No time for spruce: rapid dampening of circadian rhythms in Picea abies (L. Karst)

(1)

No Time for Spruce: Rapid Dampening of Circadian Rhythms in Picea abies (L. Karst)

Niclas Gyllenstrand

^1,

*, Anna Karlgren

²

, David Clapham

¹

, Karl Holm

²

, Anthony Hall

³

, Peter D. Gould

³

, Thomas Ka¨llman

²

and Ulf Lagercrantz

²

1Department of Plant Biology and Forest Genetics, Uppsala Biocenter, Swedish University for Agricultural Sciences, Uppsala, PO Box 7080, SE-750 07 Uppsala, Sweden

2Department of Plant Ecology and Evolution, Evolutionary Biology Center, Uppsala University, 752 36 Uppsala

3School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK

*Corresponding author: E-mail, niclas.gyllenstrand@slu.se; Fax, +46-18 673389.

(Received October 2, 2013; Accepted December 16, 2013)

The identification and cloning of full-length homologs of circadian clock genes from Picea abies represent a first step to study the function and evolution of the circadian clock in gymnosperms. Phylogenetic analyses suggest that the sequences of key circadian clock genes are conserved between angiosperms and gymnosperms. though fewer homologous copies were found for most gene families in P. abies. We detected diurnal cycling of circadian clock genes inP. abies using quantitative real-time PCR; however, cycling appeared to be rapidly dampened under free- running conditions. Given the unexpected absence of transcriptional cycling during constant conditions, we employed a complementary method to assay circadian rhythmic outputs and measured delayed fluorescence in seedlings of Norway spruce. Neither of the two approaches to study circadian rhythms in Norway spruce could detect robust 24 h cycling behavior under constant conditions. These data suggest gene conservation but fundamental differences in clock function between gymnosperms and other plant taxa.

Keywords: Circadian clock Delayed fluorescence Diurnal cycling Gene expression Picea abies Phylogeny.

Abbreviations: CT, cycle threshold; DD, constant darkness;

DF, delayed fluorescence; EC, evening complex; EST, expressed sequence tag; LD, long-day; LL, constant light; PRR, psudoresponse regulator; RACE, rapid amplification of cDNA ends; R^:^FR, red^:^far-red; RT-PCR, real-time PCR; SD, short-day.

Introduction

The ability to anticipate daily rhythms is crucial to many plant species. This requires an endogenous temporal oscillator, i.e. a self-sustained rhythm, to time physiological and developmental processes in anticipation of coming changes in, for example, light and temperature. Most species have thus evolved a circadian clock with a period of around 24 h, and data suggest that

circadian rhythms are important for fitness (Dodd et al. 2005).

Definitions of circadian rhythms include that the rhythms must be self-sustained so that they persist despite a lack of external time cues (e.g. in constant light and temperature conditions), are entrainable by zeitgebers and that they remain largely constant under a range of temperature conditions (Covington et al.

1998,Johnson et al. 1998,Harmer et al. 2001,McClung 2009).

A circadian clock is also coupled to photoperiodic responses, where interaction between external light and the endogenous rhythm is used to measure photoperiod (Thomas 1998).

In plants, photoperiodic responses are important for several developmental processes including timing of reproduction and control of growth phenology. Again, a correct timing is important for plant adaptation, and failure of synchronization with the environment can have a negative effect on fitness.

Accordingly, genes in the circadian clock have been shown to be important for adaptation to local conditions in several plant species (Michael et al. 2003,Dodd et al. 2005,Turner et al. 2005, Jones et al. 2008).

In conifers as well as other temperate trees, photoperiod has a strong impact on the annual growth cycle (Ekberg et al. 2010).

Typically, shorter days in the autumn induce growth cessation and bud set. The responses to photoperiod often show strong latitudinal clines, in particular at high latitudes, indicating strong selective forces acting through photoperiodic response.

In Picea abies, the critical night length displays a steep cline from about 2 h at latitude 67N, to 7 h at latitude 45N (Ekberg et al. 2010). Despite the importance of photoperiodic responses in conifers, limited information is available on the mechanisms controlling these responses. It is noteworthy that a recent study of sequence variation of putative conifer circadian clock genes in spruce showed that variation at some of these genes has been shaped by local adaptation to photoperiod (Chen et al.

2012).

Most of our knowledge of the molecular components of the circadian clock comes from the model organism Arabidopsis thaliana. The circadian clock in Arabidopsis is best viewed as a

Plant Cell Physiol. 0(0): 1–16 doi:10.1093/pcp/pct199, available online at www.pcp.oxfordjournals.org

!The Author 2013. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.

Regular Paper

at Uppsala Universitetsbibliotek on February 10, 2014http://pcp.oxfordjournals.org/Downloaded from

(2)

complex network including many genes (Yanovsky and Kay 2001,Hanano et al. 2008), but at its core the clock consists of interconnected feedback loops with a limited set of known components (Wang et al. 1997,Dodd et al. 2005,Locke et al.

2006,Holm et al. 2010,van Ooijen et al. 2011,Nagel and Kay 2012,Pokhilko et al. 2012). Mathematical modeling and experimental data suggest the existence of connected morning and evening oscillators. The morning oscillator consists of the closely related MYB transcription factors CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) that positively regulate the PSEUDO RESPONSE REGULATORS 7 and 9 which in turn repress CCA1 and LHY transcription (Covington et al. 1998, Johnson et al. 1998, Alabadı´ et al. 2001, Harmer et al. 2001, Farre´ et al. 2005, McClung, 2009, Nakamichi et al. 2010,Nusinow et al. 2011, van Ooijen et al. 2011,Pokhilko et al. 2012). The evening loop includesGIGANTEA (GI) and TIMING OF CAB EXPRESSION 1 (TOC1), and recent data also incorporate an evening complex (EC) consisting of LUX ARRHYTHMO (LUX), EARLY FLOWERING 3 (ELF3) and EARLY FLOWERING 4 (ELF4) (Thomas 1998, Kim et al. 2003, Michael et al. 2003, Dodd et al. 2005,Turner et al. 2005, Jones et al. 2008, Holm et al.

2010, Nusinow et al. 2011, van Ooijen et al. 2011, Pokhilko et al. 2012). The EC binds different target genes, such as TOC1, GI, PRR9 and LUX, and represses their expression (Ekberg et al. 2010, Dixon et al. 2011, Helfer et al. 2011, Schumann et al. 2011). The GI protein is hypothesized to down-regulate the EC, causing expression of EC target genes to increase since inhibition from the EC is released, implicating an indirect positive regulation by GI on evening expressed genes such as TOC1. Previously, TOC1 was thought indirectly to induce the expression ofCCA1 and LHY, thus connecting the morning and evening loops, but Huang et al. (2012) showed that TOC1 functions rather as a general repressor of several clock genes and thus prevents activation of morning loop genes (Gendron et al. 2012).

Comparative studies have revealed a large degree of conservation of key circadian clock genes among monocotyledons and dicotyledons (Hazen et al. 2005, Onai and Ishiura 2005, Murakami et al. 2006,Ekberg et al. 2010,Holm et al. 2010,Dai et al. 2011) and to some extent also among distant taxa such as the mossPhyscomitrella patens (Alabadı´ et al. 2001,Holm et al.

2010,Chen et al. 2012,Karlgren et al. 2013). The identification of circadian clock genes in common across the plant kingdom can provide important insights into the evolution of developmental processes controlled by photoperiod and diurnally controlled processes (Matsushika et al. 2000, Yanovsky and Kay 2001, Hanano et al. 2008,Pokhilko et al. 2012,Karlgren et al. 2013).

In addition, comparative studies can identify evolutionary novelties as exemplified by the evolutionary history of the Physcomitrella clock where the clock components GI, ZEITLUPE (ZTL) and TOC1 are missing (Brinker et al. 2001, Holm et al. 2010,Rugnone et al. 2013). Homologs of the GI protein seem also to be absent in Chlamydomonas and animals (Lumsden and Millar 1998, Piechulla et al. 2001,

Mizoguchi et al. 2005,Holm et al. 2010) and are thus possibly confined to higher plants, where they have important roles in, for example, shaping circadian rhythms and in temperature compensation (Oberschmidt et al. 1994, Park et al. 1999, Mizoguchi et al. 2005,Gould et al. 2006).

To date, most plant model organisms are angiosperms, with the exception of Physcomitrella and Selaginella. As an important step to investigate gymnosperm circadian clock mechanisms and their evolution, we identified putative conifer full-length homologs of angiosperm core circadian clock genes based on current spruce expressed sequence tag (EST) databases and studied phylogenetic relationships and expression patterns under different light regimes. As a means to identify a circadian output phenotype, we also applied a method based on delayed fluorescence (Burgerstein 1900,Gould et al. 2009).

Results

Identification of conifer clock gene homologs

Homologs to core clock genes as describes by Pokhilko et al.

(2012)were. Using blast searches against spruce EST databases, we identified putative homologs of PSEUDO-RESPONSE REGULATOR (PRR) genes including TOC1, LATE ELONGATED LHY/CCA1, GI, ZTL, ELF4 and LUX. None of the spruce ESTs was a full-length gene since the conifer EST databases mainly contain 3⁰-reads and fewer 5⁰-reads. We used rapid amplification of cDNA ends (RACE) reactions to identify the full-length gene in each case. The identified full-length spruce genes were aligned against homologs from other species (Supplementary fig. 1, Supplementary table 1), and used to investigate the phylogenetic relationships among plant taxa. Searches for additional copies of the identified genes in the recently published Norway spruce draft genome (Nystedt et al. 2013) did not reveal any new genes.

CCA1 and LHY are closely related transcription factors regulating circadian rhythms and contain a single Myb domain at the N-terminal end. We obtained three sequences in the spruce EST databases that gave a significant hit to the Arabidopsis CCA1/LHY Myb domain. Only one transcript showed cycling behavior under photoperiod: PaCCA1, which encodes 4a 731 amino acid protein. The other two ESTs did not cycle and encode a much shorter peptide (81 amino acids) than CCA1/

LHY and were thus excluded from further analysis. ThePaCCA1 gene shows 77% and 84% similarity to the ArabidopsisCCA1 and LHY genes across the Myb domain. Phylogenetic reconstruction suggests that PaCCA1 shares a common ancestor withCCA1/LHY in angiosperms (Fig. 1A).

The promoter sequence ofPaCCA1 contains a HUD motif (C ACATG) that recently was shown to be a binding site for TOC1 in Arabidopsis and acis-element (G-box; CACGTGKM) that are enriched in TOC1-regulated genes (Gendron et al. 2012).

PRR genes contain an N-terminal pseudo-receiver domain and a C-terminal CCT (CONSTANS, CONSTANS-LIKE and TOC1) domain. The small gene family contains five genes in

(3)

Fig. 1 Phylogenetic trees of circadian clock genes among plant taxa. Trees were constructed using PhyML 3.0 and rooted with eitherP. patens orS. moellendorffii. Trees represent (A) LHY/CCA1, (B) PRR genes, (C) GI, (D) ZTL, (E) ELF4 and (F) LUX. For a full description of genes seeSupplementary table 1.

(continued)

(4)

Fig. 1 Continued.

(5)

Fig. 1 Continued.

(6)

Arabidopsis: PRR1 (TOC1), PRR3, PRR5, PRR7 and PRR9.

Previous phylogenetic analyses in angiosperms suggest that this family is divided into three clades (PRR1, PRR3/7 and PRR5/9) that diverged before the separation of monocots and eudicots, of which PRR1/TOC1 are suggested to be ancestral (Satbhai et al. 2010,Takata et al. 2010). All the ArabidopsisPRR genes are involved in the circadian clock (Eriksson and Millar 2003,Farre´ et al. 2005).

We identified three conifer homologs:PaPRR1, PaPRR3 and PaPRR7. The PaPRR1 gene encodes a 564 amino acid protein with 65% (REC domain) and 79% (CCT domain) identity to PRR1. The predicted PaPRR3 protein is 890 amino acids with 68% (REC domain) and 81% (CCT domain) identity to the Arabidopsis PRR3 protein, while thePaPRR7 gene encodes an 814 amino acid protein with 76% (REC domain) and 95% (CCT domain) identity to PRR7. The phylogenetic positions of the conifer PRR genes suggest that thePaPRR1 gene is an ortholog of angiospermPRR1 (TOC1) while PaPRR3 and PaPRR7 both cluster with angiospermPRR3/7 genes (Fig. 1B). A comparison of identities in the REC and CCT domains suggests thatPaPRR3 andPaPRR7 are more similar to angiosperm PRR3/7 than to PRR5/9 (Supplementary table 2). Blast searches using the five Arabidopsis PRR genes as queries against the Pinus taeda genome, v. 0.6 (http://dendrome.ucdavis.edu/resources/

databases/) resulted in three significant hits, strengthening the view that conifers have only threePRR genes.

A putative conserved binding site motif for PaCCA1 was found in the promoter of the PaPRR1 gene. The motif in PaPRR1 (AAAAATCT) is similar to the evening element (EE) present inTOC1 and other cycling genes (AAAATATCT;Wang et al. 1997). As CCA1 was shown to bind the EE and negatively regulate TOC1 in Arabidopsis (Alabadı´ et al. 2001), a similar interaction is possible between P. abies homologs.

Transcriptional repressor activity of PRR5, PRR7 and PRR9 on CCA1 and LHY has recently been shown (Nakamichi et al. 2010).

A conserved motif was found to be sufficient for repressor activity and the molecular mechanism includes the protein TOPLESS and histone deacetylase activity to repress transcription and alter the circadian period (Wang et al. 2013). The required conserved repression motif contains an EAR (ethyl- ene-responsive element binding factor-associated amphiphilic repression) motif (LxLxL) that can also be found in PaPRR1, PaPRR3 and PaPRR7 (Supplementary fig. 2).

TheGI gene encodes a nuclear protein without known domains. One conifer homolog was identified: PaGI, encoding a 1,179 amino acid protein with 58% identity to the Arabidopsis GI protein. OneGI copy is found in most species, including Selaginella, but no homolog is found in the moss Physcomitrella (Holm et al. 2010). The PaGI gene is basal to angiospermGI genes (Fig. 1C).

TheZTL gene contains three main domains: an N-terminal Per-Arnt-Sim/Light, oxygen or voltage (PAS/LOV) domain including a terminal PAS-associated, C-terminal (PAC) domain, a central F-box and a C-terminal domain consisting of Kelch repeats. Although the Arabidopsis ZTL mRNA is

constitutively expressed, the protein levels are under circadian control (Kim et al. 2003). The Arabidopsis genome contains three members of this small gene family: ZTL, FLAVIN- BINDING KELCH F-BOX1 (FKF1) and LOV KELCH PROTEIN2 (LKP2). Among conifer ESTs, we could identify one gene with strong homology:PaZTL with 78% identity to the Arabidopsis ZTL (83, 85 and 85% identity for the PAS/LOV, F-box and Kelch repeat domains, respectively). ThePaZTL cDNA encodes a 621 amino acid protein. The spruce gene clusters with angiosperm ZTL genes, while FKF1/LKP2-like genes were not detected in gymnosperm or lycophytes, indicating thatFKF1 might be ex- clusive to angiosperms. This is in accordance with previous results (Ekberg et al. 2010,Schumann et al. 2011) (Fig. 1D).

We could not retrieve any significant ELF3 hit either in EST databases or in the conifer genomesP. taeda genome v.0.6 and P. abies. The best blastx hit was against the genome assembly and had an e-value of 2e-07 and score of 64.3, and encoded a putative protein of 260 amino acids, while the Arabidopsis protein is 695 amino acids. The putative conifer protein does not contain any of the proline-rich, threonine-rich, glutamine-rich regions or the nuclear targeting signal that has been described for the ArabidopsisELF3 gene (Hicks et al. 2001,Karlgren et al.

2013) and as we could not find the same sequence in ESTs from conifers it is perhaps not very likely that the identified part of the genome assembly corresponds to an expressed gene.

One putative spruce homolog ofELF4 was retrieved: PaELF4.

ThePaELF4 cDNA encodes a 119 amino acid protein containing a domain of unknown function (DUF) 1313 domain with 51%

identity to the DUF 1313 domain in the ArabidopsisELF4 gene.

The phylogenetic position ofPaELF4 suggests that it belongs rather to a sister lineage to angiosperm ELF4 (Fig. 1E). The biological functions ofELF4-L1-4 are currently unknown.

The Arabidopsis LUX gene had one putative spruce homolog: PaLUX encoding a 274 amino acid protein. The myb-SHAQKYF domain of PaLUX had 98% identity to AtLUX. The phylogenetic tree reveals three major clades.

Angiosperm genes are present in two clades, andPaLUX clusters with the Selaginella gene, Selmo1_36646, Os05g0412000, POPTR_0016s03240 and POPTR_0006s03340 rather than the clade containingLUX (Fig. 1F). Among myb-SHAQKYF genes, only three genes [AtLUX, At5g59570 and PCL1 (Os01g0971800)]

have been shown to be circadian clock genes (Hazen et al. 2005, Onai and Ishiura 2005,Dai et al. 2011). Whether genes in the clade with PaLUX are circadian clock genes is currently unknown. Conifer homologs to most but not all Arabidopsis circadian clock genes could thus be identified (Supplementary table 3).

Expression patterns of spruce clock homologs

We investigated the temporal expression patterns of the identified spruce homologs of Arabidopsis circadian clock genes, using RNA samples from needles of seedlings. Expression patterns were studied under long-day (LD) conditions with 19 h light and 5 h darkness, short-day (SD) conditions with 8 h light and 16 h darkness, constant darkness (DD) and constant

(7)

light (LL). DD and LL was studied after entrainment in a 12 h light and 12 h darkness regime. Under SD conditions, a distinct cycling expression pattern was observed for all the studied genes except for the constitutive expression ofPaZTL (Figs. 2, 3). This further supports that PaZTL is a homolog of angiosperm ZTL/LKP2 rather than FKF1. Angiosperm ZTL and LKP2 also show constitutive expression, while FKF1 shows diurnal and circadian expression rhythms.

In LD conditions, the amplitude of the expression profiles was lower for all studied genes as compared with SD and 12 : 12 h photoperiods (Figs. 2–4). Hence, there was a strong effect of photoperiod on the rhythmic expression of transcript levels of spruce homologs. Diurnal cycling of spruce clock genes was also detected in adult trees under natural light conditions (Supplementary fig. 3). Of the studied genes,PaCCA1 was the only one displaying peak expression during the dark period, a pattern similar to that observed in angiosperms. Furthermore, the expression phases of PaPRR1, PaPRR3 and PaPRR7 were largely opposite to that of PaCCA1, in line with what has been observed for the angiospermLHY/CCA1 and PRR genes (Alabadı´ et al. 2001). The transcript abundance ofPaGI showed marked cycling with steady accumulation during the light period and peak expression at dusk, in accordance with angio- spermGI expression (Fig. 2).

The expression patterns ofPaELF4 and PaLUX were diurnal with low amplitude in SD conditions, while a clear rhythm was not obvious in LD conditions (Fig. 3).

In order to test for the light response function of the putative spruce clock genes, trees were grown in a 12 h light/12 h dark photoperiod for 3 d before transfer to constant dark conditions at dusk. Replicate samples were subjected once to a 1 h light exposure at 4 h intervals. Gene expression of the clock genes was measured before and after the 1 h light exposure (Fig. 5). From these data we can conclude that none of the investigated genes showed a noticeable photo-induction following light exposure at any time.

We studied the expression patterns during LL and DD conditions for PaGI, PaCCA1, PaPRR1 and PaPRR7. These genes display a clear diurnal rhythm, and represent homologs to the genes in the morning and evening loops of the recently proposed repressilator model of the circadian clock in Arabidopsis (Pokhilko et al. 2012). Despite the robust cycling observed for these genes under SD and 12 : 12 h photoperiods, a surprising lack of rhythm was observed when entrained plants were transferred to either LL or DD conditions (Fig. 4).

Assuming that the assayed genes are part of theP. abies core clock, these experiments indicate that the circadian rhythm is rapidly dampened under free-running conditions.

Delayed fluorescence does not detect a circadian rhythm in spruce

In an attempt to capture circadian behavior in clock outputs, we also monitored delayed fluorescence (DF) over time in several individual spruce samples. DF refers to the extremely weak

light emitted in the dark by pre-illuminated photosynthetic tissue, and it has been shown to be under circadian control in a number of plant species (Gould et al. 2009). DF was measured in LL and, in order to simulate a response to DD, we used a low-light regime resembling a skeleton photoperiod with min- imal light pulses every hour to excite the photosynthetic tissue (see the Materials and Methods for details). While not consti- tuting a DD light regime proper, we refer to it as low light DF in the following discussion. However, no rhythm with an approxi- mate 24 h period could be detected in the gymnospermP. abies in the DF assay. In both LL and low light conditions (see the Materials and Methods for details),P. abies DF data indicated an ultradian rather than a circadian rhythm, with a period around 2–3 h (Fig. 6). For comparison, we also analyzed samples from other plant species with a documented circadian rhythm in the same LL and low light regimes. For samples of the moss Physcomitrella patens, a circadian rhythm was observed in low light but not in LL, which is consistent with gene expression patterns ofP. patens clock genes (Fig. 6;Holm et al. 2010). For samples ofCapsella bursa-pastoris seedlings, a robust circadian rhythm was observed in LL, and forHordeum vulgare leaves, a robust rhythm with an 24 h period was observed using our low light regime (Fig. 6). In conclusion, these results suggest that either PSII activity (Gould et al.

2009, and references therein) in spruce is not coupled to the circadian clock under constant conditions or the spruce circadian clock does not show a robust rhythm in constant conditions.

In summary, homologs to most of the key circadian clock genes of Arabidopsis were identified in Norway spruce, but fewer gene family members were generally observed. The transcript abundance of all genes except PaZTL showed diurnal cycling. Under constant conditions, the cycling was rapidly dampened, unlike for angiosperm clock genes. Finally, assays of circadian output based on DF did not detect a circadian rhythm.

Discussion

Gymnosperms comprise a similar set of circadian clock genes to angiosperms

We could identify conifer homologs to several plant circadian clock genes and conclude that these genes are conserved at the sequence level between angiosperms and gymnosperms.

However, we found fewer spruce genes within most gene families, which suggests that the existence of larger gene families in angiosperms is due to duplications within the angiosperm lineage. The sequenced genomes ofPhyscomitrella and Selaginella allow for comparison with bryophytes and lycophytes as representatives of non-seed plants. The Selaginella genome contains a similar number of putative clock genes to spruce: one homolog each ofGI, ZTL and CCA1/LHY genes and three PRR homologs. Among the three Selaginella PRR genes, one clusters with the PRR1 (TOC1) group together with the spruce genePaPRR1. Physcomitrella lacks homologs to GI, ZTL

(8)

and, most probablyTOC1, but contains two CCA1 and four PRR gene homologs (Holm et al. 2010). The GI, ZTL and TOC1 functions might thus represent novelties among vascular plants. It should be noted that aZTL copy has been annotated inOstreococcus tauri (van Ooijen et al. 2011) though this gene

lacks the important PAS/LOV domain and might thus represent an ancestor to plantZTL genes. In Physcomitrella, the two CCA1 and four PRR genes, respectively, seem to be closely related and probably represent duplications within the Physcomitrella lineage (Holm et al. 2010). In summary, analyses Fig. 2 Expression patterns of putative circadian clock genes,PaGI, PaCCA1, PaPRR7 and PaPRR1, in P. abies under a short-day (SD) regime (8 h light/16 h dark) on the left, and under a long-day (LD) regime (19 h light/5 h dark) on the right. Graphic representations of light regimes are indicated at the top of each panel. Unfilled boxes represent the light period and filled boxes represent the dark period, respectively. Plants were sampled every fourth hour during 4 d. Quantitative RT-PCR expression values were normalized using CTtarget–CTreference.

(9)

of sequence data suggest that most of the core circadian clock components present in the angiosperms were present already in lycophytes and gymnosperms. Further, functional studies of spruce circadian clock genes suggest that the protein function of theses clock components are partly conserved (Karlgren et al.

2013). However, we could not detect any quick light response of any of these clock components as shown for APRR9 (Matsushika et al. 2000). Genes that are induced by light during the night have been shown to be enriched for clock genes (Rugnone et al. 2013).

Fig. 3 Expression patterns of putative circadian clock genes,PaELF4, PaLUX, PaPRR3 and PaZTL, in P. abies under a short-day (SD) regime (8 h light/16 h dark) on the left and under a long-day (LD) regime (19 h light/5 h dark) on the right. Plants were sampled every fourth hour during 4 d. Quantitative RT-PCR expression values were normalized using CTtarget–CTreference.

(10)

Fig. 4 Expression patterns of putative circadian clock genes inP. abies under free-running conditions [continuous light (LL), and darkness (DD)].

Plants were sampled every fourth hour. Quantitative RT-PCR expression values were normalized using CTtarget–CTreference. Shown are two representative time series of transcript abundance under free-running conditions.

(11)

Circadian rhythms in Picea abies are rapidly dampened

In light of these findings, the most surprising observation in the present study was a lack of detectable circadian rhythm for the

putative circadian clock genes in spruce grown in constant conditions. Also, no evidence for a circadian rhythm in poten- tial clock outputs was detected. The output of the circadian clock in Arabidopsis is often studied using leaf movement assays (Lumsden and Millar 1998). One study of needle movement in conifers has been reported, but the amplitude of the presumed circadian rhythm was small and the period length deviated greatly from the expected 24 h rhythm under constant conditions (Brinker et al. 2001), and should perhaps not be viewed as strong evidence of a circadian rhythm. We at- tempted to repeat the above experiment and trace needle movements in Norway spruce seedlings; however, we were un- successful and could not detect any such movements. Previous expression studies of putative circadian-regulated genes in gymnosperms have also generally suggested a lack of circadian rhythm (Piechulla et al. 2001). In particular, studies of genes related to photosynthesis, such as genes coding for light- harvesting complex proteins (Lhcs), show constant levels in most gymnosperm species, despite their circadian regulation in angiosperms, as well as in algae, moss and fern (Oberschmidt et al. 1994).

A persistent circadian rhythm might not be essential in gymnosperms

In contrast to most other plant species, gymnosperms are able to synthesize Chl in the dark (Burgerstein 1900). Thus, there is no apparent need to predict dusk and dawn for Chl biosynthesis in gymnosperms. In angiosperms, correct anticipation of dusk and dawn via the circadian clock enhances photosynthesis and increases Chl content, which allow plants to fix more carbon and grow faster, and thereby convey a fitness advantage (Thomas 1998,Dodd et al. 2005). If regulation of photosynthesis in anticipation of dawn and dusk is one of the major selective forces to maintain a strong circadian rhythm in plants, the different regulation of Chl biosynthesis in gymnosperms might explain the lack of strong circadian rhythms inP. abies.

Still, gymnosperms includingP. abies show a strong photoperiodic response, in particular in the control of growth cessation and build up of dormancy and frost tolerance (Heide, 1974, Christersson 1978,Michael et al. 2003,Dodd et al. 2005,Turner et al. 2005,Jones et al. 2008). This response requires some sort of time measurement and, even though no circadian rhythm was observed for P. abies clock genes, they generally show strong diurnal cycling especially under shorter photoperiods.

The apparent rapid dampening of cycling in the transcript abundance of spruce clock genes in constant conditions would be compatible with an hourglass model of photoperiodism. The hourglass model states that a gradual accumulation of a prod- uct above a certain threshold will trigger a physiological response. The hourglass model is not dependent on an endogenous timekeeper; instead the process is reset each day by the light cycle (Rensing et al. 2001,Ekberg et al. 2010).

However, night break experiments support that photoperiodic control of budset in Norway spruce involves a circadian rhythm of light sensitivity with an 24 h period in line with the Fig. 5 Transcriptional light response of putative P. abies circadian

clock genes. Plants were entrained in 12 h light/12 h dark. RNA samples were collected at different time points in constant dark, before and after a 1 h light pulse. Filled boxes, expression before light pulse;

open boxes, expression after light pulse; open circles, differential expression (expression before light pulse – expression after light pulse).

Quantitative RT-PCR expression values were normalized using CTtarget

–CTreference. Error bars indicate the standard error.

(12)

external coincidence model (Clapham et al. 2001,Ekberg et al.

2010). These experiments were conducted using a night break technique with an extended dark period. During the extended 40 h night, two peaks of higher sensitivity to a night break with an 24 h spacing were observed. These results suggest that a circadian rhythm persists for at least 24 h in darkness. Such a rhythm was not evident in the present study, either for mRNA levels of clock genes or for outputs related to photosynthetic

activity, but it does not rule out a weak transcriptional rhythm or a rhythm at the protein level for some outputs of the circadian clock inP. abies. Such a circadian rhythm without transcription has been reported for cyanobacteria (Nakajima et al.

2005,Tomita et al. 2005). A rapidly dampened circadian clock has been proposed to explain photoperiodic responses in several insect species (Lewis and Saunders 1987, Saunders and Lewis 1987,Saunders 2011,Chen et al. 2012).

Fig. 6 Analysis of delayed fluorescence does not reveal a circadian rhythm inPicea abies. The plots represent normalized averages for DF of 6–25 samples.Picea abies seedlings (A, n = 8; B, n = 6), P. patens protonema culture (C, n = 8; D, n = 6), C. bursa-pastoris seedlings (E, n = 6) and detachedH. vulgare leaves (F, n = 25) were entrained under 12 h light/12 h dark cycles before transferring to constant RB light (A, C, E) or incubated in constant darkness for 48 h before transferring to the constant low light measurement regime (B, D, F). Error bars indicate the standard error.

(13)

Although regulation of transcription constitutes an important basis for circadian rhythms, recent data show that circadian timekeeping can sometimes be sustained in the absence of transcription (Yanovsky and Kay 2001, Locke et al. 2006, Hanano et al. 2008, Hastings et al. 2008, O’Neill et al. 2011, Nagel and Kay 2012). It is thus possible that an unknown post-translational rhythm in P. abies could be important for various circadian output pathways, including the observed circadian regulation of budset in night break experiments (Clapham et al. 2001,Pokhilko et al. 2012).

Possible reasons for rapid dampening of circadian rhythms in P. abies

Wenden et al. (2012)showed that circadian rhythms of individual cells of Arabidopsis leaves are not as strongly coupled as in animal cells. Weak intercellular coupling resulted in desyn- chronization and spatiotemporal waves of gene expression in leaves grown in constant conditions. That the desynchron- ization between noisy oscillation in single cells can explain dampened oscillations was also recently shown for Arabidopsis by including stochasticity in clock models (Murakami et al. 2006,Guerriero et al. 2012). The lack of obvious circadian gene expression rhythms in Norway spruce needles could partly be due to a weaker coupling of individual cell clocks, but the apparently very rapid dampening within 24 h is not easily explained by decoupled and desynchronized cell clocks.

An alternative reason for the lack of strong circadian rhythms in Norway spruce could be differences in components of the core clock. A potentially important difference between gymnosperm and angiosperm clocks is the apparent lack of homologs to ELF3 in gymnosperms. ELF3 is part of the EC that is required to maintain a circadian rhythm in Arabidopsis.

Mutations in any of the EC genes (ELF3, ELF4 or LUX) results in arrhythmia or comprised rhythmicity in both LL and DD (McWatters et al. 2000, Covington et al. 2001, Doyle et al.

2002,Holm et al. 2010,Herrero et al. 2012). In the case ofelf3 andlux mutants, rhythm is retained under DD conditions and absent under LL conditions (Hicks et al. 1996,Hazen et al. 2005).

Until a high quality genome of a conifer exists, it is still prema- ture to exclude the possibility of a coniferELF3 gene, but the apparent absence of conifer homologs of ELF3 proteins provides a plausible hypothesis to explain the weak circadian rhythm in conifers. ELF3 genes have been reported in P. patens and S. moellendorffi and might thus have been lost in the gymnosperm lineage (Holm et al. 2010).

Ultradian rhythms of delayed fluorescence

Analyses of DF of P. abies seedlings suggested an ultradian rather than a circadian rhythm of photosynthetic activity.

Ultradian rhythms have been observed in several plants species, affecting, for example, stomatal conductance. These synchro- nized oscillations are often accompanied by rhythmic photosynthetic CO₂ uptake, and may be initiated by, for example,

abrupt changes in water status or light (Naidoo and Von Willert 1994,Siebke and Weis 1995,Mizoguchi et al. 2005,Holm et al.

2010). Such mechanisms might result in an ultradian DF rhythm inP. abies.

Conclusions

Although the spruce circadian clock seems to consist of fewer components than the angiosperm clock, probably due to du- plication events among angiosperms, homologs to several core clock genes are present in gymnosperms. Several of these homologs display robust diurnal cycling in mRNA levels, but show rapid dampening in free-running conditions. The rapid dampening suggests that a strong circadian rhythm is not crucial in gymnosperms, possibly an effect of a limited need to anticipate dawn and dusk. As conifers can produce Chl in the dark, timing is probably of less importance. The strong photoperiodic response observed in conifers probably does not re- quire a persistent circadian rhythm. Also, conifers might lack an ELF3 homolog, which is a vital component in the angiosperm circadian clock to sustain rhythm in constant conditions.

Further functional studies are needed to elucidate the clockwork of gymnosperms, and its role in the control of diurnal processes and photoperiodic responses.

Materials and Methods

Plant material and growth conditions

The starting material was seeds collected from a Romanian population of P. abies trees growing in the eastern Carpathians at about 750 m above sea level, latitude 4630⁰, longitude 26. Seedlings were raised in a phytotron under continuous light from metal halogen lamps at 250 mmol m²s¹ (400–700 nm) with a ratio of irradiances at 660 and 730 nm [red : far-red (R : FR)] of 2.0, at 20C (400–700 nm) with a ratio of irradiances at 660 and 730 nm (R : FR) of 2.0, at 20C for 8–10 weeks before experiments were conducted. The plants were grown in 8.56 cm pots, four plants to a pot, in a mixture of peat and coarse sand, with watering every 1–2 d from a weak complete nutrient solution afterIngestad (1979), giving 100 mg N l¹. Experiments were performed in growth cabinets (Fison) with fluorescent tubes giving irradiances at plant level during the light periods of 200 mmol m²s¹, R : FR = 3.0 at 20C.

Cloning full-length cDNAs using RACE reactions

Spruce ESTs for putative conifer circadian clock genes were identified using tblastn searches with Arabidopsis genes as queries at PlantGDB (http://www.plantgdb.org/). Sequence- specific primers were designed for 5⁰- and 3⁰-RACE reactions based on partial ESTs. 5⁰and 3⁰cDNA libraries were prepared using the SMART RACE-PCR kit (Clontech) according to the manufacturer’s protocols. cDNA libraries were prepared from a mix of mRNAs taken from a photoperiodic time series. PCRs were performed using the proofreading Phusion^TMDNA poly- merase (Finnzyme). The resulting PCR fragments were blunt

(14)

end cloned using a Zero Blunt^Õ TOPO^Õ PCR Cloning Kit (Invitrogen) and One Shot^Õ TOP10 Chemically Competent Escherichia coli (Invitrogen) according to the manufacturer’s instructions. Plasmids were isolated using a QIAprep Spin MiniPrep Kit (Qiagen) and subsequently sequenced at Macrogen Inc.

The promoter sequences ofPaCCA1 and PaPRR1 were isolated using the Genome Walker Kit (Clontech) following the manufacturer’s instructions.

Sequence analysis and phylogenetic reconstruction

Blast searches for plant homologs were done using the following plant genomes: A. thaliana, Populus trichocarpa, Oryza sativa, Mesembryanthemum cristallinum, Castanea sativa, S. moellendorffii and P. patens (Supplementary table 1).

Homologs were identified using tblastn searches with Arabidopsis genes as queries against databases at JGI (http://

www.jgi.doe.gov/) and NCBI (http://ncbi.nlm.nih.gov/).

Conserved domains were identified using the Prosite database (http://prosite.expasy.org/). Predicted amino acid sequences of the spruce genes and homologous genes were aligned using Clustal W (Thompson et al. 1994,Park et al. 1999,Mizoguchi et al. 2005,Gould et al. 2006) and visualized by Boxshade 3.31 (http://www.ch.embnet.org/software/BOX_form.html).

Phylogenetic reconstructions were done using PhyML 3.0 (Gould et al. 2009,Guindon et al. 2010) and finally visualized by FigTree 1.2.3 (http://tree.bio.ed.ac.uk/software/figtree/). In all cases except for PRRs, the complete sequences were used in the analysis. For PRRs, the pseudo-receiver (REC) domain and the CONSTANS, CO-like and TOC1 (CCT) domain were used.

Spruce RNA isolation and real-time PCR (RT-PCR)

Total RNA was isolated from needles following the protocol described previously (Azevedo et al. 2003,Nystedt et al. 2013) with minor modifications. Needles were extracted from a pool of four plants. cDNAs were synthesized from 0.5 mg of total RNA using Superscript III reverse transcriptase and random hexamer primers. cDNAs were diluted 1 : 100 and 5 ml were used in dupli- cate quantitative assays using SYBR Green master mix (Molecular Probes) with an ABI 7000 system according to the manufacturer’s protocol. Primers (Supplementary table 4) for RT-PCR were designed to span introns. Expression values were calculated as CT values (CT_control– CT_target) using polyubiquitin as the endogenous control gene.

Delayed fluorescence

DF measurements were performed on 8-week-old spruce seedlings from a northern (67N) and a southern (47N) population.

The imaging system for monitoring delayed fluorescence in living plant tissue was identical to a system used for luciferase imaging (seeGould et al. 2009for a detailed description). Briefly, the system consisted of a 16-bit low light CCD camera cooled to 80C with a high-transmission lens inserted through the

top of a programmable incubator. The light source in the cab- inet consisted of a red and blue light-emitting diode array (35 mmol m²s¹). In LL, DF images were captured once every hour immediately following lights off using a 1 min exposure. For the low light DF assay, the light source was instead turned on for 3 min to excite PSII, and directly followed by lights off and a 1 min exposure once every hour. Fluctuations in DF were traced for 5–7 d. All samples measured in low light were previously incubated in DD for 48 h at 22C (n = 6, 3 northern and 3 southern), whereas plants measured in LL were grown in 12 h light/12 h dark cycles at 22C (n = 8, 4 northern and 4 southern) for 7 d before measurement. Time series image cap- tures were normalized against background intensities and detrended using the BRASS 3.0 software (Locke et al. 2005, Satbhai et al. 2010,Takata et al. 2010) to give final measurements. All experiments and data analyses were performed at the School of Biological Sciences at the University of Liverpool.

Supplementary data

Supplementary dataare available at PCP online.

Funding

This work was supported by the Carl Trygger Foundation [grants to N.G.]; the Swedish Research Council grant no.

2007-6430 and the Swedish Research Council (FORMAS) grant no. 2005-1299 [to U.L.].

Acknowledgments

Kerstin Jeppsson contributed excellent lab assistance. Martin Larsson, Svenska skogsplantor AB, supplied plant material for some experiments.

Disclosures

The authors have no conflicts of interest to declare.

References

Alabadı´, D., Oyama, T., Yanovsky, M.J., Harmon, F.G., Ma´s, P. and Kay, S.A. (2001) Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science 293:

880–883.

Azevedo, H., Lino-Neto, T. and Tavares, R. (2003) An improved method for high-quality RNA isolation from needles of adult mari- time pine trees.Plant Mol. Biol. Rep. 21: 333–338.

Brinker, M., Engelmann, W., Kellmann, J.W. and Piechulla, B. (2001) Circadian rhythms of leaf and stomatal movements in gymnosperm species.Biol. Rhythm Res. 32: 471–478.

Burgerstein, A. (1900) Ueber Das Verhalten Der Gymnospermen- Keimlinge Im Licht Und Im Dunkeln. Berichte Der Deutschen.

Botanischen Gesellschaft.

(15)

Chen, J.J., Ka¨llman, T.T., Ma, X.X., Gyllenstrand, N.N., Zaina, G.G., Morgante, M.M. et al. (2012) Disentangling the roles of history and local selection in shaping clinal variation of allele frequencies and gene expression in Norway spruce (Picea abies).Genetics 191:

865–881.

Christersson, L. (1978) The influence of photoperiod and temperature on the development of frost hardiness in seedlings of Pinus silvestris and Picea abies.Physiol. Plant. 44: 288–294.

Clapham, D.H., Vince-Prue, D., Ekberg, I. and Norell, L. (2001) Circadian timekeeping for the photoperiodic control of budset in Picea abies (Norway spruce) seedlings.Biol. Rhythm Res. 32: 479–487.

Covington, M.F., Panda, S., Liu, X.L., Strayer, C.A., Wagner, D.R. and Kay, S.A. (1998) A clockwork green: circadian programs in photosynthetic organisms.In Biological Rhythms and Photoperiodism in Plants. Edited by Lumsden, P. and Millar, A. pp. 1–34. BIOS Scientific Publishers, Oxford.

Covington, M.F., Panda, S., Liu, X.L., Strayer, C.A., Wagner, D.R. and Kay, S.A. (2001) ELF3 modulates resetting of the circadian clock in Arabidopsis.Plant Cell 13: 1305–1315.

Dai, S., Wei, X., Pei, L., Thompson, R.L., Liu, Y., Heard, J.E. et al. (2011) BROTHER of LUX ARRHYTHMO is a component of the Arabidopsis circadian clock.Plant Cell 23: 961–972.

Dixon, L.E., Knox, K., Kozma-Bogna´r, L., Southern, M.M., Pokhilko, A.

and Millar, A.J. (2011) Temporal repression of core circadian genes is mediated through EARLY FLOWERING 3 in Arabidopsis. Curr.

Biol. 21: 120–125.

Dodd, A.N., Salathia, N., Hall, A., Ke´vei, E., To´th, R., Nagy, F. et al. (2005) Plant circadian clocks increase photosynthesis, growth, survival, and competitive advantage.Science 309: 630–633.

Doyle, M.R., Davis, S.J., Bastow, R.M., McWatters, H.G., Kozma- Bogna´r, L., Nagy, F. et al. (2002) The ELF4 gene controls circadian rhythms and flowering time in Arabidopsis thaliana. Nature 419:

74–77.

Ekberg, I., Eriksson, G. and Dormling, I. (2010) Photoperiodic reactions in conifer species.Ecography 2: 255–263.

Eriksson, M.E. and Millar, A.J. (2003) The circadian clock. a plant’s best friend in a spinning world.Plant Physiol. 132: 732–738.

Farre´, E.M., Harmer, S.L., Harmon, F.G., Yanovsky, M.J. and Kay, S.A.

(2005) Overlapping and distinct roles of PRR7 and PRR9 in the Arabidopsis circadian clock.Curr. Biol. 15: 47–54.

Gendron, J.M., Pruneda-Paz, J.L., Doherty, C.J., Gross, A.M., Kang, S.E.

and Kay, S.A. (2012) Arabidopsis circadian clock protein, TOC1, is a DNA-binding transcription factor. Proc. Natl Acad. Sci. USA 109:

3167–3172.

Gould, P.D., Diaz, P., Hogben, C., Kusakina, J., Salem, R., Hartwell, J. et al.

(2009) Delayed fluorescence as a universal tool for the measurement of circadian rhythms in higher plants.Plant J. 5: 893–901.

Gould, P.D., Locke, J.C.W., Larue, C., Southern, M.M., Davis, S.J., Hanano, S. et al. (2006) The molecular basis of temperature compensation in the Arabidopsis circadian clock. Plant Cell 18:

1177–1187.

Guerriero, M.L., Pokhilko, A., Ferna´ndez, A.P., Halliday, K.J., Millar, A.J.

and Hillston, J. (2012) Stochastic properties of the plant circadian clock.J. R. Soc. 9: 744–756.

Guindon, S., Dufayard, J.-F., Lefort, V., Anisimova, M., Hordijk, W. and Gascuel, O. (2010) New algorithms and methods to estimate max- imum-likelihood phylogenies: assessing the performance of PhyML 3.0.Syst. Biol. 59: 307–321.

Hanano, S., Stracke, R., Jakoby, M., Merkle, T., Domagalska, M.A., Weisshaar, B. et al. (2008) A systematic survey in Arabidopsis

thaliana of transcription factors that modulate circadian param- eters.BMC Genomics 9: 182.

Harmer, S.L., Panda, S. and Kay, S.A. (2001) Molecular bases of circadian rhythms.Annu. Rev. Cell Dev. Biol. 17: 215–253.

Hastings, M.H.M., Maywood, E.S.E. and O’Neill, J.S.J. (2008) Cellular circadian pacemaking and the role of cytosolic rhythms. Curr.

Biol. 18: 805–815.

Hazen, S.P., Schultz, T.F., Pruneda-Paz, J.L., Borevitz, J.O., Ecker, J.R. and Kay, S.A. (2005) LUX ARRHYTHMO encodes a Myb domain protein essential for circadian rhythms. Proc. Natl Acad. Sci. USA 102:

10387–10392.

Heide, O.M. (1974) Growth and dormancy in Norway spruce ecotypes (Picea abies) I. Interaction of photoperiod and temperature.Physiol.

Plant. 30: 1–12.

Helfer, A.A., Nusinow, D.A.D., Chow, B.Y.B., Gehrke, A.R.A., Bulyk, M.L.M. and Kay, S.A.S. (2011) LUX ARRHYTHMO encodes a nighttime repressor of circadian gene expression in the Arabidopsis core clock.Curr. Biol. 21: 126–133.

Herrero, E.E., Kolmos, E.E., Bujdoso, N.N., Yuan, Y.Y., Wang, M.M., Berns, M.C.M. et al. (2012) EARLY FLOWERING4 recruitment of EARLY FLOWERING3 in the nucleus sustains the Arabidopsis circadian clock.Plant Cell 24: 428–443.

Hicks, K.A., Albertson, T.M. and Wagner, D.R. (2001) EARLY FLOWERING3 encodes a novel protein that regulates circadian clock function and flowering in Arabidopsis. Plant Cell 13:

1281–1292.

Hicks, K.A., Millar, A.J., Carre´, I.A., Somers, D.E., Straume, M., Meeks- Wagner, D.R. et al. (1996) Conditional circadian dysfunction of the Arabidopsis Early-Flowering 3 mutant.Science 274: 790–792.

Holm, K., Ka¨llman, T., Gyllenstrand, N., Hedman, H. and Lagercrantz, U.

(2010) Does the core circadian clock in the moss Physcomitrella patens (Bryophyta) comprise a single loop?.BMC Plant Biol. 10: 109.

Ingestad, T. (1979) Mineral nutrient requirements of Pinus silvestris andPicea abies seedlings. Physiol. Plant. 45: 373–380.

Johnson, C.H., Knight, M., Trewavas, A. and Kondo, T. (1998) A clockwork green: circadian programs in photosynthetic organisms. In Biological Rhythms and Photoperiodism in Plants. Edited by Lumsden, P.J. and Millar, A.J. pp. 1–34. BIOS Scientific Publishers, Oxford.

Jones, H., Leigh, F.J., Mackay, I., Bower, M.A., Smith, L.M.J., Charles, M.P.

et al. (2008) Population-based resequencing reveals that the flowering time adaptation of cultivated barley originated east of the fertile crescent.Mol. Biol. Evol. 25: 2211–2219.

Karlgren, A., Gyllenstrand, N., Ka¨llman, T. and Lagercrantz, U. (2013) Conserved function of core clock proteins in the gymnosperm Norway spruce (Picea abies L. Karst).PLoS One 8: e60110.

Kim, W.-Y., Geng, R. and Somers, D.E. (2003) Circadian phase-specific degradation of the F-Box protein ZTL is mediated by the proteasome.Proc. Natl Acad. Sci. USA 100: 4933–4938.

Lewis, R.D. and Saunders, D.S. (1987) A damped circadian oscillator model of an insect photoperiodic clock. I. Description of the model based on a feedback control system.J. Theor. Biol. 128: 47–59.

Locke, J.C.W., Kozma-Bogna´r, L., Gould, P.D., Fehe´r, B., Kevei, E., Nagy, F.

et al. (2006) Experimental validation of a predicted feedback loop in the multi-oscillator clock of Arabidopsis thaliana.Mol. Syst. Biol.

2: 59.

Locke, J.C.W., Southern, M.M., Kozma-Bogna´r, L., Hibberd, V., Brown, P.E., Turner, M.S. et al. (2005) Extension of a genetic network model by iterative experimentation and mathematical analysis.Mol.

Syst. Biol. 1: 2005.0013.

(16)

Lumsden, D.M. and Millar, A.J. (eds) (1998) Biological Rhythms and Photoperiodism in Plants. BIOS Scientific, Oxford.

Matsushika, A., Makino, S., Kojima, M. and Mizuno, T. (2000) Circadian waves of expression of the APRR1/TOC1 family of pseudo-response regulators in Arabidopsis thaliana: insight into the plant circadian clock.Plant Cell Physiol. 41: 1002–1012.

McClung, C.R. (2009) Circadian rhythms. Linking the loops.Science 323: 1440–1441.

McWatters, H.G., Bastow, R.M., Hall, A. and Millar, A.J. (2000) The ELF3 Zeitnehmer regulates light signalling to the circadian clock.Nature 408: 716–720.

Michael, T.P., Salome´, P.A., Yu, H.J., Spencer, T.R., Sharp, E.L., McPeek, M.A. et al. (2003) Enhanced fitness conferred by natur- ally occurring variation in the circadian clock. Science 302:

1049–1053.

Mizoguchi, T., Wright, L., Fujiwara, S., Cremer, F., Lee, K., Onouchi, H.

et al. (2005) Distinct roles of GIGANTEA in promoting flowering and regulating circadian rhythms in Arabidopsis. Plant Cell 17:

2255–2270.

Murakami, M.M., Tago, Y.Y., Yamashino, T.T. and Mizuno, T.T. (2006) Comparative overviews of clock-associated genes of Arabidopsis thaliana and Oryza sativa.Plant Cell Physiol. 48: 110–121.

Nagel, D.H. and Kay, S.A. (2012) Complexity in the wiring and regulation of plant circadian networks.Curr. Biol. 22: R648–R657.

Naidoo, G. and Von Willert, D.J. (1994) Stomatal oscillations in the mangrove Avicennia germinans.Funct. Ecol. 8: 651–657.

Nakajima, M., Imai, K., Ito, H., Nishiwaki, T., Murayama, Y., Iwasaki, H.

et al. (2005) Reconstitution of circadian oscillation of cyanobacterial KaiC phosphorylation in vitro.Science 308: 414–415.

Nakamichi, N., Kiba, T., Henriques, R., Mizuno, T., Chua, N.-H. and Sakakibara, H. (2010) PSEUDO-RESPONSE REGULATORS 9, 7, and 5 are transcriptional repressors in the Arabidopsis circadian clock.

Plant Cell 22: 594–605.

Nusinow, D.A.D., Helfer, A.A., Hamilton, E.E.E., King, J.J.J., Imaizumi, T.T., Schultz, T.F.T. et al. (2011) The ELF4–ELF3–LUX complex links the circadian clock to diurnal control of hypocotyl growth.Nature 475: 398–402.

Nystedt, B., Street, N.R., Wetterbom, A., Zuccolo, A., Lin, Y.-C., Scofield, D.G. et al. (2013) The Norway spruce genome sequence and conifer genome evolution.Nature 497: 579–584.

O’Neill, J.S., van Ooijen, G., Dixon, L.E., Troein, C., Corellou, F., Bouget, F.-Y. et al. (2011) Circadian rhythms persist without transcription in a eukaryote.Nature 469: 554–558.

Oberschmidt, O.O., Hu¨cking, C.C. and Piechulla, B.B. (1994) Diurnal Lhc gene expression is present in many but not all species of the plant kingdom.Plant Mol. Biol. 27: 147–153.

Onai, K. and Ishiura, M. (2005) PHYTOCLOCK 1 encoding a novel GARP protein essential for the Arabidopsis circadian clock.Genes Cells 10: 963–972.

Park, D.H., Somers, D.E., Kim, Y.S., Choy, Y.H., Lim, H.K., Soh, M.S. et al.

(1999) Control of circadian rhythms and photoperiodic flowering by the Arabidopsis GIGANTEA gene.Science 285: 1579–1582.

Piechulla, B., Brinker, M., Jaschke, K., Kellmann, J. and Wissel, K. (2001) Circadian gene expression in angiosperms and gymnosperms.

Endocytobiosis Cell Res. 14: 33–44.

Pokhilko, A., Ferna´ndez, A.P., Edwards, K.D., Southern, M.M., Halliday, K.J. and Millar, A.J. (2012) The clock gene circuit in

Arabidopsis includes a repressilator with additional feedback loops.Molec. Syst. Biol 8: 574.

Rensing, L., Meyer-Grahle, U. and Ruoff, P. (2001) Biological timing and the clock metaphor: oscillatory and hourglass mechanisms.

Chronobiol. Int. 18: 329–369.

Rugnone, M.L., Soverna, A.F., Sanchez, S.E., Schlaen, R.G., Hernando, C.E., Seymour, D.K. et al. (2013) LNK genes integrate light and clock signaling networks at the core of the Arabidopsis oscillator.Proc. Natl Acad. Sci. USA 110: 12120–12125.

Satbhai, S.B., Yamashino, T., Okada, R., Nomoto, Y., Mizuno, T., Tezuka, Y. et al. (2010) Pseudo-response regulator (PRR) homo- logues of the moss Physcomitrella patens: insights into the evolution of the PRR family in land plants.DNA Res. 18: 39–52.

Saunders, D.S. (2011) Unity and diversity in the insect photoperiodic mechanism.Entomol. Sci. 14: 235–244.

Saunders, D.S. and Lewis, R.D. (1987) A damped circadian oscillator model of an insect photoperiodic clock.J. Theor. Biol. 128: 61–71.

Schumann, N., Navarro-Quezada, A., Ullrich, K., Kuhl, C. and Quint, M.

(2011) Molecular evolution and selection patterns of plant F-box proteins with C-terminal Kelch repeats.Plant Physiol. 155: 835–850.

Siebke, K. and Weis, E. (1995) Imaging of chlorophyll-a-fluorescence in leaves: topography of photosynthetic oscillations in leaves of Glechoma hederacea.Photosynth. Res. 45: 225–237.

Takata, N., Saito, S., Saito, C.T. and Uemura, M. (2010) Phylogenetic footprint of the plant clock system in angiosperms: evolutionary processes of pseudo-response regulators.BMC Evol. Biol. 10: 126.

Thomas, B. (1998) Photoperiodism: an overview.In Biological Rhythms and Photoperiodism in Plants. Edited by Lumsden, P.J. and Millar, A.J. pp. 151–156. BIOS Scientific, Oxford.

Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTAL W:

improving the sensitivity of progressive multiple sequence align- ment through sequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res. 22K: 4673–4680.

Tomita, J., Nakajima, M., Kondo, T. and Iwasaki, H. (2005) No transcription–translation feedback in circadian rhythm of KaiC phosphorylation.Science 307: 251–254.

Turner, A., Beales, J., Faure, S., Dunford, R.P. and Laurie, D.A. (2005) The pseudo-response regulator Ppd-H1 provides adaptation to photoperiod in barley.Science 310: 1031–1034.

van Ooijen, G., Dixon, L.E., Troein, C. and Millar, A.J. (2011) Proteasome function is required for biological timing throughout the twenty- four hour cycle.Curr. Biol. 21: 869–875.

Wang, L., Kim, J. and Somers, D.E. (2013) Transcriptional corepressor TOPLESS complexes with pseudoresponse regulator proteins and histone deacetylases to regulate circadian transcription.Proc. Natl Acad. Sci. USA 110: 761–766.

Wang, Z.Y., Kenigsbuch, D., Sun, L., Harel, E., Ong, M.S. and Tobin, E.M.

(1997) A Myb-related transcription factor is involved in the phyto- chrome regulation of an Arabidopsis Lhcb gene. Plant Cell 9:

491–507.

Wenden, B., Toner, D.L., Hodge, S.K., Grima, R. and Millar, A.J. (2012) Spontaneous spatiotemporal waves of gene expression from biological clocks in the leaf.Proc. Natl Acad. Sci. USA 109: 6757–676.

Yanovsky, M.J. and Kay, S.A. (2001) Signaling networks in the plant circadian system.Curr. Opin. Plant Biol. 4: 429–435.