Remodulation of bacterial transcriptome after acquisition of foreign DNA: the case of irp-HPI high-pathogenicity island in Vibrio anguillarum

ABSTRACT The high-pathogenicity island irp-HPI is widespread in Vibrionaceae and encodes the siderophore piscibactin, as well as the regulator PbtA that is essential for its expression. In this work, we aim to study whether PbtA directly interacts with irp-HPI promoters. Furthermore, we hypothesize that PbtA, and thereby the acquisition of irp-HPI island, may also influence the expression of other genes elsewhere in the bacterial genome. To address this question, an RNAseq analysis was conducted to identify differentially expressed genes after pbtA deletion in Vibrio anguillarum RV22 genetic background. The results showed that PbtA not only modulates the irp-HPI genes but also modulates the expression of a plethora of V. anguillarum core genome genes, inducing nitrate, arginine, and sulfate metabolism, T6SS1, and quorum sensing, while repressing lipopolysaccharide (LPS) production, MARTX toxin, and major porins such as OmpV and ChiP. The direct binding of the C-terminal domain of PbtA to piscibactin promoters (PfrpA and PfrpC), quorum sensing (vanT), LPS transporter wza, and T6SS structure- and effector-encoding genes was demonstrated by electrophoretic mobility shift assay (EMSA). The results provide valuable insights into the regulatory mechanisms underlying the expression of irp-HPI island and its impact on Vibrios transcriptome, with implications in pathogenesis. IMPORTANCE Horizontal gene transfer enables bacteria to acquire traits, such as virulence factors, thereby increasing the risk of the emergence of new pathogens. irp-HPI genomic island has a broad dissemination in Vibrionaceae and is present in numerous potentially pathogenic marine bacteria, some of which can infect humans. Previous works showed that certain V. anguillarum strains exhibit an expanded host range plasticity and heightened virulence, a phenomenon linked to the acquisition of the irp-HPI genomic island. The present work shows that this adaptive capability is likely achieved through comprehensive changes in the transcriptome of the bacteria and that these changes are mediated by the master regulator PbtA encoded within the irp-HPI element. Our results shed light on the broad implications of horizontal gene transfer in bacterial evolution, showing that the acquired DNA can directly mediate changes in the expression of the core genome, with profounds implications in pathogenesis.

pathogens (4,5).However, the newly acquired genes can impose an excessive fitness cost on recipient bacteria caused by, e.g., their uncontrolled expression (6).To counterbalance these effects, bacteria have evolved mechanisms, known as xenogeneic silencers, to repress the expression of horizontally acquired DNA (7).Interestingly, DNA acquired through horizontal gene transfer usually encodes factors that ensure its own expression ad hoc (8).Thus, the acquired DNA must be integrated into the recipient genome regulatory network.
The Vibrio genus is characterized by a large genotypic diversity, which is partially attributed to HGT (9).This encompasses a wide range of niche specialization from free-living bacteria to those associated with other organisms in a mutualistic, commen sal, or pathogenic relationship (9).Vibrio anguillarum is able to cause hemorrhagic septicemia (vibriosis) in warm-and cold-water fish species, leading to high mortalities and economic losses in aquaculture worldwide (10,11).Several virulence-related factors have been identified in this species, including lipopolysaccharide (LPS) (the most virulent strains belong to serotypes O1, O2, and, to a lesser extent, O3), motility and chemo taxis, quorum sensing, extracellular products with hemolytic and proteolytic activities, and up to three siderophores (11)(12)(13).Notably, certain V. anguillarum strains exhibit high virulence in a broad host range, causing great mortality rates in warm-and cold-water-adapted fish, even at temperatures as low as 7°C, which is well below the bacterial optimal growth conditions (14,15).This expanded virulence phenotype has been associated with the acquisition of the high-pathogenicity island irp-HPI through HGT (16,17).
The irp-HPI genomic island harbors the irp genes, which are responsible for the synthesis, transport, and utilization of the siderophore piscibactin (18,19) (Fig. 1).Piscibactin production plays a key role in virulence, not only in V. anguillarum (16) but also in other fish or mollusc pathogens such as Photobacterium damselae subsp.piscicida (20), V. ordalii (21), and V. neptunius (22).irp-HPI has a broad dissemination in Vibrionaceae and is present in numerous potentially pathogenic marine bacteria, some of which can infect humans (22)(23)(24).The piscibactin system has a dual requirement for iron starvation and low temperature to be expressed and is one of the most induced virulence factors in V. anguillarum when growth temperature drops below 20°C (16).Our recent work showed that the expression of piscibactin is controlled via a regulatory cascade involving the global regulators H-NS and ToxR-S, though none of them is the main actor (24).The irp-HPI encodes an AraC-like transcriptional regulator named PbtA (Fig. 1), whose inactivation disables the expression of piscibactin biosynthesis and transport genes (24,25).PbtA expression is greatly induced with temperature decrease, making it the FIG 1 Genetic map and distribution of GC content of irp-HPI genomic island encoding piscibactin system in Vibrio anguillarum.
main activator of the piscibactin siderophore system expression under low temperature.Interestingly, PbtA deletion results in a dramatic decrease in the degree of virulence, a decrease higher than that found after the inactivation of the piscibactin system alone (24).
In this study, we aimed to investigate whether the transcriptional regulator PbtA, and thereby the acquisition of the irp-HPI island, may influence the expression of core genes in V. anguillarum.To this purpose, an RNAseq analysis was performed to identify differentially expressed genes (DEGs) after the inactivation of pbtA in V. anguillarum RV22, a highly pathogenic strain that harbors the irp-HPI island (26) (16).Moreover, PbtA C-and N-terminal domains were purified, and their binding ability to the irp-HPI promoters and to promoters of some DEGs was tested by gel mobility shift analysis.Thus, V. anguillarum was used as a model system to elucidate the interaction of the irp-HPI island with the genome of the recipient bacteria.Our results provide valuable insights into the regulatory mechanisms underlying the expression of the horizontally acquired irp-HPI island and its impact on Vibrio transcriptome.

PbtA in silico analysis, recombinant protein expression, and purification
UniProt (27) and BLAST (28) were used to perform the functional and homology analysis of V. anguillarum PbtA (UniProtKB accession: A0A289GIJ0).PbtA and its N-or C-termi nal domains (PbtA N and PbtA C , respectively) were cloned into the expression vector pET20b(+).To obtain the PbtA N fusion protein, the region encoding amino acids M1-E211 was PCR amplified, whereas to create the PbtA C fusion protein, the region encoding amino acids S213-P326 was PCR amplified and cloned into the NdeI and XhoI restriction sites of the pET20b(+) vector in frame with a C-terminal (PbtA and PbtA N ) or N-terminal His-tag (PbtA C ).The oligonucleotides used are listed in Table S1.
The E. coli expression strains (Table S2) and the induction temperature were selected based on expression tests performed to determine the optimal conditions to achieve the maximum protein yield.Therefore, PbtA and PbtA N were overexpressed in E. coli BL21 at 17°C and 37°C, respectively, whereas PbtA C was expressed in E. coli BL21 pLysS at 37°C.The correspondent E. coli cells carrying the expression vectors were grown aerobically in LB medium at 17°C or 37°C until an OD 600 of 0.5 was reached, and protein expression was induced by adding 0.5 mM of IPTG and carried out overnight.Induced cells were collected by centrifugation at 4,000 × g for 30 min at 4°C, the supernatant was discarded, and the pellet was resuspended in 40 mL of resuspension buffer (50 mM Tris-HCl pH 8.0, 500 mM NaCl).The cells were lysed by sonication, and the lysate was centrifuged at 35,000 × g for 30 min at 4°C.The resultant supernatant was loaded onto a C-50 column packed with high-density nickel resin (ABT), and the fusion proteins were step-eluted with 50 mM Tris-HCl, pH 8.0, and 500 mM NaCl containing increasing concentrations of imidazole.The elution(s) containing the desired protein was concentrated at 4°C using Vivaspin centrifugal concentrators following the manufactur er's instructions and dialyzed (Spectra/Por dialysis membrane, Spectrum Labs) against 1 L of 50 mM Tris-HCl, pH 8.0, and 500 mM NaCl overnight at 4°C.The purified proteins were analyzed by SDS-PAGE, and the concentration was determined using a NanoDrop ND-1000 Spectrophotometer, taking into account the extinction coefficient (at 280 nm), and the molecular weight was calculated using ProtParam (29).The proteins were stored at −80°C until further use.

Electrophoretic mobility shift assay (EMSA)
The DNA regions to be tested were PCR amplified and end-labeled using the Biotin 3′ End-Labeling Kit (Thermo Scientific) following the manufacturer's instructions.EMSAs were performed using the LightShift Chemiluminescent EMSA Kit (Thermo Scientific) following supplier's recommendations.Briefly, 0.2 pmol, 0.5 pmol, or 1 pmol of PbtA N or PbtA C were mixed with 25 fmol of biotin-labeled DNA in a binding buffer containing 1 mM Tris, pH 7.5, 155 mM KCl, 0.1 mM DTT, 5% glycerol, 5 mM MgCl 2 , 1 mM EDTA, and 0.3 mg/mL BSA.To ensure the specificity of the interaction, a 100-fold molar excess of specific unlabeled DNA was added as a control.Binding reactions were incubated for 30 min at 20°C, loaded on a 5% polyacrylamide gel (acrylamide:bis-acrylamide, 29:1) in 0.5 × Tris Borate EDTA buffer, electrophoresed at 100 V for 1 h at 4°C and then transfer red to a positively charged nylon membrane.The transferred DNA was crosslinked to the membrane using a UV-light Linus MiniCross equipment.Biotin-labeled DNA was detected by chemiluminescence using a Fujifilm LAS-3000.The intensity of the bands was quantified using Multi-Gauge v 3.0 software (Fujifilm), and relative quantification of the band was calculated as the ratio to labeled probe controls.At least four relative quantifications were performed per EMSA band.Statistical significance was determined by Student's t test with a threshold P value < 0.05.

Construction of pbtA and pbtA N defective mutants and complementation
In-frame deletions of pbtA and pbtA N were constructed by allelic exchange in V. anguillarum RV22 wild-type strain background, respectively.The flanking sequences of the gene region encoding the N-terminal domain of PbtA (PbtA N ) were PCR amplified and cloned into the low-copy number vector pWKS30 (30).The plasmid was digested with NotI and ApaI, and the allele was cloned into the suicide vector pNidKan (31), resulting in the creation of the plasmid named pML1287.Subsequently, the previously constructed plasmid pML118, used for generating the ΔpbtA defective mutant (Table S2), and pML1287 were conjugated with the correspondent V. anguillarum strain, and transconjugants selection was based on ampicillin and kanamycin resistance.After a second event of recombination, the mutant strain was selected based on sucrose (15%) resistance and subsequent growth on ampicillin and kanamycin plates.PCR was used to confirm the allelic exchange event.Mutant complementation was achieved by cloning the wild-type pbtA gene into the vector pSEVA651 (32) in E. coli S17-1 λpir.The plasmid was then mobilized to the mutant strain by conjugation.

Transcriptional fusions and β-galactosidase assays
The previously constructed plasmids pMB276, pMB277, and pML212 (16,25) carrying the lacZ fusions PfrpA::lacZ, PpbtA::lacZ, and PfrpC::lacZ (Fig. 1) into plasmid pPHRP309 (33) were mobilized from E. coli S17-1 λpir (34) to V. anguillarum RV22 and the derivative mutants ΔpbtA and ΔpbtA N by conjugation.The resultant constructions were confirmed by PCR amplification of the promoter regions.For the evaluation of the transcriptional activity, the V. anguillarum strains carrying the lacZ fusions were grown under weak iron-restrictive conditions in CM9 minimal medium supplemented with 25 µM 2,2′-dipyr idyl.The bacterial cultures were grown aerobically at 15°C until they reached an OD600 = 0.3.β-galactosidase activities were measured following the method of Miller (35).The results shown are the mean of three independent experiments.Statistical significance was determined by Student's t test with a threshold P value < 0.05.

Growth conditions and total RNA extraction
V. anguillarum RV22 wild-type and RV22 ΔpbtA mutant strains were grown at 15°C in CM9 minimal medium under iron-restricted conditions achieved by adding 50 µM of 2,2′-μM dipyridyl.Bacterial cultures were grown until mid-exponential phase (OD 600 ≈0.8) and harvested by centrifugation at 10,000 × g for 10 min.Total RNA from three independent cultures was isolated using TRIzol Reagent (Invitrogen) following the manufacturer's instructions.RNA integrity was visualized in a 1% agarose gel, and the concentration was determined using Qubit TM RNA BR Assay Kit (Invitrogen).RNA was stored at −80°C until further use.The RNA integrity number (RIN) of the samples, measured using an Agilent 2100 Bioanalyzer, was higher than 8.

cDNA library construction and sequencing
Triplicate biological samples were used for RNA extraction to create independent cDNA libraries for whole-transcriptome sequencing.Before library construction, residual DNA was eliminated and bacterial rRNA depletion was performed.Each biological replicate was represented by a separate library, consisting of approximately 20 million 2 × 150 bp reads.The sequencing was conducted on an Illumina MiSeq sequencing machine using the NextSeq High Output 1 × 150 pb kit.The construction of cDNA libraries and sequencing services were provided by the FISABIO Sequencing and Bioinformat ics Service (Valencia, Spain, http://fisabio.san.gva.es/secuenciacion-masiva-y-bioinformatica).RNAseq reads per replicates were deposited at SRA database (Sequence Read Archive) under BioProject id.PRJNA991769 (Table S3).

Bioinformatic analysis and gene expression quantification
The Tuxedo suite, a collection of open-source software programs (36), was utilized for RNAseq data analysis and validation.Briefly, Tophat was used to align the reads to V. anguillarum RV22 genome (GenBank acc.No. GCA_000257185.1).Reads mapped statistics are shown in Table S3.The mapped reads were then assembled into poten tial transcripts, and a final transcriptome assembly was generated using Cufflinks.Cuffdiff was used for the differential expression analysis of genes across wild-type and pbtA mutant strains.The output files from Cuffdiff were processed using the R package CummeRbund, generating volcano plot and figures.Functional classification of differentially expressed genes (DEGs) was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database.

Reverse Transcription-Quantitative PCR (RT-qPCR) validation
The RNA concentration of each sample was adjusted to 10 µg/µL, and 4 µg was subjected to DNaseI treatment to ensure the total removal of DNA.The resultant RNA was used for the quantitative analysis using One-step NZY RT-qPCR Green kit (NZYTech).Primers used to detect the expression of 16S ribosomal RNA (reference gene), frpA, wza, hcp1, and vanT are listed in Table S1.Three independent RT-qPCR reactions were performed using a CFX96 Real-Time PCR Detection System (Bio-Rad) following the cycling setup with an initial reverse transcription of 50°C for 20 min, polymerase activation at 95°C for 10 min, and finally, 40 cycles of denaturation at 95°C for 15 s and annealing/extension at 60°C for 30 s.To confirm the amplification of only one product, a melting curve was performed from 65°C to 95°C with consecutive increments of 0.5°C each 5 s.The 2 (-ΔΔct) method (37) was used to calculate the relative fold gene expression normalized with rRNA 16S.

Biofilm formation
The crystal violet staining assay was used for biofilm quantification.V. anguillarum RV22, ΔpbtA mutant, and the correspondent complemented strain were grown overnight aerobically at 25°C.The cultures were adjusted to an OD 600 of 0.5, and a final dilution of 1:20 was inoculated in CM9 minimal medium supplemented with 50 µM of 2,2′-dipyridyl in a final volume of 200 µL.The 96-well microtiter plate was incubated aerobically at 15°C until early exponential phase (OD 600 = 0.3).Subsequently, the cultures were incubated under static conditions for 48 h to allow biofilm formation.The growth achieved was quantified by measuring the optical density at 600 nm after resuspending the content of each well using a pipette.For biofilm quantification, the content of the plate was discarded, and the attached bacteria were fixed with 200 µL of 99% methanol (Panreac).After fixation for 2 min, the methanol was discarded and crystal violet was added to each well and incubated for 5 min.The excess dye was rinsed off with distilled water, and the dye bound to the biofilm was solubilized with 200 µL of 33% (vol/vol) glacial acetic acid (Panreac).Biofilm quantification was performed by measuring absorbance at 570 nm in an iMark Microplate Reader (Bio-Rad).The results shown are the mean values of three independent experiments with four technical replicates each one.Statistical significance was determined by Student's t test with a threshold P value < 0.05.

Purification of C-and N-terminal domains of PbtA
The transcriptional regulator PbtA is 326 amino acids in length and exhibits a twodomain arrangement commonly found in AraC-like transcriptional regulators such as ToxT (38,39) (Fig. 2A).The N-terminal domain of PbtA (PbtA N ) is predicted to com prise the first 210 amino acids.Following a short linker of ca 15 aa, the C-terminal domain of PbtA (PbtA C ) (region S228-P326) contains the characteristic helix-turn-helix (HTH) DNA-binding domain.BLASTP search showed that ToxT of V. cholerae (UniProtKB accession: A5F384) is the closest homologue of PbtA whose function has been studied (38).The overall sequence identity shared between the two proteins is only 14.23%, although it increases to 29% (48% similarity) when the alignment is limited to their C-terminal domains (Fig. 2C).Thus, PbtA does not exhibit significant aa similarity to any previously characterized AraC-like regulators.
DNA sequences encoding full-length PbtA or PbtA C and PbtA N domains (Fig. 2B) were cloned into the expression vector pET20b(+) in frame with His-tags, and their expression was induced in E. coli.The resulting proteins were purified using Ni 2+ -affinity chromatog raphy.PbtA C (14.25 kDa) and PbtA N (24.5 kDa) domains were successfully produced and purified (Fig. S1; Fig. 2D).However, the complete PbtA (38 kDa) protein was predomi nantly found in the insoluble fraction, suggesting low solubility.In agreement with this, the purification resulted in a low yield of the protein (Fig. S1), and several concentration attempts led to protein degradation (Fig. 2D).Given the challenges with solubility and protein stability, it was decided not to use the complete PbtA protein in subsequent experiments.Thus, the feasible production and purification of PbtA C and PbtA N allowed their utilization for in vitro studies.

PbtA C-terminal domain binds directly to the irp-HPI promoters PfrpA and PfrpC
To assess the binding ability of the two domains of PbtA (PbtA C and PbtA N ) to 3′end biotin-labeled DNA probes, EMSA was employed.In the initial set of EMSAs, DNA probes of ca.350 bp that encompassed the irp-HPI promoters PpbtA, PfrpA, and PfrpC (Fig. 1) were used.Specifically, PfrpA probe corresponds to the pbtB-frpA intergenic region of 333 bp and controls the expression of piscibactin biosynthesis and transport genes (frpAirp1-9).PfrpC controls the expression of frpC and frpB transport genes.PpbtA and PfrpC probes include the 330-and 340-bp DNA region upstream of the ATG start codon of pbtA and frpC, respectively.PbtA does not regulate its own promoter (24) so the PpbtA probe was used as EMSA-negative control.EMSA assays were performed as described in experimental procedures, and the interpretation of the results relied on two indicators: (1) the decrease in the band intensity of the free DNA compared to the lane that contained only DNA, and (2) the appearance of a band at the top of the gel.These two indicators revealed the formation of the PbtA-DNA complex without relying on a conventional shift in DNA gel mobility, which was not possible to observe due to the high isoelectric point (pI) of the PbtA C (pI = 10.43).
The results of the EMSA assays showed that PbtA C caused a reduction in the relative band intensity, denoting mobility shift, in the DNA probes that cover PfrpA and PfrpC promoter sequences.PfrpA and PfrpC showed relative quantifications of 0.49 ± 0.03 and 0.41 ± 0.03, respectively, when 1 pmol of PbtA C was added to the EMSA reaction (Fig. 3A).As expected, it did not bind to its own promoter (PpbtA) (Fig. 3A).These results indicate that PbtA binds directly to piscibactin promoters PfrpA and PfrpC.In addition, the PbtA N was unable to bind any of the DNA probes tested (Fig. 3B).This finding suggests that there are no DNA-binding determinants present in the N-terminal domain of the protein and that PbtA can bind to the target DNA sequences as a monomer.Interestingly, deletion of PbtA N leads to a significant reduction in PfrpA and PfrpC transcriptional activity, matching the expression levels observed in the PbtA null mutant (Fig. 4).This result clearly suggests that the N-terminal domain of PbtA is required for its functionality and consequently for the expression of irp genes.
Further EMSA assays were performed to identify specific region(s) within the piscibactin promoters PfrpA and PfrpC that PbtA binds to (Fig. 5).These assays aimed to narrow down the PbtA binding region.The pbtB-frpA intergenic region with a size of approximately 333 bp includes a conserved region among all variants of irp-HPI island proximal to the ATG of pbtA, and a low complexity sequence with six repeats of an AAAAT motif (between positions 137 and 167) (24) (Fig. 5A).Upon testing the low complexity region by EMSA, the formation of a PbtA-DNA complex was not detected (Probe 4 in Fig. 5A).The minimum DNA fragment of PfrpA required for a positive EMSA result was the region spanning from −13 to −150 of the ATG start codon (Probe 5 in Fig. 5A).In PfrpC, the minimum EMSA-positive region was defined as that spanning from positions −10 to −160 (Probe 12 in Fig. 5B).Interestingly, the PbtA C showed a higher affinity for the probes encompassing the larger frpC promoter regions (e.g., relative quantification of Probe 8 was 0.40 ± 0.07) than the smaller ones: Probe 10 (0.48 ± 0.03) and Probe 12 (0.70 ± 0.04) (Fig. 5B).We hypothesized that PbtA could have the property to bind to multiple binding sites in relatively large and diffuse DNA regions.Additionally, the EMSA-positive probes of PfrpA and PfrpC promoters would contain the −10 and −35 promoter elements and share a sequence of 22 bp with the motif 5′-TTTTATRCCTWATTSMGTTAGC-3′ (Fig. 6).It is interesting to note that both irp-HPI promoter regions, PfrpA and PfrpC, have an extremely low G + C content (Fig. 1).This suggests that a low G + C content may be a requirement for PbtA binding to DNA.

The inactivation of pbtA induces changes in the whole-genome expression profile
To gain insight into the putative role of PbtA in the modulation of V. anguillarum whole-genome expression, an RNAseq assay was conducted.Therefore, the transcrip tomic profile of the RV22 wild-type strain was compared to that of a pbtA null mutant (RV22 ΔpbtA).Both strains were subjected to growth at low temperature (15°C) under weak iron-restrictive conditions, to mimic the iron-starved conditions encountered during the host-pathogen interaction and to ensure the expression of virulence factors (17).
The transcriptomic analysis showed that approximately 16% of V. anguillarum RV22 genome, specifically 569 out of 3,621 genes, were differentially expressed: 370 genes were downregulated, whereas 199 genes were upregulated upon PbtA deletion (Fig. 7A) (Table S4).The differentially expressed genes (DEGs) were grouped into functional Kyoto Encyclopedia of Genes and Genomes (KEGG) categories (Fig. 7B).The results indicate that the deletion of pbtA results in a downregulation of genes related to signal transduction mechanisms (T), transcription (K), amino acid metabolism and transport (E), energy production and conversion (C), inorganic ion transport and metabolism (P), and translation, ribosomal structure and biogenesis (J).Conversely, functions related to replication, recombination, and repair (L) and cell wall/membrane biogenesis (M) were upregulated.Numerous proteins (ca.180) with unknown function were found within the differentially expressed genes.Differential expression of the down-expressed genes frpA, vanT, and hcp1, as well as the up-expressed wza, was verified by RT-qPCR, showing 2 (-ΔΔct)  values of 0.07 ± 0.02 (mean ± SD), 0.73 ± 0.18, 0.37 ± 0.18, and 1.95 ± 0.51, respectively.These results strongly suggest that PbtA exerts, directly or indirectly, a global regulatory effect in V. anguillarum transcriptome.The most representative DEGs related to metabolism and virulence are illustrated in Fig. 8. Specifically, the expression of genes napFDABC encoding periplasmic nitrate reductase (Nap system) decreased by 3.8-fold.Genes related to sulfur metabolism, including cysKDNCGJ, exhibit an overall downregulation of 2.7-fold.Additionally, genes involved in sulfate uptake, such as cysT and cysP, showed a 3.9-fold downregulation.The inactivation of PbtA also impacts arginine biosynthesis and transport.The expression of genes involved in the linear synthesis pathway of arginine, argA-H genes (40), was diminished by 5.1-fold.The arginine transport system encoded by the gene cluster argT-hisJQMP showed a significant 2.8-fold decrease in expression.In agreement with this, the pathway involved in de novo synthesis of arginine as well as its import from the extracellular environment showed a reduced expression.Complementary, the alterna tive pathway for arginine synthesis was 14.8-fold induced.This pathway involves three enzymatic steps mediated by ArgF and ArcAC (40).The gene arcA, which encodes an arginine deiminase responsible for the interconversion of citrulline and arginine, showed a significant 16.8-fold upregulation.
When examining the list of differentially expressed genes, several well-known virulence factors were found (Table 1).As previously described, PbtA is required for the expression of the piscibactin system (24).Congruently, irp-HPI genes were found within the most downregulated differentially expressed genes (−17.4-fold).A slight induction of vanchrobactin genes was also observed, with an overall fold change of 1.4.However, significant changes were only detected in vanchrobactin export (vabS), utilization (vabH), and regulatory (vabR) genes.The vanchrobactin TonB-dependent outer membrane transporter gene fvtA was constitutively expressed under both conditions.The overall expression of the MARTX toxin, encoded by the rtxACHBDE operon ( 41), showed an increase of 2.1-fold.Among these genes, the toxin acyltransferase rtxC and the toxin ABC transporters component rtxE were significantly upregulated, with a 3.4-fold and 2.0-fold increase, respectively.More notably, genes related to the T6SS1 showed a 6.4-fold downregulation, whereas the T6SS2 was 2.9-fold induced.In addition, several T6SS-rela ted genes located elsewhere in V. anguillarum genome were also differentially expressed.Specifically, the effector proteins and components of the structural puncturing device hcp1, hcp2, and vgrG showed a reduction in their expression of 143.5-, 39.7-, and 5.72-fold, respectively.Two copies of the structural component PAAR domain displayed a divergent expression, with one showing a 4.6-fold induction and the other a 2.64fold repression.Deletion of PbtA appears to also have an impact on the outer mem brane components of V. anguillarum.Specifically, the expression of two major porins, OmpV and ChiP, was induced 3.4-fold and 2.3-fold, respectively.The genes involved in lipopolysaccharide (LPS) biosynthesis showed a 2.6-fold upregulation, indicating an increase in LPS production.Moreover, the genes responsible for exopolysaccharide transport and assembly, wzbai and wbfcD operons, were twofold upregulated in the mutant strain.These findings suggest that the deletion of PbtA leads to a remodulation of V. anguillarum outer membrane, including an increased expression of some porins and LPS.
As mentioned earlier, numerous genes encoding functions grouped in the categories signal transduction mechanisms (T) and transcription (K) were differentially expressed.These include ca.70 genes encoding some diguanylate cyclases, LysR-and AraC-type transcriptional regulators (Table S4).The expression of rpoS, the RNA polymerase sigma factor S ( 42), showed a significant 7.8-fold decrease (Table 1).Notably, the inactivation of PbtA led to a significant downregulation of components of the quorum sensing system, including the master regulator VanT (8.8-fold decrease).The sensor kinase proteins, VanN and CqsS, responsible for signal recognition of autoinducers AI-1 and CAI-1 and the TetR-like transcriptional regulator LuxT showed a 2.3-, 7.9-, and 2.1-fold downregulation.Quorum sensing plays a central role in bacteria regulation and virulence, including biofilm formation (43).Congruently, a reduced capacity for biofilm formation was verified in the pbtA defective mutant compared to wild-type strain (Fig. 9).This finding raises the question of whether the observed changes in the whole-genome expression pattern associated with PbtA result from the direct interaction of PbtA with the promoters of regulated genes, or if additional players are involved.

PbtA would act as both activator and repressor in the modulation of genes located outside of the irp-HPI genomic island
We have shown above that PbtA efficiently binds to irp-HPI promoter regions, but its ability to directly modulate differentially expressed genes (DEGs) located outside of the genomic island was not elucidated.Thus, EMSA assays were performed to study whether  a Fold change values with P < 0.05 are shown; ns, non-significant differences detected; nd, non-determined; HIDATA, too many fragments in locus.
PbtA directly binds to promoters of certain virulence-related genes.The promoter regions tested included those of the downregulated genes vanT, T6SS genes vipA, hcp1, hcp2, and vgrG, as well as the promoter of LPS transport and assembly genes wziab.LPS genes were over-expressed after inactivation of PbtA (see above).DNA probes and EMSA results are shown in Fig. 10.The results showed that PbtA C binds to the promoters of vipA (0.20 ± 0.04) (Fig. 10A), hcp2 (0.28 ± 0.06), and LPS (wzi) (0.26 ± 0.06) (Fig. 10C) with higher affinity than irp promoters PfrpA (0.49 ± 0.03) and PfrpC (0.41 ± 0.05) (Fig. 3A and 10A).It exhibited a higher binding affinity to the promoter region of vgrG even at the lowest protein concentration tested (0.09 ± 0.03) (Fig. 10A) and was also capable of binding to the promoter region of vanT (0.81 ± 0.04) (Fig. 10B) and hcp1 (0.74 ± 0.02) (Fig. 10C), albeit with lower affinity.These findings greatly suggest that PbtA may play a direct role in regulating the expression of a great range of genes located elsewhere in V. anguillarum genome, including important virulence factors described in Vibrio species.Interestingly, PbtA efficiently binds to promoters of both down-and up-expressed genes, which suggests that PbtA would act as activator or repressor, depending on the target promoter.

DISCUSSION
AraC-like proteins involved in the regulation of carbon metabolism (e.g., the arabinose operon regulator [AraC] ), are active as dimers, and their activation is induced by effector molecules that bind to the N-terminal domain (44).In contrast, those involved in stress response (e.g., SoxS, MarA, and Rob) and regulation of virulence factors (e.g., ToxT) can function as monomers (45)(46)(47)(48).EMSA results provided clear evidence that the C-terminal domain of PbtA functions by directly binding to the promoters of the piscibactin biosynthesis and transport genes (PfrpA and PfrpC).The N-terminal domain is not needed for stable DNA binding of PbtA in vitro, but it is required for its function ality.Thus, it is hypothesized that the N-terminal domain of PbtA has a regulatory role, potentially functioning as a sensor for environmental signals or effector molecules  (48)(49)(50).Further studies are necessary to determine the specific effector(s) or environmental signal(s) that trigger the response of PbtA and to identify its mode of action.More notably, the results demonstrate that PbtA not only is required for the expression of the irp-HPI genes but also modulates approximately 16% of V. anguil larum transcriptome.This includes the ability to produce piscibactin (encoded by irp-HPI) and the induction of genes related to nitrate, arginine, and sulfate metabolism, Type VI Secretion System 1 (T6SS1), and quorum sensing.Simultaneously, it leads to the repression of genes associated with lipopolysaccharide (LPS) production, Type VI Secretion System 2 (T6SS2), MARTX toxin, and major porins like OmpV and ChiP.Repression of outer membrane components such as LPS would allow the pathogen to evade the host immune system response and persist within the host (51).Interestingly, most of these factors, including T6SS, were found to be essential for V. anguillarum persistence during the bacteria-host interaction (12).The contrasting effect of PbtA on the expression of each T6SS system is remarkable, as it activates T6SS1 and represses T6SS2.The expression pattern and the role in fitness and virulence of each T6SS varies among Vibrio species and also depends on the specific subset of effector proteins (52,53).The concrete role of T6SSs in V. anguillarum needs to be further studied.Overall, the results show that PbtA modulates the expression of numerous factors known to be required for virulence in Gram-negative bacteria (41,(54)(55)(56).RNAseq results suggest that PbtA primarily functions as a transcriptional activator, as most differentially expressed genes showed reduced expression after its inactivation.Specifically, we proved that PbtA directly interacts with the promoter region of down-expressed genes such as the quorum sensing master regulator vanT, and structural components of the T6SS1 (vipA) and its probable puncturing device (hcp1, hcp2, and vgrG genes).But it also binds to the promoter region of LPS transport and assemble genes (wziab), whose expression significantly increases in the PbtA mutant strain.This result suggests that PbtA may also act as a repressor.Although most characterized members of this family act as transcriptional activators, a few have been found to function exclusively or additionally as repressors (39).In this concern, ToxT requires dimerization for the activation of most target genes, such as cholera toxin gene ctxA, but it can also function as a monomer to activate other genes (57,58).In addition, it binds as a monomer to three "toxbox" sequences within the msh operon to inhibit mshA transcription (59,60).Our results altogether provide strong evidence that PbtA is directly responsible for modulating the transcription of genes located elsewhere in V. anguillarum genome and that its role as activator or repressor could depend on the specific target gene.
Due to the diverse mechanisms of action exhibited by AraC-like regulators, it can be challenging to accurately predict the binding sites and the genes they modulate (39).Several members of the family interact with the RNAP α-subunit C-terminal domain (α-CTD) for transcription activation (45,61,62).We have demonstrated that PbtA efficiently binds to DNA regions of approximately 150 base pairs, which share a 22-bp sequence with the motif 5′-TTTTATRCCTWATTSMGTTAGC-3′, and extend from position −10 of the ATG start codons of both frpA and frpC promoters.Thus, we hypothesized that the presence of this sequence motif would be an important feature for PbtA binding to these regions.However, PbtA exhibited a high affinity for the promoter region of vgrG and efficiently binds to the promoters of hcp2 and LPS production (wzi promoter region), and these promoter regions do not share the DNA motif identified in the irp-HPI.Apparently, the unique common characteristic shared by the promoter regions regulated by PbtA is a low GC content.AraC-like regulators commonly recognize specific nucleo tide motifs or structural DNA features, such as the intrinsically curved A + T rich DNA sequences, which are often located in or near the −30 region of transcription start sites (45).Thus, they can play a role in counteracting the gene silencing effects mediated by global negative H-NS regulators, one of the most studied xenogeneic silencers (8,63).A noteworthy feature of the AraC/XylS family is the capacity of certain members (e.g., AraC or MelR) to mediate DNA looping, forming the so-called "repression loops" (44,64).Further studies will be needed to investigate the functional significance of PbtA preference for AT-rich promoter regions and to elucidate the precise mechanisms by which PbtA recognizes and interacts with its target DNA sequences.
V. anguillarum possess two quorum sensing systems: an acyl-homoserin lactone (AHL) system involving the pair VanI/VanR, and a three-channel system that modulates the expression of the target genes through VanT (65).Our results show that PbtA plays a role in modulating the quorum sensing system in V. anguillarum.The ability of PbtA for binding vanT promoter region suggests that this effect could be a direct interac tion.Some virulence factors of V. anguillarum such as EmpA metalloprotease, pigment production, and biofilm formation are regulated by quorum sensing (65)(66)(67).In addition, RpoS of V. anguillarum plays important roles in bacterial adaptation to environmental stresses and pathogenicity, promoting the expression of quorum sensing and virulence factors (42,68).However, the role of quorum sensing in V. anguillarum virulence remains unclear since some studies suggest that it would not play a major role (65,66,69).Interestingly, recent research has revealed that the importance of quorum sensing in virulence varies greatly among V. anguillarum strains (70).By influencing the quorum sensing system, PbtA may have far-reaching effects on the overall gene expression patterns and behaviors of V. anguillarum.
Besides virulence factors themselves, the inactivation of PbtA also leads to the downregulation of various cellular processes that may play roles in the bacteria-host interaction.These include the periplasmic nitrate reductase, Nap operon, which act as an alternative electron acceptor in the absence of O 2 (71), or sulfate metabolism (72).The genes encoding sulfate metabolism are repressed by cysteine, the end-product of the pathway (73,74).Given that cysteine is a precursor for piscibactin synthesis (75), the non-production of the siderophore likely results in an intracellular accumulation of this amino acid, contributing to the observed downregulation of sulfate metabolism.The results suggest that the deletion of pbtA likely abolishes the de novo arginine synthesis and also its import from the extracellular environment.Bacterial arginine requirements would be met by the upregulation of arcA, an arginine deiminase required for the alternative pathway for arginine synthesis (76).Arginine importance is not restricted to protein synthesis, as this amino acid can be used as nitrogen and carbon source and, along with its precursor ornithine, is a substrate for the synthesis of polyamines as well as proline.Additionally, the involvement of arginine pathway in acid resistance under anaerobic conditions in E. coli (40) raises the possibility that this pathway may also play a role in V. anguillarum persistence within the host.
Previous results showed that certain V. anguillarum strains exhibit a remarkable host range plasticity (14,15), which was associated with the acquisition of the irp-HPI genomic island through horizontal gene transfer (16,17).The present work shows that this adaptive capability is likely achieved through comprehensive changes in the transcrip tome of the bacteria and that these changes are mediated by the master regulator PbtA.The global regulatory effect of PbtA on V. anguillarum transcriptome counterbalances the reduced growth ability of this pathogen at cold temperatures (17) by inducing the overproduction of numerous virulence traits and promoting evasion of host immunity.Our results reinforce the idea that the acquisition of a genomic island can directly mediate changes in the expression of the core genome, as it was suggested for the pathogenicity island SPI-1 of Salmonella or PAPI-1 of Pseudomonas aeruginosa (77,78).
Overall, our results provide valuable insight into the significant role of PbtA, and hence about irp-HPI acquisition, in modulating the expression of a plethora of V. anguillarum core genome genes, virulence factors, and behavior of V. anguillarum, with notable implications in adaptation to the host and pathogenesis.Further research will unravel the complete role of PbtA as a master regulator in modulating gene expression in Vibrionaceae.Indeed, the present work clearly shows that horizontally acquired DNA has the capacity to remodel the bacterial transcriptome.

FIG 2 (
FIG 2 (A) AlphaFold protein model of PbtA (UniProtKB accession: A0A289GIJ0) where colors represent per-residue confidence score (pLDDT).(B) Schematic representation of the recombinant proteins produced in this work (PbtA, PbtA C , and PbtA N ) where the numbers correspond to the amino acid positions in PbtA.(C) Amino acid alignment of PbtA and V. cholerae ToxT (UniProtKB accession: A5F384).(D) SDS-PAGE gel of PbtA, PbtA C , and PbtA N after purification and concentration process.Black triangles denote the expected migration of each recombinant protein: PbtA (38 kDa), PbtA C (14.25 kDa), and PbtA N (24.5 kDa).

FIG 3
FIG 3 Electrophoretic mobility shift assays of PbtA C (A) and PbtA N (B) binding to probes encompassing irp-HPI promoters (PpbtA, PfrpA, and PfrpC).R1 denotes the labeled probe controls and does not include protein in the reaction; R2 includes increasing concentration of PbtA C or PbtA N ; R3 denotes reaction controls that include protein and an excess of non-labeled DNA.Asterisks denote statistical significance (*P < 0.05), and the relative quantification of the band is denoted within the parentheses.

FIG 4
FIG4 Transcriptional activity of the irp-HPI promoters PpbtA, PfrpA, and PfrpC in either V. anguillarum RV22 wild type strain (WT), used as the parental strain, or its derivative ΔpbtA N and ΔpbtA defective mutants.Asterisks denote statistical significance, *P < 0.05.

FIG 5
FIG 5 Schematic representation of frpA (A) and frpC (B) promoter regions, probes used for EMSA, and representative results.Numbers correspond to nucleotides position counting from the ATG star codons.Two regions are depicted in the frpA promoter region according to their presence among Vibrionaceae versions of irp-HPI: one harboring a 6xAAAAT motif and the other containing a conserved region.R1 denotes the labeled probe controls and does not include protein in the reaction; R2 includes increasing concentration of PbtA C ; R3 denotes reaction controls that include protein and an excess of non-labeled DNA.Asterisks denote statistical significance (* P < 0.05), and the relative quantification of the band is denoted within the parentheses.

FIG 6
FIG 6 Alignment of ca. 150 bp of PfrpA and PfrpC regions.A conserved motif is highlighted in red.Gray shade denotes probable −10 promoter motif.

FIG 7 FIG 8
FIG 7 Modifications in V. anguillarum transcriptome after pbtA inactivation (expression values of ΔpbtA mutant compared to the wild-type strain).(A) Volcano plot showing differentially expressed genes (DEGs).(B) Functional classification of DEGs into KEGG categories.

FIG 9
FIG9 Biofilm formation of V. anguillarum wild-type, its derivative pbtA defective mutant, and complemented strain.Biofilm quantification was performed using the crystal violet assay.The results shown are the mean ± SD of three independent experiments with four technical replicas.Asterisk denotes statistically significant differences.Representative image of a biofilm assay is shown.

FIG 10
FIG10 Electrophoretic mobility shift assays of vipA, vgrG, vanT, wziab, hcp1, and hcp2 promoter regions.PfrpC (probe 9) was used as control.Gene cluster organization and GC content are shown for each genomic region.EMSAs were performed as described in Material and Methods: R1 denotes the labeled probe controls and does not include PbtA C protein in the reaction; R2 includes increasing concentration of PbtA C ; R3 denotes reaction controls that include protein and an excess of non-labeled DNA.Asterisks denote statistical significance (*P < 0.05), and the relative quantification of the band is denoted within the parentheses.

TABLE 1
Most relevant differentially expressed genes related to metabolic and virulence functions after deletion of pbtA in V. anguillarum a