The use of adeno-associated viruses (AAV) as gene delivery vectors has vast potential for the treatment of many severe human diseases. A small group of intensively studied AAV capsids have been propelled into pre-clinical and clinical use, and more recently, marketed products; however, many less-studied, naturally occurring capsids may also have desirable properties (e.g. potency differences, tissue tropism, reduced immunogenicity, etc.) that have yet to be thoroughly described. Within this under-studied group of capsids is AAV7, a potentially therapeutic capsid for which tropism studies are limited, and direct, head-to-head comparisons with well-described capsids have not been performed. We sought to characterize the biodistribution of this capsid more fully by several methods including quantitative PCR, immunofluorescence (IF) and immunohistochemistry (IHC), and whole-body imaging via cryofluorescence tomography (CFT) as an unbiased method to look for novel regions of transduction.
CFT of whole mice was performed three weeks after IV administration of AAV7 or AAV9 carrying an EGFP transgene driven by CAG, or co-administration of AAV7-EGFP and AAV9-tdTomato. This was compared to traditional biodistribution analysis of vector genome copies and expressed RNA transcripts. Superiority of AAV9 in mouse skeletal muscle was shown over AAV7 and was confirmed via CFT and IF. Cardiac transduction by AAV7 trended higher than AAV9 in all studies. AAV7 and AAV9 both transduce the mouse brain with equal efficiency according to PCR-based biodistribution data; however, by CFT and IHC, cell type differences were found within the brain, with AAV7 being localized mostly around blood vessels within the brain, and AAV9 widely transducing astrocytes and neurons. A surprising finding from CFT was the presence of fluorescence in facial structures, some of it highly concentrated, for both AAV7 and AAV9. As this is not typically an area sampled by traditional biodistribution methods, this finding has been largely overlooked. In particular, the mandible showed the highest intensity of EGFP fluorescence of any other tissue for both AAV7 and AAV9. The incisor also had measurable EGFP fluorescence. As the mouse mandible is surrounded by muscle tissue that likely was transduced, we conducted an IHC study to further ascertain detailed information about the transduction seen in and around mouse facial structures and investigate tropism differences between AAV7 and AAV9. Surprisingly, strong transgene expression was observed throughout the head, as verified by IHC, including regions of the cranial sinuses, teeth, and mandible for both AAV7 and AAV9, an attribute that has not been previously evaluated via traditional biodistribution methods. These data will help build a broader structure-function knowledge base in the field, present capsid engineering opportunities, and enable the use of novel capsids with unique properties. Overall, CFT represents an unbiased method to evaluate novel AAV capsids, eliminating initial bias in tissue sampling and giving precedent for further investigation of tissues of interest.
 
Key Points:
Many naturally occurring AAV capsids remain under-studied and when describing biodistribution of novel vectors, a whole-body approach, such as CFT, in conjunction with traditional PCR-based biodistribution is preferable.