Functional Characterization of a Novel SMR-Type Efflux Pump RanQ, Mediating Quaternary Ammonium Compound Resistance in Riemerella anatipestifer

Riemerella anatipestifer (R. anatipestifer) is a multidrug-resistant bacterium and an important pathogen responsible for major economic losses in the duck industry. Our previous study revealed that the efflux pump is an important resistance mechanism of R. anatipestifer. Bioinformatics analysis indicated that the GE296_RS02355 gene (denoted here as RanQ), a putative small multidrug resistance (SMR)-type efflux pump, is highly conserved in R. anatipestifer strains and important for the multidrug resistance. In the present study, we characterized the GE296_RS02355 gene in R. anatipestifer strain LZ-01. First, the deletion strain RA-LZ01ΔGE296_RS02355 and complemented strain RA-LZ01cΔGE296_RS02355 were constructed. When compared with that of the wild-type (WT) strain RA-LZ01, the mutant strain ΔRanQ showed no significant influence on bacterial growth, virulence, invasion and adhesion, morphology biofilm formation ability, and glucose metabolism. In addition, the ΔRanQ mutant strain did not alter the drug resistance phenotype of the WT strain RA-LZ01 and displayed enhanced sensitivity toward structurally related quaternary ammonium compounds, such as benzalkonium chloride and methyl viologen, which show high efflux specificity and selectivity. This study may help elucidate the unprecedented biological functions of the SMR-type efflux pump in R. anatipestifer. Thus, if this determinant is horizontally transferred, it could cause the spread of quaternary ammonium compound resistance among bacterial species.


Introduction
Riemerella anatipestifer (R. anatipestifer) is a rod or oval-shaped, spore-free, and flagellafree gram-negative bacteria of the genus Riemerella, family Weeksellaceae, order Flavobacterium [1]. The bacteria mainly infect the respiratory tract, skin wounds, or digestive tract of 1-5-weeks-old ducklings and can cause acute or chronic septicemia and serositis in ducks [2]. There are 21 recognized serotypes of R. anatipestifer, but there is no cross-protection between them, making vaccination more challenging [3]. Consequently, antibiotics have become an important means for preventing and treating R. anatipestifer infections. This would greatly increase the emergence and spread of multiple drug-resistant strains too. In addition, R. anatipestifer has a natural resistance to a variety of antibiotics, including aminoglycosides, macrolides, cephalosporins, tetracyclines, lincosamides, amino cyclic alcohols, sulfonamides, and polymyxin [4]. For this reason, it is essential to investigate mechanisms of multidrug resistance (MDR) in R. anatipestifer in order to prevent resistance from spreading and resistance from increasing.
The drug efflux pump system is a major mechanism that mediates innate and acquired drug resistance in bacteria. The study of the efflux pump enables us to better understand

Cloning of RanQ in Deleted E. coli ATCC35150 for Heterologous Studies
Genome sequence of E. coli ATCC35150 (accession number GCA_013168075.1) was retrieved from the GenBank database and used to design guide RNA and an editing template. The acrB, ydhE and hsd genes of E. coli ATCC35150 were knocked out using CRISPR/Cas9 technology as described elsewhere with slight modifications [19]. Briefly, we first constructed plasmids pTargetF-acrB sgRNA, pTargetF -YdhE sgRNA, and pTargetF-hsd sgRNA vectors encoding Cas9 endonuclease-guided RNA using primers acrB sg, YdhE sg, and hsd sg, respectively. Then left and right homologous arms of acrAB, ydhE, and hsd genes were amplified from the genome of E. coli ATCC35150 using primers acrAB-up-F/R, acrAB-down-F/R, ydhE-up-F/R, ydhE-down-F/R, hsd-up-F/R, and hsd-down-F/R, respectively. The homologous arms were fused by fusion PCR. Then E. coli ATCC35150 competent cells containing pCas were prepared. The recombinant plasmid and fusion fragment were electroporated into the competent cells and cultured in LB agar containing kanamycin (50 mg/L) and ampicillin (50 mg/L) at 30 • C overnight. The positive clones were identified by colony PCR. Finally, two plasmids were eliminated to obtain E. coli ATCC35150 gene deletion strains ∆acrB, ∆ydhE, and ∆hsd. (Primer sequences are shown in Table 2.) Table 2. This study used qRT-PCR and PCR primer.

Primers
Sequences (  The RanQ is located from nucleotide 511,851 to nucleotide 512,177 in the genome sequence of R. anatipestifer RA-LZ01. The putative efflux genes RanQ and the entire SMR operon were amplified by a standard PCR protocol using the primer pairs RanQ-p-F/RanQp-R (Table 2), and cloned into the EcoRI and PstI (TaKaRa, Dalian, China) sites of pUC18. The resulting recombinant plasmid pRanQ was transformed into deleted E. coli ATCC35150 for functional characterization.

Construction of Knockout Strain
In Table 2, we list all primers that were used in this study. Genomic DNA was extracted from LJW-2 strain using the TIANamp bacterial genome extraction kit (TIANGEN, Beijing, China), and the operation method is shown in the instruction manual. Using genomic DNA of LJW-2 strain as template, erythromycin resistance gene Erm was amplified by Erm-F/Erm-R primers. RanQ-UpF/RanQ-Up-R and RanQ-Do-F/RanQ-Do-R primers were used to amplify the upstream and downstream homologous arms of RanQ gene based on the genomic DNA of the RA-LZ01 strain. Then, the upstream and downstream fragments and Erm resistance gene were fused by fusion PCR to obtain the target DNA fragment of 2001 bp. The fused DNA fragment was linked to the ZERO vector (TransGen, Beijing, China) and sequenced by a professional sequencing company (QINGKE, Xi'an, China) to determine the exact nucleotide sequence. The fusion fragment was digested with restriction endonuclease Sac| and Sph| (TaKaRa, Dalian, China), and then ligated to the pRE112 plasmid similar to restriction endonuclease T4 DNA ligase (TaKaRa, Dalian, China) to produce the recombinant plasmid pRE112-RanQ. We selected E. coli X7213 strains carrying pRE112-RanQ on LB agar plates with Cm and DPA (25 µg/mL and 50 µg/mL, respectively). A conjugation procedure was then performed to introduce the recombinant plasmids into the R. anatipestifer RA-LZ01 strain [20]. The positive strains screened on the plate containing 5% calf serum and Erm (1 µg/mL) were identified by primers OmpA-F/OmpA-R, RanQ-F/RanQ-R, and Erm-F/Erm-R. Mutant strains resulting from gene deletions were designated RA-LZ01∆RanQ.

Construction of Complemented Strain
In our previous study, we have referred to the literature and constructed the shuttle expression vector pCPRA using pCP29 [21,22]. Using the parent strain RA-LZ01 as the template, the RanQ gene with Pst| and Sph| digested sites was amplified by primer RanQ-Co-F/RanQ-Co-R and ligated with T4 DNA ligase to the pCPRA vector digested with the same restriction endonuclease to construct the recombinant plasmid pCPRA-RanQ. Thw X7213 strain was then transfected with recombinant plasmid pCPRA-RanQ. The positive strain X7213-Pcpra-RanQ was screened on an LB solid plate containing ampicillin (100 µg/mL) and DPA (50 µg/mL). By conjugation, the pCPRA-RanQ plasmid from X7213 was introduced into the RA-LZ01∆RanQ strain [20]. The strains on the TSA plate containing 5% calf serum and CFX (1 µg/mL) were identified by primers RA-OmpA-F/RA-OmpA-R and RanQ-F/RanQ-R. The screened positive strain was named RA-LZ01c∆RanQ.

Antimicrobial Susceptibility Testing
To assess the resistance profile of RA-LZ01, deletion mutants, and the corresponding complemented strains, antimicrobial agents were routinely tested with 2-fold serial broth microdilutions to determine their minimal inhibitory concentrations (MICs). The operation procedure is carried out with reference to the literature [22]. We used E. coli ATCC 25,922 as a control and antimicrobial agents were present in concentrations ranging from 512 to 0.25 µg/mL.

Quaternary Ammonium Salt Tolerance Assay
To further investigate the tolerance of RA-LZ01, ∆RanQ, and c∆RanQ to quaternary ammonium salts, we determined the colony-forming ability of the strains in TSA plates supplemented with different types of quaternary ammonium salts at a final concentration of 3.0 µg/mL, according to the reference [23]. Briefly, the bacteria were cultured in the TSB medium to OD 600 = 1.0. A suspension adjusted to 10 8 CFU/mL was serially diluted 10 times to 10 3 CFU/mL. Then add 6 µ L of each diluted bacterial droplet to the TSA plate with quaternary ammonium salt and dry it. The results were observed after 18 h of inverted culture in a 5% CO 2 incubator at 37 • C.

Quantitative Real-Time RT-PCR
The strains were inoculated in 10 mL of TSB medium containing half MIC of each quaternary ammonium salt and cultured to OD 600 = 1.0. The total RNA of the strain was extracted with the bacterial total RNA extraction kit (TIANGEN, Beijing, China), and the RNA concentration was measured by ultraviolet spectrophotometer. Total RNA was reverse transcribed into cDNA using a reverse transcription kit (Vazyme, Nanjing, China) as a template for qRT-PCR. The total volume of the reaction was 20 µL, including 10 µL of 2× Master Mix, 0.4 µL of upstream and downstream primers, 10 ng of cDNA content, and finally water was added to 20 µL. Each sample was in triplicate. Amplification protocol consisted of an initial denaturation step at 95 • C for 30 s and 40 cycles at 95 • C for 10 s and 60 • C for 30 s. The RecA gene was used as an internal reference gene [4]. The relative gene expression levels were quantified according to the comparative 2 −∆∆CT method [24].
In addition, in order to evaluate the effects of BAC and MV alone and in combination with efflux pump inhibitors (EPIs) on RanQ gene activity, we analyzed the effects of quaternary ammonium salts alone or in combination with inhibitors PAβN (40 µg/mL) and CCCP (5 µg/mL) on RanQ gene expression. The relative expression of the RanQ gene was assessed by comparing the relative quantity of the respective mRNA in the presence of the BAC and MV, and the quaternary ammonium salt + sub-MIC concentration of the inhibitor, with those of the nonexposed strain. The method is the same as the above operation.

Bacterial Growth Curves
In order to explore whether the deletion of the RanQ gene affects the growth of RA-LZ01, we monitored the growth of RA-LZ01, ∆RanQ, and c∆RanQ, and recorded the OD 600 value every 1 h for 20 h [25]. The same method was used to determine the growth of RA-LZ01 when BAC (1 µg/mL) and methyl viologen (MV) (1 µg/mL) were used alone or in combination with the inhibitor PAβN dihydrochloride (PAβN) (40 µg/mL) and Carbonyl cyanide 3-chlorophenylhydrazone (CCCP) (5 µg/mL) to evaluate the role of the RanQ gene in the efflux of BAC and MV.

Pathogenicity Test
RA-LZ01, ∆RanQ, and c∆RanQ strains were cultured in TSB to logarithmic growth phase, and then their colony forming units (CFU) were measured. Then, the initial concentration of each strain was adjusted to 10 9 CFU/mL. Each 8-day-old duckling was challenged with an intramuscular injection of 0.1 mL bacterial suspension in the thigh muscle using a standard needle (26 gauge). The ducklings were randomly divided into 4 groups, 10 in each group. The control group was intramuscularly injected with the same amount of sterile saline [25]. After infection, continuous observation was performed for 14 days and the survival curve was drawn.

Bacterial Adherence and Invasion Assay
An adherence assay was conducted on the parent strain RA-LZ01 and the RanQ mutant to determine if deletion of the RanQ gene would affect adherence of R. anatipestifer [26,27]. We grew DBMECs to 95% confluence in 48-well cell culture plates. The cells of each well were infected with 10 8 CFU of ∆RanQ mutant or RA-LZ01 bacteria in an Endothelial Cell Medium (ECM), and incubated for 2 h at 37 • C in an atmosphere containing 5% CO 2 . In order to remove nonadherent bacteria, DBMECs were rinsed with sterile PBS 3 times and then digested with 500 µL 0.25% trypsin. This cell suspension was a 10-fold series diluted with PBS and coated on the TSA panels containing 5 µg/mL polymyxin B to determine the number of viable bacteria. In the invasion experiment, bacterial infection was established and bacteria were counted as described above for the bacterial adherence assay, except that during the 3 h incubation for bacterial infection, the extracellular bacteria were killed by incubating the monolayers with the ECM medium containing ampicillin (100 µg/mL) for another 1 h and washed thrice with PBS. At least three independent assays were conducted on separate days in duplicate.

Biofilm Quantification
An assay involving crystal violet (CV) staining was used to quantify biofilm formation by the RA-LZ01 strain [28]. Briefly, the strains were cultured overnight in TSB. It was then adjusted to 0.1 under OD 655 , and 200 µL was transferred to a 96-well microtiter plate (Corning, NY, USA). The cells were incubated in a 5% CO2 incubator at 37 • C for 6, 12, 24, and 48 h. The cell suspension was discarded and gently washed with PBS 3 times, and then stained with 0.1% CV at room temperature for 30 min. They were then rinsed with distilled water 4 times, dried, and added 100 µL of 95% ethanol to dissolve the crystal violet attached to the wall. The optical density at 595 nm (OD 595 ) was determined using a microplate reader (BioTek, Winooski, VT, USA).
The experiment was repeated three times and all samples were measured in triplicate; the mean 1 standard deviation for each sample was calculated from three independent experiments.

Observation under Transmission Electron Microscope
The overnight cultures of RA-LZ01 and ∆RanQ were collected after centrifugation at 5000× g for 10 min and washed twice with sterile PBS. After the bacterial solution was re-suspended, 10 µL of the bacterial suspension was uniformly coated on the copper net of the formvar film and dried for 10 min. Then, the formvar membrane was stained with 5 µL uranylacetate dihydrate acid for 1 min. Using a transmission electron microscope, the samples were observed and analyzed.

Glucose Metabolism Experiment
The WT strain RA-LZ01, mutant strain ∆RanQ, and c∆RanQ were grown until the OD 600 value was about 1.0, and diluted to 10 8 CFU/mL with 0.85% sterile saline. The bacterial suspensions (0.08 mL) were added to each trace of the cillin bottles and incubated at 37 • C in a CO 2 incubator for 24 h. The results were then interpreted based on specifications.
For the triple sugar iron agar test, a well-isolated colony from the nutrient agar plate was taken using an inoculation needle and inoculated in the triple sugar iron agar, first by stabbing to the bottom of the tube through the center of the medium and then by streaking the surface of the slant. The lid was half-plugged and the tube was incubated at 37 • C for 24 h and further observed for any change in the color of agar.

Statistical Analysis
GraphPad Prism version 6.0 Windows software was used to perform the statistical analysis. Students' t-test was used to determine if the difference between the two data sets was significant; a value of p < 0.05 was considered significant.

Ethics Statement
All animals were handled by following strict procedures according to the Care and Use of Laboratory Animals published by the Institute of Laboratory Animal Resources (Reference No. LVRIAEC-2020-019). All animal experimental procedures were approved by the Institutional Animal Care and Use Committee of Lanzhou Veterinary Research Institute.

Sequence Analysis of RanQ
Bioinformatics analysis of RanQ (WP_004920112.1), annotated as a multidrug efflux SMR transporter in R. anatipestifer strain RA-LZ01 (NCBI Reference Sequence: NZ_CP045564.1), indicated an open reading frame (ORF) of 327 nucleotides. According to the deduced amino acid sequence, the protein consists of 108 residues and has a molecular mass of 11.59 kDa and a theoretical isoelectric point (pI) of 8.85. The amino acid sequence of RanQ alignment, as per the results from protein-protein Basic Local Alignment Search Tool (BLASTP), indicated over 98% identity among different R. anatipestifer strains. Based on the predictions of its secondary structure and transmembrane topology, RanQ is composed of three helical transmembrane segments, with the N-termini located in the periplasm and the C-termini located in the cytoplasm ( Figure 1A). Support for this structure came from an independent analysis that revealed the three-dimensional (3D) structure of the protein ( Figure 1B). The predicted product of the RanQ gene exhibited low amino acid identity and similarity (24% and 38%, respectively) with other SMR transporters involved in the drug efflux in gram-negative bacteria, such as E. coli, Bordetella pertussis, Klebsiella pneumoniae, and Acinetobacter baumannii. These data indicated that RanQ is a new multidrug efflux SMR transporter, which differs from the alreadyknown SMR transporters by the presence of three rather than four transmembrane segments.

Reduced QACs Susceptibility by RanQ
The WT RA-LZ01 and deletion strain ∆RanQ were tested for susceptibility to forty-two antimicrobial agents belonging to twelve classes with dissimilar structures. The results showed that the ∆RanQ strain significantly increased the resistance to macrolide antibiotics because the strain carried an erythromycin-resistant cassette. Compared with the RA-LZ01 strain, the ∆RanQ strain only increased susceptibility to MV (two-fold) and BAC (fourfold), and no change was observed in other QACs (Benzyl dimethyl tetradecyl ammonium chloride (TDBAC), Dodecyl trimethyl ammonium chloride (DTAC), and Benzyl cetyl dimethyl ammonium chloride (HDBAC)). According to these results, the RanQ protein is mainly responsible for QACs resistance (Table 3).
Preliminary experiments showed that RA-LZ01 grew well in TSB medium supplemented with 3 µg/mL MV, BAc, TDBAC, DTAC, and HDBAC. Therefore, we performed spot tests on TSA plates supplemented with 3 µg/mL QACs. The results showed that all strains could form colonies on plates without QACs. RA-LZ01 and c∆RanQ could form colonies in the presence of all QACs, while the deletion strain ∆RanQ could only form a small number of colonies in the presence of MV and BAC, and the colony formation efficiency was significantly lower than that of the RA-LZ01 and c∆RanQ strains (Figure 2). These results conveyed that the RanQ gene contributed to the resistance of QACs. lecular mass of 11.59 kDa and a theoretical isoelectric point (pI) of 8.85. The amino acid sequence of RanQ alignment, as per the results from protein-protein Basic Local Alignment Search Tool (BLASTP), indicated over 98% identity among different R. anatipestifer strains. Based on the predictions of its secondary structure and transmembrane topology, RanQ is composed of three helical transmembrane segments, with the N-termini located in the periplasm and the C-termini located in the cytoplasm ( Figure 1A). Support for this structure came from an independent analysis that revealed the three-dimensional (3D) structure of the protein ( Figure 1B). The predicted product of the RanQ gene exhibited low amino acid identity and similarity (24% and 38%, respectively) with other SMR transporters involved in the drug efflux in gram-negative bacteria, such as E. coli, Bordetella pertussis, Klebsiella pneumoniae, and Acinetobacter baumannii. These data indicated that RanQ is a new multidrug efflux SMR transporter, which differs from the already-known SMR transporters by the presence of three rather than four transmembrane segments.

Reduced QACs Susceptibility by RanQ
The WT RA-LZ01 and deletion strain ΔRanQ were tested for susceptibility to forty-two antimicrobial agents belonging to twelve classes with dissimilar structures. The results showed that the ΔRanQ strain significantly increased the resistance to macrolide antibiotics because the strain carried an erythromycin-resistant cassette. Compared with the RA-LZ01 strain, the ΔRanQ strain only increased susceptibility to MV (two-fold) and BAC (four-fold), and no change was observed in other QACs (Benzyl dimethyl tetradecyl ammonium chloride (TDBAC), Dodecyl trimethyl ammonium chloride Growth of parent strain RA-LZ01, deletion strain RA-LZ01ΔranQ, and complemented strain RA-LZ01cΔRanQ were determined by spot assay on TSA plates containing the same concentration of QACs. Remarkably, the deletion strains were grown on TSA plates including 5% calf serum, and a significant difference was observed in the presence of BAC and MV. A representative result is given from the three independent experiments.

Characterization of RanQ in Deleted E. coli ATCC35150
The minimum inhibitory concentrations (MICs) for RanQ-harboring cells (deleted E. coli ATCC35150/pRanQ) were higher (MV (4-fold), BAC (8-fold), DTAC (2-fold), TDBAC (1-fold), and HDBAC (1-fold)) as compared to those for deleted E. coli ATCC35150/pUC18 (Table 4). It is significant to note that pRanQ alone showed a change Growth of parent strain RA-LZ01, deletion strain RA-LZ01∆ranQ, and complemented strain RA-LZ01c∆RanQ were determined by spot assay on TSA plates containing the same concentration of QACs. Remarkably, the deletion strains were grown on TSA plates including 5% calf serum, and a significant difference was observed in the presence of BAC and MV. A representative result is given from the three independent experiments.

Up-Regulation of RanQ Gene Transcriptions by QACs
Subinhibitory concentrations (2 µg/mL) of BAc, DTAC, TDBAC, HDBAC, and MV (8 µg/mL) were added to TSB to evaluate whether RanQ gene transcription was regulated by antibiotics. Induced by antimicrobial agents, the transcription levels of genes were measured using the qRT-PCR assay. The results indicated that the transcription levels of the RanQ gene were up-regulated from 2.5-to 2.8-fold and 2.0-to 2.2-fold ( Figure 3A), respectively, after induction with BAC and MV. However, there was no significant difference in the transcription level of the RanQ gene in wild geese treated with the other three kinds of QACs. When RA-LZ01 was exposed to BAC and MV, the expression of the RanQ gene increased significantly. When PAβN was added, the RanQ gene expression did not change significantly. However, the expression of the RanQ gene was significantly reduced by the addition of CCCP ( Figure 3B).
Microorganisms 2023, 11, x FOR PEER REVIEW 11 of 17 measured using the qRT-PCR assay. The results indicated that the transcription levels of the RanQ gene were up-regulated from 2.5-to 2.8-fold and 2.0-to 2.2-fold ( Figure 3A), respectively, after induction with BAC and MV. However, there was no significant difference in the transcription level of the RanQ gene in wild geese treated with the other three kinds of QACs. When RA-LZ01 was exposed to BAC and MV, the expression of the RanQ gene increased significantly. When PAβN was added, the RanQ gene expression did not change significantly. However, the expression of the RanQ gene was significantly reduced by the addition of CCCP ( Figure 3B). Error bars indicate standard deviation (n = 3). A representative result is given from the three independent experiments.

RanQ Gene Had No Effect on the Growth of Parent Strains
From the growth curve, When RA-LZ01 was exposed to 1 µg/mL BAC ( Figure 4A) and MV ( Figure 4B), it did not affect the growth of RA-LZ01. When CCCP was added, the growth of RA-LZ01 was significantly slowed down, whereas no significant change in the growth of RA-LZ01 was observed when PAβN was added (Figure 4). But the growth rate of the deletion strain ΔRanQ and the replenishing strain cΔRanQ was not significantly different compared to the RA-LZ01 strain ( Figure 5A).  Error bars indicate standard deviation (n = 3). A representative result is given from the three independent experiments.

RanQ Gene Had No Effect on the Growth of Parent Strains
From the growth curve, When RA-LZ01 was exposed to 1 µg/mL BAC ( Figure 4A) and MV ( Figure 4B), it did not affect the growth of RA-LZ01. When CCCP was added, the growth of RA-LZ01 was significantly slowed down, whereas no significant change in the growth of RA-LZ01 was observed when PAβN was added (Figure 4). But the growth rate of the deletion strain ∆RanQ and the replenishing strain c∆RanQ was not significantly different compared to the RA-LZ01 strain ( Figure 5A).

RanQ Gene Had No Effect on the Growth of Parent Strains
From the growth curve, When RA-LZ01 was exposed to 1 µg/mL BAC ( Figure 4A) and MV ( Figure 4B), it did not affect the growth of RA-LZ01. When CCCP was added, the growth of RA-LZ01 was significantly slowed down, whereas no significant change in the growth of RA-LZ01 was observed when PAβN was added (Figure 4). But the growth rate of the deletion strain ΔRanQ and the replenishing strain cΔRanQ was not significantly different compared to the RA-LZ01 strain ( Figure 5A).  represents the mean ± standard for triplicate samples. (**** p < 0.0001).

Pathogenicity Test
The results of pathogenicity experiments showed that: On the 14th day, the survival rate of WT RA-LZ01 and supplementary strain cΔRanQ was 20%, and that of deleted strain ΔRanQ was 25% ( Figure 5B). These data revealed that RanQ was not involved in the virulence of the RA-LZ01 strain.

RanQ Gene Is Not Involved in the Biofilm Formation of R. anatipestifer
To determine whether the RanQ gene plays a key role in R. anatipestifer biofilm formation, the biofilm-forming ability of the parent strain RA-LZ01, deletion strain ΔRanQ, and complement strain cΔRanQ were investigated at 6 h, 12 h, 24 h, and 48 h. Although there was a difference in the biofilm formed at 6 h by the deleted strain and the parent strain, no significant differences were observed at 12 h, 24 h, and 48 h among the three strains ( Figure 5C). This confirmed that the RanQ gene was not involved in the biofilm formation of R. anatipestifer.

Effect of Deletion of RanQ gene on the RA-LZ01 Morphology of WT
Using a transmission electron microscope, we observed the morphology of the RA-LZ01 and ΔRanQ strains. After negative staining with uranyl acetate, the WT strain RA-LZ01 and mutant strain ΔRanQ were observed microscopically to have capsules that completely encompass the cells and exhibit rugose formation ( Figure 5D). These data implied that deletion of the RanQ gene does not affect the RA-LZ01 morphology of WT.

Pathogenicity Test
The results of pathogenicity experiments showed that: On the 14th day, the survival rate of WT RA-LZ01 and supplementary strain c∆RanQ was 20%, and that of deleted strain ∆RanQ was 25% ( Figure 5B). These data revealed that RanQ was not involved in the virulence of the RA-LZ01 strain.

RanQ Gene Is Not Involved in the Biofilm Formation of R. anatipestifer
To determine whether the RanQ gene plays a key role in R. anatipestifer biofilm formation, the biofilm-forming ability of the parent strain RA-LZ01, deletion strain ∆RanQ, and complement strain c∆RanQ were investigated at 6 h, 12 h, 24 h, and 48 h. Although there was a difference in the biofilm formed at 6 h by the deleted strain and the parent strain, no significant differences were observed at 12 h, 24 h, and 48 h among the three strains ( Figure 5C). This confirmed that the RanQ gene was not involved in the biofilm formation of R. anatipestifer.

Effect of Deletion of RanQ gene on the RA-LZ01 Morphology of WT
Using a transmission electron microscope, we observed the morphology of the RA-LZ01 and ∆RanQ strains. After negative staining with uranyl acetate, the WT strain RA-LZ01 and mutant strain ∆RanQ were observed microscopically to have capsules that completely encompass the cells and exhibit rugose formation ( Figure 5D). These data implied that deletion of the RanQ gene does not affect the RA-LZ01 morphology of WT.

Adhesion and Invasion Assay
The adhesion and invasion experiments showed that there was no significant difference in the adhesion and invasion ability of the deleted strain ∆RanQ to DBMECs as compared with that of WT RA-LZ01 and c∆RanQ ( Figure 6A,B). Thus, it can be inferred that RanQ was not involved in the interaction between R. anatipestifer and the host cells.

Adhesion and Invasion Assay
The adhesion and invasion experiments showed that there was no significant difference in the adhesion and invasion ability of the deleted strain ΔRanQ to DBMECs as compared with that of WT RA-LZ01 and cΔRanQ ( Figure 6A,B). Thus, it can be inferred that RanQ was not involved in the interaction between R. anatipestifer and the host cells.

Effect of Deletion of RanQ Gene on the RA-LZ01 Glucose Metabolism
It was found that RA-LZ01 could utilize glucose, fructose, and maltose, but not lactose and fructose. The deletion of the RanQ gene did not affect the glucose metabolism of WT RA-LZ01 ( Figure 7A). The triple sugar iron agar test showed a color change from purplish red to yellow, indicating that all three strains could produce acid and make the culture medium yellow ( Figure 7B).

Effect of Deletion of RanQ Gene on the RA-LZ01 Glucose Metabolism
It was found that RA-LZ01 could utilize glucose, fructose, and maltose, but not lactose and fructose. The deletion of the RanQ gene did not affect the glucose metabolism of WT RA-LZ01 ( Figure 7A). The triple sugar iron agar test showed a color change from purplish red to yellow, indicating that all three strains could produce acid and make the culture medium yellow ( Figure 7B).

Effect of Deletion of RanQ Gene on the RA-LZ01 Glucose Metabolism
It was found that RA-LZ01 could utilize glucose, fructose, and maltose, but not lactose and fructose. The deletion of the RanQ gene did not affect the glucose metabolism of WT RA-LZ01 ( Figure 7A). The triple sugar iron agar test showed a color change from purplish red to yellow, indicating that all three strains could produce acid and make the culture medium yellow ( Figure 7B).

Discussion
In gram-negative bacteria, SMR efflux pumps play a significant role in MDR. Drug transport in the SMR family requires proton exchange to help drive drug efflux through a series of conformational changes. So far, the reporter EmrE in E. coli has been the most studied in gram-negative bacteria [29], followed by Acinetobacter baumannii [17], Klebsiella pneumoniae [16], Bordetella pertussis [15], etc. In this study, the genome sequence analysis suggested that the RanQ gene in R. anatipestifer RA-LZ01 encodes a putative SMR efflux pump, and further investigations on the biological function of RanQ in R. anatipestifer RA-LZ01 were carried out.
By adding efflux substrate, the expression of efflux pump genes was induced, such as the raeB gene in R. anatipestifer CH-1 [30]. The parent strain RA-LZ01 was induced by TDBAC, DTAC, HDBAC, MV, and BAC, and then fluorescence quantitative real-time RT-PCR was performed. We found that the transcriptional level of the RanQ gene was up-regulated by 1.1-2.8 times, indicating that these QACs were the efflux substrates of the RA-LZ01 strain RanQ gene.
It has been reported that multidrug efflux pumps play a role in the diverse phenotypes of bacteria, including growth and virulence, such as MmpL11 in Mycobacterium tuberculosis [31], AcrB in Salmonella typhimurium [32], and AcrB in Klebsiella pneumoniae [33]. However, the deletion of the RanQ gene did not affect the growth and virulence of RA-LZ01, indicating that the RanQ gene was not involved in the growth and virulence of R. anatipestifer. Further the adhesion and invasion abilities of the mutant strain to DBMECs were analyzed and the results manifested that the adhesion and invasion capacities of the ∆RanQ were not significantly varied compared to those of the WT strain RA-LZ01. At the same time, biochemical experiments showed that the RanQ gene did not play a role in the glucose metabolism of R. anatipestifer.
As another phenotype indicating bacterial pathogenicity, biofilms are communities of microorganisms surrounded by extracellular polymeric matrixes. Numerous pathogens, such as E. coli, Streptococcus suis, and Haemophilus influenzae, form biofilms [34]. A biofilm provides bacterial protection from biotic and abiotic stresses in most hostile environments [35]. As part of our previous analysis of our R. anatipestifer collection, we found that RA-LZ01 does not produce a strong biofilm. The disruption of RanQ does not affect the formation of biofilms, as we found in the biofilm quantification experiment. The pathogenicity of ∆RanQ strain was not significantly different from that of the wild strain RA-LZ01, which was consistent with the observation that the mutant strain could not enhance the biofilm formation of the strain.
The importance of antimicrobial compounds has been highlighted recently due to the COVID-19 pandemic outbreak [36]. QACs are antiseptics that have shown a wide spectrum of antibacterial activity. Many studies have been carried out involving the applications of QACs as antifouling agents for the inhibition of biofilm growth on medical implants and as antibacterial agents on surfaces and aquatic environments [37]. Since the outbreak of COVID-19, QACs are the most widely used disinfectants. The expression of RanQ in the heterologous host E. coli ATCC35150 caused decreased susceptibility to several agents, such as MV and BAC. It is suggested that the QACs resistance mediated by the efflux pump should be taken seriously.
Efflux pumps are transmembrane transporters, expressed in all types of prokaryotes and eukaryotes with wide substrate specificity that have common amphipathic character and ionizable groups [38]. The SMR efflux pump mediates resistance to antibiotics and compounds such as acriflavine, BAC, Betaine, chloramphenicol, cetyl trimethylamine, crystal violet, ethidium bromide, MV, tetraphenyl phosphine chloride, and so on. In this study, the MIC values of the strains to different antibiotics were determined, and it was found that the deletion strain ∆RanQ increased susceptibility to QACs MV and BAC, thereby indicating that the RanQ gene mediated resistance to QACs in the RA-LZ01 strain. Our observations corroborated well with the previously defined functions for SMR-type efflux pumps, such as in E. coli emrE [14] (betaine and choline), Acinetobacter baumannii AbeS [17] (deoxycholate, sodium dodecyl sulfate, acriflavine, BAC), Bacillus subtilis EbrAB [39] (ethidium bromide), and Staphylococcus aureus QacG [40] (acriflavine, BAC, Cetyl trimethylamine, and ethidium bromide). Reports on the role of SMR-type efflux pumps in antimicrobial resistance have been demonstrated in a few clinically significant pathogens; however, its function has never been investigated in R. anatipestifer, thus far. All these data suggest that SMR-type efflux pumps have multiple different physiological functions in different species of bacteria. Overall, we report here for the first time the single biological characterization of an SMR-type RanQ efflux system for specific efflux of QACs in R. anatipestifer.