Pear (Pyrus communis L.) cultivation, a significant component of the global fruit industry (Kang et al., 2021), faces significant challenges due to susceptibility to various pathogens, including Monilinia laxa (Aderhold & Ruhland) Honey (Petróczy et al., 2012; Martini and Mari, 2014). Commonly known as a causal agent of the brown rot disease, this necrotrophic fungus spreads rapidly in pear orchards (Ogawa and English, 1991; Petróczy et al., 2012; Sardella et al., 2016). The pathogen infects flowers during their vulnerable blooming stage and causes extensive fruit rots in late season resulting in substantial economic losses (Martini and Mari, 2014).

To defend against fungal infections, plants have developed a complex repertoire of innate immune responses (Saijo et al., 2018). Particularly upon challenge with necrotrophs, a fine-tuned layer of cell surface immune receptors initiates basal defenses known as pattern-triggered immunity (PTI) responses (Couto and Zipfel, 2016). In addition, even when PTI is overwhelmed, an alternative cell wall integrity mechanism may be activated to induce defense responses upon fungal challenge (Engelsdorf et al., 2018). Previous studies have highlighted the molecular mechanisms implemented by both M. laxa and stone fruits during the early infection stages (Balsells-Llauradó et al., 2020; Villarino et al., 2022). Thus, activation of the carbohydrate-active enzyme (CAZyme)-encoding genes is employed by M. laxa to achieve penetration (Balsells-Llauradó et al., 2020), whereas non-ribosomal peptide synthase (NRPS)-encoding genes are potentially involved in the production of fungal toxins during the colonization of fruit tissues (Villarino et al., 2022). On the other hand, the host molecular mechanisms against M. laxa were previously reported using comparative transcriptome analysis between resistant and sensitive peach genotypes, highlighting the involvement of hormone signal transduction, phenylpropanoid, flavonoid, and glutathione metabolismrelated genes in defense responses (Guidarelli et al., 2014; Balsells-Llauradó et al., 2020). Furthermore, many transcription factor (TF)-encoding genes, such as members of the WRKY, MYB, ERF, and NAC families, have been suggested to be involved in mediating the immune responses of fruits against M. laxa (Guidarelli et al., 2014; Balsells-Llauradó et al., 2020).

Despite the considerable losses due to blossom blight caused by M. laxa, the molecular interplay between this necrotrophic fungus and pear petals has not yet been deciphered. Previously, it was recorded that 19, 35, and 7 WRKY-encoding genes were involved in the regulation of defense responses in rose petals, grapevine, and strawberry flowers to Botrytis cinerea, respectively (Haile et al., 2017; Liu et al., 2018; Xiao et al., 2022). TFs are key regulators of the host transcriptional reprogramming during pathogen attack (Pandey and Somssich, 2009; Phukan et al., 2016), and among them, WRKYs are the most involved TF family in plant defense regulation (Zhang et al., 2016; Yuan et al., 2018; Zambounis et al., 2020; Xiao et al., 2022). Furthermore, the positive impact of WRKY TFs in rice and Arabidopsis responses to pathogen attacks was further highlighted as WRKYs can functionally cooperate in co-regulatory networks such as in the “COR-A” network (Berri et al., 2009).

In plant–pathogen interactions, the regulation of defense processes is usually mediated by the induction of a massive reprogramming, which is redirected to involve highly connected and complex molecular networks (Delplace et al., 2022). In order to reveal a detailed framework of the host responses to pathogen attack, the identification of the main regulators (hub genes) that orchestrate the transcriptional reprogramming beyond plant defense responses can be achieved through powerful system biology approaches such as weighted gene co-expression network analyses (WGCNA). Hub genes allow the identification of key counterparts with high connectivity degree in an interactive gene network, which usually play important roles in the regulation of several other genes and biological processes (Yu et al., 2017).

In this study, we compared the transcriptome dynamics of pear petals at two early stages of M. laxa infection between two cultivars characterized by different levels of sensitivity to this pathogen. We particularly explored the transcriptional network governing responses to M. laxa infection to uncover the molecular mechanisms underlying various aspects of immune responses. Specific differentially expressed genes (DEGs) involved in cell wall modification processes, pathogen recognition, and hormone signal transduction contributed to the differential and time-dependent defense-related responses among the two cultivars. The role of regulatory TFs associated with these responses in the relatively tolerant cultivar was also revealed by constructing co-expression regulatory networks to identify key genes regulating this interplay. These genes may be potential targets in future breeding programs toward a sustainable disease management strategy against M. laxa in pear.

In pear petals, the transcriptional responses to infection with M. laxa remain largely unknown. Thus, the overarching objective of this study was to decipher the transcriptome dynamics and regulatory mechanisms involved in the early stages of infection with M. laxa in the petals of two pear cultivars with differences in their sensitivity to fungal diseases. Our results suggest that, during these compatible interactions, both cultivars initiate a basal defense response to some extent, which is quite distinctive at different stages of the infection. However, in both cultivars, the specific time-dependent transcriptional reprogramming upon infection was accompanied by a failure to restrict fungal growth and disease progression, which was more pronounced in the susceptible cultivar Kristalli.

In both cultivars, the expression patterns of our RNA-seq data in petals suggest a dynamic reprogramming upon infection, as observed in other studies, such as in strawberry, grapevine, and rose petals infected by B. cinerea (Haile et al., 2017; Liu et al., 2018; Xiao et al., 2022). Notably, DEGs involved in plant–pathogen interaction, immune signal transduction, biosynthesis of secondary and primary metabolism, and other defense-related responses were induced in delay in the Sissy cultivar, which is relatively tolerant to M. laxa compared with the Kristalli cultivar. The expression profiles of the DEGs were also significantly different among the cultivars. Thus, in the Kristalli cultivar, the majority of DEGs were suppressed at 2 HAI, while in the Sissy cultivar, DEGs were mostly upregulated at 48 HAI. In contrast, a less abundant inventory of key genes and pathways involved in defense responses was observed at 48 HAI in the Kristalli cultivar. Significant downregulation of a high proportion of DEGs in the Kristalli cultivar is consistent with other studies on hosts infected with M. laxa (Balsells-Llauradó et al., 2020), as well as those infected with B. cinerea (Petrasch et al., 2019; Zambounis et al., 2020). This downregulated expression in Kristalli at 2 HAI was mainly attributed to the significant and rapid downregulation of specific DEGs and pathways associated with defense responses, biosynthesis of secondary metabolites, and plant–pathogen interactions. Notably, these pathways are suppressed during early responses in both compatible and incompatible interactions with B. cinerea (Smith et al., 2014; Kong et al., 2015; Petrasch et al., 2019).

The cell wall constitutes a structural barrier in response to pathogen attack (Haile et al., 2019; Wang et al., 2020b). At the same time, it is also the initial target for M. laxa during penetration as its disassembly contributes to susceptibility to pathogen invasion (Garcia-Benitez et al., 2019; Haile et al., 2019). Particularly, in the Kristalli cultivar, many DEGs involved in cell wall modification were differentially regulated at 2 HAI, as already described in strawberries upon Botrytis infection (Haile et al., 2019; Xiao et al., 2022). Thus, the constitutive upregulation of DEGs associated with hampering pathogen penetration, including CCR genes, along with the simultaneous suppression of PG genes, strongly mitigated the effect of susceptibility during M. laxa infection, as was also reported during B. cinerea infection in kiwi fruits (Cantu et al., 2008; Zambounis et al., 2020). Monilinia laxa may further stimulate petal softening by manipulating PL genes during the early infection stage, as has been reported in Botrytis (Lu et al., 2019). Indeed, in our study, we found nine PL genes that were downregulated in the K2 comparison group, implying their involvement in the induction of early defense responses. Previously, the suppression of the PL gene was reported to reduce susceptibility to B. cinerea and increase the concentration of cellulose and hemicellulose in tomatoes (Yang et al., 2017). Furthermore, Gl genes that degrade cellulose and hemicellulose (Blanco-Ulate et al., 2013) were mainly suppressed in the K2 group, indicating that they might partially contribute to the deployment of defense responses to some extent against M. laxa infection. On the other hand, in the same comparison group, the downregulation of CesA and LRR-EXT genes that influence cell wall extensibility and susceptibility (AbuQamar, 2014) might have facilitated pathogen colonization, along with the suppression of four XTH genes, which may be related to the thickening of cell walls (Liu et al., 2018; Wang et al., 2020b). The downregulation of a dirigent protein gene involved in the biosynthesis of lignans, which was previously found to be the most upregulated DEG in unripe strawberry fruits during cell wall reinforcement upon Botrytis elicitation (Haile et al., 2019), is another potential susceptibility-related response in the Kristalli cultivar that could be taken into consideration at the early infection stage. Overall, we are tempted to speculate that transcriptional changes in DEGs related to cell wall degradation and modification processes might contribute to Kristalli sensitivity to M. laxa, whereas the relative tolerance of the Sissy cultivar might not be regulated by such DEGs, at least at the early time point upon infection.

The high induction of DEGs involved in pathogen perception and signal transduction, particularly in Kristalli, highlights the important role of PTI in the pear–M. laxa pathosystem. Thus, in Kristalli, PTI seems to be mediated through the induction of an array of PRR genes, such as LRR-RLKs and LecRKs, which were however mainly suppressed at the early time point. Notably, LRR-RLK genes in the Sissy cultivar were constitutively upregulated at both time points, suggesting their putative role in higher innate immunity response. Several RLK genes have been previously shown to induce immune responses in necrotrophs (De Cremer et al., 2013; Zambounis et al., 2020). Furthermore, LecRKs are also involved in the defense responses of rose petals against Botrytis (Liu et al., 2018). It is worth mentioning that among the induced RLK genes, the G-type LecRKs are known to be involved in plant defense (Lannoo and Van Damme, 2014). In addition, in the S48 comparison group, the upregulation of a WAK receptor gene might provide further evidence of its involvement in M. laxa recognition, whereas another WAK gene, along with GLRs and PERKs, was suppressed in the K2 group. Previously, it was reported that during infection with Botrytis, WAK receptors were also upregulated in ripe strawberry fruits, lettuce, and rose petals (De Cremer et al., 2013; Liu et al., 2018; Haile et al., 2019). Our RNA-seq data also revealed that in S48, in contrast to the K48 group, various classes of branched kinases (STKs, CRKs, and CRCK) known to be involved in immune signaling pathways (Tang et al., 2017) were upregulated. Among them, STKs and CRKs are well-known plant defense regulators (Zambounis et al., 2020) and were similarly induced in lettuce after B. cinerea infection at 48 HAI (De Cremer et al., 2013).

Reprogramming of both secondary and primary metabolism putatively activates host defense responses (Agudelo-Romero et al., 2015). The upregulation of the GltS gene in the cv. Sissy at 48 HAI seems to further promote its tolerance, as it is known that glutamate triggers long-distance calcium-based plant defense signaling (Toyota et al., 2018). Secondary metabolites actively participate in defense pathways to tackle invading pathogens (Zaynab et al., 2018). Indeed, in the Kristalli cultivar, the biosynthesis of secondary metabolites was selectively activated in delay to some extent, which is consistent with previous studies (Agudelo-Romero et al., 2015; Xiong et al., 2018; Haile et al., 2019). These delayed defense responses in Kristalli were also supported by the highly enriched KEGG terms of “metabolic pathways” and “biosynthesis of secondary metabolites.” These results are consistent with previous findings in a tomato genotype susceptible to Botrytis, where the induction of defense responses and particularly the metabolic shunt for the biosynthesis of secondary metabolites were not observed until 48 HAI (Smith et al., 2014). It is known that JA has important regulating roles associated with defense responses against necrotrophic fungi (Blanco-Ulate et al., 2013; Sham et al., 2014; Agudelo-Romero et al., 2015; Haile et al., 2019). In our study, JA was also highlighted as the principal phytohormone in signaling transduction-mediated responses in Sissy based on the S48 group, as a lipoxygenase gene, which is related to JA biosynthesis, and was upregulated at 48 HAI. In this line, the activation of a CAD gene might also contribute to the elevated tolerance of the Sissy cultivar, as such genes are involved in lignin biosynthesis and resistance to pathogens (Li et al., 2022). In addition, one PAL-encoding gene, the key enzyme in the phenol biosynthesis pathway, was found upregulated in the S48 group, whereas its triggering is likely correlated with the Sissy higher competence to block M. laxa growth. In Kristalli, genes encoding 2-oxoglutarate-dependent dioxygenase, which is also involved in JA biosynthesis, were suppressed at 2 HAI and upregulated at 48 HAI. Notably, in the K2 group, the pathway of “alpha-linolenic acid metabolism” was enriched through the induction of linoleate 13S-lipoxygenase-encoding genes, suggesting a role of this pathway in the activation to some extent of primary biosynthetic pathways. Finally, we are tempted to speculate that although induction of the phenylpropanoid biosynthesis pathway was recorded in both cultivars, it was insufficient to restrict fungal growth and disease progression, as previously reported in ripe grapes after infection with B. cinerea (Agudelo-Romero et al., 2015).

Upon challenge with necrotrophic phytopathogens, a large set of TF families play important roles in the orchestration and regulation of defense mechanisms (Smith et al., 2014; Tsuda and Somssich, 2015; Liu et al., 2018; Haile et al., 2019). It is also possible that some of these TFs are involved in susceptibility (Agudelo-Romero et al., 2015). In our study, the transcriptional reprogramming of several TF-encoding DEGs was revealed by a quite different and time-dependent expression pattern between the two cultivars. Thus, in the K2 group, several TF-encoding genes belonging to the AP2/ERF family were mainly upregulated at the early time point. Considering that ERFs are responsive genes in ethylene biosynthesis and act as susceptibility factors upon challenging with pathogens (Agudelo-Romero et al., 2015; Petrasch et al., 2019), M. laxa may rapidly manipulate their induction in Kristalli, thereby accelerating further its susceptibility to M. laxa. Members of the bHLH family are also characterized as JA-mediated transcriptional regulators that act cooperatively with other TFs (Goossens et al., 2017), such as MYBs, in plant defense responses against pathogen attacks (Vailleau et al., 2002). The downregulation of MYBs and bHLHs in the K2 group further indicates the suppression of Kristalli defense mechanisms, enhancing its high susceptibility. In contrast, ZFPs and MUBs, which regulate overlapping signaling pathways and metabolic modulation toward the establishment of disease resistance, were among the most abundantly induced TFs in the S48 group.

A virulence factor exploited by M. laxa to manipulate host responses and facilitate colonization is the production of ROS, which leads to an oxidative burst that orchestrates the hypersensitive response (HR) and promotes susceptibility (Balsells-Llauradó et al., 2020). In our study, physiological indexes of pear petals among the two cultivars upon ML treatment showed higher levels of both TBARS and H₂O₂ in the sensitive cultivar than in the relatively tolerant cultivar Sissy. As no induction of any Rboh (respiratory burst oxidase homolog) genes that play an important role in redox homeostasis was observed in either cultivar, the scavenging strategy seems to be mediated by the induction of TXN and GST genes that directly participate in the ROS-scavenging pathway (Zheng et al., 2020), which were mainly upregulated in the Kristalli cultivar at both time points.

The activation of PRs and defense-related DEGs in the S48 group further indicates a delay in the immune responses. These responses were also evident to a lesser extent at the later time point in Kristalli, whereas at 2 HAI, such DEGs were primarily suppressed in the susceptible cultivar. However, in both cultivars, DEGs encoding homologs of major allergen proteins belonging to the PR-10 group family were constitutively upregulated at 48 HAI, and also in the K2 group, suggesting an enhancement of the JA-mediated transduction of defense signaling (Casañal et al., 2013). These genes were also induced in strawberry and kiwi fruits challenged with B. cinerea (Xiong et al., 2018; Zambounis et al., 2020). Similar expression patterns were recorded for another class of the PR-10 family encoding metalloendoproteinase (MMP) genes that play pivotal regulatory roles in homeostasis during PTI (Zhao et al., 2017), while a specific MMP protein is required for disease resistance against B. cinerea in tomato (Li et al., 2015). However, in Kristalli, the downregulation at the early time point of numerous disease resistance genes, such as those encoding defensin, elicitor-responsive, enhanced disease resistance, and TMV resistance proteins, further highlights the suppression of any defense-related responses promoting the susceptibility of this cultivar. However, in Sissy, all except one disease-related gene were upregulated at 48 HAI such as those encoding thaumatin, endochitinase, and peroxidase proteins.

At both stages of infection for the Sissy and Kristalli cultivars, respectively, an abundant number of DEGs encoding nutrients and ion transporters were either upregulated or downregulated. Several of these might have been utilized by the pathogen to obtain nutrients for its own needs from decayed host cells (Xiong et al., 2018). Among the upregulated DEGs encoding ABC transporters at 48 HAI, members of the G family are involved in the transport and secretion of secondary metabolites in plants challenged by pathogens, whereas fungi have developed similar transporters to export it from their cytosol (Khare et al., 2017).

Co-expression network analysis has become a powerful tool that is commonly used for the identification or prediction of new genes and TFs associated with crop pathogen resistance (Amrine et al., 2015; Liu et al., 2019; Cui et al., 2022; Kumar et al., 2022). Here, WGCNA was used to identify co-expression modules and hub genes correlated with M. laxa tolerance in pear petals. The analysis classified the pathogen-responsive genes of both cultivars into five co-expression modules. The most significant module (brown module) was positively correlated to the Sissy cultivar at 48 HAI (S48ML treatment). The functional analysis revealed that genes within this module were mainly involved in DNA- and FAD-binding molecular functions, which further supports the hypothesis that this co-expression module is involved in the transcriptional regulation of the Sissy cultivar in response to M. laxa. Five TFs and cytokinin dehydrogenase-encoding genes were identified as hubs in this module regulatory network.

Among the TF hubs, three belonged to the WRKY family: WRKY28, WRKY65, and WRKY71. A study conducted on the Arabidopsis thaliana and B. cinerea pathosystem revealed that overexpression of AtWRKY28 led to A. thaliana disease resistance through the positive regulation of JA and ET biosynthesis-related genes (Wu et al., 2011). In pear petals, WRKY28 was highly upregulated (log2FC = 8.33) in the S48 comparison group compared with the other TFs. Furthermore, WRKY65 and WRKY71 were upregulated only in this comparison group (S48). These genes are known to be involved in plant resistance to fungal and bacterial pathogens. The silencing of PlWRKY65 from Paeonia lactiflora induced a higher sensitivity of the mutant plants to Alternaria tenuissima infection (Wang et al., 2020a), while overexpression of maize ZmWRKY65 in Arabidopsis transgenic plants enhanced their resistance to B. cinerea and Pseudomonas syringae pv. tomato DC3000 (Pst) infection via the activation of PR genes (Huo et al., 2021). Both WRKY65 orthologs exerted a regulatory effect on disease hormone signaling pathways, resulting in a more resistant phenotype. The WRKY71 TF was also a hub gene and had the same expression pattern as previous TFs in the Sissy cultivar. Overexpression of such a WRKY gene in rice (OsWRKY71) also resulted in improved resistance to Xanthomonas oryzae pv. oryzae (Liu et al., 2007).

DEGs encoding TGA3-like and HSF-B1a TFs were also identified as hub genes in the regulatory network which correlated with the relatively tolerant cultivar, whereas both genes were upregulated exclusively in the Sissy cultivar at 48 HAI. HSF-B1 TFs have been demonstrated to play a pivotal role in the activation of defense priming in Arabidopsis (Pick et al., 2012). TGA3 is a member of the TGA TF family known as regulators of PR genes since they constitutively interact with the non-expresser PR gene 1 (NRP1) (Jakoby et al., 2002), while the tga3 Arabidopsis mutant is defective in basal pathogen resistance (Kesarwani et al., 2007). This suggests that TGA3-like TFs play a central role in the activation of PR genes in the cv. Sissy. In contrast, plant-derived cytokinins promote Arabidopsis resistance to Pst DC3000 through TGA3-dependent activation of ARR2, a cytokinin-activated transcription factor (Choi et al., 2010). This supports further our results, as the main hub gene in the regulatory network of the Sissy 48 HAI-related module was a cytokinin dehydrogenase 7-encoding gene. Such cytokinin-related genes have been reported to support plant responses and resistance to necrotrophic pathogens (Li et al., 2022; Zhu et al., 2022).

Gene co-expression network analyses also have limitations, even if they provide valuable information about potential genes and their correlation, as they do not indicate the nature of the relationship between genes that are co-expressed. Thus, further analyses are necessary to define which genes are regulated by the identified TFs inducing Sissy tolerance to M. laxa, such as ChIP-seq (Chen et al., 2018), which can be integrated into our data to allow the identification of the target genes of TFs.

Comparative transcriptomics is a powerful tool for identifying the key genes and pathways that are differentially expressed during pathogen invasion. In this study, the application of this technique provided a comprehensive view of the molecular mechanisms underlying the differential tolerance of the petals of the two pear cultivars, Sissy and Kristalli. Transcriptome profiling revealed cultivar-specific and time-dependent responses after M. laxa inoculation. In particular, GO terms and KEGG pathway enrichment analyses showed an earlier transcriptome reprogramming in Kristalli compared with the Sissy cultivar, whereas defense-related DEGs were mainly suppressed. DEGs involved in signal transduction, biosynthesis of secondary and primary metabolism, and other defense-related responses were mainly induced in the relatively tolerant Sissy cultivar at 48 HAI. These results allow us to better decipher the pear–M. laxa pathosystem identifying the main pathways involved in pear petal defense responses against M. laxa. The integration of a weighted gene co-expression network analysis with transcriptome profiling allowed the identification of six hub genes highly correlated to tolerance to M. laxa, including three WRKYs, one TGA, and one HSF TF, along with a cytokinin dehydrogenase-encoding gene, whose orthologs were already reported to be involved in crop pathogen resistance. The insights gained from our research may offer novel disease control strategies, such as new target genes for genome editing to develop new resistant and transgene-free pear cultivars.

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 below: https://www.ncbi.nlm.nih.gov/genbank/, PRJNA1049420.

MA: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Writing – original draft, Writing – review & editing. PT: Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. AB: Formal analysis, Investigation, Writing – original draft. AD: Investigation, Writing – original draft. MM: Investigation, Writing – original draft. CD: Writing – review & editing. DT: Conceptualization, Methodology, Writing – original draft. EP: Conceptualization, Methodology, Writing – original draft. AM: Data curation, Formal analysis, Investigation, Validation, Writing – review & editing. LS: Writing – review & editing. AZ: Conceptualization, Data curation, Formal analysis, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing.