The Role of AcrAB-TolC Efflux Pump in Mediating Fluoroquinolone Resistance in Naturally Occurring Salmonella Isolates from China

Abstract

The involvement of AcrAB-TolC efflux pump in regulating fluoroquinolone resistance of naturally occurring Salmonella isolates is insufficiently investigated. In this study, the regulatory genes, acrR, ramR, marRAB, and soxRS of AcrAB-TolC efflux pump, of 27 naturally occurring fluoroquinolone-resistant Salmonella isolates collected in China were sequenced. The expression levels of acrB, ramA, marA, and soxS were also examined using quantitative real-time polymerase chain reaction. Gene alterations were mainly observed for acrR (three mutation types) and ramR (four mutation types), not for marRAB (no mutation) or soxRS (one mutaton type). Overexpressions were also mainly observed for acrB and ramA, not for marA or soxS. Some mutations/deletions in ramR caused highly elevated expression of ramA. Complementation with wild-type ramR gene reduced mRNA levels of acrB and ramA by 1.7- to 2.2-fold and 10.5- to 30.1-fold, respectively, and lowered fluoroquinolones (FQ) minimum inhibitory concentrations by 2- to 8-fold. Neither MarA nor SoxS was found to be associated with increased FQ resistance. This study shows that the AcrAB efflux pump is playing a role in mediating fluoroquinolone resistance, and RamA may be the major global regulator of AcrAB-TolC-mediated fluoroquinolone resistance in Salmonella.

Introduction

S almonella can cause acute gastroenteritis and typhoid/paratyphoid. Many infections are due to ingestion of contaminated food, mainly meat. Since the beginning of the 1990s, Salmonella resistant to a range of antimicrobials have emerged and are threatening to become a serious public health problem (Chambers and Gong, 2011).

Fluoroquinolones (FQ) are first-line drugs to treat invasive Salmonella infections. However, resistances to FQ have developed worldwide in recent years (Hakanen et al., 2001; Biedenbach et al., 2006; Rotimi et al., 2008). Resistance to FQ is believed to arise from an interplay of different resistance mechanisms, including mutations in gyrase and topoisomerase genes (gyrA, gyrB, parC, and parE) in quinolone resistance-determining region (QRDR), increased expression of efflux pumps, and target site protection by Qnr proteins (Kehrenberg et al., 2009). QRDR mutation-mediated FQ resistance has been extensively studied in Salmonella (Piddock et al., 1998; Hansen and Heisig, 2003; Turner et al., 2006; Nishino et al., 2009), whereas there are very few reports on the role of efflux pumps in FQ resistance. Multidrug efflux pumps are internal membrane proteins that utilize cellular energy to extrude antibiotics actively out of the cell (Nishino et al., 2009). There are at least nine multidrug efflux pumps in Salmonella, and AcrAB is the most effective in causing multidrug resistances (Nishino et al., 2006; Nikaido et al., 2011). AcrAB-TolC efflux pump belongs to resistance-nodulation-cell division (RND) family. It is a tripartite complex consisting of AcrA, a fusion protein; AcrB, a cytoplasmic membrane transporter protein; and TolC, an outer membrane channel (Keeney et al., 2008). It has been shown to directly contribute to FQ resistance and multidrug resistance in Salmonella (Baucheron et al., 2004).

In Salmonella, the expression of acrAB is subject to multiple regulatory pathways. It is regulated locally by the repressor AcrR. Mutations within acrR gene have been associated with overexpression of acrAB in Salmonella and increased resistance to FQ (Olliver et al., 2004). The acrAB operon expression is also modulated by global regulators such as RamA, MarA, and SoxS. RamA and its repressor RamR are encoded by the ram locus. Mutations within ramR could cause increased ramA expression, and the association of increased expression of ramA and overexpression of acrAB has been reported (Abouzeed et al., 2008; O'Regan et al., 2009). The transcriptional activator MarA, its local repressor MarR, and two proteins with unknown functions, MarB and MarC, are encoded by the mar locus. Mutation within the marR gene leads to constitutive transcriptional activation of marRAB, resulting in an increased acrAB expression (Randall and Woodward, 2002). The transcriptional activator SoxS and a protein SoxR, whose oxidized form can activate soxS expression, are encoded by the soxRS locus. Increased expression of soxS has also been associated with AcrAB-mediated FQ resistance in Salmonella (Eaves et al., 2004; Chen et al., 2007).

To the best of our knowledge, previous reports regarding alterations in acrAB regulatory genes and overexpression of the AcrAB eflux pump in Salmonella were mainly focused on serotypes Typhimurium, Enteritidis, Typhi, and Paratyphi A (Piddock et al., 2000; O'Regan et al., 2009; Baucheron et al., 2014). However, very limited information is available for other serotypes. Furthermore, most of the previous research was on the in vitro induced mutants, and the information about the mutations and expression of the regulators in naturally occurring Salmonella isolates is rarely reported. In this study, we investigated the sequence alterations in the regulatory genes of acrR, ramR, marRAB, and soxRS; determined the expression levels of the genes they controlled; and explored the role of AcrAB efflux pump in mediating FQ resistance in various naturally occurring Salmonella serotypes isolated from foods and human beings in China. Novel regulatory gene alterations were observed in some of the tested serotypes. We also found that, apart from QRDR gene mutations, the AcrAB efflux pump contributes to FQ resistance, and RamA may be the major global regulator in AcrAB-mediated FQ resistance in Salmonella.

Materials and Methods

Bacterial isolates

Twenty-seven Salmonella isolates with different resistance levels to FQ were used in this study. Twenty were obtained from retail meats from a food investigation covering eight representative provinces of China between 2007 and 2013, and seven were human isolates kindly provided by China National Institute for Food and Drug Control. The 27 isolates were chosen from hundreds of Salmonella isolates based on their pulsed-field gel electrophoresis patterns and antimicrobial resistance patterns; therefore there were no repeat isolates in this study (Yang et al., 2010). They cover eight serotypes, including Shubra (n = 8), Indiana (n = 6), Enteritidis (n = 4), Typhimurium (n = 4), Virchow (n = 2), Djugu (n = 1), Kentucky (n = 1), and Derby (n = 1). The QRDR mutations of these isolates were determined in our previous work (Yang et al., 2012). Salmonella Typhimurium LT2 was used as a control strain in the regulatory gene sequencing and efflux pump gene expression studies. An additional strain used in this study was plasmid-free Escherichia coli DH5a. Plasmid pBR1MCS2, containing a kanamycin cassette, was used for cloning in the complementation study, since the isolates for complementation experiment were susceptible to kanamycin.

Antimicrobial susceptibility test

The FQ minimum inhibitory concentrations (MICs) of these isolates were determined in our previous work using agar dilution method (Yang et al., 2012). For efflux pump inhibition test, 10 μg/mL of carbonyl cyanide m-chlorophenylhydrazone (CCCP), which is known to inhibit RND-family efflux pumps, was added to determine the enrofloxacin and ciprofloxacin MIC values in the presence of CCCP.

Polymerase chain reaction amplification and DNA sequencing of acrR, ramR, marRAB, and soxRS genes

Identification of sequence alterations of acrR, ramR, marRAB, and soxRS in the 27 isolates and one Salmonella Typhimurium LT2 control strain was performed by polymerase chain reaction (PCR) and DNA sequencing. DNA template was extracted using the boiling method: pure bacterial colonies grown overnight on Luria-Bertani (LB) agar were suspended in sterile water, heated at 100°C for 10 min, and centrifuged at 10,000 g for 5 min. Supernatant containing DNA was collected and stored at −20°C for future use. Primers used and optimal conditions for PCR are listed in Table 1 (Olliver et al., 2004; Abouzeed et al., 2008). PCR was performed in 50 μL of distilled water with 1.5 mM Mg²⁺, 0.2 mM each dNTP, 0.2 pM each primer, and 1.25 U Taq Pfu DNA polymerase, and 5 μL DNA template. After a 10-min denaturation at 94°C, amplification was performed for 30 cycles of 1 min at 94°C, 15 s at 60 (58)°C, and 1 min at 72°C, with a final extension of 7 min at 72°C. The PCR products were sent to Aoke Biotechnology Company (Beijing, China) for nucleotide sequencing. Sequences in the amplified genes were analyzed using BLASTN (http://blast.ncbi.nlm.nih.gov).

Table 1.

Primers and Polymerase Chain Reaction Conditions Used in This Study

Primer purpose and target region	Primer sequence (5′-3′)	Annealing temp (°C)	Amplicon size (bp)	References
Regulation
acrR	CAGTGGTTCCGTTTTTAGTG	58	992	Olliver et al. (2004)
acrR	ACAGAATAGCGACACAGAAA	58	992	Olliver et al. (2004)
ramR	CGTGTCGATAACCTGAGCGG	60	934	Olliver et al. (2004)
ramR	AAGGCAGTTCCAGCGCAAAG	60	934	Olliver et al. (2004)
marRAB	ACGGTGGTTAGCGGATTGGC	58	1329	Olliver et al. (2004)
marRAB	AGCGGCGGACTTGTCATAGC	58	1329	Olliver et al. (2004)
soxRS	CGAACAGGGCGTCGTCGCTT	60	1199	Olliver et al. (2004)
soxRS	CTGGTTGCTAAAACGCGGCG	60	1199	Olliver et al. (2004)
Complementation
ramR	CATGGGATCC-CGTGTCGATAACCTGAGCGG	60	934	Abouzeed et al. (2008)
ramR	CTAGGAATTC-AAGGCAGTTCCAGCGCAAAG	60	934	Abouzeed et al. (2008)
Real-time PCR
16S rRNA	GCGGCAGGCCTAACACAT	60	182	O'Regan et al. (2009)
16S rRNA	GCAAGAGGCCCGAACGTC	60	182	O'Regan et al. (2009)
acrB	TTTTGCAGGGCGCGGTCAGAATAC	60	184	O'Regan et al. (2009)
acrB	TGCGGTGCCCAGCTCAACGAT	60	184	O'Regan et al. (2009)
ramA	CGTCATGCGGGGTATTCCAAGTG	60	107	O'Regan et al. (2009)
ramA	CGCGCCGCCAGTTTTAGC	60	107	O'Regan et al. (2009)
marA	ATCCGCAGCCGTAAAATGAC	60	180	O'Regan et al. (2009)
marA	TGGTTCAGCGGCAGCATATA	60	180	O'Regan et al. (2009)
soxS	AAATCGGGCTACTCCAAGTG	60	217	O'Regan et al. (2009)
soxS	CTACAGGCGGTGACGGTAAT	60	217	O'Regan et al. (2009)

PCR, polymerase chain reaction.

Expression of efflux transporter gene acrB and global regulators

Real-time PCR was used to measure the expression levels of acrB, ramA, marA, soxS, and 16S rRNA using SYBR Green I. Primers for each gene and optimal conditions for each primer pair are listed in Table 1 (O'Regan et al., 2009). Briefly, overnight bacterial cultures were diluted 1:100 in LB broth and grown to mid-logarithmic phase (OD₆₀₀ 0.6) with shaking at 37°C. Two milliliters aliquot of each culture was centrifuged and RNA was extracted immediately using the RNApure rapid extraction kit (Bioteke Corporation, Beijing, China). Contaminating genomic DNA was eliminated by DNase I (Fermentas) according to the manufacturer's instructions, and its absence was confirmed by a reverse transcriptase-minus control. RNA was immediately subjected to reverse transcription using RevertAid Premium First Strand cDNA Synthesis Kit (Fermentas) according to the manufacturer's instructions and diluted thrice before use. Real time PCR was performed in a MyiQ2 two-color real-time PCR detection system (Bio-Rad, Hercules, CA) and in a 25-μL reaction mixture containing 12.5 μL SYBR Green/Fluorescein qPCR Master Mix (2 × ), 0.3 pM each primer, and 2 μL cDNA template. The condition included a predenaturation of 10 min at 95°C, 35 cycles of 15 s of denaturation at 95°C, 30 s of annealing at 60°C, and 30 s of extension at 72°C, with a final melting step. 16S rRNA was used as a constant control within each isolate. Relative gene expression was calculated using the threshold cycle method.

Complementation study

The 934-bp ramR fragment generated by the primers (Abouzeed et al., 2008) from the genomic DNA of Salmonella Typhimurium LT2 was digested with EcoRI and BamHI, gel purified, and ligated into the EcoRI- and BamHI-digested broad-host-range plasmid vector pBR1MCS2 (Kovach et al., 1995). The FQ MICs and mRNA expressions of acrB and ramA transformants carrying the wild-type genes inserted into the vector were compared to those of the mutant strains.

Results

Demonstration of the involvement of efflux pumps on FQ resistance by comparing the MICs obtained in the presence and absence of the CCCP

The efflux pump inhibition experiment result showed that all the 14 isolates with higher FQ MIC values (S14–S27) had decreased enrofloxacin or/and ciprofloxacin MIC values in the presence of CCCP; on the contrary, only three (S5, S6, and S13) in the 13 isolates (S1–S13) with lower FQ MIC values showed a decrease in enofloxacin or/and ciprofloxacin MIC values in the presence of CCCP (Table 2). Consistent with efflux pump inhibition result, isolates with higher FQ MICs normally had more QRDR mutations than those with lower FQ MICs: for the 14 isolates with higher FQ MICs (S14–S27), there were 2–3 QRDR point mutations in each isolate; whereas for the 13 isolates with lower FQ MICs (S1–S13), only 0–1 QRDR point mutations were present in each isolate.

Table 2.

AcrAB-tolC Efflux Pump Gene Expression Levels, Quinolone Resistance-Determining Region Mutations, and Fluoroquinolone Resistance for the 27 Salmonella Isolates

Isolate ^a	Alteration type	Gene expression fold increase ^b				FQ MIC (μg/mL) ^c				QRDR mutations ^d (no. isolates)
Isolate ^a	Alteration type	acrB	ramA	marA	soxS	Enr	Cip	Gat	Lev	QRDR mutations ^d (no. isolates)
S1-Vir-c	None					0.5	≤0.25	≤0.25	0.5	gyrA (S83Y)
S2-Vir-c						0.5	≤0.25	≤0.25	0.5	gyrA (S83Y)
S3-Der-h						0.5	0.25	0.125	0.125	gyrA (D87N)
S4-Ken-h						2	0.25	0.5	0.25	None
S5-Dju-c			3.4 ± 1.2		2.2 ± 0.2	2 (1)	2 (1)	0.25	0.125	gyrA (S83F D87G)
S6-Typ-c		3.9 ± 0.8				1 (0.5)	1 (0.25)	0.25	0.125	gyrA (S83Y)
S7-Ent-h	acrR (V213F)+ramR (M83T)				1.6 ± 0.4	8	0.25	0.5	1	gyrA (D87N)
S8-Ent-c						4	0.5	8	2	gyrA (S83Y)
S9-Ent-h						2	0.5	0.5	1	gyrA (S83Y)
S10-Typ-h						1	0.25	0.5	4	None
S11-Ent-c				1.6 ± 0.1	3.1 ± 1.0	2	0.5	0.5	0.25	gyrA (D87N)
S12-Typ-h	acrR (S216A)					8	0.015	0.06	1	None
S13-Typ-h	ramR (deletion 1^e)	2.3 ± 0.3	13.7 ± 2.3			2 (0.25)	1 (0.25)	1	1	gyrA (S83F)
S14-Shu-c	acrR (V213A)		6.0 ± 1.6			8 (1)	≥16 (8)	16	8	gyrA (S83F D87G)+parC (S80R)
S15-Ind-c		1.8 ± 0.6				8 (1)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S16-Shu-c						8 (2)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S17-Ind-c		3.3 ± 1.1	3.1 ± 0.9	2.0 ± 0.4		8 (0.5)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S18-Shu-c						8 (1)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S19-Ind-c		1.9 ± 0.4	1.7 ± 0.2			8 (2)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S20-Shu-c				3.4 ± 0.9		8 (0.5)	8 (4)	8	4	gyrA (S83F D87G)+parC (S80R)
S21-Shu-c		5.3 ± 1.5	12.4 ± 1.1			8 (2)	≥16 (8)	≥16	8	gyrA (S83F D87G)+parC (S80R)
S22-Ind-c		1.9 ± 0.7	3.2 ± 0.4			8 (1)	16 (8)	8	4	gyrA (S83F)+parC (S80R)
S23-Ind-c	acrR (V213A)+ramR (T18P)	4.0 ± 0.8	63.5 ± 4.3			8 (0.5)	8 (2)	8	8	gyrA (S83Y)+parC (S80R)
S24-Shu-c	acrR (V213A)+ramR (deletion 2^f)	3.5 ± 0.6	45.1 ± 4.0	1.8 ± 0.5		8 (0.5)	8 (2)	16	16	gyrA (S83F D87G)+parC (S80R)
S25-Ind-c	acrR (V213A)+ramR (deletion 2^f)	2.5 ± 0.4	17.6 ± 2.5			8 (0.5)	8 (2)	16	16	gyrA (S83F D87G)+parC (S80R)
S26-Shu-c	acrR (V213A)+soxS (L107P)	3.7 ± 1.1	2.0 ± 0.3			8 (2)	≥16 (8)	8	4	gyrA (S83F D87G)+parC (S80R)
S27-Shu-c	acrR (V213A)+soxS (L107P)	4.6 ± 1.2	2.5 ± 0.6			8 (2)	8 (8)	8	4	gyrA (S83Y)+parC (S80R)

The name of the isolates means: isolate ID-serotype-source. The abbreviations of serotypes: Vir (Virchow), Der (Derby), Kentucky (Ken), Dju (Djugu), Typ (Typhimurium), Ent (Enteritidis), Shu (Shubra), Ind (Indiana). The abbreviations of source: c (chicken), h (human).

Gene expression data represent the mean ± standard deviations of three independent total RNA extractions.

Fluoroquinolones abbreviations: enrofloxacin (Enr), gatifloxacin (Gat), levofloxacin (Lev), ciprofloxacin (Cip). S1–S13 are isolates with relatively lower FQ MICs, and S14–S27 are isolates with higher FQ MICs. Data in parentheses are Enr or Cip MICs in the presence of CCCP. The Enr or Cip MICs for S5, S6, S13, and S14–S17 decreased in the presence of CCCP, whereas Enr or Cip MICs for other isolates were not affected by addition of CCCP. Gat and Lev MICs were not determined in the presence of CCCP.

Mutations in the QRDR: serine (S), phenylalanine (F), tyrosine (Y), asparagine (N), aspartic acid (D), glycine (G), arginine (R), glutamine (Q), lysine (K), isoleucine (I).

173–176 amino acids deletion.

5-bp deletion.

CCCP, carbonyl cyanide m-chlorophenylhydrazone; FQ, Fluoroquinolones; MICs, minimum inhibitory concentrations; QRDR, quinolone resistance-determining region.

Effect of alterations in regulatory genes on expression of the AcrAB-TolC efflux pump

Regulatory gene alterations were mainly found in acrR and ramR, whereas no mutations were detected in the global regulatory loci marRAB or soxRS, except a mutation in soxS in two Shruba isolates that led to the substitution of leucine by proline at codon 107 (L107P) (Table 2). Twenty of the 27 isolates carried three kinds of mutations that lead to protein changes in AcrR. All Indiana (6) and Shubra isolates (8) had a single amino acid change of valine to alanine at codon 213 (V213A); four Enteritidis isolates and one Typhimurium isolate had the change of valine to phenylalanine at codon 213 (V213F); and one Typhimurium isolate carried the change of serine to alanine at codon 216 (S216A). For ramR gene, four alteration types were found to cause RamR protein amino acid change: the most prevalent change was the substitution of methionine by threonine at codon 83 (M83T) in all four Enteritidis isolates and one Typhimurium isolate; replacement of threonine by proline at codon 18 (T18P) was found in one Indiana isolate; a 5-bp deletion was identified in one Indiana and one Shubra isolate, and resulted in an early stop codon; and one Typhimurium isolate lost four amino acids AGEY, meaning alanine, glycine, glutamic acid, and tyrosine (codons 173–176). It is worth noting that 10 isolates had both acrR and ramR/soxR gene alterations (Table 2). In addition, associations between serotypes and regulatory protein types were found (Table 2): for AcrR, V213A is prevalent in Indiana and Shubra isolates, whereas V213F is frequently observed in Enteritidis isolates. For RamR, all four Enteritidis isolates had M83T.

Consistent with regulatory gene alterations, the elevated efflux pump expressions also mainly occurred for acrB and ramA, rather than e marA or soxS (Table 2). Some alterations of ramR gene were associated with high expression levels of ramA. T18P substitution in one Indiana isolate (S23) led to a 64-fold increase in ramA expression. The 5-bp deletions in one Indiana isolate (S25) and one Shubra isolate (S24) caused 18- and 45-fold increase in ramA expression, respectively. The loss of nucleotides coding for 173–176 amino acids in one Typhimurium isolate (S13) led to a 14-fold increase of ramA mRNA expression. All of these isolates discussed above had relatively higher FQ MICs, except S13. In contrast, mutation type M83T seemed not to contribute to ramA overexpression and thus not to contribute to FQ resistance, since isolates carrying these mutations did not over express ramA and had relatively lower FQ MICs. The same phenomenon happened for acrB: among the three acrR alteration types, most acrB-overexpressed isolates harbored mutation type V213A and exhibited higher FQ MICs, whereas for isolates bearing other acrR mutation types, neither increased levels of acrB mRNA nor higher FQ MICs were observed. On the contrary, the marRA and soxRS did not seem to contribute to FQ resistance, since neither mutation in the two genes nor elevated expression levels of marA and soxS mRNA were found to be associated with high FQ MICs.

Complementation result

Complementation studies were performed in the four FQ-resistant isolates with high ramA mRNA levels harboring the three ramR alteration types: 173–176 amino acids deletion (S13), 5-bp deletion (S24 and S25), and T18P (S23), to determine if wild-type ramR would lower the expression of acrB/ramA mRNA or reduce FQ MICs. The acrB and ramA expression levels and FQ MICs for the wild-type ramR-complemented isolates are shown in Table 3. Only three isolates accepted the complementing plasmid. The gene expressions of acrB and ramA decreased by 1.7- to 2.2-fold and 10.5- to 30.1-fold, respectively, when the three isolates were complemented with wild-type ramR. The FQ MICs for the complemented wild-type strains also decreased by twofold to eightfold for enrofloxacin and twofold for ciprofloxacin, gatifloxacin, and levofloxacin, respectively, compared to their mutant strains.

Table 3.

The Efflux Pump Expression and Fluoroquinolone Minimum Inhibitory Concentrations of Wild-Type ramR-Complemented Strains Compared with Those of the Mutant Strains

Strain ^a	mRNA fold increase		MIC (μg/mL)
Strain ^a	ramA	acrB	Enr	Cip	Gat	Lev
S13-Typhimurium	13.7 ± 2.3	2.3 ± 0.3	2	1	1	1
S13-complement	1.3 ± 0.2	1.2 ± 0.1	0.25	0.5	0.5	0.5
S24-Shubra	45.1 ± 4.0	3.5 ± 0.6	8	8	16	16
S24-complement	1.5 ± 0.4	1.6 ± 0.4	4	4	8	8
S25-Indiana	17.6 ± 2.5	2.5 ± 0.4	8	8	16	16
S25-complement	1.1 ± 0.1	1.5 ± 0.3	4	4	8	8

The name of the isolates means: isolate ID-serotype.

Enr, enrofloxacin; Gat, gatifloxacin; Lev, levofloxacin; Cip, ciprofloxacin.

Discussion

In this study, CCCP inhibition result indicates that efflux pump may contribute to FQ resistance especially in high-level FQ-resistant isolates. To further investigate the involvement of AcrAB-TolC efflux pump in the regulation of FQ resistance, regulatory genes acrR, ramR, marRAB, and soxRS were sequenced for the 27 naturally occurring Salmonella isolates. In addition, the acrB, ramA, marA, and soxS mRNA were investigated, respectively, to determine the contribution of these regulatory genes to AcrAB-TolC-mediated FQ resistance in Salmonella.

As far as we know, this is the first report on acrAB regulatory gene alterations in naturally occurring isolates of serotypes Shubra, Indiana, Virchow, Djugu, Kentucky, and Derby. Consistent with previous findings on FQ-resistant Salmonella mutants, the mutations/deletions were mainly found in acrR and ramR genes, rather than marRAB or soxRS genes (Olliver et al., 2004; Zheng et al., 2009). Also, it is noteworthy that certain alteration types in ramR were correlated with highly elevated expressions of ramA and higher FQ MICs in different serotypes of Salmonella isolates, whereby for marRAB and soxRS, no such alterations were observed. Abouzeed et al. (2008) also found that inactivation of ramR resulted in an increase in ramA expression and the AcrAB-tolC efflux pump, and thus FQ resistance level, but inactivation of marR, marA, soxR, and soxS did not affect the susceptibilities of Salmonella. In addition, the complementation result showing decreased acrB and ramA mRNA levels as well as lowered FQ MICs in the wild-type ramR-complemented strains confirmed that the ramR gene is responsible for FQ resistance through upregulation of ramA and acrB in these isolates. All the above findings indicate that regulator RamA, rather than MarA or SoxS, may play a major role in AcrAB-TolC efflux pump-mediated FQ resistance in Salmonella (Piddock et al., 2000; Abouzeed et al., 2008; Zheng et al., 2009).

It has been reported previously that repression of ramA occurred through the direct binding of RamR to the ramA promoter (PramA) region in Salmonella Typhimurium (Baucheron et al., 2012; Ricci et al., 2012). In this study, several alterations in ramR caused significantly high expression levels of ramA mRNA, whereas other ramR alteration types did not have much effect, which is similar to recent findings (Fàbrega et al., 2016). It is speculated that some ramR alterations could cause three-dimensional structural changes in the RamR protein whose DNA binding affinity to PramA region may be reduced, thereby inducing overexpression of ramA, while other ramR alteration types may not have this effect (Yamasaki et al., 2013). Further genomic and proteomics studies are needed to help better understand the detailed regulation mechanism of ramRA in Salmonella. We also note that not all ramA overexpressed isolates contained mutations in their respective ramR genes. Nikaido et al. (2011) also found that both RamR-dependent and RamR-independent pathways are involved in induction of ramA. The RamR-independent pathways of ramA overexpression may also exist in these isolates and still need further investigation.

FQ-resistant mechanisms for isolates with relatively lower FQ MICs may be mainly due to the QRDR mutations, since neither their FQ MICs were affected by the addition of CCCP nor did they overexpress ramA/acrB, except S5 (Djugu), S6 (Typhimurium), or S13 (Typhimurium) (Table 2). In these three isolates, efflux pump may also play a role in mediating FQ resistance because of reduced FQ MICs in the presence of CCCP and the elevated efflux pump expression. In addition, the role of plasmid genes in mediating low-level FQ resistance also merits further investigation (Veldman et al., 2011). As for the high-level FQ-resistant isolates, both the efflux pumps and QRDR may contribute to FQ resistance, since their FQ MICs decreased by addition of CCCP. For the high FQ-resistant isolates with elevated AcrAB efflux pump expressions, the AcrAB pump may contribute to their high FQ resistances. For the high FQ-resistant isolates in which AcrAB efflux pump overexpressions were not observed, the role of efflux pumps other than AcrAB could not be excluded. Future studies on other efflux pumps may help understand the FQ resistance mechanisms of these isolates.

Conclusions

We presented novel acrAB regulatory gene alterations in naturally occurring Salmonella isolates of diverse serotypes, including Shubra, Indiana, Virchow, Djugu, Kentucky, and Derby. To our best knowledge, this is the first report about the association between serotypes and mutation types of Salmonella. Efflux pump inhibition results, generation of regulatory gene mutants and complementation experiments together with expression studies, showed that apart from QRDR mutations, AcrAB-TolC efflux pump is also playing a role in mediating FQ resistance in Salmonella mainly through the global regulator RamA.

Footnotes

Acknowledgments

This work was financially supported, in part, by National Key R&D Program of China (2016YFD0401102), research program of General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China (2016IK308), research program of Jiangsu Entry-Exit Inspection and Quarantine Bureau (2016KJ53), and Science and Technology Planning Project of Zhangjiagang (ZKS1506).

Disclosure Statement

No competing financial interests exist.

References

Abouzeed

, Baucheron

, Cloeckaert

. ramR mutations involved in efflux-mediated multidrug resistance in Salmonella enterica serovar Typhimurium. Antimicrob Agents Chemother, 2008; 52:2428–2434.

Baucheron

, Chaslus-Dancla

, Cloeckaert

. Role of TolC and parC mutation in high-level fluoroquinolone resistance in Salmonella enterica serotype Typhimurium DT204. J Antimicrob Chemother, 2004; 53:657–659.

Baucheron

, Coste

, Canepa

, et al. Binding of the RamR repressor to wild-type and mutated promoters of the RamA gene involved in efflux-mediated multidrug resistance in Salmonella enterica serovar Typhimurium. Antimicrob Agents Chemother, 2012; 56:942–948.

Baucheron

, Monchaux

, Le Hello

, Weill

, Cloeckaert

. Lack of efflux mediated quinolone resistance in Salmonella enterica serovars Typhi and Paratyphi A. Front Microbiol, 2014; 5:12.

Biedenbach

, Toleman

, Walsh

, Jones

. Analysis of Salmonella spp. with resistance to extended-spectrum cephalosporins and fluoroquinolones isolated in North America and Latin America: Report from the SENTRY Antimicrobial Surveillance Program (1997–2004). Diagn Microbiol Infect Dis, 2006; 54:13–21.

Chambers

, Gong

. The intestinal microbiota and its modulation for Salmonella control in chickens. Food Res Int, 2011; 44:3149–3159.

Chen

, Cui

, McDermott

, et al. Contribution of target gene mutations and efflux to decreased susceptibility of Salmonella enterica serovar Typhimurium to fluoroquinolones and other antimicrobials. Antimicrob Agents Chemother, 2007; 51:535–542.

Eaves

, Ricci

, Piddock

LJV

. Expression of acrB, acrF, acrD, marA, and soxS in Salmonella enterica Serovar Typhimurium: Role in multiple antibiotic resistance. Antimicrob Agents Chemother, 2004; 48:1145–1150.

Fàbrega

, Ballesté-Delpierre

, Vila

. Differential impact of ramRA mutations on both ramA transcription and decreased antimicrobial susceptibility in Salmonella Typhimurium. J Antimicrob Chemother, 2016; 71:617–624.

10.

Hakanen

, Kotilainen

, Huovinen

, Helenius

, Siitonen

. Reduced fluoroquinolone susceptibility in Salmonella enterica serotypes in travelers returning from Southeast Asia. Emerg Infect Dis, 2001; 7:996.

11.

Hansen

, Heisig

. Topoisomerase IV mutations in quinolone-resistant salmonellae selected in vitro. Microbial Drug Resistance, 2003; 9:25–32.

12.

Keeney

, Ruzin

, McAleese

, Murphy

, Bradford

. MarA-mediated overexpression of the AcrAB efflux pump results in decreased susceptibility to tigecycline in Escherichia coli . J Antimicrob Chemother, 2008; 61:46–53.

13.

Kehrenberg

, Cloeckaert

, Klein

, Schwarz

. Decreased fluoroquinolone susceptibility in mutants of Salmonella serovars other than Typhimurium: Detection of novel mutations involved in modulated expression of ramA and soxS. J Antimicrob Chemother, 2009; 64:1175–1180.

14.

Kovach

, Elzer

, Hill

, et al. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene, 1995; 166:175–176.

15.

Nikaido

, Shirosaka

, Yamaguchi

, Nishino

. Regulation of the AcrAB multidrug efflux pump in Salmonella enterica serovar Typhimurium in response to indole and paraquat. Microbiology, 2011; 157:648–655.

16.

Nishino

, Latifi

, Groisman

. Virulence and drug resistance roles of multidrug efflux systems of Salmonella enterica serovar Typhimurium. Mol Microbiol, 2006; 59:126–141.

17.

Nishino

, Nikaido

, Yamaguchi

. Regulation and physiological function of multidrug efflux pumps in Escherichia coli and Salmonella . Biochim Biophys Acta, 2009; 1794:834–843.

18.

O'Regan

, Quinn

, Pages

, McCusker

, Piddock

, Fanning

. Multiple regulatory pathways associated with high-level ciprofloxacin and multidrug resistance in Salmonella enterica serovar enteritidis: Involvement of RamA and other global regulators. Antimicrob Agents Chemother, 2009; 53:1080–1087.

19.

Olliver

, Valle

, Chaslus-Dancla

, Cloeckaert

. Role of an acrR mutation in multidrug resistance of in vitro-selected fluoroquinolone-resistant mutants of Salmonella enterica serovar Typhimurium. FEMS Microbiol Lett, 2004; 238:267–272.

20.

Piddock

, White

, Gensberg

, Pumbwe

, Griggs

. Evidence for an efflux pump mediating multiple antibiotic resistance in Salmonella enterica serovar Typhimurium. Antimicrob Agents Chemother, 2000; 44:3118–3121.

21.

Piddock

LJV

, Ricci

, McLaren

, Griggs

. Role of mutation in the gyrA and parC genes of nalidixic-acid-resistant Salmonella serotypes isolated from animals in the United Kingdom. J Antimicrob Chemother, 1998; 41:635–641.

22.

Randall

, Woodward

. The multiple antibiotic resistance (mar) locus and its significance. Res Vet Sci, 2002; 72:87–93.

23.

Ricci

, Busby

, Piddock

. Regulation of RamA by RamR in Salmonella enterica serovar Typhimurium: Isolation of a RamR superrepressor. Antimicrob Agents Chemother, 2012; 56:6037–6040.

24.

Rotimi

, Jamal

, Pal

, Sonnevend

, Dimitrov

, Albert

. Emergence of multidrug-resistant Salmonella spp. and isolates with reduced susceptibility to ciprofloxacin in Kuwait and the United Arab Emirates. Diagn Microbiol Infect Dis, 2008; 60:71–77.

25.

Turner

, Nair

, Wain

. The acquisition of full fluoroquinolone resistance in Salmonella Typhi by accumulation of point mutations in the topoisomerase targets. J Antimicrob Chemother, 2006; 58:733–740.

26.

Veldman

, Cavaco

, Mevius

, et al. International collaborative study on the occurrence of plasmid-mediated quinolone resistance in Salmonella enterica and Escherichia coli isolated from animals, humans, food and the environment in 13 European countries. J Antimicrob Chemother, 2011; 66:1278–1286.

27.

Yamasaki

, Nikaido

, Nakashima

, et al. The crystal structure of multidrug-resistance regulator RamR with multiple drugs. Nat Commun, 2013; 4:2078.

28.

Yang

, Qu

, Zhang

, et al. Prevalence and characterization of Salmonella serovars in retail meats of marketplace in Shaanxi, China. Int J Food Microbiol, 2010; 141:63–72.

29.

Yang

, Xi

, Cui

, et al. Mutations in gyrase and topoisomerase genes associated with fluoroquinolone resistance in Salmonella serovars from retail meats. Food Res Int, 2012; 45:935–939.

30.

Zheng

, Cui

, Meng

. Effect of transcriptional activators RamA and SoxS on expression of multidrug efflux pumps AcrAB and AcrEF in fluoroquinolone-resistant Salmonella Typhimurium. J Antimicrob Chemother, 2009; 63:95–102.