RpoN1 (sigma factor 54) contributes to the virulence of Paracidovorax citrulli by regulating the expression of type IV pili PilA

Abstract

The σ⁵⁴ factor (RpoN), a significant transcriptional regulatory factor, plays crucial roles in regulating virulence, motility, biofilm formation, and the utilization of carbon and nitrogen sources in pathogenic bacteria. However, the function of RpoN has not been identified in Paracidovorax citrulli (formerly Acidovorax citrulli). To investigate this, we constructed a rpoN1 deletion mutant and a corresponding complement strain in the background of P. citrulli strain xjl12. The P. citrulli rpoN1 deletion mutant displayed attenuated virulence in melon. RNA-Seq analysis revealed that rpoN1 is involved in regulating the expression of certain pathogenicity-associated genes related to the secretion system, biofilm formation, and motility. Phenotypic analysis demonstrated that the rpoN1 deletion mutant of P. citrulli significantly attenuated biofilm formation, twitch motility, swarming motility, cotyledon colonization, and seed colonization. However, swimming motility was significantly enhanced in the rpoN1 mutant. As expected, qRT-PCR assays indicated that the type IV pili-related gene Aave_4679 (pilA) was barely expressed in the rpoN1 mutant, and western blot analysis revealed that RpoN1 positively regulated the expression of pilA. Additionally, bacterial one-hybrid assays and electrophoretic mobility shift assays indicated that RpoN1 directly binds to the promoter of pilA. Our investigation revealed that RpoN1 is essential for the virulence of P. citrulli and provides valuable insights into the physiology and pathogenic mechanisms of bacterial fruit blotch.

Background

Bacterial fruit blotch (BFB) is a seed-borne disease caused by Paracidovorax citrulli (formerly Acidovorax citrulli) (Schaad et al. 1978) and poses a significant threat to cucurbit production, primarily affecting watermelons and melons, on a global scale (Willems et al. 1992). In recent years, research on the pathogenic molecular mechanisms of P. citrulli has primarily focused on the Type III secretion system (T3SS) (Jiwénez-Guerrero et al. 2020; Zhang et al. 2020; Ji et al. 2022), Type VI secretion system (T6SS) (Tian et al. 2015b; Fei et al. 2022), polar flagella (Bahar et al. 2011), and quorum sensing (Wang et al. 2016). Additionally, the Type IV pilus (T4P) is known to contribute to the pathogenicity of P. citrulli (Bahar et al. 2009; Yang et al. 2023). T4P is widely present on the surface of bacteria and can be dynamically retracted. This dynamic activity is essential for various bacterial functions, including cell adhesion, biofilm formation, twitching motility, genetic material uptake, and virulence (Craig et al. 2004; Nudleman and Kaiser 2004).

RpoN, also known as σ⁵⁴, is a subunit of RNA polymerase that meticulously controls gene expression by recognizing specific promoter elements. RpoN plays a crucial role in regulating carbon and nitrogen metabolism and is involved in various bacterial functions, including flagellar synthesis, T4P, bacterial growth, motility, biofilm formation, T3SS, and virulence regulation (Hutcheson et al. 2001; Tian et al. 2015a; Li et al. 2020; Yu et al. 2020). However, the impact of RpoN on P. citrulli remains largely unknown.

To characterize the function of RpoN in P. citrulli, we identified Aave_0419 (rpoN1) and Aave_1899 (rpoN2) from the P. citrulli AAC00-1 genome (GenBank accession number NC_008752). We constructed deletion mutants of rpoN1 and rpoN2 and investigated the transcript levels of both using RNA-seq. Interestingly, we found that the deletion of rpoN1 severely downregulated the expression of pilA (Aave_4679), which was not found in ΔrpoN2 (Additional file 1). In this study, we focused on examining the regulatory mechanism of RpoN1 on the virulence of P. citrulli. Phenotypic experiments, RNA-seq, qRT-PCR, and western blot analyses revealed that rpoN1 positively regulated the transcription of pilA. Additionally, bacterial one-hybrid and electrophoretic mobility shift assays (EMSA) demonstrated that RpoN1 directly regulates the expression of pilA, influencing biofilm formation, twitching motility, swarming motility, and virulence in P. citrulli.

Results

The presence of two σ⁵⁴ factors in Paracidovorax citrulli

Two RpoN-encoding genes, Aave_0419 (rpoN1) and Aave_1899 (rpoN2), were identified in the genome of P. citrulli AAC00-1 through comparison of their amino acid sequences with those of other pathogenic bacteria. The full length of rpoN1 is 1590 bp, encoding 529 amino acids, while the full length of rpoN2 is 1407 bp, encoding 468 amino acids. A BLASTP sequence homology analysis was conducted using NCBI BLAST. Multiple sequence alignment revealed that the amino acid sequences of RpoN1 and RpoN2 were highly conserved in Escherichia coli, Pseudomonas ogarae, Xanthomonas oryzae pv. oryzae, and Pseudomonas syringae. Compared with the above bacteria, the amino acid sequence similarities of RpoN1 (RpoN2) were 41.54% (38.01%), 41.46% (36.99%), 36.86% (33.97%), and 40.21% (38.01%), respectively. (Additional file 2: Figure S1).

RNA-Seq analysis reveals RpoN1's role in regulating secretion systems, biofilm formation, and motility

In this study, we examined the transcriptional regulation of RpoN1 in P. citrulli using RNA-Seq. Compared to the wild-type strain, ΔrpoN1 mutant exhibited 140 differentially expressed genes (DEGs), with 61 genes upregulated and 79 genes downregulated (Additional file 2: Figure S2). These DEGs are associated with motility, the secretion system, and biofilm formation in P. citrulli (Additional file 3: Table S1). The RNA-Seq results were validated through qRT-PCR using eight selected DEGs (Fig. 1). A comprehensive list of all DEGs is available in Additional file 1.

Fig. 1

Expression levels of selected genes between wild-type xjl12 and ΔrpoN1. To verify the accuracy of the RNA-Seq data, eight genes related to motility, T6SS, T2SS, and chemotaxis were selected from the list of differentially expressed genes (DEGs). These genes are identified as Aave_4679, Aave_4400, Aave_1465, Aave_3783, Aave_2725, Aave_2722, Aave_3880, and Aave_0035. The expression levels of these genes in the mutant strain ΔrpoN1 and the wild-type strain xjl12 were monitored using qRT-PCR. The values represent the average of three independent experiments. Asterisks indicate significant differences between the samples (* P < 0.05, ** P < 0.01, and *** P < 0.001)

Among the DEGs listed in Additional file 3: Table S1, the expression of several genes associated with T6SS increased, including the gene Aave_1465, which encodes the type VI secretion system tube protein Hcp and negatively regulates biofilm formation in P. citrulli (Fei et al. 2022). Additionally, the expression of genes related to Type II secretion system (T2SS) and chemotaxis decreased in ΔrpoN1. Moreover, the expression levels of genes involved in the movement, such as pilA—a key gene for the formation of type IV pili (Yang et al. 2023)—and various genes related to flagellar assembly, also declined (Additional file 3: Table S1). Notably, among all downregulated genes, the expression of the type IV pili-related gene pilA was the most significantly reduced, with a log2-fold change value of −5.613 (P < 0.05) (Additional file 3: Table S1). These findings underscore the role of rpoN1 in regulating the transcription of pilA.

Gene Ontology (GO) analysis indicated that the DEGs primarily engaged in biological functions such as bacterial pilus and flagellar assembly, endopeptidase activity, and tryptophan catabolism (Additional file 2: Figure S3). KEGG enrichment analysis identified pathways significantly enriched with DEGs, including flagellar assembly, biofilm formation, bacterial chemotaxis, and the bacterial secretion system (Additional file 2: Figure S4). Given these observations, RpoN1 appears to regulate various pathogenic factors in P. citrulli, particularly the regulation of T4P.

Δ rpoN1 and Δ pilA reduce biofilm formation in Paracidovorax citrulli

Previous studies have demonstrated that biofilms contribute to the virulence of P. citrulli and that pilA is essential for biofilm formation in this species (Yang et al. 2023). In this study, we assessed the effects of rpoN1 and pilA on biofilm formation in LB medium. As illustrated in Fig. 2, compared to the wild-type strains, the biofilms of ΔrpoN1 and ΔpilA completely disappeared but were restored in the complemented strains ΔrpoN1-C and ΔpilA-C. These findings indicate that rpoN1 and pilA positively regulate the biofilm formation of P. citrulli.

Fig. 2

rpoN1 is essential for biofilm formation in P. citrulli. a Biofilm formation in wild-type (WT), ΔrpoN1, ΔpilA, and the complemented strains ΔrpoN1-C and ΔpilA-C. b Biofilm formation was visualized using crystal violet staining and quantified by measuring absorbance at 590 nm after ethanol suspension. The values represent the averages of three independent experiments, and *** indicates a highly significant difference between wild type and the tested samples (P < 0.001)

rpoN1 and pilA regulate the motility of Paracidovorax citrulli

Previous studies have confirmed that motility is crucial to the virulence of P. citrulli and that pilA is essential for its twitching motility (Bahar et al. 2009, 2011; Yang et al. 2023). To investigate the role of RpoN1 in regulating twitching motility in P. citrulli, we measured the transparent halos surrounding the ΔrpoN1 and ΔpilA colonies on NA plates. As shown in Fig. 3a, significantly reduced twitching motility was observed in the ΔrpoN1 and ΔpilA strains compared with the wild-type and complemented strains ΔrpoN1-C and ΔpilA-C. Moreover, transmission electron microscopy (TEM) revealed that neither ΔrpoN1 nor ΔpilA could produce pili (Fig. 3b). Furthermore, T4P can mediate the swarming motility of bacteria. In Pseudomonas aeruginosa, the pilA-deficient mutant strain exhibited enhanced swarming motility compared to the wild-type strain (Shrout et al. 2006). In contrast, the swarming motility of the ΔrpoN1 and ΔpilA strains was dramatically reduced compared with that of the wild-type and complemented strains ΔrpoN1-C and ΔpilA-C. Notably, these motility phenotypes of ΔrpoN1 and ΔpilA were highly similar (Fig. 3c; Additional file 2: Figure S5). These results indicate that rpoN1 and pilA positively regulate both twitching and swarming motility in P. citrulli.

Fig. 3

The role of rpoN1 in bacterial motility of P. citrulli. a Twitching motility of the wild type (WT), ΔrpoN1, ΔpilA, and complemented strains ΔrpoN1-C and ΔpilA-C of P. citrulli. The strains were spread on NA plates containing 1.0% agar. After 72 h, the colonies were photographed and observed using a stereoscope. b Transmission electron microscopy verified the presence of pili. Pili were observed in the wild type (WT) strain and the complemented strains ΔrpoN1-C and ΔpilA-C, while no pili were observed in the ΔrpoN1 and ΔpilA strains. Solid arrows represent type IV pili, and dashed arrows represent the polar flagellum. Bars = 500 nm. c Swarming motility of the WT, ΔrpoN1, ΔpilA, and complemented strains ΔrpoN1-C and ΔpilA-C of P. citrulli. The strains were added dropwise to 0.6% (w/v) semisolid medium plates and incubated at 28°C for 48 h. d Swimming motility of the WT, ΔrpoN1, and complemented strains ΔrpoN1-C of P. citrulli. The strains were added dropwise to 0.3% (w/v) semisolid medium plates at 28°C for 48 h. e The expression levels of the flagella-related genes flhA, flgM, fliC, fliE, fliH, fliI, fliJ, and fliK in the wild-type and ΔrpoN1 strains were measured by qRT-PCR. The values represent the average of three independent experiments. Asterisks indicate significant differences between the samples (* P < 0.05, ** P < 0.01, and *** P < 0.001)

Interestingly, the swimming motility of the ΔrpoN1 strain was greater than that of the wild-type strain (Fig. 3d; Additional file 2: Figure S6). These results suggest that rpoN1 negatively regulates the swimming motility of P. citrulli. However, qRT‒PCR results indicated that the expression levels of the flagella-related genes flhA, flgM, and fliC were significantly decreased in ΔrpoN1, whereas fliE, fliJ, and fliK were significantly increased (Fig. 3e). Therefore, the regulatory role of rpoN1 on flagella warrants further investigation.

rpoN1 and pilA contribute to Paracidovorax citrulli virulence

To further investigate the regulatory mechanism of rpoN1 on the virulence of P. citrulli in melon, we conducted cotyledon injection and spray inoculation trials using the wild-type strain, ΔrpoN1, ΔpilA, and their complemented strains ΔrpoN1-C and ΔpilA-C on melon seedlings. The results from the cotyledon injection indicated that by day 5, melon leaves inoculated with ΔrpoN1 and ΔpilA exhibited only slight disease symptoms, which were significantly less severe compared to the wild-type strain (Fig. 4a). The disease indices (DIs) for the cotyledons of melon seedlings inoculated with wild-type xjl12, ΔrpoN1, ΔpilA, ΔrpoN1-C, and ΔpilA-C were 0.893, 0.301, 0.329, 0.876, and 0.872, respectively (Fig. 4b). Notably, the DIs of ΔrpoN1 and ΔpilA were significantly lower than those of the wild-type strain, ΔrpoN1-C, and ΔpilA-C.

Fig. 4

The role of rpoN1 in the virulence of P. citrulli. a Cotyledon injection inoculation assay. The cotyledons of melon seedlings were inoculated with a P. citrulli suspension (~ 1.0 × 10³ CFU/mL). Double distilled water (ddH₂O) served as the control group (CK). Bacterial fruit blotch (BFB) symptoms of the seedlings were observed 5 days post-inoculation (dpi). b Disease index of cotyledon injection. c Spray inoculation assay. P. citrulli suspension (∼1 × 10⁸ CFU/mL) was sprayed onto the true leaves of melon plants, with ddH₂O as the CK. BFB symptoms were evaluated 7 days post-inoculation (dpi). d Disease index of plants subjected to true leaf spray inoculation. e Seed-to-seedling transmission assay. The effect of rpoN1 on the seed-to-seedling transmission of P. citrulli in melons. Muskmelon seeds (n = 20) were soaked in a (~ 1.0 × 10⁶ CFU/mL) bacterial suspension. BFB symptoms were observed 7 days after planting. f Seedling disease index of the seed-to-seedling transmission assay. The values in b, d, and f represent the average of three independent experiments. *** indicates a highly significant difference between wild type and the tested samples (P < 0.001)

After seven days of spray inoculation, the disease incidence of true leaves in plants inoculated with ΔrpoN1 and ΔpilA was significantly lower compared to those inoculated with the wild-type xjl12 (Fig. 4c). The DIs for plants inoculated with the wild-type, ΔrpoN1, ΔpilA, ΔrpoN1-C, and ΔpilA-C strains were 0.421, 0.144, 0.121, 0.395, and 0.428, respectively (Fig. 4d).

In the seed-to-seedling transmission assays, seedlings from seeds inoculated with ΔrpoN1 and ΔpilA exhibited weak symptoms and significantly lower mortality than those inoculated with the wild-type xjl12 and the complemented strains ΔrpoN1-C and ΔpilA-C (Fig. 4e). The DIs of seedlings inoculated with the wild-type, ΔrpoN1, ΔpilA, ΔrpoN1-C, and ΔpilA-C strains were 0.864, 0.256, 0.243, 0.838, and 0.831, respectively (Fig. 4f). These results indicate that rpoN1 and pilA positively regulate the virulence of P. citrulli.

Deletion of rpoN1 in Paracidovorax citrulli decreased bacterial growth in melon

In this study, the colonization ability of P. citrulli was evaluated through cotyledon and seed colonization. The colonization capacity of various strains was assessed by measuring the population of P. citrulli at 2, 24, 48, 72, and 96 h post-cotyledon inoculation. At 48 h post-inoculation, the populations of the ΔrpoN1 and wild-type strains in melon cotyledons were 5.754 × 10³ CFU/cm² and 2.75 × 10⁴ CFU/cm², respectively. At 96 h post-inoculation, the population numbers of the ΔrpoN1 and wild-type xjl12 strains were 5.047 × 10⁶ CFU/cm² and 2.884 × 10⁷ CFU/cm², respectively (Fig. 5a).

Fig. 5

Impact of rpoN1 on the colonization ability of P. citrulli in melon. a Assessment of P. citrulli colonization on melon cotyledons. A bacterial suspension (~ 1 × 10³ CFU/mL) for each strain was injected into melon seedling cotyledons. The bacterial population was quantified at 0, 24, 48, 72, and 96 h post-inoculation. b Assessment of P. citrulli colonization on melon seeds. A bacterial suspension (~ 1 × 10³ CFU/mL) for each strain was injected into melon seeds. The bacterial population was quantified at 0, 24, 48, 72, and 96 h post-inoculation

In assessing seed colonization ability, the population of P. citrulli in melon seeds was dynamically monitored over a period of 4 days. After 48 h of inoculation, the colony numbers of the ΔrpoN1 and wild-type xjl12 strains were 4.753 × 10⁴ CFU/g and 1.148 × 10⁷ CFU/g, respectively. At 96 h post-inoculation, the colony numbers were 2.328 × 10⁵ CFU/g and 1.936 × 10⁸ CFU/g for the ΔrpoN1 and wild-type xjl12 strains, respectively (Fig. 5b). These results indicate that the deletion of rpoN1 significantly weakens the colonization ability of P. citrulli in melon seeds.

rpoN1 directly regulates the transcription of pilA

The results of RNA-seq and phenotypic assays established a foundation for the transcriptional regulatory effect of rpoN1 on pilA. To further investigate the relationship between rpoN1 and pilA in P. citrulli, we conducted qRT-PCR, western blot, bacterial one-hybrid, and electrophoretic mobility shift assays. The qRT-PCR results demonstrated that pilA expression in the ΔrpoN1 strain was significantly down-regulated compared to the wild-type strain (Fig. 6a). Additionally, western blot analysis revealed that the deletion of rpoN1 led to decreased PilA expression in P. citrulli (Fig. 6b). Furthermore, the bacterial one-hybrid assay confirmed the interaction between RpoN1 and the promoter of pilA (Wang et al. 2018). As shown in Fig. 6c, the cotransformed strains pBXcmT-pilA and pTRG-RpoN1 successfully grew on the screening media, whereas the negative control strains did not. Moreover, we validated the binding between RpoN1 and the pilA promoter through EMSA. As the concentration of RpoN1 increased, the amount of bound RpoN1 also increased (Fig. 6d). These findings suggest that RpoN1 interacts with the pilA promoter and directly regulates pilA transcription in P. citrulli.

Fig. 6

RpoN1 directly regulates the expression of pilA. a The expression levels of pili-related genes pilA, pilB, pilM, pilR, pilT, and pilV in both the wild-type and ΔrpoN1 strains were measured via qRT‒PCR. The values represent the averages of three independent experiments. * denotes a significant difference between the two samples (P < 0.05). b The expression of PilA in the wild-type and ΔrpoN1 strains was determined by western blotting. c The interaction between RpoN1 and the pilA promoter was assessed using a bacterial one-hybrid system. RpoN1 was cloned into vector pTRG, and the pilA promoter region as cloned into vector pBXcmT. Plasmids co-transformed into the bacterial strain are indicated on the left. d The interaction between RpoN1 and the pilA promoter was examined using an electrophoretic mobility shift assay. Purified RpoN1 (0 ~ 1.2 μM) was incubated with 200 ng of DNA (containing the pilA promoter region) at 25°C for 30 min. The product was then resolved on a 5% (w/v) polyacrylamide gel in 0.5 × Tris–borate-EDTA (TBE) buffer at 90 V for approximately 2 h

Discussion

RpoN is a transcriptional regulator crucial for various biological functions in bacteria. In Pseudomonas aeruginosa, RpoN influences the expression of virulence factor, biofilm formation, and nitrogen source utilization (Hobbs et al. 1993; Cai et al. 2015). In Vibrio cholerae, RpoN controls the expression of genes related to motility and chemotaxis (Dong and Mekalanos 2012). However, its role in Paracidovorax citrulli (formerly Acidovorax citrulli) remains unclear. Our previous studies identified two genes encoding RpoN in P. citrulli, Aave_0419 (rpoN1) and Aave_1899 (rpoN2). In this study, we investigate the virulence regulatory mechanism of rpoN1 in P. citrulli.

We analyzed the transcriptional level of ΔrpoN1 using RNA-seq. The results revealed that rpoN1 regulates the motility, the secretion system, and biofilm formation in P. citrulli. Notably, pilA, a major gene of T4Ps, was significantly down-regulated. Previous studies have shown that pilA is essential for biofilm formation and the twitching motility of P. citrulli (Bahar et al. 2009; Yang et al. 2023). In this study, P. citrulli was unable to form biofilms after the deletion of rpoN1 or pilA, and its twitching motility was also weakened. Additionally, T4Ps are involved in mediating the swarming motility of pathogenic bacteria. Mutant strains of pilA in P. aeruginosa exhibited enhanced swarming motility compared with wild type strains (Shrout et al. 2006). In P. citrulli, the swarming motility of ΔrpoN1 and ΔpilA was significantly reduced compared with the wild-type strain. Furthermore, the pili of the ΔrpoN1 and ΔpilA strains were absent, as observed by TEM. These findings suggest that rpoN1 regulates biofilm formation and motility in P. citrulli, and is involved in regulating the expression of pilA.

To elucidate the roles of rpoN1 and pilA in the pathogenicity of P. citrulli in melon, we assessed the virulence of ΔrpoN1 and ΔpilA mutants through cotyledon injection, spray inoculation, and seed-to-seedling transmission assays. Our findings revealed that rpoN1 and pilA positively regulate both the virulence and seed-to-seedling transmission capabilities of P. citrulli. Furthermore, we observed a reduction in the colonization ability of the ΔrpoN1 mutant in melon. These results indicate that both rpoN1 and pilA contribute significantly to the virulence of P. citrulli.

To further study the relationship between RpoN1 and pilA in P. citrulli, we conducted western blot analysis. The results indicated that the expression of PilA was significantly decreased in the ΔrpoN1 mutant. These findings demonstrate that rpoN1 regulates pilA transcription in P. citrulli. To determine whether RpoN1 directly influences motility and biofilm formation, thereby affecting the virulence of P. citrulli, we examined the interaction between RpoN1 and pilA using a bacterial one-hybrid system and EMSA. Our analysis revealed that RpoN1 can directly bind to the promoter of pilA. These results suggest that RpoN1 directly regulates the transcription of pilA, which impacts biofilm formation, twitching motility, swarming motility, and the virulence of P. citrulli (Fig. 7).

Fig. 7

A simple model shows that RpoN1 regulates pilA in Paracidovorax citrulli. RpoN1 regulates the expression of pilA, influencing biofilm formation, twitch motility, swarm motility, and the virulence of Paracidovorax citrulli

Flagella, a crucial pathogenic factor, influence the swimming motility and virulence of P. citrulli (Bahar et al. 2011). In this study, the swimming motility of P. citrulli was enhanced in ΔrpoN1 mutants. These results suggested that rpoN1 negatively regulates the swimming motility of P. citrulli. However, qRT-PCR results showed that not all flagella-related genes were up-regulated. On the contrary, the expression levels of key flagella-related genes such as flhA, flgM, and fliC were significantly down-regulated. Additionally, in Helicobacter pylori, flagella can hijack pili proteins to control motility, and the deletion of pilO and pilN enhances migration ability in semisolid media (Liu et al. 2024). Based on these findings, the regulatory effect of rpoN1 on the flagella of P. citrulli requires further investigation. Moreover, the σ⁵⁴ factor relies on enhancer-binding proteins (EBPs) to regulate the transcription of various genes in bacteria (Gao et al. 2020). Therefore, exploring the regulatory network of the σ⁵⁴ factor necessitates understanding EBPs. This study confirmed that RpoN1 directly regulates the expression of pilA in P. citrulli; however, identifying which EBP mediates RpoN1's regulation of pilA transcription warrants further study.

Conclusions

In this study, we identified two σ⁵⁴ factors, RpoN1 and RpoN2, in P. citrulli. Our findings reveal that rpoN1 regulates biofilm formation, bacterial motility, virulence, and colonization ability. Furthermore, we confirmed that RpoN1 directly influences biofilm formation, twitching motility, and swarming motility in P. citrulli by interacting with the promoter of pilA. Therefore, we demonstrated that RpoN1 affects virulence in P. citrulli by regulating the expression of the type IV pili-related gene pilA. In future studies, we will further investigate the regulatory network of RpoN1 in P. citrulli.

Materials and methods

Bacterial strains, growth conditions, and plant material

Bacterial strains and plasmids used in this study are listed in Table 1. P. citrulli was cultured in Luria–Bertani (LB) medium at 28°C (Chong 2001). All Escherichia coli strains were grown in LB medium at 37°C. The turbidity of the cell suspensions was quantified using optical density measurements at a wavelength of 600 nm, as determined by a spectrophotometer. The concentrations of the antibiotics used were 100 µg/mL rifampicin (Rif), 50 µg/mL kanamycin sulfate (Km), 50 µg/mL gentamycin sulfate (Gm), 100 µg/mL ampicillin (Amp), and 8 µg/mL streptomycin (Sm). Melon (cv. Huanghou) seeds were cultured in an artificial climate incubator at 25°C with 70% relative humidity (RH). The inoculated seedlings were transferred to a greenhouse set at 28°C with 80% RH.

Table 1 Bacteria and plasmids used in this study

Construction of deletion mutants and complemented strains of Paracidovorax citrulli

Deletion mutations of rpoN1 and pilA were produced in P. citrulli through homologous recombination as previously described (Liu et al. 2019). Briefly, the upstream and downstream fragments of the target gene were amplified by PCR from P. citrulli AAC00-1 genomic DNA using specific primer pairs. The Km fragment was cloned from pET30a using Km primer pairs. A recombinant vector was constructed by ligating the three fragments into the suicide vector pEX18GM, which was then transferred into E. coli BW20676 for biparental mating with wild-type xjl12. Gene deletion mutants were obtained and validated by PCR using target gene-specific primers. The promoter positions of rpoN1 and pilA were predicted using the Gene Promoter Prediction website (http://www.softberry.com/). The specific primer pair comp-F/R was employed to amplify the sequences containing the target gene and promoter. The fragment was ligated to pBBR1MCS-5 to construct a recombinant vector, which was then transferred into E. coli BW20676 for biparental mating with the P. citrulli mutants ΔrpoN1 and ΔpilA. These complemented strains were further confirmed via PCR analysis with the relevant F/R primers. All sequences of primers used in this study are listed in Additional file 3: Table S2.

Transcriptome sequencing and data analysis

To elucidate the regulatory mechanism of rpoN1 in P. citrulli, RNA-Seq analysis was conducted by Shanghai Personalbio Technology Co., Ltd. (Shanghai, China). In summary, total RNA was extracted using Trizol reagent (Invitrogen Life Technologies). The RNA quality and integrity were assessed using a nanodroplet spectrophotometer (Thermo Scientific) and a Bioanalyzer 2100 system (Agilent). rRNA was removed from the total RNA with the Zymo-Seq Ribo Free Total RNA Library Kit. The AMPure XP system (Beckman Coulter, Beverly, CA, USA) was employed to purify the library fragments, selecting DNA fragments between 400 and 500 bp. Illumina PCR primer cocktails were utilized to selectively enrich DNA fragments bearing linker molecules at both ends during 15 cycles of PCR. Following purification, the product was quantified on a Bioanalyzer 2100 system (Agilent) using Agilent's high-sensitivity DNA assay. Subsequently, the sequencing library was sequenced on the Nova Seq 6000 platform. Differentially expressed mRNAs were analyzed using DESeq (v1.38.3). Transcripts with | log₂FoldChange |> 1 and P-value < 0.05 were considered differentially expressed mRNAs. GO enrichment analysis of differentially expressed genes was executed using topGO, with the P value calculated using the hypergeometric distribution method (significance threshold: P-value < 0.05), to identify GO terms with significant enrichment and determine the primary biological functions associated with these genes. KEGG pathway enrichment analysis of differentially expressed genes was performed with Cluster Profiler (v4.6.0), focusing on significantly enriched pathways with P-values < 0.05. Each strain was analyzed in three biological repetitions.

Quantitative real-time PCR analysis

In this study, total RNA from P. citrulli was quickly extracted using Total RNA Extractor (Trizol) reagent (Shengon Biotech, Shanghai, China). The quantitative real-time (qRT)-PCR system was set up with a ChamQ Universal SYBR kit (Vazyme, Nanjing, China). Reverse transcription was performed using the Hiscript III RT SuperMix for qPCR (Vazyme, Nanjing, China). The qRT‒PCR was carried out on an ABI PRISM 7500 real-time PCR instrument (Applied Biosystems). The reaction program was as follows: 95°C for 30 s (1 cycle), 95°C for 10 s, and 60°C for 30 s (40 cycles). In this study, the 16S ribosomal RNA gene was employed as the internal reference gene, and the data were analyzed using the 2^−∆∆Ct method. The experiments were performed three times with three biological replicates per gene. The primers utilized for the selected genes in this assay are detailed in Additional file 3: Table S2.

Biofilm formation assay

The biofilm formation assay was conducted following a previously described method (Wang et al. 2022). All strains were cultured overnight in LB broth at 28°C and adjusted to an OD₆₀₀ of 1.0 after two washes with sterilized water. Then, 40 µL of the bacterial suspension was added to 4 mL of LB broth in a polystyrene 12-well plate, which was maintained at 28°C. After 48 h, the bacterial suspensions were gently aspirated, washed three times with sterile water, and incubated at 80°C for 20 min. The biofilms were stained with a 1% crystal violet solution for 50 min and subsequently dissolved in anhydrous ethanol. The OD₅₉₀ values were measured using an enzyme marker. This experiment was repeated three times.

Bacterial motility assay

Twitching motility was measured for each P. citrulli strain using a previously described method (Wang et al. 2022). Briefly, the samples were washed twice with sterile water, adjusted to an OD₆₀₀ of 1.0, diluted to 1.0 × 10⁵ CFU/mL, and then spread evenly on 1.0% (w/v) NA solid medium plates. After incubation at 28°C for 72 h, the morphological characteristics of the colonies were observed using a stereo fluorescence microscope (Nikon). These experiments were repeated three times.

The swarming motility assay was performed as described previously (Liu et al. 2016). Briefly, for the swarming assay, all strains were cultured overnight in LB broth at 28°C and adjusted to an OD₆₀₀ of 0.3. Then, 3 µL of the bacterial suspension was added dropwise to 0.6% (w/v) semisolid medium plates. The diameter of the swarming halos was measured on agar plates after 48 h of incubation at 28°C. For the swimming motility assay, the medium was changed to 0.3% (w/v) semisolid medium. These experiments were conducted three times.

Transmission electron microscopy

Transmission electron microscopy (TEM) was utilized to observe the pili of cultured bacteria. The TEM specimens were prepared as previously described, with slight modifications. The strains for observation were incubated on 1.0% (w/v) NA plates at 28°C for 12 h. Appropriate amounts of sterile water were added to the plates and gently shaken to suspend the colonies. The copper mesh was immersed in the bacterial suspension, stained with 2% phosphotungstic acid for 5 min, and then baked for 30 min. The bacterial pili were observed using a Hitachi-7650 transmission electron microscope at 80 kilovolts (kV).

Virulence assays

Three previous approaches have been employed to explore the effects of rpoN1 and rpoN2 on the virulence of P. citrulli (Liu et al. 2019).

Cotyledon inoculation assay: A gradient dilution of the cultured strains was prepared to achieve a concentration of approximately 1.0 × 10³ CFU/mL. The strains were then injected into the cotyledons of well-developed melons over a 7-day growth period. The inoculated melon seedlings were incubated at 28°C with 80% RH. Symptoms of BFB were monitored at 24-h intervals post-inoculation.

True leaf spray inoculation: Melon seedlings bearing their third true leaf were chosen for spray inoculation in this study. Each cultured strain was washed twice with sterile water, and the OD₆₀₀ was adjusted to 0.3. At least 50 mL of the bacterial suspension was transferred to sterile spray bottles and sprayed uniformly on both the front and back surfaces of the melon leaves. The inoculated seedlings were incubated at 28°C with 100% RH for 48 h, followed by incubation at 80% RH. BFB symptoms were observed seven days post-inoculation.

Seed-to-seedling transmission assay: Twenty dewy melon seeds were placed in 5 mL sterile centrifuge tubes, and 2 mL of bacterial suspensions with a concentration of approximately 1.0 × 10⁶ CFU/mL were added. The mixture was then gently shaken for 4 h. After air-drying the bacterial liquids, the seeds were planted in a greenhouse maintained at 28°C with 80% RH. Symptoms of BFB were observed seven days post-planting. These experiments were repeated three times.

Bacterial colonization assay of melon cotyledons and seed

The seedling colonization of the wild-type strain of P. citrulli and its derived mutant strains was determined by infiltrating melon cotyledons and seeds. Bacterial cells at a concentration of 1.0 × 10³ CFU/mL were injected into the cotyledons of melon plants (cv. Huanghou) using a sterile syringe. The inoculated melon seedlings were incubated in a growth chamber at 100% RH at 28°C for 0, 24, 48, 72, and 96 h. The inoculated melon cotyledons were then extracted into sterile centrifuge tubes with forceps, ground with 100 µL of sterile water, diluted in a gradient, and then evenly spread on LA plates containing the corresponding antibiotics. Colonies were counted after 24 to 96 h of incubation at 28°C. Seed colonization assays were conducted based on previous reports (Tian et al. 2015b). Melon seeds were sterilized using 70% ethanol, and 5 µL of the bacterial suspension at approximately 1 × 10³ CFU/mL was injected into the seed openings. The seeds were subsequently placed in petri dishes lined with moistened sterile filter paper and incubated at 28°C. After 2, 24, 48, 72, and 96 h of incubation, the seeds were removed, placed in a sterile centrifuge tube, and 1 mL of sterile water was added. The mixture was vortexed and shaken for 10 min, then uniformly spread on LA plates containing the corresponding antibiotics after gradient dilution. Colonies were counted after incubation at 28°C for 24 to 96 h. These experiments were repeated three times.

Protein expression and purification

The DNA fragment of rpoN1 was amplified by PCR using specific primers listed in Additional file 3: Table S2 and inserted into the pET30a plasmid to construct the recombinant vector pET30a-RpoN1-His. This vector was subsequently introduced into E. coli BL21 for protein expression. For protein purification, the bacteria were cultured in LB medium at 37°C and 220 rpm until the OD₆₀₀ reached approximately 0.4. Isopropyl β-D-thiogalactoside was then added to a final concentration of 0.4 mM to induce expression for 12 h at 16°C and 220 rpm. The bacterial cells were harvested by centrifugation at 6000 × g for 10 min at 4°C and resuspended in 10 mL of 20 mM Tris–HCl (pH 7.4). The cells were lysed using TieChui E. coli Lysis Buffer (ACE Biotechnology, Changzhou, China), and the lysates were incubated with pre-equilibrated Ni²⁺ at 4°C for 1 h. Proteins containing His tags were extensively washed with a buffer containing 20 mM Tris–HCl (pH 7.4) and 50 mM imidazole, and subsequently eluted with buffers containing 100 mM, 200 mM, and 300 mM imidazole, respectively. The purified protein was mixed with glycerol to a final concentration of 20% and stored at −80°C.

Western blot analysis

To clarify the expression of PilA in the absence of rpoN1, the plasmid pBBR-MCS5 carrying a Flag tag was ligated to the pilA fragment, which includes its native promoter, to construct recombinant vectors. These vectors were subsequently transfected into P. citrulli wild-type and rpoN1 mutants. The overnight culture was adjusted to an OD₆₀₀ of 1.0, and the cell precipitates were collected at 4°C and 6000 × g for endocrine protein detection. The cell sediments were resuspended in 900 µL of 20 mM Tris–HCl buffer solution, and the cells were lysed with 100 µL of TieChui E. coli Lysis Buffer (ACE Biotechnology, Changzhou, China) for 5 min at 4°C. Protein supernatants were separated from the precipitates at 4°C and 8000 × g, then heated at 100°C for 10 min to denature the proteins. Next, the protein supernatants were separated by sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) and transferred to a polyvinylidene difluoride membrane (Millipore, Red Bank, NJ, United States) using a semidry blot machine (Bio-RAD, CA, United States). After blocking with 5% milk containing 0.05% Tween in Tris buffer solution (TBST, pH 7.5) for 1 h at room temperature, the membrane was incubated with a monoclonal antibody specific for Flag tags (1:5000; Abmart, Shanghai, China) for 1 h. Detection was carried out using an HRP-conjugated anti-rabbit secondary antibody (No. M21002, Abmart, Shanghai, China). Immunoblots were developed with a HyGlo HRP ECL Detection Kit (MDBio Inc., Qingdao, China) and visualized using an automatic multifunction image analysis system Tanon-6,600 (Tanon, Shanghai, China). These experiments were conducted three times.

Bacterial one-hybrid assay

A bacterial one-hybrid reporter system comprising two plasmids, pTRG and pBXcmT, and the E. coli -Blue MRF' kan strain was employed to detect protein interactions between the transcriptional regulator RpoN1 and the promoter of pilA (Wang et al. 2018). Briefly, the fragment containing the pilA promoter was cloned into pBXcmT to construct the recombinant vector pBXcmT-pilA. Similarly, the fragment encoding RpoN1 was cloned into pTRG to construct the recombinant vector pTRG-RpoN1. Both recombinant vectors were then cotransfected into the E. coli XL1-Blue MRF' Kan strain. If direct physical binding occurs between RpoN1 and the pilA promoter, the transformed E. coli strains containing both pBXcmT-pilA and pTRG- RpoN1 will grow well on a selective medium. This medium is a minimal medium containing 5 mm 3-amino-1,2,4-triazole, streptomycin at 8 µg/mL, tetracycline at 12.5 µg/mL, chloramphenicol at 34 µg/mL, and Km at 30 µg/mL (Wang et al. 2018). Furthermore, cotransformants containing pBX-R2031 and pTRG-R3133 served as positive controls (Xu et al. 2016), while cotransformants containing empty pTRG and pBXcmT-pilA were used as negative controls. All cotransformants were spotted onto the selective medium and grown at 28°C for 3–4 days, and then photographed. These experiments were conducted three times.

Electrophoretic mobility shift assay

An electrophoretic mobility shift assay was conducted to determine whether RpoN1 binds to the promoter of pilA. The DNA fragment of the pilA promoter was amplified by PCR using specific primers listed in Additional file 3: Table S2. DNA binding was carried out in a 20 µL reaction system containing 4 µL of EMSA Tris binding buffer, 1 µM His₆-RpoN1, and 200 ng of the DNA fragment. The binding reaction was performed at 25°C for 30 min. Subsequently, the samples were loaded onto a native 5% (w/v) polyacrylamide gel and electrophoresed in 0.5 × Tris–borate-EDTA buffer at 90 V for approximately 2 h. The gel was then removed and soaked in developing solution for 5 min and visualized using a UV imager. These experiments were conducted three times.

Statistical analysis

All analyses were performed using SPSS 22.0 (SPSS Inc.). Analysis of variance (ANOVA) was used to determine the differences in biofilm assay, motility assay, disease index, and gene expression between treatments.