Physiological and Genetic Dissection of Sucrose Inputs to the Arabidopsis thaliana Circadian System

Circadian rhythms allow an organism to synchronize internal physiological responses to the external environment. Perception of external signals such as light and temperature are critical in the entrainment of the oscillator. However, sugar can also act as an entraining signal. In this work, we have confirmed that sucrose accelerates the circadian period, but this observed effect is dependent on the reporter gene used. This observed response was dependent on sucrose being available during free-running conditions. If sucrose was applied during entrainment, the circadian period was only temporally accelerated, if any effect was observed at all. We also found that sucrose acts to stabilize the robustness of the circadian period under red light or blue light, in addition to its previously described role in stabilizing the robustness of rhythms in the dark. Finally, we also found that CCA1 is required for both a short- and long-term response of the circadian oscillator to sucrose, while LHY acts to attenuate the effects of sucrose on circadian period. Together, this work highlights new pathways for how sucrose could be signaling to the oscillator and reveals further functional separation of CCA1 and LHY.


Introduction
Circadian clocks are endogenous timekeepers that coordinate internal physiological responses to the predicted external environment. In Arabidopsis, 30% of the transcriptome is circadian regulated, and this includes processes such as metabolism, the induction of flowering, growth, responsiveness to hormones, and biotic and abiotic stress [1][2][3][4][5][6][7][8]. Consequently, having an internal oscillator that closely matches external time enhances plant fitness [9]. Studies that investigated responsiveness to periodic stress cues or metabolism have highlighted the importance of metabolic oscillations in the regulation of circadian rhythms [3,10]. It has been proposed that endogenous timekeepers not only predict external stress cues but also provide the basis for temporal segregation between incompatible cellular metabolic processes that would otherwise be energetically futile and stressful [7,8,11,12].
Plant circadian rhythms are generated through a series of transcriptional-translational loops. At the center of the oscillator is a repressive feedback loop formed between the morning expressed MYB transcription factors CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) and the night-phased TIMING OF CAB EXPRESSION 1 (TOC1), also known as PSEUDO RESPONSE REGULATOR 1 (PRR1) [13][14][15]. This core loop is subsequently regulated by a series of morning and evening loops. In the morning, sequential expression of PRR9/7/5 starting with On day six, seedlings were transferred to 96-well imaging microtiter plates (Perkin Elmer, Juegesheim, Germany) containing 3% w/v sucrose or no sucrose as indicated and re-entrained for a further day before being transferred to the TOPCOUNT (Perkin-Elmer (Perkin-Elmer-Cetus), Norwalk, CT, USA). TOPCOUNT experiments were carried out under a constant temperature of 21 • C and either constant blue or constant red light as stated in text. Data were analyzed as previously described [20,21,39].

Confirmations That Sucrose Regulates Free-Running Period of the Oscillator
To explore the effects of sucrose on the oscillator, seedlings expressing GI::LUC (Col-0) or CAB2::LUC (Ws-2) were examined under free-running conditions in the presence or absence of 3% w/v sucrose with either monochromatic blue light (BL) or red light (RL) [14]. In both backgrounds, the free-running period (FRP) of GI::LUC decreased when sucrose was applied as previously reported (Figure 1 and Figure S1) [10,25]. Similarly, CAB2::LUC (assayed only under BL) also had an accelerated period, but this was limited to only the first time window of free-running conditions. In subsequent time windows, CAB2::LUC FRP returned to~26 h, highlighting a temporal effect of sucrose on this reporter gene ( Figure 2, Figure S1). Seeds were surface-sterilized and sown onto MS medium containing 3% w/v sucrose or no sucrose as indicated, and then stratified at 4 °C for 3 days. Seedlings were entrained for seven days under 12 h light/12 h dark photoperiods (white light at 100 µmol m −2 s −1 ) and a constant temperature of 22 °C. On day six, seedlings were transferred to 96-well imaging microtiter plates (Perkin Elmer, Juegesheim, Germany) containing 3% w/v sucrose or no sucrose as indicated and re-entrained for a further day before being transferred to the TOPCOUNT (Perkin-Elmer (Perkin-Elmer-Cetus), Norwalk, CT, USA). TOPCOUNT experiments were carried out under a constant temperature of 21 °C and either constant blue or constant red light as stated in text. Data were analyzed as previously described [20,21,39].

Confirmations That Sucrose Regulates Free-Running Period of the Oscillator
To explore the effects of sucrose on the oscillator, seedlings expressing GI::LUC (Col-0) or CAB2::LUC (Ws-2) were examined under free-running conditions in the presence or absence of 3% w/v sucrose with either monochromatic blue light (BL) or red light (RL) [14]. In both backgrounds, the free-running period (FRP) of GI::LUC decreased when sucrose was applied as previously reported (Figure 1 and Figure S1) [10,25]. Similarly, CAB2::LUC (assayed only under BL) also had an accelerated period, but this was limited to only the first time window of free-running conditions. In subsequent time windows, CAB2::LUC FRP returned to ~26 h, highlighting a temporal effect of sucrose on this reporter gene ( Figure 2, Figure S1).  In (E) BL and (F) RL, the relative amplitude error (RAE) and circadian period are paired for all plants that generated an FFT output (RAE < 0.9; see Methods section); in these graphs, the closer a population is to the origin, the more robustly it oscillates. the figures). Under BL, FFT analysis was performed during the time windows 06-96 h (1st-tw) and 31-121 h (2nd-tw). Error bars represent standard error (SE). In (E) BL and (F) RL, the relative amplitude error (RAE) and circadian period are paired for all plants that generated an FFT output (RAE < 0.9; see Methods); in these graphs, the closer a population is to the origin, the more robustly it oscillates.  Figure 1, the RAE and circadian period are paired for all plants that generated an FFT output (RAE < 0.9) under the different sucrose availabilities during entrainment or under free-running conditions. Error bars represent SE. FFT analysis was conducted as described in Figure 1 for the respective light conditions. Alongside accelerating the circadian period, we also observed that the presence of sucrose acted to stabilize the circadian period and the accuracy of these rhythms under free-running conditions. In the absence of sucrose, the FRP of GI::LUC (Col-0) and CAB2::LUC (Ws-2) gradually accelerated over successive free-running time windows, and there was an increase in the relative amplitude error (RAE), which measures the accuracy of circadian oscillations (Figures 1 and 2, Figure S1). This observed effect occurred independently of the prevailing light conditions. By contrast, seedlings treated with sucrose retained a consistently accelerated period, and RAE either stayed stable or decreased (rhythms became more accurate) for both GI::LUC (Col-0) or CAB2::LUC (Ws-2) (Figures 1 and 2, Figure S1). Therefore, sucrose accelerates FRP and stabilizes circadian rhythms under free-running conditions, although there is a reporter gene effect in some instances.

The Oscillator Dynamically Responds to Sucrose under Free-Running Conditions
To determine if the observed effects of sucrose was caused by exposure when entrained or if the oscillator was dynamically responding to sucrose availability, plants were either entrained on media  Figure 1, the RAE and circadian period are paired for all plants that generated an FFT output (RAE < 0.9) under the different sucrose availabilities during entrainment or under free-running conditions. Error bars represent SE. FFT analysis was conducted as described in Figure 1 for the respective light conditions. Alongside accelerating the circadian period, we also observed that the presence of sucrose acted to stabilize the circadian period and the accuracy of these rhythms under free-running conditions. In the absence of sucrose, the FRP of GI::LUC (Col-0) and CAB2::LUC (Ws-2) gradually accelerated over successive free-running time windows, and there was an increase in the relative amplitude error (RAE), which measures the accuracy of circadian oscillations (Figures 1 and 2, Figure S1). This observed effect occurred independently of the prevailing light conditions. By contrast, seedlings treated with sucrose retained a consistently accelerated period, and RAE either stayed stable or decreased (rhythms became more accurate) for both GI::LUC (Col-0) or CAB2::LUC (Ws-2) (Figures 1 and 2, Figure S1). Therefore, sucrose accelerates FRP and stabilizes circadian rhythms under free-running conditions, although there is a reporter gene effect in some instances.

The Oscillator Dynamically Responds to Sucrose under Free-Running Conditions
To determine if the observed effects of sucrose was caused by exposure when entrained or if the oscillator was dynamically responding to sucrose availability, plants were either entrained on media containing 0% w/v sucrose or 3% w/v sucrose and then moved to free-running conditions with media either containing 0% or 3% w/v sucrose. All experiments were carried under BL as BL results in the most stable rhythms under free-running conditions [40]. When sucrose was applied only during free-running conditions, there was a shortening of CAB2::LUC FRP (Figure 2A,B). However, in seedlings only exposed to sucrose during free-running conditions, FRP shortened more than when seedlings were exposed to sucrose during both entrainment and free-running conditions. Furthermore, seedlings only exposed to sucrose during free-running conditions did not display the temporal effect of sucrose on CAB2::LUC FRP (Figure 2A,B). Conversely, seedlings exposed to sucrose only during entrainment had a longer FRP than seedlings exposed to sucrose during free-running conditions.
As with the effect of sucrose on FRP, the ability of sucrose to stabilize circadian oscillations over successive free-running windows was also dependent upon sucrose being present during free-running conditions (Figure 2A,B). Prior exposure during entrainment was not sufficient for sucrose to stabilize circadian oscillations. Similar results were also obtained for GI::LUC (Col-0) for both the effect of sucrose on accelerating FRP and stabilizing the oscillator ( Figure 1C,E, Figure S1). However, we did observe that GI::LUC seedlings exposed to sucrose only during entrainment had a quicker FRP than seedlings never exposed to sucrose. This effect was temporal and restricted to the first-time window of the experiment. In later time windows, the period of the two sets of seedlings became similar ( Figure 1C, Figure S1). This would suggest that sucrose is dynamically regulating both periodicity and stabilizing FRP periodicity but may have distinct short-and long-term effects as highlighted previously [25].

GI Has a Light Dependent Sucrose Phenotype
Sucrose has been recently shown to signal directly to the oscillator by stabilizing GI to maintain robust oscillations during the dark [34]. To determine if GI was also involved in any light-dependent responses of sucrose, the response of gi-11 to sucrose was examined under constant BL or RL. As described previously, gi-11 GI::LUC (assayed under RL) or gi-11 CCA1::LUC (assayed under BL) with or without exposure to sucrose had weaker circadian rhythms when compared to wild type, although sucrose did increase the amplitude and robustness of these rhythms for a limited period of time ( Figure 3A,B, Figure S2) [41]. Wild-type seedlings displayed a consistent response to sucrose regardless of the light conditions or reporter gene used as described earlier ( Figure 3A,B). By contrast, gi-11 had distinct response to sucrose depending on whether BL or RL was used during free-running conditions. For BL, gi-11 seedlings grown in the presence of 3% w/v sucrose containing media had a longer circadian period than gi-11 seedlings grown in the absence of sucrose ( Figure 3C,D, Figure S2). By contrast, we found that gi-11 under RL displayed a wild-type response to sucrose in both the acceleration of free-running period and the stabilization of circadian rhythms over successive time windows ( Figure 3A,B). However, in the absence of sucrose, gi-11 GI::LUC FRP under RL did not accelerate over successive time windows but instead de-accelerated ( Figure 3A,B, Figure S2). As wild type responded consistently to BL and RL regardless of the reporter gene used, it would suggest that GI has a blue-light dependent role in sucrose signaling to the oscillator.

CCA1 and LHY Have Functionally Distinct Roles in Sucrose
Signaling to the Oscillator CCA1/LHY have been previously linked to sucrose signaling indirectly through a PIF metabolic-entrainment pathway, and separately through sugar-dependent regulation of PRR7 [10,32]. To determine if CCA1/LHY could have a direct role in sugar signaling, the response of cca1-11, lhy-21, and cca1-11/lhy-21 mutants (all harboring CAB2::LUC) to sucrose was examined. The FRP of the respective mutants was firstly examined under RL across two consecutive free-running windows with varying concentrations of sucrose. As before, sucrose accelerated the FRP of CAB2::LUC ( Figure 4A,B). This quickening of FRP occurred at even the lowest tested percentage of sucrose (0.5%), and there was an additive effect of increasing the percentage of sucrose on shortening FRP ( Figure 4A,B). However, at higher percentages of sucrose, there was a decline in the magnitude of effect sucrose caused on FRP. Across all percentages of sucrose, the FRP remained stable across the two time windows, while the FRP of non-treated seedlings decreased as observed previously for CAB2::LUC (Figure 2 and Figure 4A,B). As with wild-type plants, sucrose decreased the period of lhy-21, and this occurred in a sucrose percentage dependent manner. By contrast, cca1-11 was insensitive to sucrose application; across all tested concentrations of sucrose, there was no decrease in FRP for the first-time window and there was only a slight decrease at 2% sucrose in the second-time window. In the presence of sucrose, cca1-11/lhy-21 mutants only had detectable oscillations for the 1st time window of free-running conditions before becoming arrhythmic as previously described [15]. During the first-time window, the FRP of cca1-11/lhy-21 responded to the sucrose gradient as was seen in wild type and lhy-21, but surprisingly, at 3% sucrose, the FRP increased ( Figure 4A,B).

CCA1 and LHY Have Functionally Distinct Roles in Sucrose
Signaling to the Oscillator CCA1/LHY have been previously linked to sucrose signaling indirectly through a PIF metabolic-entrainment pathway, and separately through sugar-dependent regulation of PRR7 [10,32]. To determine if CCA1/LHY could have a direct role in sugar signaling, the response of cca1-11, lhy-21, and cca1-11/lhy-21 mutants (all harboring CAB2::LUC) to sucrose was examined. The FRP of the respective mutants was firstly examined under RL across two consecutive free-running windows with varying concentrations of sucrose. As before, sucrose accelerated the FRP of CAB2::LUC ( Figure 4A,B). This quickening of FRP occurred at even the lowest tested percentage of sucrose (0.5%), and there was an additive effect of increasing the percentage of sucrose on shortening FRP ( Figure 4A,B). However, at higher percentages of sucrose, there was a decline in the magnitude of effect sucrose caused on FRP. Across all percentages of sucrose, the FRP remained stable across the two time windows, while the FRP of non-treated seedlings decreased as observed previously for CAB2::LUC (Figures 2  and 4A,B). As with wild-type plants, sucrose decreased the period of lhy-21, and this occurred in a sucrose percentage dependent manner. By contrast, cca1-11 was insensitive to sucrose application; across all tested concentrations of sucrose, there was no decrease in FRP for the first-time window and there was only a slight decrease at 2% sucrose in the second-time window. In the presence of sucrose, cca1-11/lhy-21 mutants only had detectable oscillations for the 1st time window of free-running conditions before becoming arrhythmic as previously described [15]. During the first-time window, the FRP of cca1-11/lhy-21 responded to the sucrose gradient as was seen in wild type and lhy-21, but surprisingly, at 3% sucrose, the FRP increased ( Figure 4A,B). As GI had a distinct response to sucrose depending on whether BL or RL was used, we looked to see if cca1-11 or lhy-21 also responded to sucrose differently depending on the light condition used. Plants were either entrained with or without 3% w/v sucrose and then transferred to free-running conditions (constant BL) with or without 3% w/v sucrose. As for RL, cca1-11 seedlings exposed to BL were insensitive to sucrose application during the first-time window regardless of when plants were first exposed to sucrose ( Figure 4C,D). However, during the second time window, cca1-11 responded similarly to wild-type; seedlings exposed to sucrose during entrainment and under free-running conditions had an increase in CAB2::LUC FRP, while seedlings only exposed to sucrose during free-running conditions had no change in CAB2::LUC FRP when compared to the first time window (Figure 4C,D). As was seen for RL, sucrose supplied under free-running conditions shortened lhy-21 FRP, but lhy-21 was hypersensitive to the period shortening effects of sucrose compared to wild type. lhy-21 seedlings exposed to sucrose during entrainment but not under free-running conditions also had a much shorter period compared to lhy-21 seedlings never exposed to sucrose, as was seen previously for wild-type GI::LUC (Figure 1 and Figure 4A,B). Such a response was not seen in wild-type CAB2::LUC seedlings ( Figure 2B and Figure 4D). By the second time window, this response was no longer observed, as was the case for GI::LUC before (Figure 1 and Figure 4D). CCA1 and LHY therefore seem to have functionally independent roles to sucrose signaling to the oscillator.  A,B) The free-running period of cca1-11, lhy-21, or cca1-11/lhy-21 mutant harboring CAB2::LUC was examined across a sucrose gradient in two consecutive time windows under constant RL. Data were analyzed as described in Figure 1. (C,D) The free-running period of cca1-11 (C) or lhy-21 (D) under constant blue light was examined across two consecutive time windows. Seedlings were entrained with or no 3% w/v sucrose and then moved to media containing no or 3% w/v sucrose during free-running conditions. Data were analyzed as described in Figure 1. Error bars for (A-D) are standard error. Data represent the average of two independent experiments for both respective light conditions. As GI had a distinct response to sucrose depending on whether BL or RL was used, we looked to see if cca1-11 or lhy-21 also responded to sucrose differently depending on the light condition used. Plants were either entrained with or without 3% w/v sucrose and then transferred to free-running conditions (constant BL) with or without 3% w/v sucrose. As for RL, cca1-11 seedlings exposed to BL were insensitive to sucrose application during the first-time window regardless of when plants were first exposed to sucrose ( Figure 4C,D). However, during the second time window, cca1-11 responded similarly to wild-type; seedlings exposed to sucrose during entrainment and under free-running conditions had an increase in CAB2::LUC FRP, while seedlings only exposed to sucrose during free-running conditions had no change in CAB2::LUC FRP when compared to the first time window (Figure 4C,D). As was seen for RL, sucrose supplied under free-running conditions shortened lhy-21 FRP, but lhy-21 was hypersensitive to the period shortening effects of sucrose compared to wild type. lhy-21 seedlings exposed to sucrose during entrainment but not under free-running conditions also had a much shorter period compared to lhy-21 seedlings never exposed to sucrose, as was seen previously for wild-type GI::LUC (Figures 1 and 4A,B). Such a response was not seen in wild-type CAB2::LUC seedlings ( Figures 2B and 4D). By the second time window, this response was no longer observed, as was the case for GI::LUC before (Figures 1 and 4D). CCA1 and LHY therefore seem to have functionally independent roles to sucrose signaling to the oscillator.

Discussion
It has become established that sucrose has a critical role in regulating the period and phase of the Arabidopsis oscillator. We have confirmed that sucrose application shortens the circadian period and also shown that the magnitude of this effect is dependent on the reporter gene analyzed. For GI::LUC, sucrose "permanently" shortened FRP across the length of the experiment, and this occurs regardless of the Arabidopsis background used (Figure 1, Figure S1). However, for CAB2::LUC, when sucrose is supplied during entrainment and under free-running conditions, FRP is only shortened during the first time window. This effect is not seen when sucrose is supplied exclusively during free-running conditions with FRP becoming consistently accelerated across the experiment (Figure 2, Figure S1), highlighting a short-and long-term response of CAB2::LUC to sucrose. The mechanism regulating these distinct processes is still not fully clear [25].
We have also shown that sucrose acts to stabilize the circadian oscillator across consecutive windows of free-running conditions (Figures 1, 2 and 4, Figure S1). Regardless of the reporter gene used, FRP gradually decreased if sucrose was absent under free-running conditions and RAE increased. Exposure to sucrose during entrainment was not necessary for the buffering effect to be observed, suggesting that sucrose is dynamically stabilizing the oscillator rather than fixing the oscillator in a stable state during entrainment. It has been previously shown that sucrose stabilizes circadian rhythms during the dark, and this was dependent on GI [25]. However, our experiments were carried out under BL or RL, and we did not observe a requirement for GI. In fact, GI under RL promotes the gradual acceleration of FRP ( Figure 4B, Figure S2). This suggests that sucrose acts to stabilize the oscillator both in the light and in the dark, and this effect is regulated through two separate pathways.
It has been recently proposed that the Arabidopsis oscillator is not fixed but dynamically responds to both external and internal stimuli [23,42]. Our work here further supports such a fluid nature. When no sucrose was supplied during entrainment but then applied during free-running conditions, FRP shortened to the same extent as plants that have been entrained with sucrose (Figures 1-4, Figures S1  and S2). Conversely, if sucrose was applied during entrainment but then removed under free-running conditions, FRP was not shortened and remained similar to plants never exposed to sucrose. However, in some instances, there seemed a prior memory of exposure of sucrose application (Figures 1 and 4). The memory of sucrose was temporal and by the second time window, there was no or little difference in the FRP between the two treatment groups. This effect was also dependent on the reporter gene, highlighting again short-and long-term responses of circadian genes to sucrose availability. The oscillator also shows high sensitivity to sucrose, with concentrations as low as 0.5% w/v being sufficient to not only accelerate FRP but also to stabilize the circadian period under free-running conditions. Therefore, the effect of sucrose on the oscillator is not fixed but instead remains highly responsive to even low quantities of sucrose as availability fluctuates.
The pathways through which sucrose signals to the oscillator have begun to be uncovered. The sugar-responsive transcription factor bZIP63 directly regulates PRR7, sucrose stabilizes GI in the dark, and AKIN10 and TIC form an input pathway to regulate the phase of GI expression [26,28,34]. CCA1/LHY have been previously linked to sucrose signaling indirectly, either as a target of a PIF4 metabolic signaling pathway, or separately as a target of the SnRK1/bZIP63/PRR7 signaling module [32,34]. Here we found that CCA1 and LHY have direct but potentially independent roles in sucrose signaling to the oscillator. For both proteins, the effects of sucrose are dependent on the light condition. Under RL, cca1-11 mutants are insensitive to sucrose-mediated shortening of FRP at most percentages across multiple time windows (Figure 4A,B). By contrast, under BL, cca1-11 is insensitive to the effects of sucrose during the first time window but does respond to sucrose during the second time window in a similar manner to wild type ( Figure 4C,D). This would therefore suggest that CCA1 forms part of a long-term, red-light dependent sucrose-signaling pathway but for BL is only part of a short-term signaling pathway. By contrast, LHY acts to restrict the effects of sucrose on the oscillator under BL across multiple time windows, indicating that LHY acts in a long-term signaling pathway. Additionally, LHY may also contribute to a short-term pathway, as lhy mutants have a much shorter period when exposed to sucrose only during entrainment compared to seedlings never exposed to sucrose (Figure 4). However, lhy mutants displayed a wild-type response under RL. CCA1 and LHY have conventionally been viewed as a singular component within the Arabidopsis clock, but independence has been highlighted in both temperature compensation and entrainment of the oscillator [38,43]. However, it remains unclear how these two highly similar proteins could be acting separately.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4425/10/5/334/s1, Figure S1: The effect of sucrose on accuracy and periodicity of different luciferase markers in different ecotypes, Figure S2: Sucrose has a light specific effect in the gi-11 background.