Chronobiology International, 30(4): 559–576, (2013) Copyright © Informa Healthcare USA, Inc. ISSN 0742-0528 print/1525-6073 online DOI: 10.3109/07420528.2012.754451
Peripheral Circadian Clocks—A Conserved Phenotype? Yuval Weigl, Valerie L. Harbour, Barry Robinson, Line Dufresne, and Shimon Amir 1
Chronobiol Int Downloaded from informahealthcare.com by Hebrew University, Medical Library on 11/12/14 For personal use only.
Center for Studies in Behavioral Neurobiology/Centre de Recherche en Neurobiologie Comportementale, Concordia University, Montreal, Quebec, Canada
The circadian system of mammals regulates the timing of occurrence of behavioral and physiological events, thereby optimizing adaptation to their surroundings. This system is composed of a single master pacemaker located in the suprachiasmatic nucleus (SCN) and a population of peripheral clocks. The SCN integrates time information from exogenous sources and, in turn, synchronizes the downstream peripheral clocks. It is assumed that under normal conditions, the circadian phenotype of different peripheral clocks would be conserved with respect to its period and robustness. To study this idea, we measured the daily wheel-running activity (WRA; a marker of the SCN output) in 84 male inbred LEW/Crl rats housed under a 12 h:12 h light-dark cycle. In addition, we assessed the mRNA expression of two clock genes, rPer2 and rBmal1, and one clock-controlled gene, rDbp, in four tissues that have the access to time cues other than those emanating from the SCN: olfactory bulbs (OBs), liver, tail skin, and white blood cells (WBCs). In contrast with the assumption stated above, we found that circadian clocks in peripheral tissues differ in the temporal pattern of the expression of circadian clock genes, in the robustness of the rhythms, and possibly in the number of functional ∼24-h-clock cells. Based on the tissue diversity in the robustness of the clock output, the hepatic clock is likely to house the highest number of functional ∼24-h-clock cells, and the OBs, the fewest number. Thus, the phenotype of the circadian clock in the periphery is tissue specific and may depend not only on the SCN but also on the sensitivity of the tissue to non-SCN-derived time cues. In the OBs and liver, the circadian clock phenotypes seem to be dominantly shaped by the SCN output. However, in the tail skin and WBC, other time cues participate in the phenotype design. Finally, our study suggests that the basic phenotype of the circadian clock is constructed at the transcript level of the core clock genes. Yet, additional posttranscriptional and translational events can contribute to the robustness and periodicity of the clock output. (Author correspondence:
[email protected]) Keywords: Circadian timekeeping system, Peripheral tissues, rBmal1, rDbp, rPer2, Suprachiasmatic nucleus; Wheelrunning activity
INTRODUCTION
pathological consequences such as malignancy, physical and cognitive declines, as well as sleep disorders (Haus & Touitow, 1992; Lie et al., 2006). The hierarchical architecture of the mammalian circadian system is composed of a central circadian pacemaker located in the suprachiasmatic nucleus (SCN) of the brain and a network of clocks widely distributed throughout the brain and periphery (Kowalska & Brown, 2007; Dibner et al., 2010; Lowrey & Takahashi, 2011; Mohawk et al., 2012). The circadian timing devices in the SCN and peripheral tissues are identical in their molecular mechanism and involve virtually the same core components in most tissues. The molecular mechanism is composed of interlocked and delayed autoregulatory transcriptional and translational/posttranslational feedback loops
In mammals, the endogenous circadian timekeeping system drives the daily rhythms of most physiological and behavioral variables. The rhythms follow each other in an orderly manner that allows for optimal adaptation to predictable daily changes in the environment (Arendt et al., 1989; Haus & Touitow, 1992; Minors & Waterhouse, 1986; Rensing et al., 2001). Currently, the mechanisms underlying the orderly occurrence of these rhythms are not fully understood. It is known, however, that their temporal order can be modulated, at least in part, by changes in the illumination cycle and other external time cues such as social interactions and feeding (Dibner et al., 2010; Gachon et al., 2004; Mistlberger & Skene, 2004; Schibler & Naef, 2005; Stokkan et al., 2001). Impairment of the temporal order can have
Submitted June 29, 2012, Returned for revision August 4, 2012, Accepted November 16, 2012 Address correspondence to Dr. Shimon Amir, Center for Studies in Behavioral Neurobiology/Centre de Recherche en Neurobiologie Comportementale, Concordia University, 7141 Sherbrooke Street West, Montreal, Quebec, Canada, H4B 1R6. Tel.: (514) 848-2424, extension 2188; E-mail:
[email protected]
Chronobiol Int Downloaded from informahealthcare.com by Hebrew University, Medical Library on 11/12/14 For personal use only.
Y. Weigl et al.
consisting of positive and negative elements. BMAL1 and CLOCK/NPAS2 form the elements of the positive arm of the feedback loop, whereas PER and CRY proteins constitute the major negative elements. The expression products of Rev-Erbα and Rorα bridge between these opposite basic loops. Additional posttranscriptional/ translational events, such as SUMOylation, methylation, microRNA base pairing, contribute to the precision and robustness of the clockwork machinery (Dardente & Cermakian, 2007; Dibner et al., 2010; Ko & Takahashi, 2006; Welsh et al., 2010; Yagita et al., 2001; Zhou et al., 2011). The peripheral clocks resemble the SCN in the capacity to produce self-sustained circadian rhythms in a cell-autonomous manner (Balsalobre et al., 1998; Guilding & Piggins, 2007; Merrow & Roenneberg, 2001; Yamamoto et al., 2004; Yamazaki et al., 2000, 2009; Yoo et al., 2004). The SCN pacemaker is entrained by the environmental light cycle and, in turn, conveys time information directly to downstream peripheral clocks via neural and humoral signals emanating from the pacemaker itself or indirectly by its entrainment of the behavioral and physiological rhythms (Buijs & Kalsbeek, 2001; Delaunay & Laudet, 2002; Gachon et al., 2004; Ripperger & Schibler, 2006). As a result, the SCN orchestrates the circadian timekeeping system leading to optimal adaptation of the animal to its current surroundings. The current view of the SCN as the master pacemaker posits that peripheral clocks normally tick in accordance with a conserved phenotype defined by the period and robustness of the master clock. Yet, several studies revealed discrepancies in the circadian period and phase of clocks in different tissues (Dibner et al., 2010; Liu et al., 2007; Yoo et al., 2004). In the present study, we characterized the circadian clock phenotypes in four different tissues, namely, the olfactory bulbs (OBs), liver, tail skin, and white blood cells (WBCs), and compared them with each other and with a key output marker of the SCN master clock, wheel-running activity (WRA) (Davis & Viswanathan, 1996; Earnest et al., 1999; Reebs & Maillet, 2003). These tissues were chosen because they serve distinct biological functions and because they normally respond to different non-SCN time cues, namely odors (OBs; Amir et al., 1999), feeding-fast cycle (liver; Vollmers et al., 2009), cell cycle (tail skin; Bjarnason, 2001), and immunological cues (WBCs; Berger, 2008). Previous studies in humans and rodents demonstrated circadian rhythms in expression of several core clock genes (especially Per and Bmal1) in each of these tissues. Other experiments concluded that functional circadian clocks tick in the liver, OBs, and in most cell types in the skin (Geyfman & Andersen, 2009, 2010; Granados-Fuentes et al., 2004; Guilding & Piggins, 2007; Hamada et al., 2011; Hughes et al., 2009; Kornmann et al., 2007; Liu et al., 2007; Sandu et al., 2012; Takata et al., 2002; Tanioka et al., 2009). Contradictory evidence appears to exist for the WBCs (Boivin et al., 2003; Liu et al., 2007; Teboul et al., 2005).
For each of the tissues studied, we monitored the 24-h mRNA profiles of two core clock genes, rPer2 and rBmal1, and the clock output gene, rDbp. In all, 84 rats (7 cohorts of 12 rats each) were housed under a 12 h:12 h light-dark cycle for 3 wks. Rats from each cohort were killed at different times of day to result in 1–4 rats killed every 30 min over the 24-h day. These mRNA profiles were compared with the WRA profiles of the last 3 d. Systematic and comprehensive chronobiological analysis was applied on all of the variables. Such analysis provides a genuine and full characterization of the bestfitted rhythm function and its overall output. To the best of our knowledge, this is the first study to offer such a comprehensive analysis of the clock genes expression patterns. MATERIALS AND METHODS Animals Eighty-four 2-mo-old male inbred LEW/Crl rats (150– 200 g; Charles River Laboratories, St. Constant, QC, Canada) were housed in individual cages equipped with a metal running wheel (34 cm diameter) for a period of 3 wks to ensure adaptation to the experimental environment. The experimental procedures followed the guidelines of the Canadian Council on Animal Care and the ethical standards of this journal (Portaluppi et al., 2010), and were approved by the Animal Care Committee, Concordia University. Illumination, Temperature, and Food Regimens The rats were maintained at 22°C ± 2°C and entrained to a 12 h:12 h light-dark cycle. Light intensity at the cage level was 100 lux. The animals had a free access to Purina rat chow (catalog no. 5075; Charles River) and water. The cages were placed in ventilated, sound- and light-tight isolation chambers equipped with a computer-controlled lighting system (VitalView; Mini Mitter, Sunriver, OR, USA). Wheel-Running Activity (WRA) Each running wheel was equipped with a magnetic microswitch connected to a computer. Wheel-running activity data were recorded in 10-min bins with VitalView software (Mini Mitter). Activity data were analyzed with Actiview (Mini Mitter) and TableCurve 2D (Jandel Scientific, San Diego, CA, USA). Tissue Sampling Individual rats from each of the 7 cohorts were removed from their cages at 1 of 48 predetermined times of day and 1-cm portion of the tail was clipped and sampled immediately. In 6 out of the 7 cohorts (nos. 1–6), whole blood (250 µL) was collected, within 90 s, from the tail vein of each rat and placed in separate premade tubes of RNAlater reagent (MouseRiboPure™ Blood RNA Isolation kit, catalog no. AM1951; Ambion, Austin, TX, USA). Rats from 1 cohort (no. 7) were not sampled due Chronobiology International
Peripheral Circadian Clocks to technical problems. Individual rats from all 7 cohorts were deeply anesthetized (Somnotol; 100 mg/kg) and perfused with saline followed by paraformaldehyde. During the saline perfusion, the right lobe of the liver was dissected from each rat. Finally, the olfactory bulbs were detached from the brain of each individual. (The fixed brains were stored and used for immunohistochemical analyses of clock proteins in the brain, the results to be reported elsewhere.) Each of the skin, liver, and olfactory bulbs samples was rinsed with saline and immersed in RNAlater stabilization reagent (Ambion).
Chronobiol Int Downloaded from informahealthcare.com by Hebrew University, Medical Library on 11/12/14 For personal use only.
Total RNA Extraction OB Total RNA was isolated with the RecoverAll™ kit (catalog no. AM1975; Ambion) in accordance with the manufacturer’s instructions. In order to purify the eluted RNA from DNA, a second DNase digestion step was included with the TURBO DNA-free™ kit (catalog no. AM1907; Ambion). Liver Total RNA was isolated using the commercially available PARIS™ kit (catalog no. AM1921; Ambion) following the manufacturer’s instructions. Tail Skin Total RNA was isolated using the RNAqueous-4PCR™ kit (catalog no. AM1914; Ambion) in accordance with the manufacturer’s instructions. In order to purify the RNA from traces of DNA, the samples were treated with TURBO DNA-free™ kit (catalog no. AM1907; Ambion) following the manufacturer’s protocol. Blood Total RNA was isolated with the commercially available MouseRiboPure™ (catalog no. AM1951; Ambion) according to the manufacturer’s instructions. The recovered RNA was treated with TURBO DNA-free™ kit (catalog no. AM1907; Ambion) in order to remove DNA traces. Quality of RNA Samples The RNA integrity profile (RIN) of each of the samples and their concentrations were determined by the Agilent RNA6000 Nanokit running on a Agilent 2100 Bioanalyzer electrophoresis unit (Agilent Technologies, Santa Clara, CA, USA). The DNA and protein contamination levels in each of the samples were assessed by
measuring the 260/280 and 260/230 nm absorbance ratios on the Nanodrop2100c spectrophotometer (ThermoFisher Scientific, Wilmington, DE, USA). cDNA Library Construction Equal amounts of total RNA from each tissue of each rat were reverse transcribed to single-stranded cDNA using the commercially available High Capacity cDNA Reverse Transcription kit with RNase inhibitor (Applied Biosystems, Foster City, CA, USA) and according to the manufacturer’s protocol. Briefly, a 2× reverse transcription (RT) master mix containing RT buffer, deoxyribonucleotide triphosphate mixture, random primers, MultiScribe™ RT enzyme, and RNase inhibitor was added to the total RNA sample. Forty-microliter-sample reactions were incubated at 25°C for 10 min, followed by 37°C for 2 h and 85°C for 5 min in a CFX96 C1000 Thermal Cycler chassis (Bio-Rad, Hercules, CA, USA). Negative controls of the RT reaction were preformed in parallel. Quantitative Real-Time Polymerase Chain Reaction (PCR) For each of the peripheral tissues, the mRNA levels of the three target genes: rPer2, rBmal1, and rDbp, and five potential housekeeping genes (TopI, Ywhaz, Gapdh, B2M, and Hmbs) were determined by quantitative realtime PCR based on the Perfect Probe method (Primer Design, Southampton, UK) on a CFX96 Real Time PCR Detection System (Bio-Rad). Primer/probe sequences and amplicon lengths of the genes of interest are detailed in Table 1. Amplifications were performed in 20 µL volume reactions containing 5 µL cDNA, Fast Universal TaqMan master mix (Applied Biosystems), and Perfect Probe primer/probe mix (PrimerDesign) according to the manufacturer’s instructions. All cDNA samples were run in triplicates. The 2−ΔΔCt method (Livak & Schmittgen, 2001) was applied in order to retrieve the relative mRNA levels of rPer2, rBmal1, and rDbp in each the rats’ tissue samples. First, the target gene expression level in each rat was normalized according to a combination of the ≥3 most stable housekeeping genes within the five-gene panel, as determined by the geNorm software (http:// medgen.ugent.be/~jvdesomp/genorm). Then, the relative value was renormalized with respect to the highest expression value expressed along the 24-h time span. Rhythm Function Characterization Two-dimensional Table Curve (Jandel Scientific) was used to assess the statistically significant presence of one or multiple periodic component (compound)
TABLE 1. Primer and probe sequences used in real-time quantitative RT-PCR Gene transcript rPer2 rBmal1 rDbp
Sense primer (5’ → 3’)
Antisense primer (5’ → 3’)
Probe (5’FAM → 3’)
TTCCACCAGCAACCCCAAA ACCAGGGTTTGAAGTTAGAGTC ACCCACTCGCCCAGACTATA
CAGGAGTTATTTCAGAGGCAAGT AAGTCACTGATTGTGGAGGAAAT AGCAAGCCTCCAGTATCAGAA
CTTCCCCAGCCAGCCTCACTTTCCGggaag CCATTCTCTGGTCCGCCATTGGAAGGgaatgg CTTCAAATCCTACGAGCACTGCGGGGGttgaag
Details the sequences of the probe/primer sets used to quantify the expression levels of the three genes of interest. © Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from informahealthcare.com by Hebrew University, Medical Library on 11/12/14 For personal use only.
Y. Weigl et al.
(Weigl et al., 2004, 2010). The software derives the “basic” characteristics for each of the rhythm components: period, acrophase (waveform-approximated peak time), amplitude (half of the waveform-approximated extent of variation), and the rhythm mesor (midline estimating statistic of rhythm) with their standard errors. In the case of compound rhythms, the definition of the “major” and “minor” entities of the compound rhythm was based on the component’s contribution to the overall rhythm pattern, e.g., amplitude and the rhythm’s statistical fitness, i.e., degrees of freedom–adjusted explained variance, DF-Adj.r 2 (Lewy et al., 2005; Weigl et al., 2012). We refer in this study to periods of four domains: 24 ± 4 h (circadian); 18 ± 2 (pseudodian); 13 ± 3 h (semidian); 8 ± 1 h, 6 ± 1 h, and 4 ± 1 h (short ultradian). Additional parameter sets accompanied the above “basic” ones to obtain a more informative insight of the rhythm characteristics (A2/A1, and overall pattern phenotypes, as explained below). They were obtained either directly from the Table Curve software or by calculating ratios of the basic parameters. A2/A1 (Weigl et al., 2012) equals to the ratio between the amplitudes of the secondary and major periodic rhythm components. It estimates the robustness of the major entity of a given compound rhythm.
Overall Pattern of Phenotype Parameters This set of parameters is essential in the case of compound rhythms, since it defines the interaction outcome between the individual rhythm components. Thus, the values of these parameters depend on the amplitudes of the rhythm components and synchronization level. Major actual peak/trough (Weigl et al., 2012) refers to the time points when a variable reaches its maximal or minimal computed expression along the rhythm, respectively. More than 1-h shift separates between the major and secondary actual peaks/troughs. Relative rhythm fluctuation range (FR) (Weigl et al., 2012) reflects the variable’s relative expression range along one rhythm cycle.
In the case of simple 24-h rhythm, the values of 1Q and 4Q are similar and equal to 8 h. Any deviation from these values reflects at least a partial desynchrony between the internal rhythm components. Statistics Two-way unpaired t tests with Welch correction were utilized to assess significant differences between the “basic” parameters of the different transcripts/tissues. Differences between the level parameters: mesor, amplitude and FR of
