¹Hokkaido University Graduate School of Medicine, Department of Psychiatry, Sapporo, Japan

Background: Neuropsychiatric disorders are vulnerable to stress, and high-intensity stress has been suggested to trigger neuroinflammatory responses. Neuroinflammation is thought to contribute to cognitive dysfunction across multiple disorders including Alzheimer's disease, depression, and schizophrenia. β-hydroxybutyrate (BHB), an endogenous ketone body that inhibits NLRP3 inflammasome activation and interleukin-1β (IL-1β) release, has attracted attention as a potential modulator of neuroinflammation in stress-related disorders. Preclinical studies have shown that neuroinflammation impairs learning and memory via proinflammatory cytokines such as IL-1β and tumor necrosis factor-α (TNF-α). Memory consists of distinct stages, including encoding, consolidation, retrieval, and reconsolidation. Previous studies have demonstrated that intrahippocampal administration of TNF-α impairs hippocampus-dependent contextual fear memory when administered before retrieval but not before conditioning, suggesting that TNF-α selectively disrupts retrieval while sparing encoding and consolidation. However, the memory stage most vulnerable to stress-induced endogenous neuroinflammation remains unclear.

Aims & Objectives: This study aimed to determine (1) which phase of hippocampus-dependent memory is disrupted by stress, and (2) whether BHB can prevent stress-induced memory impairment.

Method: Male C57BL/6J mice (8 weeks old) were trained with a contextual fear conditioning consisting of a 148 s context exposure, a 2 s 0.8 mA foot shock, and a 30 s post-shock period. Memory retrieval was tested 24 h after conditioning by re-exposure to the same context for 5 min, and freezing duration was measured. To examine phase-specific effects, 2 h restraint stress (RS) was applied at different time points: 8 h before conditioning or 8 h or 4 h before retrieval test. In some groups, BHB (250 mg/kg) was subcutaneously administered prior to RS exposure. To evaluate reconsolidation, mice that underwent the initial retrieval test were re-tested 24 h later. Hippocampal expression of inflammatory cytokines was measured at different time points after RS in a separate cohort by qPCR. All experiments were approved by the Animal Research Committee and adhered to the Guide for the Care and Use of Laboratory Animals of Hokkaido University.

Results: Freezing during the retrieval test was significantly reduced when RS was applied 8 h before retrieval test, whereas RS applied 8 h before conditioning or 4 h before retrieval test had no effect. Hippocampal expression of IL-1β and TNF-α mRNA increased following RS, peaking approximately 4 h after onset in a separate cohort. Pretreatment with BHB prior to RS exposure prevented both the stress-induced cytokine upregulation and the reduction in freezing. When mice that had experienced RS 8 h before initial retrieval test were re-tested 24 h later, freezing levels recovered to control levels, indicating that the stress-induced impairment was transient.

Discussion & Conclusions: Our findings demonstrate that acute restraint stress selectively disrupts hippocampus-dependent memory retrieval in a time-dependent manner, while leaving encoding, consolidation, and reconsolidation intact. This retrieval-specific impairment coincided with elevated proinflammatory cytokine expression in the hippocampus. Pretreatment with BHB suppressed the stress-induced neuroinflammatory response and restored memory retrieval. These findings suggest that acute stress transiently impairs memory retrieval through cytokine-mediated hippocampal inflammation, and that BHB, as an endogenous metabolic modulator, may have therapeutic potential in mitigating stress-related cognitive dysfunction.