Pear Scab Disease Suppression by Pseudomonas capeferrum NFX1 Is Mediated by Direct Antagonism Against Venturia pyrina and Pear Defense Priming

TL;DR

A new strain of Pseudomonas capeferrum, NFX1, effectively controls pear scab disease by fighting the fungus and boosting pear defenses.

Contribution

Pseudomonas capeferrum NFX1 is introduced as a novel biocontrol agent against pear scab with genomic and phenotypic evidence.

Findings

Strain NFX1 inhibits Venturia pyrina growth and spore germination.

NFX1 primes pear defenses and induces systemic resistance.

New methods for measuring scab severity and fungal DNA detection were developed.

Abstract

Pear scab, caused by Venturia pyrina, poses a threat to pear cultivation, with particularly severe consequences for Portugal’s high-value Rocha pear industry. Despite its economic impact, few biological control agents are currently available. In this work, the phenotypic and genomic characterization of Pseudomonas capeferrum NFX1 is performed and its role as an effective biocontrol agent against V. pyrina is reported. Detailed genomic analysis revealed that strain NFX1 and other members of the Pseudomonas capeferrum species contain key biosynthetic gene clusters involved in pathogen antagonism, including the cyclic lipopeptide putisolvin. Phenotypic assays showed that strain NFX1 significantly inhibited V. pyrina growth, spore germination, and reduced pear scab lesion severity and fungal colonization in detached leaf assays. Moreover, strain NFX1 reprogrammed the Rocha pear leaf transcriptome to be consistent with a priming state and induced systemic resistance. A novel image-based method quantifying lesion darkening as a proxy for pear scab severity in detached leaves and a qPCR assay targeting the V. pyrina ef1-α gene and optimized for fungal DNA detection in infected pear leaves were also developed, thereby establishing a laboratory workflow specifically tailored to biocontrol evaluation against V. pyrina. Ultimately, the obtained results demonstrated the potential of P. capeferrum NFX1 for sustainable pear scab control.

Figures

Full text

1. Introduction

Rocha pear (Pyrus communis L. cv. ‘Rocha’) plays a central role in Portugal’s fruit production sector, particularly in the Oeste region, where it benefits from Protected Designation of Origin status [1]. Rocha pear cultivation spans approximately 11,000 hectares and yields an average of 173,000 tons per year, representing the vast majority (about 99%) of Portugal’s total pear production [2,3]. Around 60% of this output is exported, generating close to €85 million in annual revenue and providing employment for an estimated 10,000 people across the production and supply chain [2].

Rocha pear production is increasingly threatened by pear scab, an economically important disease affecting pears worldwide [4,5]. In Europe, this disease is caused by the ascomycete, Venturia pyrina, which is responsible for significant losses in both yield and fruit quality, estimated to reach 40–80% in affected orchards [4,6,7]. Under conducive temperature and leaf-wetness regimes, V. pyrina ascospores (commonly found in overwintered leaves) germinate, form appressoria, and penetrate host tissues, after which lesions sporulate to generate conidia that drive successive secondary infection cycles on expanding foliage and fruit [8,9]. Management of pear scab has traditionally relied on regular applications of chemical fungicides [7]; however, this approach has raised increasing concerns regarding environmental sustainability, food safety, and the emergence of fungicide-resistant V. pyrina populations [4,7]. These challenges are further compounded by evolving regulatory frameworks, such as the European Green Deal, which aims to reduce pesticide use by 50% by 2030 [10]. Collectively, these factors emphasize the urgent need for more sustainable and environmentally responsible strategies for disease control. In this context, biological control agents (BCAs) have emerged as a promising sustainable alternative. However, the effective identification of bacterial BCAs requires robust screening systems. Field-based evaluations are often impractical due to their time-consuming nature and dependence on natural infection events. To address these limitations, laboratory-based screening methods, such as in vitro inhibition assays and detached leaf tests, are widely used to evaluate the efficacy of candidate BCAs against fungal pathogens [11,12,13]. Nonetheless, the application of these methods to study the impacts of BCAs in limiting Rocha pear scab disease and its agent, V. pyrina, remains largely unexplored. Several plant-associated Pseudomonas strains are known to suppress the growth of a wide range of fungal pathogens, deploying multiple antagonistic mechanisms such as antibiotic and lipopeptide production, iron competition via siderophores, lytic enzymes, volatile compounds, and induced systemic resistance (ISR) in the host plant [14,15], thereby positioning Pseudomonas as a key resource for developing sustainable, multi-mechanism strategies that can strengthen integrated plant disease management and reduce reliance on chemical fungicides [16].

In this work, we performed the genomic characterization of Pseudomonas capeferrum NFX1 and reported its functional validation as a biological control agent against the Rocha pear scab disease agent, V. pyrina. Strain NFX1 was selected from a collection of 11 bacterial isolates preliminarily screened for pear scab biocontrol activity in detached leaf assays and prioritized for further characterization based on its previously reported phytohormone-modulating traits [17]. Several in vitro assays, including the inhibition of fungal growth, interference with spore germination, and reduction in disease symptoms on detached pear plant leaves, were performed. Moreover, to support and enhance biocontrol discovery and assessment in this pathosystem, a novel imaging-based method for quantifying lesion severity through normalized lesion pixel intensity was developed, establishing lesion darkening as a quantitative proxy for pear scab disease severity in detached leaves. This metric is consistent with agronomic descriptions of symptom progression in leaves and fruits evolving to dark brown as the infection progresses [18,19]. In addition, a real-time qPCR assay for the precise molecular quantification of V. pyrina DNA in plant tissues was developed. Transcriptomic profiling of NFX1-treated and not-treated infected leaf lesions was conducted and provided insights into the modulation of host defense responses during V. pyrina infection. To our knowledge, this is the first study to establish a comprehensive and standardized laboratory workflow for assessing bacterial BCAs against pear scab, and it identifies the P. capeferrum NFX1 strain as a promising biocontrol candidate for future application in Rocha pear protection.

2. Results

2.1. Genomic Characterization of Strain NFX1 Reveals Its Affiliation with the Pseudomonas capeferrum Species

The genome of strain NFX1 was composed of a single chromosome of 5,986,175 bp, presenting a GC% of 62.9% (Figure 1). A total of 5393 genes were annotated, of which 5220 corresponded to protein-coding sequences (CDS) and 103 to RNA-related genes.

Phylogenomic analysis revealed that the strain NFX1 genome was highly similar to that of the strain Pseudomonas capeferrum WCS358 (Average Nucleotide Identity—ANI—97.79%; digital DNA–DNA hybridization—dDDH—82.4%) (Table 1). BLAST and FastANI analysis also confirmed the increased synteny between the genomes of strains NFX1 and WCS358 (Figure 1), indicating that strain NFX1 is a member of the P. capeferrum species. In addition, phylogenomic analysis also showed that several other Pseudomonas strains, isolated from different geographical locations and habitats, and known for their fungal and bacterial antagonistic activities (described below), were identified as members of the P. capeferrum species (Table 1). These strains’ genomes presented >95% ANI and >70% dDDH values when compared to the genome of P. capeferrum WCS358. A similar genomic structure (genome size, GC%) was observed in all the members of the P. capeferrum species (Table 1).

2.2. Increased Prevalence of Multiple Genes Involved in Antagonistic Features in the Genome of NFX1 and Other Pseudomonas capeferrum Strains

AntiSMASH analysis indicated the presence of several secondary metabolite biosynthetic gene clusters in the genome of strain NFX1, including two non-ribosomal peptide (NRP) metallophores, a non-ribosomal peptide synthetase (NRPS), and a homoserine lactone (Table 1). Detailed analysis revealed that the NFX1 NRP metallophore cluster 1 (DJ563_RS00700–DJ563_RS00725) presented the highest similarity (~50%) to the functional pyoverdine-SXM cluster (BGC0002693) of Pseudomonas sp. SXM-1, while the NFX1 NRP-metallophore cluster 2 (DJ563_RS02160-DJ563_RS02220) contained a pvdL gene and other homologs of the functional pyoverdine Pf-5 cluster (BGC0000413) of P. protegens Pf-5.

Analysis of the NFX1 NRPS cluster (DJ563_RS06135–DJ563_RS06145) showed that it highly identified (98%) with the functional putisolvin biosynthesis cluster (BGC0000411) of P. putida PCL1445. Interestingly, the increased identity between the putisolvin biosynthesis cluster of P. putida PCL1445 and P. capeferrum NFX1 (Table 1) suggested their close taxonomic identity. In addition, genomic analysis also showed that the NFX1 homoserine lactone biosynthesis genes (DJ563_RS16665–DJ563_RS16680) presented a high similarity to the strain PCL1445 functional ppuI-rsaL-ppuR quorum-sensing gene. A similar cluster was previously described in strain IsoF [20], and our genomic analysis revealed that ppuI-rsaL-ppuR is also prevalent in all other P. capeferrum strains (overall identity of ~98%) (Table 1).

In addition, a recent study showed that strain IsoF, here reclassified as a member of the P. capeferrum species, presented antagonistic activities against a wide range of soil and plant-associated Gram-negative bacteria through the expression of a type IVB secretion system, encoded by the kib gene cluster, in a contact-dependent manner [21]. Genomic data showed that strain NFX1 possesses a kib gene cluster homolog (98.8% identity). Moreover, the kib gene cluster was also prevalent amongst most P. capeferrum strains, except for strain WCS358 and TDA1 (Table 1).

2.3. Pseudomonas capeferrum NFX1 Presented Antagonistic Activities Against V. pyrina

To test the antifungal properties of strain NFX1, two distinct assays were performed: a plate assay in rich solid media to assess the general abilities of strain NFX1 in decreasing V. pyrina growth (counts of V. pyrina colonies) and a more detailed spore germination assay in liquid medium to evaluate the effect of strain NFX1 on the germination of V. pyrina spores based on a three-level rating scale (Figure S1).

Plate assays revealed that the presence of P. capeferrum NFX1 led to the inhibition of V. pyrina growth, with significant impacts on colony density, when compared to the not-treated control (Figure 2A). The presence of strain NFX1 resulted in a 27.4% reduction in V. pyrina colony density, whereas the application of a chemical fungicide, Cerimonia^®^ (positive control), strongly inhibited fungal development (no visible colonies were observed) (Figure 2B).

The spore germination assays also showed the antifungal effects of P. capeferrum NFX1 (Figure 2C,D). In the not-treated control, fungal spores germinated extensively and formed dense germ tubes by 24 h post-inoculation (hpi), which developed into branched hyphal structures by 96 hpi. In contrast, the application of strain NFX1, or a chemical fungicide, significantly reduced V. pyrina spore germination and the subsequent development of hyphae (Figure 2C,D). Fisher’s exact test revealed a highly significant association between treatments and inhibition scores at both 24 hpi and 96 hpi (p < 2.2 × 10^−16^). Subsequent analysis showed that, at 24 hpi (Figure 2C,D), NFX1 application significantly inhibited V. pyrina spore germination across all tested bacterial inoculum concentrations (OD_600 nm_ = 0.05, 0.1, and 0.5), as reflected by the inhibition scores (score 2 = 30–70% inhibition; score 3 = >70% inhibition relative to the not-treated control; Figure 2D). By 96 hpi, the application of strain NFX1 resulted in a marked increase in inhibitory activity, with all tested concentrations achieving a strong suppression of V. pyrina germination (score 3, >70% inhibition) (Figure 2D). Importantly, at both 24 hpi and 96 hpi, all tested concentrations of strain NFX1 achieved levels of fungal suppression comparable to those observed with the highest tested dose of Cerimonia^®^ (Figure 2C,D).

2.4. The Application of Pseudomonas capeferrum NFX1 Significantly Reduced Pear Scab Disease Severity in Detached Leaf Assays

To assess the biocontrol potential of P. capeferrum NFX1 against V. pyrina in the Rocha pear plant, a detached leaf assay was conducted. Disease severity was evaluated using lesion pixel intensity as a proxy for symptom development (Figure S2). This approach was supported by time course analysis showing that the lesion area remained largely unchanged between 1 and 2 weeks post-infection (46.8 ± 1.44 mm^2^ vs. 48.1 ± 1.50 mm^2^; p = 0.5551, Welch’s Two-Sample t-test), whereas normalized lesion severity values significantly increased from 0.14 ± 0.08 to 0.72 ± 0.10 (p = 0.0044) (Figure S3). Additionally, leaf lesion areas were used in a quantitative PCR (qPCR) assay to quantify V. pyrina colonization.

Detached leaf assays demonstrated that the foliar application of P. capeferrum NFX1 led to a marked reduction in scab lesion formation when compared to the infected control (Figure 3A). Statistical analysis confirmed that strain NFX1 significantly reduced pear scab disease severity by nearly 52%, a level of inhibition significantly greater than that observed with the recommended dosage of the Cerimonia^®^ fungicide (Figure 3B). In addition, the NFX1 treatment did not cause any visible damage or adverse effects on non-infected control leaves (Figure S4).

The qPCR assay targeted the V. pyrina translation elongation factor 1α (ef1-α) gene and showed high linearity (R^2^ = 0.999) and 87.7% amplification efficiency, confirming reliable quantification of V. pyrina DNA in infected leaf tissues (Figure S5). The qPCR results revealed significant differences in the quantification of V. pyrina DNA, standardized by leaf lesion area, when comparing strain NFX1 and Cerimonia^®^ treatments to the not-treated infected control (Figure 3C). Infected leaves treated with strain NFX1 contained, on average, ~500 pg of fungal DNA, representing a statistically significant reduction (~4-fold) compared to the not-treated control, which contained about 2000 pg of V. pyrina DNA. Cerimonia^®^-treated leaves contained a lower fungal DNA quantity (~600 pg), also indicating a suppressive effect relative to the not-treated control. The NFX1-treated samples exhibited slightly lower fungal DNA levels than Cerimonia^®^-treated ones; however, this difference was not statistically significant (Figure 3C). Furthermore, a significant correlation between fungal DNA levels and disease severity, as measured by lesion pixel intensity, was observed, with a Pearson correlation coefficient of 0.63 (p = 0.0012).

2.5. Transcriptomic Insights into the Effects of Pseudomonas capeferrum NFX1 in Rocha Pear Leaves Under V. pyrina Infection

Transcriptomic analyses were conducted to ascertain the impacts of strain NFX1 in modulating plant gene expression under V. pyrina infection. Comparing NFX1-treated with not-treated infected leaves, a total of 717 Differentially Expressed Genes (DEGs) (311 upregulated, 406 downregulated) were identified. From these DEGs, 389 (~54%) were annotated with Gene Ontology (GO) Biological Processes terms (164 up, 225 down) (File S1, Sheet 1). The top 25 upregulated genes were mainly involved in pathways related to defense signaling, detoxification, and protein stabilization. These included genes encoding TIFY 8-like (jasmonic acid signaling), G-type lectin receptor-like kinase SD2-5 (pathogen recognition and defense signaling), dehydration-responsive element-binding protein 1D-like (stress-responsive transcription factor), and several detoxification-related proteins (e.g., Glutathione S-transferase) (File S1, Sheet 1). In contrast, the top 25 downregulated genes were linked to pathways involved in jasmonate biosynthesis, secondary metabolism, and oxidative defense. Strong repression was observed for genes encoding allene oxide synthase 3-like (jasmonic acid biosynthesis), Peroxidase P7-like and Laccase-7-like (oxidative stress and lignin catabolism), and multiple Patatin-like (lipid catabolism) proteins. Other downregulated genes included those encoding sorbitol dehydrogenase (sugar metabolism), Pectinesterase 2 (cell wall modification), berberine bridge enzyme-like 8, and (R)-mandelonitrile lyase 1-like (secondary/defense metabolism) (File S1, Sheet 1).

Gene set enrichment analysis (GSEA) showed 146 GO Biological Process gene sets (FDR ≤ 0.25) skewed toward positive enrichment (119 up vs. 27 down) (File S1, Sheet 2). The enrichment background was dominated by photosynthesis- and chloroplast-centered Biological Processes, spanning light capture, electron transport, and plastid biogenesis, alongside anabolic energy metabolism (Figure 4). Additional metabolic and organelle programs (e.g., mitochondrial gene expression/translation, isoprenoid/tetraterpenoid metabolism) also aligned with a generalized plastid-associated energy metabolism (File S1, Sheet 2). The negatively enriched sets (n = 27) centered on secondary metabolism and membrane/vesicular trafficking, with the strongest effects in glucosamine-containing compound, lignin, and phenylpropanoid metabolic processes, carbohydrate-linked processes, ER stress/ERAD and endosomal–vacuolar/Golgi–plasma membrane transport, plus defense responses to fungus (Figure 4, File S1, Sheet 2).

To provide additional context, two complementary comparisons were also performed: NFX1-treated vs. not-treated, not-infected leaves (File S1, Sheets 3 and 4) and infected vs. not-infected not-treated leaves (File S1, Sheets 5 and 6). Among the three comparisons, the largest transcriptional reprogramming was observed in NFX1-treated vs. not-treated, not-infected leaves, with 1658 DEGs identified (File S1, Sheet 3). GSEA revealed 148 GO Biological Process gene sets (133 positively enriched, 15 negatively enriched), with predominant enrichment of photosynthesis-related processes, chlorophyll and tetrapyrrole metabolism, electron transport, carbon fixation, and plastid organization. Negative enrichment was observed for hormone-related and secondary metabolic processes, including salicylic acid metabolism, response to abscisic acid, and phenylpropanoid and lignin metabolic processes (File S1, Sheet 4). The infection-only comparison identified 146 DEGs (File S1, Sheet 5) and 58 enriched GO Biological Process terms (three positively enriched, 55 negatively enriched). The most represented negatively enriched categories comprised plastid and chloroplast organization, mitochondrial ATP synthesis coupled electron transport and mitochondrial membrane organization, and ribosome assembly and cytoplasmic translation, together with nucleosome organization and nucleosome assembly (File S1, Sheet 6).

2.6. P. capeferrum NFX1-Induced Defense Priming in Pear Leaves Is Linked to Hormonal Reprogramming and Immune Defense Response Regulation

Interestingly, beyond the major transcriptional trends observed, the dataset also revealed direct evidence of hormonal reprogramming and immune defense response regulation as core components of the strain NFX1-induced transcriptional activation in pear leaves. For example, ethylene signaling components, including several ERF/AP2 transcription factors, were strongly induced, whereas negative regulators such as ABR1-like and EIN3-like were repressed (Figure 5, File S1), suggesting a shift toward rapid ethylene-driven transcriptional responses. Abscisic acid perception was enhanced through the induction of PYL4-like, while several abscisic acid-responsive effectors were reduced (Figure 5, File S1), reflecting receptor-level readiness but limited downstream activation. Auxin pathways were also reconfigured, with increased expression of transport and perception components (auxin efflux carrier component 5-like, BIG GRAIN 1-like) alongside the repression of catabolic regulators such as indole-3-acetic acid (IAA)-amino acid hydrolase ILR1-like 4 (Figure 5, File S1), consistent with altered auxin flux and growth–defense trade-offs. Jasmonate signaling displayed a more complex profile, with TIFY repressors (TIFY6B-like, TIFY8-like) upregulated, while allene oxide synthase 3-like and TIFY10B-like were downregulated, consistent with restrained jasmonate biosynthesis but primed regulatory control. In parallel, immune defense pathways were selectively modulated. Several genes encoding receptor-like kinases and co-receptors, including G-type lectin and LRR-RLKs (Leucine-Rich Repeat Receptor-Like Kinases), were induced, whereas other receptor-like proteins were repressed, suggesting the targeted reinforcement of perception modules. Intracellular surveillance followed a similar pattern: genes encoding disease resistance proteins such as TMV resistance protein N-like and enhanced disease resistance 2-like were upregulated, whereas other defense regulators, including DMR6-like, SAR deficient 1-like, and several NLRs (RPM1-like and TAO1-like) were downregulated (Figure 5, File S1), pointing to an overall reduced R-protein activity. This selective regulation was accompanied by broad transcriptional activation, with a substantial portion of upregulated genes corresponding to transcription factors from the WRKY, MYB, bHLH, and TCP families, linking defense signaling and immune regulation. Meanwhile, pathogenesis-related proteins, including several endochitinases and thaumatin-like proteins, were predominantly downregulated (Figure 5, File S1), indicating strengthened perception and transcriptional control while restraining downstream effectors.

3. Discussion

Pear scab, caused by V. pyrina, remains one of the most serious fungal diseases affecting P. communis cultivation, with particularly damaging consequences for the economically vital Rocha pear orchards of Portugal. To address the limitations of chemical control under increasing regulatory pressure, in this work, we describe the suppressive effects of P. capeferrum NFX1 on pear scab disease and its potential role as a biological control agent.

Detailed genomic analysis revealed that strain NFX1 is a member of the P. capeferrum species, a known Pseudomonas species presenting plant disease-suppressive activities [22,23]. Moreover, phylogenomic analysis revealed and confirmed that several other completely sequenced strains are members of the P. capeferrum species (Table 1), including those presenting strong plant disease-suppressive abilities. For instance, strain IsoF has been shown to display contact-dependent antagonism against a wide range of Gram-negative bacteria and to protect tomato plants against the phytopathogen Ralstonia solanacearum in a type IVB secretion system-dependent manner [21]; strain B21-047 presented the ability to directly limit the growth of the lettuce plant pathogens Xanthomonas hortorum pv. vitians, Pseudomonas cichorii, and Pectobacterium carotovorum [24]; and strains UC_21.3 A.1 and UC_21.30 A.1 presented antifungal activities against Botrytis cinerea, and their supernatants presented nematicidal activity against the tomato root knot nematodes Meloidogyne hapla and Meloidogyne incognita [25].

Interestingly, our genomic analysis indicated that all available P. capeferrum genomes, including that of strain NFX1, possessed a biosynthetic cluster involved in the production of the antifungal compound putisolvin. This cluster was first functionally described in P. putida PCL1445 [26,27]. Putisolvins (or, alternatively, capesolvins) are cyclic lipodepsipeptides with a hexanoic lipid chain connected to the N-terminus of a 12-amino-acid peptide moiety known for their surfactant and biofilm-disrupting activities and have also demonstrated antifungal effects, including zoosporicidal activity and growth inhibition of fungal pathogens such as B. cinerea, Rhizoctonia solani, and Phytophthora capsici [28,29]. In addition, our analysis also indicated that most P. capeferrum genomes contained other genetic factors potentially involved in biocontrol activities, including the ppuI-rsaL-ppuR quorum-sensing genes involved in the production of homoserine lactones and the regulation of biofilms and putisolvin biosynthesis [30]; NRP metallophore clusters involved in siderophore biosynthesis, a mechanism known to regulate the beneficial effects of P. capeferrum WCS358 and its biocontrol activities [22]; and the type IVB secretion system (kib genes) involved in contact-dependent antagonistic activities [21]. Although genomic analysis revealed biosynthetic gene clusters associated with antimicrobial activity, including putisolvin, the direct contribution of these loci to the antifungal phenotype observed against V. pyrina remains to be functionally validated. Future studies involving targeted mutagenesis of the putisolvin biosynthetic cluster, followed by complementation analyses, would be required to definitively establish causal links between specific secondary metabolite production and the antifungal effects observed in dual plate and detached leaf assays.

The assays conducted in this work revealed that P. capeferrum NFX1 significantly reduced V. pyrina growth and spore germination to a similar extent as the application of Cerimonia^®^, a known commercial antifungal chemical product. Moreover, the biocontrol potential of strain NFX1 was further confirmed in detached leaf assays, where it achieved a 52% reduction in pear scab disease severity, linked to a decrease in the fungal population, corroborating its strong disease suppressive activities. To our knowledge, this is one of the first reports of a bacterial biocontrol agent against European pear scab disease caused by V. pyrina. Based on the genomic data and phenotypic assays, P. capeferrum NFX1 likely acts as an antagonist against V. pyrina, directly competing with the pear scab agent and limiting its growth and colonization activities, perhaps through the biosynthesis of putisolvin, as well as other potential molecular mechanisms regulating the bacteria–fungal interaction, such as siderophores and the type IVB secretion system effectors.

To explore the transcriptional responses associated with strain NFX1 treatment during pear scab suppression, a detailed transcriptomic analysis of the detached leaf assay was performed. The obtained results indicated that NFX1-treated infected leaves displayed substantial transcriptional changes with upregulated energy production, detoxification, and proteostasis genes, while repressing jasmonate biosynthesis and multiple oxidative/secondary metabolic pathways; concordantly, GSEA showed the positive enrichment of photosynthesis- and plastid-centered programs and negative enrichment of the phenylpropanoid/lignin metabolism and other catabolic pathways. Notably, the comparison between infected and non-infected, untreated leaves revealed the predominant negative enrichment of plastid-, mitochondrial-, and chromatin-associated processes, indicating the repression of core metabolic and organellar functions during fungal colonization. Considering the reduced V. pyrina DNA abundance and lesion severity observed in NFX1-treated leaves, together, these observations suggest that the transcriptional configuration of NFX1-treated infected leaves partly reflects the alleviation of infection-associated metabolic disruption under reduced fungal pressure. In addition, we also found that the application of strain NFX1 greatly impacted the transcription of genes involved in plant hormonal reprogramming and the regulation of plant immune responses, suggesting that P. capeferrum NFX1 not only suppresses the pathogen but is also associated with the broader modulation of plant transcriptional responses. Indeed, NFX1-treated leaves displayed transcriptional patterns consistent with an ethylene-induced, jasmonate-restrained configuration and enhanced auxin transport/perception, suggesting a growth–defense transcriptional balance consistent with induced systemic resistance. Yet further studies, including the direct quantification of endogenous hormone levels, will be required to clarify whether gene expression changes in NFX1-treated leaves reflect functional hormonal shifts. Nevertheless, NFX1 treatment in the absence of infection also induced the enrichment of photosynthesis and carbon fixation pathways, together with the negative enrichment of hormone-related and secondary metabolic processes, consistent with an infection-independent priming state associated with enhanced energy metabolism and modulation of defense pathways. These observations are consistent with previous reports describing the ability of P. capeferrum WCS358 to induce systemic resistance in a wide variety of plants [31,32,33]. Interestingly, P. capeferrum NFX1 was isolated based on its ability to catabolize auxins such as IAA and contains the iac genes involved in IAA catabolism [34], which may be relevant in shaping plant auxin levels. Overall, the integration of genomic, phenotypic, and transcriptomic evidence supports a coherent mechanistic framework. The strong in vitro inhibition of fungal growth and spore germination is consistent with the presence of the putisolvin biosynthetic gene cluster, which is known to interfere with fungal membrane integrity and hyphal development. In planta, siderophore-associated NRP metallophore clusters, together with the type IVB secretion system, may contribute to iron sequestration and competitive interactions that restrict fungal proliferation. Together, these genomic features provide plausible mechanisms underlying the reduced fungal DNA accumulation and diminished lesion severity observed in detached leaves. Transcriptomic changes in NFX1-treated leaves further indicate the modulation of host regulatory networks consistent with induced systemic resistance and enhanced energy metabolism. However, the extent to which these transcriptional shifts represent primary drivers of resistance or secondary consequences of reduced fungal colonization remains to be fully elucidated.

Importantly, this work also established the first reproducible, standardized laboratory pipeline for evaluating a bacterial biocontrol agent specifically against V. pyrina. The pipeline was based on quantifying lesion severity by measuring the mean pixel intensity of lesion areas, as well as a novel qPCR assay targeting the V. pyrina elongation factor 1-alpha (ef1-α) gene in the total DNA obtained in planta. The primer set was designed to ensure high specificity for V. pyrina in infected plant tissue, avoiding off-target amplification in P. communis. Additionally, a statistically significant and moderately strong correlation was observed between symptom severity and fungal DNA quantity, supporting the view that more severe symptoms are generally associated with increased fungal DNA quantity. Still, the two metrics capture different dimensions of the host–pathogen interaction and should be considered complementary rather than interchangeable tools for assessing biocontrol efficacy. Furthermore, while qPCR offers a sensitive measure of fungal accumulation, potential matrix effects from plant-derived compounds cannot be entirely excluded.

By integrating genomic, phenotypic, and transcriptomic evidence, this work identified P. capeferrum NFX1 as a strong candidate for the development of a sustainable solution to control European pear scab disease in Rocha pear. Future studies evaluating NFX1 under field conditions will be essential to assess strain persistence on pear tissues, maintenance of antifungal activity under environmental variability, and potential impacts on host growth and physiological performance. Such studies will be critical to validate the stability and practical applicability of NFX1-mediated disease inhibition under orchard conditions and to support the development of suitable NFX1-based product formulations.

4. Materials and Methods

4.1. Bacterial Strain Selection and Genomic Characterization

Pseudomonas capeferrum NFX1 was previously isolated from the rhizosphere of an Eucalyptus tree in Portugal [17]. NFX1 was part of a collection of 11 bacterial isolates preliminarily screened for pear scab biocontrol activity using detached leaf assays (Section 4.5). Preliminary results (Supplementary Figure S6) identified several strains showing a reduction in lesion severity relative to the not-treated infected control, although these differences were not statistically significant (p > 0.05). Among them, NFX1 was prioritized for further characterization based on its previously reported ability to degrade indole-3-acetic acid (IAA) [17], a trait associated with phytohormone modulation and relevant to plant–microbe interactions. The strain was routinely cultured overnight at 23 °C in Tryptic Soy Broth (TSB) and used in the experiments.

The complete genome of strain NFX1 was sequenced using the services of the Microbes NG company (https://microbesng.com/, accessed on 3 March 2026) (Birmingham, UK). Sequencing libraries were generated using SQK-RBK114.96. Sequencing was performed on a GridION (Oxford Nanopore Technologies, Oxford, UK) using an R10.4.1 flow cell, with base-calling model r1041_e82_400bps_hac_v4.2.0. Reads were randomly subsampled to 50× coverage using Rasusa (V 0.7.1) [35] and assembled using Flye (V2.9.2-b1786) [36]. The assembly was polished using previously sequenced Illumina Mi-Seq data obtained as described in Urón et al. [37]. The Illumina reads were trimmed using Trimmomatic v. 0.30 [38] (Q30, sliding window 4) and used for the polishing step performed through Pilon v.1.2.3 [39] (standard parameters). The final genome sequence of strain NFX1 was submitted to the NCBI and can be found under the accession number GCF_051123555.1.

The genome circular view was created using Proksee (https://proksee.ca/, accessed on 3 March 2026) [40]. Secondary metabolite clusters were identified using antiSMASH v.7.1 [41].

4.2. Phylogenomic Analysis

Genomic sequences of Pseudomonas capeferrum strains, including P. capeferrum WCS358 (type strain), were obtained from the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/, accessed on 3 March 2026), Refseq database, in June 2025. Moreover, online standard BLASTp analysis based on the P. capeferrum WCS358 DnaA protein (WP_033702442.1) led to the identification of other potential P. capeferrum strains, from which the Refseq genomes were obtained and analyzed.

Phylogenomic analysis, namely, Average Nucleotide Identity (ANI) and digital DNA–DNA hybridization (dDDH), were performed using OrthoANI [42]. Secondary metabolite cluster homologs were identified in the P. capeferrum genomes using BLASTn v 2.6.0 (standard parameters) in the Geneious software v.9.10 [43].

4.3. V. pyrina Culture and Spore Suspension Preparation

V. pyrina (CECT 2725) was cultured for 8 days on sterile cellophane squares placed on Potato Dextrose Agar supplemented with 0.3% (w/v) yeast extract (PDAYE) [44] under 22 °C/18 °C (day/night), 60% relative humidity, and a 16 h/8 h photoperiod. After incubation, the cellophane squares were stored at −20 °C with sterile silica until use. To prepare the spore suspension, frozen cellophanes were transferred to sterile 50 mL containers, and 4 mL of sterile distilled water was added. The containers were vigorously shaken for 1 min to dislodge the spores from the cellophane, and mycelial fragments were removed using a 40 µm cell strainer. The resulting spore suspension was standardized to 1 × 10^5^ spores/mL using a Neubauer counting chamber and used in subsequent experimental assays.

4.4. Antagonism In Vitro Assays

P. capeferrum NFX1 was cultured overnight at 23 °C in Tryptic Soy Broth (TSB). The bacterial culture was centrifuged (3 min, 7142 g, 4 °C), and the resulting pellet was resuspended in 0.03 M MgSO_4_ to an optical density (OD_600 nm_) of 0.5, corresponding to 6.00 × 10^8^ colony-forming unit (CFU)/mL). A commercial chemical fungicide, Cerimonia^®^ (Ascenza, Lisbon, Portugal), was included as a positive control at a concentration of 0.15 mL/L, based on the manufacturer’s recommendation. Cerimonia^®^ is composed of Difenoconazole (23.6%, 250 g/L) and is a registered commercial fungicide used against V. pyrina in Portugal. PDAYE plates were uniformly inoculated with a V. pyrina spore suspension (1 × 10^5^ spores/mL) using a sterile swab. After drying, each plate was divided into four quadrants, and a 5 µL drop of either the bacterial suspension or the positive control was placed at the center of each quadrant. Control plates received only the V. pyrina inoculum. Plates were incubated in a growth chamber at 22 °C/18 °C (day/night), 60% relative humidity, and a 16 h/8 h photoperiod for 7 days. After incubation, plates were scanned, and the number of fungal colonies within each quadrant were counted using the ImageJ software v.1.54g. Each treatment consisted of three replicate plates, with four bacterial drops per plate. The experiment was independently repeated three times.

The effect of strain NFX1 on the germination and development of V. pyrina spores was assessed via co-culture in diluted Potato Dextrose Broth supplemented with 0.3% yeast extract (PDBYE, 1:10 v/v in sterile distilled water) in a 96-well microplate. Strain NFX1 was cultured overnight in TSB at 23 °C with agitation at 180 rpm. On the day of the experiment, bacterial suspensions were adjusted to OD_600nm_ values of 0.05, 0.1, and 0.5 in PDBYE (1:10) solution (corresponding to 1.50 × 10^8^ CFU/mL, 2.00 × 10^8^ CFU/mL, and 6.00 × 10^8^ CFU/mL, respectively). To mirror the range of bacterial concentrations tested, the recommended dose of Cerimonia^®^ (0.15 mL/L) was used as the highest concentration and additionally tested at 1:5 (0.03 mL/L) and 1:10 (0.015 mL/L) dilutions in sterile distilled water. For the confrontation assays, 132 µL of each bacterial suspension or Cerimonia^®^ dilution was mixed with 66 µL of a V. pyrina spore suspension (1 × 10^5^ spores/mL). Negative controls consisted of 66 µL of V. pyrina spores combined with 132 µL of PDBYE (1:10) without bacterial cells. Three images per well were captured at predefined positions every 12 h for 4 days at 10× magnification using the Tecan Spark Cyto^®^ imaging system (Tecan group Ltd, Männedorf, Switzerland), and three independent assays were conducted. Due to spore overlap and bacterial proliferation under the selected assay conditions, reliable quantification of individual germinated spores or germ tube length measurements was not feasible; therefore, the inhibitory effect of strain NFX1 and Cerimonia^®^ on V. pyrina spore germination was evaluated using a categorical and ordinal three-level rating scale: 0–30% inhibition (score = 1), 30–70% inhibition (score = 2), and ≥70% inhibition (score = 3). Scores were assigned to each image relative to the not-treated V. pyrina control at the corresponding time point. One well per treatment with three individual measurements (three sections per well) were performed, and all individual scores from the three independent assays were used for statistical analysis. Representative images for each score category are provided in Figure S1.

4.5. Rocha Detached Pear Leaf Assay

One-year-old Rocha pear trees (P. communis L. cv. ‘Rocha’), grafted onto quince clone EMA, were obtained from a national nursery and maintained under greenhouse conditions in 20 L pots filled with Siro Horta (Siro) substrate from April 2024 to October 2024. For the detached leaf assay, young, fully expanded leaves were excised using a razor blade, retaining the petiole, and immediately transported on ice to the laboratory. Leaves were washed under running tap water, surface-sterilized with 70% ethanol, and rinsed in sterile distilled water within a laminar flow chamber. After drying on sterile paper towels, two leaves per treatment were submerged for 30 min in 50 mL of either an NFX1 suspension (OD_600nm_ = 0.5 in 0.03 M MgSO_4_) or a freshly prepared Cerimonia^®^ solution (0.15 mL/L in sterile distilled water). The not-treated infected control consisted of leaves inoculated with V. pyrina that were not subjected to MgSO_4_ immersion prior to infection. Treated and not-treated control leaves were placed onto square agar plates (0.7% agar, 120 × 120 mm) and air-dried in the laminar flow chamber. Infection was performed by applying two 50 µL drops of V. pyrina spore suspension (1 × 10^5^ spores/mL) per leaf. Not-infected controls (non-treated and NFX1-treated) were included to evaluate the compatibility of the bacterial treatment with leaf tissue (Figure S4). Plates were incubated in a growth chamber at 22 °C/18 °C (day/night), 60% relative humidity, and a 16 h/8 h photoperiod for 7 days. After incubation, leaves were scanned and images converted to 8-bit grayscale. Lesion severity was quantified by measuring the mean pixel intensity of infected areas using ImageJ software, with lower pixel values indicating darker lesions. To account for natural variation in leaf pigmentation, values were normalized to the mean pixel intensity of an adjacent healthy tissue (representative example images are provided in Figure S2). For each leaf, the mean value of the two infection sites was used. Given the narrow range and inverse relationship between pixel values and lesion severity, normalized values were inverted so that higher values reflect greater lesion severity and min–max rescaled using the global minimum and maximum across all independent experiments to obtain a 0–1 range. Four independent assays were conducted.

4.6. qPCR Assay for V. pyrina DNA Quantification in Plant Tissue

To assess the efficacy of strain NFX1 and Cerimonia^®^ in reducing V. pyrina infection, fungal DNA was quantified in infected detached pear leaves using quantitative PCR (qPCR). After 7 days of incubation, two leaf disks (1.548 cm^2^ each) were excised from lesion areas (or from equivalent positions in control leaves), pooled, flash-frozen in liquid nitrogen, and used for nucleic acid extraction. Eight biological replicates (leaves) per treatment were used for the qPCR assay. Total nucleic acids were extracted following Assunção et al. [45], with the addition of 40 µL proteinase K (20 mg/mL) to the extraction buffer. This protocol was originally optimized for woody plant tissues and was selected to minimize co-extraction of PCR inhibitors. To obtain total DNA, one microgram of nucleic acid was treated with 1 µL RNase A (1 mg/mL) at 37 °C for 1 h to remove RNA. Candidate primers targeting different V. pyrina genes (Table S1) were designed using Primer3 [46] and screened in silico via NCBI Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/, accessed on 3 March 2026). Specificity was assessed using the NCBI nucleotide collection (nt) restricted to P. communis (taxid:23211), with a maximum product size of 4000 bp. Primers were retained only if they showed no predicted amplification in pear or if off-target potentials exceeded 1000 bp. A primer pair targeting the V. pyrina elongation factor 1-alpha-like gene (ef-1α; NCBI accession MK888820.1) was ultimately selected as the only candidate with no predicted off-target amplification in P. communis (Table S1). Primer efficiency was evaluated using standard curves generated from serial dilutions of both pure V. pyrina total DNA (extracted independently using the NucleoSpin™ Plant II Mini Kit, Macherey-Nagel, Cat. No. 740770.50, Germany, according to the manufacturer’s instructions) and DNA from V. pyrina-infected pear tissue. Primer specificity was further experimentally assessed via melt curve analysis. The qPCR reactions were run on a QuantStudio 5 Real-Time PCR System (Applied Biosystems™, Carlsbad, California, USA ) using the SensiFAST SYBR^®^ Lo-ROX Kit (Bioline Reagents Ltd., London, UK) in 20 µL reactions containing 20 ng of template DNA, quantified using the Qubit™ dsDNA BR Assay (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Thermal cycling conditions were: 95 °C for 3 min, followed by 40 cycles of 95 °C for 5 s and 60 °C for 15 s. Melt curve analysis was performed from 60 °C to 95 °C at 0.15 °C/s. Quantification was based on a standard curve ranging from 40 pg to 5000 pg of pure V. pyrina genomic DNA. The resulting pg values are interpreted as relative estimates of fungal colonization in planta and are not directly equivalent to fungal biomass or CFU. Negative controls included no-template reactions and DNA from not-infected pear leaves.

4.7. Transcriptome Analysis of Pear Leaves in Response to P. capeferrum NFX1 and V. pyrina Infection

A pool of the total nucleic acid samples previously obtained was used for total RNA extraction. RNA was extracted by pooling two leaf disks from two biological replicates (i.e., four leaf disks) per sample, and three replicated samples per treatment were used for sequencing, followed by DNase treatment using the TURBO DNA-free™ Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA), according to the manufacturer’s instructions. RNA samples were quantified using the Qubit 4 Fluorometer (broad-range assay, Thermo Fisher Scientific, Waltham, Massachusetts, USA) and their integrity assessed by 2% (w/v) agarose gel electrophoresis under standard conditions. RNA was then sequenced using the SMARTer Ultra Low RNA Kit for Illumina NovaSeq (paired end) by Macrogen Inc. (Seoul, Korea), according to the manufacturer’s specified conditions. A total of 12 libraries were generated and represented the three biological replicates per condition tested: NFX1-treated and not-treated in both V. pyrina-infected and non-infected backgrounds. Transcriptomic analyses were performed using OmicsBox (v3.4.6) [47]. Reads were trimmed using Trimmomatic v.0.30 [38] (Q30, sliding window 4) and aligned to the P. communis coding sequences (CDS) from the NCBI RefSeq assembly GCF_963583255.1 (drPyrComm1.1). Gene expression was normalized using counts per million (CPM > 1 in all triplicates per treatment) and Trimmed Mean of M-values (TMM) with the edgeR [48]. Differential expression was assessed using a Generalized Linear Model (GLM) across three pairwise comparisons: (i) NFX1-treated vs. not-treated infected leaves; (ii) NFX1-treated vs. not-treated, not-infected leaves; and (iii) infected vs. not-infected untreated leaves. DEGs were considered under the following conditions: p values after false discovery rate (FDR) < 0.05 and |log_2_ fold change (logFC)| ≥ 1, ≤-1.

The functional annotation of DEGs was performed using GO terms related to Biological Processes using OmicsBox (v3.4.6) and Diamond BLAST. Gene set enrichment analysis (GSEA) was then applied to each DEG dataset, and GO terms with FDR ≤ 0.25 were retained for interpretation. GO term enrichment plots were generated in RStudio (v3.6) using the ggplot2 v.4.0.2 [49] and ggnewscale v.0.5.2 [50] packages, and heatmaps were created with the ComplexHeatmap package v.2.26.11 [51]. Sequencing data were deposited in the NCBI database under the BioProject accession number PRJNA1299138 (available at: https://www.ncbi.nlm.nih.gov/sra/PRJNA1299138, accessed on 3 March 2026).

4.8. Statistical Analysis

Colony counts from the dual plate assay and the positive rescaled lesion pixel values were normalized to the mean value of the untreated control within each independent experiment, ensuring that biological variability across experiments did not affect the analysis. All normalized data from all experiments were then pooled and used to calculate the inhibitory effect of the treatments on V. pyrina mycelial growth and lesion development using the following formula:

$$[eqn]$$

Inhibition values and qPCR quantification data for V. pyrina DNA (ef-1α target) were analyzed with the Kruskal–Wallis rank sum test using the agricolae R package v.1.3.7 [52] after evaluation of data normality via the Shapiro–Wilk test. To evaluate the relationship between fungal DNA quantity and lesion severity, Pearson correlation analysis was performed. To optimize lesion severity assessment, normalized, positive, and min-max rescaled values of lesion pixel intensity and lesion area from non-treated samples were compared between 1 and 2 weeks post-inoculation (WPI) using Welch’s Two-Sample t-test. Contingency tables of the spore germination scores pooled from the three independent assays were evaluated using Fisher’s exact test to assess significant associations between categorical variables (i.e., scores and treatments) and the Kruskal–Wallis test, followed by Dunn’s post hoc test for multiple comparisons among treatments. All statistical analyses were conducted in RStudio v.3.6 ). Graphical representations were produced using the ggplot2 package [49], with data presented as means ± standard error (SE), and different letters indicating statistically significant differences at p < 0.01.