Comprehensive 3D mapping reveals distinct spatial gradients of genetically-identified SST, PV, and TH interneurons across the mouse caudoputamen

Introduction: In addition to spatially organized excitatory forebrain inputs along its mediolateral, dorsoventral, and anteroposterior axes, the dorsal striatum (caudoputamen) relies on cellular diversity to subserve its myriad processing functions. Distinct GABAergic interneuron subtypes, including somatostatin (SST), parvalbumin (PV), and tyrosine hydroxylase (TH) interneurons likely subserve complementary computational roles. However, a detailed understanding of how these microcircuit components are distributed across the caudoputamen remains lacking. Methods: To address this gap, we generated a comprehensive three-dimensional atlas of genetically-defined SST-, PV-, and TH- labeled interneuron populations across the mouse caudoputamen using genetic labeling, caudoputamen-wide imaging, and voxel-wise quantification. Results: We found that genetically-defined SST and TH interneurons were relatively enriched in the ventral caudoputamen, whereas PV interneurons were enriched dorsally. In addition, PV and TH interneurons exhibited opposing anteroposterior distribution patterns, with PV interneurons enriched posteriorly and TH interneurons showing a marked decline in density toward the tail of the caudoputamen. Consequently, while the three interneuron subtypes displayed comparable densities in the functionally defined lateral caudoputamen and anterior ventromedial caudoputamen, PV interneurons predominated in the dorsomedial caudoputamen and tail of the caudoputamen. While some statistically significant sex differences were detected, the overall spatial distribution patterns of interneurons were similar across sexes. Discussion: Together, these findings reinforce the view that the caudoputamen is not a monolithic structure: in addition to excitatory and neuromodulatory inputs, inhibitory microcircuits themselves are differentially distributed across the caudoputamen, providing region-specific constraints on circuit computation. By integrating interneuron organization into existing anatomical frameworks, this atlas provides a foundation for linking dorsal striatal anatomy to function across behavioral domains.