Main text

Sleep behavior is exhibited by virtually every species studied, while the precise function of sleep remains unknown. Understanding the underlying cellular and molecular mechanisms that relay sleep behavior may resolve important clues as to sleep function. Sleep is thought to be governed by two processes, a circadian system, which controls the daily timing of sleep, and a homeostatic system, which regulates sleepiness based on previous time spent awake [1, 2]. Circadian rhythms are regulated by a well-defined and phylogenetically conserved transcriptional/translational autoregulatory negative feedback loop [3], which includes the core clock gene Bmal1 [4]. Bmal1 is a basic helix-loop-helix transcription factor known to heterodimerize with core circadian factors CLOCK or NPAS2 and bind to E-box elements in the promoter of downstream target genes to influence circadian their transcriptional output and behavior [5, 6]. Deletion of Bmal1 is the only single gene deletion that disrupts circadian clock function, whereas single gene deletion of other clock components only leads to attenuated circadian rhythms [7]. However, exactly how this circadian clock shapes molecular events that in turn regulate sleep behavior are not well understood.

Fatty-acid binding proteins (Fabp) comprise a family of small (~ 15 kDa) hydrophobic ligand binding carriers with high affinity for long-chain fatty-acids for intracellular transport, and are associated with metabolic, inflammatory, and energy homeostasis pathways [8, 9]. These include three that are expressed in the adult mammalian central nervous system (CNS), and are Fabp3 (H-Fabp), Fabp5 (E-Fabp), and Fabp7 (B-Fabp). Fabp3 is predominantly expressed in neurons, Fabp5 is expressed in multiple cell types, including both neurons and glia, and Fabp7 is enriched in astrocytes and neural progenitors. Fabp7 mRNA was identified as a unique transcript elevated in multiple hypothalamic brain regions of mice during the sleep phase [10], and Fabp7 is also known to regulate sleep in flies, mice, and humans [11]. Here we were interested in determining whether circadian Fabp7 mRNA expression is disrupted in mice that lack the core circadian clock gene Bmal1.

To test whether circadian Fabp7 mRNA expression is regulated by the core molecular clock, we compared Fabp7 transcript levels in hypothalamus of Bmal1 KO to wild-type (WT) littermate control C57BL/6 J mice using qPCR analysis. We observed that baseline levels of Fabp7 mRNA expression in the Bmal1 KO mouse were higher compared to WT littermate controls, irrespective of circadian time (Fig. 1a, Fig. S1), while the circadian expression of Fabp7 mRNA was completely ablated in Bmal1 KO mice compared to controls (Fig. 1b). In order to test whether other CNS expressing Fabps have altered mRNA expression as a result of Bmal1 deficiency, we profiled Fabp3 and Fabp5 and found that neither the baseline expression or circadian pattern of these transcripts were affected by Bmal1 (Fig. 1a, b), suggesting that among Fabps expressed in the CNS, both the baseline and circadian profile of transcription affected by the core clock is Fabp7 specific.

Fig. 1
figure 1

Increased baseline Fabp7 mRNA expression, and disruption of its circadian rhythm, in Bmal1 KO. (a) Average expression of hypothalamic Fabp7 mRNA is ~7fold greater in BMAL KO compared to WT mice, while Fabp3 mRNA and Fabp5 mRNA are stable (n = 18 per group). ***p < 0.001 Fabp7 mRNA KO/WT vs. Fabp3 mRNA or Fabp5 mRNA KO/WT (t-test). (b) Normalized mRNA expression values (to mean of circadian values within each group) of various Fabps depicts the circadian oscillation of Fabp7 mRNA in C57BL/6 WT which is absent in BMAL KO mice (n = 3 per group per timepoint). Fabp7 mRNA shows a significant circadian dependent change in expression based on genotype. Two-way ANOVA (p < 0.001). ***p < 0.001, **p < 0.01, *p < 0.05 vs ZT0 (post-hoc Bonferroni), while there is no significant difference for circadian variation of Fabp7 mRNA in BMAL KO mice

Full size image

Circadian Fabp7 mRNA expression has been shown to be regulated by the nuclear hormone receptor Rev-erbα (NR1D1), a transcriptional repressor, via Rev-erbα response elements (ROREs) in the Fabp7 gene promoter [12]. Since Bmal1 deficiency greatly suppresses Rev-erbα expression, it is likely that the circadian regulation of Fabp7 mRNA by Bmal1 is through Rev-erbα. Studies determining whether Bmal1 regulates Fabp7 mRNA indirectly via Rev-erbα or whether the Fabp7 gene promoter also has functional E-box binding sites will be crucial for identifying the clock-regulated expression of this gene. Fabp7 mRNA expression was also shown to be upregulated in mice with deletion of Bmal1 in neurons and astrocytes [13]. Future studies using cell-specific knock-down of Bmal1 to determine whether disruption of core clock genes in astrocytes regulates circadian Fabp7 mRNA expression will be important for our understanding of how clock-controlled genes in non-neural cells regulate pathways affecting behavior. For example, recent studies show that astrocytes are able to regulate daily rhythms in the master circadian pacemaker of the hypothalamus, the suprachiasmatic nucleus (SCN), and circadian behavior [14, 15]. Further, post-transcriptional processing mechanisms known to operate on Fabp7 mRNA targeting to perisynaptic processes [16] will be important to characterize following circadian clock disruption in astrocytes. Determining whether circadian Fabp7 expression is necessary for daily rhythms in SCN and extra-SCN physiology [17] and related behavior are additional avenues for future study.