The necrotrophic fungus, Lasiodiplodia theobromae, is geographically widespread in the subtropical and tropical regions and is known to attack approximately 500 plant species, including crops and woody trees (Alves et al., 2008; Cipriano et al., 2015). This fungus has been regarded as a latent pathogen or an opportunistic pathogen leading to dieback, canker, or fruit rot diseases in many economically important woody crops (Slippers and Wingfield, 2007; Ali et al., 2019). In southern China, L. theobromae is also a causal agent of peach gummosis, one of the most devastating diseases of peach (Prunus persica), annually causing considerable quantity and quality losses (Beckman et al., 2003; Wang et al., 2011). A better understanding of the molecular mechanisms of the peach-L. theobromae interaction is necessary for effective control of peach gummosis.

To establish successful infections, pathogens need to overcome both preformed and induced host defenses (Qi et al., 2017). During pathogen attacks, one of the major and fastest plant defense responses is a rapid accumulation of reactive oxygen species (ROS) at the invasion site (Scheler et al., 2013). ROS, primarily superoxide (O2−) and hydrogen peroxide (H₂O₂), are produced by plasma membrane-localized NADPH oxidases, also known as respiratory burst homologs (RBOH), at the inoculation site (Suzuki et al., 2011). Due to the toxicity, ROS can cause oxidative stress and damage to biomolecules, such as DNA mutation, lipid peroxidation, and protein oxidation, eventually causing cell death of the pathogens (De Gara et al., 2003).

Additionally, as a class of signaling molecules, ROS play crucial roles in plant-pathogen interactions. Plant-derived ROS act as signaling molecules to mediate various important responses of plant cells to fight against pathogen infection and enhance plant resistance by inducing plant defense-related gene expression and activating related enzyme activity (Torres and Dangl, 2005; Baxter et al., 2014). Our previous study demonstrated that the infection by L. theobromae caused a ROS burst, and transcripts of pathogenesis-related (PR) genes were markedly induced, potentially contributing to the restriction of disease development (Zhang et al., 2020).

To survive and colonize under harsh conditions, pathogens have developed ROS scavenging systems to efficiently reclaim excess ROS (Segal and Wilson, 2018). Scavenging enzymatic and non-enzymatic compounds, either preformed or induced, include superoxide dismutase (SOD), catalases (CAT), peroxidases (POD), glutaredoxins, and thioredoxins (Kawasaki et al., 1997; Lanfranco et al., 2005; Ma et al., 2018). The glutaredoxin system has glutathione, glutathione peroxidase (GPX), glutathione reductase (GLR), and NADPH. The thioredoxin machinery includes thioredoxin peroxidase [equal to thiol-specific antioxidant protein (TSA)], thioredoxin reductase (TRR), thioredoxins (TRX), and NADPH (Ma et al., 2018; Zhang et al., 2019). In fungal pathogens, transcription factor-mediated ROS detoxification through the regulation of antioxidant genes expression is vital in plant-pathogen interactions. One of the critical regulators mediating ROS detoxification is the Activating Protein 1 (AP1) class of basic leucine zipper (bZIP) family (Segal and Wilson, 2018). AP1 is a key transcriptional activator in response to oxidative stress in yeasts and filamentous fungi (Reverberi et al., 2008; Lin et al., 2018; Segal and Wilson, 2018). In our previous study, the LtAP1 gene was consistently and highly expressed in the infection stage of L. theobromae on peach shoots, implying that LtAP1 may play a crucial role in the pathogenicity of L. theobromae (Zhang et al., 2020).

Saccharomyces cerevisiae YAP1 serves as one of the most critical determinants of yeast to oxidative stress response, which is responsible for transcriptional activation of various ROS detoxification-related genes (Mendoza-Martínez et al., 2020). Subsequently, YAP1 homologs in several fungal pathogens were identified and characterized, and found to have conserved roles in oxidative stress response and tolerance, but differed in virulence. YAP1-mediated ROS detoxification has been identified as being an essential virulence determinant in the necrotrophic fungus Alternaria alternata (Lin et al., 2009), the hemibiotrophic rice blast fungus Magnaporthe oryzae (Guo et al., 2011), and the biotrophic maize pathogen Ustilago maydis (Molina and Kahmann, 2007). However, YAP1-assisted ROS detoxification is associated with avirulence in the animal pathogen Aspergillus fumigatus (Lessing et al., 2007), the necrotrophic plant pathogen Cochliobolus heterostrophus (Lev et al., 2005), or the hemibiotrophic plant pathogen Fusarium graminearum (Montibus et al., 2013). Although many studies have examined YAP1 homologs in other fungi, their function in L. theobromae during pathogenesis remains poorly understood, particularly for canker or gummosis disease in woody fruit trees. Understanding the role of the LtAP1 gene in L. theobromae may lead to new tools to develop novel, sustainable disease management strategies against peach gummosis.

In this study, transcription factor LtAP1 was isolated and functionally characterized through genetic transformation. We examined the effects of deletion of the LtAP1 gene on mycelial growth, sensitivity to oxidative and nitrosative stresses, and pathogenicity. This study shed some light on the function of the LtAP1 gene for ROS detoxification, virulence, and suppression of plant defense responses during L. theobromae and peach interaction, which could deepen our knowledge of the role of fungal YAP1s in plant diseases.

The peach gummosis pathogen, L. theobromae, is a destructive threat to peach harvests (Wang et al., 2011), and infection events at the molecular level need deeper investigation. In a previous study, L. theobromae infection caused an oxidative burst in peach shoots and promoted expression of LtAP1 and other genes associated with the ROS scavenging system (Zhang et al., 2020). Subsequently, we attempted to uncover how plant infection is regulated by an oxidative stress regulator, YAP1, in L. theobromae.

Eukaryotic microorganisms have stress-protective functions against a variety of adverse conditions. We first compared the growth performance of the ΔLtap1 mutant and the WT grown on media supplemented with different exogenous chemicals to mimic environmental stresses. Our results suggest that in L. theobromae, LtAP1 was involved in response to various stresses. Deletion of LtAP1 led to decreased sensitivity to osmotic and cell wall inhibitors, indicating that LtAP1 negatively regulated the sensitivity to osmotic pressure and the maintenance of cell wall integrity in L. theobromae (Figure 2). Likewise, in C. gloeosporioides, ΔCgap1 mutants had higher resistance to sorbitol than the WT (Li et al., 2017). However, in F. graminearum, the Fgap1-deficiency mutant exhibited increased sensitivity to sorbitol or NaCl-induced stresses (Montibus et al., 2013). In addition, our oxidative stress tests indicated that ΔLtap1 mutants were hypersensitive to H₂O₂, cumene H₂O₂, and TBHP, as well as menadione (Figures 3A,B). These results suggest that LtAP1 plays a vital role in the regulation of fungal response to oxidative stress. This is consistent with studies on M. oryzae (Guo et al., 2011), F. graminearum (Montibus et al., 2013), A. alternata (Lin et al., 2009), and C. gloeosporioides (Sun et al., 2016), where the mycelial growth of the respective mutant was severely reduced by oxidative stress compared to their respective WT. Moreover, the expression levels of LtAP1 were significantly upregulated under the oxidant treatments (Figures 3C,D). A similar finding was observed in C. gloeosporioides (Sun et al., 2016) and Monilinia fructicola (Yu et al., 2017). The results suggest that the fungal YAP1s transcription factors are highly conserved for oxidative stress response in different species.

Interestingly, the ΔLtap1 mutant showed a significant reduction of mycelial growth with the SNP treatment. The growth suppression of the ΔLtap1 mutant after treatment with SNP plus H₂O₂ was higher than of either SNP or H₂O₂ alone (Figure 4), which showed an additive effect between SNP and H₂O₂. ROS can react with nitric oxide and generate toxic reactive nitrogen species (RNS), such as peroxynitrite (Marroquin-Guzman et al., 2017). Hence, LtAP1 is likely an essential player in oxidative and nitrosative stress adaptation.

Pathogenicity tests revealed that the ΔLtap1 mutant induced smaller necrotic lesions, less gum release, and decreased pathogen biomass than WT (Figure 5), suggesting that LtAP1 is essential for growth and virulence of the necrotrophic fungus L. theobromae on peach shoots. Similarly, in the biotrophic U. maydis and necrotrophic A. alternata, deletion of AP1 failed to incite necrotic lesions (Molina and Kahmann, 2007; Lin et al., 2009). In the hemibiotrophic pathogen M. oryzae, Moap1 is essential to the growth of invasive hyphae for successful infection (Guo et al., 2011). In hemibiotrophic C. gloeosporioides, ΔCgap1 mutant showed severely attenuated virulence on poplar leaves (Sun et al., 2016) and could not induce lesions on mango fruits (Li et al., 2017). However, in necrotrophic B. cinerea (Temme and Tudzynski, 2009) and F. graminearum (Montibus et al., 2013), the deletion of AP1 did not show noticeable effects on pathogenicity, indicating that YAP1 homologs are not necessary for virulence in all pathogenic fungi studied. This might be because fungal virulence associated with YAP1 differs in the types of associations established between specific fungi and plant hosts. A better and deeper understanding of the mechanisms of pathogen virulence associated with YAP1 homologs is needed.

It is well known that a major mechanism of plant defense is the production of ROS against pathogens attack. Therefore, fungal pathogens need robust strategies for ROS scavenging, which involves YAP1 homologs (Segal and Wilson, 2018). In the L. theobromae-infected peach shoots, the expression of LtAP1, the glutaredoxin system genes (LtGPX3 and LtGLR1), and the thioredoxin system genes (LtTRX2, LtTSA1, and LtTRR1) was markedly upregulated, which was perhaps to scavenge ROS derived from the host (Zhang et al., 2020). In this study, we found higher ROS accumulation at the inoculation site with the mutant than with the WT (Figures 6A,B). Moreover, the ΔLtap1 mutants were hypersensitive to exogenous oxidative stress (Figures 3A,B). This suggested that LtAP1-modulating oxidative stress tolerance might play a crucial role in fungal pathogenicity. To further investigate the link between LtAP1 modulation of oxidative stress tolerance and fungal pathogenicity, we used an NADPH oxidase inhibitor, DPI, to prevent ROS generation. The results clearly showed that the DPI treatment increased necrotic lesion size and enhanced gum release in the shoots inoculated with the ΔLtap1 mutants as compared to the mock control, suggesting that the pathogenicity of ΔLtap1 mutants was partially restored (Figures 7A,B). Overall, LtAP1 modulation of oxidative stress tolerance, at least in part, contributed to the pathogenicity of L. theobromae. Similarly, the ΔAaAP1 mutant of necrotrophic A. alternata was hypersensitive to oxidants, and its pathogenicity was rescued by the NADPH oxidase inhibitor treatment (Lin et al., 2009). In biotrophic U. maydis, H₂O₂ was markedly accumulated at sites inoculated with the Umap1 mutant, and inhibition of the plant NADPH oxidase decreased ROS accumulation and restored the virulence of the mutant, suggesting that Umap1 acts in neutralizing the ROS generated by the maize NADPH oxidase (Molina and Kahmann, 2007).

In filamentous fungi, YAP1 homologs are major regulators of the antioxidant response, but YAP1 homologs involve a wide array of processes by regulating genes involved in ROS scavenging (Mendoza-Martínez et al., 2020). The expression of such genes, such as the core glutaredoxin system genes (LtGPX3 and LtGLR1) and thioredoxin system members (LtTRX2, LtTSA1, and LtTRR1), was dramatically downregulated in the ΔLtap1 mutant treated with water or H₂O₂ (Figures 8A,B), indicating that LtAP1 acts as a major regulator in the antioxidant system. Similarly, the transcription factor AaAP1 could activate glutaredoxin (AaGPX3 and AaGLR1) and thioredoxin systems (AaTSA1 and AaTRR1) to cope with oxidative stress (Yang et al., 2016; Ma et al., 2018). The thioredoxin MoTrx2 was found to be a target of the transcription factor MoAP1 in M. oryzae, and ΔMotrx2 mutant displayed higher ROS levels and lower POD and laccase activities (Wang et al., 2017). However, in F. graminearum, the expression of three CAT- and two Cu/ZnSOD-encoding genes was downregulated in the Fgap1 mutant (Montibus et al., 2013). Likewise, EfAP1 in Epichloe festucae was required for expression levels of the CAT gene (Cartwright and Scott, 2013). As a whole, it suggests that YAP1 homologs could regulate/target different antioxidant system-related genes to overcome oxidative stress in different fungi.

During infection, the transcript levels of genes in glutaredoxin and thioredoxin systems were significantly downregulated in the ΔLtap1 mutant (Figure 8C). Concomitantly, the contents of O2− and H₂O₂ and transcripts of PpRBOHs were markedly higher in shoots inoculated with the ΔLtap1 mutant, demonstrating a reduced ability of the ΔLtap1 mutant to scavenge overproduced ROS during the interaction (Figure 6). It is speculated that LtAP1 is likely to modulate glutaredoxin and thioredoxin systems to scavenge host-derived ROS. Similarly, AaAP1 could also modulate glutaredoxin and thioredoxin systems to cope with oxidative stress (Lin et al., 2009; Yang et al., 2016; Ma et al., 2018).

ROS serving as the primary signaling molecule during pathogens attack can activate an array of defense responses, such as induction of defense-related genes (Qi et al., 2017; Segal and Wilson, 2018). We observed significantly higher expression levels of PR genes, such as PpPR1a, PpPR8, PpPR10-1, PpPR10-4, PpDFN1, and PpLTP1, in the ΔLtap1 mutant-infected shoots in comparison with the WT treated (Figures 9A–F). Furthermore, the transcripts of SA biosynthesis and signaling-related genes (PpPAL1, PpICS1, and PpNPR1) were also significantly upregulated after ΔLtap1 mutant inoculation than the WT (Figures 9G–I). The accumulation of PR proteins and SA-mediated plant defense response might assist in limiting disease development, which was reflected by the reduced lesion size, gum release, and fungal biomass at inoculation sites with the ΔLtap1 mutant (Figure 5). Similarly, the thioredoxin MoTrx2 regulated by MoAP1 played an essential role in the ROS scavenging during host invasion and in the suppression of the rice defense response, in which the transcript levels of plant defense genes were markedly higher in rice cells infected with the ΔMotrx2 mutant than the control (Wang et al., 2017). The rice cells inoculated with the ΔModes1 mutant exhibited strong defense responses accompanied by the accumulation of ROS and PR genes transcript in neighboring tissues, indicating that DES1 is required to suppress the host basal defenses (Guo et al., 2011). Taken together, we propose that the restricted expansion of the ΔLtap1 mutant in peach shoots is partly caused by the defect in active suppression of peach defense response.

In summary, we cloned and characterized the LtAP1 gene, which encodes a homolog of yeast YAP1. Our experiments demonstrated that LtAP1 was valuable for mycelial growth, stress response, and pathogenicity. We found that LtAP1 was a key regulator of oxidative stress response, acting in activating fungal glutaredoxin and thioredoxin systems, and suppressing plant defense responses during infection. The prevention of ROS production could partially restore pathogenicity of ΔLtap1 mutant. LtAP1 plays a central role in adjusting ROS homeostasis between fungal pathogen and plant host and is necessary for full virulence of L. theobromae. This study advances our understanding of the link between oxidative stress response, ROS detoxification, and virulence in L. theobromae. Given the critical roles of LtAP1 in L. theobromae-induced peach gummosis, it would be urgent to identify its potential targets in the downstream network, which would be helpful for future disease management.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://www.ncbi.nlm.nih.gov/genbank/, MN933613.1.

HZ, DZ, GL, and JL designed the experiments. HZ performed all the experiments with occasional help from WS, DZ, and XS. HZ, WS, and DZ analyzed the data. FW provided the analytical tools. HZ, JL, and TH wrote the manuscript. All authors read and approved the final manuscript.

This work was financially supported by the China’s National Key Research and Development Program (grant no. 2018YFD1000300) to JL, the Natural Science Foundation of China (grant no. 31471840) and the China Agriculture Research System of MOF and MARA (grant no. CARS-30) to GL, and the Central Public-interest Scientific Institution Basal Research Fund for Chinese Academy of Tropical Agricultural Sciences (grant nos. 1630092021006 and 16300920210010) to HZ.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.