Abstract
The roles of mitogen-activated protein kinases (MAPKs) in plant-fungal pathogenic interactions are less understood in crops. Here, microscopic, phenotyping, proteomic and biochemical analyses revealed that independent TALEN-based knockout lines of Hordeum vulgare MITOGEN-ACTIVATED PROTEIN KINASE 3 (HvMPK3 KO) were resistant against Fusarium graminearum infection. When co-cultured with roots of the HvMPK3 KO lines, F. graminearum hyphae were excluded to the extracellular space, the growth pattern of hyphae was considerably deregulated, mycelia development was less efficient and number of appressoria and their penetration potential were significantly reduced. Intracellular penetration of hyphae was preceded by the massive production of reactive oxygen species (ROS) in attacked cells of the wild type, but it was mitigated in the HvMPK3 KO lines. Suppression of ROS production in these lines coincided with the elevated abundances of catalase and ascorbate peroxidase. Moreover, differential proteomic analysis revealed downregulation of defense-related proteins in wild type, and the upregulation of peroxidases, lipid transfer proteins, and cysteine proteases in HvMPK3 KO lines after 24h of F. graminearum inoculation. Consistently with proteomic analysis, microscopic observations showed an enhanced suberin accumulation in roots of HvMPK3 KO lines, most likely contributing to the arrested infection by F. graminearum. These results suggest that TALEN-based knockout of HvMPK3 leads to the barley root resistance against Fusarium root rot.
Introduction
Cereals are a major food staple for the world population, however, there is a substantial reduction in the annual yield because of pathogens, weeds, temperature extremes, high salt concentrations, drought, and arid conditions (Sewelam et al., 2016). Diseases caused by fungal pathogens alone are responsible for the destruction of approximately 125 million tons of crops worldwide in the last decade (Vogelgsang et al., 2019).
To combat invading fungi, plants recognize microbe-associated molecular patterns (MAMPs), and damage-associated molecular patterns (DAMPs), which are essential for triggering plant immunity (Schwessinger and Zipfel, 2008). MAMPs and DAMPs elicit so-called pattern-triggered immunity (PTI), which is mediated by plant pattern recognition receptors (PRRs) (Huang et al., 2016). PTI is characterized by a series of events, including signaling through protein kinase cascades, generation of reactive oxygen species (ROS), calcium ion influx, transcriptional reprogramming, cell wall appositions, and hormonal changes (Kurusu et al., 2015). ROS play a dual role during plant immune responses. In addition to signaling functions (Qi et al., 2017), their over-accumulation leads to the hypersensitive response, which hinders the progression of the biotrophic pathogens to the plant tissues (Camagna and Takemoto, 2018; Camejo et al., 2016). In contrast, necrotrophic pathogens may benefit from nutrients provided by the damaged tissues, and thus ROS accumulation facilitates the invasion of the pathogen into the tissues (Barna et al., 2012; Kámán-Tóth et al., 2019).
Mitogen-activated protein kinases (MAPKs) are integrated in signaling cascades responsible for conveying signals generated by extracellular and intracellular stimuli (Komis et al., 2011). In Arabidopsis and rice, there are several numbers of MAPKs described, of which mostly AtMPK3, AtMPK4, and AtMPK6 and their rice orthologues, are responsible for plant resistance against pathogens (Bigeard et al., 2015; Kishi-Kaboshi et al., 2010; Meng and Zhang, 2013). The signaling pathway involving AtMPK3/AtMPK6 pair is responsible for the regulation of camalexin biosynthesis (a major phytoalexin found in Cruciferae plants) during infection of Arabidopsis by necrotrophic fungal pathogens (Mao et al., 2011). However, in the case of bacterial infection, AtMPK4 is responsible for its regulation (Bazin et al., 2020). MPK3/MPK6 in different plant species also regulate hypersensitive cell death responses and the generation of reactive oxygen species (Kim and Zhang, 2004; Kroj et al., 2003; Liu et al., 2007; Ren et al., 2002). The importance of MAPK signaling in plant-pathogen interactions is also supported by studies of bacterial effectors, several of which target and inhibit plant MAPK cascades (Cui et al., 2010; Zhang and Dong, 2007). Recently, it was shown that knock-out mutations of HvMPK3 prepared by TALEN technology attenuated the responsivity of barley to bacterial PAMP flagellin 22 (flg22), manifested by decreased abundance of chitinases and pathogenesis-related proteins (Takáč et al., 2021).
Fusarium graminearum causes a major loss of barley yield due to head blight and root rot diseases. This fungus is considered a hemibiotrophic pathogen, living in the host plants for a short period (hours to several days), before switching to necrotrophic form, which retrieves nutrients from dead cells (Tucker et al., 2021, 2019). Although Fusarium head blight was earlier believed as a primary disease caused by F. graminearum, the root colonization by this pathogen is currently recognized as very important for immense economic losses. Fusarium root rot causes rapid necrosis, leading to a significant reduction in root growth and biomass, which is accompanied by the progression of the pathogen to the stem base (Smiley et al., 2005). The growth-inhibiting impact of the pathogen was assigned to the production of the mycotoxin deoxynivalenol (DON) (Masuda et al., 2007). The hyphae colonize intra- and intercellular spaces in the root cortex in sensitive wheat cultivars, while the invasion in resistant cultivar is stopped at the epidermal cells (Wang et al., 2015). Barley defense mechanisms against Fusarium root rot are poorly understood. So far, these resistance strategies involved de novo biosynthesis of barley root exudates (Lanoue et al., 2010), activation of jasmonic acid (JA)- dependent defense genes and genes related to DON detoxification (Wang et al., 2018). The role of MAPKs in barley responses to Fusarium has not been elucidated, and their participation in Fusarium-induced signal transduction is unknown.
In the present study, we discovered that independent barley lines with TALEN-based knock-out mutations of HvMPK3 exhibited higher root resistance to F. graminearum and produced less ROS than wild type plants. Fusarium hyphae potential to penetrate the root cells of HvMPK3 lines significantly decreased. This was accompanied by the attenuation of defense responses in wild type plants and the induction of upregulated levels of cysteine proteases and secretory peroxidases in HvMPK3 KO lines, as documented by proteomic analysis. Our results indicate that ROS generation, might facilitate the invasion of wild type plants by F. graminearum, but elevated abundances of cytosolic ascorbate peroxidase and catalase might reduce ROS levels, thus contributing to the higher resistance of HvMPK3 KO lines. Finally, the stiffening of the cell walls by suberin deposition likely represents a barrier, which prevents pathogen invasion to the root cells of HvMPK3 KO lines.
Results
Phenotype of wild type and HvMPK3 KO plants infected by Fusarium graminearum
Fungal mycelium developed extensively after infecting the roots of five days old barley seedlings by F. graminearum conidia. Ten days after infection, densely developed mycelium surrounded seeds and basal parts of the root system in wild type (WT) plants (Figure S1A). Roots of WT plants were largely arrested in their growth (Figure S1A, black arrows). In contrast, the root system of HvMPK3 KO-A (Figure S1B), HvMPK3 KO-B (Figure S1C), and HvMPK3 KO-D (Figure S1D) lines inoculated by F. graminearum developed well without any visible reduction of root growth. In addition, the development of F. graminearum mycelia itself was inhibited, particularly around the seeds and roots (Figure S1B-D). Dark brown coloration of F. graminearum mycelia appeared close to the seeds and the basal parts of the root system of HvMPK3 KO lines (Figure S1B-D). On WT genotypes, however, F. graminearum mycelia showed pale white color and massive coverage of infected plants (Figure S1A).
Inoculation of control seedling roots (Figure 1A-D) with F. graminearum spores led to their germination and subsequent colonization of roots by growing mycelia. In a period of 24 h after inoculation (Figure 1E-H), the root apex of WT plants was massively invaded by GFP-expressing F. graminearum hyphae (Figure 1E). However, only sparse hyphae were present on the root surface of HvMPK3 KO lines (Figure 1F-H). Propidium iodide (PI) staining of cell walls in living cells of uninfected plants (Figure 1A-D) served as a marker of root tissue organization, while accumulation of PI in nuclei in dead cells of infected plants (Figure 1E-H) indicated their mortality after penetration by F. graminearum hyphae. The cell death in the epidermis of WT roots colonized by F. graminearum was evident (Figure 1E), while the amount of dead epidermal cells in inoculated roots of HvMPK3 KO lines was considerably low (Figure 1F-H). The quantitative analysis of cell death rate revealed a statistically significant increase in infected WT plants. In contrast, there was no statistical difference in the number of dead root epidermal cells between uninfected plants and plants analyzed 24h after inoculation in HvMPK3 KO lines (Figure 1I).
Examination of plants 48h after inoculation with F. graminearum showed that roots of WT plants were massively colonized by the fungal mycelium (Figure 1J), but mycelium was much less developed around the infected area on the root surface in HvMPK3 KO lines (Figure 1K-M). Surprisingly, this fact was related to different growth pattern of hyphae, including considerable changes in their density, but also in the shape of developing mycelium. On the surface of WT roots, the hyphae were growing in parallel with longitudinal root axis and closely associated with the surface of root epidermal cells (Figure 1J). In contrast, hyphae were growing away of the roots and often without touching the root surface of HvMPK3 KO lines (Figure 1K-M). In addition, mycelium developed around HvMPK3 KO roots was malformed with unusual wavy growth pattern of hyphae, and showed changed growth direction oriented perpendicularly to the axis of elongated root epidermal cells (Figure 1K-M). Using analysis of the angular distribution of growing mycelia, we quantitatively determined a degree of anisotropy of hyphae distribution at the root surface. In the WT roots, the graph shows almost uniformly longitudinal orientation of hyphae relative to the root longitudinal axis (Figure S2A). Conversely, mycelia around root surfaces of HvMPK3 KO lines showed more random organization (Figures S2B-S2D). Quantitative evaluation revealed higher average values for angular distribution of hyphae on WT roots (Figure S2E), suggesting a high degree of anisotropy, characterized by hyphae arrangement in one orientation. Opposite, statistically significant lowering of values determining a high degree of isotropy characterized by uniform distribution of hyphae with no prevalent orientation, was found in HvMPK3 KO lines (Figure S2E). We also analysed fluorescence skewness, defining a pattern of fluorescence intensity distribution on the root surface and comparing how a ratio between high and low fluorescence intensities distribution is changing. This analysis revealed a high degree of fluorescence uniformity (referred by lower values) on the surface of WT roots (Figure S2F), but a high degree of non-uniformity of the fluorescence intensity distribution (characterized by distinct dark and bright fluorescent areas and referred by higher values) on the surface of HvMPK3 KO roots (Figure S2F).
In comparison to uninfected WT plants (Figure 2A), analysis of root system phenotypes of F. graminearum-treated WT plants (10 days after inoculation) revealed a reddish-brown coloration and malformation of some roots (Figure 2E). However, a comparison of roots of uninfected HvMPK3 KO lines (Figure 2B-D) with infected ones (Figure 2F-H) showed no change. Roots in the basal part of the root systems of HvMPK3 KO lines did not exhibit a reddish-brown coloration (Figure 2F-H). The morphology (Figure S3) and the length (Figure 2I) of roots measured 10 days after inoculation revealed that F. graminearum infection significantly reduced the root length in the WT plants, but there was no significant difference between control versus infected HvMPK3 KO lines (Figure 2I). These data indicate insensitivity of HvMPK3 KO plants to F. graminearum infection in early seedling stages, and sustained root growth without deleterious effects of the pathogen.
Growth of the mycelium on the root surface precedes the hyphae invasion to the root cells by cell wall penetration. Volumetric fluorescence visualization of GFP-expressing F. graminearum hyphae on the root surface allowed not only qualitatively determine the distribution and density of mycelium, but also its penetration ability. Therefore, we documented the depth of hyphal invasion by color-coded fluorescence intensity distribution along the Z-axis of imaging. In WT roots 48h after inoculation with spores, such analysis revealed formation of appressoria and infection hyphae invading the root epidermal cells. Deeper position of appressoria on cell surface and intracellular location of invading hyphae were clearly distinguished from root surface mycelium according to distinct fluorescence colorcoding (Figure 3A, B). In sharp contrast, the same analysis did not reveal penetration and intracellular infection hyphae formation in the root epidermis of HvMPK3 KO lines (Figure 3C-H). Overall microscopic analysis revealed formation of dense mycelium on the surface of WT roots, with frequent infection hyphae penetration to the root epidermal cells (Figure S4A, B). In contrary, low-density surface mycelium with different growth pattern, and no hyphae penetration in the root epidermal cells were revealed in HvMPK3 KO lines (Figure S4C-H). Moreover, occasional F. graminearum appressoria developed on root surfaces of HvMPK3 KO lines were often abnormal in shape and size (yellow arrows in Figure 3C-H). Differences in appressoria morphology between F. graminearum mycelium growing on the surface of WT roots (Figure 3A, B, 4A) and roots of HvMPK3 KO lines (Figure 3C-H, 4B-D) prompted us to analyze appressoria formation capacity, which may directly influence also their penetration ability to the root epidermal cells. We have found significantly lower numbers of appressoria (Figure 4E) and penetrating hyphae (Figure 4F) on the surface of HvMPK3 KO roots compared to the WT. Therefore, data from phenotypical analysis revealed different responses of WT and HvMPK3 KO roots to F. graminearum infection.
Fusarium-induced ROS production in barley roots
ROS production in infected plant cells is associated with the early recognition of pathogen at the apoplast, and is accompanied by sudden oxidative burst (Qi et al., 2017). Therefore, ROS production in roots of WT and HvMPK3 KO lines was analysed by CM-H2DCFDA fluorescence detection 24h after inoculation with F. graminearum spores. This analysis revealed a substantially higher ROS accumulation in WT (Figure 5A), compared to the HvMPK3 KO lines (Figure 5B-D). Consequently, quantitative evaluation of the fluorescent staining levels showed that ROS accumulation was substantially higher in the infected WT roots than in infected roots of HvMPK3 KO lines (Figure 5E). Independent spectrophotometric examination of ROS levels in the infected WT and HvMPK3 KO lines showed results consistent with these microscopic observations (Figure S5). In WT plants, ROS production in cells attacked by fungus preceded fungal hyphae penetration (Figure 5F). Such cells in the preinvasion stage were still alive and accumulated a high amount of ROS, while other surrounding cells with absent ROS fluorescence signal were dead (Figure 5F). In the root apex of WT plants, ROS production and accumulation occurred in cells not yet penetrated with F. graminearum hyphae, while in most of the root epidermal cells ROS production was not detected (Figure 5G). Based on the staining of nuclei by PI, these cells were already dead and penetrated by F. graminearum hyphae (Figure 5G). Conversely, only a limited number of root epidermal cells appeared dead after PI staining of HvMPK3 KO lines 24h after inoculation with F. graminearum spores (Figure 5H-J). Fluorescence signal for ROS detection was not observed, which was related to the low number of F. graminearum hyphae growing around the roots and their minimal contact with the root epidermal cells of HvMPK3 KO lines (Figure 5H-J).
Differential proteomic analysis
To elucidate resistance mechanisms of HvMPK3 KO lines against F. graminearum at molecular level, root proteomes of WT and HvMPK3 KO lines infected with F. graminearum were compared to mock controls 24h after infection.
In total, we have identified 133 differentially abundant proteins in F. graminearum- treated WT plants, of which 43 were upregulated and 90 were downregulated in infected versus control plants. In HvMPK3 KO lines, 94 differentially abundant proteins were found, from which 47 were upregulated and the same number was downregulated in infected versus control plants (Tables S1 and S2). We evaluated the differential proteomes of WT and HvMPK3 KO lines using GO annotation analysis (Figure S6-S7).
Here we compared the previously published differential proteome of HvMPK3-KO lines exposed to flg22 (Takáč et al., 2021) with the differential proteome of the same lines treated with Fusarium. In contrast to flg22, Fusarium caused the differential abundance of lipid transfer proteins (LTPs), histone isoforms and proteins involved in reactive oxygen species regulation (Figure 6). In more detail, abundances of histone isoforms were upregulated in WT, but downregulated in HvMPK3 KO lines treated with Fusarium (Figure 6). LTPs, a protein family affected by Fusarium but not by flg22, were distinctly affected in KO HvMPK3 lines compared to WT (Figure 6). Putative lipid transfer-like protein DIR1 and non-specific LTP1 were upregulated while YLS3 was downregulated in infected WT lines. In HvMPK3 KO lines, Fusarium caused upregulation of two VAS lipid transfer-like proteins and downregulation of CW21 non-specific LTP.
Concerning proteins involved in redox regulation, peptide methionine sulfoxide reductase B5, thioredoxin domain-containing protein 9 and glutathione S-transferase GSTU6 were downregulated in WT (Figure 7). Peptide methionine sulfoxide reductase B3 and glutathione S-transferase GSTU1 were negatively affected by Fusarium treatment in HvMPK3 KO lines. Monodehydroascorbate reductase 5 (involved in ascorbate regeneration) was downregulated, while thioredoxin M2 (a chloroplastic redox buffering protein) was upregulated in HvMPK3 KO plants (Figure 7).
Fusarium also deregulated secretory peroxidases in barley. Proteome-wide, HvMPK3 KO lines contained five secretory peroxidases and WT only three. Notably, abundances of secretory peroxidases were decreased in WT, while three out of five secretory peroxidases were upregulated in the HvMPK3 KO lines. Using specific activity staining on native gels, we have found that HvMPK3 KO lines exhibited also increased activities of three peroxidase isoforms (Rf 0.63, 0.90, 0.98), as compared to the WT (Figure 8A, B). To assess the abundance of major H2O2-decomposing enzymes, L-ascorbate peroxidase and catalase were analyzed using immunoblot assays. We detected cytosolic APX (cAPX) in the barley root extracts employing anti-ascorbate peroxidase (APX) antibody (Figure 8C, D). Remarkably, the abundance of cAPX increased in HvMPK3 KO lines, while it decreased in WT plants. In addition, we observed an increase of catalase abundance in both analyzed HvMPK3 KO lines, while it slightly decreased in WT (Figure 8E, F). These results suggest that HvMPK3 KO lines likely exert an increased capacity to decompose H2O2 during F. graminearum infection.
Moreover, cysteine proteases were more affected by Fusarium as by flg22. They were downregulated in WT but mostly upregulated in HvMPK3 KO lines. Next we identified more proteins belonging to the family of heat shock proteins in the Fusarium-induced differential proteome. The majority of them was downregulated in HvMPK3 KO lines.
Proteins involved in cell wall modifications and abiotic stress response were more and preferentially affected by Fusarium as compared to flg22, and these protein groups mostly showed reduced abundances in WT. Finally, enzymes with ubiquitination and deubiquitination activities were also affected by Fusarium, but without differences between WT and KO HvMPK3 lines.
As HvMPK3 KO lines exhibited reduced colonization of roots by F. graminearum hyphae, we focused on proteins predicted to be localized to extracellular space (Table S3). These included three cysteine proteinases, which were upregulated in HvMPK3 KO lines, and downregulated in WT. Furthermore, PR proteins (PR-1, thionin 2, glucan endo-1,3-beta-glucosidase 4) were found downregulated in WT. Notably, PR-1 protein was upregulated in HvMPK3 KO lines. Immunoblotting analysis validated this differential abundance of PR-1 in WT and HvMPK3 KO lines (Figure 8G, H). Some of the above-mentioned LTPs were predicted as extracellular, including DIR1 putative lipid transfer-like protein, non-specific LTP1 and two VAS lipid transfer-like proteins (Table S3). LTPs together with another extracellular protein, GDSL esterase/lipase At5g33370, are involved in cutin and suberin formation in plants (Edqvist et al., 2018; Hong et al., 2017). Therefore, we examined suberin accumulation in roots of F. graminearum-treated WT and HvMPK3 KO lines using specific fluorescent dye fluorol yellow 088. This histochemical analysis showed stronger suberin deposition to the root surface cell layers of non-infected HvMPK3 KO lines (Figure 9B, C, G), but not in WT plants (Figure 9A, G). The suberin histochemical staining at root surface cell layers was considerably enhanced only in HvMPK3 KO lines after F. graminearum infection (Figure 9D, E, F, G). In addition, the suberin staining also increased in the root endodermis and stele of HvMPK3 KO lines (Figure 9B, C, E, F). In contrast, histochemical staining of lignin with basic fuchsine showed an opposite tendency, and no differences in lignin staining were observed under control conditions (Figure 10A-C, G). Nevertheless, treatment with F. graminearum caused over 54,5% decrease of lignin staining in WT roots, but only 18,5-28% decrease in HvMPK3 KO lines (Figure 10D-G). These results suggest that a suberized cell wall barrier against F. graminearum invasion was built up at the surface layers of HvMPK3 KO roots.
In conclusion, proteomic analysis showed that WT exhibited reduced abundances of PR proteins, while the resistance of HvMPK3 KO lines to F. graminearum was accompanied by upregulation of PR proteins, peroxidases, LTPs, cysteine proteinases, proteins involved in suberin formation and it was corroborated by overabundances of cAPX and catalase.
Discussion
The diversified family of MAPKs lies in the core of plant defense mechanisms due to their significant role in extracellular stimuli perception. In addition, MAPK cascades are responsible for plant resistance to abiotic stresses and exert multiple developmental functions (Bigeard et al., 2015; Komis et al., 2018). Despite a significant number of MAPK-related studies in dicots like Arabidopsis, the current understanding of MAPK roles in defense of cereal crops against pathogens is still limited. In cereals, MAPKs play either positive or negative roles in immune reactions against fungal pathogens (Křenek et al., 2015; Chen et al., 2021). In rice, the OsMKK4–OsMPK3/OsMPK6 module participates in transduction of a fungal chitin elicitor signal and regulates defense responses leading to the biosynthesis of diterpenoid phytoalexins and lignin, and progression of plant cell death (Kishi-Kaboshi et al., 2010). OsMPK3 negatively regulates the expression of defense genes and the defense reactions against hemibiotrophic blast fungus Magnaporte grisea (Xiong and Yang, 2003). It was reported that this negative regulation is mediated through interaction with, and by phosphorylation of OsMPK3 via calcium-dependent protein kinase OsCPK18 (Xie et al., 2014). Wheat TaMPK3, unlike TaMPK6, is specifically activated, and its transcript levels as well as abundance are increased during compatible interactions with necrotrophic fungus Mycosphaerella graminicola (Rudd et al., 2008). This increase is directly correlated with the onset of programmed cell death (PCD) and the generation of related PCD symptoms (Rudd et al., 2008). Barley MPK3 is upregulated in response to the rust fungus Puccinia hordei inoculation, especially during effector-triggered immunity (Křenek et al., 2015). Remarkably, HvMPK3 KO lines showed attenuated response to flg22 treatment in terms of defense-related genes such as chitinases, indicating some positive role of HvMPK3 in PTI (Takáč et al., 2021).
Although Fusarium root rot is capable to cause severe yield losses, the plant defense mechanism against this disease has not been sufficiently investigated and understood yet. Role of MAPKs in cereal responses to Fusarium were not studied before. The resistance against Fusarium root rot was previously shown to be mediated by hindering the penetration of the fungi at the wheat epidermal cells (Wang et al., 2015). Our results suggest that the exclusion of F. graminearum to the extracellular space in the roots of HvMPK3 KO lines is likely caused by suberin deposition to the root surface of these lines, which is also supported by upregulation of proteins involved in suberin formation. We also observed decrease in lignin deposition, similarly as previously published for F. oxysporum treatment of flax and F. solani treatment of soybean (Lozovaya et al., 2006; Wojtasik et al., 2016).
When compared to flg22 treatment (Takáč et al., 2021), F. graminearum infection had more obvious impact on the barley proteome, especially concerning changed abundance of extracellular proteins. This might be assigned to the hemibiotrophic pathosystem of the fungus, eliciting more complex defense mechanisms compared to bacterial elicitor flg22. Hence, this proteomic analysis allowed to gain a more detailed insight into defense mechanisms occurring in HvMPK3 KO lines and suggested that WT roots were compromised in defense response, most likely due to fungus-secreted toxin effectors. Is was illustrated by the downregulation of PR17c precursor and several isoforms of cysteine proteases, representing well-known pathogen effector targets (Zhang et al., 2012, Mueller et al., 2013). Cysteine proteases are major regulators of plant defense responses (Misas-Villamil et al., 2016) and in contrast to WT, their abundance considerably increased in HvMPK3 KO lines, indicating that they might mediate resistance of these lines against F. graminearum.
Specialized infection structures called appressoria are necessary for the F. graminearum infection. It was reported on plants like wheat, rice, corn or barley attacked by different fungal pathogens, including rusts, powdery mildew and blast diseases (Boenisch and Schäfer 2011; Qiu et al., 2019). To facilitate host penetration, appressoria of Magnaporthe oryzae, generate enormous pressure up to 8.0 MPa that drive physically rupture of the host cell wall (Talbot, 2019). F. graminearum can produce multicellular appressoria called infection cushions (Boenisch and Schäfer 2011), whereas the biology of these structures is not completely known. Here, the number of formed and invading appressoria were significantly higher in the case of WT as compared to the HvMPK3 KO lines. Twenty-four hours after infection, fungal hyphae successfully invaded the epidermal root cells of WT plants, which was preceded by the dramatic ROS increase and accumulation, leading to hyphae penetration to cells. This is entirely consistent with the previously described model of hemibiotrophic and necrotrophic fungal pathosystems (Desmond et al., 2008). In contrast, the production of ROS was substantially lower in infected HvMPK3 KO plants, partly due to prevention of the interaction between the pathogen and the root surface of transgenic plants, and partly also due to their elevated antioxidant capacity mediated by cytosolic ascorbate peroxidase and catalase.
The ROS generated and released during infection can affect both counterparts, the host and the microbe (Desmond et al., 2008; Yang and Fernando, 2021). ROS are indispensable for the activation of plant defense responses as a signaling component (Lee et al., 2020) or they can cause peroxidation of lipids, oxidation of proteins, damage to nucleic acids, enzyme inhibition, and activation of PCD pathway during hypersensitive response (Camagna and Takemoto, 2018), thus leading to the inhibition of biotrophic pathogens. However, the Fusarium infection represents another example, since ROS accumulation is induced by its toxins in the host plants (Desmond et al., 2008). Fusarium infection follows the H2O2 produced by the host cells, which causes enhanced production of fungal toxins enabling hyphae penetration into the host cells (Khaledi et al., 2016).
Secretory peroxidases were one of the most affected protein groups in both HvMPK3 KO and WT lines after F. graminearum infection. All peroxidase isoforms were downregulated in WT, but three out of five peroxidases were upregulated in HvMPK3 KO lines in response to F. graminearum. Secretory peroxidases belong to class III of plant peroxidases, having a broad spectrum of activities in plants (Almagro et al., 2009; Passardi et al., 2005; Sasaki et al., 2004). Such peroxidases belong to a PR protein 9 subfamily and are supposed to reduce the progression of pathogens by a generation of structural barriers. They are bifunctional because they exert either ROS-scavenging or ROS-producing activity (Passardi et al., 2004). Peroxidase activity contributes to the ROS generated during oxidative burst (O’Brien et al., 2012). ROS produced by peroxidases may either have signaling roles or contribute to the toxic environment for the pathogen (Camejo et al., 2016).
One of the crucial features of secretory peroxidases is to mediate cross-linking of cell wall components leading to cell wall reinforcement. They may cross-link extensins leading to the formation of large oligomers (Mishler-Elmore et al., 2021). The cross-linking of ferulic acid by covalent bonds requires the presence of cell wall peroxidases. Such cross-linked ferulic acid may create intra- or inter-polysaccharide bonds leading to cell wall stiffening (Fry, 2004). Lignin formation depends on oxidative cross-linking of monolignols (Ralph et al., 2004) and expression of peroxidases was shown to correlate with lignification (Cosio et al., 2017; Sorokan et al., 2014). Similarly, suberin poly-phenolic domain assembly requires peroxidase mediated oxidative coupling reactions (Bernards et al., 2004). Moreover, transient overexpression of HvPrx40 a class III peroxidase in barley epidermis, improves resistance against Blumeria graminis f.sp. hordei pathogen (Johrde and Schweizer, 2008). It is likely that peroxidases similar to peroxidase 12 (Arabidopsis thaliana), cationic peroxidase PNC1 (Arachis hypogaea), and peroxidase HRPN (Armoracia rusticana) contribute to the formation of suberized structural barriers which exclude F. graminearum to the extracellular space in HvMPK3 KO plants. Moreover, a genomic study reported a possible link of GDSL esterase/lipases to suberin biosynthesis (Soler et al., 2007). Protein similar to GDSL esterase/lipase At5g33370 (Arabidopsis thaliana) was significantly upregulated in HvMPK3 KO plants. According to our results, HvMPK3 KO lines activated a different set of LTPs compared to WT. A protein similar to lipid transfer-like protein VAS, implicated in suberin formation in Arabidopsis (Edstam et al., 2013), was upregulated in HvMPK3 KO lines. Also, Pitzschke et al., (2014) found that AZI1 (belonging LTPs) directly interacts with AtMPK3, and confers salinity tolerance in Arabidopsis (Pitzschke et al., 2014). In summary, our results indicate that the cell wall stiffening by suberin in surface root cells might substantially contribute to the resistance of HvMPK3 KO lines against F. graminearum. This might suggest a connection between HvMPK3 and LTPs as a part of barley defense mechanism against F. graminearum infection.
Since TALEN-based knockout of HvMPK3 causes resistance to F. graminearum it provides a reliable tool for targeted biotechnological application in important cereal crop such as barley.
Experimental procedures
Cultivation of plant material
Three independent homozygous knock-out lines of barley (Hordeum vulgare L.) designated as HvMPK3 KO-A, HvMPK3 KO-B and HvMPK3 KO-D, and control WT lines (Takáč et al., 2021) were used for the experiments. Seeds were surface sterilized using 5% (v/v) sodium hypochlorite and 70% (v/v) ethanol, followed by an additional incubation with 500 mM (v/v) H2O2 overnight. After sterilization, the seeds were imbibed and stratified on plates with agar-solidified nitrogen-free Fahräeus medium (Fåhraeus, 1957) at 4°C for synchronous germination (Perrine-Walker et al., 2007). The stratified sterile seeds on plates were transferred to phytotron (Weiss-Gallenkamp, Loughborough, United Kingdom) and cultivated at 21°C for 16 h in the light (day), and 8 h in darkness (night), 70% relative humidity with light levels of 150 μmol.m-2.s-1 provided by cool white fluorescent tubes (Philips Master tl-d 36W/840).
Preparation of fluorescent F. graminearum
Mycelia of F. graminearum were cultivated in Mung bean soup (140 rpm, 20°C, 3 days), subsequently filtered through glass wool and centrifuged (300 rpm, room temperature, 5 min). Conidia obtained as such, were resuspended in sterile distilled water and inoculated on yeast extract-peptone-dextrose (YPD) medium. After overnight cultivation (180 rpm, 30°C), the mycelium was filtered through Miracloth (Merck) and washed with sterile distilled water, and 1.2 M KCl. For protoplast isolation, mycelia were cultivated in a solution containing 250 mg driselase, 1 mg chitinase, and 100 mg lysing enzyme from Trichoderma harzianum (all from Sigma-Aldrich). After 3 h (90 rpm, 30 °C), the enzyme-protoplast solution was run over the frit and the flow-through was centrifuged (3000 rpm, 4 °C, 10 min). Obtained pellet was washed with 1.2 M KCl, then resuspended in STC buffer (1.2 M sorbitol, 50 mM CaCl2).
Plasmid (5 μg of pNDN-OGG containing GFP; Schumacher, 2012) was mixed with 2x STC buffer, F. graminearum protoplasts, and 50% (w/v) polyethylene glycol (PEG) 4000. After 25 min on ice, 50% (w/v) PEG 4000 was added. Next, the mixture was incubated 10 min at RT, followed by addition of 1x STC buffer. Prepared samples were mixed with Regeneration agar containing nourseothricin (300 μg.ml-1, Jena BioScience, pNDN-OGG) or hygromycin (100 μg.ml-1, InvivoGen, pNDH-OCT). After 5 days at 20 °C, nourseothricin- and hygromycin-resistant transformants were transferred into complete medium (CM) agar plates containing antibiotics in the final concentration of 100 μg.ml-1.
Genomic DNA was extracted from lyophilized F. graminearum mycelia according to Cenis, (1992). Diagnostic PCR was performed using GoTaq G2 Flexi DNA polymerase according to a manufacturer’s protocol (Promega). Primers were designed using Primer3 0.4.0 software and synthesized by Sigma-Aldrich. The presence of pNDN-OGG or pNDH-OCT in F. graminearum transformants was checked with primers PoliC_fw (5’-CCCGGAAACTCAGTCTCCTT-3’) and TgluC_rev (5’-GTCTTCCGCTAAAACACCCC-3’) (1,295 bp) or PoliC_fw and Ttub_rev (5’-GAGGTGTGAGCATGGAAGTGATG-3’) (1,593 bp), respectively.
Homokaryots of OE:GFP transformants were obtained by single spore isolation. Briefly, fungi were cultivated in Mung bean soup (20°C, 140 rpm) and after 5 days, the cultures were filtered through glass wool and centrifuged (10 min, 3000 rpm, 4°C). Obtained conidia were resuspended in sterile distilled water and sprayed onto CM agar plates containing appropriate antibiotics in the final concentration of 100 μg/ml.
Infection of barley plants with Fusarium
Fusarium grown in potato dextrose agar (PDA) medium for 2 weeks at 28 °C was washed with sterile distilled water containing 0.01% (v/v) Tween and filtered through sterile miracloth to obtain conidial spores (~1 × 105 spores.ml-1) (Erayman et al., 2015). Five days old barley seedlings were infected by dipping in freshly prepared conidial suspension (sterile distilled water containing 0.01% (v/v) Tween was used as control) and put back to the media for incubation for further experiments. Microscopic, biochemical and proteomic analyses were performed on control and Fusarium-treated plants 24h and 48h post-infection.
Phenotypic and microscopic analysis
For microscopic analysis, the roots of both WT and HvMPK3 KO plants 24h and 48h after inoculation with GFP-labeled F. graminearum were used. Infected plants were grown in microscopic growth chambers (Nunc™ Lab-Tek™ II Chambered Coverglass (Thermofischer scientific, USA)) and observed using 488 nm excitation for GFP and the 517-527 nm emission using confocal laser scanning microscope LSM710 (Carl Zeiss, Germany). The images were color-coded for the analysis of the penetration of the infection hyphae in root epidermal cells using appropriate Zen Blue software function. The numbers of appressoria and penetration events were counted from the images and were compared among the lines by one-way ANOVA test at significance level at p < 0.05. Roots were stained with PI (1 mg.ml-1) for 10 min, washed with liquid Fahräeus medium for 1 min, and observed under spinning disk microscope (Carl Zeiss, Germany) equipped with Plan-Apochromat 20×/0.8 NA (Carl Zeiss, Germany) at 488 nm (for GFP) and 561 nm (for PI) with emission filters BP525/50 (for GFP) and BP629/62 (for PI). Image post-processing was done using ZEN 2014 software (Carl Zeiss, Germany), and the percentage of the dead cells were calculated from the processed images using the formulae (Number of dead cells / total number of cells in the field) × 100. To document root length, both treated and untreated plants were photographed by camera once per day for 10 days after the root inoculation with GFP-labeled F. graminearum spores. The experiment was performed in three biological replicates. Ten plants were used per replicate and treatment for each line. The statistical significance of treatment vs. control was deemed by one-way ANOVA test at significance level at p < 0.05.
ROS staining
Control plants and treated plants from individual lines were used for ROS analysis 24 hrs postinoculation with GFP-labeled F. graminearum using the modified protocol of Kristiansen et al., (2009). Mock-treated and infected roots of WT and HvMPK3 KO plants were incubated in 30 μM 2’,7’-dichlorodihydrofluorescein diacetate (CM-H2DCFDA; Cat no. C6827, Invitrogen™, USA) and PI (1 mg.ml-1) diluted in Fahräeus medium. Seedlings in microscopic chambers were stained by perfusion, followed by incubation in darkness for 15 min. After incubation the residual stain was washed using sterile Fahräeus medium for 3 times using perfusion. After washing, the signal excited at 488 nm for ROS/GFP and 543 nm for PI was recorded at the emission range 517-527 nm for ROS/GFP and 610-625 nm (For PI) using confocal laser scanning microscope LSM710 (Carl Zeiss, Germany). The imaging of ROS accumulation in whole root was performed after staining with H2DCFDA using epifluorescence microscope Zeiss Axio Imager M2 (Carl Zeiss, Germany) with settings for GFP excitation and emission. ROS levels in roots were analyzed semi-quantitatively from the images with ZEN software (Carl Zeiss, Germany).
Histochemical detection of suberin and lignin
Root segments (first 25 mm from the root apex) of WT and HvMPK3 KO plants 24 h after inoculation with F. graminearum were used for suberin and lignin histochemical detection. Suberin and lignin staining was carried out according to a previously published protocol (Sexauer et al., 2021; Ursache et al., 2018). Freehand cross-sections of fixed and cleared root segments were made in the root region between 15 and 25 mm from the root apex and stained with 0,01% (v/v) Fluorol Yellow 088 in absolute ethanol (stock solution: 1% (w/v) Fluorol Yellow in DMSO) for 30 min. For lignin detection root sections were stained with 0,2% (w/v) Basic Fuchsin in ClearSee solution for overnight. Stained cross-sections of roots were observed under confocal laser scanning microscope LSM710 (Carl Zeiss, Germany) using following settings for suberin: excitation: 488 nm, detection: 500-550 nm and for lignin: excitation: 561 nm, detection: 600-650. Measurement of fluorescence intensity and post-processing were done using ZEN 2014 software (Carl Zeiss, Germany). Fluorescence mean intensities were calculated from a minimum of three biological replicates.
Western Blot Analysis
Samples were prepared and western blots performed as described before (Takáč et al., 2021). Polyvinylidene difluoride (PVDF) membranes were incubated with rabbit primary anti-L-ascorbate peroxidase antibody (#AS 08 368), diluted 1:2000, or with rabbit primary anticatalase antibody (#AS 09 501), diluted 1:2000, or with rabbit primary anti-PR-1 antibody (#AS 10 687, all from Agrisera, Sweden), diluted 1:2000, all in 1% (w/v) BSA in TBS-T at 4°C overnight. Membranes were washed thoroughly and subsequently incubated at room temperature with corresponding HRP-conjugated secondary antibody (Thermo Fisher Scientific), diluted 1:5000 in 1% (w/v) BSA in TBS-T for 1.5 h. Following three washing steps, membranes were incubated with commercial Clarity Western ECL Substrate (BioRad, Hercules, CA, United States) and documented in a ChemiDoc MP imaging system (BioRad). Experiments were performed in three biological replicates, and the statistical significance was evaluated using Student’s t test (p< 0.05).
Analysis of peroxidase activities and spectrophotometric measurement of ROS
Specific activities of peroxidases were analyzed using staining on native PAGE gels as published previously (Takáč et al., 2016). ROS levels were estimated in water extracts using xylenol orange assay as described in Takáč et al., (2014). The analyses were carried out in three biological replicates on control and Fusarium-treated roots of WT and HvMPK3 KO seedlings 24h post-inoculation. The statistical significance was evaluated using Student’s t test (p< 0.05).
Relative protein quantitation using nano-Liquid Chromatography-Tandem Mass Spectrometry Analysis (nLC-MSMS)
Barley WT, and HvMPK3 KO lines A, B, and D were used for proteomic study. Root parts from four plants of each line in control and treatment variants were pooled into one sample. Each sample was analyzed in two biological replicates, and the final differential proteomes of WT and HvMPK3 KO lines were obtained by comparing 6 biological replicates of Fusarium-treated lines vs. corresponding mock controls. Proteins were extracted using phenol extraction and methanol/ammonium acetate precipitation, as described previously (Takáč et al., 2017). Total of 50 μg of proteins dissolved in 50 μl of 6 M urea were subjected to in-solution trypsin digestion.
The peptides were then cleaned on C18 cartridges (Bond Elut C18; Agilent Technologies, Santa Clara, CA), dried using SpeedVac, and utilized for nLC-MS/MS (Takáč et al., 2017).
Two micrograms of protein tryptic digest were analyzed as published previously (Takáč et al., 2021), using the Ultimate 3000 nano-LC system and LTQ-Ortbitrap Velos mass spectrometer (both Thermo Fisher Scientific). All raw and results files were deposited to a publicly accessible database (see Data Availability Statement for details). Protein identification and label-free quantification was performed by the Proteome Discoverer 2.1 (Thermo Fisher Scientific), and an in-house script based on precursor ion intensities, as described before (Takáč et al., 2021). Statistically significant results were filtered with ANOVA p ≤ 0.05, applied to proteins exhibiting the fold change ≥ 1.5.
Proteins identified by single peptide were excluded from the results. Proteins present in all six replicates corresponding to the control proteome, and absent in all the treated replicates were considered unique for the control proteome, and vice versa.
The differential proteomes were evaluated by gene onthology (GO) annotation analysis and screening of protein domains using OmixBox Functional analysis module, as specified previously (Takáč et al., 2021).
Supporting Information
Figure S1. Colonization of roots in barley plants of WT and HvMPK3 KO lines by Fusarium graminearum mycelia 10 days after inoculation with spores.
Figure S2. Characterization of distribution and growth pattern of F. graminearum hyphae in regard of the root longitudinal axis. (A-D) Cytospectre graphs showing a distribution pattern of fluorescent hyphae on the root surface of WT (A), HvMPK3 KO-A (B), HvMPK3 KO-B (C) and HvMPK3 KO-D (D) lines. (E) Quantitative assessment of anisotropy characterizing angular distribution of growing hyphae. (F) Quantitative assessment of fluorescence distribution skewness of hyphae on the surface of measured root area. Data are represented as mean ± SD; n = 10 images/line, error bars represent standard deviations. Different lowercase letters above the error bars represent statistical significance according to one-way ANOVA and subsequent LSD test at p value < 0.05.
Figure S3. Comparison of root phenotypes in barely WT and HvMPK3 KO lines between nontreated and Fusarium graminearum-treated plants 10 days after inoculation with spores.
Figure S4. Fluorescence visualization of F. graminearum hyphae distribution on root surface of barley wild type (WT) and HvMPK3 KO lines 48h after inoculation with spores. (A, B) WT root. (C, D) Root of HvMPK3 KO-A line. (E, F) Root of HvMPK3 KO-B line. (G, H) Root of HvMPK3 KO-D line. Orthogonal view depicting root surface cell layers (A, C, E, G) and maximum intensity projection of the root surface (B, D, F, H). Appressoria are indicated by asterisk and infection hyphae penetrating root epidermal cells are indicated by white arrows. Scale bars: 20 μm (A-H).
Figure S5. Concentration of hydrogen peroxide in roots of WT and HvMPK3 KO lines 24h after inoculation with Fusarium graminearum spores as examined by Xylenol orange assay.
Figure S6. Comparison of gene onthology annotation according to biological processes, carried out in differential proteomes of WT and HvMPK3 KO plants treated by Fusarium graminearum for 24h.
Figure S7. Comparison of gene onthology annotation according to cellular compartments, carried out in differential proteomes of WT and HvMPK3 KO plants treated by Fusarium graminearum for 24h.
Figure S8. Full scan of the entire original gel whose fragments are presented in Figure 8A.
Figure S9. Full scan of the entire original blot (A) and loading control (B) whose fragments are presented in Figure 8C.
Figure S10. Full scan of the entire original blot (A) and loading control (B) whose fragments are presented in Figure 8E.
Figure S11. Full scan of the entire original blot (A) and loading control (B) whose fragments are presented in Figure 8G. The highlighted regions show sections presented in Figure 8G.
Table S1. Summary and quantification details of differentially regulated proteins found in roots of wild type plants 24h after the treatment with Fusarium graminearum
Table S2. Summary and quantification details of differentially regulated proteins found in roots of HvMPK3 KO plants 24h after the treatment with Fusarium graminearum
Table S3. Proteins predicted to be localized in extracellular space.
Funding
This work was funded by ERDF project Plants as a tool for sustainable global development (CZ.02.1.01/0.0/0.0/16_019/0000827), and NIH MS-IDeA Network of Biomedical Research Excellence award 5P20GM103476-19. The mass spectrometry proteomics analysis was performed at the Institute for Genomics, Biocomputing and Biotechnology, Mississippi State University, with partial support from Mississippi Agricultural and Forestry Experiment Station.
Authors contributions
JB, PV and MO conducted phenotypic and microscopic documentation and analysis. PV, TT and TP performed proteomics analysis. PV and PM conducted immunoblot analysis, native electrophoresis and spectrophotometric measurements. OŠ performed suberin and lignin staining and related microscopic observations and evaluation. PK selected HvMPK3 KO and WT barley lines. MK performed Fusarium transformation. JB, PV, MO, TT, GK and JŠ drafted the manuscript with input from all co-authors. JŠ conceived and supervised the project, provided infrastructure and secured funding.
Data availability
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD029964”.
Reviewer account details: Username: reviewer_pxd029964{at}ebi.ac.uk, Password: t4gfjKBY
Acknowledgement
We would like to thank Lena Studt from Department of Applied Genetics and Cell Biology, University of Natural Resources and Life Sciences, Vienna for helping us with F. graminearum transformation. We also thank technicians Petra Trčková, Pavlína Floková, Katarína Takáčová and Monika Vadovičová for their expert technical help in all stages of the presented work.