Exercise has been considered for the treatment of depression, but the mechanism by which exercise improves depression is still unclear. To clarify the mechanism, rats were randomly divided into the control, chronic unpredictable stress (CUS)/standard and CUS/running groups. The rats in the CUS/running group ran for four weeks. In this study, a sucrose preference test (SPT) was used to evaluate the depression-like symptoms in the rats, and western blot, immunohistochemical and stereological analyses were performed to study the expression of synaptic-related proteins in the hippocampus and the changes in excitatory synapses in each sub-region. The results show that sucrose preference in the CUS/standard group was significantly lower than that in the control group, but in the CUS/running group, sucrose preference was higher than that in the CUS/standard group. Surprisingly, there was no difference in the synaptic-related proteins in the hippocampus among groups. The CUS/standard group exhibited fewer spinophilin+ (Sp+) dendritic spines representing excitatory synapses in CA1, CA3 and dentate gyrus (DG) of the hippocampus than the control group, whereas the CUS/running group exhibited significantly more Sp+ dendritic spines in these regions than the CUS/standard group, indicating that excitatory synapses were reduced in depressed rats and that running exercises could reverse this change. We hypothesize that the changes in the number of excitatory synapses better reflect the changes in depressive symptoms than the level of synaptic proteins and that the effect of exercise on excitatory synapses in the sub-regions of the hippocampus may be an important structural indicator of the improvement of depressive symptoms.

Chronic exposure to stressful conditions may affect spatial learning and memory abilities and the brain structure, and disruptions in oligodendrocyte function may cause cognitive dysfunction. Leucine-rich repeat and immunoglobulin-like domain-containing protein 1 (LINGO-1) is a potent negative regulator of oligodendrocytes and axon myelination. However, the questions we sought to answer in this study are whether hippocampal oligodendrocytes are involved in the pathological process of spatial learning and memory impairments induced by chronic stress (CS) and whether antibodies targeting LINGO-1 improve stress-induced spatial learning and memory impairments by protecting the hippocampal oligodendrocytes in stressed rats. After 4 weeks of CS, rats were randomly divided into either the CS standard group or anti-LINGO-1 group. The anti-LINGO-1 group was treated with an anti-LINGO-1 antibody (8 mg/kg) for 3 weeks; all rats were assessed in the Morris water maze. Immunohistochemical staining and modern stereological methods were used to precisely quantify the total number of 2',3'-cyclic nucleotide 3'-phosphodiesterase-positive (CNPase+) oligodendrocytes in each subregion of the hippocampus. At the behavioural level, after three weeks of treatment, the anti-LINGO-1 group displayed significantly more platform crossings in the Morris water maze test than the CS standard group. The total swimming distance and swimming speed were not significantly different. In the open field test, the percentage of distance travelled in the central region did not differ between the CS standard group and control group or between the anti-LINGO-1 group and the CS standard group. Unbiased stereological analyses revealed significantly greater total numbers of CNPase+ cells in the CA3 and dentate gyrus (DG) areas of the hippocampus in the anti-LINGO-1 group than in the CS standard group. A significant difference in the total number of CNPase+ cells was not observed in the hippocampal CA1 region between the anti-LINGO-1 and CS standard groups. Based on the results of the present study, the anti-LINGO-1 antibody alleviated spatial memory impairments and protected oligodendrocytes in the hippocampus of chronically stressed rats.