Unmoving and uninflamed: Characterizing neuroinflammatory dysfunction in the Wistar‐Kyoto rat model of depression

Abstract

AbstractReductionistic research on depressive disorders has been hampered by the limitations of animal models. Recently, it has been hypothesized that neuroinflammation is a key player in depressive disorders. The Wistar‐Kyoto (WKY) rat is an often‐used animal model of depression, but no information so far exists on its neuroinflammatory profile. As such, we compared male young adult WKY rats to Wistar (WS) controls, with regard to both behavioral performance and brain levels of key neuroinflammatory markers. We first assessed anxiety‐ and depression‐like behaviors in a battery consisting of the Elevated Plus Maze (EPM), the Novelty Suppressed Feeding (NSFT), Open Field (OFT), Social Interaction (SIT), Forced Swim (FST), Sucrose Preference (SPT), and Splash tests (ST). We found that WKY rats displayed increased NSFT feeding latency, decreased OFT center zone permanence, decreased EPM open arm permanence, decreased SIT interaction time, and increased immobility in the FST. However, WKY rats also evidenced marked hypolocomotion, which is likely to confound performance in such tests. Interestingly, WKY rats performed similarly, or even above, to WS levels in the SPT and ST, in which altered locomotion is not a significant confound. In a separate cohort, we assessed prefrontal cortex (PFC), hippocampus and amygdala levels of markers of astrocytic (GFAP, S100A10) and microglial (Iba1, CD86, Ym1) activation status, as well as of three key proinflammatory cytokines (IL‐1β, IL‐6, TNF‐α). There were no significant differences between strains in any of these markers, in any of the regions assessed. Overall, results highlight that behavioral data obtained with WKY rats as a model of depression must be carefully interpreted, considering the marked locomotor activity deficits displayed. Furthermore, our data suggest that, despite WKY rats replicating many depression‐associated neurobiological alterations, as shown by others, this is not the case for neuroinflammation‐related alterations, thus representing a novel limitation of this model.