Skip to main content
NCBI home page
Molecular Plant logoLink to Molecular Plant
. 2012 Feb 10;5(3):152–161. doi: 10.1093/mp/sss005

GIGANTEA and EARLY FLOWERING 4 in Arabidopsis Exhibit Differential Phase-Specific Genetic Influences over a Diurnal Cycle

Yumi Kim a, Miji Yeom a, Hyunmin Kim a, Junhyun Lim a, Hee Jung Koo b, Daehee Hwang b,c,1, David Somers d,e,1, Hong Gil Nam a,b,d,f,1
PMCID: PMC3355345  PMID: 22328721

Abstract

The endogenous circadian clock regulates many physiological processes related to plant survival and adaptability. GIGANTEA (GI), a clock-associated protein, contributes to the maintenance of circadian period length and amplitude, and also regulates flowering time and hypocotyl growth in response to day length. Similarly, EARLY FLOWERING 4 (ELF4), another clock regulator, also contributes to these processes. However, little is known about either the genetic or molecular interactions between GI and ELF4 in Arabidopsis. In this study, we investigated the genetic interactions between GI and ELF4 in the regulation of circadian clock-controlled outputs. Our mutant analysis shows that GI is epistatic to ELF4 in flowering time determination, while ELF4 is epistatic to GI in hypocotyl growth regulation. Moreover, GI and ELF4 have a synergistic or additive effect on endogenous clock regulation. Gene expression profiling of gi, elf4, and gi elf4 mutants further established that GI and ELF4 have differentially dominant influences on circadian physiological outputs at dusk and dawn, respectively. This phasing of GI and ELF4 influences provides a potential means to achieve diversity in the regulation of circadian physiological outputs, including flowering time and hypocotyl growth.

Keywords: microarray, LHY, endogenous clock, GI, ELF4

INTRODUCTION

The Earth’s rotation is responsible for the day–night cycle. Living organisms recognize the external changes produced by this cycle and generate their own endogenous daily rhythms known as a circadian rhythm. This endogenous rhythm having almost 24-h period can be entrained by day–night cycles and be sustained without external stimuli. Living organisms use circadian rhythms to adapt themselves to the environment (Dodd et al., 2005). The endogenous clock is composed of a series of interlocking molecular feedback loops that are conserved in most organisms (Harmer, 2009; Young and Kay, 2001).

Plant clock recognizes photoperiods and uses daily rhythms in many developmental processes from seedling growth to flowering, as demonstrated in Arabidopsis thaliana (de Montaigu et al., 2010). A recent model for the Arabidopsis clock described a series of multiple interlocking feedback loops referred to as the morning, core, and evening loops. These loops are interlocked in a complex manner: (1) the core loop including TIMING OF CAB EXPRESSION 1 (TOC1), CIRCADIAN CLOCK ASSOCIATED 1 (CCA1), and LATE ELONGATED HYPOCOTYL (LHY) (Alabadi et al., 2001; Locke et al., 2005a, 2005b); (2) the morning loop, inducing PSEUDO RESPONSE REGULATOR 9 (PRR9) and PSEUDO RESPONSE REGULATOR 7 (PRR7), which are linked to CCA1/LHY (Locke et al., 2006; Zeilinger et al., 2006); and (3) the evening loop, including GI and ZEITLUPE (ZTL), which are connected to TOC1 in the core loop (Pokhilko et al., 2010). In addition, EARLY FLOWERING 3 (ELF3) acts as a component of the circadian clock input pathway (McWatters et al., 2000) and EARLY FLOWERING 4 (ELF4) has been suggested as another component in the core loop (Doyle et al., 2002; McWatters et al., 2007).

Genes such as TOC1, CCA1, and LHY were isolated via direct circadian screens. However, GI, ELF3, and ELF4 were identified using non-circadian forward genetic screens, such as for altered flowering time (Hall and McWatters, 2005). GI was initially isolated as a photoperiodic flowering regulator and gi mutants show an aberrant flowering phenotype caused by altered circadian rhythms (Fowler et al., 1999; Park et al., 1999). Although GI encodes a protein with domains of unknown functions, recent reports have suggested that GI stabilizes the photoreceptor protein, ZTL, via the formation of a blue light-dependent complex (Kim et al., 2007). GI also interacts with ELF3, which recruits CONSTITUTIVELY PHOTOMORPHOGENIC1 (COP1), leading to GI protein degradation during the night (Yu et al., 2008). Additionally, GI can be found at CONSTANS (CO) and FLOWERING LOCUS T (FT) promoters where interactions between GI and FLAVIN-BINDING, KELCH REPEAT F-BOX 1 (FKF1) modulate CO mRNA expression through degradation of the CO repressor, CYCLING DOF FACTOR 1 (CDF1) (Sawa et al., 2007). GI interacts with SHORT VEGETATIVE PHASE (SVP), TEMPRANILLO (TEM) 1, and TEM2 in vivo and controls expression at the FT promoter (Sawa and Kay, 2011). Furthermore, GI also functions in hypocotyl growth at the seedling stage via an unknown molecular mechanism (Huq et al., 2000; Nozue et al., 2007).

ELF4-deficient mutants show an early flowering phenotype with increased CO expression, while ELF4 overexpressors exhibit delayed flowering (Doyle et al., 2002; McWatters et al., 2007). ELF4 is also involved in PHYTOCHROME B (PHYB)-dependent seedling growth (Khanna et al., 2003) and a recent study revealed that an ELF3/ELF4/LUX complex binds to PHYTOCHROME INTERACTING FACTOR (PIF) 4 and PIF5 promoters to control their expression (Nusinow et al., 2011). ELF4 is involved in many of the same physiological processes as GI but the ELF4/GI interaction has been rarely investigated. Studies in pea (Pisum sativum) suggest that DIE NEUTRALIS (DNE) and LATE BLOOMER 1 (LATE1), orthologs of ELF4 and GI, respectively, interact genetically to regulate flowering time (Liew et al., 2009). late1 is epistatic to dne in regulating flowering time, but flowering by both DNE and LATE 1 seemed not to regulate CO-like genes in pea (Hecht et al., 2007; Liew et al., 2009).

In this study, we investigated the genetic interactions between GI and ELF4 in the regulation of the circadian clock, flowering time, and hypocotyl growth in Arabidopsis. The results show diverse genetic interactions between GI and ELF4 in regulation of these and other processes. Furthermore, differential dominant influences between GI and ELF4 were confirmed by genome-wide gene expression analysis of gi, elf4, and gi elf4 mutants at dawn and dusk. The results suggest that differences in the functional dominance of GI and ELF4 during a day may provide a means by which a diverse set of circadian physiological outputs can be coordinated.

RESULTS AND DISCUSSION

gi Is Epistatic to elf4 in Photoperiodic Flowering Time Regulation

In plants, flowering represents the transition from vegetative to reproductive development. The determination of flowering time is regulated by multiple pathways including the aging, autonomous, vernalization, gibberellin, ambient temperature, and photoperiodic pathways (Mouradov et al., 2002; Fornara et al., 2010). Plants with defects in maintaining circadian rhythms often show an altered flowering time in response to photoperiod (Schaffer et al., 1998; Park et al., 1999; Strayer et al., 2000). In Arabidopsis, GI and ELF4 mutants have late and early flowering time phenotypes, respectively. The flowering time phenotypes in both mutants were ascribed to the altered circadian rhythms (Park et al., 1999; Doyle et al., 2002). The orthologs of GI and ELF4 in Pea, LATE1, and DNE, respectively, control flowering time through a linear hierarchy in which dne acts upstream of late1 (Liew et al., 2009).

We attempted to confirm whether this interaction is conserved in Arabidopsis using gi-2 and elf4-209 alleles, respectively (Fowler et al., 1999; Kolmos et al., 2009). The elf4-209 mutant (elf4) contains a premature stop codon and produces a 26 amino acid polypeptide (Kolmos et al., 2009). However, the elf4-209 mutant was generated by Targeting Induced Local Lesions IN Genomes (TILLING) so, to exclude the possible effects of the other mutations, we backcrossed elf4-209 with Col-0 (WT) for three generations and then introduced this allele into the gi-2 background to generate gi elf4 double mutants.

The elf4 and gi single mutants showed early- and late-flowering phenotypes, respectively (Figure 1A), as reported previously (Fowler et al., 1999; Doyle et al., 2002). The flowering time in gi elf4 double mutants was similar to that of gi single mutant under LD (Figure 1A). We further quantitatively evaluated the flowering time of all mutant combinations by counting total leaf number (rosette and cauline) at the appearance of the first flower (Figure 1B). The early flowering of elf4 in SD was statistically significant, but not as dramatic as previously described in Ws (Doyle et al., 2002; Khanna et al., 2003). The late-flowering phenotype in gi mutants was much more obvious in LD than that in SD as previously described (Fowler et al., 1999; Park et al., 1999; Mizoguchi et al., 2005). The flowering times in gi elf4 double mutants under both LD and SD were not statistically different from those of gi single mutants. Therefore, the results indicate that gi is epistatic to elf4 in photoperiodic flowering time regulation, thus forming a hierarchy similar to that described in pea (Hecht et al., 2007; Liew et al., 2009).

Figure 1.

Figure 1.

Open in a new tab

Photoperiodic Flowering Time in WT and Mutants.

(A) Whole plant images. Plants were grown under LD for 4 weeks: Col (WT), elf4, gi, and gi elf4 mutants. The scale bar represents 5 cm.

(B) Total numbers of leaves. Total leaves including rosette and cauline leaves were counted at the first flower bloom. White and black bars represent LD and SD, respectively. Data represent the means ± 95% confidence interval (CI) from at least 16 plants. Asterisks indicate statistically significant differences (Col versus elf4; P < 0.01).

(C, E) CO mRNA expression levels under LD (C) and SD (E).

(D, F) FT mRNA expression levels under LD (D) and SD (F). Total RNA was isolated from 7-day-old seedlings, and CO and FT mRNA were measured by quantitative PCR and normalized against Actin 2 (ACT). White and black bars represent day and night, respectively. Data represent the means ± standard error (SE) from biological triplicates.

CO is an integrator in the photoperiodic pathway that accelerates flowering in response to long days, and its expression is under the control of the endogenous clock (Suarez-Lopez et al., 2001). During daylight, the CO protein directly induces FT as a florigen signal, and FT then promotes flowering (Valverde et al., 2004). To confirm and extend the epistatic effects of gi over elf4 in the regulation of flowering time at the molecular level, we examined CO and FT mRNA levels under LD and SD conditions in the single and double mutants. In WT plants, the CO expression level was lower under SD than under LD, and the shoulder shape of CO expression near dusk in LD disappeared under SD (Figure 1C and 1E). The FT expression level in WT was also reduced under SD, compared to that under LD (Figure 1D and 1F). CO and FT expression levels in gi mutants were almost completely eliminated under both LD and SD conditions. The CO and FT expression patterns in gi elf4 double mutants mirrored those in gi single mutants. The data showed a mild increase in expressions of CO and FT in elf4 mutant, relative to WT, in contrast to the data previously reported (Doyle et al., 2002). Nonetheless, these results indicate that gi is epistatic to elf4 in the regulation of CO and FT expression.

elf4 Is Epistatic to gi-2 in Hypocotyl Growth Regulation

Hypocotyl growth is coordinately controlled by external stimuli and the endogenous clock (Dowson-Day and Millar, 1999). The circadian clock regulates the rhythmic growth of seedlings and gates the activity at night. PIF4 and PIF5 are key factors in controlling these phenomena (Nozue et al., 2007). GI and ELF4 also play roles in seedling growth (Huq et al., 2000; Khanna et al., 2003; Nozue et al., 2007). Thus, we examined the genetic interactions of GI and ELF4 in the regulation of hypocotyl growth. When 7-day-old seedlings were grown under LD, elf4 mutants showed significantly longer hypocotyls than WT plants and gi mutants showed slightly longer hypocotyls than the WT (Figure 2A). The hypocotyl length in gi elf4 double mutants was similar to that of elf4 single mutants. When the seedlings were grown under SD, hypocotyl lengths in WT were longer than those under LD (Figure 2B). Similarly to the LD condition, elf4 seedlings under SD had significantly increased hypocotyl lengths, compared to WT, and hypocotyl lengths in gi elf4 double mutants were similar to those in the elf4 single mutants. The results imply that elf4 is epistatic to gi in regulating seedling growth.

Figure 2.

Figure 2.

Open in a new tab

Seedling Growth in WT and Mutants.

(A) Seedling images of Col (WT), elf4, gi, and gi elf4 seedlings. Plants grown under LD for 7 d were imaged. The scale bar represents 1 cm.

(B) Relative hypocotyl lengths under LD and SD. Hypocotyl lengths were measured and normalized to the mean hypocotyl length under continuous dark (DD) conditions. Data represent the means ± 95% CI from at least 20 seedlings.

(C, D) PIF4 mRNA expression levels under LD (C) and SD (D). Total RNA was isolated from 7-day-old seedlings, and PIF4 mRNA expression levels were measured by quantitative PCR and normalized to that of ACT. White and black bars represent day and night, respectively. Data represent the means ± SE from experiments performed in triplicate. Asterisks indicate statistically significant differences (Col versus elf4; P = 0.02 in LD and P = 0.12 in SD).

Hypocotyl growth increases dramatically late at night. Growth at night is controlled by the endogenous clock, which induces PIF4 expression (Nozue et al., 2007). The gi and elf4 mutants exhibited defects in photoperiod dependent hypocotyl growth. To examine the relationship between elf4 and gi in the regulation of hypocotyl growth at the molecular level, we examined PIF4 expression levels under diurnal conditions (Figure 2C and 2D). gi and elf4 mutants showed differential effects on PIF4 expression level during day and night. We first focused on their effects during the night when hypocotyl growth is most apparent. In the elf4 mutant, increased PIF4 expression levels at night were significantly increased under SD (P = 0.02 between WT and elf4 at ZT22; Figure 2D), but not significantly so under LD (P = 0.12 between WT and elf4 at ZT22; Figure 2C), indicating that ELF4 suppresses PIF4 expression mostly at night. In contrast, the PIF4 expression levels in gi mutants during the night were similar to those in WT under LD (Figure 2C), and only marginally higher under SD (Figure 2D). Interestingly, the PIF4 expression levels in gi elf4 double mutants at night were much higher than those in either gi or elf4 single mutants under both photoperiods (Figure 2C and 2D). The elevated expression levels of PIF4 in elf4 appear to be sufficient to accelerate hypocotyl growth, such that the much higher levels of PIF4 in the gi elf4 background do not contribute significantly to increasing hypocotyl elongation (Figure 2A and 2B). Therefore, while our results show that elf4 is epistatically dominant to gi in the regulation of hypocotyl growth, the synergistic effect on PIF4 expression suggests that there is another mechanism governing hypocotyl expansion, independently of PIF4 levels.

GI and ELF4 Function Together in Circadian Regulation

GI is known to function in both the core circadian clock and input pathways (Fowler et al., 1999; Park et al., 1999; Kim et al., 2007), and ELF4 also has a role in the core pathway (Doyle et al., 2002; Kikis et al., 2005; McWatters et al., 2007). To better understand interactions between these two genes in circadian clock regulation, we measured the period and robustness of the clock by a leaf movement assay and a luminescence assay using COLD, CIRCADIAN RHYTHM, and RNA BINDING 2 (CCR2) promoter-luciferase reporters under the free-running conditions.

Plant leaves close at night and open during the daytime repetitively and, after light/dark entrainment, these movements persist under continuous light (LL) or dark (DD) conditions. Compared to WT, gi mutants showed a circadian period shortened by about 1 h. Moreover, they showed an irregular circadian rhythm with reduced robustness as indicated by the relative amplitude error (RAE) values larger than those of WT (Figure 3A and 3B, and Table 1) (Plautz et al., 1997). Similarly, elf4 mutants also exhibited a dramatic loss in robustness, with high RAE values as previously reported (Doyle et al., 2002; McWatters et al., 2007). In the gi elf4 double mutants, the circadian amplitude is further diminished, indicating a synergistic or additive effect of GI and ELF4 on the regulation of circadian rhythmicity (Figure 3B and Table 1). Approximately 50% of the gi elf4 double mutants appear arrhythmic, and thus the period and RAE could not be assessed. The remaining 50% of plants showed irregular rhythms with significantly reduced robustness (Figure 3A and Table 1). These results indicate that GI and ELF4 together contribute strongly to the regulation of circadian rhythmicity.

Figure 3.

Figure 3.

Open in a new tab

Endogenous Clock Activities in WT and Mutants.

(A) Leaf movements in LL. Tip-to-tip distances between first and second leaves were measured using the leaf movement assay (LMA) program.

(B) Relative amplitude errors (RAE) analyzed by FTT-NLLS.

(C, E) CCR2 promoter activities in LL (C) and DD (E). Luminescence intensities were measured every hour, and the absolute luminescence intensities were normalized to the mean intensity in each background. Data represent the means ± SE.

(D, F) RAEs analyzed by FTT-NLLS. Period lengths computed from each experiment and statistics are shown in Table 1.

(G, H) LHY expression levels in LD (G) and SD (H). Total RNA was isolated from 7-day-old seedlings, and LHY mRNA levels were measured by quantitative PCR and then normalized to that of ACT. White and black bars represent day and night, respectively. Data represent the means ± SE from experiments performed in triplicate.

Table 1.

Period Lengths and Relative Amplitude Errors in gi elf4 Double Mutants.

Period (h) RAE Number of rhythmic (total) plants
Leaf movement assay
Col 24.64 ± 0.26 0.23 ± 0.02 16 (17)
gi 23.20 ± 0.27 0.38 ± 0.04 18 (18)
elf4 24.50 ± 0.93 0.52 ± 0.07 16 (18)
gi elf4 24.55 ± 2.01 0.61 ± 0.06 9 (17)
P-value 0.49 <0.0001
CCR2pro::LUC activity (LL)
Col 24.95 ± 0.13 0.33 ± 0.02 16 (16)
gi 27.85 ± 0.37 0.60 ± 0.03 13 (16)
elf4 26.82 ± 0.88 0.71 ± 0.09 8 (16)
gi elf4 30.12 ± 0.87 0.82 ± 0.05 9 (21)
P-value <0.0001 <0.0001
CCR2pro::LUC activity (DD)
Col 26.34 ± 0.80 0.41 ± 0.03 16 (16)
gi 29.29 ± 1.65 0.37 ± 0.03 8 (16)
elf4 31.22 ± 0.77 0.52 ± 0.06 10 (16)
gi elf4 30.95 ± 1.06 0.5 ± 0.04 4 (19)
P-value 0.003 0.07
Open in a new tab

Period lengths and Relative Amplitude Errors (RAEs) were statically analyzed in Col, gi, elf4, and gi elf4 using one-way ANOVA and presented in the table. Data represent mean ± 95% CI.

To evaluate the collective regulation of circadian rhythm at the molecular level by GI and ELF4, we further measured the endogenous clock activity by tracking the promoter activity of the clock-controlled gene, CCR2, in WT, gi, elf4, and gi elf4 plants. In constant light, the CCR2 promoter showed maximum activity at dusk in WT plants (Figure 3C). However, both gi and elf4 mutants showed significantly irregular rhythms with reduced robustness (Figure 3D), consistent with the reduced robustness shown by leaf movement (Figure 3A). Furthermore, gi elf4 double mutants had a much longer period and showed a further severe loss of robustness than either the gi or elf4 single mutants (Figure 3C and 3D, and Table 1). In constant dark (DD), the robustness of rhythmic plants was not statistically different among the three types of mutants (Figure 3E and 3F, and Table 1). However, the numbers of rhythmic plants were reduced in both gi and elf4 mutants, compared to those of WT plants, and further reduced in gi elf4 double mutants (Figure 3F and Table 1). Also, both gi, elf4, and gi elf4 mutants exhibited longer period lengths, compared to WT (Figure 3E and 3F, and Table 1), which were not significantly different among the three types of mutants. The results (the RAE values under LL and the reduced numbers of rhythmic plants under DD) indicate that GI and ELF4 have the synergistic or additive effect on the circadian rhythmicity of CCR2 promoter activity in both LL and DD conditions.

Finally, we measured the mRNA expression levels of a core circadian clock gene, LHY, under light/dark cycles to further examine the molecular effects of GI and ELF4 on circadian rhythmicity (Figure 3G and 3H). LHY is a core component of the central oscillators of the Arabidopsis circadian clock in the morning feedback loop (Alabadi et al., 2001; Locke et al., 2006; Zeilinger et al., 2006). LHY expression in gi and elf4 mutants was lower than that of WT as previously reported (Park et al., 1999; Khanna et al., 2003; Mizoguchi et al., 2005; McWatters et al., 2007), and it decreased further to near undetectable levels in gi elf4 double mutants (Figure 3G and 3H). Taken together, all the data above collectively indicate that GI and ELF4 function either additively or synergistically in Arabidopsis circadian clock regulation.

Gene Expression Profiling Reveals Differential Temporal Predominance of GI and ELF4 on the Regulation of Flowering Time, Hypocotyl Growth, and Circadian Rhythmicity

Recent studies revealed that GI induces CO at dusk (Sawa et al., 2007) and ELF3/ELF4/LUX complex directly suppresses PIF4/PIF5 expression at night (Nusinow et al., 2011). These results imply that flowering time regulation by GI is modulated during the daytime and hypocotyl growth is regulated by ELF4 complex during the night. Based on these results, we hypothesized that the genetic interactions between GI and ELF4 could be circadian phase-dependent and that the influence of the two genes, relative to each other, might vary throughout a day. To test this hypothesis, we performed gene expression profiling of WT, gi, elf4, and gi elf4 plants at dawn (ZT1) and dusk (ZT16). Plants grown under LD for 7 d were used. For each time point, we first identified differentially expressed genes (DEGs; FDR < 0.01) between each type of mutant and WT using a previously reported integrative method (Lee et al., 2010) (see ‘Methods’). 1482 and 1766 DEGs at ZT1 and ZT16, respectively, were identified from the three comparisons of (1) elf4 versus WT (684 and 785 DEGs at ZT1 and ZT16, respectively), (2) gi versus WT (570 and 723 DEGs at ZT1 and ZT16, respectively), and (3) gi elf4 versus WT (918 and 969 DEGs at ZT1 and ZT16, respectively) (Supplemental Figure 1A and Supplemental Table 1). At each time point, the DEGs were categorized into the 26 groups based on their differential expression patterns (Supplemental Figure 1B). Among the 26 groups, three pairs of groups were selected that showed epistatic and synergistic interactions between gi and elf4. Figure 4 shows the three groups of genes representing gi epistasis (Figure 4A), elf4 epistasis (Figure 4B), and the synergistic effects of gi and elf4 (Figure 4C), respectively. Assessment of the relative epistatic interactions based on differential expression patterns was scored as done for flowering time (Figure 1A), hypocotyl growth (Figure 2A), and circadian rhythmicity (Figure 3A). For example, a group of genes with the same expression changes in gi versus WT and gi elf4 versus WT, but with no expression change in elf4 versus WT was considered to indicate that gi is epistatic to elf4 for that particular gene’s expression (Figure 4A).

Figure 4.

Figure 4.

Open in a new tab

Differential Expression Patterns between Col-0 and Each of elf4, gi, and gi elf4 Mutants.

(A) Gene expression patterns representing the relative dominance of gi to elf4.

(B) Gene expression patterns representing the relative dominance of elf4 to gi.

(C) Gene expression patterns reflecting the synergistic effects of gi and elf4 mutations. Red and green colors represent log2-fold-changes for up- and down-regulation in gi, elf4, and gi elf4 mutants relative to WT, respectively. Patterns were numbered from one to six according to their differential expression. The number of genes in each pattern is denoted in the box. In each pattern, the genes associated with the regulations of flowering time, hypocotyl growth, and circadian rhythmicity.

The three groups of genes support the relative genetic relationships between GI and ELF4 as presented above in the regulation of flowering time, hypocotyl growth, and circadian rhythmicity. First, 91 genes were epistatically regulated by GI at ZT1 and the number of GI-regulated genes is greatly increased up to 301 genes at ZT16, reflecting that the influence of GI relative to ELF4 is more significant at dusk than at dawn. Patterns 1 and 2 at ZT16 contain many genes involved in flowering and photosynthesis pathways, such as FKF1, CO, FT, and LHCB2.3 (Figure 4A). Enrichment analysis of Gene Ontology Biological Processes (GOBPs) using the Database for Annotation, Visualization and Integrated Discovery (DAVID) (Huang da et al., 2009) also showed that ‘flower development’ is significantly represented by the genes in Pattern 2 at ZT16 (Figure 5B). These data suggest that GI plays a stronger role than ELF4 in the regulation of dusk-active genes required for flowering.

Figure 5.

Figure 5.

Open in a new tab

Biological Processes Represented by the Genes in the Six Patterns at Dawn and Dusk.

(A, B) The heat maps of the enrichment scores. A high enrichment score (see ‘Methods’) in a coordinate (x-axis and y-axis) of the heat map indicates that the corresponding GOBP (y-axis) is significantly represented by the genes in the corresponding pattern (x-axis). The red color represents the enrichment score as described in the color bar. Blue letters indicate GOBPs associated with the regulations of flowering time, hypocotyl growth, and circadian rhythmicity.

Second, 193 genes were epistatically regulated by ELF4 at ZT1 while that number is decreased by about 30% to 129 genes at ZT16. Patterns 3 and 4 include the genes involved in circadian clock and light signaling pathways, such as TOC1, LUX, HFR1, and PIF4 (Figure 4B). GOBP enrichment analysis also showed that the GOBP term ‘response to light stimulus’ that is related to hypocotyl growth is significantly represented by the genes in Pattern 3 at ZT1 (Figure 5A). These data suggest that ELF4 has a stronger influence, relative to GI, in the regulation of seedling growth at dawn.

Third, 180 and 119 genes at ZT1 and ZT16 were regulated synergistically/additively by GI and ELF4, respectively (Figure 4C). These patterns contain clock-regulated genes, such as PRR5, CCL, LHY, PIF7, PIF4, and CABs (Figure 4C). GOBP enrichment analysis further showed that the GOBP term ‘response to light stimulus’ that is related to circadian rhythm is significantly represented by the genes in Pattern 5 at ZT16 and this biological process is also highly enriched in both Pattern 1 and Pattern 3 (Figure 5B). These data suggest that GI and ELF4 function synergistically at both dawn and dusk in the regulation of circadian rhythmicity. Taken together, the results suggest that the relative functional relationship between GI and ELF4 in the regulation of a variety of physiological processes varies over time along a diurnal cycle. Since expression of both genes is evening-phased (David et al., 2006; Kim et al., 2007; Nusinow et al., 2011), the observation that the mutants affect both evening and morning-phased processes indicates that ELF4 and GI play both repressive and inductive roles in plant development and physiology.

CONCLUSIONS

The endogenous clock in plants regulates diverse circadian physiological outputs. How the clock achieves this diversity still remains elusive. To answer this question, complex genetic interactions between clock genes need to be elucidated. In this study, we investigated genetic interactions between two clock genes, GI and ELF4, in the regulation of flowering time, hypocotyl growth, and circadian rhythmicity. The results showed that GI is epistatic to ELF4 in the regulation of flowering time, whereas ELF4 is epistatic to GI in the regulation of hypocotyl growth. In addition, GI and ELF4 have a synergistic effect on the regulation of the circadian clock. Gene expression profiling of gi, elf4, and gi elf4 mutants further confirmed our qPCR results (Supplemental Figure 2) and extended our analysis to elucidate the differential regulation of GI and ELF4 on circadian outputs at the systems level. Our results further indicate that the relative differential influences of GI and ELF4 on a wide range of processes are achieved in a time-dependent manner: (1) GI is epistatic to ELF4 predominately at dusk, (2) ELF4 is epistatic to GI predominately at dawn, and (3) GI and ELF4 collectively regulate circadian rhythmicity at both dawn and dusk. In summary, our results suggest that the differential relative influences of GI and ELF4 over a diurnal cycle may provide a means to coordinate the regulation of diverse circadian outputs.

METHODS

Plant Materials

We generated gi-2 mutants as previously described (Koornneef et al., 1991). To generate elf4 mutants, elf4-209 (Col) (Kolmos et al., 2009) was backcrossed three times to remove other possible mutations with the standard A. thaliana laboratory strain, Col-0 (WT), and selected based on long hypocotyl phenotype. We then introduced elf4-209 into the gi-2 background and selected gi-2 elf4-209 double mutants from F3 segregating lines using PCR. The elf4-209 mutation was identified using primers 5'-AGG CAG AGC AGG GAG AGC CAC CGG CGA T-3' and 5'-CTT CCA TGG AGC TCT AGT TCC GGC AGC-3', followed by digestion with XcmI; in this reaction, the elf4-209 mutation remained undigested. The gi-2 mutants were isolated using PCR as described previously (Park et al., 1999).

Flowering Time Measurement

To measure day-length-dependent flowering time, seeds were sown on soil following stratification (4°C for 2 d). Plants were grown under either long-day (LD) (16L/8D) or short-day (SD) conditions (10L/14D). Flowering time was measured by counting the number of rosette and cauline leaves when the first flower opened. The data represent the means ± 95% confidence interval (CI) from over 10 plants.

Seedling Growth Measurement

To measure photoperiod-dependent seedling growth, seeds were sown on soil following stratification (4°C for 2 d). Plants were grown under either LD or SD for 7 d. Hypocotyl lengths were measured by Scion Image software (Scion Corp. Frederick, MD, USA). The data represent the means ± 95% CI from over 15 seedlings.

Measurements of Endogenous Clock Activity

To measure leaf movement as a metric of endogenous clock activity, 10-day-old seedlings grown under 12L/12D were transferred to constant light (LL), and images were taken every hour for 7 d. Leaf movement was determined by measuring the tip-to-tip distance between the first and second leaves using Leaf Movement Analysis (LMA) software. The data represent the means ± 95% CI from plants described in Table 1. Alternatively, we measured the clock activity by measuring activities of the COLD-CIRCADIAN RHYTHM-RNA BINDING (CCR2) gene promoter. Seven-day-old seedlings were transferred to 96-well microplates containing 500 μM luciferin (SYNCHEM, Felsberg/Altenburg, Germany), and images were taken every hour for 4 d. Luminescence intensities from each plant were imported into the Biological Rhythms Analysis Software System (BRASS) (Southern and Millar, 2005), and periods were computed using the FFT-NLLS suite (Plautz et al., 1997).

Measurement of mRNA Expression Levels

To measure CO and FLOWERING LOCUS T (FT) mRNA expression levels, 7-day-old seedlings, grown under LD or SD, were harvested in liquid nitrogen every 3 h. Total mRNA was extracted using WelPrep™ (JOIN BIO-INNOVATION, Daegu, Korea), and DNA was digested by treatment with DNase I (Ambion, Austin, TX). For each sample, 0.75 μg of total mRNA was reverse-transcribed using ImProm II Reverse Transcriptase (Promega, Madison, WI). The amounts of the transcripts were measured by Real-Time PCR, using SYBR Premix Extaq (Takara, Shuzo, Kyoto, Japan) and the ABI 7300 Real-Time PCR system (Applied Biosystems, Foster city, CA). The following primers previously reported were used: CO, FT, LHY, and PIF4 primers (Mockler et al., 2004); GI primers (Edwards et al., 2006); ELF4 primers (Kim et al., 2008); and ACT primers (Hall et al., 2003).

Microarray Experiments

Seven-day-old seedlings grown under LD were harvested at ZT1 and ZT16. Total RNA was isolated and used for microarray experiments. The integrity of total RNA was evaluated using the Bioanalyzer 2100 (Agilent, Santa Clara, CA). The RNA integrity in all samples was sufficient for gene expression analysis (RNA integrity number ≥ 9.5). RNA was reverse-transcribed, amplified, and then hybridized into Customized Arabidopsis Gene Expression Microarrays, containing 43 803 probes corresponding to 25 945 annotated genes, according to the standard Agilent protocols. The levels of mRNAs were measured for three biological replicates of each transgenic plant (wild-type (Col-0), gi, elf4, and gi elf4) at ZT1 and ZT16. Log2-intensities were normalized using quantile normalization (Bolstad et al., 2003).

Identification of Differentially Expressed Genes

Using the normalized intensities, differentially expressed genes (DEGs) between Col-0 and each mutant were determined using the previously reported integrative method: (1) two independent tests were performed: t-test and log2-median-ratio test; (2) false discovery rates (FDRs) from each test were computed using an empirical distribution of the null hypothesis that the means of the genes are not different, which was obtained from random permutations of the samples; (3) the individual FDRs were combined to compute the overall FDR using Stouffer’s method (Hwang et al., 2005); and (4) DEGs were selected as the genes with the overall FDR < 0.01. Finally, potential false positives were further removed by selecting the DEGs fold-change ≥ 97.5 percentiles or ≤ 2.5 percentiles of fold-changes obtained from the randomly permuted samples (i.e. the empirical distribution of fold-changes).

Functional Enrichment Analysis

For the genes identified for each pattern, we performed the enrichment analysis of GO biological processes (GOBPs) using DAVID software (Huang da et al., 2009). Among all the resulting GOBPs, we selected those that were significantly enriched (P < 0.05) in at least one of the six patterns at ZT1 or ZT16. The enrichment scores were computed as –log10(P) where P is the P-value from the enrichment analysis.

SUPPLEMENTARY DATA

Supplementary Data are available at Molecular Plant Online.

FUNDING

This work was supported by grants from the National Research Foundation of Korea (NRF) funded by the Korea government (MEST): the National Researcher Support Program (No. 20100020417), the National Core Research Center (No. 2009–0091504), the World Class University Program (No. R31-2008–000–10105–0), and the National Institute of Health (R01GM093285).

Supplementary Material

Supplementary Data
supp_5_3_152__index.html (1.3KB, html)

Acknowledgments

We thank K.H. Suh, Y.S. Park, and B.H. Kim for technical assistance. No conflict of interest declared.

References

  1. 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;293:880–883. doi: 10.1126/science.1061320. [DOI] [PubMed] [Google Scholar]
  2. Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics. 2003;19:185–193. doi: 10.1093/bioinformatics/19.2.185. [DOI] [PubMed] [Google Scholar]
  3. David KM, Armbruster U, Tama N, Putterill J. Arabidopsis GIGANTEA protein is post-transcriptionally regulated by light and dark. FEBS Lett. 2006;580:1193–1197. doi: 10.1016/j.febslet.2006.01.016. [DOI] [PubMed] [Google Scholar]
  4. de Montaigu A, Toth R, Coupland G. Plant development goes like clockwork. Trends Genet. 2010;26:296–306. doi: 10.1016/j.tig.2010.04.003. [DOI] [PubMed] [Google Scholar]
  5. Dodd AN, et al. Plant circadian clocks increase photosynthesis, growth, survival, and competitive advantage. Science. 2005;309:630–633. doi: 10.1126/science.1115581. [DOI] [PubMed] [Google Scholar]
  6. Dowson-Day MJ, Millar AJ. Circadian dysfunction causes aberrant hypocotyl elongation patterns in Arabidopsis. Plant J. 1999;17:63–71. doi: 10.1046/j.1365-313x.1999.00353.x. [DOI] [PubMed] [Google Scholar]
  7. Doyle MR, et al. The ELF4 gene controls circadian rhythms and flowering time in Arabidopsis thaliana. Nature. 2002;419:74–77. doi: 10.1038/nature00954. [DOI] [PubMed] [Google Scholar]
  8. Edwards KD, et al. FLOWERING LOCUS C mediates natural variation in the high-temperature response of the Arabidopsis circadian clock. Plant Cell. 2006;18:639–650. doi: 10.1105/tpc.105.038315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fornara F, de Montaigu A, Coupland G. SnapShot: control of flowering in Arabidopsis. Cell. doi: 10.1016/j.cell.2010.04.024. 141, 550, 550, e551-e552. [DOI] [PubMed] [Google Scholar]
  10. Fowler S, et al. GIGANTEA: a circadian clock-controlled gene that regulates photoperiodic flowering in Arabidopsis and encodes a protein with several possible membrane-spanning domains. EMBO J. 1999;18:4679–4688. doi: 10.1093/emboj/18.17.4679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hall A, et al. The TIME FOR COFFEE gene maintains the amplitude and timing of Arabidopsis circadian clocks. Plant Cell. 2003;15:2719–2729. doi: 10.1105/tpc.013730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hall AJW, McWatters H. Endogenous Plant Rhythms. Oxford and Ames, IA: Blackwell Pub; 2005. [Google Scholar]
  13. Harmer SL. The circadian system in higher plants. Annu. Rev. Plant Biol. 2009;60:357–377. doi: 10.1146/annurev.arplant.043008.092054. [DOI] [PubMed] [Google Scholar]
  14. Hecht V, et al. Pea LATE BLOOMER1 is a GIGANTEA ortholog with roles in photoperiodic flowering, deetiolation, and transcriptional regulation of circadian clock gene homologs. Plant Physiol. 2007;144:648–661. doi: 10.1104/pp.107.096818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature Protocols. 2009;4:44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
  16. Huq E, Tepperman JM, Quail PH. GIGANTEA is a nuclear protein involved in phytochrome signaling in Arabidopsis. Proc. Natl Acad. Sci. U S A. 2000;97:9789–9794. doi: 10.1073/pnas.170283997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hwang D, et al. A data integration methodology for systems biology. Proc. Natl Acad. Sci. U S A. 2005;102:17296–17301. doi: 10.1073/pnas.0508647102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Khanna R, Kikis EA, Quail PH. EARLY FLOWERING 4 functions in phytochrome B-regulated seedling de-etiolation. Plant Physiol. 2003;133:1530–1538. doi: 10.1104/pp.103.030007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kikis EA, Khanna R, Quail PH. ELF4 is a phytochrome-regulated component of a negative-feedback loop involving the central oscillator components CCA1 and LHY. Plant J. 2005;44:300–313. doi: 10.1111/j.1365-313X.2005.02531.x. [DOI] [PubMed] [Google Scholar]
  20. Kim J, Kim Y, Yeom M, Kim JH, Nam HG. FIONA1 is essential for regulating period length in the Arabidopsis circadian clock. Plant Cell. 2008;20:307–319. doi: 10.1105/tpc.107.055715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kim WY, et al. ZEITLUPE is a circadian photoreceptor stabilized by GIGANTEA in blue light. Nature. 2007;449:356–360. doi: 10.1038/nature06132. [DOI] [PubMed] [Google Scholar]
  22. Kolmos E, et al. Integrating ELF4 into the circadian system through combined structural and functional studies. HFSP J. 2009;3:350–366. doi: 10.2976/1.3218766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Koornneef M, Hanhart CJ, van der Veen JH. A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana. Mol. Gen. Genet. 1991;229:57–66. doi: 10.1007/BF00264213. [DOI] [PubMed] [Google Scholar]
  24. Lee HJ, et al. Direct transfer of alpha-synuclein from neuron to astroglia causes inflammatory responses in synucleinopathies. J. Biol. Chem. 2010;285:9262–9272. doi: 10.1074/jbc.M109.081125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liew LC, et al. DIE NEUTRALIS and LATE BLOOMER 1 contribute to regulation of the pea circadian clock. Plant Cell. 2009;21:3198–3211. doi: 10.1105/tpc.109.067223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Locke JC, et al. Experimental validation of a predicted feedback loop in the multi-oscillator clock of Arabidopsis thaliana. Mol. Syst. Biol. 2006;2:59. doi: 10.1038/msb4100102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Locke JC, et al. Extension of a genetic network model by iterative experimentation and mathematical analysis. Mol. Syst. Biol. 1, 0013. 2005a doi: 10.1038/msb4100018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Locke JC, Millar AJ, Turner MS. Modelling genetic networks with noisy and varied experimental data: the circadian clock in Arabidopsis thaliana. J. Theor. Biol. 2005b;234:383–393. doi: 10.1016/j.jtbi.2004.11.038. [DOI] [PubMed] [Google Scholar]
  29. McWatters HG, Bastow RM, Hall A, Millar AJ. The ELF3 zeitnehmer regulates light signalling to the circadian clock. Nature. 2000;408:716–720. doi: 10.1038/35047079. [DOI] [PubMed] [Google Scholar]
  30. McWatters HG, et al. ELF4 is required for oscillatory properties of the circadian clock. Plant Physiol. 2007;144:391–401. doi: 10.1104/pp.107.096206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mizoguchi T, et al. Distinct roles of GIGANTEA in promoting flowering and regulating circadian rhythms in Arabidopsis. Plant Cell. 2005;17:2255–2270. doi: 10.1105/tpc.105.033464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mockler TC, et al. Regulation of flowering time in Arabidopsis by K homology domain proteins. Proc. Natl Acad. Sci. U S A. 2004;101:12759–12764. doi: 10.1073/pnas.0404552101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mouradov A, Cremer F, Coupland G. Control of flowering time: interacting pathways as a basis for diversity. Plant Cell. 2002;14 Suppl:S111–S130. doi: 10.1105/tpc.001362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Nozue K, et al. Rhythmic growth explained by coincidence between internal and external cues. Nature. 2007;448:358–361. doi: 10.1038/nature05946. [DOI] [PubMed] [Google Scholar]
  35. Nusinow DA, et al. The ELF4-ELF3-LUX complex links the circadian clock to diurnal control of hypocotyl growth. Nature. 2011;475:398–402. doi: 10.1038/nature10182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Park DH, et al. Control of circadian rhythms and photoperiodic flowering by the Arabidopsis GIGANTEA gene. Science. 1999;285:1579–1582. doi: 10.1126/science.285.5433.1579. [DOI] [PubMed] [Google Scholar]
  37. Plautz JD, et al. Quantitative analysis of Drosophila period gene transcription in living animals. J. Biol. Rhythms. 1997;12:204–217. doi: 10.1177/074873049701200302. [DOI] [PubMed] [Google Scholar]
  38. Pokhilko A, et al. Data assimilation constrains new connections and components in a complex, eukaryotic circadian clock model. Mol. Syst. Biol. 2010;6:416. doi: 10.1038/msb.2010.69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sawa M, Kay SA. GIGANTEA directly activates Flowering Locus T in Arabidopsis thaliana. Proc. Natl Acad. Sci. U S A. 2011;108:11698–11703. doi: 10.1073/pnas.1106771108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sawa M, Nusinow DA, Kay SA, Imaizumi T. FKF1 and GIGANTEA complex formation is required for day-length measurement in Arabidopsis. Science. 2007;318:261–265. doi: 10.1126/science.1146994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Schaffer R, et al. The late elongated hypocotyl mutation of Arabidopsis disrupts circadian rhythms and the photoperiodic control of flowering. Cell. 1998;93:1219–1229. doi: 10.1016/s0092-8674(00)81465-8. [DOI] [PubMed] [Google Scholar]
  42. Southern MM, Millar AJ. Circadian genetics in the model higher plant, Arabidopsis thaliana. Methods Enzymol. 2005;393:23–35. doi: 10.1016/S0076-6879(05)93002-4. [DOI] [PubMed] [Google Scholar]
  43. Strayer C, et al. Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog. Science. 2000;289:768–771. doi: 10.1126/science.289.5480.768. [DOI] [PubMed] [Google Scholar]
  44. Suarez-Lopez P, Wheatley K, Robson F, Onouchi H, Valverde F, Coupland G. CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature. 2001;410:1116–1120. doi: 10.1038/35074138. [DOI] [PubMed] [Google Scholar]
  45. Valverde F, Mouradov A, Soppe W, Ravenscroft D, Samach A, Coupland G. Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science. 2004;303:1003–1006. doi: 10.1126/science.1091761. [DOI] [PubMed] [Google Scholar]
  46. Young MW, Kay SA. Time zones: a comparative genetics of circadian clocks. Nat. Rev. Genet. 2001;2:702–715. doi: 10.1038/35088576. [DOI] [PubMed] [Google Scholar]
  47. Yu JW, et al. COP1 and ELF3 control circadian function and photoperiodic flowering by regulating GI stability. Mol. Cell. 2008;32:617–630. doi: 10.1016/j.molcel.2008.09.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Zeilinger MN, Farre EM, Taylor SR, Kay SA, Doyle FJ., III A novel computational model of the circadian clock in Arabidopsis that incorporates PRR7 and PRR9. Mol. Syst. Biol. 2006;2:58. doi: 10.1038/msb4100101. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_5_3_152__index.html (1.3KB, html)
supp_sss005_00009464_file001.tif (4.1MB, tif)
supp_sss005_00009464_file002.tif (1.5MB, tif)
supp_sss005_00009464_file003.xlsx (16MB, xlsx)
supp_sss005_00009464_file004.doc (11.6KB, doc)

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

RESOURCES