Adeno-Associated Virus 8 and 9 Myofibre Type/Size Tropism Profiling Reveals Therapeutic Effect of Microdystrophin in Canines

Abstract

Background: Adeno-associated virus (AAV) 8 and 9 are in clinical trials for treating neuromuscular diseases such as Duchenne muscular dystrophy (DMD). Muscle consists of myofibres of different types and sizes. However, little is known about the fibre type and fibre size tropism of AAV in large mammals.

Methods: We evaluated fibre type- and size-specific transduction properties of AAV8 and AAV9 in 17 dogs that received systemic gene transfer (dose 1.94 ± 0.52 × 10¹⁴ vg/kg; injected at 2.86 ± 0.30 months; harvested at 20.79 ± 3.30 months). For AAV8, two DMD dogs and three carrier dogs received an alkaline phosphatase (AP) reporter vector, and five DMD dogs received a four-repeat microdystrophin (uDys) vector. For AAV9, one normal and one DMD dog received the AP vector, and five DMD dogs received a five-repeat uDys vector. Association between AAV transduction and the fibre type/size was studied in three muscles that showed mosaic transgene expression, including the biceps femoris, teres major and latissimus dorsi.

Results: Transgene expression was detected in 30%-45% of myofibres. In the AP reporter vector-injected dogs, neither AAV8 nor AAV9 showed a statistically significant fibre type preference. Interestingly, AP expression was enriched in smaller fibres. In uDys-treated DMD dogs, slow and fast myofibres were equally transduced. Notably, uDys-expressing myofibres were significantly larger than uDys-negative myofibres irrespective of the AAV serotype (p < 0.0001). In AAV8 uDys vector-injected dogs, the mini-Feret diameter was 15%, 16% and 23% larger in uDys-positive slow, fast and hybrid fibres, respectively; the cross-sectional area was 30%, 34% and 46% larger in uDys-positive slow, fast and hybrid fibres, respectively. In AAV9 uDys vector-injected dogs, the mini-Feret diameter was 12%, 13% and 25% larger in uDys-positive slow, fast and hybrid fibres, respectively; the cross-sectional area was 25%, 28% and 59% larger in uDys-positive slow, fast and hybrid fibres, respectively.

Conclusions: Our studies suggest that AAV8 and AAV9 transduce fast and slow myofibres at equivalent efficiency. Importantly, uDys therapy effectively prevented dystrophic myofibre atrophy. Our study provides important insight into systemic muscle AAV delivery in large mammals and supports further development of uDys gene therapy for DMD.

Keywords: Duchenne muscular dystrophy (DMD); adeno‐associated virus (AAV); canine model; microdystrophin; myofibre size; myofibre type.

FIGURE 1

Quantification of transgene expression in muscles from systemic AAV‐injected dogs. (A) Cartoon illustration of AAV vectors used in the study. In the alkaline phosphatase (AP) vector, the heat‐resistant human placental AP gene was expressed from the ubiquitous RSV promoter. In the microdystrophin (uDys) vector, the uDys gene was expressed from the muscle‐specific CK8e promoter. (B) Quantification of transgene‐positive myofibres. Five dogs (two affected and three carrier) received systemic AAV8 AP vector injection. Two dogs (one affected and one normal) received systemic AAV9 AP vector injection. Five affected dogs received systemic AAV8 uDys vector injection. Five affected dogs received systemic AAV9 uDys vector injection. Two muscles (biceps femoris and latissimus dorsi) were collected from the AAV9 AP vector–injected affected dog. Three muscles (biceps femoris, teres major and latissimus dorsi) were collected from each dog for the remaining dogs. Each data point represents the result from one muscle in one dog. ns, not significant.

FIGURE 2

Evaluation of myofibre type composition and AP expression following systemic injection of the AAV8 AP reporter vector. Five dogs (two affected and three carrier dogs) received the vector. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in each dog. (A) Representative photomicrographs of myofibre type immunofluorescence staining and AP histochemical staining. Asterisk, the same myofibre in serial sections. MyHC, myosin heavy chain. Blue, slow Type I myofibre; red, fast Type IIa/IIx myofibre; purple, hybrid myofibre with both Type I and Type IIa/IIx fibres; green, Type IIb myofibre (note: no Type IIb fibre was detected); white, laminin. (B) Myofibre type composition. Each point represents data from one muscle in one dog. (C) Fibre type composition of AP‐positive myofibres. Each point represents data from one muscle in one dog. (D) Fibre type normalized AP expression. Each point represents data from one muscle in one dog. ns, not significant, ****p < 0.0001.

FIGURE 3

Evaluation of myofibre type composition and AP expression following systemic injection of the AAV9 AP reporter vector. Two dogs (one affected and one normal) received the vector. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in the normal dog. Two muscles (biceps femoris and latissimus dorsi) were examined in the affected dog. (A) Representative photomicrographs of myofibre type immunofluorescence staining and AP histochemical staining. Asterisk, the same myofibre in serial sections. MyHC, myosin heavy chain. Blue, slow Type I myofibre; red, fast Type IIa/IIx myofibre; purple, hybrid myofibre with both Type I and Tpe IIa/IIx fibres; green, Type IIb myofibre (note: no Type IIb fibre was detected); white, laminin. (B) Myofibre type composition. Each point represents the percentage of the corresponding myofibre type from one muscle. (C) Fibre type composition of AP‐positive myofibres. Each point represents data from one muscle in one dog. (D) Fibre type normalized AP expression. Each point represents data from one muscle in one dog. ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 4

Evaluation of myofibre type composition and uDys expression following systemic injection of the AAV8 uDys vector. Five affected dogs received the vector. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in each dog. (A) Representative photomicrographs of myofibre type and uDys immunofluorescence staining. Asterisk, the same myofibre. MyHC, myosin heavy chain. Blue in the cytosol, slow Type I myofibre; red in the cytosol, fast Type IIa/IIx myofibre; purple in the cytosol, hybrid myofibre with both Type I and Type IIa/IIx fibres; green in the cytosol, Type IIb myofibre (note: no Type IIb fibre was detected); green on the sarcolemma, uDys; white on the sarcolemma, laminin. (B) Myofibre type composition. Each point represents data from one muscle in one dog. (C) Fibre type composition of uDys‐positive myofibres. Each point represents data from one muscle in one dog. (D) Fibre type normalized uDys expression. Each point represents data from one muscle in one dog. ns, not significant, *p < 0.05, **p < 0.01, ****p < 0.0001.

FIGURE 5

Evaluation of myofibre type composition and uDys expression following systemic injection of the AAV9 uDys vector. Five affected dogs received the vector. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in each dog. (A) Representative photomicrographs of myofibre type and uDys immunofluorescence staining. Asterisk, the same myofibre. MyHC, myosin heavy chain. Blue in the cytosol, slow Type I myofibre; red in the cytosol, fast Type IIa/IIx myofibre; purple in the cytosol, hybrid myofibre with both Type I and Type IIa/IIx fibres; green in the cytosol, Type IIb myofibre (note: no type IIb fibre was detected); green on the sarcolemma, uDys; white on the sarcolemma, laminin. (B) Myofibre type composition. Each point represents data from one muscle in one dog. (C) Fibre type composition of uDys‐positive myofibres. Each point represents data from one muscle in one dog. (D) Fibre type‐normalized uDys expression. Each point represents data from one muscle in one dog. ns, not significant, *p < 0.05, ****p < 0.0001.

FIGURE 6

Evaluation of fibre size distribution in AP‐positive and AP‐negative myofibres following systemic injection of the AP reporter vector. Five dogs (two affected and three carrier dogs) received the AAV8 AP vector. Two dogs (one affected and one normal) received the AAV9 AP vector. Two muscles (biceps femoris and latissimus dorsi) were examined in the AAV9 AP vector–injected affected dog. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in the remaining dogs. (A) Scatter plots of the mini‐Feret diameter (left panel) and cross‐sectional area (right panel) of individual myofibres in dogs that received the AAV8 AP vector. AP‐positive slow fibres, N = 899; AP‐negative slow fibres, N = 1677; AP‐positive fast fibres, N = 1639; AP‐negative fast fibres, N = 3622; AP‐positive hybrid fibres, N = 80; AP‐negative hybrid fibres, N = 262. (B) Scatter plots of the mini‐Feret diameter (left panel) and cross‐sectional area (right panel) of individual myofibres in dogs that received the AAV9 AP vector. AP‐positive slow fibres, N = 314; AP‐negative slow fibres, N = 263; AP‐positive fast fibres, N = 301; AP‐negative fast fibres, N = 945; AP‐positive hybrid fibres, N = 9; AP‐negative hybrid fibres, N = 45. Data are presented as mean ± 95% confidence interval. Each point represents data from one myofibre. ns, not significant, **p < 0.01, ****p < 0.0001.

FIGURE 7

Evaluation of fibre size distribution in uDys‐positive and uDys‐negative myofibres following systemic injection of the uDys vector. Five affected dogs received the AAV8 uDys vector. Five affected dogs received the AAV9 uDys vector. Three muscles (biceps femoris, teres major and latissimus dorsi) were examined in each dog. (A) Scatter plots of the mini‐Feret diameter (left panel) and cross‐sectional area (right panel) of individual myofibres in dogs that received the AAV8 uDys vector. uDys‐positive slow fibres, N = 946; uDys‐negative slow fibres, N = 1261; uDys‐positive fast fibres, N = 1762; uDys‐negative fast fibres, N = 2870; uDys‐positive hybrid fibres, N = 109; uDys‐negative hybrid fibres, N = 444. (B) Scatter plots of the mini‐Feret diameter (left panel) and cross‐sectional area (right panel) of individual myofibres in dogs that received the AAV9 uDys vector. uDys‐positive slow fibres, N = 573; uDys‐negative slow fibres, N = 908; uDys‐positive fast fibres, N = 2205; uDys‐negative fast fibres, N = 2293; uDys‐positive hybrid fibres, N = 78; uDys‐negative hybrid fibres, N = 329. Data are presented as mean ± 95% confidence interval. Each point represents data from one myofibre. ****p < 0.0001.