The Period 2 Enhancer Nobiletin as Novel Therapy in Murine Models of Circadian Disruption Resembling Delirium

Abstract

Objectives: Delirium occurs in approximately 30% of critically ill patients, and the risk of dying during admission doubles in those patients. Molecular mechanisms causing delirium are largely unknown. However, critical illness and the ICU environment consistently disrupt circadian rhythms, and circadian disruptions are strongly associated with delirium. Exposure to benzodiazepines and constant light are suspected risk factors for the development of delirium. Thus, we tested the functional role of the circadian rhythm protein Period 2 (PER2) in different mouse models resembling delirium.

Design: Animal study.

Setting: University experimental laboratory.

Subjects: Wildtype, Per2 mice.

Interventions: Midazolam, lipopolysaccharide (lipopolysaccharide), constant light, nobiletin, or sham-treated animals.

Measurements and main results: Midazolam significantly reduced the expression of PER2 in the suprachiasmatic nucleus and the hippocampus of wild-type mice. Behavioral tests following midazolam exposure revealed a robust phenotype including executive dysfunction and memory impairment suggestive of delirium. These findings indicated a critical role of hippocampal expressed PER2. Similar results were obtained in mice exposed to lipopolysaccharide or constant light. Subsequent studies in Per2 mice confirmed a functional role of PER2 in a midazolam-induced delirium-like phenotype. Using the small molecule nobiletin to enhance PER2 function, the cognitive deficits induced by midazolam or constant light were attenuated in wild-type mice.

Conclusions: These experiments identify a novel role for PER2 during a midazolam- or constant light-induced delirium-like state, highlight the importance of hippocampal PER2 expression for cognitive function, and suggest the PER2 enhancer nobiletin as potential therapy in delirium-like conditions associated with circadian disruption.

Figure 1. Midazolam downregulates PER2 in the SCN and the hippocampus

Wildtype mice (C57BL6/J) were injected with either midazolam (10 mg/kg i.p.) or LPS (100 μg/kg i.p.) and compared to saline treated controls. Brain tissue was harvested, mRNA was isolated using RNeasy Mini Kit (Qiagen), cDNA was generated using miScript RT II kits (Qiagen), and transcript levels were determined by quantitative real-time RT-PCR (iCycler or iCycler IQ; Bio-Rad Laboratories Inc.). In a subset of experiments immunohistochemistry staining for PER2 in the SCN or hippocampus was performed. (A) Brain PER2 transcript levels (n=4 animals) 2 hours following midazolam administration. (B) Brain PER2 transcript levels (n=3 animals) 2 hours following LPS administration. (C, D) Immunohistochemistry of PER2 in the SCN or hippocampus 24 hours following midazolam administration (original magnification, X200, scale bars indicate 100 μm, n=8 animals). All data are presented as mean ± SD; SCN=suprachiasmatic nucleus, LPS=Lipopolysaccharide.

Figure 2. Midazolam induces cognitive deficits and reduces locomotion in wildtype mice

Wildtype mice (C57BL6/J) were injected with either midazolam (10 mg/kg i.p.) or LPS (100 μg/kg i.p.) and compared to saline treated controls. 24 or 27 hours later mice underwent behavioral studies using a T-maze alternation test or open-field line-crossing determination. (A) T-maze alternation in %, 24 or 72 hours after a single dose of midazolam. (B) T-maze alternation in %, 24 or 72 hours after a single dose of LPS. (C) Numbers of squares crossed for 10 minutes (line crossing) 24 or 72 hours after a single dose of midazolam. (D) Numbers of squares crossed for 10 minutes (line-crossing) 24 or 72 hours after a single dose of LPS. All data have n=5 individual animals and are presented as mean ± SD, LPS=Lipopolysaccharide, P value denotes t-test and * denotes one-way ANOVA (see also Supplemental Digital Content 1).

Figure 3. Midazolam induces loss of memory in wildtype mice

Wildtype (C57BL6/J) were injected with either midazolam (10 mg/kg i.p.), LPS (100 μg/kg i.p.) or midazolam and LPS and compared to saline treated controls. 24 or 27 hours later mice underwent behavioral studies using novel-object-recognition tests. (A) Preference in % for a novel object after 2 days habituation to two old objects 24 hours after a single dose of saline in wildtype mice. (B) Preference in % for a novel object after 2 days habituation to two old objects 24 or 72 hours after a single dose of midazolam in wildtype mice. (C) Preference in % for a novel object after 2 days habituation to two old objects 24 or 72 hours after a single dose of LPS in wildtype mice. (D) Preference in % for a novel object after 2 days habituation to two old objects 24 or 72 hours after a single dose of LPS + midazolam in wildtype mice. All data have n=5 individual animals and are presented as mean ± SD, LPS=Lipopolysaccharide, P value denotes t-test (see also Supplemental Digital Content 1).

Figure 4. Midazolam induced delirium is PER2 dependent

Wildtype (C57BL6/J) or Per2^−/− mice were injected with either midazolam (10 mg/kg i.p.), LPS (100 μg/kg i.p.) or midazolam and LPS and compared to saline treated controls. 24 hours later mice underwent behavioral studies using T-maze alternation test, open-field line-crossing determination or novel-object-recognition tests. (A) T-maze alternation in % 24 hours after a single dose of midazolam or LPS + midazolam. (B) Numbers of squares crossed for 10 minutes (line-crossing) 24 hours after a single dose of midazolam or LPS + midazolam. (C, D) Preference in % for the novel object after 2 days habituation to the old object 24 hours after a single dose of midazolam. All data have n=5 individual for wildtype mice and an n=8 individual animals for Per2^−/− mice and are presented as mean ± SD, LPS=Lipopolysaccharide P value denotes t-test and * denotes one-way ANOVA (see also Supplemental Digital Content 1).

Figure 5. Nobiletin abolished midazolam or constant light induced cognitive deficits

Wildtype (C57BL6/J) mice were injected with either midazolam (10 mg/kg i.p.) or midazolam + nobiletin (1 mg/kg i.p.) and compared to saline treated controls. 24 hours later mice underwent behavioral studies. In a subset of experiments mice were exposed to 7 days of constant light and tested for behavioral changes 24h later. (A) T-maze alternation in % 24 hours after a single dose of midazolam or midazolam + nobiletin. (B) Numbers of squares crossed for 10 minutes (line crossing) 24 hours after a single dose of midazolam or midazolam + nobiletin. (C) Preference in % for a novel object after 2 days habituation to two old objects 24 hours after a single dose of midazolam or midazolam + nobiletin. (D) T-maze alternation in % 24 hours after 7 days at constant light conditions with and without nobiletin. All data on behavioral tests have n=5 individual wildtype mice and are presented as mean ± SD, NOB=nobiletin, P value denotes t-test and * denotes one-way ANOVA (see also Supplemental Digital Content 1).