PIF transcriptional regulators are required for rhythmic stomatal movements

Stomata govern the gaseous exchange between the leaf and the external atmosphere, and their function is essential for photosynthesis and the global carbon and oxygen cycles. Rhythmic stomata movements in daily dark/light cycles prevent water loss at night and allow CO2 uptake during the day. How the actors involved are transcriptionally regulated and how this might contribute to rhythmicity is largely unknown. Here, we show that morning stomata opening depends on the previous night period. The transcription factors PHYTOCHROMEINTERACTING FACTORS (PIFs) accumulate at the end of the night and directly induce the guard cell-specific K+ channel KAT1. Remarkably, PIFs and KAT1 are required for blue light-induced stomata opening. Together, our data establish a molecular framework for daily rhythmic stomatal movements under well-water conditions, whereby PIFs are required for accumulation of KAT1 at night, which upon activation by blue light in the morning leads to the K+ intake driving stomata opening.

protein phosphatase (PP1) and its regulatory subunit PRSL1 to activate the H + -ATPase, causing activation of the inward-rectifying K + channels, and the influx of K + leading to stomatal opening. The blue-light pathway saturates at low fluence rates and drives stomatal opening at dawn 16,29 .
Importantly, whether light regulates stomata movements at the transcriptional level, and how this might contribute to stomatal dynamics under diurnal conditions through the night-day cycle, is still largely unexplored. Here, we present data showing that under diurnal cycles the transcription factors PIFs are required for blue light-induced stomatal opening in the morning.
We show that the night period preceding dawn, when PIFs accumulate, is necessary for robust stomatal movements. We identify the GC-specific inward-rectifying K + -channel KAT1 as a direct PIF-induced target gene, and show that mutants defective in PIFs or KAT1 are impaired in morning stomata opening under diurnal conditions. Our data establish a novel molecular framework for the rhythmic diurnal movements of stomata under non-restrictive water conditions, whereby high PIF levels at the end of the night allow for PIF-dependent accumulation of KAT1, which, upon activation by blue light phototropin-mediated signaling leads to the GC K + intake that drives morning stomata opening.

Stomata opening under short days oscillates and requires the night period
Stomata movements in wild-type Col-0 cotyledons were measured over 24 hours during the fourth day of seedling growth from germination onwards under SD conditions (8h light + 16 h dark; SI Fig. 1) (Fig. 1a-c). At the end of the night at Zeitgeber time (ZT) ZT0, stomatal pores had a mean area of ~18 µm 2 . Between 0-3 hours after dawn (ZT0-ZT3), stomata opened and reached an aperture of ~37 µm 2 . At the end of the light period (ZT6), stomata started to close, resulting in maximum closure 1 hour after the transition from light to dark at ZT9 (~12 µm2).
During the night (ZT8-ZT24), stomata remained closed with a slight increase in stomata pore area, from to ~12 µm 2 at ZT9 to ~16-20 µm 2 at ZT24 (Fig. 1a). Similar stomatal dynamics were obtained when opening was measured using stomata width (Fig. 1b). Together, these results suggest that the dark-to-light transition in SD induces stomata aperture, whereas the subsequent light period during the day promotes closing of the stomata, which remain closed during the night period. These daily oscillations in stomata movements are in accordance with and extend previous observations in older plants 33 .

PIFs are necessary to induce stomata opening at dawn
The results above prompted us to hypothesize that PIFs, which accumulate during the night in SD, and are necessary under these conditions for dynamic responses such as hypocotyl elongation 7-10 , might be involved in stomata movements. To test this, we measured the stomata pore in pifq mutants over 24 hours in 4-day-old seedlings ( Fig. 1a-c). Small differences in stomata aperture were detected in pifq compared to Col-0 at the end of the night (ZT0) (~13 µm 2 compared to ~17 µm 2 ), but most strikingly pifq was not able to open the stomata three hours after dawn, in clear contrast to Col-0 (~13 µm 2 compared to ~37 µm 2 ) (Fig. 1a,c). Similar results were obtained when measuring stomata width (Fig. 1b). Before the light-to-dark transition (ZT6), Col-0 started to close the stomata, reaching the same aperture as pifq.
Importantly, under LL conditions, where PIF3 and likely other PIFs do not accumulate 8,9 , Col-0 was not able to open the stomata in the morning, similarly to pifq in SD (Fig. 1d). These results led us to conclude that under SD, PIFs are required for morning stomata opening. In addition, Col-0 seedlings grown for 3 days under SD conditions and then transferred to an extended night (DD conditions; SI Fig. 1), did not open their stomata in the subjective morning (SI Fig. 2). Together, our data indicate that under diurnal conditions morning stomata opening requires (1) PIF accumulation during the previous night, and (2) the transition to light in the morning. This indicates that darkness during the night period (to allow for PIF accumulation) and the transition to light at dawn, are both necessary for stomata movements under diurnal conditions.

Endogenous basal ABA prevents early stomata opening during the night under SD
ABA promotion of stomata closure is well established under drought conditions 21 or when applied exogenously 34 , whereas the contribution of endogenous basal ABA to diurnal stomata dynamics has been less studied. To address ABA's role under well-watered diurnal conditions, we measured stomata pore in the ABA-biosynthesis-deficient mutant aba2 under our diurnal setup 35 . At ZT3, aba2 stomata were open to an extent similar to Col-0 (~34 µm 2 ), and were closed at the end of the light period, although not as much as Col-0 (~21 µm 2 compared to ~12 µm 2 ). Remarkably, in sharp contrast to the WT, aba2 stomata then opened progressively during the night to display fully open stomata at ZT24 (~34 µm 2 ) (Fig. 1a-c). These results suggest that the main role of basal ABA under SD conditions is to maintain the stomata closed during the night hours. This finding is consistent with previous data showing that endogenous ABA accumulates to a maximum during the dark period 19,36 , and with results showing that ABAinsensitive plants display more open stomata in the dark 37 . In addition, our results show that endogenous basal ABA is necessary for full stomatal closure during dusk, although other factor(s) are probably involved given that aba2 seedlings at ZT9 display more open stomata than Col-0 (~21 µm 2 compared to ~12 µm 2 ) ( Fig. 1 a-c). Interestingly, in LL conditions, aba2 stomata remained completely open during ZT0-ZT9, in contrast to SD and also to DD, suggesting that the unknown factor(s) acting together with ABA at dusk might not accumulate or be active in LL (SI Fig. 3).

PIF3 localizes in the nucleus of guard cells at the end of the dark phase under SD and is degraded by light at dawn
Our analysis suggests that the previously described accumulation of PIFs at the end of the night under diurnal conditions 7-9 is required for the subsequent stomata aperture at dawn. To address whether PIF-mediated stomata opening activity could be done locally, we next study whether PIF3 accumulates in guard cells. Visualization of 3-day-old SD-grown pPIF3::YFP:PIF3 (YFP-PIF3) at the end of the night (ZT0) by confocal microscopy showed PIF3 in the nucleus of guard cells, localized in the characteristic speckles ( Fig. 2a) 38 . Notably, YFP fluorescence was almost undetectable after 1 hour of white light treatment (ZT1) (Fig. 2b). Together, these results indicate that PIF3 accumulates in guard cells at the end of the dark period and is rapidly degraded by light. This pattern of PIF3 protein accumulation in the dark and its degradation upon light exposure is in accordance with the described presence of phytochrome in guard cells 39,40 , and it is comparable to that described in hypocotyl cell nuclei of seedlings from the same pPIF3::YFP:PIF3 line 41 , or in PIF3 overexpressing seedlings 38 .

PIFs are necessary for blue-light induced stomatal opening
Light-induced stomatal opening is driven mainly by two distinct pathways induced by red and blue light, respectively 16,29 . The red or mesophyll/photosynthetic pathway takes place at high fluence rates and is thought to coordinate stomatal opening with photosynthesis. The blue-light pathway, on the other hand, is phototropin-mediated, considered guard cell-specific and independent of photosynthesis, saturates at low fluence rates and drives stomatal opening at dawn. To further define the role of PIFs in stomatal opening and understand PIF contribution in the red and/or blue pathways, we next performed an experiment following the set up reported by Papanatsiou et al. 42 . Col-0 and pifq SD-grown seedlings were first exposed to 3 h of red light (40 µmols/m 2 s) at dawn to provide a photosynthetic energy input that reduces CO2 concentration 42 . We observed partial transient stomata opening in Col-0 after 1h of red light treatment (ZT0 to ZT1) (from 18 to 25µm 2 ), and no significant differences in stomatal opening between Col-0 and pifq (Fig. 3). At ZT3, after 3h of red light, pore areas in Col-0 and pifq stomata had closed again similar to ZT0. In this background, adding blue light (10 µmols/m 2 s) rapidly (in 2 h) elevated stomata opening in Col-0 from 18 to 33 µm 2 . Remarkable, this response was completely absent in pifq (Fig. 3). After 3h of blue light treatment, Col-0 stomata continued open (32 µm 2 ) indicating that the stomata response was already at a maximum after 2 h, whereas pifq started opening between ZT5 and ZT6. Controls in red light remained closed ( Fig. 3). Together, these results indicate that PIFs are necessary for fast blue light stomatal opening, and strongly suggest that the impaired stomatal opening in pifq in the morning is a consequence of a deficiency in the blue light pathway. Given the described PIF accumulation during the night hours and the rapid phytochrome-mediated degradation upon light exposure 8 , we hypothesize that PIFs might be necessary to regulate the expression during the previous night of a key component in the guard-cell specific phototropin pathway leading to stomatal opening in the morning.

ABA-repressed guard cell-specific gene under SD conditions
To explore the possibility that PIFs regulate expression of a necessary component for stomatal opening in the phototropin-mediated pathway, and given the co-regulation of stomata movements by PIFs and ABA, we next aimed to identify ABA-responsive genes in guard cells that might be PIF targets. We reasoned that these genes could encode proteins involved in stomatal dynamics downstream of the PIFs. Because ABA induces stomatal closure whereas PIFs are necessary for aperture, we hypothesized that relevant PIF-regulated genes in stomatal dynamics could correspond to GC genes inversely regulated by PIFs and ABA. To this end, we compared previously defined gene sets of PIF-regulated genes in SD at the end of the night (538 genes, 331 induced and 207 repressed by PIFs) 43 with ABA-responsive guard-cell specific genes (906 genes, 515 induced and 384 repressed by ABA) (responsive to a treatment of 50 µM of exogenous ABA for 3 hours) 27 . Forty-one common genes were identified, and of these, 9 genes (AT5G46240) encodes the KAT1 voltage-dependent potassium channel predominantly expressed in GC (Fig. 4b). KAT1 mediates the potassium influx that leads to stomata swelling and opening 14,44,45 , through a unique gating mechanism that is activated upon the membrane hyperpolarization triggered by phototropin-mediated blue light signaling 29,46 . The KAT1 gene is induced by PIFs and repressed by ABA (Fig. 4 a,b): KAT1 regulation showed a ~14-fold decrease in expression in GC upon treatment with ABA 27 , whereas KAT1 expression was 2.3fold higher at ZT24 in WT compared to pifq 43 (SI Table 1). In addition, publicly available diurnal data (http://diurnal.mocklerlab.org) showed a peak of KAT1 expression in SD at the end of the night. This peak is absent in LL conditions (Fig. 4c), an expression pattern that is characteristic of PIF-regulated genes under diurnal conditions 47 . To test the possibility that in our conditions PIFs induce KAT1 expression before dawn, we checked KAT1 expression in WT and pifq at ZT0, and found that indeed PIFs are necessary for the elevated expression of KAT1 at the end of the night. After 3 h of light at ZT3, expression levels in the WT decreased to levels similar to pifq at ZT0 and ZT3 (Fig. 4d). Importantly, in LL KAT1 levels at ZT0 were similar to levels in SD at ZT3 (Fig. 4d), whereas in DD levels remained high at ZT3 (SI Fig. 4).
Together, these results indicate that PIFs induce KAT1 expression at the end of the night, whereas the transition to light triggers a rapid decrease in KAT1 transcript levels that correlates with light-induced PIF degradation. Regarding the regulation of KAT1 expression by ABA, although the repression by application of exogenous ABA is well defined, we wanted to test to what extent endogenous ABA in our conditions might contribute to the transcriptional regulation of KAT1. To this end, we analyzed KAT1 expression in aba2 mutant seedlings grown in SD and LL. Under SD conditions, no significant differences in KAT1 expression were observed between Col-0 and aba2 at the end of the night period (ZT0), whereas at ZT3 the KAT1 levels in aba2 were slightly more elevated compared to Col-0, differences that were not detected in LL (Fig. 4d) or DD (SI Fig. 4). These data suggest that endogenous basal ABA does not significantly repress PIF-mediated induction of KAT1 expression at night, and might contribute to the repression of KAT1 expression in SD upon exposure to light. Taken together, we conclude that in SD PIFs are necessary to induce the expression of KAT1, a GC specific gene that encodes the inward-rectifying K + channel driving the K + uptake that leads to stomata opening upon activation by blue light. The induction of KAT1 is dark-dependent and peaks at the end of the night, whereas light exposure leads to a repression in KAT1 expression, likely as a result of PIF degradation with a minor contribution of endogenous ABA.

KAT1 expression is directly regulated by PIF3 at the end of the dark period under SD
We next addressed whether KAT1 might be a direct PIF target. We found several G-box and PBE binding motifs approximately 1 kb upstream of the transcription start site (TSS) in the KAT1 promoter that could act as potential PIF binding sites (Fig. 5a). Indeed, available chromatin immunoprecipitation (ChIP)-seq data 48 showed binding of PIF1, PIF3 and PIF4 (statistically significant for PIF1 and PIF4) to this region in 2-day-old dark-grown seedlings ( Fig. 5a). To test whether PIF3 could bind to the KAT1 promoter in our diurnal conditions, we performed ChIP followed by qPCR (ChIP-qPCR) analysis at the end of the night period (ZT0) using the PIF3-tagged line driven by the endogenous PIF3 promoter (YFP-PIF3) (the same line used previously in Fig. 2). After ChIP, primer pairs P1 and P2, encompassing the G-boxes and PBE in this region, were used for qPCR. Statistically significant YFP-PIF3 binding was observed for P2, whereas a primer combination (P3) that binds to the last exon was used as a negative control (Fig. 5b). We conclude that PIF3, and possibly other PIFs, directly bind to the KAT1 promoter under diurnal conditions.

KAT1 is essential to induce morning stomata opening
Previous reports have described contrasting results as to whether KAT1 deficiency might be enough to impair stomata movements [49][50][51] . To evaluate the relevance of the PIF-KAT1 module to mediate stomata dynamics in our conditions, we obtained two independent mutant lines lacking KAT1 (kat1-1 and kat1-2) 52 . Stomata aperture analyses showed that at the end of the night (ZT0), kat1-1 and kat1-2 displayed slightly more closed stomata compared to Col-0 (13 µm 2 compared to 20 µm 2 ). Strikingly, 3 hours after dawn (ZT3), kat1-1 and kat1-2 stomata remained closed, in contrast to Col-0, which displayed fully opened stomata (33 µm 2 ). At ZT6, Col-0 seedlings had closed stomata, similar to kat1-1 and kat1-2. No differences were observed at ZT9 (Fig. 6a). These results indicate that, under day/night cycles, KAT1 is essential for stomata opening in the morning. Next, we tested a KAT1OX line. Compared to Col-0, KAT1OX had slightly more open stomata in the dark at ZT0, reached similar light-induced stomata opening at ZT3, and displayed slower stomata closing at the end of the day (ZT6). At ZT9, KAT1OX stomata were as closed as the Col-0 (Fig. 6b). Together, these data indicate that under diurnal conditions: (1) alteration in KAT1 levels had a small but significant effect on stomata movements at the end of night; (2) KAT1 is required to induce stomata opening at dawn; and (3) ectopic expression of KAT1 resulted in similar light-induced stomata pore area that remained opened for longer time during the day period.

Discussion
We have shown here that the phytochrome-interacting transcription factors PIFs are required for establishing the rhythmicity in stomatal dynamics under diurnal conditions, through direct induction of the expression of KAT1, a guard-cell specific gene encoding an inward-rectifying potassium channel that is necessary for stomatal opening. Our work unveils a novel regulatory link between light/dark cycles and stomata movements, whereby PIF accumulation during the night results in KAT1 transcript accumulation at dawn, in preparation for stomata opening in the morning (Fig. 7, SI Fig. 5). In the beginning of a new day, when the blue/red light ratio is high, light triggers phototropin-mediated blue-light induced hyperpolarization of the membrane, which drives ion influx through KAT1, leading to water intake, turgor increase and stomata opening 13 . Light also induces phytochrome-mediated PIF degradation, which ensures that KAT1 induction is not maintained during the day, which could result in slower stomata closure as seen with KAT1OX lines (Fig. 6). Moreover, using the ABA-deficient mutant aba2, our results show that under diurnal cycles in non-restrictive water availability, endogenous ABA levels are necessary for full dark-induction of stomata closure, and are required to prevent stomata opening during the night (Fig. 7, SI Fig. 5).
These results, together with previous data, support a model whereby a novel interplay between PIF function and endogenous ABA signaling in the guard cell establishes the dynamic regulation of stomata aperture under diurnal cycles: at dusk and in the dark, endogenous ABA is produced 21 , which induces stomata closure and is necessary to maintain stomata closed during the night. Before dawn, accumulated PIFs induce KAT1 expression (Fig. 7, SI Fig. 5).
ABA likely prevents KAT1 accumulation at this time through post-transcriptional mechanisms such as KAT1 endocytosis and sequestration 21 . In the morning, a reduction in ABA is necessary for light-induced stomata opening 53  (2) Both are promoted by PIFs (Fig. 1) 9 ; (3) In both, timing at dawn likely correlates with the highest PIF levels throughout the day/night cycle, as determined at whole-seedling level 7,8 and here in guard cells during the night-to-day transition (Fig. 2), in accordance to the described pattern of PIF accumulation in the dark and light-induced degradation 38,41 ; And (4), in both, PIF-mediated transcriptional regulation of key components underlies the response, shown here for KAT1 and stomata movement (Fig. 4-6), and previously for gene networks involved in cell elongation and hypocotyl growth (like auxin or cell wall remodeling) 43,47,55,56 . Interestingly, we and others previously demonstrated that under diurnal conditions, PIF activity during the night is gated to pre-dawn by the circadian clock components and transcriptional repressors Timing of CAB expression 1 (TOC1 or pseudo response regulator (PRR) 1), PRR5, 7 and 9, by direct interaction and co-binding with PIFs on the G-and PBE-box DNA motifs found in target genes involved in cell elongation 47,[57][58][59] . Whether the clock contributes to gate PIF activation of KAT1 induction is currently unknown, although ChIP-seq experiments did not find binding of PRR proteins to the KAT1 promoter region bound by PIFs (Fig. 5) 60,61 .
The diurnal setup used here favored blue-light mediated stomata opening in the morning, and this has allowed us to uncover a novel role for PIFs as promoters of stomata opening under these conditions. Interestingly, recent studies have reported a role for PIFs as negative  (Fig. 1)), a possible explanation for these differences is that the growth conditions and/or sample preparation (for example epidermal peel manipulation in the light) could have led the authors to overlook the stomata opening phenotype of KAT1deficient mutants.
Our findings, together with the described regulation by ABA 27 , suggest that KAT1 might be a fundamental hub for the regulation of stomata dynamics. In accordance, previous reports showed that KAT1 expression is enhanced by the ABA-regulated AKS1 transcription factor 50 , by auxins 63 and by brassinosteroids 64 . Although the implication of these regulatory levels in stomata opening in day/night cycles is currently unknown, the emerging picture seems to suggest that KAT1 can integrate information of diverse environmental and endogenous factors including photoperiod, time of day, and hormone levels (like ABA, auxin and BRs), which together would provide a mechanism for fast stomata response to a diversity of light conditions and water contents. Interestingly, although KAT1 expression has been well described to be repressed by ABA 27 , our observation that KAT1 levels are not significantly affected in aba2 seedlings under non-restrictive water availability (Fig. 4d) indicates that endogenous basal ABA is likely not sufficient to interfere with PIF-mediated induction of KAT1. Instead, repression of KAT1 expression by ABA may take place only when ABA levels increase, which is mimicked by the exogenous application of high ABA levels in most studies, including the transcriptomic work used here to identify KAT1 27 . Under stress conditions, such as those of low humidity or drought, ABA accumulation could override the PIF-mediated signal by repressing KAT1 expression to prevent morning stomatal opening and preserve water. How ABA might repress PIF-mediated induction of KAT1 is currently unknown. Intriguingly, a recent report has shown that the ABA receptors PYL8 and PYL9 can physically interact with PIFs and interfere with PIF activity to regulate ABI5 expression 65 . Future work will be necessary to evaluate whether this mechanism is active in the regulation of other genes targeted by PIFs and ABA such as KAT1.
In conclusion, our work proposes a conceptual framework whereby the dynamic accumulation of PIFs and endogenous basal ABA throughout the day/night cycles provides the plant with an exquisite mechanism to adjust stomata movements to the precise time of the day. Correct timing and speed of stomatal movements through the day/night cycle is critical for optimized carbon uptake, photosynthesis, water use efficiency, and control of plant physiology 23,42,66 .
Therefore, understanding how the day/night cycle regulates stomatal movements can provide targets to optimize plant yield with improved water use 42 . Indeed, and to illustrate the potential relevance of our findings, a recent work has revealed the importance of K + inward-rectifying channels in determining plant biomass production, and plant adaptation to fluctuating and stressing natural environments 67 .

Seedling growth and measurements
Arabidopsis thaliana seeds used in this manuscript include the previously described aba2/gin1-3 (aba2) 35  measurements, cotyledons (n=6 seedlings) were wet mounted with water on a microscope slide with a cover glass, and pictures were taken with and optical microscope (AixoPhot DP70) with the oil immersion 63X objective lens. Dark samples were mounted in the dark before imaging the abaxial epidermis. Stomata (n=40-60) pore area or pore width was measured using NIH image software (Image J, National Institutes of Health).

Fluorescence microscopy
Stomata of 3-day-old SD-grown pPIF3::YFP:PIF3 seedlings at ZT0, and same stomata maintained in white light for 1h after dawn (ZT1), were visualized using a confocal laser scanning microscope Olympus FV1000 (Emission window: 500nm -660nm). pif3 was used as a negative control. Nuclei were stained using DAPI.

Statistical analysis
Differences in stomata area or width between two genotypes, or between two different light conditions for the same genotype, were analyzed by a pairwise Mann-Whitney test to assess mean differences in non-parametric. Significantly different pairs (P<0.05) were represented by asterisks. In the ChIP experiment, statistical differences between mean fold change values relative to Col-0 were log2 transformed and analyzed by Student t-test (P<0.05). To identify gene expression differences taking into account global variation across all genotypes and light conditions, and given the parametric nature of the gene expression measurements, data were analyzed using two-way ANOVA. A post-hoc Tukey test was performed to identify significant differences between pairs of genotypes or light conditions, and significantly different pairs were represented by letters. In specific sample pairs, a t-test was performed and asterisks indicate statistically significant differences.

Gene expression analysis
RNA was extracted using Mawxell RSC plant RNA Kit (Promega). 1µg of total RNA extracted were treated with DNase I (Ambion) according to the manufacturer's instructions. First-strand cDNA synthesis was performed using the NZYtech First-strand cDNA Synthesis Kit (NZYtech). 2 µl of 1:25 diluted cDNA with water was used for real-time PCR (LightCycler 480 Roche) using SYBR Premix Ex Taq (Takara) and primers at 300nM concentration. Gene expression was measured in three independent biological replicates, and at least two technical replicates were done for each of the biological replicates. PP2A (AT1G13320) was used for normalization 70 . Primers are listed in SI Table 2.

Chromatin Immunoprecipitation (ChIP) and ChIP Assays
Chromatin immunoprecipitation (ChIP) and ChIP-qPCR assays were performed as previously  (Table S3) spanning the region containing the predicted binding sites for the PIFs and an intergenic region as a negative control 48 . Three independent biological replicates were performed and PIF3 binding was represented as % of input and relative to Col-0 set at unity.

Diurnal profile expression
Transcript abundance were analyzed using the publicly available genome-wide expression data DIURNAL5 (http://diurnal.mocklerlab.org) 71 using a cut-off of 0.2 for the following conditions: SD (Col-0_SD) names as Col-0 SD and free running (LL23_LDHH) named as Col-0 LL.    seedlings in 3-day-old SD-or LL-grown at ZT0 and ZT3. Data are the means ± SE of biological triplicates (n = 3). Letters denote the statistically significant differences using 2-way Anova followed by posthoc Tukey's test (P < 0.05), and asterisks indicate statistically significant differences between specific samples (t-test; *, P < 0.05).   KAT1 expression in Col-0, pifq, and aba2 seedlings in 3-day-old SD-or DD-grown at ZT0 (common for SD and DD) and ZT3. Data are the means ± SE of biological triplicates (n = 3).

Figure Legends
Letters denote the statistically significant differences using 2-way Anova followed by posthoc Tukey's test (P < 0.05).  Fig. 1 for a diagram of light treatments). Seedlings were grown under SD conditions for 2 days, then at ZT8 of the third day they were either kept under SD as a control (SD) or they were transferred to continuous white light (LL). Stomata measurements were performed during the fourth day at ZT0, 3, 6 and 9h. (a, b, d) Statistical differences relative to Col-0 (a, b) for each time point, or (d) for each time and condition, are indicated by an asterisk (Mann-Whitney test. P<0.05).  . Letters denote the statistically significant differences using 2-way Anova followed by posthoc Tukey's test (P < 0.05), and asterisks indicate statistically significant differences between specific samples (t-test; *, P < 0.05).   Endogenous ABA represses activity of KAT1, as well as that of the plasma membrane H + pump (PM H + -ATPase). Anion and K + efflux reduces the GC turgor causing stomata to close. At dawn, blue light activates phototropins, which initiate a signaling cascade to activate the plasma PM H + -ATPase that transports H + across the membrane, causing a hyperpolarization that activates the KAT1 channel and induces an influx of K + and accumulation of K + and counteranions (Cl − and malate) into the GC and its vacuole. Accumulation of these ions leads to water uptake into the vacuole and turgor increase, triggering stomatal opening. In the morning, red light activated phytochromes degrade PIFs and prevent KAT1 overexpression.