Erratum: Exosome-associated AAV vector as a robust and convenient neuroscience tool

Correction to: Gene Therapy (2016) 23, 380–392; doi:10.1038/gt.2016.11

The initial Figure 2a was erroneously generated from a file from a mouse injected with conventional AAV9-GFP and not exo-AAV9-GFP, as described in the manuscript. We have therefore reformatted this specific panel, and corrected Figure 2a and the figure legend accordingly, now showing an image of the GFP signal detected by 2-photon microscopy after intravenous injection of exo-AAV9-GFP in a mouse. The main conclusions of the paper remain unchanged, and both conventional AAV9-GFP and exo-AAV9-GFP mediated detectable in vivo transduction of astrocytes.

Figure 2

Detection of direct GFP fluorescence by in vivo 2-photon imaging and ex vivo serial 2-photon tomography (STP). BALB/c mice were injected with 7 × 10¹¹ g.c. of exo-AAV9-CBA-GFP. After 3 weeks, the GFP fluorescent signal was detected in vivo by multiphoton imaging or ex vivo by STP (in absence of immunostaining). (a) Representative 2-photon images of two GFP-transduced astrocytes (white arrows, identified with the astrocytic marker SR101 topically applied on the brain) in the living animal after cranial window implantation. A nearby astrocyte not expressing detectable GFP is shown (blue arrow). Scale bar: 50 μm. (b) Three-dimensional reconstruction of the entire cerebellum by post-mortem STP tomography imaging, showing the whole vascular tree as well as the direct fluorescent signal across this particular region of the brain. On the right panel, numerous GFP-transduced cells could be identified on a higher magnification image of a small region of the cerebellum. Scale bar: 1000 μm. (c) GFP signal detected in one section of the cerebellum imaged by 2-photon before three-dimensional reconstruction. Scale bar: 1000 μm. On the right, two cropped regions of the initial image show GFP-positive astrocytes (yellow arrow), vascular endothelium (blue arrow) and Purkinje cells (purple arrow). Scale bar: 200 μm.