A comparative study of N-glycolylneuraminic acid (Neu5Gc) and cytotoxic T cell (CT) carbohydrate expression in normal and dystrophin-deficient dog and human skeletal muscle

Abstract

The expression of N-glycolylneuraminic acid (Neu5Gc) and the cytotoxic T cell (CT) carbohydrate can impact the severity of muscular dystrophy arising from the loss of dystrophin in mdx mice. Here, we describe the expression of these two glycans in skeletal muscles of dogs and humans with or without dystrophin-deficiency. Neu5Gc expression was highly reduced (>95%) in muscle from normal golden retriever crosses (GR, n = 3) and from golden retriever with muscular dystrophy (GRMD, n = 5) dogs at multiple ages (3, 6 and 13 months) when compared to mouse muscle, however, overall sialic acid expression in GR and GRMD muscles remained high at all ages. Neu5Gc was expressed on only a minority of GRMD satellite cells, CD8⁺ T lymphocytes and macrophages. Human muscle from normal (no evident disease, n = 3), Becker (BMD, n = 3) and Duchenne (DMD, n = 3) muscular dystrophy individuals had absent to very low Neu5Gc staining, but some punctate intracellular muscle staining was present in BMD and DMD muscles. The CT carbohydrate was localized to the neuromuscular junction in GR muscle, while GRMD muscles had increased expression on a subset of myofibers and macrophages. In humans, the CT carbohydrate was ectopically expressed on the sarcolemmal membrane of some BMD muscles, but not normal human or DMD muscles. These data are consistent with the notion that altered Neu5Gc and CT carbohydrate expression may modify disease severity resulting from dystrophin deficiency in dogs and humans.

Figure 1. Neu5Gc expression in Golden Retriever cross (GR) and Golden Retriever Muscular Dystrophy (GRMD) dog skeletal muscle relative to mouse.

(A) Neu5Gc-specific affinity-purified chicken IgY was used to stain skeletal muscles from 6 month-old normal Golden Retriever (GR) or Golden Retriever Muscular Dystrophy (GRMD) dogs. An example of staining from a severely affected and a mildly affected GRMD dog are shown. (B) Time-matched images in A were compared to normal mouse skeletal muscle immunostained with the same reagents. Non-immune chicken IgY was used as a control for background staining in A and B. Bar is 200 µm for all panels in A and B.

Figure 2. Quantification of Neu5Ac and Neu5Gc levels in normal and dystrophin-deficient mouse and dog muscles.

Tibialis anterior (TA) and soleus (Sol) muscles were analyzed from 11 wild type (WT) mice, TA and gastrocnemius (GS) muscles were analyzed from 8 mdx mice, and vastus lateralis (VL) and cranial sartorius (CS) muscles were analyzed from 2 golden retriever (GR) and 5 golden retriever muscular dystrophy (GRMD) dogs. Average Neu5Gc as a percentage of total sialic acid is shown. Errors are standard deviation (SD). ***P<0.001, for all dog vs. mouse muscle comparisons.

Figure 3. Neu5Gc expression in GRMD muscle at different ages.

Three different GRMD cases (GRMD1, 2, and 3) were biopsied and muscles stained for Neu5Gc at 3, 6 and 13 months of age. Time-matched images of staining of the cranial sartorius (A) and vastus lateralis (B) muscles are shown. Bar is 50 µm for all panels in A and B.

Figure 4. Neu5Gc immunostaining of satellite cells, T lymphocytes and macrophages in GRMD skeletal muscle.

GRMD muscle (cranial sartorius) was triple stained for Neu5Gc (green), Pax7 (A), CD11b (B), CD4 (C), CD8 (D) (all red), and DAPI (blue). Arrows indicated co-staining of Neu5Gc and Pax7 (in A), CD11b (in B) or CD4 (in C). Bar is 100 µm (A) and 50 µm (B–D) for all panels.

Figure 5. Quantification of CD4−, CD8−, CD11b− and Pax7−stained cells in GR and GRDM muscle and of Neu5Gc co-staining in GRMD muscle.

(A) Cranial sartorius muscle sections from GR and GRMD dogs were stained with markers for satellite cells (Pax7), T lymphocytes (CD4 or CD8) or macrophages (CD11b) and quantified for numbers of cells stained per 40X visual field. (B) The percentage of cells co-stained for Neu5Gc and CD4, CD8, CD11b or Pax7 in GRMD muscles was quantified. Errors are SEM. ***P<0.001, for each GR vs. GRMD comparison in A.

Figure 6. Neu5Gc co-staining with β spectrin, CD11b, CD8 or Pax7 in normal, BMD or DMD human muscle.

(A) Otherwise normal, Becker muscular dystrophy (BMD) and Duchenne muscular dystrophy (DMD) muscle biopsy sections were stained for Neu5Gc (green), β spectrin (red) and DAPI (blue). Arrows indicate Neu5Gc puncta in cytoplasmic or perimembranous regions of BMD and DMD skeletal myofibers. (B) DMD muscle was co-stained for Neu5Gc and CD11b, CD8 or Pax7 (and DAPI). Bar is 50 µm for all panels in A and B.

Figure 7. Co-localization of Neu5Gc staining with markers for endosomes and Golgi in BMD and DMD muscle.

(A) BMD muscle co-stained for Neu5Gc (green) and clathrin (red), a marker of endosomes. Merged image on right shows overlap of Neu5Gc and clathrin expression in yellow. Arrow marks several examples of co-staining. (B) DMD muscle co-stained for Neu5Gc (green) with LAMP1, a lysosomal marker, 58K Golgi, a Golgi marker, or calnexin, and endoplasmic reticulum marker, all in red, and DAPI (blue). Arrow marks region of coincident staining (yellow) for Neu5Gc and 58K Golgi. Bar is 50 µm for all panels in A and B.

Figure 8. Co-staining of CT carbohydrate with macrophages in GRMD muscle.

(A) Single and merged staining for CT2 (green) with CD11b (red, to mark macrophages) and DAPI (blue). Arrow shows co-incident staining for CT2 and CD11b. (B) Single and merged staining for CT2 (green) with Pax7 (red, to mark satellite cells) and DAPI (blue). Arrow shows lack of coincident staining for CT2 and Pax7. (C) Examples of WFA staining (green) with embryonic myosin (red) and DAPI (blue) (two left panels). Examples of control staining for staining shown in A and B using only secondary antibodies with DAPI (two right panels). Bar is 50 µm for all panels in A, B and C.

Figure 9. Comparison of CT1 and CT2 immunostaining in DMD and normal human muscle.

CT1 and CT2 were used to immunostain two different DMD (A) and two different normal human (B) muscle biopsies, compared to secondary antibody control. These were compared to time-matched images taken from mdx muscles (C), diaphragm (left panel) or tibialis anterior (right panel), that had been infected with rAAV(rh.74).MCK.GALGT2 for 12 weeks. Bar is 200 µm for all panels in A, B and C.

Figure 10. Co-staining of CT1 with βspectrin in BMD muscles.

Sections from DMD, BMD and normal human muscle biopsies sections were co-stained with CT1 (green), β spectrin(red) and DAPI (blue). Merged sarcolemmal membrane staining for CT1 and β spectrin is orange-yellow. PN represents intramuscular peripheral nerve. Bar is 50 µm for all panels.