Phytosphingosine induces systemic acquired resistance through activation of sphingosine kinase

Abstract Phytosphingosine (PHS) is a naturally occurring bioactive sphingolipid molecule. Intermediates such as sphingolipid long‐chain bases (LCBs) in sphingolipid biosynthesis have been shown to have important roles as signaling molecules. PHS treatment caused rapid cell damage and upregulated the generation of reactive oxygen species (ROS) and ethylene in tobacco plants. These events were followed by the induction of sphingosine kinase (SphK) in a biphasic manner, which metabolized PHS to phytosphingosine‐1‐phosphate (PHS‐1‐P). On the other hand, a PHS treatment with a virulent pathogen, Phytophthora parasitica var. nicotianae (Ppn), alleviated the pathogen‐induced cell damage and reduced the growth of Ppn. A Ppn infection increased the PHS and PHS‐1‐P levels significantly in the upper part of the leaves at the infection site at the later stage. In addition, Ppn increased the transcription levels of serine palmitoyltransferase (LCB1 and LCB2) for sphingolipid biosynthesis at the later stage, which was enhanced further by PHS. Moreover, the PHS treatment increased the transcription and activity of SphK, which was accompanied by prominent increases in the transcription levels of ROS‐detoxifying enzymes and PR proteins in the later phase of the pathogen infection. Overall, the PHS‐induced resistant effects were prominent during the necrotic stage of this hemibiotrophic infection, indicating that it is more beneficial for inhibiting the pathogenicity on necrotic cell death. Phosphorylated LCBs reduced the pathogen‐induced cell damage significantly in this stage. These results suggest that the selective channeling of sphingolipids into phosphorylated forms has a pro‐survival effect on plant immunity.


| INTRODUCTION
Sphingolipids are essential structural components of the cellular membrane system, which includes the plasma membrane, tonoplast, and other endomembranes of plant cells that provide mechanical stability (Markham et al., 2006). It is estimated that sphingolipids constitute up to 30% of the tonoplast and plasma membrane lipids in plants (Lynch & Dunn, 2004). Further, sphingolipids are reported to participate as bioactive molecules in the regulation of intracellular processes including cell proliferation, differentiation, development, apoptosis, angiogenesis, and immunity in all eukaryotic and some prokaryotic cells (Aguilera-Romero et al., 2013). Moreover, sphingolipid metabolites have drawn attention as second messengers for stomatal closure, programmed cell death (PCD), and defense against pathogen attack in plants (Glenz et al., 2019;Magnin-Robert et al., 2015). Although the physiological roles of sphingolipid have been extensively studied in animals (Mashima et al., 2020), it remains unclear what mechanism is responsible for their physiological action in plants.
Long-chain bases (LCBs) are unique building blocks in all sphingolipids including ceramide, sphingomyelin, dihydrosphingosine, and phytosphingosine (PHS). Sphingolipid biosynthesis is initiated in the endoplasmic reticulum (ER) by condensation of serine with palmitoyl-coenzyme A catalyzed by serine palmitoyltransferase (SPT), resulting in the production of the LCB 3-ketodihydrosphinganine, which is then reduced to dihydrosphingosine (d18:0) (Chao et al., 2011). Dihydrosphingosine is further modified to PHS (t18:0) by the introduction of a double bond into the LCB in plants and fungi (Berkey et al., 2012), as well as to sphingosine, a free form of longchain sphingoid base, in mammals (Mashima et al., 2020). The highest proportion of PHS, a trihydroxylated LCB, has been reported in the leaves of tobacco (Nicotiana tabacum and Nicotiana benthamiana) with large proportions of d18:2 and t18:1 (Cacas et al., 2012). Therefore, the composition of sphingolipids in plants is notably different from those in animals and fungi, in which the major LCBs are t18:1 and d18:2.
One very important sphingolipid metabolite is sphingosine-1-phosphate (S1P), which is formed from sphingosine by sphingosine kinase (SphK). The activity of S1P has been described in a wide spectrum of organisms, ranging from Arabidopsis thaliana through Saccharomyces cerevisiae to Homo sapiens (Bourquin et al., 2010;Michaelson et al., 2009). S1P has received attention due to its regulation of many biological processes such as cell growth and survival, proliferation, differentiation, and immune function in eukaryotes (Proia & Hla, 2015;Spiegel & Milstien, 2003). S1P functions as an intracellular second messenger and a ligand for G-protein-coupled cell-surface receptors belonging to the lysophospholipid receptor family in mammals, which functions that have been implicated in cell growth and inhibition of apoptosis (Spiegel & Milstien, 2003). It was recently reported that S1P signaling via sphingosine 1-phosphate receptor-1 (S1PR1), a cellsurface receptor, can enhance tumorigenesis and stimulate growth, expansion, angiogenesis, metastasis, and survival of cancer cells (Cartier et al., 2020). Therefore, potential uses of S1P signaling modulators as pharmaceutical and therapeutic targets in cancer therapy have been suggested.
The evidence for S1P as a signaling molecule in mammals has been extended to plants. S1P is involved in abscisic acid (ABA)mediated guard cell signal transduction in response to chilling (Dutilleul et al., 2012) in Arabidopsis. Phytosphingosine-1-phosphate (PHS1P) is active in stress signaling, which is mediated by the Gprotein α-subunit in Arabidopsis (Coursol et al., 2005). More recent studies have observed that they participate as signaling molecules in the defense pathways of the hypersensitive response (HR) (Glenz et al., 2019).
The final step in the sphingolipid degradative pathway is mediated by S1P lyase (DPL1), which irreversibly converts S1P to hexadecenal and phosphoethanoamine (Magnin-Robert et al., 2015;Nishikawa et al., 2008). It has been suggested that DPL1 silences the alarms from the immune system by removing the available S1P signaling pools. However, there is the reversible dephosphorylation of S1P back to sphingosine, which is mediated by S1P phosphatase (SPP) (Zhang et al., 2012). Recent studies showing that both DPL1 and SPP regulate LCB/LCBP homeostasis have suggested new opportunities for the use of S1P-metabolizing enzymes as antiviral drugs by targeting the host enzymes (Wolf et al., 2019). Similar findings that modification of sphingolipid metabolite content can affect plant defense responses have been reported in Arabidopsis (Magnin-Robert et al., 2015).
PHS is a major LCB in some plants and is involved in cell signaling.
Sphingolipidomic profiling revealed that PHS accumulates as early as 1 h after pathogen inoculation with virulent and avirulent strains of Pseudomonas syringae pv. tomato in Arabidopsis leaves (Peer et al., 2010). We previously reported that PHS levels rapidly increased in susceptible tobacco (N. tabacum L. cv Wisconsin 38) plants at 1 and 48 h after shoot inoculation with the hemibiotrophic pathogen P. parasitica var. nicotianae (Ppn), which was determined by performing ultraperformance liquid chromatography-quadrupole-time of flight/ mass spectrometry (Cho et al., 2012). In this study, we tried to establish a pathophysiological link between pathogen infection and sphingolipid metabolites such as PHS and PHS1P in tobacco plants.
Further, we analyzed their contribution to plant defense.

| PHS-induced rapid responses
Although the physiological roles of sphingolipids are not fully described, several studies have indicated that they have crucial roles in the induction of apoptotic-like cell death in plants (Berkey et al., 2012). Initially, we determined the rate of cell damage in PHStreated tobacco plant stems with five to six leaves, followed by photography after staining with lactophenol trypan blue. The leaves were treated with PHS for 12 h at concentrations of 1-10 μM. The results showed that the PHS treatment induced rapid cell death from 1-μM PHS ( Figure 1a). An investigation of cell damage after the PHS treatment revealed severe damage to the leaves, even at 1-μM concentration, indicating that PHS initiates PCD through the HR response.
Therefore, we investigated the expression of genes related to cell damage after treatment with 1-μM PHS. We first determined the effects of PHS treatment on the expression of the metacaspase type II gene (MC2), which mediates biotic and abiotic stress-induced PCD (Watanabe & Lam, 2011). Following PHS treatment, the transcription level of NtMC2 gradually increased to a maximum level at 12 h (Figure 1b), after which it decreased somewhat until 48 h. Therefore, the absence of a further increase in NtMC2 transcription after 12 h indicates that cell damage did not progress into a further severe state, which is in accordance with the PHS-induced cell death pattern ( Figure 1a). These results indicate that a rapid upregulation of NtMC2 expression is responsible for the PHS-mediated HRs at an early stage.
F I G U R E 1 Effect of phytosphingosine (PHS) treatment on cell damage and transcription of sphingolipid biosynthesis-related genes in tobacco leaves. (a) Mature tobacco leaves were treated with different concentrations of PHS for 12 h, and after the indicated time, necrotic areas were stained with trypan blue and then imaged with a digital camera. (b) transcription levels of tobacco NtMC2 gene in tobacco plants after PHS treatment for 48 h. results of real-time qRT-PCR analysis of NtMC2 transcription after 1-μM PHS treatment using total RNAs from tobacco leaves. (c-f) Transcription levels of LCB1 (c) and LCB2 (d) of serine palmitoyltransferase and ORM1 (e) and ORM2 (f) of orosomucoid in tobacco leaves after 1-μM PHS treatment. (g and h) Transcription level (g) and activity (h) of sphingosine kinase in tobacco leaves after 1-μM PHS treatment. (i) Transcription levels of DPL1 of sphingosine-1-phoshpate lyase in tobacco leaves after 1 μM PHS treatment. Transcription levels are expressed relative to the reference gene β-actin after real-time qRT-PCR. Relative mRNA expression levels are expressed as means AE SD. One asterisk (P < .05) or two asterisks (P < .01) indicate a significant difference between mock-and PHS-treated cases at the same indicated time We next determined whether alteration of sphingolipid biosynthesis-related gene expression occurs after PHS treatment.
LCB molecules, which are unique components of sphingolipids, are formed by condensation of serine and palmitoyl-coenzyme A (Dietrich et al., 2008). This reaction is the first distinctive step in sphingolipid biosynthesis, which is catalyzed by the pyridoxal phosphatedependent enzyme SPT (EC 2.3.1.50). SPT, an ER-associated heterodimeric protein consisting of LCB1 and LCB2 subunits, is thought to be a rate-limiting enzyme in the sphingolipid biosynthetic pathway (Dietrich et al., 2008). It was recently reported that PHS is sufficient to induce ER stress surveillance phenotypes in budding yeast, S. cerevisiae, which in turn elevates the PHS level, suggesting the presence of feedback activation in pathogen-induced sphingolipid biosynthesis.
Gene expression levels of LCB1 and LCB2 subunits were determined in order to elucidate the effects of PHS in sphingolipid biosynthetic pathways (Figure 1c localization, but may directly or indirectly inhibit its activity (Han et al., 2019). In particular, although the ORM proteins negatively regulate SPT, the regulatory mechanisms of these proteins in the sphingolipid biosynthetic pathway are unclear (Alsiyabi et al., 2021).
We next investigated the effects of PHS treatment on the transcription and activity of SphK, which catalyzes PHS to PHS1P, which is abundant in fungi and plants (Dutilleul et al., 2012). PSH1P has a fundamental role as an intracellular signaling molecule in development and stress responses in eukaryotic cells (Piña et al., 2018). Although the transcription level of SphK was immediately increased after PHS treatment, it returned to the basal level after 12 h ( Figure 1g). The amount of SphK transcripts began to increase again after 24 h of PHS treatment, reaching a maximum at 96 h and then decreasing. However, no change was observed for the entire period following mock treatment.
The level of SphK activity responded to PHS treatment in a biphasic manner (Figure 1h). SphK activity began to increase rapidly and peaked at 1 h, after which it returned to the basal level. However, it increased again after 24 h and reached a peak at 96 h, similar to the maximum pattern level of the SphK transcript ( Figure 1h). These results suggest that the PHS-induced activation of SphK occurs at the transcriptional level. It is suggested that exogenously added PHS can be metabolized to PHS1P at a later stage, which is known to be involved in intracellular signaling (Coursol et al., 2005).
The final step in sphingolipid metabolism is mediated by DPL1, which degrades long-chain base-1-phosphates (LCBPs) such as PHS1P and is the only reported route for the destruction of sphingolipids (Magnin-Robert et al., 2015). The amount of DPL1 mRNA induced after PHS treatment was biphasic (Figure 1i). The amount of DPL1 mRNA rapidly increased to a peak at 0.5 h after PHS treatment; after which, it lowered to basal levels before gradually increasing up to 120 h.

| NADPH oxidase-dependent transient ROS accumulation at an early stage after PHS treatment
It has been recognized that reactive oxygen species (ROS) generation and signaling can activate HR-related cell death in plants (Zurbriggen et al., 2010). PHS is considered a powerful contributor to oxidative damage in eukaryotic cells. To investigate ROS generation in tobacco leaves following treatment with 1-μM PHS, we histochemically monitored the levels of two important ROS, superoxide anion and hydrogen peroxide, by using 3,3 0 -diaminobenzidine (DAB) and nitrotetrazolium blue chloride (NBT) as chromogenic substrates, respectively, in tobacco leaves for up to 48 h. NBT reacts with O 2˙À to form a dark blue insoluble formazan compound, and DAB is oxidized by hydrogen to produce a reddish-brown precipitate (Kumar et al., 2014). Both ROS were significantly detected from 15 min after PHS treatment and abundance peaked at 1 h, suggesting the ROS were rapidly and transiently generated (Figure 2a,b). Surprisingly, only very low levels of both ROS were produced after 3 h ( Figure S1).
These results imply that PHS-induced ROS generation is only HRrelevant at an early stage. NtRbohD is more responsive than NtRbohF during PHS-induced ROS generation.

| PHS-induced ethylene production is dependent on HR-related NtACS2 and NtACS4 expression
Ethylene is implicated as a virulence factor of pathogens as well as a signaling molecule in disease resistance . Pathogen infection was shown to induce typical responses, including biphasic production of ROS and ethylene, in which synergism between ROS and ethylene constitutes an important regulatory mechanism in tobacco leaves . Therefore, we propose that ethylene production could be a central component of a self-amplifying loop wherein transient biphasic ROS bursts have a critical regulatory role.
To further determine whether or not ethylene functions as a physiological amplifier in PHS-induced cell death, we measured ethylene production after PHS treatment. Monophasic ethylene production was observed in tobacco leaves after PHS treatment (Figure 3a), F I G U R E 2 Effects of phytosphingosine (PHS) treatment on reactive oxygen species (ROS) accumulation and NADPH oxidase gene (NtRbohD and NtRbohF) transcription in tobacco leaves. (a and b) Histochemical analysis of intracellular ROS accumulation in PHS-treated tobacco leaves. After mature tobacco leaves were treated with 1-μM PHS for 48 h, superoxide anion level was determined by nitrotetrazolium blue chloride (NBT) staining (a), and H 2 O 2 level was detected by 3,3 0 -diaminobenzidine (DAB) staining (b). Staining images of leaves were photographed by a digital camera. (c and d) Relative mRNA levels of NtRbohD and NtRbohF genes in mature tobacco leaves treated with 1-μM PHS. Transcription levels of NtRbohD (c) or NtRbohF (d) are expressed as means AE SD. Transcription levels are expressed relative to the reference gene β-actin after qRT-PCR. An asterisk indicates a significant difference between mock-and PHS-treated cases (one asterisk [P < .05] or two asterisks at the same time point [P < .01]) which was in accordance with the observation of monophasic ROS production ( Figure 2a,b). Ethylene production rapidly increased at 30 min, reached a transient peak at about 1 h, and declined thereafter.
After 30 h of PHS treatment, ethylene production had completely returned to the basal level. Taken together, the results show that PHSinduced ROS generation is followed by PHS-induced ethylene production, suggesting that ROS generation functions upstream of ethylene generation in response to PHS treatment.
Abiotic stresses are known to induce ethylene production through gene-specific expression of 1-aminocyclopropane-1-carboxylic acid (ACC) synthase (ACS) members in a time-dependent manner (Wi et al., 2010). HR-related NtACS2 and NtACS4 expressions have been shown to induce early ethylene production, as early as after 1 h, in response to either abiotic/oxidative or biotic stress induced by the compatible pathogen Ppn, whereas necrosis-or senescence-related NtACS1 expression was shown to be responsible for late-phase ethylene production at 24 or 72 h for induction of oxidative or biotic stress, respectively (Wi et al., 2010. In this study, we measured the transcription levels of NtACS1, NtACS2, and NtACS4 in response to PHS treatment by performing qRT-PCR. Although NtACS2 transcription was detected at a relatively low level in mock-treated plants at 0.5 and 48 h, it was upregulated by about 3.8-fold at 0.5 h in PHS-treated tobacco plants compared with that in mock-treated plants ( Figure 3b). However, it returned to the basal level at 48 h, at which time there was no significant difference in NtACS2 expression between PHS-and mock-treated plants.
On the other hand, NtACS4 expression was remarkably upregulated, by approximately 59-fold, at 3 h compared with mock-treated tobacco plants. However, HR-related NtACS4 transcription was almost absent at 48 h in both mock-and PHS-treated plants. These results are consistent with our previous finding that apoptotic-like cell death, which was dependent on NtACS2 and NtACS4, was not further induced at 48 h after PHS treatment.
We previously reported that expression of NtACS1, a senescence-or necrosis-related ACS gene member, increased starting  Real-time qRT-PCR is a reliable technique for the detection and quantification of plant pathogens, and it is increasingly being used in plant pathology investigations. We previously reported that qRT-PCR was a highly sensitive method for the quantification of Ppn in plants . We designed primers specific to 5.8S rRNA (GenBank AY769953) as an internal control for monitoring Ppn growth. Using qRT-PCR, expression of the 5.8S rRNA gene was compared with that of tobacco β-actin, which was used as the internal plant reference gene. Pathogen growth was continuous until 96 h, revealing an S-shaped growth curve after Ppn inoculation, which was significantly inhibited by PHS treatment; moreover, the inhibitory effect was more prominent during the necrotrophic stage ( Figure 4d).

| Induced resistance by the treatment with PHS and PHS1P under virulent pathogen infection
Although PHS caused rapid cell death, PHS exhibited pathogen resistance at the necrotic stage, which was the later stage of a Ppn infection, but the reason for this is unclear. This study examined the significant increases in the transcript level and enzyme activity of SphK at 96 h after a PHS treatment (Figure 1g,h). Therefore, under a F I G U R E 5 Contents of phytosphingosine (PHS) and phytosphingosine-1-phosphate (PHS1P) and necrosis in tobacco leaves in Ppninoculated tobacco leaves. (a) Amount of PHS and PHS1P produced by the upper third and fourth leaves of the tobacco plants at 1 and 96 h after inoculating the stem with Ppn. (b) Ppn-inoculated tobacco plants determined using trypan blue staining and imaging with a digital camera F I G U R E 6 Effect of sphingosine-1-phosphate (S1P) and phytosphingosine-1-phosphate (PHS1P) treatment on cell damage in Ppn-inoculated tobacco leaves. Necrotic areas in Ppn-inoculated tobacco plants were determined after 1 μM of S1P or PHS1P treatment by using trypan blue staining and imaging with a digital camera. Cell damage was detected after shoot inoculation with Ppn for 24 and 72 h, during which 100 μl of 1-μM PHS was slowly dropped on the shoot area using a micropipette tip  (Figures 9 and S2). This result is contrary to the PHS-induced upregulation of NtRbohD transcription in the HRrelevant stage at 1 h without Ppn infection (Figure 2c). In addition, PHS did not affect Ppn-induced expression of NtACS2 and NtACS4 for ethylene production at the early stage but slightly reduce that of NtACS1 at the later stage ( Figure S3).
ROS production at a later stage might be a byproduct or harmful substance of cell damage in the necrotic stage of compatible pathogen infection . Therefore, we investigated whether PHS contributes to the expression of ROS-detoxifying enzymes during the later stage of Ppn infection. An enzymatic dismutation reaction converts superoxide into a more stable, membrane-permeable hydrogen peroxide (H 2 O 2 ) derivative, which is required for cell-to-cell signaling. Sphingolipids, as bioactive molecules, have been shown to have physiological functions in the stress responses and PCD in plants (Berkey et al., 2012). Sphingolipid accumulation is an intrinsic early step in HR activation during the pathogen response. Sphingolipids derived from the fungal pathogen Magnaporthe grisea have been shown to induce phytoalexin accumulation, PCD, and resistance to infection in rice (Koga et al., 1998). It was reported that sphingolipids beneficial regulator in the pathogenesis process (Berkey et al., 2012). Therefore, it is necessary to examine how PHS plays this role. It can be suggested that the rapid accumulation of sphingolipid bases such as sphingosine and PHS is an important step for the resistance response against pathogen attack. However, exactly how pathogens trigger PHS accumulation has not been described, and whether or not PHS accumulation has a role in the restriction of bacterial growth during the pathogen response also remains to be elucidated. Further, the physiological significance of PHS accumulation in pathogen-infected plants requires explanation. On that basis, our study first focused on the pathophysiological significance of PHS.

ROS
Although PHS induced acute transient accumulation of ROS and ethylene (Figures 2 and 3), which are associated with HR-related cell death (Figure 1a,b), the transcription and activation of SphK increased in a biphasic pattern under PHS treatment (Figure 1g,h). In the later stage of the Ppn infection, pathogen growth and pathogen-induced cell damage were significantly suppressed by PHS treatment (Figure 4b-f). In addition, the levels of PHS and PHS1P were increased dramatically at 96 h after Ppn inoculation ( Figure 5). These observations are consistent with the results showing that an exogenous treatment with PHS1P and S1P can suppress pathogen-induced plant damage significantly ( Figure 6).
A biphasic defense response against the hemibiotroph, such as Ppn, initially induces Ca 2+ signaling and the ROS-mediated pathway before 6 h, which is followed by ROS detoxification and PR generelated defenses from 18 to 24 h after pathogen infection (van den Berg et al., 2018;Wi et al., 2012). This suggests that these inductions of a later stage could play a key role in the resistance to hemibiotroph pathogen. Therefore, it could be suggested that these characteristics were related to the abrogation of the late phase of ROS accumulation and necrotic cell death. As PHS and its phosphorylated form have been reported to mediate pathogen resistance in several plants against necrotrophic pathogens (Magnin-Robert et al., 2015), these results suggest that PHS might be involved in the development of F I G U R E 9 Effects of phytosphingosine (PHS) treatment on reactive oxygen species (ROS) production and NADPH oxidase gene (NtRbohD and NtRbohF) transcription in pathogen-infected tobacco leaves. (a) After histochemical analysis of intracellular ROS accumulation in Ppn-infected tobacco leaves after PHS treatment, the intensity of fluorescence was quantified by ImageJ software. After mature tobacco leaves were treated with 1-μM PHS for the indicated time, ROS was determined by incubation with DCFH-DA for 10 min. Staining images of leaves were obtained by confocal microscopy and then quantified by ImageJ software. (b and c) Relative mRNA levels of NtRbohD and NtRbohF genes in mature tobacco leaves infected with Ppn and then treated with PHS. Transcription levels of NtRbohD (b) or NtRbohF (c) are expressed as means AE SD. Transcription levels are expressed relative to the reference gene β-actin after qRT-PCR. Relative mRNA expression levels are expressed as means AE SD. Asterisks indicate a significant difference in PHS-treated or co-treated cases with PHS and Ppn infection from mock-treated cases at the same indicated time (one asterisk [P < .05] or two asterisks [P < .01]) protective machinery against pathogen-induced cell death through conversion of PHS into other protective compounds such as PHS1P.
Interestingly, alteration of the balance between PHS and phosphorylated PHS is commonly known to affect sphingolipid-mediated PCD or cell survival in plants and animals (Sánchez-Rangel et al., 2015). These results indicate that the magnitude of SphK transcription is related to the progression of plant tolerance against pathogenicity. Moreover, PHS upregulated the expression of ROS-detoxifying enzymes, which may be responsible for decreasing ROS levels, and the expression of PR genes at the later stage of Ppn infection (Figures 10 and 11). Elevation of PR gene expression resulted in disease resistance such as SAR in the necrotic stages of Ppn infection in PHS-treated tobacco leaves.
A previous study reported that the PHS level increased biphasically (by 1.8-fold at 1 h and 2.36-fold at 48 h) (Cho et al., 2012), and the SphK activity increased continuously up to 96 h after virulent pathogen inoculation of tobacco plants (Figure 7b). In another study, the ceramide kinase mRNA level was upregulated by about fivefold at 24 h in plants infected with virulent P. syrangae compared to the level in an uninfected control (Liang et al., 2003). Collectively, these observations indicate that PHS can act as a signaling molecule by activating SphK transcription for the suppression of necrotic cell death when infected by the hemibiotrophic pathogen P. parasitica. Therefore, based on the profile of SphK transcription, increased induction of SphK is notably advantageous to plant immunity.
Interesting evidence has indicated that phosphorylated sphingosines such as S1P can act as potent bioactive lipids for the survival of cancer cells (Riccitelli et al., 2013). Other studies have shown that SphK inhibition results in apoptosis in animal xenografts (Kapitonov et al., 2009). Further, exogenous S1P supplementation can propel murine splenocytes toward a pro-survival outcome in response to hypoxia-induced injury (Chawla et al., 2014). Moreover, PHS1P is known to increase the cell viability of human dermal fibroblasts via the c-Jun N-terminal kinase/Akt pathway (Lee et al., 2012).
It has also been reported that SphK is activated by ABA in A. thaliana for the inhibition of stomatal opening and the promotion of stomatal closure, suggesting that S1P is a signaling molecule involved in ABA regulation of guard cell turgor (Coursol et al., 2005).
In addition, overexpression of the rice S1P lyase gene OsDPL1 in transgenic tobacco was shown to result in reduced tolerance to salt F I G U R E 1 0 Relative transcription levels of endogenous reactive oxygen species (ROS) detoxification enzymes, CAT1, CAT2, MnSODmi, CuZnSODc, APXc, and GSTF, after 1-μM phytosphingosine (PHS) treatment for 96 h in leaves of tobacco plants. PHS-treated whole leaves were subjected to real-time qRT-PCR analysis. Transcription levels are expressed relative to the reference gene β-actin after qRT-PCR. Relative mRNA expression levels are expressed as means AE SD. Asterisks indicate a significant difference in PHS-treated or co-treated cases with PHS and Ppn infection from mock-treated cases at the same time point (one asterisk [P < .05] or two asterisks [P < .01]) and oxidative stresses compared to that in WT plants (Zhang et al., 2012). ROS generation induced by LCBs is specifically blocked by their phosphorylated forms, indicating that maintenance of homeostasis between a free sphingolipid base and a phosphorylated derivative is critical to determining cell fate (Shi et al., 2007).
In conclusion, these observations in this study suggest that elevated gene expressions of long-chain sphingolipid bases for de novo synthesis of a sphingolipid is an important determinant for preventing necrotic cell death in response to pathogen attack. These observations also suggest that an elevated ratio of phosphorylated/ unphosphorylated long-chain sphingolipid bases induced by SphK may be a more significant determinant for promoting resistance during plant-pathogen interactions. Therefore, the rapid induction of PHS is beneficial for the activation of SphK and inhibition of pathogenicity in the necrotic stage of a hemibiotrophic infection, resulting in the development of SAR in plant immunity. Taken together, our results indicate that the selective channeling of sphingolipids into their phosphorylated forms in conjunction with detoxification of stress-induced ROS has physiological pro-survival effects related to resistance in plants exposed to biotic stresses. sphingosine-1-P (S1P) was purchased from Sigma-Aldrich (St. Louis, Missouri, USA. Solutions of PHS, PHS1P, and S1P with 20-mM MES buffer (pH 6.1) were applied to whole leaves through petioles or stems with five leaves. Tobacco shoots with five to six leaves were inoculated directly with a pathogen plug (1 cm diameter) in a culture bottle containing solid half-strength MS medium. The mock treatment was the control group in the absence of PHS (Figures 1-3 and 4a). In addition, the mock treatment was a control tobacco shoot on the MS media plug without Ppn (Figures 4b-f and 5-11).
F I G U R E 1 1 Effects of phytosphingosine (PHS) treatment on the transcription of PR genes. Relative transcription levels of PR proteins under pathogen infection in response to PHS treatment for 96 h. transcription levels of the PR genes PR-1a, PR-3, PR-4b, taumatin-like proteins (TLP), and osmotin (OSM) for PR5, PR-6, SAR8.2, and NPR1 were determined in tobacco leaves after treatment with 1-μM PHS. Transcription levels are expressed relative to the reference gene β-actin after qRT-PCR. Relative mRNA expression levels are expressed as means AE SD. Asterisks indicate a significant difference in PHS-treated or co-treated cases with PHS and Ppn infection from mock-treated cases at the same time point (one asterisk [P < .05] or two asterisks [P < .01]) 4.2 | RNA isolation and real-time qRT-PCR Total RNA isolation was performed as described previously (Seo et al., 2020). After 1 μg of total RNA from leaves was reverse transcribed using a High Fidelity PrimeScript™ RT-PCR Kit (Takara, Japan), and real-time qRT-PCR was performed using a Thermal Cycler Dice ® Real Time System III T950 (Takara, Japan) with gene-specific PCR primers (Table S1). Relative expression levels in each cDNA sample were normalized to the reference gene β-actin.

| Ethylene measurement
Ethylene production by PHS-treated plants was measured by gas chromatography (Hewlett Packard 5890 Series II, Wilmington, DE, USA) using an activated alumina column at 250 C and a flame ionization detector.

| Trypan blue staining
To monitor plant cell death, tobacco leaves were stained as previously described (Seo et al., 2020). PHS-treated tobacco whole leaves were immersed for 1 min in a boiling solution of 10 ml of lactic acid, 10 ml of glycerol, 10 g of phenol, and .4% (w/v) trypan blue. Stained plants were decolorized overnight and then photographed using a digital camera.

| Quantitation of sporangia
Pathogen isolation from Ppn-infected tobacco leaves was performed following a modified method from that in a previously reported protocol (Sharma & Ghosh, 2016). The pathogen-infected tobacco leaves were sterilized outside using a 2% NaClO (v/v) solution. After sterilized leaves were flooded with phosphate-buffered saline (PBS) at 4 C for 30 min and then incubated at 25 C for 16-20 h. After sonication, the number of sporangia was determined by using a hemocytometer.

| Quantification of PHS and PHS1P in tobacco leaves
The leaves samples (100 mg) were homogenized in 300 μl of isopropyl alcohol using a Tissuelyser (Qiagen, Hilden, Germany) and then incubated at À20 C for 60 min. After centrifugation of the mixture at 15,000Âg for 15 min, 5 μl of the supernatants were injected to UPLC-MS/MS system. The chromatographic system consisted of a Waters mode. The data were processed by using MassLynx™ 4.1 software.