Sánchez‑Sanuy et al. Rice (2022) 15:68 https://doi.org/10.1186/s12284-022-00609-w RESEARCH Open Access Iron Induces Resistance Against the Rice Blast Fungus Magnaporthe oryzae Through Potentiation of Immune Responses Ferran Sánchez‑Sanuy1, Roberto Mateluna‑Cuadra1, Keisuke Tomita2, Kazunori Okada2, Gian Attilio Sacchi3, Sonia Campo1,5* and Blanca San Segundo1,4* Abstract Iron is an essential nutrient required for plant growth and development. The availability of iron might also influence disease resistance in plants. However, the molecular mechanisms involved in the plant response to iron availability and immunity have been investigated separately from each other. In this work, we found that exposure of rice plants to high iron enhances resistance to infection by the fungal pathogen Magnaporthe oryzae, the causal agent of blast disease. RNA-Seq analysis revealed that blast resistance in iron-treated rice plants was associated with superinduc‑ tion of defense-related genes during pathogen infection, including Pathogenesis-Related genes. The expression level of genes involved in the biosynthesis of phytoalexins, both diterpene phytoalexins and the flavonoid phytoalexin sakuranetin, was also higher in iron-treated plants compared with control plants, which correlated well with increased levels of phytoalexins in these plants during M. oryzae infection. Upon pathogen infection, lipid peroxidation was also higher in iron-treated plants compared with non-treated plants. We also show that M. oryzae infection modulates the expression of genes that play a pivotal role in the maintenance of iron homeostasis. Histochemical analysis of M. oryzae-infected leaves revealed colocalization of iron and reactive oxygen species in cells located in the vicinity of fun‑ gal penetration sites (e.g. appressoria) in rice plants that have been exposed to iron. Together these findings support that ferroptosis plays a role in the response of iron-treated rice plants to infection by virulent M. oryzae. Understanding interconnected regulations between iron signaling and immune signaling in rice holds great potential for developing novel strategies to improve blast resistance in rice. Keywords: Blast, Ferroptosis, Immune response, Iron, Magnaporthe oryzae, Oryza sativa, Phytoalexins Background (Connorton et al. 2017). Although iron is abundant in Iron (Fe) is an essential nutrient element required for most agricultural soils, its bioavailability to plants is important processes in plant growth and development, usually very low. In alkaline soils, Fe is predominantly such as photosynthesis and mitochondrial respiration. found in the form of insoluble ferric form ­(Fe3+), form- It is commonly associated to cellular reduction–oxi- ing hydroxide complexes poorly available for the plant. dation (redox) reactions and electron transfer chains In contrast, under reductive conditions, Fe is present as the soluble ferrous form (­Fe2+). Both deficiency and excess of Fe can be toxic to the plant. In particu- *Correspondence: soniacampo@fundaciomiquelagusti.cat; blanca. lar, excess Fe leads to the generation of reactive oxygen sansegundo@cragenomica.es; blanca.sansegundo@csic.es species (ROS) through redox reactions between F ­ e3+ 2+ 1 Centre for Research in Agricultural Genomics (CRAG), CSIC-IRTA-UAB-UB, and ­Fe (Haber–Weiss and Fenton reactions) which Campus Universitat Autónoma de Barcelona (UAB), Bellaterra (Cerdanyola del can be harmful to the plant. ROS can damage lipids Vallés), Barcelona, Spain Full list of author information is available at the end of the article (e.g. membrane lipids), proteins and nucleic acids and © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 2 of 23 eventually, cell death (Mansoor et al. 2022). Therefore, The iron nutritional status might have an effect on plants have evolved homeostatic mechanisms to regu- disease resistance in plants (Herlihy et al. 2020; Liu late Fe accumulation and provide enough amounts for et al. 2007). Considering that foliar pathogens entirely the plant metabolism while preventing Fe accumulation depend on the host tissue to acquire this vital ele- causing oxidative stress. ment, during pathogen infection there is a competi- Plants use two main approaches to take up Fe from tion between the host and the pathogen for Fe. In the soil, referred to as Strategy I and Strategy II (also plant/pathogen interactions, the plant might use dif- known as reducing and chelating strategies, respec- ferent strategies to reduce pathogen virulence. On the tively) (Marschner and Römheld 1994). The main dif- one side, plants might use iron-withholding strate- ference between both strategies is the oxidation state of gies by sequestering iron from pathogens, a phenom- Fe when taken up by the plant, ferrous ­Fe2+ for Strategy enon originally described in animals and referred to as I and ferric ­Fe3+ for Strategy II. The strategy I (reduc- “nutritional immunity” (Weinberg 1975). On the other tion strategy) is found in all plants except those from hand, the host plant might capitalize the toxicity of the Poaceae family, and has been characterized in the iron through local accumulation of Fe that can be toxic model plant Arabidopsis thaliana. It consists in lower- for the invading pathogen (Herlihy et al. 2020). Plant ing soil pH by extrusion of protons to increase Fe solu- pathogens might also deploy strategies for Fe acquisi- bility which relies on plasma membrane H ­ +-ATPases tion from the host plant, such as the secretion of high- 3+ (Santi and Schmidt 2009). The F ­ e is then reduced to affinity Fe-binding siderophores (Herlihy et al. 2020). ­Fe2+ by a ferric reductase-oxidases (FROs) located in The deployment of these strategies by the host and/ the plasma membrane and imported into the root epi- or the pathogen might determine the outcome of the dermal cells by iron-regulated transporters (IRTs) (Eide interaction. et al. 1996; Robinson et at. 1999). Additional mecha- Sensing of Fe depletion has been proposed as a mech- nisms used in Strategy I plants involves the release of anism by which plants recognize a pathogen threat Fe chelating compounds, such as phenolic compounds, for the activation of immune responses (Herlihy et al. to the rhizosphere (Rodríguez-Celma et al. 2013). In 2020). Supporting this notion, resistance to infection Strategy II (chelation strategy), which is found in gram- by Dickeya dadantii and Botrytis cinerea was observed inaceous species, the roots produce phytosiderophores in Fe-starved Arabidopsis plants (Kieu et al. 2012). (PS) which are transported to the rhizosphere by TOM In other studies, Fe-starved maize plants were found (Transporter of mugineic acid) that chelate F ­ e3+ in the to be unable to produce ROS during Colletotrichum 3+ soil (Nozoye et al. 2011). The PS-Fe complexes are graminicola infection which correlated with increased then transported into the root cells by yellow stripe-like susceptibility to this fungal pathogen (Ye et al. 2014). (YSL) transporters (Nozoye et al. 2011). Rice, unlike On the other hand, local excess iron can activate ROS most graminaceous crops, is well adapted for growth burst for pathogen resistance as recently described in under submerged conditions (paddy fields) where F ­ e2+ maize plants during infection by Curvularia lunata 3+ is frequently more abundant than ­ Fe . Under such causing leaf spot (Fu et al. 2022). At present, crosstalk conditions, the absorption of excessive ­Fe2+ by the rice mechanisms that adjust iron homeostasis and immune roots might cause oxidative stress and toxicity. Rice, responses in plants are not fully understood, particu- in spite of being a grass, uses a combined strategy for larly in crop species. Fe uptake which has features of Strategy II (PS release Rice (Oryza sativa L.) is one of the most impor- through TOM and PS-Fe3+ uptake through YSL), as tant crops in the world, and its production is severely well as features of Strategy I ­(Fe2+ uptake using IRT affected by the fungus Magnaporthe oryzae. This fun- transporters) (Ishimaru et al. 2006; Inoue et al. 2009; gus is the causal agent of the rice blast disease (Wilson Nozoye et al. 2011). and Talbot 2009; Fernandez and Orth 2018). In previ- Due to its potential toxicity, plants develop mechanisms ous studies, it was described that F ­ e3+ and ROS (e.g. to regulate the transport, utilization and storage of Fe. ­H2O2) accumulate in rice leaves during infection with For translocation, Fe is complexed to chelators, namely an avirulent strain of M. oryzae (Dangol et al. 2019). In the phytosiderophore deoxymugineic acid (DMA). In the this work, we investigated the effect of treatment with xylem, ­Fe3+-citrate is the dominant form of Fe transport high Fe on blast resistance, at the molecular and cellu- (Connorton et al. 2017). Fe storage mechanisms in plants lar level. RNA-Seq analysis revealed that short exposure include sequestration into vacuoles as well as binding to of rice plants with high Fe is accompanied by stronger Ferritins (Aung and Masuda 2020). Compared to roots, induction of defense responses, including Pathogenesis- our knowledge on mechanisms of regulation of Fe trans- Related (PR) gene expression. Phytoalexin biosynthesis port and homeostasis in leaf tissues is still limited. genes were also strongly induced by M. oryzae infection Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 3 of 23 ◂ Fig. 1 Resistance to M. oryzae infection in rice plants grown under high Fe supply. A Experimental design used for iron treatment and infection with M. oryzae (5 × ­105 spores/ml). Rice plants were grown in soil for 16 days under sufficient Fe supply (50 µM Fe, Control plants) and allowed to continue growth under sufficient conditions (Ctrl) or high Fe conditions (1 mM Fe; High Fe) for 5 days more (left panel). Then, the rice plants were inoculated or not with M. oryzae spores and assayed for blast resistance. B Appearance of rice plants that have been treated with high Fe for 5 days. C Shoot and root fresh weight (FW, left and middle panels), and chlorophyll content (right panel) of plants treated with high Fe for 5 days. Differences were not statistically significant. D Total Fe content estimated by the ferrozine colorimetric method in roots, stems and leaves of Control and High Fe-treated plants (5 days of treatment). Data are mean ± SEM (n = 10). Asterisks indicate statistical significance calculated by t-test (*, P < 0.05; **, P < 0.01). E Resistance to M. oryzae in Fe-treated rice plants. Control and High Fe-treated rice plants were inoculated with a conidial suspension of M. oryzae spores (5 × ­105 spores/ml), or mock-inoculated. Left panel, Disease symptoms in youngest developed leaves (3rd leaf ) at 7 days post-inoculation (dpi) are shown. Results from one representative experiment of four independent experiments are presented. Middle panel, percentage of diseased area at 7 dpi (n = 10). Right panel, fungal biomass was quantified by qPCR using specific primers of the M. oryzae 28S ribosomal gene relative to the rice Ubiquitin 1 gene (Os06g46770) at 7 dpi, (−, mock-inoculated; +, M. oryzae-inoculated). Data are mean ± SEM (n = 10). Asterisks indicate statistically significant differences calculated by t-test (***, P < 0.001) Results Treatment with High Fe Enhances Resistance to M. oryzae Infection in Rice Plants In this work, we investigated the effect of treatment with high Fe on resistance to infection by M. oryzae in rice plants. Rice (O. sativa cv. Nipponbare) plants were grown in soil under sufficient Fe supply (50 µM Fe) for in Fe-treated rice plants, which correlates well with a 16 days. Then, half of the plants were supplied with 1 mM higher accumulation of phytoalexins in these plants. Fe while the other half of the plants continued growth Superactivation of defense genes is associated to resist- under 50 µM Fe condition (henceforth High-Fe and ance to M. oryzae infection in Fe-treated rice plants. Control Fe plants, respectively). Treatment with high Fe Pathogen infection also modulates the expression of was carried out for 5 days (Fig. 1A). No phenotypic dif- genes involved in iron homeostasis in rice leaves and ferences were observed between Control and High-Fe provokes alterations in Fe content (total Fe, apoplastic plants after 5 days of Fe treatment (Fig. 1B), these plants Fe). Histochemical staining revealed co-localization in showing similar shoot biomass, root biomass and chloro- ROS and Fe during infection of Fe-treated rice plants phyll content (Fig. 1C). High-Fe plants had a higher con- with a virulent M. oryzae isolate (compatible interac- tent of Fe in root, stem and leaf tissues compared with tion), these plants also showing increased levels of lipid Control Fe plants (Fig. 1D). Under the same experimen- peroxidation during M. oryzae infection. Treatment tal conditions, cupper, manganese or calcium levels were with the ferroptosis inhibitor Ferrostatin-1 (Fer-1) sup- not significantly altered in High-Fe plants compared with presses Fe accumulation suggesting that ferroptosis Control-Fe plants (Additional file 1: Fig. S1A). When underlies blast resistance in Fe-treated rice plants. Col- using a longer period of Fe treatment (e.g. 19 days of lectively, these results further support a relationship treatment with 1 mM Fe), plants developed symptoms of between the Fe signaling and immune signaling in rice Fe toxicity in their leaves, these plants also showing a sig- while highlighting the relevance of ferroptosis in confer- nificant reduction in root and shoot biomass and chloro- ring blast resistance in rice. phyll content (Additional file 1: Fig. S1B, C). To avoid that Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 4 of 23 toxic effects caused by Fe accumulation could confound Transcriptome analysis also indicated that treatment our results on disease resistance, in this study we used a with high Fe results in alterations in the expression of short-term Fe treatment (5 days of treatment with 1 mM genes involved in ROS metabolism, in particular per- Fe), followed by a M. oryzae infection for 7 days (Fig. 1A). oxidase (prx) genes (Fig. 2B, Additional file 2: Table S1a). Most importantly, upon infection with M. oryzae, Compared with Control plants, the expression of prx High-Fe plants consistently exhibited resistance to genes was either up-regulated (OsPrx62, OsPOX22.3) or infection as revealed by visual inspection of disease down-regulated (OsPrx11, OsPrx15, OsPrx16, OsPrx38, symptoms, quantification of diseased leaf area and meas- OsPrx81, OsPrx83, OsPrx89) in High-Fe plants. Peroxi- urement of fungal biomass in leaves of pathogen-infected dases participate in a broad range of processes, such as plants (Fig. 1E). These results were substantially similar cross-linking of cell wall components, synthesis of phyto- to those previous reported in hydroponically-grown rice alexins, and ROS production in plant defense responses. plants (Peris-Peris et al. 2017) supporting that treatment They catalyze oxidation of various substrates concomi- with high Fe increases resistance to infection by the rice tant with the decomposition of ­H2O2. Among peroxi- blast fungus in rice. dases, Class III peroxidases can either produce ­H2O2 in oxidase cycle or scavenge H ­ 2O2 to ­H2O and ­O2 through Transcriptional Responses of Rice Leaves to Treatment peroxidase activity. POX22.3 was previously identified with High Fe among proteins accumulating in M. oryzae-infected rice Transcriptional changes occurring in rice leaves in leaves (Sun et al. 2004). A Metallothionein-like protein response to treatment with high Fe were examined. For (OsMT3a), and a glycolipid transfer protein involved in this, RNA-Seq analysis was carried out in leaves from ceramide transport (OsGLTP) were also up-regulated in High-Fe and Control plants. Differentially expressed High-Fe plants compared with Control plants (Fig. 2B). genes (DEGs) were identified based on significance level Metallothioneins are metal-binding proteins implicated (FDR ≤ 0.05) and l­og2 fold change (FC) with a threshold in metal detoxification and scavenging of ROS (Hassinen of FC ≥  + 0.5 and FC ≤  − 0.5 for up-regulated and down- et al. 2011), while ceramides regulate the Arabidopsis regulated genes, respectively. defense response by promoting ­H2O2 generation (Li et al. When comparing leaf transcriptomes of High-Fe and 2022). Control plants, only 239 genes were found to be regu- Genes related to responses to biotic stress were also lated by treatment with high Fe (48 genes upregulated; regulated by Fe treatment under non-infection condi- 191 genes down-regulated) (Fig. 2A, Additional file 2: tions, including certain Pathogenesis-Related (PR) genes. Table S1a). Genes whose expression is known to be Their expression was up-regulated (OsPR1b) or down- induced by Fe deficiency where found to be repressed regulated (OsCht11 and OsChib3a encoding rice chi- in leaves of High-Fe plants relative to Control plants tinases; Jasmonate-Inducible PR10, or JIOsPR10; and (Fig. 2B, Additional file 2: Table S1b) which is con- Germin-like Protein, or OsGLP genes) in High-Fe plants sistent with an increase in Fe content in these plants (Fig. 2B). (see Fig. 1C). Down-regulated genes in High-Fe plants Results obtained by RNA-Seq analysis were validated included: Iron deficiency-inducible peptide-IRON MAN by RT-qPCR for genes showing either down-regulation (OsIMA1, OsIMA2), Iron-related transcription factors (OsIRO2, OsIRO3, OsNRAMP1, OsChib3a, OsJIPR10) or (OsIRO2, OsIRO3) and Iron-binding Haemerythrin RING up-regulation (OsFER2, OsMT3a, OsPR1b) by treatment ubiquitin ligases (OsHRZ1, OsHRZ2). Other genes down- with high Fe (Fig. 2C). regulated in leaves of High-Fe rice plants were those Together, these findings indicated that treatment encoding plasma membrane iron transporters, such with Fe provokes transcriptional alterations in genes as Iron-deficiency-regulated oligopeptide transporter 7 involved in iron homeostasis and defense responses in (OsOPT7), Natural Resistance-Associated Macrophage the absence of pathogen infection. Down regulation of Protein 1 (OsNRAMP1), Vacuolar Membrane Ferric- defense-related genes by treatment with high Fe (e.g. PR Chelate Reductase (OsFRO1), and Vacuolar mugineic genes) is in apparent contradiction with the phenotype acid transporter (OsVMT) (Fig. 2B, Additional file 2: of blast resistance that is observed in High-Fe plants. Table S1a). OsFER2 (FERRITIN 2) was up-regulated in We then reasoned that blast resistance in rice plants that leaves of High-Fe plants (Fig. 2B). As previously men- have been treated with high Fe might rely more on the tioned, Ferritins bind Fe (mainly in the form of ­Fe3+), stronger induction of defense responses upon patho- thus, contributing to protection of plants against Fe- gen infection rather than on constitutive expression of mediated oxidative stress. defense genes (see below). Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 5 of 23 Fig. 2 Differentially expressed genes in leaves of High-Fe plants relative to Control plants. A Number of genes that are up-regulated and down-regulated genes in leaves of rice plants treated with high Fe for 5 days. B Heatmap showing the expression of genes involved in Fe homeostasis, ROS metabolism and biotic stress identified as differentially expressed by RNA-Seq analysis. The expression level (row scaled FPKM, fragments per kilobase per million reads) is represented from pale yellow (less expressed) to brown (more expressed). Fold change (FC) values are represented on the right, in purple (up-regulated genes, ­Log2 FC ≥  + 0.5) and green (down-regulated genes, ­Log2 FC ≥  − 0.5). Data represented correspond to the mean of two biological replicates, each biological replicate consisting in a pool of 5 leaves from individual plants. The full gene name and ID of DEGs are indicated in Additional file 2: Table S1. C RT-qPCR analysis of DEGs in High-Fe plants relative to Control plants. Data are mean ± SEM (n = 4 biological replicates, including the two replicates used for RNA-Seq analysis). Asterisks indicate statistically significant differences (t-test, *, P < 0.05; **, P < 0.01). Gene-specific primers are listed in Additional file 9: Table S8 Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 6 of 23 Transcriptional Reprogramming in Leaves of High‑Fe Rice in each Fe condition (­log2 FC ≥ 0.5 or ≤  − 0.5 for up- Plants During M. oryzae Infection regulated and down-regulated genes, respectively; To explore the possibility that treatment of rice plants P ≤ 0.05). Pair-wise comparisons of DEGs (M. oryzae- with high Fe has an effect on the induction of defense inoculated vs mock-inoculated) revealed 5996 genes genes during pathogen infection, we compared the leaf regulated by M. oryzae infection in Control Fe plants transcriptomes of Control and High-Fe plants, inocu- (2879 up-regulated; 3117 down-regulated) (Fig. 3A, lated with M. oryzae spores or mock-inoculated (at Additional file 3: Table S2a). In High-Fe plants, 6470 48 h post-inoculation, hpi). The same criteria described genes were regulated by M. oryzae infection (3293 up- above were used to identify pathogen-regulated genes regulated; 3177 down-regulated) (Fig. 3A, Additional Fig. 3 GO enrichment of genes up-regulated by M. oryzae infection in Control and High-Fe plants (48 hpi). The categories of Biological Processes (A) and Molecular Function (B). GO terms were visualized using REVIGO after reducing redundancy and clustering of similar GO terms in the O. sativa database. GO terms are represented by circles and are clustered according to semantic similarities (more general terms are represented by larger size circles, and adjoining circles are most closely related). Circle size is proportional to the frequency of the GO term, whereas color indicates the enrichment derived from the AgriGO analysis (red higher, blue lower). Full data sets of DEGs and lists of GO terms are presented in Additional file 3: Table S2 and Additional file 4: Table S3, respectively) Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 7 of 23 file 3: Table S2b). Most of the genes regulated by patho- examined co-expression patterns. Genes up-regulated gen infection, either up-regulated or down-regulated by M. oryzae infection classified into 6 hierarchical clus- genes, were shared between Control and High-Fe ters (Fig. 4). GO terms in co-expressing genes identified plants (Additional file 1: Fig. S2). terms related to plant defense in clusters I, II, IV, and V, Gene ontology (GO) enrichment analysis was used to whereas genes playing a role in protein phosphoryla- identify GO terms enriched in the set of pathogen-reg- tion grouped in cluster VI (Fig. 4). In particular, Cluster ulated genes in High-Fe or Control plants. Redundant V contained the GO terms of “Diterpene phytoalexin GO terms were removed using the REVIGO tool (Supek biosynthetic process” and “Response to fungus/bacte- et al. 2011; http://​revigo.​irb.​hr). In Biological Processes, rium/oomycetes”. This analysis also revealed that High- the most abundant subcategories were “Diterpene Phyto- Fe plants exhibited stronger induction in the expression alexin Metabolism” followed by “Secondary Metabolism” of pathogen-responsive genes compared with Control and “Response to Biotic Stimulus/Defense Responses”, plants (Fig. 4). which were enriched in both High-Fe and Control-Fe plants (Fig. 3A, Additional file 4: Table S3a, b). The GO Enhanced Defense Responses of Iron‑Treated Rice Plants category of “Phosphorus Metabolism” was also enriched Against the Blast Fungus in the set of genes that are up-regulated by pathogen The induction of PR genes is regarded as a ubiquitous infection (Fig. 3A). Indeed, an enrichment of GO terms response to pathogen infection in plants. Presently, PR for “ATP binding” and “Protein kinase activity” annota- proteins are grouped into 17 families based on their pro- tions were found in the Molecular Function category, tein sequence similarities and function (Ali et al. 2018). both in High-Fe and Control plants (Fig. 3B). Genes As expected, our RNASeq analysis revealed that M. ory- involved in “Iron binding” and “Signaling receptor activ- zae infection activated the expression of an important ity” were also over-represented in the set of up-regulated number of PR genes (Fig. 5A, Additional file 5: Table S4). genes by pathogen infection in both High-Fe and Con- They included genes belonging to different PR families, trol plants, while genes with “Chitinase activity” were such as PR1, PR2 (β-1,3 glucanases), PR3, PR4 and PR8 over-represented in High-Fe plants, but not in Control (chitinases), PR5 (thaumatin-like proteins, TLPs), PR6 plants (Fig. 3B). In this respect, transgenic expression of (proteinase inhibitors), PR9 (peroxidases), PR10 (ribonu- chitinase genes has long been recognized to confer resist- clease-like proteins), PR14 (lipid transfer proteins, LTPs) ance to fungal diseases in different plant species, includ- and PR15 (germin-like proteins) genes. Of them, OsPR1 ing rice blast disease (Ali et al. 2018). As for genes that and OsPBZ1 (PR10 family) are routinely used as marker are down-regulated by M. oryzae infection, in Biologi- genes for the induction of the rice response to M. oryzae cal Processes, they categorized in “Photosynthesis” and infection (Midoh and Iwata 1996; Agrawal et al. 2001). “Pigment biosynthetic processes” in both High-Fe and Although PR gene expression was induced by M. ory- Control plants (Additional file 1: Fig. S3A). In Molecu- zae infection in both High-Fe and Control plants, their lar Function, GO terms for “Iron-sulfur cluster binding” expression was induced to a greater extent in High-Fe and “Zinc ion binding” were represented in the group plants compared with Control plants (Fig. 5A, Additional of down-regulated genes in Control and High-Fe plants file 5: Table S4). Here it is worth mentioning that trans- (Additional file 1: Fig. S3B). genic expression of distinct PR genes, such as chitinase or Taken together, GO enrichment analysis of genes regu- PR10 genes, has proven to enhance resistance to M. ory- lated by M. oryzae infection indicated similar GO terms zae infection (Wu et al. 2016; Ali et al. 2018). in High-Fe and Control rice plants, indicating that similar RNA-Seq analysis also revealed regulation of an impor- biological processes and mechanisms appear to be acti- tant number of peroxidase (Prx) genes during M. oryzae vated by pathogen infection in both conditions. Then, infection (PR9 family), these genes exhibiting stronger quantitative rather than qualitative differences in defense induction in High-Fe plants compared with Control gene expression would explain the phenotype of blast plants (Fig. 5A, Additional file 5: Table S4). It is tempt- resistance that is observed in High-Fe plants. Indeed, ing to hypothesize that M. oryzae-regulated peroxidases hierarchical clustering of DEGs expression levels in Con- might contribute to control the spatial and temporal trol and High-Fe plants (infected and mock) revealed that accumulation of ­H2O2 in rice plants. the transcriptional response to M. oryzae infection in Among the genes regulated by both Fe treatment and High-Fe plants was not qualitatively different from that M. oryzae infection there were several lipoxygenases in Control plants, but quantitatively different (Additional (Fig. 5B). Lipoxygenases (LOXs) are crucial for lipid per- file 1: Fig. S4). oxidation processes during plant defense responses to To further investigate differences in the response of pathogen infection. For instance, LOX are involved in the Control and High-Fe plants to M. oryzae infection, we conversion of polyunsaturated fatty acids into oxylipins Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 8 of 23 Fig. 4 Hierarchical clustering of genes that are regulated by M. oryzae infection in Control and High-Fe plants. The expression level is depicted as row scaled FPKM (fragments per kilobase per million reads) and is represented from pale yellow (less expressed) to brown (more expressed). GO enrichment analyses of DEGs in each individual cluster revealed over-represented GO terms (shown on the right) Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 9 of 23 Fig. 5 Expression of genes involved in defense responses in leaves of Control and High-Fe plants. Heat maps (A and B) show the expression pattern of DEGs identified by RNA-Seq analysis (Mock and infection conditions, 48 hpi) as row scaled FPKM (fragments per kilobase per million reads), from pale yellow (less expressed) to brown (more expressed). A Expression of Pathogenesis-Related (PR) genes. B Expression of lipoxygenase and phenylpropanoid genes. Gene names, locus IDs, FPKMs and fold change values are indicated in Additional file 5: Table S4. (C) Transcript levels were determined by RT-qPCR analysis (−, mock-inoculated; +, M. oryzae-inoculated). The expression values were normalized to the rice Ubiquitin1 gene. Four independent biological replicates were examined. Data are mean ± SEM. Asterisks indicate statistical significance calculated by two-way ANOVA (*, P < 0.05; **, P < 0.01, *** P < 0.001). Gene-specific primers are listed in Additional file 9: Table S8 and jasmonates (Viswanath et al. 2020). Lipoxygenases Defensive functions of phenylpropanoid compounds have been also implicated in the initiation and execution have long been recognized, these metabolites confer- of ferroptotic cell death in plants (Conrad et al. 2018). ring broad-spectrum disease resistance in different plant Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 10 of 23 species (Yadav et al. 2020). To note, genes involved in the sakuranetin (Okada et al. 2007; Hasegawa et al. 2010, general phenylpropanoid pathway were found to be reg- 2014). ulated by both iron treatment and M. oryzae infection, Diterpenoid phytoalexins are synthesized from gera- most of these genes showing a stronger response in High- nylgeranyl diphosphate (GGDP), the end product of the Fe plants than in Control plants (Fig. 5B, Additional file 5: methylerythritol phosphate (MEP) pathway (Fig. 6A). Table S4). A schematic representation of these genes in Rice produces a variety of diterpene phytoalexins, namely the rice phenylpropanoid pathway is presented in Addi- momilactones (A, B), phytocassanes (A to E), and oryz- tional file 1: Fig. S5. They included genes that participate alexins (A to F, S). Our RNA-Seq analysis revealed that in the general phenylpropanoid pathway (phenylala- the expression of genes implicated in the MEP and diter- nine ammonia lyase, PAL; cinnamic acid 4-hydroxylase, pene phytoalexin biosynthesis pathways was induced to C4H; 4-coumarate:CoA ligase, 4CL), as well as genes in a greater extend in High-Fe plants than in Control plants branches leading to flavonols and anthocyanins (chalcone (Fig. 6B, Additional file 6: Table S5). They included genes synthase, CHS; chalcone isomerase, CHI). The phenyl- in the biosynthetic pathway leading to the production of propanoid pathway also leads to the production of nar- oryzalexins (A–F), phytocassanes (A–E) and momilac- ingenin, the precursor of flavonols. Naringenin is also tones (A and B) whose expression was strongly induced the precursor molecule for sakuranetin, the only flavo- by pathogen infection in High-Fe plants compared with noid phytoalexin so far identified in rice (Hasegawa et al. Control plants (Fig. 6B, Additional file 6: Table S5). 2014). Sakuranetin has been shown to accumulate in rice Results obtained by RNA-Seq analysis were confirmed leaves during M. oryzae infection and to exhibit anti- by RT-qPCR analysis for selected genes (Additional file fungal activity against M. oryzae (Hasegawa et al. 2014). 1: Fig. S6). As previously mentioned, RNA-Seq analysis We noticed that OsNOMT1 (Naringenin 7-O-methyl- also revealed stronger induction of OsNOMT1 involved transferase), responsible of methylation of naringenin in sakuranetin (phenolic phytoalexin) biosynthesis in to sakuranetin (Shimizu et al. 2012) reached a higher the response of High-Fe plants to M. oryzae infection expression in High-Fe compared with Control plants compared with Control plants (Fig. 6, Additional file 6: (Fig. 5B). RT-qPCR confirmed higher expression levels of Table S5). defense-related genes in response to pathogen infection Next, we investigated whether superinduction of phy- in High-Fe plants compared with infected Control plants toalexin biosynthesis genes in High-Fe plants during (OsPR1a, OsPR1b, OsPBZ1, PR10, JIOsPR10, ROsPR10, infection has an effect on phytoalexin content. As shown OsLOX, OsLOX6, OsLOX8) (Fig. 5C). in Fig. 6C, M. oryzae infection is accompanied by the Altogether, these results indicated that treatment with accumulation of diterpene phytoalexins (Momilactone high Fe is associated with a stronger induction of defense A and B, Phytocassane B, C and E) in leaves of Control gene expression during M. oryzae infection, which corre- and High-Fe plants. Most importantly, their accumu- lates well with resistance to M. oryzae in High-Fe plants. lation was found to be significantly higher in High-Fe plants than in Control plants (Fig. 6C). Equally, the phe- Treatment with High Fe Promotes the Accumulation nolic phytoalexin Sakuranetin accumulated at a higher of Phytoalexins in Rice Leaves level during infection of High-Fe plants compared with Phytoalexins are low-molecular-weight antimicrobial Control plants (Fig. 6C). These results are in agreement compounds that are produced by plants in response to with those obtained on the expression of phytoalexin bio- pathogen infection (Ahuja et al. 2012). Major phytoalex- synthetic genes indicating that treatment with high-Fe is ins in rice accumulating during M. oryzae infection are associated to a higher accumulation of phytoalexins in diterpene phytoalexins and the flavonoid phytoalexin rice plants, both diterpene and phenolic phytoalexins. (See figure on next page.) Fig. 6 Accumulation of phytoalexins in leaves of rice plants that have been grown under high Fe supply. A Methylerythritol phosphate (MEP) and diterpenoid phytoalexin biosynthesis pathways in rice. Genes whose expression is up-regulated by M. oryzae infection in control and high-Fe plants are indicated in purple color. The full name and details on the expression of these genes can be found in Additional file 6: Table S5. B Heat map showing expression levels (row scaled FPKM, fragments per kilobase per million reads) in leaves of Control and High-Fe plants, mock-inoculated and M. oryzae-inoculated plants (48hpi). Gene expression is represented from pale yellow (less expressed) to brown (more expressed). C Accumulation of phytoalexins, diterpenoid phytoalexins and sakuranetin, in leaves of Control and High-Fe plants at 48 hpi and 72 hpi with M. oryzae spores: momilactones (A and B), phytocassanes (B, C, and E), and the flavonoid phytoalexin sakuranetin (−, mock-inoculated; +, M. oryzae-inoculated). Data are mean ± SEM of three biological replicates each with five plants. Asterisks indicate statistically significant differences calculated by two-way ANOVA (*** P ≤ 0.001). G3P, glyceraldehyde-3-phosphate; DXP, 1-deoxy-D-xylulose 5-phosphate; MEP, 2- C-methyl-D-erythritol 4-phosphate; CDP-ME, 4-(cytidine 5′-diphospho)-2-C-methyl-D-erythritol; CDP-ME2P, 2-phospho-4-(cytidine 5′- diphospho)-2-C-methyl-D-erythritol; MECDP, 2-C-methyl-D-erythritol 2,4-cyclodiphosphate; HMBDP, 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate; IPP, isopentenyl diphosphate; DMAPP, dimethylallyl diphosphate; GGDP, geranylgeranyl diphosphate; and CDP, copalyl diphosphate Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 11 of 23 Fig. 6 (See legend on previous page.) Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 12 of 23 Phytoalexin content was also measured in rice plants Regarding vacuolar iron transporters, we found that grown under Fe limiting conditions. For this, the rice the Vacuolar Iron Transporter 1 (OsVIT1) involved in Fe plants were grown for 16 days under control Fe condi- sequestration into vacuoles, was repressed during infec- tions followed by 5 days under low Fe (Low-Fe) or Con- tion in both Control and High-Fe plants. As for the Vacu- trol Fe condition (Ctrl-Fe), and then inoculated with M. olar Mugineic acid Transporter (OsVMT) responsible for oryzae spores (Additional file 1: Fig. S7A, left panel). As sequestration of DMA into the vacuoles, its expression expected, treatment with high Fe results in a decrease in was down-regulated by infection only in Control plants total Fe content in rice tissues (Additional file 1: Fig. S7A, (Fig. 7A). To note, M. oryzae infection strongly induced right panels) Under these experimental conditions, there FERRITIN2 (FER2) in High-Fe plants (Fig. 7A). As previ- were no statistically significant differences in phytoalexin ously mentioned, Ferritin is considered to be the major content (momilactone A, momilactone B, phytocassane Fe-storage protein in plants. C, phytocassane B, phytocassane E sakuranetin) in M. Other Fe-related genes that are also regulated by patho- oryzae-infected Low Fe plants compared with M. oryzae- gen infection were those involved in Fe transport, such as infected Control plants (Additional file 1: Fig. S7B). members of the NRAMP (Natural Resistance-Associated Knowing that rice phytoalexins have been described to Macrophage Protein) family (OsNRAMP1, OsNRAMP3, exhibit antifungal activity against M. oryzae (Umemura OsNRAMP5, OsNRAMP6, OsNRAMP7), and the ZIP et al. 2003; Hasegawa et al. 2010, 2014), it is tempting to (Zn-regulated transporter, IRT-like protein) family hypothesize that blast resistance in High-Fe plants might (OsZIP2, OsZIP4, OsZIP8) (Fig. 7A). Additionally, up to result, as least in part, from the superactivation of phyto- 7 genes encoding Yellow Stripe-Like (YSL) transporters alexin biosynthesis genes, and subsequent accumulation functioning in the import of ­Fe3+-DMA complexes into of phytoalexins in these plants. the plant cell were found to be regulated during M. ory- zae infection in rice leaves. They were: OsYSL5, OsYSL12, M. oryzae Infection Modulates Iron Accumulation in Rice OsYSL13, OsYSL16 (up-regulated), OsYSL6, OsYSL14 Leaves (down-regulated), and OsYSL9 (up-regulated in Control Comparative transcriptome analysis of Control and plants but down-regulated in High-Fe plants) (Fig. 7A). High-Fe plants revealed that, in addition to defense- Changes in the expression of selected iron homeosta- related genes, M. oryzae infection has an impact on the sis-related genes in response to M. oryzae infection in expression of genes that function in iron homeostasis in High-Fe and Control plants were confirmed by RT-qPCR leaves. Among them, there were genes that are down- analysis (Additional file 1: Fig. S8). regulated by Fe treatment (e.g. in the absence of patho- Altogether, this study revealed that M. oryzae infection gen infection) that are also down-regulated by pathogen triggers alterations in the expression of genes with differ- infection in Control plants. This is the case of the tran- ent functions in maintenance of iron homeostasis, includ- scription factors OsIRO2, OsIRO3, OsIMA1, OsIMA2, ing transcriptional regulators of Fe-responsive genes, Fe OsHRZ1, OsHRZ2, the OsOPT7 Oligopeptide transporter, transporters, and Fe chelators. Presumably, these genes and Ferric Reductase Oxidase 2 (OsFRO2) (Fig. 7A, Addi- would be of importance in maintaining proper spatio- tional file 7: Table S6). temporal accumulation of Fe in the host tissue during As chelators of Fe, rice plants produce phytosidero- pathogen infection. phores of the mugineic acid (MA) family, DMA (2-deoxy- Having established that pathogen infection regulates mugineic acid) being the sole MA produced in rice. DMA the expression of genes involved in Fe homeostasis, we is synthesized from S-adenosyl-methionine (SAM). To determined whether M. oryzae infection provokes alter- note, OsSAMS1, OsSAMS2 (encoding SAM synthases 1 ations in Fe content in rice leaves. Total Fe content was and 2) responsible for the conversion of L-methionine to determined by inductively coupled plasma-mass spec- SAM were found to be strongly up-regulated by patho- trometry (ICP-MS) in leaves of Control and High-Fe gen infection in High-Fe plants (Fig. 7A). A priori, the plants, that have been inoculated with M. oryzae spores up-regulation of these genes would ensure a sufficient or mock-inoculated. As expected, treatment with high supply of SAM for DMA conversion. Other genes in the Fe results in an increase in total Fe content (Fig. 7B, grey biosynthesis of DMA that are regulated by M. oryzae bars). Upon pathogen challenge, there was a significant infection were OsNAAT1 (nicotianamine aminotrans- increase in Fe content in leaves of Control plants, but its ferase 1) and OsDMSA1 (deoxymugineic acid synthase level was not significantly altered in leaves of High-Fe 1). Phytosiderophore efflux transporter genes (OsTOM2, plants (a tendency to a lower content in response to M. OsTOM3) were also up-regulated by infection, these oryzae infection could be, however, observed in High-Fe genes being implicated in the transport of DMA out of plants) (Fig. 7B). the cell (Fig. 7A). Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 13 of 23 Fig. 7 Effect of pathogen infection on Fe homeostasis-related genes and pathogen-induced alterations on Fe content in leaves. A Heat map showing the expression level of iron-related genes (row scaled FPKMs) in leaves of Control and High-Fe plants (mock-inoculated and M. oryzae-inoculated) at 48 hpi. Gene expression is represented from pale yellow (less expressed) to brown (more expressed). The full name and details on the expression of these genes are indicated in Additional file 7: Table S6. B Total Fe content in the youngest developed leaf (third leaf ) (−, mock-inoculated; +, M. oryzae-inoculated). Five independent biological replicates (three technical replicate each) were analyzed. C ­Fe2+ and ­Fe3+ content was determined by the colorimetric Ferrozine method in the youngest developed leaf of Control and High-Fe plants. D Fe content in the apoplast of leaves (youngest developed leaves) determined by ICP-MS. Five independent biological replicates (three technical replicate each) were analyzed. Data are mean ± SEM. Asterisks indicate statistically significant differences calculated by two-way ANOVA *, P ≤ 0.05). DW, dry weight; FW, fresh weight Next, it was of interest to investigate whether M. ory- infection, ­Fe3+ content increased in Control plants, while zae infection has an effect on the accumulation of each its level decreased in High-Fe plants (infected vs non- Fe ion in rice leaves, ­Fe3+ and ­Fe2+. Accordingly, the infected plants) (Fig. 7C, left panel, pink bars), a pattern ferrozine assay was used to assess ­Fe3+ and ­Fe2+ accu- that resembled that of total Fe content (see Fig. 7B). From mulation in leaves of Control and High-Fe plants, mock- these results, it appears that F­ e3+ content in M. oryzae- inoculated and M. oryzae-inoculated. Treatment with infected High-Fe plants reached a level similar to that in high Fe increased both F ­ e3+ and ­Fe2+ content in mock- non-infected Control plants. As for F ­ e2+, its level was not inoculated plants (Fig. 7C, grey bars). Upon pathogen Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 14 of 23 significantly altered by pathogen infection either in Con- trol or High-Fe plants (Fig. 7C, right panel). To gain more information about effect of pathogen infection on Fe distribution in rice leaves, we measured apoplastic Fe levels. Malate dehydrogenase (MDH) activ- ity, an indicator of cytosolic contamination, was not detected in apoplast samples, thus, excluding cytosolic contamination in apoplastic samples (Additional file 1: Fig. S9). This study revealed that treatment with high-Fe was accompanied by a significant increase in apoplastic Fe concentrations (Fig. 7D, grey bars). Upon M. oryzae infection, apoplastic Fe decreased in both Control and High-Fe plants (Fig. 7D, pink bars). In this way, M. oryzae infection decreased Fe accumulation in the leaf apoplast of High-Fe plants to a level similar to that in non-infected Control plants. Collectively, transcriptome analysis in combination with Fe determinations revealed that M. oryzae infection provokes alterations in the expression of genes involved in Fe homeostasis leading to alterations in Fe con- tent, total Fe and apoplastic Fe, in rice leaves. Different responses to pathogen infection are, however, observed depending on the Fe supply condition. In High-Fe plants, there is a reduction on total Fe, F ­ e3+ and apoplastic Fe during pathogen infection, whereas pathogen infection increases total and F­ e3+ content, but not apoplastic Fe, in control plants. Clearly, a tight control of Fe homeostatic Fig. 8 Accumulation of ROS and Fe in leaves of Control and High-Fe mechanisms occurs in rice plants at the intracellular and rice plants. A ROS ­(H2O2) was detected in leaves using the fluorescent extracellular levels during infection by M. oryzae, which probe ­H2DCFDA. Leaves (youngest developed leaves) from is also dependent on the Fe status of the rice plant. mock-inoculated (mock) and M. oryzae-inoculated plants (48 hpi) were examined. Bars correspond to 2 mm. Right panel, quantification of ROS fluorescence using Image J software (three independent ROS and Fe Accumulation in High‑Fe Plants During experiments, 100 fields, each). Asterisks indicate statistically significant Pathogen Infection differences calculated by two-way ANOVA (*P ≤ 0.05). B Fe detection The production of ROS is a generalized plant defense in leaves of Control-Fe (upper panels) and High-Fe (lower panels) response against pathogen attack (Torres 2010). Among plants that have been mock-inoculated or infected with M. oryzae (48 hpi). The youngest developed leaf (third leaf ) was subjected to DAB ROS, ­H2O2 is considered to be an important signaling staining (brown) followed by Perls staining (blue). Bars correspond to molecule in regulating plant immunity. It is also known 50 μm. St, stomata, Ap: appressorium that pathogen infection triggers ROS accumulation and localized cell death around the site of infection, thus, limiting the spread of the pathogen. Additionally, Fe- mediated ROS production (Fenton reaction) is a central infection, ROS was not detected in leaves of either Con- executioner of ferroptosis, an iron-dependent form of cell trol or High-Fe plants (Fig. 8A, mock). Upon pathogen death which is also associated with increased lipid per- challenge, however, discrete regions accumulating ROS oxidation (Distéfano et al. 2017; Conrad et al. 2018). The could be observed in both Control and High-Fe plants, observation that M. oryzae infection was accompanied but the H ­ 2DCFDA-fluorescent signals were more intense by alterations in Fe accumulation in rice leaves prompted and abundant in High-Fe plants relative to Control plants us to investigate Fe and ROS accumulation in leaves of (Fig. 8A, infected). Quantification of H­ 2DCFDA fluo- High-Fe plants during M. oryzae infection at the histo- rescence using ImageJ software confirmed higher path- chemical level. ogen-induced accumulation of ROS in High-Fe plants Histochemical detection of ROS was carried out using compared with Control plants (Fig. 8A, right panel). the fluorescent probe H ­ 2DCFDA (2′, 7′ dichlorofluores- Regions accumulating ROS, most probably, correspond cein diacetate) which mainly detects H ­ 2O2 accumula- to the M. oryzae infection sites. tion (Fichman et al. 2019). In the absence of M. oryzae Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 15 of 23 To further investigate Fe accumulation during infection of High-Fe plants we used an histochemical procedure based on 3, 3′-diaminobenzidine (DAB) staining followed by Perls staining. In mock-inoculated leaves from Con- trol and High Fe plants, ­Fe3+ ions were detected in sto- mata (Fig. 8B, blue color). Upon M. oryzae infection of Control-Fe plants, ROS was found to accumulate at the invaded leaf cells (Fig. 8B, brown color; Ctrl. Fe infected). Interestingly, in M. oryzae-infected High Fe plants, ROS (brown) and F ­ e3+ ions (blue) accumulated at the infec- tion sites with the surrounding cells also accumulating ­Fe3+ ions (Fig. 8B, High Fe infected). Thus, transcriptome analysis in combination with his- tochemical detection of iron indicated that pathogen infection causes a redistribution of Fe in the rice leaf. Fe preferentially accumulates around the infection sites, most probably, as a consequence of alterations in the expression of genes involved in iron homeostasis-related genes during pathogen infection. The higher content of Fe in leaves of High-Fe plants would contribute to a higher accumulation of Fe (hence, ROS) at the infection sites in High-Fe plants. Alterations in iron distribution have been previously described in Arabidopsis tissues infected by Dickeya dadantii, and in maize plants infected by Cur- vularia lunata (Aznar et al. 2014; Fu et al., 2022). How- Fig. 9 Fe detection in leaves of Low-Fe, Control-Fe, and High-Fe ever, additional investigation needs to be carried out to rice plants that have been inoculated with M. oryzae spores (48 hpi). determine the precise spatio-temporal accumulation of Leaves were treated with the ferroptosis inhibitor Ferrostatin (Fer-1) Fe during M. oryzae infection at both tissue and subcel- (+ Fer-1) or not (− Fer-1). Rice leaves (youngest developed leaves) lular levels. were subjected to Perls staining for detection of Fe (blue color). The Finally, trypan blue staining revealed a pattern of cell images shown are representative of two independent experiments. Bars correspond to 100 µm death in leaves of M. oryzae-infected High-Fe plants that was absent in leaves of M. oryzae-infected Control plants (Additional file 1: Fig. S10). A closer examination of these regions revealed dead cells clustered in the vicinity of rice plants. In other studies, the use of ferroptosis induc- fungal penetration sites (appressoria) in the M. oryzae- ers was found to prevent invasive growth of M. oryzae in infected High-Fe plants (Additional file 1: Fig. S10). These leaf sheaths of an otherwise susceptible indica-type rice results confirmed that cell death occurs during infection cultivar (O. sativa cv CO 39) (Shen et al. 2020). of Fe-treated rice plants. As noted above, lipid peroxidation is also a hallmark of ferroptosis which is mainly caused by iron-mediated ROS Treatment with the Ferroptosis Inhibitor Ferrostatin‑1 accumulation (Yang et al. 2016; Conrad et al. 2018). In Suppresses Fe Accumulation During M. oryzae Infection this work, we measured lipid peroxidation as malondial- The ferroptosis inhibitor Ferrostatin-1 (Fer-1) was used to dehyde (MDA) content, MDA being a typical breakdown examine Fe accumulation during infection of rice plants product of peroxidized fatty acids in plant membranes. that have been grown under high, control or low Fe sup- Under no infection conditions, the MDA levels were sim- ply. Perls staining of M. oryzae-infected leaves indicated ilar in Control and High-Fe plants (Fig. 10A, grey bars). that treatment with Fer-1 suppresses Fe accumulation at Under infection conditions, however, MDA accumulated the infection sites, hence ferroptosis (Fig. 9). Brown spots at a higher level in High-Fe plants than in Control plants observed in non-treated High-Fe plants (-Fer-1) points to in response to pathogen infection, a response that was a HR reaction in these plants. Treatment with Fer-1 of M. more evident at 24 hpi (Fig. 10A, pink bars). oryzae-infected leaf sheaths gave similar results. Here, Fe To further investigate lipid peroxidation in High- accumulation was also suppressed upon infection with Fe plants, we used C11-BODIPY581/591 (from now on M. oryzae (Additional file 1: Fig. S11). These findings BODIPY), a fluorescent lipid peroxidation reporter mol- confirm ferroptosis mediating blast resistance in High-Fe ecule that allows monitoring lipid peroxidation in cell Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 16 of 23 near-complete loss of accumulation of oxidized BODIPY (Fig. 10B, right panels). These findings further support lipid peroxidation and ferroptosis during M. oryzae infec- tion in leaves of High-Fe plants. Collectively, results here presented indicated that rice plants that have been exposed to treatment with high Fe develop a ferroptotic cell death during infection with a virulent isolate of M. oryzae, a response that also occurs during infection of rice leaf sheaths with an incompatible M. oryzae isolate (Dangol et al. 2019). Discussion In this study, we provide evidence that short-term expo- sure to iron enhances resistance to infection by the blast fungus M. oryzae in rice plants. Pathogen resistance in Fe-treated rice plants is associated with stronger induc- tion of defense responses during pathogen infection, including higher production of ROS and superinduction of PR genes. Treatment with high Fe also provokes super- activation of genes for the biosynthesis of phytoalexins during pathogen infection, both diterpene phytoalexins and sakuranetin. This, in turn, results in higher accu- mulation of phytoalexins in M. oryzae-infected leaves of High-Fe plants. Knowing that rice phytoalexins exhibit antifungal activity against M. oryzae (Okada et al. 2007; Hasegawa et al. 2014), a higher accumulation of phyto- alexins would explain, at least in part, the phenotype of blast resistance in High-Fe plants. In previous stud- ies, it was described that growing rice plants in low Fe increases blast susceptibility (Peris-Peris et al. 2017). The observed pathogen-induced transcriptional alterations in High-Fe plants and Control plants were not qualitatively different each other, but the magnitude of change in gene expression was greater in High-Fe plants compared with that in Control plants. Overall, these findings support Fig. 10 Lipid peroxidation in leaves of Low-Fe, Control-Fe, and priming for enhanced defense in Fe-treated rice plants. High-Fe rice plants upon infection with M. oryzae. A MDA content Defense priming has been shown to occur in plants in leaves (youngest developed leaves) of mock-inoculated and M. upon treatment with natural or synthetic compounds oryzae-inoculated plants (−, mock-inoculated; +, inoculated with M. oryzae spores). B Confocal microscopy was performed on M. (Martinez-Medina et al. 2016). In a more general sense, oryzae-infected leaves (48 hpi) treated, or not, with the ferroptosis these findings indicated that Fe can regulate the expres- inhibitor Fer-1, and stained with C11-BODIPY581/591 to monitor sion of a range of genes through which rice immunity is oxidized lipid peroxides. Lower panels show the corresponding controlled. bright field images. Representative results from two independent Histochemical staining revealed that, under non-infec- experiments are shown. Ap, Appressorium. Bars correspond to 50 µm tion conditions, Fe preferentially accumulates at stomata whereas during pathogen infection, Fe accumulates in cells that are in the proximity to the infection sites. Most membranes associated to pro-oxidant activity during fer- probably, a redistribution of Fe occurs in the leaf tissue roptosis (upon oxidation, BODIPY shows green fluores- during pathogen infection. Knowing the capacity of Fe cence). As shown in Fig. 10B (left panels), leaves of High to produce ROS and the general requirement of ROS Fe plants showed the characteristic green fluorescence of production for the induction of defense responses, the oxidized BODIPY at the infected regions, which was not accumulation of Fe, and subsequent ROS production observed in M. oryzae-infected leaves of either Control at the infection sites might well trigger the activation of plants or Low-Fe plants. Treatment with Fer-1 showed a immune responses in High-Fe plants (Distéfano et al. Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 17 of 23 2021). Trypan blue staining confirmed cell death in cells NRAMP genes. Together, these findings support that located in the vicinity of the sites of fungal penetration plants have evolved mechanisms to sustain Fe homeosta- in the host tissue. The phenotype of cell death induced sis and immune responses in concerted action. The iron- by treatment with high Fe closely resembles the hyper- regulated and pathogen-regulated genes identified in this sensitive response induced by infection with an avirulent study are good candidates to be components of regula- pathogen which would explain the phenotype of blast tory connections between Fe signaling and immune sign- resistance that is observed in High-Fe plants. Also, ROS aling in resistance to infection by the blast fungus in rice. accumulation at the sites of infection can be potentially We show that whereas M. oryzae infection causes toxic to the pathogen while avoiding Fe toxicity caused by an increase in Fe content (attributable to an increase in a generalized accumulation of Fe. ­Fe3+) in Control Fe, non-treated rice plants, the level of Several lines of evidence support that ferroptosis con- apoplastic Fe is reduced in Control Fe plants during M. tributes to blast resistance in a susceptible rice cultivar oryzae infection. On the other hand, in High-Fe rice that has been subjected to short-term treatment with plants, total Fe (and F­ e3+) as well as apoplastic Fe con- high Fe. As previously mentioned, the accumulation of tent decreased in response to M. oryzae infection, thus, Fe-dependent ROS is a hallmark of ferroptosis, a phe- reaching levels similar to those in non-infected Control nomenon that is observed in M. oryzae-infected High- Fe plants. While ­Fe3+ levels are regulated by M. oryzae Fe plants. In addition to Fe-dependent ROS production, infection, ­Fe2+ accumulation appears not to be affected lipid peroxidation is also common feature for ferropto- by pathogen infection, either in Control- or High-Fe sis. In this respect, several lipoxygenases were found to plants. A redistribution of intracellular and apoplastic Fe, be strongly induced in High-Fe plants, these plants also as well as changes in the spatial and temporal accumula- accumulating higher levels of MDA than Control plants tion of Fe, might well be the consequence of pathogen- during M. oryzae infection. Treatment with the canoni- induced alterations in the expression of genes involved cal ferroptosis inhibitor Fer-1 prevented Fe accumulation in iron homeostasis. Thus, a regulatory network of iron and lipid peroxidation in High-Fe plants which further chelation, transport and trafficking operating in M. ory- supports ferroptosis during infection of High-Fe rice zae-infected rice leaves would facilitate recruitment of Fe plants. towards the infection sites (as revealed by histochemical In plants, ferroptosis was first reported in Arabidopsis analysis in leaves of M. oryzae-infected plants). Whether thaliana under heat stress conditions (Distéfano et al. mobilization of Fe from non-infected leaf regions 2017). Later on, Dangol et al. (2019) reported ferroptosis towards the infection sites occurs during infection is an occurring during infection of rice plants with an aviru- interesting issue that deserves further investigation. lent M. oryzae strain (incompatible interaction), but not Several studies gathered evidence of connections upon infection with a virulent strain of M. oryzae. In this between the Fe status of plants and disease resistance way, ferroptosis appears to occur not only in incompat- (Kieu et al. 2012; Verbon et al. 2017; Herlihy et al. 2020). ible rice/M. oryzae interactions (Dangol et al. 2019) but For instance, Fe deficiency was found to confer resistance also in susceptible rice plants upon Fe treatment (present to Dickeya dadantii and Botrytis cinerea in Arabidopsis work), or by treatment of susceptible plants with ferrop- (Kieu et al. 2012). In maize, an adequate Fe nutritional tosis inducers (Shen et al. 2020). status was found to reduce infection by Colletotrichum In addition to defense-related genes, our results indi- graminicola (Ye et al. 2014). These authors also reported cated that genes that are known to be involved in iron that the beneficial role of Fe was dependent on the homeostasis were also regulated by M. oryzae infection. strength of Fe treatment (Ye et al. 2014). Very recently, They included genes with functions related to the chela- iron redistribution was reported to induce oxidative burst tion/transport of Fe, import/export of Fe between sub- and resistance against the fungal pathogen Curvularia cellular compartments (plasma membrane, tonoplast), lunata in maize (Fu et al. 2022). In wheat plants infected or Fe sequestration (e.g. Ferritin). As ferritin is known to with B. graminis f.sp. tritici, ­Fe3+ accumulated at cell wall bind Fe to avoid cellular Fe toxicity (Aung and Masuda appositions to mediate an oxidative burst, intracellular 2020), it is tempting to hypothesize that ferritin might iron depletion and induction of PR gene expression (Liu sequester intracellular Fe caused by pathogen infection et al. 2007). From the information gained in maize, wheat in High-Fe plants, if in excess. In other studies, trans- and rice, it appears that Fe accumulation, and subsequent genic tobacco plants constitutively expressing alfalfa fer- ROS production, might be a critical response to arrest ritin were more resistant to infection by fungal pathogens pathogen infection in cereal species. (Alternaria alternata, Botrytis cinerea) (Deák et al. 1999). Although treatment with Fe has proven to be effec- Other genes playing a role in iron homeostasis that are tive to protect rice plants from the rice blast fungus, it also regulated by M. oryzae infection were ZIP, YSL, and is anticipated that the iron status of a plant might have Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 18 of 23 a different effect on disease resistance depending on the Treatments with the ferroptosis inhibitor, ferrostatin-1 two partners, host plant and pathogen. Being a foliar (Fer-1) was carried out as previously described (Dangol pathogen, M. oryzae entirely depends on the host tis- et al. 2019). sue to acquire essential nutrients required for pathogen growth, and both partners compete for Fe. In this mul- Blast Resistance Assays tifaceted process, the plant can deploy strategies that The fungus M. oryzae (strain Guy11, courtesy of Ane restrict fungal invasion, either by sequestering Fe away Sesma) was grown in complete media agar (CMA) con- from the pathogen (withholding strategy) or by accu- taining 30 µg/ml chloramphenicol) for 15 days at 28 °C mulating Fe which can be toxic for the pathogen. On under a 16 h/8 h light/dark photoperiod. M. oryzae the other hand, pathogens might secrete iron-chelating spores were collected by adding sterile water to the sur- siderophores to acquire Fe from their host (Herlihy et al. face of the mycelium and adjusted to the appropriate 2020). Further studies are required to elucidate the bio- concentration. logical significance of the observed alterations in apo- Rice plants, control plants and Fe-treated plants (at plastic Fe content in response to M. oryzae infection in 5 days of treatment), were spray-inoculated with M. rice plants. The pathogen lifestyle (necrotroph, biotroph, oryzae spores (5 × ­105 spores/ml; 0.3 ml/plant) using an hemibiotroph) might also hold important implications to aerograph at 2 atm of pressure (see Fig. 1A, left panel, understand how Fe accumulation and distribution (intra- treatment with high-Fe plants; Additional File 1: Fig. S7, cellular and apoplastic Fe) might affect pathogen coloni- treatment with low-Fe) and maintained overnight in the zation in the host tissue. Then, it will be then of interest dark under high humidity. At the time of fungal infec- to investigate whether treatment with Fe confers resist- tion, control and iron-treated rice plants were at the ance to pathogens with different lifestyles in rice. 3–4 leaf stage. After inoculation with fungal spores, the plants were allowed to continue growth under the same Conclusion Fe regime (control, high Fe, or low Fe), and photoperiod Results presented in this study demonstrated that, upon condition. The development of disease symptoms was short-term exposure to Fe, rice plants develop a more followed over time. The percentage of leaf area affected robust defense response during infection with the rice by blast lesions was determined on the youngest devel- blast fungus M. oryzae and confers blast resistance. oped leaf (third leaf ) of the M. oryzae-infected plants Our results suggest that priming of plant defense by Fe using digital imaging software (APS Assess 2.0 program; treatment might find value for improving rice resistance n = 10) (Lamari 2008). Fungal biomass was quantified by against blast. The information gained in this study also quantitative PCR (qPCR) using specific primers for the illustrates the relevance of crosstalk between iron signal- M. oryzae 28S DNA gene (Qi and Yang 2002). ing and immune signaling pathways in controlling blast resistance in rice. A better understanding on the mecha- Phenotypical Analysis and Chlorophyll Content nisms underlying iron-immunity crosstalk will allow the Shoot and root fresh weight (FW) was determined in development of appropriate strategies for protection of three-week-old rice plants treated, or not, with high Fe rice against the blast fungus. (1 mM Fe) at 5 days or 19 days after the onset of treat- ment. At least 10 biological replicates were examined. Materials and Methods Chlorophyll content was determined using the SPAD 502 Plant Growth Conditions Plus Chlorophyll Meter (Spectrum Technologies). SPAD Rice (O. sativa cv. Nipponbare) plants were grown at readings were obtained in the youngest developed leaf of 28 °C with a 14 h/10 h light/dark cycle. For Fe treatment, rice plants grown in different Fe concentrations (n = 10). plants were grown in soil (turface:vermiculite:quartz sand [2:1:3]), and watered (bottom watering) with a Plant Tissue Staining half-strength Hoagland solution for 16 days (2.5 mM H2O2 determination was carried out using ­H2DCFDA ­KNO3, 2.5 mM Ca(NO3)2·4H2O, 1 mM M ­ gSO4·7H2O, (2′,7′-dichlorodihydrofluorescein diacetate) staining fol- 0.5 mM ­NH4NO3, 25 μM ­KH2PO4, 23.15 μM ­H3BO3, lowing the procedure described by Kaur et al. (2016). 4.55 μM ­MnCl2·4H2O, 0.38 μM ­ZnSO4·7H2O, 0.1 μM Briefly, the rice leaves were cut into pieces (3 cm) and ­CuSO4·5H2O, 0.14 μM ­ Na2MoO4·2H2O, 50 μM Fe- incubated in a solution containing 10 μM ­H2DCFDA, EDDHA, pH 5.5). To assess the effect of high Fe supply, vacuum infiltrated during 5 min, and then maintained in the same nutrient solution was used but with a higher Fe darkness at room temperature for 10 min. The samples concentration (1 mM Fe-EDDHA). The same procedure were washed three times with distilled water and stored was used to examine the effect of low Fe treatment but in 20% glycerol and immediately visualized. The fluo- using a nutrient solution containing 0.1 µM Fe-EDDHA. rescence emitted by ­H2DCFDA was observed at 488 nm Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 19 of 23 in an Leica DM6 microscope under GFP fluorescence (156 ­mM−1 ­cm−1), and the results are expressed as mM (488 nm excitation, 509 nm emission). Fluorescence was MDA/g fresh weight (FW). quantified using ImageJ software. C11-BODIPY581/591 (Thermo Fisher Scientific) was For DAB staining, the rice leaves were cut in 3 cm used to visualize lipid peroxidation (10 µM C11- pieces, incubated in a solution containing 1 mg/ml DAB, BODIPY581/591) following the procedure described by vacuum infiltrated during 30 min, and then maintained Shen et al. (2020). Epifluorescence of fluorescent dye in the dark at room temperature overnight. Leaves were was visualized using a Leica DM6 and the fluorescence washed with 75% ethanol. Following DAB staining, the emitted by oxidized C11-BODIPY581/591 was observed at rice leaves were subjected to Perls staining. 488 nm in an Leica DM6 microscope under GFP fluores- Perls staining was conducted on leaves and leaf sheaths cence (488 nm excitation, 509 nm emission). of High-Fe, Control Fe and Low-Fe rice plants that have been inoculated with M. oryzae spores (48 hpi) as described by Stacey et al. (2008) with some modifica- RNA Isolation, RNA‑seq and Analysis of RNA‑seq Data tions. Briefly, rice leaves were vacuum-infiltrated in a Total RNAs were obtained using Maxwell(R) RSC Plant fixing solution (chloroform:methanol:glacial acetic acid; RNA Kit (Promega). For RNA-Seq, two biological repli- 6:3:1, v/v) for 1 h and incubated overnight at room tem- cates for each condition were examined, each biological perature. After washing with distilled water (three times), replicate consisting of leaves from five individual plants samples were vacuum infiltrated with a pre-warmed (youngest developed leaf ). RNA concentration and purity (37 °C) staining solution (4% HCl and 4% K-ferrocyanide were checked using a spectrophotometer (NanoDrop, at equal volumes) for 1 h, incubated 4 h more at 37 °C ND-1000). RNA quality and integrity was examined on in the same solution without vacuum, and then washed an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.; three times with distilled water. The Perls reagent (potas- RNA integrity number (RIN) ≥ 8). A total of eight librar- sium ferrocyanide), in the presence of reactive ­ Fe3+, ies were subjected to RNA-Seq analysis (125 paired-end produces an insoluble blue precipitate (Prussian blue) reads), yielding an average of 24,812,610 clean reads/ (Roschzttardtz and Curie 2010). Sections were mounted library (Additional file 8: Table S7). Raw reads were pro- in glycerol 50%. All the images were taken in an Aixo- cessed and analyzed as previously described (Sánchez- Phot DP70 microscope under bright light. Three bio- Sanuy et al. 2019). Reads were mapped against the logical replicates with 4 technical replicates each were reference genome, Oryza sativa sp. japonica (IRGSP-1.0 performed. Ensembl release 41). To identify differentially expressed Trypan blue staining was performed to detect cell genes, a FDR cutoff < 0.01 and log2FC − 0.5 ≤ or ≥ 0.5 death in M. oryzae-infected rice leaves. For this, detached was applied. Gene Ontology (GO) enrichment analy- rice leaves were boiled in alcoholic lactophenol (96% eth- sis (GOEA) was performed using the AgriGO web tool anol-lactophenol (1:1 v/v) containing 0.1 mg/ml trypan (Parametric Analysis of Gene Set Enrichment) (Du et al. blue for 8 min and cleared in 70% chloral hydrate solu- 2010; https://​bioin​fo.​cau.​edu.​cn/​agriG​Ov2/). Enriched tion overnight. Samples were stored in glycerol 70% until GO terms were clustered and plotted with the online microscopic observation. analysis tool ReviGO (https://​revigo.​irb.​hr/) (Supek et al. 2011). Cluster analysis was used to determine the expres- Lipid Peroxidation Assay and BODIPY Staining sion patterns of differentially expressed genes identified The level of lipid peroxidation was determined by meas- by RNA-Seq using the Panther gene ontology web tool uring the amount of MDA produced by the thiobarbi- (http://​geneo​ntolo​gy.​org/). The FPKM value was used as turic acid reaction as described by Campo et al. (2014). the expression level. Heatmaps were drawn by R pheat- Briefly, samples (0.05 g) were homogenized in 1 ml of map package. 80% (v/v) ethanol and centrifuged at 16,000g for 20 min at 4 °C. The supernatant (0.5 ml) was recovered and mixed with 0.5 ml of 20% (w/v) trichloroacetic acid con- Expression Analysis by qRT‑PCR taining 0.65% (w/v) thiobarbituric acid. The mixture was Total RNA was extracted from plant tissues using TRi- incubated at 85 °C for 30 min and then quickly cooled in zol reagent (Invitrogen). For quantitative RT-PCR an ice bath. After centrifugation (10,000g, 10 min), the (RT-qPCR), the first complementary DNA was synthe- absorbance of the supernatant was measured at 532 nm sized from DNase-treated total RNA (0.5 μg) with High and 600 nm (Spectramax M3 reader, Molecular Devices, Capacity cDNA Reverse Transcription (Life technology, USA). The value for nonspecific absorption at 600 nm Applied Biosystems). Amplification involved cDNA (2 μl, was substracted from the 532 nm reading. The MDA con- 5 ng/μl) in optical 384-well plates (Roche Light Cycler tent was calculated using its molar extinction coefficient Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 20 of 23 480; Roche Diagnostics, Mannheim, Germany) with and PMSF, 0.1 mM each). Protein concentrations of total SYBR Green I dye and gene-specific primers (Additional and apoplastic fluid extracts was quantified using Brad- file 9: Table S8). The Ubiquitin1 gene (Os06g0681400) ford. For MDH assay, protein samples (10 μl of either was used to normalize transcript levels. Results from one total protein or apoplastic fluid extracts, 0.1 μg/μl) were of three independent experiments with similar results are added to the reaction mixture containing 0.4 mM NADH, shown. Four biological replicates which included RNA 0.2 mM oxaloacetic acid, 1 mM MOPS, pH 7.5, in 96 samples used for RNASeq analysis (two technical repli- micro titer plates. The absorbance at 340 nm at 25 °C was cates each) were analyzed. Asterisks indicate significant recorded during 10 min. MDH activity was calculated as differences calculated by two way ANOVA *, P ≤ 0.05; **, (Δabsorbance 340 nm/min/mg). Malate Dehydrogenase P ≤ 0.01; ***, P ≤ 0.001). (MDH) activity and was found to be in the range of 3.5– 4% compared to total leaf protein extracts. Analysis of Total and Apoplastic Fe Apoplastic Fe concentrations weredetermined by ICP- For total Fe content, the youngest developed leaves (third MS technique (see above) after diluting 30 µL of the leaf ) were air-dried and then 25–50 mg were mineralized apoplastic solutions to 1.2 mL with 2% ­HNO3 in (v/v) by a microwave digester system (MULTIWAVE-ECO, bidistilled water. Calibration curve were obtained using Anton Paar GmbH) in 65% (v/v) ­HNO3. Aliquots of the multi-ICP-MS standard solutions (Ultra Scientific, USA). mineralized samples were adequately diluted in Milli-Q Five biological replicates (three technical replicates each) water and the concentrations of total Fe were measured were analyzed. Differences were calculated by one way by ICP-MS technique (Bruker Aurora M90 ICP-MS, ANOVA *, P ≤ 0.05; **, P ≤ 0.01). Bruker Daltonik GmbH). In order to check the nebuli- zation performance, an aliquot of 2 mg ­L−1 of an inter- Quantification of Ferrous (­ Fe2+) and Ferric Fe ­(Fe3+) nal standard solution (72Ge, 89Y, and 159Tb) was added For quantification of ­Fe2+ and ­Fe3+, we used the fer- to the samples and calibration standards to give a final rozine assay described by Adolfsson et al. (2017) with concentration of 20 µg ­L−1. Possible polyatomic interfer- some modifications. Specifically, dried leaf samples ences were removed by using CRI (Collision–Reaction- (50–70 mg of the youngest developed leaves of three- Interface) with an H ­ 2 flow of 80 mL/min flown though week-old plants, 5 plants each condition) were pulver- skimmer cone. Calibration curve were obtained using ized in a Tissue Lyser and incubated in 1 M HCl (700 μl) multi-ICP-MS standard solutions (Ultra Scientific, USA). overnight at room temperature under gentle shaking. Five biological replicates (three technical replicates each) After centrifugation (10,000 rpm, 5 min, room tempera- were analyzed. Differences were calculated by one way ture), the supernatant was recovered. For determination ANOVA *, P ≤ 0.05; **, P ≤ 0.01). of ­Fe2+, 15 μl of 2 mM Ferrozine in 50 mM ammonium The apoplast wash fluid (AWF) was obtained from the acetate pH 9.5 were added to 150 μl of plant extract. F ­ e3+ 2+ same plant material used for the analysis of total Fe con- was determined after reduction to F ­ e with 50% (v/v) tent. For this, the youngest developed leaves of three- hydroxylamine hydrochloride 6.25 M (a reducing agent week-old plants (1 gr of fresh weight) were cut in 5 cm that converts F ­ e3+ to F ­ e2+) added to 150 μl of the sam- pieces and placed in a tube containing 30 ml of 200 mM ples. After 20 min at room temperature, 15 μl of 2 mM ­CaCl2, 5 mM Na-acetate (pH 4.3 adjusted with gla- Ferrozine were added to the samples. Ferrozine, in the cial acetic acid) and 0.1 mM TPCK and 0.1 mM PMSF presence of ferrous ions gives a pink-purple color which (freshly prepared), and kept on ice under constant agita- can be measured spectrophotometrically. For each sam- tion for 1 h. Then, vacuum was applied for 30 min and ple, the background was determined in the ammonium leaves were removed and carefully dried. Dried leaves acetate buffer and subtracted from values obtained in were rolled and placed in a tube for AWF recovery. the corresponding Ferrozine-treated samples. Equivalent Leaves were centrifuged a 3000 rpm, 15 min at 4 °C. AWF volumes of 1 M HCl and hydroxylamine hydrochloride was recovered from the bottom of the tube and kept (in 1 M HCl) were used as blanks for non-reduced and in − 20 until use. reduced samples, respectively. Absorbance was read at To assess that the apoplast fluid was devoid of cyto- 562 nm. ­Fe3+ concentration was estimated by subtract- plasmic contaminants we assayed malate dehydrogenase ing reduced hydroxylamine values with ­ Fe2+ values. (MDH) activity (Goulet et al. 2010). For each material, Standard curves using increasing Fe concentrations were MDH activity in apoplast samples was compared with performed. Five biological replicates with two technical that of total protein extracts. Total protein extracts were replicates each were examined Asterisks indicate sig- prepared from rice leaves using 1 mM MOPS, pH 7.5, nificant differences calculated by two-way ANOVA *, containing 5 mM NaCl, and protease inhibitors (TPCK P ≤ 0.05; **, P ≤ 0.01). Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 21 of 23 Quantification of Rice Phytoalexins Acknowledgements We thank Maria Ribaya and Héctor Martín for assistance in the preparation of For quantification of rice phytoalexins, leaf segments apoplast samples and infection assays on leaf sheath, respectively, and Mani (aprox. 2 cm in length, 200–300 mg) were mixed with Deepika Mallavarapu for critical reading of this manuscript. We also thank 40 vol. of 70% methanol and incubated at 4 °C overnight Martí Bernardo and Victor González for their assistance in submitting RNASeq data to the Gene Expression Omnibus (GEO) at the NCBI. with constant rotation. A 1 ml aliquot was centrifuged at maximum speed to remove cell debris and used for quan- Author Contributions tification of phytoalexins. Phytoalexins were quantified SC and BSS designed the research, FS-S performed most experiments and analyzed the data. RMC contributed with experiments for Fe and ROS by liquid chromatography-tandem mass spectrometry detection, MDA determination and BODIPY staining. KT and KO performed (LC–MS/MS) as previously described (Miyamoto et al. phytoalexin analysis. GAS performed iron content determinations. BSS, SC and 2016). Three biological replicates with two technical rep- FS-S wrote the paper with contributions from other authors. All authors have read and approved the manuscript. licates each were performed. Significant differences in phytoalexin accumulation were evaluated with two-way Funding ANOVA *, P ≤ 0.05; **, P ≤ 0.01). Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This research was supported by Projects RTI2018-101275-B- I00 and PID2021-128825OB-I00 funded by MCIN/AEI/and by “ERDF A way of Supplementary Information making Europe” to BSS. We acknowledge financial support from the MCIN/ The online version contains supplementary material available at https://​doi.​ AEI/10.13039/501100011033 through the “Severo Ochoa Programme for org/​10.​1186/​s12284-​022-​00609-w. Centres of Excellence in R&D” (SEV-2015-0533 and CEX2019-000902-S) and the CERCA Programme/“Generalitat de Catalunya”. R. M-C is a recipient of a Ph.D. Grant from the “Ministerio de Ciencia, Tecnología, Conocimiento e Innovación Additional file 1: Fig. S1. Characterization of rice plants grown under del Gobierno de Chile (ref. 72210076). different Fe conditions. Fig. S2. Venn diagrams representing transcrip‑ tional changes in response to M. oryzae infection in leaves of High-Fe and Availability of Data and Materials Control plants. Fig. S3. GO enrichment analysis of genes down-regulated The RNA sequence datasets generated in this study can be found at the by M. oryzae infection in Control and High-Fe plants (48 hpi). Fig. S4. Hier‑ National Center for Biotechnology Information (NCBI) Gene Expression Omni‑ archical clustering of differentially expressed genes by RNA-Seq analysis. bus (GEO) with the accession number “GSE202997” (https://​www.​ncbi.​nlm.​nih.​ Fig. S5. Phenylpropanoid biosynthesis pathway. Genes whose expres‑ gov/​geo/​query/​acc.​cgi?​acc=​GSE20​2997). sion is regulated by M. oryzae infection are indicated. Fig. S6. Expression of diterpene phytoalexin biosynthetic genes in leaves of Control and High-Fe plants (−, mock-inoculated; +, M. oryzae- inoculated). Fig. S7. Declarations Accumulation of phytoalexins in leaves of Control and Low-Fe plants. Fig. S8. Expression of genes involved in Fe homeostasis in leaves of Control Ethics Approval and Consent to Participate and High-Fe plants (−, mock-inoculated; +, M. oryzae- inoculated). Fig. S9. Not applicable. Malate dehydrogenase (MDH) assay in total protein extracts and apoplast fluid from Control and High-Fe plants (mock-inoculated and M. oryzae- Consent for Publication inoculated). Fig. S10. Trypan blue staining for cell death in M. oryzae- Not applicable. infected leaves of Control and High-Fe rice plants. Fig. S11. Accumulation of Fe in sheaths of Low-Fe, Control-Fe, and High-Fe in rice plants that have Competing Interests been treated with the ferroptosis inhibitor Ferrostatin-1 (+ Fer-1), or not The authors declare no competing interests. (− Fer-1). Additional file 2: Table S1. Differentially expressed genes (DEGs) in Author details 1 leaves of High-Fe plants relative to Control plants identified by RNASeq Centre for Research in Agricultural Genomics (CRAG), CSIC-IRTA-UAB-UB, analysis (no infection conditions). Campus Universitat Autónoma de Barcelona (UAB), Bellaterra (Cerdanyola del Vallés), Barcelona, Spain. 2 Biotechnology Research Center, The University Additional file 3: Table S2. List of differentially expressed genes (DEGs) in of Tokyo, Tokyo, Japan. 3 Dipartimento di Scienze Agrarie e Ambientali, Univer‑ leaves of High-Fe (S2a) and Control (S2b) plants at 48 hpi with M. oryzae sità degli Studi di Milano, Milan, Italy. 4 Consejo Superior de Investigaciones spores, relative to the corresponding mock-inoculated plants. Científicas (CSIC), Barcelona, Spain. 5 Present Address: Fundació Miquel Agustí, Additional file 4: Table S3. GO enrichment analysis of genes regulated Campus Baix Llobregat, Castelldefels, Barcelona, Spain. by M.oryzae infection in High-Fe and Control plants relative to the cor‑ responding mock-inoculated plants. Received: 5 June 2022 Accepted: 30 November 2022 Additional file 5: Table S4. Expression data of defense-related genes in leaves of Control and High-Fe plants during M. oryzae infection (48 hpi). Additional file 6: Table S5. Expression data (FPKM and ­Log2FC) of genes involved in the biosynthesis of diterpene phytoalexins and sakuranetin References in leaves of Control and High-Fe plants (M. oryzae-infected and mock- Adolfsson L, Nziengui H, Abreu IN, Šimura J, Beebo A, Herdean A, Aboaliza‑ inoculated plants). deh J, Široká J, Moritz T, Novák O, Ljung K, Schoefs B, Spetea C (2017) Additional file 7: Table S6. Expression data (FPKM and ­Log2FC) of genes Enhanced secondary- and hormone metabolism in leaves of arbuscu‑ involved in Fe homeostasis in leaves of Control and High-Fe plants (M. lar mycorrhizal Medicago truncatula. Plant Physiol 175:392–411 oryzae-infected and mock-inoculated plants). Agrawal GK, Rakwal R, Jwa NS, Agrawal VP (2001) Signalling molecules and blast pathogen attack activates rice OsPR1a and OsPR1b genes: Additional file 8: Table S7. Statistics for RNAseq analysis of leaves from a model illustrating components participating during defence/stress Control and High-Fe plants, mock-inoculated and M. oryzae-inoculated response. Plant Physiol Biochem 39(12):1095–1103 (48 hpi). Ahuja I, Kissen R, Bones AM (2012) Phytoalexins in defense against patho‑ Additional file 9: Table S8. List of oligonucleotides used in this study. gens. Trends Plant Sci 17:73–90 Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 22 of 23 Ali S, Ganai BA, Kamili AN, Bhat AA, Mir ZA, Bhat JA, Tyagi A, Islam ST, Kieu NAMP, Aznar A, Segond D, Rigault M, Simond-côte E, Kunz C, Soulie Mushtaq M, Yadav P, Rawat S, Grover A (2018) Pathogenesis-related M, Expert D, Dellagi A (2012) Iron deficiency affects plant defence proteins and peptides as promising tools for engineering plants with responses and confers resistance to Dickeya Dadantii and Botrytis multiple stress tolerance. Microbiol Res 212:29–37 cinerea. Mol Plant Pathol 13:816–827 Aung MS, Masuda H (2020) How does rice defend against excess iron? Lamari L (2008) Assess 2.0: image analysis software for plant disease quanti‑ Physiological and molecular mechanisms. Front Plant Sci 11:1–8 fication. APS Press, St. Paul Aznar A, Patrit O, Berger A, Dellagi A (2015) Iron localization in infected tissues. Li J, Yin J, Wu JX, Wang LY, Liu Y, Huang LQ, Wang RH, Yao N (2022) Cera‑ Mol Plant Patho 16:521–528. https://​doi.​org/​10.​1111/​mpp.​12208 mides regulate defense response by binding to RbohD in Arabidopsis. Campo S, Baldrich P, Messeguer J, Lalanne E ,Coca M, San Segundo B (2014) Plant J 109:1427–1440 Overexpression of a Calcium-Dependent Protein Kinase Confers Salt and Liu G, Greenshields DL, Sammynaiken R, Hirji RN, Selvaraj G, Wei Y (2007) Drought Tolerance in Rice by Preventing Membrane Lipid Peroxidation. Targeted alterations in iron homeostasis underlie plant defense Plant Physiol 165:688–704 responses. J Cell Sci 120:596–605 Connorton JM, Balk J, Rodríguez-Celma J (2017) Iron homeostasis in plants- Mansoor S, Wani OA, Lone JK, Manhas S, Kour N, Alam P, Ahmad A, Ahmad P a brief overview. Metallomics 9:813–823 (2022) Reactive oxygen species in plants: from source to sink. Antioxi‑ Conrad M, Kagan VE, Bayir H, Pagnussat GC, Head B, Traber MG, Stockwell dants 11(2):225 BR (2018) Regulation of lipid peroxidation and ferroptosis in diverse Marschner H, Römheld V (1994) Strategies of plants for acquisition of iron. species. Genes Dev 32(9–10):602–619 Plant Soil 165(2):261–274 Dangol S, Chen Y, Hwang BK, Jwa NS (2019) Iron- and reactive oxygen Martinez-Medina A, Flors V, Heil M, Mauch-Mani B, Pieterse CMJ, Pozo MJ, species-dependent ferroptotic cell death in rice-Magnaporthe oryzae Ton J, van Dam NM, Conrath U (2016) Recognizing plant defense prim‑ interactions. Plant Cell 31:189–209 ing. Trends Plant Sci 21:818–822 Deák M, Horváth GV, Davletova S, Török K, Sass L, Vass I, Barna B, Király Z, Midoh N, Iwata M (1996) Cloning and characterization of a probenazole- Dudits D (1999) Plants ectopically expressing the iron-binding protein, inducible gene for an intracellular pathogenesis-related protein in rice. ferritin, are tolerant to oxidative damage and pathogens. Nat Biotech‑ Plant Cell Physiol 37(1):9–18 nol 17(2):192–196 Miyamoto K, Fujita M, Shenton MR, Akashi S, Sugawara C, Sakai A, Horie Distéfano AM, Martin MV, Córdoba JP, BellidoAM DS, Colman SL, Soto D, K, Hasegawa M, Kawaide H, Mitsuhashi W, Nojiri H, Yamane H, Kurata Roldán JA, Bartoli CG, Zabaleta EJ, Fiol DG, Stockwell BR, Dixon SJ, N, Okada K, Toyomasu T (2016) Evolutionary trajectory of phytoalexin Pagnussat GC (2017) Heat stress induces ferroptosis-like cell death in biosynthetic gene clusters in rice. Plant J 87(3):293–304 plants. J Cell Biol 216:463–476 Nozoye T, Nagasaka S, Kobayashi T, Takahashi M, Sato Y, Sato Y, Uozumi N, Distéfano AM, López GA, Setzes N, Marchetti F, Cainzos M, Cascallares M, Nakanishi H, Nishizawa NK (2011) Phytosiderophore efflux transport‑ Zabaleta E, Pagnussat GC (2021) Ferroptosis in plants: triggers, pro‑ ers are crucial for iron acquisition in graminaceous plants. J Biol Chem posed mechanisms, and the role of iron in modulating cell death. J Exp 286(7):5446–5454 Bot 72(6):2125–2135 Okada A, Shimizu T, Okada K, Kuzuyama T, Koga J, Shibuya N, Nojiri H, Du Z, Zhou X, Ling Y, Zhang Z, Su Z (2010) agriGO: a GO analysis toolkit for Yamane H (2007) Elicitor induced activation of the methylerythritol the agricultural community. Nucleic Acids Res 38(suppl_2):W64–W70 phosphate pathway toward phytoalexins biosynthesis in rice. Plant Mol Eide D, Broderius M, Fett J, Lou GM (1996) A novel iron-regulated metal Biol 65:177–187 transporter from plants identified by functional expression in yeast. Peris-peris C, Serra-cardona A, Campo S, Arin J (2017) Two NRAMP6 isoforms Proc Natl Acad Sci U S A 93(11):5624–5628 function as iron and manganese transporters and contribute to dis‑ Fernandez J, Orth K (2018) Rise of a cereal killer: the biology of Magnaporthe ease resistance in rice. Mol Plant-Microbe Interact 30:385–398 oryzae biotrophic growth. Trends Microbiol 26:582–597 Qi M, Yang Y (2002) Quantification of Magnaporthe grisea during infection Fichman Y, Miller G, Mittler R (2019) Whole-plant live imaging of reactive of rice plants using real-time polymerase chain reaction and northern oxygen species. Mol Plant 12(9):1203–1210 blot/phosphoimaging analyses. Phytopatology 92:870–876 Fu D, Li J, Yang X, Li W, Zhou Z, Xiao S, Xue C (2022) Iron redistribution Robinson NJ, Procter CM, Connolly EL, Lou GM (1999) A ferric-chelate reduc‑ induces oxidative burst and resistance in maize against Curvularia tase for iron uptake from soils. Nature 397(6721):694–697 lunata. Planta 256:46 Rodríguez-Celma J, Chun Pan I, Li W, Lan P, Buckhout TJ, Schmidt W (2013) Goulet C, Goulet C, Goulet MC, Michaud D (2010) 2-DE proteome The transcriptional response of Arabidopsis leaves to Fe deficiency. maps for the leaf apoplast of Nicotiana benthamiana. Proteomics Front Plant Sci 4:276 10(13):2536–2544 Roschzttardtz H, Curie C (2010) Straightforward histochemical staining of Hasegawa M, Mitsuhara I, Seo S, Imai T, Koga J, Okada K, Yamane H, Ohashi Fe by the adaptation of an old-school technique. Plant Signal Behav Y (2010) Phytoalexin accumulation in the interaction between rice and 5:56–57 the blast fungus. Mol Plant-Microbe Interact 23:1000–1011 Sánchez-Sanuy F, Peris-Peris C, Tomiyama S, Okada K, Hsing YI, San Segundo Hasegawa M, Mitsuhara I, Seo S, Okada K, Yamane H, Iwai T, Ohashi Y (2014) B, Campo S (2019) Osa-miR7695 enhances transcriptional priming Analysis on blast fungus-responsive characters of a flavonoid phyto‑ in defense responses against the rice blast fungus. BMC Plant Biol alexin sakuranetin; accumulation in infected rice leaves, antifungal 19:1–16 activity and detoxification by fungus. Molecules 19(8):11404–11418 Santi S, Schmidt W (2009) Dissecting iron deficiency-induced proton extru‑ Hassinen VH, Tervahauta AI, Schat H, Kärenlampi SO (2011) Plant metal‑ sion in Arabidopsis roots. New Phytol 183:1072–1084 lothioneins—metal chelators with ROS scavenging activity? Plant Biol Shen Q, Liang M, Yang F, Deng YZ, Naqvi NI (2020) Ferroptosis contributes to 13(2):225–232 developmental cell death in rice blast. New Phytol 227:1831–1846 Herlihy JH, Long TA, McDowell JM (2020) Iron homeostasis and plant Shimizu T, Lin F, Hasegawa M, Okada K, Nojiri H, Yamane H (2012) Purifica‑ immune responses: recent insights and translational implications. J Biol tion and identification of naringenin 7-O-methyltransferase, a key Chem 295:13444–13457 enzyme in biosynthesis of flavonoid phytoalexin sakuranetin in rice. J Inoue H, Kobayashi T, Nozoye T, Takahashi M, Kakei Y, Suzuki K, Nakazono Biol Chem 287:19315–19325 M, Nakanishi H, Mori S, Nishizawa NK (2009) Rice OsYSL15 is an iron- Stacey MG, Patel A, McClain WE, Mathieu M, Remley M, Rogers EE, regulated iron (III)-deoxymugineic acid transporter expressed in the Gassmann W, Blevins DG, Stacey G (2008) The Arabidopsis AtOPT3 roots and is essential for iron uptake in early growth of the seedlings. J protein functions in metal homeostasis and movement of iron to Biol Chem 284:3470–3479 developing seeds. Plant Physiol 146:589–601 Ishimaru Y, Suzuki M, Tsukamoto T, Suzuki K, Nakazono M, Kobayashi T, Wada Sun TK, Sang GK, Du HH, Sun YK, Han JK, Byung HL, Jeung JL, Kyu YK Y, Watanabe S, Matsuhashi S, Takahashi M, Nakanishi H, Mori S, Nishi‑ (2004) Proteomic analysis of pathogen-responsive proteins from rice zawa NK (2006) Rice plants take up iron as an Fe3+-phytosiderophore leaves induced by rice blast fungus. Magnaporthe Grisea Proteomics and as F­ e2+. Plant J 45:335–346 4(11):3569–3578 Kaur N, Sharma I, Kirat K, Pati P (2016) Detection of reactive oxygen species Supek F, Bosnjak M, Skunca N, Smuc T (2011) REVIGO Summarizes and visu‑ in Oryza sativa L. (Rice). Bio-Protoc 6:1–9 alizes long lists of gene ontology terms. PLoS ONE 6:1–9 Sánchez‑Sanuy et al. Rice (2022) 15:68 Page 23 of 23 Torres MA (2010) ROS in biotic interactions. Physiol Plant 138:414–429 Umemura K, Ogawa N, Shimura M, Koga J, Usami H, Kono T (2003) Pos‑ sible role of phytocassene, rice phytoalexin, in disease resistance of rice against the blast fungus Magnaporthe grisea. Biosci Biotechnol Biochem 67:899–902 Verbon EH, Trapet PL, Stringlis IA, Kruijs S, Bakker PAHM, Pieterse CMJ (2017) Iron and immunity. Annu Rev Phytopathol 55:355–375 Viswanath KK, Varakumar P, Pamuru RR, Basha SJ, Mehta S, Rao AD (2020) Plant lipoxygenases and their role in plant physiology. J Plant Biol 63(2):83–95 Weinberg ED (1975) Nutritional immunity: host’s attempt to withhold iron from microbial invaders. JAMA J Am Med Assoc 231(1):39–41 Wilson R, Talbot NJ (2009) Under pressure: investigating the biology of plant infection by Magnaporthe oryzae. Nat Rev Microbiol 7:185–195 Wu J, Kim SG, Kang KY, Kim JG, Park SR, Gupta R, Kim YH, Wang Y, Kim ST (2016) Overexpression of a pathogenesis-related protein 10 enhances biotic and abiotic stress tolerance in rice. Plant Pathol J 32:552–562 Yadav V, Wang Z, Wei C, Amo A, Ahmed B, Yang X, Zhang X (2020) Phenyl‑ propanoid pathway engineering: An emerging approach towards plant defense. Pathogens 9(4):312 Yang WS, Kim KJ, Gaschler MM, Patel M, Shchepinov MS, Stockwell BR (2016) Peroxidation of polyunsaturated fatty acids by lipoxygenases drives fer‑ roptosis. Proc Natl Acad Sci U S A 113(34):E4966–E4975 Ye F, Albarouki E, Lingam B, Deising HB, von Wirén N (2014) An adequate Fe nutritional status of maize suppresses infection and biotrophic growth of Colletotrichum graminicola. Physiol Plant 151:280–292 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations.