Differential modulation of clock gene expression in the suprachiasmatic nucleus, liver and heart of aged mice

Highlights

• Advanced age affects the clock gene network at central and peripheral levels.

• In SCN the decline of the rhythmic function of CCGS impairs circadian output.

• At the peripheral level a deterioration of the rhythm of core clock genes prevails.

Abstract

Studies on the molecular clockwork during aging have been hitherto addressed to core clock genes. These previous investigations indicate that circadian profiles of core clock gene expression at an advanced age are relatively preserved in the master circadian pacemaker and the hypothalamic suprachiasmatic nucleus (SCN), and relatively impaired in peripheral tissues. It remains to be clarified whether the effects of aging are confined to the primary loop of core clock genes, or also involve secondary clock loop components, including Rev-erbα and the clock-controlled genes Dbp and Dec1. Using quantitative real-time RT-PCR, we here report a comparative analysis of the circadian expression of canonical core clock genes (Per1, Per2, Cry1, Cry2, Clock and Bmal1) and non-core clock genes (Rev-erbα, Dbp and Dec1) in the SCN, liver, and heart of 3 month-old vs 22 month-old mice. The results indicate that circadian clock gene expression is significantly modified in the SCN and peripheral oscillators of aged mice. These changes are not only highly tissue-specific, but also involve different clock gene loops. In particular, we here report changes of secondary clock loop components in the SCN, changes of the primary clock loop in the liver, and minor changes of clock gene expression in the heart of aged mice. The present findings outline a track to further understanding of the role of primary and secondary clock loop components and their crosstalk in the impairment of circadian output which characterizes aging.

Introduction

The mammalian circadian system consists of multiple oscillators distributed throughout the organism and a central pacemaker in the suprachiasmatic nucleus (SCN) of the hypothalamus which coordinates the entire system making it coherent. The molecular mechanisms underlying circadian function are similar in central and peripheral oscillators and consist of highly conserved genes, the so-called clock genes. Together with their protein products, clock genes are organized in transcription–translation feedback loops able to complete a cycle in ~ 24 h (Ko and Takahashi, 2006).

The primary feedback loop, or core loop, consists of different components. The positive transcriptional activators CLOCK and BMAL1 (or its paralog NPAS2) heterodimerize and promote transcription of the Period genes (Per1 and Per2) and Cryptochrome genes (Cry1 and Cry2). The negative components include PER and CRY which heterodimerize, translocate back to the nucleus and repress their own transcription by acting on the CLOCK–BMAL1 complex (Sato et al., 2006). The CLOCK–BMAL1 complex is involved in an additional loop, interlocked with the core loop, that induces the expression of the nuclear hormone receptors Rev-erbα and Rorα. Their products, REV-ERBα and RORα, bind competitively to Bmal1 and determine the cycling of Bmal1 expression by suppressing and activating, respectively, its transcription (Guillaumond et al., 2005, Preitner et al., 2002).

A key function of the molecular clock machinery is represented by the circadian output, whose efficacy depends on the coordinated circadian behavior and physiology of the entire organism (see for reviews Takahashi et al., 2008, Albrecht, 2012). The transcriptional regulation of this function is achieved not only directly via core loop transcription factors, but also indirectly via auxiliary loops interlocking with the core loop, which include the transcription factors Dbp and Dec1. These interlocking loops and their constituents, the so-called first order “clock-controlled genes” (CCGS) (Reppert and Weaver, 2002) play a crucial intermediate role in conveying circadian signals to downstream local clock-controlled genes, specific for each tissue (Liu et al., 2008).

The cyclic transcription of Dbp is under the control of CLOCK–BMAL1, and is repressed by PER/CRY heterodimers. In turn, DBP can regulate the core clock machinery activating Per1 transcription. Dec1 rhythmic transcription is positively regulated by CLOCK–BMAL1, while DEC1, interacting with BMAL1 for E-box occupation, negatively acts on CLOCK/BMAL1 activity (Honma et al., 2002, Kawamoto et al., 2004; see also for review Bonaconsa et al., 2013).

Evidence in experimental animals and in humans indicates that the circadian timing system is progressively modified with advancing age, as shown by the reduced amplitude, phase and period length of circadian rhythms, increased tendency towards internal desynchronization and deficient responsiveness to phase-shifting stimuli with loss of synchronization to the environment (Gibson et al., 2009, Hofman and Swaab, 2006). Circadian rhythms of locomotor activity, core body temperature, blood pressure, and the sleep–wake cycle are especially affected (Reppert and Weaver, 2002). The possibility that these aging-related changes are due to impairment of the molecular clockwork in the SCN and/or in peripheral tissues has been repeatedly investigated.

In the SCN, Per1 and Cry1 circadian rhythms have been reported to be preserved at an advanced age in mice (Weinert et al., 2001), rats (Asai et al., 2001), and hamsters (Kolker et al., 2003). A decrease of Per2 circadian oscillation has been reported in aged mice (Weinert et al., 2001), but not in rats (Asai et al., 2001) and hamsters (Kolker et al., 2003). An aging-related decline in Clock and Bmal1 circadian rhythm has been reported in hamsters (Kolker et al., 2003).

Studies exploring the impact of aging on clock gene expression in peripheral tissues have indicated that Per1 expression is preserved in aged rats (Asai et al., 2001, Yamazaki et al., 2002). Moderate aging-related changes of Per2 and Bmal1 circadian rhythms have been reported in the macaque pituitary gland (Sitzmann et al., 2010). Analyses conducted in rats at only two time points for 24 h have indicated a significant age-related decrease of Per1–3 mRNAs and unchanged Cry1–2, Clock and Bmal1 mRNAs in the liver, associated with Per1–3 decrease and Bmal1 increase in the heart (Claustrat et al., 2005).

It remains to be clarified whether a definite dysfunction of the molecular clockwork occurs in the circadian impairment that characterizes aging. The matter is still open not only because the available data are limited and in some instances controversial, but also because they mainly focus on clock genes of the core loop, while information on the clock output function mediated by CCGS is limited.

On this basis, using quantitative real-time RT-PCR, we here assessed in young and aged mice the circadian expression of canonical core clock genes (Per1–2, Cry1–2, Clock and Bmal1), of the transcription factor Rev-erbα which, as mentioned above, forms with Bmal1 a central regulatory loop, and of the first-order CCGS Dec1 and Dbp. In order to compare the clock properties of central and peripheral oscillators, we here investigated the SCN, liver and heart.