Recently, AAV2.retro, a new capsid variant capable of efficient retrograde transport in brain, was generated in mice using a directed evolution approach. However, it remains unclear to what degree transport will be recapitulated in the substantially larger and more complex nonhuman primate (NHP) brain. Here, we compared the biodistribution of AAV2.retro with its parent serotype, AAV2, in adult macaques following delivery into the caudate and putamen, brain regions which comprise the striatum. While AAV2 transduction was primarily limited to the injected brain regions, AAV2.retro transduced cells in the striatum and in dozens of cortical and subcortical regions with known striatal afferents. We then evaluated the capability of AAV2.retro to deliver disease-related gene cargo to biologicallyrelevant NHP brain circuits by packaging a fragment of human mutant HTT, the causative gene mutation in Huntington's disease. Following intra-striatal delivery, pathological mHTT-positive protein aggregates were distributed widely among cognitive, motor, and limbic cortico-basal ganglia circuits. Together, these studies demonstrate strong retrograde transport of AAV2.retro in NHP brain, highlight its utility in developing novel NHP models of brain disease and suggest its potential for querying circuit function and delivering therapeutic genes in the brain, particularly where treating dysfunctional circuits, versus single brain regions, is warranted.Adeno-associated viruses (AAVs) are small, non-enveloped viruses capable of packaging single-stranded DNA genomes up to ~5 kb in length 1,2 . Originally discovered in 1965 3 , AAVs have become attractive agents for safely and effectively delivering gene cargo to a range of biological tissues (e.g. liver, muscle, retina, brain, kidney) in a wide variety of species (e.g. mouse, rat, cat, dog, pig, rabbit, horse, non-human primate, human) 1 . A number of AAV serotypes (e.g. AAV1-9, rh10, DJ, DJ/8) have been identified to date, as well as hundreds of naturally occurring capsid variants of each of these "parent" serotypes 4,5 . Differences between these AAVs are reflected in unique capsid structures, receptor specificity and tissue tropism 6 . For example, AAV6 and AAV8 transduce liver and skeletal muscle with high efficiency 4,7-9 , whereas AAV1, AAV2, AAV5 and AAV9 have been shown to transduce several types of cells in the central nervous system (CNS) including neurons, astrocytes, and photoreceptors 2,10-13 .Over the past decade, AAV-based gene therapies have begun to establish a track record of safety and success in human studies. For example, Luxturna (AAV2-RPE65) received Food and Drug Administration (FDA) approval in 2017 for the treatment of inherited retinal disease 14,15 and Zolgensma (AAV9-SMN1) was FDA-approved in May of 2019 to treat spinal muscular atrophy type 1 16 . Additionally, there are ongoing early-stage clinical trials (recruitment and active phases) evaluating AAV-based gene therapies for neurological indications