PsDMAP1/PsTIP60-regulated H4K16ac is required for ROS-dependent virulence adaptation of Phytophthora sojae on host plants

Significance

Our findings underscore the crucial involvement of epigenetic-associated proteins in the intricate interplay between chromatin remodeling and DNA repair mechanisms under reactive oxygen species (ROS) stress conditions. Specifically, PsDMAP1/PsTIP60 mediated H4K16ac acts as an epigenetic mark that mediates the transmission of oxidative stress adaptation to asexual and sexual progeny in Phytophthora sojae, enhancing offspring resilience to host plant defense mechanisms and increasing their resistance to fungicides. Moreover, the epigenetically inherited adaptation to ROS allows P. sojae on plants to exhibit greater resistance to fungicides compared to P. sojae cultured on artificial media under field conditions. Significantly, our results offer fresh insights and strategic avenues for mitigating oomycete diseases and addressing challenges associated with agricultural fungicide resistance.

Abstract

Host plants and various fungicides inhibit plant pathogens by inducing the release of excessive reactive oxygen species (ROS) and causing DNA damage, either directly or indirectly leading to cell death. The mechanisms by which the oomycete Phytophthora sojae manages ROS stress resulting from plant immune responses and fungicides remains unclear. This study elucidates the role of histone acetylation in ROS-induced DNA damage responses (DDR) to adapt to stress. Mechanistically, the P. sojae DNA methyltransferase 1-associated protein (PsDMAP1) binds Tat-interactive protein 60 (PsTIP60) to comediate histone H4 acetylation on lysine 16 (H4K16ac). This regulation affects RNA polymerase II (pol II) recruitment, transcriptional induction of DDR-related genes, and the enrichment of histone H2Ax phosphorylated on serine 137 (γH2Ax) in response to both plant immunity and fungicide stress. The resulting H4K16ac serves as a crucial transgenerational epigenetic signal for virulence adaptation of P. sojae on plants, as a result of adaptation to ROS stress.

Reactive oxygen species (ROS), including a range of oxygen-containing highly reactive species such as oxygen anions, free radicals, and hydrogen peroxide (H₂O₂), are by-products of aerobic respiration that can damage DNA, proteins, and lipids (1). Eukaryotic organisms have developed multiple systems to balance the production and response to ROS. In the arms race between hosts and pathogens, ROS are generated and accumulated by hosts upon the perception of pathogen-related molecular patterns, triggering immune responses such as pattern-triggered immunity and effector-triggered immunity in animals and plants (2, 3). In addition to their roles as secondary messengers, ROS are also considered primary antimicrobial agents that inhibit pathogens from invading plants (2). Furthermore, fungicides like azoxystrobin (AZX), acting on the respiratory electron transport chain, are widely used in agricultural production to control fungal or oomycete diseases (4). AZX inhibits oxidative phosphorylation by targeting mitochondrial respiratory complex III, leading to ROS overproduction and DNA damage in target organisms (4, 5). However, the mechanisms by which plant pathogens such as oomycetes, have evolved to counter ROS stress in response to host defense responses and fungicides remain poorly understood.

ROS exposure can cause oxidative damage to DNA, leading to the formation of single- or double-strand breaks (DSB) in severe circumstances (6). ROS-induced DNA damage responses (DDR) in yeast and mammals include base excision repair, double-strand break repair, nucleotide excision repair, and mismatch repair pathways (7). These processes are primarily regulated by DDR proteins including the poly (ADP-ribose) polymerase (PARP) family (8, 9). PARP1 has been shown to mediate the initial enrichment of the MRE-Rad50-Nbs1 (MRN) complex at DNA damage sites (10). Recruitment of ATM by MRN and PARP1 subsequently leads to the activation of histone H2Ax by its phosphorylation on serine 139 within a serine-glutamine motif (forming γH2Ax), and recruitment of complexes of DDR factors at sites of DNA damage (11–13). The presence of γH2Ax allows the DNA damage signal to spread along the chromatin covering large regions of the chromosome surrounding each DNA lesion, thus providing a platform for recruitment of other proteins that participate in DDR (14, 15) However, the regulatory mechanisms of ROS-induced DDR and repair pathways remain largely unknown in oomycetes.

Microbial pathogens encounter dynamic and challenging microenvironments during infection of fields protected by spraying fungicides. In general, comprehensive transcription reprogramming enables pathogens to precisely cope with varying host environments (16, 17). Histone acetylation modifications, especially histone H4 acetylation on lysine 16 (H4K16ac), are crucial epigenetic regulatory mechanisms promoting transcriptional activation in eukaryotic cells by reducing chromatin condensation (18, 19). The human Tat-interacting protein-60 kD (TIP60) is a core catalytic subunit of the highly conserved nucleosome histone H4 acetyltransferase (NuA4) complex (20), which is a predominant histone acetyltransferase (HAT). TIP60 plays vital roles in DNA repair, checkpoint control of the p53-regulated cell cycle, apoptosis, autophagy, senescence, and tumorigenesis and metastasis (20–24). Recently, histone acetylation has been shown to play a critical role in transcriptional reprogramming during activation of plant defense systems against pathogens (24, 25). In Drosophila and mammals, maternal H4K16ac can provide an instructive function to the progeny, priming future gene activation (19). Although it has been well documented that histone acetylation is critical for regulation of gene transcription during the responses of eukaryotes to changing environments, the potential molecular functions of Tip60/H4K16ac, especially in pathogens during stress tolerance and infection, remain to be determined.

The DNA methyltransferase 1-associated protein (DMAP1) was first recognized as a key activator of 5-methylcytosine DNA methyltransferase 1 (DNMT1) that is responsible for regulating gene expression by interacting with histone deacetylase 2 (HDAC2) (26). Subsequently, DMAP1 was identified to be a component of the TIP60-P400 HAT complex (27, 28), but its biochemical and physiological significance in this complex is yet to be determined. In humans, DMAP1 participates in repairing DNA DSB and suppressing tumors (29, 30).

Phytophthora spp. is a group of pathogens causing destructive plant diseases (31–34). Among these, Phytophthora sojae is a particularly notorious member, causing root and stem rot in soybeans, resulting in annual economic losses of approximately $1 billion (32, 35). Recently, the Soybean-P. sojae interaction has emerged as an important pathosystem for studying the pathogenesis of oomycetes. Previous studies have shown that knockdown of PsDMAP1 caused a pronounced reduction in mycelial growth, production of sporangia and zoospore, cystospore germination, ROS and osmotic stress tolerance, and virulence of P. sojae (36). Therefore, it was of interest to determine whether DMAP1 plays a role in ROS-induced DNA damage repair and its related mechanisms in plant pathogenic oomycetes, such as P. sojae. Here, we present a mechanism by which P. sojae can tolerate stress when exposed to high levels of ROS generated by host plants and fungicides. Specifically, we found that PsDMAP1-regulated acetylation of histone H4K16 by PsTIP60 maintains an “open” chromatin structure and upregulates DNA repair-related genes in response to ROS-induced DNA damage. The PsDMAP1 and PsTIP60 proteins work together to mediate H4K16 acetylation, which then regulates downstream DDR-related genes. This work demonstrates how oomycetes can adapt to ROS stresses derived from the host and fungicides at an epigenetic.

Results

PsDMAP1 Is A regulator of DDR in P. sojae.

In order to understand the role of ROS in host defense against pathogen invasion, we first determined ROS production in soybean seedlings after P. sojae infection using the ROS-sensitive fluorescent probe DCFH-DA (2′,7′-Dichlorodihydrofluorescein diacetate) (37). As shown in Fig. 1A, abundant ROS was observed in infected soybean seedlings at 24 h postinoculation (hpi) with P. sojae mycelia. When treated with the NADPH oxidase inhibitor diphenyleneiodonium (DPI) at 1 μM, which selectively inhibits plasma membrane NADPH oxidase and generation of extracellular ROS in both plant and mammalian cells (38), ROS accumulation was abolished in infected soybean seedlings inoculated with P. sojae (Fig. 1A). These results indicated that P. sojae infection provokes ROS burst in host tissues.

Fig. 1.

PsDMAP1 mediates the response of P. sojae to ROS stress. (A) ROS produced in P. sojae-infected soybean seedlings. Seven-day-old soybean seedlings were inoculated with mycelial plugs of P. sojae and then stained at 24 hpi by DCFH-DA with or without 1 μM DPI as ROS scavenger. Bar: 50 μm. (B) 1 µg/mL (ppm) AZX induced ROS production in P. sojae. Each strain was cultured in V8 liquid medium with 0.1% DMSO, 1 ppm AZX, or 1 ppm AZX+1 μM DPI for 48 h. The resulting mycelia of each strain were stained with DCFH-DA. ROS fluorescence was detected by confocal microscopy. Bar: 20 μm. (C) PsDMAP1 knockdown mutants showed increased sensitivity to H₂O₂ and AZX. A 5-mm mycelial plug from the wild-type P6497, empty vector strain (EV), the PsDMAP1 knockdown mutants (DMAP1-1 and DMAP1-2), and the PsDMAP1-overexpressed strain (DMAP1-OE) was incubated on V8 plates supplemented with 0.1% DMSO, 10 mM H₂O₂, 1 ppm AZX, 0.1% DMSO+1 μM DPI, 10 mM H₂O₂+1 μM DPI, or 1 ppm AZX+1 μM DPI for 5 d. (D) Mycelial growth inhibition of the H₂O₂ or AZX relative to the DMSO-treated control and mycelial inhibition by H₂O₂+DPI or AZX+DPI relative to the DMSO+DPI-treated control. Relative inhibition was calculated as (CK−Growth rate on plates with treatment)/CK×100%. Statistical significance of the mycelial inhibition of above strains compared to the wild type P6497 was determined using Student’s t test (**P < 0.01, ns: not significant). (E) The treatment with 1 μM DPI restored the impaired virulence of PsDMAP1 knockdown strains. Left panels: soybean seedlings were inoculated with mycelial plugs of each strain and then stained at 24 hpi by DCFH-DA with or without 1 μM DPI as ROS scavenger. Right panels: disease symptoms at 48 hpi of above strains when soybean seedlings were treated with H₂O (control) or DPI. (F) Lesion length and (G) pathogen biomass of each strain under treatments with H₂O (control) or DPI, measured at 48 hpi. Statistical significance of the lesion length and biomass of above strains compared to the P6497 at H₂O or DPI treatment was determined using Student’s t test (**P < 0.01, ns: not significant). Data in D, F, and G are presented as the mean ± SD from three biological replicates.

Fungicides, such as the widely used AZX, target the respiratory electron transport chain to combat fungal or oomycete diseases. Their primary protective effects occur through excessive ROS release within the pathogen and subsequent DNA damage that ultimately results in cell death (4, 5). Understanding how pathogens adapt to drug stress is crucial for effectively controlling harmful microorganisms and managing fungicide resistance. To determine the potential involvement of ROS in adaptation to stress induced by AZX in oomycetes, we measured ROS production in P. sojae mycelia under AZX treatment. In P. sojae, AZX also induced ROS accumulation (Fig. 1B). Previous studies have shown that PsDMAP1 is crucial for P. sojae’s response to both ROS and osmotic stresses (36). Thus, it was pertinent to explore whether DMAP1 contributes specifically to host defense mechanisms and AZX-induced ROS responses. Utilizing the recently updated P. sojae genome database (v3.0), we identified one ortholog of Homo sapiens DMAP1 encoded in the P. sojae genome (39) by bidirectional blastp searches with an E-value cut-off of 10⁻¹⁰. This gene has been designated as PsDMAP1 (protein ID: 559488), comprising a length of 491 amino acids. To determine whether PsDMAP1 participates in the host defense- and AZX-induced ROS responses, we aimed to delete this gene from the P. sojae genome by using CRISPR/Cas9 technology (40, 41). Although 297 heterozygous individuals were obtained, we never succeeded in obtaining homozygous full-length PsDMAP1 gene knockout mutants. This result suggested that complete deletion of PsDMAP1 might be lethal in P. sojae. Therefore, we generated PsDMAP1 knockdown mutants (DMAP1-1 and DMAP1-2) and PsDMAP1-overexpressing transformants (DMAP1-OE) (SI Appendix, Fig. S1 A–C). Notably, the relative expression of PsDMAP1 in DMAP1-OE strain was increased when under the treatment of H₂O₂ and AZX compared to CK (SI Appendix, Fig. S1C). PsDMAP1 knockdown mutants showed hypersensitivity to H₂O₂ and AZX compared to the wild-type (WT) P6497, EV, and DMAP1-OE strains (Fig. 1 C and D). DPI treatment restored the insensitivity of PsDMAP1 knockdown mutants to H₂O₂ and AZX (Fig. 1 C and D). To confirm the possible relationship between the reduced virulence of the PsDMAP1 knockdown mutants and their increased sensitivity to H₂O₂, we conducted DPI inhibition tests. DPI treatment significantly inhibited ROS accumulation during infection stages (Fig. 1E) and substantially compensated for the virulence deficiency of PsDMAP1 knockdown mutants (Fig. 1 E–G). These results suggested that the decreased virulence of PsDMAP1 knockdown mutants may be associated with their increased sensitivity to H₂O₂. PsDMAP1 is essential for resistance to host defense-produced and AZX-induced ROS stress.

Many enzymes are involved in antagonizing ROS levels. For example, catalase (CAT) mediates H₂O₂ breakdown, and superoxide dismutase (SOD) specifically scavenges noxious oxygen and nitrogen free radicals (6). To explore the mechanism by which PsDMAP1 regulates the ROS tolerance, we performed the RT-qPCR, and found that DDR related genes rather than antioxidation genes were significantly down-regulated in the DMAP1-1 mutant compared to P6497 following H₂O₂ treatment (Fig. 2A). To determine whether PsDMAP1 indirectly controlled the resistance to host defense-produced and AZX-induced ROS stress by regulating ROS clearance, assays for SOD enzyme activity, and RT-qPCR assays were performed. We observed that transcript levels of CAT and SOD homologs and SOD catalytic activity in PsDMAP1 knockdown mutants were not decreased compared to P6497, EV, and DMAP1-OE strains (SI Appendix, Fig. S2 A and B). However, the transcript levels of genes involved in DNA damage repair pathways (Dataset S1), such as the ADP-ribosyltransferase, helicase/hydrolase family, and DNA-directed DNA polymerase, specifically the genes PsPARP1 (protein ID: Ps337921), PsASF1 (protein ID: Ps354581), PsOGG1 (protein ID: Ps319596), PsRad52 (protein ID: Ps532481), PsFPG (protein ID: Ps309246), and PsRecQ (protein ID: Ps485995), were significantly induced at the infection stage and treatment of AZX, but this induction was significantly decreased in the PsDMAP1 knockdown mutants compared to the P6497, EV, DMAP1-OE strains (Fig. 2B). Furthermore, These DDR-related genes were not induced during infection of plants treated with DPI and at the treatment of AZX and DPI (SI Appendix, Fig. S2C). These data suggested that the lower tolerance to H₂O₂ of the PsDMAP1 knockdown mutant did not result from the down regulation of the catalase and SOD activity but to reduced capability to repair H₂O₂-induced oxidative damage of DNA. Indeed, upon treatment with DNA-damaging agents, such as alkylating agent mitomycin C (MMC), DNA double-strand break inducer zeocin, and DNA synthesis inhibitor hydroxyurea (HU), and Ultraviolet (UV) the PsDMAP1 knockdown mutants exhibited hypersensitivity compared to P6497, EV, and DMAP1-OE (Fig. 2C and SI Appendix, Fig. S3 A and B). The transcript level of PsDMAP1 was significantly induced by MMC, Zeocin, HU, and UV (SI Appendix, Fig. S3C). Notably, the relative expression of PsDMAP1 in DMAP1-OE strain was increased when under the treatment of H₂O₂ and AZX compared to CK (SI Appendix, Fig. S3C). Then we also determined the protein stability of PsDMAP1 using synthesis inhibitor 10 μM cycloheximide (CHX). When CHX treatment suppressed new protein synthesis, the degradation rate of PsDMAP1 under H₂O₂ and MMC treatment was not significantly different from that under control treatment (SI Appendix, Fig. S3D). To elucidate whether these PsDMAP1-dependent DDR-related genes are genetically involved in relieving ROS, single knockout mutants of these genes were constructed (SI Appendix, Fig. S4 A–C). The resulting mutants were tested for sensitivity to H₂O₂, AZX, and MMC on V8 medium. We found that these mutants exhibited significantly increased sensitivity to H₂O₂, AZX, and MMC (SI Appendix, Fig. S4D), and exhibited reduced virulence on soybean seedlings (SI Appendix, Fig. S4E). Furthermore, we also respectively generated the overexpression strain for PsPARP, PsASF, PsOGG, PsRad52, PsFPG, and PsRecQ in the PsDMAP1-silenced mutant DMAP1-1 (SI Appendix, Fig. S4 F and G). The sensitivity of these strains to H₂O₂, AZX, and MMC was subsequently evaluated. The results indicate that overexpressing these DDR-related genes dependent on PsDMAP1 in the PsDMAP1-silenced mutant either fully or partially rescued their sensitivity to H₂O₂, AZX, and MMC (SI Appendix, Fig. S4H), thereby confirming their roles in DDR.

Fig. 2.

PsDMAP1 mediates activation of DDR-related genes by oxidative stress, DNA damage, and plant infection. (A) Hierarchical clustering of nine DNA repair-related genes and five antioxidative stress–related genes at treatment of 10 mM H₂O₂. Colors represented the log₂ (Fold difference of expression) of the genes (Upper). A list of GO terms enriched in the clusters is shown (Lower). (B) The transcript levels of DNA damage repair genes were evaluated by RT-qPCR in the P6497, EV, DMAP1-1, DMAP1-2, and DMAP1-OE in soybean seedlings at 48 hpi, after treatment with 0.1% DMSO in culture (mock) or with 1 ppm AZX in culture. The PsACTB gene was used as internal references. Transcript levels of DDR genes in wild-type P6497 under DMSO treatment were set as 1.0. Different letters represent significant differences by one-way ANOVA (P < 0.05). (C) PsDMAP1 knockdown mutants exhibited increased sensitivity to DNA-damaging agents. Upper panel: Mycelial growth of P6497, EV, DMAP1-1, DMAP1-2, and DMAP1-OE on V8 agar medium in the presence of 10 ppm MMC or 10 ppm Zeocin. Mycelial growth of above strains on V8 agar medium with 0.1% DMSO was used as control. Lower panel: Inhibition ratio was calculated as (Control−Growth rate on plates with treatment)/Control growth × 100% after 5 d. Statistical significance compared to the wild-type P6497 was determined using Student’s t test (**P < 0.01, ns: not significant). (D) PsDMAP1 knockdown inhibited γH2Ax accumulation in response to DNA damage signals (10 Mm H₂O₂, 1 ppm AZX, and 10 ppm MMC). γH2Ax levels were determined by western blotting using anti-γH2Ax antibodies. Histone H2A protein was used as the loading control. The intensity of the γH2Ax band from P6497 treated with DMSO was set as 1.00; and the relative intensity of γH2Ax band from each treatment was quantified with Image J. (E) 8-hydroxy-2 deoxyguanosine (8-OHDG) levels were elevated in DMAP1-1 and DMAP1-2 compared that in wild-type P6497, EV, and DMAP1-OE after treatment with 10 mM H₂O₂ (Upper) or 1 ppm AZX (Lower). Statistical significance compared to P6497 was determined using Student’s t test (**P < 0.01). Data in B, C, and E are presented as the mean ± SD from three biological replicates.

One fundamental mechanism for maintaining genome integrity is γH2Ax activation and enrichment at DNA damage sites in an ATM-dependent manner, recruiting repair factors to the break sites (42, 43). To determine whether the high sensitivity of PsDMAP1 mutants to DNA-damaging agents was related to the accumulation of DNA damage and loss of their repair ability, γH2Ax abundance was examined in WT and mutant strains exposed to H₂O₂, AZX, and MMC. Notably, we found that H₂O₂, AZX, and MMC significantly induced accumulation of γH2Ax in comparison to the untreated wild-type P6497 strain. PsDMAP1 knockdown resulted in a significant decrease in γH2Ax enrichment (Fig. 2D). In nuclear and mitochondrial DNA, 8-hydroxy-2′-deoxyguanosine (8-OHdG) is one of the predominant forms of free radical-induced oxidative lesions, and has therefore been widely used as a biomarker for oxidative stress (44). Therefore, we detected the 8-OHdG content in the above transformants using the 8-OHdG ELISA Kit. The results showed that significantly more 8-OHdG was accumulated in the PsDMAP1 knockdown mutants compared to P6497, EV, and DMAP1-OE following H₂O₂ or AZX treatment at various time points (Fig. 2E). These results indicated that PsDMAP1 plays an important role in responses to ROS-induced DDR.

PsDMAP1 Interacts with PsTIP60 in Response to ROS-Induced DDR.

To understand the molecular mechanism of PsDMAP1 in maintaining the expression of DDR-related genes, we conducted yeast two-hybrid (Y2H) screening to identify potential PsDMAP1-interaction proteins using a P. sojae cDNA library as prey and PsDMAP1 as a bait. Given that PsDMAP1 regulated gene expression, we targeted candidate PsDMAP1-interacting proteins that were potentially involved in epigenetic regulation processes. Among the 24 candidates (Dataset S2), we found that PsDMAP1 might interact with PsTIP60 (protein ID: Ps501103), a homolog of Drosophila melanogaster TIP60, which is a core component of a histone acetyltransferase (HAT) NuA4/TIP60 complex (45). The interaction between PsDMAP1 and PsTIP60 was further validated using yeast two-hybrid (Y2H), GST pull-down, and bimolecular fluorescence complementation (BiFC) assays (Fig. 3 A–C). Additionally, Y2H assays with a series of truncated PsDMAP1 constructs and GST pull-down experiments revealed that the SANT domain of PsDMAP1 is essential for its interaction with PsTIP60. Coimmunoprecipitation (Co-IP) assays provided further evidence that PsDMAP1 associates with PsTIP60 via its SANT domain in vivo, and that the interaction is enhanced upon treatment with H2O2, AZX, and MMC (Fig. 3D and SI Appendix, Fig. S5A). In addition, we generated P6497 transformants overexpressing PsDMAP1-GFP and PsTIP60-mCherry and examined the intensity of GFP and mCherry fluorescence and protein levels during the treatment of H₂O₂ and MMC. Confocal microscopy and western blot assays showed that fluorescence intensities and protein levels of both PsDMAP1-GFP and PsTIP60-mCherry were obviously elevated after treatments with H₂O₂ and MMC compared to DMSO treatment (Fig. 3E).

Fig. 3.

PsDMAP1 physically interacts with PsTIP60. (A) Y2H assay of PsDMAP1 interaction with PsTIP60. Yeast cells carried indicated vectors. Interactions were determined by growth on synthetic defined (SD) medium lacking tryptophan (Trp), leucine (Leu), histidine (His), and adenine (Ade) (SD-Trp-Leu-His-Ade). pGBKT7-53 and pGADT7 were used as positive control. pGBKT7-Lam and pGADT7 were used as negative control. The experiment was repeated three times independently with similar results. (B) The SANT domain of PsDMAP1 was required for PsDMAP1–PsTIP60 interaction in vitro in GST pull-down assay. GST, PsDMAP1-GST, or PsDMAP1^ΔSANT-GST immobilized on GST beads was incubated with PsTIP60-His proteins. Red asterisk to indicate our target band. (C) Bimolecular fluorescence complementation (BiFC) assay showing that PsDMAP1–PsTIP60 complexes localize in the nucleus. Nuclear localization was confirmed by simultaneous nuclear staining with DAPI. [Scale bar: 10 μm (Upper)]. Colocalization of proteins was analyzed by line-scan graph analysis. The mean line indicates the analyzed area. y axis: the intensity of GFP and DAPI signals quantified with ImageJ; x axis: the distance (μm) (Lower). (D) The association of PsDMAP1 with PsTIP60 was assessed by coimmunoprecipitation (Co-IP). Proteins were extracted from the P6497 transformants coexpressing PsDMAP1-GFP and PsTIP60-FLAG constructs or either single construct during growth in liquid V8, and were immunoprecipitated with anti-GFP agarose beads; then immunoblotted with anti-FLAG or anti-GFP antibodies. Western blots were anti-β-tubulin antibodies were used to monitor input loading and contamination of precipitated proteins. (E) PsTIP60-mCherry colocalizes with PsDMAP1-GFP within the nucleus. Fluorescence of dual-labeled strains was observed under a confocal microscope after incubation in V8 medium with 0.1% DMSO, 10 mM H₂O₂, and 10 ppm MMC, respectively. Left panels, GFP; Center panels, mCherry; Right panels, merge with differential interference contrast (Left). Bar = 20 μm. The levels of PsDMAP1-GFP and PsTIP60-mCherry proteins were detected by western blotting with anti-GFP and anti-mCherry antibodies. The relative intensity band from each treatment was quantified with Image J, and the intensity of the DMSO treatment was set as 1.00. GAPDH served as a loading control (Right).

PsTIP60 Is Important for Growth, Sporangium Production, and Virulence of P. sojae.

To confirm the role of PsTIP60 in regulating tolerance to respiratory suppressive fungicides and host ROS stress, we first examined the contribution of PsTIP60 to P. sojae development and virulence. The PsTip60 gene encodes a 453 amino acid protein composed of three named domains, including Tudor, Zf-MYST, and MOZ/SAS (SI Appendix, Fig. S5B). To determine whether PsTIP60 involved in the host defense- and AZX-induced ROS responses, we aimed to delete this gene from the P. sojae genome using CRISPR/Cas9 technology (40, 41). Although 341 heterozygous individuals were obtained, we never succeeded in obtaining homozygous full-length PsTIP60 gene knockout mutants. This result suggested that complete deletion of PsTIP60 might be lethal in P. sojae. Therefore, we generated PsTIP60 knockdown mutants (TIP60-1 and TIP60-2) and PsTIP60-overexpressing transformant (TIP60-OE) (SI Appendix, Fig. S5C). Notably, the relative expression of PsTIP60 in TIP60-OE strain was increased when under the treatment of H₂O₂, MMC, and AZX compared to CK (SI Appendix, Fig. S5B). Then we also determined the protein stability of PsTIP60 using synthesis inhibitor 10 μM cycloheximide (CHX). When CHX treatment suppressed new protein synthesis, the degradation rate of PsTIP60 under H₂O₂ and MMC treatment was not significantly different from that under control treatment (SI Appendix, Fig. S5D). In addition to the knockdown lines, we produced homozygous S322A substitution mutants via CRISPR-Cas9-mediated gene editing (TIP60^S322A); serine 322 occurs in a crucial active site and is conserved in eukaryotes (SI Appendix, Fig. S5 E and F). PsTIP60 knockdown or point-mutation dramatically reduced acetylation of multiple lysine residues of histone H4 as well as histone H2 lysine 5 (H2AK5) in vivo (SI Appendix, Fig. S6A). As shown in SI Appendix, Fig. S6B, purified PsTIP60-6× His was able to catalyze the acetylation of H4 and of H2AK5 in vitro, while PsTIP60^S322A exhibited attenuated histone acetyltransferase activity. Previous studies have demonstrated that phosphorylation of serine at positions 86 and 90 in Tip60 is critical for regulating its enzymatic activity (46). To verify whether phosphorylation occurs at site 322 and whether phosphorylation has an effect on TIP60 enzyme activity, we constructed strains expressing either phosphor-inactive or phosphorymimetic variants of PsTIP60. Specifically, the predicted phosphorylation residue S322 was replaced with alanine (A) to simulate a dephosphorylation state, or aspartic acid (D) to mimic a phosphorylation condition. Phos-tag assays revealed that PsTIP60 was normally phosphorylated, however, PsTIP60^S322A could not be phosphorylated (SI Appendix, Fig. S6C). Interestingly, PsTIP60^S322D exhibited significantly enhanced histone acetyltransferase activity in vitro compared to PsTIP60^S322A (SI Appendix, Fig. S6B). Furthermore, we employed AlphaFold3 to predict the protein structures of PsTIP60, PsTIP60^S322A, and PsTIP60^S322D. Subsequent visual analyses (including pocket changes and solvent accessible surface area) were performed using PyMOL 2.6 (SI Appendix, Fig. S6C). The results indicated that PsTIP60^S322D exhibits the largest combined pocket size and solvent-accessible surface area, followed by wild-type PsTIP60, while PsTIP60^S322A displays the smallest dimensions, implying that phosphorylation at PsTIP60 S322 may enhance enzyme activity by promoting its binding to the substrate (SI Appendix, Fig. S6D). In addition, the phenotypic results indicated that the TIP60-1, TIP60-2, and TIP60^S322A lines showed reduced colony growth and had significantly decreased sporangia numbers, zoospore production, and cystospore germination rates compared to the P6497, EV, and TIP60-OE strains (SI Appendix, Fig. S6 E–H). The virulence of the TIP60-1, TIP60-2, and TIP60^S322A lines also was significantly impaired compared to the P6497, EV, and TIP60-OE strains (SI Appendix, Fig. S6I).

PsTIP60 Plays a Critical Role in the ROS-Induced DDR.

To characterize the contribution of PsTIP60 to tolerance against fungicides and host defense stresses, we next examined the roles of PsTIP60 in DDR. In line with the PsDMAP1 results, the TIP60-1, TIP60-2, and TIP60^S322A lines showed a greater sensitivity to H₂O₂, AZX, and MMC than P6497, EV, and TIP60-OE strains (Fig. 4A). DPI treatment restored the insensitivity of Tip60-1 and TIP60^S322A to H₂O₂ and AZX (SI Appendix, Fig. S7A). Accordingly, γH2Ax elevation was nearly doubled reduced in the Tip60-1 and TIP60^S322A lines compared to P6497 (Fig. 4B). Tip60-1 and TIP60^S322A lines also generated more 8-OHdG compared to P6497, EV, and TIP60-OE following H₂O₂ or AZX treatment at various time points (SI Appendix, Fig. S7B). More importantly, the reduced virulence of the Tip60-1 and TIP60^S322A lines were mostly restored by DPI treatment (Fig. 4 C and D). Similarly, the induction of transcript levels of DDR-associated genes at the infection stage and treatment of AZX and MMC was markedly reduced in the Tip60-1 and TIP60^S322A mutants compared with P6497 (SI Appendix, Fig. S7C). In summary, our results suggested that PsDMAP1 and PsTIP60 function closely together in mediating upregulation of DDR-related gene expression in response to oxidative stresses resulting from host immunity and fungicide exposure.

Fig. 4.

PsTIP60 mediates the response of P. sojae to oxidative stress and DNA damage. (A) Mycelial growth of the WT P6497, EV, PsTIP60 knockdown (TIP60-1, and TIP60-2), PsTIP60 S322A point-mutant (TIP60^S322A), and PsTIP60-overexpressing transformant (TIP60-OE) on V8 agar in the presence of 10 mM H₂O₂, 10 ppm MMC, and 1 ppm AZX. 0.1% DMSO was used as the control (Left). Mycelial growth inhibition by H₂O₂, MMC, or AZX relative to DMSO control (Right). Lower panel: Inhibition ratio was calculated as (Control−Growth rate on plates with treatment)/Control growth × 100% after 5 d. Data are the mean ± SD from three biological replicates. Statistical significance compared to the wild-type P6497 was determined using Student′s t test (**P < 0.01, ns: not significant). (B) The elevation of γH2Ax levels in response to 10 mM H₂O₂, 10 ppm MMC, or 1 ppm AZX was inhibited in Tip60-1 and TIP60^S322A mutants compared to P6497. γH2Ax levels were determined by western blotting using anti-γH2Ax antibody. Detection of H2A protein was used as the loading control. The intensity of the γH2Ax band from P6497 treated with DMSO was set as 1.00; and the relative intensity of H4K16ac and γH2Ax band from each treatment was quantified with Image J. (C) The treatment with 1 μM DPI restored the impaired virulence of TIP60-1 and TIP60^S322A mutants. Left panels: soybean seedlings were inoculated with mycelial plugs of each strain and then stained at 24 hpi by DCFH-DA with or without 1 μM DPI as ROS scavenger. Right panels: disease symptoms at 48 hpi of above strains when soybean seedlings were treated with H₂O (control) or DPI. (D) Lesion length (Upper) and pathogen biomass (Lower) of each strain under treatments with H₂O (control) or DPI, measured at 48 hpi. Data are presented as the mean ± SD from three biological replicates. Statistical significance of the lesion length and biomass of above strains compared to the P6497 at H₂O or DPI treatment was determined using Student’s t test (**P < 0.01, ns: not significant).

PsDMAP1 Enhances Histone Acetyltransferase Activity of PsTIP60 in Response to DNA Damage.

Because of the physical interaction between PsDMAP1 and PsTIP60 and their common role in mediating DDR, we next examined the effect of PsDMAP1 knockdown on PsTIP60-mediated histone acetylation in P. sojae. Therefore, we evaluated histone H4 acetylation levels in the PsDMAP1 knockdown mutants. Notably, we found that H4K5ac, H4K8ac, H4K12ac, and H4K16ac levels were weakened in PsDMAP1 knockdown mutants as compared to those in the wild-type P6497 (Fig. 5 A and B). In addition, acetylation of H2AK5 also was reduced in the knockdown lines (Fig. 5 A and B). We also detected the H3Kac levels in the mutants. However, no significant differences were observed in H3 acetylation among DMAP1-1, TIP60-1, and TIP60^S322A mutants compared to that in P6497 (SI Appendix, Fig. S8A). To investigate the potential roles of H4 acetylation in regulating DDR-related genes, we first determined which histone residues were acetylated following DNA damage. Notably, H2AK5ac and H4K16ac were elevated in response to H₂O₂ and MMC treatment. However, PsDMAP1 and PsTIP60 knockdown largely suppressed H2AK5ac and H4K16ac elevation resulting from treatment with H₂O₂ and MMC (Fig. 5 C and D). These reductions in histone acetylation implied that PsDMAP1 may be required for the full HAT activity of PsTIP60. To test whether PsDMAP1 could enhance PsTIP60 activity in vitro, we evaluated the effect of PsDMAP1 on the acetyltransferase activity of PsTIP60 on histone proteins H2A and H4 in vitro. The PsDMAP1-, H4-, and H2A-6 × His proteins were purified with a Ni-NTA column and PsTIP60-GST protein was purified with GST-tag Purification Resin. As shown in Fig. 5C, both the levels of H2K5ac and H4K16ac were notably increased in the presence of PsDMAP1 at different molar ratios (2×, and 4 × in relative to PsTIP60), indicating that PsDMAP1 was able to directly enhance the HAT activity of PsTIP60 in vitro (Fig. 5 E and F).

Fig. 5.

PsDMAP1 enhances histone acetyltransferase activity of PsTIP60 in response to DNA damage. (A) PsDMAP1 knockdown resulted in the reduction of H2A and H4 acetylation. Effects of PsDMAP1 knockdown on H2AK5ac, H4K5ac, H4K8ac, H4K12ac, and H4K16ac levels in the indicated strains compared to control strains were determined by western blotting using the corresponding antibodies. H2A and H4 protein was used as the loading control. (B) The relative intensities of the acetylated histone bands from each treatment were quantified with Image J, and the intensities of the bands from P6497 grown in V8 medium were set as 1.00. (C) PsDMAP1 and PsTIP60 knockdown (DMAP1-1 and TIP60-1) inhibited H2AK5ac and H4K16ac elevation in response to 10 mM H₂O₂ and 10 ppm MMC treatments. Effect of PsDMAP1 and PsTIP60 knockdown on H2AK5ac, H4K5ac, H4K8ac, H4K12ac, and H4K16ac levels were determined by western blotting using corresponding antibodies in different treatments (0.1% DMSO, 10 mM H₂O₂, and 10 ppm MMC) of strains P6497 (WT), DMAP1-1 (PsDMAP1 knockdown), and TIP60-1 (PsTIP60 knockdown). (D) The relative intensities of the acetylated histone bands from each treatment were quantified with Image J, and the intensities of the bands from P6497 grown in V8 medium with DMSO were set as 1.00. (E) In vitro assays of PsDMAP1 enhancement of the HAT activity of PsTIP60 toward H2AK5 (Left) and H4K16 (Right). Purified H2A- or H4-6×His were incubated with purified PsTIP60-GST in the absence or the presence of PsDMAP1-6×His at different molar ratios (twofold or fourfold relative to PsTIP60). Protein loadings are shown by coomassie blue staining (CBB). (F) The relative intensities of the acetylated histone bands from each treatment were quantified with Image J, and lane 1 in the Left and Right panels of E were set as 1.00, respectively. Red asterisk indicates our target band. Data in B, D, and F are presented as the mean ± SD from three biological replicates. Different letters in B, D, and F represent significant differences by one-way ANOVA (P < 0.05).

PsDMAP1 Promotes Expression of DDR-Related Genes by Enhancing PsTIP60-Mediated Histone Acetylation Following DNA Damage.

To investigate the potential roles of PsTIP60-mediated histone acetylation in expression of DDR-related genes, we performed the Chromatin IP quantitative PCR (ChIP-qPCR) assays. The results indicated that H4K16ac was consistently elevated in the promoters of genes involved in the DDR, namely PsPARP1, PsASF1, PsOGG1, PsRad52, PsFPG, and PsRecQ, while H2AK5ac was elevated only in the promoters of PsOGG1 and PsRecQ (SI Appendix, Fig. S8 B and C). Given that DDR-related genes regulated by PsDMAP1 exhibited elevated PsTIP60-mediated H4K16ac during DDR, we speculated that PsDMAP1 and PsTIP60 promote expression of these genes by upregulating H4K16 acetylation at the promoters of the genes. To test this hypothesis, we examined the impact of PsDMAP1 and PsTIP60 knockdown on H4K16ac elevation at the promoters of the six DDR-related genes. The results showed H₂O₂ and MMC exposure greatly elevated H4K16ac at the promoters of the DDR-related genes (SI Appendix, Fig. S9A). Furthermore, H4K16ac elevation at the DDR-related genes was reduced in the PsDMAP1- and PsTIP60 knockdown mutants (DMAP1-1 and TIP60-1) (SI Appendix, Fig. S9A). A recent study reported that lack of H4K16ac resulted in deficient RNA polymerase II recruitment to the affected promoters (19). To further understand the mechanism of PsDMAP1/PsTIP60-dependent gene regulation, we assessed whether PsDMAP1/PsTIP60-regulated H4K16ac levels affected RNA pol II recruitment. ChIP-qPCR assays with anti-pol II antibodies indicated that the RNA pol II occupancy rate of the promoters of the DDR-related genes during H₂O₂ or MMC treatment was significantly reduced in the PsDMAP1- and PsTIP60 knockdown mutants compared to P6497 (SI Appendix, Fig. S9B). These results suggest that H4K16ac might reduce nucleosome residence and enhance RNA pol II recruitment at targeted genomic regions to reinforce gene transcriptional activation in response to DDR.

H4K16ac Is A Transgenerational Epigenetic Signal for Adaptation to Stress from Fungicides and Plant Defense-Derived ROS.

Given the critical role of H4K16 acetylation (H4K16ac) in the ROS-induced DDR in P. sojae and its function as a key epigenetic marker, we propose an intriguing hypothesis: H4K16ac may serve as an epigenetic information carrier, transmitting adaptive responses to ROS stress from P. sojae to its progeny, thereby enhancing their resilience to ROS stress. The adaptation of organisms to ever-changing and challenging environments is assisted by the transmission of epigenetic marks across generations, which thus serve as carriers of inheritable information (47). For example, maternal H4K16ac was shown to provide instructive information to offspring by promoting future gene activation in D. melanogaster (19). Therefore, we first sought to determine whether exposure of parental generations of P. sojae to fungicides might make their progeny more stress adapted. The life cycle of P. sojae can include both asexual and sexual spore formation (Fig. 6A and SI Appendix, Fig. S10A), with new generations defined by the emergence of mycelia from germinating asexual or sexual spores (zoospores and oospores, respectively). To test for transgenerational effects, we exposed parental cultures to 1 ppm AZX (concentration for 80% of maximal effect, EC 80), then measured the AZX sensitivity of immediate next generation cultures. Parental cultures (F0) were either left untreated or were exposed to AZX at each step in the absence or presence of DPI under one of two protocols. Under “no prior exposure”, the parental culture was never exposed to AZX and only subcultures initiated by zoospores or oospores were tested for sensitivity (Fig. 6 B, Upper). Under “sequential exposure” the parental culture (F0) and each subsequent zoospore- or oospore-initiated generation (F1, F2, etc) was exposed to AZX, and its sensitivity assessed (Fig. 6 B, Lower). We assessed the inhibition of mycelial growth at each stage and used this metric as an indicator adaptation to AZX stress. Growth inhibition under the sequential exposure” protocol exhibited a gradual downward trend (Fig. 6C and SI Appendix, Fig. S10 B, Upper Left), which indicated a progressive intergenerational adaptation to AZX stress, while the no prior exposure protocol produced no such adaptation. Interestingly, the treatment of DPI abolished stress adaptation in zoospore- or oospore-initiated generation (Fig. 6C and SI Appendix, Fig. S10 B, Upper Right). These results demonstrate the critical role of ROS stress in triggering AZX adaptation of P. sojae.

Fig. 6.

H4K16ac is a transgenerational epigenetic signal for adaptation to fungicide- and plant-imposed stress. (A) The life cycle of P. sojae can be divided into asexual reproductive stages. Each generation in asexual reproduction lasts about 10 d. After collecting zoospores released by P. sojae, the concentration was adjusted to 10⁴/mL, and then the zoospores were evenly coated on a water AGAR plate, and the single spores that grew mycelia were selected and inoculated on V8 plate for subsequent sensitivity determination. (B) Transgenerational inheritance of fungicide stress adaptation in P. sojae. Under no prior exposure, the parental culture was never exposed to AZX and only subcultures initiated by zoospores or oospores were tested for sensitivity (Upper). Under sequential exposure the parental culture (F0) and each subsequent zoospore-initiated generation (F1, F2, etc) were exposed to AZX, and its sensitivity assessed (Lower). (C) In asexual reproductive stages, Transgenerational inheritance tested in P6497, PsDMAP1 knockdown mutant (DMAP1-1), and PsTIP60 knockdown mutant (TIP60-1) under the no prior exposure or sequential exposure at each step in the absence or presence of DPI. Mycelial growth inhibition by AZX relative to the DMSO-treated control was calculated as (Control−Growth rate on plates with treatment)/Control. (D) In sequential exposure conditions, H4K16ac transgenerational inheritance was assessed across generations of P6497, DMAP1-1, and TIP60-1 at each step in the absence or presence of DPI by western blotting using anti-H4K16ac antibody. H4 protein was used as the loading control (Left). The relative intensities of H4K16ac bands from each treatment were quantified with Image J, and the intensities of the bands from lane 1 were set as 1.00, respectively (Right). (E) In sequential exposure conditions, H4K16ac transgenerational inheritance at the promoters of PsPARP1 were assessed across generations of P6497, DMAP1-1, and TIP60-1 at each step in the absence or presence of DPI by ChIP-qPCR. H4K16ac enrichment signals were normalized to the tubulin gene. (F) Infection cycle protocol. Generation F(n+1) (e.g. F2) was defined as the mycelia isolated after Fn (e.g. F1) generation mycelia had colonized soybean seedlings. (G) Disease symptoms produced by each successive infection generation - F1, F2, F3, and F4 - of P6497, DMAP1-1, and TIP60-1. Infected seedlings were photographed at 48 hpi. (H) Lesion lengths produced by each strain in G, measured at 48 hpi. (I) Relative biomass produced by each strain in G, measured by genomic DNA qPCR. Data in C, E, H, and I are presented as the mean ± SD from three biological replicates. Different letters represent statistically significant differences according to the one-way ANOVA test (P < 0.05).

Then to test whether the persistence of AZX insensitivity across generations involved in H4K116ac, we assessed persistence in a PsDMAP1 knockdown mutant (DMAP1-1) and a PsTIP60 knockdown mutant (TIP60-1). We found that the AZX adaptation of progeny was abolished in these mutants in the both absence and presence of DPI under the sequential exposure protocol and no prior exposure (Fig. 6C and SI Appendix, Fig. S10 B, Middle and Right). Next, we examined H4K16ac levels in each generation of the WT P6497 and mutant cultures. We found that the sequential exposure protocol produced a gradual elevation of H4K16ac levels in the P6497 (Fig. 6D), while no prior exposure protocol and the treatment of DPI under sequential exposure protocol both abolished gradual elevation of H4K16ac in progeny (Fig. 6D and SI Appendix, Fig. S11). However, no elevation in H4K16ac levels was observed in the DMAP1-1 or TIP60-1 mutants in the both absence and presence of DPI under the sequential exposure (Fig. 6D). To further definite the inheritance of the acetylation mark in the progeny, we performed the ChIP-qPCR. In sequential exposure, the relevant DDR genes are hyperacetylated in the next generation of P6497 in the absence of DPI, while the level of H4K16ac on the promoter of the DDR genes did not exhibit significant differences between parents and progeny in the presence of DPI (Fig. 6E and SI Appendix, Fig. S12 A–E, Left). Nevertheless, no elevation in H4K16ac levels was observed on the promoter of the DDR genes in DMAP1-1 or TIP60-1 mutants, both in the absence and presence of DPI under the condition of sequential exposure (Fig. 6E and SI Appendix, Fig. S12 A–E, Middle and Right). Furthermore, we determined the duration of epigenetic effects in the absence of AZX. Under the condition labeled no prior exposure, the parental culture was never exposed to AZX; only subcultures initiated by zoospores were tested for sensitivity (SI Appendix, Fig. S13 A, Upper). In contrast, under the condition “treated with AZX”, the designation “F4” comes from the fourth generation resulting from sequential exposure shown in Fig. 2B. Each subsequent generation initiated by zoospores (F5, F6, etc) was not exposed to AZX, and its sensitivity was assessed (SI Appendix, Fig. S13 A, Lower). Notably, it was not until the F6 generation that strain P6497 regained the same sensitivity to AZX comparable to that observed under no prior exposure (SI Appendix, Fig. S13B). Similarly, the abundance of H4K16ac remained consistent with those observed in the F1 generation of P6497 up until the F6 generation (SI Appendix, Fig. S13C). These results suggest that epigenetic effects can persist for approximately two generations in the absence of AZX. These results suggested PsDMAP1/PsTIP60-regulated H4K16ac plays an important role in the ROS-dependent AZX adaptation.

In the field, P. sojae hyphae produced by germinating zoospores or oospores penetrate the soybean root resulting in damping off of soybean seedlings and root and stem rot of established plants (33). Zoospores and oospores released by infected plants can also be spread to the plants by splashing spreading the disease. To test the potential roles of H4K16ac-mediated ROS adaption in the disease cycle, we simulated the process of disease cycle in the field. Namely, the soybean seedlings were colonized with P. sojae, then the pathogens were isolated from the infected soybean seedlings 3 d later, and the P. sojae was subcultured one more time on basic medium before being used to inoculate plants at each step in the absence or presence of DPI (Fig. 6F). Interestingly, with the increasing infection generations, the virulence of wild type P6497 exhibited a gradual increase in the absence of DPI, while the treatment of DPI caused P6497 lose this gradual increase of virulence (Fig. 6 G–I). However, this increase was not observed in DMAP1-1 or TIP60-1 mutants both in the absence and presence of DPI (Fig. 6 G–I). Given that both adaptation to AZX and infection adaptation in P. sojae represent mechanisms of resistance to ROS stress, we further investigated potential cross-adaptation between these responses. Initially, P. sojae strains were cultured on V8 agar plates either with or without H₂O₂ (Mock or H₂O₂) for a single generation before being inoculated onto plates containing AZX to assess the inhibition rates of AZX against this pathogen in both absence or presence of DPI. Concurrently, pathogens isolated from infected soybeans were also inoculated on V8 plates (Isolation) supplemented with AZX, and the inhibition rates of AZX were determined under similar conditions regarding DPI absence or presence (SI Appendix, Fig. S14A). Interestingly, wild-type strain P6497 exhibited a lower inhibition rate by AZX under the “H₂O₂” or “Isolation,” compared to that under “Mock” conditions without DPI; however, the treatment with DPI eliminated this difference (SI Appendix, Fig. S14B). Furthermore, we found that the AZX inhibition rate exhibited no significance between the H₂O₂ or Isolation and Mock in DMAP1-1 or TIP60-1 mutants both in the absence and presence of DPI (SI Appendix, Fig. S14B). Collectively, these results suggested PsDMAP1/PsTIP60-regulated H4K16ac is a required for virulence adaptation on plants, as a result of adaptation to ROS stress.

Discussion

ROS, generated by the innate immune system in response to pathogen invasion, represent a crucial mechanism underlying plant disease resistance. Plant cell-surface pattern recognition receptors (PRRs) recognize pathogen-associated molecular patterns (PAMPs) and extracellular effector molecules, thereby initiating the activation of NADPH oxidase, encoded by the Respiratory Burst Oxidase Homologue D (RBOHD) gene (2, 3). During host infection, pathogens deploy diverse strategies to counteract host-derived ROS. These mechanisms include secreting effectors that suppress ROS production, releasing antioxidant enzymes such as catalase to neutralize ROS effects, enhancing cellular tolerance to oxidative stress, and repairing DNA damage induced by plant-derived ROS (6, 48, 49). Here, we identify PsDMAP1, a key regulator in response to ROS stress, from a family of proteins containing a SANT domain that is conserved across oomycete species and other eukaryotes. DMAP1 was initially recognized as an activator of DNMT1 and a crucial component of the TIP60-p400 complex in humans (26, 29). DMAP1 serves as an important molecular regulator in epigenetic processes due to its SANT domain’s specific binding affinity for histone tails, functioning effectively as a histone interaction module (50). The role of DMAP1 in DNA damage repair within mammalian cells is well documented. Upon the occurrence of DNA damage, the activation of ATM/ATR kinases triggers downstream signaling cascades essential for repair processes. DMAP1 plays a pivotal role in this mechanism by associating with chromatin-modifying factors, thereby contributing to the stabilization and regulation of these repair signaling pathways (29). Furthermore, mutant yeast strains lacking Eaf2, the yeast homolog of DMAP1, exhibit pronounced sensitivity to DNA breaks induced by genotoxic agents. This heightened susceptibility underscores a pivotal role for Eaf2 in cellular response to DNA damage, particularly within pathways mediating DNA repair (51). Although limited studies have directly examined the involvement of DMAP1 regulators in cellular response to ROS, extensive research on ROS-induced DNA damage (7–9) and the significant role played by DMAP1 in DNA repair underscores the potential importance of investigating its function concerning pathogen adaptation to host-derived ROS stress. This study provides a perspective on pathogen ROS tolerance mechanisms by demonstrating the cooperation between PsDMAP1 and PsTIP60 to comediate histone H4 acetylation, regulating DNA repair-related gene expression to repair genome damage, in order to adapt to ROS stress due to host defense responses as well as to fungicidal respiration inhibitors.

PsDMAP1 directly interacts with PsTIP60, substantially enhancing its HAT activity (Fig. 2). However, the elaborate molecular assembly mechanisms involving PsDMAP1, PsTIP60, and other potential NuA4 complex subunits as well as substrate nucleosome recognition remain unknown in oomycetes. NuA4 includes two major modules: the catalytic HAT module and the transcription activator-binding TRA module (52). Recent mechanistic research of NuA4 in yeast showed that it selectively acetylated histone H4, indicating that Eaf2 (DMAP1 ortholog) and Epl1 comprise the TRA module’s poly-basic surface that interacts with nucleosomal linker DNA and is crucial for optimal NuA4 activity (45). In contrast, core catalytic subunit Esa1 (TIP60 ortholog) and other components, including Epl1’s N-terminal EPc-A fragment, PHD-domain-containing Yng2, and Eaf6, form the HAT module’s core Piccolo subcomplex (45, 48). These studies suggest that Eaf2 and Esa1 appear to belong to two different submodules in the NuA4 complex, and they do not directly interact during assembly or function in yeast (45). Nevertheless, our results support a direct interaction between PsDMAP1 and PsTIP60, thus demonstrating the complexity of subunit division and cooperation in the NuA4 complex of different organisms. Further studies are required to determine whether the molecular assembly of the histone H4 acetyltransferase complex in oomycetes generates functions that differ from yeast.

Previous studies have shown that acyl groups added during histone acetylation mask the positive charge of lysine residues, reducing the affinity of the histone tail for chromatin and exposing the underlying DNA (51). Lysine acetylation also acts as a binding site for proteins containing bromodomains, which usually have protranscriptional functions. Together, these activities promote a more open, less compressed chromatin form and facilitate nucleosome access (53, 54). A recent postzygotic study in Drosophila and mammals showed that sustained H4K16ac promotes transcriptional activation by regulating nucleosome accessibility and efficient RNA Pol II recruitment (19). Here, we have demonstrated a transgenerational inheritance phenotype in which fungicide stress in the parental generation has the ability to transmit epigenetic information to progeny to make the progeny more fungicide insensitive.

In summary, our data support a model (Fig. 7) in which ROS induce the transcription of PsDMAP1 and PsTIP60, then PsDMAP1 binds to PsTIP60 and stimulates PsTIP60-dependent acetylation of histone H4K16, activating the transcription of DDR-related genes, and enhancing recruitment of the DDR marker protein, γH2Ax. These processes enable P. sojae to tolerate genomic damage produced by plant ROS and fungicide-induced oxidative stress. Furthermore, acetylation of histone H4K16, acts as an epigenetic mark that mediates the transmission of oxidative stress adaptation to asexual and sexual progeny. An important application of this study is the identification of PsDMAP1 as a potential target for anti-oomycete fungicide development, based on its homozygous knockout lethality and essential roles in development, virulence, and adaptation to plant and fungicide-induced oxidative stress. Our findings thus provide insights and strategies for preventing and controlling oomycete diseases and managing fungicide resistance in plant protection.

Fig. 7.

Proposed model for the role of PsDMAP1/PsTIP60-regulated H4K16ac in ROS-dependent virulence adaptation of P. sojae on host plants. ROS induce the transcription of PsDMAP1 and PsTIP60. Subsequently, PsDMAP1 binds to PsTIP60, thereby stimulating PsTIP60-dependent H4K16ac. This acetylation event activates the transcription of DDR-related genes and enhances the recruitment of the DDR marker protein, γH2Ax. These orchestrated mechanisms equip P. sojae with the capability to withstand genomic damage induced by host defense- and fungicide- induced ROS stress. Moreover, H4K16ac serves as an epigenetic signature facilitating the transmission of oxidative stress adaptation to both asexual and sexual progenies.

Materials and Methods

Gene deletion mutants were generated using the CRISPR-mediated gene replacement strategy (40, 41, 55). An in vitro histone acetylation assay was carried out following a previous publication with minor modifications (56). The 8-OHdG assay was performed using an 8-OHdG ELISA kit (D751009, Sangon Biotech, Shanghai, China). To assess ROS production, fungal hyphae or plant tissue was stained using 10 μM DCFH-DA (S0033S, Beyotime, Shanghai, China). ChIP was performed as previously described (6). Proteins were isolated from vegetative hyphae as previously described (57). Additional details of materials and methods can be found in SI Appendix, Materials and Methods. The amplified primer information is provided in Dataset S3, and the tag sequences are provided in Dataset S4.

Supplementary Material

Appendix 01 (PDF)

Dataset S01 (XLSX)

Dataset S02 (XLSX)

Dataset S03 (XLSX)

Dataset S04 (XLSX)

Dataset S05 (XLSX)

Data, Materials, and Software Availability

The information about gene involved in this article can be found in Dataset S5. All other data are included in the manuscript and/or supporting information.