Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2005 Nov;139(3):1557–1569. doi: 10.1104/pp.105.067173

Independent Roles for EARLY FLOWERING 3 and ZEITLUPE in the Control of Circadian Timing, Hypocotyl Length, and Flowering Time1

Woe-Yeon Kim 1, Karen A Hicks 1, David E Somers 1,*
PMCID: PMC1283789  PMID: 16258016

Abstract

The circadian clock regulates many aspects of plant development, including hypocotyl elongation and photoperiodic induction of flowering. ZEITLUPE (ZTL) is a clock-related F-box protein, and altered ZTL expression causes fluence rate-dependent circadian period effects, and altered hypocotyl elongation and flowering time. EARLY FLOWERING 3 (ELF3) is a novel protein of unknown biochemical function. elf3 mutations cause light-dependent circadian dysfunction, elongated hypocotyls, and early flowering. Although both genes affect similar processes, their relationship is unclear. Here we show that the effects of ZTL and ELF3 on circadian clock function and early photomorphogenesis are additive. The long period of ztl mutations and ELF3 overexpressors are more severe than either alone. Dark-release experiments showing additivity in phase advances suggest that the arrthymicity caused by ZTL overexpression and that of the elf3-1 mutation arise through independent pathways. A similar additive effect on hypocotyl elongation in red and blue light is also observed. In contrast, ELF3 and ZTL overexpressors act similarly to control flowering time in long days through the CONSTANS/FLOWERING LOCUS T (CO/FT) pathway. ZTL overexpression does not delay flowering through changes in GIGANTEA or FLAVIN-BINDING, KELCH REPEAT, F-BOX levels, but through a ZTL-mediated reduction in CO expression. In contrast, ELF3 negatively regulates CO, FT, and GIGANTEA transcript levels, as the expression of all three genes is increased in elf3-1. The elf3-1 co-1 double mutant flowers much earlier in long days than co-1, although FT message levels remain very low. These results show that elf3-1 can derepress late flowering through a CO-independent mechanism. ELF3 may act at more than one juncture, possibly posttranscriptionally.

Plant development is strongly affected by the light quality and intensity. EARLY FLOWERING 3 (ELF3) and ZEITLUPE (ZTL) are two genes that function in light signaling to the plant. Although each was initially identified through very different genetic screens, analysis has subsequently shown that both strongly affect plant development and physiology in similar and also in contrasting ways.

The elf3-1 mutation was identified in a screen for early flowering under short days (Zagotta et al., 1992, 1996). Subsequent analysis showed that CONSTANS (CO) expression was derepressed in elf3 loss-of-function mutations, suggesting that ELF3 controls flowering by regulating CO (Suarez-Lopez et al., 2001). Early photomorphogenesis is also affected, with elf3-1 plants showing long hypocotyls and petioles under red and blue light (Zagotta et al., 1996). Unexpectedly, elf3 mutants are also defective in circadian clock function. Clock-regulated rhythms of leaf movement, hypocotyl elongation, and gene transcription become arrhythmic in elf3 seedlings in constant light (LL; Hicks et al., 1996; Dowson-Day and Millar, 1999; Reed et al., 2000). In contrast, rhythms in elf3 mutants persist in constant dark (DD; Hicks et al., 1996). These results and others, together with the observation that it can interact physically with phytochrome B (phyB), has lead to the notion that ELF3 is involved in the gating of photic input to the clock (McWatters et al., 2000; Covington et al., 2001; Liu et al., 2001). However, the molecular role of the protein remains obscure, and it is unclear whether all the light and photoperiod-dependent phenotypes derive from the defect in circadian function.

ZTL is the founding member of the three-member ZTL gene family, first identified as a long-period circadian clock mutant, ztl-1 (Somers et al., 2000). Hypocotyl elongation in ztl mutants is hypersensitive to red light, though little affected in blue light. ZTL, LKP2 (LOV KELCH PROTEIN 2), and FKF1 (FLAVIN-BINDING, KELCH REPEAT, F-BOX) uniquely possess a light, oxygen, and voltage (LOV) domain at the N terminus, followed by an F-box domain and six carboxy-terminal kelch repeats (Somers, 2001, 2005). Fluence rate strongly affects period length in ztl mutants, implicating ZTL in the light input pathway to the clock (Somers et al., 2000). Flavin binding of the LOV domains of the phototropins and of the Neurospora crassa WHITE COLLAR-1 protein have implicated these polypeptides as blue-light photoreceptors (Briggs and Christie, 2002; Froehlich et al., 2002; He et al., 2002). Similarly, the bacterially expressed LOV domain of FKF1 can bind flavin mononucleotide in vitro (Cheng et al., 2003; Imaizumi et al., 2003) and can, in part, functionally substitute for the LOV domain in WHITE COLLAR-1 (Cheng et al., 2003). Similar results with the isolated ZTL and LKP2 LOV domains suggest that this gene family may constitute a novel type of blue-light photoreceptor. As an F-box protein, ZTL participates in an Skp/Cullin1/F-box-type E3 ligase and facilitates the proteasome-dependent proteolysis of the clock-associated protein TIMING OF CAB 1 (TOC1), targeting it for degradation via a proteasome-dependent pathway (Mas et al., 2003b; Han et al., 2004) The abundance of the ZTL protein itself is posttranscriptionally under circadian clock control, and its turnover is also proteasome dependent (Kim et al., 2003). Overexpression of ZTL down-regulates transcription of CO to delay flowering and lengthens hypocotyl (Somers et al., 2004).

Altered circadian period often correlates with abnormal hypocotyl growth and flowering time, and ELF3 and ZTL levels strongly affect these three processes. Here we investigate the genetic and physiological interactions between ZTL and ELF3. We show that the effects of ZTL and ELF3 on circadian clock function and early photomorphogenesis are additive. In contrast, ELF3 and ZTL overexpressors act similarly to control flowering time, but through different mechanisms. ZTL overexpression delays flowering entirely through a ZTL-mediated reduction in CO expression. ELF3 acts more broadly, negatively regulating CO, FLOWERING LOCUS T (FT), and GIGANTEA (GI) transcript levels. Tests of elf3-1 co double mutants show that elf3-1 can derepress late flowering through a CO-independent mechanism. ELF3 appears to act at more than one juncture and possibly through a posttranscriptional mechanism.

RESULTS

Effects of Altered ZTL and ELF3 Expression on Free-Running Circadian Period

Previous reports have shown that reduced levels of ZTL lengthens the free-running period of circadian gene expression in continuous light and darkness, whereas increasing levels of ZTL dosage shorten the pace of the clock, to the point of arrhythmicity at very high levels of expression (Fig. 1A). The fluence rate dependence of period is also altered by ZTL expression level, indicating a role in modulating phototransduction to the oscillator (Somers et al., 2000, 2004).

Figure 1.

Figure 1.

Open in a new tab

Effects of ZTL and ELF3 expression levels on free-running period. A and B, Rhythmicity of CAB:LUC expression in elf3-1, ZTL OX (C24 ecotype), elf3-1 ztl-3, and elf3-1 ZTL OX plants (A), and in ELF3 OX, ztl-1 (Col), and ELF3 OX ztl-1 double mutant plants (B). Plants were entrained in 12-h-white light /12-h-dark cycles and transferred to constant red light (RR; 20 μmol m−2 s−1), and bioluminescence was monitored every 2 h for 4 to 5 d. Traces represent averages from at least 20 seedlings from two independent experiments.

The elf3-1 mutation abolishes the rhythms of all circadian outputs tested, including CAB2:LUCIFERASE (CAB:LUC) expression (Fig. 1A), hypocotyl elongation, and cotyledon movement in LL, but has little effect on clock function in darkness (Hicks et al., 1996, 2001). Overexpression of ELF3 increases period length in LL, in support of the notion that, like ZTL, it normally acts to modulate the photosensitivity of the clock system (Covington et al., 2001). However, ELF3 appears to be a negative regulator of phototransduction, whereas ZTL acts positively, if the tendency toward arrhythmicity is interpreted as a hypersensitivity to light. We therefore examined the effects of coordinated changes in ELF3 and ZTL expression on free-running period.

When the elf3-1 mutation is combined with the ztl-3 mutation (ztl-3 elf3-1) or the ZTL overexpressor (elf3-1 ZTL OX), the free-running period of the clock-controlled CAB2::LUC reporter is arrhythmic, under either constant red light (Fig. 1A) or blue light (data not shown). These results are similar to those observed in the elf3-1 single mutant (Fig. 1A), indicating that elf3-1 is epistatic to ZTL activity.

Loss of rhythmicity can be difficult to interpret on its own, as it may mask a continued, underlying clock activity. Therefore, we also compared the phenotypes of the ELF3 overexpressor (ELF3 OX) and ztl-1 single mutants with ELF3 OX ztl-1 plants. When entrained seedlings were transferred to constant red light, the rhythms in the single mutants were robust, with the period of ELF3 OX seedlings close to wild type (24.7 ± 0.3 h) and ztl-1 plants showing the expected long period (28.9 ± 0.6 h; Covington et al., 2001; Somers et al., 2004). However, the ELF3 OX ztl-1 double mutant phenotype was more severe than either mutant alone, with period length longer than either single mutant (30.2 ± 0.6 h) and a severely reduced amplitude (Fig. 1B). Results were similar in blue light (data not shown). These data indicate an additive effect of the two genes on clock function.

Circadian cycling of the CAB:LUC reporter was also assessed in the various mutant combinations during a dark period extension into the 12-h subjective light period that occurs during standard light/dark (LD) entrainment (Fig. 2). This approach removes the light-dependent stimulation of CAB:LUC activity that normally occurs at lights on, and reports the phase of expression in each mutant combination as determined solely by the light-to-dark and dark-to-light transitions of the previous entrainment cycle. This protocol can also test for the phase of rhythmic activity in lines that are arrhythmic in LL but still cycle in darkness (e.g. elf3-1). Plants entrained in LD cycles were released into DD, and the phase of the first peak of each single and double mutant was recorded using the CAB::LUC reporter.

Figure 2.

Figure 2.

Open in a new tab

Effects of ZTL and ELF3 expression levels on clock-controlled expression in extended dark. A to C, Timing of peak CAB:LUC expression in wild type (Col WT), elf3-1, ztl-3, and elf3-1 ztl-3 plants (A); elf3-1, ZTL OX (C24), and elf3-1 ZTL OX plants (B); and ELF3 OX, ztl-1 (Col), and ELF3 OX ztl-1 plants (C). D, Peak CAB:LUC expression in Col and C24 wild type and an F2 population of a cross between Col and C24 are shown as controls. Seedlings were entrained in 12-h-light (white)/12-h-dark cycles, transferred to DD, and measured for luminescence expression every hour. Mean values (±SEM [A, C, and D]) and mean values normalized to mean expression level of 1 for each genotype (B) are shown.

Relative to wild type, the first peak in darkness is phase advanced by 4 to 5 h in elf3-1 (Fig. 2A; Reed et al., 2000) and ZTL OX plants (Fig. 2B). This is consistent with the extremely short period occasionally observed in LL in strong ZTL overexpressors (Somers et al., 2004; Fig. 1A) and suggests that the elf3-1 mutation may similarly cause arrhythmicity in LL through extreme shortening of period. Similarly, the phase delays of 8 h in ztl-1 Columbia (Col; Fig. 2C) and ztl-3 (Fig. 2A) and 2 h in ELF3 OX (Fig. 2C) are consistent with the longer period observed in LL in these backgrounds. Interestingly, the ZTL OX elf3-1 double mutant causes a reduction in peak amplitude but is clearly further phase advanced by 4 to 5 h relative to the two single mutants (Fig. 2B). Thus, despite the apparently similar arrhythmic phenotypes in LL, this analysis suggests that the two genes act separately on the same pathway, or in parallel on two independent pathways to affect circadian clock function.

The phase of CAB:LUC expression in the elf3-1 ztl-3 double is similar to that of elf3-1alone (Fig. 2A), indicating that ELF3 is required for the strong phase delay observed in the absence of ZTL under these conditions. If ELF3 acts to suppress photic input to the clock, this appears to happen independently of the hyposensitivity to light caused by the loss of ZTL. Conversely, in the ELF3 OX ztl-1 double mutant, ELF3 overexpression added little to the strong phase delay caused by ztl-1 alone (Fig. 2C), although in LL ELF3 OX enhanced the effects of the absence of ZTL. Taken together, our data suggest that ZTL and ELF3 act to modulate clock activity largely independently of each other, with ELF3 acting to suppress photic input to the clock and ZTL acting to promote.

The Regulation of Hypocotyl Elongation by ZTL and ELF3

ZTL and ELF3 can each interact with phyB in vitro and in the yeast two-hybrid system, and both have been proposed to play a role in phyB-mediated signaling in early photomorphogenesis (Jarillo et al., 2001; Liu et al., 2001). To test whether ELF3 and ZTL act in the same or different phytochrome-mediated signaling pathway during photomorphogenesis, we examined the fluence rate responsiveness of hypocotyl elongation in ELF3 and ZTL mutant and overexpression lines.

In constant red light, the ztl-3 mutant (Fig. 3B) and ELF3 OX (Fig. 3A) showed hypersensitivity to red light. In contrast, elf3-1 and ZTL OX are both hyposensitive to red light at all intensities tested, with elf3-1 slightly more effective at lengthening hypocotyl length (Fig. 3A). The elf3-1 ztl-3 double mutant was intermediate in length compared to the single mutants. Interestingly, the elf3-1 ztl-3 double mutant was more similar to elf3-1 at high red light fluences, while being more similar to ztl-3 at lower fluence rates (Fig. 3B). Similarly, the hypocotyl length of elf3-1 ZTL OX plants was longer than the two single mutants alone, under all intensities tested (Fig. 3A). These results show an additive effect of the two mutations, consistent with each gene acting independently to control hypocotyl elongation.

Figure 3.

Figure 3.

Open in a new tab

ZTL and ELF3 control of hypocotyl growth during early photomorphogenesis. The wild-type (Col), elf3-1, ztl-3, ELF3 OX, ZTL OX (Col), elf3-1 ztl-3, and elf3-1 ZTL OX plants were grown for 7 to 10 d under constant red light (RL; A and B) or constant blue light (BL; C and D) at the fluence rates indicated and measured for hypocotyl length. Unconnected data points show lengths for dark-grown seedlings. Values (mean hypocotyl length ± SEM) are representative of two trials; n = 16 to 25.

Arabidopsis (Arabidopsis thaliana) cryptochrome 1 is the major photoreceptor mediating blue-light inhibition of hypocotyl elongation. Although ZTL interaction with cryptochrome 1 has been shown in vitro and in yeast two-hybrid tests, ztl loss-of-function mutations have no effect on blue light-mediated hypocotyl inhibition, and strong ZTL overexpression only modestly lengthens hypocotyl length (Somers et al., 2000, 2004). In contrast, elf3-1 has long hypocotyls in constant blue light (Zagotta et al., 1996). To examine possible interactions between ZTL and ELF3 in blue light-mediated hypocotyl inhibition, we tested the fluence rate responsiveness of different ZTL and ELF3 mutant and overexpression combinations. Strong overexpression of ELF3 had little effect on hypocotyl length, very similar to the effects of the ztl-3 mutation, over the entire range of fluence rates tested (Fig. 3, C and D). Both the elf3-1 mutation and ZTL overexpression lengthen hypocotyls in blue light (Fig. 3C). The hypocotyl length of the elf3-1 ztl-3 double mutant was very similar to hypocotyl length of the elf3-1 single mutant itself (Fig. 3D), which would be expected if the effects of the two single mutants were simply additive. Similarly, the elf3-1 ZTL OX showed hypocotyl lengths longer than each of single mutants alone (Fig. 3C). These data are consistent with an additive effect of each single mutant, suggesting that ZTL and ELF3 act in separate blue light-signaling pathways that converge to control hypocotyl elongation early plant development. Taken together, our results of testing under red and blue light support similar conclusions.

Interactions between ZTL and ELF3 in the Control of Flowering Time

ZTL overexpression significantly delays flowering in long days, and this effect is strongly dependent on increasing ZTL dosage (Somers et al., 2004). ztl loss-of-function mutations have only a modest effect on flowering time, dependent on the ecotype (Fig. 4A; Somers et al., 2000, 2004). The effect on flowering time of changing ELF3 levels is more severe. elf3-1 causes significantly earlier flowering in short days and also shortens flowering time in long days (Hicks et al., 1996, 2001), whereas ELF3 overexpression greatly delays flowering time in long days (Fig. 4A; Liu et al., 2001). We tested whether these two genes act in the same or separate pathways to control flowering time under long days.

Figure 4.

Figure 4.

Open in a new tab

Control of flowering time by ZTL and ELF3 expression levels. A, Total number of leaves (rosette + cauline) produced at flowering under long days (16 h light/ 8 h dark) in wild-type (Col), elf3-1, ztl-3, ELF3 OX, ZTL OX (Col), elf3-1 ztl-3, and elf3-1 ZTL OX plants. Values (±SEM) are representative of two independent trials; n = 13 to 16. B and C, Expression levels of CO (B) and FT (C) transcripts under long days were determined by semiquantitative PCR for the lines described in A. Values are expressed relative to ACTIN2 (ACT2) control. The same appropriate wild-type data are plotted in both sections of B and C to facilitate comparisons. White and black boxes represent light and dark periods, respectively. Data are representative of three independent trials.

When placed in combination with elf3-1, both the ztl-3 and ZTL OX lines flowered much earlier than the single mutants alone in long days. The elf3-1 ztl-3 double mutant flowered the same as elf3-1 alone, significantly earlier than the ztl-3 single mutant (Fig. 4A). More strikingly, the elf3-1 ZTL OX double mutant flowered slightly earlier than wild type, with approximately eight leaves, in contrast to the ZTL OX single mutant which flowered with about 40 leaves under long days. These results indicate that elf3-1 is largely epistatic to ZTL, regardless of ZTL expression level (Fig. 4A).

The late flowering effects of ZTL overexpression occur through the strong reduction of CO and FT message levels (Fig. 4, B and C; Somers et al., 2004). ztl-3 has little effect on either CO or FT levels in long days, relative to wild type, consistent with a wild-type flowering time (Fig. 4, B and C). The elf3-1 mutation acts, in part through the same mechanism, to moderately elevate CO expression and greatly elevate FT transcript levels, leading to accelerated flowering (Suarez-Lopez et al., 2001; Fig. 4, B and C). Conversely, ELF3 overexpression causes a strong reduction in transcript levels of both genes, leading to strong delays in flowering in long days (Fig. 4, A–C).

To determine how elf3-1 causes early flowering in the presence of high ZTL expression, we examined CO and FT transcript levels in the ELF3 and ZTL double mutants under long days. CO and FT levels in the elf3-1 ztl-3 double mutant were essentially the same as the elf3-1 single mutant (Fig. 4, B and C). These results indicate that the acceleration of flowering in elf3 mutants does not require ZTL protein. However, the elf3-1 ZTL OX double mutant also showed reduced CO mRNA levels during the photoperiod, which is the critical time during which high CO expression activates FT transcription (Valverde et al., 2004). In this background, the level of CO transcript was as low or lower than wild type and ztl-3 during the photoperiod, and almost as low as ZTL OX during this time. This was unexpected considering the much earlier flowering of elf3-1 ZTL OX, relative to ZTL OX, and two to three leaves earlier than wild type (Fig. 4). We next examined the level of FT transcript in the elf3-1 ZTL OX line. Surprisingly, FT levels were lower than wild type at all time points except during the early photoperiod (Fig. 4C). In this background, transcript levels of FT peaked early, 4 h after lights on, and then dropped to well below wild type for the rest of the photoperiod and throughout the short skotoperiod. This sharp peak, and then drop, in relative FT levels early in the photoperiod was observed in all biological (three) and technical (twice for each biological trial) repeats in this background (data not shown). Hence, high expression of ZTL is able to suppress the normally strong up-regulation of FT in the elf3-1 single mutant, except during a short period early in the photoperiod.

Effects of ELF3 and ZTL Mutations on GI and FKF1 Expression

Mutations in GI and FKF1 also delay flowering, and both have been proposed to up-regulate CO expression to control flowering (Suarez-Lopez et al., 2001; Imaizumi et al., 2003; Tseng et al., 2004). To examine whether ZTL regulation of CO expression occurs via changes in GI or FKF1expression, transcript levels of both genes were tested in ZTL OX and ztl-3 backgrounds and compared to wild type. Neither overexpression nor absence of ZTL affected the mRNA levels of either gene very strongly, indicating that ZTL does not modulate CO or FT transcript levels by reducing FKF1 or GI expression (Fig. 5, A and B).

Figure 5.

Figure 5.

Open in a new tab

Effects of ZTL and ELF3 expression levels on GI and FKF1 expression. Expression levels of GI (A) and FKF1 (B) transcripts under long days were determined by semiquantitative PCR for the same lines tested for flowering-time effects in Figure 4. Values are expressed relative to ACTIN2 (ACT2) control. White and black boxes represent light and dark periods, respectively. Data are representative of three independent trials.

We also tested the effects of ELF3 absence and overexpression on GI and FKF1 levels. elf3-1 caused a consistently strong increase in GI levels at all time points during the photo- and skotoperiods, effectively eliminating the normal cyclic expression of GI (Fig. 5A). The absence or strong overexpression of ZTL had no consistent effect on this derepression of GI in the elf3-1 background (Fig. 5A). ELF3 overexpression reduced GI transcript levels only modestly (to 60% to 70% of wild type at peak expression). FKF1 expression in elf3-1 is slightly increased at all time points under long days, but interestingly it remained cyclic, with peak expression at ZT 8. In contrast, FKF1 transcripts in the ELF3 OX were essentially the same as wild type at all time points in long days. Consistent with the lack of effect of ZTL levels on FKF1 expression, FKF1 expression in the elf3-1 ztl-3 and elf3-1 ZTL OX lines was very similar to the elf3-1 single mutant (Fig. 5B).

The moderate rise in CO expression in elf3-1 (Fig. 4B) correlates with our observed moderate increase in FKF1 expression in this background (Fig. 5B), consistent with ELF3 acting through FKF1 to affect CO expression. Recently a DOF transcription factor, CDF1, has been shown to repress CO expression. The F-box protein, FKF1, phase-specifically degrades this factor (Imaizumi et al., 2005). Hence, ELF3 may also act on CDF1 expression or activity.

Taken together, these results indicate that ZTL overexpression does not delay flowering through changes in GI or FKF1 message levels. Most likely late flowering arises through a ZTL-mediated reduction in CO expression, resulting in lower FT expression during the photoperiod. In contrast, ELF3 may act through multiple pathways. It negatively regulates CO, FT, GI, and FKF1 transcript levels, as the expression of all four genes is increased in elf3-1. Surprisingly, the elf3-1 co-1 double mutant flowers much earlier in long days than co-1, though still later than the elf3-1 single mutant and the wild type (Fig. 6A). These results suggest that elf3-1 can derepress the inhibition of flowering caused by absence of CO through a CO-independent mechanism.

Figure 6.

Figure 6.

Open in a new tab

Effect of the elf3-1 co-1 double mutant on flowering and FT expression. A, Total number of leaves (rosette + cauline; ±SEM) produced at flowering under long days (16 h light/ 8 h dark) in wild-type (Ler), elf3-1 (2× introgressed into Ler), elf3-1 co-1, co-1 (Ler), and ELF3 co-1 segregants isolated from the elf3-1 × co-1 F2 population (co-1 [seg]). n = 8 to 20. B, Expression levels of FT transcripts under long days were determined by quantitative PCR for the lines described in A. Values are expressed relative to ACTIN2 (ACT2) control. White and black boxes represent light and dark periods, respectively. Data are representative of two independent trials.

We tested whether elf3-1 can act on FT expression by examining FT mRNA levels in the elf3-1 co-1 double mutant. Quantitative reverse transcription (RT)-PCR was performed on four independent elf3-1 co-1 segregants over a long-day time course and compared to wild type and co-1 genotypes. Surprisingly, in all four double mutant isolates, FT message levels were markedly lower than wild type during the second half of the photoperiod and throughout the skotoperiod, and very similar to the co-1 single mutant (Fig. 6B). During the first 8 h of the photoperiod, FT levels in only one of the elf3-1 co-1 lines was near wild type, whereas FT expression in the remaining three double mutant lines were at or near co-1 levels. Taken together, these data show that the elf3-1 suppression of late flowering in co-1 is not due to an increase in FT message levels.

Additionally, in light of new reports showing that FD acts together with FT to promote flowering (Abe et al., 2005; Wigge et al., 2005), we tested whether FD expression was elevated in the elf3-1 co-1 double mutants, relative to wild type. Over a long-day time course, FD expression in the double mutants, co-1 single mutant, and wild type were not significantly different at any point (data not shown), nor was there evidence of diurnal or LD variation in FD expression in any of the genetic backgrounds tested (data not shown; Abe et al., 2005).

DISCUSSION

Increasingly, many genes that control the function of the circadian oscillator are also being linked to the control of photomorphogenesis and to the timing of flowering. The molecular mechanisms that link clock function to these developmental phenomena are poorly understood. Here we have shown that for these three processes two genes, ELF3 and ZTL, largely function independently of each other, rather than within the same pathway or biochemical complex. The extent of the effect of one gene, relative to the other, varies with the process, and in some cases the two act in opposition to each other. This has proved useful in unexpectedly revealing a more diverse role for ELF3 in the control of flowering.

ZTL and ELF3 Control Clock Function through Different Pathways

Previous evidence from single mutants and overexpressors of ZTL and ELF3 suggested that the genes act oppositely on clock function. Loss-of-function elf3 mutations are similar to a strong ZTL overexpressor in that both cause circadian arrhythmicity in LL. Similarly, ztl mutants (e.g. ztl-1 and ztl-3) and strong ELF3 overexpression lengthen circadian period. From these results alone one could hypothesize that ELF3 negatively regulates ZTL activity, or vice versa. However, arrhythmicity persists in elf3-1 ztl-3 double mutants, indicating that elf3-1 does not abolish cycling through a derepression of ZTL levels. In addition, when examined in extended darkness, the phase of the first CAB:luc peak in the elf3-1 ztl-3 double mutant falls later than in elf3-1 but much earlier than ztl-3 (Fig. 2A). This indicates that the two loss-of-function mutants can act, to a limited extent, to counter the effects of the other. Additionally, if ELF3 affects period solely through ZTL, ztl-1 should be epistatic to the effects of ELF3 OX. The strongly additive effect of the ELF3 OX ztl-1 double mutant in LL suggests a convergence on period control from different pathways.

Similarly, if ZTL controls period through the repression of ELF3, the period of ZTL OX elf3-1 plants should appear the same as elf3-1 mutants. Instead, the phenotype in extended DD is more severe than either mutant alone (Fig. 2B), indicating additivity in their effects.

ZTL regulates the degradation of TOC1 and controls circadian period, at least in part, through this mechanism (Mas et al., 2003b). In contrast, the manner of ELF3 control of circadian period is still unknown. Our genetic interaction data suggests that ELF3 does not act to positively regulate TOC1 expression, which could be expected of a gene that has the opposite phenotype of a negative regulator of TOC1 (i.e. ZTL). If ELF3 were a positive regulator of TOC1, elf3-1 ztl-3 plants might be predicted to be near wild type in period, as the higher levels of TOC1 caused by a ZTL deficiency (Mas et al., 2003b) could be negated by the absence of ELF3. Instead, elf3-1 arrhythmicity is epistatic to ztl-3. By similar reasoning, ELF3 ZTL double overexpressing plants could also be expected to antagonistically balance TOC1 levels to cause near wild-type period, but such plants are arrhythmic, similar to ZTL OX alone (data not shown). Additionally, TOC1 transcription is increased in elf3-1, demonstrating a role for ELF3 in repressing TOC1 expression (Alabadi et al., 2001).

ZTL and ELF3 protein levels show rhythmic expression in LD, with similar phases of peak expression at or near dusk (approximately ZT 12–13, Liu et al., 2001; Kim et al., 2003). The intracellular location of both proteins also overlaps, with ELF3 (Liu et al., 2001) and ZTL (W.-Y. Kim and D.E. Somers, unpublished data) both present in the nucleus. Despite this temporal and spatial coincidence of expression, our data indicate that ELF3 and ZTL act additively, probably through different mechanisms and molecular partners, to control circadian function. Identification of molecular mechanism of ELF3 action will clarify this relationship.

ZTL and ELF3 in Hypocotyl Control

The manner of ELF3 control of hypocotyl length is unclear. In red light, elf3 and phyB mutations act additively to control elongation, suggesting independent or partially redundant mechanisms. However, in vitro interactions between ELF3 and phyB demonstrate the potential for complex formation and indicate that in some circumstances they may act together (Reed et al., 2000; Liu et al., 2001). The long hypocotyl phenotype of the elf3 mutants in blue light cannot be easily understood through a phy-based mechanism, suggesting that ELF3 acts downstream of where blue- and red-light photoreceptors intersect, or in a blue light-specific pathway, parallel to its role in red-light signaling.

The red- and blue-light hypocotyl length phenotypes of ztl mutants are similar to those observed in TOC1 overexpressors and toc1 mutants (Mas et al., 2003a). This suggests that the effects we observe in manipulating ZTL levels are largely due to its effects on TOC1 levels. Since our double mutant analysis shows that ZTL acts independently of ELF3 in mediating photoinhibition of hypocotyl elongation, it follows that ELF3 probably acts through a mechanism separate from that of TOC1. However, until more is known of the primary action of ELF3 and TOC1, it is still not possible to know if the hypocotyl and clock effects of these two proteins are through the same or different mechanisms.

ELF3 Acts Multiply and Separately from ZTL to Control Flowering Time

The circadian clock regulates flowering through an output pathway that includes CO and FT. Both ELF3 and ZTL OX affect flowering time by acting within the photoperiodic pathway. ELF3 and ZTL OX can act as negative regulators upstream of CO and FT in this pathway, because ectopic expression of both genes negatively controls the abundance of CO and FT message levels. This relationship is unlike their effects on circadian period and hypocotyl length, where ZTL and ELF3 act oppositely with respect to their dosage.

Although ZTL normally targets TOC1 to control circadian period, our data do not allow a positioning of TOC1 into a flowering time scheme. It appears that ZTL overexpression is not acting only through effects on TOC1 with respect to flowering time. If it did, toc1 loss-of-function mutants should exhibit very late flowering, which they do not (Somers et al., 1998; Mas et al., 2003b). The molecular mechanism of ZTL OX in reducing CO message levels is currently unclear.

When overexpression of ZTL was paired with the absence of ELF3 (ZTL OX elf3-1), we observed a surprisingly high level of FT message early in the photoperiod that was not proportional to the low level of CO expression, when compared to wild type over the same time period (Fig. 4, B and C). The very high level of FT message normally seen in the elf3-1 background was effectively suppressed throughout the photo- and skotoperiods by high ZTL expression, except during this short window of time. This suggests that ZTL-mediated suppression is limited during this early part of the day, allowing the derepressing effects of the elf3 mutation to act and raise FT message levels. This action may occur posttranscriptionally, as the CO message levels in ZTL OX elf3-1 are the same or lower than wild type, yet FT message levels are 4 to 5 times higher than wild type. This notion is further supported by the suppression of late flowering in co-1 by elf3-1. In the elf3-1 co-1 double mutant we found that FT message levels are maintained well below that of wild type during the late photoperiod, although flowering time is very similar to wild type. This suggests that the moderately high levels CO and very high levels of FT present in elf3-1 contribute to, but are not solely responsible for, the early flowering phenotype. The elf3-1 co-1 result is consistent with ELF3 acting posttranscriptionally to deactivate or destabilize FT protein. It is also possible that an additional factor, acting in parallel to or downstream of, the CO/FT pathway, is under ELF3 or GI control (see below). Clearly, ELF3 can act independently of CO to control flowering time.

Figure 7 illustrates how ELF3 may act at more than one stage in the flowering time pathway. Chou and Yang (1999) have shown that elf3-1 gi-1and elf3-1 ft double mutants flower late in long days, indicating that GI and FT are required for the early flowering effects of elf3-1. However, our results indicate that CO is not required for elf3-1 to promote early flowering, suggesting that GI can act independently of CO (Mizoguchi et al., 2005). Hence, ELF3 may act solely through an effect on GI expression. Our observed increases in CO and FT expression in the elf3-1 mutant could be indirect, through the derepression of GI. This model implies a direct regulation of FT by GI, in addition to its role in CO regulation (Fig. 7A). In addition, it would require a role for GI in posttranscriptional stabilization of FT protein. Alternatively, GI may act through a second route, in parallel with the CO/FT pathway (Fig. 7A), as recently suggested by Mizoguchi et al. (2005). In the elf3-1 co-1 mutant, high levels of this factor, induced through the high levels of GI, could partially compensate for loss of signaling via the CO/FT pathway, resulting in earlier flowering. Alternatively, our results and those of Chou and Yang (1999) are consistent with ELF3 acting directly on FT protein, as noted earlier (Fig. 7B). These two models are not mutually exclusive, and others are also possible. Together, these results suggest a floral-induction pathway that is alternate or parallel to that through CO.

Figure 7.

Figure 7.

Open in a new tab

ELF3 may regulate flowering through multiple flowering time-related genes in the CO/FT pathway. ELF3 negatively regulates GI, CO, and FT expression; modest derepression of FKF1 expression by elf3-1 is not shown. Epistasis tests show elf3-1 early flowering requires GI and FT, but not CO, indicating ELF3 may regulate CO/FT expression via GI (Fig. 7A). The dotted line leading from GI to X (Mizoguchi et al., 2005) is supported by our data, but our results additionally suggest a posttranslational control of FT by GI (Fig. 7A), based on low FT levels in the elf3-1 co-1 double mutant. Alternatively, this result plus high FT expression in the elf3-1 ZTL OX background, when CO expression is wild type or lower, indicates ELF3 may regulate FT expression posttranscriptionally (Fig. 7B). Other models are also possible; see text for more details.

MATERIALS AND METHODS

Plant Materials

All mutations used in this study were in the Col-0 ecotype or C24 background of Arabidopsis (Arabidopsis thaliana), with the exception of co-1, which was in the Landsberg erecta (Ler) ecotype (Putterill et al., 1995). The ztl-1 (Col) line, ztl-3 (Col), and ZTL OX (C24; ZTL overexpressor, C24 ecotype) were described previously (Somers et al., 2004), as was the ZTL overexpressor (ZTL OX [Col]; Han et al., 2004). The elf3-1 mutation was identified using the primers 5′-TGTTGGTCAGTCTTCTCCGA-3′ and 5′-TCCCTACTGTCATTCAAGGG-3′, followed by digestion with HincII. The ELF3 overexpression line (ELF3 OX) is as described previously (Covington et al., 2001).

Construction of Double Mutants

The plasmid pZP221-35S:ZTL-EGFP was transformed into elf3-1 to generate ZTL OX elf3-1 double mutants using standard techniques (Clough and Bent, 1998) and selected lines were confirmed via PCR-based scoring of transformants. Other double mutant combinations (elf3-1 ztl-3 and ELF3 OX ztl-1) were obtained via crosses with the appropriate mutant in the appropriate ecotypic background, with the following exceptions: For imaging experiments, the ZTL OX (C24) elf3-1 line is in a mixed background of C24 and Col ecotypes. This did not affect the arrhythmic phenotype described in Figure 1A, and the results of the period of F2 seedlings of a cross between C24 wild type and Col wild type are shown in Figure 2D as a control for the results of Figure 2B. For the elf3-1 co-1 double mutant, elf3-1 was introgressed twice into the Ler ecotype before mating with co-1 mutants, resulting in a mixed background consisting predominantly of Ler.

Plant Growth Conditions and Rhythm Analysis

Seedlings were grown on Murashige and Skoog medium (GIBCO BRL) + 3% Suc (0.8% agar) under 12-h-light/12-h-dark white fluorescent light (50–60 μmol m−2 s−1) for 7 d, then sprayed with 3 mm luciferin (Biotium) before transfer to constant red light (peak wavelength 670 nm ± 15 nm half-peak bandwidth; Quantum Devices), blue light (Bili Blue [Interlectric] filtered through Rohm and Haas 2424 plexiglass 5 mm thick), or darkness and imaged 25 min every 2 h using a Peltier-cooled CCD slow-scan camera (Nightowl; Berthold Technologies). Postimaging luminescence quantitation used WinLight software (Berthold Technologies). Period estimates were obtained using fast Fourier transform nonlinear least-squares analysis (Plautz et al., 1997). Mean period lengths and associated error metrics were variance weighted and are reported as standard error of the mean (SEM; Millar et al., 1995).

Hypocotyl Length Assays

Seeds were stratified in the dark at 4 C for 4 d, exposed to white light (70 μmol m−2 s−1) for 1 to 2 h on Murashige and Skoog medium (GIBCO BRL) + 3% Suc (0.8% agar), then placed under the appropriate light quality and fluence rate (using varying layers of neutral density filters; Roscolux 397 [Rosco Laboratories]) for 7 to 10 d. Hypocotyl length was measured using SCION Image software.

Flowering-Time Analyses

Seeds were grown under long days (16 h light/8 h dark; 60–70 μmol m−2 s−1) for 7 to 10 d on Murashige and Skoog medium (GIBCO BRL) + 3% Suc (0.8% agar), then transplanted to soil. Total number of rosette and cauline leaves were counted.

Gene Expression Analyses

Seedlings were grown for 7 d (70 μmol m−2 s−1 white fluorescent light) in 16-h-light/8-h-dark cycles and harvested on day 8 at the appropriate times. Total RNA was extracted using Trizol reagent (Invitrogen) according to the manufacturer's instruction. Except in Figure 6B, transcripts of Actin2, CO, FT, FKF1, and GI were quantified by RT-PCR, followed by DNA gel-blot analysis as described previously (Somers et al., 2004). The primer sequences and annealing temperatures used to amplify each gene are as follows: CO, 48°C, 5′-ACGCCATCAGCGAGTTCC-3′ and 5′-AAATGTATGCGTTATGGTTAATGG-3′; FT, 55°C, 5′-ACAACTGGAACAACCTTTGGCAATG-3′ and 5′-ACTACTATAGGCATCATCACCGTTCGTTACTCG-3′; GI, 52°C, 5′-CTGTCTTTCTCCGTTGTTTCACTGT-3′ and 5′-TCATTCCGTTCTTCTCTGTTGTTGG-3′; and FKF1, 55°C, 5′-GTCGTAACTGTCGATTCCTACA-3′ and 5′-ATCTCCAGTGTTCCAGTTATCT-3′. A portion of At3g18780 (ACTIN2; ACT2) was amplified using oligonucleotides 5′-AAAACCACTTACAGAGTTCGTTCG-3′ and 5′-GTTGAACGGAAGGGATTGAGAGT-3′ with the annealing temperature of 55°C and used as an internal control to normalize the amount of cDNA. The exponential range of amplification was empirically determined for each gene, and 18 cycles were used for Actin2; 22 cycles were used for GI, FKF1, and FT; and 23 cycles for CO. For Figure 6B, transcripts of Actin2 and FT were measured by quantitative RT-PCR, essentially as previously described (Mockler et al., 2004). cDNAs were prepared from DNase-treated (Turbo DNAfree, Ambion) RNA samples using the Omniscript RT kit (Qiagen). Oligonucleotide primers were designed using PRIMER EXPRESS V2.0 software (Applied Biosystems) and were as follows: FT, 5′-CCTTTGGCAATGAGATTGTGTG-3′ and 5′-TTCCTGCAGTGGGACTTGG-3′; FD, 5′-TCCCGCGCTAGGAAACAG-3′ and 5′-CCTGCAAGTGAGCAACTTCAAG-3′; and ACT2, 5′-ACCTTTAACTCTCCCGCTATGTATGT-3′ and 5′-GGCACAGTGTGAGACACACCAT-3′. Expression level was calculated based on standard curves constructed for each primer set and normalized to ACT2 in arbitrary units.

Acknowledgments

We thank A.J. Millar for suggesting the extended DD experiments, T. Michael for the FT primer information, and Jung Na for excellent technical assistance.

1

This work was supported by the National Science Foundation (grant nos. MCB–0080090 and IBN–0344377 to D.E.S., and RUI–0215504 to K.A.H.).

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: David E. Somers (somers.24@osu.edu).

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.067173.

References

  1. Abe M, Kobayashi Y, Yamamoto S, Daimon Y, Yamaguchi A, Ikeda Y, Ichinoki H, Notaguchi M, Goto K, Araki T (2005) FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex. Science 309: 1052–1056 [DOI] [PubMed] [Google Scholar]
  2. Alabadi D, Oyama T, Yanovsky MJ, Harmon FG, Mas P, Kay SA (2001) Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science 293: 880–883 [DOI] [PubMed] [Google Scholar]
  3. Briggs WR, Christie JM (2002) Phototropins 1 and 2: versatile plant blue-light receptors. Trends Plant Sci 7: 204–210 [DOI] [PubMed] [Google Scholar]
  4. Cheng P, He Q, Yang Y, Wang L, Liu Y (2003) Functional conservation of light, oxygen, or voltage domains in light sensing. Proc Natl Acad Sci USA 100: 5938–5943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chou ML, Yang CH (1999) Late-flowering genes interact with early-flowering genes to regulate flowering time in Arabidopsis thaliana. Plant Cell Physiol 40: 702–708 [DOI] [PubMed] [Google Scholar]
  6. Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743 [DOI] [PubMed] [Google Scholar]
  7. Covington MF, Panda S, Liu XL, Strayer CA, Wagner DR, Kay SA (2001) Elf3 modulates resetting of the circadian clock in Arabidopsis. Plant Cell 13: 1305–1316 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dowson-Day MJ, Millar AJ (1999) Circadian dysfunction causes aberrant hypocotyl elongation patterns in Arabidopsis. Plant J 17: 63–71 [DOI] [PubMed] [Google Scholar]
  9. Froehlich AC, Liu Y, Loros JJ, Dunlap JC (2002) White Collar-1, a circadian blue light photoreceptor, binding to the frequency promoter. Science 297: 815–819 [DOI] [PubMed] [Google Scholar]
  10. Han L, Mason M, Risseeuw EP, Crosby WL, Somers DE (2004) Formation of an SCF complex is required for proper regulation of circadian timing. Plant J 40: 291–301 [DOI] [PubMed] [Google Scholar]
  11. He Q, Cheng P, Yang Y, Wang L, Gardner KH, Liu Y (2002) White Collar-1, a DNA binding transcription factor and a light sensor. Science 297: 840–843 [DOI] [PubMed] [Google Scholar]
  12. Hicks KA, Albertson TM, Wagner DR (2001) EARLY FLOWERING3 encodes a novel protein that regulates circadian clock function and flowering in Arabidopsis. Plant Cell 13: 1281–1292 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hicks KA, Millar AJ, Carré IA, Somers DE, Straume M, Meeks-Wagner R, Kay SA (1996) Conditional circadian dysfunction of the Arabidopsis early-flowering 3 mutant. Science 274: 790–792 [DOI] [PubMed] [Google Scholar]
  14. Imaizumi T, Schultz TF, Harmon FG, Ho LA, Kay SA (2005) FKF1 F-box protein mediates cyclic degradation of a repressor of CONSTANS in Arabidopsis. Science 309: 293–297 [DOI] [PubMed] [Google Scholar]
  15. Imaizumi T, Tran HG, Swartz TE, Briggs WR, Kay SA (2003) FKF1 is essential for photoperiodic-specific light signalling in Arabidopsis. Nature 426: 302–306 [DOI] [PubMed] [Google Scholar]
  16. Jarillo JA, Capel J, Tang RH, Yang HQ, Alonso JM, Ecker JR, Cashmore AR (2001) An Arabidopsis circadian clock component interacts with both CRY1 and phyB. Nature 410: 487–490 [DOI] [PubMed] [Google Scholar]
  17. Kim WY, Geng R, Somers DE (2003) Circadian phase-specific degradation of the F-box protein ZTL is mediated by the proteasome. Proc Natl Acad Sci USA 100: 4933–4938 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Liu XL, Covington MF, Fankhauser C, Chory J, Wagner DR (2001) ELF3 encodes a circadian clock-regulated nuclear protein that functions in an Arabidopsis PHYB signal transduction pathway. Plant Cell 13: 1293–1304 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Mas P, Alabadi D, Yanovsky MJ, Oyama T, Kay SA (2003. a) Dual role of TOC1 in the control of circadian and photomorphogenic responses in Arabidopsis. Plant Cell 15: 223–236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Mas P, Kim WY, Somers DE, Kay SA (2003. b) Targeted degradation of TOC1 by ZTL modulates circadian function in Arabidopsis thaliana. Nature 426: 567–570 [DOI] [PubMed] [Google Scholar]
  21. McWatters HG, Bastow RM, Hall A, Millar AJ (2000) The ELF3 zeitnehmer regulates light signalling to the circadian clock. Nature 408: 716–720 [DOI] [PubMed] [Google Scholar]
  22. Millar AJ, Straume M, Chory J, Chua N-H, Kay SA (1995) The regulation of circadian period by phototransduction pathways in Arabidopsis. Science 267: 1163–1166 [DOI] [PubMed] [Google Scholar]
  23. Mizoguchi T, Wright L, Fujiwara S, Cremer F, Lee K, Onouchi H, Mouradov A, Fowler S, Kamada H, Putterill J, et al (2005) Distinct roles of GIGANTEA in promoting flowering and regulating circadian rhythms in Arabidopsis. Plant Cell 17: 2255–2270 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mockler TC, Yu X, Shalitin D, Parikh D, Michael TP, Liou J, Huang J, Smith Z, Alonso JM, Ecker JR, et al (2004) Regulation of flowering time in Arabidopsis by K homology domain proteins. Proc Natl Acad Sci USA 101: 12759–12764 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Plautz JD, Straume M, Stanewsky R, Jamison CF, Brandes C, Dowse H, Hall JC, Kay SA (1997) Quantitative analysis of Drosophila period gene transcription in living animals. J Biol Rhythms 12: 204–217 [DOI] [PubMed] [Google Scholar]
  26. Putterill J, Robson F, Lee K, Simon R, Coupland G (1995) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell 80: 847–857 [DOI] [PubMed] [Google Scholar]
  27. Reed JW, Nagpal P, Bastow RM, Solomon KS, Dowson-Day MJ, Elumalai RP, Millar AJ (2000) Independent action of ELF3 and phyB to control hypocotyl elongation and flowering time. Plant Physiol 122: 1149–1160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Somers DE (2001) Clock-associated genes in Arabidopsis: a family affair. Philos Trans R Soc Lond B Biol Sci 356: 1745–1753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Somers DE (2005) ZEITLUPE and the control of circadian timing. In M Wada, K Shimazaki, M Iino, eds, Light Sensing in Plants. Springer, Tokyo, pp 347–354
  30. Somers DE, Kim WY, Geng R (2004) The F-box protein ZEITLUPE confers dosage-dependent control on the circadian clock, photomorphogenesis, and flowering time. Plant Cell 16: 769–782 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Somers DE, Schultz TF, Milnamow M, Kay SA (2000) ZEITLUPE encodes a novel clock-associated PAS protein from Arabidopsis. Cell 101: 319–329 [DOI] [PubMed] [Google Scholar]
  32. Somers DE, Webb AAR, Pearson M, Kay S (1998) The short-period mutant, toc1-1, alters circadian clock regulation of multiple outputs throughout development in Arabidopsis thaliana. Development 125: 485–494 [DOI] [PubMed] [Google Scholar]
  33. Suarez-Lopez P, Wheatley K, Robson F, Onouchi H, Valverde F, Coupland G (2001) CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature 410: 1116–1120 [DOI] [PubMed] [Google Scholar]
  34. Tseng TS, Salome PA, McClung CR, Olszewski NE (2004) SPINDLY and GIGANTEA interact and act in Arabidopsis thaliana pathways involved in light responses, flowering, and rhythms in cotyledon movements. Plant Cell 16: 1550–1563 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Valverde F, Mouradov A, Soppe W, Ravenscroft D, Samach A, Coupland G (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science 303: 1003–1006 [DOI] [PubMed] [Google Scholar]
  36. Wigge PA, Kim MC, Jaeger KE, Busch W, Schmid M, Lohmann JU, Weigel D (2005) Integration of spatial and temporal information during floral induction in Arabidopsis. Science 309: 1056–1059 [DOI] [PubMed] [Google Scholar]
  37. Zagotta MT, Hicks KA, Jacobs CI, Young JC, Hangarter RP, Meeks-Wagner DR (1996) The Arabidopsis ELF3 gene regulates vegetative photomorphogenesis and the photoperiodic induction of flowering. Plant J 10: 691–702 [DOI] [PubMed] [Google Scholar]
  38. Zagotta MT, Shannon S, Jacobs C, Meeks-Wagner DR (1992) Early-flowering mutants of Arabidopsis thaliana. Aust J Plant Physiol 19: 411–418 [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES