Fig. 4: Correlated PharmacoSTORM molecular and anatomical imaging visualizes Fluo-CAR binding surrounding dopaminergic nerve terminals.

a, b Comparative analysis of the distribution of fluo-cariprazine-binding sites and well-known dopaminergic signaling proteins in the ventral part of the mouse forebrain. a Fluo-CAR-treated slices (300 nM) were immunostained for DARPP-32. Neither the hilar subregion (outlined by yellow Fluo-CAR binding) nor the cell bodies in the granular subregion (visualized by blue DAPI staining) of the Islands of Calleja contain the striatal marker protein DARPP-32 (purple). b In contrast to the characteristic DARPP-32 immunonegativity, immunostaining for tyrosine hydroxylase (TH) in the Islands of Calleja suggests that dopaminergic axons densely innervate this brain region. c Correlated confocal and PharmacoSTORM imaging of TH-containing-varicosities and Fluo-CAR binding, respectively. Although the dopaminergic varicosities are surrounded by a high density of Fluo-CAR binding sites, Fluo-CAR apparently avoids TH-immunopositive dopaminergic nerve terminals. d Quantitative nanoscale spatial analysis of Fluo-CAR binding in relation to dopaminergic afferents of the Islands of Calleja. Data are normalized to the average density of PharmacoSTORM localization points (LPs) on each image. Note that if the same number of Fluo-CAR LPs would be randomly distributed then TH-containing varicosities would bear the average density on each image. Two-tailed Wilcoxon signed-rank test was used to test if median is different from 1 (n = 12, data are from five animals, presented as mean ± SEM; P = 0.0048 for Fluo-CAR LPs, P = 0.7836 for random distribution). e Dual-color PharmacoSTORM and ImmunoSTORM imaging of Fluor-CAR labeling and TH-immunostaining. f Random nanoscale distribution of Fluo-CAR-binding sites around dopaminergic axon terminals. To test the nanoscale relationship between the site of dopamine synthesis and dopamine binding sites, the nearest-neighbor distance of each Fluo-CAR-binding site was measured from a 2D convex hull fitted onto the TH ImmunoSTORM LPs. The identical cumulative distribution of Fluo-CAR LPs and randomized LPs (Kolmogorov–Smirnov test, two-tailed P = 0.8749) indicates the lack of specific Fluo-CAR binding site nanoclusters in the proximity of TH⁺ boutons as one would expect in case of synaptic specializations and rather suggests volume transmission as the primary mode of dopaminergic signaling.