ELF4 as a Central Gene in the Circadian Clock

Abstract

The light-dark cycle of the environment serves as one of the major Zeitgebers in entrainment of the circadian clock. The circadian system consists of interconnected feedback loops in which the CCA1/LHY-TOC1 loop has a central position. Genetic analyses of the elf4 mutant suggested that it is a positive regulator of CCA1 and LHY expression. Recently, we refined the mode-of-action of ELF4 in entrainment of the clock, and here hypothesize that ELF4 expression is interlocked with the CCA1/LHY-TOC1 loop.

The circadian clock is an internal pacemaker. Primary qualities of a circadian clock include anticipation of daily events at dawn and dusk and a free-running rhythmicity of approximately 24 hours. In addition, circadian organisms can set internal rhythms to match the environmental light-dark cycle. This adjustment occurs daily and is known as entrainment.

Various groups have arranged the molecular components of the Arabidopsis thaliana circadian system into a model of transcription/translation feedback loops. Each loop consists of elements specific to certain time points (phases) of the 24 h day. In the most central loop, the expression peak times of the two MYB-like transcription factors CCA1 and LHY are in the morning, and the pseudo response regulator TOC1 peaks in the evening.^1–3 This loop is required for sustaining rhythms.⁴ Several other clock genes—GI, ELF3, ELF4, PRR7 and PRR9—are known to be important for proper periodicity.^5–11 Recently, mathematical modeling has assisted expression analyses and expanded the understanding of the circadian system.¹² Two additional feedback loops containing GI and PRR7-PRR9, respectively, were proposed to interlock with the CCA1/LHY-TOC1 feedback loop.^13,14 In these studies, the position of the clock-controlling gene ELF4 was not fully defined. In previous studies, it was found that ELF4 genetically acts as a positive regulator of CCA1.^5,15 It remained a challenge to further define the mode-of-action of ELF4 in connection to the performance of the circadian system.

We previously showed that ELF4 is a clock gene whose expression peaks in the evening.⁵ ELF4 loss-of-function leads to an impaired photoperiod response, circadian dysfunction (imprecision and arrhythmicity) under free-running conditions, and very low amplitude of the light-induced clock gene CCA1. Thus, ELF4 function was suggested to be important for input of light signaling to the clock, which is the process of clock resetting termed entrainment. Two other papers supported this conclusion.^15,16 What was missing from these efforts was a definition of ELF4's placement within the clock network. Accordingly, we sought to expand ELF4 genetic analyses to address the role of ELF4 in sensing the photoperiodic signals of the environment and its connection to core clock genes.

In the Plant Physiology paper, to which we make this Addendum, we assessed clock performance of genotypes altered in ELF4 expression and connected this to the expression phenotypes of the core clock genes CCA1, LHY, and TOC1 in relation to the light-dark cycle. We found that the elf4 mutant had elevated levels of TOC1 expression implicating ELF4 as a negative regulator of TOC1. In contrast, ELF4 promotes the expression of the morning genes CCA1 and LHY, because the expression of these two genes was markedly attenuated in the elf4 loss-of-function mutant. A reciprocal effect was seen in ELF4 overexpresssion lines: TOC1 levels were low and CCA1 and LHY levels were high. In addition, it was previously reported that ELF4 expression was increased in the cca1 lhy double mutant.¹⁵ Collectively, these expression data leads to a model where ELF4 has a central position in the interconnected feedback loops (Fig. 1). Our conclusion is thus that ELF4 is a core-clock gene working in the evening phase of the circadian cycle to drive morning expression of CCA1 and LHY.

Figure 1.

ELF4 in the three-loop model of the circadian clock. ELF4 expression is interlocked with the CCA1/LHY-TOC1 loop in the three-loop model of the central circadian network.¹³,¹⁴

The interrelationship between ELF4 and TOC1 is still unclear, but as both are evening genes, they appear to act coordinately to control the clock. Under light-dark cycles, we found an early phase of ELF4 expression in the toc1 mutant, in agreement with the short-period phenotype of toc1. The mean expression level of ELF4 under a diurnal cycle was unaltered in toc1. We hypothesize that ELF4 expression would eventually dampen in toc1 lines subjected to an extension of light period, as was seen for ELF4 expression in the cca1 lhy mutant.¹⁵ Thus, it is likely that core-oscillator genes are all codependent. Here, we suggest that ELF4 is a “full” component of the central-clock machinery. That said, reporter levels driven by the ELF4 promoter in the elf4 mutant lead to mean levels of expression. Why ELF4 is not in an autoregulatory loop to regulate its own overall expression level is an open question. As well, the transcripts of ELF4 and TOC1 both dampen in constant darkness and thus how evening genes are required for rhythm persistence in the absence of their transcript accumulation is unknown. We note that the TOC1 profile has a bimodal peak, but ELF4 does not.^2,5 Therefore, we wonder if circadian-controlling differences exist in the fine structure of TOC1 and ELF4 rhythmic expression.

The importance of ELF4 activity in the circadian clock was corroborated by expression phenotypes under light-dark cycles (diurnal conditions). In contrast to wild type, the promoter: luciferase profiles in elf4 were driven by “lights on” and “lights off.” Furthermore, elf4 lost anticipation of the phase of the environmental light-dark cycle and displayed an unusual, fast-resetting behavior after a simulated “jet lag.” The light-induced peaks of CCA1 and LHY expression can directly explain these diurnal patterns of gene expression in elf4 confirming the position of ELF4 in the central loop of the clock. In addition, the rhythm of gene expression of several clock markers in an ELF4-overexpression genotype displayed a long period. This suggests to us that ELF4 functions in a dose-dependent manner as a repressor of periodicity. This overexpression-phenotype is remarkably different from that seen in other clock-gene overexpression lines (e.g., CCA1-ox, lhy-1 and TOC1-ox),^13,17,18 and might indicate differences in post-translational performance among the central clock elements.

Analysis of overexpression lines has been supplemented by loss-of-function studies and it was found that mutations in the clock-oscillator genes differ in phenotype (arrhythmicity vs. short period).^2,19 Notably, when released into continuous light, the clock does not arrest in the cca1 lhy double, and in the elf4 single mutant, there is no arrest until after a half-full circadian cycle.^5,19 This can be interpreted as that other “cogwheels” of the clock continue for a limited period of time when one wheel is blocked. Our idea is that ELF4 function is analogous to an hourglass, where the oscillator stops after a fixed amount of time (after 12 h, subjective dusk) in absence of ELF4 activity. Subsequently, the clock enters an arrhythmic state where the amplitude of output rhythms is lost, and this is probably due to the low levels of CCA1 and LHY expression in elf4. An unknown factor (X) was modeled to function as an inducer of CCA1 and LHY expression.^12,13 Further studies are needed to elucidate whether ELF4 possess some features of X (e.g., the effect of TOC1 on ELF4).

Various studies implicate ELF4 in distinct input nodes of light signaling to the oscillator. In an earlier report, increased ELF4 expression was found in the elf3 mutant and therefore ELF3 was defined as a negative regulator of ELF4 expression.¹⁵ We found that the elf4 mutant had a minor role in gating the acute induction of CAB expression, and this was around subjective dusk. This means that both ELF3 and ELF4 have roles in controlling light input to the clock.²⁰ Similarly, elf4 arrhythmicity is evident under all wavelengths of light (McWatters H, unpublished), supporting the concept of ELF4 in both light signaling and in entrainment. Furthermore, ELF4 acts predominantly in response to the light-dark Zeitgeber, as elf4 mutants can entrain (albeit weakly) to temperature cycles. Temperature as an entrainment cue remains to be tested for many of the clock genes. PRR7 and PRR9 have been found to mediate in temperature signaling to the clock,²¹ and in another aspect of clock buffering, CCA1 and LHY were found to participate in specific roles of temperature compensation.²² The poor free-running rhythm of elf4 after entrainment to temperature cycles indicates that more is yet to be learned about the role of ELF4 in perception of environmental Zeitgeber signals.

The timing of dawn and dusk, which predictably fluctuates in most environments over the year, represents the fixed points for clock entrainment. We provided results supporting the idea that ELF4 has a central function in the circadian clock and in day-length perception. In future studies, the post-translational relationship among and within the CCA1/LHY-TOC1/ELF4 loops should be explored in order to gain a more detailed view of the feedback mechanism of the circadian clock.

Addendum to: McWatters HG, Kolmos E, Hall A, Doyle MR, Amasino RM, Gyula P, Nagy F, Millar AJ, Davis SJ. ELF4 is Required for Oscillatory Properties of the Circadian Clock. Plant Physiol. 2007;144:391–401. doi: 10.1104/pp.107.096206.