Corticosteroid treatment during sepsis influences hippocampal function in survivors

Millions of sepsis survivors annually face long-term neuropsychiatric sequelae of their illness. Corticosteroids are frequently administered in sepsis, and their use affects neuropsychiatric outcomes, but the mechanisms are unknown. We used the cecal ligation and puncture method to induce acute infection in mice and test the hypothesis that corticosteroid treatment during illness has long-term effects on hippocampal function. Functional phenotyping and hippocampal RNA-sequencing were performed in the same survivor mice to identify underlying mechanisms of behavioral and neuroendocrine outcomes. Long-term CLP survivors exhibited anxiety-like behavior, increased central hypothalamic-pituitary-adrenal (HPA) axis activity, and persistent systemic and neuro-inflammation. The relationship between gene expression and behavior suggested a protective role for inflammation and oxidative metabolism after CLP. Corticosterone treatment during illness had distinct effects on hippocampal function in survivors, including object memory impairment. The long-term behavioral effects of corticosterone treatment were associated with persistent downregulation of activity-dependent gene expression in the hippocampus. The results suggest that corticosteroid treatment for sepsis influences hippocampal function in survivors via long-lasting changes to basal hippocampal activity. Neural activity, inflammation, and oxidative metabolism should be explored as future treatment targets to modify neuropsychiatric outcomes in sepsis survivors. Overall design: Overall Design: C57/Bl6 mice used in this experiment were obtained from Jackson Laboratories and were aged 8-9 weeks old. They were housed 5 per cage on ventilated racks with a controlled light:dark cycle, temperature, and humidity. Food and water were available ad libitum. Mice were moved into the housing rooms and left undisturbed to acclimate to the environment for at least one week before starting experiments. A total of 80 mice were initially used in this experiment in two different cohorts. Each cohort consisted of 40 mice, 20 male and 20 female mice. Mice were assigned to one of four experimental groups based on a preliminary measure of locomotor activity such that each group had similar average locomotion. Within each cohort, therefore, there were 5 male and 5 female mice per group. Mice either received a cecal ligation and puncture (CLP) surgery or a sham control surgery (SHAM). Mice were given either a dose of 13 mg/kg of corticosterone or vehicle control subcutaneously for five days following surgery. Only mice from the first cohort, animal IDs 368-407, were used to obtain hippocampal samples. This cohort started with 40 mice, 20 male and 20 female mice. Following surgery, 2 female mice died resulting in a final n=38. The mice that died included one female from the CLP+saline group and one female from the CLP+cort group. Both mice died 3 days post-surgery. This resulted in the following distribution of animals among the four groups for behavioral tests and tissue collection: SHAM+saline: 5 males, 5 females SHAM+cort: 5 males, 5 females CLP+saline: 5 males, 4 females CLP+cort: 5 males, 4 females Beginning 15 days after CLP or SHAM surgery, mice underwent the following behavioral procedures prior to collection of hippocampal for RNA-sequencing on day 21: open field test (day 15), novel object recognition (day 16), and elevated zero maze (17). Euthanasia and Dissection: Mice were sacrificed at baseline in the morning 21 days post-surgery and 3 days post-testing for behavior by rapid decapitation. Trunk blood was collected and the brain was removed from skull and placed onto an inverted glass petri dish in ice. The brain was cut into two hemispheres and the entire hippocampus was then dissected out of one hemisphere and cut into dorsal and ventral halves. The same was repeated for the second hemisphere. Each mouse's dorsal hippocampal samples (2/mouse) or ventral hippocampal samples (2/mouse) were then placed into the bottom of 1.5 mL sterile Eppendorf tubes and SNAPfrozen on dry ice. Samples were stored at -80 degrees C until use for RNA extraction. Only one sample was not able to be obtained, and that was for mouse #390 ventral hippocampus (female, CLP+saline). For three mice, only one of two ventral hippocampal samples were obtained: #381 (male, CLP+cort), #402 (female, CLP+cort), and #404 (female, SHAM+cort). Dorsal hippocampus collection for those mice was completely successful.