Ultrasound-Mediated Blood-Brain Barrier Opening Improves Whole Brain Gene Delivery in Mice

Abstract

Gene therapy represents a powerful therapeutic tool to treat diseased tissues and provide a durable and effective correction. The central nervous system (CNS) is the target of many gene therapy protocols, but its high complexity makes it one of the most difficult organs to reach, in part due to the blood-brain barrier that protects it from external threats. Focused ultrasound (FUS) coupled with microbubbles appears as a technological breakthrough to deliver therapeutic agents into the CNS. While most studies focus on a specific targeted area of the brain, the present work proposes to permeabilize the entire brain for gene therapy in several pathologies. Our results show that, after i.v. administration and FUS sonication in a raster scan manner, a self-complementary AAV9-CMV-GFP vector strongly and safely infected the whole brain of mice. An increase in vector DNA (19.8 times), GFP mRNA (16.4 times), and GFP protein levels (17.4 times) was measured in whole brain extracts of FUS-treated GFP injected mice compared to non-FUS GFP injected mice. In addition to this increase in GFP levels, on average, a 7.3-fold increase of infected cells in the cortex, hippocampus, and striatum was observed. No side effects were detected in the brain of treated mice. The combining of FUS and AAV-based gene delivery represents a significant improvement in the treatment of neurological genetic diseases.

Keywords: AAV9; blood-brain barrier; focused ultrasound; gene therapy; microbubbles.

Figure 1

Scanning trajectory (in red) over the mouse brain.

Figure 2

Ultrasound transmission loss through the mouse skull (n = 12) at 1.5 MHz as a function of the animal body mass. Each measurement was repeated 3 times.

Figure 3

Safety studies 24 h after FUS treatment did not show any apoptosis or hemorrhages in mice brains. (A–C): Hematoxylin-eosin staining of mice that had FUS treatment after AAV9-GFP and microbubbles (MB) injections (B) showed no difference with control mice (only AAV9-GFP and MB injections) (A). Images are representative pictures at the corpus callosum and CA1 levels. Red blood cells are indicated with an arrow on the positive control image (C). (D–F): Cleaved-caspase 3 immuno-fluorescent red staining is absent in both control (D) and FUS-treated mice (E). Images are representative pictures in the striatum. Liver sections were used as a positive control (F). Cell nuclei are counterstained by DAPI staining (blue color). (G–I): TUNEL assay did not show signs of apoptotic cells in both control (G) and FUS-treated mice (H). Images are representative pictures showing TUNEL staining (green color), DNAse-treated brain sections were used as a positive control (I). Images are representative pictures in the striatum. Cell nuclei are counterstained by DAPI staining (blue color). The scale bar represents 100 μm (n = 3 for control mice and n = 3 for the FUS-treated mice).

Figure 4

Efficiency studies 1 month after FUS treatment showed an increase of vector quantity, vector expression and number of infected cells in the brain. (A): FUS treatment increased AAV9-GFP transduction in the brain (cohen’s d effect size = 3.47). (B): GFP cDNA expression was increased in the brain of FUS-treated mice (cohen’s d effect size = 7.28). (C): Western blot quantification showed an increase in GFP expression in the brain after FUS treatment (cohen’s d effect size = 10.98). Representative western blot images with the GFP protein stained in black are shown above the graph. (D–H): Immunohistochemistry anti-GFP showed an increase in the number of GFP-positive cells in the brain of FUS-treated mice. Representative mages with the GFP protein stained in violet are shown above the graph for control mice (D) and FUS-treated mice (E). The number of GFP-positive cells was counted in the striatum (F), the hippocampus (G), and the cortex (H). Mean + SEM are represented by vertical bars, ****: p < 0.0001 Mann–Whitney rank-sum test, and (n) number of animals per group.

Figure 5

Long term safety study 1 month after FUS treatment did not show any inflammation, white matter damage or microglial activity. (A–D): GFAP immuno-fluorescent red staining in cortex (A,B) and in hippocampus (C,D) of mice that had FUS treatment after AAV9-GFP and MB injections (B,D) showed no difference with control mice (only AAV9-GFP and MB injections) (A,C). (E–H): Iba1 immuno-fluorescent green staining in cortex (E,F) and in hippocampus (G,H) of mice that had FUS treatment (F,H) showed no difference with control mice (E,G). (I–L): Olig2 immuno-fluorescent red staining in cortex (I,J) and in hippocampus (K,L) of mice that had FUS treatment (J,L) showed no difference with control mice (I,K). Cell nuclei are counterstained by DAPI staining (blue color). The scale bar represents 100 μm (n = 3 for control mice and n = 3 for the FUS-treated mice). Images *, **, and *** are low magnification of half slide of the brain stained in GFAP (*), Iba1 (**), and Olig2 (***) in order to localize where the big magnification have been taken.