Although the serotypes and promoters we tested expressed strongly in cortical pyramidal neurons, cerebellar Purkinje cells, olfactory granule neurons, and striatal interneurons, they produced very little expression in cortical interneurons and granule neurons of the dentate gyrus and cerebellum. Expression in these cell types might be attained using different serotypes and promoters, but must be tested empirically. Finally, there is a strict temporal window during which this technique can be used. Injections must be performed within the first PI3K inhibitor 12–24 h after birth for AAV1, and within the first few days for AAV8. The timing of AAV injection may also limit which cell types can be transduced, as several neuronal populations

are generated after birth. After injection, however, expression of viral transgenes can be readily delayed

using temporal control elements such as Cre recombinase – estrogen receptor and tTA. By optimising its natural mosaic transduction pattern, we discovered that neonatal viral transgenesis opens a wide range of experimental opportunities that are not possible with existing isocitrate dehydrogenase inhibitor methods. Cell-autonomous and cell-extrinsic effects can now be readily distinguished. Purkinje neurons can now be easily manipulated and imaged in vivo. New constructs can be rapidly screened without germline transgenesis. The final advantage of the approach is the rising availability of compatible off-the-shelf viral preparations (e.g. Penn Vector Core and UNC Gene Therapy Center) and vectors (e.g. Addgene) that can be custom packaged into a variety of serotypes. These

resources for viral manipulation complement Fludarabine nmr a growing community of mouse repositories where newly characterised mutant strains can be purchased online (e.g. Jackson Laboratories, MMRRC, GENSAT, EMMA). As both the pattern and expression level of viral-delivered transgenes can depend on a number of factors including the transgene itself, construct design (i.e. promoters and enhancers), capsid serotype, quality of the viral preparation, and viral titer, each new application will require some optimisation. However, the richness of viral manipulation and the rate at which it has recently advanced suggest that, with additional experimentation, a wide range of cell type specificities and novel applications are within reach. We thank Kazuhiro Oka and the Baylor College of Medicine Viral Vector Core for AAV production, Anna Gumpel, Carolyn Allen, Yuanyuan Zhang, and Bryan Song for mouse care, Bernard Lee and Bernard Kuecking from Zeiss for microscope support, Ben Arenkiel for sharing the EF1α-iCre-2A-tdTomato AAV vector, and Roy Sillitoe and Ben Arenkiel for helpful comments on the manuscript. Grant support was from American Health Assistance Foundation Alzheimer’s Disease Research Grant A2010097, National Institute of Aging R21 AG038856, and National Institutes of Health Office of the Director New Innovator Award DP2 OD001734. None.