¹University of Edinburgh, Institute for Neuroscience and Cardiovascular Research (INCR), Edinburgh, United Kingdom, ²University of Edinburgh, Simons Initiative for the Developing Brain (SIDB), Edinburgh, United Kingdom, ³University of Edinburgh, Centre for Clinical Brain Sciences, Edinburgh, United Kingdom, ⁴University of Strathclyde, Biomedical Engineering, Glasgow, United Kingdom

Background: Excitatory synapses are highly diverse, differing in their proteome composition, nanoscale organization, and protein turnover. This enables the classification of synapses into types and subtypes, each with a distinct spatial distribution, thus creating the synaptome architecture. Synapse dysfunction is a shared mechanism underlying psychiatric disorders, whose symptoms are managed by psychotropic drugs. Previous work has shown that olanzapine, a D2 and 5-HT2A/C receptor antagonist, modulates the excitatory synaptome architecture in mice.

Aims & Objectives: However, it is not known whether psychotropic drug classes differentially affect the synaptome. To address this, we examined individual synapses on a brain-wide scale in mice chronically administered with alprazolam, a benzodiazepine; sertraline, a selective serotonin reuptake inhibitor; or quetiapine, a D2 and 5-HT2A receptor antagonist.

Method: Adult mice (5–8.5 months) were assigned to four groups (N=83, 40F) and received sertraline (31mg/kg), alprazolam (1.5mg/kg), quetiapine (15.4mg/kg), or vehicle control (1% Tween 80 and 40% condensed milk, diluted in distilled water) via daily oral micropipette for 28 days. For each experimental group, we mapped the synaptome architecture of PSD95 and SAP102, two postsynaptic scaffolding proteins widely expressed in excitatory synapses, at single-synapse resolution across 122 brain regions. We used machine-learning algorithms developed in-house to detect, colocalize, and classify excitatory synapses into three types and 37 subtypes.

Results: We hypothesized that the three drug classes would differentially modulate the excitatory synaptome architecture in mice. We observed global, region-dependent alterations in synapse subtype densities across groups, with each drug class producing a distinct synaptome signature. Unexpectedly, the synaptome architectures of the sertraline and quetiapine groups were similar across most regions and differed from that of the alprazolam group. The magnitude of change in synapse density was greatest in the quetiapine group, followed by sertraline and alprazolam. In all three drug classes, molecular composition was a determinant of synaptic response, as SAP102-containing synapses were affected in the opposing direction to PSD95-only synapses. When the extent of change in synapse density was summed across all subtypes, each drug class displayed a unique pattern across cortical regions and layers. In particular, quetiapine exhibited marked rostral-caudal and dorsal-ventral cortical gradients.

Discussion & Conclusions: These results provide further evidence for pharmacological modulation of the synaptome architecture and demonstrate the potential to profile and classify psychotropic drugs based on their effects on brain-wide synapse composition.