Adeno-associated virus serotypes 9 and rh10 mediate strong neuronal transduction of the dog brain

Abstract

Canine models have many advantages for evaluating therapy of human central nervous system (CNS) diseases. In contrast to nonhuman primate models, naturally occurring canine CNS diseases are common. In contrast to murine models, the dog's lifespan is long, its brain is large and the diseases affecting it commonly have the same molecular, pathological and clinical phenotype as the human diseases. We compared the ability of four intracerebrally injected adeno-associated virus vector (AAV) serotypes to transduce the dog brain with green fluorescent protein as the first step in using these vectors to evaluate both delivery and efficacy in naturally occurring canine homologs of human diseases. Quantitative measures of transduction, maximum diameter and area, identified both AAV2/9 and AAV2/rh10 as significantly more efficient than either AAV2/1 or AAV2/5 at transducing cerebral cortex, caudate nucleus, thalamus and internal capsule. Fluorescence co-labeling with cell-type-specific antibodies demonstrated that AAV2/9 and AAV2/rh10 were capable of primarily transducing neurons, although glial transduction was also identified and found to be more efficient with the AAV2/9 vector. These data are a prerequisite to evaluating the efficacy of recombinant AAV vectors carrying disease-modifying transgenes to treat naturally occurring canine models in preclinical studies of human CNS disease therapy.

Figure 1

Injection sites in the dog brain. A dorsal plane image (a) and transverse images (b–d) of the dog brain are shown. The injection sites in the left cerebral hemisphere are identified by Xs. A Hamilton syringe and needle were used to deliver each serotype to the brain. Each dog was injected with one rAAV serotype, and each serotype was injected into two dogs. Transduction of the caudate nucleus (CN) was assessed at site 1 (b); transduction of the thalamus (THAL) was assessed at sites 2 (c) and 3 (d), and transduction of the white matter (internal capsule (IC)) was assessed at site 4 (d).

Figure 2

GFP fluorescence in a dog brain injected with AAV2/9. (a) A transverse section of the dog brain is shown at the level of injection site 4 in the left cerebral hemisphere. The vertical rectangle shows the region of brain photographed in b. The smaller horizontal rectangle shows the region photographed in c. (b) Transduction of somata (arrowheads) characterized by large punctate areas of the thalamus (THAL), and fluorescence of white matter axons (arrows) characterized by linear regions of the internal capsule are present. Dorsal is at the top of the image and ventral at the bottom. The scale bar for the photomicrograph is 1 mm. (c) GFP fluorescence in fibers of the corpus callosum of the right cerebral hemisphere. Fluorescent axons (arrows) were seen extending from the white matter of the injected hemisphere, crossing the corpus callosum (CC) and entering the contralateral uninjected hemisphere.

Figure 3

Boxplots showing the median, and 25th and 75th percentile of the data distribution for the maximum diameter (a) and maximum area (b) of transduction for each of the four rAAV serotypes. The Kruskal–Wallis test was used to compare data between groups and the Mann–Whitney test was used for pairwise comparison. *P<0.05. Significantly increased diameter was identified in both AAV2/9 and AAV2/rh10 compared with AAV2/1, whereas only AAV2/9 was significantly greater than AAV2/5. Significantly increased area of transduction was also identified in both AAV2/9 and AAV2/rh10 compared with either AAV2/1 or AAV2/5. Finally, AAV2/9 and AAV2/rh10 were not significantly different from each other in either maximum diameter or area.

Figure 4

Neuronal transduction by AAV2/9. NeuN-positive neurons of the thalamus transduced by GFP in dog brain injected with AAV2/9 vector. (a) × 10 magnification showing GFP fluorescence (green), NeuN fluorescence (red), 4',6-diamidino-2-phenylindole (blue) and regions of co-labeling (orange). Rectangle shows × 100 magnification (b–d) with orange co-labeled neurons (b) red NeuN-positive cells alone (c) and GFP-positive cells alone (d). Arrowheads show astrocytes.

Figure 5

Astrocytic transduction by AAV2/9. GFAP-positive astrocytes of the caudate nucleus of the dog brain transduced with the AAV2/9 vector. Left column shows × 20 magnification showing GFP fluorescence (green), GFAP fluorescence (red) and regions of co-labeling (orange). Region within rectangle is present in second column and shows × 100 magnification with orange co-labeled astrocytes, red GFAP-positive cells alone and GFP-positive cells alone.

Figure 6

Oligodendrocytic transduction by GFP. GFP-positive, OLIG2-positive oligodendrocytes of the thalamus of the dog brain transduced with the AAV2/rh10 vector (a), × 10 magnification showing GFP fluorescence (green), nuclear OLIG2 fluorescence (red), 4',6-diamidino-2-phenylindole (blue) and regions of co-labeling (orange). Rectangle shows × 40 magnification (b-d) with co-labeled oligodendrocytes (b), OLIG2-positive cells alone (c) and GFP-positive cells alone (d).