Evaluating Fatty Acid Amide Hydrolase as a Suitable Target for Sleep Promotion in a Transgenic TauP301S Mouse Model of Neurodegeneration

Abstract

Sleep disruption is an expected component of aging and neurodegenerative conditions, including Alzheimer's disease (AD). Sleep disruption has been demonstrated as a driver of AD pathology and cognitive decline. Therefore, treatments designed to maintain sleep may be effective in slowing or halting AD progression. However, commonly used sleep aid medications are associated with an increased risk of AD, highlighting the need for sleep aids with novel mechanisms of action. The endocannabinoid system holds promise as a potentially effective and novel sleep-enhancing target. By using pharmacology and genetic knockout strategies, we evaluated fatty acid amide hydrolase (FAAH) as a therapeutic target to improve sleep and halt disease progression in a transgenic Tau P301S (PS19) model of Tauopathy and AD. We have recently shown that PS19 mice exhibit sleep disruption in the form of dark phase hyperarousal as an early symptom that precedes robust Tau pathology and cognitive decline. Acute FAAH inhibition with PF3845 resulted in immediate improvements in sleep behaviors in male and female PS19 mice, supporting FAAH as a potentially suitable sleep-promoting target. Moreover, sustained drug dosing for 5-10 days resulted in maintained improvements in sleep. To evaluate the effect of chronic FAAH inhibition as a possible therapeutic strategy, we generated FAAH-/- PS19 mice models. Counter to our expectations, FAAH knockout did not protect PS19 mice from progressive sleep loss, neuroinflammation, or cognitive decline. Our results provide support for FAAH as a novel target for sleep-promoting therapies but further indicate that the complete loss of FAAH activity may be detrimental.

Keywords: Alzheimer’s disease; FAAH; anandamide; endocannabinoids; sleep; tau.

Figure 1

PS19 mice have decreased CB1 and FAAH expression. (A) Western blot analysis of cortical homogenate samples showing CB1 and FAAH in WT and PS19 mice at 3, 6, and 9 months. (B) Quantification of cortical CB1 and FAAH proteins in WT and PS19 mice at 3, 6, and 9 months. N = 8 per genotype. * p < 0.05. Unpaired two-tailed Student’s t-test. Error bars indicate ± SEM.

Figure 2

Selective increased AEA promotes dark phase sleep behavior in PS19 mice. (A) Experimental design; (B) 24 h traces of hourly sleep of female and male WT (blue line) PS19 (pink line) mice at 3 months (pre-pathology), 6 months (early phase), and 9 months (symptomatic phase). Grey bars in sleep traces indicate dark phase. (C,D) Quantification of average hourly dark phase sleep in females (C) and males (D). (E,F) Quantification of average dark phase sleep bout length in seconds in females (E) and males (F). N = 5–16/age/sex/genotype. * p < 0.05, ** p < 0.01, *** p < 0.001, ns: not significant. Paired two-tailed Student’s t-test. Error bars indicate ± SEM.

Figure 3

Sustained increase in AEA promotes dark phase sleep behavior in PS19 mice. (A) Experimental design. (B,C) Quantification of average hourly sleep (B) and sleep bout length (C) in female WT and PS19 mice. (D,E) Quantification of average hourly sleep (D) and sleep bout length (E) in male WT and PS19 mice. Data separated into 12 h of dark and light phases. N = 5–12/sex/genotype. * p < 0.05, ** p < 0.01, *** p < 0.001. Repeated measure two-way ANOVA with Šídák’s multiple comparisons test. Error bars indicate ± SEM.

Figure 4

Validation of PS19/FAAH KO. Western blot analysis of cortical homogenate samples showing FAAH, CB1, and AT8 in FAAH+/−, FAAH−/−, FAAH+/−/PS19, and FAAH−/−/PS19 mice at 9 months.

Figure 5

Loss of FAAH does not protect against sleep loss in PS19 mice; (A) 24 h trace of average hourly sleep of female FAAH+/− (blue line), FAAH−/− (pink line), FAAH+/−/PS19 (blue dashed line), and FAAH−/−/PS19 (pink dashed line) mice at 3, 6, and 9 months. Grey bars in sleep traces indicate dark phase. (B,C) Quantification of average hourly sleep (B) and sleep bout length in seconds (C). Data separated into 12hrs of dark and light phases; (D) 24 h trace of average hourly sleep of male FAAH+/− (blue line), FAAH−/− (pink line), FAAH+/−/PS19 (blue dashed line), and FAAH−/−/PS19 (pink dashed line) mice at 3, 6, and 9 months. (E,F) Quantification of average hourly sleep (E) and sleep bout length in seconds (F). Data separated into 12 h of dark and light phases. N = 5–17/age/sex/genotype. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Two-way ANOVA with Šídák’s multiple comparisons test. Error bars indicate ± SEM.

Figure 6

Pro-inflammatory cytokines are increased with loss of FAAH. (A) Quantification of TNFα mRNA levels in male and female FAAH+/−, FAAH−/−, FAAH+/−/PS19, and FAAH−/−/PS19 mice at 9 months. (B) Quantification of IL-1β mRNA levels in male and female FAAH+/−, FAAH−/−, FAAH+/−/PS19, and FAAH−/−/PS19 mice at 9 months. N = 3–6/genotype. ** p-value < 0.01; other p values are indicated. Unpaired Student’s t-test was performed. Bonferroni corrected. Error bars indicate mean ± SEM.

Figure 7

Loss of FAAH does not protect against cognitive decline in PS19 mice. (A) Experimental design. (B) Swim speed in WT, FAAH−/−, PS19, and FAAH−/−/PS19. (C) Acquisition of spatial learning. Escape latencies during acquisition in WT, FAAH−/−, PS19, and FAAH−/−/PS19 mice. (D,E) Spatial memory retention during 1 min probe trial, % time in target or opposite quadrant (D) or number of crosses of the target location (E) in WT, FAAH−/−, PS19, and FAAH−/−/PS19 mice. Target indicates the quadrant where the platform had been located versus the opposite quadrant. N = 6–29 per group (genotype, treatment). Data are shown as means (± SEM) of 4 trials per day in a Morris water maze. * p < 0.05, ** p < 0.01, comparison to WT. # p < 0.05, ## p < 0.01, ### p < 0.001, comparison to KO. Δ p < 0.05, comparison to PS19.

Figure 8

Loss of FAAH leads to differences in contextual learning, but not cue-dependent learning. (A,B) Percent freezing in the first 5 min context test (Test 1) and test 2 completed 2 weeks later in WT, FAAH−/−, PS19, and FAAH−/−/PS19 mice. (C,D) Percent freezing after an 80-decibel acoustic stimulus (3 min) was presented 2 min after mice were placed in the modified conditioned fear chambers (C). Cue test 2 was conducted 2 weeks following test 1 (D). N = 6–29 per group (genotype, treatment). ** p < 0.01, comparison to WT. ## p < 0.01, #### p < 0.0001, comparison to KO.