The Role of the Two-Component System BaeSR in Disposing Chemicals through Regulating Transporter Systems in Acinetobacter baumannii

Bacterial two-component regulatory systems (TCSs) facilitate changes in gene expression in response to environmental stimuli. TCS BaeR regulons influence tigecycline susceptibility in Acinetobacter baumannii through positively regulating the pump genes adeA and adeB. In this study, we demonstrate that an additional two transport systems, AdeIJK and MacAB-TolC, are also regulated by BaeSR. In the wild type and clinical tigecycline-resistant A. baumannii strains, gene expression of AdeIJK and MacAB-TolC increased after tigecycline induction, implicating their importance to tigecycline resistance in addition to AdeABC. Phenotypic microarray results showed that A. baumannii is vulnerable to certain chemicals, especially tannic acid, after deleting baeR, which was confirmed using the spot assay. The wild-type strain of A. baumannii also exhibited 1.6-fold and 4.4-fold increase in gene expression of adeJ and macB in the medium with 100 μg/mL tannic acid, but the increase was fully inhibited by baeR deletion. An electrophoretic motility shift assay based on an interaction between His-BaeR and the adeA, adeI and macA promoter regions did not demonstrate direct binding. In conclusion, A. baumannii can use the TCS BaeSR in disposing chemicals, such as tannic acid and tigecycline, through regulating the efflux pumps.


Introduction
Efflux pumps actively export antibiotics from the bacterial cell and are, thus, one of the mechanism that contribute to multidrug resistance in bacteria [1]. Four categories of efflux pumps, including the resistance-nodulation-cell division (RND) superfamily, the major facilitator superfamily (MFS), the multidrug and toxic compound extrusion (MATE) family and the small multidrug resistance (SMR) family of transporters, reportedly relate to antimicrobial resistance in A. baumannii [2,3]. CraA [4] and Tet [2], belonging to the MFS efflux pumps, are conferring to chloramphenicol and tetracycline resistance respectively. AbeM [5], the only

Materials and Methods
Bacterial strains, growth conditions, antimicrobial susceptibility test and DNA manipulation The bacterial strains used in this study are listed in Table 1. The cells were commonly grown at 37°C in LB broth and agar. For inducing tigecycline resistance, serial passaging was performed as our previous study [19]. To determine the minimal inhibitory concentration (MIC) of tigecycline, a broth microdilution method according to the 2012 CLSI guidelines [31] was used. The MIC was defined as lowest tigecycline concentration that completely inhibited bacterial growth, and bacterial growth was determined by unaided eyes and by measuring optical density (OD) using a spectrophotometer. The provisional MIC breakpoints for tigecycline are 2, 4, and 8 μg/mL to designate susceptible, intermediate, and resistant strains, respectively [32]. A. baumannii genomic DNA was extracted as described previously [33]. The DNA was PCRamplified with a Hybaid PXE 0.2 HBPX02 Thermal Cycler (Thermo Scientific, Redwood, CA) using ProTaq DNA Polymerase (Protech, Taipei, Taiwan) or the KAPA HiFi PCR Kit (Kapa Biosystems, Boston, MA). The DNA fragments were extracted from agarose gels and purified as previously described [19]. The PCR products were verified by DNA sequencing.

RNA isolation and quantitative reverse transcription (qRT)-PCR
Total RNA was isolated by the phenol-chloroform-isoamyl alcohol (PCIA) method. Briefly, the A. baumannii ATCC 17978 strain was grown overnight in LB broth at 37°C, 220 rpm for 16 hours. The overnight cultures (OD 600~6 .5) were sub-cultured at a 1:100 dilution in 25 mL fresh LB medium in the absence of tigecycline. The cells were grown to mid-log phase (the OD 600 values for ATCC 17978, AB1026, AB1027, AB1028, ABhl1 and ABhl1tc were all~3 and the OD 600 values for both ABtc and ABtcm were~2) and harvested by centrifugation at 4°C. The cell pellets were resuspended in 200 μL ice-cold lysis buffer (0.1 M Tris-Cl [pH 7.5], 0.1 M LiCl, 0.01 M ethylenediaminetetraacetic acid [pH 8.0], 5% sodium dodecyl sulfate [SDS], and 2% β-mercaptoethanol). Then, 200 μL ice-cold PCIA ([25:24:1], pH 4.5) was added. The cells were lysed by vortexing for 2 minutes. Supernatants were collected by centrifugation, extracted with 200 μL ice-cold PCIA. This step was repeated three times. Total RNA was precipitated with ethanol at −80°C overnight and collected by centrifugation at maximum speed for 5 minutes and dissolved in dissolved in 25-100 μL RNase-free water. DNA contaminants were removed using Ambion TURBO DNase (Life Technologies, Grand Island, NY). For cDNA synthesis, RNAs were reverse transcribed using High-Capacity cDNA Reverse Transcriptase Kit (Applied Biosystems). The cDNA samples were used in PCR reactions with different primers listed in Table 1.
For qRT-PCR, a StepOne Real-Time PCR System (Life Technologies) was used with the primers listed in Table 1. Briefly, each 15-μL reaction mixture contained 25 ng cDNA, 7.5 μL Power SYBR green PCR master mix (Life Technologies), and 300 nM each forward and reverse primer. The reactions were performed with 1 cycle at 95°C for 10 minutes, followed by 40 repeated cycles of 95°C for 15 seconds and 60°C for 1 minute. The 16S rRNA transcript was used as an endogenous control for the qRT-PCR. StepOne Software v2.1 (Life Technologies) was used in data analysis.

Phenotype microarray analysis
The phenomes of the A. baumannii ATCC 17978 and its baeR mutant strains to various chemical compounds were assayed using the Biolog Phenotype MicroArray (PM) system (Biolog, Hayward, CA). Microplates PM15B, PM17A, and PM19 (http://www.biolog.com/pdf/pm_lit/ PM11-PM20.pdf) containing total 72 compounds were used. All phenotypic experiments were performed per the manufacturer's protocol (Biolog). Following inoculation, all PM plates were incubated at 37°C for 24 hours. The bacterial growth was considered positive where the tetrazolium-based dye (colorless) reduced to formazan (violet), which was determined visually and using a Microplate Reader (Bio-Rad, Hercules, CA).
Confirmation of the phenotype microarray data using a spot assay To confirm the phenotype microarray analysis, spot assays were performed. A. baumannii ATCC 17978 and the baeR mutant strains were grown overnight in LB broth with or without kanamycin (37°C, 220 rpm, 16 h). The overnight cultures were sub-cultured in 3 mL LB broth (initial cell density, OD 600 = 0.3), grown to mid-log phase and harvested through centrifugation 6000 rpm for five minutes. The cell pellets were resuspended in 1 mL phosphate buffered saline (PBS). A ten-fold serial dilution of the bacterial suspension was performed using PBS. The bacteria were then spotted (10 μL/per spot) onto the LB agar plates containing different concentrations of tannic acid (Avantor, Center Valley, PA) and incubated at 37°C for 24 hours.

BaeR protein purification
BaeR protein purification was performed as previously described [27] with certain modifications. Briefly, the His-tagged baeR gene (640 bp) was digested with restriction enzymes NheI and XhoI and ligated to the pET-23a(+) plasmid. The pET-23a(+)-His-BaeR was then transformed into the Escherichia coli BL21 (DE3) pLysS strain by heat shock method. A single colony of the transformant was inoculated into LB broth containing ampicillin (50 μg/mL) as well as chloramphenicol (30 μg/mL) and incubated at 37°C overnight (16 hours, with shaking 220 rpm). Cells from the overnight culture were sub-cultured (1:100 dilution) to a fresh LB broth, grown to OD 600~0 .6, which was followed by adding 0.5 M isopropyl β-D-1-thiogalactopyranoside (IPTG) (final concentration 0.5 mM); the samples were then further incubated for 4 hours at 37°C and 180 rpm. Cell pellets were collected using centrifugation at 3000 x g and 4°C for 20 minutes, re-suspended in lysis buffer (20 mM NaPB [pH 7.5], 500 mM NaCl) containing 1 mg/mL hen egg white lysozyme (Sigma, St. Louis, MO), and incubated at 4°C for 10 minutes. Thereafter, the cells were lysed through sonication. His-tagged BaeR proteins were collected using centrifugation and maintained on ice before use. The His-BaeR proteins seemed to be in inclusion bodies (i.e., protein aggregates). The inclusion bodies were washed with lysis buffer, resuspended in solubilization buffer (50 mM NaPB [pH 8], 1 M NaCl, 8 M urea, 10% glycerol, and 5 mM imidazole), and centrifuged at 3000 x g and 4°C for 10 minutes. The insoluble materials were discarded, and the supernatant was then mixed with Ni-NTA agarose (Qiagen, Valencia, CA) that was equilibrated with the solubilization buffer. After shaking at room temperature for 1 hour and centrifugation (3000 x g, 4°C for 10 minutes), the agarose pellet was washed twice with 40 mL washing buffer (50 mM NaPB [pH 7] and 4 M urea) containing 0.1% triton X-114 and twice with 20 mL washing buffer. The His-BaeR protein was then purified using Nickel chelate agarose. To elute the Ni-chelated BaeR protein, elution buffer (50 mM NaPB [pH 7], 4 M urea, and 250 mM imidazole) was added to the agarose and mixed at room temperature for 20 minutes. All of the eluted supernatants were pooled together and dialyzed with modified elution buffer containing less urea (50 mM NaPB [pH 7] and 3 M urea, followed with the same buffer containing 2M urea) to eliminate the imidazole. The protein concentration was determined using the bicinchoninic acid (BCA) assay. To refold the His-BaeR protein, the solution was mixed with refolding buffer (20 mM NaPB [pH 7.3] and 10 mg/mL n-octyl-b-D-glucopy-ranoside [OG]) at 42°C for 16 hours and concentrated using an Amicon Ultra-4 centrifugal filter units (Millpore, Billerica, MA) in accordance with the manufacturer's instruction.

Electrophoretic mobility shift assay (EMSA)
For the EMSA, a lightshift chemiluminescent EMSA kit (Thermo Scientific, Rockford, IL) was used, and the manufacturer's protocol was followed. We designed appropriate oligonucleotide pairs that included the putative binding regions for each pump gene. DNA labeling was performed using a Biotin 3´end DNA labeling kit (Thermo Scientific). The DNA probes were prepared through PCR amplification. DNA (20 fmol) and His-BaeR protein (1 μg) were mixed in binding buffer (10 mM Tris-HCl [pH 7.5], 50 mM KCl, and 1 mM DTT) and poly deoxyinosinic-deoxycytidylic acid (poly (dI-dC)) (50 ng/μL). The reaction mixtures (20 μl) were loaded onto a 5% polyacrylamide gel in 0.5 X TBE (45 mM Tris, 45 mM boric acid, and 1 mM EDTA at pH 8.3). The DNA-protein complexes were separated at 100 V for 2 hours in 0.5 X TBE buffer and then transferred to a Biodyne B nylon membrane (Pall Corporation, Port Washington, NY). Crosslinking and detection of the His-labeled DNA-protein complexes were performed using a UV lamp and chemiluminescence, respectively.

Statistical analysis
Depending on the suitability of the different samples, the difference in susceptibility was analyzed using chi-square or Fisher's exact test. The differences between two groups of isolates were considered significant at P < 0.05. Data entry and analyses were performed using the Statistical Package for the Social Sciences (SPSS) software version 15.0 (SPSS Inc., Chicago, IL, USA).

Expression analyses of AdeIJK and MacAB-TolC pump genes in the tigecycline-resistant A. baumannii clinical isolate
The expression levels of adeI, adeJ, and adeK in ABhl1 exhibited a statistically significant decrease compared with the wild-type strain (64%, 59%, and 41% reduction) (Fig 2B), whereas the expression levels of tolC and macB increased 1.7 and 2.3-fold, respectively. To verify these findings, the gene expression of AdeIJK and MacAB-TolC pumps was also examined in the clinical isolate with induced high tigecycline resistance (ABhl1tc). Although the gene expression of adeI, adeJ, and adeK in ABhl1 showed marked decrease compared with the wild-type strain, these genes increased 9.4-, 27-, and 37-folds in the ABhl1tc strain respectively compared with the ABhl1 strain. The increased gene expression of tolC, macB and macA (15-, 4.6-, 3.1-folds respectively) was also observed in ABhl1tc compared with ABhl1.
Phenotype microarray experiment using the baeR mutant Susceptibility to some of the 72 studied compounds was examined using the ATCC 17978 and baeR mutant strains (referred to as ABwt and ABΔbaeR in Fig 3). After a 24 h incubation, no ABΔbaeR cell growth was observed at low concentrations of procaine, alexidine, and puromycin (in PM 15B); aminopyridine, oxycarboxin, caffeine, ethionamide, tannic acid (in PM17A); and difulphiram, iodonitro tetrazolium violet, and thioglycerol (in PM 19) compared with ABwt. In contrast, the ABwt cell growth was inhibited completely at low concentrations of oleandomycin and methyl viologen (in PM15B); niaproof and compound 48/80 in (PM17A); josamycin and FCCP (in PM19) compared with ABΔbaeR. Although the colors of some chemical compounds, such as tannic acid, cefperazone, and gallic acid, somehow interfere with the The reduced expression could be restored through trans-complementation with pWH1266-kan r -baeR. Introducing pWH1266-kan r -baeR into the ATCC 17978 strain mildly increased expression of these genes. The results are shown as the means ± SD from three independent experiments. *, P < 0.05 and **, P < 0.01 and ***, P < 0.001 between ATCC 17978 and AB1026; #, P < 0.05 and ##, P < 0.01 and ###, and P < 0.001 between AB1026 and AB1027. determination of bacterial growth by Microplate Reader, with the help of subsequent bacterial cultures, tannic acid resistance was still shown as the most markedly compromised chemical after baeR deletion. Therefore, tannic acid was used for further study to determine the relationship between tannic acid and expression of baeR as well as pump genes.

Confirmation of the phenotype microarray results using a spot assay
The ATCC 17978 strain can tolerate tannic acid as high as 250 μg/mL (Fig 4A). However, we did not observe growth of 20 μL 10 3 cells/mL baeR mutant bacterial solution in the LB plate containing 50 μg/mL tannic acid, whereas 100 μg/mL tannic acid fully inhibited 10 4 cells/mL baeR mutant strain. In the presence of 150 μg/mL tannic acid, no diluted bacterial solutions exhibited growth, except 10 7 cells/mL baeR mutant strain, which exhibited slight growth With an increasing tannic acid concentration, none of the studied bacterial baeR mutant strain solutions grew. Gene expression analyses of baeR and pump genes after tannic acid exposure To understand the AdeAB, AdeIJK, MacAB-TolC, and BaeRS system responses upon tannic acid exposure, qRT-PCR of adeB, adeJ, macB and baeR was performed for the ATCC 17978, its baeR mutant strain and the clinical strain ABhl1. At 50 μg/mL tannic acid, only macB exhibited a statistically significant increase (1.7-fold) in the A. baumannii ATCC 17978 strain (Fig 4B).
With an increase in tannic acid up to 100 μg/mL, baeR, adeJ and macB exhibited increased gene expression (3.7-, 1.6-and 4.4-fold respectively) and were compared with samples that were not exposed to tannic acid. If the medium contained 500 μg/mL tannic acid, the expression of the genes investigated in the ATCC 17978 strain, baeR, adeB, adeJ, and macB clearly increased (6.8-, 5.0-, 3.1-and 6.2-fold respectively). However, the baeR mutant strain cannot survive in the medium with 500 μg/mL tannic acid. The gene expression for adeB, adeJ, and macB in the baeR mutant strain significantly increased (3.3-, 2.9-and 1.6-fold respectively) at 50 μg/mL tannic acid (Fig 4C). Upon increasing the tannic acid concentration to 100 μg/mL, the gene expression decreased to the levels without tannic acid exposure.
In the clinical strain ABhl1, the baeR gene expression increased 2.5 to 2.8 folds and the adeB gene expression increased 2.6-fold upon being exposed to 50, 100 and 500 μg/mL tannic acid (Fig 4D). The gene expression of macB increased 3.3-fold at 100 μg/mL tannic acid and increased 5.2-fold at 500 μg/mL tannic acid. However, the adeJ gene expression did not show significant fold change while being exposed to tannic acid. Phenotype microarray using the baeR mutant. Differential susceptibility to some of the 72 compounds studied was observed between the ABwt and ABΔbaeR strains. Four different concentrations of each compound were placed in individual wells next to each other in a row with increasing concentrations from the left to the right. The growth of A. baumannii was determined by measuring the optical density at 595 nm (OD 595 ). Compared to ABwt, growth of ABΔbaeR was much sensitive with increasing concentrations of several compounds, including procaine, alexidine, and puromycin (in PM 15B); aminopyridine, oxycarboxin, caffeine, ethionamide, tannic acid (in PM17A); and difulphiram, iodonitro tetrazolium violet, and thioglycerol (in PM 19). All these compounds inhibited ABΔbaeR cell growth were marked with #. Among them, tannic acid sensitivity was the most markedly compromised after deleting baeR. In contrast, ABΔbaeR exhibited better tolerance to oleandomycin, methyl viologen, niaproof, compound 48/80, josamycin and FCCP compared to ABwt (All these compounds were marked with "). *: the color of the chemical compounds somehow interferes with the observation of bacterial growth. Electrophoretic mobility shift assays EMSA was performed using different combinations of 10 fmol/μL biotin-labeled DNA (243-bp, 144-bp, and 329-bp DNA fragments upstream of adeA, adeI and macA, respectively), 2 pmol/μL unlabeled DNA and 0.05 μg/μL purified His-BaeR. Used as a nonspecific competitor, 50 ng/μL Poly (dI-dC) was also added. Adding His-BaeR protein to each biotin-labeled DNA probe did not cause band shift compared to the reaction with the probe only (no His-BaeR added) (S1 Fig). Therefore, we concluded that no protein-DNA complexes formed between His-BaeR protein and each of the adeA, adeI and macA promoter region. In addition to the purified His-BaeR, we also used total protein extract from A. baumannii ATCC 17978 and it's baeR-deletion mutant. Band shift was observed from the reaction containing protein extract and each of the adeA, adeI and macA promoter region. The band shift did not appear in the reaction lacking of protein extract. Moreover, the band shift occurred in the reaction The gene expression levels of the baeR and pump genes in the wild-type strain. At 50 μg/mL tannic acid, only macB showed a statistically significant increase in the wild-type strain. Through increasing medium tannic acid up to 100 μg/mL, baeR, adeJ and macB exhibited increased gene expression. If the medium contained tannic acid 500 μg/mL, the expression of each gene investigated in the wild-type strain increased dramatically. (C) The gene expression levels of the pump genes in the baeR mutant strain. The gene expression of adeB, adeJ, and macB in the baeR mutant strain increased significantly at the tannic acid concentration 50 μg/mL. Through increasing the tannic acid concentration to 100 μg/mL, expression of each gene decreased to levels without tannic acid exposure. (D) The gene expression levels of the baeR and pump genes in the clinical strain ABhl1. The gene expression of baeR, adeB, and macB demonstrated a statistically significant increase upon being exposed to 100 and 500 μg/mL tannic acid. The results are shown as the means ± SD from three independent experiments. *, P < 0.05 and, **, P < 0.01 and ***, P < 0.001. Role of BaeSR in Disposing Chemicals using protein extract from either ATCC 17978 or the baeR-deletion strains. These results suggest that adeA, adeI and macA genes are controlled by regulator other than BaeR.

Discussion
Bacterial two-component regulatory systems (TCSs), which consist of a sensor histidine kinase and a response regulator, facilitate changes in gene expression in response to environmental stimuli [34]. To ensure cell survival in harsh conditions, such as being exposed to hazardous chemicals, histidine kinase can sense the environmental signal and autophosphorylate. The phosphate is then transferred to an aspartic acid residue of the corresponding response regulator. The phosphorylated response regulator can thus elicit many diverse responses, including enhancing its DNA binding ability to modulate target gene expression [35]. Nineteen TCSs were identified in a clinical isolate of multidrug resistant A. baumannii by Adams et al., most of which were also in other clinical isolates, including A. baumannii AYE as well as ACICU, and 17 TCSs are conserved in the A. baumannii ATCC 17978 strain. However, functional studies on the TCSs and their downstream target genes are currently limited in A. baumannii. Only the TCSs PmrAB [36,37], BfmRS [38,39], AdeRS [15,16], BaeSR [19] and GacSA [40] have been characterized. Colistin resistance in A. baumannii can be due to adding phosphoethanolamine to cell wall lipopolysaccharide, which is mediated by pmrAB mutations [36]. Overexpression of the AdeABC efflux pump stimulated by the mutated AdeRS results in antimicrobial resistance, including tigecycline resistance, in multidrug-resistant A. baumannii (MDRAB) [16]. BfmRS TCS is related to biofilm formation [38] and mediates virulence in A. baumannii [39], whereas GacSA acts as a global virulence regulator and involves pili formation, motility and biofilm structure [40]. Nevertheless, the role of A. baumannii TCSs in disposing environmental chemical compounds has not been clarified.
Similar to other TCSs, BaeSR can detect environmental signals and respond by altering the bacterial envelope [25]. We have shown that the BaeSR regulons not only respond to high osmotic stress but also influence the tigecycline susceptibility of A. baumannii through positively regulating the RND efflux pump genes adeA and adeB [19]. In this study, an additional two transport systems, AdeIJK and MacAB-TolC, may have also been regulated by BaeSR because baeR deletion reduced gene expression in the two pump systems. This result is consistent with transcriptional data from an LPS-deficient A. baumannii strain, which showed the increased expression of BaeSR, AdeIJK and MacAB-TolC [30].
Tigecycline is a glycycline and is one of the few available effective antibiotics for MDRAB infections [41]. Several previous studies have demonstrated that tigecycline resistance is mainly facilitated through efflux pumps, including AdeABC [7], AdeIJK [10] and AdeFGH [9]. The AdeABC and AdeIJK efflux systems' tigecycline resistance is greater than an additive contribution [10]. In this study, qRT-PCR data also demonstrated a potential role for AdeIJK and MacAB-TolC in tigecycline resistance of the laboratory-induced tigecycline-resistant A. baumannii strains (ABtc and ABhl1tc). In 2011, Coyne et al. declared that AdeABC was the only system involved in clinical isolate tigecycline resistance [42]. Besides, AdeIJK is considered to play a role in the intrinsic low-level resistant phenotype of A. baumannii [10] and over-expression of this pump in A. baumannii is toxic, suggesting the presence of a tight regulation mechanism to maintain low expression levels of AdeIJK [43] in ABhl1 compared with the wild type strain.
In addition to conferring clinically relevant resistance to antibiotics, these multidrug-resistance efflux pumps encoded by bacteria can also confer resistance to natural substances produced by the host [13]. To elucidate the TCS BaeSR response after being exposed to environmental chemical compounds through the efflux pumps, we used a phenotypic microarray. A. baumannii was vulnerable to certain chemicals, especially tannic acid, after baeR deletion.
Tannins comprise a large group of natural products distributed in the vegetable kingdom and has been classified into two groups, condensed and hydrolysable [44]. Tannic acid has long been used as a topical agent for burn wounds. Tannic acid was further proposed as an adjuvant therapy with β-lactam antibiotics for Staphylococcus aureus infections [45]. Susceptibility of MDRAB to a variety of antibiotics was also enhanced in the presence of tannic acids [46]. The antioxidant capacity and antimicrobial activity of tannic acid can be enhanced by thermal processing [47]. Despite tannin antimicrobial activities, many tannin-resistant bacteria have been isolated. However, the mechanisms underlying this resistance remain unclear. The TCS BaeSR has been thought to mediate tannic acid resistance through up-regulating the multidrug transporter-encoding operon mdtABCD in E. coli [26]. In our previous and present studies, we found that both tigecycline and tannic acid resistance are associated with the BaeR regulon in A. baumannii using qRT-PCR data. baeR gene deletion influenced gene expression for the AdeAB, AdeIJK and MacAB-TolC pump systems. Increased gene expression of adeB, adeJ, and macB after being exposed to tigecycline or high concentration of tannic acid was also demonstrated. These results suggested a modified role for BaeR, as Appia-Ayme et al. proposed, in up-regulating certain pump systems in response to specific envelop damaging agents as a belt and braces approach to protect the cell through waste disposal [23].
As shown in Fig 4B, the baeR gene expression did not increase upon exposure to 50 μg/mL tannic acid, whereas macB expression showed a slight increase. In addition, similar to the baeR gene, the adeB and adeJ gene expression was also no change in the presence of 50 μg/mL tannic acid. These results suggest that macB gene expression may be controlled by a regulator other than BaeR. Interestingly, the expression of adeB, adeJ and macB pump genes was increased in the baeR-deletion background with 50 μg/mL tannic acid, whereas decreased with 100 μg/mL tannic acid (Fig 4C). One possible explanation is that the A. baumannii efflux pumps were mainly controlled by local regulators (e.g., AdeRS controls adeAB and AdeN controls adeIJK) upon exposure to low tannic acid concentrations. However, the BaeR regulon includes these pump genes at higher tannic acid concentrations. This hypothesis highlights the complex transcription regulatory networks in A. baumannii and requires further study. Moreover, EMSAs of the interaction between BaeR-His6 and the adeA, adeI and macA promoter regions did not exhibit direct binding, which implies that their mutual interaction is possibly indirect. One previous paper indicates that BaeSR is part of a cross-regulation system that includes the PhoBR and CreBC TCSs in E. coli [29]; it suggests that constructing the BaeSR mutants may not only affect the expression levels of genes under direct control of BaeSR but may disrupt a complex regulatory network.
Despite the few results for TCS in A. baumannii, it remains a possible target for therapeutics, which is supported by the conclusion from Gram-positive pathogenic bacteria [48]. A class of antibacterial agents that inhibit two-component signal transduction systems was developed in the laboratory and may represent a breakthrough in antibacterial therapy [49]. Moreover, a small molecule adjuvant was found capable of suppressing colistin resistance in MDRAB by interfering with the expression of TCS PmrCAB [50]. In conclusion, A. baumannii can use the TCS BaeSR in disposing chemicals, such as tannic acid and tigecycline, through regulating efflux pumps without direct DNA binding. Our findings may help to understand TCS and efflux pumps-associated antimicrobial resistance mechanisms and provide a basis for the future development of antimicrobials against drug-resistant A. baumannii.