Role of Two Metacaspases in Development and Pathogenicity of the Rice Blast Fungus Magnaporthe oryzae

Magnaporthe oryzae causes rice blast disease that threatens global food security by resulting in the severe loss of rice production every year. A tightly regulated life cycle allows M. oryzae to disarm the host plant immune system during its biotrophic stage before triggering plant cell death in its necrotrophic stage. ABSTRACT Rice blast disease caused by Magnaporthe oryzae is a devastating disease of cultivated rice worldwide. Infections by this fungus lead to a significant reduction in rice yields and threats to food security. To gain better insight into growth and cell death in M. oryzae during infection, we characterized two predicted M. oryzae metacaspase proteins, MoMca1 and MoMca2. These proteins appear to be functionally redundant and can complement the yeast Yca1 homologue. Biochemical analysis revealed that M. oryzae metacaspases exhibited Ca2+-dependent caspase activity in vitro. Deletion of both MoMca1 and MoMca2 in M. oryzae resulted in reduced sporulation, delay in conidial germination, and attenuation of disease severity. In addition, the double ΔMomca1mca2 mutant strain showed increased radial growth in the presence of oxidative stress. Interestingly, the ΔMomca1mca2 strain showed an increased accumulation of insoluble aggregates compared to the wild-type strain during vegetative growth. Our findings suggest that MoMca1 and MoMca2 promote the clearance of insoluble aggregates in M. oryzae, demonstrating the important role these metacaspases have in fungal protein homeostasis. Furthermore, these metacaspase proteins may play additional roles, like in regulating stress responses, that would help maintain the fitness of fungal cells required for host infection. IMPORTANCE Magnaporthe oryzae causes rice blast disease that threatens global food security by resulting in the severe loss of rice production every year. A tightly regulated life cycle allows M. oryzae to disarm the host plant immune system during its biotrophic stage before triggering plant cell death in its necrotrophic stage. The ways M. oryzae navigates its complex life cycle remain unclear. This work characterizes two metacaspase proteins with peptidase activity in M. oryzae that are shown to be involved in the regulation of fungal growth and development prior to infection by potentially helping maintain fungal fitness. This study provides new insights into the role of metacaspase proteins in filamentous fungi by illustrating the delays in M. oryzae morphogenesis in the absence of these proteins. Understanding the mechanisms by which M. oryzae morphology and development promote its devastating pathogenicity may lead to the emergence of proper methods for disease control.

IMPORTANCE Magnaporthe oryzae causes rice blast disease that threatens global food security by resulting in the severe loss of rice production every year. A tightly regulated life cycle allows M. oryzae to disarm the host plant immune system during its biotrophic stage before triggering plant cell death in its necrotrophic stage. The ways M. oryzae navigates its complex life cycle remain unclear. This work characterizes two metacaspase proteins with peptidase activity in M. oryzae that are shown to be involved in the regulation of fungal growth and development prior to infection by potentially helping maintain fungal fitness. This study provides new insights into the role of metacaspase proteins in filamentous fungi by illustrating the delays in M. oryzae morphogenesis in the absence of these proteins. Understanding the mechanisms by which M. oryzae morphology and development promote its devastating pathogenicity may lead to the emergence of proper methods for disease control.
KEYWORDS Magnaporthe oryzae, rice blast, disease, metacaspase, pathogenesis C aspases are a family of conserved cysteine-dependent, aspartate-specific proteases that play an essential role in metazoan programmed cell death. They regulate multiple cellular behaviors that contribute to organism fitness and pathology (1). This family of proteases contains a unique a/b hemoglobinase fold that consists of a large subunit (p20) containing the catalytic histidine/cysteine dyad and a small subunit (p10) mutant strain, but not the single mutant strains, was impaired in developing typical wild-type (WT) lesions on rice leaves compared to the WT strain, suggesting that MoMca1 and MoMca2 are redundant and that both are required for full pathogenicity. Consistent with this observation, both enzymes can functionally complement the yeast Dyca1 strain. Interestingly, we observed that these proteins promote the clearance of insoluble aggregates to maintain the fitness of M. oryzae cells. Collectively, our study demonstrates that the MoMca genes play important roles in growth, conidiation, appressorium development, and pathogenesis for M. oryzae.

RESULTS
M. oryzae genome contains two putative metacaspase proteins. To investigate critical roles of metacaspases in the M. oryzae life cycle, we first searched the Magnaporthe genome for metacaspase-related genes. We chose to analyze this group of proteins because they have been extensively studied in yeast but remain largely uncharacterized in filamentous, plant-pathogenic fungi. Using the protein sequence of yeast Yca1 for BLASTP searches, we identified two putative metacaspase genes, MGG_04626 and MGG_13530 (now named MoMca1 and MoMca2, respectively), in the M. oryzae genome database. MoMca1 and MoMca2 genes are located on chromosomes Chr3 and Chr4, respectively. These putative proteases are predicted to encode a typical type I metacaspase containing an N-terminal prodomain with a Q/N-rich repeat motif and a C-terminal peptidase-C14 caspase domain (Fig. 1A). MoMca1 and MoMca2 encode 396-and 410-amino-acid proteins, respectively. Clan clustering analysis of 2,712 metacaspase protein sequences showed that M. oryzae metacaspases clustered in the same clan as yeast Yca1, with their catalytic domains sharing approximately 45% sequence similarity (Fig. 1B). Bioinformatics analysis also revealed MoMca1 and MoMca2 share 67% similarity between their catalytic domains. Using ClustalW, the sequence alignment of MoMca1 and MoMca2 with Yca1 illustrated that M. oryzae proteins contain the predicted histidine and cysteine as the catalytic dyad specific for clan CD proteases ( Fig. 1A and C). In Fig. S1A in the supplemental material, the tertiary structure of fulllength M. oryzae metacaspases is modeled using the structural template of yeast Yca1 according to Swiss-Model (PDB entry 4F6P) (22).
MoMca1 and MoMca2 are Ca 2+ -dependent proteases. For most of the metacaspases studied so far, autocatalysis of an inactive zymogen is responsible for generating an active enzyme monomer (4,5,22). Therefore, we addressed the question of whether MoMca1 and MoMca2 undergo autocatalytic activation in a manner similar to that of yeast Yca1. Due to the insolubility of both M. oryzae metacaspase proteins in bacterial cells, we decided to use an in vitro transcription/translation approach to enable the rapid expression of the full-length proteins MoMca1 and MoMca2. A recent study demonstrated that Arabidopsis metacaspase AtMCP2d and yeast Yca1 strictly require Ca 21 for their proteolytic activity in vitro (5). We observed that recombinant MoMca1 and MoMca2 are already autocatalytically processed upon translation, yielding an ;35-kDa intermediate common to both metacaspases and the p10 subunit, at ;12 kDa in MoMca1 and ;10 kDa in MoMca2 ( Fig. 2A). To examine whether treatment with divalent cations such as Ca 21 positively affects M. oryzae metacaspase autoprocessing, we added 1 mM CaCl 2 to the translation mixture and observed further autocatalysis, yielding an ;25-kDa fragment of the large subunit ( Fig. 2A). These results demonstrate that Ca 21 specifically enhanced the autocatalytic processing of MoMca1 and MoMca2.
Sequence alignment of MoMca1 and MoMca2 with Yca1 showed the predicted catalytic Cys/His dyad residues in both M. oryzae metacaspases (Fig. 1C). The catalytic cysteine (Cys 242 for MoMca1 and Cys 257 for MoMca2) was replaced by an alanine residue in their coding sequences. Figure 2A shows that in the presence of Ca 21 , the catalytic dead proteins are unable to autoprocess and generate catalytic subunits, remaining inactive zymogens. These results demonstrated that the predicted conserved cysteines are essential for metacaspase autoproteolysis.
Expression of MoMca1 and MoMca2 in a heterologous system. Previous studies demonstrated that low doses of H 2 O 2 promote the expression of Yca1 in yeast, leading to the induction of apoptosis (4). The Dyca1 deletion strain exhibits higher rates of survival under oxidative stress conditions than the wild-type strain (4). To determine if MoMca1 and MoMca2 functionally complement Yca1 in H 2 O 2 -mediated apoptosis, we cloned the M. oryzae metacaspases into the Dyca1 yeast strain and grew these strains in liquid galactose inducing medium containing 1.2 mM H 2 O 2 for 24 h. Figure 2B shows a significant reduction in cell survival observed upon complementation of the Dyca1 strain containing either MoMca1 or MoMca2 compared to the Dyca1 strain carrying the empty vector. These results indicate that MoMca1 and MoMca2 can functionally complement Yca1 when expressed in S. cerevisiae. CLANS clustering analysis is represented graphically as the network of BLAST-derived sequence similarities between 2,712 metacaspase protein sequences. Dots represent sequences, with the shade of connecting lines denoting similarity from gray (low) to black (high). (C) The catalytic sites of MoMca1 and MoMca2 are aligned with Yca1, highlighting conserved active-site residues (black), calcium binding residues (red), Yca1 cleavage site (magenta), predicted MoMca1/MoMca2 cleavage site (Blue), mainly hydrophobic positions (light yellow), and mainly small positions (gray). Beta-strands (arrow) and alpha-helices (cylinder) from the Yca1 structure are indicated above the alignment and colored cyan/yellow before and pink/orange after the cleavage site (italics).
MoMca1 and MoMca2 are required for the virulence of M. oryzae. To determine if MoMca1 and MoMca2 play a role in M. oryzae pathogenicity, we generated single DMomca1 and DMomca2 deletion strains, a double mutant strain, DMomca1mca2, and a complemented strain, DMomca1mca2-C, containing pBGt MoMCA1MCA2 expression vector ( Fig. S1B and C). These strains were evaluated for pathogenicity on rice leaves. We inoculated susceptible rice plants with spore suspensions of the wild-type (WT) or mutant strains. The WT strain exhibited the expected necrotic lesions on rice leaves. In contrast, the DMomca1mca2 strain displayed a significant reduction in disease severity compared to the WT and the single mutants. (Fig. 3A and B). These data suggest that   M. oryzae metacaspases are required for sporulation. To investigate the role of MoMca1 and MoMca2 in growth and development of M. oryzae, we assessed the WT, mutant, and complemented strains for colony diameter on complete medium (CM; Fig. S2A and B). However, the DMomca1mca2 strain displayed severely reduced sporulation compared to the WT and single mutant strains (Fig. S2C). Moreover, none of these strains showed differences in sensitivity to cell wall assembly inhibitor or osmotic stress ( Fig. S3A and B). Taken together, these data suggest that metacaspases are individually dispensable for spore development and appropriate vegetative growth under axenic conditions and are redundant in providing an essential function for spore development.
MoMca1 and MoMca2 are expressed under oxidative conditions in vivo. To further verify whether MoMca1 and MoMca2 impact M. oryzae survival under oxidative stress as Yca1 does in yeast, we first measured the transcript levels of MoMca1 and MoMca2 under different stress conditions. A previous study showed that H 2 O 2 induces the expression of Yca1, leading to the activation of programmed cell death in yeast (4). Therefore, we used quantitative PCR (qPCR) to analyze the expression of MoMca1 and MoMca2 in the WT M. oryzae strain under H 2 O 2 and the free radical generator, menadione. Only MoMca1 expression was induced in the presence of H 2 O 2 , whereas both metacaspases were induced with menadione, suggesting that the enzymes play distinct roles in responding to cell stress (Fig. 4A).
To measure the sensitivity of the mutants and survival under oxidative stress conditions, we grew the WT and mutant strains on CM plates containing H 2 O 2 or menadione and compared their radial growth to that of the WT strain. Interestingly, we observed that both the single and double mutants were less sensitive to H 2 O 2 and menadione conditions than the WT and complemented strains (Fig. 4B). The mutant strains also exhibited higher radial growth under stress than the WT and complemented strains ( Fig. 4C and D). These results indicate that the M. oryzae metacaspases play an important role in stress-induced programmed cell death.
Deletion of MoMca1 and MoMca2 leads to a delay in conidial germination and appressorium formation. Next, we addressed whether MoMca1 and MoMca2 play a role in conidial germination or appressorium development. We collected spores from 10-day-old CM plates of the WT and mutant strains and placed spores in suspension on a hydrophobic surface. At 3 and 5 h postincubation (hpi), the WT strain germinated and began to form immature appressoria ( Fig. 5A and B). In contrast, the DMomca1mca2 strain was impaired in germination and appressorium morphogenesis ( Fig. 5A to C). The DMomca1 and DMomca2 single mutants showed nonsignificant reduction in germination rates. However, the DMomca2 mutant showed a reduction in appressorium development at 3 hpi but appeared similar to the WT at 5 hpi ( Fig. 5A to C). At 12 and 20 hpi, all the strains were able to develop appressoria (Fig. 5C). Interestingly, at 20 hpi we observed that approximately 20% of the germinated spores in the DMomca1mca2 strain showed elongated germ tubes compared to those of the WT and single mutants ( Fig. 6A and B). These data suggest that M. oryzae metacaspases play an important, but redundant, role in conidial germination, as there is a delay in appressorium development in their absence.
M. oryzae metacaspases play a role in the clearance of protein aggregates. Based on previous studies implicating the role of Yca1 in maintaining protein homeostasis, we next investigated whether the mutants alter protein stability during conidial germination and appressorium formation (6). In M. oryzae, the conidium contents are recycled into the appressorium, contributing to the next stages of appressorium development (19,20). To observe the cytoplasmic protein content during appressorium formation, we used the KV1 strain that harbors a cytosolic enhanced green fluorescence protein (eGFP) reporter protein (23). Spores from WT KV1 and DMomca1mca2 strains were collected and incubated on a hydrophobic surface for 20 hpi. Normally, on hydrophobic surfaces, M. oryzae spores germinate and form a short germ tube with an immature appressorium within 6 h. During this process, the conidial contents migrate into the incipient appressorium (19,24). Using eGFP as a marker, we can observe the movement of proteins into the conidium during this developmental process. For the WT KVI strain, Fig. 6B and C show the expected cytoplasmic movement of proteins from the conidium into the appressorium. In contrast, in the DMomca1mca2 strain, there is a delay of eGFP delivery due to the delay in appressorium formation.
In yeast, Yca1 plays an essential role in protein quality control by promoting the removal of insoluble protein aggregates to maintain the fitness of new cells (6). The Dyca1 mutant showed a significant increase of insoluble aggregates in the cells compared to the WT yeast strain (6). Therefore, we next asked if a similar phenotype could be observed in M. oryzae. We measured the insoluble protein aggregate fractions in M. oryzae WT, DMomca1mca2 mutant, and the DMomca1mca2-C complemented strain under normal gel electrophoresis. We observed an increase of insoluble aggregates in the DMomca1mca2 mutant compared to the WT and a restoration of the WT phenotype in the complemented strain ( Fig. 7A and Fig. S3C and D). The observations support the hypothesis that M. oryzae metacaspases play an important role in preventing protein aggregates from accumulating during appressorium formation. Previous , and 20 hpi, the percentage of appressorium development was measured and compared with that of the WT. Results are means from three independent biological measurements. A total of 150 spores were counted for each replicate. Images were taken at each time point. Error bars denote standard deviations. Asterisks indicate statistically significant differences (*, P , 0.05; **, P , 0.01; ***, P , 0.001; ****, P , 0.0001; two-way ANOVA with Tukey's multiple-comparison test using GraphPad Prism 8). Scale bar is 10 mm.

Fernandez et al.
® reports demonstrated that the Hsp70s aggregate-remodeling chaperones are overexpressed as a compensatory response to yeast Yca1 deletion (6). To further study the protein aggregation phenotype associated with metacaspase deletion, we assessed the expression of the 70-kDa heat shock protein, Hsp70, in the spores and mycelial lysates. Using Western blot analysis, we detected the expression of Hsp70 in the soluble and insoluble fractions of WT, DMomca1mca2, and DMomca1mca2-C strains (Fig. 7B). As Fig. 7 shows, Hsp70 was found to accumulate in the insoluble fraction of the DMomca1mca2 mutant but not the WT and complemented strains. Moreover, these results may lead to a better understanding of metacaspase roles in maintaining protein homeostasis in M. oryzae cells throughout development and pathogenesis.

DISCUSSION
Metacaspases are multifunctional proteases essential for the normal physiology and pathology of nonmetazoan species. These proteases have been extensively characterized in programmed cell death and nonapoptotic processes. Although the Shown is the percentage of cytoplasmic eGFP movement into the appressorium in the DMomca1mca2 compared to the WT strain. Results are means from three independent measurements. A total of 150 germinated spores with appressoria were counted for each replicate. Sp, spores; Gt, germ tube; App, appressorium. Error bars denote standard deviations. ns, not significant. Asterisks indicate statistically significant differences (*, P , 0.05; ***, P , 0.001; ****, P , 0.0001; oneway ANOVA with Tukey's multiple-comparison test using GraphPad Prism 8). Scale bar is 10 mm. We bioinformatically identified two type 1 metacaspase proteins, MoMca1 and MoMca2, in the M. oryzae genome. These proteins contain an N-terminal prodomain rich in Q/N residues and a C-terminal peptidase C14 caspase domain. MoMca1 and MoMca2 adopt the structure of a caspase fold similar to that of the yeast Yca1 core (see Fig. S1A in the supplemental material).
Due to the bioinformatic similarities of MoMca1 and MoMca2 with Yca1, we looked at the potential link in Ca 21 zymogen activation. In contrast to canonical caspases, metacaspases do not undergo dimerization for their activation. Instead, the activity of metacaspases depends on the presence of calcium ions (22,25), with the only known exception being Arabidopsis thaliana AtMC9, whose activity was shown to be Ca 21 independent (5,26). However, the Arabidopsis AtMCP2d metacaspase exhibits a strict Ca 21 dependence in millimolar concentrations for its catalytic activation (26). In Yca1, the addition of Ca 21 activates the enzyme and leads to autoprocessing events (22). Our biochemical studies have identified MoMca1 and MoMca2 to be Ca 21 -activated proteases, similar to the Yca1 metacaspase. First, we observed that the autocatalytic processing of MoMca1 and MoMca2 is stimulated by the presence of Ca 21 in vitro. In addition to the activation requirements for Ca 21 , we found two cysteine residues that were essential for Yca1 activity were also important for MoMca1 and MoMca2 activity.
Similar to the role of canonical caspases, yeast Yca1 metacaspase positively regulates apoptosis under several stress conditions (27). In this report, we provide evidence that high levels of oxidative stress inducers, such as H 2 O 2 and menadione, upregulate the expression of M. oryzae metacaspase genes. These observations lead us to investigate the functional similarities between M. oryzae metacaspases and yeast Yca1. In early studies, Madeo and colleagues demonstrated that low concentrations of H 2 O 2 facilitate the activation of Yca1 promoting yeast apoptosis (4). Lack of yeast Yca1 activity confers resistance to oxidative stress conditions and a higher survival rate under them (4). Here, we provide evidence that both MoMca1 and MoMca2 can substitute for the Yca1 function in mediating oxidative stress-induced apoptosis in yeast. Although MoMca1 and MoMca2 functionally complement the oxidative stress phenotype in yeast, whether the M. oryzae metacaspases respond to similar external stimuli remains unclear. Moreover, the DMomca1mca2 strain shows resistance to H 2 O 2 and menadione, similar to the Dyca1 strain.
Additionally, the deletion of either MoMca1 or MoMca2 in M. oryzae impairs developmental processes while having no significant reduction in vegetative growth rate on solid media. We demonstrate that both conidiation and symptom development are metacaspase-dependent processes in M. oryzae, suggesting a role for these proteins in cell development. The DMomca1mca2 strain was significantly impaired in proper conidiation and pathogenicity.
To be a successful pathogen, M. oryzae undergoes several morphological changes to colonize and proliferate inside the rice cells (18). Once the conidium attaches to the leaf cuticle, it germinates and develops a germ tube that will lead to the formation of the melanized appressorium. This morphogenesis process is tightly regulated by the cell cycle checkpoints and occurs within hours (2 to 24 h) after conidium attachment (28). Surprisingly, we discovered that the DMomca1mca2 mutant exhibits delayed conidial germination under inducible conditions, suggesting that these proteins contribute to initial processes that precede appressorium morphogenesis. This delayed phenotype in the DMomca1mca2 mutant disrupts the timing of the appressorium maturation process. Moreover, we observed a delay in the movement of conidial contents into the appressorium. It is well known that during appressorium development, the conidial contents are recycled and delivered to the appressorium, leading to conidial degradation (24). This multistage process is highly regulated by cell cycle progression and autophagy-dependent cell death (24,28,29). However, how M. oryzae metacaspases cross talk with autophagy processes to regulate appressorium development remains largely unknown.
In other systems, metacaspases display varied roles in pathogenicity and oxidative stress resistance. For instance, the filamentous fungus Podospora anserina contains two metacaspases, and the deletion of both metacaspase genes leads to lower growth rate and fertility, indicating a function in developmental processes (14). However, a single deletion strain, PaMca1, has a life span-prolonging effect and resistance response to oxidative stress (14). In contrast, the CasA/CasB metacaspases in the human pathogen Aspergillus fumigatus showed no resistance to oxidative stress or to other apoptosisinduced agents. However, the DcasA DcasB mutant displays a growth defect under endoplasmic reticulum stress conditions, indicating that these proteases confer vital cellular functions under certain stress conditions rather than an involvement in cell death processes (13). Recent studies identified a single metacaspase gene in Ustilago maydis that exhibits impaired growth under either normal growth or oxidative stress conditions (16). The U. maydis mca1 mutant also displayed a reduction in pathogenicity compared to the wild type, demonstrating that Mca1 in U. maydis is required for cellular homeostasis during cell development (16). In this study, we found that the M. oryzae DMomca1mca2 mutant exhibited a reduction in pathogenicity on rice plants. With these observations, we can infer that metacaspase functions vary among different organisms and have partial redundancy or antagonistic functions under various settings.
In S. cerevisiae, Yca1 metacaspase appears to be critical for the removal of insoluble protein aggregates during physiological growth conditions. The deletion of yca1 was associated with an accumulation of insoluble aggregates during logarithmic growth that correlated with an enrichment on vacuolization and stress-response chaperones (6).
Due to previous findings about Yca1 promoting clearance of protein aggregates, we decided to investigate the involvement of our metacaspases in this process. It is noteworthy that in the DMomca1mca2 mutant we observed the accumulation of insoluble protein aggregates during normal growth conditions not present in the control. These findings clearly suggest that M. oryzae metacaspases contribute to the removal of insoluble aggregates to maintain optimal fitness during fungal growth. However, we are not clear how M. oryzae mediates the clearance of insoluble aggregates during physiological growth. We initially hypothesized that the accumulation of the insoluble aggregates has a direct effect in delaying conidial germination in M. oryzae, although a recent study on protein aggregates in M. oryzae found no effect on germination when inducing protein aggregates in the spores (30). Although protein aggregates were present in large quantities in catalytic mutants, these spores did not have delayed germination (30). These data suggest that germination itself in the DMomca1mca2 mutant is affected in other ways, independent of protein aggregate accumulation, potentially through some other metacaspase regulatory function. The discovery of metacaspase substrates will give us new insights into the mechanism of activation in vivo and the downstream events.
Our work characterizing the role of the M. oryzae metacaspases MoMca1 and MoMca2 has demonstrated that biochemically there are several shared properties with the Yca1 metacaspase. MoMca1 and MoMca2 are Ca 21 -dependent peptidases. While MoMca1 and MoMca2 are dispensable during vegetative growth, they are critical for inducing apoptosis under oxidative stress conditions. More importantly, the deletion of these genes leads to an accumulation of protein aggregates. The most striking phenotype of the M. oryzae metacaspases is that they are critical for sporulation and, subsequently, pathogenicity in rice blast disease progression.

MATERIALS AND METHODS
Strains and culture conditions. The M. oryzae strain KV1 was used as the wild-type strain throughout this research (23) (see Table S1 in the supplemental material). This strain and its transformants were cultured on complete medium (CM; 1% [wt/vol] glucose, 0.2% [wt/vol] peptone, 0.1% [wt/vol] yeast extract, 0.1% [wt/vol] Casamino Acids, 0.1% trace elements, 0.1% vitamin solution, and 1Â nitrate salts) and incubated at 25°C under 12-h light/dark cycles for 5 to 12 days. For DNA and RNA extraction, all strains were grown on CM liquid medium at 25°C with agitation for 48 h. For sporulation rates, strains were grown on at least three independent CM plates. After 12 days of growth, spores were harvested and counted using a hemocytometer (Corning). To observe vegetative mycelial growth under stress, 10 mM H 2 O 2 and 150 mM menadione were individually added to CM agar medium. A mycelial disc (3.5 mm in diameter) of each strain was inoculated on stress medium containing either H 2 O 2 or menadione, and the growth rate was assessed by measuring culture diameters after 5 days of growth (unless otherwise stated). Plate images were taken with an Epson perfection V700 photo scanner. For routine cloning, the Escherichia coli DH5a strain was grown in 2ÂYT broth at 37°C. The Saccharomyces cerevisiae yeast strain BY4741 was grown in synthetic dropout (SD) medium as described previously (31)  For yeast expression, MoMca1 and MoMca2, full-length and variant versions, were cloned into the XhoI/HindIII sites on a galactose-driven yeast expression vector, pESC-Leu (kind gift from Vincent S. Tagliabracci).
Sequence analysis. The MoMca1 and MoMca2 coding gene sequences were obtained from the Magnaporthe oryzae database, http://fungi.ensembl.org/Magnaporthe_oryzae/Info/Index. Metacaspase sequences were collected using the MoMca1 (NCBI accession no. A4QTY2.2) sequence as a query for PSI-BLAST (33) (5 iterations, E value cutoff of 0.005) against a sequence database containing the RefSeq representative prokaryotic genome set (1,684 genomes, from 28 August 2018) and latest representative eukaryotic genomes (from 28 August 2018). Collected sequences (2,712) were clustered using CLANS (34) and colored according to taxonomy. A multiple-sequence alignment of representative sequences was generated using MAFFT (35) and colored according to conservation: mainly hydrophobic (yellow), mainly small (gray), active site (black), and calcium activation (red). To identify a structure template for MoMca1, the sequence (A4QTY2.2) was submitted to the HHPRED server (36) to search against the PDB database, which confidently identified the Yca1 structure (PDB entry 4F6O) as a top template (100% probability; score, 2.2e238). The HHPRED pairwise alignment between MoMca1 and Yca1 was used to generate a structure model for MoMca1 using SwissModel (37).
Fungal transformation and complementation. Targeted gene replacements for MoMca1 and MoMca2 were carried out using the PCR-based split marker method (38) in which the HPH gene (1.4 kb), conferring resistance to hygromycin, and ILV1 gene (2.8 kb), conferring resistance to sulfonylurea, replaced the coding sequence of each gene. Approximately 1 kb upstream and downstream of the gene was used for homologous recombination. The flanking sequences (;1 kb) of the MoMca1 and MoMca2 genes were amplified and fused to the selectable marker by PCR (Fig. S1A). PCR products were transformed into M. oryzae WT protoplasts. Positive transformants carrying homologous gene replacement of the gene of interest were confirmed by PCR using the nested primers shown in Table S2.
To generate the complemented strain, full-length MoMca1 and MoMca2 genes, including their native promoter, were amplified and inserted into pBGt (39) fungal expression vector by using the Gibson Assembly approach (NEB BioLabs). The pBGt-MoMca1Mca2 expression vector was transformed into DMomca1mca2 protoplasts and selected on gentamicin-containing plates. The complementation strain was identified by PCR (Table S2). Single-spore isolation was performed for all strains.
Spore germination and appressorium development assays. Spores were harvested from 10-dayold cultures, filtered through two layers of Miracloth (Millipore Sigma), and resuspended to a concentration of 5 Â 10 4 spores/ml in sterile water. For conidial germination, 20 ml of conidial suspension was placed on glass coverslips (Fisher Scientific, St. Louis, MO, USA) and incubated at 24°C in a humid chamber. The percentages of conidial germination and appressorium formation were assessed at 3, 5, 8, 12, and 20 h postincubation. Average mean values were determined from 150 conidia and performed in three independent experiments. Germ tube lengths were measured using ImageJ. All imaging was performed on a Zeiss LSM 800 confocal microscope, and images were converted using ImageJ (NIH).
Pathogenicity assays. Spores from the WT and transformants were harvested from 10-day-old cultures on 0.2% gelatin as previously described (40). Three-week-old rice seedlings (YT16) were spray inoculated with 2 Â 10 4 spores/ml conidial suspension. Inoculated plants were placed in a humidity chamber for 24 h and then transferred back to the growth chamber with 12-h light/dark cycles. Disease severity was assessed at 7 days postinoculation. Photographs of diseased rice leaves were taken. The number of pixels under lesion areas and healthy areas of diseased leaves were calculated by using ImageJ.
Yeast transformation. Yeast transformations were performed using the lithium acetate (LiAc) method as described previously (31). Briefly, yeast cells were grown overnight at 30°C in yeast extract-peptone-dextrose (YPD) medium. The overnight cultures were diluted in 50 ml of fresh YPD (adjusted to an optical density at 600 nm [OD 600 ] of 0.1) and then cultured until reaching an OD of 0.5 to 0.7. After incubation, the yeast cells were collected by centrifugation and washed twice with solution 1 (1Â Tris-EDTA [TE], 1 M lithium acetate, pH 7.5, with acetic acid). The pellet was resuspended with 500 ml of solution 1. In 100ml of yeast cells, 1 mg of DNA, 5ml DNA carrier, and 5 ml of 100% dimethyl sulfoxide (DMSO) were added and gently mixed. A volume of 700 ml of solution 2 (solution 1 supplemented with 50% polyethylene glycol) was added and incubated at 30°C for 30 min. The yeast cells were heat shocked at 42°C for 15 min and washed with TE, pH 7.5. Samples were plated on SD medium minus leucine or uracil and incubated at 30°C for 2 to 3 days.
Survival assay. Yeast survival assay was performed as described previously (4). Briefly, yeast strains harboring the expression vector were grown overnight in 3 ml liquid SD-Leu glucose medium at 30°C. The overnight cultures were diluted in 10 ml of fresh medium (adjusted to an OD 600 of 0.05) and then cultured until reaching an OD 600 of 0.4 to 0.6. The yeast cells were collected and resuspended in SD-Leu supplemented with 2% galactose and 1% raffinose. Cell counts were equalized based on OD 600 before treatment. For stimulation of yeast apoptosis, cells were treated with the final concentration of 1.2 mM H 2 O 2 and incubated for 24 h at 30°C. An aliquot of each culture was 10-fold diluted in sterile water and plated in SD glucose plates. Colonies were counted after 2 to 3 days, and treated samples were compared with nontreated samples to calculate survival rate. This experiment was conducted in biological triplicates.
RNA extraction, cDNA synthesis, and qRT-PCR analysis. To quantify metacaspase expression under oxidative stress conditions, the fungal strains were grown on liquid CM for 48 h at 25°C with agitation (200 rpm) before treatment with or without 10 mM H 2 O 2 and 150 mM menadione for 2 h. Mycelia were harvested and ground with a pestle and mortar in liquid nitrogen and stored at 280°C. A total of 100 mg of ground mycelia was used to perform RNA extractions. Total RNA was isolated using the RNeasy plant mini plus kit from Qiagen. RNA concentration was measured via NanoDrop, and cDNA was generated using the iScript cDNA synthesis kit (Quanta). Primers were designed to amplify 100 to 150 bp of each target gene (Table S2) and tested for efficiency. Transcripts were quantified on a CFX384 Touch real-time PCR detection system using the iTaq universal SYBR green supermix (Quanta) and 500 nM primers. Relative gene expression for each target gene was calculated by the 2DDC T method. The expression of each gene was normalized against the M. oryzae actin gene (ACT1). Results are given as the averages from three technical replications and three biological replications.
In vitro transcription/translation and purification. Coupled transcription/translation reactions were carried out per the manufacturer's instructions (L1170; Promega). Briefly, rabbit reticulocyte lysate was incubated with 1 mg of the appropriate expression plasmid and 2 ml of [ 35 S]L-methionine (1,000 Ci/ mMol at 10 mCi/ml) (NEG709A; Perkin Elmer) for 90 min at room temperature. Reaction mixtures were diluted in 1 ml of 50 mM Tris-HCl (pH 7.5), 10 mM NaCl, and 1 mM dithiothreitol (DTT). A volume of 100 ml of DEAE-Sepharose (DFF100; Millipore Sigma) slurry was added to the reaction mixtures and incubated on a nutator for 30 min at room temperature. The slurry was washed three times with 1 ml of 50 mM Tris-HCl (pH 7.5), 10 mM NaCl, and 1 mM DTT and eluted with 500 ml of 50 mM Tris-HCl (pH 7.5), 500 mM NaCl, and 1 mM DTT. Eluents were collected and incubated with 50 ml of packed nickel-nitrilotriacetic acid resin (His-Pur 88222; Bio-Rad) at room temperature with nutating for 30 min. The beads were washed three times with 1 ml of 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 20 mM imidazole, and 1 mM DTT. Proteins were eluted using 100 ml of 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 300 mM imidazole, and 1 mM DTT.
Protease assays. Protease reactions were carried out in a buffer solution containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM DTT, with or without 1 mM CaCl 2 . Reaction mixtures were incubated at 30°C overnight and stopped by the addition of Laemmli buffer containing b-mercaptoethanol and boiling. Reaction products were resolved on a 12% polyacrylamide gel by electrophoresis and autoradiography.
Protein aggregate assays. Protein fractions from spores and mycelial cell lysates were collected as described previously (6), with some modifications. Briefly, spores were harvested from 10-day-old cultures, filtered through two layers of Miracloth (Millipore Sigma), and resuspended to a concentration of 5 Â 10 5 spores/ml in sterile water. Spores were recovered by centrifugation, and pellets were frozen down and stored at 280°C for later processing. Mycelium was harvested and ground with a pestle and mortar in liquid nitrogen and stored at 280°C. A total of 100 mg of ground mycelia was used to performed protein extractions. Pellet samples were thawed in equal volumes of lysis buffer (0.1% Triton X-100, 50 mM Tris, pH 7.4, 1 mM EDTA, and 1% glycerol supplemented with 5 mM Na 3 VO 4 and 1 mM phenylmethylsulfonyl fluoride) and acid-washed glass beads (Sigma-Aldrich). Samples were vortexed for 20 min with 1-min on/off cycles. To remove the cell debris, samples were centrifuged at 2,000 rpm. The total cell lysate was collected and protein concentration was measured and normalized. The cell lysate was centrifuged at 15,000 Â g for 15 min at 4°C to separate the soluble (supernatant) and insoluble (pellet) fractions. Pellets were washed twice with lysis buffer containing 10% Triton X-100. Pellets were treated with 4 M urea and boiled in 2Â Laemmli sample buffer for 5 min. Proteins were loaded on 12.5% SDS-polyacrylamide gel electrophoresis for one-dimensional (1D) PAGE separation for silver staining or transferred to polyvinylidene difluoride membranes. HSP70 expression was detected with mouse anti-HSP70 (1:5,000; Abcam) antibody, and b-tubulin (polyclonal anti-b-TUB; Santa Cruz) was used as a loading control.
Statistical analysis. All data are given as means 6 standard deviations from at least three independent experiments (unless stated otherwise). Each experiment was conducted in triplicate. Statistical analyses were performed by using one-way analysis of variance (ANOVA) and Tukey's multiple-comparison test. A P value of ,0.05 was considered significant.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.