Background: The natural compound asperosaponin VI has shown potential as an antidepressant, but how it works is unclear. Here, we explored its effects on mice exposed to chronic mild stress (CMS) and the underlying molecular pathways. Methods: Mice were exposed to CMS for 3 weeks followed by asperosaponin VI (40 mg/kg) or imipramine (20 mg/kg) for another 3 weeks. Depression-like behaviors were assessed in the forced swimming test (FST), sucrose preference test (SPT), tail suspension test (TST). Microglial phenotypes were evaluated using immunofluorescence staining, real-time quantitative PCR and enzyme-linked immunosorbent assays in hippocampus of mice. In some experiments, stressed animals were treated with the PPAR-γ antagonist GW9662 to examine its involvement in the effects of asperosaponin VI. Blockade of PPAR-γ in asperosaponin VI-treated primary microglia in the presence of lipopolysaccharide (LPS) was executed synchronously. The nuclear transfer of PPAR-γ in microglia was detected by immunofluorescence staining in vitro and in vivo. A co-cultured model of neuron and microglia was used for evaluating the regulation of ASA VI on the microglia–neuron crosstalk molecules. Results: Asperosaponin VI ameliorated depression-like behaviors of CMS mice based on SPT, TST and FST, and this was associated with a switch of hippocampal microglia from a pro-inflammatory (iNOS-Iba1) to neuroprotective (Arg-1-Iba1) phenotype. CMS reduced the expression levels of PPAR-γ and phosphorylated PPAR-γ in hippocampus, which asperosaponin VI partially reversed. GW9662 treatment prevented the nuclear transfer of PPAR-γ in asperosaponin VI-treated microglia and inhibited the induction of Arg-1 microglia. Blockade of PPAR-γ signaling also abolished the ability of asperosaponin VI to suppress pro-inflammatory cytokines while elevating anti-inflammatory cytokines in the hippocampus of CMS mice. The asperosaponin VI also promoted interactions between hippocampal microglia and neurons by enhancing CX3CL1/CX3CR1 and CD200/CD200R, and preserved synaptic function based on PSD95, CamKII β and GluA levels, but not in the presence of GW9662. Blockade of PPAR-γ signaling also abolished the antidepressant effects of asperosaponin VI in the SPT, TST and FST. Conclusion: CMS in mice induces a pro-inflammatory microglial phenotype that causes reduced crosstalk between microglia and neuron, inflammation and synaptic dysfunction in the hippocampus, ultimately leading to depression-like behaviors. Asperosaponin VI may ameliorate the effects of CMS by inducing microglia to adopt a PPAR-γ-dependent neuroprotective phenotype.