Pseudomonas aeruginosa strain UCBPP-PA14 (PA14) was used as the wild-type strain. All Escherichia coli and Pseudomonas strains were grown at 37°C in lysogeny broth (LB) or lysogeny agar (Lennox) unless otherwise indicated in the method. The following concentrations of antibiotic(s) were added: carbenicillin (Cb) at 200 μg/ml and gentamicin (Gm) at 75 μg/ml for P. aeruginosa; gentamicin (Gm) at 15 μg/ml and ampicillin (Ap) at 100 μg/ml for E. coli. All bacterial strains and plasmids used in this study are listed in Table 1.

A gidA mutant strain was constructed using the pKNOCK suicide vector (Alexeyev, 1999). A 200 bp gidA fragment was amplified from the genomic DNA of P. aeruginosa PA14 using polymerase chain reaction (PCR) with the primer pair BT4409 (5′-CACATCGGCCTGGAGAAC-3′) and BT4410 (5′-GAACCGAGGAAGGACATCAC-3′). The amplified fragment was ligated into the SmaI site of the pKNOCK vector, which carried a Gm^R cassette (pKNOCK-Gm^R). The ligation reaction was transformed into E. coli strain BW20767, which was screened for the gentamicin resistance phenotype (Metcalf et al., 1996). The resulting plasmid was named pKN-gidA and transferred to the recipient P. aeruginosa PA14 by biparental mating using E. coli BW20767 as the donor strain. The gidA mutant was verified using colony PCR and confirmed using Southern blot analysis.

A full-length DNA fragment of the gidA gene was amplified from PA14 using primers BT5113 (5′-CCGAGGTGCGTGGTGGATT-3′) and BT5114 (5′-TGGGTTACCGCAGACATCAAG-3′). The gidA full-length DNA fragment was ligated into the SmaI restriction site of the pBBR1MCS-4 vector (Kovach et al., 1995), which contains an Amp^R cassette, to produce a pBB-gidA plasmid. To obtain the gidA complemented strain, the verified pBB-gidA was transformed into a gidA mutant by electroporation. The gidA complemented strain was then screened on LB agar plates containing carbenicillin and gentamicin and confirmed using colony PCR analysis.

Total tRNA in vivo was purified as previously described (Jaroensuk et al., 2016; Thongdee et al., 2019). Total RNA was purified from logarithmic phase cultures using Trizol reagent (Invitrogen). Large RNA species were separated from small RNA species using 35% ethanol precipitation. tRNA was purified from the small RNA pool using size-exclusion high performance liquid chromatography (SEC-HPLC) with an Agilent SEC3 300 Å, 7.8 × 300 mm column. The SEC-HPLC was operated under isocratic elution mode using 100 mM ammonium acetate as the mobile phase. The concentration and quality of total tRNA were assessed using a bioanalyzer (Agilent Technologies). To produce ribonucleoside products, the purified tRNA was treated with benzonase nuclease (Sigma), bacterial alkaline phosphatase (Invitrogen), and phosphodiesterase in the presence of deaminase inhibitors (0.5 μg/ml coformycin and 5 μg/ml tetrahydrouridine) and antioxidants (50 μM desferrioxamine and 50 μM butylated hydroxytoluene) at 37°C overnight. The ribonucleosides were fractionated using reverse phase chromatography (Thermo Hypersil Gold a Q column, 100 × 2.1 mm, 1.9 μm particle size) with a gradient of 0.1% (v/v) formic acid in water (solvent A) and 0.1% (v/v) formic acid in acetonitrile (solvent B). The HPLC column was directly connected to a triple quadrupole mass spectrometer (Agilent LC/QQQ 6490) operated in the positive ion mode. Ribonucleosides were identified by comparing their m/z, CID fragmentation patterns, fragmentor voltages, collision energies, and HPLC retention times with chemical synthetic standards.

MIC assays were performed to determine the minimum inhibitory concentration of H₂O₂ - that is, the lowest concentration that inhibits bacterial growth (Wiegand et al., 2008). Briefly, an overnight cell culture was diluted in Mueller Hinton Broth (MHB) and grown to a turbidity of 0.5 McFarland standard. The culture was then diluted by a factor of 1:20 with 0.9% sterile normal saline. The diluted culture was used to inoculate each well of a 96-well plate containing different concentrations of a test chemical solution prepared in MHB. MHB without cells were used as a negative control, while MHB with cells inoculated but without any test chemical was used as a positive control. The 96-well plate was incubated at 37°C. After 16–20 h, the turbidity level in each well was measured using a microplate reader.

Exponential phase cultures were treated with or without H₂O₂ at 5 mM or 10 mM for 25 min at 37°C with shaking. The cell pellets were washed three times with 50 mM phosphate buffer at pH 7.4 before being subjected to sonication. Crude protein concentration was then measured using the Bradford assay. Specific catalase activity (U/mg protein) was spectrophotometrically measured (A₄₂₀, light path = 1 cm). U was defined as one unit of catalase decomposing 1.0 μmole of H₂O₂ per minute at pH 7.0 at 25°C. 10 mM H₂O₂ in 50 mM phosphate buffer (pH 7.0) was used as the substrate.

Exponential phase cultures were treated with or without 5 mM H₂O₂ for 25 min at 37°C with shaking. 50 μg of crude protein was separated using 12.5% SDS-PAGE. The separated proteins were transferred onto the polyvinylidene difluoride (PVDF) membranes (GE Healthcare). The membranes were blocked with 5% skim milk in TBST buffer (20 mM Tris–HCl pH 7.5, 150 mM NaCl, and 0.05% Tween 20) for at least 1 h before incubating with antibody solution (horse radish peroxidase-conjugated anti-His tag antibody diluted in 3% skim milk and 1% BSA in TBST) for 1 h. The membranes were washed three times for 30 min in TBST buffer. The antibody reaction was observed using a chemiluminescent (ECL) detection reagent kit (GE Healthcare) according to the company’s recommendation. KatA and KatB were expressed from the pBB-katA-6xHis plasmid and the pBB-katB-6xHis plasmid, respectively. The pBB-katB 6XHis plasmid was previously described (Thongdee et al., 2019). pBB-katA 6XHis plasmid was constructed in a similar fashion to the pBB-katB 6XHis plasmid. The full-length katA gene was amplified from PA14 chromosomal DNA using PCR with the primer pair BT5117 (5′-GTTCTCCGTGGTCGCCC-3′) and BT6458 (5′-TTAGTGGTGGTGGTGGTGGTGGTCCAGCTTCAGGCCGAGG-3′), which contains hexahistidine (6XHis) tag. The katA-6XHis fragment was ligated into the BamHI and EcoICRI restriction sites of the pBBR1MCS-4 vector (Kovach et al., 1995) resulting in pBB-katA 6XHis plasmid for gene expression in PA14. The verified pBB-katA 6XHis plasmid was introduced into the wild-type and gidA mutant strains for Western blot analysis.

Exponential phase cultures grown in LB medium and treated with or without 5 mM H₂O₂ for 10 min at 37°C were used to determine the level of gene expression. A hot acid phenol extraction protocol was used for total RNA extraction (Chomczynski and Sacchi, 1987). Following DNase treatment, cDNA was synthesized using random hexamer primers and a RevertAid^™ M-MuLV Reverse Transcriptase Kit (Thermo Fisher Scientific). Semiquantitative real-time PCR was performed using KAPA SYBR FAST reagent (KAPA Biosystems) and a StepOnePlus^™ real-time PCR system (Thermo Fisher Scientific) according to the manufacturers’ instructions. The specific primers for the genes of interest were as follows: BT5637 (5′-TCTCCATGCGTTTCTACACC-3′) and BT5638 (5′-CGCATTGATGAAGCTGAAGG-3′) for katA; BT5639 (5′-CGACGCTTCGATTTCTTCTC-3′) and BT5640 (5′-TTCGGATCGAGGTTCTTCTG-3′) for katB; BT5641 (5′-CTACGGCGAGTTCCAGAAAG-3′) and BT5642 (5′-AGTGGATCTCGACGGTCTTG-3′) for ahpC; BT8259 (5′-CGCGAAGAAATACGACGCCG-3′) and BT8260 (5′-GTCGCTGAGGATGCCGTAGTA-3′) for fliA; BT8261 (5′-AAGGTCGAGGTCAGCGATGAC-3′) and BT8262 (5′-TCACCACGGTCTGTTCGTTGAT-3′) for fliD; BT8263 (5′-AACCCGCACCGTCTGATCC-3′) and BT8264 (5′-GCCTCGACCAGACGAGCGA-3′) for fliS; BT8253 (5′-TGGACAAGCAGACCGGCGAC-3′) and BT8254 (5′-TACATCCTTGGGCAGGCAGG-3′) for algD; BT8255 (5′-GGTGCCGCTCATCGTGCTCTA-3′) and BT8256 (5′-CGCGTTGCCCTGCATCTGGTA-3′) for algK; BT8257 (5′-CTACCCTGCCGAAGCTGGAT-3′) and BT8258 (5′-TCGCTGGACGAGGAGTTGGT-3′) for mucA; and BT2781 (5′-GCCCGCACAAGCGGTGGAG-3′) and BT2782 (5′-ACGTCATCCCCACCTTCCT-3′) for 16 s RNA gene. The 16S ribosomal RNA gene was used as the internal control to normalize gene expression.

Motility was examined on BM2-swimming plates (0.3% bacto agar, 62 mM potassium phosphate buffer pH 7, 7 mM (NH₄)₂SO₄, 2 mM MgSO₄, 10 μM FeSO₄, 0.4% glucose) and BM2-swarming plates (0.5% bacto agar, 62 mM potassium phosphate buffer pH 7, 0.1% casamino acid, 2 mM MgSO₄, 10 μM FeSO₄, 0.4% glucose; Overhage et al., 2007, 2008; Filloux and Ramos, 2014). Homogenized colonies were stabbed into the center of the BM2-swimming plate while 5 μl of the overnight cultures was inoculated on the surface of each BM2-swarming plate. The swimming and swarming plates were incubated overnight at 37°C. The relative motility of each strain was determined by comparing the diameters of the swim and swarm colonies.

Two methods were used to observe the biofilm formation, i.e., dye straining and scanning electron microscope (SEM). For the dye staining method, 100 μl of a diluted overnight culture (OD_600nm ~ 0.2) was added to a well of a 96-well plate and incubated overnight at 37°C. After overnight incubation, the plate was gently rinsed with distilled water before staining with 150 μl of 1% crystal violet for 10 min. Then, the crystal violet solution was removed. The plate was rinsed with distilled water again and allowed to air dry for at least 15 min. 100 μl of absolute ethanol was added into the well to dissolve the remaining crystal violet. Then, a spectrophotometer measurement was conducted at an optical density of 600 nm. The SEM method was performed using the method of Jin et al. with some adaptations (Jin et al., 2005). The biofilm of wild-type, gidA mutant, and complemented strains were developed on 1 × 1 cm plastic slide. The plastic slides were pretreated with concentrated sulfuric acid and 95% ethanol. 100 μl of cell suspensions (10⁷ CFU/ml) in PBS were dropped on the slide and were incubated for 90 min at 37°C to promote the bacterial cell adhesion. The plastic slides were washed twice with 200 μl PBS to remove nonadherent cells and filled with 100 μl Brain Heart Infusion (BHI) culture medium. The bacterial cultures were incubated at 37°C for 36 h and the BHI culture medium was replenished daily. The biofilm samples were fixed with 2.5% glutaraldehyde for 2 h and serially dehydrated with ethanol (70%, 95%, and absolute ethanol). The samples were kept in absolute ethanol until analysis. Before observing under the SEM, all samples were dried using critical-point drying in an EMITECH K850 dryer (Quorum Technologies Ltd.) at 39.1°C and 1,000 psi. The dried samples were then coated with gold, and the images were obtained using a TESCAN MIRA3 FEG-SEM (Field Emission Scanning Electron Microscope).

Alginate measurement was adapted from methods of Knutson and Jeanes (1968) and Zheng et al. (2016). Wild type, gidA mutant and the complement strain were spread onto BHI agar to promote alginate production. The cells were incubated at 37°C for 40–48 h, harvested and resuspended in 5 ml PBS (pH 7.4). Approximately 10⁷ cells were centrifugation at 12,000 g for 20 min to remove the cell pellet. Alginate was precipitated by adding equal volume of 2% cetylpyimidine to the cell free medium followed by centrifugation at 12,000 g for 15 min. The pellets were resuspended in 5 ml of 1 M NaCl. Then, 5 ml ice-cold isopropanol (−20°C) was added before centrifugation at 12,000 g for 15 min to collect the alginate pellet. Completely air-dried alginate pellets were then resuspended in 1 ml 0.9% NaCl. Measurement of alginate in the solutions using the standard alginic acid (0.01–1.0 mg/ml in 0.9% NaCl) was performed in 1.5 ml microtubes. 70 μl of alginate suspension was mixed using vortex with freshly prepared ice-cold 0.1 M H₃BO₃ in concentrated H₂SO₄. Then, 20 μl of 0.1% carbazole in ethanol was added. The solution was mixed and incubated at 55°C for 30 min. The absorbance at 530 nm was measured. The amount of alginate in each sample was calculated against alginate standards and normalized with OD₆₀₀ of the bacterial cell suspension.

The phagocytosis assay was modified from previously described protocols (Kuang et al., 2011; Charoenlap et al., 2012; Gao et al., 2016). Macrophage cells (RAW264.7) were activated with mouse IFN-gamma. Activated macrophage cells (RAW264.7) were seeded (2 × 10⁶ cells/well) in six-well tissue culture plates overnight. The cells were infected with mid-log phase P. aeruginosa cultures (MOI 1:10). The live and dead macrophage cells were counted at 0, 2, 4, and 8 h.

The virulence of P. aeruginosa was tested using Caenorhabditis elegans as the animal model (Tan et al., 1999). P. aeruginosa strains were grown in 5 ml of King’s broth at 37°C overnight with shaking. 10 μl of each overnight culture was spread onto high osmotic strength PGS agar (peptone-glucose-sorbital; 1% Bacto-peptone, 1% NaCl, 1% glucose, 0.15 M orbital, 1.7% Bacto agar) in a 60 × 15 mm plate. The plates were incubated overnight at 37°C and then subsequently transferred to 22°C for 8–12 h. Each bacterial lawn plate was seeded with 20–200 L4 stage larvae of C. elegans (Bristol-N2 wild-type strain from BP. Braeckman’s lab), incubated at room temperature, and scored for live and dead worms at time intervals over 3 days.

The exponential phase cells were snap-frozen in liquid nitrogen. Total protein was isolated using 0.5% SDS solution. Protein content was measured using the Lowry assay with bovine serum albumin as a standard (Lowry et al., 1951). 5 μg of protein sample was subjected to in-solution digestion. All samples were completely dissolved in 10 mM ammonium bicarbonate (AMBIC). Disulfide bonds were reduced using 5 mM dithiothreitol (DTT) in 10 mM AMBIC at 60°C for 1 h. Alkylation of sulfhydryl groups was done by incubating them with 15 mM iodoacetamide (IAA) in 10 mM AMBIC at room temperature for 45 min in the dark. Then, the samples were mixed with 50 ng/μL of sequencing grade trypsin (1:20 ratio; Promega, United States) and incubated at 37°C overnight. The digested samples were dried and redissolved with 0.1% formic acid before injection into an Ultimate3000 Nano/Capillary LC System (Thermo Scientific, United Kingdom) coupled to an HCTUltra LC–MS system (Bruker Daltonics Ltd.; Hamburg, Germany) equipped with a nano-captive spray ion source. 5 μl of peptide digest was enriched on a μ-Precolumn 300 μm i.d. × 5 mm C18 Pepmap 100, 5 μm, 100 A (Thermo Scientific, United Kingdom), separated on a 75 μm i.d. × 15 cm and packed with Acclaim PepMap RSLC C18, 2 μm, 100 Å, nanoViper (Thermo Scientific, United Kingdom). The C18 column was enclosed in a thermostatted column oven set at 60°C. Solvents A and B containing 0.1% formic acid in water and 0.1% formic acid in 80% acetonitrile, respectively, were used as mobile phases. A gradient of 5–55% solvent B was used to elute the peptides at a constant flow rate of 300 nl/min for 30 min. Electrospray ionization was carried out at 1.6 kV using CaptiveSpray. Nitrogen was used as a drying gas at a flow rate of about 50 l/h. Collision-induced dissociation (CID) product ion mass spectra were obtained using nitrogen gas as the collision gas. Mass spectra (MS) and MS/MS spectra were obtained in the positive-ion mode at 2 Hz over a range of 150–2,200 m/z. The collision energy was adjusted to 10 eV as a function of the m/z value. Liquid chromatography-mass spectrometry (LC–MS) analysis of each sample was done in triplicate. The raw MS/MS spectra data are available in ProteomeXchange: JPST001884 and PXD037219.¹

For protein quantitation, DeCyder MS Differential Analysis software (DeCyderMS, GE Healthcare) was used (Johansson et al., 2006). The analyzed MS/MS data from DeCyderMS were submitted for database searches using Mascot software (Matrix Science, London, United Kingdom). The data were searched against the NCBI database for protein identification. Database interrogation was performed for taxonomy (Pseudomonas aeruginosa); enzyme (trypsin); variable modifications (carbamidomethyl, oxidation of methionine residues); mass values (monoisotopic); protein mass (unrestricted); peptide mass tolerance (1.2 Da); fragment mass tolerance (±0.6 Da); peptide charge state (1+, 2+ and 3+); and missed cleavages. The level of protein in each sample was expressed as a log2 value. Gene ontology annotation was performed using Panther (Mi et al., 2019).

P. aeruginosa GidA displays 70% identity to the E. coli MnmE protein (Yim et al., 2006). To assess whether P. aeruginosa GidA is responsible for the formation of cmnm⁵(s²)U or nm⁵(s²)U in the mnm⁵(s²)U biosynthetic pathway (Figure 1A), tRNA modification profiles of the wild type, a gidA insertional inactivation mutant, and a complemented strain were analyzed using LC–MS/MS. The result of this study showed that the inactivation of gidA specifically altered the level of tRNA modifications linked to the mnm⁵(s²)U biosynthetic pathway (Figure 1B). Among the 24 modifications detected in P. aeruginosa’s tRNAs, only the s²U, cmnm⁵U, and mnm⁵U levels were found to change due to the loss of gidA (Table 2; Figure 1B). The s²U levels increased around fourfold, while the cmnm⁵U and mnm⁵U levels significantly reduced in the gidA mutant compared to those in the wild-type and complemented strains (Figure 1B; Table 2). These findings denote the crucial function of gidA in the biosynthesis of 5–carboxymethylaminomethyl uridine (cmnm⁵U) and 5-methylaminomethyl uridine (mnm⁵U) tRNA modifications in P. aeruginosa. In addition, hypomodifications of cmnm⁵U and mnm⁵U also resulted from the disruption of P. aeruginosa trmE (Supplementary Figure S1), indicating that the formation of cmnm⁵U and mnm⁵U is dependent on gidA as well as trmE.

Cell response to oxidative stress is an important mechanism of Pseudomonas for counteracting the toxic effects of oxidizing agents such as hydrogen peroxide, superoxide, and hypochlorous acid, which are produced by the host immune system during infection. In our H₂O₂ sensitive phenotype screening from the mutant library of Pseudomonas aeruginosa, gidA mutant is one of mutant strains showing such phenotype (unpublished data). To further determine the role of gidA in H₂O₂ stress response, the sensitivity of a gidA mutant strain to H₂O₂ was tested using a minimal inhibition concentration (MIC) assay. The results revealed that the gidA mutant (78 nM) was more susceptible to H₂O₂ than the wild-type and complemented strains (156 nM). The same result was found in the experiment involving the use of a plate sensitivity assay (data not shown). Therefore, it is possible that gidA is necessary for full resistance against H₂O₂.

The detoxification of H₂O₂ into harmless H₂O and O₂ molecules is primarily catalyzed by catalase enzymes such as KatA and KatB (Zamocky et al., 2010). Disruption of gidA may affect the capability of P. aeruginosa to detoxify H₂O₂. To test this hypothesis, the total catalase activity of the gidA mutant, the wild-type, and complemented strains were compared. Under unexposed conditions, the total catalase activities were not statistically different (143.3 ± 11.1, 121.6 ± 23.0, and 109.3 ± 5.8 units/mg protein, respectively; Figure 2A). Upon exposure to 5 mM H₂O₂, the total catalase activities of all strains increased from the basal level (Figure 2A). However, the catalase activity of the gidA mutant (462.9 ± 13.4 units/mg protein) was significantly lower than those of the wild-type (801.1 ± 72.1 units/mg protein) and complemented strains (690.6 ± 45.2 units/mg protein). This observation shows that the disruption of gidA has a negative effect on catalase-mediated H₂O₂ detoxification in P. aeruginosa. The reduced catalase activity of the gidA mutant may not be sufficient for cell survival during H₂O₂ treatment, resulting in a higher H₂O₂ sensitivity of the gidA mutant than those of the wild-type and complemented strains.

P. aeruginosa possesses three catalase enzymes; KatA, KatB, and KatE. To further explain how the loss of gidA affects catalase activity, the impact of gidA disruption on catalase gene regulation was determined. The regulation of P. aeruginosa katA and katB, which encode major catalases expressed during exponential phase growth, was investigated. The results of a semiquantitative real-time PCR assay demonstrated that the transcript levels of katA and katB increased in response to H₂O₂ exposure in all strains, as expected, and that the transcript levels were not statistically different. Moreover, no significant differences in transcript levels were observed between the strains in the absence of H₂O₂ (Figures 2B,C).

Given that total catalase activity was significantly reduced due to the absence of gidA (Figure 2A), it is thus highly possible that gidA exerts its effect on katA and katB gene expression at the posttranscriptional level. Western blot analysis was used to investigate the impact of gidA on the protein level of KatA and KatB during exposure to 5 mM H₂O₂. We investigated the level of KatA and KatB expressed from the plasmid copy, where gene is expressed using the lac promoter and ribosome binding site from the plasmid. The result demonstrated that KatA protein expression was not detectable in the gidA inactivation mutant compared to the wild-type strain under both normal and 5 mM H₂O₂ treatment conditions (Figure 2D). In addition, the level of KatB protein was markedly decreased in the gidA mutant compared with that in the wild type (Figure 2D). The decreased level of both KatA and KatB protein, but not katA and katB mRNA level, indicates that gidA exerts a posttranscriptional effect on katA and katB expression.

It is known that gidA is associated with virulence mechanisms and the expression of a wide range of virulence genes in bacteria, including quorum sensing and stress responses (Gupta et al., 2009; Shippy and Fadl, 2014; Zhang et al., 2014; Gao et al., 2016). In P. aeruginosa PA14, gidA is involved in the Rhl-quorum sensing system affecting the bacterium’s growth and proteolytic activities (Gupta et al., 2009). Therefore, genes associated with bacterial motility and biofilm formation assays were performed in this study. Figures 3A,B present the results for both motility activities, i.e., swimming and swarming. With respect to swimming motility (Figure 3A), the wild-type and complemented strains both presented swimming behaviors on the semisolid swimming plates by 5.7 ± 0.1 cm and 5.1 ± 0.1 cm, respectively, compared to 3.7 ± 0.1 cm in the gidA inactivation mutant indicating a decreased in the swimming ability of the gidA mutant strain. In addition, the swarming plates revealed that the loss of gidA reduced the swarming ability of P. aeruginosa PA14 in covering the semisolid plate (Figure 3B). Furthermore, the lack of gidA was also found to affect biofilm production in P. aeruginosa PA14. Biofilm formation in the gidA mutant significantly decreased in relation to biofilm formation in the wild-type and complemented strains as determined by the dye staining method (Figure 3C). Biofilm formation was also observed using scanning electron microscope. The results demonstrated that gidA mutant was unable to form the confined biofilm compared to the wild-type and complemented strains (Figure 3D). In addition, alginate concentration was measured as it is an important component of biofilm (Gardner et al., 1987; Tatnell et al., 1994; Keiski et al., 2010). Consistent with the biofilm formation data, the alginate production was lower in the gidA mutant by 21.2 ± 1.8 folds. Taken together, these observations indicate that gidA contributes to various virulence genes in P. aeruginosa PA14, alter the flagellum-mediated motility activities (swimming and swarming) and biofilm formation.

To further investigate the changes in the protein profile of the gidA mutant in comparison to the wild-type strain, proteomic analysis using liquid chromatography–tandem mass spectrometry was performed. A total of 1,106 protein identities were detected, 377 of which were differentially produced between the gidA mutant and wild-type strains (Supplementary Figure S2). Among these differentially produced proteins, 267 were upregulated and 110 were downregulated in the gidA mutant relative to the wild type. The abundance levels of several proteins involved in bacterial pathogenesis were altered (Table 3). A lower abundance of some components of the flagellar system, including FliD, FliS, and FliA, was seen in the gidA mutant than in the wild-type strain (Table 3). In addition, the levels of some proteins in the alginic acid biological process, which plays a role in biofilm formation, were found to decrease in the gidA mutant when compared with the levels in the wild-type strain; these included GDP-mannose dehydrogenase (AlgD) and multiple tetratricopeptide-like (TPR-like) repeats lipoprotein (AlgK). The level of MucA, an anti-sigma factor involved in the regulation of alginic acid biosynthesis, was higher in the gidA mutant than in the wild type. These data support the defects in mobility and biofilm formation observed in the gidA mutant.

The mRNA levels of fliA, fliD, and fliS, which are involved in flagella formation, were investigated using semiquantitative RT-PCR analysis (Figure 4). The results revealed reduced mRNA levels in the gidA mutant strain compared to the wild-type and complemented strains (Figures 4A–C). With respect to genes involved in biofilm formation, the mRNA levels of the algD and algK genes were slightly lower in the gidA mutant strain than in the wild-type strain, while the mRNA level of the mucA gene remained unchanged in the gidA mutant but not in the wild-type strain (Figures 4D–F). The mRNA levels of the algD and algK genes changed in the same direction as the corresponding protein levels, indicating that such changes might not involve regulation at the translation level of those genes. The unchanged mucA mRNA level and upregulated protein level indicate the effect of gidA on mucA expression at the posttranscriptional level.

In this study, the inactivation of gidA affected the oxidative stress response, motility, and biofilm formation capability of P. aeruginosa PA14 (Figures 2, 3), suggesting that the loss of gidA attenuates its virulence. Therefore, in vitro phagocytosis assays were performed using a mouse macrophage cell line (RAW264.7). The cells were infected with the wild-type, gidA mutant, and complemented strains of P. aeruginosa PA14 at a ratio of 10 bacterial cells per macrophage (MOI 1:10). Macrophages infected with the gidA inactivation mutant showed a significantly high survival rate compared to those infected with either the wild-type or complemented strains (Figure 5A). Further, the ability of the wild type, gidA inactivation mutant, and complemented strain to kill Caenorhabditis elegans was evaluated in vivo using a slow-killing assay. The results mirrored the macrophage assays. P. aeruginosa was clearly attenuated when gidA was disrupted. After 72 h of infection, the C. elegans fed with the gidA mutant had a significantly lower % of death (* p ≤ 0.05) than those fed with either the wild type or complemented strain (Figure 5B). Overall, these results confirmed that the inactivation of gidA attenuates the virulence of P. aeruginosa PA14.