Special Reviews

Circadian Clocks in Daily and Seasonal Control of Development

See allHide authors and affiliations

Science  18 Jul 2003:
Vol. 301, Issue 5631, pp. 326-328
DOI: 10.1126/science.1085935


The rotation of the earth results in regular changes in the light environment, and organisms have evolved a molecular oscillator that allows them to anticipate these changes. This daily molecular oscillator, known as the circadian clock, regulates a diverse array of physiologies across a wide variety of organisms. This review highlights a few of the insights we have into circadian clock regulation of development, in both plants and animals. A common thread linking plants and animals is the use of the circadian clock to sense changes in day length and to mediate a diverse number of photoperiodic responses.

Plants are intimately associated with their environments and use light as a major signal to modulate growth and development throughout their life cycles. Given the periodicity of light:dark cycles in the environment, it is not surprising that the circadian clock impinges upon several key aspects of plant development. Rhythmic changes in hypocotyl (seedling stem) growth and photoperiodic control of flowering time are two circadian-regulated developmental processes that have been described in some detail.

Daily Regulation of Growth in Plants

Hypocotyl elongation is targeted by a large number of both environmental (light and temperature) and endogenous (the circadian clock, giberellic acid, brassinosteroids, ethylene, abscisic acid, and cytokinins) stimuli and has been used extensively as a reporter of numerous signaling pathways. A connection between the circadian clock and hypocotyl length has been noted in two general ways. First, hypocotyl elongation exhibits overt rhythmicity, making it an easily measured output of the circadian clock. Rhythms in stem elongation under constant light were described more than 100 years ago in classical studies and more recently in several plants, including the model system Arabidopsis thaliana (1). Under constant light, Arabidopsis hypocotyls elongate with periodic pauses occurring near subjective dawn. Growth inhibition at dawn is not surprising, because light signaling pathways are known to inhibit hypocotyl growth. These periodic pauses persist under constant conditions and are capable of being entrained with reverse light:dark cycles. Second, many clock mutants exhibiting altered circadian rhythms also have hypocotyl phenotypes. A current model for the molecular framework of the central oscillator in Arabidopsis is based on a feedback loop between three proteins: two highly conserved myb-like DNA binding proteins, CCA1 and LHY, and the TOC1 protein (2). In addition to strong circadian phenotypes, loss-of-function toc1 mutants and mutants that overexpress either CCA1 or LHY all exhibit long hypocotyl phenotypes (35). The correlation between mutants with altered clock function and hypocotyl length extends beyond putative central clock factors and includes the ZTL, FKF1, ELF3, ELF4, GI, and SRR1 genes. Although the correlation between the circadian clock and hypocotyl length phenotypes is strong, it is not absolute; an allele of TOC1 (toc1-1) exhibits a short period but does not have a hypocotyl phenotype under any condition tested (6). This observation indicates that these phenotypes are separable, even though they correlate very strongly.

The conclusion from these observations is that hypocotyl elongation is a clock-controlled process. The next challenge is to define the molecular and biochemical basis for this rhythmic elongation pattern. The leading hypothesis for explaining the role of the clock in the regulation of hypocotyl growth is the concept that the clock “gates” light signaling pathways that target hypocotyl elongation (Fig. 1). When the gate is closed, the clock inhibits light signaling pathways even in the presence of light, allowing the hypocotyl to grow. When the gate is open, these light signaling pathways then inhibit growth, resulting in pauses in the growth pattern. Evidence for gating has been described for other light-regulated events. The acute induction of CAB gene expression in response to light is also gated by the circadian clock (7), indicating that circadian gating of light signaling pathways may be a generalized mechanism. Clues as to the molecular nature of this gating phenomenon may lie in work done on ELF3. Loss-of-function elf3 mutant alleles result in a loss of rhythmic hypocotyl elongation (1), a loss of rhythmic CAB gene expression in constant light, and an attenuation of gating of the acute response of CAB to light (8). These results all point to a central role for ELF3 in the attenuation of light signal transduction in a rhythmic fashion. The observation that ELF3 is able to physically and genetically interact with the red light photoreceptor PhyB provides a mechanistic model for ELF3 regulation of light signaling pathways (9), although further analysis is needed.

Fig. 1.

Schematic representation of circadian clock regulation of hypocotyl elongation and flowering time in plants. Hypocotyl growth is a target of both the circadian clock and light signaling pathways. ELF3 is clock-regulated and is able to attenuate light signaling, providing a potential mechanism for clock regulation of growth. CO is a key regulator of flowering time and is also clock-regulated. During long days, CO expression is coincident with light, leading to expression of FT and flowering, whereas during short days, CO expression is limited to the dark, resulting in no FT activation.

Seasonal Control of Reproductive Development in Plants

The transition from a vegetative mode of development to a reproductive mode represents a major step during the life cycle of higher plants, and the time to flowering is responsive to a number of environmental signals, including seasonal variations in day length (10). Arabidopsis is a facultative long-day plant, meaning that it flowers more rapidly during long days (16 hours light:8 hours dark) than during short days (8 hours light:16 hours dark). Changes in response to day length are known as photoperiodic responses. Early studies on photoperiodism in plants proposed that plants use their circadian clocks as a mechanism for detecting changes in day length. Recent genetic analysis of flowering time in Arabidopsis has yielded insights into the mechanism of photoperiodic regulation of flowering time, and we now have a basis for understanding the molecular interactions between the circadian clock and the regulation of flowering.

Two key genes mediating the transition to flowering have been identified, FLOWERING LOCUS T (FT) and CONSTANS (CO). In Arabidopsis, expression of FT is induced specifically in long days and appears to be the last regulatory step before the plant is committed to initiating an inflorescence (11). FT appears to be a direct target of CO (12), another flowering-time gene identified in genetic screens for late-flowering mutants. CO expression is clock-regulated, yet overexpression of CO does not result in circadian phenotypes, indicating that CO is an output from the clock. The waveform of CO expression is complex and appears to be critical for the photoperiodic induction of flowering (13). In long days, CO expression is coincident with light at the end of the day, whereas in short days the majority of its expression occurs in the dark. CO levels in the presence of light appear to be the key to activation of FT and induction of flowering. The significance of the phase of CO expression has been clearly demonstrated in the short period toc1 mutant (14). CO expression in the toc1 mutant is phase-shifted such that expression occurs during the light in both long days and short days, resulting in early flowering and no photoperiodic control over flowering. However, photoperiodic control was restored when the toc1 mutant plants were grown under light:dark cycles that matched their endogenous period of 21 hours.

These results support a model where clock control over the phase of CO expression is the key to photoperiodic control of flowering time; at the end of the day on long days, CO expression is coincident with light and induces expression of FT, whereas in short days CO expression begins to peak after lights-off and FT levels remain low. Two interesting questions are posed by this model. First, how does the circadian clock regulate CO expression? CO is clearly a key mediator between the clock and flowering time, but other factors must occupy positions between the clock and CO. Second, how does light interact with CO to regulate FT? Photoreceptors such as PhyA and Cry2 are clearly involved in flowering time, yet how they interface with CO has yet to be determined. Finally, it will be interesting to determine how these seasonal signals are integrated with other signaling pathways, such as temperature (15), that regulate FT and the transition to flowering.

Rice (Oryza sativa) is a short-day plant and requires short days in order to flower. Several quantitative trait loci for heading (flowering) have been described in rice and three genes have been identified; two appear to be homologs of CO and FT (16, 17), and the third appears to encode a casein kinase II (CK2) subunit (18). CK2 in Arabidopsis has been shown to be clock-associated and can shorten period length when overexpressed (19). These results reinforce the conclusion that the circadian clock plays a key role in photoperiodic control of flowering, yet raise the question of how the same genes promote flowering in long-day (Arabidopsis) and short-day (Oryza) plants.

The Intersection Between Circadian Control and Developmental Processes in Animals

Although the proteins forming the oscillator in animal clocks bear little resemblance to those found in plants, animals also use their circadian clocks to mediate photoperiodic responses. Many birds and mammals exhibit photoperiodic control of reproductive development, which is similar to the situation described for flowering. Classical experiments suggested that multiple oscillator systems may be involved, but it is in the Siberian hamster where the connection between bona fide circadian clocks and photoperiodic control has been unequivocally established (20). This is due to the availability of the tau mutant hamster, which bears a mutation in its CK1 gene that is an essential clock component (21). In Syrian hamsters, exposure to short photoperiods or constant darkness results in a decrease in gonadotrophin secretion, leading to overt gonadal regression. In the tau mutant, however, the circadian control of photosensitivity that regulates gonadal regression was affected by the same proportion and in the same direction as tau affects the circadian period (22). This strongly suggests that the clock residing in the suprachiasmatic nucleus (SCN) of the hamster that is responsible for overt circadian rhythms is the same one that plays a key role in regulating testis maintenance in response to photoperiod. Similarly, in Drosophila, the photoperiodic induction of ovarian diapause appears to be controlled by the same PERIOD/TIMELESS–dependent circadian clock that controls overt circadian rhythms such as locomotor activity (23, 24). Further evidence for the role of circadian clocks in reproductive development has also been demonstrated for Drosophila, in arrhythmic flies that exhibit aberrant sperm release and reduced reproductive fitness (25). Taken together, a wealth of evidence suggests the adaptive advantage of circadian clocks for both the seasonal and daily control of reproductive capacity (Fig. 2).

Fig. 2.

The major neuroendocrine pathway underlying circadian and light-modulated physiology in rodents. Photoperception at the retina is transmitted to the master circadian oscillator in the SCN. The SCN subsequently transmits the light signal to various nuclei around the hypothalamic paraventricular nucleus (PVN). The PVN can influence corticosterone secretion from the adrenal gland through corticotropin-releasing hormone (CRH) neurons as well as through PVN projections to the intermediolateral column (IML). The superior cervical ganglion (SCG) relays signals from the IML to the pineal gland and regulates melatonin release from the pineal. ACTH, adrenocorticotropic hormone.

Although later developmental events such as eclosion have also been shown to be clock-regulated through endocrine mechanisms (26), the impact of circadian clocks upon early developmental processes is less clear in the animal kingdom. The ontogeny of clock-regulated processes has only been systematically studied in a handful of cases, and the precise onset of embryonic circadian regulation appears likely to occur at different times in different tissues and to depend on the light cycle to which it has been exposed. One of the earliest events measured has been cyclic melatonin release by the pineal gland in Xenopus. Pineal glands are capable of making measurable melatonin in culture soon after they evaginate from the diencephalon at developmental stage 26. By stage 41, the eyes release melatonin rhythmically, indicating that Xenopus embryos develop functional photoresponsive circadian clocks within the first few days of life. In embryonic zebrafish, a circadian oscillator that regulates melatonin synthesis becomes functional and light-responsive between 20 and 26 hours postfertilization (27). However, there is little evidence yet that suggests these embryonic clocks are gating key patterning events within the spatial axes. Clocks that play that role are more likely to have periods similar to those of the somite clocks that exhibit a Hes1 gene–dependent oscillation with a period of about 2 hours (28, 29). Recent studies showing that a mouse strain that bears mutations at the PER2 locus exhibits enhanced susceptibility to cancer (30) may implicate a role for circadian rhythmicity in cell proliferation, although it has not been ruled out that PER2 may have a role independent from circadian control that is yet to be elucidated.

Circadian clocks have a well-defined role in regulating physiological and behavioral events on a 24-hour basis and have extended that role into seasonal timing and photoperiodism. It remains to be elucidated what early developmental patterning events might be gated by circadian clocks, although a major role here seems unlikely given the relatively normal morphology and lack of heterochronic events exhibited by organisms that bear strong mutations in their circadian systems.

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article