Plant Physiol. (1999) 121: 9-20
UPDATE ON DEVELOPMENT
The Physiology and Molecular Bases of the
Plant Circadian
Clock
David E. Somers*
Department of Cell Biology, The Scripps Research Institute, La
Jolla, California 92037
 |
INTRODUCTION |
Periodicity in biology comes in many
shapes and sizes. Among the very rapid are ultradian rhythms in the
courtship song of the Drosophila male, which recur every 50 to 60 s (Edmunds, 1988
). Running at a slightly faster pace is the
glycolytic oscillator in yeast, which exhibits self-sustained
oscillations in NADH fluorescence. These are typically between 2 and 70 min long, though periods of up to 6 h have been obtained by
manipulating metabolite concentrations (Edmunds, 1988). Hypocotyl
circumnutations with period lengths ranging from 25 min to 8 h are
part of the process of stem elongation in Arabidopsis and
Sinapis (Engelmann and Johnsson, 1998).
Infradian rhythms (>24 h) on the order of 4 to 5 d constitute the
estrus cycles of some rodents, whereas the menstrual cycle in higher
primates is generally between 25 and 35 d (Moore-Ede et al.,
1982). Finally, some of the longest cycles occur as circannual rhythms
that manifest as certain activities in animals (e.g. mating or
hibernation) or changes in plant development (e.g. flowering or
dormancy) that occur at a specific time just once a year (Sweeney, 1987; Edmunds, 1988).
Lying between these temporal extremes are processes that oscillate with
a nearly 24-h (circadian) periodicity. This type of rhythm, with a
period so closely matched to the rotation rate of the earth, occurs
ubiquitously in both prokaryotes (e.g. cyanobacteria) and many
eukaryotes, and regulates a wide range of physiological and
developmental processes. The recent molecular cloning of a number of
novel components of the circadian clock system has changed the
landscape of the field and greatly improved our view of this timing
mechanism. In this Update I will focus on how these and other findings have advanced our understanding of the molecular basis
of the circadian clock in plants.
| |
WHAT ARE CIRCADIAN RHYTHMS? |
Two hundred and seventy years ago, the French astronomer Jean
Jacques d'Ortous de Mairan initiated, with a single page report, the
experimental approach to the study of endogenous biological rhythms (de
Mairan, 1729; Sweeney, 1987). He used a "sensitive" plant (mimosa)
that was already known to fold its leaves and leaflets closed at night
and reopen them during the day. When he placed the plants in constant
darkness, he found that leaf opening and closing persisted just as if
the plants were seeing the day and night. With this came the first
experimental evidence for the persistence of an endogenous rhythmicity
in the absence of environmental cues.
Prior to de Mairan's report, and for many years after, rhythmic
movements in plants were assumed to be caused by the daily cycles of
light and dark. Indeed, in 1751 Linneaus designed a garden consisting
of flowers that opened and closed their petals at specific but
different times of day (Fig. 1). In this
scheme, by simply looking out the window and noting which species were open or closed one might tell the hour of the day (Moore-Ede et al.,
1982).
View larger version (79K):
[in this window]
[in a new window]
| Figure 1.
Depiction of the flower clock (eine Blumen-Uhr)
designed by Linneaus. The left half of the figure (6 AM-12
PM) shows when the petals of different species are opening;
the right half (12 PM-6 PM) shows times of
petal closing (except evening primrose, which starts to open its
flowers after 5 PM). Note that some species can act to time
both morning and afternoon events. (From Moore-Ede et al.,
1982.)
|
|
Skeptics of the idea of an endogenous source of the rhythmicity
suggested that subtle, unknown environmental signals were being
detected by the plants and were responsible for the maintenance of
rhythmicity in the dark. However, 100 years after de Mairan, a second
Frenchman, Augustin de Candolle, again working with mimosa, observed that in continuous dark the period length of the leaf opening
and closing rhythm was not exactly 24 h, but closer to 22 to
23 h (Moore-Ede et al., 1982). Although he did not note it at the
time, this result argued against an environmental signal based on the
24-h rotation of the earth as the source for maintaining rhythmicity.
It also anticipated by another 100 years the results of Erwin
Bünning, who identified two variants of common bean that differed
in their endogenous period length by 3 h. When crossed, the
progeny exhibited period lengths ranging between the extremes of the
two parents, suggesting that this property of circadian rhythms is a
genetically based polygenic trait (Bünning, 1935).
From the results of numerous studies from these initial observations to
the present, a number of characteristics have come to define a
circadian rhythm (Fig. 2) (Edmunds,
1988). First and foremost is an endogenous period that is approximately
24 h long (circadian). Only when a 24-h
environmental cycle is imposed on the system does the period become
exactly 24 h. Indeed, within limits, a non-24-h environmental
cycle (e.g. 10 h of light/10 h of dark) will constrain the clock
to oscillate with just that period length (i.e. 20 h). This
adjustment arises from a second primary feature of circadian rhythms:
entrainability. This is the process by which the clock is synchronized
to the outside world. In all organisms studied to date, the primary,
though not exclusive, entraining stimuli are temperature and light.
Diurnal oscillations in temperature (high/low) or light (light/dark)
are the cues that adjust the circadian system with each cycle. This occurs because a step up or step down in light or temperature can alter
the position, or phase, of the oscillation. In some organisms (e.g.
Drosophila) a short 15-min light pulse is sufficient to
fully entrain the clock, whereas plants often require 3 to 4 h of
illumination.
View larger version (13K):
[in this window]
[in a new window]
| Figure 2.
Four characteristics of circadian rhythms (see
text). An idealized rhythm is shown under entraining conditions
(light/dark cycles) and during free-run in constant conditions
(continuous light). In this example the endogenous free-running period
is longer than the cycle of the entrainment schedule. As long as
light/dark cycles continue, the period of the oscillator is identical
to the period length of the entrainment schedule cycle, resulting in a
consistent difference in time (i.e. phase relationship) between any
chosen point on the curve (e.g. peak expression) and a given phase
marker (e.g. the dark-to-light transition; dotted line). During
free-run, this constant relationship breaks down as the oscillator
reverts to its longer, endogenous period and the peak of expression
drifts further from "subjective" dawn. Dark (shaded bar) and light
(white bar) are indicated below the trace. (Adapted from Edmunds,
1988.)
|
|
A third characteristic of circadian rhythms that is not shared with
other biological or biochemical periodicities is temperature compensation. Although a temperature shift can reset the phase of the
clock, when stabilized, this new temperature regime has very little
effect on endogenous period length. Most biochemical processes are
sensitive to temperature, with reaction rates doubling or tripling with
each 10°C change in temperature (Q10 = 2-3)
(Johnson et al., 1998). In contrast, the Q10
values of circadian rhythms lie between 0.8 and 1.4. For example, we
have tested Arabidopsis over a 20°C temperature range and found no
more than a 2.5-h change in period length (Q10 = 1.0-1.1) (Somers et al., 1998b).
The fourth feature, the persistence of rhythmicity in the absence of
periodic input, is perhaps the most intriguing aspect of circadian
biology. The primary focus of most approaches to understanding the
mechanism of the circadian clock has been to identify the components
and their interactions that allow the maintenance of a self-sustained
oscillation in a non-periodic environment.
| |
THE CIRCADIAN CLOCK IS A SIGNALING SYSTEM LINKING THE ENVIRONMENT
TO PHYSIOLOGY AND DEVELOPMENT |
An oscillator alone is not a clock. To be meaningful to the
organism a clock must be linked to the outside world (Fig.
3). The daily occurrence of light/dark
transitions provides the time-setting information by which
clock-controlled processes are appropriately phased. For example,
processes required for photosynthesis are phased early in the day,
whereas other genes, such as catalase, show peak expression at night
(Fejes and Nagy, 1998). In the absence of external input to the clock,
oscillations from a pacemaker can control various processes, but there
is no synchronization with the environment and no temporal information
is relayed. A familiar example is the periodic beating of the heart,
which continues to pump blood at a fairly even pace day and night.
View larger version (17K):
[in this window]
[in a new window]
| Figure 3.
Model of a simple circadian system. The three
primary components include: an input (entrainment) pathway(s), the
central oscillator, and an output pathway(s). The phytochromes (PHY)
and the cryptochromes (CRY) are two classes of photoreceptors known to
mediate the first step of the light entrainment pathway (Somers et al.,
1998a). Interactions among the components (A-D) of the central
oscillator create the autoregulatory negative-feedback loop that
generates the approximately 24-h oscillations. Three different
hypothetical couplings of the central oscillator to possible output
pathways are shown to indicate that differently phased overt rhythms
with the same period (E-G) can arise from a single pacemaker.
|
|
A clock must also control the rhythmicity of events in the organism. An
oscillator that is not coupled to other processes cannot act as a
meaningful timer. In plants a wide variety of processes are controlled
by the clock. Gene transcription, Ca2+ levels,
and some enzyme activities are examples of intracellular processes
under circadian clock regulation (Sweeney, 1987; Johnson et al., 1998).
At a higher level of organization, rhythms in stomatal opening, leaf
movement, and hypocotyl expansion occur through coordination of cell
and tissue level events by the circadian oscillator (Engelmann and
Johnsson, 1998; Webb, 1998; Dowson-Day and Millar, 1999). Finally, the
circadian pacemaker can mediate fundamental changes in plant
development. The photoperiodic control of flowering time and the onset
of bud dormancy are two examples of how a 24-h clock can underlie
processes that occur on a non-circadian time scale (Thomas and
Vince-Prue, 1996).
Both the input pathway(s) to the oscillator and the output pathway(s)
leading to the control of the overt rhythms are signal transduction
pathways, and are linked by the central oscillator (Fig. 3). Current
research on the circadian clock is concerned with identifying candidate
components and positioning them within one of the three realms of the
system. As noted below, the borders between these are becoming fuzzy as
new information emerges, but the model has provided the conceptual
framework from which most clock research has proceeded (Johnson et al.,
1998).
| |
THE "HANDS" OF THE CLOCK: THE OUTPUT PATHWAY LEADS TO MANY AND
VARIED CLOCK-REGULATED PROCESSES IN PLANTS |
Most model circadian systems have been developed around a limited
number of clock-regulated outputs. For example, until recently, all of
the progress in understanding the Drosophila circadian clock
has come through the observation of eclosion (adult emergence) and
locomotor activity. Work in rodents has relied largely on wheel-running
activity, and circadian-regulated sporulation is still the primary
morphological assay for clock function in Neurospora.
In contrast, research into the circadian clock in plants has benefited
from studying a large number of different output processes. These
afford the opportunity to probe the nature of the clock itself, as well
as to understand the effects of the clock on plant growth and
development. Each "hand" of the clock becomes an additional assay
of the activity of the oscillator. As noted above, there are different
classes of outputs and each can tell us something different about the
organization of the clock system.
Gene Regulation
The list of genes under the control of the circadian clock is
constantly growing (Kreps and Kay, 1997; Fejes and Nagy, 1998). There
are examples of circadian regulation of gene expression at each step:
transcription (Millar and Kay, 1991; Liu et al., 1996), transcript
abundance (Fujiwara et al., 1996; Zheng et al., 1998), translation
(Mittag et al., 1994), and posttranslational processing (Nimmo, 1998).
Since the central oscillator itself is strongly based on temporal
variation in gene expression (see below), it is possible that the
output pathway from the oscillator to clock-controlled gene expression
will be very short compared with more complex processes such as cell
expansion and flowering.
The first examples of clock-controlled gene regulation in plants were
mRNA levels of light-induced genes involved in photosynthesis (Kreps
and Kay, 1997). Among these are the chlorophyll a/b-binding proteins, part of the light-harvesting complex in chloroplasts that are
coded for by a small gene family (CAB and LHCb).
In most species the circadian peak of expression of these genes is a
few hours after dawn, as would be expected for proteins required for photosynthesis.
Other genes related to photosynthesis are also expressed rhythmically.
These include the small subunit of Rubisco (Pilgrim and McClung, 1993),
Rubisco activase (Liu et al., 1996), and, in CAM plants, PEP
carboxylase kinase (Nimmo, 1998). In contrast to the phasing of genes
of the light reactions, peak expression of PEP carboxylase kinase mRNA
occurs at midnight, accurately reflecting the requirement of the enzyme
at that time (Nimmo, 1998).
Non-photosynthetic enzymes under the control of the circadian clock
include the catalases, which are primarily involved in eliminating
toxic H2O2 from the cell.
The circadian regulation of the small catalase gene families in maize
and Arabidopsis has been intensively studied (Zhong and McClung, 1996;
Polidoros and Scandalios, 1998). It is particularly interesting that
the peaks of expression of CAT2 and CAT3 in
Arabidopsis are 12 h out of phase with each other (Zhong and
McClung, 1996). This difference probably reflects different roles for
each in plant metabolism, but also raises the question of how they are
coupled to an oscillator. Although it is possible that entirely
different clocks control these two genes, there is evidence (see below)
that at least some differently phased outputs are under the control of
the same clockwork.
The transcript levels of two closely related putative RNA-binding
proteins from Arabidopsis, CCR1 (Atgrp8) and
CCR2 (Atgrp7), cycle with a peak expression 8 to
12 h after dawn in the wild type (Kreps and Simon, 1997). Although
the function of these proteins is unclear, overexpression of
CCR2 in Arabidopsis strongly depresses the cycling of the
endogenous CCR2 transcript (Heintzen et al., 1997). This
finding, together with the observation that endogenous CCR2 protein
levels normally oscillate with a 4-h phase delay relative to its
transcript, strongly suggests that the transcript and protein act
together as components of an autoregulatory negative feedback loop.
However, CCR2 overexpression has no effect on the period or
levels of other unrelated circadian-regulated transcripts tested, such
as CAB3 and CAT3. In addition, the
toc1 (timing of cab) mutation, which shortens the period
length of a wide range of clock-controlled processes (Somers et al.,
1998b), also shortens the cycling of the CCR2 transcript in
constant light (Kreps and Simon, 1997). Therefore, circadian control of
CCR2 gene expression most likely lies downstream of the
effects of toc1 on the clock and may define a "slave"
oscillator that is still subject to temporal control from a
"master" oscillator, but may itself control the amplitude and phase
of an undefined subset of clock-controlled outputs (Heintzen et al.,
1997).
Ca2+ Signaling
Much less is known about the effect of the clock on subcellular
processes unrelated to gene expression in plants. One recent insight
came from the constitutive expression of the
Ca2+-dependent photoprotein aequorin in tobacco
and Arabidopsis. In both species, this assay revealed circadian
oscillations in cytosolic [Ca2+] in constant
light and constant dark after entrainment in light/dark cycles (Johnson
et al., 1995). When the reporter protein is targeted to the
chloroplast, similar rhythms are observed, but only in the dark and
they damp rapidly. These results demonstrate the potential for
circadian regulation of Ca2+-dependent enzymes
such as calmodulin and Ca2+-dependent protein
kinases. Such enzymes could be part of output pathways, causing or
contributing to, for example, a circadian cycle of activity of key
metabolic enzymes. Ca2+-dependent regulation of
the input pathway might also occur, since Ca2+
has been shown to participate in phytochrome-mediated light regulation (Neuhaus et al., 1997). This potential for Ca2+
oscillations in a dual role of mediator of output and input signaling has precedence in the animal field. Pharmacological and
electrophysiological studies in the sea slug (Bulla) show
Ca2+ as an important component that mediates
signaling to and from the central oscillator (Block et al., 1995).
Cell and Tissue Phenotypes
Circadian control of cell expansion/contraction is the basis of
rhythms in leaf movements, hypocotyl elongation, and stomatal aperture
size. In legumes, endogenously controlled leaf movements arise through
alternate shrinking and swelling of specialized extensor and flexor
cells that lie on opposite sides of the pulvinus. Circadian rhythms in
the membrane potentials of protoplasts made from these flexor and
extensor cells are 12 h out of phase with each other, and result
directly from changes in the state of the K+
channels in the two cell types (Kim et al., 1993). In the intact pulvinus, flexor and extensor cells are discretely localized but closely adjacent to each other. This recapitulation of their endogenous activity in cell culture is one of the best pieces of evidence to
suggest that the clock can act cell autonomously in higher plants.
In contrast, circadian leaf movements (and rhythms in hypocotyl growth)
in Arabidopsis and other plants probably arise from differential cell
expansion in the organ, since rhythmicity ceases after leaf (or
hypocotyl) growth is complete (Engelmann and Johnsson, 1998; Dowson-Day
and Millar, 1999). Using simple video technology, leaf movement rhythms
can be followed for more than a week in a single seedling, and are a
rapid way to assess the effects of a mutation or treatment on the
circadian clock (Millar et al., 1995a; Hicks et al., 1996; Schaffer et
al., 1998). The cellular bases of these movements, however, are not
understood.
The size of the stomatal aperture is influenced by a wide range of
environmental and physiological factors, as well as by the endogenous
circadian clock (Assman, 1993; Webb, 1998). Together with the accessory
cells, the paired guard cells form a discrete, specialized unit
imbedded in the epidermis. When the epidermis is peeled away from the
blade in fava bean, rhythms in stomatal aperture persist, strongly
suggesting that circadian regulation of the stomatal apparatus is cell
autonomous (Gorton et al., 1989). These data also support the notion of
complete circadian clock systems running independently in each cell.
A comparison of stomatal conductance rhythms and leaf movement rhythms
in the same plant (common bean) raised the question of whether the same
type of clock controls these two outputs. The period length of the
rhythm in stomatal opening and photosynthesis, both occurring in the
leaf blade, is shorter (approximately 24 h) than the
pulvinus-based leaf movement rhythm (approximately 27 h)
(Hennessey et al., 1992). One interpretation of these results is that
separate oscillators control the rhythms in these two tissues, each
with intrinsically different free-running periods and possibly
comprised of distinctly different components. Alternatively, both
organs may differ only in the nature of the light input pathway leading
to the oscillator, since the free-running circadian period can be
modulated by the intensity of the light input to the oscillator (see
below). If the pulvinar region is less sensitive to illumination than
the blade, identical oscillators in the two tissues might run at
slightly different rates. Ways to address this question would be to
observe the period length of two or more very different outputs in the
same cell or tissue (Roenneberg, 1996), or to determine how extensively
a period length mutation in a known oscillator component acts on
various outputs throughout the plant (see below and Millar [1998] for
further discussion).
Flowering and Dormancy
The processes by which the circadian clock is involved in the
control of flowering and dormancy are too complex and too ambiguous to
fully review here (see Vince-Prue, 1994; Thomas and Vince-Prue, 1996).
Unlike most circadian phenotypes that manifestly cycle with a 24-h
rhythm, the transition from vegetative to reproductive growth occurs as
the endpoint of a series of processes to which the circadian clock
contributes an element of timing. Nonetheless, much of the primary
evidence for a role of the clock in photoperiodic timing comes from
experiments in which light pulses were administered over 3- to 4-d time
courses. In Arabidopsis and barley, there was a 24-h rhythm in the
acceleration of flowering time caused by 6-h far-red light treatments
given over a 72-h period (Deitzer et al., 1982; Deitzer, 1984;
Carré, 1998). This rhythmic change in the sensitivity to light is
found in both short- and long-day plants (Vince-Prue, 1994), suggesting
that the circadian clock is involved in regulating flowering time in
both types of reproductive strategies. However, it is also apparent
that the way in which the clock is incorporated into the entire
flowering process differs greatly both within and between these two
flowering time systems (Vince-Prue, 1994; Thomas and Vince-Prue, 1996).
Physiological studies such as these cannot easily address another
version of the same question raised earlier: is the same type of
oscillator responsible for both photoperiodic timing and circadian
timing of daily rhythmic processes? In Arabidopsis, the answer has been
approached by examining day-neutral flowering-time mutants, lines that
flower equally rapidly or slowly in both short and long days (Koornneef
et al., 1991; Carré, 1998). This strategy assumes that if the
circadian clock is severely disrupted (or eliminated, resulting in
arrythmicity), the timing information inherent in day-length
differences will not be perceived or processed by the plant and
flowering time will be the same in both short and long days.
elf3 (early-flowering 3) (Hicks et al., 1996), lhy (long hypocotyl) (Schaffer et al., 1998), and
toc1 (in ecotype Landsberg erecta) (Somers et
al., 1998b) are three such day-neutral lines and share the additional
feature of being disrupted in other circadian phenotypes. Although
these findings suggest that the same circadian clock controls
photoperiodic timing and at least a subset of other circadian
processes, the phenotypes of these mutants are still sufficiently
ambiguous or incomplete to discourage conclusive positioning of these
genes within the system (see below).
| |
ENTRAINMENT OF THE CENTRAL OSCILLATOR OCCURS VIA THE LIGHT INPUT
PATHWAY |
Since light is a universal entraining stimulus for all known
circadian systems, there is intense interest in tracing the signaling pathway that provides the ultimate timing cues that entrain the clock.
In addition to the phase-setting signal that a light pulse provides,
changing the incident light intensity can vary the period length of a
free-running rhythm. As a general rule, increasing light intensity
tends to lengthen the period in nocturnal organisms and shorten the
period in diurnal organisms (including plants) (Aschoff, 1979; Millar
et al., 1995b; Somers et al., 1998a, 1998b). However, the relationships
can be more complex than that. In the dinoflagellate
Gonyalaux, increasing the red-light intensity lengthens the
period, whereas increasing blue or white light decreases period length
(Roenneberg, 1996). Together these results suggest that at least two
types of photoreceptors mediate light input to this alga (Roenneberg
and Deng, 1997).
Similar experiments in Arabidopsis have begun to sort out the roles of
the eight or more photoreceptors in higher plants that might mediate
light signaling to the circadian clock. To facilitate easy monitoring
of the activity of the oscillator, the cab2-luciferase reporter (cab2::luc) was used as a bioluminescent
assay of clock-controlled gene transcription. When the luciferin
substrate is applied, this promoter drives robust rhythms of
luminescence from intact seedlings that can be monitored with a highly
sensitive video camera (Millar et al., 1992). With this system it was
found that the period length (
) of cab2::luc
expression in Arabidopsis significantly lengthens when plants are moved
from continuous light ( = 24.5 h) to continuous darkness
( = approximately 30 h) (Millar et al., 1995b). Using this
assay, previous evidence (Johnson et al., 1998) for a role of
phytochromes in the entrainment of the clock has been verified using
phytochrome-deficient mutants of Arabidopsis. hy1-6,
deficient in all five of the photoreversible phytochromes
(PHYA, PHYB, PHYC, PHYD,
and PHYE), shows a slightly longer period in red
and white light than the wild type, as if the plants were partially
blind (Millar et al., 1995b).
Subsequent studies using type-specific phytochrome mutants show that
over a nonoverlapping range of light intensities, both phyA and phyB
mediate red-light signaling to the clock (Somers et al., 1998a). Under
dim-red light, plants deficent in phyA show a 2-h period lengthening
compared with wild type, whereas a loss of phyB lengthens period only
in high-fluence red light. PhyA also acts to mediate blue light input
to the clock but, again, only at low light intensities. Not
surprisingly, the dedicated blue-light photoreceptors cryptochrome 1 (cry1) and cryptochrome 2 (cry2) also play roles. A deficiency in the
more abundant species, cry1, causes period lengthening in both high-
and low-fluency blue light, whereas the absence of cry2 alone has only
a minor effect on period length (Somers et al., 1998a).
In an ecological context these results make sense. Together the
phytochromes and cryptochromes sense the full spectrum of visible
light. This allows a sensitive monitoring of the changes in light
quality and quantity that occur at dusk and dawn. In addition, a green
leaf canopy and the proximity of neighboring plants both have marked
effects on the ratio of red and far-red light that reaches a plant
(Ballaré et al., 1987). The above findings suggest that the plant
recruits a diversity of photoreceptors to ensure that the pace of the
circadian oscillator remains unaffected by this wide range of fluence
rates and spectral qualities it encounters in the environment.
Photoreceptors only initiate light signaling to the clock. The next
steps are to identify the components that lie downstream and determine
which are specific to the circadian clock and how many are shared with
other light-responsive processes. One fruitful approach has been to
examine the effects of constitutive photomorphogenic mutants of
Arabidopsis on period length. Mutants of this class (det/cop/fus) act downstream of photoreceptors and are the
converse of the partially blind photoreceptor mutants, since they
develop some of the features of light-grown plants in complete darkness (Fankhauser and Chory, 1997). However, the circadian phenotype of
mutants at three loci, DET1, DET2, and
COP1, differ in important respects. The free-running period
of det1-1 is much shorter than that of the wild
type, both in continuous light and constant dark (Millar et al.,
1995b). Surprisingly, it is even shorter in the dark (approximately
18 h) than in the light (approximately 20 h), suggesting that
light may antagonize or moderate the action of this non-null-mutant
form of DET1 on the clock, possibly through a parallel signaling
pathway.
On the other hand, the extremely short period is consistent with the
notion that this mutation mimics constitutive very high light input,
although periods this short have never been observed in wild-type
plants under high irradiance. A partially active allele of
COP1, cop1-6, shows a similar, though
less extreme phenotype, maintaining a period in continuous darkness
(23.5 h) similar to wild type in continuous light (Millar et al.,
1995b). Both DET1 and COP1 are nuclear-localized proteins that
negatively regulate a host of light-dependent phenotypes (Fankhauser
and Chory, 1997; Torii and Deng, 1997), and their circadian phenotypes
in continuous darkness are consistent with these genes acting within
the light input pathway to the clock at a point still common to many
other light-regulated processes. In contrast, DET2 is an enzyme of the brassinosteroid biosynthetic pathway (Fankhauser and Chory, 1997) and
has little to no effect on period in continuous darkness ( = approximately 30 h) (Millar et al., 1995b), suggesting that this
hormone does not play a primary role in regulating the pace of the
oscillator.
The elf3 mutant was originally identified through a screen
for early-flowering mutants. This recessive mutation causes a long hypocotyl in white light and equally rapid flowering in long and short
days. Subsequently, it was found that the rhythms of
cab2::luciferase activity and leaf movement are
absent in the light in this mutant. However, there are still
oscillations in luminescence in dark-grown seedlings, indicating that
the arrhythmicity is light dependent (Hicks et al., 1996). These data
strongly suggest that ELF3 acts as part of the light input pathway to
the clock. But it is also possible that separate oscillators operate in
the light and dark and that ELF3 acts only as a key component of the
former. Alternatively, ELF3 may be part of the output pathway, with
elf3 masking the activity of the oscillator only in the
light. This ambiguity in knowing where ELF3 acts within the circadian
clock system derives in part from its arrhythmic phenotype.
| |
ARRHYTHMICS: WHERE DO THEY BELONG? |
Positioning a component within a signaling pathway first requires
the appropriate assignment to a place within the circadian system.
There is no single criterion that allows one to unequivocally determine
if a candidate is an input, output, or oscillator component, but any
decision requires that cycling be observable under at least some
conditions. Therefore, a mutation that causes the loss of rhythmicity
itself can confound analysis because the essential assay is lost. In
addition, arrhythmicity may arise from disruptions in any of the three
subsystems of the clock.
The Oscillator
A loss-of-function mutation in a non-redundant rhythmic component
of the central oscillator will stop the clock. This condition, when
coupled with the isolation of both long- and short-period alleles of
the locus, has been one of the more reliable identifiers of a bona fide
oscillator component (Dunlap, 1996). In Drosophila and
Neurospora, constitutive high expression of some oscillator genes also disrupts cycling (Aronson et al., 1994; Zeng et al., 1994). In either case, it is the loss of rhythmicity in the component that brings the pacemaker to a halt.
The Output Pathway
Two alterations in the output pathway that can lead to
arrhythmicity are decoupling and masking. In the former, a mutation eliminates or impairs signaling from the oscillator to the overt rhythm, resulting in constitutively high or low expression
(arrhythmicity). Deletions within the promoters of clock-controlled
genes that eliminate rhythms in transcription or transcript abundance
are one example (Carré and Kay, 1995; Bell-Pedersen et al.,
1996). In masking, the signaling system is intact, but other processes override circadian regulation. This might arise from constitutive overexpression or misexpression of a normally circadian-regulated component (Heintzen et al., 1997). In both cases, the central pacemaker
may be functioning normally but may simply not be visible because the
overt rhythms that normally report its activity have been disrupted.
The Input Pathway
As noted above, the conditional light-dependent arrhythmicity of
the elf3 mutation suggests that this lesion is in a
component of the light input pathway. Furthermore, many organisms that
show robust rhythmicity in continuous darkness become arrhythmic with increasing light, presumably due to excessive or inappropriate light
input to the oscillator (Aschoff, 1979; Konopka et al., 1989).
Therefore, a lesion in the input pathway that mimics high-intensity light signaling in such systems could cause arrhythmicity. Finally, formal modeling of clock systems has shown that if the level of an
input component itself cycles, as influenced by feedback from the
pacemaker, then the loss of this component may result in a more rapidly
damped rhythm in constant conditions (Roenneberg and Merrow, 1998).
Clearly, arrhythmicity per se gives no clues as to how and where a
component acts within the circadian system. Along with elf3,
two recently described arrhythmic lines of Arabidopsis, CCA-1-ox
(circadian clock associated) and lhy, have brought this discussion to the forefront. Although the aberrant expression of these
two genes affects many clock-controlled processes, their normal roles
within the circadian clock remain unclear (see below).
| |
THE CIRCADIAN OSCILLATOR IS AN AUTOREGULATORY NEGATIVE FEEDBACK
LOOP |
Recently, there has been extremely rapid progress in understanding
the molecular nature of the circadian oscillator in animals and
cyanobacteria (Golden et al., 1997; Ishiura et al., 1998; Wilsbacher
and Takahashi, 1998; Dunlap, 1999). These advances have built upon the
model of the oscillator as an autoregulatory negative-feedback loop. At
its simplest, this scheme requires two interacting components that rise
and fall in abundance or activity and between which there is a time
delay (Merrow et al., 1997). Classical and molecular genetics have
helped to identify genes in Drosophila,
Neurospora, and mammals that fit these criteria, and a
common molecular mechanism has begun to emerge (Fig.
4). In all three systems, both the mRNA
and protein products of key genes (state variables) cycle in abundance
and/or activity. The 24-h delay is caused in part by the time it takes
for the protein product(s) to enter the nucleus, where they contribute
to the repression of their own transcription, and/or in the time
required for these proteins to degrade in the nucleus (Merrow et al.,
1997; Dunlap, 1999).
View larger version (32K):
[in this window]
[in a new window]
| Figure 4.
Current working model of a eukaryotic circadian
clock. This generalized scheme is drawn largely from work with
Drosophila and mouse, although all of the available
evidence from Neurospora supports a similar mechanism.
PROT1 and PROT2 are two nuclear-localized proteins that act to
negatively regulate their own transcription by blocking the activity of
two bHLH-containing DNA-binding proteins (PP1 and PP2). PP1 and PP2
share a common domain (PAS) that facilitates protein-to-protein
interactions and may contribute to their heterodimerization and binding
to the circadian clock element (CCE) present on the promoters of PROT1
and PROT2. In the absence of PROT1 and PROT2, PP1 and PP2 activate the
transcription of both PROT genes. This results in increased protein
levels of both PROT gene products in the cytoplasm, which are further
regulated by light (PROT2 is light labile) and/or phosphorylation state
(PROT1 is destabilized by phosphorylation). Association of PROT2 with
PROT1 helps stabilize PROT1 levels and facilitates the transport of the
heterodimer to the nucleus. PP1/PP2-dependent transcription is
inhibited by the pair until turnover of PROT1 and PROT2 depletes
levels, allowing transcription to proceed, and then the cycle begins
again. Degraded proteins are indicated as small, grouped ovals. P,
Phosphorylation.
|
|
Other components that do not cycle but are critical to the proper
functioning of the oscillator are termed parameters, and mutations in
these components can alter or stop cycling altogether (Kloss et al.,
1998). The five molecular components known to participate in this
feedback loop in Drosophila include genes that code for transcription factors (dClock, dbmal1), a protein
kinase (doubletime), and novel proteins (period,
timeless) (Wilsbacher and Takahashi, 1998). Between flies
and mammals there has been a remarkable conservation of function among
these proteins, suggesting a common origin of the circadian oscillator
in these two groups (Dunlap, 1999). Fewer components of the oscillator
have been identified in Neurospora (frequency,
white collar-1, white collar-2), but, apart from
a motif (the PAS domain) that facilitates protein-to-protein
interactions (Young, 1998; Dunlap, 1999), none of the proteins appears
to be an ortholog of other known clock components. The positive
components of the loop are the transcription factors and kinase
activities, while the negative factors (period,
timeless) facilitate the disruption of the transcription of
their own genes and others (Dunlap, 1999).
In cyanobacteria (Synechococcus), all known period-length
mutations occur within a three-gene cluster and range between 16 and
60 h (Ishiura et al., 1998). An autoregulatory feedback loop is
likely the basis of this clock system too, although the requisite lag
in the 24-h rhythm probably arises differently than in the eukaryotic
system, since there is no cytoplasmic/nuclear partition to cross. These
genes have been cloned and none shows similarity to known eukaryotic
clock components (Ishiura et al., 1998).
In plants, no unequivocal oscillator components have been identified.
Attempts to use known components from other organisms to probe plant
genomes have generally not been successful. Two untested potential
exceptions are the recently identified Arabidopsis homolog of a
Drosophila casein kinase I (doubletime) (Kloss et al., 1998), and a putative Arabidopsis timeless gene (Zylka
et al., 1998), which in flies plays a critical role in the nuclear import of the period protein (Young, 1998). Otherwise, the
two cloned loci in Arabidopsis, CCA-1 and LHY,
which have been shown to disrupt clock function, are unrelated to known
animal, fungal, or bacterial clock components.
CCA-1 and LHY are closely related proteins, each with a single
myb-related DNA-binding motif (Wang et al., 1997; Schaffer et al.,
1998; Wang and Tobin, 1998). CCA-1 was originally identified by its
ability to bind a phytochrome-responsive region of the cab1
promoter in Arabidopsis, which was later noted to fall within a 38-bp
region sufficient to confer circadian regulation (Carré and Kay,
1995; Wang et al., 1997). In wild-type plants, the mRNA levels of both
genes cycle with a 24-h period and with a very similar early-morning
peak phase. When overexpressed at constitutively high levels, both
cause arrhythmicity in all circadian outputs so far examined. The
levels of endogenous CCA-1 and LHY mRNAs are
clamped low when CCA-1 is overexpressed (CCA1-ox), whereas LHY mRNA abundance is intermediate in an
LHY-overexpressing mutant (lhy).
The cycling of transcript levels of other clock-regulated genes such as
CAB, CAT3 (catalase), and CCR2 (an
RNA-binding protein) are also reduced or eliminated in both
overexpressing lines. CAB expression peaks in the late
morning, whereas CAT3 and CCR2 transcript levels
peak in the late afternoon. That such differently phased rhythms are
affected suggests that both CCA-1 and LHY act on, or close to, the
fundamental processes of the clock, and not just on one branch of the
output pathway. This conclusion is strengthened by the finding that
leaf movements are arrhythmic (lhy), and hypocotyls are
lengthened (lhy and CCA-1-ox) in these lines. Flowering time under long days is delayed in both overexpressors, which is consistent with their both having roles in the circadian clock system (Schaffer et
al., 1998; Wang and Tobin, 1998).
These observations fulfill many of the criteria previously established
to help identify a candidate as a bona fide state variable of the
circadian oscillator. These criteria include: (a) the component (or
process) itself oscillates with circadian periodicity, (b) the
component feeds back to control its own level and rhythmicity, (c)
clamping the level of the component to any constant value (null to
constitutive overexpression) stops the clock and any overt rhythmicity,
and (d) an induced change in the level (or activity) of the component
causes a phase shift in the cycling of the component and all overt
rhythms within less than one circadian cycle (Aronson et al., 1994).
The mRNA oscillations of both LHY and CCA-1
satisfy criterion a, and in the case of CCA-1 the protein level has
also been shown to cycle (Schaffer et al., 1998; Wang and Tobin, 1998). Overexpression of either transcript eliminates rhythmic expression of
the respective endogenous gene, suggesting that each protein normally
feeds back to control its own transcription (criterion b). Since all
other rhythmic activity is blocked as well, overexpression of these
proteins appears to stop the clock (criterion c). However, because of
possible complications arising from ectopic expression and potential
cross-talk between closely related family members, overexpression data
must be interpreted cautiously. For example, aberrantly high levels of
an output component may mask oscillator activity, as noted above, and
may also affect output pathways not normally controlled by that
component. A more ideal approach would be to use an
inducible/repressible promoter-gene fusion system akin to the one
developed and implemented in Neurospora (Aronson et al.,
1994) to induce a short pulse of light-independent expression of the
candidate component. A phase shift in the peak of both the endogenous
level of the induced gene and other outputs would fulfill criterion d.
The strong degree of sequence similarity between LHY and
CCA-1 raises the possibility that they belong to a gene
family of proteins with redundant or partially redundant functions
(Schaffer et al., 1998; Wang and Tobin, 1998). A null mutation in
CCA-1 (cca-1) shortens the free-running period of
the circadian oscillations of three clock-controlled genes
(CAB2, LHY, and CAT2), suggesting that
LHY and CCA-1 are not fully redundant proteins (Green and Tobin, 1999).
This result also shows that CCA-1 is not an essential or unique
oscillator component, since the clock continues to function in its
absence. However, this change in free-running period does suggest that
CCA-1 is not a simple output regulator, but may instead be part of an
input pathway to the clock, linking light signaling to the oscillator
itself (Green and Tobin, 1999).
Most of the genes identified as coding for components of the oscillator
in non-plant organisms arose from genetic screens for period length
changes or as homologs of these genes (Wilsbacher and Takahashi, 1998).
The strong genetics infrastructure of Arabidopsis has made it a good
choice for the same approach in plants. Using the
cab2::luc luminescence reporter system described
above, Millar et al. (1995a) conducted a screen for mutants in which
the peak of free-running cycling bioluminescence was shifted either
later or earlier than wild type. More than 20 individuals were
recovered and displayed period lengths ranging from 21 to 27 h
(wild type = 24.5 h). The best characterized of these lines,
toc1-1, runs with a period of 21 h in
continuous white light. The morphology of the mutant is wild type,
consistent with the normal phenotypes observed for period length
mutants in other organisms. This simple observation suggests that that
clock is not based on metabolic processes that are fundamental to the
maintenance of the organism, but, rather, arises from interactions
between processes that are at least in part specific to the clockwork
itself. At the same time, close examination of the
toc1-1 line shows wide-ranging effects of this
semidominant mutation on circadian-related phenotypes throughout plant
growth and development.
The toc1-1 mutant affects all clock-controlled
processes so far examined. At the cellular level in green seedlings,
the period of cycling in transcription (CAB2) and mRNA
abundance (CAB2 and CCR2) of two differently
phased genes are both shorter than wild type by 2 to 3 h in the
toc1 background (Millar et al., 1995a; Kreps and Simon,
1997). CAB2 expression peaks about 4 to 6 h after dawn,
whereas peak CCR2 mRNA abundance occurs 5 to 6 h later. This different
phasing, together with the aforementioned effects of CCR2
overexpression, suggests that different signaling pathways lead from
the oscillator to control of each of these two genes, and that TOC1
acts upstream of the divergence. toc1-1 also
shortens the cab2::luc rhythm induced by a
red-light pulse in etiolated seedlings, indicating that neither
photosynthesis nor the photosynthetic apparatus is necessary for an
intact circadian clock. Leaf movement and stomatal conductance rhythms
in toc1-1 are also proportionately shorter than
in the wild type (Millar et al., 1995a; Somers et al., 1998b).
The severity of the effects of toc1-1 on
flowering time depends on the ecotype. In the C24 background the
difference between short- and long-day-plant flowering time was reduced
relative to wild-type, and this was accentuated in the Laer ecotype to the extent that the plants flowered nearly as early in short- as in
long-day conditions. Clearly, toc1-1 interacts in
an allele-specific way with other loci that affect the transition to
flowering. These results also demonstrate a link between quantitative
changes in the pace of the clock and the processing of the day-length
information that lead to the switch to reproduction (Somers et al.,
1998b).
The evidence is strong, but circumstantial, that TOC1
encodes a clock component. A shorter period length could arise through a mutation in the light input pathway, causing an increased sensitivity to light relative to the wild-type. However, over a 500-fold range of
red-light intensity the period of toc1-1 is
consistently 2 to 3 h shorter than in the wild type, suggesting
that toc1-1 acts constitutively and independently
of light intensity (Somers et al., 1998b). Cloning and manipulation of
the protein, as described above, should resolve the question.
Together, the comprehensive effect that toc1-1 has on
cellular, tissue-level, and developmental processes strongly suggests that one type of oscillator mediates most, if not all, of the circadian
responses in this plant. Alternatively, different but closely related
oscillators that share one or more components (such as TOC1)
may be present in different tissues and organs in the plant. These
possibilities will become more easily addressed as more
period-affecting mutations are recovered and characterized.
| |
PERSPECTIVE |
The above considerations make it clear that the model shown in
Figure 3 is oversimplified. Feedback loops that act on only a subset of
circadian responses (e.g. CCR2 overexpression) indicate that a clock
system may be built from combinations of master/slave oscillator
interactions, which may even be tissue or organ specific. Coupling of
two or more master oscillators within a cell or whole organism have
been inferred from physiological studies with algae and humans (see
Moore-Ede et al., 1982; Roenneberg, 1996).
The levels of cryptochrome transcript and protein abundance in flies
and mouse cycle with a 24-h rhythm (Emery et al., 1998; Miyamoto
and Sancar, 1998). These results suggest a very close connection
between photoperception and the oscillator, leading some to suggest
that CRY participates directly in the animal clockwork, and blurring
the distinction between input pathway and the oscillator. Although
there are no published reports of clock-regulated expression of any
higher-plant photoreceptors, light pulse experiments have shown a
circadian rhythmicity in the rapid light induction of light-regulated
CAB expression in Arabidopsis (Millar and Kay, 1996). If
such clock-regulated light signaling also affects the light input
pathway to the oscillator itself, then a feedback loop formed by the
central oscillator controlling its own input (Roenneberg and Merrow,
1998) will obscure the neat borders outlined in Figure 3. As with much
of biology, the complete story will likely be complex. A combination of
molecular genetics and experimental physiology currently appears to be
the best approach to sorting out what may be a network of feedback
loops and interconnected oscillators, large and small, that we now
simply term the circadian clock.
| |
FOOTNOTES |
*
E-mail dsomers{at}scripps.edu; fax 858-784-9840.
Received January 21, 1999;
accepted June 8, 1999.
| |
ACKNOWLEDGMENTS |
I wish to thank members of the Kay lab for critical reading of
this manuscript. Thanks also to Steve Kay, under whose auspices I have
been working, for his support throughout, and to Martin Moore-Ede who
provided the color image of the flower clock. My apologies to the many
researchers whose work could not be cited here because of space
limitations.
| |
LITERATURE CITED |
Aronson BD,
Johnson KA,
Loros JJ,
Dunlap JC
(1994)
Negative feedback defining a circadian clock: autoregulation of the clock gene frequency.
Science
263:
1578-1584
[Abstract/Free Full Text]
Aschoff J
(1979)
Circadian rhythms: influences of internal and external factors on the period measured in constant conditions.
Z Tierpsychol
49:
225-249
[Web of Science][Medline]
Assman SM
(1993)
Signal transduction in guard cells.
Annu Rev Cell Biol
9:
345-375
[CrossRef][Web of Science]
Ballaré CL,
Sanchez RA,
Scopel AL,
Casal JJ,
Ghersa CM
(1987)
Early detection of neighbour plants by phytochrome perception of spectral changes in reflected sunlight.
Plant Cell Environ
10:
551-557
Bell-Pedersen D,
Dunlap JC,
Loros JJ
(1996)
Distinct cis-acting elements mediate clock, light, and developmental regulation of the Neurospora crassa eas (ccg-2) gene.
Mol Cell Biol
16:
513-521
[Abstract/Free Full Text]
Block GD,
Geusz ME,
Khalsa SBS,
Michel S
(1995)
A clockwork Bulla: cellular study of a model circadian system.
Semin Neurosci
7:
37-42
Bünning E
(1935)
Zur Kenntnis der erblichen Tagesperiodizität bei den Primärblattern von Phaseolus multiflorus.
Jarhb Wiss Bot
81:
411-418
Carré IA (1998) Genetic dissection of the
photoperiod-sensing mechanism in the long-day plant Arabidopsis
thaliana. In PJ Lumsden, AJ Millar, eds, Biological Rhythms and
Photoperiodism in Plants. BIOS Scientific Publishers, Oxford, pp
257-269
Carré IA,
Kay SA
(1995)
Multiple DNA-protein interactions at a circadian-regulated promoter element.
Plant Cell
7:
2039-2051
[Abstract]
Deitzer GF
(1984)
Photoperiodic induction in long-day plants.
In
D Vince-Prue,
KE Cockshull,
eds, Light and the Flowering Process.
Academic Press, London, pp 51-63
Deitzer GF,
Hayes RG,
Jabben M
(1982)
Plant Physiol
69:
597-601
[Abstract/Free Full Text]
de Mairan J (1729) Observation botanique. In Histoire
de l'Academie Royale des Sciences, pp 35-36
Dowson-Day MJ,
Millar AJ
(1999)
Circadian dysfunction causes aberrant hypocotyl elongation patterns in Arabidopsis.
Plant J
17:
63-71
[CrossRef][Web of Science][Medline]
Dunlap JC
(1996)
Genetic and molecular analysis of circadian rhythms.
Annu Rev Genet
30:
579-601
[CrossRef][Web of Science][Medline]
Dunlap JC
(1999)
Molecular bases for circadian clocks.
Cell
96:
271-290
[CrossRef][Web of Science][Medline]
Edmunds LN
(1988)
. Cellular and Molecular Bases of Biological Clocks.
Springer-Verlag, New York
Emery PT,
So WV,
Kaneko M,
Hall JC,
Rosbash M
(1998)
CRY, a Drosophila clock and light-regulated cryptochrome, is a major contributor to circadian rhythm resetting and photosensitivity.
Cell
95:
669-679
[CrossRef][Web of Science][Medline]
Engelmann W, Johnsson A (1998) Rhythms in organ movement.
In PJ Lumsden, AJ Millar, eds, Biological Rhythms and
Photoperiodism in Plants. BIOS Scientific Publishers, Oxford, pp 35-50
Fankhauser C,
Chory J
(1997)
Light control of plant development.
Annu Rev Cell Dev Biol
13:
203-229
[CrossRef][Web of Science][Medline]
Fejes E, Nagy F (1998) Molecular analysis of circadian
clock-regulated gene expression in plants: features of the 'output'
pathways. In PJ Lumsden, AJ Millar, eds, Biological Rhythms
and Photoperiodism in Plants. BIOS Scientific Publishers, Oxford, pp
99-118
Fujiwara S,
Ishida N,
Tsuzuki M
(1996)
Circadian expression of the carbonic anhydrase gene, Cah1, in Chlamydomonas reinhardtii.
Plant Mol Biol
32:
745-749
[CrossRef][Web of Science][Medline]
Golden SS,
Ishiura M,
Johnson CH,
Kondo T
(1997)
Cyanobacterial circadian rhythms.
Annu Rev Plant Physiol Plant Mol Biol
48:
327-354
[CrossRef][Web of Science]
Gorton HL,
Williams WE,
Binns ME,
Gemmell CN,
Leheny EA,
Shepherd AC
(1989)
Circadian stomatal rhythms in epidermal peels from Vicia faba.
Plant Physiol
90:
1329-1334
[Abstract/Free Full Text]
Green RM,
Tobin EM
(1999)
Loss of the circadian clock-associated protein 1 in Arabidopsis results in altered clock-regulated gene expression.
Proc Natl Acad Sci USA
96:
4176-4179
[Abstract/Free Full Text]
Heintzen C,
Nater M,
Apel K,
Staiger D
(1997)
AtGRP7, a nuclear RNA-binding protein as a component of a circadian-regulated negative feedback loop in Arabidopsis thaliana.
Proc Natl Acad Sci USA
94:
8515-8520
[Abstract/Free Full Text]
Hennessey TL,
Freeden AL,
Field CB
(1992)
Evidence of multiple circadian oscillators in bean plants.
J Biol Rhythms
7:
105-113
[Abstract/Free Full Text]
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
[Abstract/Free Full Text]
Ishiura M,
Kutsuna S,
Aoki S,
Iwasaki H,
Andersson CR,
Tanabe A,
Golden SS,
Johnson CH,
Kondo T
(1998)
Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria.
Science
281:
1519-1523
[Abstract/Free Full Text]
Johnson CH, Knight M, Trewavas A, Kondo T (1998) A clockwork
green: circadian programs in photosynthetic organisms. In PJ
Lumsden, AJ Millar, eds, Biological Rhythms and Photoperiodism in
Plants. BIOS Scientific Publishers, Oxford, pp 1-34
Johnson CH,
Knight MR,
Kondo T,
Masson P,
Sedbrook J,
Haley A,
Trewavas A
(1995)
Circadian oscillations of cytosolic and chloroplastic free calcium in plants.
Science
269:
1863-1865
[Abstract/Free Full Text]
Kim HY,
Cote GC,
Crain RC
(1993)
Potassium channels in Samanea saman protoplasts controlled by phytochrome and the biological clock.
Science
260:
960-962
[Abstract/Free Full Text]
Kloss B,
Price JL,
Saez L,
Blau J,
Rothenfluh A,
Wesley CS,
Young MW
(1998)
The Drosophila clock gene double-time encodes a protein closely related to human casein kinase I
.
Cell
94:
97-107
[CrossRef][Web of Science][Medline]
Konopka RJ,
Pittendrigh CS,
Orr D
(1989)
Reciprocal behavior associated with altered homeostasis and photosensitivity of Drosophila clock mutants.
J Neurogenet
6:
1-10
[Web of Science][Medline]
Koornneef M,
Hanhart CJ,
Van der Veen JH
(1991)
A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana.
Mol Gen Genet
229:
57-66
[Web of Science][Medline]
Kreps JA,
Kay SA
(1997)
Coordination of plant metabolism and development by the circadian clock.
Plant Cell
9:
1235-1244
[CrossRef][Web of Science][Medline]
Kreps JA,
Simon AE
(1997)
Environmental and genetic effects on circadian clock-regulated gene expression in Arabidopsis.
Plant Cell
9:
297-304
[Abstract]
Liu Z,
Taub CC,
McClung CR
(1996)
Identification of an Arabidopsis thaliana ribulose-1,5-bisphosphate carboxylase/oxygenase activase (RCA) minimal promoter regulated by light and the circadian clock.
Plant Physiol
112:
43-51
[Abstract]
Merrow MW,
Garceau NY,
Dunlap JC
(1997)
Dissection of a circadian oscillation into discrete domains.
Proc Natl Acad Sci USA
94:
3877-3882
[Abstract/Free Full Text]
Millar AJ (1998) The cellular organization of circadian rhythms in
plants: not one but many clocks. In PJ Lumsden, AJ Millar,
eds, Biological Rhythms and Photoperiodism in Plants. BIOS Scientific
Publishers, Oxford, pp 51-68
Millar AJ,
Carré IA,
Strayer CA,
Chua N-H,
Kay SA
(1995a)
Circadian clock mutants in Arabidopsis identified by luciferase imaging.
Science
267:
1161-1163
[Abstract/Free Full Text]
Millar AJ,
Kay SA
(1991)
Circadian control of cab gene transcription and mRNA accumulation in Arabidopsis.
Plant Cell
3:
541-550
[Abstract/Free Full Text]
Millar AJ,
Kay SA
(1996)
Integration of circadian and phototransduction pathways in the network controlling CAB gene transcription in Arabidopsis.
Proc Natl Acad Sci USA
93:
15491-15496
[Abstract/Free Full Text]
Millar AJ,
Short SR,
Hiratsuka K,
Chua N-H,
Kay SA
(1992)
Firefly luciferase as a reporter of regulated gene expression in higher plants.
Plant Mol Biol Rep
10:
324-337
[CrossRef]
Millar AJ,
Straume M,
Chory J,
Chua N-H,
Kay SA
(1995b)
The regulation of circadian period by phototransduction pathways in Arabidopsis.
Science
267:
1163-1166
[Abstract/Free Full Text]
Mittag M,
Lee D-H,
Hastings W
(1994)
Circadian expression of the luciferin-binding protein correlates with the binding of a protein to the 3
untranslated region of its mRNA.
Proc Natl Acad Sci USA
91:
5257-5261
[Abstract/Free Full Text]
Miyamoto Y,
Sancar A
(1998)
Vitamin B2-based blue-light photoreceptors in the retinohypothalamic tract as the photoactive pigments for setting the circadian clock in mammals.
Proc Natl Acad Sci USA
95:
6097-6102
[Abstract/Free Full Text]
Moore-Ede MC, Sulzman FM, Fuller CA (1982) The Clocks That Time
Us. Harvard University Press, Cambridge, MA
Neuhaus G,
Bowler C,
Hiratsuka K,
Yamagata H,
Chua NH
(1997)
Phytochrome-regulated repression of gene expression requires calcium and cGMP.
EMBO J
16:
2554-2564
[CrossRef][Web of Science][Medline]
Nimmo HG
(1998)
Circadian regulation of a plant protein kinase.
Chronobiol Int
15:
109-118
[Medline]
Pilgrim ML,
McClung CR
(1993)
Differential involvement of the circadian clock in the expression of genes required for ribulose-1,5-bisphosphate carboxylase/oxygenase synthesis, assembly and activation in Arabidopsis thaliana.
Plant Physiol
103:
553-564
[Abstract]
Polidoros AN,
Scandalios JG
(1998)
Circadian expression of the maize catalase Cat3 gene is highly conserved among diverse maize genotypes with structurally different promoters.
Genetics
149:
405-415
[Abstract/Free Full Text]
Roenneberg T
(1996)
The complex circadian system of Gonyaulax polyedra.
Physiol Plant
96:
733-737
[CrossRef]
Roenneberg T,
Deng TS
(1997)
Photobiology of the Gonyaulax circadian system. 1. Different phase response curves for red and blue light.
Planta
202:
494-501
[CrossRef][Web of Science]
Roenneberg T,
Merrow M
(1998)
Molecular circadian oscillators: an alternative hypothesis.
J Biol Rhythms
13:
167-179
[Abstract/Free Full Text]
Schaffer R,
Ramsay N,
Samach A,
Corden S,
Putterill J,
Carré IA,
Coupland G
(1998)
The late elongated hypocotyl mutation of Arabidopsis disrupts circadian rhythms and the photoperiodic control of flowering.
Cell
93:
1219-1229
[CrossRef][Web of Science][Medline]
Somers DE,
Devlin PF,
Kay SA
(1998a)
Phytochromes and cryptochromes in the entrainment of the Arabidopsis circadian clock.
Science
282:
1488-1490
[Abstract/Free Full Text]
Somers DE,
Webb AAR,
Pearson M,
Kay S
(1998b)
The short-period mutant, toc1-1, alters circadian clock regulation of multiple outputs throughout development in Arabidopsis thaliana.
Development
125:
485-494
[Abstract]
Sweeney BM
(1987)
Rhythmic Phenomena in Plants.
Academic Press, San Diego
Thomas B,
Vince-Prue D
(1996)
Photoperiodism in Plants.
Academic Press, London
Torii KU,
Deng XW
(1997)
The role of COP1 in light control of Arabidopsis seedling development.
Plant Cell Environ
20:
728-733
Vince-Prue D
(1994)
The duration of light and photoperiodic responses.
In
RE Kendrick,
GHM Kronenberg,
eds, Photomorphogenesis in Plants.
Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 447-490
Wang ZY,
Kenigsbuch D,
Sun L,
Harel E,
Ong MS,
Tobin EM
(1997)
A Myb-related transcription factor is involved in the phytochrome regulation of an Arabidopsis Lhcb gene.
Plant Cell
9:
491-507
[Abstract]
Wang ZY,
Tobin EM
(1998)
Constitutive expression of the CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) gene disrupts circadian rhythms and suppresses its own expression.
Cell
93:
1207-1217
[CrossRef][Web of Science][Medline]
Webb AAR (1998) Stomatal rhythms. In PJ Lumsden, AJ
Millar, eds, Biological Rhythms and Photoperiodism in Plants. BIOS
Scientific Publishers, Oxford, pp 69-79
Wilsbacher LD,
Takahashi JS
(1998)
Circadian rhythms: molecular basis of the clock.
Curr Opin Genet Dev
8:
595-602
[CrossRef][Web of Science][Medline]
Young MW
(1998)
The molecular control of circadian behavioral rhythms and their entrainment in Drosophila.
Annu Rev Biochem
67:
135-152
[CrossRef][Web of Science][Medline]
Zeng H,
Hardin PE,
Rosbash M
(1994)
Constitutive expression of the Drosophila period protein inhibits period mRNA cycling.
EMBO J
15:
3590-3598
[Web of Science][Medline]
Zheng CC,
Porat R,
Lu PZ,
O'Neill SD
(1998)
PNZIP is a novel mesophyll-specific cDNA that is regulated by phytochrome and a circadian rhythm and encodes a protein with a leucine zipper motif.
Plant Physiol
116:
27-35
[Abstract/Free Full Text]
Zhong HH,
McClung CR
(1996)
The circadian clock gates expression of two Arabidopsis catalase genes to distinct and opposite phases.
Mol Gen Genet
251:
196-203
[Web of Science][Medline]
Zylka MJ,
Shearman LP,
Levine JD,
Jin X,
Weaver DR,
Reppert SM
(1998)
Molecular analysis of mammalian timeless.
Neuron
21:
1115-1122
[CrossRef][Web of Science][Medline]
Top
Introduction
References