Mechanism of kinetin-induced death of Vicia faba ssp. minor root cortex cells

Cell death (CD) may be induced by endogenous or exogenous factors and contributes to all the steps of plant development. This paper presents results related to the mechanism of CD regulation induced by kinetin (Kin) in the root cortex of Vicia faba ssp. minor. To explain the process, 6-(2-hydroxy-3-methylbenzylamino)purine (PI-55), adenine (Ad), 5′-amine-5′-deoxyadenosine (Ado) and N-(2-chloro-4-piridylo)-N′-phenylurea (CPPU) were applied to (i) block cytokinin receptors (CKs) and inhibit the activities of enzymes of CK metabolism, i.e., (ii) phosphoribosyltransferase, (iii) kinases, and (iv) oxidases, respectively. Moreover, ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), lanthanum chloride (LaCl3), ruthenium red (RRed) and cyclosporine A (CS-A) were applied to (i) chelate extracellular calcium ions (Ca2+) as well as blocks of (ii) plasma-, (iii) endoplasmic reticulum- (ER) membrane Ca2+ ion channels and (iv) mitochondria- (MIT) Ca2+ ions release by permeability transition por (PTP), respectively. The measured physiological effectiveness of these factors was the number of living and dying cortex cells estimated with orange acridine (OA) and ethidium bromide (EB), the amounts of cytosolic Ca2+ ions with chlortetracycline (CTC) staining and the intensity of chromatin and Ca2+-CTC complex fluorescence, respectively. Moreover, the role of sorafenib, an inhibitor of RAF kinase, on the vitality of cortex cells and ethylene levels as well as the activities of RAF-like kinase and MEK2 with Syntide-2 and Mek2 as substrates were studied. The results clarified the previously presented suggestion that Kin is converted to appropriate ribotides (5′-monophosphate ribonucleotides), which cooperate with the ethylene and Ca2+ ion signalling pathways to transduce the signal of kinetin-programmed cell death (Kin-PCD). Based on the present and previously published results related to Kin-PCD, the crosstalk between ethylene and MAP kinase signalling, as well as inhibitors of CK receptors and enzymes of their metabolism, is proposed.


Results
Fluorescence intensity measurements of acridine orange (AO) and ethidium bromide (EB) in the nuclei showed that approximately 4% of Ctrl cells were dying or dead after 72 h, while the number of these cells was significantly (p < 0.05) greater after Kin treatment, at approximately 48% (Figs. 1A, 2A,C, 3A, 5A,C, 6A,C, 7A). The cytophotometric measurement showed that the amount of cytosolic Ca 2+ ions in Ctrl cells was approximately 10,000 a.u., while the amount Ca 2+ ions after Kin treatment was significantly (p < 0.05) greater, at approximately 14,000 a.u. (Figs. 1B, 2B The amount of Ca 2+ ions was significantly increased after treatment with 5 µM PI-55 (PI-55-5 series) (p < 0.05), by approximately 2.5-and 2.0-fold compared to the Ctrl and Kin series, respectively, while the amount of Ca 2+ ions after treatment with 10 µM PI-55 (PI-55-10 series) was approximately 2.0-and 0.5-fold greater (p < 0.05). Thus, 5 µM PI-55 was used to study the kinetin effects on the metabolism of cortex cells. In the Kin/ PI-55-5 series, the amount of Ca 2+ ions was significantly greater (p < 0.05), by about 3.0-and 2.0-fold than that in the Ctrl and Kin series, respectively (Fig. 1B).
Ad, at both 50 µM and 100 µM concentrations, did not change (p > 0.05) the CD index compared to Ctrl cells ( Fig. 2A). The amounts of Ca 2+ ions in the Ad-50 and Ad-100 series were approximately 2.0-and 1.3-fold greater (p < 0.05) than those in the Ctrl and Kin series, respectively. Thus, 100 µM Ad was used to study Kin. In the Kin/ Ad-100 series, the number of dying cells was similar to that in the Ctrl series (p > 0.05), while the amount of Ca 2+ ions was similar to that in the Kin series (p > 0.05) (Fig. 2B).
Ado at 10 µM and 50 µM did not have a significant (p > 0.05) impact on the CD index in comparison to Ctrl series; the number of dying cells was approximately 12% and 5%, respectively. However, 50 µM Ado was used in Kin experiments. In this series, the number of dying cells was similar in Ctrl cells (Fig. 2C), while the amounts of Ca 2+ ions in the Ado-10 and Ado-50 series were significantly greater, by approximately 2.0-and 3.0-fold, respectively, in comparison to Ctrl cells and by approximately 1.5-and 2.5-fold, respectively, in comparison with the Kin series. Meanwhile, in the Kin/Ado-50 series, the amount of Ca 2+ ions was similar (p > 0.05) in the Kin series (Fig. 2D).
CPPU at 5 µM (CPPU-5 series) did not change (p < 0.05) the CD index compared to Ctrl cells, but after treatment with a 10 µM concentration (CPPU-10 series), the number of dying cells was similar to (p > 0.05) the Kin series. Thus, to study the effect on Kin-PCD, the 5 µM CPPU (Kin/CPPU-5) was used. In the Kin/CPPU-5 series, the number of dying cells was significantly lower (p < 0.05), by approximately 30%, than that in the Kin series (Fig. 3A). The levels of Ca 2+ ions in the CPPU-5 and CPPU-10 series were significantly greater in comparison to Ctrl cells (p < 0.05), by approximately 2.0-and 1.5-fold, respectively, and by approximately 1.5-and 1.3-fold   3B).
Measurements of ETH showed that, in the Kin series, its levels were significantly greater than in the Ctrl series (p < 0.05), by approximately 30% (Fig. 4).
Seedlings treated with CPPU at 5 μM (CPPU-5) produced five times (p < 0.05) more ETH than the Ctrl seedlings, while those treated with Kin and CPPU (Kin/CPPU-5) produced approximately three times more ETH (p < 0.05) than the Kin seedlings. However, its levels were 30% lower in this series than in CPPU-5 (Fig. 4).
Ad at a concentration of 100 µM or with Kin decreased the amount of ETH (Ad-100) by approximately three times (p < 0.05) compared to the Ctrl and Kin series, respectively (Fig. 4). Ado at 50 µM evoked similar effects to the Ad series (Fig. 4).
Effects of EGTA and La 3+ , RRed and CS-A, chelators, and blockers of plasma-, ER-, and MIT membrane-dependent channels, respectively, on the number of dying cells induced by kinetin (Kin-PCD), the amount of cytosolic Ca 2+ in root cortex cells and the amount of ETH in the atmosphere of growing seedlings. EGTA at 10 µM (EG-10) and 50 µM (EG-50) concentrations did not change (p > 0.05) the number of living or dying cells compared to Ctrl cells. To study its effect on Kin-PCD, 10 µM EGTA was used (Kin/EG-10). The results showed that, in the Kin/EG-10 series, the number of dying cells was significantly lower (p < 0.05) than that in the Kin series, at approximately 8% (Fig. 5A). The amounts of Ca 2+ in the EG-10 and EG-50 series were approximately 1.5-and twofold greater (p < 0.05) than those in the Ctrl group, respectively, and approximately 1.3-and 1.5-fold greater (p < 0.05) than those in the Kin group. Meanwhile, in the Kin/EG-50 group, the amount of Ca 2+ was similar (p > 0.05) to that in the Ctrl group (Fig. 5B).
La 3+ ions at 5 µM (La-5) and 25 µM (La-25) concentrations did not change the PCD index compared to Ctrl cells. To study its effect on Kin-PCD, a 5 µM concentration of La 3+ was used. It was observed that in comparison with Kin, the number of Kin-PCD-dying cells in the Kin/La-5 series was significantly lower (p < 0.05), by approximately 38% (Fig. 5C). The amounts of Ca 2+ in the La-5 and La-25 series were significantly greater (p < 0.05), by approximately 2-and 1.5-fold, than those in the Ctrl and Kin series, respectively. In the Kin/La-5 series, the amount of Ca 2+ was significantly greater (p < 0.05), by approximately 1.5-fold, than that in the Kin series (Fig. 5D).
RRed at a 10 µM concentration (RRed-10) did not increase (p > 0.05) the number of Kin-PCD-dying cells in comparison with Ctrl cells, while at a 25 µM concentration (RRed-25), it increased the number of dying cells by approximately 20-and 1.5-fold (p < 0.05) compared to the Ctrl and Kin series, respectively. Thus, a 10 µM concentration of RRed was used to study its effects on Kin-PCD. The results showed that 10 µM RRed inhibited Kin-induced CD; thus, the CD index was similar to that observed for the Kin series (Fig. 6A). The amounts of Ca 2+ in the RRed-10 and RRed-25 series were significantly greater than the Ctrl and Kin series (p < 0.05),   www.nature.com/scientificreports/  Measurements of ETH levels in the atmosphere of growing seedlings showed that seedlings in the EG-10, La-5 and Red-10 series had similar amounts of ETH (p > 0.05) and were approximately four times lower (p < 0.05) than those in the Ctrl series. In Kin/EG-10, Kin/La-5 and Kin/RRed-10, the amounts of ETH were six times lower (p < 0.05) than those in the Kin series. Treatment of Ctrl seedlings with CA-S at 25 μM concentration (CS-A-25) elevated the level of ETH by approximately three times (p < 0.05), while CS-A-and Kin-treated seedlings at the same concentration (Kin/CS-A-25) produced approximately four times greater (p < 0.05) levels of ETH compared to the Ctrl and Kin series (Fig. 7).

Effect of sorafenib, an inhibitor of RAF kinase, on the number of living and dying cells during Kin-PCD and ETH as well as the activities of RAF-like and MEK2 enzymes.
To estimate the number of living and dying cells, 1 µM sorafenib was used. The number of living cells in the sorafenib and Kin/ Sorafenib series remained at the same level (p > 0.05) as in the Ctrl series. In the Kin series, the proportions of living and dying cells were approximately 45% and 55%, respectively (Fig. 8A).
In the Ctrl series, the amount of ETH per six seedlings was approximately 5 ppm, while after treatment with Kin for 72 h, its amount was approximately 30% greater (p < 0.05). Treatment of Ctrl seedlings with sorafenib (Sorafenib series) showed that the ETH amount was approximately 20% lower (p < 0.05) than that of the Ctrl seedlings. In the combination of 1 µM sorafenib with Kin (Kin/Sorafenib series), the amount of ETH was approximately 30% and 40% greater (p < 0.05) than that in the Ctrl and sorafenib series, respectively, and was similar (p > 0.05) to that in the Kin group (Fig. 8B).
Analyses of kinase activities showed that the activities of RAF-like kinase, with Syntide-2 as the substrate, and MEK2, with Mek2 as the substrate, in the Ctrl series were approximately 2.5 mg and 1 mg of ATP per mg of protein, respectively. After treatment with Kin for 72 h, their activities were approximately 20% greater (p < 0.05) and 20% lower (p < 0.05), respectively, than those of the Ctrl series (Fig. 8C).
Studies of Kin-PCD have shown that Kin in faba bean roots induces two responses 6 . The first is related to the development of protective mechanisms that act against death. Kin is an antioxidative factor that strongly inhibits oxidative and glycoxidative protein damage 5 . Such an antioxidative effect of Kin is observed in the apical parts of faba bean seedlings during Kin-PCD as increases and then decreases in ROS levels as a result of increased metabolic enzyme activities 11 . CK in Kin form can delay the senescence of detached Raphanus sativus leaf discs 50 . CKs inhibit leaf senescence via activation of cytokinin receptor (AHK3), the type-B response regulator (ARR2) and cytokinin response factor 6 (CRF6) 51 .
The second response 6 of Kin in faba bean roots is related to the induction of PCD [6][7][8][9][10][11] . Kin may be degraded to Ad and ribosylated to its ribosides by APRT 14,[32][33][34][35] . This enzyme was identified in wheat (Triticum aestivum) germs 14 . To obtain phosphorylated forms of CKs, adenosine kinase catalyses the formation of adenosine monophosphate from adenosine. Adenosine kinase was found in tobacco BY-2 cells 17 as well as in Arabidopsis 27 . The next step of adenine metabolism is related to conversion of the 5′-monophosphate of zeatin to its trans or cis forms 51 . The fact that Ad and Ado, adenosine phosphorylase 14,32-35 and adenosine kinase 36-38 inhibitors, The number of living and dying cells as well as the amounts of Ca 2+ were fluorescently estimated after staining with acridine orange, ethidium bromide and chlortetracycline. Kinase activities were determined using the luciferase technique and ATP as a substrate. Bars indicate ± SE of the results from three biological replicates. The means of the results with the same letter above the column are not significantly different at p ≤ 0.05. Ctrl, control; Kin, kinetin; Sorafenib, RAF kinase inhibitor; sorafenib with Kin, Kin/Sorafenib; Syntide-2, synthetic peptide recognized as a substrate by Ca 2+ /calmodulin-dependent protein kinase II. www.nature.com/scientificreports/ respectively, completely suppressed Kin-PCD in the root cortex confirmed the occurrence of such enzymes in faba bean roots and their functions. Ad inhibits the formation of ribotides of adenine (AMP) in humans 16 although not in A. thaliana. Its application to faba bean seedlings inhibited the CD induced by Kin. This fact suggested that the Ad-inhibiting activity of APRTs disrupts the CK synthesis pathway; therefore, lower amounts of CK monophosphates, which are suggested as ligands for CK receptors 16,17 , are released in the CK-dependent pathway.
Moreover, inhibition of adenosine phosphorylase and cytokinin kinase activities might enhance the inhibitory effect on Kin-PCD because an inhibitor of cytokinin oxidase, i.e., CPPU, suppressed Kin-PCD. Adenine, which is a direct product of CK degradation by cytokinin oxidase and an inhibitor of APRT activities [32][33][34][35] , acts against Kin-PCD 6 . This result indicated that the level of Kin or other CK riboside monophosphates, as a result of the activity of APRT, may be lowered. Thus, its CD-inducing effects can be decreased.
Based on the results, the authors suggested that riboside monophosphates of CKs could activate faba bean HK4 receptors in endoplasmic reticulum (ER) membranes 18,19 . HK4 has a greater affinity for monophosphorylated forms of cytokinin ribosides and free cytokinin bases than HK319; however, both receptors can interact with each other 22 . This might result from the fact that PI-55 completely inhibited the Kin-induced death of cortex cells. It is known that dimerization and autophosphorylation of HKs lead to the translocation of the phosphate group from histidine to aspartic acid, the regulatory domain of the receptor. Then, histidine HPT carriers and type B RR, discovered in A. thaliana 26 , whose homologues may exist in faba bean, can activate the cytokinindependent response (Fig. 9).
During Kin-PCD, a decrease in the levels of ATP 7 , activities of cellular dehydrogenases 9 and histone kinase activities 7 is observed, suggesting that ATP is used for the synthesis of CK riboside monophosphates and/or cytotoxic N 6 -furfuryladenosine (Kin-riboside), as in human cancer cell lines 52 .
The idea that Kin-PCD depends on the cooperation of Kin with ETH comes from the results of studies with aminooxyacetic acid (AOA), CoCl 2 and 2,5-norbornadiene (NBD), inhibitors of the synthesis and conversion of ACC to ETH as well as its receptor, respectively, which suppress Kin-PCD 6 , and from results showing that Kin elevated the levels of ACC 9 and ETH (present paper). Loss of HK4 function in the ahk4 mutant decreased the ETH response to CKs, indicating that the A. thaliana HK4 receptor is probably a primary contributor to ETH biosynthesis in response to CKs 29,53 . Therefore, HPT of A. thaliana can transduce signals from CKs 29,53,54 ( Fig. 9) to ETH receptors (ethylene triple response, ETR1 and 2; ethylene insensitive 4, EIN4; ethylene response sensor 1 and 2 ERS1 and 2 in ER membranes 29,53 . Moreover, CKs can posttranscriptionally increase the activity of ACS 4,5 gene products 55 , leading to an increase in the activities of these enzymes in ETH synthesis. Thus, it can synergistically increase the ETH amount in faba plants (Figs. 4, 7). ETH can activate EIN2-and ER-and nuclear ETH-dependent membrane receptors and induce ETH response elements (EREs) 29,53,56 . The fact that aerenchyma was formed after Kin-6 and ACC-induced 13 death of cortex cells of faba bean root confirmed the hypothesis because ETH seemed to be a direct hormonal factor involved in the process 42 . This was also confirmed by the results presented in the paper showing that Ad, Ado and CPPU, inhibitors of CK synthesis enzymes, i.e., APRT, ADK and CKO, respectively, reduced the Kin-induced death of cells, but the level of ETH only increased in the CPPU series.
CPPU could participate in the alternative ETH-CK pathway, which was postulated by Kin-PCD authors. The results showed that sorafenib, an inhibitor of ATP-competitive kinase 40 , suppressed Kin-PCD, which was manifested by a low number of dying cells, similar to the Ctrl series. However, the RAF-like CTR1 kinase activity, the mitogen-activated protein kinases (MAP kinases), a structural equivalent of serine/threonine-specific protein kinase (RAF kinase), element of the MAP-kinase-dependent cascade, the regulator of ETH signal transduction pathway 57 , was greater in roots of Kin than in Ctrl series, while the MEK2 activity was lower compared to Ctrl series. The differences in the activities of these kinases may have resulted from the fact that sorafenib can also affect the activity of other kinases, e.g., ERK (extracellular signal-regulated kinase) 40 . These facts allow us to suppose that both kinases exist in faba bean roots.
However, the results indicated that a decrease in the number of dying cells depended on the effect of an increase in RAF-like kinase activity and a decrease in MEK2 activity. Moreover, the results confirmed that Kin-PCD is the process controlled by cooperation among ETH-dependent MAP kinase signalling pathways. The crosstalk between Kin and ETH might also take place at the level of CK and ETH receptors (the two-component systems) [25][26][27][28]54 because both receptors are members of the HK family proteins [18][19][20]22,29,53 .
Results presented by Scharein and Groth 27 also confirmed the existence of ETH-CK crosstalk. The phosphorylation between ETR1 and histidine-containing phosphotransfer protein 1 (AHP1) complexes plays a crucial role in this process. The affinity between the two partners is greater when one of them is phosphorylated and the other is not 27 .
The present results also showed that the amounts of cytosolic Ca 2+ in the root cortex cells in faba bean seedlings in PI-55-5 and CPPU-5 were similar to Kin/PI-55-5 and Kin/CPPU-5, but they were greater in Ad-100 and Ado-50 in comparison with Kin/Ad-100 and Kin/Ado-50. The levels of cytosolic Ca 2+ in Kin/PI-55-5 and Kin/CPPU-5 were greater than in Kin, while the levels were similar between the Kin/Ad-100 and Kin/Ado-50 and Kin series.
The application of EGTA, La 3+ , RRed and CS-A, i.e., chelators of extracellular Ca 2+ ions 43 , inhibitors of plasma-43 , in a nitrate 3 or chloride form, ER-44 and MIT-45 membrane Ca 2+ ion channels, respectively, inhibited the Kin-PCD process. Additionally, the levels of cytosolic Ca 2+ in cortex cells in Kin/La-5, Kin/RRed-10 and Kin/CS-A-25 were greater in comparison with Ctrl and in Kin series (Figs. 5, 6), but the amounts of Ca 2+ in Kin/ EG-10 series were similar to the Ctrl (Fig. 5).
The idea that Ca 2+ ions are important for Kin-PCD induction was suggested previously 7 . The results of the paper clearly indicated that inhibition of Kin-PCD takes place when plasma-, ER-membrane Ca 2+  www.nature.com/scientificreports/ not available, leading to limitations of the migration and maintenance of noninductive cell death levels of these ions (Fig. 9). Therefore, blocked MIT permeability pores protected against the release of ROS and the swelling of MIT, some of the most important factors 58,59 related to the hallmark 58,59 of cell death. Both phenomena were observed during Kin-PCD 6,11 . Doniak et al. 3 and Kunikowska et al. 6 suggested an alternative role of Ca 2+ in the transduction of Kin-dependent signals. A greater amount of Ca 2+ ions was also important for the activation of Kin-PCD via calciumdependent ACC synthase (ACS) and ACC oxidase (ACO), enhancing 55,60 ACC and ETH synthesis, respectively. It is also possible that ETH initiated the expression of protein kinases and/or their activities, especially H1, and core histones 7 induced ETH-positive feedback on their own synthesis via the same enzymes (Fig. 9).
The existence of one or more alternative pathways inducing and transducing Kin-PCD was confirmed by the results showing that the elevated levels of ETH during Kin-PCD were lowered by two CK metabolism inhibitors Figure 9. Schematic of the mechanism of the Kin-PCD signalling pathway based on the suggestions presented by Doniak et al. 6 and Kunikowska et al. 9 and the results described in the present study on the root cortex of Vicia faba ssp. minor seedlings. Kin transported to the cell through purine transporter (PUP) is converted to riboside and then to ribonucleotide (ribotide). The ribotide activates histidine kinase (HK) receptors (i.e., HK4), which activates the ethylene (ETH)-dependent signalling pathway using histidine-containing phosphotransmitter (HPT) and Ca 2+ -dependent channels. These channels are needed to release Ca 2+ (also from MIT), which is important for the activation of Ca 2+ -dependent ACC (1-aminocyclopropane-1-carboxylic acid) synthase (Ca 2+-ACS). Then, the action of serine and cysteine proteases and exo-/endonucleases of the execution/dying phase lead to protoplast and cell wall degradation, appearing as aerenchyma formation. Simultaneously, HPT may stimulate a cytokinin response regulator (RR-B) and synthesis of Ca 2+ -dependent ACC oxidase (ACO), which may enhance the PCD effect by activating ETH-dependent genes through ethylene response elements (EREs). Moreover, by stimulating reactive oxygen species (ROS) metabolism enzymes with HPT (CK-dependent genes), the protective role of Kin against Kin-PCD might be activated. www.nature.com/scientificreports/ (AD and Ado), EGTA, a chelator of extracellular Ca 2+ ions, and by two Ca 2+ ion channel blockers, i.e., La 3+ and RRed. In this process, CPPU and CS-A elevated the Kin-PCD-dependent ETH levels.
The other alternative pathway of Kin-PCD in relation to Ca 2+ ions and ETH may depend on phosphatidylinositol 4,5-bisphosphate and inositol 1,4,5-trisphosphate 4 because the phosphoinositide cycle is required for the regulation of ACS activity 61 .
The participation of ETH in Kin-PCD can also be confirmed by the fact that this process is accompanied by an ETH-dependent triple response 3,6,10 , i.e., shortening and thickening of roots as the effect changes in the same way as cortex cell expansion and root apical hook formation 3,6 , as well as by aerenchyma formation in Kin-PCD 3,6 . Thus, aerenchyma formation should be added as the next feature of ETH-dependent but not triple or quadruple responses.

Conclusions
The results based on the usage of reagents, such as PI-55, Ad, Ado, CPPU, EGTA, La 3+ , RRed and CS-A, sorafenib, Syntide-2 and Mek2, the inhibitor of RAF kinase as well as RAF-like and MEK2 substrate related to the animals, plants and bacteria were crucial for explaining Kin-PCD progression.
First and foremost is the fact that Kin controls cell differentiation in both plants and animals. Second, CTR1 kinase, known in animals and bacteria as RAF kinase, forms transmembrane receptors for CKs, and ETH also exists in plants as RAF-like receptors.
Taking all the results into account, we can conclude that the Kin-dependent signalling pathways by the phosphorylated forms of Kin or other CK ribosides may activate CD-dependent HK4 (strongly suggested as the cell death receptor activated by CKs) receptors and generate pathways via HPT, enhancing the expression of ACS4,5 genes and the activities of Ca 2+ -dependent ACSs. Then, by activation of Ca 2+ -dependent ACOs, the ACC and ETH amounts increase (Fig. 9).
Then, the signal is transferred onto specific ETH-dependent genes, and it induces: (i) expression of elements initiating the executive phase of Kin-PCD, including serine and caspase-like proteolytic and nucleolytic machinery; and (ii) cell wall compound metabolism, which finally leads to complete degradation of some cells in the degradation phase of Kin-PCD, resulting in aerenchyma formation.
We demonstrated that perception and metabolism of CK inhibitors completely suppressed Kin-PCD. However, PI-55, Ado and CPPU increased amounts of cytosolic Ca 2+ , but Ad did not reduce the Kin-induced changes. Ad and Ado decreased while CPPU increased the Kin-elevated amounts of ETH.
Moreover, the chelator of extracellular Ca 2+ ions as well as plasma-, ER-and MIT-membrane Ca 2+ channel inhibitors also suppressed Kin-PCD. However, except for EGTA, they increased the Kin-elevated amounts of cytosolic Ca 2+ in the cortex cells of seedling roots. EGTA, La 3+ and RRed decreased and CS-A increased the amount of ETH.
Thus, Ca 2+ ions are the central link of cooperation between ETH and Kin in the studied Kin-PCD.

Materials and methods
Plant material, chemicals, and experiments. V. faba ssp. minor cv. Bobas (Danko, Sobiejuchy 2, 88-400 Żnin, Sobiejuchy 2, Poland; http:// www. danko. pl) seeds (20) were germinated for 3 days in Petri dishes (15 cm in diameter and 3 cm high) on two blotting papers with distilled water (the seeds were half submerged) in a dark breeding room. For analyses, 6 of the 3-d-old seedlings with nearly equal root length (2.0 ± 0.3 cm) were transferred into a glass container (8 cm in diameter and 4 cm high) with two blotting papers moistened with 10 cm 3 of water (Ctrl) or adequate solutions of chemicals and cultivated at 23 ± 1 °C and 92% ± 2% relative humidity for exactly 72 h and then used for analyses. The types, sources and solvents of the factors originally used in the studies are presented in Table 1. First, the impact of CK perception and metabolism regulators as well as Ca 2+ channel inactivators at two selected concentrations (Table 1) without Kin on CD and cytosolic Ca 2+ amounts in the faba bean seedling root cortex was tested. Then, the concentration of the factor that did not induce or slightly induced cell death compared to the other was used to analyse its respective influence on the vitality and amount of cytosolic Ca 2+ in cortex cells during Kin-induced CD. The viability (Supplementary Fig. S1) of cells was also analysed after treatment with sorafenib, a RAF-like kinase inhibitor, at 1.0 µM concentration in cells treated and untreated with Kin V. faba ssp. minor seedlings (Supplementary Fig. S2).
Additionally, the effects of solvents used for PI-55, CS-A, and Kin as well as for CPPU preparation, i.e., 0.05% DMSO (Avantor), 0.01 N HCl (Avantor) and 0.1% methanol and a mixture of HCl and DMSO as well as HCl with methanol (Avantor) in distilled water, respectively, on the viability of cortex cells were tested, and the average values were used to complete the figures. Moreover, the effects of the mentioned solvents on the fluorescence To detect and measure the intensity of CD (percentage of dying cells) in the root cortex, apical fragments of roots were cut off from the seedlings, washed twice with 0.1 M PHB (Na phosphate buffer pH 7.4; Avantor), stained with a mixture of 100 μg cm −3 acridine orange (AO; Sigma-Aldrich) and 100 μg cm −3 ethidium bromide (EB; Sigma-Aldrich) in PHB, washed in PHB two times and fixed with a 2% solution of glutaraldehyde in Naphosphate buffer (PHB). Then, thin sections of the long axes of 2-cm apical parts of roots were analysed and photographed under blue light from an Optiphot-2 fluorescence filter (B2A) (Nikon, Japan) equipped with a DXM1200 digital camera and Act-1 (Precoptic, Poland) software.
Estimation of the numbers of living, dying and dead cortex cells was carried out according to the method described in Doniak et al. 6 , Byczkowska et al. 13 and Kunikowska et al. 9 using the specially prepared calibration curve, which presents relationships between the fluorescence intensity (FI) of dyes and the amount of nuclear chromatin. Measurements of FI were carried out using Scion Image (Scion Corporation) software. During measurements, each stained nucleus was separately outlined using the threshold function, and the values of fluorescence intensity in arbitrary units were then read and compared to the calibration curve of Byczkowska et al. 13 .
This method uses the properties of EB-migration, in which the amount in nuclei increases proportionally with the CD-induced increase in plasma and nuclear membrane permeabilization. AO migration through all types of membranes does not depend on their conditions. Thus, the changing colour of nuclear chromatin ranging from green to orange-red is related to increasing fluorescence intensity (FI; Supplementary Fig. 3S; A) 13 .
Bright-orange ( Fig. S3; A, a1) and orange-red ( Fig. S3; A, a2) colours indicate dead cells (FI values > 46 a.u.), yellow ( Fig. S3; A,a3) and yellow-orange ( Fig. S3; A,a4) indicate dying ones (FI values 34-55 a.u.), while green-yellow ( Fig. S3; A,a2) and green indicate living cells, with resultant fluorescence intensity (RFI) values of 10-35 a.u. (Fig. S3; A,a1). To measure the cytosolic Ca 2+ amount in cortex cells, 2-cm-long apical parts of faba bean roots (between the 4th and 20th mm from the apex) were fixed with a 2% solution of glutaraldehyde (POCH) in PHB for 1 h and stained with 100 μM chlortetracycline (CTC; Merck-Sigma), and longitudinal handmade (approximately 300-400 μm thick) sections were prepared. Then, analyses were carried out under a B2A filter of an epifluorescence microscope ( Fig. S3; B, b1-b6), and the total green fluorescence intensity (TFI) of Ca 2+ -CTC complexes was cytophotometrically measured using Scn Image software using photos 6,9 . Table 1. The concentrations, solvents, experimental destinations and origin of the modulators of cytokinin reception and metabolisms as well as blockers of calcium ions channels and its release from mitochondria as well as inhibitor of Raf-like kinase activities and RAF and MEK2 kinase substrates applied in the studies. www.nature.com/scientificreports/ During measurements, each stained cell was separately outlined using the threshold option, and then the values of fluorescence intensity in a.u. were read and used to calculate the relative Ca 2+ amounts. The decreasing amount of Ca 2+ was related to the values of green TFI of Ca 2+ -CTC complexes ( Fig. S3; B,b1-b6). The data represent the means ± SE of two replicates of three independent experiments (n = 3) from approximately 500-600 cells.
Estimation of ETH amount, RAF-like kinase and MEK2 activities and protein amount. ETH measurements were carried out in Erlenmeyer flasks before sample preparations. Erlenmeyer flasks were sealed with aluminium foil (to keep seedlings in the dark) and tightly closed with caps with clogged pipette tips. After 30 min of incubation, a handheld ETH analyser (SCS56, Storage Control System, United Kingdom) equipped with a pump was used. It was connected to pipette tips via a flexible tube directly before measurement (Fig. 2S). Then, the pump was turned on, and measurements were conducted for 30 s. The results were read between 20 and 30 s, when the values reached the plateau; the five readings from the monitor were written in a spreadsheet of MS Excel and taken to calculate the ETH amount in ppm per six seedlings. ETH was measured in Ctrl and Kin series and after treatment with 100 μM Ad (series Ad-100), 50 μM Ado (series Ado-50), 5 μM CPPU (series CPPU-5), 10 μM EGTA (series EG-10), 5 μM La 3+ (series La-5), 10  To estimate RAF-like and MEK2 kinase activities, one-third of the length of apical parts of roots was homogenized and reextracted in 0.04 M Tris-HCl pH 7.5 buffer (Sigma-Aldrich) containing 20 mM MgCl 2 , 10 µg ml -1 BSA (bovine serum albumin; Sigma-Aldrich) and 1 mM PMSF (phenylmethylsulfonyl fluoride; Sigma-Aldrich) in 1.5 ml Eppendorf-like tubes with a plastic mortar and centrifuged at 5000g for 10 min 7 .
The reaction mixture for kinase activity analyses was prepared by sequentially adding 2-ml tubes: the extract (20 µl), extraction buffer (1035 µl), ATP (5 µl), Syntide-2 or Mek2 (25 µl; 1 µg protein per 1 ml of buffer) as substrates and Kinase-Glo Reagent (50 µl; Promega), containing luciferase and luciferin. After mixing, the luminescent signals were measured in a semimicrofluorometer cell with a Teflon Stopper by a fluorescence/luminescence spectrophotometer F -2500 (Hitachi) at 458 nm wavelength every for 2.5 min for 30-s intervals. The kinase activities were calculated as the difference between the luminescence of the samples without substrates and the luminescence of the amount of ATP not utilized by kinase substrate phosphorylation. The kinase activity was expressed in relative light units (RLU), indicating the RLU amount of ATP utilized by kinases in 1 mg of protein.
To estimate the protein amounts, the apical parts of roots were homogenized with 100 mM Tris-HCl (pH 7.4) buffer in Eppendorf-like tubes using a plastic pestle (4-8 °C) and then centrifuged at 5000×g for 10 min (4 °C); residues were then re-extracted. Combined supernatants were used to measure protein amounts in the reaction mixture containing 2-ml Eppendorf-like tubes, extract (10-100 µl), extraction buffer (90-0 µl) and Coomassie Brilliant Blue G-250 reagent 56 (1.4 ml). Absorbance was measured at 595 nm after 10 min of incubation (Amersham Biosciences Ultrospec 1100 Pro UV-VIS spectrophotometer with semi-micro cell).
To calculate the protein amount, a standard, i.e., BSA dissolved in PHB, was prepared in a range of 5 to 100 μg of protein in a 100 µl volume, and measurements were carried out according to the above description. The prepared calibration curve was used to read the amount of protein in the sample and final calculations.
Protein estimation reagent was prepared with Coomassie Brilliant Blue G-250 (100 mg) diluted in 95% ethanol (50 ml), 85% H3PO4 (100 ml) and 950 ml of distilled water and stored in a dark bottle.
Statistics and software. Three biological replicates, at least in tri-, duplicate and more random samples, were analysed. The samples were prepared from at least six plants. The results of measurements were statistically verified by the Mann-Whitney U test and/or Student's t-test using MS EXCEL software (licenced) by independent step-by-step analyses of each column of results. Significant differences between results were observed at p ≤ 0.05. Calculations, all charts, and tables were prepared using MS EXCEL software (licenced).
To estimate the vitality of cells, by counting the number of live, dying, and dead cells, an Optiphot-2 epifluorescence microscope (Nikon) with a blue filter (B2A) equipped with a digital camera (DXM1200) and objectives (10×, 20×, 40×) ACT-1 software (Precoptic, Poland) and SCION IMAGE (Scion Corporation) software (open source) were used.
CORELDRAW GRAPHICS SITE X7 EDULIC and/or INSCAPE (open source) were used to prepare figures and image planes in tiff extensions. Images of seedlings were taken using a Canon 100 (Japan) camera in jpg extension.
BIORENDER software was used to prepare Fig. 9. Declaration. The authors confirm that all experiments with seedlings of nongenetically modified Vicia faba ssp. minor were performed in accordance with relevant guidelines and regulations using operation instructions for the laboratory equipment and measuring instruments. All applied study methods were performed in accordance with the relevant guidelines and regulations using protection equipment against hazards. Moreover, all images have been taken by the authors of the manuscript of the paper.