The Essential Role of OmpR in Acidithiobacillus caldus Adapting to the High Osmolarity and Its Regulation on the Tetrathionate-Metabolic Pathway

Acidithiobacillus spp. are prevalent in acid mine drainage, and they have been widely used in biomining for extracting nonferrous metals from ores. The osmotic stress generated by elevated concentrations of inorganic ions is a severe challenge for the growth of Acidithiobacillus spp. in the bioleaching process; however, the adaptation mechanism of these bacteria to high osmotic pressure remains unclear. In this study, bioinformatics analysis indicated that the osmotic stress response two-component system EnvZ-OmpR is widely distributed in Acidithiobacillus spp., while OmpRs from Acidithiobacillus spp. exhibited a far more evolutionary relationship with the well-studied OmpRs in E. coli and Salmonella typhimurium. The growth measurement of an Acidithiobacillus caldus (A. caldus) ompR-knockout strain demonstrated that OmpR is essential in the adaptation of this bacterium to high osmotic stress. The overall impact of OmpR on the various metabolic and regulatory systems of A. caldus was revealed by transcriptome analysis. The OmpR binding sequences of differentially expressed genes (DEGs) were predicted, and the OmpR box motif in A. caldus was analysed. The direct and negative regulation of EnvZ-OmpR on the tetrathionate-metabolic (tetH) cluster in A. caldus was discovered for the first time, and a co-regulation mode mediated by EnvZ-OmpR and RsrS-RsrR for the tetrathionate intermediate thiosulfate-oxidizing (S4I) pathway in this microorganism was proposed. This study reveals that EnvZ-OmpR is an indispensable regulatory system for the ability of A. caldus to cope with high osmotic stress and the significance of EnvZ-OmpR on the regulation of sulfur metabolism in A. caldus adapting to the high-salt environment.

In the natural mine habitats, Acidithiobacillus spp. can metabolize the sulfur and ferrous iron deposited in minerals, generating the ion-enriched environments. For example, a sulfate concentration up to 7.9 M was detected in the water of the abandoned Richmond mine at Iron Mountain, California [11]. Metal ions, such as Fe 3+ , Cu 2+ , and Zn 2+ , are released from ores as the bioleaching reaction progresses. Dissolved inorganic ions exert

Construction of the ompR Knockout Strain of A. caldus
First, homologous 5 and 3 flanking regions of ompR were amplified using primers P1-F/R and P2-F/R, respectively. The products were further digested and ligated into the vector pSDUDI, generating the suicide plasmid pSDUDI-∆ompR. Second, this constructed plasmid was conjugated from E. coli SM10 to A. caldus according to a published protocol [38]. Single crossovers were selected on kanamycin selective Starkey-Na 2 S 2 O 3 solid plates and confirmed by colony PCR using primers oriT-F/R. Third, the I-Sce I-expressing plasmid (pSDU1-I-Sce I) was conjugated into single crossovers to facilitate the second crossover and generate mutants. Finally, the ompR knockout strains were screened by PCR using two sets of primers, ompR-F/R and IHA-F/R, specific for the targeted ompR gene and the interior regions of the two homologous arms, respectively. Identification of the ompR-knockout strain was performed by PCR using purified genomic DNA (TIANamp Bacteria DNA kit, Tiangen, China) as the templates with three sets of primers (ompR-F/R, IHA-F/R, and LHA-F/R). The PCR fragments amplified by primers LHA-F/R specific for the lateral regions of the two homologous arms were sequenced for verification. Primers used in the construction of A. caldus ompR-knockout strain were listed in Table 2.

RNA Extraction and Transcriptional Analysis
A. caldus wild-type and mutant strains were cultivated to the 6th day in S 0 media. Na 2 SO 4 was added to the cultures at a final concentration of 0.3 M. After an additional 3-h-cultivation, the cultures were filtered through the filter paper to remove the sulfur granules, and all cells were collected by centrifugation at 8000× g for 5 min. Preparation of RNA for RNA-seq was according to [39], and the subsequent construction of cDNA libraries and RNA-seq were performed by Novogene (Tianjin, China) [39]. Genes with an adjusted p value ≤ 0.05, FDR (false discovery rate) ≤ 0.05 and fold change (FC) ≥ 1.5 were selected as DEGs.
Real-Time Quantitative PCR (RT-qPCR) was used to confirm the DEGs detected by RNA-seq. All RT-qPCR reactions in this study were performed with three biological replicates. The experimental method and statistical approach are performed according the previous report in [39]. Genes with FC ≥ 1.5, p ≤ 0.05 and FC ≤ 0.67, p ≤ 0.05 were considered significantly up-regulated and down-regulated, respectively. The standard deviation (SD) value was calculated by using the Origin software, and the p value was determined by an unpaired t-test using GraphPad Prism software. A Bland-Altman limit of an agreement (LOA) plot was used to carry out agreement comparison of the results from RNA-seq and RT-qPCR. Primers used in RT-qPCR were listed in Table S1.
Restriction sites were indicated with underlines.

Expression and Purification of the OmpR Protein
The ompR gene of A. caldus MTH-04 was amplified with primers ompR 02628 -F and ompR 02628 -R ( Table 2). The purified PCR product and pET-28a were digested with BamH I and Hind III, and the generated products were ligated together to obtain plasmid pET-28a-ompR. The sequences of the inserted ompR gene were verified by sequencing, and pET-28a-ompR with the correct sequences was transformed into E. coli BL21 (DE3). The recombinant OmpR was purified with HisTrap HP column (GE Healthcare, Chicago, IL, USA). The concentration of purified OmpR was determined using the Bradford assay.

Isothermal Titration Calorimetry (ITC)
To determine the Kd (dissociation constant) value and thermodynamic parameters of the interaction between OmpR and predicted OmpR boxes within the tetH promoter region of A. caldus, OmpR was titrated with the DNA sequence using a MicroCal iTC200 system (GE Healthcare). The purified protein was thoroughly dialyzed at 4 • C against TE buffer (100 mM Tris-HCl, 50 mM EDTA [pH 8.0]). The protein solution (24 µM) was titrated at 30 • C with a double-stranded DNA (dsDNA) solution containing 2 µM OmpR box fragments, with 19 injections of 2 µL after the first injection of 0.5 µL and a time interval of 120 s between injections. The heat effects from a blank experiment (injection of OmpR into the TE buffer) were subtracted before the titration curves were fit to a nonlinear least-squares function. From these curve fits, Kd, the change in binding enthalpy (∆H), and the binding stoichiometry were determined. The change in Gibbs free energy (∆G) and the change in entropy (∆S) were calculated using the equations ∆G = RTlnKd and ∆G = ∆H − T∆S, where R is the universal molar gas constant and T is the temperature (in kelvins). DNA fragments were obtained by PCR amplification using the different sets of primers listed in Table 2.

Sequence Analysis and Comparison of OmpRs in Acidithiobacillus spp.
Annotations of published genome sequences for the Acidithiobacillus genus revealed the presence of OmpR and EnvZ homologues and the existence of ompR-envZ operons ( Figure 1A). Comparison of amino acid sequences showed that OmpR and EnvZ in Acidithiobacillus are distantly related to the respective protein in E. coli. OmpRs and EnvZs from Acidithiobacillus show approximately 50% and 30% identities to the proteins in E. coli, respectively. An unrooted phylogenetic tree was constructed for the predicted OmpRs from some Acidithiobacillus strains and several other identified OmpR homologues ( Figure 1B). OmpRs from Acidithiobacillus spp. exhibit a close relationship between Acidithiobacillus species, but a far evolutionary relationship with the well-studied OmpRs in E. coli and S. typhimurium. By analysing the OmpRs from Acidithiobacillus spp. and E. coli K12, OmpRs in acidophilic autotrophic Acidithiobacillus strains were found to carry the typical signal receiver (REC) and helix-turn-helix (HTH) domains responsible for receiving the phosphoryl group from their cognate EnvZs and binding to regulatory sequences, respectively ( Figure 1C). The phosphorylation site (D) and dimerization interface (KPF) in the REC domain are also predicted to be present in the OmpRs from Acidithiobacillus ( Figure 1C). The presence of a similar ompR-envZ operon structure, and the discovery of the typical domains and conserved amino acid residues among OmpRs indicated potential similar biological functions of the EnvZ-OmpR system in the acidophilic autotrophic Acidithiobacillus spp. and in E. coli.

Construction of the A. caldus ompR-Knockout Strain
The ompR gene in A. caldus MTH-04 was successfully deleted using the "in-out" markerless gene knockout strategy developed by our laboratory [35]. As shown in Figure 2A, the "in" step was achieved with the suicide plasmid pSDUDI-∆ompR containing two homologous arms located upstream and downstream of the ompR gene. The suicide plasmid integrates into the A. caldus genome via allelic exchange, resulting in the generation of a single crossover mutant. The "out" step was performed by introducing plasmid pSDU1-I-SceI expressing the I-SceI endonuclease. The enzyme produces a double-stranded break (DSB) at the I-SceI site in the genome of the single crossover mutant. The DSB stimulates a second allelic exchange, producing a mutant or a wild-type sequence. The ∆ompR strain was confirmed by PCR using the different sets of primers ( Figure 2B,C). No band was amplified from the ompR knockout strain when using primers ompR-F/R specific for the targeted gene ompR. A smaller fragment was obtained from the mutant using primers UHA-F/DHA-R specific for the homologous arms, and a 2.7 kb fragment was amplified from ∆ompR compared to a 3.4 kb fragment from the wild-type using primers LHA-F/R specific for the lateral regions of the two homologous arms. Finally, the 2.7 kb PCR fragment amplified from ∆ompR was sequenced for verification. The results indicated that the ∆ompR strain carries a markerless in-frame ompR mutation with deletion of a 720 bp sequence from the start (ATG) to the stop (TGA) codes.

The Influence of Ion Concentration on the Growth of A. caldus and its ompR Mutant
To investigate the effect of osmotic stress on A. caldus cells, we examined the growth of ∆ompR and wild-type strains in liquid starkey-S 0 medium with different concentrations of inorganic ions. When sodium sulfate was used as the reagent for generating osmotic pressure, the mutant exhibited an obvious disadvantage in growth compared to that of the wild-type strain with an increase in the concentration. The ∆ompR strain displayed a slightly weaker growth pattern compared to that of the wild-type strain at Na 2 SO 4 concentration of 0 and 0.15 M ( Figure 3A,B). When the concentration of Na 2 SO 4 was increased to 0.30 M, the growth of the mutant was obviously suppressed compared to that of the wild-type strain ( Figure 3C). ∆ompR was unable to grow when the concentration of Na 2 SO 4 reached 0.45 M ( Figure 3D). Similar phenomena were observed when K 2 SO 4 was used to test the sensitivity of the mutant to osmotic stress ( Figure S1). Although the growth of ∆ompR was suppressed by NaCl at concentrations of 0.15, 0.30 or 0.45 M, the mutant could grow at 0.45 M ( Figure S1). Therefore, the poor growth of ∆ompR in the presence of elevated concentrations of inorganic salts, indicated the critical role of OmpR in A. caldus adaptation to the osmotic stress caused by inorganic ions.

Construction of the A. caldus ompR-Knockout Strain
The ompR gene in A. caldus MTH-04 was successfully deleted using the "in-out" markerless gene knockout strategy developed by our laboratory [35]. As shown in Figure  2A, the "in" step was achieved with the suicide plasmid pSDUDI-ΔompR containing two homologous arms located upstream and downstream of the ompR gene. The suicide plasmid integrates into the A. caldus genome via allelic exchange, resulting in the generation of a single crossover mutant. The "out" step was performed by introducing plasmid pSDU1-I-SceI expressing the I-SceI endonuclease. The enzyme produces a double- specific for the targeted gene ompR. A smaller fragment was obtained from the mutant using primers UHA-F/DHA-R specific for the homologous arms, and a 2.7 kb fragment was amplified from ΔompR compared to a 3.4 kb fragment from the wild-type using primers LHA-F/R specific for the lateral regions of the two homologous arms. Finally, the 2.7 kb PCR fragment amplified from ΔompR was sequenced for verification. The results indicated that the ΔompR strain carries a markerless in-frame ompR mutation with deletion of a 720 bp sequence from the start (ATG) to the stop (TGA) codes.

The Influence of Ion Concentration on the Growth of A. caldus and its ompR Mutant
To investigate the effect of osmotic stress on A. caldus cells, we examined the growth of ΔompR and wild-type strains in liquid starkey-S 0 medium with different concentrations of inorganic ions. When sodium sulfate was used as the reagent for generating osmotic pressure, the mutant exhibited an obvious disadvantage in growth compared to that of the wild-type strain with an increase in the concentration. The ΔompR strain displayed a slightly weaker growth pattern compared to that of the wild-type strain at Na2SO4 concentration of 0 and 0.15 M ( Figure 3A,B). When the concentration of Na2SO4 was increased to 0.30 M, the growth of the mutant was obviously suppressed compared to that of the wild-type strain ( Figure 3C). ΔompR was unable to grow when the concentration of Na2SO4 reached 0.45 M ( Figure 3D). Similar phenomena were observed when K2SO4 was used to test the sensitivity of the mutant to osmotic stress ( Figure S1). Although the growth of ΔompR was suppressed by NaCl at concentrations of 0.15, 0.30 or 0.45 M, the mutant could grow at 0.45 M ( Figure S1). Therefore, the poor growth of ΔompR in the

The Influence of OmpR Absence on the Transcriptome Profile of A. caldus under Osmotic Stress
Total RNA from the wild-type and mutant strains under an osmotic pressure of 0.3 M Na 2 SO 4 was extracted to carry out RNA-seq for detecting DEGs. A total of 109 DEGs, including 75 down-regulated and 34 up-regulated DEGs in the OmpR knockout strain (see Supplementary Table S2), were selected using the criteria described in the Methods Section 2.3. Reverse transcription-quantitative PCR (RT-qPCR) using the primers listed in supplementary Table S1 was performed on 35 of the 109 DEGs; fold changes were calculated (Table S2), and these fold change values based on RT-qPCR were compared to those of RNA-seq. The results indicated that among the 35 genes, data for 33 were within the 95% confidence interval (see Supplementary Figure S2), suggesting high consistency between the results generated by RNA-seq and RT-qPCR. Thus, the fold changes of genes determined by RNA-seq are reliable data.
The 109 DEGs are mainly involved in the metabolic processes of sulfur, nitrogen and carbon, membrane and channel proteins, conjugal transfer system and DNA modification system (Table S2 and Figure 4). Deletion of ompR resulted in up-regulation of tetH and tqo and down-regulation of soxY (F0726_02556), soxZ (F0726_02557), soxB (F0726_02558), soxA (F0726_02562) and sdo, indicating the influence of EnvZ-OmpR on sulfur oxidation in A. caldus. Many channels and membrane proteins showed obvious down-regulation in the ompR mutant, such as the TonB-dependent receptor (F0726_03003), ABC-type Mn 2+ /Zn 2+ transport system (F0726_00324, 00325 and 00326) and RND efflux system(F0726_01023), suggesting regulation of membrane permeability by OmpR. Expression of flagellar synthesis genes (F0726_01512, 02316 and 02317) was down-regulated in the mutant, implying the influence of EnvZ-OmpR on cell motility in A. caldus adaptation to osmotic stress. In addition, the absence of OmpR resulted in down-regulation of the restriction-modification

The Influence of OmpR Absence on the Transcriptome Profile of A. caldus under Osmotic Stress
Total RNA from the wild-type and mutant strains under an osmotic pressure of 0.3 M Na2SO4 was extracted to carry out RNA-seq for detecting DEGs. A total of 109 DEGs, including 75 down-regulated and 34 up-regulated DEGs in the OmpR knockout strain (see Supplementary Table S2), were selected using the criteria described in the Methods Section 2.3. Reverse transcription-quantitative PCR (RT-qPCR) using the primers listed in supplementary Table S1 was performed on 35 of the 109 DEGs; fold changes were calculated (Table S2), and these fold change values based on RT-qPCR were compared to those of RNA-seq. The results indicated that among the 35 genes, data for 33 were within the 95% confidence interval (see Supplementary Figure S2), suggesting high consistency between the results generated by RNA-seq and RT-qPCR. Thus, the fold changes of genes determined by RNA-seq are reliable data.
The 109 DEGs are mainly involved in the metabolic processes of sulfur, nitrogen and carbon, membrane and channel proteins, conjugal transfer system and DNA modification system (Table S2 and Figure 4). Deletion of ompR resulted in up-regulation of tetH and tqo and down-regulation of soxY (F0726_02556), soxZ (F0726_02557), soxB (F0726_02558), soxA

OmpR Binding Sequences in A. caldus
To discover OmpR binding sequences (OmpR boxes) in A. caldus, the 700 bp sequence upstream of the DEGs were used for scanning OmpR boxes using the matrix-scan program at http://rsat.ulb.ac.be/rsat, with a cutoff score value of 7. As shown in Table 3, OmpR consensus-like sequences were discovered upstream of 25 DEGs, including 10 up-regulated and 15 down-regulated genes. Furthermore, these predicted OmpR binding sequences in A. caldus MTH-04 were used to create the logo motif of OmpR boxes ( Figure 5A). The OmpR-binding sequence motif from A. caldus displayed 71.4% and 65.0% similarity to that from E. coli and Salmonella enterica (S. enterica), respectively ( Figure 5B). Overall, the discovery of OmpR binding sequences upstream of the DEGs and the high similarities of OmpR box motifs between A. caldus and other bacteria, indicate the conservation of OmpR during its evolution in these chemoautotrophic and acidophilic bacteria. the influence of EnvZ-OmpR on cell motility in A. caldus adaptation to osmotic stress. In addition, the absence of OmpR resulted in down-regulation of the restriction-modification system, transposases, regulators and cyclic di-GMP metabolism-related proteins. The majority of DEGs are down-regulated in ΔompR, indicating that the mutant had to reduce metabolism, membrane permeability and cell motility to adapt to the osmotic pressure. Therefore, EnvZ-OmpR played a global regulatory role in the osmoregulation of A. caldus.

The Binding Ability of OmpR to the tetH Promoter Fragment
The two co-transcribed genes tetH (encoding tetrathionate hydrolase) and tqo (encoding thiosulfate: quinol oxidoreductase), combined with a two-component system (RsrR and RsrS), are arranged in a cluster in A. caldus [40]. Up-regulation of tetH and tqo in ∆ompR and the predicted OmpR boxes at upstream of the tetH gene suggest a negative and direct regulation of tetH by OmpR (Tables S2 and 3). To confirm this, assays, including isothermal titration calorimetry (ITC) and electrophoretic mobility shift assay (EMSA), were performed to confirm binding between OmpR and the tetH promoter sequence. In the ITC analysis, no obvious differences in the integration effect were found between OmpR and the TE buffer ( Figure 6A). An obvious reaction signal was detected using the E. coli ompC fragment containing the conserved OmpR box ( Figure 6B), indicating that OmpR from A. caldus is able to bind to the OmpR box sequence. OmpR showed a significant integration effect with the tetH fragment but not the gapdH fragment from A. caldus ( Figure 6C,D), showing the ability of OmpR to bind to the tetH promoter region. Furthermore, the binding ability was confirmed by EMSA. The expected gel shift of OmpRwith the tetH fragment and no shift of OmpRwith the gapdH fragment were observed ( Figure 6E). The consistent results from ITC and EMSA indicated that OmpR is able to bind directly to the tetH promoter region to achieve negative regulation of tetrathionate metabolism of A. caldus ( Figure 6F). Fold Change ≥ 1.5, p-value ≤ 0.05: up-regulation; Fold Change ≤ 0.67, p-value ≤ 0.05: down-regulation. The numbers indicate the nucleotide positions upstream of the transcription start sites; -: negative and direct regulation; +: positive and direct regulation.

The Binding Ability of OmpR to the tetH Promoter Fragment
The two co-transcribed genes tetH (encoding tetrathionate hydrolase) and tqo (encoding thiosulfate: quinol oxidoreductase), combined with a two-component system (RsrR and RsrS), are arranged in a cluster in A. caldus [40]. Up-regulation of tetH and tqo in ΔompR and the predicted OmpR boxes at upstream of the tetH gene suggest a negative and direct regulation of tetH by OmpR (Tables S2 and 3). To confirm this, assays, including isothermal titration calorimetry (ITC) and electrophoretic mobility shift assay (EMSA), were performed to confirm binding between OmpR and the tetH promoter sequence. In the ITC analysis, no obvious differences in the integration effect were found between OmpR and the TE buffer ( Figure 6A). An obvious reaction signal was detected using the E. coli ompC fragment containing the conserved OmpR box ( Figure 6B), indicating that OmpR from A. caldus is able to bind to the OmpR box sequence. OmpR showed a significant integration effect with the tetH fragment but not the gapdH fragment from A. caldus ( Figure 6C,D), showing the ability of OmpR to bind to the tetH promoter region. Furthermore, the binding ability was confirmed by EMSA. The expected gel shift of OmpRwith the tetH fragment and no shift of OmpRwith the gapdH fragment were observed ( Figure  6E). The consistent results from ITC and EMSA indicated that OmpR is able to bind directly to the tetH promoter region to achieve negative regulation of tetrathionate metabolism of A. caldus ( Figure 6F).

Discussion
Our study revealed that the EnvZ-OmpR system is an essential osmoregulation mechanism for Acidithiobacillus spp. to adapt to a high-salt environment. The discovery of EnvZ-OmpR homologues and the ompR-envZ operon in different species of Acidithiobacillus indicate the presence of the EnvZ-OmpR system in these chemoautotrophic sulfuroxidizing bacteria ( Figure 1A). While the OmpRs from Acidithiobacillus strains are distantly related to those from E. coli and S. typhimurium (Figure 1B), the typical domains and

Discussion
Our study revealed that the EnvZ-OmpR system is an essential osmoregulation mechanism for Acidithiobacillus spp. to adapt to a high-salt environment. The discovery of EnvZ-OmpR homologues and the ompR-envZ operon in different species of Acidithiobacillus indicate the presence of the EnvZ-OmpR system in these chemoautotrophic sulfur-oxidizing bacteria ( Figure 1A). While the OmpRs from Acidithiobacillus strains are distantly related to those from E. coli and S. typhimurium (Figure 1B), the typical domains and conserved amino acid residues of OmpRs in these acidophilic autotrophic bacteria are almost identical to that of E. coli. The significant inhibitory effect of high ion concentrations on the growth of ∆ompR demonstrates the essential role of OmpR for A. caldus to adapt to high osmotic stress ( Figure 3). Overall, the prevalence and conservation of EnvZ-OmpR in Acidithiobacillus spp. as well as the determination of the indispensable role of OmpR in the growth of A. caldus at high concentrations of inorganic salts suggest that EnvZ-OmpRs are also employed by these chemoautotrophic and acidophilic bacteria to cope with the osmotic stress induced by elevated salinity.
The regulation of membrane permeability by OmpR might be a strategy for A. caldus to achieve osmoregulation in a high-salt environment. Unlike OmpR-regulated osmoregulation in heterotrophic neutrophils [19,41], the homologues of the outer membrane porins OmpF and OmpC are absent in these autotrophic and acidophilic bacteria. Acidophiles can regulate the permeability of the cytoplasmic membrane to adapt to osmotic pressure [42,43]. The strong regulating effect of OmpR on membrane and channel proteins (Table S2) confirmed that A. caldus employs EnvZ-OmpR to modulate its membrane permeability according to the concentration of ions in the environment. Although chloride salts have a far greater inhibitory effect on Acidithiobacillus spp. than do sulfate salts [43], ∆ompR showed a relatively good growth capacity in 0.45 M NaCl, in contrast to its failure to grow at the same concentration of sulfate salts (Figures S1 and 3). This growth difference suggested that the type of salt has an obvious influence on the adaptation of the A. caldus ompR-knockout strain to osmotic stress. In many halophilic archaea, intracellular accumulation of chloride and potassium is an important mechanism for dealing with high osmotic pressure [44]. Thus, the better growth capability of ∆ompR at high concentrations of NaCl indicate the presence of a similar osmoregulatory mechanism in A. caldus.
Regulation of sulfur oxidation by OmpR may help A. caldus adapt to osmotic stress. It has been reported that a high concentration of salt (200 mM) lowers the sulfur oxidation rate of A. thiooxidans [12]. Sulfur-metabolic processes in A. caldus include activation and oxidation of elemental sulfur in the outer membrane, thiosulfate-metabolic pathways in the periplasm, and sulfur-oxidizing enzymes in the cytoplasm. The two periplasmic thiosulfate-metabolic pathways, the tetrathionate intermediate thiosulfate oxidation (S 4 I) pathway and sulfur oxidizing enzyme (Sox) system, were both affected by the absence of OmpR under exposure to high salt (Table S2). The S 4 I pathway consists of a thiosulfate: quinol oxidoreductase (Tqo) and a tetrathionate hydrolase (TetH), which are responsible for oxidizing thiosulfate to tetrathionate and hydrolysing tetrathionate to thiosulfate and other products, respectively [40,45]. As thiosulfate is incompletely oxidized by the Sox system, this compound is considered a joint substrate for the S 4 I and Sox pathways. Tetrathionate in the periplasm is proposed to be transferred by DsrE/TusA into cytoplasm for further oxidation [46]; thus, the concentration of tetrathionate might affect periplasmic and cytoplasmic sulfur-oxidizing processes. Moreover, thiosulfate is unstable at pH < 4, whereas tetrathionate is acid stable [47]. Therefore, S 4 I pathway-mediated conversion between S 2 O 3 2− and S 4 O 6 2− , not only influences periplasmic and cytoplasmic sulfur-metabolizing pathways, but also the concentration and species of sulfur substrates in acidic environments. The direct and negative regulation of OmpR on the S 4 I pathway was discovered based on the up-regulation of tetH and tqo in ∆ompR and the binding ability of OmpR to the tetH promoter region (Table S2 and Figure 6). This discovery suggests that A. caldus invokes EnvZ-OmpR to control the expression of the S 4 I pathway directly, further influencing the expression of other sulfur-oxidizing enzymes and the sulfur-metabolic processes. Thus, the EnvZ-OmpR-mediated regulation of sulfur metabolism might be an adaptation of A. caldus to osmotic pressure.
A co-regulation mode of the S 4 I pathway mediated by EnvZ-OmpR and RsrS-RsrR was discovered in A. caldus. A previous study revealed that RsrS-RsrR positively regulates the S 4 I pathway via the binding of RsrR to a 19 bp inverted repeat sequence (IRS) in the tetH promoter region [35]. In this study, the direct and negative regulation of the S 4 I pathway by EnvZ-OmpR was revealed for the first time. While RsrS-RsrR is considered an EnvZ-OmpR like two-component system [35,40], the binding sites for the RsrR and OmpR are not identical and show significant distinction in base composition ( Figure 6F). Thus, it is speculated that RsrR and OmpR bind to IRS and OmpR boxes to achieve positive and negative regulation of the S 4 I pathway, respectively. Thus, a model for the co-regulation mode for the tetrathionate-metabolic pathway mediated by EnvZ-OmpR and RsrS-RsrR, is proposed to illustrate the significance of this regulation to the osmotic and acidic adaptations of A. caldus (Figure 7). Under favorable conditions, A. caldus modulates the expression of sulfur-oxidizing gene via positive regulation of the S 4 I pathway by RsrS-RsrR. As sulfur substrates are oxidized and the ion concentration increases in the solution, the environmental stresses of osmolality and acidity emerges, and A. caldus readjusts the expression of sulfur-oxidizing genes via RsrS-RsrR-mediated negative regulation of the S 4 I pathway. In summary, we demonstrated that EnvZ-OmpR is a fundamental system for A. caldus to adapt to high osmotic pressure. OmpR functions as a global regulator to endow A. caldus with adaption in environments of high salinity. The discovery of direct and negative regulation of the S4I pathway by EnvZ-OmpR reveals the novel regulatory function of EnvZ-OmpR in sulfur-oxidizing bacteria and the significance of sulfur-metabolic regulation in the ability of A. caldus to cope with osmotic stress. The discovery of the co-regulation mode for S4I pathway mediated EnvZ-OmpR and RsrS/RsrR in A. caldus provides new insights into the sulfur-metabolic regulation and environmental adaptation mechanism in these chemoautotrophic sulfur-oxidizing bacteria.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: Growth analysis of ΔompR and wild-type strains of A. caldus at different concentrations of NaCl or K2SO4 in S0 medium; Figure S2: Bland-Altman plot of differences against averages of fold changes for RNA-seq and RT-qPCR, with 95% confidence limit (the upper and lower solid lines); Table S1: Primers used for RT-qPCR; Tables S2: Differentially expressed genes In summary, we demonstrated that EnvZ-OmpR is a fundamental system for A. caldus to adapt to high osmotic pressure. OmpR functions as a global regulator to endow A. caldus with adaption in environments of high salinity. The discovery of direct and negative regulation of the S 4 I pathway by EnvZ-OmpR reveals the novel regulatory function of EnvZ-OmpR in sulfur-oxidizing bacteria and the significance of sulfur-metabolic regulation in the ability of A. caldus to cope with osmotic stress. The discovery of the co-regulation mode for S 4 I pathway mediated EnvZ-OmpR and RsrS/RsrR in A. caldus provides new insights into the sulfur-metabolic regulation and environmental adaptation mechanism in these chemoautotrophic sulfur-oxidizing bacteria.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms11010035/s1, Figure S1: Growth analysis of ∆ompR and wild-type strains of A. caldus at different concentrations of NaCl or K 2 SO 4 in S0 medium; Figure S2: Bland-Altman plot of differences against averages of fold changes for RNA-seq and RT-qPCR, with 95% confidence limit (the upper and lower solid lines); Table S1: Primers used for RT-qPCR; Table S2: Differentially expressed genes detected by RNA-seq and RT-qPCR.

Data Availability Statement:
The genome sequence of A. caldus MTH-04 was deposited in NCBI with GenBank: CP043926.1. The nucleotide sequences of rsrR and rsrS have been deposited with GenBank accession numbers KX161704 and KX161705, respectively. The raw data of RNA-seq is deposited in NCBI with accession number SRA1121784.