Abstract

   Hydrogen sulfide (H[2]S) regulates cellular activities in plants and
   mammals through S-sulfhydration, a post-translational modification of
   proteins. The role of H[2]S and its molecular targets in fungi,
   however, remains unclear. Here we show that H[2]S, synthesized by
   cystathionine γ-lyase (CSE1) in the rice blast fungus Magnaporthe
   oryzae, is essential for optimal fungal infection. Excessive H[2]S,
   through S-sulfhydration, impairs fungal infectivity by inhibiting
   autophagy. Using quantitative proteomics, we identify numerous
   S-sulfhydrated proteins in M. oryzae, including the autophagy-related
   protein ATG18. S-sulfhydration of a cysteine residue (Cys78) in ATG18
   is essential for its binding to phosphatidylinositol 3-phosphate,
   thereby maintaining the protein’s structural stability and regulating
   autophagy. Thus, our study reveals a mechanism by which H[2]S-mediated
   S-sulfhydration controls autophagy in the rice blast fungus and
   suggests the potential use of H[2]S donors as a strategy to control
   fungal diseases by targeting fungal development and infection
   structures.

   Subject terms: Fungal pathogenesis, Pathogens, Post-translational
   modifications
     __________________________________________________________________

   Hydrogen sulfide regulates cellular activities in some organisms
   through posttranslational S-sulfhydration of proteins. Here, Hu et al.
   reveal mechanisms by which S-sulfhydration controls infectivity and
   autophagy in a phytopathogenic fungus.

Introduction

   Hydrogen sulfide (H[2]S) has been recognized as the third gas signaling
   molecule after nitric oxide (NO) and carbon monoxide (CO)^[54]1,[55]2.
   Emerging data over recent years suggest that it is as important as NO
   and H[2]O[2]. H[2]S is involved in many physiological and pathological
   processes in animals, and it regulates cell cycle, apoptosis, and
   oxidative stress. Additionally, H[2]S has been implicated in various
   diseases such as cardiovascular disorders, neurodegenerative diseases,
   and cancer^[56]3–[57]5. In plants, H[2]S controls flowering, stomatal
   closure, and other developmental activities^[58]6,[59]7. More
   significantly, a plethora of recent studies have elucidated the
   regulatory properties of H[2]S on drought tolerance, oxidative stress,
   and osmotic stress^[60]6,[61]8,[62]9. However, the intricate regulatory
   mechanism of H[2]S during these processes remains largely elusive.

   In mammals, endogenous H[2]S is mainly catalyzed by cystathionine
   γ-lyase (CSE1) and cystathionine β-synthase (CBS), which are mainly
   present in the cytoplasm^[63]10,[64]11. Mice with deletion of CSE1 gene
   exhibit a decrease in the ability to produce H[2]S, leading to a
   significant reduction in endogenous H[2]S levels in the heart, blood
   vessels, and many other tissues, ultimately resulting in the
   development of hypertension^[65]12. [L]-cysteine desulfhydrase 1 (DES1)
   is a component of the plant ABA signaling pathway in guard cells and
   responsible for cysteine degradation and the concomitant production of
   H[2]S^[66]13,[67]14. The study of the Arabidopsis des1 mutant, which
   has a deficiency in producing H[2]S from cysteine in the cytosol, has
   revealed that H[2]S acts as an inhibitor of autophagy triggered by
   nutrient scarcity, furthermore, the deficiency of DES1 not only
   promotes the accumulation of the Autophagy Related 8 (ATG8) protein but
   also enhances its lipidation, thereby influencing the autophagy
   process^[68]15.

   S-sulfhydration, an emerging post-translational modification (PTM), is
   a key mechanism for cellular sensing of H[2]S signaling^[69]13. This
   modification involves the covalent addition of thiol groups to cysteine
   residues on target proteins, forming persulfides^[70]16. As a redox
   modification, S-sulfhydration can regulate protein function similarly
   to NO-induced S-nitrosylation, by inhibiting or activating proteins and
   affecting their activity and structure^[71]17,[72]18, as well as their
   subcellular localizations^[73]19. High-throughput proteomics has
   identified ATG-related proteins such as ATG4, ATG18a, ATG3, ATG5, and
   ATG7 as susceptible to S-sulfhydration, highlighting its role in
   autophagy regulation^[74]15. H[2]S-mediated S-sulfhydration has been
   shown to inhibit ATG4’s proteolytic activity, negatively regulating
   autophagy^[75]20.

   Quantitative data reveal that S-sulfhydration, a persulfur-based
   signaling process, is prevalent in both animal and plant cells,
   indicating its universal role in cellular regulation^[76]16,[77]21.
   This has led to increased interest in understanding its mechanisms.
   S-sulfhydrated proteins have been identified in these systems using
   biotin labeling and proteomics^[78]22–[79]24. Up to 25% of mouse liver
   proteins, including actin, β-tubulin, and GAPDH, are in vivo
   S-sulfhydrated^[80]16, and similar proteins have been found in plants.
   In Arabidopsis thaliana, at least 10% of the proteome may be
   S-sulfhydrated, participating in carbon metabolism, stress responses,
   and RNA translation^[81]25. Despite these findings, the role and
   regulatory mechanisms of S-sulfhydration in fungi remain poorly
   understood.

   Autophagy is a major catabolic process in eukaryotic cells that
   degrades organelles and proteins for recycling^[82]26,[83]27. It has
   conserved functions in development, cell homeostasis, and stress
   response in plants, animals, and fungi. The autophagy process is
   characterized by the formation of a double-membrane structure called an
   autophagosome by the phagophore^[84]28. To date, more than 30 known
   autophagy related genes (ATGs) in yeast have been characterized^[85]29.
   Recent research has shown that ATG proteins are subject to various
   post-translational modifications (PTMs), including phosphorylation,
   acetylation, and S-nitrosylation^[86]30–[87]32. However, the specific
   impact of S-sulfhydration on these ATG proteins remains to be fully
   elucidated.

   Rice blast caused by the filamentous fungus Magnaporthe oryzae is the
   most devastating disease in cultivated rice^[88]33. During infection,
   M. oryzae has shown to develop a specialized structure known as an
   appressorium within the host, which forms a strong turgor inside and
   penetrates the host^[89]33,[90]34. During this process, the regulation
   of autophagy-related genes in M. oryzae plays a critical role in both
   conidium cell death and appressorium maturation, which are essential
   for effective fungal infection^[91]35,[92]36. In this study, we sought
   to determine the role of H[2]S-mediated S-sulfhydration in the
   regulation of autophagy in M. oryzae. We discovered that H[2]S-mediated
   S-sulfhydration negatively regulates the autophagy process in M. oryzae
   by modulating the binding activity of the ATG18 protein to its
   substrates.

Results

Excessive H[2]S is harmful to M. oryzae

   H[2]S, a signaling molecule of equal importance to carbon monoxide and
   nitric oxide, may play a crucial role in the infection process of
   pathogenic fungi. To explore the possible role of H[2]S in the
   infection process of the rice blast fungus, we utilized a donor of
   H[2]S (sodium hydrosulfide, NaHS) and a scavenger of H[2]S
   (hypotaurine, HT) to evaluate the impact of H[2]S on the fungal
   infection structures and infection process. The rate of appressorium
   formation was not affected when 100 μM NaSH was used to treat the
   conidium suspension on the hydrophobic surface (Fig. [93]1a). However,
   it did affect the accumulation of turgor pressure in the appressoria
   and the formation of the septin ring (Fig. [94]1b–e), suggesting that
   high levels of H[2]S negatively regulate fungal infection.
   Additionally, treatment with NaHS impeded the infection process of M.
   oryzae in rice (Fig. [95]1f–h). In contrast, 100 μM Hypotaurine did not
   significantly impact appressorium formation but enhanced the
   development of infection structures during the infection process
   (Fig. [96]1a–h). Collectively, these findings indicate that excessive
   H[2]S is harmful to the development and infection of M. oryzae.

Fig. 1. Excessive H[2]S is harmful to the development and invasion of
infected structure.

   [97]Fig. 1
   [98]Open in a new tab

   a The appressorium formation rate of wild-type (WT) strains treated
   with H[2]S donors (NaHS) and H[2]S scavengers (Hypotaurine, HT). The
   spore suspension treated with 100 µM NaHS and HT was cultured on
   hydrophobic surfaces, and the appressorium formation rate was measured
   after 24 hours. The data represent the mean ± standard error of three
   biological replicates (n = 3). b Observation of turgor pressure in
   appressorium of wild-type (WT) strains treated with 100 µM NaHS or HT.
   The collapse of appressorium was observed in environments with
   different concentrations of glycerol at 24 h. Bars, 20 μm. c Statistics
   on the collapse rate of appressorium under different treatments in
   various concentrations of glycerol. The data represent the
   mean ± standard error of three biological replicates (n = 3). d
   Observation on the formation of appressorium septin ring at 12 hpi
   (hours post-inoculation) after treatment with 100 µM NaHS or HT. Bar, 5
   μm. e Percentage of septin ring formation in wild-type (WT) strains
   treated with 100 µM NaHS or HT. Data presented are the mean ± standard
   errors from three biological replicates (n = 3). f Observation of
   invasive hyphae growth of the wild-type strain in rice sheath cells
   after 100 µM NaHS or HT treatment at 24 hpi and 36 hpi (hours
   post-inoculation). Bar, 10 μm. Statistical analysis of invasive hyphae
   growth in rice sheath cells at 24 hpi (g) and 36 hpi (h). The data
   represent the mean ± standard error of three biological replicates
   (n = 3). Data are displayed as box and whisker plots with individual
   data points: median, first and third quantiles, maximum and minimum (c,
   e). Data are analyzed by one-way ANOVA (a, e) or two-way ANOVA (c, g,
   h) followed by Tukey’s test. Adjusted p values are shown in figures.
   Source data are provided as a Source data file.

Identification of H[2]S synthesizing enzyme CSE1 in M. oryzae

   In order to investigate the function of H[2]S in M. oryzae, we
   identified, MGG_10380, the homologous protein of cystathionine γ-lyase
   (CSE1) in M. oryzae. The CSE1 gene has a full length of 1869 bp and
   encodes a protein consisting of 426 amino acids. This protein contains
   a conserved Aminotran_1_2 domain (Fig. [99]2a), which is essential for
   catalyzing amino acid transfer reactions, engaging in metabolic
   pathways and synthesis, and preserving protein structure stability.
   Phylogenetic tree analysis reveals the conservation of the CSE1 protein
   across different species (Fig. [100]2b). Homology modeling comparisons
   revealed that the structure of CSE1 in M. oryzae is highly similar to
   that of the homologous protein in Saccharomyces cerevisiae
   (Fig. [101]2b). The qRT-PCR data reveal a notable decline in CSE1 gene
   expression during the appressorium development stages of M. oryzae
   (AP_3H and AP_12H), as shown in Supplementary Fig. [102]1. This
   suggests that H[2]S potentially exerts a negative regulatory effect on
   appressorium function in vivo.

Fig. 2. CSE1 controls H[2]S production and S-sulfhydration levels in
Magnaporthe oryzae.

   [103]Fig. 2
   [104]Open in a new tab

   a Structural domain and phylogenetic tree analysis of M. oryzae CSE1
   protein using molecular evolutionary genetics analysis v.5 (MEGA5) and
   simple modular architecture research tool (SMART). b Comparison of
   homology modeling of CSE1 protein in M. oryzae and S. cerevisiae. c
   Assessment of H[2]S levels at various developmental stages of the M.
   oryzae using the fluorescent probe 7-azido-4-methylcoumarin (AzMC),
   with representative images shown. Bars, 10 μm. d Quantification of
   fluorescence intensity shown in c. Samples from different strains were
   measured using ImageJ software. The data represent the mean ± standard
   error of three biological replicates (n = 3). e Measurement of
   endogenous H[2]S levels at different developmental stages of the M.
   oryzae using a microplate reader. Data presented are the
   mean ± standard errors from three biological replicates (n = 3). f
   Determination of Cys content in different strains using a cysteine
   (Cys) content measurement kit. Data presented are the mean ± standard
   errors from three biological replicates (n = 3). g Detection of H[2]S
   levels in appressorium treated with NaHS and HT using the fluorescent
   probe 7-azido-4-methylcoumarin (AzMC), accompanied by representative
   images. Bars, 10 μm. h Quantification of fluorescence intensity shown
   in g. Samples from different strains were measured using ImageJ
   software. The data represent the mean ± standard error of three
   biological replicates (n = 3). i Detection of S-sulfhydration levels in
   wild-type (WT) and CSE1_OE strains using the tag-switch method. Total
   protein was incubated with a specified concentration of NaHS or
   dithiothreitol (DTT) (10 mM) for 20 minutes. Anti-biotin antibody was
   used for immunoblotting. j Quantification of S-sulfhydration levels
   shown in i. The data represent the mean ± standard error of three
   biological replicates (n = 3). k Detection of S-sulfhydration levels in
   wild-type (WT) after 100 µM NaHS or HT treatment using tag-switch
   method. Anti-biotin antibody was used for immunoblotting. l
   Quantification of S-sulfhydration levels shown in k. The data represent
   the mean ± standard error of three biological replicates (n = 3). Data
   are analyzed by two-way ANOVA followed by Tukey’s test (d, e, h, j, l)
   or Student’s t tests (f). Exact p (f) or adjusted p (d, e, h, j, l)
   values are shown in figures. Source data are provided as a Source data
   file.

   Given that both excessive and insufficient levels of H[2]S have an
   impact on the infection of M. oryzae, we successfully obtained two CSE1
   deletion mutants (KO1 and KO2) through homologous recombination and PCR
   screening (Supplementary Fig. [105]2a–c). Complementary transformants
   (cCSE1) were successfully obtained by introducing CSE1, driven by its
   native promoter, into the Δcse1 mutant. Additionally, we created three
   CSE1 overexpression strains, designated as CSE1_OE-5, CSE1_OE-8, and
   CSE1_OE-12, to further explore their biological roles. qRT-PCR analysis
   validated the overexpression of CSE1 in the three CSE1_OE mutants, with
   transcriptional levels increased by about 30 to 40-fold compared to the
   wild-type strain (Supplementary Fig. [106]3). Moreover, the GFP-CSE1
   fusion protein was detected in high levels within the cytoplasm
   throughout various developmental stages (Supplementary Fig. [107]4).
   This pattern of localization, which is consistent with that observed in
   animals and plants, further implies the protein’s role in diverse
   biological processes within the fungus.

CSE1 is responsible for H[2]S production and S-sulfhydration levels of M.
oryzae

   In animals and plants, H₂S synthesized by cystathionine γ-lyase (CSE1)
   mediates S-sulfhydration and takes part in various biological processes
   within cells. CSE1 facilitates the transformation of cysteine (Cys) and
   homocysteine into α-oxopropanoic acid and α-ketoglutaric acid,
   respectively, and in the process, releases H[2]S^[108]16. To determine
   if CSE1 similarly controls H[2]S production in M. oryzae, we utilized
   the 7-azido-4-methylcoumarin (AzMC) dye to assess H[2]S levels across
   different developmental phases of M. oryzae. Using confocal microscopy,
   we observed a substantial increase in fluorescence intensity in the
   CSE1_OE mutants as opposed to the wild-type strain (Fig. [109]2c, d).
   Consistent with this, compared with the wild-type, the H[2]S content of
   the CSE1_OE mutant was significantly increased at every developmental
   stage (Fig. [110]2e). Overexpression of CSE1 led to a roughly 20%
   decrease in substrate cysteine content relative to the wild-type
   (Fig. [111]2f). We also observed that the levels of H[2]S in ∆cse1 were
   significantly lower than those in the wild-type (Supplementary
   Fig. [112]5a–c). Moreover, the application of NaHS and HT led to
   respective increases and decreases in H[2]S levels in M. oryzae
   (Fig. [113]2g, h). Consequently, our findings confirm that CSE1 is the
   key enzyme for H[2]S production in M. oryzae.

   Previous studies have indicated that elevated H[2]S levels lead to
   increased S-sulfhydration modification^[114]37. To determine if H[2]S
   plays a similar role in S-sulfhydration in M. oryzae, we employed the
   tag-switch method^[115]38 to assess the difference in total
   S-sulfhydration modification levels between the wild-type and CSE1_OE
   mutants. The results showed that the S-sulfhydration level of total
   proteins in the CSE1_OE mutants was markedly higher compared to the
   wild-type (Fig. [116]2i, j), suggesting that CSE1 and H[2]S contribute
   to the regulation of total protein S-sulfhydration in M. oryzae.
   Additionally, NaSH and HT treatments respectively elevated and lowered
   S-sulfhydration levels (Fig. [117]2k, l), demonstrating that H[2]S
   modulates signaling in fungi through the post-translational
   modification of S-sulfhydration.

H[2]S signal negatively regulates the virulence of M. oryzae

   To assess the contribution of CSE1 to the virulence of M. oryzae, we
   inoculated barley and rice leaves with spore suspensions of the
   wild-type (WT) and Δcse1 strains. The results showed that, compared to
   WT, the Δcse1 strains produced more lesions on barley and rice leaves
   (Fig. [118]3a, b), and spread more rapidly on wounded rice leaves
   (Fig. [119]3c). We found that the Δcse1 strains exhibited no
   significant difference in colony growth rate compared to the WT
   (Supplementary Fig. [120]6a, b), but its infection hyphae spread faster
   than those of the WT (Supplementary Fig. [121]6c, d). Subsequently, we
   performed pathogenicity tests on barley and rice leaves using both the
   wild-type and CSE1_OE mutants. The results demonstrated that the
   CSE1_OE mutants had significantly lower pathogenicity on both barley
   and rice in comparison to the wild-type (Fig. [122]3d, e).
   Additionally, wounded rice leaves displayed smaller lesions induced by
   the CSE1_OE mutants (Fig. [123]3f). To understand the cause of the
   diminished pathogenicity in the CSE1_OE mutants, we examined their
   infection process within rice sheath cells, discovering that these
   mutants had impairments in both penetration and invasive growth
   (Fig. [124]3g–i). Overall, our findings indicate that CSE1 negatively
   regulates the virulence of M. oryzae. Consequently, we hypothesize that
   H[2]S stress within the fungus might reduce its infection efficacy in
   host cells.

Fig. 3. CSE1 negatively regulates the pathogenicity of M. oryzae.

   [125]Fig. 3
   [126]Open in a new tab

   a Barley seedling infection assay. Wild-type (WT) and ∆cse1 were
   inoculated onto barley leaves to form lesions, and the number of
   lesions was counted 5 days (dpi) after inoculation. b Rice seedling
   infection assay. Wild-type (WT) and ∆cse1 were inoculated onto rice
   leaves to form lesions, and the number of lesions was counted 5 days
   (dpi) after inoculation. c Wounded rice infection assay. The length of
   lesions was counted 5 days (dpi) after inoculation. d Barley seedling
   infection assay. Wild-type (WT) and CSE1_OE were inoculated onto barley
   leaves to form lesions, and the number of lesions was counted 5 days
   (dpi) after inoculation. e Rice seedling infection assay. Wild-type
   (WT) and CSE1_OE were inoculated onto rice leaves to form lesions, and
   the number of lesions was counted 5 days (dpi) after inoculation. f
   Wounded rice infection assay. The length of lesions was counted 5 days
   (dpi) after inoculation. g Observation of invasive hyphae growth of
   wild-type (WT) and CSE1_OE strains at 24 hpi and 36 hpi (hours after
   inoculation) in rice sheath cells. Bars, 10 μm. statistical analysis of
   invasive hyphae growth in rice sheath cells at 24 hpi (h) and 36 hpi
   (i). j Detection of ROS levels in wild-type and CSE1_OE mutant strains
   using dihydrorhodamine 123 (DHR 123) dye. Bars, 5 μm. k Quantification
   of fluorescence intensity shown in j. Samples from different strains
   were measured using ImageJ software. The data represent the
   mean ± standard error of eight biological replicates (n = 8). Data in
   a–f are displayed as box and whisker plots with individual data points:
   center line, median, first and third quantiles, maximum and minimum (a,
   n = 8 biologically repetitions; b, n = 9 biologically repetitions; c,
   f, n = 4 biologically repetitions; d, e, n = 13 biologically
   repetitions). Data are analyzed by one-way ANOVA (a–f) or two-way ANOVA
   (h, i) followed by Tukey’s test, or Student’s t tests (k). Exact p (k)
   or adjusted p (a–f, h, i) values are shown in figures. Source data are
   provided as a Source data file.

   In mammals, CSE1 is the predominant enzyme in H[2]S synthesis. CBS
   (cystathionine beta-synthase) coordinates the allocation of sulfur
   atoms between cysteine biosynthesis and degradation reactions, thereby
   indirectly regulating H[2]S production^[127]39–[128]42. To explore the
   contributions of other H[2]S pathway genes to fungal virulence, we
   obtained knockout mutants (KO1 and KO2) and complementary transformants
   of DES1 (cystathionine beta-synthase, MGG_04744) (Supplementary
   Fig. [129]7a–c). However, we were unable to obtain a CBS (MGG_07384)
   knockout mutants, suggesting that CBS may be essential for the normal
   life activities of M. oryzae. DES1 expression was downregulated during
   the infection stage compared to the mycelial stage (Supplementary
   Fig. [130]7d), and its deletion had no significant impact on fungal
   growth (Supplementary Fig. [131]8a, b). We noticed that deletion of
   DES1 in M. oryzae did not affect the production of H[2]S (Supplementary
   Fig. [132]8c). Importantly, pathogenicity assays revealed that DES1 is
   dispensable for M. oryzae virulence (Supplementary Fig. [133]8d–f).
   Collectively, these findings highlight CSE1 as the crucial H[2]S
   synthase for M. oryzae virulence.

   To elucidate the biological role of CSE1, we initially evaluated the
   vegetative growth of the CSE1_OE mutants. Following five days of
   inoculation, the growth rate of these mutants on OTA medium was found
   to be not significantly different from that of the wild-type
   (Supplementary Fig. [134]9a, b). Additionally, CFW staining indicated
   that the cell length of the CSE1_OE mutants was comparable to the
   wild-type (Supplementary Fig. [135]9c, d). Consistently, the spore
   production in the CSE1_OE mutants was found to be similar to that of
   the wild-type (Supplementary Fig. [136]9e, f). Collectively, these
   outcomes suggest that CSE1 does not influence the vegetative growth and
   conidiation of M. oryzae.

   To determine if the CSE1 gene of the rice blast fungus is pivotal in
   evading or suppressing host defense responses, we inoculated spore
   suspensions of both wild-type and CSE1_OE mutants onto barley leaves.
   After 30 hours, the leaves were subjected to DAB staining. We observed
   a significant increase in the accumulation of reactive oxygen species
   (ROS) in host cells infected with the CSE1_OE mutants compared to the
   wild type, a phenomenon that was reversed by hypotaurine (HT) treatment
   (Supplementary Fig. [137]10a, b). This indicates that H[2]S stress
   caused by the overexpression of CSE1 in M. oryzae induces a ROS burst
   within the host. S-sulfhydration alters the redox status and function
   of target proteins through a reversible oxidative post-translational
   modification of protein thiol groups^[138]43. The cellular redox
   environment is typically assessed by the GSH/GSSG ratio. Consequently,
   we compared the GSH/GSSG ratios in the WT and CSE1_OE mutants and found
   that the ratio in the CSE1_OE mutants was significantly lower than in
   the WT (Supplementary Fig. [139]10c). We also investigated whether
   H[2]S stress in M. oryzae leads to an increase in intracellular ROS and
   discovered significantly higher ROS levels in the CSE1_OE mutants than
   in the WT using a dihydrorhodamine 123 (DHR 123) dye assay, suggesting
   that H[2]S accumulation in M. oryzae triggers a surge in cellular ROS
   (Fig. [140]3j, k). These findings suggest that H[2]S signaling is a
   crucial regulator of redox homeostasis in M. oryzae.

H[2]S signal negatively regulates the autophagy process of M. oryzae

   It has been demonstrated that elevated levels of H[2]S are harmful to
   M. oryzae (Fig. [141]1). To explore the link between the reduced
   infectivity of the CSE1_OE mutants and H[2]S levels within their
   infection structures, we measured H[2]S levels throughout the
   development of M. oryzae appressoria. Our findings showed that in the
   wild-type, peak H[2]S levels occurred during the germination phase,
   with H[2]S signaling progressively moving from the conidia to the
   maturing appressoria (Fig. [142]4a, b). This pattern of H[2]S migration
   was similar to that observed for reactive oxygen species (ROS) and
   reactive nitrogen species (RNS) signaling in appressoria. The CSE1_OE
   mutants, however, exhibited higher H[2]S levels than the wild-type
   during appressorial development (Fig. [143]4a, b). These results
   suggest that an overabundance of H[2]S during appressorial maturation
   hinders the infection ability of M. oryzae.

Fig. 4. H[2]S-mediated S-sulfhydration negatively regulates autophagy of M.
oryzae.

   [144]Fig. 4
   [145]Open in a new tab

   a Detection of endogenous H[2]S levels in the M. oryzae using the
   fluorescent probe 7-azido-4-methylcoumarin (AzMC), with representative
   images presented. Bars, 15 μm. b Quantification of fluorescence
   intensity shown in a. Samples from different strains were measured
   using ImageJ software. The data represent the mean ± standard error of
   three biological replicates (n = 3). c Statistics of appressorium
   formation rates of WT and CSE1_OE strains at 24 h. The data represent
   the mean ± standard error of six biological replicates (n = 6). d
   Observation of turgor pressure formation in wild-type (WT) and CSE1_OE
   strains in 2 M glycerol. Bars, 30 μm. e Statistical analysis of the
   appressorium collapse rates of different strains in 1 M or 2 M
   glycerol. Data are displayed as box and whisker plots with individual
   data points: center line, median, first and third quantiles, maximum
   and minimum (n = 6 biologically repetitions). f Autophagy processes
   detected by observation of ATG8 subcellular localization. Green
   fluorescent protein (GFP)-Atg8 was transformed into wild-type (WT) and
   CSE1_OE mutants and cultured in MM-N medium for 5 h, with arrows
   pointing to the vacuoles. Bars, 5 μm. g Autophagy intensity assessed by
   the proportion of GFP-Atg8 translocated to vacuoles. The data represent
   the mean ± standard error of three biological replicates (n = 3). h
   Immunoblot analysis conducted using anti-GFP methods after nitrogen
   deprivation treatment, with anti-GAPDH antibody used as a control. i
   The extent of autophagy was estimated by calculating the ratio of free
   GFP to the total amount of intact GFP-Atg8 and free GFP. The data
   represent the mean ± standard error of three biological replicates
   (n = 3). j Cellular localization of autophagosomes during conidial
   development. Different strains of conidia were inoculated onto a
   hydrophobic interface and observed at different time points using
   fluorescence microscopy. Bars, 10 μm. k Statistical analysis of the
   average number of autophagosomes in conidia, germ tubes, and
   appressorium at 0, 2, and 4 hours after germination. The data represent
   the mean ± standard error of three biological replicates (n = 3). Data
   are analyzed by two-way ANOVA (b, e, g, i, k) or one-way ANOVA (c)
   followed by Tukey’s test. Adjusted p-values are shown in figures.
   Source data are provided as a Source data file.

   Given the impaired penetration ability of the CSE1_OE mutants in host
   cells, we hypothesize that their appressoria may not accumulate
   adequate turgor pressure for penetration. We found no significant
   difference in appressorium formation between the wild-type and CSE1_OE
   mutants (Fig. [146]4c), leading us to suspect that the appressoria of
   the CSE1_OE mutants might not be fully functional. To test this, we
   exposed the CSE1_OE mutants’ appressoria to varying concentrations of
   glycerol and measured their turgor pressure. The findings indicated
   that the appressoria of the CSE1_OE mutants were more susceptible to
   collapse under glycerol treatment (Fig. [147]4d, e), which suggests
   that CSE1 plays a role in negatively regulating the maturation of M.
   oryzae appressoria.

   Autophagy, similar to MAPK and cAMP signaling pathways, is a crucial
   process for appressorium turgor accumulation^[148]44,[149]45. Given the
   defects in turgor accumulation observed in the CSE1_OE mutants, and the
   fact that sulfide generally negatively regulates autophagy across
   different species^[150]46, we explored the potential of CSE1 to
   regulate autophagy in M. oryzae. We introduced GFP-Atg8 into both the
   wild-type and CSE1_OE strains and subjected the transformed mycelia to
   nitrogen starvation (MM-N). Following a 5-hour incubation in MM-N, we
   noticed a reduced localization of GFP-Atg8 in the vacuoles of CSE1_OE
   cells as compared to the wild-type cells (Fig. [151]4f, g). This
   finding indicates that the autophagy process is disrupted in the
   CSE1_OE mutants.

   Additionally, we assessed the level of autophagy using western blot
   analysis. Autophagy can be quantified by determining the ratio of free
   GFP to the sum of GFP-Atg8 and free GFP (GFP/[GFP + GFP-Atg8]). Upon
   treating the wild-type (WT) mycelia with nitrogen starvation medium
   (MM-N), this ratio notably increased from 0.31 at 0 hours to 0.82 at
   5 hours. Conversely, in the CSE1_OE strains, the ratio only slightly
   rose from 0.32 at 0 hours to 0.59 at 5 hours (Fig. [152]4h, i).
   Collectively, these outcomes illustrate that H[2]S-mediated
   S-sulfhydration significantly regulates autophagy in M. oryzae. Given
   the importance of S-sulfhydration in both plants and animals, our data
   imply that the role of S-sulfhydration in regulating autophagy is
   highly conserved among different species. To evaluate autophagy within
   appressoria, we analyzed the formation of autophagosomes in both WT and
   CSE1_OE strains. Initially, autophagosomes labeled with GFP-Atg8
   predominantly accumulated in WT spores. Throughout appressorium
   formation and maturation, the number of autophagosomes marked with
   GFP-Atg8 was typical in the WT strain but considerably higher in
   CSE1_OE (Fig. [153]4j, k). We therefore propose that the accumulation
   of H[2]S due to CSE1 overexpression leads to an increase in
   autophagosomes, ultimately impacting autophagic homeostasis.

Quantitative proteomics unveils numerous S-sulfhydrated proteins in M. oryzae

   To elucidate the role of S-sulfhydration in fungal pathogenesis and to
   pinpoint the S-sulfhydrated proteins within fungi, we established a
   technology aimed at identifying proteins that have undergone
   S-sulfhydration. We profiled global protein S-sulfhydration
   modifications, and total proteins extracted from wild-type and CSE1_OE
   mutants were subjected to sequential processing. Free thiols were first
   alkylated with 1 mM IAA-PEO-biotin in darkness at 25 °C for 1 h,
   followed by overnight tryptic digestion at 37 °C. Peptide samples
   (0.5 mg per replicate) underwent two rounds of thiol-affinity
   enrichment using iodoacetyl resin to capture S-sulfhydrated peptides.
   After stringent washing, bound peptides were eluted and re-alkylated
   with 30 mM IAM for 30 min to prevent disulfide reformation. Finally,
   the peptides were analyzed using liquid chromatography-tandem mass
   spectrometry (LC-MS/MS) (Fig. [154]5a). Across three biological
   replicates, the number of quantifiable sites was 6986, with a total of
   10348 sites identified, yielding a reproducibility rate of
   approximately 67.51%. In the WT group, the number of reproducible sites
   was 5925, accounting for about 57.26%, while in the CSE1_OE group, the
   number of reproducible sites was 7117, representing approximately
   68.78% (Fig. [155]5b). We identified a total of 1054 S-sulfhydrated
   proteins (Supplementary Data [156]1), with 997 being upregulated
   modifications (94.59%), and 57 being downregulated proteins (5.41%).
   Among the targeted proteins we identified, there were 1482
   S-sulfhydration modification sites, with 1419 being upregulated sites
   (95.75%), and 63 being downregulated sites (4.25%) (Fig. [157]5c, d).

Fig. 5. Identification of S-sulfhydrated proteins in M. oryzae.

   [158]Fig. 5
   [159]Open in a new tab

   a Quantitative proteomics workflow diagram for the identification of
   H[2]S-mediated S-sulfhydrated proteins. For detailed notes on these
   proteins, see Supplementary Data [160]1. b Overview of identification
   of S-sulfhydration modification sites. c Box plot of RSD of protein
   quantitation values between three biological replicates samples of
   wild-type (WT) and CSE1_OE strains. Data are displayed as box and
   whisker plots with individual data points: center line, median, first
   and third quantiles, maximum and minimum. d Statistics on the number of
   S-sulfhydrated target proteins and the number of S-sulfhydrated
   modification sites. e Statistics of the number of differentially
   up-regulated and down-regulated S-sulfhydrated proteins in different
   subcellular structural species. f Heat map was used to show the
   frequency changes of amino acids near modification sites. g Analysis of
   the enrichment of protein domains with S-sulfhydration. The enrichment
   bubble plot illustrates the most significantly enriched functions, with
   the color of each bubble indicating the significance level (p value) of
   the enrichment. The Pfam database was employed to analyze the
   enrichment of functional domains in differentially expressed modified
   proteins. Fisher’s exact test was utilized to assess the significance
   of domain enrichment in differentially expressed modified proteins,
   using the identified proteome as the background. A p value of less than
   0.05 was considered statistically significant. h GO analysis of
   significantly altered S-sulfhydrated proteins by biological process,
   cellular component and molecular function. i Analysis of the enrichment
   of S-sulfhydrated proteins within KEGG signaling pathways. The
   enrichment bubble plot depicts the most significantly enriched
   functions, with the color of each bubble indicating the corresponding p
   value for enrichment significance. The Kyoto Encyclopedia of Genes and
   Genomes (KEGG) database was used to perform pathway enrichment analysis
   of differentially expressed modified proteins. Fisher’s exact test was
   employed to evaluate the significance of pathway enrichment for
   differentially expressed modified proteins, using the identified
   proteome as the background. A p value of less than 0.05 was considered
   statistically significant. j S-sulfhydrated proteins associated with
   appressorium and pathogenicity of M. oryzae are listed in Table.

   Statistical analysis of the number of differentially upregulated and
   downregulated proteins in various subcellular compartments showed that
   there were 353 upregulated and 21 downregulated proteins in the
   cytoplasm; 280 upregulated and 13 downregulated proteins in the
   mitochondria; 144 upregulated and 15 downregulated proteins in the
   nucleus; 84 upregulated and 3 downregulated proteins in the
   extracellular; and 49 upregulated and 1 downregulated proteins in the
   plasma membrane (Fig. [161]5e). To characterize sequence preferences