TY - JOUR
AU1 - Hiraki,, Hayato
AU2 - Uemura,, Matsuo
AU3 - Kawamura,, Yukio
AB - Abstract Environmental adaptability is essential for plant survival. Though it is well known that a simple cooling or cold shock leads to Ca2+ signals, direct evidence has not been provided that plants use Ca2+ signals as a second messenger in the cold acclimation (CA) process in the field. By developing a technique to analyze Ca2+ signals using confocal cryomicroscopy, we investigated Ca2+ signals under several temperature conditions by combining the start temperature, cooling rate and cooling time duration. In both root and leaf cells, Ca2+ signals rapidly disappeared after cooling stopped, and thereafter under a constant low temperature no Ca2+ signal was observed. Interestingly, under the cooling regime from 2°C to −2°C, non-acclimated plants grown at 23°C hardly showed Ca2+ signals, but cold-acclimated plants at 2°C were able to form Ca2+ signals in root cells. These findings suggest that plants sense temperature decreases with Ca2+ signals while adjusting the temperature sensitivity to their own temperature environment. Furthermore, if the temperature is constant, no Ca2+ signal is induced even during CA. Then, we also focused on the CA under field conditions, rich in temperature fluctuations. In CA under field conditions, the expression patterns of CBF/DREB1 genes were distinctly different from those in artificial CA. Pharmacological studies with Ca2+ channel blockers showed that the Ca2+-induced expression of CBF/DREB1 genes was closely correlated with the amplitude of temperature fluctuation, suggesting that Ca2+ signals regulate CBF/DREB1 gene expression during CA under natural conditions. Introduction Ca2+ is a common signaling element in living organisms. In plant cells, Ca2+ acts as a signaling element along with other ions, reactive oxygen species and phytohormones (Dat et al. 2000, Xiong et al. 2002, Wasternack 2007). Because biotic and abiotic stresses induce transient increases in the Ca2+ concentration in plant cells, Ca2+ signals are considered to be a second messenger during stress responses (McAinsh and Pittman 2009, Martí et al. 2013). Several characteristics of Ca2+ signals were reported using simple temperature changes (Knight et al. 1996, Plieth et al. 1999, Knight and Knight 2000). In this study, to understand the natural phenomenon of ‘cold acclimation (CA)’ in plants, we focused on the hypothesis that Ca2+ acts as a second messenger to induce C-repeat (CRT)-binding factor/dehydration-responsive element-binding protein 1 (CBF/DREB1) transcription factors in low-temperature-sensing systems in thermal regimes found under field conditions. CA is the phenomenon by which plants sense the cold at non-freezing low temperatures in the autumn and enhance freezing tolerance to survive severe winter (see Levitt 1980). In terms of gene expression, the transcriptional regulatory cascade during the CA process is mainly mediated by the CBF/DREB1 transcription factors, which bind to the DRE/CRT cis-element and induce cold-responsive (COR) genes (Fowler and Thomashow 2002, Wang and Hua 2009). CBF/DREB1s are considered to be mediated by Ca2+ signals, for example through the calmodulin-binding transcription activator (Doherty et al. 2009, Kidokoro et al. 2017). In fact, rapid increases in intracellular Ca2+ levels induce the expression of KIN1, which is one of the CBF/DREB1 regulon genes (Knight et al. 1996). Several studies have revealed that cooling or cold shock induces transient increases in Ca2+ levels (Plieth et al. 1999, Knight and Knight 2000, Kiegle et al. 2000, Krebs et al. 2012). The cold- induced Ca2+ signals were observed not only in Arabidopsis, but also in wheat (Nagel-Volkmann et al. 2009). Cooling treatments from room temperature to 0°C induce Ca2+ signals with two peaks (Knight and Knight 2000). In addition, the magnitude of Ca2+ signals is affected by the cooling rate rather than the absolute temperature, and no Ca2+ signal is observed by heating (Plieth et al. 1999, Nagel-Volkmann et al. 2009). Interestingly, several types of artificially induced Ca2+ signals at room temperature activate specific promoter motifs, including cold-responsive elements (Whalley et al. 2011, Whalley and Knight 2013). Thus, plants encode environmental information in Ca2+ signals, which are decoded to start and/or regulate the CA process (Chinnusamy et al. 2010, Miura and Furumoto 2013). However, when considering the field condition, it is not sufficiently clear which cooling factors (e.g. temperature at the start of cooling, cooling rate, cooling duration and their combination) cause and form the Ca2+ signals. In almost all CA studies, the treatment has been conducted by placing plants at a constant low temperature, such as 2°C. However, in the field, the air and soil temperatures are affected not only by the daily and seasonal sun movement, but also by cloud, rain and wind, and therefore the temperature change around plants is continuous and complex (Supplementary Fig. S1). Plants have evolved under the daily or seasonal temperature changes from several combinations of temperature factors, and the cold response Ca2+ signal has co-evolved in the context of these natural conditions. As the first step to understand this mechanism, the relationship between these cooling factors and Ca2+ signals must be quantified. In this study, we developed an experimental system that combined confocal cryomicroscopy and an Arabidopsis thaliana line expressing Yellow Cameleon 3.60 (YC3.60), which is a fluorescence resonance energy transfer (FRET)-based Ca2+ sensor (Nagai et al. 2004). Our cryomicroscope has a cryostage (Supplementary Methods S1) or cryochamber (Supplementary Methods S2) in which the temperature in intact plants can be arbitrarily and accurately controlled. The research with this newly developed system has three advantages over previous studies. First, this system has better resolution, and our mathematical analysis by the latest software provided a more in-depth understanding of the Ca2+ signals than previous studies. Secondly, to quantify Ca2+ signals correctly in any absolute temperature, we corrected the temperature characteristic of YC3.60 (Supplementary Methods S3). This is because the activities of Ca2+ sensor proteins are affected by temperature (Felber et al. 2004), although this temperature effect has not been accounted for in any previous reports. Thirdly, while a number of characteristics (e.g. peak number and response to the continuous temperature change) of Ca2+ were shown (Plieth et al. 1999, Knight and Knight 2000, Nagel-Volkmann et al. 2009), the cooling regimes found under natural conditions have not been investigated. Understanding of the Ca2+ behavior in the field is necessary to consider CA in the field. Here, our results showed that not only the cooling rate but also several cooling factors, plant conditions and their combination determine the final profile of Ca2+ signals. In addition, we provided a new aspect of the CA process in the field from the point of view of Ca2+ signals. Focusing on the Ca2+ signals under natural conditions, we pharmacologically analyzed the cold-responsive gene expression level under field CA conditions. Taken together, our results indicated that the expression of CBF/DREB1 genes may be differentially regulated by cold-responsive Ca2+ signals in the field compared with controlled-environment growth chambers at constant 2°C. Results and Discussion Cold-induced Ca2+ signals of root cells in simple cooling conditions Several studies have reported that cold shock causes transient Ca2+ signals, and that the magnitude of Ca2+ signals is affected by the cooling rate rather than the absolute temperature (Plieth et al. 1999, Nagel-Volkmann et al. 2009). We constructed a confocal cryomicroscope with a cryostage on which the sample temperature can be arbitrarily and accurately controlled if samples are thin, as in the case of roots (Fig. 1a). In this system, we confirmed that faster cooling rates resulted in higher Ca2+ signals (Plieth et al. 1999, Nagel-Volkmann et al. 2009). In addition, the correction formula of temperature for the YC3.60 fluorescence was established to compare the value of the ratio [fluorescence value of cpVenus divided by that of enhanced cyan fluorescent protein (ECFP)] in any absolute temperature (see the Materials and Methods). Here, the combination of several parameters, such as temperatures at the start or end of cooling, the cooling rate and cooling duration, were studied. Fig. 1 View largeDownload slide The effect of different starting and nadir temperatures on Ca2+ signals in root cells. (a) An example of the ratio image of fluorescence of cpVenus divided by that of ECFP in Arabidopsis root expressing Yellow Cameleon 3.60. (b) The temperature settings. (c) Averaged Ca2+ signals when cooled from 20°C (red), 16°C (black) or 12°C (yellow) to 0°C at 2°C min−1 (n = 5 or 6). (d) Name of the Ca2+ signal phase. (e) The peak time (before the last decrease phase) of Ca2+ signal was plotted against the temperature. (f) Boxplots were used to indicate the timing of the Ca2+ signal peaks, or (g) the normalized FRET value which indicates the Ca2+ concentration changes. The boxplots present the first to third quartile values, and the bars represent the minimum and maximum values. Significant differences between each data set were analyzed using the Tukey–Kramer test [P < 0.01 (e), P < 0.05 (f)]. (h) Temperature settings. (i) Averaged Ca2+ signals when cooled from 20°C (black) or 10°C (gray) at a rate of 4°C min− 1 (n = 5). (j) Boxplots were used to indicate the timing of the Ca2+ signal peaks (k) as well as the normalized FRET. The boxplots present the first to third quartile values, and the bars represent the minimum and maximum values. Ratio values were normalized by the value at t = 0. The vertical broken lines in (a–c) or in (h) and (i) indicate the time point where the cooling rate is 0°C min− 1. Significant differences between data sets were analyzed using the t-test (**P < 0.01). Fig. 1 View largeDownload slide The effect of different starting and nadir temperatures on Ca2+ signals in root cells. (a) An example of the ratio image of fluorescence of cpVenus divided by that of ECFP in Arabidopsis root expressing Yellow Cameleon 3.60. (b) The temperature settings. (c) Averaged Ca2+ signals when cooled from 20°C (red), 16°C (black) or 12°C (yellow) to 0°C at 2°C min−1 (n = 5 or 6). (d) Name of the Ca2+ signal phase. (e) The peak time (before the last decrease phase) of Ca2+ signal was plotted against the temperature. (f) Boxplots were used to indicate the timing of the Ca2+ signal peaks, or (g) the normalized FRET value which indicates the Ca2+ concentration changes. The boxplots present the first to third quartile values, and the bars represent the minimum and maximum values. Significant differences between each data set were analyzed using the Tukey–Kramer test [P < 0.01 (e), P < 0.05 (f)]. (h) Temperature settings. (i) Averaged Ca2+ signals when cooled from 20°C (black) or 10°C (gray) at a rate of 4°C min− 1 (n = 5). (j) Boxplots were used to indicate the timing of the Ca2+ signal peaks (k) as well as the normalized FRET. The boxplots present the first to third quartile values, and the bars represent the minimum and maximum values. Ratio values were normalized by the value at t = 0. The vertical broken lines in (a–c) or in (h) and (i) indicate the time point where the cooling rate is 0°C min− 1. Significant differences between data sets were analyzed using the t-test (**P < 0.01). First of all, to understand the effect of absolute temperature on Ca2+ signals, we focused on the temperature at the start of cooling . At 30 s after starting to capture the image, roots of non-cold-acclimated (NA) plants were cooled from 20, 16 or 12 to 0°C at a cooling rate of 2°C min−1, and then were maintained at 0°C (Fig. 1b). From fluorescence images of ECFP (FRET donor) and cpVenus (FRET acceptor), the fluorescence ratio image, which is proportional to the concentration of free cytosolic Ca2+, was obtained from the root cells (Fig. 1a). The fluorescence ratio value at each time point (i.e. at the specified temperature) was temperature corrected, averaged and then was plotted (see the Materials and Methods). Ratio values were normalized by the value at t = 0. The normalized ratio was used as the y-axis to show small Ca2+ concentration changes. The temperature change from 20 to 0°C produced two Ca2+ signal peaks, and the second peak appeared at about the 630 s point, the time corresponding to the end of cooling. When plants were cooled from 16 to 0°C, the Ca2+ signal profile was a flat peak, and the peak decreased quickly after the cooling treatment was stopped at about 500 s. This flat peak was formed by the overlap of two major peaks shown by analysis with the software IGOR Pro (Supplementary Fig. S2b). The temperature change from 12 to 0°C resulted in one sharp peak at about 400 s, and this peak also decreased just after stopping the cooling treatment (Fig. 1c). In addition, the Ca2+ signals had one or two peaks (Fig. 1c), and those signals consisted of four or five peaks (Supplementary Fig. S2). While it has been reported that the apoplast and vacuole were the candidate Ca2+ pools for cold shock (Knight et al. 1996, Knight and Knight 2000), our results suggest that the Ca2+ signal is formed by release of Ca2+ from several Ca2+ pools and/or through several Ca2+ channels in the same Ca2+ pool. To characterize the shape of the Ca2+ signals quantitatively, the following four technical terms were defined: (i) the latent phase, i.e. the period from the start of cooling until the Ca2+ increase; (ii) the increase phase, i.e. the period during the increase in the Ca2+ signal of the first peak; (iii) the decrease phase, i.e. the period during the decrease in the Ca2+ signal of the last peak; and (iv) the stable phase, i.e. the period after the Ca2+ signal (Fig. 1d). In all conditions, the point when Ca2+ signals were in the decrease phase corresponded to the period just after stopping cooling or the time when the temperature reached approximately 0°C (Fig. 1e). There were two possibilities for the cue for the decrease phase: (i) stopping cooling or (ii) absolute temperature (i.e. 0°C). To test these possibilities, plants were cooled (i) from 20 to 0°C or (ii) from 10 to −10°C (Fig. 1h). In our experimental conditions, the buffer solution remained super-cooled even at −10°C. In both cases, two peaks of Ca2+ signal were observed regardless of differences in the initial and nadir temperatures of cooling (Fig. 1i), suggesting that ‘stopping the cooling’ decreases the Ca2+ signal. Besides, it should not be overlooked that the basic Ca2+ signal patterns such as the number of peaks did not change when the temperature decrease range was the same (Fig. 1c, i). The temperature change from 12 to 0°C induced a Ca2+ signal with only one peak (Fig. 1c). This means that if the cooling was stopped during the first increase of the Ca2+ signal, the first peak decreased, and no further Ca2+ increase was induced. Even in the case of cooling conditions from 10 to −10°C, the decrease phase was caused by stopping the cooling. This phenomenon seems to have the function of creating multiple Ca2+ signal peaks in the field. For example, the temperature fluctuations which are produced by the movement of the sun, clouds and wind may cause the sharp temperature decrease/increase of the leaves. The temperature decrease induces the increase phase of the Ca2+ signal, and the temperature increase (or stopping the cooling) causes the decrease phase of Ca2+ signals. Therefore, the temperature fluctuations may result in multiple Ca2+ signal peaks in the field. Although Plieth et al. (1999) mentioned that lower absolute temperature induced greater increases of Ca2+ signal, we showed that there were no significant differences among signal peak heights under different absolute temperatures (Fig. 1g, k). Plieth et al. (1999) used step-wise cooling, which may have caused the larger Ca2+ increase depending on the absolute temperature. Furthermore, the luminescence of aequorin was not corrected by temperature in their studies. On the other hand, we carefully established a correction formula to compare the Ca2+ concentration level at different absolute temperatures. If temperature correction is not performed, the Ca2+ concentration seems to decrease just by lowering the temperature even if the Ca2+ concentration does not actually decrease in our experimental system (see Supplementary Methods S3). In some cases, it becomes difficult to distinguish signal and noise without temperature correction. Thus, we would like to emphasize here that we introduced the temperature correction of Ca2+ signals (see Supplementary Methods S3). Furthermore, in the observation of root cells, we concluded that the absolute temperature affected the period from the start of cooling to the generation of the Ca2+ peak but did not affect the peak height (see Fig. 1). These conclusions are quite important, given that Plieth et al. (1999) concluded that their results indicate that temperature sensing is mainly dependent on the cooling rate, dT/dt, whereas the absolute temperature T is of less importance. Without a temperature correction, it would be difficult to draw a valid conclusion. We also found that there was an effect of absolute temperature on the latent phase of Ca2+ signal. A lower start temperature resulted in a longer latent phase (Fig. 1c, i). Similarly, the first peak time was delayed by lowering the start temperature. For example, the first peak time when cooling from 12°C caused at least a 100 s delay when compared with cooling from 20 °C (Fig. 1c, f). Thus, we concluded that the absolute temperature affects the time of the latent phase and peak time. The cooling from 12 to 0°C induced a Ca2+ signal which had only one peak, but it was unclear whether the peak was the first or the second peak (Fig. 1c). To clarify this, we determined the Ca2+ signal with LaCl3, which is known to affect the appearance of the first peak but not that of the second peak due to the inhibition of Ca2+ influx from the extracellular space (Knight and Knight 2000). When plants were cooled from 20 to 0°C (Fig. 2a), LaCl3 mainly inhibited the appearance of the first peak in the roots but did not markedly inhibit the appearance of the second peak (Fig. 2b). When cooled from 12 to 0°C when only one Ca2+ peak arose (Fig. 2c), LaCl3 almost eliminated the cold-induced Ca2+ peaks (Fig. 2d). On the slope of the increase phase of Ca2+ signals and influx rate of Ca2+, LaCl3 significantly changed only the first peak in both cooling conditions from 20 and 12°C (Fig. 2e). Thus, these results indicate that the Ca2+ source of the first peak is almost entirely the extracellular space regardless of absolute temperature. Fig. 2 View largeDownload slide The source of Ca2+ signal under low temperature in root cells. Averaged Ca2+ signals when cooled from (a and b) 20 to 0°C or (c and d) 12 to 0°C at a rate of 2°C min−1 with/without LaCl3 (n = 3–5). The black line indicates the Ca2+ signals of untreated plants (controls) and the gray line indicates those of plants treated with 100 µM LaCl3 for 1 h prior to being analyzed. The FRET values were calculated and normalized against the initial values. (e) The slopes between 100 and 200 s, and between 500 and 600 s, were calculated for each first and second Ca2+ signal peak induced by a temperature change from 20 to 0°C. The slope between 300 and 400 s for the temperature change from 12 to 0°C was also calculated. Ratio values were normalized by the value at t = 0. Significant differences between each data set were analyzed using the t-test (*P < 0.05) (n = 3–5). The bars indicate the SE. Fig. 2 View largeDownload slide The source of Ca2+ signal under low temperature in root cells. Averaged Ca2+ signals when cooled from (a and b) 20 to 0°C or (c and d) 12 to 0°C at a rate of 2°C min−1 with/without LaCl3 (n = 3–5). The black line indicates the Ca2+ signals of untreated plants (controls) and the gray line indicates those of plants treated with 100 µM LaCl3 for 1 h prior to being analyzed. The FRET values were calculated and normalized against the initial values. (e) The slopes between 100 and 200 s, and between 500 and 600 s, were calculated for each first and second Ca2+ signal peak induced by a temperature change from 20 to 0°C. The slope between 300 and 400 s for the temperature change from 12 to 0°C was also calculated. Ratio values were normalized by the value at t = 0. Significant differences between each data set were analyzed using the t-test (*P < 0.05) (n = 3–5). The bars indicate the SE. Effect of temperature fluctuation on Ca2+ signals To estimate the shape of Ca2+ signals which are caused by natural temperature changes in the field, we analyzed the influence of the small temperature fluctuations which generally occur in the field (Supplementary Fig. S1). Temperature change patterns with two different starting temperatures were tested: one was cooling from 20 to 19, 18, 17 or 16°C and then immediately increasing the temperature to 20°C (Fig. 3a), and the other was cooling from 2 to 1, 0, −1 or −2°C and then immediately increasing the temperature to 2°C (Fig. 3b). In all the patterns, cooling and warming rates were 2°C min−1. In NA plants, the cooling from 20 to 19, 18 and 17°C induced Ca2+ signals with a small but significant peak, while only cooling from 20 to 16°C induced a sharper and larger Ca2+ signal peak (Fig. 3c, e). In contrast, in the case of a starting temperature of 2°C, a distinct Ca2+ peak arose only when cooling from 2 to −1 or −2°C, and the peaks were smaller than when cooled from 20 to 17 or 16°C (Fig. 3c–f). Regardless of cooling from 20 or 2°C, warming did not cause a Ca2+ increase, but decreased the Ca2+ concentration. Fig. 3 View largeDownload slide Ca2+ signals of non-acclimated or cold-acclimated YC3.60-expressing Arabidopsis roots. (a) Temperature conditions were from 20 to 19°C and then to 20°C (brown), from 20 to 18°C and then to 20°C (orange), from 20 to 17°C and then to 20°C (green), from 20 to 16°C and then to 20°C (red), and 20°C constant as a control (blue). (b) Other settings were from 2 to 1°C and then to 2°C (black), from 2 to 0°C and then to 2°C (orange), from 2 to −1°C and then to 2°C (green), and from 2 to −2°C and then to 2°C (purple), and 2°C constant as a control (sky blue). Samples in all the experiments were cooled/heated at a rate of 2°C min−1. (c and d) The NA plants were analyzed by using the cooling series of (a) and (b). (e and f) Statistical processing was performed based on the area formed between the baseline (i.e. Ca2+ signals of 20 or 2°C constant) and the Ca2+ signal induced by the cooling treatments from 50 to 200 s (n = 255∼380). (g and h) CA plants were analyzed by using the cooling series of (a) and (b). (i and j) Statistical processing was performed in the same way as in (e) and (f). The FRET values were calculated and normalized against the initial values (n = 4–5). Ratio values were normalized by the value at t = 0. Significant differences between each data set were analyzed by using the Tukey–Kramer test (P < 0.05). Fig. 3 View largeDownload slide Ca2+ signals of non-acclimated or cold-acclimated YC3.60-expressing Arabidopsis roots. (a) Temperature conditions were from 20 to 19°C and then to 20°C (brown), from 20 to 18°C and then to 20°C (orange), from 20 to 17°C and then to 20°C (green), from 20 to 16°C and then to 20°C (red), and 20°C constant as a control (blue). (b) Other settings were from 2 to 1°C and then to 2°C (black), from 2 to 0°C and then to 2°C (orange), from 2 to −1°C and then to 2°C (green), and from 2 to −2°C and then to 2°C (purple), and 2°C constant as a control (sky blue). Samples in all the experiments were cooled/heated at a rate of 2°C min−1. (c and d) The NA plants were analyzed by using the cooling series of (a) and (b). (e and f) Statistical processing was performed based on the area formed between the baseline (i.e. Ca2+ signals of 20 or 2°C constant) and the Ca2+ signal induced by the cooling treatments from 50 to 200 s (n = 255∼380). (g and h) CA plants were analyzed by using the cooling series of (a) and (b). (i and j) Statistical processing was performed in the same way as in (e) and (f). The FRET values were calculated and normalized against the initial values (n = 4–5). Ratio values were normalized by the value at t = 0. Significant differences between each data set were analyzed by using the Tukey–Kramer test (P < 0.05). The reaction of the NA plant to the cold is the first incidence of CA. NA plants showed a Ca2+ increase when cooled from 20 to 19°C (Fig. 3c). Cooling from 20 to 18 or 17°C induced very similar Ca2+ signals, but these Ca2+ signals were larger than that induced by cooling from 20 to 19°C (Fig. 3e). The Ca2+ signal induced by cooling from 20 to 16°C (4°C cooling), the lowest temperature of this experiment, was quantitatively and qualitatively different from the others. The Ca2+ concentration was about double that of the 2 or 3°C cooling. Furthermore, the speed of the Ca2+ signal increase/decrease was faster compared with the other cooling conditions. In the field, the temperature rarely falls >4°C in a few minutes (e.g. see Supplementary Fig. S1) and, more often, the degree of a continuous temperature change is <2°C (e.g. see Fig. 5b, c). Therefore, a sharp Ca2+ signal as observed may be a type of emergency alert for the plant cell. Fig. 5 View largeDownload slide The expression levels of CBF/DREB1 genes during the CA treatment at 2°C or in the field. (a) The expression levels of CBF1/DREB1B, CBF2/DREB1C and CBF3/DREB1A were measured by real-time PCR at 0, 4 (except during November), 8, 22, 32, 46 and 56 h after starting the CA treatments (n = 3 or 4). Plants were cold acclimated in a constant 2°C chamber, or outside in November 15–17 and December 6–8, 2016, and March 21–23, 2017, in Morioka, Japan. The bar indicates the SE. (b) The air temperature was measured by the data loggers. The white and colored bars indicate day and night time, respectively, during mid-November (orange), early December (gray) and late March (purple). (c) The rate of temperature change at 2 °C min−1 was calculated. The bars under the graph indicate day (white) or night (black) time. Fig. 5 View largeDownload slide The expression levels of CBF/DREB1 genes during the CA treatment at 2°C or in the field. (a) The expression levels of CBF1/DREB1B, CBF2/DREB1C and CBF3/DREB1A were measured by real-time PCR at 0, 4 (except during November), 8, 22, 32, 46 and 56 h after starting the CA treatments (n = 3 or 4). Plants were cold acclimated in a constant 2°C chamber, or outside in November 15–17 and December 6–8, 2016, and March 21–23, 2017, in Morioka, Japan. The bar indicates the SE. (b) The air temperature was measured by the data loggers. The white and colored bars indicate day and night time, respectively, during mid-November (orange), early December (gray) and late March (purple). (c) The rate of temperature change at 2 °C min−1 was calculated. The bars under the graph indicate day (white) or night (black) time. Plants cold-acclimated for 7 d were also analyzed using the same patterns of temperature changes. In contrast to the results of NA plants, even when cooled from 20 to 16°C, the distinct Ca2+ signal was not observed (Fig. 3g, i). In addition, a large Ca2+ signal peak arose when cooled from 2 to −2°C, and the peak was the highest among the temperature treatments for CA plants (Fig. 3h, j). Thus, while cooling with a temperature difference of about 4°C is necessary to generate a distinct Ca2+ signal, its signal generation is dependent on the combination of absolute temperature and the plant CA condition. It had been reported that CA treatment enhanced the second peak of Ca2+ signals (Knight and Knight 2000). Here, we found a novel characteristic of the Ca2+ signal, i.e. CA changed the response of the first peak to cooling. It was confirmed by using LaCl3 that this first peak was identical to the first peak of NA and CA plants (Supplementary Fig. S3). When NA plants were cooled from 2°C, plants could not respond to 1 or 2°C cooling (Fig. 3d). In contrast, CA plants did not show an increase in Ca2+ concentration when cooled from 20°C, but did when cooled from 2°C (Fig. 3g, h). For NA plants, cooling from 2°C would be an unexpected temperature change and, thus, a Ca2+ signal is not induced unless the temperature change is large (e.g. greater than a few degrees). CA plants, in contrast, did not show an induction of Ca2+ signal even in the face of a large temperature drop from 20°C. These results may indicate that the Ca2+ signal is induced under conditions to which plants have been exposed for a certain period (i.e. at 20°C for NA plants and 2°C for CA plants). CA is known to induce the reconstruction of the plasma membrane (Uemura et al. 1995) and an increase in plasma membrane fluidity (Yoshida 1983). Ca2+ channels are regulated by membrane fluidity, which depends on the temperature (Örvar et al. 2000, Sangwan et al. 2001). Taken together with these studies, our results support the hypothesis that plants adjust their plasma membrane fluidity depending on the ambient temperature, and make the more distinct Ca2+ signal responses to regulate downstream activities during/after CA. Comparison of Ca2+ signals between leaf and root cells The environmental condition around leaves is quite different from that around roots. In fact, the temperature changes of the leaves were quite different from those of the soil (Supplementary Fig. S1). In addition, the temperature of the roots depends on the temperature of the soil that has a large heat capacity, and the temperature of the leaves is influenced by such factors as air temperature, wind and sunlight. Thus, we assume that there is a difference between leaves and roots in their low temperature sensitivities. To test this assumption, we constructed a cryochamber for cryomicroscopy, in which an intact plant could be cooled in air (Supplementary Methods S2), although accurate temperature control with this system is somewhat more difficult that with the cryostage (Fig. 4a, b). The air temperature surrounding the leaves was cooled from 20°C to approximately 2.5°C at cooling rates of up to about 1.6°C min−1 (Fig. 4a, b). In this case, leaf cells induced the Ca2+ signal with three peaks, while the Ca2+ signal of root cells induced by the temperature change of 20 to 0°C had only two peaks (Fig. 4c). One of the differences between leaf cells and root cells is the presence or absence of chloroplasts. The calcium-sensing protein (CAS) of thylakoid membrane (Peltier et al. 2004, Nomura et al. 2008, Vainonen et al. 2008) induces an extracellular Ca2+-induced cytosolic Ca2+ increase (Han et al. 2003, Nomura et al. 2008, Rocha and Vothknecht 2012). In addition, it has been reported that the chloroplast has ion channels such as ACA1, HMA1, MSL2/3 and GLR3.4 including the putative Ca2+ channels (Huang et al. 1993, Seigneurin-Berny et al. 2005, Haswell and Meyerowitz 2006, Teardo et al. 2011, Hochmal et al. 2015). Therefore, there is a possibility that the third Ca2+ peak observed only in leaf cells is formed by the Ca2+ influx from the chloroplast through the several Ca2+ channels. Fig. 4 View largeDownload slide Ca2+ signal characteristics in leaf cells. (a) Comparison of the Ca2+ signal in leaf and root cells. Both leaf and root cells were cooled from 20°C to around 0°C at about 2°C min−1, or kept at 20°C for leaf cells. (b) The cooling rate was calculated from the temperature setting (root) and temperature data (leaf). The temperature was measured every 5 s. The black arrowhead indicates the time point where the cooling rate is the maximum and the white arrowhead indicates the time point where the cooling rate is 0°C min− 1. (c) Ca2+ signals induced in the root cells (broken black line) and the leaf cells (solid black line) by cooling. Ca2+ signals at a constant 20°C are also shown as the baseline (gray line) (n = 4–5). (d) Leaves were cooled from 20°C to about 2.5°C or from 12°C to about 1.5°C at 1.6°C min−1. (e) The cooling rate was calculated from the temperature data. The temperature was measured every 5 s. The gray and black arrowheads indicate the time point where the cooling rate is the maximum. (f) Ca2+ signals were observed in each condition every 5 s (n = 4–5). (g) The correlation between the time point where the cooling rate is the maximum and the starting time of the increase phase in Ca2+ signals. Each plot was calculated from the results of (e) and (f) (12 to 1.5°C at 1.6°C min−1), (h) and (i) (the first peak) and Supplementary Fig. S4. (h) Temperatures were changed from 20 to 16 to 20°C at 0.47°C min−1 for three cycles. (i) The Ca2+ signals were observed every 5 s (n = 5). The vertical broken lines in (h) and (i) indicate the time point where the cooling rate is 0°C min− 1. (j) Temperatures were changed from 23 to 2.6°C at 1°C min−1, which mimicked CA treatment at a constant low temperature. (k) Fluorescence images were captured at intervals of 1 min for 72 h (n = 3). All ratio values were normalized by the value at t = 0. Each plot of normalized FRET values represents the average of independent experiments. Fig. 4 View largeDownload slide Ca2+ signal characteristics in leaf cells. (a) Comparison of the Ca2+ signal in leaf and root cells. Both leaf and root cells were cooled from 20°C to around 0°C at about 2°C min−1, or kept at 20°C for leaf cells. (b) The cooling rate was calculated from the temperature setting (root) and temperature data (leaf). The temperature was measured every 5 s. The black arrowhead indicates the time point where the cooling rate is the maximum and the white arrowhead indicates the time point where the cooling rate is 0°C min− 1. (c) Ca2+ signals induced in the root cells (broken black line) and the leaf cells (solid black line) by cooling. Ca2+ signals at a constant 20°C are also shown as the baseline (gray line) (n = 4–5). (d) Leaves were cooled from 20°C to about 2.5°C or from 12°C to about 1.5°C at 1.6°C min−1. (e) The cooling rate was calculated from the temperature data. The temperature was measured every 5 s. The gray and black arrowheads indicate the time point where the cooling rate is the maximum. (f) Ca2+ signals were observed in each condition every 5 s (n = 4–5). (g) The correlation between the time point where the cooling rate is the maximum and the starting time of the increase phase in Ca2+ signals. Each plot was calculated from the results of (e) and (f) (12 to 1.5°C at 1.6°C min−1), (h) and (i) (the first peak) and Supplementary Fig. S4. (h) Temperatures were changed from 20 to 16 to 20°C at 0.47°C min−1 for three cycles. (i) The Ca2+ signals were observed every 5 s (n = 5). The vertical broken lines in (h) and (i) indicate the time point where the cooling rate is 0°C min− 1. (j) Temperatures were changed from 23 to 2.6°C at 1°C min−1, which mimicked CA treatment at a constant low temperature. (k) Fluorescence images were captured at intervals of 1 min for 72 h (n = 3). All ratio values were normalized by the value at t = 0. Each plot of normalized FRET values represents the average of independent experiments. Ca2+ signals in leaf cells also decreased after stopping the cooling. When the cooling rate reached 0°C min−1 (Fig. 4b, c: white arrow), the decrease phase of the Ca2+ signal was induced. In order to investigate whether this phenomenon occurs under other cooling conditions, the Ca2+ signals were observed by using three cooling rates (Supplementary Fig. S4a). Under all the cooling conditions, when the cooling was stopped during the third increase of Ca2+ signal, the third peak decreased immediately (Supplementary Fig. S4b, c: outlined arrowheads). While there were some differences in the Ca2+ signals between leaf and root cells, continuous cooling was required to sustain the Ca2+ signals in both types of cells. We found that in terms of responsiveness to cooling, the leaf cells were different from the root cells. In leaf cells, there was the longer latent phase of Ca2+ signal in leaf cells (Fig. 4c). However, root cells showed a much shorter latent phase (Fig. 4c). It is reported that the sensitivity to the cold is higher in root cells than in aerial parts in tobacco seedlings (Campbell et al. 1996). Thus, these factors may cause the delay in the decrease phase and the longer latent phase in leaf cells. Furthermore, we found several Ca2+ characteristics related to cold response in leaf cells. We focused on the effect of the temperature at the start of cooling on Ca2+ signals as we tested in the root cells (Fig. 1b, c). The three Ca2+ signal peaks were induced in leaf cells when the leaves were cooled from 12 to 1.5°C (Fig. 4d–f), while only one Ca2+ signal peak was induced in root cells when cooled from 12 to 0°C, (Fig. 1c). The reason why three peaks were observed on the leaves is that our cryochamber cannot stop the cooling immediately like the cryostage (Fig. 4e). Additionally, the latent phase of the Ca2+ signal was shorter in leaf cells when cooled from 12 to 1.5°C than when cooled from 20 to 2.5°C (Fig. 4d–f) although lower temperature induced a longer latent phase in root cells (Fig. 1c, f). In leaf cells, the increase in Ca2+ concentration appeared to begin immediately after the cooling rate started to decrease from the maximum value (Fig. 4e, f). The time point where the cooling rate started to decrease from the maximum value was determined experimentally at about 520 s (Fig. 4e: gray arrowhead) in the cooling from 12 to 1.5°C, and at about 740 s in the cooling from 20 to 2.5°C (Fig. 4e: black arrowhead), which is consistent with the time at which the Ca2+ signal began to increase in both cases. In order to investigate whether this phenomenon is common in leaves, the cooling from 20 to 2.5°C with three kinds of cooling rate was tested (see Supplementary Fig. S4), and a strong correlation was found between the time point where the cooling rate started to decrease from the maximum value and the start time of the increase phase in Ca2+ signals (Fig. 4g). Therefore, leaf cells may need a change in the cooling rate from fast to slow in order to induce the increase phase from the latent phase. On the other hand, at a slower cooling rate, after the time point where the cooling rate started to decrease from its maximum value, it tended to take more time until the time point where the Ca2+ signal began to increase (e.g. comparison among three cooling rates in Supplementary Fig. S4). It is difficult now for us to judge whether this is an essential phenomenon or not, because, at very slow cooling rates, there is a limit to the temperature control and measurement in our cryochamber system. In contrast, in root cells, any cooling rate induced the increase phase in Ca2+ signal, but the length of the latent phase was dependent on the absolute temperature (Fig. 1c, f). The difference in factors necessary for the initiation of the Ca2+ increase may influence the length of the latent phase in Ca2+ signal to the same cooling in root and leaf cells (Fig. 4a–c). This different initiation of the Ca2+ increase in root and leaf cells might be caused by the higher sensitivity to the same cooling of the Ca2+ increase to cold in root cells than in aerial parts (Campbell et al. 1996). It was noteworthy that Ca2+ signal peaks could be detected during cooling with a cooling rate of 0.47°C min−1 in leaf cells. In the root cells, the small difference (i.e. 1 to 4°C) and slower rate (i.e. 2°C min–1) of cooling induced Ca2+ signals (Fig. 3c, e). For accurate and stable control of the temperature in the cryochamber, the cooling and heating temperature cycle from 20 to 16 to 20°C with a 0.47°C min−1 cooling rate was used and repeated three times (Fig. 4h). This setting was based on the temperature changes in the field in order to understand the behavior of Ca2+ signals in leaf cells in the field. While a previous study reported that the Ca2+ peak declined with each rapid cooling cycle (about 2 to 3.5°C with about 10°C min–1) (Plieth et al. 1999), our slow cooling cycle (i.e. from 20 to 16°C with a 0.47°C min−1 cooling rate) induced a Ca2+ increase with constant peak heights (Fig. 4i). It should be emphasized that we succeeded in the detection of the Ca2+ signal induced by the slow cooling (at 0.47°C min−1). Plieth et al. (1999) also reported that the only prolonged Ca2+ signal without a peak was observed with a cooling rate of 0.03°C s−1 (i.e. 3°C min−1), while a 3°C min−1 cooling rate is too fast under natural conditions (e.g. Fig. 5c). On the other hand, we succeeded in observing the peaks in the slow cooling cycle. Thus, our result clearly showed that Ca2+ signals could be used in response to small and slow temperature changes in the field. Expression of CBF/DREB1 genes in the field and Ca2+ signals Many researchers use the constant low temperature condition (e.g. 2 or 4°C) for CA treatment (Uemura et al. 1995, Knight and Knight 2000, Doherty et al. 2009). Since cooling was required for the generation of Ca2+ signals, conventional CA treatments, such as transfer of plants from 23°C to constant 2°C, caused the Ca2+ signals to occur only once (Fig. 4j, k). In contrast, when plants are cold-acclimated in the field, the frequent rise and fall of temperature may induce the Ca2+ signals several times (Fig. 4h, i). Therefore, we determined the role of Ca2+ signals in CA-associated gene expression of CA plants in the field during the winter with experimental control plants that were subjected to CA at constant 2°C. Experiments in the field were conducted during mid-November, early December and late March at Morioka, Japan, for 56 h using 2-week-old Arabidopsis (Supplementary Methods S4). Since KIN1 was down-regulated by the Ca2+ channel chelator (Knight et al. 1996), we focused on the expression of CBF/DREB1 genes which regulate KIN and COR genes. The conventional CA treatment at constant 2°C induced the single and earliest expression peak of CBF/DREB1 genes at 4 h (first day at 15:00 h) (Fig. 5a: black line). In contrast, the field CA treatment tended to induce two peaks at 22 h (second day at 09:00 h) and 46 h (third day at 09:00 h) (Fig. 5a: orange, gray and purple lines). The maximum CBF/DREB1 gene expression level was larger in the order of: the experiment in December, at constant 2°C, or in March (depending on the gene) and in November (Fig. 5a). The highest/lowest air temperature in the field between 0 and 46 h in the experiment was 14/1.3°C, 5.1/−3.8°C and 10.7/−1.1°C in November, December and March, respectively. Thus, lower temperature tended to induce greater gene expression. Interestingly, the constant 2°C treatment does not have temperature oscillations in the day and night, but temperature fluctuations in minutes induced higher gene expression than that of November. Thus, absolute temperature itself may also affect the gene expression level. The second expression peak of CBF1/DREB1B and CBF2/DREB1C was found only in CA in the field (Fig. 5a). The CBF/DREB1 genes are known to exhibit a transient expression pattern when exposed to low temperature (Fowler and Thomashow 2002, Agarwal et al. 2006). In the field, there were differences in absolute temperature, temperature fluctuation in minutes, and day length (Fig. 5b, c). In particular, there were few temperature fluctuations in December (Fig. 5c), but the second expression peak in CBF/DREB1 genes was induced (Fig. 5a). The only one difference between artificial CA at 2°C and the three field CA conditions was the temperature difference between day and night. This suggests that the temperature difference of day and night is important for the second induction of CBF/DREB1 gene expression. The Ca2+ channel blockers, 1 mM LaCl3 and 50 µM ruthenium red (RR), were used to ascertain whether multiple and large expression levels of CBF/DREB1 genes were induced by a Ca2+ signal. To visualize how the Ca2+ signals were inhibited by the Ca2+ channel blockers, a cryomicroscopy system was used. LaCl3 strongly inhibited both the speed of increase and the height of the first Ca2+ signal peak, and the speed of increase in the second and third Ca2+ signal peaks (Fig. 6). Since LaCl3 inhibited Ca2+ influx from the extracellular space (Knight and Knight 2000), the cold-induced Ca2+ signal peaks in the leaf cells may be mainly composed of the Ca2+ influx from the extracellular space. However, it had been reported that the second peak was the Ca2+ release from internal sources (Knight and Knight 2000). Thus, another possibility is that the first Ca2+ influx from the extracellular space induces the release of Ca2+ from intracellular organelles, which forms the second and third Ca2+ signal peaks. In other words, LaCl3 inhibited only the first peak, but it affects the second and the third peak in indirect ways. In fact, the slow vacuolar ion channels have been reported possibly to release Ca2+ from vacuoles depending on the increase in the cytosolic Ca2+ concentration (Ward and Schroeder 1994, Bewell et al. 1999). Another Ca2+ channel blocker, RR, was known as a putative inhibitor of mitochondrial and endoplasmic reticulum Ca2+ channels (Knight et al. 1992, Monroy and Dhindsa 1995). We observed that RR weakly inhibited all Ca2+ signal peaks in leaf cells (Fig. 6). RR inhibited the speed of increase for the second and third Ca2+ signal peaks and the height and rate of increase for the first Ca2+ signal peak (Fig. 6c, d), suggesting that all Ca2+ signal peaks included the Ca2+ influx from organelles. Fig. 6 View largeDownload slide The effect of Ca2+ channel blockers on Ca2+ signals in leaf cells in the cryochamber. (a) The plants were cooled from 20 to 1.9°C at a 1.6°C min−1 cooling rate. (b) The Ca2+ signals were induced by cooling with a chemical treatment, i.e. 0.05% of Triton X-100 solution was sprayed on the plants as control (black line), and 1 mM LaCl3 with 0.05% Triton X-100 solution (gray line) or 50 µM RR with 0.05% Triton X-100 solution (broken black line) was sprayed on plants to inhibit their Ca2+ signals. Ratio values were normalized by the value at t = 0. Each plot of normalized FRET values represents the average of independent experiments (n = 5). (c) The slopes between peak time and 60 s before peak time were calculated to observe the effect of the chemicals. (d) The boxplots present the first to third quartile values and the bars represent the minimum and maximum values of the first to third Ca2+ signal peaks. Significant differences between each data set were analyzed by using a t-test (***P < 0.001, **P < 0.01, *P < 0.05) (n = 4–5). Fig. 6 View largeDownload slide The effect of Ca2+ channel blockers on Ca2+ signals in leaf cells in the cryochamber. (a) The plants were cooled from 20 to 1.9°C at a 1.6°C min−1 cooling rate. (b) The Ca2+ signals were induced by cooling with a chemical treatment, i.e. 0.05% of Triton X-100 solution was sprayed on the plants as control (black line), and 1 mM LaCl3 with 0.05% Triton X-100 solution (gray line) or 50 µM RR with 0.05% Triton X-100 solution (broken black line) was sprayed on plants to inhibit their Ca2+ signals. Ratio values were normalized by the value at t = 0. Each plot of normalized FRET values represents the average of independent experiments (n = 5). (c) The slopes between peak time and 60 s before peak time were calculated to observe the effect of the chemicals. (d) The boxplots present the first to third quartile values and the bars represent the minimum and maximum values of the first to third Ca2+ signal peaks. Significant differences between each data set were analyzed by using a t-test (***P < 0.001, **P < 0.01, *P < 0.05) (n = 4–5). Subsequently, we focused on the inhibition of gene expression levels by the Ca2+ channel blockers. We focused on the effect of Ca2+ channel blockers on the gene expression level of CBF/DREB1 genes, in particular the second expression peak (i.e. at 46 h), because the first peak of the CBF/DREB1 genes was possibly caused by the temperature change when plants were transferred from the chamber at 23°C to the CA treatment (e.g. outside or the 2°C chamber). In conventional CA at 2°C, there was no significant difference in the second expression peak of CBF1/DREB1 between control and chemical-treated plants (Fig. 7a). However, in the field CA in mid-November, the expression levels of CBF1/DREB1B and CBF2/DREB1C were significantly higher in the control than in the LaCl3- and RR-treated plants, except for the expression of CBF1/DREB1B with LaCl3. In contrast, the expression level of CBF3/DREB1A was not inhibited by the Ca2+ channel blockers (Fig. 7b). In the CA under field conditions in early December, the expression levels of CBF/DREB1 genes were also not inhibited by the Ca2+ channel blockers (Fig. 7c). Conversely, in CA under field conditions in March, the expression level of CBF1/DREB1B was inhibited by both Ca2+ channel blockers, but CBF2/DREB1C was inhibited only by LaCl3 (Fig. 7d). Fig. 7 View largeDownload slide Inhibition of the expression levels of CBF/DREB1 genes by Ca2+ channel blockers at 46 h CA treatment. (a–d) The expression levels of CBF1–3 were measured in CA plants with 1 or 5 mM LaCl3, or 50 µM ruthenium red. The chemicals were sprayed 1 h prior to the initiation of CA treatment. (a) CA at 2°C in the chamber. (b) CA in the field on November 15–17, 2016. (c) CA in the field, on December 6–8, 2016. (d) CA in the field on March 21–23, 2017. Significant differences between each value were analyzed by using the Tukey–Kramer test (P < 0.05) (n = 3 or 4). The bars indicate the SE. Fig. 7 View largeDownload slide Inhibition of the expression levels of CBF/DREB1 genes by Ca2+ channel blockers at 46 h CA treatment. (a–d) The expression levels of CBF1–3 were measured in CA plants with 1 or 5 mM LaCl3, or 50 µM ruthenium red. The chemicals were sprayed 1 h prior to the initiation of CA treatment. (a) CA at 2°C in the chamber. (b) CA in the field on November 15–17, 2016. (c) CA in the field, on December 6–8, 2016. (d) CA in the field on March 21–23, 2017. Significant differences between each value were analyzed by using the Tukey–Kramer test (P < 0.05) (n = 3 or 4). The bars indicate the SE. Although the changes in temperature were similar in November, December and March, the extent of temperature fluctuation was larger in mid-November and late March than in early December (Fig. 5c). The SDs of the temperature changes (Fig. 5c; °C 2 min–1) between 22 and 46 h of CA in three experimental months were calculated as an index of temperature fluctuation. The correlations between the SDs of temperature changes and the expression suppression rates by the Ca2+ channel blockers were calculated and plotted (Fig. 8). We found a strong correlation between the suppression level of CBF1/DREB1B and CBF2/DREB1C at the second expression peak and the magnitude of temperature fluctuation (Fig. 8). Therefore, this suggests Ca2+ signal induction in mid-November and late March as a result of the greater magnitude of temperature fluctuation than in early December. Fig. 8 View largeDownload slide Correlation between the effect of Ca2+ channel blockers and fluctuation of temperature during CA treatment. (a and b) SDs were calculated from the temperature change between 22 and 46 h at the start of CA. The value of CA at 2°C was substituted as 0. Percentage inhibition of gene expression was calculated by subtracting each gene expression level of chemical-treated plants from the average value of the gene expression level of the control. (a) Effect of LaCl3 on the expression of CBF1–3. (b) Effect of ruthenium red on the expression of CBF1–3 (n = 3 or 4). Fig. 8 View largeDownload slide Correlation between the effect of Ca2+ channel blockers and fluctuation of temperature during CA treatment. (a and b) SDs were calculated from the temperature change between 22 and 46 h at the start of CA. The value of CA at 2°C was substituted as 0. Percentage inhibition of gene expression was calculated by subtracting each gene expression level of chemical-treated plants from the average value of the gene expression level of the control. (a) Effect of LaCl3 on the expression of CBF1–3. (b) Effect of ruthenium red on the expression of CBF1–3 (n = 3 or 4). LaCl3 tended to inhibit CBF2/DREB1C and RR inhibited CBF1/DREB1B, which suggests that CBF/DREB genes may be regulated by different Ca2+ signal parameters. As shown in Fig. 6, LaCl3 and RR inhibited Ca2+ signals at different levels. Without an inhibitor, the first and second Ca2+ signal peaks were almost the same height, and the third Ca2+ signal peak was the smallest (Fig. 6d). In the case of LaCl3 treatment, the second Ca2+ signal peak became the highest because the first peak was inhibited more strongly. In the case of RR treatment, however, the first Ca2+ signal peak became the highest because the second and third peaks were inhibited more strongly. Previous reports have discussed Ca2+ signal decoding. In guard cells, the frequency, transient number, duration and amplitude of Ca2+ signals are important for controlling stomatal aperture (Allen et al. 2001). Three different amplitudes and frequencies of Ca2+ signals caused by electrical stimulation induce very different gene expression profiles (Whalley et al. 2011, Whalley and Knight 2013). On the other hand, we demonstrated that Ca2+ channel blocker treatment affected the amplitude and ratio of Ca2+ signal peak heights (Fig. 6). In particular, RR did not strongly decrease the amplitude of Ca2+ signal peaks (Fig. 6), but inhibited CBF1/DREB1B expression significantly (Figs. 7, 8). This result indicates that the ratio of Ca2+ signal peak heights (e.g. ratio of first/second peak) and/or that of extracellular and intracellular influx may also be decoded to modulate gene expression. It had been reported that calmodulin-binding transcription activator 3 (CAMTA3) and CAMTA5 respond to a temperature decrease, and induced the expression of CBF1/DREB1B and CBF2/DREB1C via the Ca2+ signaling pathway (Doherty et al. 2009, Kidokoro et al. 2017). These reports are consistent with our results showing that the gene expression of CBF1/DREB1B and CBF2/DREB1C was inhibited by the Ca2+ channel blockers only under conditions with temperature fluctuations (Fig. 8). This suggests that Ca2+ signals induced by temperature regulate the gene expression of CBF1/DREB1B and CBF2/DREB1C in the field. In addition, LaCl3 and RR tended to suppress the first expression peak of CBF1/DREB1B and CBF2/DREB1C in each CA treatment, including conventional CA (Supplementary Fig. S5). Since the rapid temperature decrease occurred in all the conditions when plants were transferred outside or to the 2°C chamber, this suppression could be explained by the Ca2+–CAMTA pathway. The expression of CBF3/DREB1A was, however, not suppressed by LaCl3 and RR (Figs. 7, 8). CBF3/DREB1A does not have a binding site for the CAMTA protein (Kidokoro et al. 2017). CBF3/DREB1A expression is also regulated by clock factors, such as CIRCADIAN CLOCK-ASSOCIATED 1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) (Dong et al. 2011, Kidokoro et al. 2017). Therefore, we concluded that CBF3/DREB1A is controlled not only by Ca2+ signaling but also by many other factors. Moreover, the channel blockers tended to enhance the expression level of CBF1/DREB1B and CBF2/DREB1C under lower thermal amplitude fluctuations in December. These results suggest that different Ca2+ signals, which are dependent on the amplitude of the temperature fluctuation or on the absolute temperature, may assist gene expression positively and negatively dependent on the environment. Conclusions In this study, by using a newly developed Ca2+ signal determination system and carefully formulated temperature correction, we revealed several novel findings that (i) there was only one transient Ca2+ signal when plants are subjected to a rapid temperature drop at the beginning of conventional CA treatment (i.e. such as at constant 2°C), but multiple Ca2+ signals were induced in the temperature fluctuation during CA, which probably occurs under field-like CA conditions; (ii) Ca2+ signals were adapted to ambient temperatures and sensed temperatures even during or after CA; and (iii) the expression of CBF/DREB1 genes was regulated by the Ca2+ signals which were induced by the temperature fluctuation in the field during the winter. These results suggest that Ca2+ signals act as a messenger for temperature fluctuation in minutes in the field instead of large diurnal temperature changes from day to night. Materials and Methods Plant materials Arabidopsis thaliana (Col-0) plants were grown in Petri dishes containing modified Hoagland’s nutrient solution solidified with 1% agar (Uemura et al. 1995), and grown at 23°C under a 16 h photoperiod (photon flux rate: 80 µmol m−2 s−1) for 7 d (for root samples) or 14 d (for leaf samples). Plants were incubated in a vertical position for the root study. Some plants were cold-acclimated at 2°C under a 12 h photoperiod (photon flux rate: 80 µmol m−2 s−1) or in the field. The samples were covered with moistened paper and a plastic sheet to maintain humid conditions during CA treatment. Sample preparation for microscopic observations For root observation, YC3.60-expressing Arabidopsis seedlings were placed on a 24 mm×18 mm cover glass, with only the roots covered by an 18 mm×18 mm cover glass. Silicone (Sin-Etsu Chemical Co., Ltd.) was used to ensure that space was maintained between the cover glasses (i.e. a shallow well was constructed). The well was filled with 100 µl of 1 mM MES/KOH buffer (pH 5.6). The aerial plant parts were covered with plastic wrap, and a drop of water was added to prevent the sample from dehydration (Supplementary Methods S1a). The prepared samples were then placed on the cryostage and temporarily fixed with thin silicone to improve thermal conductivity. Before starting all experiments, samples were placed on the cryostage and maintained at a constant temperature for at least 10 min (i.e. starting temperature). Fluorescence images of the root elongation zone were observed at intervals of 1– 2 s. For leaf observation, YC3.60-expressing Arabidopsis seedlings were placed in a Petri dish (diameter: 40 mm) containing modified Hoagland’s nutrient medium and 1 ml of 1 mM MES/KOH (pH 5.6) buffer. The surface of the medium was covered with plastic wrap (Supplementary Methods S2a). The dish was attached to the bottom of a glass-jacketed beaker with silicone, and the samples were incubated at the starting temperature for at least 10 min. Fluorescence images of the fifth leaf (except the cotyledon) were observed at 5 s intervals. Cryomicroscopy Cryomicroscopy-based experimental systems were employed to observe samples at a microscopic level during temperature changes. The cryomicroscopy system for root analyses consisted of a confocal microscope (C2si; Nikon Co.; http://www.nikon-instruments.jp/jpn/bioscience-products/confocal/c2/index.html), a cryostage (THMS600; Linkam Scientific Instruments, Ltd.; http://www.linkam.co.uk/thms600-features/; Supplementary Methods S1c), an objective inverter (300T; LSM TECH; http://lsmtech.com/our-products/#InverterScope), and an iron plate to adjust the height of the cryostage. The cryostage temperature was regulated by a heater and liquid N2 following the program edited by Linksys for Windows (Linkam Scientific Instruments). The cryostage was covered with a plastic bag filled with dry air to prevent the misting of the lens and cover glass (Supplementary Methods S1d). The cryomicroscopy system for leaf analyses consisted of a confocal microscope, a cooling circulator (Ministat 230 with Pilot ONE; Huber Kältemaschinenbau GmbH; http://www.huber-online.com/en/product_datasheet.aspx? no=2015.0012.01), a glass-jacketed beaker (Asahi Glassplant Inc.; http://www.asahiglassplant.com/products/) (as a cooling chamber), an objective inverter, an adjustment cylinder (to extend the objective inverter) and an iron plate. The glass-jacketed beaker was covered with an expanded polystyrene box (to stabilize the internal temperature) and a plastic box containing a rubber sheet at the base (to enable the attachment of the iron plate) (Supplementary Methods S2). To monitor air and medium temperatures, four thermocouples were attached to the medium and the beaker. Temperature data were collected using a thermocouple data logger (MCR-4TC; T&D Co.; https://www.tandd.co.jp/product/mcr-4_series.html). Ca2+ signal measurement, correction for temperature and data analysis The cryomicroscopy systems to analyze transgenic Arabidopsis plants expressing a FRET-based Ca2+ sensor (i.e. YC3.60) (Nagai et al. 2004) enabled the observation of fluctuations of the cytosolic Ca2+ concentration. Fluorescence images were captured with a 32 channel spectral detector or 472/30 nm (for ECFP fluorescence) and 527/20 nm (cpVenus fluorescence) dichroic filters used for a confocal laser scanning microscope. The fluorescence ratio was calculated according to the following equation using the fluorescence intensity values of cpVenus at t s [VF(t)] and ECFP at t s [CF(t)]. Ratio (t)=[VF(t)][CF(t)] The normalized FRET value for each sample was determined with this equation by using the NIS-elements C program (Nikon). The data for the experiment assessing the effects of temperature changes needed to be corrected because the fluorescence intensities were influenced by temperature (Felber et al. 2004). We established a correction formula for temperature changes in leaf and root cells. Leaf protoplasts and whole roots were used for making the correction formula. Since the leaf is covered with a cuticle layer which prevents the liquid chemicals from entering into the cell, leaf protoplasts were prepared. Leaf protoplasts and roots were treated with 2.5 µM ionomycin, 0.01% Triton X-100 and 250 nM CaCl2 for 1 h to increase the intracellular Ca2+ concentration. The FRET values at 25, 20, 15, 10, 5, 0 and −5°C were measured (Supplementary Method S3). The correction formulae for leaf and root cells were based on an approximation formula. Finally, the normalized FRET values indicating the Ca2+ concentration were calculated using the following equations and the smoothed FRET values at t s [SR(t)] and 0 s [SR(0)]: Corrected Relative Ratio of Root Cells (t) = [SR(t)]/1.0772e0.0122×T(t)[SR(0)]/1.0772e0.0122×T(0) Corrected Relative Ratio of Leaf Cells (t)= [SR(t)]/3.147e0.0235×T(t)[SR(0)]/3.147e0.0235×T(0) where T represents the temperature at a specific time point. Additionally, each data set was subjected to binominal smoothing using the IGOR Pro 6.3.6.0 program (WaveMetrics, Inc.; https://www.wavemetrics.com/) to detect the Ca2+ signal peaks. This program was also used for analyzing the Ca2+ signals and determining the timing and height of the peaks. The outliers caused by changes in focus during microscopic analyses were replaced with values calculated with IGOR Pro. Averaged rather than individual Ca2+ signals are presented in the figures. Sometimes, the averaged Ca2+ signals do not show the actual peak shape due to the differential signal initiation. To clearly visualize the data, we modified the x-axis (time) for the Ca2+ signals, and the first peak time was replaced with the averaged first peak time in Figs. 1c, 2b, d, 4c and 6b. Untransformed data are also shown in Supplementary Methods S5. This was performed only for visualizing the Ca2+ signals, but not for detailed analysis such as peak time and height. By this method, we prevented the Ca2+ signals from deforming due to the averaging treatment. Measurement of mRNA accumulation level Total RNA was isolated from aerial parts of the Arabidopsis (Col-0) plant using ISOSPIN Plant RNA (NIPPON GENE CO., LTD.). ReverTra Ace qPCR RT Master Mix (TOYOBO Co.) was used for the cDNA synthesis. Real-time PCR was performed on the corresponding cDNA synthesized from each sample. PDF2 was used as a reference gene. The primers which were used for real-time PCR are listed in Supplementary Table S1. Cold acclimation treatment in the field The samples were covered with moistened paper and a plastic sheet to maintain humid conditions during CA in the field. The field cold acclimation was performed on November 15–17 and December 6–8, 2016, and on March 21–23, 2017, in Morioka, Japan. Temperature data were collected every 2 min using a thermocouple data logger (TR-55i; T&D Co.). We avoided measuring the leaf temperature directly in order not to change the heat conductivity of the leaf. Also, we confirmed that there were few differences between leaf and air temperature (Supplementary Methods S6). Chemical treatments A solution of 0.05% Triton X-100 was used as the control, and 0.05% Triton X-100 and 1.5 mM LaCl3 or 50 µM RR were used to inhibit the Ca2+ signaling during CA treatment or Ca2+ observation. About 200 µl of solution was sprayed on each plant 1 h prior to the experiment. Funding This work was supported by the Japanese Society for the Promotion of Science [Grants-in Aid Nos. 25292205 and 18K19319 to Y.K. and Nos. 22120003 and 17H03961 to M.U.] and the Dean’s fund of the United Graduate School of Agricultural Sciences (UGAS), Iwate University [to H.H.]. Acknowledgments We would like to express our thanks to Professor Karin Schumacher (Heidelberg University, Heidelberg, Germany) for Arabidopsis plants expressing YC3.60, and Ms. Mei Ogata for the data of the air temperature in the field. We greatly appreciate technical support from Ms. Yuko Suzuki, Michiko Saito and Nozomi Yokota. We thank Professor Tatsumi Hiraki for English editing of this manuscript. We would like to thank Professor Karen Tanino for reviewing this manuscript. Disclosures The authors have no conflicts of interest to declare. Footnotes Subject Areas: (2) Environmental and stress responses References Agarwal M. , Hao Y. , Kapoor A. , Dong C.H. , Fujii H. , Zheng X. , et al. . ( 2006 ) A R2R3 type MYB transcription factor is involved in the cold regulation of CBF genes and in acquired freezing tolerance . J. Biol. Chem. 281 : 37636 – 37645 . Google Scholar Crossref Search ADS PubMed Allen G.J. , Chu S.P. , Harrington C.L. , Schumacher K. , Hoffmann T. , Tang Y.Y. , et al. . ( 2001 ) A defined range of guard cell calcium oscillation parameters encodes stomatal movements . Nature 411 : 1053 – 1057 . Google Scholar Crossref Search ADS PubMed Bewell M.A. , Maathuis F.J.M. , Allen G.J. , Sanders D. ( 1999 ) Calcium-induced calcium release mediated by a voltage-activated cation channel in vacuolar vesicles from red beet . FEBS Lett . 458 : 41 – 44 . Google Scholar Crossref Search ADS PubMed Campbell A.K. , Trewavas A.J. , Knight M.R. ( 1996 ) Calcium imaging shows differential sensitivity to cooling and communication in luminous transgenic plants . Cell Calcium 19 : 211 – 218 . Google Scholar Crossref Search ADS PubMed Chinnusamy V. , Zhu J.K. , Sunkar R. ( 2010 ) Gene regulation during cold stress acclimation in plants . Methods Mol. Biol. 639 : 39 – 55 . Google Scholar Crossref Search ADS PubMed Dat J. , Vandenabeele S. , Vranová E. , Van Montagu M. , Inzé D. , Van Breusegem F. ( 2000 ) Dual action of the active oxygen species during plant stress responses . Cell. Mol. Life Sci . 57 : 779 – 795 . Google Scholar Crossref Search ADS PubMed Doherty C.J. , Van Buskirk H.A. , Myers S.J. , Thomashow M.F. ( 2009 ) Roles for Arabidopsis CAMTA transcription factors in cold-regulated gene expression and freezing tolerance . Plant Cell 21 : 972 – 984 . Google Scholar Crossref Search ADS PubMed Dong M.A. , Farré E.M. , Thomashow M.F. ( 2011 ) Circadian clock-associated 1 and late elongated hypocotyl regulate expression of C-repeat binding factor (CBF) pathway in Arabidopsis . Proc. Natl. Acad. Sci. USA 108 : 7241 – 7246 . Google Scholar Crossref Search ADS Felber L.M. , Cloutier S.M. , Kündig C. , Kishi T. , Brossard V. , Jichlinski P. , et al. . ( 2004 ) Evaluation of the CFP–substrate–YFP system for protease studies: advantages and limitations . Biotechniques 36 : 878 – 885 . Google Scholar Crossref Search ADS PubMed Fowler S. , Thomashow M.F. ( 2002 ) Arabidopsis transcriptome profiling indicates that multiple regulatory pathway are activated during cold acclimation in addition to the CBF response pathway . Plant Cell 14 : 1675 – 1690 . Google Scholar Crossref Search ADS PubMed Han S. , Tang R. , Anderson L.K. , Woerner T.E. , Pei Z.M. ( 2003 ) A cell surface receptor mediates extracellular Ca2+ sensing in guard cells . Nature 425 : 196 – 200 . Google Scholar Crossref Search ADS PubMed Haswell E.S. , Meyerowitz E.M. ( 2006 ) MscS-like proteins control plastid size and shape in Arabidopsis thaliana . Curr. Biol. 16 : 1 – 11 . Google Scholar Crossref Search ADS PubMed Hochmal A.K. , Schulze S. , Trompelt K. , Hippler M. ( 2015 ) Calcium-dependent regulation of photosynthesis . Biochim. Biophys. Acta 1847 : 993 – 1003 . Google Scholar Crossref Search ADS PubMed Huang L. , Berkelman T. , Franklin A.E. , Hoffman N.E. ( 1993 ) Characterization of a gene encoding a Ca2+-ATPase-like protein in the plastid envelope . Proc. Natl. Acad. Sci. USA 90 : 10066 – 10070 . Google Scholar Crossref Search ADS Kiegle E. , Moore C.A. , Haseloff J. , Tester M.A. , Knight M.R. ( 2000 ) Cell-type-specific calcium response to drought, salt and cold in the Arabidopsis root . Plant J. 23 : 267 – 278 . Google Scholar Crossref Search ADS PubMed Kidokoro S. , Yoneda K. , Takasaki H. , Takahashi F. , Shinozaki K. , Yamaguchi-Shinozaki K. ( 2017 ) Different cold-signaling pathway function in the responses to rapid and gradual decrease in temperature . Plant Cell 29 : 769 – 774 . Google Scholar Crossref Search ADS Knight H. , Knight M.R. ( 2000 ) Imaging spatial and cellular characteristics of low temperature calcium signature after cold acclimation in Arabidopsis . J. Exp. Bot . 51 : 1679 – 1686 . Google Scholar Crossref Search ADS PubMed Knight H. , Trewavas A.J. , Knight M.R. ( 1996 ) Cold calcium signaling in Arabidopsis involves two cellular pools and a change in calcium signature after acclimation . Plant Cell 8 : 489 – 503 . Google Scholar Crossref Search ADS PubMed Knight M.R. , Smith S.M. , Trewavas A.J. ( 1992 ) Wind-induced plant motion immediately increases cytosolic calcium . Proc. Natl. Acad. Sci. USA 89 : 4967 – 4971 . Google Scholar Crossref Search ADS Krebs M. , Held K. , Binder A. , Hashimoto K. , Herder G.D. , Parniske M. , et al. . ( 2012 ) FRET-based genetically encoded sensors allow high-resolution live cell imaging of Ca2+ dynamics . Plant J. 69 : 181 – 192 . Google Scholar Crossref Search ADS PubMed Levitt J. ( 1980 ) Responses of Plants to Environmental Stress: Chilling, Freezing and High Temperature Stresses . Academic Press , New York . Martí M.C. , Stancombe M.A. , Webb A.A.R. ( 2013 ) Cell- and stimulus type-specific intracellular free Ca2+ signals in Arabidopsis . Plant Physiol. 163 : 625 – 634 . Google Scholar Crossref Search ADS PubMed McAinsh M.R. , Pittman J.K. ( 2009 ) Shaping the calcium signature . New Phytol. 181 : 275 – 294 . Google Scholar Crossref Search ADS PubMed Miura K. , Furumoto T. ( 2013 ) Cold signaling and cold response in plants . Int. J. Mol. Sci. 14 : 5312 – 5337 . Google Scholar Crossref Search ADS PubMed Monroy A.F. , Dhindsa R.S. ( 1995 ) Low-temperature signal transduction: induction of cold acclimation-specific genes of alfalfa by calcium at 25°C . Plant Cell 7 : 321 – 331 . Google Scholar PubMed Nagai T. , Yamada S. , Tominaga T. , Ichikawa M. , Miyawaki A. ( 2004 ) Expanded dynamic range of fluorescent indicators for Ca2+ by circularly permuted yellow fluorescent proteins . Proc. Natl. Acad. Sci. USA 101 : 10554 – 10559 . Google Scholar Crossref Search ADS Nagel-Volkmann J. , Plieth C. , Becker D. , Lüthen H. , Dörffling K. ( 2009 ) Cold-induced cytosolic free calcium ion concentration changes in wheat . Plant Physiol . 166 : 1955 – 1960 . Google Scholar Crossref Search ADS Nomura H. , Komori T. , Kobori M. , Nakahira Y. , Shiina T. ( 2008 ) Evidence for chloroplast control of external Ca2+-induced cytosolic Ca2+ transients and stomatal closure . Plant J . 53 : 988 – 998 . Google Scholar Crossref Search ADS PubMed Örvar B.L. , Sangwan V. , Omann F. , Dhindsa R.S. ( 2000 ) Early steps in cold sensing by plant cells: the role of actin cytoskeleton and membrane fluidity . Plant J. 23 : 785 – 794 . Google Scholar Crossref Search ADS PubMed Peltier J.B. , Ytterberg A.J. , Sun Q. , Wijk K.J. ( 2004 ) New functions of the thylakoid membrane proteome of Arabidopsis thaliana revealed by simple, fast, and versatile fractionation strategy . J. Biol. Chem. 279 : 49367 – 49383 . Google Scholar Crossref Search ADS PubMed Plieth C. , Hansen U.P. , Knight H. , Knight M.R. ( 1999 ) Temperature sensing by plants: the primary characteristics of signal perception and calcium response . Plant J. 18 : 491 – 497 . Google Scholar Crossref Search ADS PubMed Rocha A.G. , Vothknecht U.C. ( 2012 ) The role of calcium in chloroplast—an intriguing and unresolved puzzle . Protoplasma 249 : 957 – 966 . Google Scholar Crossref Search ADS PubMed Sangwan V. , Foulds I. , Singh J. , Dhindsa R.S. ( 2001 ) Cold-activation of Brassica napus BN115 promoter is mediated by structural changes in membranes and cytoskeleton, and requires Ca2+ influx . Plant J. 27 : 1 – 12 . Google Scholar Crossref Search ADS PubMed Seigneurin-Berny D. , Gravot A. , Auroy P. , Mazard C. , Kraut A. , Finazzi G. , et al. . ( 2005 ) HMA1, a new Cu-ATPase of the chloroplast envelope, is essential for growth under adverse light conditions . J. Biol. Chem. 281 : 2882 – 2892 . Google Scholar Crossref Search ADS PubMed Teardo E. , Formentin E. , Sefalla A. , Fiacometti G.M. , Marin O. , Zanetti M. , et al. . ( 2011 ) Dual localization of plant glutamate recepror AtGLR3.4 to plastid and plasmamembrane . Biochim. Biophys. Acta 1807 : 357 – 367 . Uemura M. , Joseph R.A. , Steponkus P.L. ( 1995 ) Cold acclimation of Arabidopsis thaliana: effect on plasma membrane lipid composition and freeze-induced lesions . Plant Physiol. 109 : 15 – 30 . Google Scholar Crossref Search ADS PubMed Vainonen J.P. , Sakuragi Y. , Stael S. , Tikkanen M. , Allahverdiyeva Y. , Paakkarinen V. , et al. . ( 2008 ) Light regulation of CaS, a novel phosphoprotein in the thylakoid membrane of Arabidopsis thaliana . FEBS J . 275 : 1767 – 1777 . Google Scholar Crossref Search ADS PubMed Wang Y. , Hua J. ( 2009 ) A moderate decrease in temperature induces COR15a expression through the CBF signaling cascade and enhances freezing tolerance . Plant J . 60 : 340 – 349 . Google Scholar Crossref Search ADS PubMed Ward J.M. , Schroeder J.I. ( 1994 ) Calcium-activated K+ channels and calcium-induced calcium release by slow vacuolar ion channels in guard cell vacuoles implicated in the control of stomatal closure . Plant Cell 6 : 669 – 683 . Google Scholar PubMed Wasternack C. ( 2007 ) Jasmonates: an update on biosynthesis, signal transduction and action in plant stress response, growth and development . Ann. Bot . 100 : 681 – 697 . Google Scholar Crossref Search ADS PubMed Whalley H.J. , Knight M.R. ( 2013 ) Calcium signatures are decoded by plants to give specific gene responses . New Phytol. 197 : 690 – 693 . Google Scholar Crossref Search ADS PubMed Whalley H.J. , Sargeant A.W. , Steele J.F. , Lacoere T. , Lamb R. , Saunders N.J. , et al. . ( 2011 ) Transcriptomic analysis reveals calcium regulation of specific promoter motifs in Arabidopsis . Plant Cell 23 : 4079 – 4095 . Google Scholar Crossref Search ADS PubMed Xiong L. , Schumaker K.S. , Zhu J.K. ( 2002 ) Cell signaling during cold, drought, and salt stress . Plant Cell 14 : 165 – 183 . Google Scholar Crossref Search ADS PubMed Yoshida S. ( 1983 ) Studies on freezing injury of plant cells. I. Relation between thermotropic properties of isolated plasma membrane vesicles and freezing injury . Plant Physiol . 75 : 38 – 42 . Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Calcium Signaling-Linked CBF/DREB1 Gene Expression was Induced Depending on the Temperature Fluctuation in the Field: Views from the Natural Condition of Cold Acclimation
JF - Plant and Cell Physiology
DO - 10.1093/pcp/pcy210
DA - 2019-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/calcium-signaling-linked-cbf-dreb1-gene-expression-was-induced-50y5FfZ9FP
SP - 303
VL - 60
IS - 2
DP - DeepDyve
ER -