For CN measurements, time-lapse photography recordings were made from 09:00 a.m. on the fourth day to 09:00 a.m. on the seventh day of seedling growth. A monochromatic camera (Mintron MTV-1368CD, Mintron Enterprise Co. Ltd, Taipei, Taiwan) was used to record the circumnutation trajectory of the hypocotyls apex. The plants were filmed from the top. Time-lapse images were recorded one frame per 5 min by Gotcha! Multicam software (Prescient System Inc., West Chester, PA, USA). The system was calibrated in a millimeter scale. The time-lapse images were digitized using Circumnutation Tracker (Stolarz et al., 2014) and Microsoft Excel programs. Experimental points (coordinates x, y of the stem apex on the horizontal plane) were determined at 5-min intervals. In the geographic direction plane, the single circumnutation cycle is determined by two subsequent maximum northward bends of the organ (Stolarz et al., 2014). Supplementary Video S1 shows subsequent CN cycles of sunflower seedlings. The distance covered by the hypocotyls apex during one CN cycle was used to calculate the CN rate. CN intensity was the rate of CN divided by the length of the hypocotyls. The CN period was the time required for the hypocotyls apex to trace a single CN cycle (time between two subsequent maximum northward bends of the organ) (Stolarz et al., 2014).

The electrical measurements were carried out in a Faraday cage on 7- to 8-day-old seedlings or 3-week-old sunflowers. The changes in the electrical potential were measured with two or four extracellular Ag/AgCl electrodes (a silver wire, 0.2 mm diameter, World Precision Instruments, Sarasota, FL, USA) inserted across the sunflower hypocotyl or stem and then interfaced with a multi-channel data acquisition system composed of a differential amplifier (ME-4600 Meilhaus, Germany) and RealView software (Abacom, Germany) (Supplementary Figure S1). During the preparation of the Ag/AgCl electrodes, the silver wire was electrolytically coated with silver chloride. The electrical potential was recorded from tissues adjacent to the electrode, i.e., vascular bundles, parenchyma, and epidermis (Dziubinska et al., 2001). The reference electrode (Ag/AgCl) was placed in the hydroponic medium or soil. The frequency of sample recording was 1 Hz. For registration of SAPs (spontaneous action potentials), seedlings grew in the Faraday cage in a hydroponic solution for 3 days. Glutamate (Glu) injection was applied between 8:00 a.m.–2:00 p.m. in 7–8-day-old seedlings growing in different nutrient solutions for 3 days.

The data were analyzed using Statistica ver. 12 software (StatSoft, Inc., 2014). The data set was first tested for normality using Shapiro-Wilk test and homogeneity of variance by Levene's test. The non-parametric Kruskal-Wallis ANOVA and non-parametric Mann-Whitney U-test for pairwise analysis were used when the data had non-normal distribution or unequal variance. The one-way ANOVA and post-hoc Tukey test for pairwise analysis were used when the data had normal distribution and equal variance. The level of statistical significance for all tests was set at p < 0.05.

The sunflower seedlings growing in the control nutrient solution (23 mOsm) had 10 ± 0.6 cm (n = 20) long hypocotyls. Both the hypoosmotic and hyperosmotic nutrient solutions reduced the hypocotyl length in a statistically significant way by ~10–30% (Figure 1A, Supplementary Figure S2). Simultaneously, the CN parameters changed differently in the osmotic and salt stress conditions. Sunflower seedlings growing in distilled water (0 mOsm) had significantly shortened hypocotyls and drastically reduced CN intensity in relation to those growing in the control nutrient solution (Figures 1A,B). Similarly, the high osmotic potential (160 mOsm) evoked by D-sorbitol significantly shortened the hypocotyl length and drastically reduced CN intensity (Figures 1A,B). In turn, the increase in the osmotic potential to 160–300 mOsm induced by salt (KCl or NaCl) did not influence the CN intensity although the hypocotyl length was significantly reduced (Figures 1A,B). Sunflower seedlings with similar hypocotyl length exhibited completely different CN intensity in osmotic and salt stress. The CNs were strongly inhibited in distilled water and 160 mM D-sorbitol (160 mOsm) but were vigorous (as in the control seedlings) in the salt stress inducing solutions (160–300 mOsm KCl and NaCl). The osmotically induced growth inhibition was accompanied by a decrease in the CN intensity but the salt-induced growth inhibition had little effect on the CN intensity.

In the sunflower seedlings, the mean CN period in the control solution was 174 ± 6 min (n = 20, Figure 1C). The CN period was lengthened in a statistically significant way in seedlings growing in distilled water (0 mOsm) to 245 ± 19 min (n = 10). The CN period was longer by 70 min. Simultaneously, the CN period in seedlings growing in the D-sorbitol, KCl, and NaCl solutions (160–300 mOsm) was the same as in the control conditions. In 120 mM NaCl and 160 mM KCl, it was significantly longer but only by ~15 min. The CN period was the same as in the control in the mild salt stress and even in the strong stress (160 mM NaCl). The hypocotyls of the same length had a completely different CN period in the osmotic and salt solution, which proves the absence of a strict connection between the CN and hypocotyl growth.

Given the above results, it can be assumed that only the osmotic stress disturbs growth and CN (intensity and period); in turn, the salt stress disturbs the growth mechanism only and does not influence the CN mechanism. This shows that there is a salt stress-resistant ultradian pacemaker in the sunflower and confirms that the presence of ions in the nutrient solution is basic for the CN mechanism.

K⁺, Cl⁻, and Na⁺ are also essential for membrane potential maintenance and excitability of organisms; therefore, we expect that the sunflower seedlings growing under different K⁺, Cl⁻, and Na⁺ contents in the nutrient solution could have a different excitability.

The number, velocity, and direction of propagation of SAPs were determined in salt and osmotically stressed sunflower seedlings (Figure 2, Table 1). An example of SAPs in sunflower seedlings is shown in Figure 2 and Supplementary Video S2. In the control conditions, 3 ± 1 SAPs 24 h⁻¹ plant⁻¹ (n = 12) were registered, which propagated mainly basipetally (75%) with a mean velocity 14 ± 1 cm min⁻¹. This was similar in the strong salt stress conditions (300 mOsm KCl and NaCl) (Table 1). In the seedlings growing in distilled water and in D-sorbitol (160 mOsm), SAPs were completely silenced. The number of SAPs significantly increased in mild KCl stress where the number of SAPs increased two and three times to 10 SAPs 24 h⁻¹ plant⁻¹, compared to the control and 300 mOsm salt solution. Similarly, in mild NaCl stress, the SAP number slightly increased to 7 SAPs 24 h⁻¹ plant⁻¹ and SAP propagation velocity significantly increased to 20 ± 1 cm min⁻¹ in those seedlings. The percentage of basipetally and acropetally propagating SAPs remained similar in all conditions (Table 1). The complete lack of SAPs in distilled water and the 160 mOsm D-sorbitol solution as well as the significantly increased number of SAPs and their propagation velocity in mild salt stress showed that the K⁺, Cl⁻, and Na⁺ concentration in the nutrient solution modulated spontaneous excitability. The silencing of SAPs and the reduced vigor of CNs indicates that SAPs and CNs were repressed by the lack of ions or ion uptake disturbance in distilled water and the hyperosmotic D-sorbitol solution. Mild and strong salt stress nutrient conditions maintain CNs and maintain or even enhance SAPs; thus, these results show an essential role of K⁺, Cl⁻, and Na⁺ ions in both phenomena.

Besides SAPs, Glu-induced series of APs in osmotically and salt-stressed seedlings were studied. The series of APs evoked by injection of the 50 mM Glu solution in the control plants consisted of 10 ± 1 APs (n = 10) (Figure 2D, Table 2). Similar series were obtained in mild stress in the D-sorbitol, KCl and NaCl (160 mOsm) treated seedlings; however, in strong KCl and NaCl (320 mOsm) stress, the series were significantly inhibited to 2–3 APs in a series (Figure 2D, Table 2). Seedlings growing in distilled water exhibited reduced growth, CNs (Figures 1A,B), and Glu-induced series of APs (only 30% of the seedlings were excitable, Table 2), compared to the control plants. Seedlings growing in D-sorbitol (160 mOsm) had shorter hypocotyls and reduced CN but generated Glu-induced series, likewise the control and mild stress-treated plants, which had shorter hypocotyls and unchanged vigor of CN. Additionally, sunflower seedlings treated with strong salt stress (300 mOsm) had a reduced hypocotyl length and unchanged CN, but the Glu-induced series of APs were strongly inhibited.

The spontaneous and glutamate-induced excitability changes in seedlings growing in the osmotic and salt stress conditions encouraged us to check the APs appearing after injection of the hypoosmotic and hyperosmotic solution in 3-week old sunflower plants.

Injection of distilled water or 500 mOsm solutions (D-sorbitol, KCl) into the sunflower stem evoked series of APs presented in Figure 3. The parameters of sunflower excitation after injection of the osmotic and salt solutions are presented in Table 3. Distilled water, D-sorbitol, and KCl solutions evoked series of 3 to 24 APs in 70–100% of treated plants. APs propagated mainly acropetally from the site of injection toward the stem apex, at an ~15 cm long distance. The injection of the NaCl 250 mM solution did not evoke APs (Supplementary Figure S3). These results show that a series of long-distance propagating (~15 cm) APs in the sunflower stem can be induced by distilled water, D-sorbitol and KCl but not by NaCl.

The problem of CN dependence on growth has been considered many times (Spurny, 1975; Johnsson, 1979, 1997; Millet and Badot, 1996; Stolarz et al., 2008). In general, CNs are dependent on growth, but in some environmental conditions, for example under lithium, red light, or ethylene treatment, growth and CN uncoupling was observed (Spurny, 1968; Zachariassen and Johnsson, 1988; Millet and Badot, 1996; Yoshihara and Iino, 2005; Binder et al., 2006; Stolarz et al., 2008, 2015). This study has shown that the osmotic and salt stress treatments distinguish between growth and CN behavior (Figures 1A,B). Significant inhibition of elongation of hypocotyls was observed when distilled water, D-sorbitol, KCl, and NaCl were applied (Figure 1A). The CN intensity decreased significantly only in seedlings growing in distilled water and D-sorbitol (Figure 1B). Seedlings with a similar hypocotyl length had different CN intensity in osmotic and salt stress. These results showed that the mechanism of elongation is influenced differently than the mechanism of CN. The concentration of 80–160 mM K⁺, Na⁺, and Cl⁻ decreased growth but did not affect the CN intensity. The absence of ions or ion uptake disturbance in distilled water or D-sorbitol inhibited growth and CN.

Some other studies showed the different physiological effect of the “osmotic” and “ionic” component of salt stress. The “osmotic” and “ionic” component of salt stress differently modulated net ion fluxes in leaf mesophyll of Vicia faba (Shabala, 2000). Osmotic stress in barley regulated expression of a different set of genes than salt stress (Ueda et al., 2004). The effect of osmotic stress was also shown in Beta vulgaris when D-sorbitol-induced osmotic stress strongly reduced its growth and led to a significant decrease in shoot osmotic potential, water content, and K⁺ concentrations (Wu et al., 2015a).

Circumnutation, a rhythmic phenomenon can be considered as a “cue” internal ultradian oscillator (Millet and Badot, 1996; Shabala, 2003; Lloyd, 2006). The CN period and it shortening or lengthening in various environmental conditions is important in a study of the ultradian pacemaker mechanism (Johnsson, 1979, 1997; Zachariassen and Johnsson, 1988; Stolarz, 2009; Hinrichsen et al., 2012). In this study, a stable period, despite the growth reduction, was shown (Figures 1A,C). The period of CNs in the mild and strong salt stress (163–188 min) was close to the CN period of the control plants (174 min). The CN period was lengthened significantly only in distilled water (245 min). This confirms a strong resistance of the CN period to osmotic and salt stress (Erdei et al., 1998). The independence of the CN period of the growth of hypocotyls confirmed some degree of independence between growth and CN behavior, as in the case of lithium treatment (Stolarz et al., 2015).

The results presented above confirmed again the autonomy of CN behavior in relation to elongation and special physiological and ecological functions of CN being not merely a strict result of growth (Darwin and Darwin, 1880; Inoue et al., 1999; Larson, 2000; Kosuge et al., 2013; Migliaccio et al., 2013).

In plants, beside the stimulus-induced APs (electrical and mechanical stimuli, light, temperature, chemicals), there are SAPs. The SAPs appear in the absence of external stimuli and their endogenous source is unknown. SAPs were previously observed in 3-week old Helianthus and Lycopersicum plants growing in pots with standard soil (Zawadzki et al., 1995; Macedo et al., 2015). In this study, we have shown for the first time SAPs in sunflower seedlings growing in a hydroponic medium (Figure 2, Supplementary Video S2). This offered a new opportunity to modify the nutrient composition and investigate SAPs in plants in a specified nutrient environment.

In this work, we have described the number of SAPs per 24 h⁻¹plant⁻¹ as a parameter of endogenous “firing,” which we have shown to be able to appear in control conditions (no external stimuli) but were modulated by osmotic and ionic stress. In the controlled nutrient conditions, the sunflower seedlings generated 3 ± 1 SAPs 24 h⁻¹plant⁻¹ (n = 12). This number of SAPs persisted even under strong salt stress but completely disappeared in the distilled water and D-sorbitol (160 mM, 160 mOsm) treatment (Table 1). The number of SAPs increased significantly to 10 ± 2 SAPs 24 h⁻¹ plant⁻¹ (n = 12) in 80 mM KCl (160 mOsm). This showed that the “ionic” but not “osmotic” component of salt stress increased the number of SAPs. The number of SAPs 24 h⁻¹ plant⁻¹ differed in the different nutrient conditions. These results demonstrate that sunflower seedlings are capable of generating many APs, and in a sufficiently long time, we can consider this as a specific frequency code of APs. In slowly moving hypocotyls of sunflower seedlings, this approach is justified. Slow motion of hypocotyls is accompanied by low frequency electrical signals (from 1 to 45 of APs during 24 h per seedling). In such a long time span, there are various numbers of SAPs and this amount can be modulated by the nutrient environment; therefore, there is endogenous electrical activity with very low frequency coding in sunflower seedlings. SAPs are known to propagate acropetally and basipetally. Basipetally propagating SAPs dominated in all nutrient solutions. They appeared in the upper part of the seedlings and propagated downwardly. This showed that the upper part of the seedlings generated more SAPs than the lower part. The amplitudes were in the range of 5–60 mV and the rate of propagation was 12–20 cm min⁻¹. In evoked APs, the amplitude is relatively constant (50–60 mV). Significantly lower amplitudes of some SAPs indicate that the excitation is transmitted along individual phloem vessels but not along the transmitting system as in the case of evoked APs. Seedlings growing in the mild NaCl stress exhibited significantly higher SAP propagation velocity than the control plants. The amplitude and velocity of propagation was similar to that in 3-week old Helianthus and Lycopersicum plants growing in pots with standard soil (Zawadzki et al., 1995; Macedo et al., 2015).

In the research of changes in excitability in sunflowers, we used KCl and NaCl. A treatment with the KCl solution was applied because K⁺ is an important component of the AP mechanism and transmembrane potential maintenance (Trebacz et al., 1989, 1994; Shabala, 2003; Johansson et al., 2006; Shabala and Cuin, 2008; Aleman et al., 2011; Dreyer and Uozumi, 2011; Shabala and Pottosin, 2014; Salvador-Recatala, 2016). In turn, NaCl was used because it is the most commonly occurring salt in the environmental salt stress (Munns and Tester, 2008; Maathuis, 2014; Foster and Miklavcic, 2015). The mechanism of the action of environmental salt (NaCl) stress on plants also includes K⁺ homeostasis disturbance (Shabala and Cuin, 2008; Foster and Miklavcic, 2015). The effect of KCl and NaCl on SAP appearance was different (Table 1). The mild stress induced by KCl raises the number of SAPs 24 h⁻¹ plant⁻¹ in a statistically significant way, while NaCl slightly increases this value. The strong stress caused by NaCl and KCl keeps SAPs at a control level. This shows that only the mild stress induced by KCl increases the appearance of SAPs in sunflower seedlings. Potassium ions are an important part of the mechanism of maintaining the excitability of the cell membrane and we can assume that its slight increase in the environment affects the membrane transport (Trebacz et al., 1989, 1994; Shabala, 2003; Johansson et al., 2006; Shabala and Cuin, 2008; Shabala and Pottosin, 2014; Salvador-Recatala, 2016). The mild stress induced by NaCl slightly increases the number of SAPs and increases SAP velocity in a statistically significant way. An increase in SAP velocity occurs also in strong stress induced by KCl. It is presumed that both these effects are due to the involvement of K⁺ and Na⁺ ions in membrane transport and the function of cation channels. The increased velocity of SAP propagation is in accordance with the cable theory, which postulates such an effect after lowering of the external (apoplast) electrical resistance caused by the applied ions.

Additionally, we have shown changes in the growth and intensity of the endogenous movement of the hypocotyl in spontaneously excited seedlings. The complete lack of SAPs in distilled water and D-sorbitol (with an ability to generate Glu-induced APs) growing seedlings with strongly inhibited CN may be associated with CN inhibition. In our recent work, the relationship between SAPs and CN was shown, i.e., the number of SAPs decreased with the restricted CN (Stolarz and Dziubinska, 2017). On the other hand, the similarly circumnutating seedlings had a significantly lower number of SAPs in the strong salt stress [4 ± 1 SAPs 24 h⁻¹ plant⁻¹ (n = 8)] than in the mild salt stress [10 ± 2 SAPs 24 h⁻¹ plant⁻¹ (n = 12)]. This confirms a possible complex relation between endogenous APs and endogenous hypocotyl movements in sunflower seedlings.

A treatment with KCl combined with prior pricking induced series of APs in C. conicum (Favre et al., 1999) and A. thaliana (Favre et al., 2011). Recently, Salvador-Recatala (2016) used NaCl and KCl solutions to evoke depolarization of root cells of A. thaliana. The salts were also administered in Vicia faba and Hordeum vulgare to evoke system potentials (SPs) (Zimmermann et al., 2009). An electrical response after the hypo- or hyperosmotic solution injection (Figure 3, Table 3, Supplementary Figure S3) was dependent on the kind of treatment. Usually, a series of APs appeared, but we assume that some of the recorded spikes (Figure 3, electrode 3 and 4) are SPs. A similar recording was shown in Vicia faba, Hordeum vulgare, and Nicotiana tabacum after wounding as well as abiotic and biotic stress (Zimmermann et al., 2009, 2016). The injection of NaCl did not evoke series of APs (Supplementary Figure S3). It is probable that the presence of K⁺ (but not Na⁺) decreases the membrane potential difference and the excitation threshold, which facilitates the generation of APs.

In this study, osmotically driven APs propagating for many centimeters along the sunflower stem have been shown for the first time. Moreover, we have shown that not only salt stimuli but also distilled water and osmotically active solutions (D-sorbitol) induce electrical signals propagating along stem.

Our working hypothesis is that osmotically active solutions and salt solutions affect membrane transport and thus evoke propagated electrical signals and changes in CN. APs could result from adjustment of membrane polarization of some cells over the threshold by ions (K⁺, Na⁺, and Cl⁻) contained in the solutions (in the nutrient or injected). At first, the salt stress affected the membrane transport (Foster and Miklavcic, 2015) via primary active proton pumps, secondary active antiporters and symporters, as well as passive ion channels. High salinity inhibits also plant growth via increasing osmotic pressure disrupting the plant ability to take up water and hence nutrients. Next, accumulation of salt ions (usually Na⁺) leads to concentrations that are toxic (Munns and Tester, 2008). The maintenance of a high cytosolic K⁺/Na⁺ ratio appears to be critical to plant salt tolerance (Shabala and Cuin, 2008). Exposure to high concentrations of Na⁺ can also disrupt the homeostasis of K⁺, which is essential for many physiological processes including excitability and organ movement.

Besides ions, nitrogen is a major factor of plant growth (Debouba et al., 2006). The sensitivity to nitrogen sources in soil and capability of nitrogen uptake are basic features of growing plants. Glutamate is a major amino acid involved both in nitrogen metabolism and in nitrogen signaling (Forde and Lea, 2007; Mousavi et al., 2013). Previously, we showed that lithium modulated Glu-induced excitability and CN in sunflower seedlings (Stolarz et al., 2015). Thus, we expected that also the external concentration of K, Na, and Cl could modulate the Glu-induced series of APs. Here, we have shown for the first time the modulation of Glu-induced excitation by salinity. Seedlings growing in distilled water did not generate SAPs under 3-day observations (Table 1), but these plants were able to generate Glu-induced series of APs (Figure 2D, Table 2). Strong salt stress disturbed significantly the Glu-induced series of APs (Figure 2D, Table 2). Thus, it has been shown that only strong salt stress, but not osmotic stress, decreased Glu-induced series of APs. These results suggested that Glu-induced APs could be a part of the signaling pathway in nitrogen metabolism.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.