Abstract

The plant circadian clock generates rhythms with a period close to 24 h, and it controls a wide range of physiological and developmental oscillations in habitats under natural light/dark cycles. Among clock-controlled developmental events, the best characterized is the photoperiodic control of flowering time in Arabidopsis thaliana. Recently, it was also reported that the clock regulates a daily and rhythmic elongation of hypocotyls. Here, we report that the promotion of hypocotyl elongation is in fact dependent on changes in photoperiods in such a way that an accelerated hypocotyl elongation occurs especially under short-day conditions. In this regard, we provide genetic evidence to show that the circadian clock regulates the photoperiodic (or seasonal) elongation of hypocotyls by modulating the expression profiles of the PIF4 and PIF5 genes encoding phytochrome-interacting bHLH (basic helix–loop–helix) factors, in such a manner that certain short-day conditions are necessary to enhance the expression of these genes during the night-time. In other words, long-day conditions are insufficient to open the clock-gate for triggering the expression of PIF4 and PIF5 during the night-time. Based on these and other results, the photoperiodic control of hypocotyl elongation is best explained by the accumulation of PIF4 and PIF5 during the night-time of short days, due to coincidence between the internal (circadian rhythm) and external (photoperiod) time cues. This mechanism is a mirror image of the photoperiod-dependent promotion of flowering in that plants should experience long-day conditions to initiate flowering promptly. Both of these clock-mediated coincidence mechanisms may coordinately confer ecological fitness to plants growing in natural habitats with varied photoperiods.

Introduction

A major focus in the field of plant biology is the mechanism by which the circadian clock generates rhythms with a period close to 24 h. In fact, the expression of numerous clock-controlled genes oscillates diurnally at the level of transcription in the model plant Arabidopsis thaliana (Covington et al. 2008, Michael et al. 2008). More specifically, over half of expressed genes are diurnally regulated (Michael et al. 2008), and/or nearly one-third of expressed genes are circadian regulated in seedlings (Covington et al. 2008). This fact implies that the circadian clock plays roles in a wide range of physiological processes. During the last decade, a number of Arabidopsis clock-associated protein components have been identified, providing unprecedented general insights into the plant clock system (McClung 2006, McClung 2008). The representative clock components are a pair of homologous transcription factors, CIRCADIAN CLOCK-ASSOCIATED 1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) (Schaffer et al. 1998, Wang and Tobin 1998, Mizoguchi et al. 2002), and five members of the small PSEUDO-RESPONSE REGULATOR (PRR) family consisting of PRR9, PRR7, PRR5, PRR3 and PRR1/TIMING OF CAB2 EXPRESSION 1 (TOC1) (Makino et al. 2000, Matsushika et al. 2000, Strayer et al. 2000). In addition to an initial single feedback loop model involving CCA1/LHY and TOC1 (Alabadi et al. 2001), several improved versions of interlocking multiloop clock models are currently envisaged by taking many other clock components into consideration (Gardner et al. 2006, McClung 2006, McClung 2008, Mizuno and Nakamichi 2005). Computational models were also proposed to simulate the expression patterns of some clock-controlled genes successfully (Locke et al. 2006, Zeilinger et al. 2006).

Most of those current clock-oriented experiments in plants were carried out under continuous light (or dark) conditions, in order to understand the characteristics of free-running rhythms in the absence of an external time cue. However, the circadian clock is implicated in a wide range of biological oscillations in plants growing in natural habitats, including movement of organs such as leaves and petals, opening of stomata, and photosynthetic activities in response to diurnal changes under ambient light/dark cycles (Mas 2005, Yakir et al. 2006, Hotta et al. 2007). Hence, it is crucial to understand how the biological clock regulates downstream output events under natural light/dark cycles.

In this context, there has been the best-characterized precedent, namely the clock-mediated seasonal control of flowering time. Arabidopsis thaliana is a facultative long-day plant, meaning that it flowers promptly in response to a long-day photoperiod (Yanovsky and Kay 2003, Searle and Coupland 2004). Clock-defective mutants almost exclusively display an altered phenotype of early or late flowering time, implying a key role for the circadian clock in the photoperiodic control of flowering time (see Table 1). The so-called photoperiodic flowering pathway has been well characterized, in which the key player is the CONSTANS (CO) gene (Imaizumi and Kay 2006). The expression of CO is under the control of the circadian clock, and shows biphasic expression peaks both in the late daytime and at night-time in long days. Coincidentally, CO protein is stabilized in the presence of external light, and actively promotes the transcription of FLOWERING LOCUS T (FT), the protein product of which moves to the shoot apical meristem and induces differentiation of the floral meristem (Corbesier and Coupland 2006, Kobayashi and Weigel 2007). Hence, the essence of the photoperiodic flowering pathway is best explained by coincidence between internal (circadian clock) and external (seasonal photoperiod) time cues.

Table 1

Arabidopsis plants used in this study

Mutants and transgenic linesa Period of rhythms (continuous light) Hypocotyl length (red light) Flowering (long or short days) References 
Wild type (Col)    Laboratory stock 
toc1-2 Short Long Early Mas et al. (2003
prr9-10 prr7-11 Very long Long Late Nakamichi et al. (2005
prr7-11 prr5-11 Very short Long Late Nakamichi et al. (2007
prr9-10 prr7-11 prr5-11 Arrhythmic Long Late Nakamichi et al. (2007
prr7-11 prr5-11 phyB-9 Very short Long NDb Ito et al. (2007
cca1-1 lhy-11 Very short Short Early Mizoguchi et al. (2002 
phyB-9 Long in red light Long Early Reed et al. (1994
pif4-101 pif5-1 No phenotype Short NDb de Lucas et al. (2008
phyB-9 pif4-101 pif4-1 NDb See Fig. 6NDb de Lucas et al. (2008
35S::TOC1ox (PRR1ox) Arrhythmic Short NDb Makino et al. (2002
35S::CCA1ox Arrhythmic Long Late Wang and Tobin (1998), Matsushika et al. (2002
35S::PIF4ox No phenotype Long Early Fujimori et al. (2004
35S::PIF5ox No phenotype Long Early Fujimori et al. (2004
35S::PIF5-HAox NDb Long Early This study 
Mutants and transgenic linesa Period of rhythms (continuous light) Hypocotyl length (red light) Flowering (long or short days) References 
Wild type (Col)    Laboratory stock 
toc1-2 Short Long Early Mas et al. (2003
prr9-10 prr7-11 Very long Long Late Nakamichi et al. (2005
prr7-11 prr5-11 Very short Long Late Nakamichi et al. (2007
prr9-10 prr7-11 prr5-11 Arrhythmic Long Late Nakamichi et al. (2007
prr7-11 prr5-11 phyB-9 Very short Long NDb Ito et al. (2007
cca1-1 lhy-11 Very short Short Early Mizoguchi et al. (2002 
phyB-9 Long in red light Long Early Reed et al. (1994
pif4-101 pif5-1 No phenotype Short NDb de Lucas et al. (2008
phyB-9 pif4-101 pif4-1 NDb See Fig. 6NDb de Lucas et al. (2008
35S::TOC1ox (PRR1ox) Arrhythmic Short NDb Makino et al. (2002
35S::CCA1ox Arrhythmic Long Late Wang and Tobin (1998), Matsushika et al. (2002
35S::PIF4ox No phenotype Long Early Fujimori et al. (2004
35S::PIF5ox No phenotype Long Early Fujimori et al. (2004
35S::PIF5-HAox NDb Long Early This study 

aThe mutants and transgenic lines used in this study are all in the Col background (Ito et al. 2007, de Lucas et al. 2008).

bThe phenotypes have not yet been determined.

In A. thaliana, the mechanisms underlying the regulation of hypocotyl elongation have been documented comprehensively in the literature, in which the light- and/or phytohormone-mediated signaling pathways have mainly been emphasized (Vandenbussche et al. 2005, Nozue and Maloof 2006, Alabadi and Blazquez 2009, and references therein). In this respect, another hallmark phenotype common to many circadian-defective mutants is abnormally long or short hypocotyls (Ito et al. 2007, see Table 1). Since this phenotype was originally observed under continuous red and/or blue light, it was not clear if there was a direct connection between circadian rhythm and altered hypocotyl length (Huq et al. 2000, Sato et al. 2002, Mas et al. 2003). Based on the current consistent background, introduced below, here we explore the molecular mechanisms by which the circadian clock regulates the elongation of hypocotyls, in the hope of addressing the issue of how the circadian clock system controls downstream output pathways under natural light/dark cycles.

The most recent progress that is closely relevant to the issue of this study is summarized briefly, as follows. (i) The major red light photoreceptor phytochrome B (phyB) plays a prominent role in the regulation of hypocotyl elongation in response to light (Quail 2002). The light-activated (Pfr) form of phyB interacts directly with its downstream target proteins, PHYTOCHROME INTERACTING FACTOR 4 and 5 (PIF4 and PIF5) (Huq and Quail 2002, Khanna et al. 2004). PIF5 is also known as PIL6 (PIF-LIKE 6) (Fujimori et al. 2004). They are homologous basic helix–loop–helix (bHLH) transcription factors that promote elongation of hypocotyls (Fujimori et al. 2004, Duek and Fankhauser 2005, Monte et al. 2007). The light-activated phyB induces protein degradation of PIF4 and PIF5, thereby inhibiting elongation of hypocotyls in the presence of light. (ii) Among many plant hormones that are involved in hypocotyl elongation of seedlings, the light-regulated effect of gibberellin on hypocotyl elongation is prominent (Vandenbusssche et al. 2005, Nozue and Maloof 2006, Newhauser 2008). There is a direct cross-regulation between the light–phyB–PIF4/5 signaling pathway and the light–gibberellin action on hypocotyl elongation during early photomorphogenesis (Djakovic-Petrovic et al. 2007, Achard et al. 2008, Alabadi et al. 2008, de Lucas et al. 2008, Feng et. 2008). The gibberellin-dependent modulators (GAI, RGA, RGL1, RGL2 and RGL3, collectively designated as DELLA) are capable of interacting directly with the PIF4 and PIF5 proteins, in such a manner that DELLA inhibits their activities (for a recent review, see Davier et al. 2008). Together with the fact that gibberellin induces the protein degradation of DELLA in a manner dependent on the light signal (Alabadi et al. 2008), it is postulated that this phytohormone promotes hypocotyl elongation of seedlings through activating the PIF4 and PIF5 proteins. (iii) Finally, it should also be mentioned that Nozue et al. (2007) described a phenomenon of rhythmic hypocotyl elongation of young Arabidopsis seedlings (for a review, see Breton and Kay 2007). They found that hypocotyl elongation occurs diurnally at the end of every night. They also found that the circadian clock regulates the transcription of PIF4 and PIF5, so as to be coincidentally enhanced at that time. Taken together, it was currently believed that the PIF4 and PIF5 proteins are capable of integrating the light, hormone (gibberellin) and clock signals to regulate hypocotyl elongation properly during early photomorphogenesis.

Based on the current background with regard to the regulation of hypocotyl elongation, here we provide compelling genetic evidence to propose the novel view that the circadian clock regulates not only the daily response but also the seasonal response of hypocotyl elongation. At the molecular level, we show that the circadian clock modulates the expression profiles of PIF4 and PIF5 in such a manner that certain lengths of dark period are necessary to enhance the expression of PIF4 and PIF5 during the night-time. Hence, the photoperiodic control of hypocotyl elongation was best explained by the stable accumulation of the proteins PIF4 and PIF5 during the night-time, due to coincidence between the internal (circadian rhythm) and external (photoperiod) time cues specifically in short days. In general, these proposed views of this study substantially extend our knowledge with regard to the circadian clock-controlled output pathways. The physiological and/or ecological relevance of our views as to clock-controlled plant growth will also be discussed.

Results

Hypocotyl elongation of seedlings under light/dark cycles

Arabidopsis seedlings (accession Columbia, Col) were grown on MS gellan gum plates under 24 h light/dark (LD) cycles with varied photoperiods, including continuous light (LL) and continuous dark (DD) conditions. After incubation for 3, 5 and 7 days after germination (DAG), hypocotyl lengths were measured and plotted against the length of the dark period experienced by the plants (Fig. 1A). When grown under 3 h dark/21 h light (3D/21L) cycles (or long-day conditions), hypocotyls were short and similar to those of seedlings grown in LL. When grown under 21D/3L cycles (or short-day conditions), conversely, the hypocotyls were long and comparable with those of etiolated seedlings in DD. We considered that if the dark periods were the sole determinant of the hypocotyl length of seedlings, the observed responsive curve would (or should) be linearly proportional to altered dark periods (Fig. 1A). However, the fact was that the resulting response curve of hypocotyl elongation was non-linear. This curious phenomenon was commonly observed for other Arabidopsis accessions, Landsberg erecta (Ler), C24 and Wassilewskija (WS) (Fig. 1B). We assumed that the acceleration of hypocotyl elongation is a short-day-specific event. This is reminiscent of the circadian clock-mediated photoperiodic control of flowering time, which is a long-day-specific event (see Introduction; also see Pouteau et al. 2008). In this study, we wanted to understand the molecular mechanism underlying this curious regulation of hypocotyl elongation.

Fig. 1

Photoperiodic response of hypocotyl elongation in Arabidopsis seedlings. (A) Wild-type Columbia (Col) seedlings were grown on MS gellan gum plates containing sucrose (1%) in the following photoperiodic conditions: continuous light (LL), 3 h dark (D)/21 h light (L), 6D/18L, 9D/15L, 12D/12L, 15D/9L, 18D/6L, 21D/3L and continuous darkness (DD). The light intensity used was 100 μmol m–2 s–1. At 3 days after germination (3 DAG), 5 DAG and 7 DAG, hypocotyl lengths of seedlings were measured (n > 20), and the averaged values with the SD were plotted against the photoperiod applied. Inset at bottom: representative seedlings in a given photoperiod were photographed (5 DAG). To emphasize the intriguing results of this experiment, a simple linear response of hypocotyl elongation to the dark period was schematically illustrated. (B) Essentially the same experiments were concomitantly repeated with other Arabidopsis accessions, namely Col, Ler (Landsberg erecta), C24 and Ws (Wassilewskija), as indicated. Hypocotyl lengths were analyzed at 5 DAG. (C) Essentially the same experiments were carried out with Col, Ler and C24, with different combinations of light intensities and photoperiods, as indicated. Hypocotyl lengths were analyzed at 5 DAG. (D) Photographs were taken for seedlings (7 DAG), grown for the indicated photoperiods, in order to show the characteristic morphologies.

Fig. 1

Photoperiodic response of hypocotyl elongation in Arabidopsis seedlings. (A) Wild-type Columbia (Col) seedlings were grown on MS gellan gum plates containing sucrose (1%) in the following photoperiodic conditions: continuous light (LL), 3 h dark (D)/21 h light (L), 6D/18L, 9D/15L, 12D/12L, 15D/9L, 18D/6L, 21D/3L and continuous darkness (DD). The light intensity used was 100 μmol m–2 s–1. At 3 days after germination (3 DAG), 5 DAG and 7 DAG, hypocotyl lengths of seedlings were measured (n > 20), and the averaged values with the SD were plotted against the photoperiod applied. Inset at bottom: representative seedlings in a given photoperiod were photographed (5 DAG). To emphasize the intriguing results of this experiment, a simple linear response of hypocotyl elongation to the dark period was schematically illustrated. (B) Essentially the same experiments were concomitantly repeated with other Arabidopsis accessions, namely Col, Ler (Landsberg erecta), C24 and Ws (Wassilewskija), as indicated. Hypocotyl lengths were analyzed at 5 DAG. (C) Essentially the same experiments were carried out with Col, Ler and C24, with different combinations of light intensities and photoperiods, as indicated. Hypocotyl lengths were analyzed at 5 DAG. (D) Photographs were taken for seedlings (7 DAG), grown for the indicated photoperiods, in order to show the characteristic morphologies.

Photoperiodic response of hypocotyl elongation

It was first suspected that hypocotyl elongation might be regulated merely by the total intensity of light applied within a day, and the effective light intensities may have already saturated the maximal threshold under certain long-day conditions. Alternatively, the determinant of hypocotyl elongation may be the photoperiod itself. To test these possibilities, hypocotyl elongation was measured under varied light intensities and photoperiods (Fig. 1C). The results indicated that photoperiods, but not total light intensities, were the determinant of the response curve. This conclusion was confirmed by replicated experiments with different light intensities and photoperiods (Supplementary Fig. S1). We tentatively termed this phenomenon ‘photoperiodic response of hypocotyl elongation’. In these experiments, we also observed cotyledon expansion and chloroplast development (or greening) (Fig. 1D). Characteristic differences were evident in overall morphologies of seedlings with cotyledons and juvenile leaves, depending on the photoperiod applied. However, the traits appeared to be too complex to deal with quantitatively in this study. For simplicity, hereafter we focused mainly on the molecular basis of the photoperiodic response of hypocotyl elongation.

PhyB-mediated pathway in the photoperiodic response of hypocotyl elongation

Based on current knowledge about the early photomorphogenesis of Arabidopsis seedlings (see Introduction), some experiments were carried out in order to gain preliminary insights into the molecular components implicated in the photoperiodic response of hypocotyl elongation. We first employed a loss-of-function mutant of the red light photoreceptor phyB, which plays a prominent role in the regulation of hypocotyl elongation in response to red light. The phyB-9 mutant seedlings had longer hypocotyls than the wild-type seedlings in every photoperiod tested, with the exception of DD (Fig. 2A). In contrast to the wild-type seedlings, the hypocotyl lengths of phyB-9 seedlings appeared to be rather proportional to the dark period, suggesting that PHYB-dependent light signal transduction is implicated in the mechanism underlying the photoperiodic response of hypocotyl elongation.

Fig. 2

Characterization of the photoperiodic response of hypocotyl elongation with a set of Arabidopsis mutant and transgenic lines. (A and B) Analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG by employing the phyB-9 loss-of-function mutant (A), and both the PIF4ox and PIF5ox transgenic lines (B). (C) Analyses of the photoperiodic response of hypocotyl elongation were carried out with Col on MS gellan gum plates containing gibberellin (GA3). Hypocotyl lengths were analyzed at 3 DAG. Other details were the same as those given in the legend to Fig. 1.

Fig. 2

Characterization of the photoperiodic response of hypocotyl elongation with a set of Arabidopsis mutant and transgenic lines. (A and B) Analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG by employing the phyB-9 loss-of-function mutant (A), and both the PIF4ox and PIF5ox transgenic lines (B). (C) Analyses of the photoperiodic response of hypocotyl elongation were carried out with Col on MS gellan gum plates containing gibberellin (GA3). Hypocotyl lengths were analyzed at 3 DAG. Other details were the same as those given in the legend to Fig. 1.

According to current understanding, the light-activated (Pfr) form of phyB interacts directly with the downstream target proteins PIF4 and PIF5, both of which act as transcriptional factors that promote elongation of hypocotyls (Fujimori et al. 2004, Duek and Fankhauser 2005, Monte et al. 2007). Therefore, the impact of PIF4 and PIF5 on hypocotyl elongation was confirmed with transgenic lines overexpressing the PIF4 and PIF5 transcripts (Fig. 2B). Seedlings of these transgenic lines (PIF4ox and PIF5ox) showed markedly elongated hypocotyls in every photoperiod, supporting the current idea that PIF4 and PIF5 serve as positive regulators for hypocotyl elongation.

Gibberellin-mediated pathway in the photoperiodic response of hypocotyl elongation

Among plant hormones important for hypocotyl elongation of seedlings, we specifically focused on gibberellin because there is a direct link between the light–phyB–PIF4/5 signaling pathway and the action of gibberellin on hypocotyl elongation (see Introduction). In fact, the gibberellin-dependent modulators (or DELLA) are thought to interact directly with PIF4 and PIF5 to inhibit their activity. Together with the fact that gibberellin induces the protein degradation of DELLA, it also promotes hypocotyl elongation of seedlings through activation of PIF4 and PIF5 (Alabadi et al. 2008, de Lucas et al. 2008, Feng et al. 2008, for a review, see also Davier et al. 2008). As expected, gibberellin promoted hypocotyl elongation of seedlings in every photoperiod tested (Fig. 2C). In contrast to the phyB-9 photoreceptor mutant, the photoperiod-dependent profile was still evident in the gibberellin-treated seedlings. These results suggest that the GA–DELLA–PIF4/5 pathway regulates hypocotyl elongation in a manner parallel to the phyB–PIF4/5 pathway.

Preliminary insights into the mechanism underlying the photoperiodic regulation of hypocotyl elongation

The above results provided us with preliminary insights into the mechanism underlying the photoperiodic regulation of hypocotyl elongation, as follows. (i) The light–phyB–PIF4/5 pathway plays a role in the photoperiodic control of hypocotyl elongation (Fig. 2A). (ii) However, this light–phyB pathway at the post-transcriptional PIF4/5 level is not sufficient to explain the phenomenon in question fully, because it is known that the expression of PIF4 and PIF5 fluctuates markedly diurnally at the transcriptional level (Fujimori et al. 2004, see also the microarray database at http://diurnal.cgrb.oregonstate.edu). (iii) Indeed, this diurnal regulation of PIF4 and PIF5 at the level of transcription appears to be crucial for the photoperiodic response of phyocotyl elongation, because the constitutive transcription of PIF4 and PIF5 resulted in a complete abolishment of the non-linear response curve (Fig. 2B). (iv) In this respect, phyB does not directly affect the diurnal transcription profiles (or amplitudes) of PIF4 and PIF5 (see the transcription profiles of PIF4 and PIF5 in the phyB-9 mutant in the database at http://diurnal.cgrb.oregonstate.edu/). Taken together, it was reasonably deduced that there must be another critical determinant(s) which controls the PIF4 and PIF5 activities in response to the photoperiod in a non-linear manner at the level of transcription. (v) In this regard, the possible candidate gibberellin was dismissed (Fig. 2C). Hence, the most probable candidate that could be presumed to have such a sophisticated role is the circadian clock.

Photoperiodic response of hypocotyl elongation in clock-defective mutants

To test if the circadian clock is implicated in the photoperiodic control of hypocotyl elongation, the most rational and direct approach is to characterize certain mutant seedlings lacking the clock function. Here, we employed a large set of clock-defective mutants or transgenic lines (Table 1), which are all in the unified Col background (Ito et al. 2007). First, the toc1-2 single and cca1-1 lhy-11double mutants were examined (Fig. 3A). In the toc1-2 seedlings, the photoperiodic response curve was altered in such a way that its effective threshold of the photoperiod was shifted towards a shorter dark period. In contrast, the photoperiodic response curve in the cca1-1 lhy-11 mutant was altered in such a way that its effective threshold of the photoperiod was shifted toward a longer dark period. These results are consistent with the idea that the circadian clock plays a role in the photoperiodic response of hypocotyl elongation, at least in part.

Fig. 3

Characterization of the photoperiodic response of hypocotyl elongation with a set of clock-defective mutants and transgenic lines. Analyses of the photoperiodic response of hypocotyl elongation were carried out at 3 DAG by employing a set of clock-defective mutant and transgenic lines, as indicated. Seedlings characterized are: (A) both the toc1 single and cca1 lhy double mutants; (B) both the PIF4ox and PIF5ox transgenic lines; and (C) the prr9-10 prr7-11 prr5-11 triple mutant. Other details were the same as those given in the legend to Fig. 1.

Fig. 3

Characterization of the photoperiodic response of hypocotyl elongation with a set of clock-defective mutants and transgenic lines. Analyses of the photoperiodic response of hypocotyl elongation were carried out at 3 DAG by employing a set of clock-defective mutant and transgenic lines, as indicated. Seedlings characterized are: (A) both the toc1 single and cca1 lhy double mutants; (B) both the PIF4ox and PIF5ox transgenic lines; and (C) the prr9-10 prr7-11 prr5-11 triple mutant. Other details were the same as those given in the legend to Fig. 1.

If the circadian clock does indeed play a role, certain clock-defective seedlings should show a markedly altered (or linear) photoperiodic response curve (see Fig. 1A). To test this, we characterized the CCA1ox and TOC1ox transgenic lines, which overexpress the CCA1 and TOC1 transcripts, respectively. These clock-defective seedlings exhibited considerably altered photoperiod response curves (Fig. 3B). Notably, the profiles observed for CCA1ox and TOC1ox contrasted with those observed for, respectively, the cca1-1 lhy-11 double and toc1-2 single mutants (see Fig. 3A). We further employed the clock-defective prr9-10 prr7-11 prr5-11 triple mutant (Fig. 3C). The clock function of this triple mutant is severely defective in the sense that this mutant shows an arrhythmia phenotype even under light/dark and hot/cold cycles (Yamashino et al. 2008). Like the CCA1ox seedlings, the prr9-10 prr7-11 prr5-11 seedlings also showed longer hypocotyls in proportion to the dark period. From these results, we concluded that the circadian clock is involved in the photoperiodic response of hypocotyl elongation.

Furthermore, it is worth mentioning that the promotion of hypocotyl elongation in these clock mutants was not observed when they were examined in LL (Fig. 3). This is in sharp contrast to the results observed for the phyB-9 mutant, the PIF4ox transgenic line and the gibberellin-treated seedlings, in which accelerated hypocotyl elongation was evident even in LL (Fig. 2). This suggests that the light/dark cycle is a cue for the photoperiodic response of hypocotyl elongation, and this is consistent with the view that the entrained circadian clock is involved in the event.

Photoperiodic response of hypocotyl elongation and clock-controlled expression of PIF4 and PIF5

The next question was: how does the circadian clock regulate the hypocotyl elongation in response to the photoperiod in a non-linear fashion? As reported previously, the circadian clock controls the diurnal expression of PIF4 and PIF5 in such a way that the transcripts exhibit rhythmic peaks every morning (Fujimori et al. 2004). Therefore, we examined the diurnal expression profiles of PIF4 and PIF5 in the clock-defective mutants, compared with the wild type. In CCA1ox, PIF4 and PIF5 were expressed at higher levels throughout the day and night, if not constitutively (Fig. 4A). We also employed the prr9-10 prr7-11 double and prr9-10 prr7-11 prr5-11 triple mutants (Table 1). In both the mutants, the altered and linear photoperiodic response patterns were confirmed (Fig. 4B, C, upper panels). In these mutants, the expression levels of PIF4 and PIF5 were highly constitutive even during the night-time (Fig. 4B, C, lower panels). Taken together, we hypothesized that the circadian clock controls the photoperiodic response of hypocotyl elongation by modulating the rhythmic expression profiles of PIF4 and PIF5.

Fig. 4

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profiles of PIF4 and PIF5. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with the CCA1ox transgenic line at 5 DAG. Both the Col and CCA1ox plants were grown in the 14 h dark/10 h light photoperiod conditions for 10 d, and then RNA samples were prepared for every 3 h interval. They were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. The qRT-PCR data were the averages with the SD of three experimental replicates, and they were plotted against the timings of sampling (gray rectangle, dark period; open rectangle, light period). (B) Essentially the same analyses were carried out with the prr9-10 prr7-11 double mutant. (C) Essentially the same analyses were also carried out with the prr9-10 prr7-11 prr5-11 triple mutant. Other details were given in Materials and Methods.

Fig. 4

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profiles of PIF4 and PIF5. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with the CCA1ox transgenic line at 5 DAG. Both the Col and CCA1ox plants were grown in the 14 h dark/10 h light photoperiod conditions for 10 d, and then RNA samples were prepared for every 3 h interval. They were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. The qRT-PCR data were the averages with the SD of three experimental replicates, and they were plotted against the timings of sampling (gray rectangle, dark period; open rectangle, light period). (B) Essentially the same analyses were carried out with the prr9-10 prr7-11 double mutant. (C) Essentially the same analyses were also carried out with the prr9-10 prr7-11 prr5-11 triple mutant. Other details were given in Materials and Methods.

Examination of diurnal expression profiles of PIF4 and PIF5 under different photoperiodic conditions

We needed to verify the above hypothetical view in the wild-type background (Fig. 5A). To this end, diurnal expression profiles of PIF4 and PIF5 were examined in Col seedlings, grown for 15 d in 5D/19L (Fig. 5B), 14D/10L (Fig. 5C) and 19D/5L (Fig. 5D). The toc1-2 seedlings were also analyzed in 14D/10L to compare them with the results from Col (Fig. 5C). When these results were compared, it was revealed that the expression levels of PIF4 and PIF5 in Col fluctuate during the dark period such that higher expression levels become more pronounced when seedlings were grown in a longer dark period. Coincidentally, the seedlings showed longer hypocotyls in proportion to the levels of PIF4 and PIF5 during the night-time. Also, the night-time expression levels of PIF4 and PIF5 were significantly elevated in toc1-2 in 14D/10L; under these conditions, the mutant exhibited considerably longer hypocotyls than Col (Fig. 5A). These results were consistent with the view that the circadian clock controls the photoperiodic response of hypocotyl elongation by modulating the rhythmic expression patterns of PIF4 and PIF5.

Fig. 5

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profies of PIF4 and PIF5 in the wild-type seedlings. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with Col and toc1-2 at 5 DAG. (B–D) Col seedlings were grown under the following conditions, 5D/19L for 14 d, 14D/10L for 10 d and 19D/5L for 14 d, respectively. toc1-2 seedlings were also grown in 14D/10L for 10 d concomitantly with Col. From these plants, RNA samples were prepared, and they were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. Other details were the same as those given in the legend to Fig. 5.

Fig. 5

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profies of PIF4 and PIF5 in the wild-type seedlings. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with Col and toc1-2 at 5 DAG. (B–D) Col seedlings were grown under the following conditions, 5D/19L for 14 d, 14D/10L for 10 d and 19D/5L for 14 d, respectively. toc1-2 seedlings were also grown in 14D/10L for 10 d concomitantly with Col. From these plants, RNA samples were prepared, and they were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. Other details were the same as those given in the legend to Fig. 5.

General consideration as to the mechanism underlying the photoperiodic response of hypocotyl elongation

The results of this study taken together (Figs. 1–5) allowed us to deduce a signaling framework underlying the photoperiodic control of hypocotyl elongation (Fig. 6A). According to this framework, the photoperiod-dependent clock pathway, the light-induced phyB pathway and the gibberellin-mediated DELLA pathway are all converged onto the same hub proteins, PIF4 and PIF5. However, these pathways each modulate PIF4 and PIF5 activities through a different mode of action (denoted by a, b and c in Fig. 6A). The gibberellin–DELLA pathway regulates PIF4 and PIF5 activity through direct protein–protein interaction (Djakovic-Petrovic et al. 2007, Alabadi et al. 2008, de Lucas et al. 2008). The phyB pathway triggers degradation of PIF4 and PIF5 by binding directly to them in response to light (Huq and Quail 2002, Khanna et al. 2004, Monte et al. 2007). Finally, the clock pathway controls the diurnal expression profiles of PIF4 and PIF5 at the level of transcription in a manner dependent on the photoperiod, as demonstrated in this study. In this last respect, it should be recalled that the periodic light/dark cycle (not the duration of the dark period) is an important cue for the circadian clock-controlled hypocotyl elongation (see Figs. 2, 3). This is consistent with the general concept that the periodic light/dark cycles with 24 h intervals are necessary to make it possible for the circadian clock to keep entraining to local and seasonal times (McClung 2006).

Fig. 6

A proposed framework of signal transduction pathways involved in the photoperiodic control of hypocotyl elongation. (A) The clock-controlled, the phyB-regulated and the gibberellin–DELLA-mediated pathways converge onto the same hub proteins PIF4 and PIF5, which are known to promote hypocotyl elongation. Arrows indicate the positive effects on downstream components, whereas T-bars indicated negative effects. Gray lines indicate the existence of functional interactions, regardless of positive or negative effects. The photoperiodic control of hypocotyl elongation is best explained by the stable accumulation of PIF4 and PIF5 during the night-time of short days, due to their degradation by light. Other details were given in the text. (B) Wild-type (Col), phyB-9 and phyB-9 pif4-101 pif5-1 seedlings (5 DAG), grown under 12L/12D conditions, were characterized with reference to their hypocotyl lengths. Representative seedlings were photographed. Scale bar, 5 mm. (C) Similarly, wild-type (Col) and pif4-101 pif5-1 seedlings (5 DAG) were characterized with reference to their hypocotyl lengths, after being grown in the presence and absence of gibberellin (50 μM) under 12L/12D conditions. Scale bar, 5 mm. (D) Analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG with a set of mutant and transgenic lines, as indicated. Other details were the same as those given in the legend to Fig. 1.

Fig. 6

A proposed framework of signal transduction pathways involved in the photoperiodic control of hypocotyl elongation. (A) The clock-controlled, the phyB-regulated and the gibberellin–DELLA-mediated pathways converge onto the same hub proteins PIF4 and PIF5, which are known to promote hypocotyl elongation. Arrows indicate the positive effects on downstream components, whereas T-bars indicated negative effects. Gray lines indicate the existence of functional interactions, regardless of positive or negative effects. The photoperiodic control of hypocotyl elongation is best explained by the stable accumulation of PIF4 and PIF5 during the night-time of short days, due to their degradation by light. Other details were given in the text. (B) Wild-type (Col), phyB-9 and phyB-9 pif4-101 pif5-1 seedlings (5 DAG), grown under 12L/12D conditions, were characterized with reference to their hypocotyl lengths. Representative seedlings were photographed. Scale bar, 5 mm. (C) Similarly, wild-type (Col) and pif4-101 pif5-1 seedlings (5 DAG) were characterized with reference to their hypocotyl lengths, after being grown in the presence and absence of gibberellin (50 μM) under 12L/12D conditions. Scale bar, 5 mm. (D) Analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG with a set of mutant and transgenic lines, as indicated. Other details were the same as those given in the legend to Fig. 1.

Specific consideration as to the mechanism underlying the photoperiodic response of hypocotyl elongation

Obviously, the proposed network (Fig. 6A) is only a part of the highly sophisticated circuitry that is responsible for the complicated processes of early photomorphogenesis. For instance, the proposed framework does not include the well-characterized CONSTITUTIVE PHOTMORPHOGENIC 1–ELONGATED HYPOCOTYL 5 (COP1–HY5) pathway (Newhauser 2008). Hence, we needed to test whether or not the proposed framework is central to the photoperiodic response of hypocotyl elongation, as follows.

We first evaluated the light–phyB–PIF4/PIF5 pathway by employing a phyB-9 pif4-101 pif5-1 triple loss-of-function mutant (Table 1). Hypocotyl lengths of the mutant seedlings were examined under the 12L/12D conditions in comparison with those of the phyB single mutant (Fig. 6B). The results were consistent with the current view that the light–phyB-mediated hypocotyl elongation is partly dependent on the functions of PIF4 and PIF5 (Fujimori et al. 2004, Duek and Fankhauser 2005, Monte et al. 2007). We then evaluated the light–gibberellin–PIF4/PIF5 pathway by employing a pif4-101 pif5-1 double mutant. The mutant seedlings were examined in the presence and absence of gibberellin under the 12L/12D conditions (Fig. 6C). The results were also consistent with the current view that the light–gibberellin-mediated hypocotyl elongation is dependent on the functions of PIF4 and PIF5 (Alabadi et al. 2008, Davier et al. 2008, de Lucas et al. 2008, Feng et al. 2008). Nevertheless, these results also indicated that the phyB and gibberellin pathways are not absolutely dependent on the functions of PIF4 and PIF5, because both the pif4-101 and pif5-1 lesions were not fully epistatic over the phyB mutant and the effect of gibberellin application. It was thus suggested that as yet unidentified factors (Xs) appear to be implicated downstream, as also integrated into Fig. 6A.

The circadian clock regulates the photoperiodic regulation of hypocotyl elongation in a manner independent of the phyB–PIF4/5 pathway

To verify the importance of the clock-mediated pathway, a prr7-11 prr5-11 phyB-9 triple mutant was examined (Fig. 6D, left). It was shown that both the prr7-11/prr5-11 mutations affect the hypocotyl elongation in an additive manner to the phyB-9 mutation. This result indicated that the clock-mediated pathway plays a critical role in the regulation of hypocotyl elongation in a manner additive to, and independent from, the phyB-mediated pathway. Furthermore, the prr7-11 prr5-11 phyB-9 triple mutant was examined in the presence of gibberellin. Under the experimental conditions, since all of the three negative pathways (i.e. a, b and c) must largely be canceled out, the resulting hypocotyls were expected to be extremely long in any photoperiod, and this was indeed the case (Fig. 6D, right). We also examined the hypocotyl length of PIF4ox seedlings in the presence of gibberellin (Fig. 6D, right). The lengths of the PIF4ox hypocotyls in LL were extremely long, like those of etiolated seedlings in DD, in a manner almost independent of the photoperiod (see also Supplementary Fig. S2). Taken together, these results strongly supported the idea that the inferred framework in Fig. 6A is central to the photoperiodic regulation of hypocotyl elongation, in which the circadian clock plays a crucial role.

Evidence for a clock-mediated coincident mechanism underlying the photoperiodic response of hypocotyl elongation

Based on the compiled data and interpretation mentioned above, the most plausible scenario for the photoperiodic control of hypocotyl elongation was hypothesized. As schematically illustrated in Fig. 7, regardless of photoperiodic conditions, and also regardless of PIF4 and PIF5 mRNA abundance, it is assumed that PIF4 and PIF5 proteins are degraded during the daytime through the phyB pathway. This light-dependent degradation of PIF5 in plants was confirmed in this study with PIF5-HAox seedling (see the inset in Fig. 6A). In addition, residual PIF4 and PIF5 activity must also be inhibited through interaction with DELLA in the presence of light (or in the absence of gibberellin). Hence, hypocotyl elongation is attenuated whenever seedlings are in the presence of light. In this respect, the circadian clock closes the gate for the transcription of PIF4 and PIF5 during the night in long days, as demonstrated (Fig. 5B), and as illustrated (Fig. 7, upper left). In short days, however, the clock opens the gate at the end of every night, so that PIF4 and PIF5 mRNA is accumulated in the dark, as demonstrated (Fig. 5D), and as illustrated (Fig. 7, upper right). Consequently, PIF4 and PIF5 proteins accumulate specifically in short days because the active forms of phyB and DELLA are not present coincidentally at that time, thereby resulting in a coincident elongation of hypocotyls. In short, the PIF4/PIF5-dependent photoperiodic response of hypocotyl elongation is best explained by coincidence between the clock-controlled internal and the photoperiod-mediated external time cues.

Fig. 7

A proposed coincident model to explain the molecular mechanism underlying the photoperiodic control of hypocotyl elongation. The scenario is schematically depicted, and details were explained in the text. It may also be worth mentioning that Nozue et al. (2007) reported the diurnal rhythmic elongation of hypocotyls during the late night under short-day conditions. This diurnal event appears to be a facet of the proposed mechanism, which manifests particularly in a short-day-specific manner, as indicated. To investigate the involvement of the function of PIF4 and PIF5 proposed above in the photoperiodic control of hypocotyl elongation, analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG by employing the pif4-101 pif5-1 double loss-of-function mutant.

Fig. 7

A proposed coincident model to explain the molecular mechanism underlying the photoperiodic control of hypocotyl elongation. The scenario is schematically depicted, and details were explained in the text. It may also be worth mentioning that Nozue et al. (2007) reported the diurnal rhythmic elongation of hypocotyls during the late night under short-day conditions. This diurnal event appears to be a facet of the proposed mechanism, which manifests particularly in a short-day-specific manner, as indicated. To investigate the involvement of the function of PIF4 and PIF5 proposed above in the photoperiodic control of hypocotyl elongation, analyses of the photoperiodic response of hypocotyl elongation were carried out at 5 DAG by employing the pif4-101 pif5-1 double loss-of-function mutant.

We evaluated this proposed model experimentally by employing the clock-defective prr7-11 prr5-11 mutant, together with the pif4-101 pif5-1 double mutant. According to the model, the photoperiod-dependent gating effect is severely perturbed in certain prr multiple mutants, so that the PIF4 and PIF5 transcripts are highly (or constitutively) accumulated even during the night-time of long days (see Fig. 4). In such clock-defective mutants, the elongation of hypocotyls is accelerated in a manner simply proportional to the dark period, as demonstrated (Fig. 3, and see the lower part of Fig. 7). Furthermore, the result from the pif4-101 pif5-1 double mutant indicated that the photoperiodic response of hypocotyl elongation is largely, if not absolutely, dependent on the functions of PIF4 and PIF5 (Fig. 7, and Supplementary Fig. S3) (later, see also Fig. 9D). Taking the results of Fig. 6B and C also into consideration, these results suggested that the clock–PIF4/5 pathway is the major determinant for the photoperiodic control of hypocotyl elongation (Fig. 7, lower part). However, it should be noted that the results also suggested that an as yet unidentified factor other than PIF4 and PIF5 most probably plays the redundant role (see also Fig. 6A), because the photoperiodic response of hypocotyl elongation is still observed to some extent even in the pif4-101 pif5-1 double mutant.

Fig. 8

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profiles of PIF4 and PIF5. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with the cca1-1 lhy-11 double mutant at 5 DAG. Both the Col and mutant plants were grown under the 19 h dark/5 h light photoperiod conditions for 14 d, and then RNA samples were prepared for every 3 h interval. They were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. The qRT-PCR data were the averages with the SD of three experimental replicates. (B) Essentially the same analyses were carried out with the TOC1ox transgenic line. For RNA preparation, plants were grown in 14D/10L for 14 d, as indicated. Other details were the same as those given in the legend to Fig. 4

Fig. 8

Characterization of the photoperiodic response of hypocotyl elongation with reference to the diurnal expression profiles of PIF4 and PIF5. (A) As described in Fig. 3, the photoperiodic response of hypocotyl elongation was reproducibly analyzed with the cca1-1 lhy-11 double mutant at 5 DAG. Both the Col and mutant plants were grown under the 19 h dark/5 h light photoperiod conditions for 14 d, and then RNA samples were prepared for every 3 h interval. They were analyzed by qRT-PCR with special reference to the diurnal expression profiles of PIF4 and PIF5. The qRT-PCR data were the averages with the SD of three experimental replicates. (B) Essentially the same analyses were carried out with the TOC1ox transgenic line. For RNA preparation, plants were grown in 14D/10L for 14 d, as indicated. Other details were the same as those given in the legend to Fig. 4

Fig. 9

Photoperiodic response of Arabidopsis plant growth. (A) Wild-type Columbia (Col), prr9-10 prr7-11 prr5-11 triple mutant and CCA1ox plants were grown for 25 d under the indicated photoperiodic conditions. Photographs were taken of representative plants to characterize their photomorphologies. Scale bar (upper right), 1 cm. (B and C) For these plants (n > 10), their hypocotyl lengths were measured, as indicated. (D) A similar examination was carried out for pif4-101 pif5-1 double mutant plants (21-days-old) (pictures of these plants are not shown).

Fig. 9

Photoperiodic response of Arabidopsis plant growth. (A) Wild-type Columbia (Col), prr9-10 prr7-11 prr5-11 triple mutant and CCA1ox plants were grown for 25 d under the indicated photoperiodic conditions. Photographs were taken of representative plants to characterize their photomorphologies. Scale bar (upper right), 1 cm. (B and C) For these plants (n > 10), their hypocotyl lengths were measured, as indicated. (D) A similar examination was carried out for pif4-101 pif5-1 double mutant plants (21-days-old) (pictures of these plants are not shown).

Additional supportive evidence for the clock-mediated photoperiodic response of hypocotyl elongation

We showed earlier that both the cca1 lhy and TOC1ox seedlings exhibited shorter hypocotyls, relative to Col (see Fig. 3). These events were opposite to those observed for CCA1ox and toc1-2. We were challenged by the last and critical argument that the altered photoperiod responses of cca1 lhy and TOC1ox should also be explained by the alteration of PIF4 and PIF5 expression profiles. To test this, the cca1 lhy double mutant was grown in 19D/5L (Fig. 8A), while the TOC1ox line was grown in 14D/10L (Fig. 8B). Under the photoperiodic conditions tested, both of these clock-defective seedlings display shorter hypocotyls, as compared with Col (see upper panels). In these clock-defective seedlings with shorter hypocotyls, the results were consistent with the coincidence model in that the expression levels of PIF4 and PIF5 during the night-time were considerably reduced, relative to Col.

Discussion

The circadian clock regulates hypocotyl elongation of seedlings by measuring altered photoperiods in a short-day-specific manner

In this study, we found that the rates of hypocotyl elongation in seedlings are considerably dependent on the photoperiod in a non-linear manner (Fig. 1). We provided extensive genetic evidence to show that the circadian clock is involved in this photoperiodic (or seasonal) regulation of hypocotyl elongation by modulating the expression profiles of the PIF4 and PIF5 genes (Figs. 3–8). It should be emphasized that this clock-mediated pathway (Fig. 6, pathway a) is independent of the phyB-mediated pathway (Fig. 6, pathway b), at least partly, as clearly shown in Fig. 6B. It appears that the circadian clock plays an essential role to detect short-day time cues that are necessary to enhance the expression of PIF4 and PIF5 during the night-time, thereby promoting the elongation of hypocotyls in a short-day-specific manner. Based on these results, we propose a mechanism by which the elongation of hypocotyls is controlled through the circadian clock. In essence, the PIF4/PIF5-dependent photoperiodic response of hypocotyl elongation is best explained by coincidence between the clock-controlled internal and photoperiodic external time cues, as explained (Fig. 7). According to this coincidence model, the circadian clock properly controls opening/closing of the gate during the dark period for PIF4/5 expression in the wild type. In other words, the wild-type clock normally acts as a gatekeeper in such a manner that the short day conditions are required to open the gate to promote the elongation of hypocotyls. As a result, the wild-type clock regulates the hypocotyl elongation in a short-day-specific non-linear manner (Fig. 1A). In the prr9 prr7 prr5 mutant, the clock is out of order so that the gate opens overnight, resulting in longer hypocotyls in proportion to the dark period (Fig. 3C). In the cca1 lhy mutant, the clock is defective in such a way that a much longer dark period is required to open the gate, resulting in shorter hypocotyls at a given photoperiod (Fig. 8A).

Final verification of this proposed model must await further experimentation, including characterization of PIF4 and PIF5 at the intrinsic protein level. At present, unfortunately, there is no means to do so properly with the wild-type plant (Huq and Quail 2002, Khanna et al. 2004, Nozue et al. 2007, Alabadi et al. 2008, de Lucas et al. 2008, Feng et al. 2008, Lorrain et al. 2008, this study). As mentioned earlier (see Fig. 7), it should also be noted that an as yet unidentified factor other than PIF4 and PIF5 most probably plays the redundant role in this model (see also Fig. 6A), because the photoperiodic response of hypocotyl elongation is still observed to some extent even in the pif4-101 pif5-1 double mutant. These critical problems remain to be resolved in the future. In the meantime, it will be of interest to address the physiological and ecological significance of our findings, as follows.

Daily and seasonal regulation of hypocotyl elongation

Daily and rhythmic elongation of seedling hypocotyls at the end of every night has recently been demonstrated (Nozue et al. 2007). In this respect, it should be emphasized that the main issue in the previous study was about the clock-controlled daily elongation of hypocotyls under a given light/dark cycle, whereas the issue we address in this study was whether the clock-controlled seasonal elongation of hypocotyls varied under light/dark photoperiod cycles. Taken together, it was revealed that the circadian clock-dependent coincidence mechanism is responsible for both the daily and seasonal control of plant growth in natural habitats, as also integrated into the model (Fig. 7).

Comparison with the clock-dependent photoperiodic control of flowering time

The essence of this study is quite similar to the coincidence model underlying the photoperiodic control of flowering time, in which the CO protein (instead of PIF4/5) serves as the critical integrator for both the clock and light signals (see Introduction). In fact, the set of clock-defective mutants exhibit an altered phenotype of flowering time (Table 1). These facts imply that the circadian clock is capable of controlling multiple output pathways in a distinct fashion, depending on when the clock opens the gate for a given pathway. In the photoperiodic control of flowering time, the clock opens the gate during extended evening in long days for CO expression so as to be stabilized coincidentally at that time in the presence of light (Valverde et al. 2005, Niwa et al. 2007). The proposed mechanism underlying the photoperiodic control of hypocotyl elongation is a mirror image in that the same timer opens the gate during extended pre-dawn in short days for PIF4/5 expression so as to be stabilized coincidentally at that time in the absence of light (Fig. 7, and see also Niwa et al. 2007). From the evolutionary viewpoint, therefore, the photoperiodic regulation of flowering time and plant growth might be examples of sophisticated applications of the circadian clock system for adaptive responses, as further considered below.

Physiological significance of clock-dependent photoperiodic control of hypocotyl elongation and plant growth

The circadian clock-mediated photoperiodic regulation was assumed to be important, not only for the control of hypocotyl elongation in young seedlings but also for wider aspects of photomorphogenesis (Fig. 1D). This mechanism might be implicated in a more global growth regulation of adult plants, which includes regulation of leaf expansion, petiole elongation, chloroplast development and perhaps production of plant mass. Here, we attempted to examine this assumption experimentally, as follows.

The wild type (Col), prr9 prr7 prr5 mutant and CCA1ox transgenic plants were grown for a prolonged time (25-days-old) in varied photoperiods (Fig. 9A). For these adult plants, their hypocotyl lengths were measured (Fig. 9B, C). Notably, the characteristic photoperiodic response curves were observed, the profiles of which were essentially the same as those observed for young seedlings (see Fig. 3B, C). Furthermore, the pif4-101 pif5-1 double mutant plants (21-days-old) showed considerably shorter hypocotyls in every photoperiod tested (Fig. 9D), the results being consistent with those of pif4-101 pif5-1 young seedlings (see Fig. 7).

More interestingly, it was apparent that the overall plant growth including leaf development in the clock-defective mutants was attenuated (or affected) more rapily and severely in response to changes in photoperiods, as compared with the case of wild-type plants (see the images in Fig. 9A). In other words, the morphological development of wild-type plants was less affected by changes in photoperiod. Hence, it is conceivable that the circadian clock modulates overall plant growth by functioning as a sophisticated buffer against changes in photoperiod. Such a clock-dependent prevention of rapid attenuation of plant growth in short-day conditions may be an important adaptive response. Keeping this in mind, it will be of interest to consider the ecological significance (or evolutionarily selective advantage) for the clock-dependent photoperiodic control of plant growth.

Ecological significance of clock-dependent photoperiodic control of plant growth

To speculate on the ecological significance of the photoperiodic response of plant growth (or hypocotyl elongation), the following general point should first be considered from the Darwinian viewpoint, because nothing in biology makes sense except in the light of evolution, as Dobzhansky has emphasized in his report (1973). The question is, namely, which aspect of the clock-controlled photoperiodic response is ‘evolutionarily adaptive’? (i) The acceleration of hypocotyl in short days may be an important adoptive response. (ii) In contrast, the prevention of rapid elongation of hypocotyls and/or rapid attenuation of growth in response to seasonal changes may be more important, as discussed in Fig. 9. In any case, as recently considered by Robertson et al. (2009), preferential cell elongation (or hypocotyl elongation) during the night of short days may have certain adaptive advantages. They pointed out the following: (i) turgor pressure of the cell, which is the driving force for cell elongation, is highest prior to dawn when there is minimal loss of water from the closed stomata; and (ii) preferential growth during the night of short days might also allow plants to use carbon sources judiciously (e.g. starch), which are stored in the daytime in anticipation of night-time demand for growth (Smith and Stitt 2007). In this context, both the analogous coincidence mechanisms underlying the photoperiodic regulation of plant growth and flowering time might be advantageous for proper seasonal switching of vegetative growth to reproductive seed and/or fruit development (Hotta et al. 2007, Pouteau et al. 2008). More specifically, the clock-controlled season-dependent switching ‘from source to sink’ might be potentially advantageous for the plant life cycle, because this will maximize accumulation of carbon source, body growth, seed set and successful survival and revival of descendants, eventually (Robertson et al. 2009).

The photoperiodic control of plant growth might also be relevant to shade avoidance, through which the elongation of hypocotyls and internodes is accelerated to escape from the canopy (or competitive neighborhoods) (Franklin 2008, Lorrain et al. 2008), and shade avoidance also promotes the flowering time through modulating the photoperiodic pathway (Johnson et al. 2008). Through both the coincidence mechanisms, alternatively, plants might be able to adapt to their native habitat in high-latitude areas where photoperiods vary across a very wide rage. In fact, it was previously reported that CCA1ox seedlings are less viable under very short-day conditions (Green et al. 2002). In any case, the circadian clock-dependent photoperiodic controls of plant growth and development may confer ecological fitness to plants growing in natural habitats.

Materials and Methods

Plant lines and growth conditions

Arabidopsis thaliana mutants and transgenic lines used in this study are listed (Table 1, also see Yamashino et al 2008). They are all in the Col background. Seeds were sown on gellan gum plates containing MS salts with sucrose (1%) and kept at 4°C for 48 h in the dark. After the seeds were exposed to white light for 3 h to enhance germination, they were kept at 22°C for 21 h in the dark. They were then incubated in a multiple growth chamber consisting of eight compartments, all of which had been set equivalently at 22°C with 70 μmol m–2 s–1 (CCFL white light). Each compartment was adjusted to varied photoperiods.

Preparation of RNA and qRT-PCR

Plants were grown on MS plates containing 1% sucrose under conditions of the light/dark cycle, as described previously. They were harvested at every appropriate interval to prepare total RNA. Samples were purified with an RNeasy plant mini kit (Qiagen, Venlo, The Netherlands). To synthesize cDNA, RNA (1 μg of each) was converted to cDNA with ReverTra Ace (TOYOBO, Osaka, Japan) and oligo(dT) primer. The synthesized cDNAs were amplified with SYBR Premix Ex TaqII (TAKARA SHUZO CO., LTD., Kyoto, Japan) and a primer set using a MiniOpticon real-time PCR system (Bio-Rad Laboratories, Hercules, CA, USA). The primer sets used in this study were, for PIF4, 5′-ATCATCTCCGACCGGTTTGC and 5′-AGTGGCTCACCAACCTAGTG; for PIF5, 5′-GATGCAGACCGTGCAACAAC and 5′-CTTTTATGCTTGCTTAGGCG; and for APX3, 5′-CTCCGTTCTCTCATCGC and 5′-CAGAGATCGAGAGCGATC. The APX3 encoding an ascorbate peroxidase isozyme was used an internal reference. The following standard thermal cycling program was used for all PCRs: 95°C for 120 s, 40 cycles of 95°C for 10 s and 60°C for 60 s. The equipment used is a Stepone Plus™ Real-Time PCR System (Applied Biosystems, Foster City, CA, USA).

Funding

The Ministry of Education, Culture, Sports, Science, and Technology of Japan Grants-in-Aid for Scientific Research on Priority Areas (No. 19039013, to T.M., and No. 20061016, to T.Y.) and a Grant-in-Aid for the GCOE Programs (Systems Biology).

Acknowledgements

We thank Dr. N. Nakamichi (RIKEN Plant Science Center) and Dr. S. Ito (Nagoya University) for some preliminary genetic studies, including construction of clock-defective mutant strains. Thanks are also due to Dr. E. Tobin (University of California, Los Angeles, CA, USA), Dr. J. Chory (The Salk Institute, La Jolla, CA, USA), Dr. S. Kay (Scripps Research Institute, La Jolla, CA, USA), Dr. C. Fankhauser (University of Lausanne, Switzerland) and Dr. S. Prat (Campus University, Spain), also the Arabidopsis Biological Resources Centre and Kazusa DNA Research Institute for seeds.

References

Achard
P
Renou
JP
Berthome
R
Harberd
NP
Genschik
P
Plant DELLAs restrain growth and promote survival of adversity by reducing the levels of reactive oxygen species.
Curr. Biol.
 , 
2008
, vol. 
6
 (pg. 
656
-
660
)
Alabadi
D
Blazquez
MA
Molecular interactions between light and hormone signaling to control plant growth.
Plant Mol. Biol.
 , 
2009
, vol. 
69
 (pg. 
409
-
417
)
Alabadi
D
Gallego-Bartolome
J
Orlando
L
Garcia-Carcel
L
Rubio
V
Martinez
C
, et al.  . 
Gibberellins modulate light signaling pathways to prevent Arabidopsis seedling de-etiolation in darkness.
Plant J
 , 
2008
, vol. 
53
 (pg. 
324
-
335
)
Alabadi
D
Oyama
T
Yanovsky
MJ
Harmon
FG
Mas
P
Kay
SA
Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock.
Science
 , 
2001
, vol. 
293
 (pg. 
880
-
883
)
Breton
G
Kay
SA
Plant biology: time for growth.
Nature
 , 
2007
, vol. 
448
 (pg. 
265
-
266
)
Corbesier
L
Vincent
C
Jang
S
Fornara
F
Fan
Q
Searle
I
, et al.  . 
FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis.
Science
 , 
2007
, vol. 
316
 (pg. 
1030
-
1033
)
Covingtone
MF
Maloof
JN
Straume
M
Kay
SA
Harmer
SL
Global transcriptome analysis reveals circadian regulation of key pathways in plant growth and development.
Genome Biol
 , 
2008
, vol. 
9
 pg. 
R13
 
Davier
J
de Luca
M
Prat
S
Transcriptional factor interaction: a central step in DELLA function.
Curr. Opin. Genet. Dev
 , 
2008
, vol. 
18
 (pg. 
1
-
9
)
de Lucas
M
Daviere
JM
Rodriguez-Falcon
M
Pontin
M
Iglesias-Pedraz
JM
Lorrain
S
, et al.  . 
A molecular framework for light and gibberellin control of cell elongation.
Nature
 , 
2008
, vol. 
451
 (pg. 
480
-
484
)
Djakovic-Petrovic
T
de Wit
M
Voesenek
LA
Pierik
R
DELLA protein function in growth responses to canopy signals.
Plant J
 , 
2007
, vol. 
51
 (pg. 
117
-
126
)
Dobzhansky
Th
Nothing in biology makes sense except in the light of evolution.
Amer. Biol. Teach
 , 
1973
, vol. 
35
 (pg. 
125
-
129
)
Duek
PD
Fankhauser
C
bHLH class transcription factors take centre stage in phytochrome signalling.
Trends Plant Sci
 , 
2005
, vol. 
10
 (pg. 
51
-
54
)
Feng
S
Martinez
C
Gusmaroli
G
Wang
Y
Zhou
J
Wang
F
, et al.  . 
Coordinated regulation of Arabidopsis thaliana development by light and gibberellins.
Nature
 , 
2008
, vol. 
451
 (pg. 
475
-
479
)
Franklin
KA
Shade avoidance.
New Phytol
 , 
2008
, vol. 
179
 (pg. 
930
-
944
)
Fujimori
T
Yamashino
T
Kato
T
Mizuno
T
Circadian-controlled basic/helix–loop–helix factor, PIL6, implicated in light-signal transduction in
Arabidopsis thaliana. Plant Cell Physiol
 , 
2004
, vol. 
45
 (pg. 
1078
-
1086
)
Gardner
MJ
Hubbard
KE
Hotta
CT
Dodd
AN
Webb
AA
How plants tell the time.
Biochem. J
 , 
2006
, vol. 
397
 (pg. 
15
-
24
)
Green
RM
Tingay
S
Wang
ZY
Tobin
EM
Circadian rhythms confer a higher level of fitness to Arabidopsis plants.
Plant Physiol.
 , 
2002
, vol. 
129
 (pg. 
576
-
584
)
Hotta
CT
Gardner
MJ
Hubbard
KE
Baek
SJ
Dalchau
N
Suhita
D
, et al.  . 
Modulation of environmental responses of plants by circadian clocks.
Plant Cell Environ
 , 
2007
, vol. 
30
 (pg. 
333
-
349
)
Huq
E
Quail
PH
PIF4, a phytochrome-interacting bHLH factor, functions as a negative regulator of phytochrome B signaling in Arabidopsis.
EMBO J
 , 
2002
, vol. 
15
 (pg. 
2441
-
2450
)
Huq
E
Tepperman
JM
Quail
PH
GIGANTEA is a nuclear protein involved in phytochrome signaling in Arabidopsis.
Proc. Natl Acad. Sci. USA
 , 
2000
, vol. 
97
 (pg. 
9789
-
9794
)
Imaizumi
T
Kay
SA
Photoperiodic control of flowering: not only by coincidence.
Trends Plant Sci
 , 
2006
, vol. 
11
 (pg. 
550
-
558
)
Ito
S
Nakamichi
N
Nakamura
Y
Niwa
Y
Kato,
T
Murakami
M
, et al.  . 
Genetic linkages between circadian clock-associated components and phytochrome-dependent red light signal transduction in Arabidopsis thaliana.
Plant Cell Physiol
 , 
2007
, vol. 
48
 (pg. 
971
-
983
)
Johnson
AC
Strasser
B
Cerdan
PD
Amasino
RM
Acceleration of flowering during shade-avoidance in Arabidopsis thaliana alters the balance between FLOWERING LOCUS C-mediated repression and photoperiodic induction of flowering.
Plant Physiol
 , 
2008
 
10.1104/pp.108.125468.
Khanna
R
Huq
E
Kikis
EA
Al-Sady
B
Lanzatella
C
Quail
PH
A novel molecular recognition motif necessary for targeting photoactivated phytochrome signaling to specific basic helix–loop–helix transcription factors.
Plant Cell
 , 
2004
, vol. 
16
 (pg. 
3033
-
3044
)
Kobayashi
Y
Weigel
D
Move on up, it's time for change—mobile signals controlling photoperiod-dependent flowering.
Genes Dev
 , 
2007
, vol. 
21
 (pg. 
2371
-
2384
)
Locke
JC
Kozma-Bognar
L
Gould
PD
Feher
B
Kevei
E
Nagy
F
, et al.  . 
Experimental validation of a predicted feedback loop in the multi-oscillator clock of Arabidopsis thaliana.
Mol. Syst. Biol
 , 
2006
, vol. 
2
 pg. 
59
 
Lorrain
S
Allen
T
Duek
PD
Whiteleam
GC
Fankhauser
C
Phytochrome-mediated inhibition of shade avoidance involves degradation of growth-promoting bHLH transcription factors.
Plant J
 , 
2008
, vol. 
53
 (pg. 
312
-
323
)
Makino
S
Kiba
T
Imamura
A
Hanaki
N
Nakamura
A
Suzuki
T
, et al.  . 
Genes encoding pseudo-response regulators: insight into His-to-Asp phosphorelay and circadian rhythm in Arabidopsis thaliana.
Plant Cell Physiol
 , 
2000
, vol. 
41
 (pg. 
791
-
803
)
Makino
S
Matsushika
A
Kojima
M
Yamashino
T
Mizuno
T
The APRR/TOC1 quintet implicated in circadian rhythms of Arabidopsis thaliana: I. characterization with APRR1-overexpressing plants.
Plant Cell Physiol
 , 
2002
, vol. 
43
 (pg. 
58
-
69
)
Mas
P
Circadian clock signaling in Arabidopsis thaliana: from gene expression to physiology and development.
Int. J. Dev. Biol
 , 
2005
, vol. 
49
 (pg. 
491
-
500
)
Mas
P
Alabadi
D
Yanovsky
MJ
Oyama
T
Kay
SA
Dual role of TOC1 in the control of circadian and photomorphogenic responses in Arabidopsis.
Plant Cell
 , 
2003
, vol. 
15
 (pg. 
223
-
236
)
Matsushika
A
Makino
S
Kojima
M
Mizuno
T
Circadian waves of expression of the APRR1/TOC1 family of pseudo-response regulators: insight into the plant circadian czlock in Arabidopsis thaliana.
Plant Cell Physiol
 , 
2000
, vol. 
41
 (pg. 
1002
-
1012
)
Matsushika
A
Makino
S
Kojima
M
Yammashino
T
Mizuno
T
The APRR1/TOC1 quintet implicated in circadian rhythms of Arabidopsis thaliana: II. Characterization with CCA1-overexpressing plants.
Plant Cell Physiol
 , 
2002
, vol. 
43
 (pg. 
118
-
122
)
McClung
CR
Plant circadian rhythms.
Plant Cell
 , 
2006
, vol. 
18
 (pg. 
792
-
803
)
McClung
CR
Comes a time.
Curr. Opin. Plant Biol
 , 
2008
, vol. 
11
 (pg. 
1
-
7
)
Michael
TP
Mockler
TC
Breton
G
McEntee
C
Byer
A
Trout
JD
, et al.  . 
Network discovery pipeline elucidates conserved time-of-day-specific cis-regulatory modules.
PLoS Genet
 , 
2008
, vol. 
4
 pg. 
: e14
 
Mizoguchi
T
Wheatley
K
Hanzawa
Y
Wright
L
Mizoguchi
M
Song
HR
, et al.  . 
LHY and CCA1 are partially redundant genes required to maintain circadian rhythms in Arabidopsis.
Dev. Cell
 , 
2002
, vol. 
2
 (pg. 
629
-
641
)
Mizuno
T
Nakamichi
N
Pseudo-response regulators (PRRs) or true oscillator components (TOCs).
Plant Cell Physiol
 , 
2005
, vol. 
46
 (pg. 
677
-
685
)
Monte
E
Al-Sady
B
Leivar
P
Quail
PH
Out of the dark: how the PIFs are unmasking a dual temporal mechanism of phytochrome signalling.
J. Exp. Bot
 , 
2007
, vol. 
58
 (pg. 
3125
-
3133
)
Nakamichi
N
Kita
M
Ito
S
Yamashino
T
Mizuno
T
PSEUDO-RESPONSE REGULATORS, PRR9, PRR7, and PRR5, play together an essential role close to the circadian clock of Arabidopsis thaliana.
Plant Cell Physiol
 , 
2005
, vol. 
46
 (pg. 
686
-
698
)
Nakamichi
N
Kita
M
Niinuma
K
Ito
S
Yamashino
T
Mizoguchi
T
, et al.  . 
Arabidopsis clock-associated pseudo-response regula-tors PRR9, PRR7 and PRR5 coordinately and positively regulate flowering time through the canonical CONSTANS-dependent photoperiodic pathway.
Plant Cell Physiol
 , 
2007
, vol. 
48
 (pg. 
822
-
832
)
Newhauser
JL
Dawning of a new era: photomorphogenesis as an integrated molecular network.
Curr. Opin. Plant Biol
 , 
2008
, vol. 
11
 (pg. 
4
-
8
)
Niwa
Y
Ito
S
Nakamichi
N
Mizoguchi
T
Niinuma
K
Yamashino
T
, et al.  . 
Genetic linkages of the circadian clock-associated genes, TOC1, CCA1 and LHY, in the photoperiodic control of flowering time in Arabidopsis thaliana.
Plant Cell Physiol
 , 
2007
, vol. 
48
 (pg. 
925
-
937
)
Nozue
K
Covington
MF
Duek
PD
Lorrain
S
Fankhauser
C
Harmer
SL
, et al.  . 
Rhythmic growth explained by coincidence between internal and external cues.
Nature
 , 
2007
, vol. 
448
 (pg. 
358
-
361
)
Nozue
K
Maloof
JN
Diurnal regulation of plant growth.
Plant Cell Environ
 , 
2006
, vol. 
29
 (pg. 
396
-
408
)
Pouteau
S
Carre
I
Gaudin
V
Ferret
V
Lefebvre
D
Wilson
M
Diversification of photoperiodic response patterns in a collection of early flowering mutants of Arabidopsis.
Plant Physiol
 , 
2008
, vol. 
148
 (pg. 
1465
-
1473
)
Quail
PH
Photosensory perception and signalling in plant cells: new paradigms?
Curr. Opin. Cell Biol
 , 
2002
, vol. 
14
 (pg. 
180
-
188
)
Reed
JW
Nagatani
A
Elich
TD
Fagan
M
Chory
J
Phytochrome A and phytochrome B have overlapping but distinct functions in Arabidopsis development.
Plant Physiol
 , 
1994
, vol. 
104
 (pg. 
1139
-
1149
)
Robertson
FC
Skeffigton
AW
Gardner
MJ
Webb
AA
Interactions between circadian and hormonal signalling in plants.
Plant Mol. Biol
 , 
2009
, vol. 
69
 (pg. 
419
-
427
)
Sato
E
Nakamichi
N
Yamashino
T
Mizuno
T
Aberrant expression of the Arabidopsis circadian-regulated APRR5 gene belonging to the APRR1/TOC1 quintet results in early flowering and hypersensitiveness to light in early photomorphogenesis.
Plant Cell Physiol
 , 
2002
, vol. 
43
 (pg. 
1374
-
1385
)
Schaffer
R
Ramsay
N
Samach
A
Corden
S
Putterill
J
Carre
IA
, et al.  . 
The late elongated hypocotyl mutation of Arabidopsis disrupts circadian rhythms and the photoperiodic control of flowering.
Cell
 , 
1998
, vol. 
93
 (pg. 
1219
-
1229
)
Searle
I
Coupland
G
Induction of flowering by seasonal changes in photoperiod.
EMBO J
 , 
2004
, vol. 
23
 (pg. 
1217
-
1222
)
Smith
AM
Stitt
M
Coordination of carbon supply and plant growth.
Plant Cell Environ
 , 
2007
, vol. 
30
 (pg. 
1126
-
1149
)
Strayer
C
Oyama
T
Schultz
TF
Raman
R
Somers
DE
Mas
P
, et al.  . 
Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog.
Science
 , 
2000
, vol. 
289
 (pg. 
768
-
771
)
Valverde
F
Mouradov
A
Soppe
W
Ravenscroft
D
Samach
A
Coupland
G
Photoreceptor regulation of CONSTANS protein in photoperiodic flowering.
Science
 , 
2005
, vol. 
303
 (pg. 
1003
-
1006
)
Vandenbussche
F
Verbelen
JP
Van Der Straeten
D
Of light and length: regulation of hypocotyl growth in Arabidopsis.
Bioessays
 , 
2005
, vol. 
27
 (pg. 
275
-
284
)
Wang
ZY
Tobin
EM
Constitutive expression of the CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) gene disrupts circadian rhythms and suppresses its own expression.
Cell
 , 
1998
, vol. 
93
 (pg. 
1207
-
1217
)
Yakir
E
Hilman
D
Harir
Y
Green
RM
Regulation of output from the plant circadian clock.
FEBS J
 , 
2006
, vol. 
274
 (pg. 
335
-
345
)
Yamashino
T
Ito
S
Niwa
Y
Kunihiro
A
Nakamichi
N
Mizuno
T
Involvment of Arabidopsis clock-associated pseudo-response regulators in diurnal oscillation of gene expression in the presence of environmental time cues.
Plant Cell Physiol
 , 
2008
, vol. 
49
 (pg. 
1839
-
1850
)
Yanovsky
MJ
Kay
SA
Living by the calendar: how plants know when to flower.
Nat. Rev. Mol. Cell Biol
 , 
2003
, vol. 
4
 (pg. 
265
-
275
)
Zeilinger
MN
Farre
EM
Taylor
SR
Kay
SA
Doyle
F.J.
III
A novel computational model of the circadian clock in Arabidopsis that incorporates PRR7 and PRR9.
Mol. Syst. Biol
 , 
2006
, vol. 
2
 (pg. 
58
-
64
)

Abbreviations:

    Abbreviations:
  • BHLH

    basic helix–loop–helix

  • CCA1

    CIRCADIAN CLOCK-ASSOCIATED 1

  • CO

    CONSTANS

  • Col

    Columbia

  • COP1

    CONSTITUTIVE PHOTMORPHOGENIC 1

  • DAG

    days after germination

  • DD

    continuous darkness

  • FT

    FLOWERING LOCUS T

  • PhyB

    phytochrome B

  • Hy5

    ELONGATED HYPOCOTYL 5

  • LD

    light/dark

  • Ler

    Landsberg erecta

  • LHY

    LATE ELONGATED HYPOCOTYL

  • LL

    continuous light

  • PIF

    PHYTOCHROME INTERACTING FACTOR

  • PRR

    PSEUDO-RESPONSE REGULATOR

  • TOC1

    TIMING OF CAB2 EXPRESSION 1

  • WS

    Wassilewskija.