• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Oct 16, 2012; 109(42): 17129–17134.
Published online Oct 1, 2012. doi:  10.1073/pnas.1209148109
PMCID: PMC3479464
Plant Biology

CIRCADIAN CLOCK-ASSOCIATED 1 regulates ROS homeostasis and oxidative stress responses


Organisms have evolved endogenous biological clocks as internal timekeepers to coordinate metabolic processes with the external environment. Here, we seek to understand the mechanism of synchrony between the oscillator and products of metabolism known as Reactive Oxygen Species (ROS) in Arabidopsis thaliana. ROS-responsive genes exhibit a time-of-day–specific phase of expression under diurnal and circadian conditions, implying a role of the circadian clock in transcriptional regulation of these genes. Hydrogen peroxide production and scavenging also display time-of-day phases. Mutations in the core-clock regulator, CIRCADIAN CLOCK ASSOCIATED 1 (CCA1), affect the transcriptional regulation of ROS-responsive genes, ROS homeostasis, and tolerance to oxidative stress. Mis-expression of EARLY FLOWERING 3, LUX ARRHYTHMO, and TIMING OF CAB EXPRESSION 1 affect ROS production and transcription, indicating a global effect of the clock on the ROS network. We propose CCA1 as a master regulator of ROS homeostasis through association with the Evening Element in promoters of ROS genes in vivo to coordinate time-dependent responses to oxidative stress. We also find that ROS functions as an input signal that affects the transcriptional output of the clock, revealing an important link between ROS signaling and circadian output. Temporal coordination of ROS signaling by CCA1 and the reciprocal control of circadian output by ROS reveal a mechanistic link that allows plants to master oxidative stress responses.

Keywords: redox homeostasis, transcriptional coordination

Circadian rhythms are directed by day–night cycles so that organisms can synchronize external conditions with internal metabolism to allow temporal separation of incompatible metabolic events (1). Plants undergo aerobic metabolism, e.g., photosynthesis and respiration, which results in the generation of toxic by-products of oxygen (O2) known as Reactive Oxygen Species (ROS) (2). The photoreduction of O2 to H2O gives rise to singlet oxygen, superoxide anion (O2−), hydrogen peroxide (H2O2), and hydroxyl radical (·OH) (3). If the production of ROS is left unmanaged, plants may experience oxidative stress due to an imbalance in cellular redox state that eventually leads to cell death (2). Thus, plants have evolved enzymatic and nonenzymatic scavenging machineries to keep ROS at physiologically permissive levels (4).

The activation and monitoring of stress-responsive pathways are energetically demanding. Not surprisingly, circadian gating of stress pathways has been found to confer maximal tolerance to stress while minimizing the use of resources (5, 6). The Arabidopsis circadian clock consists of a core feedback loop that connects morning- and evening-phase circuits (7). The core loop is made up of two morning-expressed Myb transcription factors (TFs)—CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY)—that inhibit the expression of an evening-expressed pseudoresponse regulator TIMING OF CAB EXPRESSION 1 (TOC1) through the association of CCA1 to the Evening Element (EE) motif in the TOC1 promoter (8, 9). The morning loop consists of TOC1 homologs, the PSEUDO-RESPONSE REGULATOR (PRR) 7 and PRR9, which are partially redundant in repressing the transcription of CCA1 and LHY (10). In the evening loop, EARLY FLOWERING (ELF) 3, ELF4, and LUX ARRHYTHMO (LUX) drive CCA1 and LHY expression and are required for clock function under constant light (LL) (1113).

Because ROS can be used as important secondary messengers during stress (3, 14), it may be advantageous for ROS homeostasis to be in tune with daily light–dark cycles to enhance fitness. Moreover, as photosynthesis is driven by the sun, ROS levels would fluctuate across the day, and organisms may evolve systems to cope with periodic increase in such toxic products. However, the underlying mechanisms and the biological importance of restricting stress responses to certain times of the day have not been fully elucidated. Here, we show that the circadian clock coordinates ROS homeostasis and transcriptional response. The mechanisms for signaling circadian time to a proposed clock-controlled output, the ROS pathway, are discussed.


ROS Production and Scavenging Are Regulated by Diurnal Cycles.

Photosynthetic genes peak at Zeitgeber time (ZT) 4 (15), suggesting that, at this time, light-harvesting capacity peaks, and such metabolic changes may tilt the balance of ROS production and scavenging. Thus, we hypothesized that ROS production and scavenging may exhibit time-of-day–specific changes. To test this, H2O2 levels were quantified from plants grown under 12-h light/12-h dark (LD) photocycles. Indeed, H2O2 production peaks at noon, ZT7 [0.117 ± 0.009 μmol/mg fresh weight (FW)] and reaches trough levels (0.015 ± 0.008 μmol/mg FW) at midnight, ZT19 (Fig. 1 A and C). We also determined the time-of-day activity of the H2O2 scavenger catalase in LD-grown plants. As expected, catalase activity peaks at ZT7 (0.611 ± 0.021 U/mL) and dips at ZT19 (0.156 ± 0.039 U/mL; Fig. 1B), coinciding with the peak and trough of H2O2.

Fig. 1.
ROS homeostasis is regulated by diurnal cycles. (A) H2O2 levels and (B) catalase enzyme activity from 16-d-old Col-0 plants entrained in LD. Error bars represent SEM of n = 20. One-way ANOVA (effect of time) for LD profiles of H2O2 and catalase was significant ...

As we observed diurnal rhythms in ROS production and scavenging, we tested whether genes from the ROS network (16) are regulated similarly. We obtained time-course expression profiles of 167 ROS-responsive genes (Dataset S1), selected from a group of general oxidative stress response markers (16), by quantitative PCR (qPCR). Of the 167 genes, 140 genes display time-of-day–specific phases (one-way ANOVA, P < 0.0001) under long-day photocycles (Fig. 1D). As expected, the largest gene cluster peaks at noon (Fig. 2E), coinciding with the peak of ROS levels (Fig. 1 A and C). To ascertain other biological processes that might be coregulated by these genes, we analyzed Gene Ontology (GO) enrichment and found that genes associated with stresses and abiotic stimuli are overrepresented (Dataset S2).

Fig. 2.
The circadian clock regulates ROS homeostasis. (A) H2O2 levels and (B) catalase enzyme activity quantified from 16-d-old Col-0 plants entrained in LD and transferred to LL before sample collection. Error bars represent SEM of n = 20. One-way ANOVA (effect ...

Circadian Clock Regulates ROS Production, Scavenging, and Transcription of ROS-Responsive Genes.

The phase relationship between ROS production, scavenging, and ROS-driven transcription suggests coordinated regulation of this network within the diel cycle. We next determined whether the observed effect persists in LL conditions, indicating clock regulation. It is possible that H2O2 levels remain high in LL due to the overreduction of electron acceptors (17). Indeed, H2O2 levels are elevated in both subjective day and night in LL (Fig. 2 A and C). Interestingly, we still observed a time-of-day–specific peak at midday (ZT7; one-way ANOVA, P < 0.001) albeit with a lower peak:trough ratio of 1.49 (Fig. 2A) compared with the peak:trough ratio in LD of 7.86. Likewise, catalase activity, although elevated in LL (Fig. 2B), still shows a significant time-of-day peak with a reduced peak:trough ratio in LL (1.59) compared with LD (3.92). To test if the circadian clock exerts transcriptional coordination over ROS signaling, expression profiling of the same 167 genes was performed on LD entrained plants that were transferred to LL. We observed that ROS genes are phased to the subjective midday (ZT10; one-way ANOVA, P < 0.0001) under LL (Fig. 2D), and this implies a clock-regulated effect. Under LD condition, only 39% of the cycling ROS genes exhibit a noon phase (ZT10) whereas in LL, over 75% of cycling genes peak at the subjective midday (Fig. 2E). Taken together, these persistent time-of-day signals in LL (Fig. 2D) suggest that the circadian clock regulates the ROS transcriptional response network.

Functional Clock Is Required for the Time-of-Day–Specific Regulation of ROS Production.

To further investigate the role of the clock in ROS signaling, we examined the sensitivity of clock mutants to ROS-generating agents. Plants with mutations in CCA1, LHY, ELF3, ELF4, LUX, TIME FOR COFFEE (TIC) (18), PRR5, PRR9, and PRR7 are hypersensitive (Fig. 3 and Fig. S1) to the application of 5 μM methyl viologen (MV), which causes an increase in superoxide levels (19). In contrast, plants overexpressing CCA1 (CCA1-ox) (8) are hyposensitive to MV (Fig. 3). The observed hypersensitivity of cca1-1, lhy-11, and cca1-1/lhy-11 mutants to ROS-generating agents could be a result of impaired ROS homeostasis in these genotypes. Under both LD and LL conditions, cca1-1/lhy-11 mutants exhibit higher H2O2 and lower catalase levels compared with WT plants (Fig. 4 AC). Interestingly, overexpression of CCA1 suppresses H2O2 levels under both LD and LL (Fig. 4 A and C). However, catalase activity in CCA1-ox plants is found to be lower than WT levels only during the day (in LD) and the subjective day (in LL; Fig. 4B). Both cca1-1 and lhy-11 single mutants exhibit high H2O2 levels in the evening and night in LD but not in LL (Fig. S2 A and C). Catalase activity is also decreased in cca1-1 and lhy-11 single mutants during the day in LD and the subjective day in LL (Fig. S2B). As previously described (20), all three catalase genes—CAT1, CAT2, and CAT3—display time-of-day–specific phases in LL. CAT1 and CAT3 peak at noon (ZT7) whereas CAT2 peaks at dawn (ZT3; Fig. 4D). Similar to previous findings, we observed that either CCA1 overexpression or mutations in CCA1 and LHY result in altered CAT1, CAT2, and CAT3 expressions (Fig. 4D) (8, 21). Loss of function in clock genes may cause circadian arrhythmia (22), and, to investigate global clock effects on ROS production, H2O2 levels were determined in mutants that have defective clocks. Indeed, H2O2 levels are high in elf3-1, elf4-101, and tic-1 (Fig. S2C). Temporal expression profiling revealed that CAT1, CAT2, and CAT3 expression is arrhythmic in elf3-1 (Fig S2D) and that the expression of CAT genes is changed in both lux-1 and toc1-1 (Fig. S2C). Thus, our results are consistent with the suggestion that the differential sensitivity of clock mutants to oxidative stress reflects perturbations in the basal cellular ROS homeostasis in these mutants, although the relationship with MV sensitivity and circadian effects appears complex.

Fig. 3.
Mutations in CCA1 and LHY resulted in ROS hypersensitivity. Plants were entrained in LD and transferred to LL for a day before 5 μM MV treatment was administered at ZT3 day 16. (A) Mean number of wilted leaves was scored after 24 h of treatment. ...
Fig. 4.
H2O2 and catalase rhythms are regulated by CCA1. (A) H2O2 levels and (B) catalase enzyme activity from 16-d-old WT, CCA1-ox, and cca1-1/lhy-11 plants entrained in LD (Left) and transferred to LL (Right). Error bars represent SEM of n = 20. One-way ANOVA ...

CCA1 Regulates the Coordinated Transcription of ROS Genes in the Absence of Oxidative Stress.

To determine whether the circadian clock regulates the ROS network at the transcriptional level, we investigated whether genes grouped under various ROS-related GO categories (referred to as ROS GO genes hereafter) in Arabidopsis exhibit circadian rhythmicity. We observed that, of the 517 ROS GO genes, on average 73 and 39% of the genes are rhythmic in two or more diurnal and circadian conditions, respectively (Dataset S3). Of these rhythmic genes, the midday phase (ZT5) is found to be overrepresented in ROS GO genes (Z-score > 1.96; Fig. S3A) when cycling calls were made on two LD datasets (LDHH_SM and LDHH_ST). Furthermore, cycling calls on two short-day (8 h light/16 h dark) datasets (COL_SD and LER_SD) revealed that the ZT4 phase is enriched (Z-score > 1.96; Fig. S3B). These results are consistent with an independent study where 34% of ROS genes were found to be clock-regulated (5).

To investigate whether this time-of-day expression of ROS genes is regulated by central components of the circadian clock, we examined the expression of a subset of ROS genes in circadian clock mutants. We performed a 48-h expression profiling on 32 selected transcripts in LD entrained plants released to LL in elf3-1, lux-1, toc1-1, cca1-11/lhy-11, and CCA1-ox. These 32 ROS genes are selected on the basis of their involvement in ROS signaling (Dataset S4). Of the 32 genes, 24 genes exhibit time-of-day–specific phases in WT plants (one-way ANOVA, P < 0.001; Figs. S4A and S5A; Dataset S5). No overt phases could be detected in the remaining eight genes (Fig. S6). These 24 genes can be grouped into three clusters based on the time when they peak, i.e., midday (ZT7), evening (ZT11), and predawn (ZT23; Figs. S4A and S5A). Genes of the phenylpropanoid pathway are known to peak at predawn (15). Indeed, we observed that the ascorbate biosynthesis gene VITAMIN C 2 (VTC2) (3) has a predawn phase (Fig. S4B). Phase-specific expression of VTC2 is altered in CCA1-ox and cca1-1/lhy-11 (Fig. S4B). Another ROS scavenger is ascorbate peroxidase (APX), and deficiency in this enzyme results in light-induced necrosis (23). The arrhythmic expression profiles of APX4 in CCA1-ox and cca1-1/lhy-11 are distinct from the expression profile in WT plants where a midday phase (ZT7) can be observed (Fig. S4B). The phase of APX4 appears to be shifted in elf3-1 and lux-1 (Fig. S5B). Other well-studied genes that exhibit phase-specific expression are the predawn phased (ZT23) HEAT SHOCK PROTEIN 18.2 (HSP18.2), the midday phased (ZT7) PHENYLALANINE AMMONIA-LYASE 1 (PAL1), AT2G22420 (PEROXIDASE) and HEAT SHOCK TRANSCRIPTION FACTOR A4A (HSFA4A), and the evening-phased (ZT11) MYB DOMAIN PROTEIN 59 (MYB59; Fig. S4B) (2427). All five genes show dramatic changes in time-of-day expression in CCA1-ox, cca-1/lhy-11, and elf3-1 (Fig. S4B and S5B). In all elf3-1, lux-1, and toc-1, midday-phased genes show altered expression (Fig. S5A). Genes from the evening and predawn cluster appear to display altered amplitudes in toc1-1 (Fig. S5A). In elf3-1, evening and predawn genes are arrhythmic (Fig. S5A). Evening-phased genes show altered phase of expression in lux-1, whereas the predawn cluster still appears to show a broadened peak (Fig. S5A). In general, phase-specific expressions of the remaining genes from all three clusters are altered in CCA1-ox and cca1-1/lhy-11 (Fig. S4A), suggesting that rhythmic oscillation in CCA1 levels may be essential for transcriptional coordination of ROS genes.

Seven of the eight ROS genes did not display any specific time-of-day phase in the conditions tested. Nevertheless, they exhibit altered expression in CCA1-ox, cca1-1/lhy-11, elf3-1, lux-1, and toc1-1 plants (Fig. S6). This may be the cause of altered ROS homeostasis in these mutants (Figs. 3 and 4 AC; Figs. S1 and S2C). This also suggests a potential for regulation of the ROS network that is beyond circadian regulation of rhythmic ROS genes. PYRIDOXINE BIOSYNTHESIS 1 (PDX1) encodes an enzyme involved in production of the ROS quencher pyridoxine that increases in levels upon UV-B irradiation (28). Notably, the expression of PDX1 is elevated in cca1-1/lhy-11 (Fig. S6). It is also interesting that the expression of RESPIRATORY BURST OXIDASE HOMOLOG C (RBOHC) (29) is elevated in CCA1-ox and toc1-1 mutants (Fig. S6). Collectively, the results suggest that the circadian clock mediates the expression of ROS-signaling genes under regular growth conditions.

CCA1 Regulates Responses to Oxidative Stress.

The circadian clock in plants may regulate phase-specific expression of ROS genes to allow the anticipation of oxidative stress according to a diurnal schedule. Of the 32 genes tested in the previous section, 28 had a putative EE and/or CCA1-binding site (CBS) in their upstream promoter regions (Dataset S5). This, combined with the altered expression in CCA1-ox and cca1-1/lhy-11 mutants, prompted us to investigate if these responses were mediated directly by CCA1. If CCA1 mediates this response, it would depend on the time of CCA1 expression, i.e., at dawn. We focused on seven genes (Dataset S5), five of which are ROS TFs that rapidly respond to ROS treatments (30, 31), and first examined their expression in LL. Time-of-day specific phasing is observed in six of the seven genes, and these genes also display altered expression in CCA1-ox and cca1-1/lhy-11 plants (Fig. 5A). To determine if CCA1 mediates the response to oxidative stress, we induced oxidative stress in WT plants, using 2 μM MV, at three different times in a single LD cycle, i.e., morning (ZT3), evening (ZT11), and midnight (ZT19). For all genes tested, treatments in the morning, but not evening or night, result in significant inductions (Student’s t test, P < 0.0001; Fig. 5B), consistent with the prediction that the system is the most responsive in the morning when CCA1 is expressed. Interestingly, for six genes, down-regulation (Student’s t test, P < 0.01) of gene expression is observed in evening-treated plants (Fig. 5B). In the CCA1-ox background, five of the genes that are repressed at ZT11 in WT plants show significant inductions in CCA1-ox (Fig. 5C). Furthermore, the diurnal response to ROS, observed in WT plants with the peak response at ZT3 (Fig. 5B), is abolished in CCA1-ox (Fig. 5C). The response to different concentrations of MV is also attenuated in CCA1-ox (Fig. 5C and Fig. S7). Moreover, we noted that the expression of all seven genes is, in part, also affected by mutations in ELF3, LUX, and TOC1 under diurnal conditions (Fig. S8). These mutants have altered circadian clocks; thus, the observed changes in gene expression support the importance of circadian regulation of these genes. In short, the observed diurnal response of ROS TFs suggests that plants may be most responsive to ROS treatments in the morning when CCA1 is expressed (Fig. 5B).

Fig. 5.
Response to ROS is regulated by diurnal cycles and is dependent on the time of CCA1 expression. (A) Expression profiles of seven ROS TFs obtained from plants (n = 15) entrained in LD and transferred to LL before sampling. One-way ANOVA (effect of time) ...

WRKY11, MYB59, PAL1, and ZAT12 Are Direct Targets of CCA1 in Vivo.

We next determined whether CCA1 could physically associate with the promoters of ROS genes in vivo. WRKY11, MYB59, PAL1, and ZAT12 have a time-of day–specific expression that is altered in plants with mis-regulated CCA1 (Fig. S4B and Fig. 5A). This observation and the presence of EE and/or CBS in their promoter regions (Fig. S9 A and C) suggest that CCA1 may be a direct regulator of these genes. We performed chromatin immunoprecipitation (ChIP)–qPCR assays using pCCA1::CCA1-GFP transgenic lines at the peak of CCA1 expression (ZT1) with CAT3 (32) as a positive control (Dataset S5). In addition to the four aforementioned genes, six additional ROS genes involved in transcriptional regulation were assayed, and enrichments are detected for EE/CBS-containing promoter fragments of COR27 (33), JUMONJI DOMAIN PROTEIN 5 (JMJD5) (34), ZAT12, MYB59, WRKY11, and PAL1 and not in negative control sequences (Fig. S9B). However, enrichment is not observed in PUP1, PEROXIDASE, and METHYL ESTERASE 18 (MES18) although all three genes contain putative CBS sequences (Fig. S9D).

ROS Signals Feed-Back to Affect a Clock Output.

Because interlocking feedback loops are common in other metabolic processes (35), we hypothesized that ROS signals may feed-back to affect processes controlled by the oscillator. Such feedback may allow the plant to reset the ROS-signaling cascade through crosstalk with other pathways, and plants could continuously monitor the changes in ROS levels under various physiological conditions. To investigate the effects of ROS on circadian output, CHLOROPHYLL A/B-BINDING PROTEIN::LUCIFERASE (CAB2::LUC) and FLAVIN BINDING, KELCH REPEAT, F-BOX1 (FKF1::LUC) were used. These output genes were selected on the basis of both being likely targets of CCA1 and containing the EE in their promoters (8, 3638). ROS treatments were achieved by administering different doses of MV, H2O2, the catalase inhibitor 3-aminotriazole (39), and the peroxidase inhibitor salicylhydroxamic acid (40). As ROS signaling is mediated by both production and scavenging (14), inhibition of ROS production can be expected to have similar effects to inducing ROS production. Diphenyleneiodonium chloride and potassium iodide (41) were used to inhibit NADPH oxidase activity and to scavenge H2O2, respectively. We observed that FKF1 shows a dramatic response to chronic ROS treatments. The FKF1 reporter shows phase delays and dose-dependent lengthening of the FKF1 period (Fig. S10A). The period and phase of CAB2 (Fig. S10B) are, however, not significantly affected by either ROS induction or inhibition. Our results suggest that the clock is involved in coordinated transcriptional regulation of ROS genes with CCA1 being a likely master regulator of ROS networks. In addition, we observed that ROS could feed-back to affect the transcription of a clock-regulated output, FKF1.


Here we examine the potential crosstalk between the circadian clock and the ROS transcriptional network. Our results provide evidence that ROS production, response, and transcriptional regulation of ROS genes are controlled by the circadian clock. ROS genes show time-of-day–specific expression patterns that persist in constant conditions and that this temporal regulation of ROS is also reflected in the enzymatic activity of catalase and the production of H2O2 (Figs. 1 and and2).2). The changes in phenotypic responses to oxidative stress in plants containing mutations in the components of the circadian clock reflect the importance of the circadian clock in regulating this response (Fig. 3 and Fig. S1). Further investigation reveals that the loss of CCA1 and LHY impairs time-of-day–specific ROS production and scavenging (Figs. 4 and and5B).5B). Moreover, the expression pattern of ROS genes is coordinated in a diurnal and circadian manner that is dependent on components of the circadian clock (Figs. S4 and S5). This coordinated regulation, combined with the hypothesis that ROS levels may reflect the metabolic state of the plant (42), suggests that metabolic needs may be partitioned to different times of the diel cycle. Notably, transcriptional coordination of ROS genes may be driven in part by CCA1 rhythms per se (Fig. S4).

Although the sensitivity to MV and the ROS production levels in circadian mutants were altered compared with WT plants, a direct correlation between period effects in the mutants and ROS responses was not found. The complex feedback between the circadian components themselves could contribute to such lack of linearity. For example, in a short-period mutant, CCA1 will be expressed at different times of the day than expected, and sensitivity of plants depends on the time-of-day of CCA1 expression (Fig. 5B). Combined with this, the pleiotropic phenotypes of some of the circadian mutants (elf3-1 and lux-1) could cause the connection to be skewed, making it complicated to untangle the relationship between ROS response and the circadian clock. Indeed, the lack of a linear relationship between ROS sensitivity and circadian period is reflected by the differential effects of ROS treatments on the circadian output genes CAB2 and FKF1. This suggests a complicated role potentially involving multiple oscillators and tissue specificity. If this is the case, endogenous ROS production and scavenging may become out of sync in different ways in the various circadian mutants.

A metaphorical “gate” governs the sensitivity of the clock to resetting signals presented at different times of the day (43). CCA1 regulates the expression of ROS genes only when it is expressed at dawn, and, if oxidative stress is administered at night when CCA1 levels are low, this effect is not observed (Fig. 5B). However, the response is not limited to dawn in CCA1-ox plants, indicating that CCA1 is an important component of this gate. Overexpression of CCA1 appears to enable a continually permissive state for ROS responses (Fig. 5C) and could therefore account for the low basal ROS levels in this mutant (Fig. 4 A and C). These lower basal ROS levels or the overexpression of CCA1 could cause the attenuated induction of ROS genes as observed in CCA1-ox in response to oxidative stress (Fig. 5C and Fig. S7). Further analysis will be necessary to determine the causative signaling pathway regulating these transcriptional responses.

The circadian clock may receive inputs from multiple metabolic pathways to use this information to fine-tune clock function (44, 45). Thus, it is possible for ROS signals to exert indirect effects on clock output pathway(s) when the oscillator is perturbed. Notably, transcription of FKF1 is altered by ROS treatments (Fig. S10A). Both FKF1 and CAB2 were selected on the basis of both being CCA1-regulated outputs. The identification of the effects of ROS on the expression of FKF1, a known regulator of flowering time (46), could also provide a potential link between previously reported studies on antioxidants and flowering. For example, the vtc1 mutant has alterations in flowering time that are photoperiod-dependent (47), perhaps due to the accumulation of ROS and its effect on FKF1 expression. Also, when the antioxidant ascorbic acid was increased artificially, flowering could be delayed, and this correlates with lower mRNA levels of circadian clock and photoperiodic genes (47). FKF1 is also involved in the degradation of TOC1 through the interaction with GIGANTEA (GI). Indeed, the role of GI in oxidative stress tolerance has been implicated (48). PRR5, PRR7, and PRR9 have been linked to the regulation of ROS production (49) where the prr5-1, prr9-1, prr7-3, and prr5-1/prr9-1 mutants are hypersensitive to ROS treatments (Fig. S1). Interestingly, these mutants also have altered levels of CCA1 (50), which possibly explains the mechanism of this effect.

The observed effect of ROS on the expression FKF1 but not on CAB2 suggests that the effects of ROS-generating agents on the circadian output are not uniform. This disconnection could result from the presence of multiple oscillators or tissue-specific differences in the expression of FKF1 and CAB2 (51, 52). FKF1 is expressed in both vascular bundles and mesophyll cells whereas CAB2 is expressed in mesophyll cells and epidermal guard cells (53, 54). It was previously observed that CAB2 and CAT3 are regulated by two oscillators within the same tissue (51). Both genes respond differently to temperature signals although the spatial expression patterns of CAB2 and CAT3 overlap in the mesophyll. The toc1-1 mutant has short period rhythms of CAB2 but has a wild-type period for cytosolic Ca2+ oscillation whereas the toc1-2 mutant has a short period for both CAB2 and Ca2+ rhythms, which suggests that different mechanisms regulate the rhythms of CAB2 and Ca2+ (52). We contemplated the possibility that because CAB2 is involved in photosynthetic responses and may be exposed to ROS fluctuations, its rhythms are buffered from ROS responses by some mechanism—perhaps a second oscillator. It is also noteworthy that CAB2 is repressed by glucose and fructose (55), and therefore it is possible for CAB2 rhythms to be buffered from other products of photosynthesis. Alternatively, the differential responses between FKF1 and CAB2 reporter constructs could also be due to the time of day at which the genes are expressed. Because CCA1 represses many targets like FKF1 that are expressed at night, whereas CCA1 appears to be a positive regulator of CAB2 (8, 37, 38), the differences in ROS response could be due to different mechanisms of CCA1 action on CAB2 versus FKF1.

Recently, an enzyme that senses H2O2 has been proposed to be the ancestor of all biological clocks where it is assumed that clocks either evolved to confer metabolic advantage to anticipate the presence of ROS or to cope with periodic fluctuations in UV radiation from the sun (56). Our results have shown the mechanistic relationship between ROS homeostasis and biological timekeeping, which may have coevolved as a result of the Great Oxidation Event (56). We demonstrate that CCA1 is a central regulator of the ROS-responsive transcriptional network where it is essential for the coordinated response to oxidative stress and the regulation of ROS production and scavenging.

Materials and Methods

Details are described in SI Materials and Methods. This includes information on plant materials, growth conditions, H2O2 and catalase assays, ROS hypersensitivity assays, ROS treatments for transcript analysis, ChIP-qPCR assays, bioluminescence assays, and bioinformatics analyses. Primers used for expression analysis are provided in Dataset S6.

Supplementary Material

Supporting Information:


We thank D. Mertten and J. Jayaraman for technical assistance and C. R. McClung, D. Hincha, and E. Tobin for gifts of seeds. We thank the Institute of Molecular Biosciences for funding (A.G.L.) and the Bundesministerium für Bildung und Forschung - Golm Forschungseinheit zur Systembiologie Systems Biology Research Initiative for funding (Grant FKZ 0313924 to B.M.-R.).


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1209148109/-/DCSupplemental.


1. Hotta CT, et al. Modulation of environmental responses of plants by circadian clocks. Plant Cell Environ. 2007;30(3):333–349. [PubMed]
2. Apel K, Hirt H. Reactive oxygen species: Metabolism, oxidative stress, and signal transduction. Annu Rev Plant Biol. 2004;55:373–399. [PubMed]
3. Foyer CH, Noctor G. Oxidant and antioxidant signalling in plants: A re-evaluation of the concept of oxidative stress in a physiological context. Plant Cell Environ. 2005;28:1056–1071.
4. Mittler R. Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci. 2002;7(9):405–410. [PubMed]
5. Covington MF, Maloof JN, Straume M, Kay SA, Harmer SL. Global transcriptome analysis reveals circadian regulation of key pathways in plant growth and development. Genome Biol. 2008;9(8):R130. [PMC free article] [PubMed]
6. Legnaioli T, Cuevas J, Más P. TOC1 functions as a molecular switch connecting the circadian clock with plant responses to drought. EMBO J. 2009;28(23):3745–3757. [PMC free article] [PubMed]
7. Pruneda-Paz JL, Kay SA. An expanding universe of circadian networks in higher plants. Trends Plant Sci. 2010;15(5):259–265. [PMC free article] [PubMed]
8. Wang ZY, Tobin EM. Constitutive expression of the CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) gene disrupts circadian rhythms and suppresses its own expression. Cell. 1998;93(7):1207–1217. [PubMed]
9. Alabadí D, et al. Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science. 2001;293(5531):880–883. [PubMed]
10. Farré EM, Harmer SL, Harmon FG, Yanovsky MJ, Kay SA. Overlapping and distinct roles of PRR7 and PRR9 in the Arabidopsis circadian clock. Curr Biol. 2005;15(1):47–54. [PubMed]
11. Hicks KA, et al. Conditional circadian dysfunction of the Arabidopsis early-flowering 3 mutant. Science. 1996;274(5288):790–792. [PubMed]
12. Doyle MR, et al. The ELF4 gene controls circadian rhythms and flowering time in Arabidopsis thaliana. Nature. 2002;419(6902):74–77. [PubMed]
13. Hazen SP, et al. LUX ARRHYTHMO encodes a Myb domain protein essential for circadian rhythms. Proc Natl Acad Sci USA. 2005;102(29):10387–10392. [PMC free article] [PubMed]
14. D’Autréaux B, Toledano MB. ROS as signalling molecules: Mechanisms that generate specificity in ROS homeostasis. Nat Rev Mol Cell Biol. 2007;8(10):813–824. [PubMed]
15. Harmer SL, et al. Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science. 2000;290(5499):2110–2113. [PubMed]
16. Gadjev I, et al. Transcriptomic footprints disclose specificity of reactive oxygen species signaling in Arabidopsis. Plant Physiol. 2006;141(2):436–445. [PMC free article] [PubMed]
17. Velez-Ramirez AI, van Ieperen W, Vreugdenhil D, Millenaar FF. Plants under continuous light. Trends Plant Sci. 2011;16(6):310–318. [PubMed]
18. Hall A, et al. The TIME FOR COFFEE gene maintains the amplitude and timing of Arabidopsis circadian clocks. Plant Cell. 2003;15(11):2719–2729. [PMC free article] [PubMed]
19. Bowler C, Van Montagu M, Inzé D. Superoxide dismutase and stress tolerance. Annu Rev Plant Physiol Plant Mol Biol. 1992;43:83–116.
20. Zhong HH, McClung CR. The circadian clock gates expression of two Arabidopsis catalase genes to distinct and opposite circadian phases. Mol Gen Genet. 1996;251(2):196–203. [PubMed]
21. Green RM, Tobin EM. Loss of the circadian clock-associated protein 1 in Arabidopsis results in altered clock-regulated gene expression. Proc Natl Acad Sci USA. 1999;96(7):4176–4179. [PMC free article] [PubMed]
22. Nakamichi N. Molecular mechanisms underlying the Arabidopsis circadian clock. Plant Cell Physiol. 2011;52(10):1709–1718. [PMC free article] [PubMed]
23. Giacomelli L, Masi A, Ripoll DR, Lee MJ, van Wijk KJ. Arabidopsis thaliana deficient in two chloroplast ascorbate peroxidases shows accelerated light-induced necrosis when levels of cellular ascorbate are low. Plant Mol Biol. 2007;65(5):627–644. [PubMed]
24. Stracke R, Werber M, Weisshaar B. The R2R3-MYB gene family in Arabidopsis thaliana. Curr Opin Plant Biol. 2001;4(5):447–456. [PubMed]
25. Rohde A, et al. Molecular phenotyping of the pal1 and pal2 mutants of Arabidopsis thaliana reveals far-reaching consequences on phenylpropanoid, amino acid, and carbohydrate metabolism. Plant Cell. 2004;16(10):2749–2771. [PMC free article] [PubMed]
26. Nishizawa A, et al. Arabidopsis heat shock transcription factor A2 as a key regulator in response to several types of environmental stress. Plant J. 2006;48(4):535–547. [PubMed]
27. Mu R-L, et al. An R2R3-type transcription factor gene AtMYB59 regulates root growth and cell cycle progression in Arabidopsis. Cell Res. 2009;19(11):1291–1304. [PubMed]
28. Ristilä M, Strid H, Eriksson LA, Strid Å, Sävenstrand H. The role of the pyridoxine (vitamin B6) biosynthesis enzyme PDX1 in ultraviolet-B radiation responses in plants. Plant Physiol Biochem. 2011;49(3):284–292. [PubMed]
29. Foreman J, et al. Reactive oxygen species produced by NADPH oxidase regulate plant cell growth. Nature. 2003;422(6930):442–446. [PubMed]
30. Davletova S, Schlauch K, Coutu J, Mittler R. The zinc-finger protein Zat12 plays a central role in reactive oxygen and abiotic stress signaling in Arabidopsis. Plant Physiol. 2005;139(2):847–856. [PMC free article] [PubMed]
31. Journot-Catalino N, Somssich IE, Roby D, Kroj T. The transcription factors WRKY11 and WRKY17 act as negative regulators of basal resistance in Arabidopsis thaliana. Plant Cell. 2006;18(11):3289–3302. [PMC free article] [PubMed]
32. Michael TP, McClung CR. Phase-specific circadian clock regulatory elements in Arabidopsis. Plant Physiol. 2002;130(2):627–638. [PMC free article] [PubMed]
33. Mikkelsen MD, Thomashow MF. A role for circadian evening elements in cold-regulated gene expression in Arabidopsis. Plant J. 2009;60(2):328–339. [PubMed]
34. Lu SX, et al. The Jumonji C domain-containing protein JMJ30 regulates period length in the Arabidopsis circadian clock. Plant Physiol. 2011;155(2):906–915. [PMC free article] [PubMed]
35. Dodd AN, et al. The Arabidopsis circadian clock incorporates a cADPR-based feedback loop. Science. 2007;318(5857):1789–1792. [PubMed]
36. Carre IA, Kay SA. Multiple DNA-protein complexes at a circadian-regulated promoter element. Plant Cell. 1995;7(12):2039–2051. [PMC free article] [PubMed]
37. Imaizumi T, Tran HG, Swartz TE, Briggs WR, Kay SA. FKF1 is essential for photoperiodic-specific light signalling in Arabidopsis. Nature. 2003;426(6964):302–306. [PubMed]
38. Mizoguchi T, et al. Distinct roles of GIGANTEA in promoting flowering and regulating circadian rhythms in Arabidopsis. Plant Cell. 2005;17(8):2255–2270. [PMC free article] [PubMed]
39. Gechev T, et al. Hydrogen peroxide protects tobacco from oxidative stress by inducing a set of antioxidant enzymes: Cell Mol Life Sci. 2002;59(4):708–714. [PubMed]
40. Brouwer KS, van Valen T, Day DA, Lambers H. Hydroxamate-stimulated O2 uptake in roots of Pisum sativum and Zea mays, mediated by a peroxidase: Its consequences for respiration requirements. Plant Physiol. 1986;82(1):236–240. [PMC free article] [PubMed]
41. Tsukagoshi H, Busch W, Benfey PN. Transcriptional regulation of ROS controls transition from proliferation to differentiation in the root. Cell. 2010;143(4):606–616. [PubMed]
42. Mittler R, Vanderauwera S, Gollery M, Van Breusegem F. Reactive oxygen gene network of plants. Trends Plant Sci. 2004;9(10):490–498. [PubMed]
43. McWatters HG, Devlin PF. Timing in plants: A rhythmic arrangement. FEBS Lett. 2011;585(10):1474–1484. [PubMed]
44. Stephan FK. The “other” circadian system: Food as a Zeitgeber. J Biol Rhythms. 2002;17(4):284–292. [PubMed]
45. Gibon Y, et al. Integration of metabolite with transcript and enzyme activity profiling during diurnal cycles in Arabidopsis rosettes. Genome Biol. 2006;7(8):R76. [PMC free article] [PubMed]
46. Nelson DC, Lasswell J, Rogg LE, Cohen MA, Bartel B. FKF1, a clock-controlled gene that regulates the transition to flowering in Arabidopsis. Cell. 2000;101(3):331–340. [PubMed]
47. Kotchoni SO, Larrimore KE, Mukherjee M, Kempinski CF, Barth C. Alterations in the endogenous ascorbic acid content affect flowering time in Arabidopsis. Plant Physiol. 2009;149(2):803–815. [PMC free article] [PubMed]
48. Kurepa J, Smalle J, Van Montagu M, Inzé D. Oxidative stress tolerance and longevity in Arabidopsis: The late-flowering mutant gigantea is tolerant to paraquat. Plant J. 1998;14(6):759–764. [PubMed]
49. Fukushima A, et al. Impact of clock-associated Arabidopsis pseudo-response regulators in metabolic coordination. Proc Natl Acad Sci USA. 2009;106(17):7251–7256. [PMC free article] [PubMed]
50. Nakamichi N, et al. PSEUDO-RESPONSE REGULATORS 9, 7, and 5 are transcriptional repressors in the Arabidopsis circadian clock. Plant Cell. 2010;22(3):594–605. [PMC free article] [PubMed]
51. Michael TP, Salome PA, McClung CR. Two Arabidopsis circadian oscillators can be distinguished by differential temperature sensitivity. Proc Natl Acad Sci USA. 2003;100(11):6878–6883. [PMC free article] [PubMed]
52. Hotta CT, et al. Are there multiple circadian clocks in plants? Plant Signal Behav. 2008;3(5):342–344. [PMC free article] [PubMed]
53. Sawa M, Nusinow DA, Kay SA, Imaizumi T. FKF1 and GIGANTEA complex formation is required for day-length measurement in Arabidopsis. Science. 2007;318(5848):261–265. [PMC free article] [PubMed]
54. Millar AJ, Carré IA, Strayer CA, Chua NH, Kay SA. Circadian clock mutants in Arabidopsis identified by luciferase imaging. Science. 1995;267(5201):1161–1163. [PubMed]
55. Cho YH, Yoo SD. Signaling role of fructose mediated by FINS1/FBP in Arabidopsis thaliana. PLoS Genet. 2011;7(1):e1001263. [PMC free article] [PubMed]
56. Edgar RS, et al. Peroxiredoxins are conserved markers of circadian rhythms. Nature. 2012;485(7399):459–464. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...