bar = 100 μm. B: Quantification of the percentage of total area with dye uptake in the diaphragm (Dia), gastrocnemius (Gas), quadriceps (Quad), tibialis anterior (TA) and the triceps (Tri) muscle. Dye uptake was reduced in all Sgca−/−CT muscles compared with Sgca−/− (P < 0.05 for all). Errors are SD for n = 3 to 4 animals per condition.

Figure 4

Whole animal measures of muscular dystrophy are reduced in Galgt2 transgenic Sgca−/− mice. Galgt2 transgenic (CT) and non-transgenic Sgca+/− and Sgca−/− mice were compared for levels of (A) serum creatine kinase activity and (B) locomotor activity. A: Serum creatine kinase activity was increased by an order of magnitude in Sgca−/− animals and this was significantly reduced, to wild-type levels, in Sgca−/−CT mice (P < 0.01 for Sgca−/− vs. Sgca+/− and P < 0.05 for Sgca−/−CT versus Sgca−/−). B: Sgca−/− mice had significantly reduced locomotor activity, and this too was significantly reversed in Sgca−/−CT mice (P < 0.001 for Sgca−/− vs. Sgca+/− and for Sgca−/−CT versus Sgca−/−). Errors are SEM for n = 11(Sgca+/−), 7 (Sgca+/−CT), 6 (Sgca−/−), or 17 (Sgca−/−CT) in (A) and n = 8 (Sgca+/− and Sgca−/−) or 7 (Sgca+/−CT and Sgca−/−CT) animals in (B).

Figure 5

Postnatal overexpression of Galgt2 inhibits the development of muscle pathology in Sgca−/− mice. A: Sgca−/− muscles infected with AAV1-CMV-Galgt2 show a range of increased Galgt2 mRNA expression, as measured by QRT-PCR, in four separate experiments. TA, tibialis anterior; Gas, gastrocnemius. B: Infected muscles overexpressing Galgt2 were visualized by immunostaining with CT2, an antibody to the CT carbohydrate (green). Central nuclei were evident on counterstaining with an antibody to β dystroglycan (red). Scale bar =100 μm. C: Myofiber diameters were measured in Galgt2-overexpressing myofibers and compared with muscles not overexpressing transgene. D: Muscles overexpressing Galgt2 had significantly fewer myofibers with central nuclei than muscles not overexpressing transgene (P < 0.001 for comparison in Gastroc and TA). Errors are SD for n = 3 to 4 animals per condition in C and D. Errors in (A) are SEM for n = 9 measurements per condition.

Figure 6

Loss of α sarcoglycan does not affect glycosylation of α dystroglycan with the CT carbohydrate in Galgt2 transgenic mice. Gastrocnemius muscle was solubilized in non-ionic detergent and precipitated with Wheat germ agglutinin (WGA), a control lectin known to bind α dystroglycan, and Wisteria floribunda agglutinin, a βGalNAc-binding lectin that binds the CT-glycosylated form of α dystroglycan. α dystoglycan was glycosylated by Galgt2 such that it could be precipitated by WFA equally well in Sgca+/−CT and Sgca−/−CT muscle. In non-CT muscle, WGA bound as much α dystroglycan as WFA did in CT muscle. Precipitates resolved on a low percentage gel (6%) and blotted with CT2 show equivalent glycoforms of α dystroglycan precipitated by WFA in Sgca+/−CT and Sgca−/−CT muscle. β dystroglycan was co-precipitated with α dystroglycan in all muscles, but was relatively reduced in WFA precipitates from Sgca−/−CT, as were β and γ sarcoglycan, due to their reduced overall expression in these lysates (see Figure 7). Full-length dystrophin (427 kDa) and a utrophin protein fragment (51 kDa) were also enriched in CT muscles. Data are representative of three experiments with similar results.

Figure 7

β, γ, δ, and ε sarcoglycan are not overexpressed in Galgt2 transgenic Sgca−/− muscle. A: Immunostaining of gastrocnemius muscle in Galgt2 transgenic (CT) and non-transgenic Sgca+/− and Sgca−/− mice. β, γ, and ε sarcoglycan were not increased in expression along myofibers in Sgca−/−CT muscle, while (B) immunostaining of α dystroglycan, β dystroglycan, dystrophin, and utrophin was high in Sgca+/−CT and Sgca−/−CT muscle. Scale bar =100 μm (A and B). C: 40 μg of SDS whole muscle lysate from gastrocnemius is loaded per lane. β, γ and δ sarcoglycan protein levels were not increased in Sgca−/−CT muscle as they were in Sgca+/−CT muscle. Utrophin, α dystroglycan, and plectin 1, by contrast, were similarly increased in both Sgca−/−CT and Sgca+/−CT muscle. Actin is shown as a control for protein loading and transfer.