Characterization of a unique phosphopantetheinyl transferase (PPTase) in the fish pathogen Pseudomonas plecoglossicida

TL;DR

This study investigates a unique enzyme in a fish pathogen, revealing its role in bacterial survival and virulence.

Contribution

The paper is the first to characterize the function and virulence significance of PPTase in Pseudomonas plecoglossicida.

Findings

The PPTase mutant showed reduced growth, increased sensitivity to stress, and lower virulence in host fish.

RNA-seq analysis revealed 179 differentially expressed genes linked to iron transport, stress response, and secretion systems.

The PPTase enzyme is essential for fatty acid synthesis and pathogenicity in P. plecoglossicida.

Abstract

Phosphopantetheinyl transferase (PPTase) is a key enzyme that catalyzes 4’-phosphopantetheinylation at conserved serine residues of fatty acid synthases, polyketide synthases, and nonribosomal peptide synthetases. It is involved in the biosynthesis of primary and secondary metabolites and has been well characterized as an important virulence factor in many pathogenic microorganisms. A unique Sfp-type PPTase was annotated in Pseudomonas plecoglossicida NB2011, the causative agent of visceral granulomas syndrome in the large yellow croaker (Larimichthys crocea). However, the biological function and virulence significance of this enzyme in the pathogen have not yet been studied. In this paper, we aligned the target sequence with other Sfp-type PPTases, constructed a deletion mutant using the method of double homologous recombination, investigated the biological phenotypes, and compared the intracellular survival and virulence of the mutant and wild-type strains. Additionally, we performed RNA-seq and analyzed the transcription data. The results indicated that the PPTase of P. plecoglossicida shares the highest sequence similarity with PapcpS of P. aeruginosa. A conditional mutant was successfully constructed when pET22b-entD was present. Compared to the wild-type strain, the mutant was lysine indispensable, exhibited poorer growth during the late log phase, was more sensitive to H2O2 exposure, displayed lower siderophore activity, and exhibited a significant reduction in total fatty acid synthesis. Furthermore, fewer bacterial cells of the mutant persisted in mouse macrophage J774A.1, and its toxicity to the host fish was substantially attenuated. RNA-seq analysis revealed 179 differentially expressed genes (DEGs), including 105 up-regulated and 74 down-regulated genes. Genes involved in iron transportation, oxidative stress response, and ABC transporters were down-regulated, while genes in the type III and type VI secretion systems were predominantly up-regulated. The transcriptomic results are consistent with the phenotypic observations, suggesting that the PPTase is essential for bacterial survival and plays important roles in the pathogenicity process. This is the first paper on the PPTase from P. plecoglossicida.

Figures

Sequence alignment of Sfp-type PPTases.The protein sequences used for alignment are as follows: Pseudomonas plecoglossicida NB2011 PPTase, Pseudomonas aeruginosa Pcps (WP_003112468.1), Salmonella enterica typhimurium EntD (AAA91937.1), Bacillus substilis Sfp (ACG68433), and Bacillus cereus Sfp (WPQ43272.1). Conserved amino acids are shaded in black; gray shading indicates that 100% of the residues are similar at that position. The symbol “*” indicates a number 10 less than the following number.

Growth curves of mutant and wild type strain.Measure the strain concentration within 14h under the condition of OD600 and draw the growth curve diagram.

Swarming and swimming ability of the mutant and wild type strain.A, Swarming motility of the mutant and WT; B, Swimming ability of the mutant and WT. Data was mean±SD of three independent experiments.

Lysine autotrophy observation, sensitivity to H2O2 and siderophore activity of the mutant and WT strain.A, lysine autotrophy observation, after 12 hours, the OD600 of the bacterial culture was measured; B, sensitivity to H2O2 treatment; C, siderophore activity of the mutant and WT strain, after 24 hours cultivation, the culture supernatants were collected and siderophore activity measured; D, Csaky test for hydroxamte siderophore of the mutant and WT strain, the 24h hours culture supernatant was detected by Csaky test for hydroxamate type of siderophores. A, B, C, Data was mean±SD of three independent experiments. Asterisk “*” and “” denotes a significant difference at the level of P < 0.05 and P < 0.01, respectively. D, Representative data was shown from three independent replicate tests.

Adhesion, internalization and replication of the mutant and WT strain in the macrophage J774A.1.CFU level of the strains at −3 h (adhesion), 0 h (internalization) and 0-24 h (replication), respectively. The data is presented as mean ± standard deviation (SD) from 3 replicates for each sample.

Full text

Introduction

In recent decades, Pseudomonas plecoglossicida has emerged as a significant pathogen affecting the large yellow croaker (Larimichthys crocea), responsible for visceral granulomas syndrome that persistently threatens aquaculture populations. Extensive research has elucidated key pathogenic mechanisms and virulence factors of this bacterium, including its ability to proliferate intracellularly within macrophages, deploy toxic secretion systems (T6SS1 and T3SS), and upregulate siderophore production under low-temperature conditions [1–4]. Of particular interest is phosphopantetheinyl transferase (PPTase), a critical enzyme that catalyzes the 4’-phosphopantetheinylation of conserved serine residues in fatty acid synthases, polyketide synthases (PKS), and non-ribosomal peptide synthetases (NRPS). This post-translational modification is essential for the biosynthesis of both primary and secondary metabolites. PPTase has been established as a crucial virulence factor in various pathogenic microorganisms, including EntD in Escherichia coli [5], PapcpS in Pseudomonas aeruginosa [6], PptT in Mycobacterium tuberculosis [7], and sfp-type Ppt1 in Fusarium fujikuroi [8]. Notably, the genome of P. plecoglossicida NB2011 [1] encodes a predicted protein exhibiting significant amino acid sequence homology with EntD, an Sfp-type PPTase involved in enterobactin-mediated iron acquisition in E. coli. Repeated unsuccessful attempts to knock out this gene suggest its essential role in bacterial viability. Despite these findings, the specific contributions of PPTase to the pathogenicity of P. plecoglossicida remain poorly characterized, highlighting a critical knowledge gap in our understanding of its virulent mechanisms.

In this study, we generated a conditional deletion mutant of P. plecoglossicida NB2011 by targeting the open reading frame of the PPTase gene. Following the introduction of the recombinant plasmid pET22b-entD, we systematically characterized the mutant strain by analyzing its biological phenotypes, evaluating cytotoxicity in in vitro cultured cell models, assessing virulence in host fish infection assays, and analyzing the transcription data by RNA-seq. These investigations aim to elucidate the functional role of PPTase in bacterial pathogenesis, thereby advancing our understanding of the virulence mechanisms of this pathogen.

Materials and methods

Bacterial strains and plasmids

P. plecoglossicida NB2011 was isolated and identified in our previous study [1]. E. coli DH5αand E.coli WM3064 was used for plasmid transformation. The entD containing plasmid pETt22b-entD was constructed by Sangon (Sangon Biotech, Co, Ltd, Shanghai, China). The suicide plasmid pK18mobsacB-ERY was kindly offered by Dr. Xiaoxue Wang (South China Sea Institute of Oceanology, Guangzhou, China).

Sequence alignment

The amino acid sequence identity of the PPTase homologs was aligned by BlastP [9,10]. The NCBI protein database was used as the source of sequences. Homologous sequences were aligned with Clustal X2 [11].

Generation of conditional deletion mutant

Double homogenous recombination was used to construct a conditional deletion mutant of the target gene, with the recombination system of plasmid pK18mobsacB-ERY which encodes a counter selectable marker sacB [12]. At first, the recombinant plasmid pET22b-entD was constructed containing the target coding sequence of entD of E. coli (Gene ID: 945194); and the plasmid was then transferred to P. plecoglossicida NB2011 resulted in the recombinant strain P. plecoglossicida/pET22b-entD, used as the parent strain for deletion of the target gene. The primers used were listed in Table 1. Briefly, the PPTase upstream and downstream flanking regions (780 bp and 742 bp, respectively) were amplified by PCR with primers pairs MF1/MR1, MF2/MR2. The upstream and downstream fragments were then linked by fusion PCR with the primers MF1/MR2, the resulting fragment was then ligated to the plasmid pK18mobsacB-ERY and transferred to E. coli WM3064. The positive transferred cells were selected on the LB (Luria Broth) agar with erythromycin (25 μg/mL) and kanamycin (25 μg/mL). Then the recombinant E. coli was conjugated with P. plecoglossicida NB2011 on the counter selecting medium with 20% sucrose. Deletion of targeted sequence was confirmed by PCR and DNA sequencing.

Table 1: Primers used in this study.

Analyses of bacterial growth of the mutant

To observe growth rate of the wild type (WT) and mutant strain, bacteria were grown at 28℃ in 5 mL LB with shaking (180 rpm), the optical density at 600 nm were determined every 1 h over a period of 14 h.

Swarming and swimming ability assay

Swarming and swimming ability assay were performed as previously described [13]. In brief, the strains were incubated in LB agar at 28℃ overnight, then single colonies were inoculated in LB broth, and the overnight culture were normalized to OD600 = 1; for swarming motility observation, 10 μL of the bacterial suspension were spotted on the center of swarming LB plate (with the agar concentration of 0.7%) respectively. After 24h, 48h and 72h incubation, the migrated distances %of the colonies were measured. The means of three independent measures were calculated and accessed. For the swimming assay, 2 μL of bacterial cells grown overnight were inoculated onto LB agar (with the agar concentration of 0.3%).

Analysis of the sensitivity to H2O2 treatment in vitro

H_2_O_2_ sensitivity of the strains were detected as Li et al. [13]. The WT strain and mutant were incubated in 100mL LB broth at 28℃ for 24h to the stationary phase (OD_600_ = 1.0). Bacterial cells were collected by centrifugation (6000 × g, 10 min), washed twice with 0.2M phosphate buffer (pH7.0), then suspended and diluted with the same buffer. Each culture with an initial population of 10^7^cfu/mL was treated with 20mM H_2_O_2_. After 0 min, 10 min, 20 min, 30 min and 40 min of treatments, 200 μL treated cultures from each sample were plated on fresh LB agar to quantify survival bacterial population. Viable counts were observed after 48h of incubation at 28℃.

Lysine dependance observation

Lysine auxotrophy was investigated as previously described [14]. The wild type and mutant strain were inoculated and collected as the above described, the bacterial concentration was adjusted with sterile PBS to OD_600_ = 0.6. Then the bacterial suspension was transferred to the fresh prepared M9 minimum medium (with or without 50μM lysine). After 12 hours, the OD_600_ of the bacterial culture was measured, and the mean value of three measurements was taken.

Determination of siderophore activity in the supernatant

P. plecoglossicida strains were grown in the iron-restricted succinate medium [15] for 7 h. The culture supernatants were collected by centrifuge at 4℃, 6000 × g, 10 min. Siderophore activity was measured according to the methods of Schwyn and Neilands [16] with some modifications. Briefly, the supernatants were mixed with an equal volume of chrome azurol sulphonate (CAS) solution containing 0.6mM hexadecyltrimethylammonium, 1.5 × 10 − 2mM FeCl_3_·6H_2_O, 0.15mM CAS, 50mM anhydrous piperazine, and 0.75M HCl. After the mixture was incubated for 60 min at room temperature in the dark, an increased absorption in orange color was then measured using a spectrophotometer at 630 nm. The relative quantity of siderophore activity (%) was calculated by (Acontrol−Asample)/Acontrol×100%. Further tests were performed to determine the specific type of siderophores. The Csaky assay [17] for hydroxamate type of siderophores was prepared as follows: the siderophore solution or supernatant (1mL) is hydrolyzed with 1 ml of 6N H_2_SO_4_ in a boiling water bath for 6h. The solution is then buffered by adding 3mL sodium acetate solution. Add 1mL sulfanilic acid solution followed by 0.5mL iodine solution. After 5 min, excess iodine is destroyed with 1mL sodium arsenite solution. Add 1mL α-naphthylamine solution. Add water to 10mL, allow color to develop for 20–30 min, and measure the absorbance at 400–600nm; the hydroxamates appear a maximum absorbance at 526nm. catechol type of siderophores was detected by Arnow’s test [17]. Pyoverdine was measured by UV-Visible Spectrofluorimetry as described by Hoegy et al. [15], with the wavelength range from 380–600 nm. If pyoverdine is present in the sample, a specific band with a maximum of absorption at 380nm should appear. Experiments were performed in triplicate.

Cellular fatty acid analysis

For cellular fatty acid analysis, the wild type and mutant strain were grown aerobically in modified Marine Agar (BD) (CH_3_COONa, 1.0g/L, trisodium citrate, 0.5g/L, pH7.5) at 28℃ for 72h. Bacterial cells were extracted, then fatty acids were saponified, methylated and extracted using the standard protocol of the midi system (Sherlock Microbial Identification System, version 6.2). Fatty acids were analyzed by gas chromatography using a 6850 instrument (Agilent Technologies) and identified using TSBA6.0. Experiments were performed in triplicate.

Macrophage J774A.1 infection observation

J774A.1 cells were infected as described by Jin et al. [18]. Briefly, P. plecoglossicida NB2011 was grown overnight in LB at 28°C with shaking and then diluted into sterilized PBS with a concentration of 10^8^ cells/ml. The cells (2.5 × 10^5^ cells/ml) were seeded and grown overnight in a 24-well plate. Harvested bacterial suspensions were added to the cells at a multiplicity of infection (MOI) of 10:1. Plates were then centrifuged at 600 × g for 10 min. After two hours of incubation, the cells were washed three times with PBS and then incubated in growth medium containing 200 μg/ml gentamicin for 1 h to kill the extracellular bacteria. Infected cells were washed three times with PBS and lysed at 2 and 24 h post-infection with 1% (vol/vol) Triton X-100, and serial dilutions of the lysates were plated onto LB agar and incubated at 28°C for 24 hr. For enumeration of viable extracellular bacteria, cell supernatants were collected, and viable bacteria (CFU) were counted.

Virulence of the mutants to the large yellow croaker

To test the virulence of the mutant and WT, the strains were cultured in LB broth at 28°C overnight, and bacteria were collected with centrifuge at 4°C (6000 × g, 10 min). The bacterial suspension was diluted to 5.0 × 10^5^ cells mL^-1^. After 7 days acclimation period, the croakers with a body weight of 50 ± 5g were randomly divided into 20 fish per group. Tested groups were intraperitoneally (i.p.) injected with 0.2 ml bacterial suspension; control fish were injected with 0.2 ml of sterile saline. Signs of visceral granulomas and death of the infected fish were recorded during the following 7 days.

RNA-seq, transcription analysis and data validation by qRT-PCR

P. plecoglossicida NB2011 (WT) and mutant strain were used for RNA-Seq analysis. Total RNA was extracted from the tissue using TRIzol® Reagent according to the manufacturer’s instructions (Invitrogen) and genomic DNA was removed using DNase I (TaKara). Then RNA quality was determined by 2100 Bioanalyser (Agilent) and quantified using the ND-2000 (NanoDrop Technologies). Only high-quality RNA sample (OD260/280 ≥ 1.8, OD260/230 ≥ 1.0, RIN ≥ 6.5, 28S:18S≥1.0, ≥ 50ng/μl, ≥ 1 μg) was used to construct sequencing library. RNA-seq transcriptome library was prepared following TruSeq^TM^ RNA sample preparation Kit from Illumina (San Diego, CA) using 2 μg of total RNA. High-throughput sequencing was performed with the Illumina Novaseq Platform. The processing of original images to sequences, base-calling, and quality value calculations were performed using the Illumina GA Pipeline (version 1.6), in which 150 bp paired-end reads were obtained. The data generated from Illumina platform were used for bioinformatics analysis. All the analyses were performed using the free online platform of Majorbio Cloud Platform (www.majorbio.com) from Shanghai Majorbio Bio-pharm Technology Co., Ltd. Reference to the whole genome sequencing information of the P. plecoglossicida NB2011 strain for gene alignment analysis, gene expression level analysis and differential expression analysis. Genes were significantly up-/down-regulated when the absolute value of fold change (relative to the control) was greater than 1.5, with a false discovery rate (FDR) less than 0.05. The raw data for analysis were deposited in SRA (Sequence Read Archive, http://www.ncbi.nlm.nih.gov/sra) with the accession number PRJNA1251296.

Transcription of six DEGs in the transcription analysis of the WT and mutant strain, including exoU (L321_RS18235), tssA (L321_RS18140), tssE (L321_RS18120), katE (L321_RS16920), PPTase (L211_RS03080) and fliS (L321_RS14690) were selected to be assayed for validation of RNA-seq results. Three flasks of bacterial culture were sampled for each strain, and the cells collected, mRNA was extracted with Trizol (Invitrogen, San Diego, USA), and the cDNA was synthesized for qRT-PCR. 16S rRNA encoding gene was designed as the internal reference gene for the test. Detection primers were listed in S1 Table. The transcription of each gene was normalized to that of 16S rDNA and expressed as fold change relative to the level of control.

Statistical analysis

T-test was used to analysis the bacterial motility, lysine autotrophy, H_2_O_2_ resistance and other phonotypic data. One way analysis of variance (ANOVA) was used for multi-group comparison of the gene transcription data. Statistical analysis was performed using SPSS v17.0 statistical software (SPSS Inc., Chicago, IL, USA). P values < 0.05 were considered statistically different.

Ethics statement

The animal experiments in this study were all approved by the Ethics Review Committee for Laboratory Animals of Zhejiang Wanli University (Approval No. 20250417001).

Results

Identification of the Sfp-type PPTase gene in P. plecoglossicida NB2011

The amino acid sequence of the expected PPTase of P. plecoglossicida was aligned with the other Sfp-type PPTases in prokaryotes. As revealed in Fig 1, the sequence showed the highest similarity of 59.5% with PcpS of P. aeruginosa, and less with entD of E. coli (24.9%). Three high conservative motifs were shared with all the sequences, including the L/VTH, LGL/ID and KES/A motif. The PPTase of P. plecoglossicida belongs to the F/KES subfamily.

Sequence alignment of Sfp-type PPTases.The protein sequences used for alignment are as follows: Pseudomonas plecoglossicida NB2011 PPTase, Pseudomonas aeruginosa Pcps (WP_003112468.1), Salmonella enterica typhimurium EntD (AAA91937.1), Bacillus substilis Sfp (ACG68433), and Bacillus cereus Sfp (WPQ43272.1). Conserved amino acids are shaded in black; gray shading indicates that 100% of the residues are similar at that position. The symbol “” indicates a number 10 less than the following number.*

Construction of a conditional deletion mutant ΔPPT::EntD P. plecoglossicida

At first, the wild-type strain was transferred with the recombinant plasmid pET22b-entD, then a normal procedure of double homologous recombination targeting of the PPTase ORF sequence was completed (S1 Fig). The upstream and downstream homologous fragments were amplified, respectively (S1 Fig A). Then the fusion fragment with expected length of 1520 bp was produced with primers MR1/MF2 (S1 Fig B). After ligation and transferring to E. coli DH5α, the positive clones were selected on LB agar with erythromycin and kanamycin. Then the recombinant plasmid was transformed from E. coli DH5α to P. plecoglossicida NB2011 through conjugation. PCR analysis demonstrated the expected 5’-end and 3’-end recombined fragments of the positive insertion clones (S1 Fig C and D). After double recombination the deletion mutants were screened on LB containing 20% sucrose and confirmed by PCR with the primers PPT-wA/dS (S1 Fig E).

Growth comparison of the mutants and WT

The growth curves of the mutant and WT strain were pictured in Fig 2. As disclosed, during the lag phase and early log growth phase, no apparent difference was observed between the two strains; as the cultivation time lengthened, the mutant showed poorer growth than the WT, although it’s not statistically significant (P > 0.05).

Growth curves of mutant and wild type strain.Measure the strain concentration within 14h under the condition of OD600 and draw the growth curve diagram.

Mobility assay of the mutant and WT strain

The motility of the mutant and WT strain was observed in 72h cultivation, as revealed in Fig 3. The strains swarmed similarly during the whole experiment span; while the mutant swam a little slower than the WT from the first 24h, and the difference was significant at 72h (P < 0.05).

Swarming and swimming ability of the mutant and wild type strain.A, Swarming motility of the mutant and WT; B, Swimming ability of the mutant and WT. Data was mean±SD of three independent experiments.

Conditional deletion mutation leads to autotrophy for lysine, hypersensitivity to hydrogen peroxide (H2O2), and declined siderophore activity

The conditional deletion mutant ΔPPT showed autotrophy for lysine under minimum media, the growth was very poor without additional supplement of lysine (Fig 4A), while the WT strain grew well on both the minimum media and complement media. On comparison to the WT, the mutant was more sensitive to H_2_O_2_ (Fig 4B). The mutant showed poorer siderophore activity than the WT (Figs 4C), the difference was highly significant (P < 0.01). The Csaky test exhibited a typical absorbance peak at around 520nm, with the maximum value occurring at 526nm in both the strains, while no peak appeared in the blank control without the supernatants, indicative of the presence of hydroxamate-type siderophores. When the supernatant was measured by Arnow’s test or UV-Visible Spectrofluorimetry, negative results were produced by both the strains, suggesting the bacteria did not secret catechol or pyoverdine-type siderophore.

Lysine autotrophy observation, sensitivity to H2O2 and siderophore activity of the mutant and WT strain.A, lysine autotrophy observation, after 12 hours, the OD600 of the bacterial culture was measured; B, sensitivity to H2O2 treatment; C, siderophore activity of the mutant and WT strain, after 24 hours cultivation, the culture supernatants were collected and siderophore activity measured; D, Csaky test for hydroxamte siderophore of the mutant and WT strain, the 24h hours culture supernatant was detected by Csaky test for hydroxamate type of siderophores. A, B, C, Data was mean±SD of three independent experiments. Asterisk “” and “**” denotes a significant difference at the level of P < 0.05 and P < 0.01, respectively. D, Representative data was shown from three independent replicate tests.*

Cellular fatty acid profiles

As revealed in the following Table 2, the conditional mutation of PPTase in P. plecoglossicida NB2011 did not alter the core framework of fatty acid metabolism. However, through direct downregulation of key enzyme (PPTase) activities and indirect metabolic flux redistribution, it resulted in a distinct fatty acid profile characterized by decreased unsaturated fatty acids, accumulated saturated fatty acids, reduced long-chain/cyclic fatty acids, and increased hydroxy fatty acids, accompanied by a significant reduction in total fatty acid synthesis.

Table 2: Fatty acid profiles of P. plecoglssicida NB2011 and the mutant ΔPPT::EntD.

PPTase is involved in intracellular reproduction in mouse Macrophage J774A.1

Adhesion, internalization and replication of the mutant and WT strain in the macrophage J774A.1 were investigated, and the results were shown in Fig 5. As revealed, adhesion, internalization and the early replication of the mutant were not affected by the deletion as the bacterial counts were similar during these processes; then from 8h to 24h post internalization, the intracellular bacterial load of the mutant was apparently lower than that of the WT, although the difference was not statistically significant (P > 0.05).

Adhesion, internalization and replication of the mutant and WT strain in the macrophage J774A.1.CFU level of the strains at −3 h (adhesion), 0 h (internalization) and 0-24 h (replication), respectively. The data is presented as mean ± standard deviation (SD) from 3 replicates for each sample.

Virulence of the conditional mutant against the large yellow croaker

To test the virulence of the strains, the same dose of bacterial suspension of the mutant and wild-type strain were intraperitoneally injected into the large yellow croaker, symptoms were observed, mortality were recorded in the following 7 days, and the results were displayed in Table 3. As revealed, on comparison to the WT infected group, death occurred latter in the mutant injected group, the cumulative mortality of the mutant infection group was 25%, much lower than the control (75%), indicating apparently attenuated virulence.

Table 3: Virulence test of the mutant and wild-type strain against the large yellow croaker.

RNA-seq transcription analysis of RNA-seq data

RNA-seq analysis revealed 179 differently regulated genes between the mutant and wild-type strain, including 104 up-regulated and 75 down-regulated genes. As listed in Table 4, almost all the top twenty of up-regulated DEGs are involved in GO term of membrane secretion, including type Ⅲ and type Ⅵ secretion system; while down-regulated genes are involved in a variety of GO terms, including oxidative stress response, metal iron transportation, ABC transporters, flagellar assembly, etc.

Table 4: Differently expressed genes (DEGs) in the mutant ΔPPT::EntD P. plecoglossicida.

The transcription analysis was validated by qRT-PCR (S2 Fig). The expression levels of six DEGs were revealed, including three up-regulated genes and three down-regulated genes, the tendency agrees with the results of RNA-seq.

Discussion

The unique PPTase in P. plecoglossicida NB2011 plays an essential role in the survival of the bacteria

Usually several PPTases present in a bacteria, including type ⅠPPTase involved in FAS pathway in primary metabolism, and type Ⅱ PPTase activating PKP and NRPS pathways, such as Sfp from B. subtilis and EntD from E. coli [19]. However, there are a few bacteria which encode a unique PPTase, including PapcpS from P. aeruginosa [20], XcHetI from Xanthomonas campestris pv. campestris [21], and a Sfp-PPTase in Legionella spp. [22]; such PPTase must be involved in both primary metabolism and secondary metabolism, so plays a crucial role in bacterial survival. The amino acid sequence of the PPTase in P. plecoglossicida shared a high similarity of 59.5% with PapcpS, suggesting the enzyme plays a similar role as PapcpS in P. aeruginosa, which majorly dedicates to modifying the apo-acyl carrier proteins (ACPs) to activate FAS pathways [6]. Apart from activating PCPs in the biosynthesis of the siderophore enterobactin, EntD of E. coli also catalyzes ACPs in long-chain n-3 polyunsaturated fatty acid synthesis when highly expressed, suggesting the PPTase is involved in fatty acid biosynthesis to a certain extent [23]. In our study, growth observation and cellular fatty acid profiling revealed that the mutant displayed a significant reduction in total fatty acid synthesis. However, the conditional mutation did not disrupt the core fatty acid metabolism framework or severely impair bacterial growth, suggesting that E. coli entD can partially compensate for the function of PPTase in P. plecoglossicida, notwithstanding their low amino acid sequence identity (24.9%) and potentially distinct mechanisms of action. The main function of the PPTase in our study should be involved in primary metabolism, such as FAS pathway, since routine manipulation of deletion mutation failed several times, until the recombinant plasmid pET22b-entD was transferred to the WT strain, suggesting the enzyme was independent for survival of the pathogen.

Generally, the deletion of a bacterium’s sole PPTase gene is feasible only when an alternative functional PPTase is provided in trans. This has been demonstrated in systems such as X. campestris pv. campestris complemented with E. coli EcAcpS, or P. aeruginosa PAO1 complemented with either XcHetI or EcAcpS, where cell viability was maintained despite impaired secondary metabolism (e.g., pigment or siderophore production) [21]. These observations support the notion that while type I PPTases (e.g., EcAcpS) are primarily dedicated to primary metabolism, some type II enzymes (e.g., PaPcpS, XcHetI) play dual roles in both primary and secondary metabolic pathways. Given that the PPTase of P. plecoglossicida shares low but significant identity with EntD of E. coli, and EntD also appears to participate in primary metabolism, we proceeded to delete the target sequence in an E. coli entD‑conditional background. Nevertheless, replacement with acpS should also be examined in future studies, which would likely yield distinct phenotypic outcomes and further clarify the biological functions of the PPTase in P. plecoglossicida.

In this study, RNA-seq data demonstrate basal expression of pET22b-entD in P. plecoglossicida NB2011. In the conditional mutant, EntD (corresponding to L321_RS03080) expression was approximately half of the PPTase in the wild-type (WT) strain (Table 3). However, the mechanism underlying this background expression remains unclear. While low basal expression from similar vectors (containing T7 promoters and LacI repressors) has been observed in E. coli BL21(DE3) [24], comparable findings in other bacteria have not been reported. Further research is needed to elucidate this mechanism.

The PPTase in P. plecoglossicida is involved in siderophore production, oxidative stress resistance and lysine synthesis

Several Sfp-type PPTases have been found to be involved in the production of siderophore, or other NRPS/PKP derived metabolites [5,20,25,26]. In this study, the PPTase in P. plecoglossicida also plays a role in siderophore production as disclosed in the phenotype investigation and transcription analysis. As revealed in Fig 4D, P. plecoglossicida NB2011 produced hydroxamate siderophore, which is consistent with the prediction of antiSMASH [27]; a possible NRPS gene cluster presents in P. plecoglossicida NB2011, which is composed of five core genes, encoding an ACP synthase, a TauD/TfdA family dioxygenase, two NRPSs, and a NRPS/PKS, respectively. (S3 Fig). The PPTase may catalyze these NRPSs and then activate the production of the siderophore. Nevertheless, no pyoverdine was detected in both the mutant and wild type strain, which is different from other Pseudomonads species [28], and whether the bacteria produce other types of siderophores, more research work is needed.

The down-regulated genes involved in oxidative stress response provided a good explanation for the sensitivity of the conditional mutant to H_2_O_2_ treatment, like that observed in the bacteria Lactoccus lactis [29], Streptococcus mutans [30], and the Cochliobolus fungal phytopathogens [31,32]. Researchers have suggested the iron homeostasis and oxidative stress resistance seemed to be regulated by a common transcription factor [8].

Lysine auxotrophy has been noticed in some plant pathogenic fungus, such as Ppt1 from Fusarium fujikuroi [8], PptA from Asperigillus fumigatus [14], a Sfp-type PPTase from Cochliobolus sativus [31], Ppt of Zymoseptoria tritici [33] and PPT1 from Colletotrichum graminicola [34]. It has been demonstrated that Sfp-PPTase activates α-aminoadipate reductase (AAR) involved in lysine biosynthesis in filamentous fungi [34,35]. Lysine-dependent growth of the mutant strain was also observed in our study, however, the involvement of PPTase in lysine biosynthesis has not been exactly revealed in bacteria, the precise mechanism remains unknown. It should be noted that RNA-seq analysis did not show down regulation of lysine synthesis in the mutant, which may owe to the culture media, the bacteria prepared for RNA-seq were cultivated in LB broth which is rich in amino acids.

As for the motility of the conditional mutant, bacterial swarming has not changed, while swimming ability declined in comparison to the WT, suggestion of an association between the PPTase and swimming motility. Downregulated expression of four flagellar genes in the mutant may provide a reasonable explanation. The results seemed different from the investigation in P. aeruginosa, although the swarming ability was abolished when replacing the acyl-carrier protein acpP1 (which is phosphopantetheinylated by PcpS to function in fatty acid biosynthesis) with E. coli acpP [36], the swimming motility was not affected. How is the mobility affected by the PPTase in P. plecoglossicida? That’s an interesting question that should be answered in further research.

PPTase associated iron uptake may be co-regulated with toxic secretion systems

RNA-seq analysis showed significant downregulated expression of genes involved in iron transportation but upregulated in T3SS and T6SS of the mutant strain, which suggests the two toxic secretion systems are enhanced under iron starvation, just like the previous investigation [4]. On the other hand, a significant decline of T3SS and T6SS was observed in a two-component system (TCS) deletion mutant ΔpvgAS P. plecoglossicida, suggesting the two systems being positively regulated by the PvgAS system [37]. In other pathogenic bacteria, such as enteroaggregative E. coli, P. aeruginosa, and E. tarda, ferric uptake proteins (Fur) negatively regulate iron uptake and toxic secretion systems have been well documented [38–40]. However, how the two systems are regulated with iron availability, is there a common transcription regulator of iron uptake and secretion systems in this pathogen, the problems are waiting for a solution.

The PPTase plays an important role in bacterial pathogenicity

In many bacterial and fungal pathogens, PPTases play important roles in pathogenicity [7,8,14,31,33,41], we achieved a similar result in this study. The conditional mutant showed declined persistence in the mouse macrophage and attenuated virulence against the host fish, which well indicates the target gene being involved in bacterial pathogenicity. Decreased resistance to oxidative stress may partially account for the less bacterial load of the mutant in the mouse macrophages. Substantially attenuated virulence in the croakers indicates iron depletion being the most major virulence factor of the pathogen. While the mutant showed lysine auxotroph when grew in minimum media, lysine deficient may not play important roles in bacterial virulence, as proteins nutrition may be rich during in vivo infection.

It’s worthy of mention that in this study the temperature value 28℃ was chosen for bacterial growth. There are several reasons: at first, it is almost the highest water temperature of the local aquaculture environment for croakers; and the bacteria reaches its fastest growth at this point, but lower at higher temperature of 35–37℃ [18]; lastly, the bacteria secret major virulence factors (including T3SS and T6SS）and develop infection after artificial challenge at this temperature [2]. So, this temperature is widely used for cultivation of P. plecoglossicida.

Conclusion

In this study, we identified and characterized a Sfp-type PPTase in the fish pathogen P. plecoglossicida, which is essential to bacterial growth, involved in the siderophore synthesis, oxidative stress resistance and lysine biosynthesis, and critical to the pathogenicity of the bacteria. This is the first paper on PPTase of this bacterial species and therefore provides some insight into this field.

Supporting information

S1 TablePrimers used for qRT-PCR detection.(DOCX)

S1 FigConstruction of a conditional deletion mutant ΔPPTase::entD *Pseudomonas plecoglossicida.*A, PCR amplification of upstream and downstream homogenous fragment of the target sequence; M，DL2000 DNA marker; lane 1, upstream homologous fragment (780 bp); lane 2, downstream homologous fragment (742 bp); B, PCR amplification the fusion fragment of the targeting sequence; M, DL5000 DNA marker; lane 1, the fusion fragment of upstream and downstream fragment (1520 bp); C, screening of the insertion strain by wS/-wA amplification; D, screening of the insertion strain by dS/-dA amplification; M: DL5000 DNA marker; lane 1 ~ 3, 8 ~ 10: 5’-end recombination clones with the expected fragment of 1649/1112 bp; lane 4 ~ 7, 3’-end recombination clones with the expected fragment of 1005/1801 bp; lane 11, negative control of the WT with the expected fragment 1649/1801 bp. E, primers PPTase-wA/dS assay of the deletion mutant. M: DL5000 DNA Marker; Lane 1–3: Positive clones with expected fragment of 702 bp; Lane 4: negative control of the WT with expected fragment of 1264 bp.(TIF)

S2 FigRT-PCR analysis to validate expression of genes in Pseudomonas plecoglossicida NB2011 and ΔPPT::entD.Total RNA was isolated from bacterial cultures grown in LB with Trizol RNA isolation kit (Invitrogen), and cDNA was synthesized for RT-PCR. RT-PCR analysis was performed with RNA obtained from three independent isolations and the figure shows results of one such experiment. Graph shows normalized fold difference of genes compared to 16S rRNA expression levels.(TIF)

S3 FigNRPS biosynthetic gene cluster and domain architecture of NRPS modules (antiSMASH analysis).The genome sequence of Pseudomonas plecoglossicida NB2011 was uploaded to antiSMASH homepage https://antismash.secondarymetabolites.org/#!/start. A, a predicted NRPS gene cluster was displayed. The fragments in pink and brown indicate additional biosynthetic genes and core biosynthetic genes, respectively. The number below indicates the genome location of the sequence. TE, thioesterase; ACP: acyl carrier protein; TauD: TauD/TfdA family dioxygenase; NRPS, non-ribosomal peptide synthetase (NRPS); PKS, polyketide synthase (PKS). B, domain-organization of the NRPS modules. CAL: condensing-like domain; CP, carrier protein; A, AMP-binding, C, condensation; I, interface; E, epimerization.(TIF)