Aequorin-Based Luminescence Imaging Reveals Stimulus- and Tissue-Specific Ca²⁺ Dynamics in Arabidopsis Plants

SUMMARY

An Aequorin-based Film Adhesive Seedling (FAS) Ca²⁺ recording system was developed for monitoring Ca²⁺ in response to various stimuli in Arabidopsis. This system revealed stimulus- and tissue-specific Ca²⁺ signatures in seedlings with a simple, sensitive, and robust Ca²⁺ recording.

Abstract

Calcium ion is a versatile second messenger for diverse cell signaling in response to developmental and environmental cues. The specificity of Ca²⁺-mediated signaling is defined by stimulus-elicited Ca²⁺ signature and downstream decoding processes. Here, an Aequorin-based luminescence recording system was developed for monitoring Ca²⁺ in response to various stimuli in Arabidopsis. With the simple, highly sensitive, and robust Ca²⁺ recording, this system revealed stimulus- and tissue-specific Ca²⁺ signatures in seedlings. Cellular Ca²⁺ dynamics and relationship to Aequorin-based Ca²⁺ recording were explored using a GFP-based Ca²⁺ indicator, which suggested that a synchronous cellular Ca²⁺ signal is responsible for cold-induced Ca²⁺ response in seedlings, whereas asynchronous Ca²⁺ oscillation contributes to osmotic stress-induced Ca²⁺ increase in seedlings. The optimized recording system would be a powerful tool for the identification and characterization of novel components in Ca²⁺-mediated stress-signaling pathways.

INTRODUCTION

Calcium ion has been adopted as a primary signal element during evolution. To respond to changing developmental and environmental cues, plant cells rapidly change cytosolic free Ca²⁺ concentration in time and space. The spatiotemporal patterning of cellular Ca²⁺ dynamics has been formulated as a concept of Ca²⁺ signature. Ca²⁺ signature, namely the alterations in amplitude, duration, frequency, and spatial distribution of Ca²⁺ signal, encodes information of the type and the strength of the stimuli (Knight et al., 1996; Webb et al., 1996; McAinsh and Hetherington, 1998; Trewavas and Malho, 1998). Such Ca²⁺-encoded stimulus-specific information is decoded by downstream effectors. Most downstream effectors are Ca²⁺ sensing proteins, which bind to Ca²⁺ to initiate or regulate biochemical processes and ultimately translate information into specific end responses (Harper et al., 2004; Kim et al., 2009; Luan, 2009; DeFalco et al., 2010; Kudla et al., 2010). Therefore, the specificity of stimulus–response coupling is believed to be achieved by stimulus-defined Ca²⁺ encoding and decoding processes.

Ca²⁺ signaling regulates diverse cellular responses to abiotic and biotic stresses, plant hormones, light, and mechanical stimulation (Kudla et al., 2010). However, it remains unknown how plant cells generate stimulus-specific Ca²⁺ signals. A number of Ca²⁺-signaling components have been identified in plant cells that comprise Ca²⁺-signaling toolkits (Harper, 2001; McAinsh and Pittman, 2009; Kudla et al., 2010). Such cellular Ca²⁺-signaling toolkits enable the cell to use both extracellular and internal sources of Ca²⁺. However, in plant cells, most Ca²⁺ channels that control the entry of extracellular Ca²⁺ or the release of Ca²⁺ from internal store are still largely unidentified. So far, three families of proteins have been demonstrated to have Ca²⁺-influx activity. Ligand-gated receptors, cyclic nucleotide-gated channels (CNGCs), and glutamate receptor-like channels (GLRs) have been shown to possess Ca²⁺-influx activities. CNGCs function in defense signaling and thermosensing (Ali et al., 2007; Ma and Berkowitz, 2011; Finka et al., 2012), whereas GLRs are proposed to be responsible for selected amino acid-triggered Ca²⁺ influx, functioning in pollen tube growth (Michard et al., 2011), stomatal closure (Cho et al., 2009), root development (Li et al., 2006; Qi et al., 2006), and defense signaling (Kwaaitaal et al., 2011; Vatsa et al., 2011). Unlike CNGCs and GLRs, two pore channel (TPC1) is a vacuolar Ca²⁺ release channel, regulating ABA-sensitive germination and stomatal opening (Peiter et al., 2005; Islam et al., 2010; Hedrich and Marten, 2011). In addition to the Ca²⁺-influx system mentioned above, electrophysiological studies have shown Ca²⁺ currents when plasma membrane is hyperpolarized or depolarized, indicating the existence of voltage-gated Ca²⁺ channels in plants. In addition, the ligands 1,4,5-trisphosphate and cyclic ADP-ribose likely trigger Ca²⁺ release from internal stores through ligand receptor-operated channels in plants (Allen et al., 1995; Martinec et al., 2000; Navazio et al., 2000; Lemtiri-Chlieh et al., 2003), although the molecular identities of these channels have yet to be identified. On the other hand, most Ca²⁺ pumps and exchangers that rapidly transport cytosolic Ca²⁺ back to the extracellular space or internal membrane stores have been well characterized. Two types of Ca²⁺ pumps and exchangers, P-type Ca²⁺ ATPase (ECAs, ER-type Ca²⁺ ATPase and ACAs, autoinhibited Ca²⁺ ATPase) and Ca²⁺ exchanger (CAXs), are located in the plasma membrane and different organelles (Sze et al., 2000; Mäser et al., 2001; Shigaki and Hirschi, 2006). They bring Ca²⁺ to the resting level and are important in shaping Ca²⁺ signatures (McAinsh and Pittman, 2009).

Unveiling the mechanism by which plant cells initiate stimulus-specific Ca²⁺ signals is crucial to understanding how plants perceive and transduce signals to adapt to various stresses. A precise measurement of cellular Ca²⁺ dynamics is a prerequisite for identifying genes and defining their functions in Ca²⁺ signaling. So far, the measurements of Ca²⁺ dynamics at the cellular or whole plant level are performed in various plant species using different Ca²⁺ indicators, including ratiometric fluorescent dyes and genetically encoded Ca²⁺ indicators (Campbell et al., 1996; Knight et al., 1996, 1997; Cessna et al., 1998; Gong et al., 1998; Mori et al., 1998; Blume et al., 2000; Allen et al., 2001; Kosuta et al., 2008; Monshausen et al., 2009; Krebs et al., 2011; Munemasa et al., 2011; Ranf et al., 2012). However, stimuli-specific Ca²⁺ signatures have not been systematically defined in Arabidopsis at the level of either the single cell or the whole plant. Furthermore, the identification of the components, especially Ca²⁺-influx channels and their regulators in Ca²⁺-signaling toolkits, necessitate well-designed genetic screens in Arabidopsis.

To bridge these gaps, we developed an Aequorin-based FAS Ca²⁺ recording system that allows a systematic comparison of Ca²⁺ dynamics in response to stimuli of various strengths and types, including abiotic and biotic stresses, plant hormones, and amino acids in Arabidopsis seedlings. Using the GFP-based Ca²⁺ indicator Case12, we demonstrated cellular Ca²⁺ dynamics in response to cold and osmotic stresses, and provided experimental evidence that the whole seedling or organ-level Ca²⁺ patterns reflect combined single-cell Ca²⁺ dynamics of populations of cells. We further demonstrated that the FAS recording was able to detect alterations of Ca²⁺ signals by Ca²⁺ channel blockers. As a result, FAS recording could be a powerful system for the isolation of mutants with altered stimulus-specific Ca²⁺ signals, which may help reveal the molecular identities of Ca²⁺-signaling components that are important in abiotic stress-signaling pathways.

RESULTS

Aequorin-Based FAS Recording for Tracking Spatiotemporal Ca²⁺ Signals in Arabidopsis Seedlings

In order to simultaneously monitor tissue-specific Ca²⁺ responses in multiple Arabidopsis seedlings, we developed an Aequorin-based FAS luminescence recording system. We used transgenic Arabidopsis plants expressing cytosolic Aequorin, referred to as the Aeq or Aeq wild-type plants. Aeq seedlings were grown vertically on MS agar in a square Petri dish at a density of about 100 seedlings per plate until 7 d old, and were then transferred onto an adhesive-backed film. A sheet of FAS was incubated with h-CTZ for the reconstitution of Aequorin. Stimuli-induced Ca²⁺ signals in FAS were recorded using a photon-counting camera (Supplemental Figure 1). Sequential luminescence images over the time course are shown in Figure 1, which demonstrate spatiotemporal dynamic changes of Ca²⁺ in response to different abiotic stress stimuli. The free cytosolic Ca²⁺ level in the cell is correlated with the luminescence intensity emitted from Aequorin (Gilroy et al., 1989). A comparison of the amplitude and duration of Ca²⁺ increase is made based on the average photon density, namely total photon density divided by the area of region of interest (ROI). As shown in Figures 1 and 2, abiotic stress stimuli-induced Ca²⁺ signals vary greatly in terms of the speed of signal initiation, and signal amplitude and duration, in addition to differences in tissue distribution patterns. In all cases, the amplitude, but not always the duration, of Ca²⁺ signals was correlated with the strength of the stimulus. For example, the 0°C water-induced Ca²⁺ amplitude was 1.4-fold higher than that induced by 4°C water. However, the duration of Ca²⁺ increase was the same under 0°C or 4°C, and both Ca²⁺ signals show only one sharp spike. In contrast, heat stress triggered a longer duration of Ca²⁺ increase, although the Ca²⁺ amplitudes were very similar to that triggered by 4°C water. As shown in Figures 1 and 2, seedlings responded differently to different levels of acidic pH. A delayed initiation of Ca²⁺ signals was observed in response to both pH 3.5 and pH 4.5 acidic stimuli, but the stronger acidic stimulus (pH 3.5) initiated an earlier and greater Ca²⁺ increase than the more moderate acidic stimulus (pH 4.5). In this study, mannitol, hydrogen peroxide, and sodium chloride were used for mimicking osmotic, oxidative, and salt stress stimuli, respectively. The concentrations used in this study were much lower than those used in previously published studies, which indicates a high sensitivity of our FAS recording. The lowest concentrations tested here for mannitol, NaCl, and hydrogen peroxide were 200, 50, and 1mM, in which Ca²⁺ signals were clearly detectable, although the amplitudes were 2.5-, 1.4-, and 1.06-fold lower than those triggered by 400, 75, and 2mM NaCl, mannitol, and H₂O₂, respectively. The results indicated that the Ca²⁺ amplitudes in whole seedlings were correlated with the strengths of osmotic, oxidative, and salt stresses (Figure 2). Interestingly, various tissue patterns of Ca²⁺ dynamics were observed, and the tissue patterns were not only related to the type, but also to the strength, of the stress stimuli. Moderate cold and heat stresses, such as 4°C and 37°C water, triggered strong Ca²⁺ signals in leaves, but they triggered only weak Ca²⁺ signals in roots. In contrast, severe cold and heat stress, such as 0°C and 45°C water, triggered strong Ca²⁺ signals in roots, in addition to the leaves. Such tissue-specific patterns were even more pronounced in the seedlings responding to the osmotic stress agent mannitol, in which Ca²⁺ signal in roots was not even detectable in response to 200mM mannitol (Figure 1). The speed of initiation of Ca²⁺ signal was also correlated with the strength of the stimuli, as exemplified by delayed responses to 200mM mannitol and pH 4.5, compared to the responses to 400 mM mannitol and pH 3.5 medium, respectively.

Figure 1.

Abiotic Stress-Induced Spatiotemporal Patterns of Ca²⁺ Signals in Arabidopsis Seedlings.

Time course of luminescence images of FAS. Sequential images were collected using a photon-counting camera with an interval of 40-s photon-counting integrations upon application of stimuli. The type and strength of abiotic stress stimuli are indicated on the left of each panel, and time points are indicated on the top of each panel. The display range is 100–200 for all images.

Figure 2.

Temporal Patterns of Abiotic Stress Stimulus-Induced Ca²⁺ Signal of the Whole Seedlings.

Stimulus-induced Ca²⁺ increases are illustrated by the averaged photon intensity over time. The total photon was counted for 160 s for cold (0°C and 4°C water), oxidative (1 and 2 mM H₂O₂), osmotic (200 and 400 mM mannitol), and salt (50 and 75 mM NaCl) stress stimuli, and 400 s for heat (37°C and 45°C water), acidic (pH 3.5 and 4.5) stress stimuli, and control (room temperature water, pH 7.6).

Taken together, the Aequorin-based FAS recording system provided a platform for a highly sensitive and robust Ca²⁺ signal recording. Such Ca²⁺ recording revealed diverse abiotic stress stimuli-specific Ca²⁺ spatiotemporal patterns of Ca²⁺ signals.

A GFP-Based Ca²⁺ Indicator for Monitoring Cellular Ca²⁺ Dynamics

Aequorin-recorded Ca²⁺ signal is a readout of populations of cells. It is unclear how Aequorin-recorded Ca²⁺ signature in whole seedlings or tissues of seedlings is related to cellular Ca²⁺ dynamics. To understand cellular Ca²⁺ dynamics underlying the FAS-recorded Ca²⁺ signature, we used a genetically encoded Ca²⁺ indicator, Case12, for tracking cellular Ca²⁺ dynamic changes in response to cold and osmotic stresses. Case12 is a GFP-based Ca²⁺ indicator, for which changes in fluorescent intensity are directly correlated with changes in Ca²⁺ concentration (Souslova et al., 2007; Zhu et al., 2010). In this study, we generated transgenic Arabidopsis plants constitutively expressing Case12 in the cytosol. Ca²⁺ signals in the root cells were recorded upon application of 0°C water or 400mM mannitol under a confocal microscope. As shown in Figure 3A and 3B, 0°C water triggered synchronous Ca²⁺ spikes in root cells (Supplemental Movie 1). The amplitude and duration of 0°C water-induced Ca²⁺ spikes were various among single cells, and the average amplitude and duration were 218.54±53.67 and 19.27±1.98 s, respectively. In contrast, as shown in Figure 3C and 3D, mannitol-induced Ca²⁺ oscillations in most cells (Supplemental Movie 2) have an average frequency of 39.95±15.07 s. The average amplitude and duration of mannitol-induced Ca²⁺ spike were 60.44±21.66 and 17.28±1.54 s, respectively, which were lower than that of 0°C water-induced Ca²⁺ spikes. It appears that the amplitudes and frequencies varied while the durations were similar among cold and mannitol-induced Ca²⁺ spikes. Overall, there were great differences in cold and osmotic stress stimuli-induced single-cell Ca²⁺ dynamics. Our results suggest that Aequorin-based FAS-recorded Ca²⁺ signals in response to cold and osmotic stress stimuli reflect distinct cellular Ca²⁺ dynamics under the different stimuli. Synchronous Ca²⁺ spikes contributed to cold induced overall Ca²⁺ increase in the root, whereas osmotic stress stimulus induced overall Ca²⁺ changes in the root were attributed to asynchronous Ca²⁺ oscillations in individual root cells.

Figure 3.

Cellular Ca²⁺ Dynamics in Arabidopsis Root Cells.

(A) Selected time series of images of 0°C water treated root of 5-day-old CaFP300 seedling. Time points are indicated on the bottom of each panel.

(B) Enlarged image of root cells showing a Ca²⁺ response to 0°C water. The responsive cells are indicated by the numbers.

(C) Changes in fluorescence intensity of the single cells responding to 0°C water over the time course. Colored lines tagged by the numbers represent corresponding cells in the root as indicated in (B).

(D) Selected time series of images of 400mM mannitol treated root of 5-day-old CaFP300 seedlings. Time points are indicated on the bottom of each panel.

(E) Enlarged image of root cells showing a Ca²⁺ response to 400mM mannitol. The responsive cells are indicated by the numbers.

(F) Changes in fluorescence intensity of the single cells responding to 400mM mannitol over the time course. Colored lines represent corresponding cells in the root as indicated in (E).

FAS Recording for Monitoring Ca²⁺ Response of Arabidopsis Seedlings to Hormones, Pathogen Elicitor, and Amino Acids

Aiming at adapting the FAS system for a simple, highly sensitive, and robust approach for recording Ca²⁺ response to any given stimuli, we further evaluated the use of the FAS system in recording Ca²⁺ signals in response to a number of stimuli, including the plant hormones salicylic acid (SA), Jasmonic acid (JA), Abscisic acid (ABA), and brassinosteriod (BR), the pathogen elicitor flagellin, and the amino acids L-glutamate, L-glycine, L-cystine, L-alanine, L-serine, and L-aspargine. As shown in Figure 4, 180 µM SA and 450 µM JA, the lowest concentrations tested in this study, induced Ca²⁺ increases that could be visualized by FAS recording. The highest amplitudes of SA- and JA-induced Ca²⁺ signals occurred in the second or third 40 s, depending on the concentration of SA and JA. High concentrations of SA triggered earlier and higher Ca²⁺ response than low concentrations of SA, indicating that the initiation and amplitude, but not the duration, of SA-induced Ca²⁺ were closely related to SA concentration. Interestingly, it appeared that the roots were more sensitive to SA, which was evident by the initial strong and prolonged Ca²⁺ signals induced by SA in roots. In contrast, JA-induced Ca²⁺ signals were relative weak, and were detected mostly in leaves. No significant differences in the initiation and amplitude of Ca²⁺ responses were observed in response to low and high concentrations of JA. ABA- and BR-induced Ca²⁺ signals could not be detected in 7-day-old seedlings at the highest concentration of 20 µM ABA and 2 µM BR tested in this study (data not shown). However, prolonged Ca²⁺ increases were observed in the leaves of 14-day-old seedlings in response to 7 µM ABA and 1 µM BR. The highest amplitudes varied greatly among the leaves of 14-day-old seedlings in response to SA, JA, ABA, and BR, as shown in Supplemental Figure 2.

Figure 4.

Spatiotemporal Ca²⁺ Responses of Arabidopsis Seedlings to Plant Hormones and Pathogen Elicitor.

Sequential integrated luminescence images of FAS over the time course upon application of SA, JA, and flagellin. The type and concentration of the stimuli are indicated on the left of each panel, and time points are indicated on the top of each panel. The display range is 100–200 for all images.

Similarly, FAS recording revealed that flagellin induced prolonged Ca²⁺ increases that were mostly detected in the leaves, showing a tissue-specific sensing of the pathogen elicitor (Figure 4). The lowest concentration of flagellin used in this study was 0.125 µM, which still induced clearly detectable Ca²⁺ increases. Flagellin at 0.5 µM induced a Ca²⁺ signal with a higher amplitude and longer duration compared to that induced by 0.125 µM flagellin. Six amino acids have been considered to be agonists of GRL3.3-mediated Ca²⁺ influx (Qi et al., 2006). We visualized Ca²⁺ response to these six amino acids using FAS recording. As shown in Figure 5, all six amino acids induced strong Ca²⁺ responses during the first 40 s, and the amplitudes were correlated with the concentrations. The highest Ca²⁺ amplitude was observed for L-cystine, and the lowest one was for L-asparigine (Table1). The order of Ca²⁺ amplitude induced by each amino acid was very similar to that reported previously with the same concentration using a luminometer which is supposed to be more sensitive than imaging (Qi et al., 2006). The results suggest that our FAS Ca²⁺ recording is highly sensitive and reliable.

Figure 5.

Spatiotemporal Ca²⁺ Responses of Arabidopsis Seedlings to Amino Acids.

Integrated luminescence images of FAS at first and second 40 s upon application of amino acids. Amino acids are indicated in the first 40 s integrated images, and the concentration of amino acids and the time points are indicated on the top of each panel.

Table 1.

Ca²⁺ Response of Arabidopsis Seedlings to Amino Acids.

Amino acid	L-Glu		L-Gly		L-Cys
Time/conc.	10 mM	1 mM	10 mM	1 mM	10 mM	1 mM
40 s	157.59±16.61	120.67±6.32	161.21±2.54	127.34±4.32	177.09±11.07	141.44±5.32
80 s	93.78±3.41	94.45±2.56	93.71±2.14	93.59±1.02	99.99±1.15	94.76±2.14
120 s	92.79±2.11	94.13±1.34	92.93±2.33	93.06±0.59	93.64±1.05	93.62±1.12
160 s	92.65±1.05	94.31±0.89	92.83±1.09	93.13±0.89	92.28±0.75	92.68±0.98

Amino acid	L-Ala		L-Ser		L-Asn
Time/conc.	10 mM	1 mM	10 mM	1 mM	20 mM	10 mM
40 s	158.95±8.97	102.85±5.43	110.39±5.69	94.42±1.03	101.57±5.33	96.26±2.64
80 s	93.55±3.59	94.59±2.12	93.41±2.06	93.22±1.12	94.22±2.08	93.32±3.15
120 s	92.87±1.33	93.76±1.45	92.38±1.08	92.70±0.23	92.15±1.13	92.12±2.08
160 s	92.92±0.16	94.06±0.32	92.20±1.15	92.79±0.43	91.87±1.08	91.67±1.10

Note: values are the average photon intensity ± SD of three independent experiments.

FAS Recording Reveals Tissue-Specific Ca²⁺ Responses to Metal Ions

We chose two heavy metal ions, Cu²⁺ and Cd²⁺, as the other stress stimuli to examine the spatiotemporal patterns and sources of metal ion-triggered Ca²⁺ signals. It has been unclear whether these two metal ions can trigger Ca²⁺ signals in Arabidopsis. Only recently, it has been reported that Cu²⁺ ion could activate ER Ca²⁺ channels in the marine algae Ulva compressa to release Ca²⁺ from the ER (Gonzalez et al., 2010). Ruthenium Red (RR) and neomycin are membrane-permeable Ca²⁺ channel blockers that predominantly inhibit ryanodine receptors (RYRs)-like intracellular Ca²⁺ channels and Ca²⁺ release through IP3-regulated Ca²⁺ channels, respectively. Gadolinium (Ga³⁺) has been widely used as a plasma membrane Ca²⁺ channel blocker to inhibit Ca²⁺ influx from the extracellular space (Tracy et al., 2008). To examine the effect of blocking Ca²⁺ entry or release on potential Cu²⁺- and Cd²⁺-induced Ca²⁺ signals, we compared Ga²⁺, RR, and Neomycin-treated and -untreated FAS recording in response to Cu²⁺ and Cd²⁺. Images of Ca²⁺ channel blocker-untreated and -treated FAS were acquired simultaneously upon application of Cu²⁺ and Cd²⁺. As shown in Figure 6 and Supplemental Figure 3, both Cu²⁺ and Cd²⁺ induced strong and prolonged Ca²⁺ increases in leaves, whereas Ca²⁺ signals in roots were undetectable. Compared to Cu²⁺, Cd²⁺ initiated Ca²⁺ signal later but with a longer duration (Figure 6C and 6D). Application of Ga²⁺ reduced Cu²⁺ and Cd²⁺-induced Ca²⁺ amplitude by approximately 92% and 94%, respectively, indicating that Cu²⁺ and Cd²⁺ trigger Ca²⁺ increase mostly by activating Ca²⁺ influx through the plasma membrane. In the case of Cu²⁺, RR and neomycin reduced Ca²⁺ amplitude by 49% and 66%, respectively. However, RR almost abolished Ca²⁺ signals after 280 s (Figure 6A and Supplemental Figure 3A). Similarly, RR and neomycin also greatly inhibited Cd²⁺-induced Ca²⁺ signal, reducing the amplitude by 64% and 79%, respectively. We found that neomycin initially enhanced Cu²⁺ and Cd²⁺-induced Ca²⁺ signals at the first 40 s (Figure 6B and Supplemental Figure 3B), which is similar to observations reported previously (Tracy et al., 2008).

Figure 6.

Cu²⁺ and Cd²⁺-Induced Ca²⁺ Signals in the Leaves of Arabidopsis Seedlings.

(A, B) Effect of 100 µM RR and 600 µM neomycin on Cu²⁺-triggered Ca²⁺ increase. Time series of integrated luminescence images of FAS were collected with an interval of 40-s photon-counting integrations upon application of 10mM CuCl₂. Time points are indicated on each image panel. 100 µM RR (A) and 600 µM neomycin (B)-treated and -untreated FAS are indicated on the top of each panel.

(C, D) Temporal pattern of effects of RR, neomycin and Ga³⁺ on Cu²⁺ (C) and Cd²⁺ (D)-induced Ca²⁺ increases. Changes of the averaged photon intensity over times are indicated by the colored lines with open circles for RR, neomycin, and Ga³⁺ treatments, and the same colored lines with solid circles are corresponding controls treated with only CuCl₂ and CdCl₂.

Taken together, our results show that Cu²⁺ and Cd²⁺ induced strong Ca²⁺ signals specifically in leaves. The FAS-recorded spatiotemporal patterns of these two divalent metal ion-induced Ca²⁺ signals are different from the patterns induced by the other abiotic stress stimuli. It appears that Cu²⁺ and Cd²⁺ initiate Ca²⁺ signals by activating Ca²⁺ influx across the plasma membrane. Because internal Ca²⁺ channel blockers also significantly reduced Cu²⁺- and Cd²⁺-induced Ca²⁺ signals, Ca²⁺ release from internal stores may also play a role in Ca²⁺ signaling in Arabidopsis.

DISCUSSION

Genetically encoded Ca²⁺ indicators have been perfected to meet the needs of accurate measurement of Ca²⁺ dynamics in cells. Aequorin (McCombs and Palmer, 2008), an early version of Ca²⁺ indicator, contributed to the formulation of the concept of Ca²⁺ signature in plants (Knight et al., 1991). However, low photon production and less-than-robust measurement have hindered the use of Aequorin in high-resolution Ca²⁺ imaging and genetic screens in plants (Plieth, 2001). In this study, we revisited Aequorin and developed an Aequorin-based FAS luminescence Ca²⁺ recording system in Arabidopsis. Using the FAS recording system, we compared various stimuli, including abiotic and biotic stresses, hormones, and amino acids in their induction of Ca²⁺ responses in Arabidopsis seedlings. The FAS Ca²⁺ recording revealed tissue-specific Ca²⁺ signatures that are associated with the type and strength of stimuli. To our knowledge, such a systematic comparison, which is important for the concept of Aequorin-recorded Ca²⁺ signatures, has not been made in the Arabidopsis model plant previously. The optimized FAS recording system may also enable a high-throughput genetic screen for the identification of potentially novel Ca²⁺-signaling components in stress responses. Therefore, the FAS system can be a useful tool for studying Ca²⁺ signaling in Arabidopsis plants.

The non-invasive Case12-based Ca²⁺ recording revealed the complexity of cellular Ca²⁺ dynamics in individual living cells. The readout of FAS recording represents the averaged signals from cell populations, which masks the behavior of individual cells. In many cases, such as under osmotic, oxidative, and salt stresses, there are unsynchronized Ca²⁺ oscillations or transients in individual cells. This cellular heterogeneity in Ca²⁺ oscillations could be overcome by cold stress. Thus, FAS-recorded Ca²⁺ signal in cold stress-treated seedlings represent the collective behavior of synchronized Ca²⁺ oscillations in single cells. The observations imply that different cellular machinery in the cell is responsible for responding to different stimuli. Unsynchronized cellular Ca²⁺ dynamics could originate from cell-specific Ca²⁺-signaling toolkits that are regulated by the stimuli, and may also be attributed to cell–cell communication and the accessibility of individual cells to the stimuli. Mathematic simulations of oscillatory behaviors of cell populations suggested that Aequorin-recorded Ca²⁺ increases results from in-phase oscillations whereas the Ca²⁺ decrease is due to the out-phase oscillation of cell populations (Dodd et al., 2006; Plieth, 2010; Batistic and Kudla, 2012). However, our results suggest that such models of Ca²⁺ signaling may not be applicable to diverse stimulus-specific Ca²⁺ dynamics. In addition, experiments with Ca²⁺ channel blockers suggested that Aequorin-recorded initial Ca²⁺ increase is attributed to single-cell Ca²⁺ amplitudes in which both in-phase or out-phase oscillations might occur. In the present study, we could not correlate the Case12 recording of cellular Ca²⁺ dynamics with FAS recording of tissue-level Ca²⁺ dynamics in response to a given stimulus due to a lack of mutants that express both Ca²⁺ indicators. Nevertheless, we observed the effects of Ca²⁺ channel blockers on Ca²⁺ signals at the tissue level, suggesting that FAS-recorded signals are relevant to Ca²⁺ dynamic changes at the cellular level.

Aequorin-recorded Ca²⁺ dynamics to a great extent depends on not only the sensitivity of the assay system, but also the state of the plants. For the FAS system, 7-day-old seedlings are ideal for detecting Ca²⁺ responses of both roots and leaves to most stimuli tested in this study. However, for some stimuli that are known to trigger Ca²⁺ responses in the leaves, such as the plant hormones ABA and BR, the pathogen elicitor flagellin and heavy metal ions, characterization of Ca²⁺ responses or screening for the mutants with altered Ca²⁺ responses would ideally be conducted in older seedlings. The high sensitivity and robustness are also a prerequisite for high-throughput screens. In particular, the high sensitivity of the system allows the use of relatively mild stress stimuli for triggering Ca²⁺ responses. Severe stress treatments likely cause damage to the cells, leading to more artifacts. For all stimuli tested in this study, the FAS system was able to image Ca²⁺ responses to the stimuli applied at levels used for Ca²⁺ reading by luminometers (Knight et al., 1997; Kawano et al., 1998; Tracy et al., 2008; Munemasa et al., 2011; Pan et al., 2012; Ranf et al., 2012), which normally is more sensitive than Ca²⁺ luminescence imaging. Therefore, the FAS system has advantages over other systems reported previously not only for its highly sensitive Ca²⁺ recording, but also for its detection of spatial Ca²⁺ responses.

In conclusion, the Aequorin-based FAS system developed in this study provides a powerful tool for studying Ca²⁺ signaling in response to various stimuli. Importantly, the FAS system offers a platform for future identification of Ca²⁺-signaling components in stress-signaling pathways. In combination with other techniques, the FAS system would become a very useful tool for studying the specificity and crosstalk in plant-signaling networks.

METHODS

Plant Material and Growth Conditions

The Arabidopsis thaliana Col-0-expressing cytosolic Aequorin (referred to as Aeq wild-type) was obtained from Dr Marc R. Knight (Knight et al., 1991). Seedlings were grown in the growth chamber conditioned with a standard light–dark cycle of 16 h of light and 8 h of dark at 23°C. Ca²⁺ indicator Case12 (Souslova et al., 2007) constructed in a binary vector was introduced into Arabidopsis ecotype Col-0 by Agrobacterium-mediated transformation. An independent transgenic line, CaFP300, expressing a high level of Case12 was used in this study.

Establishment of Aequorin-Based FAS Luminescence Recording System

For establishing FAS system, Aeq wild-type seeds were planted in a square plate (10 × 10 cm square Petri dish with grid, Fisher Scientific, USA) containing full- strength MS medium, 1% sucrose, and 1.2% agar. Aeq seedlings were grown vertically on the plate at a density of about 100 seedlings per plate and transferred onto a clear adhesive film (Adhesive PCR Plate Seals, Thermo Scientific, USA) on the seventh day after germination. A sheet of FAS (about 100 seedlings) was incubated in the plate containing 15 ml water and 30 µg h-CTZ (coelenterazine, NanoLight Technolgies, Arizona, USA) for 5 h overnight for reconstitution of Aequorin.

Aequorin Luminescence Imaging and Analysis

A sheet of FAS was placed on the stage of the luminescence imaging system (Princeton Instruments, New Jersey, USA) and images were acquired immediately upon application of stimuli. The time gap between applying stimuli and recording image was about 2–4 s. For EMS screening, two integrated images were collected at first and second 40 s. For other experiments, the time series of Ca²⁺ recording was as indicated in the figures or figure legends. Qualitative analysis of luminescence images was performed using ImageJ. An average photon intensity of stimulus-induced luminescence signals at any given time was defined as the total photon intensity of luminescence signals emitted by FAS divided by the area of ROI. Plant hormones, SA, JA, and ABA were dissolved in 100% ethanol, whereas eBL was dissolved in DMSO. Therefore, a negative control of ethanol or DMSO in the same final concentrations as in the working solution of hormones was included in hormone-induced Ca²⁺ response. The stocks of 1 mM flagellin and 100 mM amino acid were dissolved in water and pHs of the working solutions of these two stimuli were unadjusted. For pretreatments with Ca²⁺ channel blockers, the FAS sheets were cut and divided equally into two parts. Half of the FAS sheets were pretreated with 50 µM GaCl₃, 100 µM RR, or 600 µM neomycin in h-CTZ incubation solution for 20 min, respectively, while other half of the FAS sheets remained in the h-CTZ incubation solution without adding blockers. Treated and untreated FAS sheets were the assembled into one piece and imaged for 400 s with a 40-s interval of integration upon application of 10 mM CuCl₂ or 10 mM CdCl₂. For cam mutant analysis, Aeq wild-type and cam mutant seeds were planted on MS agar in the same plate with a similar density, occupying the left or the right half of the plate. Seven-day-old Aeq and cam seedlings were transferred onto an adhesive film and incubated with h-CTZ, and images were acquired for 160 s as described above. Stimuli-induced Ca²⁺ amplitudes were compared at the time point in which the highest Ca²⁺ amplitudes were observed. The highest Ca²⁺ amplitudes were normalized by total photon intensity of luminescence signals emitted from discharged FAS in the solution of 1 M CaCl₂ and 20% ethanol.

Confocal Ca²⁺ Imaging

Five-day-old seedlings of CaFP300 were placed in a customized chamber, which was created by placing a small piece of clay between a coverslip and a slide. A polyethylene tube (0.58 mm diameter, PerkinElmer, USA) was connected to a 1-ml syringe and the other end of the tube was fixed on the slide next to the chamber. Stimulus solution was slowly injected into the chamber just prior to acquiring images. For cold stimulus, CaFP300 seeds were directly planted in a chambered cover glass (Nalge Nunc International, USA) containing a thin layer of MS medium, and grown vertically for 5 d. 0°C water was added into the chamber just prior to acquiring images. Ca²⁺ imaging was performed using an inverted Nikon A1R confocal laser-scanning microscope with a 20× water immersion lens (numerical aperture 0.75). Time series images were collected at an interval of 4 s with excitation and emission wavelengths of 488 nm and 500–550 nm at a pixel resolution of 512 × 512. Data were exported and processed in DataGraph (Visual Data Tools Inc., Chapel Hill, North Carolina, USA).

SUPPLEMENTARY DATA

Supplementary Data are available at Molecular Plant Online.

FUNDING

This work was supported by National Institutes of Health grant R01GM059138 to J.-K.Z.