Evidence for the Involvement of Electrical, Calcium and ROS Signaling in the Systemic Regulation of Non-Photochemical Quenching and Photosynthesis

In contrast to the function of reactive oxygen species, calcium, hormones and small RNAs in systemic signaling, systemic electrical signaling in plants is poorly studied and understood. Pulse amplitude-modulated Chl fluorescence imaging and surface electrical potential measurements accompanied by pharmacological treatments were employed to study stimuli-induced electrical signals in leaves from a broad range of plant species and in Arabidopsis thaliana mutants. Here we report that rapid electrical signals in response to a local heat stimulus regulate systemic changes in non-photochemical quenching (NPQ) and PSII quantum efficiency. Both stimuli-induced systemic changes in NPQ and photosynthetic capacity as well as electrical signaling depended on calcium channel activity. Use of an Arabidopsis respiratory burst oxidase homolog D (RBOHD) mutant (rbohD) as well as an RBOH inhibitor further suggested a cross-talk between ROS and electrical signaling. Our results suggest that higher plants evolved a complex rapid long-distance calcium-dependent electrical systemic signaling in response to local stimuli that regulates and optimizes the balance between PSII quantum efficiency and excess energy dissipation in the form of heat by means of NPQ.

Introduction

Being constantly tied to one habitat, plants developed complex physiological and molecular mechanisms that enable them to communicate efficiently between their different leaves and organs. Systemic signaling pathways such as systemic acquired resistance (SAR) and systemic acquired acclimation (SAA) (Métraux et al. 1990, Jabs et al. 1996, Karpinski et al. 1999, Mühlenbock et al. 2008, Szechynska-Hebda et al. 2010) have been intensively studied, with many reports supporting a substantial role for reactive oxygen species (ROS), hormones and, recently, small RNA molecules in long-distance signaling. Nevertheless, the role of electric signals has in recent years been neglected. Electric signals in plants were discovered in the 19th century in Dionaea muscipula (Burdon-Sanderson 1872). Subsequent research revealed that in contrast to animals, plants evolved only a few types of electrical communication, which have distinct characteristics. At least three types of electrical signals have been described to date: action potential (AP), variation potential (VP) and system potential (SP) (Zimmermann et al. 2009).

An AP consists of depolarization, repolarization and hyperpolarization phases, and requires a stimulus strong enough to reach a particular threshold. The mechanism of AP generation relies on ion transport across the membranes, particularly Ca²⁺, K⁺ and Cl^–, and it is the fastest known form of electrical communication in plants. It is propagated, within a few seconds, over a long distance and is responsible, for example, for the immediate reaction of traps in carnivorous plants. The preferred potential medium for AP signals are sieve elements, as they ensure continuous, good communication from cell to cell via plasmodesmata (van Bel and Ehlers 2005). AP generation is usually associated with non-damaging stimuli, which can ultimately affect phloem transport, gas exchange or gene expression (Fromm and Bauer 1994, Fromm and Lautner 2007).

A VP is a slow propagating type of signal, which is generated upon an injurious stress treatment. The membrane depolarization and repolarization cycle takes several minutes. In contrast to AP, VP amplitude positively correlates with the stimulus strength. VP propagation is correlated with changes in the hydraulic pressure mainly in xylem vessels, and fades away with distance from the point of origin (Malone 1992, Stankovic et al. 1997, Stahlberg et al. 2005). The underlying mechanism of VP employs mainly perturbations of H⁺-ATPase activity (Stahlberg et al. 2006). VP affects hormone emission and gene expression (Wildon et al. 1992, Dziubinska et al. 2003).

Unlike the initial depolarization that accompanies the generation of AP and VP, SP primary polarity is reversed (Zimmermann et al. 2009). SP can be evoked by wounding as well as heat stimulation (scorching), and its induction and spread depend mainly on cations (Zimmermann and Mithöfer 2013). It is noteworthy that a close relationship between SP propagation and NADPH oxidase was recently reported, as plants devoid of functional NADPH oxidase (rbohD) had a suppressed capability to mediate SP (Miller et al. 2009, Suzuki et al. 2013). This finding suggested a close link and cross-talk between ROS and this type of electrical signaling.

There are two main approaches for measuring plant electrical signaling, which deliver different information. The intracellular method uses a glass microelectrode, filled with electrolyte. The microelectrode is carefully inserted into a living cell and can be used for the measurement of absolute values of cell voltage potential (Krol et al. 2004, Szechynska-Hebda et al. 2010, Sukhov et al. 2014). In contrast to intracellular measurements, the extracellular approach delivers only relative values of electrical potentials. Extracellular measurements include invasively inserting the electrode directly into the plant material, usually employed with woody plants (Ríos-Rojas et al. 2014), or non-invasively placing the electrode on the surface of the plant with an intermediate ensuring good electrical contact (Mousavi et al. 2013, Suzuki et al. 2013). The second method is widely used, as it eliminates wounding and is relatively easy.

Systemic changes in Chl fluorescence parameters were reported for the first time during the discovery of SAA (Karpiński et al. 1999). However, the cause and the effects of these changes are poorly understood. For example, it is unknown whether the transient systemic suppression of photosynthesis upon local stimulation is just a symptom of stress in plants, or whether it carries any information to the unstressed parts of the plant, alerting them to an upcoming danger. An association between electrical signals induced by heat stimulation of a leaf and transient photosynthesis changes in mimosa, poplar and tobacco were previously reported (Koziolek et al. 2004, Lautner et al. 2005, Hlavácková et al. 2006). Although these affected leaf gas exchange and PSII functioning, little is known about the mechanism that links these two phenomena.

Here we present evidence for a physiological and molecular link between electric signaling, calcium channel activity, ROS and photosynthetic parameters, and propose their reciprocal dependence.

Results and Discussion

We assessed values of PSII quantum efficiency [Y(II)], non-photochemical quenching (NPQ) and photochemical quenching (qP) upon a local and punctual, damaging heat stimulus. Y(II) values decreased in a concentric manner, while NPQ and qP values increased starting from the site of treatment. Changes in these parameters were initially observed in veinal tissues of leaves and then spread through the mid-veinal regions (Fig. 1; Supplementary Video S1).

Fig. 1

Changes in Chl a fluorescence of a dandelion leaf following heat stimulation. (a) Spatiotemporal changes of NPQ, Y(II) and qP assessed by Chl fluorescence imaging. The arrow indicates the area stimulated for 1 s with a flame-heated metal wire. Time is shown in min:s format (heat stimulation was performed at 0:00). Scale bar = 1 cm. The false color scale represents values of assessed parameters. (b) NPQ, (c) Y(II) and (d) qP changes in stimulated (Heat stress) and non-stimulated (Control) leaves measured by Chl fluorescence imaging at representative areas [shown in (a) as a black circle] approximately 1 cm from the stimulated region or the region touched with the unheated wire in the control measurement, respectively. Representative data of five independent experiments are shown. Persistence of (e) NPQ and (f) Y(II) changes caused by a heat stress. Additional Chl fluorescence measurements were performed, and NPQ and Y(II) values were recorded up to 55 min after stimulation. Values are means ± SEM (n = 5). Asterisks indicate a significant difference in comparison with the control (**P < 0.01, ***P < 0.001; P-values calculated using Tukey’s test).

Proper functioning of PSII, reflected in normal Chl a fluorescence parameters, depends on accurate electron transport via the photosynthetic electron transport (PET) chain. The electron transport chain is in turn susceptible to temporary perturbations, and at the same time has a broad range of plasticity to ensure efficient operation of further photosynthetic processes. Due to the many carriers involved, there are several different factors which can truncate PET (for reviews, see DeEll and Toivonen 2003, Foyer et al. 2012).

To test whether electrical signaling accompanied the changes in photosynthetic parameters, surface potential (Mousavi et al. 2013, Suzuki et al. 2013) and photosynthesis changes were monitored in dandelion (Figs. 1, 2). The results showed that following the initial hyperpolarization recorded, repolarization occurred at different time points with respect to the proximal and distal electrodes (in reference to the site of stress; while the electrodes were on the same leaf; Fig. 2a). Potential changes recorded by electrodes placed at different distances from the site of stress application indicated that the electrical signal spread within a few minutes, and that the hyperpolarization, and then repolarization was delayed on the distal electrode. This, together with a reduced amplitude of hyperpolarization detected by the distal electrode in comparison with that recorded on the proximal electrode (Fig. 2a), suggested that the tracked electrical events are SPs. Similar measurements were performed with Arabidopsis. Despite no indication, or weak insignificant changes of fluorescence disturbance upon local treatment, surface potential changes were clearly detectable with the same trends as in dandelion (Fig. 2c).

Fig. 2

Leaf surface electrical potential changes following heat stimulation of the tip of the leaf. (a and b) Effect of lanthanum chloride (LaCl₃) on electrical signal propagation. Representative potentials recorded with two electrodes placed on the same leaf irritated by heat: the proximal electrode on the area not treated directly (Non-treated) and the distal electrode on the area directly treated by solution: (a) without inhibitor (Mock) or (b) with LaCl₃. (c) Comparison of electrical signals in Arabidopsis and dandelion. Representative potentials recorded on Arabidopsis and dandelion leaves. (d) Effect of Arabidopsis RBOHD gene mutation on electrical signal propagation. Representative potentials recorded on rbohD plants vs. the WT. The arrows indicate the moment of heat stimulation of the tip of the leaf for 1 s by the flame of a lighter. Statistical significance of observed changes is shown in Supplementary Fig. S1.

A link between surface potential changes and photosynthesis was studied using LaCl_3, diphenyleneiodonium chloride (DPI) and DCMU. LaCl₃ is a widely known Ca²⁺ channel blocker, and is commonly used in electrophysiology (Sharma et al. 1992, Friedman et al. 1998). Moreover the ROS-generating activity of respiratory burst oxidase homolog D (RBOHD) is regulated by Ca²⁺ ions directly (Takeda et al. 2008, Kimura et al. 2012) as well as via post-translation modification by kinase activity (Drerup et al. 2013, Li et al. 2014, Monaghan et al. 2014). Our results clearly show that regulation of PSII operation in response to the focused heat stimuli depended on calcium channel activity, as LaCl₃ application to the leaf led to a significant delay in systemic changes of Chl a fluorescence (Fig. 3; Supplementary Fig. S1a–c), that was accompanied by surface potential delay and amplitude reduction (Fig. 2a, b; Supplementary Fig. S1d).

Fig. 3

Changes in Chl a fluorescence of a dandelion leaf treated with a calcium channel blocker (LaCl₃) following heat stimulation. (a) Spatiotemporal changes of NPQ, Y(II) and qP assessed by Chl fluorescence imaging. Arrow indicates the area stimulated for 1 s with a flame-heated metal wire. The black rectangle shows the area treated with LaCl₃. The time is shown in min:s format (heat stimulation was performed at 0:00). Scale bar = 1 cm. The false color scale represents values of assessed parameters. (b) NPQ, (c) Y(II) and (d) qP changes measured by Chl fluorescence imaging at representative areas [shown in (a) as black circles] of treated (Area 2 represented by circle 2) and untreated (Area 1 represented by circle 1) parts of the leaf at a similar distance from the stimulated region. Representative data of five independent experiments are shown. The statistical significance of the observed changes is shown in Supplementary Fig S1.

To confirm that the surface potentials measured in our experiments were SPs, rbohD mutants were exposed to a local heat stimulus. Similar to the findings of Miller et al. (2009) and Suzuki et al. (2013), electrical signals in rbohD mutants were significantly less pronounced than in the wild type (WT), but definitely detectable (Fig. 2d; Supplementary Fig. S1e). These observations indicated that the stimulus applied in our work generated a similar type of electrical signal associated with RbohD function. In order to examine the RbohD function in dandelion, we used the flavin oxidase inhibitor DPI. The application of DPI blocked the spreading of Chl fluorescence changes following heat stimulation (Fig. 4); however; it did not stop electrical signal spreading in the leaf (Supplementary Fig. S1d). These results suggest that extracellular propagation of the systemic electrical signaling depends on plasma membrane calcium channels, while intracellular retransmission of these signals into PSII reaction centers may depend on ROS signaling. Such an interpretation is the simplest one to explain cross-talk between electrical and ROS systemic signaling and is in agreement with a recently published integrated model of the ROS, calcium and electrical wave-like systemic signaling presented by Gilroy et al. (2016).

Fig. 4

Changes in Chl fluorescence of a dandelion leaf treated with a RBOHD inhibitor (DPI) following heat stimulation. (a) Spatiotemporal changes of NPQ, Y(II) and qP assessed by Chl fluorescence imaging. The arrow indicates the area stimulated for 1 s with a flame-heated metal wire. The black rectangle shows the area treated with DPI. Time is shown in min:s format (heat stimulation was performed at 0:00). Scale bar = 1 cm. The false color scale represents values of assessed parameters. (b) NPQ, (c) Y(II) and (d) qP changes measured by Chl fluorescence imaging at representative areas [shown in (a) as black circles] of treated (Area 2 represented by circle 2) and non-treated (Area 1 represented by circle 1) parts of the leaf at a similar distance from the stimulated region. Representative data of five independent experiments are shown.

In a previous work, we showed that partial illumination of leaves with excess light led to the generation of intercellular electrical potential changes from the exposed leaf to systemic shaded leaves (Szechynska-Hebda et al. 2010). To exclude any contribution of PET activity to the measured surface potential, we employed a pharmacological treatment with DCMU, a known and specific PET inhibitor. DCMU binds to the quinone B- (Q_B) binding pocket and blocks the oxidation of quinone A (Q_A) and subsequent reduction of plastoquinone (PQ). It results in a disturbance in F_m and F₀, which leads to strong reduction of NPQ (Escoubas et al. 1995, Kulasek et al. 2016). In our experiments, DCMU application also resulted in a strong decrease of NPQ in the leaf area exposed to this inhibitor; however, it did not disturb spreading of Chl fluorescence changes across the treated region (Fig. 5). This result suggests that the specific inhibition of PET at the Q_B side does not inhibit propagation of calcium- and RbohD-dependent systemic electrical and ROS signaling from local to systemic PSII reactions centers.

Fig. 5

Changes of the Chl fluorescence in dandelion leaf treated with a photosynthetic electron transport inhibitor (DCMU). (a) Spatiotemporal changes of NPQ, Y(II) and qP assessed by Chl fluorescence imaging. The arrow indicates the area stimulated for 1 s with a flame-heated metal wire. The black rectangle shows the area treated with DCMU. Time is shown in min:s format (heat stimulation was performed at 0:00). Scale bar = 1 cm. The false color scale represents values of assessed parameters. (b) NPQ, (c) Y(II) and (d) qP changes measured by Chl fluorescence imaging at representative areas shown in (a) as a black circles (Area 1 and Area 2 represented by circles 1 and 2, respectively) on the two sides of the region treated with DCMU. Representative data of five independent experiments are shown.

In an attempt to examine how widespread the link is between electrical signaling and changes in photosynthetic parameters, we measured it in a variety of different plants. In these studies, an effort was made to check many species from different families for the response of photosynthesis to the stimulus mediated by electrical signals. Using the same experimental design, photosynthetic parameters were evaluated in plants from natural and laboratory conditions exposed to local heat stress. The velocity and area of the plants’ reaction varied in different species. Clear similar effects were observed in Aesculus hippocastanum, Hieracium pilosella, Plantago major, Populus maximowiczii, Rumex acetosa, Syringa vulgaris, Taraxacum officinale, Tilia cordata and Tilia platyphyllos (Supplementary Fig. S2). In contrast, there was weak or no detectable reaction in A. thaliana, Capsicum annuum, Cucurbita pepo, Ginkgo biloba, Nicotiana benthamiana and Nicotiana tabacum (Supplementary Fig. S2). Using the same experimental design for different plant species from evolutionarily distant families, we revealed that there are substantial differences between various species of higher plants in PSII sensitivity or in response to the local stimulation that triggered systemic electric, calcium and ROS signaling that regulates NPQ changes in non-directly stressed PSII reaction centers. We further examined whether systemic fluorescence changes are characteristic for plants grown in natural, variable conditions.

We found that Arabidopsis and tobacco did not exhibit spreading of transient photosynthesis signals when grown in the field (data not shown). In contrast, dandelion retained the ability to mediate systemic changes of fluorescence parameters upon local stimulus even when grown in a controlled, laboratory environment (data not shown). These results may suggest that plants such as Arabidopsis and tobacco lost their ability to mediate systemic changes in photosynthesis as a result of many years of cultivation of subsequent generations in non-natural, controlled conditions. It is also possible that differences in leaf architecture (i.e. thickness, cuticle layer, presence of trichomes, etc.), photosynthetic apparatus composition and chloroplast morphology could have led to the observed differences in systemic signaling while responding to a local stimulus or affected the surface potential recordings. In addition, it is possible that the methods used in our study were not sufficient to detect significant responses of insensitive plants.

Speculations on the mechanism of calcium influence on photosynthesis are rather difficult. For example, it is hard to explain how the disturbance of NPQ occurs, because the precise mechanism of quenching is still under discussion (Duffy and Ruban 2015). The agricultural significance of these processes is extremely important, as recent studies clearly showed that adequate manipulations of the quenching machinery may lead to a substantial increase in biomass production (Kromdijk et al. 2016).

The nature of electrical signaling in plants is very complex, and its role remains poorly understood. It seems to interact with some of the other main players in rapid signaling such as ROS and calcium waves (Gilroy et al. 2016). Its involvement in SAA and SAR and impact on photosynthesis and systemic communication between photosystems of remote chloroplasts and cells strongly suggests that proper electrical communication as well as cross-talk with ROS signaling is essential for plant survival under conditions of natural multivariable stresses and stimuli. Our research shows that at the molecular level calcium ion channels directly link stimuli-induced transient changes in photosynthesis with electrical signaling and these findings are in agreement with the recently presented model (Gilroy et al. 2016).

Materials and Methods

Plant material and growth conditions

Arabidopsis thaliana Col-0 and the rbohD mutant (Torres et al. 2002) were grown on Jiffy pots (Jiffy products) in a growth chamber (23°C, 8 h light/16 h dark with a light intensity of 100 ± 15 µmol photons m^–2 s^–1). Nicotiana benthamiana, N. tabacum and T. officinale were grown on soil mixed with perlite (2 : 1) in a growth chamber (23°C, 16 h light/8 h dark with a light intensity of 100 ± 15 µmol photons m^–2 s^–1). Plants at 6–8 weeks old were used for experiments. Whole, fully developed and undamaged detached leaves of A. hippocastanum, C. annum, C. pepo, G. biloba, H. pilosella, P. major, P. maximowiczii, R. acetosa, S. vulgaris, T. officinale, T. cordata and T. platyphyllos growing in their native environment or field conditions (collected from June to August) were used for Chl a fluorescence imaging. After detachment, the petiole of the leaf was put in tap water for 30 min to 1 h before the beginning of the experiment.

Chl a fluorescence imaging

Spatiotemporal Chl a fluorescence was measured using an imaging Chl fluorometer (IMAGING-PAM MINI, Walz). Imaged leaf area was 32 mm × 24 mm. The plants were kept in darkness for 30 min, then blue (470 nm) actinic light (60 µmol photons m^–2 s^–1) was switched on for 12 min. To measure F_m and F_m′, saturating pulses were applied (6,000 µmol photons m^–2 s^–1, duration 800 ms). Non-photochemical quenching, NPQ = (F_m – F_m′)/F_m′, effective quantum yield of PSII, Y(II) = (F_m′ – F)/F_m′ and photochemical quenching, q_P = (F_m′ – F)/(F_m′ – F₀′) were determined according to the manufacturer’s instructions and as described previously (Baker 2008, Gawroński et al. 2014). Changes in these parameters during the experiment were monitored by applying saturating pulses at 10 s intervals. Four minutes following the beginning of the measurement the leaf was stimulated by touching for about 1 s with a 1 mm thick metal wire heated with the flame of a lighter for approximately 3 s. For control measurement, an unheated metal wire was used. To determine the transience of the fluorescence changes after heat stimulation, the same measurement procedure was carried out twice after stimulation, each preceded by 15 min of dark adaptation.

Pharmacological treatments

LaCl₃ (1 mM; Acros), 500 µM DPI (DPI, Sigma Aldrich) or 500 µM DCMU (Sigma Aldrich) solutions were made in deionized water and contained in addition 0.2% (v/v) dimethyl sulfoxide (DMSO; Sigma Aldrich) and 0.1% (v/v) Tween-20 (Sigma Aldrich). Half of the surface of the leaf was submerged in LaCl₃ solution for 18 h prior to the experiment. DPI and DCMU were sprayed on the leaf surface 30 min prior to the experiment. DPI was used on half of the surface of the leaf and DCMU was used on a smaller area in the center of the leaf. For control experiments, a mock solution that contained only DMSO and Tween-20 was used. The concentrations of agents were chosen based on the literature and adjusted to our particular experimental design.

Surface potential recordings.

Surface potential was measured using a glass microelectrode filled with 1 M KCl connected to the leaf through a drop (10 µl) of 1 M KCl in 1% (w/v) agar placed 25 mm from the tip of the leaf, so that the electrode did not damage the cuticle. The ground electrode (Ag/AgCl) was placed in the soil. An Axoclamp 900A amplifier (Molecular Devices) was used to record the surface potential. Experiments were conducted in a Faraday cage in room temperature (22–25°C). A few minutes after connection of the electrodes to the leaf, the tip of the leaf was heat stimulated by the flame of the lighter for 1 s. The experimental set-up for electrophysiological measurements made it very difficult to touch the leaf during electrical potential recordings using the metal wire, avoiding any disturbance during assessment of the the parameters. Additionally, an ImagingPAM device was used for Chl fluorescence measurements which hampered the access of microelectrodes to the leaf.

Statistics and data analysis

All data analysis and statistics were performed using R (R Core Team, 2015). For data visualization, the R package ‘ggplot2’ was used (Wickham 2009).

Supplementary data

Supplementary data are available at PCP online.

Funding

This study was supported by funding from the National Science Centre project [UMO-2012/07/B/NZ3/00228 and UMO-2014/14/A/NZ1/00218 to S.K., M.B. and M.G.]; the POKL.04.03.00-00-042/12-00 programme co-financed by the European Social Fund [to M.B.]; the USA National Science Foundation [IOS-1353886, IOS-0639964 and IOS-0743954 to R.M.] and the University of North Texas, College of Arts [to R.M.]. The funders had no role in the design, data collection, analysis, decision to publish, or preparation of the manuscript.

Disclosures

The authors have no conflicts of interest to declare.

Abbreviations

Abbreviations
 
• AP

  action potential

 
• DMSO

  dimethyl sulfoxide

 
• DPI

  diphenyleneiodonium chloride

 
• NPQ

  non-photochemical quenching

 
• PAM

  pulse amplitude-modulated

 
• PET

  photosynthetic electron transport

 
• PQ

  plastoquinone

 
• Q_A

  quinone A

 
• Q_B

  quinone B

 
• qP

  photochemical quenching

 
• RBOHD

  respiratory burst oxidase homolog D

 
• ROS

  reactive oxygen species

 
• SAA

  systemic acquired acclimation

 
• SAR

  systemic acquired resistance

 
• SP

  system potential

 
• VP

  variation potential

 
• WT

  wild type

 
• Y(II)

  PSII quantum efficiency