Plant electrome: the electrical dimension of plant life

  • Gabriel R. A. de ToledoEmail author
  • André G. Parise
  • Francine Z. Simmi
  • Adrya V. L. Costa
  • Luiz G. S. Senko
  • Marc-Williams Debono
  • Gustavo M. Souza


The sessile lifestyle of plants exposes them to many challenges that must be overcome to ensure their survival. Despite their apparent lack of movement and behaviour, plants are very active organisms that constantly respond to external cues and internal signals, making fine tunings at the metabolic, cellular and tissue levels. To accomplish this, plants largely rely on their relationship between phenotypic plasticity and cognitive abilities, which requires an information (or signalling) system. This information system is composed of many different signals, and among them are the electrical ones. Besides, in plants, there are many electrical activities that are not necessarily considered electrical signals. However, the sum of all electrical activity of a plant, considering all levels of organization, engenders the electrical dimension of their life. This dimension of life that encompasses all the living beings has recently been named “electrome”, in analogy with the other ‘omics’ from system biology approach. In the present review, we will explore this concept to describe the plant electrical activities. Furthermore, some basic aspects that are underlying the plant electrome will be presented, including the principles of biophysics that sustain the electrical activity of the cells. Moreover, some examples of electrical activities and signals that are vital to the plants will be exposed. Then, our most relevant and recent breakthroughs about plant electrome dynamics under different environmental conditions will be presented. Finally, some considerations about the possible involvement of the electrome plasticity and cognition features associated with the adaptive behaviour of plants will be discussed.


Plant electrome Electrical activity Electrical signalling Complexity Ecophysiology Proto-neural systems Cognition 

1 Introduction

Plants, like any other biological organism, can be considered as electric systems. It means that they work like living circuits and need electrical activity for their proper functioning (Volkov et al. 2009; Stavrinidou et al. 2015). Part of this electrical activity is called electrical signalling. It is characterised by ionic imbalances generated and propagated with the purpose of signalling, with the aim of informing something to other cells or parts (near or distant) of the plant’s body (Baluška and Mancuso 2013a; Vodeneev et al. 2016a, b). All the plants’ electrical activity dynamics, including electrical signalling, can be considered as the electrical dimension of plant life. This fundamental dimension of life, in the context of the omics sciences, can be named “electrome” (de Loof 2016; Souza et al. 2017).

According to de Loof (2016), electrome “stands for the totality of all ionic currents of any living entity, from the cellular to the organismal level” which is vital for life and its own definition, because the “death of any cell ensues at the very moment that it irreversibly (excluding regeneration) loses its ability to realize its electrical dimension”.

The self-generated electrical dimension of cells is the consequence of the movement of electrically charged species, and the electrical fields generated by these charge flows. In plant cells, among many other eukaryotic cells, the ionic flows are predominantly constituted by ions such as H+, K+, Cl, Na+, and Ca2+, that are translocated from one side to the other of the membranes. However, other ions such as Fe3+, Al3+, Mg2+, Zn2+, Mn2+, and Hg2+ might contribute to ionic currents in plant cells and tissues in less extensive ways (Higinbotham 1973; Nobel 1999; Volkov 2006).

Moreover, some others mobile charged species such as nutrients (e.g. NH4+, NO3, H2PO4, or SO42−), amino acids (lysine, glutamate, aspartate and arginine), organic acids and small proteins, can also be included as part of the electrical dimension of plant life since they are electrically charged (Higinbotham 1973; Baluška and Mancuso 2013a; Volkov 2006). Moreover, inner or external immobile charged species of the plant cells (e.g. DNA, RNA, cytoskeleton and cell wall filaments), interacting with mobile ions and charged mobile molecules, are other source of electrical activity of the electrome (Nobel 1999). The electron flow of the electron transport chain of chloroplasts and mitochondria, and its proton gradients, also must be accounted as components of the plant electrome (Nobel 1999; Volkov 2006; Stavrinidou et al. 2015; Taiz et al. 2014).

Although all these electrical activities are relevant for the electrome, the major part of the plant electrome dynamics is reflex of few ionic species (H+, K+, Na+, Cl+ and Ca2+) which cross the membranes mainly by channels and electrogenic pumps (Nobel 1999; Volkov 2006; Baluška and Mancuso 2013a; de Loof 2016). Hence, our main focus in the present review will be the ionic flows on the membranes that participate in the electrical signalling. However, firstly, we will offer a brief presentation of the general aspects of the electrical dimension of life, including its general biophysical properties. We will also give some examples of electrical activities that compose the electromic dynamics, focusing on electrical signalling and its physiological importance. Lastly, we will present our findings on plant electrome studies and discuss about future perspectives accounting for possible interrelations between plant electrome and plant cognition. For a brief description of plant electrical data collecting and usual methods employed in plant electrophysiology, see Online Appendix 1.

2 Underlying the electrome: biophysics and physiology

In plants, there is a great amount and diversity of polymers and other macromolecules that are immobile and electrically charged, both inside and outside the cells. These immobile charged molecules are generally surrounded by an aqueous phase containing small mobile ions that generates an electrical potential between them, which is known as Donnan potential. Using the plant cell wall as an example, the macromolecules rich in carboxyl groups, like pectin, tend to dissociate their H+, rendering cell walls as a negative charged system that attracts other cations, commonly Ca2+. The attraction of these positively charged species can increase the concentration of solutes. Consequently, it generates a higher osmotic pressure near the pectin of the cell wall at equilibrium. Donnan potential can also be regarded as diffusion potential because mobile ions tend to diffuse away from the immobile charges, generating little charge separations that create an electrical potential gradient. Hence, Donnan potential can be considered as both osmotic and diffusion potentials (Nobel 1999).

The region containing immobile charges, which electrostatically attract mobile ions, is generally referred to as Donnan phase. Electrostatic interactions between two charged particles vary inversely as the square of the distance between them. Thus, the electrical force greatly increases as the ions get closer. In the cytoplasm the immobile charged species are most represented by proteins and large polymers including RNA, DNA, and cytoskeleton filaments. These molecules have many carboxyl and phosphate groups from which protons can dissociate resulting in a net negative charge. Moreover, the plasmalemma and endomembranes also generally act as charged Donnan phases (Nobel 1999).

At equilibrium, the Donnan potentials (electric potential differences) on each side of a membrane are in opposite directions. So it is assumed that they cancel each other when a diffusion potential across the membrane is calculated. It is supposed that there is no net flux of mobile ions at equilibrium, because it is assumed that no chemical potential differences across membranes exist. Then, the Donnan potential can be calculated using the constancy of the chemical potential, that is the same principle used to derive Nernst potential (EN) between two aqueous compartments. Walther Hermann Nernst was the first to model the membrane electrical potentials at equilibrium by the Nernst equation (Eq. 1), taking into account the relationship between electrical potential differences across membrane with the distribution of ions across the same membrane at equilibrium. The Nernst potential is given by
$$E_{{N_{j} }} = E^{i} - E^{o} = \frac{RT}{{z_{j} F}}ln\left( {\frac{{a_{j}^{o} }}{{a_{j}^{i} }}} \right),$$
where \(E_{{N_{j} }}\) is the Nernst potential for a certain ion j, Ei is the electric potential for the j inside membrane and Eo is the electric potential of ion j outside membrane, R is the gas constant, T is the temperature in Kelvin, zj is the electric charge of the ion j, F is the Faraday constant, ln is the natural logarithms, \(a_{j}^{o}\) is the ionic activity of ion j out of the membrane, and \(a_{j}^{i}\) is the ionic activity of ion j in of the membrane. The activities values can be replaced by ion concentrations inside and outside cells. In other words, Nernst equation states that internal and external activities (or concentrations) of a certain ionic species are associated with the electrical potential across the membranes. Thus, the electrical potential difference (E) across a membrane for a specific ion at equilibrium is the Nernst potential for this ion (Nobel 1999).

The difference of the electrical potential between two locations is a measure of the amount of electrical work involved in moving a charge from one place to another. Electric potentials variations are very important for many physiological and cellular processes essential to life insofar they realize work, besides signalling purposes. Ions have different motilities and activities in solution, diffusing towards a lower concentration or opposite charges with different velocities (Higinbotham 1973; Nobel 1999; Volkov 2006). The more mobile a specific ion species is—in relation to its counter-ion—more it will tend to diffuse away from its opposite charged ion species, resulting in a microscopic charge separation. This slight separation of charges leads to an electrical potential gradient, engendering what is known as diffusion potential (Higinbotham 1973; Nobel 1999).

Membrane potentials depend on the different mobility of many ions and their concentration gradients (as represented by the Nernst model). The ‘solution’ where diffusion takes place, in this case, is the membrane itself, including its phospholipid bilayer and associated proteins. Under usual biological conditions, most anions and cations have very low mobility through membranes because lipid phase acts as a dielectric of a capacitor. Many divalent cations, like Ca2+, cannot even cross the membranes passively, therefore these ions have minor contributions to ionic fluxes across membranes, affecting weakly the diffusion potentials (Higinbotham 1973; Nobel 1999).

Since passive fluxes of ions across membranes are caused by gradients in chemical potentials that lead to electrical potentials (diffusion potentials) across the membrane, it is possible to calculate such electrical potentials taking into account the contribution of all ionic fluxes across membrane and the condition of electroneutrality (Nobel 1999). For many plant cells, the total ionic fluxes can be given mainly by K+, Na+ and Cl because these ions generally have the highest mobility and the highest concentrations near the plant cells’ membranes, and are the main contributors for the total ionic flux density. In order to calculate membrane diffusion potentials, the Goldman equation is generally used, also referred to as Goldman–Hodgkin–Katz (GHK) equation (Eq. 2):
$$E_{M} = \frac{RT}{F}ln\left( {\frac{{P_{K} c_{K}^{o} + P_{Na} c_{Na}^{o} + P_{Cl} c_{Cl}^{i} }}{{P_{K} c_{K}^{i} + P_{Na} c_{Na}^{i} + P_{Cl} c_{Cl}^{o} }}} \right),$$
where EM is the membrane potential, PK is the permeability coefficient for K+ across a specific membrane, \(c_{K}^{o}\) is the outside concentration of K+, \(c_{K}^{i}\) is the K+ concentration inside membrane, \(P_{Na}\) is the permeability coefficient for Na+ cross a specific membrane, \(c_{Na}^{o}\) is the outside concentration of Na+, \(c_{Na}^{i}\) is the Na+ concentration inside membrane, \(P_{Cl}\) is the permeability coefficient for Cl cross a specific membrane, \(c_{Cl}^{o}\) is the external concentration of Cl, \(c_{Cl}^{i}\) is the Cl concentration inside membrane.

Despite this simple model considering only K+, Na+ and Cl, it approaches very well the values measured in controlled biological conditions. For example, the electrical potential measured across the plasmalemma of Nitella translucens (Persoon) C.Agardh in KCl bath solution is -138 mV and the calculated membrane potential using GHK equation -140 mV (Spanswick and Williams 1964). This proximity between calculated and measured values using GHK equation reinforces that the membrane potential is a diffusion potential. One can verify this by variating the external concentration of K+, Na+ or Cl and observing that membrane potentials changes accordingly with the Goldman equation (Nobel 1999).

Despite being very useful, the modelled membrane potentials given by GHK equation are valid only for passive fluxes of ions and it is not valid when active processes are taking place. Generally, protons (H+) play the major contribution for electrical potentials across plant cell membranes when active ion transport is going on. When the ionic active transport has no charge compensation, the process is called electrogenic. In fact, in many plant cells the electrogenic process is based on H+ pumping by different H+-ATPases. These protein complexes are the main players in the generation of electrical potential differences across the plant cell membranes (Higinbotham 1973; Spanswick 1981; Nobel 1999; Baluška and Mancuso 2013a; Vodeneev et al. 2016a, b).

It is worth to note that electrochemical gradients developed by electrogenesis are used for the transport of a variety of solutes through the membrane (charged or not), and for the control of osmotic pressure and cell turgor. The electrogenic H+ pumping outward of the cells is important to control the pH of the cytoplasm. It is also important to provoke the external environment acidification that is used to nutrient uptake in root cells, as well as for cell wall expansion. Hence, for plant growing and development (Higinbotham 1973; Rayle and Cleland 1992; Taiz et al. 2014). Electrogenesis is fundamental for each self-generated cells’ electrome since it is the principal force that maintains the cells’ homeostasis far from thermodynamic equilibrium. When a cell is in electrochemical equilibrium with its external milieu, it is dead. This equilibrium state can be caused by different situations such as membrane disruption, ion pumps stopping or ion channels blocking. Therefore, cell death can be characterised by the absence of the ionic/voltage gradient through the plasma membrane (Nobel 1999; Baluška and Mancuso 2013a, b; de Loof 2016).

Some particular characteristics of the membrane composition are fundamental for electrochemical gradients, or for the emergence of voltage gradients across the cell membranes. The lack of significant electrostatic interaction between polar and nonpolar charged sites of membrane lipid phase is what underlies the capacity of the membranes to limit the passage of these charged hydrophilic solutes (Higinbotham 1973; Nobel 1999). Hence, the lipid bilayer of membranes represents a dielectric phase of a capacitor, which separates the two aqueous conducting phases on either side (Fig. 1). Thus, the ions and charged molecules cannot cross freely through membranes, and some of them cross the membranes only by special protein complexes (pores, channels, pumps and transporters) under normal biological conditions. Therefore, transmembrane electrogenic pumps are especially important for the voltage gradient generation insofar they allow cells to concentrate charges against the diffusion potential that is used to make work. This potential is used for a diversity of physiological processes and for plant behaviours, including some movements (Higinbotham 1973; Spanswick 1981; Nobel 1999; Scott 2008; Baluška and Mancuso 2013a, Taiz et al. 2014; de Loof 2016).
Fig. 1

Cell membrane as a capacitor. The lipid bilayer works as a dielectric of a capacitor and the aqueous phase in both sides is the conducting phase. The cytoplasm of living plant cells usually has a negative net charge and external side is usually positive. (Color figure online)

The lipid bilayers, mainly of the plasmalemma, including their pumps and channels—these ones anchored at and connected by the cytoskeleton—constitute the most relevant components of a single cell electrome apparatus, which is the basic and elementary unit of electrome of any living system (Fig. 2) (de Loof 1986, 2016; Davies 1993; Hedrich 2012; Honigmann and Pralle 2016; Krapf 2018). Although ion fluxes occur in other membranes of plant cells, the plasmalemma is the main structure in which the majority of ion fluxes take place. Among many different membrane proteins and complexes that are permeable to ions, the most important are pumps, selective and non-selective gated channels (that can be or not sensitive to ligands, such as metabolites or ions), besides voltage-dependent gated channels and pressure (or mechanic) sensitive channels (Fig. 2).
Fig. 2

Representation of the most important components of each cell electrome. The plasmalemma is the membrane of the plant cells that offers more resistivity to ions, but is the membrane in which most ionic flow occurs. Thus, the plasmalemma is assumed to be more relevant for the cell’s self-generated electrome than other inner cellular membranes. However, the tonoplast and vacuole are also major components of the plant cell’s electrome, since they also stock ions and generate voltage gradients. There is a great diversity of membrane proteins and protein complexes that allow most of the ionic flows through the membranes. These proteins can be classified as channels (gated or not), pumps, and transporters. The channels that do not have control of pore aperture are not regulated; they are always opened (1, 2). However, there are also many different gated channels, usually regulated by molecules (metabotropic—3), by ions (ionotropic—4), by pressure (mechanosensitive—5) and by voltage oscillations (voltage-depend channels—6). These channels can be selective or not, transporting only a specific ion or a certain class of them (i.e. cation or anion, divalent or monovalent ions). There is also a great quantity and diversity of proton pumps, which transport H+ against concentration gradients using metabolic energy. The most common plant pumps are H+-ATPases that pump H+ out of the cells (7), and V-H+-ATPase that pumps H+ into the vacuole (8). Another important component of the electrome are the transporters (uniporter—9, symporter—10, antiporter—11), which use proton gradient generated by H+-ATPases to cotransport other ions, even against the concentration gradient (in the case of cotransporters). All these different membrane proteins mentioned are usually anchored and interconnected by cytoskeleton filaments (represented by white lines in cytoplasm), which allows a heterogeneous distribution of such set of proteins that, by turn, allows the generation of ion fluxes across different sites of the cell membrane. These electrical fields are represented by dashed black lines with arrows indicating the direction of the ionic flows from one side to the other of one cell (12). (Color figure online)

Different molecules inside and outside the cells regulate metabotropic-gated channels or pumps. Some of the protein families of plant metabotropic channels, like GLRs (glutamate-like receptors) and CNGCs (cyclic nucleotide-gated channel) are homologous to animal ion channels and regulated by the same metabolites (glutamate and cyclic nucleotides), indicating that a primitive common ancestor already had a similar signalling system. Ionotropic-gated channels or pumps are regulated by specific ions, while voltage-gated channels and pumps are regulated by voltage status of the membrane (membrane potential). These voltage and ionotropic channels are fundamental for fast membrane depolarisation and repolarisation that characterise, for instance, the action potentials of plant and animal cells. There are also mechanosensitive channels and pumps that are regulated by distention or extension over the membrane and cytoskeleton, caused by changes in the cells turgor pressure and/or mechanical stimuli (Davies 1993; Fletcher and Mullins 2010; Hedrich 2012, Baluška and Mancuso 2013a; Baluška and Miller 2018).

The temporal and spatial distribution of these membrane proteins, as well as their permeability control by a multitude of different ligands and biophysics status, together with their patterns of mobilisation and transport between membrane and cytoplasm, creates a very complex arrangement that is the basis for the emergence of complex electrome dynamics in its close interaction with environment (Hedrich 2012; de Loof 2016; Baluška and Mancuso 2013a; Baluška and Miller 2018). The asymmetrical distribution of channels, pores and pumps along time and space through membranes is essential for the generation and alteration of electrical currents based on ionic fluxes across the cells. Consequently, such ionic currents generate an electrical field around cells that likely has important roles in the life of cells and organisms (Scott 1967; Baluška and Mancuso 2013a, b; de Loof 1986, 2016; Baluška and Miller 2018).

Probably, the best example of the importance of electrical fields for life forms are the electrical fishes, which use self-generated electrical fields to forage, for mate recognition, to kill their preys and to defend themselves from predators (Stoddard 1999; Catania 2016). Another notorious phenomenon in the natural world is the sharks’ sensitivity to electrical fields by their ampullae of Lorenzini, which helps them to detect their preys (Josberger et al. 2016). It has been demonstrated that bumblebees use electrical fields in the air to make foraging decisions which help them to find and land on flowers (Clarke et al. 2013). Moreover, it is possible that endogenous electrical fields act in brain physiology, morphogenesis and anatomy (Frohlich and McCormick 2010; Levin et al. 2017). The list of examples of electrical fields in life processes is extensive and had just begun to be explored (Baluška and Mancuso 2013a; Baluška and Miller 2018; de Loof 2016).

Less known is that electrical fields generated by ionic currents are important for root and pollen tube growth (Weisenseel et al. 1975, 1979; Miller and Gow 1989), for root gravitropic responses (Iwabuchi et al. 1989; Collings et al. 1992), for morphogenesis and development (Scott 1967; Nuccitelli et al. 1986; Miller and Gow 1989; Tseng and Levin 2013), for localisation and orientation of cells components (Jaffe et al. 1974), for transport, signalling and communication (Olivotto et al. 1996), for cell electrophoresis (Jaffe 1977; de Loof 1986) among many other biological processes.

The ionic flows across membranes (especially the electrogenic ones) also influence the transport of other solutes. Together, they alter the osmotic status of the cell, causing variations in osmotic and turgor pressure of the plant cells by water and solute flows across a semipermeable membrane. These variations are indispensable for basic plant physiology and behaviour, like nutrient uptake (Fig. 3a), growing and development, besides many types of movements, including tropisms and nastic movements (Higinbotham 1973; Scott 2008; Segal 2016). The cell expansion in general, and especially the growth of the pollen tubes are good examples of important processes for plants that depend on variations in osmotic/turgor status of cells associated with ionic flows (Higinbotham 1973; Rayle and Cleland 1992; Kroeger et al. 2011; Taiz et al. 2014; Wudick et al. 2018).
Fig. 3

Examples of general plant electrical activity. One common example of electrical activity fundamental for plant life, is nutrient uptake, here represented by NO3 uptake, which involves many ion channels and pumps (a). Other common electrical activity in plant life is stomata control, which also depends on different channels and pumps to regulate pore aperture (b). Electron transport chains of chloroplasts, which involve the transport of electrons made by different carriers and the ion transport made by different pumps and channels (c). A special type of electrical activity very common in plant life is electrical signalling, here represented by an action potential generated and transported along phloem cells (d). (Color figure online)

The control of stomata aperture (Fig. 3b) and the consequent regulation of leaf gas exchange, as well as the control of leaf movements, including those made by the pulvini, are other great examples of the important processes mediated by ionic fluxes across the membranes (Higinbotham 1973; MacRobbie 2006; Scott 2008; Taiz et al. 2014; Segal 2016). Root growing and its foraging behaviour also depend on the electrical activity of the ion flows across the membranes, and the consequent flow of water induced by them (Burbach et al. 2012; Baluška and Mancuso 2013b; Yokawa et al. 2014). Hence, osmotic, hydraulic and ionic (voltage) dynamics are interlinked and underpin a plethora of plant behaviours and physiological processes (Baluška et al. 2006; Volkov 2006; Baluška and Mancuso 2013a, b; Taiz et al. 2014).

Another example of electrical activity vital to plant life is the electron transport chain at the membranes of chloroplasts and mitochondria. These chains are composed of a set of proteins that transport the electrons from the water molecule to the ferredoxin-NADPH reductase (during the photochemical phase of photosynthesis), and from the NADH or NADPH atoms to the water molecule (during respiration). In both cases, there is literally a flux of electrons running from one site to the other, which can be characterised as an electric current (Fig. 3c) (Taiz et al. 2014; Buchanan et al. 2015).

The protein complexes both in the photosynthetic and in the respiratory apparatus are partially responsible for keeping the electrochemical gradient within the lumen of thylakoid and the mitochondrial matrix. Again, in both cases, the electrochemical gradient is utilised for the synthesis of ATP by the ATP synthases, which benefits from the proton gradient generated (Buchanan et al. 2015). Hence, the transport chains have a vital importance for the plants insofar they allow one of the most relevant processes of ATP synthesis within the cells. This is a good example of an electrical activity that is not necessarily considered as an electrical signal because there is an electrical current (activity) that is not carrying meaning to trigger activities in other sites.

Other examples of electrical activity are the variations on the membrane potential caused by changes in light availability. MacRobbie (1962) and Spanswick and Williams (1964) detected that the membrane potentials of Nitella translucens (Persoon) C.Agardh were affected by light/dark alternations, observing that by switching on the lights the electrogenic process was activated. Hence, light leads to an increase in cytoplasmic ATP concentration, which allows the functioning of the ATPases on the membrane. This increase in ATPases activity keeps the potential more negative compared to plants kept in the dark (Spanswick and Williams 1964). In the light presence, the active transport of K+ and Cl (inward) and Na+ (outward) increases. The decreased availability of energy in dark leads to the weakening of the transport processes that keep the membrane potential, decreasing it. This explains why the plants’ membrane electrical potential is more negative in the presence of light (Spanswick and Williams 1964; Nobel 1999).

In the first half of the 20th century, pioneering studies conducted by Bose, Brauner, Clark, and Shrank, demonstrated that the perception of gravity and growth reorientation of a tilted plant was related to a transient electrical activity (Stanković 2006). The link between electrical activity and gravitropism was not known. Behrens et al. (1985) found transient voltage variations of the cell membrane in the root tip of Lepidium sativum L. associated with gravistimulation. However, they did not find any correlation of “excitatory voltage changes (action potentials) in signal transmission in the gravitropism of Lepidium roots” (Behrens et al. 1985).

Many other researchers have observed this electrical signalling associated with changes in the gravity vector, both in the roots and in the shoots, and this is called “gravielectric phenomena” (Stanković 2006). It is still not clear for many researchers if this electrical activity has any physiological significance, or if it is just an artefact generated by other subcellular processes related to the gravity change, such as the reorganisation of the organelles or changes on cytoskeleton tensions, which could induce the activation of mechanosensitive ion channels (Stanković 2006).

Perhaps the best-studied aspects of the plant electrome until now are the electrical signalling processes, mainly the evoked ones (i.e. induced by external stimuli) (Fig. 3d). Therefore, from now on we will focus on these signals that are already known for many years.

3 Electrical signalling in plants

The first researcher who considered the importance of electrical signalling for the plants was the Indian polymath Jagadish Chandra Bose (1850–1937). However, the first record of an extracellular plant electrical signal was carried out by the human physiologist Burdon-Sanderson (1828–1905). Following a Charles Darwin’s suggestion, Burdon-Sanderson (1873) measured the electrical response of the Venus flytrap (Dionaea muscipula Ellis.) traps after a mechanical stimulation on its leaves.

In 1926, Bose demonstrated that the fast movements performed by the leaves of Mimosa sp. and Desmodium sp. were caused by long-distance electrical signalling. Using isolated vascular bundles of ferns, Bose also showed that electrical excitation was transmitted in a similar way as electrical signals in animal nerves, confirming that plants produce continuous systemic electrical pulses that spread throughout the plant (Bose 1926; Fromm and Lautner 2007).

Initially, it was believed that electrical signals were present only in plants considered ‘sensitive’, such as the Venus flytrap or the Mimosa pudica L. However, already at Bose’s times, it was known that action potentials, among other electrical signals, were also generated in ‘ordinary’ plants, other than the ‘sensitive’ ones. Bose himself used species such as Musa sp., Brassica sp., Phoenix dactylifera L., Chrysanthemum sp., Ficus sp., Nauclea sp., Mangifera sp., among others, to demonstrate that (Shepherd 2005, 2012).

Umrath (1930) performed the first intracellular measurements of an action potential (AP) in living vegetal cells of Nitella sp. using microelectrodes. This experiment was made before the first intracellular record of an AP in animal cells (Baluška et al. 2006; Volkov 2006; Fromm and Lautner 2007). From this time on, research demonstrated that voltage variations in animal and plant cells relies on similar processes, and now we know that they share a similar set of ionic channels and pumps, many of them encoded by homolog genes (Trewavas 2016, 2017; Canales et al. 2018).

Likewise, studies have demonstrated that neurotransmitters, among other neuronal molecules, also play important roles in plant signalling transmission and integration (Baluška et al. 2006; Brenner et al. 2007). Furthermore, there is increasing evidence that neurotransmitters act together with Ca2+ ions and Ca2+ channels sensitive to glutamate in a particular zone of the roots’ tip, modulating and synchronising electrical pulses along hundreds of cells (Masi et al. 2009). Altogether, this evidence is indicating that it is possible that plants possess a simple neural network (or proto-neural) dependent on neurotransmitters and electrical signals generated by ion fluxes (Brenner et al. 2007; Baluška and Mancuso 2009a, b, 2013a, b; Debono 2013a, b; Calvo 2016; Calvo et al. 2017).

Nowadays, it is known that all plants use electrical signals to coordinate the activity of different modules, as well as the diversity of physiological functions required to improve the plants’ adaptive responses to the environment (Fromm and Lautner 2007; Szechyńska-Hebda et al. 2017). Probably, the need to readily respond to environmental fluctuations has been the cause by which plants developed paths for the transmission of electrical signals (Baluška et al. 2006; Fromm and Lautner 2007).

Environmental fluctuations might induce electrical signals in a continuous manner throughout the symplast and/or apoplast. Until now, basically, four different kinds of electrical signal propagation have been reported in the plant kingdom: action potentials (APs), variation potentials (VPs), local electric potentials (LEPs), and systemic potentials (SPs) (Fromm and Lautner 2007; Huber and Bauerle 2016; Sukhova et al. 2017; Szechyńska-Hebda et al. 2017). In the following sections, each one of these signals is described in more details.

3.1 Action potential

The action potential (AP) was the first electrical signal recorded in plants and it is characterised by a rapid depolarisation phase of the cell membrane followed by a phase of repolarisation likewise fast (Burdon-Sanderson 1873; Pickard 1973; Vodeneev et al. 2016a; Sukhova et al. 2017). The APs are triggered by non-invasive stimuli such as electrical stimulation, irradiation, thermal shock, among others, although they have already been observed after damage caused by caterpillars feeding on Vicia faba L. and Hordeum vulgare L. (Zimmermann et al. 2016). The first phase of the AP, the membrane depolarisation, is caused by the increase of cytoplasmic calcium concentration due to the opening Ca2+ channels, diffusing calcium into the cytoplasm from the apoplast, as well as from the vacuole and endoplasmatic reticulum (Huber and Bauerle 2016; Novikova et al. 2017).

As a response to the calcium influx, Cl channels open and chloride ions diffuse outside the cell due to the electrochemical gradient until the voltage-dependent K+ channels are activated and the repolarisation begins by the efflux of K+. Moreover, the re-establishment of the resting potential is promoted by H+ pumps (Fromm and Lautner 2007; Huber and Bauerle 2016; Sukhova et al. 2017; Trebacz et al. 2006).

The stimulus for triggering the AP must reach a critical intensity (excitation threshold) to initiate. If the stimulus does not achieve this critical intensity, the AP will not be commenced. This is known as the all-or-none principle, which is also characteristic of neuron APs. As soon as the threshold is reached the signal perpetuates itself through the plasmodesmata of the phloem parenchymal cells with constant speed and amplitude, similar to the AP propagation on nerves (Fromm and Lautner 2007; Trebacz et al. 2006; Szechyńska-Hebda et al. 2017).

APs are the most rapid electrical signals detected in plants until now, travelling along the phloem with velocities that can reach up to 3000 cm s−1, a value similar to that of animal neurons, which range about 1 and 100 m s−1 (Volkov et al. 2000; Fromm and Lautner 2007, de Loof 2016). However, the velocities generally vary around 1–10 cm s−1 (Huber and Bauerle 2016).

There is increasing evidence pointing that sieve tubes have some similarities with nerves, when it comes to transmission of electrical signals across the organism. The earliest evidences are Bose’s works in the first half of the last century (Bose 1926; Stahlberg 2006; Brenner et al. 2007; Fromm and Lautner 2007; Shepherd 2012; Calvo 2016, Calvo et al., 2017). Today, it is accepted that the phloem is the highway for fast long-distance electrical signal transmission by AP in numerous species (Baluška and Mancuso 2009a, b, 2013a, b; Hedrich et al. 2016; Huber and Bauerle 2016).

Because action potential signals are so uniform and rapid, they are suitable for a quick signalling throughout the plant but do not have the potential for encoding much information, since they are very similar, independently of the stimuli that initiated them. For delivering more detailed systemic information, it seems that variation and systemic potentials, and their combinations with APs, are more adequate (Fromm and Lautner 2007; Sukhov et al. 2012, 2014; Huber and Bauerle 2016; Vodeneev et al. 2016a, b).

3.2 Variation potential

Variation potential (VP) is a characteristic plant electrical signal, absent in animals. It is characterised by an initial phase of rapid depolarisation followed by a slower repolarisation when compared to the AP (Huber and Bauerle 2016; Vodeneev et al. 2016b). VPs are induced mainly by biotic and abiotic stimuli such as mechanical wounds, tissues burning, herbivore attack, drought or flooding, among others (Gallé et al. 2015; Yan et al. 2009).

The VPs depolarisation is rapid and it is caused by the same mechanism of ionic unbalance of APs, while the slow phase of repolarisation commences with a transitory inactivation of the H+-pumps that can last over 30 min (Stahlberg et al. 2006; Vodeneev et al. 2016b). It is largely accepted that VPs are propagated in the xylem and do not follow the all-or-none principle as APs. VPs velocity and amplitude decreases as long as it distances from the stimulus site whilst its magnitude and shape vary according to the intensity of the stimulus itself, differently from AP that is more regular and uniform (Stahlberg et al. 2006; Vodeneev et al. 2016a).

3.3 Systemic potential

Systemic potentials (SP) were first recorded by Zimmermann et al. (2009) after injuring the leaves of Vicia faba and Hordeum vulgare and next stimulating the wound with inorganic cations. The signals detected in leaves dozens of centimetres away from the stimulus site were self-propagable as the APs, but they did not follow the all-or-none principle. Furthermore, different of what occurs in APs and VPs, these signals were characterised by a hyperpolarisation of the apoplast instead of a depolarisation. Whereas the former signals are caused by the continuous Ca2+ influx through the plasma membrane, the SPs are initiated by the activation of the H+ pumps that led to the hyperpolarisation of the membrane. These signals propagate throughout the plant at a speed of 5–10 cm min−1. The amplitude depends on the stimulus intensity. Even stimuli weak enough to not initiate the APs, by not reaching the critical intensity, can trigger SPs (Zimmermann et al. 2009). SPs were found being triggered by wounding originated by caterpillars feeding on leaves of H. vulgare, V. faba and Nicotiana tabacum L., spreading systemically through the plants (Zimmermann et al. 2016).

3.4 Local electrical potential

The local electric potential (LEP) is generated at the stimulus site under the influence of various factors such as light, changes of temperature, water, soil nutrients, air humidity, phytohormones, and pathogen infection (Sukhova et al. 2017). Its amplitude depends on the intensity and duration of the stimulus. It is generated by the transitory inactivation of the H+-ATPases, which are the main active electrogenic transporters of the plasma membrane. Besides, LEPs can also be generated by changes in the activity of ionic channels (Yan et al. 2009; Vodeneev et al. 2016a).

3.5 Propagation pathways of electrical signals

Environmental cues perceived by plants’ tissues must be encoded in signals and forwarded to other parts of the body in order to inform the entire plant about what is happening around. These signals might be, for example, the nutrients that the roots found in their underground foraging or other compounds related to wounding, pathogen infection or oxidative stresses, among others (Kwaaitaal et al. 2011; Baluška and Mancuso 2013a, Canales et al. 2018). Since plants are sessile organisms, they cannot afford to have a centralised system to which the signals are sent to be processed before being shared with other parts of the body (Trewavas 2016). Hence, the information they perceive is processed locally and spread diffusely through the plant (Jones and Dangl 2006; Huber and Bauerle 2016).

Different stimuli are encoded in signals as diverse as reactive oxygen species (ROS), micro RNAs and other nucleotides, proteins, hormones, and electrical signals. Particularly, the latter ones are highly accomplished for the long-distance signalling task because of its rapid velocity of propagation (several centimetres per second), as already mentioned. Notwithstanding, all these different signals might work together to encode complex meaningful messages that form the plant proto-neural system (Jones and Dangl 2006; Debono 2013a, b; Choi et al. 2016, 2017).

One of the simplest model organisms to study the nature of the electrical signalling in plants is the Characean algae. In the cells of these plants, APs are generated by the entrance of Ca2+, which causes the depolarisation of the membrane, making voltage-gated chloride channels to open, allowing the Cl to diffuse inside the cell. This unbalance opens voltage-gated potassium channels, which make potassium leak out of the cell, thus restoring the membrane resting potential. In higher plants, the mechanism is similar, but there is also the involvement of H+-ATPase pumps for the restoration of the resting potential (in APs and VPs) or to the membrane hyperpolarisation (in SPs) (Vodeneev et al. 2016a, b; Zimmermann et al. 2016).

Chara algae are structurally simple since their inter-nodal stems are composed of single rows of giant cells connected by small isodiametric cells. Though, the propagation of an AP generated in some part of the plant is quite straightforward. Once triggered, the APs flow down the cell with constant amplitude and velocity until they reach the next cell. Many calcium channels are voltage-dependent and, once the AP arrives with amplitude higher than the excitation threshold, it causes the opening of the channel, thus propagating the signal (Vodeneev et al. 2016a, b).

In higher plants, the process of signal propagation is more complex. APs, SPs and VPs travel through different pathways. Since APs and SPs are active self-propagated processes, they need to travel along the living tissue. Locally, in the parenchyma, they spread from cell to cell via plasmodesmata. But for spreading systemically it would not be feasible since the small diameter of these pores represents a non-negligible source of resistance (Zimmermann et al. 2009; Hedrich et al. 2016; Vodeneev et al. 2016a).

The phloem network consists of a continuous set of cells that connects the entire plant body. It is the main pathway by which sugars, hormones, RNAs, peptides and all sorts of signalling molecules traffic in the plant, from the highest leaf to the deepest root (Baluška and Mancuso 2009a, b, 2013a, b; Hedrich et al. 2016; Vodeneev et al. 2016a). In addition, the cells of the sieve tubes, besides their relatively wide size, have large pore plates in the connection between them. Those cells also suffer a partial apoptosis that devoid them of vacuole, nucleus, and plastids. All these properties make them not only highly suitable for the quick transport of solute and signalling molecules, but for the conductivity of electrical signals as well (Baluška and Mancuso 2009a, b; Hedrich et al. 2016).

APs and SPs travelling along the sieve tube elements can carry information to distant parts of the plant, exciting adjacent parenchymal cells via plasmodesmata. These electrical signals can be, then, spread by tissues others than the phloem. Hence, long-distance communication is ensured (Vodeneev et al. 2016a).

The mechanism of propagation of the VPs is less elucidated and more puzzling. To be transmitted by the xylem, two hypotheses have been proposed: a hydraulic one, in which the signal is transmitted by a hydraulic wave, and a chemical one, by which some chemical substance is responsible for the propagation of the signal. Each hypothesis has its strengths and weaknesses, and Vodeneev et al. (2016b) propose a combination of both. Interestingly, Bose, at the beginning of the 20th century, proposed similar hypotheses (Bose 1926; Baluška and Mancuso 2009a, b; Shepherd 2005, 2012).

With respect to the modulation of plant electrical signals, their generation and propagation, in any case, is caused by ionic flows through cell membranes crossing by ionic channels and ionic pumps. Such structures are highly diverse, where different stimuli trigger different channels and pumps, generating different signals (Canales et al. 2018). For example, AP generation is caused by voltage-dependent Ca2+ channels, whilst for VP, due to its chemical nature, the triggering of the signal is related to ligand-controlled ion channels (Vodeneev et al. 2016a).

According to Canales et al. (2018), “electrical proprieties of cells derive from the expression and control of ion channels, and pumps.”. These channels and pumps are expressed in different concentrations at the cell membranes of different tissues of the plant depending on the site of the cell, the development stage and the ecophysiological status of the plant (Baluška and Mancuso 2013a, b; de Loof 2016; Canales et al. 2018). Ion pumps and channels can also be modulated, being quickly expressed and/or mobilised from internal cells’ compartments up to the membrane just after stimulation (Matzke et al. 2010; Baluška and Mancuso 2013a, b; Canales et al. 2018).

Canales et al. (2018) showed that there are 56 different calcium channels only for Arabidopsis. Each one is excited and opened by a different stimulus, such as chemicals, mechanical stimulation and voltage variations. The localisation of each one of these channels is finely regulated by a diversity of molecules such as anchoring proteins, lipids, and cytoskeleton elements (Lee 2015). Indeed, in a recent study, it has been demonstrated that the localisation of specific glutamate receptor-like (GRL) calcium channels is sorted and regulated by CORNICHON HOMOLOG proteins, which brings different GLRs to different membranes of the cell, allowing a very precise regulation of the Ca2+ homeostasis. In animal cells, CORNICHON proteins have similar roles (Wudick et al. 2018).

In plants, there are different ‘checkpoints’ where the electrical signals can be filtered and regulated to ensure a reliable communication. For example, parenchymal cells of the leaves have a lower excitation threshold when compared to the phloem sieve tubes (Huber and Bauerle 2016; Vodeneev et al. 2016a). This is interesting to the plant because slight stimuli will excite the cells only locally, and only strong stimuli that ought to be communicated throughout the plant will reach the sieve tubes. Once there, the APs generated will spread with continuous amplitude and velocity by the phloem, which connects all the plant (Hedrich et al. 2016; Huber and Bauerle 2016).

Interestingly, in Arabidopsis, as well as for many other species, there is no connection via plasmodesmata between the parenchymal and epithelial cells of the roots and shoot, making them electrically uncoupled. The only pathway for the transmission of APs is the sieve tubes of the phloem (Canales et al. 2018). The region of the hypocotyl might constitute, then, another site for the sorting and modulation of the electrical signals that are shared by both the aerial and underground parts of the plant.

3.6 Integration of the electrical signals with ROS, NO, and Ca2+ signalling

Plants possess numerous different signalling pathways, which integrate stimuli received by the sensorial system, crossing information at connections of biochemical pathways of the metabolism network. This signalling network ‘crosstalk’ allows systemic responses to localised stimulus (Bray 1995; Genoud and Métraux 1999; Baluška and Mancuso 2013). Many of these signalling pathways derivate from molecules that were only known in the plant defence system, as is the case of the hydrogen peroxide (H2O2) and the nitric oxide (NO). More recently, their roles as signalling molecules acting in various biotic and abiotic stimuli have been described. In addition to their function in the stress response, the involvement of H2O2 and NO in developing processes, such as the germination of seeds, roots growth, and gravitropism, indicate that these signalling molecules are key-regulators for a plethora of endogenous and exogenous stimuli, in a similar way as abscisic acid and auxin (Neill et al. 2002a, b; Choi et al. 2017).

It is already consolidated that the transmission of electrical signals in living systems provokes systemic responses (including gene regulation) in distant tissues from where the perception of the signal occurred. It allows the whole plant to protect itself from a second occurrence of the same stress. In such a manner, the electrical signals can act as a starting signal preparing the plant to respond in a more selective fashion to subsequent specific signals, which is a kind of memory (Wildon et al. 1992, Thellier and Lüttge 2012; Mousavi et al. 2013, Choi et al. 2017; Szechyńska-Hebda et al. 2017). It is thought that the electrical signals act together with specific signalling mechanisms such as Ca2+ waves, waves of reactive oxygen species (ROS waves), hormones, and small RNA molecules, beyond neurotransmitters (Baluška 2013; Gilroy et al. 2016). These different signalling mechanisms in plants operate in diverse timescales and codify information about the specific nature of a given stimulus (Szechyńska-Hebda et al. 2017).

The presence of a great number of Ca2+ channels in the plasmalemma, together with the Ca2+-sensitive proteins, promotes the triggering of a multifactorial signalling cascade, which causes changes in the metabolism and in gene expression, enabling adaptive reactions (Wildon et al. 1992, Matzke et al. 2010; Baluška and Mancuso 2013a, b; Mousavi et al. 2013, Szechyńska-Hebda et al. 2017; van Bel et al. 2014). Soon after the detection of an external cue, an influx of Ca2+ occurs in the cytoplasm by channels permeable to Ca2+ on the plasma membrane, increasing the cytosolic levels of this ion. This increase occurs rapidly, at a speed of approximately 400 µm s−1 (Shabala et al. 2016; Szechyńska-Hebda et al. 2017).

As for the Ca2+ propagation, the spread of ROS waves occurs in a relatively rapid speed after receiving the stimulus: approximately 8 cm min−1 (Miller et al. 2009). The ROS waves propagation might elicit action potentials since ROS regulate a number of ionic channels located in the cell membranes. These channels can rapidly regulate the cell ionic profile in face of environmental condition changes. This relation indicates the possible involvement of electrical signals with ROS (Szechyńska-Hebda et al. 2017; Shabala et al. 2016). Furthermore, the ROS signalling is intimately linked with the Ca2+ signalling, since the increase of cytosolic Ca2+ induces the production of ROS (Gilroy et al. 2014; Choi et al. 2017). In plants, ROS can act directly as signalling molecules, or indirectly as secondary metabolites (Szechyńska-Hebda et al. 2017).

The different systemic signalling pathways in plants and responses to different stimuli have been widely studied (Suzuki et al. 2013; Szechyńska-Hebda et al. 2017; Shabala et al. 2016; Choi et al. 2017; Gilroy et al. 2016; Evans et al. 2016). In a study performed by Suzuki et al. (2013), a spatiotemporal interaction was described between ROS waves and abscisic acid (ABA) which is linked to the regulation of the rapid systemic acclimation to thermal stress in Arabidopsis plants. In their work, it was also demonstrated that ROS waves are associated with the propagation of electrical signals after exposure to abiotic stresses. In such case, the electrical signalling can be attenuated or amplified by the activation of the production of ROS by means of protein phosphorylation or by the Ca2+ signalling via Respiratory Burst Oxidase Homologs (RBOHD) proteins (Gilroy et al. 2016; Choi et al. 2017).

Besides, the ROS signalling might be intimately linked to the Ca2+ signalling, as it was demonstrated in a study conducted by Evans et al. (2016), in which a mathematical modelling was combined with experimental analysis to understand the triggering of the Ca2+ waves provoked by saline stress, and its connection with ROS in both wild-type Arabidopsis and a mutant super-expressing the TWO PORE CHANNEL 1 (TPC1) genes. It was possible to demonstrate that the propagation of Ca2+ and ROS are interlinked, since the application of the saline stress triggered ROS production by the RBOHD proteins in the stimulated site, and the ROS diffused by the apoplast activating, therefore, the Ca2+ channels sensitive to ROS present in the plasma membrane. This lead to a systemic distribution of both Ca2+ and ROS. However, the diffusion of Ca2+ occurred via plasmodesmata, and the diffusion of ROS occurred by the apoplast.

A general model integrating all the signalling pathways in plants was proposed by Gilroy et al. (2016). This model suggests that a local stimulus originates Ca2+ waves, and these waves would be integrated to the ROS waves by means of activation of RBOHD proteins by direct binding of Ca2+ ions. Also, it could take place by means of Ca2+-dependent protein kinases (CPK) phosphorylation, thus causing a wider ROS production. In turn, it might be integrated into an electrical wave by activation of ionic channels that depolarise the plasmalemma and regulates the electrical signals. The Ca2+ waves can, additionally, be integrated into the electrical waves via GLRs and (two-pore channel) TPC1. Finally, these three waves could amplify and regulate each other mutually, spreading different information through plant organs or even to the whole body.

3.7 Electrical signals and plant physiology: the case of photosynthesis

As mentioned above, electrical signals are underlying many plant physiological processes. For sake of example, herein we describe how electrical signalling is related to one of the most fundamental plant’s traits: the photosynthesis.

Many studies describe the influence of electrical signalling on photosynthetic processes. Different photosynthetic responses might be induced by a local stimulus, which in turn can trigger responses in distant parts from the stimulated site. The electrical signals play an important role in the regulation of photosynthetic responses, such as the rapid inactivation of photosynthesis. APs and VPs are related to long-term activation and inactivation of photosynthesis (Sukhov 2016).

The rapid inactivation of photosynthesis by electrical activity is the best-studied phenomenon of the photosynthetic responses. To elucidate this kind of response, two mechanisms have been proposed: (i) high concentrations of the cytosolic Ca2+ results in an increase in the concentration of this same ion at the chloroplast, causing inactivation of the enzymes of the Calvin–Benson cycle (Krupenina and Bulychev 2007); (ii) the alterations in the intra- and extracellular pH, caused mainly by the inactivation of the H+-ATPases pumps, interfere with the diffusion of CO2 in the mesophyll. Nonetheless, the mechanisms of such effects are not clear yet (Gallé et al. 2013; Sukhov et al. 2014; Sukhov 2016).

In some studies, a few minutes after the occurrence of different stimuli (e.g. burning, mechanical damage, electrical current) different effects have been reported, such as rapid reduction on the assimilation rate of CO2, decreasing in photochemical quantum yield of the photosystems I and II, and an increase in non-photochemical processes of energy dissipation (Pavlovic et al. 2011; Hlavinka et al. 2012; Sukhov et al. 2015; Surova et al. 2016).

The photosynthesis activation was discussed in some studies, in which a correlation was observed between the generation of the AP related to the re-irrigation after drought, and the increase in the assimilation rate of CO2 in maize leaves under high light conditions. In this case, the response started 10 min after the stimulation and reached its maximum amplitude within 20–30 min. However, when the response was inhibited by the interruption of the AP propagation by cooling, it became clear that the AP induced the photosynthetic response (Grams et al. 2007). The transient activation of photosynthesis was observed after the induction of a local stimulus with high temperature. Interestingly, this kind of stimulus is normally related to the generation of VPs, which in turn are also capable of reducing photosynthesis (Vodeneev et al. 2016a, b).

The long-term inactivation responses of photosynthesis, including diminished CO2 assimilation rate, the reduction of the photochemical quantum yield and the increase in non-photochemical processes of energy dissipation, were observed a few minutes after local stimulation, with a duration of inactivation up to 5 h or more (Herd et al. 1999).

There is a hypothesis that injuries could induce the production of systemin, jasmonic and abscisic acids, which are transported to other parts of the plant and are capable of suppressing photosynthesis. The propagation of electrical signals was observed before the changes in the phytohormones contents, confirming the influence of the electrical signal for the induction of long-term responses (Hlavinka et al. 2012). According to Sukhov (2016), the prolonged inactivation induced by the electrical signals also can be measured by the change in the cytoplasm and apoplast pH, similarly to the rapid inactivation of photosynthesis.

The photosynthetic response presents a variable velocity depending on the mesophyll site, with the strongest and most rapid responses observed in the sites near the vascular bundles, where the cells of the photosynthetically active mesophyll are localised, suggesting that the propagation of the electrical signal runs throughout the vascular vessels to the other cells of the leaf (Bialasek et al. 2017).

The mechanism of photosynthetic electron transport is associated with the electrical signalling since light induces voltage variations, which in turn influences photosynthesis. According to Szechyńska-Hebda et al. (2017), the excess of light promotes alterations on the membrane potential, which induces responses to a systemic level on the plant. However, the use of specific inhibitors of the electrical transport of photosynthetic electrons like DCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea) and DBMIB (2,5-dibromo-3-methyl-6-isopropylbenzoquinone) deregulated the systemic propagation of the electrical signals. This demonstrates the importance of this interaction for the promotion of an efficient mechanism of control to keep the plant’s homeostasis.

4 Plant electrome as a complex self-organised multi-coloured ‘noisy’ system

As described in the sections above, there are different kinds of electrical activities and signals in plants (e.g. APs, VPs, SPs and LEPs) which act together with hormones, neurotransmitters, ROS and Ca2+ waves carrying information through plant modules about local stimuli. Besides, it is known that some electrical signals can trigger other electrical signals of different natures in different cells, interchanging electrical information among them (i.e. APs that trigger VPs and vice versa) (Hedrich et al. 2016; Huber and Bauerle 2016; Vodeneev et al. 2016a; Choi et al. 2017).

This blend of electrical signals, among many other electrical activities, creates a complex web of systemic information in which various electrical signals might be overlapped, hindering a proper analysis of specific signals (Choi et al. 2016, 2017). This complexity of electrical oscillations in plants, which emerges from numerous different processes in time and space, has been named by Souza et al. (2017) as ‘plant electrome’, since it represents the sum of all electrical activity of plant tissues along time, as proposed by de Loof (2016). These overlapped electrical signals have been registered as electrophytograms (EPGs) (Debono and Bouteau 1992; Saraiva et al. 2017; Souza et al. 2017).

The word ‘electrome’, then, can be used to describe the totality of the ionic currents in different levels of organisation of any biological system. All scales of the electrome, from cells to organisms, depend on the set of ions, ions channels and pumps anchored by cytoskeleton in the plasma membranes (de Loof 2016). All cells, from bacteria to humans, including plant cells, have a dynamic DNA-cytoskeleton-membrane complex that engenders cells electrome complexity and its dynamics, which are possibly involved with cells memory and cognition (Matzke et al. 2010; Trewavas and Baluška 2011; de Loof 2016; Calvo et al. 2017; Baluška and Miller 2018).

In previous studies, the plant’s electrome has been accessed as a micro voltage potential variation measured by electrophytographic technique (EPG), which is basically an adaptation of the classical EEG and ECG methods used to monitor the electrome of the heart or brain of humans and non-human animals (Karlsson 1972; Brenner 1985; Debono and Bouteau 1992; Debono 2013a, b; Saraiva et al. 2017). Below, we summarise some of our current findings on plant electrome, describing its statistical properties, special traits, methods for automatic classification, and ongoing issues.

Our main model studied until now has been soybean plants subjected to different external stimuli. The microvolt variations measurements are registered by needle electrodes inserted between root and shoot with sampling rate ranging from 62.5 up to 125 Hz, depending on the specific aims of the experiments. This specific part of the plant has been chosen because it is halfway between shoots and roots, where the evoked long-distance electrical signals must pass through. The time series scored are then analysed by standard mathematical methods, including auto-correlation and cross-correlation, probability density function, fast Fourier transform (FFT) and power spectral density (PSD) analyses, besides complexity measurements as approximate entropy (ApEn) (Saraiva et al. 2017; Souza et al. 2017).

In a typical electrome signal record (Saraiva et al. 2017) the auto-correlation analysis has shown that, contrary to the white (random) noise of the device, all sampled experimental time series had long-range correlations. This was our first indication that plants voltage variations (ΔVs) are distinct from open electrodes ΔVs. However, the pattern decay of the auto-correlation function (ρ(τ)) of the signal measured before and after the stimuli exhibited different behaviours (the trend of decay of ρ(τ) before stimuli was continuous, but after stimuli it was oscillatory). In order to investigate the observed differences in the ρ(τ) between the signals before and after stimuli, a cross-correlation analysis has been carried out, often showing no correlation between them (Saraiva et al. 2017), indicating that the electrome dynamics of each plant had a completely different pattern before and after osmotic stimulation.

The β exponents calculated from the power spectral density function (PSD ~ 1/fβ) indicated a range of coloured noises (Cuddington and Yodzis 1999). Electrome recordings before stimulation have shown an average exponent β = 1.5 ± 0.3, a typical reddened noise. However, after osmotic stimulus, the average β exponent was 2.6 ± 0.2, with values between brown (β = 2) and black (β = 3) noises (Saraiva et al. 2017). This range of β values with β ≠ 0 and β ≠ 2 indicates that the runs have a long-range temporal correlation (Gao et al. 2006; Petchey et al. 1997), supporting the results of auto-correlation, and reinforcing memory effect content of plant electrome dynamics.

Moreover, the analysis of auto-correlation together with PSD have shown that plant electrome dynamics is not random, rather it has shown a long-range temporal correlation and likely meaning information. In other words, plant electrome ∆Vs co-varies not only with its most recent value but also with its long-term history, showing persistence (memory effect) (Cuddington and Yodzis 1999; Gao et al. 2006; Saraiva et al. 2017).

This long-range correlation and persistence could be a reflex of the many kinds of electrical oscillations present in plant’s electrome. Such oscillations shall be involved with long-distance electrical signalling (Hedrich et al. 2016; Saraiva et al. 2017); first, because in this case the osmotic stimulation was given to roots and electrome records were made in shoots; second, because non-random oscillations and coloured noises are associated with information content and transduction (Cuddington and Yodzis 1999; Gao et al. 2006; Vodeneev et al. 2016a, b; Saraiva et al. 2017).

Notwithstanding, the visual analysis of each original time series (before and after osmotic stimuli) has allowed the observation of spikes up to 500 μV only after the stimulus (Saraiva et al. 2017). The tail of the distribution of the ∣∆V∣ magnitudes of time series after stimuli, exhibited average μ = 2.15 ± 0.08 in the log–log plot of the pdf (D (∣∆V∣) = ∣∆V−μ), after osmotic stimulation (Saraiva et al. 2017), which indicates a power law distribution (Edwards et al. 2007).

The distribution of ∆Vs following a power law suggests some kind of collective behaviour underpinning the electrome, probably the collective behaviour of the  cells during their membrane potential oscillations (Jensen 1998; Masi et al. 2009; Baluška and Mancuso 2013b, Debono 2013a, b; Saraiva et al. 2017). It could even be a consequence of a certain degree of cells synchronisation (Masi et al. 2009; Saraiva et al. 2017). 1/f-like power law distributions, specifically when 1 < μ < 3, are signatures of scale invariance (Edwards et al., 2007). In that case, this means that ∆Vs of plant electrome have not a characteristic size under osmotic stress, i.e. it is scale-free (Schroeder 1991; Saraiva et al. 2017). Al Khazaaly et al. (2009) observed 1/f2 distribution in membrane potential fluctuation of Chara australis R.Br. cells under saline conditions, indicating that scale-free distribution of voltage potential variations is not an uncommon phenomenon among plants under circumstances like high salinity and high osmotic potential.

It is widely assumed that the presence of many components interacting over a range of time or space scales could be a cause for the 1/f-like spectrum in the fluctuations of dynamic systems (Schroeder 1991; Ivanov et al. 2009). According to the Bak-Tang-Wiesenfeld theory (BTW theory) (Bak et al. 1987), a possible explanation for 1/f-like power law distribution is self-organised criticality (SOC). SOC is a ubiquitous phenomenon in nature regardless of the details of the physical system under study, from earthquake magnitudes distribution to population dynamics and species evolution (Bak 1996; Jensen 1998; Lapenna et al. 1998). Human brain EEGs analysis also has shown evidence of SOC (Rubinov et al. 2011; Meisel et al. 2012).

According to Bak et al. (1987), SOC systems are barely stable. Consequently, perturbations might cause a cascade of energy dissipation in all length scales, following a power law. The SOC phenomenon is also related to system responses to constant and frequent external perturbations, for example, tectonic plates stressing each other resulting in earthquake magnitude distribution following a power law (Bak and Tang 1989; Schroeder 1991; Jensen 1998; Lapenna et al. 1998). Our results can be insufficient to indubitably assign that plant electrome ΔV distribution following a power law is a consequence of the SOC phenomenon. However, we speculate that the bursts of spikes could dissipate the tensions caused by constant osmotic stress through the system, enhancing system stability while avoiding its collapse (Saraiva et al. 2017).

Since power law distributions and SOC are associated with complex dynamic systems, the complexity level of the time series (before and after stimuli) was measured by approximate entropy (ApEn). This is a robust method to measure biological time series complexity in terms of irregularity level. It gives a non-negative number to a time series, in which larger values correspond to larger randomness or irregularity, and smaller values correspond to more instances of recognisable features, or patterns, in the data rows (Pincus 1991; Pincus and Goldberger 1994; Souza et al. 2005).

The analysis of ApEn showed a decrease in irregularity of electrome dynamics after stimuli, with higher complexity before stimuli. For instance, Saraiva et al. (2017) observed that ApEn mean values were 1.12 ± 0.21 before stimuli and 0.67 ± 0.39 after stimuli (n = 20). This indicated that soybean plants’ electrome has some complexity level and that this level decreases under osmotic stress stimulation. The complex dynamics of plant low voltage electrical signals was also identified in other situations and with different species (Masi et al. 2009; Cabral et al. 2011; Debono 2013a, b). Thus, the complexity underlying electrical signalling in plants could be a ubiquitous phenomenon, likely carrying information with physiological meaning like APs and VPs among other signals (Saraiva et al. 2017; Fromm and Lautner 2007).

Stability usually is strongly linked with complex dynamic systems (May 1973; Souza and Lüttge 2015). In a biological point of view, it is expected that more complex dynamics allows system stability, providing higher resilience to a system under external disturbances and facilitating real-time fine tuning adjustment (Souza et al. 2005; Souza and Lüttge 2015; Saraiva et al. 2017). Thus, the higher the complexity of the system, the ‘healthier’ it should be. For example, plants that have more complex dynamics in leaf gas exchange under control conditions show better recovery after a water deficit situation (Souza et al. 2004, 2005). Studies with EEG in human brains have also found a correlation between complexity and health. For example, drastic reductions in EEG complexity measured by ApEn have been associated with seizures attacks in humans (Srinivasan et al. 2007). Similarly, analysis of the temporal dynamics of EEG using ApEn revealed that Alzheimer patients had reductions in irregularity of brain waves (Abásalo et al. 2005).

In order to verify the observations by Saraiva et al. (2017), a more extensive study with three different environmental stimuli (osmotic stress, low light and low temperature) was performed (Souza et al. 2017). A naked-eye inspection of the original runs showed notable differences between the electrical signals before and after environmental stimuli. While the baseline of all signals was, in module, 9.6 ± 1.2 µV, after stimuli, spikes up to 500 µV were observed, mainly in the treatments with low light and osmotic stress, confirming previous observations of Saraiva et al. (2017).

The spectral analysis by FFT showed a substantial difference between the runs before and after stimuli. Before the treatments, fundamental frequencies were observed around 0.9 ± 0.6 Hz, a range of mean frequencies of 4.8 ± 0.9 Hz, and higher frequencies around 10.2 ± 0.9 Hz with lower amplitudes. However, after the 3 external stimuli, only the fundamental frequencies remained (Souza et al. 2017). So, it appears that a kind of ‘tuning’ or attention happens in plant electrome, selecting only fundamental frequencies after stimulation. The reduction of the population of frequencies after the stimulation can be a consequence of an increase in the synchronisation of cell membrane potential variation, which turns the signals more organised and regular (Baluška and Mancuso 2013b, Marder 2012, 2013; Neto et al. 2015; Saraiva et al. 2017). This is probably related to complexity decreasing like presented by a reduction in ApEn values after osmotic stimulation (Burioka et al. 2005; Saraiva et al. 2017).

Possibly, selection of fundamental frequencies and complexity decrease are signatures of plant electrome under stressing situations. We had speculated that fundamental frequencies are prioritised because they can carry more distant information (lower frequencies = longer electric oscillations), being part of plant long-distance electric signalling (Hedrich et al. 2016; Huber and Bauerle 2016; Vodeneev et al. 2016a).

The β exponents calculated from power spectral density function (PSD ~ 1/fβ) displayed a range of coloured noises as in Souza et al. (2017). The electrical signals recorded in the plants before stimuli showed a typical reddened noise (β = 1.51 ± 0.21). However, the runs under the different environmental conditions varied between brown (β = 2) and black noise (β = 3). Under cold conditions β = 2.85 ± 0.69, while under low light β = 1.96 ± 0.30, and under osmotic stimuli β = 2.58 ± 0.34 (Souza et al. 2017).

Low light treatment exhibited the lowest β values among the three stimuli. However, there was no significant difference in β values between cold and osmotic stress treatments. Here again, PSD’s β values varying from 1.5 until 3 showed that plant electrome ΔVs have long-range correlations both before and after stimulation (memory effect for β ≠ 0 and β ≠ 2) (Hausdorf and Peng 1996; Cuddington and Yodzis 1999; Gao et al. 2006; Saraiva et al. 2017; Souza et al. 2017).

The tendency of darker noises (β > 2) after different stimulations indicates higher memory effect since persistence is associated with darker noises (Cuddington and Yodzis 1999). Thus, we have speculated about increasing correlations signal persistence as a consequence of an increase in long-distance electrical signalling in plants under stress situations (Hedrich et al. 2016; Vodeneev et al. 2016a; Saraiva et al. 2017; Souza et al. 2017). As plant electrome is a nonlinear dynamic system, it shall be an emergent property of cells electrome dynamics oscillating their membrane voltage potentials in a collective way, transmitting information by electrical waves and spikes to distant and local parts of plants, maybe displaying an essential role in whole plant signalling integration (May 1973; Jensen 1998; Hedrich et al. 2016; Vodeneev et al. 2016a; Choi et al. 2017; Saraiva et al. 2017; Souza et al. 2017).

The log–log plots of the tail of the probability density function (pdf, f (|ΔV|)) of electrome dynamics displayed different distributions for each plant condition. While the histogram before stimuli showed distributions varying from an exponential to Gaussian, there was no unique type of distribution after 3 different stimulations (Souza et al. 2017). However, no run followed indeed a typical Gaussian distribution as expected, since the runs showed a long-range time correlation. Still, it was possible to observe differences in the density of spikes among 3 treatments, with a higher density of spikes under osmotic stress and lower density under cold conditions (Souza et al. 2017).

Furthermore, analysing closer the distribution of spikes (100 ≥ ΔV ≤ 500 µV) an exponential distribution was observed under low temperature; while under low light and osmotic stress the spikes followed a power law distribution, with µ = 2.1 ± 0.3 and µ = 2.5 ± 0.6, respectively (Souza et al. 2017). Thus, these results have supported the hypothesis that plant electrome can be pushed to a self-organised critical state under stimulation, not only osmotic but by others stimuli like, for instance, low-ligh. Low-temperature stimulation probably did not push electrome to a critical state because this situation reduces the activity of ions and their channels, pumps and transporters, consequently affecting the emergence and generation of spikes, remembering that low-temperature stimulation had the lowest spike density (Souza et al. 2017).

As already mentioned, self-organised criticality (SOC) is associated with many different nonlinear dynamic systems, mainly with complex systems under circumstances that involve energy accumulation/dissipation, like earthquakes distributed under a power law (Schroeder 1991). Many biological cases of SOC systems come from neurobiology. A classical example is the distribution of neuron connections of the nematode Caenorhabditis elegans Maupas, 1900 which has many neurons with few connections and few neurons with many connections (Varshney et al. 2011; Towlson et al. 2013). A power law distribution of connections in any biological signalling network seems to be very important for plasticity, stability, efficiency, learning, memory and cognition, including single cells, plants and slime-molds (Trewavas 2016, 2017). Human neurophysiology also shows many examples of SOC, like brain operation under normal states or under cognition tasks as learning or memory (Ivanov et al. 2009; de Arcangelis and Herrmann 2010). Deviations of brain SOC states are usually associated with illnesses such as Parkinson, Alzheimer and seizure attacks (Rubinov et al. 2011; Meisel et al. 2012). On the other hand, in our case, the plant electrome presented SOC only after stressing stimulation (Saraiva et al. 2017; Souza et al. 2017).

Accordingly, we have explored the possibility that constant stress leads to tensions accumulation on electrome that are dissipated in a catastrophic way, in ‘spike-ways’ which spreads through the plant body, likely carrying information too (Saraiva et al. 2017; Souza et al. 2017). However, we are not sure if the plant electrome ΔVs distribution under a power law is involved with information transmission/processing nor if it is just a consequence of energy accumulation/dissipation by stress tensions. Maybe both occur together and are interdependent, similar as calcium, ROS, Ca2+ and electric waves that are at the same time cause and consequence of each other and still have a signalling role in leaf physiology (Gilroy et al. 2016; Trewavas 2016; Choi et al. 2017). However, there are several indications that plant electrome has information (coloured noises, persistence, spectral changes, complexity reduction), that reinforces the signalling role hypotheses. We also considered the possibility of the involvement of plant electrome with plant physiology stability and its interplay with energy efficiency, since SOC appears only after stressing situations (Saraiva et al. 2017; Souza et al. 2017).

In Fig. 4 it is possible to see some examples of plant electrome dynamics before and after three different stimuli (irrigation, salt and osmotic stress). The electrome recordings were done in our laboratory (LACEV) in 2017.
Fig. 4

Example of the records of Phaseolus vulgaris L. electrome dynamics before and after irrigation (distilled water), salt stress (NaCl solution with—2 MPa), and osmotic stress (PEG solution with—2 MPa) stimulations. All three stimuli were given by applying 20 mL of each solution to the root substrate. Each time series was recorded along 2 h with 62 Hz of sampling rate. The measurements ‘before’ and ‘after’ of each treatment were carried with the same individual (same plant). Graphics were plotted using matplotlib

5 The information in plant electrome is classifiable

Considering that plant electrome could be an electrophysiological correspondent of plant physiological status, the existence of stimulus-specific signatures in plant electrome was tested, allowing algorithmic classification. In order to accomplish this task, Pereira et al. (2018) have used two different methods for the automatic classification of plant stress based on the plant electrome. The first approach used Interval Arithmetic to reduce the dimensionality of the input signal, and the second used deep learning for unsupervised feature learning.

Chen et al. (2016) have studied the efficiency of some deep learning methods, such as Artificial Neural Network and Support Vector Machine, comparing with template matching method to classify plant action potentials (APs), obtaining really satisfactory results. In our case, was tested different methods of automatic classification in order to identify when different environmental cues cause specific changes in the plant electrome original raw data (without filters).

Five different supervised classification algorithms were used: Artificial Neural Networks (ANN), Convolutional Neural Network (CNN), Optimum-Path Forest (OPF), k-Nearest Neighbors (K-NN) and Support Vector Machine (SVM). In approach A, the Interval Arithmetic (IA) mapping experiment was used. Afterwards, it was applied the ANN, OPF, k-NN, and SVM classifiers, in order to overcome the problems presented by machine learning application in our kind of data. In approach B, the signal encoding into images by Visual Rhythm as Almeida et al. (2015) for later apply CNN was used (Pereira et al., 2018).

In Pereira et al. (2018), SVM obtained the best results for all datasets, followed by k-NN that obtained statistically similar results in the ‘cold’ and ‘osmotic’ datasets. OPF classifier achieved good accuracies for three datasets, and CNN also obtained good results for ‘cold’ and ‘low light’ datasets, although it obtained the worst results for the ‘osmotic’ dataset. It is possible to observe that AI combined with supervised classifiers can provide better results than deep learning techniques.

The recognition rates presented by Pereira et al. (2018) point to promising results concerning the task of automatically identifying stress-like patterns in plant electrome. In the confusion matrices using 90% of the datasets for training purposes with SVM classifier, it is possible to observe that plant electrome after cold stress has a peculiar pattern that is easier to differentiate from other stress conditions. However, the ‘low light’ and ‘osmotic’ time series presented similar results. Additionally, the experiment with ‘all’ dataset showed that is possible to detect patterns among different stress signals, despite the apparent confusion between ‘low light’ and ‘no stress’ conditions, which may affect the final classifier’s accuracy (Pereira et al. 2018).

Chen et al. (2016), using similar strategies to classify plant’s action potentials had the best performance with template matching algorithm (96.0% of accuracy), while ANN and SVM reached a maximum accuracy of 84.1% and 75.8%, respectively. Our results were pretty close to these ones (Pereira et al. 2018). However, different from Chen et al. (2016), Pereira and colleagues’ (2018) classification algorithms were applied over a continuous time series of low voltage variation that has very complex dynamics as presented before (Saraiva et al. 2017; Souza et al. 2017). Such ‘raw’ complex time series let such kind of analysis more difficult, leading tendencies to decrease pattern recognition by the classifiers used herein. Such results are pointing to possible uses of plant electrome dynamics and classification algorithms to identify physiological disturbs caused by external perturbations.

6 Plants as electromic cognitive systems: conclusions and further perspectives

Considering the complex traits of the electrome presented above, we have speculated that emergent global states of receptivity conducting to cognitive states could be supported by the bioelectrical continuum ensured by the electrome using SOCs and common sensory networks, particularly at the level of anastomoses and stem or root tissues. It also gives a solid working hypothesis as to the effectiveness of the background EPG activities sustained by the plant electrome (Debono 2013a, b; Calvo et al. 2017).

Concerning the generating source for the electrome, Masi et al. (2009) have clearly validated the permanent emission of spontaneous EPG low voltage signals at the plant cell surface with the multi-electrode array (MEA) technology, and also have shown the presence of spatiotemporal synchronised oscillatory activities at maize root apices. Following this hypothesis, modular bioelectrical activities of collective groups of interconnected cells that could be synchronised by internal or external stimuli, together with plant electrome managed interfaces, would constitute the dynamic proto-neural networks ensuring constant informational, learning, and cognitive processing at the whole plant level (Baluška, 2013; Baluška and Mancuso 2013a, b; Debono 2013a, b; de Loof 2016; Trewavas 2016).

Several modules could favour this continuity and participate in the emergence of other complex systems of plants. For instance, the epidermis, sometimes multi-stratified, which continuity is interrupted by stomata, prevents mechanical damage and limits water loss, while allowing gas exchange between the plant and the surrounding O2/CO2 and transpiration. The network of semi-autonomous stomata often shows a very complex pattern of open-closing dynamics, named as patch stomatal conductance, optimising the gas exchange interface with the external air space (Mott and Buckley 2000; Souza et al. 2004, Mousavi et al. 2014).

Another example is roots managing nutrients and water spatiotemporal distribution for the best development of the plant. In the root tips, where synchronised oscillations are described specifically in the apical transition zone, they are assumed to drive sensory inputs or sensory-motoric networks enabling the growing apex to permanently monitor environmental data and gravity, growing in an intentional way, even making decisions in mazes (Masi et al. 2009, 2015; Baluška 2013; Baluška and Mancuso 2013b; Marder 2012, 2013; Yokawa et al. 2014).

The complex and scale-invariant nature of typical low voltage spontaneous and evoked EPG patterns (Karlsson 1972; Debono and Bouteau 1992; Masi et al. 2009), as well as that of the release of evoked spikes followed by a power law of distribution (Saraiva et al. 2017), reflect together with other cellular events (local, systemic, variation potentials, Ca2+ signalling), the whole plant electrome signature. This signature is assumed to cover all the bioelectrical spectrum of the embodied-cognitive plant structure in its environment. We can thus consider that the billions of years of eukaryotic proto-neural organisation, during the course of evolution, have probably determined, or pre-built, the conditions for the emergence of future neural structures of plants and animals (Debono 2013a, b; de Loof 2016; Trewavas 2016, 2017).

This opens new ways of investigation for the study of biosensors using bioelectricity as main parameters. Indeed, as previously stated by Souza et al. (2017), at the era of omics, or signalomics (Vian et al. 2015), the consideration of the plant electrome permits to specify the electrical profile of a species, an organism or even a given individual (De Loof 2016). It could be generalised to the metaplasticity of living systems (Debono 2008), taking into account their different degrees of interaction with the environment (perception levels, phenotypic and epigenetic plasticity, etc.) or with their natural habitat (mesological plan, Berque 2017). In other terms, we suggest that the electrome could be the key to understand the cognitive nature of plants by attributing to its eco-perceptual nature a continuity, or unity, that overcomes its organic modularity.

This assumption has two implications: 1- the consideration of plants as conducting organisms able to coordinate their physiological activities by means of electrical excitations, and transmissions, and also to integrate the perception–action loop in all their behaviours without the need of central nervous systems; 2- the consideration of the evolutionary plan and the plant’s use of common receptor gene superfamilies from bacteria to humans, so integrating previous generic processes becoming faster, or more specialised, but above all, more differentiated.

The electrome could act as a filter by discriminating critically thresholds or windows modulating complex network signalling at the whole plant level. The result can be better emergent behaviours, caused by enhancement of the correlation between modules (de Kroon et al. 2005), as well as improving the learning levels by plant-plant or interspecies communication, and finally an adequate or more optimal response to the complexity of environmental stimuli surrounding plants.

Finally, we offered an ‘insightful food for thought’, for plant learning and cognition issues, e.g. pointing a likely analogy between the SOC behaviour in human brain electrical activity related with learning process, and the SOC behaviour observed in plant electrome during stressful situations, suggesting a common electrical dynamic for the learning process, regardless taxa (de Arcangelis and Herrmann 2010; Gagliano et al. 2016; Souza et al. 2017). Notwithstanding, we think that plant electrome could be used to test hypotheses in plant neurobiology, beyond being useful for identification and monitoring of plant physiological status, or even likely consciousness states, similarly to what EEGs and ECGs are for humans (Brenner 1985; Brenner et al. 2007; Gardiner 2012, 2013; Calvo 2016; Calvo et al. 2017).



This work was supported by CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), Capes (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) and Stoller do Brasil.

Supplementary material

40626_2019_145_MOESM1_ESM.docx (34 kb)
Electronic supplementary material 1 (DOCX 35 kb)


  1. Abásalo D, Hornero R, Espino P, Poza J, Sánches C, de La Rosa R (2005) Analysis of regularity in the EEG background activity of Alzheimer’s disease patients with approximate entropy. Clin Neurophys 116:1826–1834Google Scholar
  2. Al Khazaaly S, Walker NA, Beilby MJ, Shepherd VA (2009) Membrane potential fluctuations in Chara australis: a characteristic signature of high external sodium. Eur Biophys J 39:167–174Google Scholar
  3. Almeida JG, dos Santos JA, Alberton B, Morellato LPC, Torres RS (2015) Phenological visual rhythms: compact representations for fine-grained plant species identification. Pattern Recogn Lett. CrossRefGoogle Scholar
  4. Bak P (1996) How nature works: the science of self-organized criticality. Springer, New YorkGoogle Scholar
  5. Bak P, Tang C (1989) Earthquakes as a self-organized critical phenomenon. J Geophys Res 94:635–637Google Scholar
  6. Bak P, Tang C, Wiesenfield K (1987) Self-organized criticality: an explanation of 1/f noise. Physic Rev Lett 59:381–384Google Scholar
  7. Baluška F (ed) (2013) Long-distance systemic signaling and communication in plants. Springer, BerlinGoogle Scholar
  8. Baluška F, Mancuso S (2009a) Plant neurobiology: from sensory biology, via plant communication, to social plant behavior. Cogn Process 10(Suppl 1):S3–S7. Google Scholar
  9. Baluška F, Mancuso S (2009b) Deep evolutionary origins of neurobiology: turning the essence of ‘neural’ upside-down. Commun Integr Biol 2(1):60–65Google Scholar
  10. Baluška F, Mancuso S (2013a) Ion channels in plants: from bioelectricity, via signaling, to behavioral actions. Plant Signal Behav 8(1):e23009Google Scholar
  11. Baluška F, Mancuso S (2013b) Root apex transition zone as a oscillatory zone. Front Plant Sci 4(354):1–15Google Scholar
  12. Baluška F, Miller WB (2018) Senomic view of the cell: senome versus genome. Commun Integr Biol 11:1–9. PubMedPubMedCentralGoogle Scholar
  13. Baluška F, Mancuso S, Volkamann D (2006) Communication in plants: neuronal aspects of plant life. Springer, BerlinGoogle Scholar
  14. Behrens HM, Gradmann D, Sievers A (1985) Membrane-potential responses following gravistimulation in roots of Lepidium sativum L. Planta 163:463–472Google Scholar
  15. Berque A (2017) Qu’est-ce que la mésologie, Blog Mésologiques. Accessed 14 Oct 2018Google Scholar
  16. Bialasek M, Górecka M, Mittler R et al (2017) Evidence for the involvement of electrical, calcium and ROS signaling in the systemic regulation of non-photochemical quenching and photosynthesis. Plant Cell Physiol 58:207–215. Google Scholar
  17. Bose JC (1926) The nervous mechanism of plants. Longman, Green & Co, New YorkGoogle Scholar
  18. Bray D (1995) Protein molecules as computational elements in living cells. Nature 376:307–312Google Scholar
  19. Brenner RP (1985) The electroencephalogram in altered states of consciousness. Neurol Clin 3:615–631Google Scholar
  20. Brenner ED, Stahlberg R, Mancuso S, Vivanco JM, Baluška F, van Volkenburgh E (2007) Plant neurobiology: an integrated view of plant signaling. Trend Plant Sci 11(8):413–419Google Scholar
  21. Buchanan BB, Gruissem W, Jones RL (2015) Biochemistry and molecular biology of plants, 2nd edn. Wiley-Blackwell, Hoboken, NJGoogle Scholar
  22. Burbach C, Markus K, Zhang Y, Schlicht M, Baluška F (2012) Photophobic behavior of maize roots. Plant Signal Behav 7:874–878Google Scholar
  23. Burdon-Sanderson J (1873) Note on the electrical phenomena which accompany stimulation of the leaf of Dionea muscipula. Proc R Soc 21:495–496Google Scholar
  24. Burioka N, Miyata M, Cornelissen G, Halberg F, Takeshima T, Kaplan DT, Suyama H, Endo M, Maegaki Y, Nomura T, Tomita Y, Nakashima K, Shimuzu E (2005) Approximate entropy in the electroencephalogram during wake and sleep. Clin EEG Neurosc 36:21–24Google Scholar
  25. Cabral EF, Pecora PC, Arce AIC, Tech ARB, Costa EJX (2011) The oscillatory bioelectrical signal from plants explained by a simulated electrical model and tested using Lempel-Ziv complexity. Comput & Electron in Agric 76:1–5Google Scholar
  26. Calvo P (2016) The philosophy of plant neurobiology: a manifesto. Synthese 193:1323–1343. Google Scholar
  27. Calvo P, Sahi VP, Trewavas A (2017) Are plants sentient? Plant, Cell Environ 40:2858–2869Google Scholar
  28. Canales J, Henriquez-Valencia C, Brauchi S (2018) The integration of electrical signals originating in the root of vascular plants. Front Plant Sci 8:2173. Google Scholar
  29. Catania KC (2016) Leaping eels electrify threats, supporting Humboldt’s account of a battle with horses. Proc Natl Acad Sci USA 113:6979–6984. Google Scholar
  30. Chen Y, Zhao DJ, Wang ZY, Wang ZY, Tang G, Huang L (2016) Plant electrical signal classification based on waveform similarity. Algorithms. CrossRefGoogle Scholar
  31. Choi W-G, Hilleary R, Swanson SJ, Kim S-H, Gilroy S (2016) Rapid, long-distance electrical and calcium signaling in plants. Annu Rev Plan Biol 67(1):287–307. Google Scholar
  32. Choi WG, Miller G, Wallace I et al (2017) Orchestrating rapid long-distance signaling in plants with Ca2+, ROS and electrical signals. Plant J 90:698–707. Google Scholar
  33. Clarke D, Whitney H, Sutton G, Robert D (2013) Detection and learning of floral electric fields by bumblebees. Science 340:66–69Google Scholar
  34. Collings DA, White RG, Overall RL (1992) Ionic current changes associated with the gravity-induced bending response in roots of Zea mays L. Plant Phys 100:1417–1426Google Scholar
  35. Cuddington KM, Yodzis P (1999) Black noise and population persistence. Proc R Soc Lond B 266:969–973. Google Scholar
  36. Davies E (1993) Intercellular and intracellular signals and their transduction via the plasma membrane—cytoskeleton interface. Semin Cell Biol 4:139–147Google Scholar
  37. de Arcangelis L, Herrmann HJ (2010) Learning as phenomenon occurring in a critical state. Proc Nat Acad Sci USA 107:3977–3981. Google Scholar
  38. de Kroon H, Huber H, Stuefer JF, van Groenendael JM (2005) A modular concept of phenotypic plasticity in plants. New Phytol 166:73–82Google Scholar
  39. de Loof A (1986) The electrical dimension of cells: the cell as a miniature electrophoresis chamber. lnt Rev Cytol 88:251–352Google Scholar
  40. de Loof A (2016) The cell’s self-generated “electrome”: the biophysical essence of the immaterial dimension of life? Commun Integr Biol 1:e1197446Google Scholar
  41. Debono MW (2008) Transdisciplinarity: a new approach to metadynamics and consciousness. In: Nicolescu B (ed) Transdisciplinarity: theory & practice. Hampton Press Creskill, New JerseyGoogle Scholar
  42. Debono MW (2013a) Dynamic proto-neural networks in plants: a new approach of spontaneous extracellular potential variations. Plant Signal Behav 8:e24207Google Scholar
  43. Debono M (2013b) Perceptive levels in plants: a transdisciplinary challenge in living organism’s plasticity. Transdiscipl J Eng Sci 4:21–39Google Scholar
  44. Debono MW, Bouteau F (1992) Spontaneous and evoked surface potentials in Kalanchoe tissues. Life Sci Adv Plant Physiol 11:107–117Google Scholar
  45. Edwards AM, Phillips RA, Watkins NW, Freeman MP, Murphy EJ, Afanasyev V, Buldyrev SV, da Luz MGE, Raposo EP, Stanley HE (2007) Revisiting Lévy flight search patterns of wandering albatrosses, bumblebees and deer. Nature 449:1044–1049Google Scholar
  46. Evans MJ, Choi WG, Gilroy S, Morris RJA (2016) ROS-assisted calcium wave dependent on the AtRBOHD NADPH oxidase and TPC1 cation channel propagates the systemic response to salt stress. Plant Phys 171(3):1535–1539. Google Scholar
  47. Fletcher DA, Mullins RD (2010) Cell mechanics and the cytoskeleton. Nature 463:485–492. Google Scholar
  48. Frohlich F, McCormick DA (2010) Endogenous electric fields may guide neocortical network activity. Neuron 67:129–143Google Scholar
  49. Fromm J, Bauer T (1994) Action potentials in maize sieve tubes change phloem translocation. J Exp Bot 45:463–469Google Scholar
  50. Fromm J, Lautner S (2007) Electrical signals and their physiological significance in plants. Plant, Cell Environ 30:249–257Google Scholar
  51. Gagliano M, Vyazovskiy VV, Borbély AA, Grimonprez M, Depczynski M (2016) Learning by association in plants. Sci Rep 6:38427. Google Scholar
  52. Gallé A, Lautner S, Flexas J et al (2013) Photosynthetic responses of soybean (Glycine max L.) to heat-induced electrical signaling are predominantly governed by modifications of mesophyll conductance for CO2. Plant, Cell Environ 36:542–552. Google Scholar
  53. Gallé A, Lautner S, Flexas J et al (2015) Environmental stimuli and physiological responses: the current view on electrical signaling. Environ Exp Bot 114:15–21. Google Scholar
  54. Gao J, Hu J, Tung W-W, Cao Y, Sarshar N, Roychowdhury VP (2006) Assessment of long-range correlation in time series: how to avoid pitfalls. Phys Rev E 73:016117Google Scholar
  55. Gardiner J (2012) Insights into plant consciousness from neuroscience, physics and mathematics: a role for quasicrystals? Plant Signal Behav 7:1049–1055. Google Scholar
  56. Gardiner J (2013) Fractals and the irreducibility of consciousness in plants and animals. Plant Signal Behav 8(8):e25296Google Scholar
  57. Genoud T, Métraux J-P (1999) Crosstalks in plant cell signaling: structure and function of the genetic network. Trends Plant Sci 4:503–507Google Scholar
  58. Gensler W, Yan TL (1988) Investigation of the causative reactant of the apoplast electropotentials of plants. J Electrochem Soc Electrochem Sci Technol 135:2991–2995Google Scholar
  59. Gilroy S, Suzuki N, Miller G, Choi WG, Toyota M, Devireddy AR, Mittler R (2014) A tidal wave of signals: calcium and ROS at the forefront of rapid systemic signaling. Trends Plant Sci 19:623–630. Google Scholar
  60. Gilroy S, Białasek M, Suzuki N, Górecka M, Devireddy AR, Karpínski S, Mittler R (2016) ROS, calcium, and electric signals: key mediators of rapid systemic signaling in plants. Plant Physiol 171:1606–1615Google Scholar
  61. Grams T, Koziolek C, Lautner S et al (2007) Distinct roles of electric and hydraulic signals on the reaction of leaf gas exchange upon re-irrigation in Zea mays L. Plant, Cell Environ 30:79–84. Google Scholar
  62. Hausdorf JM, Peng C-K (1996) Multiscaled randomness: a possible source of 1/f noise in biology. Phys Rev E 54(2):2154–2157. Google Scholar
  63. Hedrich R (2012) Ion channels in plants. Physiol Rev 92:1777–1811. Google Scholar
  64. Hedrich R, Salvador-Recatalà V, Dreyer I (2016) Electrical wiring and long-distance plant communication. Trend Plant Sci 21(5):376–387Google Scholar
  65. Herd O, Peña-Cortés H, Fuss H et al (1999) Effects of mechanical wounding, current application and heat treatment on chlorophyll fluorescence and pigment composition in tomato plants. Physiol Plant 105:179–184. Google Scholar
  66. Higinbotham N (1973) Electro potentials of plant cells. Ann Rev Plan Phys 24:25–46Google Scholar
  67. Hlavinka J, Nozková-Hlavácková V, Floková K et al (2012) Jasmonic acid accumulation and systemic photosynthetic and electrical changes in locally burned wild type tomato, ABA-deficient sitiens mutants and sitiens pre-treated by ABA. Plant Physiol Biochem 54:89–96. Google Scholar
  68. Honigmann A, Pralle A (2016) Compartmentalization of the cell membrane. J Mol Biol 428:4739–4748Google Scholar
  69. Huber AE, Bauerle TL (2016) Long-distance plant signaling pathways in response to multiple stressors: the gap in knowledge. J Exp Bot 67:2063–2079Google Scholar
  70. Ivanov PC, Ma QDY, Barstsch RP, Hausdorf JM, Amaral LAN, Schulte-Frohlinde V, Stanley HE, Yoneyama M (2009) Levels of complexity in scale invariance neural signals. Phys Rev E 79:041920–1–041920-12Google Scholar
  71. Iwabuchi A, Yano M, Shimizu H (1989) Development of extracellular electric pattern around Lepidium roots: its possible role in root growth and gravitropism. Protoplasma 148:94–100Google Scholar
  72. Jaffe LF (1977) Electrophoresis along cell membranes. Nature 265:600–602Google Scholar
  73. Jaffe LF, Robinson KR, Nuccitelli R (1974) Local cation entry and self-electrophoresis as an intracellular localization mechanism. In: Electrically mediated growth mechanisms in living systems, vol 238. Annals of the New York Academy of Sciences USA, pp 372–389Google Scholar
  74. Jensen HJ (1998) Self-organized criticality: emergent complex behaviour in physical and biological systems. Cambridge Lecture Notes in Physics. Cambridge University Press, Cambridge, UKGoogle Scholar
  75. Jones JGD, Dangl JL (2006) The plant immune system. Nature 444:323–329Google Scholar
  76. Josberger EE, Hassanzadeh P, Deng Y, Sohn RMJ, Amemiya CT, Rolandi M (2016) Proton conductivity in ampullae of Lorenzini jelly. Sci Adv 2:e1600112. Google Scholar
  77. Karlsson L (1972) Nonrandom bioelectrical signals in plant tissue. Plant Physiol 49:982–986Google Scholar
  78. Krapf D (2018) Compartmentalization of the plasma membrane. Curr Opin Cell Biol 53:15–21Google Scholar
  79. Kroeger JH, Zerzour R, Geitmann A (2011) Regulator or driving force? The role of turgor pressure in oscillatory plant cell growth. PLoS ONE 6(4):e18549. Google Scholar
  80. Krupenina N, Bulychev A (2007) Action potential in a plant cell lowers the light requirement for non-photochemical energy-dependent quenching of chlorophyll fluorescence. Biochim Biophys Acta 1767:781–788. Google Scholar
  81. Kwaaitaal M, Huisman R, Maintz J et al (2011) Ionotropic glutamate receptor (iGluR)-like channels mediate MAMP-induced calcium influx in Arabidopsis thaliana. Biochem J 440:335–365Google Scholar
  82. Lapenna V, Macchiato M, Telesca L (1998) 1/f fluctuations and self-similarity in earthquake dynamics: observation evidences in southern Italy. Phys Earth Planet Inter 106:115–127Google Scholar
  83. Lee JY (2015) Plasmodesmata: a signaling hub at the cellular boundary. Curr Opin Plant Biol 27:133–140. Google Scholar
  84. Levin M, Pezzulo G, Finkelstein JM (2017) Voltage gradients for control of growth and form. Annu Rev Biomed Eng 19:353–387Google Scholar
  85. MacRobbie EAC (1962) Ionic relations in Nitella translucens. J Gener Phys 45:861–878Google Scholar
  86. MacRobbie EAC (2006) Control of volume and turgor in stomatal guard cells. J Membr Biol 210:131–142. Google Scholar
  87. Mancini R (2002) Op amps for everyone. Texas Instrum Des Ref. CrossRefGoogle Scholar
  88. Marder M (2012) Plant intentionality and the phenomenological framework of plant intelligence. Plant Signal Behav 7(11):1–8Google Scholar
  89. Marder M (2013) Plant intelligence and attention. Plant Signal Behav 8(5):e23902. Google Scholar
  90. Masi E, Ciszak M, Stefano G, Renna L, Azzarello E, Pandolfi C, Mugnai S, Baluška F, Arecchi FT, Mancuso S (2009) Spatio-temporal dynamics of the electrical network activity in the root apex. Proc Natl Acad Sci USA 106:4048–4053. Google Scholar
  91. Masi E, Ciszak M, Comparini D, Monetti E, Pandolfi C, Azzarello E, Mugnai S, Baluška F, Mancuso S (2015) The electrical network of maize root apex is gravity dependent. Sci Rep 5:7730. Google Scholar
  92. Matzke AJM, Weiger TM, Matzke M (2010) Ion channels at the nucleus: electrophysiology meets the genome. Mol Plant 3(4):642–652Google Scholar
  93. May RM (1973) Stability and complexity in model ecosystems. Princeton University Press, PrincetonGoogle Scholar
  94. Meisel C, Storch A, Hallmeyer-Elgner S, Bullmore E, Gross T (2012) Failure of adaptive self-organized criticality during epileptic seizure attacks. PLoS Comput Biol 8:e1002312Google Scholar
  95. Miller AL, Gow NAR (1989) Correlation between root-generated ionic currents, pH, fusicoccin, indoleacetic acid, and growth of the primary root of Zea mays. Plant Phys 89:1198–1206Google Scholar
  96. Miller G, Schlauch K, Tam R, Cortes D, Torres MA, Shulaev V (2009) The plant NADPH oxidase RBOHD mediates rapid systemic signaling in response to diverse stimuli. Sci Signal 2:ra45. CrossRefGoogle Scholar
  97. Mott KA, Buckley TN (2000) Patchy stomatal conductance emergent collective behaviour. Trends Plant Sci 5:258–262Google Scholar
  98. Mousavi SAR, Chauvin A, Pascaud F, Kellenberger S, Farmer EE (2013) GLUTAMATE RECEPTOR-LIKE genes mediate leaf-to-leaf wound signaling. Nature 500:422–440Google Scholar
  99. Mousavi SAR, Nguyen CT, Farmer EE, Kellenberger S (2014) Measuring surface potential changes on leaves. Nat Protocol 9(8):1997–2004Google Scholar
  100. Neill SJ, Desikan R, Clarke A, Hancock JT (2002a) Nitric oxide is a novel component of abscisic acid signaling in stomatal guard cells. Plant Physiol 128:13–16. Google Scholar
  101. Neill SJ, Desikan R, Hancock J (2002b) Hydrogen peroxide signaling. Curr Opin Plant Biol 5:388–395. Google Scholar
  102. Neto E, Allen EA, Aurlien H, Nordby H, Eichele T (2015) EEG spectral features discriminate between Alzheimer’s and vascular dementia. Front Neurol 6(25):1Google Scholar
  103. Nobel PS (1999) Physicochemical & environmental plant physiology, 2nd edn. Academic press, San DiegoGoogle Scholar
  104. Novikova EM, Vodeneev VA, Sukhov VS (2017) Mathematical model of action potential in higher plants with account for the involvement of vacuole in the electrical signal generation. Membr Cell Biol 11(2):151–167Google Scholar
  105. Nuccitelli R, Robinson K, Jaffe L (1986) On electrical currents in development. BioEssays 5:292–294Google Scholar
  106. Olivotto M, Arcangeli A, Carlà M, Wanke E (1996) Electric fields at the plasma membrane level: a neglected element in the mechanisms of cell signalling. BioEssays 18:495–504Google Scholar
  107. Pavlovic A, Slováková L, Pandolfi C et al (2011) On the mechanism underlying photosynthetic limitation upon trigger hair irritation in the carnivorous plant Venus flytrap (Dionaea muscipula Ellis). J Exp Bot 62:1991–2000. Google Scholar
  108. Pereira DN, Papa JP, Saraiva GFR, Souza GM (2018) Automatic classification of plant electrophysiological responses to environmental stimuli using machine learning and interval arithmetic. Comput Electron Agric 145:35–42Google Scholar
  109. Petchey OL, Gonzalez A, Wilson HB (1997) Effects on population persistence: the interaction between environmental noise color, intraspecific competition and space. Proc R Soc Lond 264:1841–1847Google Scholar
  110. Pickard BG (1973) Action potentials in higher plants. Bot Rev 39:172–201Google Scholar
  111. Pincus SM (1991) Approximate entropy as a measure of system complexity. Proc of Nat Acad Sci USA 88:2297–2301Google Scholar
  112. Pincus SM, Goldberger AL (1994) Physiological time-series analysis: what does regularity quantify? Am J Phys 266:1643–1656Google Scholar
  113. Rayle DL, Cleland RE (1992) The acid growth theory of auxin-induced cell elongation is alive and well. Plant Phys 99:1271–1274Google Scholar
  114. Rubinov M, Sporms O, Thivierge J-P, Breakspear M (2011) Neurobiologically realistic determinants of self-organized criticality in networks of spiking neurons. PLoS Comput Biol 7:e1002038Google Scholar
  115. Saraiva GRF, Ferreira AS, Souza GM (2017) Osmotic stress decreases complexity underlying the electrophysiological dynamic in soybean. Plant Biol. CrossRefGoogle Scholar
  116. Schroeder M (1991) Fractals, chaos, power laws: minutes from an infinite paradise. W.H. Freeman, New YorkGoogle Scholar
  117. Scott BIH (1967) Electric fields in plants. Annu Rev Plant Phys 18:409–418Google Scholar
  118. Scott P (2008) Physiology and behaviour of plants. Wiley, West SussexGoogle Scholar
  119. Segal AW (2016) NADPH oxidases as electrochemical generators to produce ion fluxes and turgor in fungi, plants and humans. Open Biol 6:160028. Google Scholar
  120. Shabala S, White RG, Djordjevic MA, Ruan YL, Mathesius U (2016) Root-to-shoot signaling: integration of diverse molecules, pathways and functions. Funct Plant Biol 43:87–104. Google Scholar
  121. Shepherd VA (2005) From semi-conductors to the rhythms of sensitive plants: the research of J.C. Bose. Cell Mol Biol 51:607–619Google Scholar
  122. Shepherd VA (2012) At the roots of plant neurobiology. In: Volkov AG (ed) Plant electrophysiology. Springer, Berlin, pp 3–43. Google Scholar
  123. Souza GM, Lüttge U (2015) Stability as a phenomenon emergent from plasticity–complexity–diversity in eco-physiology. Prog Bot 76:211–239Google Scholar
  124. Souza GM, Oliveira RF, Cardoso VJM (2004) Temporal dynamics of stomatal conductance of plants under water deficit: can homeostasis be improved by more complex dynamics. Arq de Biol e Tecnol 47:423–431Google Scholar
  125. Souza GM, Pincus SM, Monteiro JAF (2005) The complexity–stability hypothesis in plant gas exchange under water deficit. Braz J Plant Phys 17:363–373Google Scholar
  126. Souza GM, Ferreira AS, Saraiva GFR, Toledo GRA (2017) Plant “electrome” can be pushed toward a self-organized critical state by external cues: evidences from a study with soybean seedlings subject to different environmental conditions. Plant Signal Behav 12(3):e1290040. Google Scholar
  127. Spanswick RM (1981) Electrogenic ion pumps. Rev Plant Phys 32:67–89Google Scholar
  128. Spanswick RM, Williams EJ (1964) Electrical potentials and Na, K, and Cl concentrations in the vacuole and cytoplasm of Nitella translucens. J Exp Bot 15:193–200Google Scholar
  129. Srinivasan V, Eswara C, Sriraam N (2007) Approximate entropy-based epileptic EEG detection using artificial neural networks. IEEE Trans Inf Technol Biomed 11:288–295Google Scholar
  130. Stahlberg R (2006) Historical overview on plant neurobiology. Plant Signal Behav 1(1):6–8Google Scholar
  131. Stahlberg R, Cleland RE, van Volkenburgh E (2006) Slow wave potentials: a propagating electrical signal unique to higher plants. In: Baluška F, Mancuso S, Volkmann D (eds) Communication in plants: neuronal aspects of plant life. Springer, New York, pp 291–308Google Scholar
  132. Stanković B (2006) Electrophysiology of plant gravitropism. In: Volkov AG (ed) Plant electrophysiology: theory and methods, 1st edn. Springer, BerlinGoogle Scholar
  133. Stavrinidou E, Gabrielsson R, Gomez E, Crispin X, Nilsson O, Simon DT, Berggren M (2015) Electronic plants. Sci Adv 1(10):e1501136Google Scholar
  134. Stoddard PK (1999) Predation enhances complexity in the evolution of electric fish signals. Nature 400:254–256Google Scholar
  135. Sukhov V (2016) Electrical signals as mechanism of photosynthesis regulation in plants. Photosynth Res 130:373–387. Google Scholar
  136. Sukhov V, Orlova L, Mysyagin S et al (2012) Analysis of the photosynthetic response induced by variation potential in geranium. Planta 235:703–712. Google Scholar
  137. Sukhov V, Sherstneva O, Surova L et al (2014) Proton cellular influx as a probable mechanism of variation potential influence on photosynthesis in pea. Plant, Cell Environ 37:2532–2541. Google Scholar
  138. Sukhov V, Surova L, Sherstneva O et al (2015) Variation potential influence on photosynthetic cyclic electron flow in pea. Front Plant Sci 5:766. Google Scholar
  139. Sukhova E, Akinchits E, Sukhov V (2017) Mathematical models of electrical activity in plants. J Membr Biol 250:407–423Google Scholar
  140. Surova L, Sherstneva O, Vodeneev V et al (2016) Variation potential propagation decreases heat-related damage of pea photosystem I by 2 different pathways. Plant Sign Behav 11:e1145334. Google Scholar
  141. Suzuki N, Miller G, Salazar C, Mondal HA, Shulaev E, Cortes DF, Shuman JL, Luo X, Shah J, Schlauch K, Shulaev V, Mittler R (2013) Temporal-spatial interaction between reactive oxygen species and abscisic acid regulates rapid systemic acclimation in plants. Plant Cell 25:3553–3569. Google Scholar
  142. Szechyńska-Hebda M, Lewandowska M, Karpiński S (2017) Electrical signaling, photosynthesis and systemic acquired acclimation. Front Physiol 8:684. Google Scholar
  143. Taiz E, Zeiger E, Møller IM, Murphy A (2014) Plant physiology and development, 6th edn. Sinauer Associates, SunderlandGoogle Scholar
  144. Thellier M, Lüttge U (2012) Plant memory: a tentative model. Plant Biol 15:1–12Google Scholar
  145. Towlson EK, Vertes PE, Ahnert SE, Schafer WR, Bullmore ET (2013) The rich club of the C. elegans neuronal connectome. J Neurosci 33:6380–6387. Google Scholar
  146. Trebacz K, Dziubinska H, Król E (2006) Electrical signals in long-distance communication in plants. In: Baluška F, Mancuso S, Volkmann D (eds) Communication in plants. Neuronal aspects of plant life. Springer, Berlin, pp 277–290Google Scholar
  147. Trewavas A (2016) Intelligence, cognition, and language of green plants. Front Psychol 7:588. Google Scholar
  148. Trewavas A (2017) The foundations of plant intelligence. Interface Focus 7:20160098Google Scholar
  149. Trewavas A, Baluška F (2011) The ubiquity of consciousness. EMBO Rep 12(12):1221–1225Google Scholar
  150. Tseng A-S, Levin M (2013) Cracking the bioelectric code: probing endogenous ionic controls of pattern formation. Commun Integr Biol 6:1–8Google Scholar
  151. Umrath K (1930) Untersuchungen über Plasma und Plasmaströmungen an Characeen. IV. Potentialmessungen an Nitella mucronata mit besonderer Berücksichtigung der Erregungserscheinungen. Protoplasma 9:576–597Google Scholar
  152. van Bel AJE, Furch ACU, Will T, Buxa SV, Musetti R, Hafke JB (2014) Spread the news: systemic dissemination and local impact of Ca2+ signals along the phloem pathway. J Exp Bot 65:1761–1787. Google Scholar
  153. Varshney LR, Chen BL, Paniagua E, Hall DH, Chklovskii DB (2011) Structural properties of the Caenorhabditis elegans neuronal network. PLoS Comput Biol 7:e1001066. Google Scholar
  154. Vian A, Stankovic B, Davies E (2015) Signalomics: diversity and methods of analysis of systemic signals in plants. In: Plantomics: the omics of plant science. Springer, Berlin, pp 459–490Google Scholar
  155. Vodeneev VA, Katicheva LA, Sukhov VS (2016a) Electrical signals in higher plants: mechanisms of generation and propagation. Biophysics 61(3):505–512Google Scholar
  156. Vodeneev V, Akinchits E, Sukhov V (2016b) Variation potential in higher plants: mechanisms of generation and propagation. Plant Signal Behav 10:e1057365. Google Scholar
  157. Volkov AG (2006) Plant electrophysiology: theory and methods, 1st edn. Springer, BerlinGoogle Scholar
  158. Volkov AG, Collins DJ, Mwesigwa J (2000) Plant electrophysiology: pentachlorophenol induces fast action potentials in soybean. Plant Sci 153:185–190Google Scholar
  159. Volkov AG, Carrell H, Markin VS (2009) Biologically closed electrical circuits in venus flytrap. Plant Phys 149:1661–1667Google Scholar
  160. Weisenseel MH, Nuccitelli R, Jaffe L (1975) Large electrical currents traverse growing pollen tubes. J Cell Biol 66:556–567Google Scholar
  161. Weisenseel MH, Dorn A, Jaffe L (1979) Natural H+ currents traverse growing roots and root hairs of barley (Hordeum vulgare L.). Plant Phys 64:512–518Google Scholar
  162. Wildon DC, Thain JF, Minchn PEH, Gubb IR, Reily AJ, Skipper YD, Doherty HM, O’Donnel PJ, Bowles DJ (1992) Electrical signaling and systemic proteinase inhibitor induction in the wounded plant. Nature 360:62–65Google Scholar
  163. Wudick MM, Portes MT, Michard E et al (2018) CORNICHON sorting and regulation of GLR channels underlie pollen tube Ca2+ homeostasis. Science 360(80):533–536. Google Scholar
  164. Yan X, Wang Z, Huang L, Wang C, Hou R, Xu Z, Qiao X (2009) Research progress on electrical signals in higher plants. Progr Nat Sci 19:531–541Google Scholar
  165. Yokawa K, Derrien-Maze N, Mancuso S, Baluška F (2014) Binary decisions in maize root behavior: Y-maze system as tool for unconventional computation in plants. Int J Unconv Comput 10:381–390Google Scholar
  166. Zimmermann MR, Maischak H, Mithoefer A, Boland W, Felle HH (2009) System potentials, a novel electrical long-distance apoplastic signal in plants, induced by wounding. Plant Phys 149:1593–1600Google Scholar
  167. Zimmermann M, Mithofer A, Will T et al (2016) Herbivore-Triggered electrophysiological reactions: candidates for systemic signals in higher plants and the challenge of their identification. Plant Physiol 170:2407–2419. Google Scholar

Copyright information

© Brazilian Society of Plant Physiology 2019

Authors and Affiliations

  1. 1.Laboratory of Plant Cognition and Electrophysiology (LACEV)Federal University of Pelotas (UFPel)PelotasBrazil
  2. 2.PSA Research GroupPalaiseauFrance

Personalised recommendations