A chitin deacetylase Ps CDA2 from Puccinia striiformis f. sp. tritici confers disease pathogenicity by suppressing chitin‐triggered immunity in wheat

Abstract

Plants have the ability to recognize the essential chitin molecule present in the fungal cell wall, which stimulates the immune response. Phytopathogenic fungi have developed various strategies to inhibit the chitin‐triggered immune response. Here, we identified a chitin deacetylase of Puccinia striiformis f. sp. tritici (Pst), known as PsCDA2, that was induced during the initial invasion of wheat and acted as an inhibitor of plant cell death. Knockdown of PsCDA2 in wheat enhanced its resistance against Pst, highlighting the significance of PsCDA2 in the host–pathogen interaction. Moreover, PsCDA2 can protect Pst urediniospores from being damaged by host chitinase in vitro. PsCDA2 also suppressed the basal chitin‐induced plant immune response, including the accumulation of callose and the expression of defence genes. Overall, our results demonstrate that Pst secretes PsCDA2 as a chitin deacetylase involved in establishing infection and modifying the acetyl group to prevent the breakdown of chitin in the cell wall by host endogenous chitinases. Our research unveils a mechanism by which the fungus suppresses plant immunity, further contributing to the understanding of wheat stripe rust control. This information could have significant implications for the development of suitable strategies for protecting crops against the devastating effects of this disease.

The chitin deacetylase PsCDA2 is able to alter the type of chitin in the cell wall of Puccinia striiformis f. sp. tritici and enhances its virulence by suppressing chitin‐triggered immunity in wheat.

1. INTRODUCTION

During plant–pathogen interactions, the plant immune system typically recognizes pathogen‐associated pattern molecules (PAMPs) by pattern recognition receptors (PRRs) to initiate pattern‐triggered immunity (PTI) (Dodds & Rathjen, 2010). PAMPs are usually conserved molecules specific to pathogens (Zhang & Zhou, 2010). For fungi, the polysaccharides from their cell wall provide the most abundant and efficient source of PAMPs (Gong et al., 2020; Liu et al., 2013; Oliveira‐Garcia & Valent, 2015). As one of the most abundant polysaccharides, chitin is a long‐linear homopolymer of β‐1,4‐linked N‐acetylglucosamine (GlcNAc), which is crucial for maintaining cell wall integrity (Liu et al., 2016; Thomma et al., 2011). It is believed that plants secrete chitinases into the apoplast region to disrupt the fungal cell wall during fungal attack, leading to release of chitin oligomers that act as PAMPs recognized by PRRs, thereby initiating PTI (Lee et al., 2014; Shimizu et al., 2010). Chitin elicitor‐binding protein and the chitin elicitor receptor kinase, which are localized on the cell surface, possess a lysin motif (LysM) structural domain and are capable of sensing chitin oligomers, consequently triggering the immune response (Desaki et al., 2018; Miya et al., 2007). The typical PTI response initiates activation of mitogen‐activated protein kinase cascades, production of reactive oxygen species (ROS), callose deposition and expression of immune response genes (Kinoshita & Seki, 2014; Li et al., 2016). The acetyl group plays a significant role in binding chitin oligomers to the receptor (Liu et al., 2012; Zipfel, 2014).

To successfully establish infection in plants, fungal pathogens produce virulence effectors to suppress or evade the host plant's PTI (Wang et al., 2022). Suppression of host recognition is a key infection strategy used by various pathogens. Fungi employ multiple mechanisms to dampen the immune response triggered by chitin oligomers, such as deacetylation, binding or degradation (Sanchez‐Vallet et al., 2015). The first strategy is to protect PAMPs from recognition by host receptors. For example, Verticillium dahliae produces VdPDA1, a chitin deacetylase, that modifies chitin oligomers in cotton (Gao et al., 2019). Pst_13661 from Puccinia striiformis f. sp. tritici (Pst) can modify the fungal cell wall through deacetylation to prevent recognition. Similar effectors have been found in other fungi, including chitin deacetylases (CDAs) from Ustilago maydis and PoCda7 from Pyricularia oryzae (Dai et al., 2021; Rizzi et al., 2021). Another strategy is to produce chitin‐binding proteins that can bind long‐chain chitin or chitin oligomers, such as Avr4 from Cladosporium fulvum. Avr4 protects chitin in the fungal cell wall from plant chitinases. ECP6, a LysM effector from C. fulvum, can bind to chitin oligomers released by plant chitinases, thereby preventing these oligomers from being recognized by chitin receptors (Bolton et al., 2008; de Jonge et al., 2010; van den Burg et al., 2006). In addition, effectors with chitinase activity degrade chitin oligomers and thus alter the perception of chitin by the host plant, such as Podosphaera effector candidate (PEC) proteins from Podosphaera xanthii (Martinez‐Cruz et al., 2021). These successful pathogens can evade host recognition and suppress the PTI response.

Wheat stripe rust, caused by Pst, is one of the most important diseases of wheat (Schwessinger, 2017). As an obligately parasitic filamentous fungus, Pst can only exist in the living host cells, where it forms haustoria, specialized infection structures vital for its development and survival in plant tissues (Kang et al., 2017). However, when Pst attacks the host, the haustoria are exposed to the apoplast containing plant defence factors, such as enzymes and receptors, that release and recognize PAMPs of the fungal cell wall, respectively. In response, Pst secrets virulence factors known as effectors through its haustoria and hyphae into the host cells to evade or disrupt this surveillance mechanism (Mapuranga et al., 2022; Xu, Tang, et al., 2020).

Little is known about the mechanism of action of CDA in wheat–Pst interactions. Previous studies have experimentally validated Pst_13661 as the first effector protein with CDA activity (Xu, Wang, et al., 2020). In this study, we conducted genomic analysis of Pst and identified Pst_02914 with CDA activity. This second experimentally validated CDA from Pst was named PsCDA2. To investigate the role of PsCDA2 in the pathogenesis of Pst, we employed various methods, including transient gene expression, host‐induced gene silencing (HIGS), protein expression and microscale thermophoresis (MST) analysis. Our results show that PsCDA2 is a CDA that can bind to chitin in the cell wall, preventing its breakdown by chitinases. This study is therefore essential for understanding the pathogenesis of wheat stripe rust and developing new and durable strategies for controlling stripe rust.

2. RESULTS

2.1. Identification of PsCDA2

In our previous analysis, we conducted a BLAST search on the Pst CYR32 genome using 10 chitin effectors reported from different fungi, resulting in the identification of 49 chitin‐associated proteins. To understand the evolution and sequence diversity of chitin‐related proteins in Pst, we constructed a phylogenetic tree that used ClustalW alignment of amino acid sequences belonging to all of the 49 candidate effectors, along with 10 previously reported effectors using the maximum‐likelihood method. These proteins were divided into three categories: chitinases, chitin‐binding protein and CDAs (Figure S1a). We obtained the expression profiles for all 49 effectors from publicly available RNA sequencing (RNA‐seq) data gathered from two groups of wheat–rust interaction combination named 32R and 32S (Table S1). Five genes were found to be expressed during disease development (Figure S1b). Intriguingly, all five of these genes belong to the CDA family, which suggests that this family plays a vital role in the Pst infection of wheat. An alignment of the five mentioned members with PSTCY_13661 of Pst CYR32, VdPDA1 of V. dahliae and CDA of Colletotrichum lindemuthianum revealed high motif similarity. All five effectors contained five motifs in the structural domain of the polysaccharide deacetylase that were conserved when they were compared to the characteristic polysaccharide deacetylases or CDAs of other fungal species (Figure 1a). These eight proteins shared numerous structural similarities, with five conserved motifs associated with CDAs, namely, motif 1 (TFDD), motif 2 (HTWSH), motif 3 (RPPY), motif 4 (DSGD) and motif 5 (LNH) (Aragunde et al., 2018; Grifoll‐Romero et al., 2018). PsCDA was predicted to be a noneffector by EffectorP v. 3.0. We also predicted the presence of signal peptides in the five candidate genes from CYR32 mentioned above. We discovered that only PsCDA2 and Pst_14974 were predicted to contain a signal peptide (Figure S2). Because effector proteins usually have fewer than 300 amino acids, we decided to select PsCDA2 as the object of our study.

FIGURE 1.

Multiple protein sequences alignment of PsCDA2. (a) Alignment of five candidate effectors with VdPDA1 from Verticillium dahliae, Pst_13661 from Puccinia striiformis f. sp. tritici (Pst) and CDA from Colletotrichum lindemuthianum. GenBank accession nos. V. dahliae PDA1 (this work, NW_009276934), Pst_13661 (KNE92947), C. lindemuthianum CDA (AY633657). (b) PsCDA2 is induced in wheat leaves inoculated with virulent Pst pathogen CYR32. CK, mock‐inoculated control; hpi, hours postinoculation. Data shown are the means ± standard deviation calculated from three independent biological replications (ns p > 0.05, **p < 0.01).

To further determine the expression changes of PsCDA2 during Pst infestation in wheat, we assessed the transcript levels of PsCDA2 in wheat leaves inoculated with Pst. Urediniospores of Pst CYR32 served as a control. Our results showed considerable upregulation of PsCDA2 transcript levels at early stages of Pst CYR32 infection, at 12, 24 and 48 hours postinoculation (hpi). Of note, the highest upregulation of PsCDA2 occurred at 12 hpi, which coincides with the establishment of primary haustoria (Figure 1b).

2.2. PsCDA2 is secreted into the apoplast

To ascertain whether or not the signal peptide of PsCDA2 functions as a secretory signal, we employed the yeast secretion system. Like the positive control TaSBT1.7SP, the yeast strain YTK12 transformed with the pSUC2 vector inserted into PsCDA2SP grew well on the YPRAA medium, whereas yeast strain YTK12 with empty vectors did not grow on this medium. We investigated the enzymatic activity of the secreted convertase in the medium via a colourimetric reaction. The presence of transformase led to the reduction of 2,3,5‐triphenyltetrazolium chloride (TTC) to the insoluble red 1,3,5‐triphenylformazan (TPF), indicating that transformants of both TaSBT1.7SP and PsCDA2SP secreted transformase into the medium (Figure 2a).

FIGURE 2.

Validation of the secretory function of PsCDA2 signalling peptide. (a) Yeast secretion system confirmed the secretory function of PsCDA2 signal peptide. Yeast YTK12 strain carrying empty vector, PsCDA2SP and TaSBT1.7SP was grown on CMD−W and YPRAA plates, respectively. Enzymatic activity of the convertase was assayed by reducing 2,3,5‐triphenyltetrazolium chloride (TTC) to insoluble red 1,3,4‐triphenylformazan (TPF). The negative control was a YTK12 strain carrying an empty vector and the positive control was a YTK12 strain carrying a TaSBT1.7 signal peptide. (b) Transiently expressed PsCDA2:YFP localized in the apoplast of Nicotiana benthamiana. A construct carrying CD3‐1007 (AtPIP2A:mCherry fusion) was used as a marker for plasma membrane (PM) localization. Plasmolysis was induced by 200 mM NaCl. Arrows represent plasmolysis regions. Bar = 20 μm. (c) The western blot analysis on apoplastic fluid isolated from leaves co‐expressing PsCDA2:YFP and PM marker. Staining by Ponceau S to show equal loading of total protein in the immunoblot. AF, apoplastic fluid; TCE, total cell extracts.

Due to the presence of signal peptides in the protein, we used the full length of PsCDA2 fused with yellow fluorescent protein (YFP) and detected the presence of PsCDA2 in the apoplast after NaCl‐induced plasmolysis. Remarkably, the signals for PsCDA2:YFP were observed in the apoplast, while the plasma membrane (PM) marker was solely restricted to the PMs after plasmolysis (Figure 2b). The apoplastic signal for PsCDA2 was confirmed by western blot analysis. The PM marker was only present in the total cell extracts (TCE) from Nicotiana benthamiana. Notably, using anti‐green fluorescent protein (GFP) antibodies, the presence of PsCDA2:YFP in apoplastic fluid (AF) and TCE from N. benthamiana was detectable, consistent with the observations made using confocal microscopy (Figure 2c). These results suggest that PsCDA2 is secreted into the apoplast region.

2.3. PsCDA2 suppresses Bax‐induced cell death in N. benthamiana

To test whether PsCDA2 has a potential regulatory role in suppressing programmed cell death (PCD), PsCDA2 was transiently overexpressed in N. benthamiana leaves via agro‐infiltration. The negative control YFP did not suppress cell death. However, like the positive control, Pst_13661 from Pst, PsCDA2 was able to inhibit Bax‐induced PCD in N. benthamiana (Figure 3a). These results were consistent across multiple biological replicates (Figure S3). Furthermore, western blot analysis demonstrated that PsCDA2 and Bax protein were expressed in N. benthamiana. Western blot analysis with anti‐GFP and anti‐FLAG antibodies confirmed the expression of YFP (27 kDa), Pst_13661:YFP (55 kDa), PsCDA2:YFP (55 kDa) and Bax (22 kDa) in vivo (Figure 3b). The intensity of the hypersensitive response (HR) in leaves expressing YFP, Pst_13661 and PsCDA2 alone was almost zero. As the negative control, the intensity of the HR of leaves co‐expressing Bax and YFP reached 4. Like the positive control, the leaves co‐expressing Bax and Pst_13661, the leaves co‐expressing Bax and PsCDA2 also showed no HR symptoms in N. benthamiana (Figure 3c). The ion leakage of plants co‐expressing PsCDA2 and Bax was lower than that of control plants co‐expressing YFP and Bax (Figure 3d). These results suggest that PsCDA2 possesses the ability to suppress plant defence responses.

FIGURE 3.

Transient expression of PsCDA2 suppresses Bax‐induced cell death in Nicotiana benthamiana. (a) The recombinant vectors (pCNF3:YFP, pCNF3:PsCDA2:YFP, pCNF3:Pst_13661:YFP, pGR106:Bax) were transformed into Agrobacterium tumefaciens GV3101 for transient expression. Yellow fluorescent protein (YFP) was used as a negative control. Images were taken 4 days after inoculation with Agrobacterium carrying the corresponding vector. The leaves were photographed under visible (left) and UV (right) light. (b) Total N. benthamiana proteins were extracted 72 h after Agrobacterium infiltration and the proteins YFP (27 kDa), Pst_13661:YFP (55 kDa) and PsCDA2:YFP (55 kDa) were detected by western blot with anti‐GFP antibody. Bax (22 kDa) was detected with anti‐FLAG antibody. Staining by Ponceau S to show equal loading of total protein in the immunoblot. (c) Hypersensitive response (HR) index scored at 6 days postinfiltration (dpi) is plotted. Data shown are the means ± standard deviation calculated from three independent biological replications. (d) N. benthamiana leaves were infiltrated with Agrobacterium cultures expressing YFP, Pst_13661, PsCDA2 alone or in combination with cultures expressing Bax. Leaf discs were excised and assayed for electrolyte leakage at 5 dpi. Data represent the means ± standard errors of five replicates, each consisting of four plants that received the same treatment. Different letters indicate significant differences at p < 0.05.

2.4. Knocking down Ps CDA2 impairs the pathogenicity of Pst

To analyse the role played by PsCDA2 in the pathogenic process of Pst, we used the barley stripe mosaic virus (BSMV)‐mediated HIGS system to knock down PsCDA2. Approximately 12 days after BSMV inoculation, chlorotic mosaic symptoms were observed in all wheat samples. The functionality of the BSMV‐mediated silencing system was confirmed by the significant photobleaching of wheat samples inoculated with BSMV:TaPDS (Figure 4a). Reverse transcription‐quantitative PCR (RT‐qPCR) analysis revealed that the transcript levels of PsCDA2 were successfully suppressed below 30% at 24, 48 and 72 hpi in leaves of plants inoculated with BSMV:PsCDA2 (Figure 4c). Fewer urediniospore pustules were produced on leaves inoculated with BSMV:PsCDA2 in comparison to control plants inoculated with BSMV:γ (mock, Figure 4b), and this phenotype was consistent throughout the leaves (Figure S4). Furthermore, the infection area was reduced in plants inoculated with BSMV:PsCDA2 as compared to the mock‐inoculated plants (Figure 4d). The fungal biomass was significantly decreased in plants in which PsCDA2 was knocked down, as evidenced by the results obtained at 120 hpi (Figure 4e). These results suggest that silencing of PsCDA2 hinders infection and development of Pst.

FIGURE 4.

BSMV‐mediated silencing of PsCDA2 in wheat. (a) Mild chlorotic mosaic symptoms were observed on fourth leaves inoculated with BSMV:TaPDS at 12 days postinoculation (dpi). Bar = 0.15 cm. (b) Disease symptoms were observed at 14 dpi on fourth leaves of wheat plants that were inoculated with avirulent Puccinia striiformis f. sp. tritici (Pst) pathogen CYR32. (c) Relative transcript levels of PsCDA2 assayed in knocked‐down wheat leaf after inoculation with CYR32. Asterisks indicate significant difference in samples with PsCDA2 silenced by host‐induced gene silencing in comparison with the control at 24, 48, and 72 h postinoculation (hpi) (****p < 0.0001). (d) The development of Pst was assessed as infection areas at 14 days after inoculation with CYR32. Standard deviation and means were calculated from three independent biological replications. Asterisks indicate significant differences (****p < 0.0001; ns, not significant, p > 0.05). (e) Pst biomass was estimated by quantitative PCR at 120 hpi in the silenced plants. Means and standard deviations were calculated from three independent replicates. Asterisks indicate significant differences (**p < 0.01; ns, not significant, p > 0.05). TaEF‐1α and PstEF‐1α were used to normalize RNA levels in wheat and Pst, respectively.

2.5. PsCDA2 has an affinity for chitin and chitooctaose with six GlcNAc moieties (A6)

Due to the CDA structural domain of PsCDA2, we tested its ability to bind polysaccharides by MST assay. Four polysaccharides, namely chitin, A6, cellulose, and chitosan, were employed to evaluate the binding capacity of PsCDA2. Our results revealed that PsCDA2 has a strong affinity for chitin and A6 (K _d = 15.99 and 3.80 μM, respectively; Figure 5a). Capillary scanning showed compound‐induced changes in protein fluorescence (Figure S5). Moreover, the binding capacity of PsCDA2 to chitohexaose was investigated, revealing only minimal binding ability to cellulose and chitosan.

FIGURE 5.

PsCDA2 has a high affinity for chitin and chitooctaose with six GlcNAc moieties (A6), and possesses chitin deacetylase activity. (a) Microscale thermophoresis assays show that PsCDA2 with chitin, A6, cellulose, and chitosan in comparison with their affinities. The recombinant proteins were contained in NT standard capillaries. The solid curve is the fit of the data points to the standard K _d‐fit function. (b) The deacetylase activity of PsCDA2 (100 μg/mL) was measured using the chitin deacetylase ELISA kit. The graph on the left shows the colour change of the substrate solution in the presence of PsCDA2 protein and standard samples. Green fluorescent protein (GFP) is the negative control. The standard samples are 0, 50, 100, 200, 400 and 800 U/L. The graph on the right shows the standard curve of deacetylase activity according to the dilution of the standard samples. The blue circles indicate PsCDA2.

To further confirm the function of PsCDA2, its deacetylase activity was assessed using a chitin deacetylase ELISA kit, which revealed a colour change in the substrate solution when PsCDA2 was present (Figure 5b). These results indicate that PsCDA2 may have a role as a deacetylase, binding to the fungal cell wall to enhance pathogenicity.

2.6. PsCDA2 can protect Pst urediniospores germination from chitinase

To determine the protective role of PsCDA2 against chitinase‐mediated hydrolysis of fungal mycelia, a spore protection assay was conducted. The crude extract of wheat chitinase was employed to inhibit the germination of urediniospores by breaking down the fungal cell wall. In the absence of any treatment, the germination rate of spores was 81.5%, indicating normal viability (Figure 6a). Conversely, the presence of chitinase completely prevented Pst germination (Figure 6b). However, interestingly, when both PsCDA2 and chitinase were present, the urediniospores of Pst were still able to germinate at a rate of 55.3% (see Figure 6c). PsCDA2 provided significant protection to Pst against chitinase degradation (Figure 6d). Due to the substrate specificity of CDA, we speculate that PsCDA2 deacetylates the chitin in the Pst cell wall, thereby preventing its breakdown by endogenous plant chitinases.

FIGURE 6.

PsCDA2 protects the germination of Puccinia striiformis f. sp. tritici (Pst) CYR32 urediniospores from chitinase action. Urediniospores were subjected to dark conditions at 22–25°C for 12 h. (a) Sterile water, (b) wheat plastid exosome crude extract, (c) wheat plastid exosome crude extract and PsCDA2. (d) The germination rate of urediniospores was significantly different under different treatments. Data shown are the means ± standard deviation calculated from three independent biological replications. Asterisks indicate significant difference compared to the sample of negative control (CK) (**p < 0.01; ****p < 0.0001; ns, not significant, p > 0.05).

2.7. PsCDA2 suppresses chitin‐induced plant immunity

To investigate whether PsCDA2 can suppress chitin‐triggered plant immunity, PsCDA2 and YFP were transiently expressed in N. benthamiana. The leaves were treated with 200 μg chitin/mL at 48 h postinoculation with Agrobacterium, and at 24 h postinfiltration with chitin. Aniline blue staining showed less callose accumulation in leaves expressing PsCDA2 than expressing YFP alone in N. benthamiana (Figure 7a). Application of double‐deionized water as a control resulted in almost no callose accumulation (Figure S6). Compared with YFP control, PsCDA2 inhibited chitin‐induced callose accumulation (Figure 7b). Furthermore, staining with 3,3′‐diaminobenzidine (DAB) and nitroblue tetrazolium (NBT) indicated that the accumulation of H₂O₂ decreased in leaves expressing PsCDA2 when compared to the YFP control (Figure 7c). We also assessed the expression of marker genes for plant immunity in leaves expressing PsCDA2. The transcript levels of NbPR1, NbPR2 and NbRBOHD were found to be fourfold, twofold and twofold lower, respectively, in leaves expressing PsCDA2 compared to control plants expressing YFP (Figure 7d). These results indicate that PsCDA2 suppresses chitin‐induced plant immunity.

FIGURE 7.

PsCDA2 suppresses chitin‐induced plant immunity. (a) Callose deposition induced by 200 μg/mL chitin in Nicotiana benthamiana transiently expressing PsCDA2 or yellow fluorescent protein (YFP) alone. Images were obtained at 24 h after infiltration by staining with aniline blue. Bar = 200 μm. (b) PsCDA2 suppresses callose spots in N. benthamiana. The number of callose spots was assessed with ImageJ software. Mean and standard deviations were calculated from three biological replicates. Asterisks mark significant difference based on Student's t test (**p < 0.01). (c) 3,3′‐diaminobenzidine (DAB) and nitroblue tetrazolium (NBT) staining in N. benthamiana leaves expressing yellow fluorescent protein (YFP) or PsCDA2. (d) Expression levels of NbPR1, NbPR2, NbRBOHD and NbWRKY12 in N. benthamiana leaves transiently expressing PsCDA2 after infiltration with chitin were examined by reverse transcription‐quantitative PCR, with NbAct1 as a reference gene for normalization. Means and standard deviations were calculated from three biological replicates. Asterisks indicate significant differences based on Student's t test (****p < 0.0001; ns, not significant, p > 0.05).

3. DISCUSSION

Avoiding recognition by the host plant is one of the most important factors for successful colonization by pathogenic fungi (de Jonge & Thomma, 2009). When pathogenic fungi attack plants, chitinases in the apoplastic space break down the chitin present in their cell walls, producing chitin oligomers. These chitin oligomers are then recognized by plant cell surface LysM receptors, resulting in the stimulation of plant defence responses. The acetyl portion plays a key role in LysM binding (Fiorin et al., 2018; Kaku & Shibuya, 2016). It is essential for pathogenic fungi to establish an effective mechanism to mask PAMP‐active cell wall fragments for fungi.

Previous studies have shown that CDAs prevent the recognition of chitin oligomers by the plant immune system mainly through two mechanisms. First, during host invasion, bioconversion of chitin to chitosan by deacetylation can protect hyphae of pathogenic fungi from plant chitinases (Gubaeva et al., 2018; Hayafune et al., 2014; Sanchez‐Vallet et al., 2015). Chitosan is a fully or partially deacetylated derivative of chitin and does not induce an immune response. For example, C. lindemuthianum secretes endochitin de‐N‐acetylase (ClCDA), which functions in the conversion of chitin to chitosan and masks the fungal cell wall against the host chitinase during infection, leading to fungal pathogenesis (Blair et al., 2006). Similarly, Pst_13661 of Pst can enhance the virulence through this mechanism. Second, fungi can modify immunogenic oligomers produced by chitinase (Cord‐Landwehr et al., 2016; Liu et al., 2017). VdPDA1 from V. dahliae and CDA from Pestalotiopsis sp. (PesCDA) are able to act on the chitin oligomers in cotton and rice, respectively, effectively inactivating their exciton activity. Preventing chitin recognition by the host plant is not achieved by a single protein but rather by multiple proteins working together. About 49 proteins have been identified in Pst and divided into three groups based on function; proteins of the same function in each group may act in concert. Furthermore, CDAs are widely distributed in fungi (Martinez‐Cruz et al., 2021). We identified two CDAs by performing a BLAST analysis on the genomes of Puccinia sorghi and Puccinia polysora using PsCDA2 as the query sequence. The protein sequence comparison revealed that PsCDA2 shows great similarity to the CDAs of P. sorghi and P. polysora, with 54.46% and 57.46% similarity, respectively (Figure S7).

Pst_13661 has been identified as an effector protein with deacetylase activity, and PsCDA2 is the second experimentally validated effector protein with the same activity. Protein alignment of PsCDA2 and Pst_13661 demonstrated that they have different sequences but both share a NodB homology domain (Figure S8). The NodB homology domain is a conserved region specific to members of the carbohydrate esterase family 4 (CE‐4), which includes CDAs (Aragunde et al., 2018). In this study, we tested PsCDA2's ability to bind polysaccharides and chitooligosaccharides. The results indicate that PsCDA2 has deacetylase activity and a strong affinity for chitin. Knocking down PsCDA2 in plants reduced expression of genes related to plant immunity, as well as disease symptoms. Furthermore, PsCDA2 inhibited chitin‐induced basal plant immunity, suggesting a role for CDA in Pst infection and colonization. These results align with the results obtained from Pst_13661. Importantly, PsCDA2 was found to protect the urediniospores of Pst from chitinase in vitro. This result suggests that PsCDA2 may deacetylate the chitin in the cell wall of Pst and thus avoid its hydrolysis. We therefore speculate that PsCDA2 provides virulence and acts synergistically as a CDA during Pst infection and colonization.

It is worth noting that resistant varieties of wheat are susceptible to the variability of stripe rust pathogenicity, which presents a continuous threat of stripe rust to global wheat production and food security. Pst is a strictly biotrophic pathogen and there is a lack of reliable genetic transformation systems, limiting our knowledge of the gene function of Pst and its specific pathogenic mechanisms. Identifying antifungal targets of CDAs or other important virulence factors is vital in developing strategies to control wheat rust. It was demonstrated that PsCDA2 was able to alter the type of chitin in the cell wall of the pathogen and weaken the virulence of Pst CYR32 in PsCDA2 knockdown plants. Future studies will focus on using gene editing or overexpression technology to obtain broad‐spectrum disease resistance to advance novel strategies for protecting the wheat crop.

4. EXPERIMENTAL PROCEDURES

4.1. Fungal and bacterial strains and plant materials

The Pst CYR32 was used in this study. For rust inoculation, urediniospores of virulent race CYR32 were produced on the second leaf of wheat variety Mingxian 169 at 12 ± 2°C for 16 h in the light and 8 h in the dark. For agro‐infiltration, Agrobacterium tumefaciens GV3101 was used. This strain was usually grown in lysogenic broth (LB) at 28°C. The antibiotics rifampicin (50 mg/mL) and kanamycin (50 mg/mL) were used when needed. Wheat variety Suwon 11 (T. aestivum) and N. benthamiana used for pathogenicity and cell death assays were grown in a greenhouse (16 h:8 h, 12°C:22°C, day:night).

4.2. Sequence and phylogenetic analyses

Transcript levels of all chitin‐related genes were performed by time‐series double RNA‐seq data. We sequenced two groups of wheat–rust interaction, named NIL_R versus CYR32 and NIL_S versus CYR32, and selected time points of 0, 18, 24, 48, 96 and 168 hpi. The specific method was the same as our previous study. Evolutionary tree analysis was performed with MEGA X (www.megasoftware.net) software. Multiple sequence comparisons were performed with ClustalX v. 2.1 (http://www.clustal.org/download/current/) software, followed by Jalview v. 2.11 (http://www.jalview.org/getdown/release/) software. Protein conserved motif analysis was done through the MEME online website (https://meme‐suite.org/meme/). Prediction of signal peptides was made using SignalP v. 5.0 (https://services.healthtech.dtu.dk/service.php?SignalP‐5.0), Phobius (https://phobius.sbc.su.se/) and DeepTMHMM (https://dtu.biolib.com/DeepTMHMM).

4.3. RNA extraction and cDNA synthesis

Total RNA was extracted from the wheat leaves infected by Pst CYR32. First‐strand cDNA was synthesized with an oligo(dT) primer using the EZ‐first strand cDNA synthesis kit (Biological Industries). qPCR was performed on a real‐time PCR system with SYBR Green PCR SuperMix and the real‐time PCR instrument was purchased from Analytik Jena, model qTOWER³G. NbAct1, PstEF‐1α, and TaEF‐1α were used as the internal reference genes for RT‐qPCR. PCR conditions were used as follows: 95°C for 60 s, then 40 cycles of 95°C for 10 s, 60°C for 20 s and 72°C for 40 s. All primers used in this study are listed in Table S2.

4.4. Validation of signal peptide secretion function

To function as a signal peptide of PsCDA2, PsCDA2SP (PsCDA2 signal peptide) and TaSBT1.7SP (TaSBT1.7 signal peptide) encoding regions were inserted into plasmid pSUC2 and transformed into yeast strain YTK12 using the lithium acetate method (Jacobs et al., 1997). Positive transformants were grown on CMD−W medium (0.67% yeast N base without amino acids, 0.075% tryptophan dropout supplement, 0.1% glucose, 2% sucrose, 2% agar) and then moved to YPRAA medium (2% raffinose, 2% peptone, 1% yeast extract, 2 μg/mL antimycin A) for assay of transformase secretion. Transformase activity was detected by reduction of TTC to insoluble red TPF. The signal peptide of TaSBT1.7 was used as positive control (Yang et al., 2021).

4.5. Transient expression in N. benthamiana

In the Bax inhibition assay, recombinant vectors (pCNF3, pCNF3:YFP, pCNF3:PsCDA2:YFP, PGR106:Bax:FLAG) were constructed and transformed into A. tumefaciens GV3101 for transient expression. The transformed strain was incubated for 16 h and then centrifuged at 4000 g for 20 min. The bacterial pellet was suspended in permeation buffer (10 mM MgCl₂, 10 mM 2,4‐morpholinoethanesulfonic acid [MES] and 150 nM acetosyringone) and the OD₆₀₀ was adjusted to 0.5. After 3 h of dark treatment, the bacterial resuspension was injected into 4‐week‐old N. benthamiana leaves using a syringe and kept in a greenhouse at 22°C. The intensity of the response was scored on a scale of 0 (no HR‐like phenotype) to 10 (infiltrated area confluent necrosis).

For localization, the recombinant vector and PM marker were co‐injected as above, and photographs were taken with confocal microscopy 2 days after injection. A construct carrying CD3‐1007 (AtPIP2A:mCherry fusion) was used as a marker for PM localization (Nelson et al., 2007). For plasmolysis, the epidermis of injected leaves was treated with 200 mM NaCl for 5–10 min and then observed by fluorescence microscopy.

For chitin‐induced plant immunity, recombinant A. tumefaciens GV3101 strains (pCNF3:YFP, pCNF3:PsCDA2:YFP) were separately injected as above. Infiltration with chitin was performed 48 h after injection. Samples were taken 12, 24 and 48 h after the infiltration of chitin for the detection of defence genes.

4.6. Apoplastic fluid isolation

Six full‐size infiltrated leaves were cut off and immersed in phosphate‐buffered saline. A vacuum was applied for 3–5 min, and then, after the vacuum was released, the leaf surfaces were dried with paper, wrapped in sealing film and placed in a centrifuge tube. After centrifugation (1000 g, 10 min, 4°C), approximately 300 μL of plasma exosomes was obtained.

4.7. Western blot analysis

The leaves were ground to a fine powder in liquid nitrogen after 48 h of Agrobacterium infiltration and dissolved in plant protein extraction buffer containing 1 mM phenylmethanesulfonyl fluoride. The mixture was centrifuged at 12,000 g for 20 min at 4°C. The supernatant was mixed with the sample buffer and boiled for immunoblot analysis. For immunoblotting, proteins were separated by SDS‐PAGE, then transferred to polyvinylidene fluoride (PVDF) membranes and detected using anti‐GFP antibody (ABclonal) or anti‐FLAG antibody (ABclonal).

4.8. Measurement of ROS accumulation

The level of H₂O₂ was measured using DAB and NBT. In brief, N. benthamiana leaves were excised at the base with a razorblade and soaked in 2 mg/mL DAB or 0.5 mg/mL NBT solution for 8 h under a vacuum. The leaves were then immersed in boiling decolourization solution (ethanol:acetic acid:glycerol, 3:1:1 vol/vol/vol) for 15 min.

4.9. Aniline blue staining

To detect callose deposits, aniline blue staining was performed 12 h after infiltration of leaves with chitin. Leaves were decolourized (ethanol:acetic acid, 3:1 vol/vol) and immersed overnight in chloral hydrate. Transparent leaf segments were stained with 0.01% aniline blue in 150 mM K₂HPO₄ (pH 8.5). Staining was performed under dark conditions for 4 h. The stained leaves were stored in sterilized 50% glycerol for fluorescence observation. Callose deposits were analysed in fields of 1 mm² using ImageJ software.

4.10. Cell death measurements

Cell death was measured by quantifying ion leakage using conductivity measurements of N. benthamiana leaf discs. For the test, six discs measuring 1 cm in diameter were taken from each leaf and placed in 8 mL of distilled water at room temperature for 4 h with shaking at 70 rpm. Conductivity was measured using a conductivity meter (DDS‐11A; INESA). A final measurement made after 20 min of boiling. The experiment was repeated three times using independently treated samples.

4.11. BSMV‐mediated gene silencing in wheat

BSMV‐mediated gene silencing was used to knock down the expression of the PsCDA2 gene to gain insight into the function it plays in response to Pst invasion. At the two‐leaf stage, the second leaf of wheat seedlings was inoculated with BSMV transcripts alone, and 12 days after virus inoculation the fourth leaf of each virus‐infected plant was inoculated with fresh urediniospores of CYR32. Wheat leaves were excised for RNA extraction at 0, 12, 24 and 48 h after inoculation and the silencing efficiency of selected genes was assessed by RT‐qPCR. The phenotype of BSMV infection was examined at 14 days postinoculation, and three independent biological replicates of each experiment were repeated.

4.12. In vitro protein expression and purification

PsCDA2 was inserted into the vector pCold_TF with a 6 × His tag and transformed into Escherichia coli BL21. PsCDA2 was induced at 18°C for 16 h. After 16 h, the bacterial pellet obtained by centrifugation at 10,000 g for 20 min was resuspended with phosphate‐buffered saline. The resuspended cells were crushed by an ultrasonic crusher at 20% power, crushing for 3 s and stopping for 5 s for 25 min. The broken cells were centrifuged at 15,000 g for 30 min, and SDS‐PAGE analysis demonstrated the presence of induced proteins in the supernatant, which was used for protein purification. All operations were carried out under low‐temperature conditions. The crude protein of PsCDA2 was purified by Ni column chelate affinity chromatography according to manufacturer's instructions (Sangon).

4.13. Detection of deacetylase enzyme activity

For detection of deacetylase enzyme activity, PsCDA2 (about 100 μg/mL) was used with the chitin deacetylase (CDA) ELISA kit (Shanghai Renjie Biotechnology Co., Ltd) according to the manufacturer's instructions. CDA, when combined with a horseradish peroxidase (HRP)‐labelled antibody, formed an antibody–antigen–enzyme–antibody complex. After complete washing, tetramethylbenzidine (TMB) substrate solution was added, which caused the TMB substrate to turn blue due to the catalytic activity of the HRP enzyme. The reaction was subsequently halted by adding a sulphuric acid solution, and the resulting colour change was evaluated using a microplate reader at a wavelength of 450 nm.

4.14. MST analysis

6× His‐tagged PsCDA2 was labelled according to the manufacturer's instructions for the labelling kit (Monolith His‐Tag Labeling Kit RED‐tris‐NTA). For the ligand dilution series, a 50 mg/mL solutions of chitin (Sangon Biotech), A6 (56/11‐0010, IsoSep), cellulose (Coolaber), or chitosan (Sangon Biotech) were prepared using phosphate‐buffered saline with Tween buffer. First, 16‐step 1:1 (vol/vol) ligand serial dilutions in the respective binding buffer were prepared, thereby reducing the ligand concentration by 50% at each dilution step. MST analysis was performed on a NanoTemper instrument (Monolith NT.115) using advanced coated capillaries. Three independent biological replicates were performed to calculate K _d values. All measurements were analysed centrally using the Monolith Affinity Analysis v. 2.3 software provided by Nano Temper.

4.15. Fungal cell wall protection assay

Two grams of wheat leaf tip tissue were added to 10 mL of prepared acetic acid‐sodium acetate buffer (pH 6.0) and then ground into a homogenate by adding quartz sand at low temperature. The homogenate was centrifuged at 15,000 g for 30 min at 4°C, and the supernatant was obtained as the crude extract of chitinase. PsCDA2 was added to the conidial suspension at a final concentration of 200 μM, followed by 4 μL of crude extract containing wheat chitinase, and incubated at 12°C for 24 h. The fungal growth was then observed under the microscope.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no competing interests.

Supporting information

FIGURE S1 Analysis of a family of chitin‐related proteins from Puccinia striiformis f. sp. tritici (Pst). (a) Phylogenetic tree analysis of 49 chitin‐related proteins from Pst CYR32, EWCA8 and EWCA5 from Podosphaera xanthii, Avr4 and Ecp6 from Cladosporium fulvum, ELP1 and ELP2 from Colletotrichum higginsianum, Chia1 and Slp1 from Magnaporthe oryzae, PDA1 from Verticillium dahliae, and Pst_13661 from Pst CYR32. (b) Heat map of affinity reciprocal and nonaffinity expression of 49 candidate genes on Pst CYR32 infection.

FIGURE S2 Prediction of signal peptides. The signal peptide sequences predicted by SignalP v. 5.0, DEEPTMHMM and Phobius were consistent. PsCDA2 encodes a 21 amino acid signal peptide at the N‐terminus.

FIGURE S3 The biological replicates of the Bax suppression assay.

FIGURE S4 The phenotype of the whole leaves of the plants infected with BSMV:PsCDA2 and BSMV:γ under CYR32 infection.

FIGURE S5 Fluorescence intensity of PsCDA2 in microscale thermophoresis (MST) analysis.

FIGURE S6 Callose accumulation revealed by aniline blue staining in Nicotiana benthamiana leaves expressing yellow fluorescent protein (YFP) and PsCDA2.

FIGURE S7 Full sequence comparison of PsCDA2, chitin deacetylases from Puccinia sorghi and Puccinia polysora.

FIGURE S8 Genome structure of PsCDA2 and Pst_13661, and their full sequence comparison. (a) The protein primary structure of PsCDA2. PsCDA2 encodes a chitin deacetylase (CDA) homologous structural domain that contains five conserved motifs forming the active enzyme site. (b) Full sequence comparison of PsCDA2 and Pst_13661.

TABLE S1 The expression profiles of 49 effectors.

TABLE S2 A list of PCR primers used in this work.

Xiao, M. , Chen, D. , Liu, S. , Chen, A. , Fang, A. , Tian, B. et al. (2023) A chitin deacetylase Ps CDA2 from Puccinia striiformis f. sp. tritici confers disease pathogenicity by suppressing chitin‐triggered immunity in wheat. Molecular Plant Pathology, 24, 1467–1479. Available from: 10.1111/mpp.13381

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.