The bacterial strains, plasmids and primers used in this study are listed in Supplementary Tables 1 , 2 . Bacterial cells were grown in Luria–Bertani (LB) medium (5 g/L NaCl, 5 g/L yeast extract, and 10 g/L tryptone) or on LB agar plates (LB medium containing 15 g/L agar) at 37 °C. To maintain plasmids, appropriate antibiotics were supplemented into the medium at the following concentrations: for P. aeruginosa, tetracycline at 50 μg/mL and carbenicillin at 150 μg/mL; for E. coli, tetracycline at 10 μg/mL, kanamycin at 25 µg/mL, and ampicillin at 100 μg/mL. When needed, IPTG (isopropyl β-D-1-thiogalactopyranoside) was added to the medium at a final concentration of 1 mM. For inducing T3SS of P. aeruginosa, EGTA was added to the medium at a final concentration of 5 mM.

For complementation of the mvaT gene, the mvaT gene was amplified by PCR using specific primers ( Supplementary Table 2 ) with PAK genomic DNA as the template. The resulting PCR product was digested with BamHI-HindIII and then cloned into pMMB, generating pMMB-mvaT. P _exsC mut-lacZ (EcoRI-BamHI), pMMB-exsA-His (EcoRI-HindIII) and pET28a-mvaT (NcoI-XhoI) were constructed with similar strategies. To delete the mvaT gene, two DNA fragments upstream and downstream of the mvaT gene were amplified by PCR using the specific primer pairs mvaT-UF/UR and mvaT-DF/DR ( Supplementary Table 2 ), digested with appropriate restriction enzymes and directionally cloned into pEX18Tc, resulting in the deletion construct pEX18Tc-mvaT. To obtain the mvaT deletion mutant in the PAK and PAKΔexsA strains, pEX18Tc-mvaT was transferred into the corresponding bacterial cell via conjugation, and gene deletion was performed with a SacB-based strategy as previously described (Hoang et al., 1998).

The DNA pull-down assay was carried out as described previously with minor modifications (Dolan et al., 2020; Modrzejewska et al., 2021). A biotinylated DNA fragment containing the P _exsC promoter was amplified by PCR using primers P _exsC -biot-F/R ( Supplementary Table 2 ) with PAK genomic DNA as a template. The amplified DNA segment was purified and incubated with 100 μL of streptavidin magnetic beads (Streptavidin Mag Sepharose, Cytiva) at room temperature for 20 min. Then, the magnetic beads were washed twice with TBS buffer (50 mM Tris, 150 mM NaCl, pH 7.5) to remove the unbound DNA. In parallel, PAK cells from 200 mL cultures at the exponential phase were harvested by centrifugation at 8,000 × g for 10 min, resuspended in 3 mL TBS buffer and lysed by sonication on ice. After centrifugation at 12,000 × g for 10 min at 4°C, the supernatant was mixed with the immobilized DNA fragments on the magnetic beads and incubated at 4°C for 1 h. After that, the beads were washed three times with TBS buffer to remove nonspecifically bound proteins, and the DNA-bound proteins were released by treatment with increasing concentrations of NaCl buffer (0.2 M, 0.5 M, 1 M). PAKΔexsA/pMMB-exsA-His lysates incubated with P _exsC -coupled magnetic beads served as a positive control, while PAK cell lysates incubated with DNA-uncoupled magnetic beads served as a negative control. The eluates from the ExsA-His-overproducing strain were subjected to Western blot analysis using anti-His antibodies to confirm the predicted binding of the ExsA-His protein to the P _exsC promoter, validating the pull-down procedure. The eluates from the PAK strain with or without DNA-coupled beads were examined by SDS-PAGE and stained with Coomassie blue. The protein band of interest (the band observed in DNA-coupled beads, but not in the negative control) was excised from SDS-PAGE gel (from 0.2 M NaCl buffer elution) and identified by mass spectrometry analysis.

The β-galactosidase assay was carried out as described previously with minor modifications (Miller, 1972). Overnight cultures of bacteria were 50-fold diluted into fresh LB medium with 0 (for E. coli or P. aeruginosa T3SS non-inducing condition) or 5 mM EGTA (for P. aeruginosa T3SS inducing condition) and grown at 37°C with continuous agitation at 200 rpm. When the OD₆₀₀ reached 1.0, 500 μL bacterial cells were collected by centrifugation and resuspended in 1.5 mL Z-buffer (60 mM Na₂HPO₄, 60 mM NaH₂PO₄, 50 mM β-mercaptoethanol, 10 mM KCl, 1 mM MgSO₄). One milliliter suspension was used to measure the OD₆₀₀, and 10 μL chloroform and 10 μL 0.1% SDS were added to the remaining 500 μL suspensions, followed by vortexing for 10 s. After that, 100 μL ONPG (4 mg/mL) was added to the reaction mixture and incubated at 37°C. The reaction was stopped by the addition of 500 μL of 1 M Na₂CO₃. The reaction time was recorded, and OD₄₂₀ was measured after centrifugation at 16,000 × g for 5 min. β-galactosidase activity (Miller units) was calculated as 1000×OD₄₂₀/T_min/0.5/OD₆₀₀; T_min represents reaction time (minutes).

Overnight bacterial cultures were diluted 50-fold into 3 mL fresh LB medium containing 0 or 5 mM EGTA and then grown to an OD₆₀₀ of 1.0 with continuous shaking at 200 rpm. Equivalent numbers of bacteria were harvested by centrifugation at 12,000 × g for 3 min. Samples from bacterial cells and supernatant were mixed with loading buffer, boiled for 10 min at 99 °C, and then separated on a 12% SDS-PAGE gel. After transfer onto a polyvinylidene difluoride (PVDF) membrane, the proteins were probed with the primary antibody against ExoS or RpoA (RNAP, Abcam) for 1 h at room temperature and then with the corresponding horseradish peroxidase (HRP)-conjugated secondary antibody, anti-rabbit IgG (Promega) or anti-mouse IgG (Promega), for 1 h at room temperature. All antibodies were 2000-fold diluted into PBST buffer with 5% non-fat milk. The signals were detected using an ECL Plus kit (Millipore) and visualized in a Bio-Rad molecular imager (ChemiDoc XRS+).

To express the recombinant C-terminal His-tagged MvaT, the E. coli strain BL21 (DE3) containing pET28a-mvaT was cultured at 37°C to an OD₆₀₀ of 0.6. Overexpression of the recombinant MvaT-His protein was induced by the addition of 1 mM IPTG and further grown overnight at 16°C. Harvested cells were resuspended in lysis buffer (50 mM sodium phosphate, 0.3 M NaCl, pH 8.0) and lysed by sonication on ice. After centrifugation at 12,000 × g for 10 min, the supernatant was incubated with Ni-NTA resin (Qiagen) at 4°C for 1 h. Nonspecific binding proteins were washed away with lysis buffer containing 20 mM imidazole, and MvaT-His was eluted with the lysis buffer containing 200 mM imidazole. The purified protein was examined by SDS-PAGE and stained with Coomassie blue.

EMSA was carried out following a previous description with minor modifications (Williams McMackin et al., 2019b). Briefly, the P _exsC DNA fragment was amplified by PCR using specific primers ( Supplementary Table 2 ). Fifty or 20 nanogram DNA probe was incubated with increasing concentrations of purified MvaT-His protein in a 20-μL reaction (20 mM Tris, 100 mM KCl, 1 mM DTT, 10% glycerol, pH 7.5) on ice for 30 min. Each sample was loaded onto an 8% or 12% native polyacrylamide gel in 0.5 × TBE (Tris-borate-EDTA) buffer (44.5 mM Tris base, 44.5 mM boric acid, 1 mM EDTA, pH 8.0) and electrophoresed on ice at 10 mA for 1 h. The gel was then stained in 0.5 × TBE buffer containing 0.5 μg/mL ethidium bromide (EB) and imaged with a molecular imager ChemiDoc XRS+ (Bio-Rad, CA, USA).

To identify proteins that can modulate exsA transcription via the P _exsC promoter and consequently T3SS expression, we used a biotin-labeled P _exsC promoter fragment as bait in a DNA pull-down assay using P. aeruginosa PAK cell lysates. Proteins copurified with the P _exsC fragment by streptavidin-coated beads were separated by SDS-PAGE and stained with Coomassie blue. As shown in Figure 1 , a specific protein band was detected in the samples in the presence but not in the absence of the P _exsC fragment. As a positive control, lysates of bacterial cells overproducing ExsA (PAKΔexsA/pMMB-exsA-His) were pulled down by the P _exsC fragment and subjected to Western blot against His antibody. ExsA-His was easily detected ( Supplementary Figure 1 ), validating our approach. Mass spectrometric analysis of the protein band collection revealed a mixture of 65 proteins ( Supplementary Table 3 ). Among them, MvaT, the histone-like nucleoid-structuring DNA-binding protein, was detected. Interestingly, a previous study showed that MvaT regulates the T3SS via direct silencing of the P _exsA promoter (Williams McMackin et al., 2019b).

To determine whether MvaT, identified in the pull-down analysis, can indeed bind to the P _exsC promoter, we carried out electrophoretic mobility shift assays (EMSAs) using purified MvaT-His and DNA fragments corresponding to the P _exsC promoter region. The DNA fragments corresponding to P _exsC , but not the negative control P _algD , were shifted upon incubation with MvaT-his ( Figures 2A, B ), indicating that MvaT directly binds to the promoter of P _exsC in P. aeruginosa.

To further examine the influence of MvaT on the activity of the P _exsC promoter, the mvaT gene was expressed under the control of P_T7 in vector pET28a and introduced into DH5α and BL21 harboring the P _exsC -lacZ transcriptional fusion reporter plasmid. Measurement of β-galactosidase activity showed that expression of MvaT significantly repressed P _exsC activity in both DH5α and BL21 ( Figures 2C, D ). However, the β-galactosidase activity of P _exsC -lacZ in BL21 is much higher than that in DH5α, which may be due to the absence of OmpT protease in BL21. These results suggested that MvaT acted directly as a repressor of the P _exsC promoter. Furthermore, we wanted to investigate whether MvaT modulates P _exsC activity in P. aeruginosa. A previous study demonstrated that MvaT binds to and represses the P _exsA promoter (Williams McMackin et al., 2019b). To exclude the effect of MvaT on P _exsC activity derived from the altered ExsA protein, P _exsC -lacZ and pMMB-mvaT were co-introduced into the PAKΔexsA strain. Consistent with the P _exsC -lacZ reporter results in E. coli, expression of MvaT resulted in significant decreases in β-galactosidase activity under both T3SS inducing and noninducing conditions ( Figure 2E ). All these results demonstrated that MvaT represses the P _exsC promoter activity directly.

To determine the influence of MvaT on T3SS gene expression, we generated a mvaT clear deletion mutant and monitored the ExoS amounts using Western blot assay. Strains were cultured under T3SS-inducing (with 5 mM EGTA) and noninducing (without EGTA) conditions for T3SS gene expression. As shown in Figure 3 , the expression and secretion of ExoS were highly induced under inducing conditions in both the wild-type PAK and mvaT mutant strains. Deletion of mvaT resulted in increased expression and secretion of ExoS compared to the wild-type PAK strain, and complementation with a mvaT gene recovered the expression and secretion of ExoS in the mvaT mutant background ( Figure 3A ).

To examine the effect of MvaT on ExoS expression and secretion independent of ExsA, Western blot was carried out to compare the ExoS amounts between PAKΔexsA and PAKΔexsAΔmvaT. However, in the absence of exsA, ExoS was undetectable in both PAK and PAKΔmvaT ( Figure 3B ). Since the β-galactosidase activity of P _exsC -lacZ was detectable in the absence of exsA in PAK ( Figure 2E ), we further examined and compared the P _exsC promoter activities. As shown in Figure 3C , mvaT deletion resulted in increased β-galactosidase activity in PAKΔexsA/P _exsC-lacZ under both T3SS-inducing and noninducing conditions. In addition, the PAKΔmvaT strain displayed a much higher β-galactosidase activity than the PAKΔexsAΔmvaT strain ( Figure 3C ), hinting at the master activator role of ExsA and a repressive role of MvaT on P _exsA promoter (Williams McMackin et al., 2019b). Overall, these data suggest that MvaT represses P _exsC transcription independent of ExsA. Furthermore, we examined the promoter activity of another T3SS effector ExoT with P _exoT -lacZ transcriptional fusion reporter plasmid (Ha and Jin, 2001). Consistent with the P _exsC promoter activities, absence of mvaT resulted in increased β-galactosidase activities of P _exoT -lacZ in the PAKΔexsA strain under both T3SS inducing and non-inducing conditions ( Figure 3D ).

ExsA binds to upstream of the transcriptional start site and facilitates transcription by recruiting RNAP-σ⁷⁰ to the P _exsC promoter (Vakulskas et al., 2009). MvaT represses P _exsC promoter activity independent of ExsA. Therefore, we wanted to determine the binding site of MvaT on the P _exsC promoter. A variety of DNA fragments within P _exsC ( Figure 4A ) were PCR amplified and used as DNA probes in the EMSA. As shown in Figure 4B , MvaT binds to the DNA fragments containing -429 to -380 bp relative to the transcription start site of the exsC gene. To examine the function of the MvaT binding site in the MvaT-mediated repression of exsC gene, we constructed an additional P _{exsC mut}-lacZ transcriptional fusion, where a 397-bp fragment upstream of the exsC coding region without the MvaT binding site was fused to the promoterless lacZ gene. As shown in Figure 4C , removal of the MvaT binding site resulted in no difference of the β-galactosidase activity between PAKΔexsAΔmvaT and PAKΔexsA strain. This result further supports that MvaT binds to and represses P _exsC independent of the master activator ExsA.