Sulfane Sulfur Regulates LasR-Mediated Quorum Sensing and Virulence in Pseudomonas aeruginosa PAO1

Sulfane sulfur, such as inorganic and organic polysulfide (HSn− and RSn−, n > 2), is a common cellular component, produced either from hydrogen sulfide oxidation or cysteine metabolism. In Pseudomonas aeruginosa PAO1, LasR is a quorum sensing master regulator. After binding its autoinducer, LasR binds to its target DNA to activate the transcription of a suite of genes, including virulence factors. Herein, we report that the production of hydrogen sulfide and sulfane sulfur were positively correlated in P. aeruginosa PAO1, and sulfane sulfur was able to modify LasR, which generated Cys188 persulfide and trisulfide and produced a pentasulfur link between Cys201 and Cys203. The modifications did not affect LasR binding to its target DNA site, but made it several-fold more effective than unmodified LasR in activating transcription in both in vitro and in vivo assays. On the contrary, H2O2 inactivates LasR via producing a disulfide bond between Cys201 and Cys203. P. aeruginosa PAO1 had a high cellular sulfane sulfur and high LasR activity in the mid log phase and early stationary phase, but a low sulfane sulfur and low LasR activity in the declination phase. Thus, sulfane sulfur is a new signaling factor in the bacterium, adding another level of control over LasR-mediated quorum sensing and turning down the activity in old cells.


Introduction
Hydrogen sulfide (H 2 S) has been proposed as a new gaseous signaling molecule, mediating various biological functions in mammals, including humans [1][2][3][4][5]. Owing to its signaling role in eukaryotes, H 2 S was suggested as a "clandestine microbial messenger" in 2006 [6]. Since then, several bacterial transcription factors, including FisR [7], SqrR [8], CstR [9], and CsoR [10], have been identified that indirectly respond to H 2 S, activating sulfur-oxidizing genes. First, sulfide:quinone oxidoreductase (SQR) converts H 2 S to sulfane sulfur, including inorganic and organic polysulfide (HS n − and RS n − , n ≥ 2) and elemental sulfur (S 0 ) [11,12]. Sulfane sulfur is then sensed by these transcription factors to turn on the transcription of sulfur-oxidizing genes, including pdo coding for persulfide dioxygenase (PDO). A synthetic gene circuit that combines SQR and CstR allows the host Escherichia coli to oxidize self-produced H 2 S to sulfane sulfur and then sense the latter, resulting in gene regulation in a manner similar to quorum sensing (QS) [13]. These examples highlight how H 2 S is converted by SQR to sulfane sulfur that is then sensed by these gene regulators.
Other gene regulators that are not involved in sulfur metabolism may also sense sulfane sulfur. OxyR, the H 2 O 2 -response gene regulator, senses cellular sulfane sulfur to turn on many genes, including those coding for the removal of excessive sulfane sulfur in E. coli [14]. MexR senses sulfane sulfur, which is maximally accumulated in late log phase by

Gene Knockout and Complementation
The primers used for Pacbs, Pacse, Pamst, PacysI, Pasqr1, Pasqr2, Papdo, and PalasR inactivation are listed in Table S2. The deletions in P. aeruginosa PAO1 were performed according to a published method [44,45]. Briefly, about 1000-bp fragments upstream and downstream of the target gene were amplified from the PAO1 genomic DNA via PCR, linked, and cloned into pK18mobsacB tet at the EcoRI site. The resulting plasmid was first transformed into E. coli and then transferred into P. aeruginosa PAO1 via conjugation. The integration into the chromosome of P. aeruginosa PAO1 by a homologous crossover was selected on agar plates of a chemically defined medium, with sodium gluconate as the sole carbon source containing tetracycline. The selection of the double crossover with 12% sucrose led to gene-deletion. For multiple deletions, the process was repeated. The deletion mutants, including Pa∆cbs∆cse∆mst∆cysI∆sqr1∆sqr2∆pdo (Pa7K), Pa∆cbs∆cse∆mst∆cysI (Pa∆H 2 S), Pa∆sqr1∆sqr2∆pdo (Pa3K), and Pa∆lasR, were confirmed by using colony PCR and DNA sequencing. For complementation, the target genes were amplified by PCR and cloned into linearized pBBR1MCS5 by using an In-Fusion HD cloning kit (Clontech, Mountain View, CA, USA). The resulting plasmids were then transferred into the PAO1 strain via electroporation.

Detection of H 2 S and Sulfane Sulfur
The production of H 2 S by PAO1 and its related mutant strains was detected with a paper strip with lead-acetate [Pb(Ac) 2 ] (Shanghai, China) and a monobromobimane (mBBr) method [46]. Briefly, single colonies were innoculated in 2 mL of LB medium in a 15-mL glass tube, and a paper strip with lead-acetate was affixed at the top of the tube with a rubber stopper. The paper strip was examined and photographed to detect any lead(II)-sulfide black precipitates, indicting the production of H 2 S. H 2 S in the liquid culture was detected using the mBBr method [46]. Briefly, 5 µL of 25 mM mBBr was reacted with 50 µL of a sample at room temperature for 30 min in the dark, and an equal volume of 10% acetic acid in acetonitrile was added. The samples were centrifuged at 12,500× g for 2 min and the supernatant was analyzed using HPLC (LC-10AT, Shimadzu) with a fluorescence detector, as reported [11].
SSP4 (sulfane sulfur probe 4, 3 ,6 -di(O-thiosalicyl)fluorecein) that reacts with sulfane sulfur to become fluorescent was used to check the relative levels of cellular sulfane sulfur in P. aeruginosa cells [47]. The cells were collected, washed with phosphate buffer saline, and resuspended in phosphate buffer saline at an OD 600nm of 1. Then 10 µM SSP4 and 0.5 mM hexadecyltrimethylammonium bromide (CTAB) were added to the sample, which was incubated in the dark at 30 • C, with shaking for 20 min. Cells were harvested by centrifugation and washed twice with phosphate buffer saline. The fluorescence of the resuspended cells (OD 600nm = 1) was detected by using a Synergy H1 microplate reader with excitation of 482 nm and emission of 515 nm. We were unable to quantify sulfane sulfur with SSP4. For quantification, cellular sulfane sulfur in P. aeruginosa PAO1 at different growth stages was reacted with sulfite to produce thiosulfate, which was then quantified according to a reported method [17]. Briefly, samples were mixed with the reaction buffer with sulfite to convert sulfane sulfur to thiosulfate by incubating at 95 • C for 20 min; buffer without sulfite was used as the control. The produced thiosulfate was detected by using the mBBr method [46], as briefly described above.

Rhamnolipids Production Measurement
Rhamnolipid production was measured by following a reported method [48]. Briefly, 5 µL of an overnight culture was placed onto a M8 minimal agar plate supplemented with 0.0005% (m/v) methylene blue and 0.02% (m/v) CTAB; the plate was incubated at 37 • C for 48 h before checking for the clear zone around a colony. For the H 2 S supplemental experiment, 10 drops of 1 mM NaHS in H 2 O were separately dropped on the plate cover. The plates were inverted, and H 2 S entered the medium via evaporation.

Pyocyanin Quantitation Assay
Pyocyanin concentration was determined by using a reported method [42]. Briefly, strains were grown at 37 • C in 5 mL of PB medium to stationary phase. The culture supernatants were extracted with 3 mL of chloroform, which was mixed with 1 mL of 0.2 M HCl to develop a pink to deep red color in the organic phase for measurements at 520 nm. The concentration was calculated and normalized to the cell density (OD 600nm ).

Lettuce Leaf Model of Infection
A lettuce leaf virulence assay was performed, as described [48,49]. Briefly, P. aeruginosa strains were grown in PB medium at 37 • C overnight with shaking (200 rpm); the cells were harvested, washed, and resuspended in sterile 10 mM MgSO 4 to a bacterial density of 1 × 10 9 cells/mL. Romaine lettuce leaves were washed with sterile distilled H 2 O with 0.1% bleach and then inoculated with the strains on the midribs of the leaves, which were placed in containers underlined with wet filter paper containing 10 mM MgSO 4 . The inoculated leaves were kept in a growth chamber at 37 • C. Symptoms were monitored daily. As a control, lettuce leaves were inoculated with 10 mM MgSO 4 .

Transcriptomic Analysis of PAO1 and Pa∆H 2 S
For transcriptome sequencing (RNA-seq) analysis, PAO1 and Pa∆H 2 S were cultured in PB medium at 37 • C with shaking at 200 rpm to late log phase. Cells were centrifuged, and the pellets were frozen in liquid nitrogen, and shipped on dry ice to Beijing Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). The subsequent analysis was made by the company. Total RNA was extracted by using a TRIzol TM RNA Purification Kit (12183555, Invitrogen, Waltham, MA USA). Then, 3 µg of total RNA was treated with a Ribo-Zero rRNA Removal Kit (MRZMB 126, Epicentre Biotechnologies). First-strand cDNA was synthesized by using random hexamer primers and M-MuLV Reverse Transcriptase, and second-strand cDNA synthesis was subsequently performed by using DNA polymerase I. RNase H was used to remove RNA. NEBNext index adaptor oligonucleotides were ligated to the cDNA fragments. The cDNA fragments of 150-200 bp in length were purified and amplified via PCR with universal PCR primer and index (X) primer. The library was quantified using an Agilent High Sensitivity DNA assay on an Agilent 267 Bioanalyzer 2100 system and sequenced on the Illumina Hiseq 2500 platform. Trimmed sequence reads were aligned to the P. aeruginosa PAO1 genome sequence using Bowtie2-2.2.3. Gene expression was analyzed with the reads per kilobase of coding sequence per million reads (RPKM) algorithm. Differential expression analysis of the two strains was performed using the DESeq R package (1.18.0). Genes with a change fold >2 and a p-value <0.05 were considered as significantly differentially expressed.

Reporter Plasmids Construction and Fluorescence Assays
Red fluorescence protein (mkate) reporter assays-E. coli BL21(DE3) cells were transformed with a plasmid derived from Ptrc99a, containing lasR, the promoter of rhlR, and mkate (Ptrc-P lacI -lasR-P rhlR -mkate). Site-directed mutagenesis to convert lasR Cys 79 , Cys 188 , Cys 201 , and Cys 203 to serine was achieved according to a reported method [50]. All primers used in the experiments are listed in Table S2. Using this reporter strain, LasR activity was quantified. Starting with a 1/1000-fold dilution of an overnight culture, the reporter strain was grown in LB medium supplemented with 20 µM 3O-C 12 -HSL and ampicillin (100 µg/mL) at 37 • C with shaking. When cultures were grown to an OD 600nm of 2, 300 µM, polysulfide was added. After incubating at 37 • C for an additional 2 h to produce mkate, 0.2 mL of the cells was transferred to a 96-well plate and the mkate fluorescence was measured by using a SynergyH1 microplate reader. The excitation wavelength was set at 588 nm and the emission wavelength was set at 633 nm.

Protein Expression and Purification
E. coli BL21(DE3) carrying the expression plasmid PET30-LasR was grown in LB to an OD 600nm of 0.4-0.6, IPTG was added to 0.1 mM. Growth was continued at 18 • C, overnight. Cells were harvested by centrifugation, washed twice with ice-cold lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl and 20 mM imidazole, pH 8.0), and disrupted using a high pressure crusher SPCH-18 (STANSTED). The sample was centrifuged and the supernatant was loaded onto nickel-nitrilotriacetic acid (Ni-NTA) agarose resin (Invitrogen). The target protein was purified according to the supplier's recommendations. The eluted protein was loaded onto PD-10 column (GE) for buffer exchange to 20 mM sodium phosphate buffer (pH 7.6). The purity of the proteins was checked via SDS-PAGE.

Electrophoretic Mobility Shift Assay (EMSA)
A 300-bp DNA probe containing the rhlR promoter sequence was obtained using PCR from the genomic DNA. For quantitative binding assays, different amounts of purified LasR, DNA probe, and binding buffer (10 mM Tris, 50 mM KCl, 5% glycerin, pH 8.0) were mixed and incubated at 30 • C for 30 min. The reaction mixture was then loaded onto a 6% native polyacrylamide gel and electrophoresed at 180 V for 1.5 h. The gel was subsequently stained with SYBR green I and photographed with a FlourChemQ system (Alpha Innotech, San Jose, CA USA). For EMSA experiments, LasR purification was done in an anaerobic glove box (YQX-II, Xinmiao Medical Instruments, Shanghai, China), filled with a gas mixture (85% N 2 , 10% H 2 , and 5% CO 2 ). The O 2 level was maintained at <0.1% via palladium catalysis of H 2 reaction with O 2 and monitored using gas a detector (ADKS-4, EDKORS, Changzhou, China). The buffers for LasR purification were all degassed, and 10 mM dithiothreitol (DTT) was added, when necessary, and removed by passing a PD-10 desalting column before HS n − treatment.

In Vitro Transcription-Translation Analysis
An S30 T7 High-Yield Protein Expression System (Promega #L1110) was used for in vitro transcription-translation analysis. The reactions contained 20 µL of S30 Premix Plus, 18 µL of T7 S30 extract, 2 µL of E. coli RNA polymerase (NEB #M0551), 1 µL of RNase inhibitor, 500 ng of DNA template containing P rhlR -mkate, and 800 ng of LasR; RNase-free water was added to bring the volume to 50 µL. LasR was used as untreated, 160 µM HS n − -treated, and HS n − -treated LasR, which was then reduced by 30 mM DTT. After being incubated with vigorous shaking at 37 • C for 2 h, the translated mkate was diluted and assayed by using an Synergy H1 microplate reader. The excitation wavelength was set at 588 nm, and the emission wavelength was set at 633 nm. Fluorescence intensities from other groups were divided by that of the untreated LasR to calculate the relative expression levels.

LC-MS/MS Analysis of LasR
Three samples, untreated LasR, HS n − -treated LasR, and DTT-treated LasR, were prepared for mass spectral analysis. Untreated LasR was used as purified and diluted to 1 µg/µL in the 20 mM phosphate buffer (pH 7.6). For DTT-treated LasR, 1 mL of the purified LasR was reacted with 30 µL of 1 M DTT. For HS n − -treated LasR, 1 mL of the purified LasR (1 µg/µL) was mixed with 8 µL of 20 mM polysulfide. All the samples were incubated at room temperature for 1 h. Then denaturing buffer (0.5 M Tris-HCl, 2.75 mM EDTA, 6 M guanidine-HCl, pH 8.1) and iodoacetamide (IAM) were added to denaturalize LasR and block free thiols. Samples were digested by trypsin (Promega) for 12 h at 37 • C and were subjected to C18 Zip-Tip (Millipore) purification for desalting before analysis by HPLC-tandem mass spectrometry (LC-MS) using a Prominence nano-LC system (Shimadzu, Nishinokyo, Japan) and LTQ-OrbitrapVelos Pro CID mass spectrometer (Thermo Scientific, Waltham, MA, USA). A linear gradient of solvent A (0.1% formic acid in 2% acetonitrile) and solvent B (0.1% formic acid in 98% acetonitrile) from 0% to 100% Antioxidants 2021, 10, 1498 6 of 16 of solvent B in 100 min was used for elution. Full-scan MS spectra (from 400 to 1800 m/z) were detected with a resolution of 60,000 at 400 m/z.

Real-Time Quantitative Reverse Transcription PCR (RT-qPCR)
For RT-qPCR, the cells were collected at a defined incubation time, total RNA was extracted by using a TRIzol TM RNA Purification Kit (12183555, Invitrogen), and cDNA was synthesized by the HiScript ® II Reverse Transcriptase (Vazyme, Nanjing, China). RT-qPCR was performed by using a Bestar SybrGreen qPCR Mastermix (DBI Bioscience, Shanghai, China) and LightCycler 480II (Roche, Penzberg, Germany). For calculation of the relative expression levels of tested genes, rplS was used as the reference gene.

H 2 S and Sulfane
Sulfur Production by P. aeruginosa PAO1 and Its Mutants P. aeruginosa PAO1 contains two sqr genes (Pasqr1 and Pasqr2) and one pdo (Papdo) [16]. In LB medium, the wild type did not release H 2 S, but its mutant (Pa3K) with the three H 2 S-metabolic genes (sqr1, sqr2, and pdo) being knocked out released H 2 S, as detected by filter paper strips containing lead acetate in the gas phase ( Figure 1A). P. aeruginosa PAO1 also contains four genes (cbs, cse, mst, and cysI) capable of producing H 2 S. The four genes were deleted from PAO1 and Pa3K to generate Pa∆H 2 S and Pa7K, respectively; both mutants did not release detectable H 2 S into the gas phase ( Figure 1A). However, there was trace sulfide detectable by using the mBBr method in the culture supernatants of the mutants ( Figure S1). Intracellular sulfane sulfur was detected with SSP4. Pa∆H 2 S and Pa7K clearly contained less sulfane sulfur than PAO1 and Pa3K, especially at 24 h of culturing in LB medium ( Figure 1B). The mutants grew equally well compared to the wild type in LB medium. The results indicate that CBS, CSE, MST, and CysI are involved in generating H 2 S and intracellular sulfane sulfur, and the produced H 2 S is oxidized by SQR and PDO.

Real-Time Quantitative Reverse Transcription PCR (RT-qPCR)
For RT-qPCR, the cells were collected at a defined incubation time, total RNA was extracted by using a TRIzol TM RNA Purification Kit (12183555, Invitrogen), and cDNA was synthesized by the HiScript ® II Reverse Transcriptase (Vazyme, Nanjing, China). RT-qPCR was performed by using a Bestar SybrGreen qPCR Mastermix (DBI Bioscience, Shanghai, China) and LightCycler 480II (Roche, Penzberg, Germany). For calculation of the relative expression levels of tested genes, rplS was used as the reference gene.

H2S and Sulfane Sulfur Production by P. aeruginosa PAO1 and Its Mutants
P. aeruginosa PAO1 contains two sqr genes (Pasqr1 and Pasqr2) and one pdo (Papdo) [16]. In LB medium, the wild type did not release H2S, but its mutant (Pa3K) with the three H2S-metabolic genes (sqr1, sqr2, and pdo) being knocked out released H2S, as detected by filter paper strips containing lead acetate in the gas phase ( Figure 1A). P. aeruginosa PAO1 also contains four genes (cbs, cse, mst, and cysI) capable of producing H2S. The four genes were deleted from PAO1 and Pa3K to generate Pa∆H2S and Pa7K, respectively; both mutants did not release detectable H2S into the gas phase ( Figure 1A). However, there was trace sulfide detectable by using the mBBr method in the culture supernatants of the mutants ( Figure S1). Intracellular sulfane sulfur was detected with SSP4. PaΔH2S and Pa7K clearly contained less sulfane sulfur than PAO1 and Pa3K, especially at 24 h of culturing in LB medium ( Figure 1B). The mutants grew equally well compared to the wild type in LB medium. The results indicate that CBS, CSE, MST, and CysI are involved in generating H2S and intracellular sulfane sulfur, and the produced H2S is oxidized by SQR and PDO.

Virulence Factors and Pathogenicity of PAO1 and Its Mutant Strains
The production of two virulence factors, rhamnolipids and pyocyanin, by P. aeruginosa PAO1 and its mutants was assayed. The deletion of H 2 S-oxidizing genes did not affect the production of rhamnolipids and pyocyanin, but the deletion of H 2 S-producing genes decreased, obviously, the production of rhamnolipids and pyocyanin. Pa∆H 2 S and Pa7K restored rhamnolipid production with added sulfide, implying that H 2 S participates in the regulation of the production of virulence factors in PAO1 (Figure 2A-D). The deletion of H 2 S-producing genes also decreased the pathogenicity, as Pa∆H 2 S and Pa7K almost lost the ability to infect lettuce leaves ( Figure 2E).

Virulence Factors and Pathogenicity of PAO1 and Its Mutant Strains
The production of two virulence factors, rhamnolipids and pyocyanin, by P. aeruginosa PAO1 and its mutants was assayed. The deletion of H2S-oxidizing genes did not affect the production of rhamnolipids and pyocyanin, but the deletion of H2S-producing genes decreased, obviously, the production of rhamnolipids and pyocyanin. PaΔH2S and Pa7K restored rhamnolipid production with added sulfide, implying that H2S participates in the regulation of the production of virulence factors in PAO1 (Figure 2A-D). The deletion of H2S-producing genes also decreased the pathogenicity, as PaΔH2S and Pa7K almost lost the ability to infect lettuce leaves ( Figure 2E).

Linking H2S/Sulfane Sulfur to LasR
The RNA-seq results showed that >3000 genes were differentially expressed in PAO1 and PaΔH2S, indicating that H2S has an immense influence on PAO1 gene expression (Figure S2). The genes related to the production of rhamnolipid and pyocyanin were clearly suppressed in PaΔH2S, relative to PAO1, consistent with the observed phenotypes. The

Linking H 2 S/Sulfane Sulfur to LasR
The RNA-seq results showed that >3000 genes were differentially expressed in PAO1 and Pa∆H 2 S, indicating that H 2 S has an immense influence on PAO1 gene expression ( Figure S2). The genes related to the production of rhamnolipid and pyocyanin were clearly suppressed in Pa∆H 2 S, relative to PAO1, consistent with the observed phenotypes. The potential role of LasR in regulating RhlR and PqsR that control the production of rhamnolipids and pyocyanin was noticed, as the related genes were downregulated in Pa∆H 2 S (Table S3). Upregulated genes are summarized in Table S4 and were dominated by transporters and hypothetical proteins. PAO1 lasR null mutant (Pa∆lasR) grew as well as the wild type in LB ( Figure S3A), but it decreased infection on the lettuce leaves ( Figure S3B).

LasR Senses H 2 S through Sulfane Sulfur
The transcription regulator LasR binds to specific DNA sequences, called lux boxes [30,32]. We constructed a reporter plasmid Ptrc-P lacI -lasR-P rhlR -mkate, containing the lux box of the rhlR upstream region fused to mkate, and introduced it into E. coli BL21(DE3). The induction required 3O-C 12 -HSL [40], and the addition of NaHS did not enhance the mkate expression ( Figure 3A). However, the addition of HS n − significantly enhanced the production of mkate ( Figure 3B). Furthermore, when SQR, which converts H 2 S to sulfane sulfur [51], was also cloned into E. coli containing the reporter system, the addition of NaHS enhanced the mkate expression ( Figure 3A). The results indicate that LasR does not directly sense H 2 S, but senses sulfane sulfur.
porters and hypothetical proteins. PAO1 lasR null mutant (Pa∆lasR) grew as well as the wild type in LB ( Figure S3A), but it decreased infection on the lettuce leaves ( Figure S3B).

LasR Senses H2S through Sulfane Sulfur
The transcription regulator LasR binds to specific DNA sequences, called lux boxes [30,32]. We constructed a reporter plasmid Ptrc-PlacI -lasR-PrhlR-mkate, containing the lux box of the rhlR upstream region fused to mkate, and introduced it into E. coli BL21(DE3). The induction required 3O-C12-HSL [40], and the addition of NaHS did not enhance the mkate expression ( Figure 3A). However, the addition of HSn − significantly enhanced the production of mkate ( Figure 3B). Furthermore, when SQR, which converts H2S to sulfane sulfur [51], was also cloned into E. coli containing the reporter system, the addition of NaHS enhanced the mkate expression ( Figure 3A). The results indicate that LasR does not directly sense H2S, but senses sulfane sulfur.

Characterization of LasR Modification
LasR was purified with 3O-C 12 -HSL because this apoprotein is insoluble [29,52]. An electrophoretic mobility shift assay (EMSA) showed that HS n − -treated LasR did not affect its binding to target DNA ( Figure 4). However, the HS n − -treated LasR had a 3.4-fold higher expression of the mkate gene than DTT-treated or untreated LasR in a coupled transcription and translation assay ( Figure 5), in agreement with the whole-cell reporter assay (Figure 3). These results suggest that the HS n − -modified LasR is more effective when working with RNA polymerase to initiate transcription.

HSn − Modifies Cys 188 , Cys 201 , and Cys 203 of LasR.
In DTT-treated LasR, Cys 188 in Peptide 1a and Cys 201 , and Cys 203 in Peptide 2a were unmodified (Figures 6, S4A and S6A). The thiol groups were blocked by iodoacetamide, indicating that Peptide 1a and Peptide 2a were unmodified. A small fraction of the oxidized form (Figures 6 and S4B, Peptide 1b) containing Cys 188 -SOH was present in untreated LasR and HSn − -treated LasR, but the peak area of Cys 188 -SOH was relatively small. According to the area of the mass spectrogram, HSn − treatment extensively modified Cys 188 with 18% Cys 188 -SSH for Peptide 1c (Figures 6 and S5A) and 31% Cys 188 -SSSH for Peptide 1d (Figures 6 and S5B); 20% Cys 201 and Cys 203 formed a pentasulfur link between the two

HSn − Modifies Cys 188 , Cys 201 , and Cys 203 of LasR.
In DTT-treated LasR, Cys 188 in Peptide 1a and Cys 201 , and Cys 203 in Peptide 2a were unmodified (Figures 6, S4A and S6A). The thiol groups were blocked by iodoacetamide, indicating that Peptide 1a and Peptide 2a were unmodified. A small fraction of the oxidized form (Figures 6 and S4B, Peptide 1b) containing Cys 188 -SOH was present in untreated LasR and HSn − -treated LasR, but the peak area of Cys 188 -SOH was relatively small. According to the area of the mass spectrogram, HSn − treatment extensively modified Cys 188 with 18% Cys 188 -SSH for Peptide 1c (Figures 6 and S5A) and 31% Cys 188 -SSSH for Peptide 1d (Figures 6 and S5B); 20% Cys 201 and Cys 203 formed a pentasulfur link between the two  Figure S4A and Figure S6A). The thiol groups were blocked by iodoacetamide, indicating that Peptide 1a and Peptide 2a were unmodified. A small fraction of the oxidized form ( Figure 6 and Figure S4B, Peptide 1b) containing Cys 188 -SOH was present in untreated LasR and HS n − -treated LasR, but the peak area of Cys 188 -SOH was relatively small. According to the area of the mass spectrogram, HS n − treatment extensively modified Cys 188 with 18% Cys 188 -SSH for Peptide 1c (Figure 6 and Figure S5A) and 31% Cys 188 -SSSH for Peptide 1d (Figure 6 and Figure S5B); 20% Cys 201 and Cys 203 formed a pentasulfur link between the two Cys residues (RS-SSS-SR) ( Figure 6 and Figure S6B). A peptide containing a disulfide bond between Cys 201 and Cys 203 was not found in these samples. The observed and calculated masses of corresponding peptides are given in Table S5. These results show that HS n − readily modifies LasR.
Antioxidants 2021, 10, 1498 10 of 16 Cys residues (RS-SSS-SR) (Figures 6 and S6B). A peptide containing a disulfide bond between Cys 201 and Cys 203 was not found in these samples. The observed and calculated masses of corresponding peptides are given in Table S5. These results show that HSn − readily modifies LasR.

The Expression of lasB, rhlR, and lasI Was Affected by Cellular Sulfane Sulfur
We detected the sulfane sulfur content according to growth stages of P. aeruginosa PAO1 in LB medium. The sulfane sulfur contents were high in the mid log phase to stationary phase, and decreased in the declination phase ( Figure 7). The expression of lasB, rhlR, and lasI, activated by LasR [31], reached the maximum in the stationary phase and decreased sharply in the declination phase. These results indicate that the intracellular level of sulfane sulfur, which is associated with growth phases, regulates LasR activity. LasR ( Figures S4-S6).

The Expression of lasB, rhlR, and lasI Was Affected by Cellular Sulfane Sulfur
We detected the sulfane sulfur content according to growth stages of P. aeruginosa PAO1 in LB medium. The sulfane sulfur contents were high in the mid log phase to stationary phase, and decreased in the declination phase ( Figure 7). The expression of lasB, rhlR, and lasI, activated by LasR [31], reached the maximum in the stationary phase and decreased sharply in the declination phase. These results indicate that the intracellular level of sulfane sulfur, which is associated with growth phases, regulates LasR activity.

Discussion
In bacteria, well-documented examples of signaling mediated by H2S and sulfane sulfur are usually related to sulfur metabolism [7,14,53]. The demonstration that H2S and sulfane sulfur affect quorum sensing in P. aeruginosa PAO1 provides direct support for the previous prediction that H2S is a signaling message in bacterium [6]. P. aeruginosa PAO1 maintains a high level of cellular sulfane sulfur from the mid log phase to stationary phase (Figure 7). One mL of E. coli cells at an OD600nm of 1 corresponds to 3.6 μL of cell volume on average, as determined under 22 growth conditions [54]. By using this conversion factor, the maximum cellular sulfane sulfur content in LB grown E. coli cells was calculated to be about 111 μM, from previously reported data [17], and that in P. aeruginosa PAO1 can be calculated to be to around 250 μM from the data in Figure 7. The relatively low level of sulfane sulfur in E. coli explains why the LasR activity was enhanced by adding H2S when the E. coli cells contained both SQR (sulfide:quinone oxidoreductase) and the reporter system ( Figure 3A) or by adding HSn − when the E. coli cells contained only the reporter system ( Figure 3B).
Bacteria can metabolize L-cysteine directly into H2S by cysteine desulfhydrases or to produce sulfane sulfur via the concerted actions of L-cysteine aminotransferase and 3mercaptopyruvate sulfurtransferase [18]. When bacteria possess SQR, they will also oxidize H2S to sulfane sulfur. Sulfane sulfur can be reduced to H2S by cellular thiols, such as glutathione, or by thioredoxin and glutareduxin [20,55], or be further oxidized by PDO (persulfide dioxygenase) to sulfite, which spontaneously reacts with sulfane sulfur to produce thiosulfate [11,16]. Since the Pa3K strain and the wild type maintained similar amounts of cellular sulfane sulfur ( Figure 1B), the concerted action of SQR and PDO for the oxidation of self-produced H2S did not increase the cellular sulfane sulfur in the wild type. In the Pa3K mutant without SQR and PDO, the self-produced H2S evaporated into the gas phase, as detected by the filter paper containing lead acetate ( Figure 1A). E. coli without SQR and PDO has been shown to mainly use L-cysteine metabolism to maintain cellular sulfane sulfur [18]. Thus, the maintenance of cellular sulfane sulfur requires an active metabolism, and it is not a surprise that the cells in the declination phase contain low sulfane sulfur (Figure 7).

Discussion
In bacteria, well-documented examples of signaling mediated by H 2 S and sulfane sulfur are usually related to sulfur metabolism [7,14,53]. The demonstration that H 2 S and sulfane sulfur affect quorum sensing in P. aeruginosa PAO1 provides direct support for the previous prediction that H 2 S is a signaling message in bacterium [6]. P. aeruginosa PAO1 maintains a high level of cellular sulfane sulfur from the mid log phase to stationary phase ( Figure 7). One mL of E. coli cells at an OD 600nm of 1 corresponds to 3.6 µL of cell volume on average, as determined under 22 growth conditions [54]. By using this conversion factor, the maximum cellular sulfane sulfur content in LB grown E. coli cells was calculated to be about 111 µM, from previously reported data [17], and that in P. aeruginosa PAO1 can be calculated to be to around 250 µM from the data in Figure 7. The relatively low level of sulfane sulfur in E. coli explains why the LasR activity was enhanced by adding H 2 S when the E. coli cells contained both SQR (sulfide:quinone oxidoreductase) and the reporter system ( Figure 3A) or by adding HS n − when the E. coli cells contained only the reporter system ( Figure 3B).
Bacteria can metabolize L-cysteine directly into H 2 S by cysteine desulfhydrases or to produce sulfane sulfur via the concerted actions of L-cysteine aminotransferase and 3-mercaptopyruvate sulfurtransferase [18]. When bacteria possess SQR, they will also oxidize H 2 S to sulfane sulfur. Sulfane sulfur can be reduced to H 2 S by cellular thiols, such as glutathione, or by thioredoxin and glutareduxin [20,55], or be further oxidized by PDO (persulfide dioxygenase) to sulfite, which spontaneously reacts with sulfane sulfur to produce thiosulfate [11,16]. Since the Pa3K strain and the wild type maintained similar amounts of cellular sulfane sulfur ( Figure 1B), the concerted action of SQR and PDO for the oxidation of self-produced H 2 S did not increase the cellular sulfane sulfur in the wild type. In the Pa3K mutant without SQR and PDO, the self-produced H 2 S evaporated into the gas phase, as detected by the filter paper containing lead acetate ( Figure 1A). E. coli without SQR and PDO has been shown to mainly use L-cysteine metabolism to maintain cellular sulfane sulfur [18]. Thus, the maintenance of cellular sulfane sulfur requires an active metabolism, and it is not a surprise that the cells in the declination phase contain low sulfane sulfur (Figure 7).
LasR is a QS master regulator in P. aeruginosa PAO1, and it activates the production of several extracellular products that benefit the population as a whole. Since it is a costly process, P. aeruginosa PAO1 develops strategies to regulate LasR activity, including 3O-C 12 -HSL [35,56], RpoS [57,58], and QscR [38]. Our results suggest that LasR activity is further controlled by cellular sulfane sulfur. Both in vitro and in vivo data indicate that sulfane sulfur-modified LasR is significantly more active than unmodified LasR in the presence of 3O-C 12 -HSL (Figures 3 and 5). The involvement of sulfane sulfur in regulating LasR-mediated QS makes sense, as it varies during growth, becoming significantly lower in the declination phase (Figure 7), when cells enter survival mode and no long need LasR activity. A model whereby LasR activity is regulated by both 3O-C 12 -HSL and cellular sulfane sulfur is proposed (Figure 8), allowing the QS to coordinate with the growth phases. When LasR binds 3O-C 12 -HSL, it may bind to the promoter to recruit RNA polymerase for transcription. However, sulfane sulfur-modified LasR is more active than unmodified LasR in promoting transcription. As several virulence factors are activated by LasR, adequate levels of cellular sulfane sulfur are likely important for the pathogenicity of P. aeruginosa PAO1 (Figures 2 and 8). LasR is a QS master regulator in P. aeruginosa PAO1, and it activates the production of several extracellular products that benefit the population as a whole. Since it is a costly process, P. aeruginosa PAO1 develops strategies to regulate LasR activity, including 3O-C12-HSL [35,56], RpoS [57,58], and QscR [38]. Our results suggest that LasR activity is further controlled by cellular sulfane sulfur. Both in vitro and in vivo data indicate that sulfane sulfur-modified LasR is significantly more active than unmodified LasR in the presence of 3O-C12-HSL (Figures 3 and 5). The involvement of sulfane sulfur in regulating LasR-mediated QS makes sense, as it varies during growth, becoming significantly lower in the declination phase (Figure 7), when cells enter survival mode and no long need LasR activity. A model whereby LasR activity is regulated by both 3O-C12-HSL and cellular sulfane sulfur is proposed (Figure 8), allowing the QS to coordinate with the growth phases. When LasR binds 3O-C12-HSL, it may bind to the promoter to recruit RNA polymerase for transcription. However, sulfane sulfur-modified LasR is more active than unmodified LasR in promoting transcription. As several virulence factors are activated by LasR, adequate levels of cellular sulfane sulfur are likely important for the pathogenicity of P. aeruginosa PAO1 (Figures 2 and 8). Sulfane sulfur has been shown to modify several gene regulators in different forms. FisR of Cupriavidus pinatubonensis JMP134 forms a tetrasulfide crosslinking [7], while CstR of Staphylococcus aureus generates a mixture of di-, tri-, and tetra-sulfur crosslinked species, respectively [9,53]. MexR of Pseudomonas aeruginosa PAO1 mainly forms a disulfide bond, with a small portion of trisulfide [15]. OxyR of Escherichia coli responds to sulfane sulfur stress via persulfidation of OxyR at Cys 199 [14]. The LasR modification by HSn − is similar, but different, with persulfidation and trisulfidation of Cys 188 and a pentasulfur link between Cys 201 and Cys 203 ( Figure 6).
Sulfane sulfur affects gene repressors and activators in different ways in the reported examples to date. The modified repressors, such as CstR, SqrR, and MexR, no longer bind to the target site for the de-repression of the controlled genes [8,9,15]. The modification of activators often leads to increased transcription of the target genes. When FisR is modified by sulfane sulfur, it activates σ 54 -dependent transcription of sulfide-oxidizing genes for sulfide removal [7]. When OxyR is modified, it increases the transcription of several genes Sulfane sulfur has been shown to modify several gene regulators in different forms. FisR of Cupriavidus pinatubonensis JMP134 forms a tetrasulfide crosslinking [7], while CstR of Staphylococcus aureus generates a mixture of di-, tri-, and tetra-sulfur crosslinked species, respectively [9,53]. MexR of Pseudomonas aeruginosa PAO1 mainly forms a disulfide bond, with a small portion of trisulfide [15]. OxyR of Escherichia coli responds to sulfane sulfur stress via persulfidation of OxyR at Cys 199 [14]. The LasR modification by HS n − is similar, but different, with persulfidation and trisulfidation of Cys 188 and a pentasulfur link between Cys 201 and Cys 203 ( Figure 6).
Sulfane sulfur affects gene repressors and activators in different ways in the reported examples to date. The modified repressors, such as CstR, SqrR, and MexR, no longer bind to the target site for the de-repression of the controlled genes [8,9,15]. The modification of activators often leads to increased transcription of the target genes. When FisR is modified by sulfane sulfur, it activates σ 54 -dependent transcription of sulfide-oxidizing genes for sulfide removal [7]. When OxyR is modified, it increases the transcription of several genes for the removal of high levels of cellular sulfane sulfur [14]. LasR is also a gene activator [34], and it becomes more active in initiating transcription of controlled genes ( Figure 5). LasR responds to both H 2 O 2 and sulfane sulfur with opposite effects. H 2 O 2 treatment of LasR produces a disulfide bond between Cys 201 and Cys 203 , disrupting the LasR binding to its target DNA [40], explaining why LasR is sensitive to oxidative stress [59,60]. On the other hand, sulfane sulfur treatment of LasR results in a pentasulfur link between Cys 201 and Cys 203 , which does not affect the DNA binding ( Figure 4) and makes LasR more active in transcription initiation (Figures 3 and 5). The H 2 S n -modified LasR may be more resistant to H 2 O 2 damage, but the detailed activation mechanism warrants further investigation.
Garlic extracts that contain organosulfur compounds, such as diallyl disulfides, have antimicrobial activities [61]. The inhibitory activities of diallyl disulfides are in the order tetrasulfide > trisulfide > disulphide > monosulfide [62], implying that the long chain compounds with additional sulfane sulfur are more inhibitory. Furthermore, diallyl disulfide reduces the pathogenicity and biofilm development of P. aeruginosa PAO1 by targeting its QS systems [63]. The inhibitory effect of diallyl disulfide may be in part due to its ability to release sulfane sulfur [64]. As the added diallyl disulfide is up to 1.28 mg per mL (about 8.8 mM) [63], the higher concentration may lead to increased levels of sulfane sulfur in P. aeruginosa PAO1. At high concentrations, sulfane sulfur is toxic to microorganisms [14].

Conclusions
The level of cellular sulfane sulfur varies along with the growth phases of P. aeruginosa PAO1. The relatively high level of sulfane sulfur enhances the LasR activity for QS, and a low level inhibits LasR activity, which is likely further reduced by increased oxidative stress, when P. aeruginosa PAO1 enters the declination phase. These findings add a new level of control of LasR activity, beside the autoinducer, H 2 O 2 , and other factors [34][35][36][37][38][39][40][41], representing a fine-tuned activity regulated by various cellular factors. The discovery solidifies a key step in establishing the general signaling role of sulfide and sulfane sulfur in bacteria.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/antiox10091498/s1. Figure S1. The production of H 2 S in PAO1 and its mutants. Figure S2. Transcriptomic analysis of PAO1 and the Pa∆H 2 S mutant. Figure S3. The effect of lasR deletion in P. aeruginosa PAO1 on the growth and pathogenicity. Figure S4. LTQ-Orbitrap tandem mass spectrometry analysis of LasR (Peptides 1a and 1b). Figure S5. LTQ-Orbitrap tandem mass spectrometry analysis of LasR (Peptides 1c and. 1d). Figure S6. LTQ-Orbitrap tandem mass spectrometry analysis of LasR (Peptides 2a and 2b). Table S1. Strains and plasmids used in this study. Table S2. Primers used in this study. Table S3. List of genes significantly downregulated in Pa∆H 2 S (≥5-fold) relative to the wild-type strain PAO1. Table S4. List of genes significantly upregulated in Pa∆H 2 S (≥5-fold) relative to the wild-type strain PAO1. Table S5. Mass data from LTQ-Orbitrap tandem mass spectrometry.
Author Contributions: G.X. and C.L. acquired and analyzed the data. K.L. and H.X. contributed to the gene deletion and plasmid construction. H.L. and Y.X. supervised the research. G.X., Y.X. and L.X. designed the study and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding:
The work was financially supported by grants from the National Natural Science Foundation of China (91951202).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data is contained within the article and supplementary material.

Conflicts of Interest:
The authors declare no conflict of interest.