JAC Advance Access originally published online on June 6, 2006
Journal of Antimicrobial Chemotherapy 2006 58(1):37-46; doi:10.1093/jac/dkl202
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bacteroides fragilis BmeABC efflux systems additively confer intrinsic antimicrobial resistance
1 Greater Los Angeles Veterans Administration Healthcare Systems Los Angeles, CA, USA 2 Department of Medicine, University of California Los Angeles, CA, USA 3 Department of Oral Microbiology, Matsumoto Dental University Shiojiri, Japan 4 Department of Microbiology, School of Dentistry, Aichi-Gakuin University Nagoya, Japan
*Corresponding author. Tel: +1-310-268-3404; Fax: +1-310-268-4458; E-mail: hwexler{at}ucla.edu
Received 16 December 2005; returned 22 February 2006; revised 16 April 2006; accepted 26 April 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: To determine the prevalence of expression and function(s) of Bacteroides fragilis RND family efflux transport systems (bmeABC1-16).
Methods: The mRNA transcripts of bmeB efflux pump genes were detected in a wild-type strain ADB77 by RT–PCR and expression in different strains was quantified by comparative quantitative real-time RT–PCR. In order to determine independent or additive functions, BmeB 1, 3, 12 and 15 (the first efflux pumps identified) were deleted as singles, doubles, triples or quadruples by the double cross-over technique with pADB242 and antimicrobial susceptibility was assayed by the spiral gradient endpoint technique.
Results: All efflux pumps except bmeB9 were expressed in the wild-type parental strain. Susceptibility to ß-lactams, fluoroquinolones, ethidium bromide, SDS and triclosan was increased in ADB77
bmeB3 (up to 3-fold) and ADB77bmeB1bmeB3bmeB12 (up to 5-fold). Expression of bmeB9 was increased and that of bmeB11 repressed in the latter deletant. A quadruple deletant (ADB77bmeB1bmeB3bmeB12bmeB15) had similar changes as well as a 2-fold increase in expression of bmeB16 and norfloxacin resistance. Expression of bmeB3 was increased in two triple deletants ADB77bmeB1bmeB12bmeB15-type I (2-fold) and ADB77bmeB1bmeB12bmeB15-type II (5.8-fold). Antimicrobial MICs were also increased in the latter deletant; ampicillin (2.6-fold), cefoperazone (3.4-fold), cefoxitin (1.8-fold), tetracycline (36.4-fold), SDS (1.7-fold) and triclosan (2-fold).
Conclusions: These data demonstrate that constitutive bmeB expression is prevalent in B. fragilis. At least seven BmeB efflux pumps are functional in transporting antimicrobials and have overlapping substrate profiles, and at least four confer intrinsic resistance.
Keywords: membrane proteins , co-expression , susceptibility
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
The importance of efflux pumps in antimicrobial resistance, either alone or via interplay with decreased outer membrane porin expression, has been described for many aerobic Gram-negative bacteria.1 In many bacteria, efflux pumps expressed at basal levels confer low-level protection and enable the bacteria to survive exposure to sub-clinical levels of antimicrobial agents.2–4 Expression of efflux systems with overlapping substrate profiles in Pseudomonas and Escherichia coli appears to be regulated in a coordinated fashion.5,6
Bacteroides fragilis is intrinsically resistant to several classes of structurally unrelated antibiotics,7 and the mechanisms are often poorly understood. Clinically, B. fragilis has exhibited increasing resistance to many antibiotics including cefoxitin, clindamycin, metronidazole, carbapenems and the newer fluoroquinolones (including gatifloxacin, levofloxacin and moxifloxacin). Data from other Gram-negative bacteria including Pseudomonas aeruginosa have shown that RND-efflux systems can be a major cause of clinically relevant multidrug resistance (MDR).8 In contrast, very little is known about efflux pumps in anaerobic bacteria. Aside from our recent work describing RND pumps in B. fragilis,9 there are two reports indirectly implicating efflux pumps in B. fragilis in antimicrobial resistance including resistance to norfloxacin;10,11 also, a multidrug and toxic compound extrusion (MATE)-type efflux system has been characterized in Bacteroides thetaiotaomicron.12
When we began these studies, we had identified four RND pump homologues in B. fragilis and named them bme (Bacteroides multidrug efflux) pumps. The presence of multiple pumps made analysis of their function difficult, so we decided to construct unmarked single and combined pump deletants of these strains in order to analyse the contributions of each pump to antimicrobial resistance. The specific bmeB efflux pump genes (bme1, bme5, bme12 and bme15) were targeted for deletion as they were identified. Since BmeB1 is the efflux pump most homologous to the Pseudomonas MexB efflux pump (an RND pump often associated with antimicrobial resistance), it was reasonable to postulate that it might be similarly involved in B. fragilis. When we later identified bmeB3 and found that disruption of the gene caused changes in MICs, we included bmeB3 in our scheme of constructing deletants. Subsequently, we identified a total of 16 homologues of the P. aeruginosa mexAB-oprM operon in B. fragilis.9 We have named them BmeABC1–16, respectively, where A is the membrane fusion protein for each of the sixteen systems, B is the efflux pump protein and C is the outer membrane protein (OMP). The usual nomenclature of RND efflux systems uses different letters for different systems.2,4 However, due to the sheer number of RND family efflux pump systems in B. fragilis, we have opted to use the same letters for all systems with numeric differences for different systems. Although we are aware that this nomenclature deviates slightly from what is conventionally used, we believe that it is logical and that it makes identifying the RND system in question and its genomic localization easier. All the RND efflux systems can thus be distinguished numerically ranging from BmeABC1 to BmeABC16. Our preliminary data using the broad-spectrum efflux pump inhibitors (EPIs) carbonyl cyanide m-chlorophenylhydrazone (CCCP) and reserpine to determine MICs for a single bmeB3 deletant, laboratory mutants overexpressing bmeB efflux pump genes and multidrug-resistant clinical isolates have demonstrated that BmeB efflux pumps can accommodate structurally unrelated antimicrobial agents to cause resistance, which is reducible by EPIs.
The aims of this study were: (i) to measure expression of bmeB efflux pump genes in a wild-type strain of B. fragilis; (ii) to create single and multiple deletion mutants in order to determine their contributions to intrinsic antibiotic resistance and/or expression of other RND pumps; and (iii) to correlate patterns of efflux pump expression with specific antibiograms.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Bacterial strains, media and growth conditions
The bacterial strains and the plasmids used in this study are listed in Table 1. The eleven B. fragilis test strains used for experiments were derived from strain ADB77 [Wadsworth Anaerobe Laboratory Collection (WAL) # 108], a derivative of B. fragilis 638R that was optimized for use in constructing deletion mutants using a two-step recombination procedure.13 B. fragilis 638R is the strain generally used in research laboratories for genetic manipulations and has been sequenced at the Sanger Genome Campus (Hinxton, UK) as part of the B. fragilis genome sequencing project. The B. fragilis ATCC type strain (ATCC 25285 = WAL 3501) has also been sequenced by the Sanger Centre and was used as a control for MIC determinations as specified by the Clinical Laboratory Standards Institute (formerly NCCLS).14 Strains were cultured under anaerobic conditions (5% CO2, 10% H2 and 85% N2) at 37°C for 24–48 h in brain heart infusion broth (MP Biomedical, Aurora, OH, USA) supplemented with 0.5% yeast extract and 15 mg/L haemin (BHIS)13 or in anaerobic minimal medium with 0.5% glucose (AMMgluc).13 Haemin stock solution (0.5%) was prepared by dissolving 0.5 g of haemin chloride (Fisher Scientific, Fairlawn, NJ, USA) in 100 mL of 0.1 M NaOH (Sigma, St Louis, MO, USA). Thymine (50 mg/L; Sigma) was added for growth of Thy– strains (i.e. WAL 108 and its derivatives). For susceptibility testing, strains were resuspended in brucella broth and cultured on brucella blood agar plates. All antimicrobial agents used were obtained from Sigma. All test agents were dissolved and used according to the manufacturers' instructions.
|
Bioinformatic analysis of efflux operons
The 16 RND-type efflux operons were identified on the genome and proteome sequences of B. fragilis (http://www.sanger.ac.uk/Projects/B_fragilis/; http://www.expasy.org/sprot/hamap/BACFR.html) by tBLASTN searches against the P. aeruginosa mexAB-oprM operon, as described previously.9 The putative pump proteins were aligned with each other to determine their conserved motifs using GeneDoc v2.5.010; their secondary structure folding was determined by TMHMM v.2.0 and their tertiary structure by threading the sequence on the AcrB-crystal structure using SWISS-MODEL (http://au.expasy.org/tools/).
DNA extraction
Total cellular chromosomal DNA was isolated from strains cultured in BHI broth containing 5% thymine (3 mL). The strains were incubated for 24 h under anaerobic conditions. Aliquots of the cultures (1 mL) were centrifuged at 5000 g. The resulting pellet was resuspended in 180 mL of buffer ATL supplied with the kit, and the extraction continued according to the DNeasy Tissue kit (Qiagen, Valencia, CA, USA).
Construction of efflux pump deletion mutants
In-frame deletions of bmeB genes were constructed by a two-step double cross-over technique with either the pYT102 or pADB242 suicide vectors (both gifts from Dr Michael Malamy, Tufts University School of Medicine, Boston, MA, USA).13 Briefly, 
800 bp fragments of the upstream and downstream regions (including 50–100 bp of the beginning and end of the gene to be deleted) of the gene in question were amplified using specific primers to which appropriate restriction sites were added for subsequent cloning into the suicide vector (Table 1). Oligonucleotide sequences were based on sequence data obtained from the B. fragilis NCTC 9343 (ATCC 25285) preliminary genome sequence produced by the Pathogen Sequencing Group at the Sanger Centre (http://www.sanger.ac.uk/Projects/B_fragilis).
Competent E. coli DH5
were prepared by the RbCl method,15 transformed with the plasmid construct by heat-shock and transformants selected with chloramphenicol. The resultant clone containing the plasmid with the ‘up-down’ sequence was mobilized into B. fragilis ADB77 using E. coli DH5 and the broad host range mobilizer plasmid pK2317 in a three-part mating protocol.13 The suicide vector pYT102 contains the B. fragilis thyA gene and tetR; recombination at the specific efflux site results in thymine prototrophy (and consequent trimethoprim susceptibility) and tetracycline resistance in the recipient. Cointegrants were selected on plates containing gentamicin (50 mg/L), rifampicin (50 mg/L) and tetracycline (2 mg/L) and confirmed by colony PCR using primers designed to detect the recombinant junction. Cointegrant strains were maintained on media with tetracycline.
Cointegrants were subcultured for overnight growth in BHIS broth without antimicrobial selection and then plated on minimal media containing thymine and trimethoprim to select for the second recombination event. Trimethoprim-resistant colonies were screened to confirm that they were tetracycline susceptible and further screened by PCR with sets of both internal and junction primers to confirm that they were the desired deletion resolution products. Putative deletants were sequenced to confirm that they had the expected deletion junction.13,16 Four sets of deletants were created: single deletants of bmeB3, bme12 and bme15, double deletants of bmeB1 and bmeB12, and bmeB12 and bmeB15, triple deletants of bmeB1, bmeB12 and bmeB15, and bmeB1, bmeB3 and bmeB12 and a quadruple deletant of bmeB1, bmeB3, bmeB12 and bmeB15. Deleted genes were verified by DNA sequencing of the deletion junction and RT–PCR to verify that there was no expression of the gene in question.
RNA extraction
Total cellular RNA was isolated from strains cultured in BHI broth containing 5% thymine using the RNeasy-RNA ProtectTM (Qiagen) method with on-column DNase treatment. The strains were incubated for 2 h under anaerobic conditions to a mid-log phase of growth (OD600 = 0.4). Aliquots (3 mL) were mixed with an equal volume of RNA-ProtectTM and the extraction continued according to manufacturer's instructions. The absence of any contaminating DNA was verified by a standard PCR reaction performed on the RNA samples. The integrity of the extracted RNA was checked on a 1.5% agarose gel and by measuring the OD260/280 ratio. The samples were quantified by OD260 measurement and the measurement converted into ng/µL.
Detection of gene expression by reverse transcription (RT)–PCR
Gene expression was detected by PCR end-point analysis on an agarose gel. Briefly, first strand synthesis of the RT–PCR reactions was performed with Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) with 200 ng of RNA and 50 ng of random primers (Invitrogen). Second strand synthesis was performed using gene-specific primers (Table 2) at a final concentration of 1.0 µM each. The PCR parameters were: initial denaturation at 95°C for 5 min, followed by 30 cycles of a three-step reaction of denaturing at 95°C for 1 min, annealing at 55°C for 1 min and extension at 72°C for 1 min, ending with a final extension at 72°C for 10 min.
|
Quantification of gene expression by comparative quantitative real time RT–PCR
Gene expression was quantified by threshold analysis of SYBR Green dye incorporated during the exponential phase of PCR. Briefly, a one-step real-time RT–PCR was performed with the Cepheid SmartCycler® using the Quantitect® SYBR® Green one-step RT–PCR kit (Qiagen). RNA expression was normalized to the parental strain by using 16S rRNA. Primers were designed to amplify products between 130 and 170 bp in size (Table 2) and were added to the reactions at a final concentration of 1.0 µM each. For a one-step real-time RT–PCR (i.e. cDNA synthesis and second strand synthesis), RNA was added to the reaction tubes at a final amount of 200 ng/reaction, except for the 16S rRNA reaction tube where a final amount of 200 pg/reaction was used. Expression levels were measured as an amount of cDNA as extrapolated by a cycle threshold (Ct) value from the real-time PCR standard growth curve. The Ct was the cycle number at which the growth curve attained exponential growth and was thus the highest concentration of RNA template. In order to rule out any non-specific products resulting from primer dimers, melting curve analysis of the amplified products was performed. Expression results were quantified by the comparative threshold (Ct) approximation method,17 using the assumption that the PCR growth curve efficiency for all reactions is 100% and that the DNA concentration doubled at each cycle:
 |
2-fold change in expression compared with the parental strain was considered significant. Susceptibility testing
The spiral gradient endpoint (SGE) method, which can differentiate small changes in the MIC,18 was used to measure the antibiotic susceptibilities of the strains. Susceptibility patterns of the parental strain and the deletants were measured for 27 antimicrobial agents. Selected stock concentrations of antimicrobial agents were deposited onto brucella blood agar plates in a spiral pattern using the spiral plater (Autoplate 4000, Advanced Instruments; Norwood, MA, USA) resulting in a radially decreasing concentration gradient from the centre to the outside of the plate. Bacterial strains were resuspended in brucella broth to a density equivalent to that of a 0.5 McFarland standard and deposited on the plates. The plates were incubated for 24–48 h at 37°C under anaerobic conditions. The distance from the centre of the plate to the point where growth began was measured and the SGE computer software was used to convert these values into MIC values (mg/L). Susceptibility studies were performed on at least three independent occasions. A 1.5-fold difference in susceptibility was considered significant. Since intrinsic antibiotic resistance often involves interplay between efflux and at least one other mechanism, MIC changes due to only one of these mechanisms may cause changes as small as 1.5- to 2-fold.19 Such small changes are usually missed in the NCCLS doubling dilution method but are detectable with the SGETM method since the degree of accuracy is ±0.26 of a 2-fold dilution (compared to the ± one 2-fold dilution accuracy of the standard NCCLS agar dilution method).20 In order to reflect the accuracy of the MIC differences as determined by the SGE method, data were recorded to 1 or 2 decimal places.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Bioinformatic analysis of RND efflux systems in B. fragilis
Bioinformatic analysis revealed that all the efflux pumps had the typical 12 transmembrane (TM) helical domains, except for BmeB16, which had 16 TM helical domains. The efflux pumps also had conserved amino acids between residues 1500 and 2140, including Gly-1503, Pro-1528, Ser-1573, Phe-1611, Pro-1612, Leu-1655, Gly-1680, Glu-1929, Pro-1980, Gly-1989 and Pro-2016. All the predicted BmeB proteins also had an AcrB-like tertiary structure (>25% homology) as shown by SWISS-MODEL. The bmeABC11 operon also had an additional efflux pump gene bmeB11'. Since the genome sequence of B. fragilis has been shown to have a degree of DNA sequence repeats,21,22 we wondered if some of these efflux pumps might be simply duplications of others. Comparative analysis of the amino acid sequences and genetic organization of these efflux systems revealed that none of the RND efflux systems was repeated (data not shown).
RND efflux pump deletion mutants
Single deletants of bmeB3 (ADB77bmeB3) (WAL 243), bmeB12 (ADB77bmeB12) (WAL 165) and bmeB15 (ADB77bmeB15) (WAL 140) were created. Double and triple deletants, respectively, were constructed from the single and double deletant strains as indicated in Table 1. Strain WAL 189-tetR (ADB77bmeB1bmeB12bmeB15-type II) arose from WAL 189-tetS (ADB77bmeB1bmeB12bmeB15-type I) spontaneously upon repeated subculture in the laboratory without any antibiotic pressure (Table 1). Construction of all possible deletion combinations of the four pumps studied were attempted, usually at least twice. In general, if we were able to obtain the intermediate construct in which the modified suicide vector was integrated into the chromosome, we were able to carry through the second recombination step and obtain the deletant strain.
Expression of RND efflux pumps in B. fragilis
All RND efflux genes except bmeB9 were expressed at detectable levels in the parental strain ADB77 (WAL 108). The mRNA of the efflux pumps was detectable at varying but roughly comparable levels; RT–PCR product amounts (quantified by gel analysis) were 89.13, 88.94, 95.27, 93.27, 84.54, 84.63, 87.44, 86.12, 0.00, 105.10, 78.54, 85.20, 84.78, 92.12, 77.67 and 77.40 ng, respectively, for bmeB1–16. The transcript for bmeB11' mRNA was also detectable in the parental strain (data not shown).
No expression changes of other bmeB genes were observed in single or double deletants of bmeB1, bmeB3, bmeB12 and bmeB15 (these include the double deletants ADB77bmeB1bmeB12 and ADB77bmeB12bmeB15). However the triple deletant WAL 189-tetS (ADB77bmeB1bmeB12bmeB15-type I) had a 2-fold increase in bmeB3 expression, and WAL 189-tetR (ADB77bmeB1bmeB12bmeB15-type II) had a 5.8-fold increase in bmeB3 expression (P < 0.05) (Figure 1a). This effect was not seen in the other two single deletants or in the two double deletants. Thus, apparently when up to any two of these three pumps are missing, no compensating increase of the bmeB3 pump is triggered. (Note: we were not able to obtain the bme1/15 double mutant, and we cannot rule out that this combination might result in an increase in bmeB3.) However, when all three pumps are missing, increased expression of bmeB3 is triggered, presumably to compensate for the reduced combined pump activity.
|
The simultaneous absence of bmeB1, bmeB3 and bmeB12 resulted in an increase in bmeB9 and a decrease in bmeB11 (Figure 1b) in two constructs [WAL 215 (ADB77
bme1bme3bmeB12) and WAL 219 (ADB77bme1bmeB3bmeB12bmeB15)]. Again, the single deletants had no changes in pump expression. It appears that the double deletion of bmeB1 and bmeB12 can be compensated for by the presence of bmeB3, but when all three are missing, the expression of other pumps is affected. Since we were not able to obtain the bmeB1/B3 mutant, we cannot rule out the possibility that it is the absence of these two pumps alone that caused the change in bmeB9 and bmeB11. Also, in the quadruple mutant, in which bmeB15 was missing along with bmeB1, bmeB3 and bmeB12, bmeB16 was increased 2-fold (P < 0.05, Figure 1c). Again, the ADB77bmeB15 single mutant (WAL 140), the double mutant ADB77bmeB12bmeB15 (WAL 166) and the triple deletant ADB77bmeB1/12/15 (WAL 189) do not exhibit this effect. We also noted that in both strains in which expression of bmeB11 was decreased, expression of bmeB11' was not affected (data not shown). The two triple deletants (WAL 189 and WAL 215) have different phenotypes: WAL 189 had an increase in bmeB3, and WAL 215 had an increase in bmeB9 and decrease in bmeB11. Although they were constructed separately, one can view these strains as having the same genetic background (missing bmeB1 and bmeB12) and differing only in the third pump deleted (bmeB15 in WAL 189 and bmeB3 in WAL 215). Again, while deletion of either of these pumps alone had no effect, the respective deletion of these two pumps had different effects, even in the same genetic background (i.e. the absence of bmeB1 and bmeB12).
Susceptibilities of parental and deletant strains to antibiotics, detergents and dyes
MICs for parental and deletant strains of a variety of antimicrobials, detergents and dyes were determined using the SGE method.20 The MICs are listed in Table 3. The MICs for the single deletants WAL 140 (ADB77bme15) and WAL 165 (ADB77bmeB12) and the double mutants WAL 166 (ADB77bme12bmeB15) and WAL 187 (ADB77bme1bmeB15) were also tested but were not significantly different from those of the parental strain. The triple deletant WAL 215 (ADB77bmeB1bmeB3bmeB12) and the quadruple deletant strain WAL 219 (ADB77bmeB1bmeB3bmeB12bmeB15) were 1.5-fold more susceptible than the parental strain to cefoxitin, doripenem, ethidium bromide, faropenem, imipenem, meropenem, SDS and triclosan. Strain WAL 219 was also 2-fold more resistant to norfloxacin than the wild-type strain, possibly due to overexpression of bmeB16.
|
The single bmeB3 deletant ADB77
bmeB3 (WAL 243) was the only single deletant for which there were significant changes in some MICs. This mutant was 1.5-fold more susceptible to ampicillin, cefoxitin, cefoperazone, doripenem, faropenem, imipenem, gatifloxacin, meropenem, ethidium bromide, SDS and triclosan than the parental strain ADB77 (WAL 108). The triple deletant, strain WAL 189-tetS (ADB77bmeB1bmeB12bmeB15 type I), had no significant MIC changes compared with the parental strain, possibly due to the counteractive effect of bmeB3 overexpression. A spontaneous mutant of WAL 189 arose during these studies [189-tetR (ADB77bmeB1bmeB12bmeB15-type II)]. This strain had increases in MICs of ampicillin (2.6-fold), cefoperazone (3.4-fold), cefoxitin (1.8-fold), tetracycline (36.4-fold), SDS (1.7-fold) and triclosan (2-fold). This strain had extremely high tetracycline resistance but did not overexpress tet(A)Q2, the other likely causative gene (data not shown).
Generally, all of the strains (wild-type and all deletants) were highly resistant to aminoglycosides, bile, macrolides, and dyes and detergents (MICs 256 mg/L); therefore, changes in susceptibility, if any, were impossible to determine.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Bacterial RND family efflux transport systems (particularly the Pseudomonas Mex systems) have been intensely studied in the last decade.8 The MexPQ-OpmE and MexMN-OprM systems were recently identified and added to the seven systems already described as functional RND efflux transporter systems in P. aeruginosa.23–30 Of these, one system (MexAB-OprM) is constitutively expressed, two are inducibly expressed (MexCD-OprJ and MexXY) and the remaining six systems are silent or very weakly expressed in wild-type P. aeruginosa.23,31 These pumps can confer low-level protection that facilitates the initial survival of the organism in the presence of an antimicrobial agent and thus provides it with the opportunity to subsequently acquire high-level and clinically relevant resistance.
The RND efflux pump genes in B. fragilis (bmeB1-16) differ from those in Pseudomonas in several respects: (i) all of the bmeB efflux pump genes are transcribed; (ii) all of the bmeB efflux pump genes have a unique associated OMP channel gene,9 [in contrast to Pseudomonas RND (mex) efflux pump genes];31 and (iii) one efflux system operon (bmeABC11) has two efflux pump genes (bmeB11 and bmeB11') separated by an omp gene (one of the mex operons in Pseudomonas also has two pump genes,31 but they abut each other on the operon). We also found that one outer membrane channel gene is actually contiguous with its associated pump gene (bmeB10) without a separate start codon.9 To our knowledge this has not been reported in any other bacterium. Also, bmeB11' was not differentially expressed along with bmeB11 (data not shown) suggesting that the two genes are not co-expressed. We do not know if the two contiguous mex genes in Pseudomonas are co-expressed.
Functions of the various efflux pumps have been studied either by cloning and overexpressing single pumps (or pump operons) in a supersusceptible host32 or by constructing sets of strains in which one or several pump component(s) are deleted.6,33–37 The first approach provides a clearer description of the substrate profile of the efflux pump. The second approach allows for the determination of the contributory effects of efflux pumps to a resistance profile and detection of compensatory effects of other efflux pumps which may be important in clinical resistance and it has usually been found that expression of any one pump may have an impact on the expression of others.5 In our studies, single and double deletions of the selected bmeB genes did not affect expression of other bmeB genes or MIC profiles (except for bmeB3). When more than two efflux pump genes were deleted, however, others were differentially expressed and the MIC profile altered. A schematic summary of the various deletions, changes in expression and changes in MIC is presented in Figure 2. The complexity of the data does not allow for elucidation of the cause and effect relationships at this point. Rather, several possibilities can be considered: (i) increases and decreases in particular antimicrobial MICs are coordinated with increased expression or absence of bmeB3, strongly suggesting that BmeB3 can pump out these agents. The simultaneous absence of pumps BmeB1, 12 and 15 in the triple deletant WAL 189 strain may cause an increase in BmeB3, perhaps to remove toxic substances from the cell. This may also explain why WAL 189-tetS spontaneously gave rise to WAL 189-tetR, without any apparent selection pressure. We considered that the extremely high level of tetracycline resistance might be due to co-induction with other tetracycline resistance elements, but tet(A)Q2 was not overexpressed (data not shown); (ii) the simultaneous absence of bmeB1, bmeB3 and bmeB12 in WAL 215 may directly cause decreased MICs of certain antimicrobials, or the absence of these three pumps genes might cause an increase in bmeB9 expression and a compensatory decrease in expression of bmeB11, which in turn results in reduced MICs of these agents. The decreased MICs may also be due to a combination of these changes; (iii) the additional removal of bmeB15 (to the strain already missing bmeB1, 3 and 12) in WAL 219 resulted in an increase in bmeB16 expression with a corresponding increase in norfloxacin MICs; and (iv) the MIC profiles for the various pump deletants indicate substrate overlap, similar to the Pseudomonas pumps.
|
Compensatory expression changes have been previously reported in other bacteria such as P. aeruginosa, Mycobacterium smegmatis and Salmonella enterica serovar Typhimurium,3,36,38,39 and Li et al.36 suggest that these changes may serve to maintain net expression of these pumps at a certain level of expression. Lee et al.37 proposed a model in which pumps of the same class work in parallel with additive effects, while pumps of different classes work in series with multiplicative enhancement of efflux activity. In pumps with substrate overlap, inhibition of all pumps would be necessary to achieve a substantial effect. As more pumps are implicated in antimicrobial efflux, the predictions become even more complex.
Like other Gram-negative bacteria, the B. fragilis genome also contains efflux systems from other classes including ATP-binding cassette (ABC) (n = 10), major facilitator superfamily (MFS) (n = 3) and MATE (n = 15) pumps (http://www.sanger.ac.uk/projects/B_fragilis/). Understanding the nature of the interplay among the RND pumps, as well as their interaction with pumps in other classes, has important implications for drug design, both of antimicrobials and of EPIs as potentiating agents.40
In conclusion, these data demonstrate that: (i) all sixteen bmeB genes are expressed constitutively; (ii) BmeB1, BmeB3, BmeB9, BmeB11, BmeB12, BmeB15 and BmeB16 are functional efflux pumps and have overlapping substrate profiles; (iii) BmeABC pumps can cause intrinsic resistance; (iv) increased expression of bmeABC pumps can cause high-level multiple antibiotic resistance; and (v) deletion of bmeB3 can singly reduce MIC values of some antimicrobial agents. Taken together, the data strongly suggest a complex regulatory feedback system simultaneously involving multiple pumps.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We are grateful to Dr Michael Malamy and Dr Anthony Baughn (Tufts University School of Medicine, Boston, MA, USA) for providing us with the strains and plasmids for the gene deletion protocol, as well as their generous support and advice. We would also like to thank Dr Ian Paulsen at the Institute of Genomic Research (TIGR) for providing additional information on other efflux transporters present in B. fragilis and Dr Christopher Skilbeck (Greater Los Angeles Veterans Administration Healthcare Systems) for help with data analysis. This study was supported by Merit Review Funds from the Department of Veterans Affairs, USA.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
1 Li XZ and Nikaido H. (2004) Efflux-mediated drug resistance in bacteria. Drugs 64:159–204.[CrossRef][ISI][Medline]
2
Li XZ, Livermore DM, Nikaido H. (1994) Role of efflux pump(s) in intrinsic resistance of Pseudomonas aeruginosa: resistance to tetracycline, chloramphenicol, and norfloxacin. Antimicrob Agents Chemother 38:1732–41.
3
Li XZ, Zhang L, Nikaido H. (2004) Efflux pump-mediated intrinsic drug resistance in Mycobacterium smegmatis. Antimicrob Agents Chemother 48:2415–23.
4 Pumbwe L and Piddock LJ. (2002) Identification and molecular characterisation of CmeB, a Campylobacter jejuni multidrug efflux pump. FEMS Microbiol Lett 206:185–9.[CrossRef][ISI][Medline]
5 Lynch AS. (2006) Efflux systems in bacterial pathogens: an opportunity for therapeutic intervention? An industry view. Biochem Pharmacol 71:949–56.[CrossRef][ISI][Medline]
6
Chuanchuen R, Murata T, Gotoh N, et al. (2005) Substrate-dependent utilization of OprM or OpmH by the Pseudomonas aeruginosa MexJK efflux pump. Antimicrob Agents Chemother 49:2133–6.
7 Hecht DW. (2004) Prevalence of antibiotic resistance in anaerobic bacteria: worrisome developments. Clin Infect Dis 39:92–7.[CrossRef][ISI][Medline]
8 Poole K and Srikumar R. (2001) Multidrug efflux in Pseudomonas aeruginosa: components, mechanisms and clinical significance. Curr Top Med Chem 1:59–71.[CrossRef][Medline]
9
Ueda O, Wexler HM, Hirai K, et al. (2005) Sixteen homologues of the mex-type multidrug resistance efflux pump in Bacteroides fragilis. Antimicrob Agents Chemother 49:2807–15.
10
Miyamae S, Nikaido H, Tanaka Y, et al. (1998) Active efflux of norfloxacin by Bacteroides fragilis. Antimicrob Agents Chemother 42:2119–21.
11
Ricci V, Peterson ML, Rotschafer JC, et al. (2004) Role of topoisomerase mutations and efflux in fluoroquinolone resistance of Bacteroides fragilis clinical isolates and laboratory mutants. Antimicrob Agents Chemother 48:1344–6.
12
Miyamae S, Ueda O, Yoshimura F, et al. (2001) A MATE family multidrug efflux transporter pumps out fluoroquinolones in Bacteroides thetaiotaomicron. Antimicrob Agents Chemother 45:3341–6.
13
Baughn AD and Malamy MH. (2002) A mitochondrial-like aconitase in the bacterium Bacteroides fragilis: implications for the evolution of the mitochondrial Krebs cycle. Proc Natl Acad Sci USA 99:4662–7.
14 National Committee for Clinical Laboratory Standards. (2001) Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria—Fifth Edition: Approved Standard M11-A5 (NCCLS, Wayne, PA, USA).
15 Hanahan D, Jessee J, Bloom FR. (1991) Plasmid transformation of Escherichia coli and other bacteria. Methods Enzymol 204:63–113.[ISI][Medline]
16
Baughn AD and Malamy MH. (2003) The essential role of fumarate reductase in haem-dependent growth stimulation of Bacteroides fragilis. Microbiology 149:1551–8.
17
Stintzi A, Marlow D, Palyada K, et al. (2005) Use of genome-wide expression profiling and mutagenesis to study the intestinal lifestyle of Campylobacter jejuni. Infect Immun 73:1797–810.
18
Wexler HM, Molitoris E, Jashnian F, et al. (1991) Comparison of spiral gradient with conventional agar dilution for susceptibility testing of anaerobic bacteria. Antimicrob Agents Chemother 35:1196–202.
19
Masuda N, Gotoh N, Ishii C, et al. (1999) Interplay between chromosomal ß-lactamase and the MexAB-OprM efflux system in intrinsic resistance to ß-lactams in Pseudomonas aeruginosa. Antimicrob Agents Chemother 43:400–2.
20 Wexler HM, Molitoris E, Murray PR, et al. (1996) Comparison of spiral gradient endpoint and agar dilution methods for susceptibility testing of anaerobic bacteria: a multilaboratory collaborative evaluation. J Clin Microbiol 34:170–4.[Abstract]
21
Kuwahara T, Yamashita A, Hirakawa H, et al. (2004) Genomic analysis of Bacteroides fragilis reveals extensive DNA inversions regulating cell surface adaptation. Proc Natl Acad Sci USA 101:14919–24.
22
Cerdeno-Tarraga AM, Patrick S, Crossman LC, et al. (2005) Extensive DNA inversions in the B. fragilis genome control variable gene expression. Science 307:1463–5.
23 Mima T, Sekiya H, Mizushima T, et al. (2005) Gene cloning and properties of the RND-type multidrug efflux pumps MexPQ-OpmE and MexMN-OprM from Pseudomonas aeruginosa. Microbiol Immunol 49:999–1002.[Medline]
24
Chuanchuen R, Narasaki CT, Schweizer HP. (2002) The MexJK efflux pump of Pseudomonas aeruginosa requires OprM for antibiotic efflux but not for efflux of triclosan. J Bacteriol 184:5036–44.
25 Kohler T, Michea-Hamzehpour M, Henze U, et al. (1997) Characterization of MexE-MexF-OprN, a positively regulated multidrug efflux system of Pseudomonas aeruginosa. Mol Microbiol 23:345–54.[CrossRef][ISI][Medline]
26
Li Y, Mima T, Komori Y, et al. (2003) A new member of the tripartite multidrug efflux pumps, MexVW-OprM, in Pseudomonas aeruginosa. J Antimicrob Chemother 52:572–5.
27 Poole K, Gotoh N, Tsujimoto H, et al. (1996) Overexpression of the mexC-mexD-oprJ efflux operon in nfxB-type multidrug-resistant strains of Pseudomonas aeruginosa. Mol Microbiol 21:713–24.[CrossRef][ISI][Medline]
28
Sekiya H, Mima T, Morita Y, et al. (2003) Functional cloning and characterization of a multidrug efflux pump, mexHI-opmD, from a Pseudomonas aeruginosa mutant. Antimicrob Agents Chemother 47:2990–2.
29
Mine T, Morita Y, Kataoka A, et al. (1999) Expression in Escherichia coli of a new multidrug efflux pump, MexXY, from Pseudomonas aeruginosa. Antimicrob Agents Chemother 43:415–7.
30
Poole K, Krebes K, McNally C, et al. (1993) Multiple antibiotic resistance in Pseudomonas aeruginosa: evidence for involvement of an efflux operon. J Bacteriol 175:7363–72.
31 Schweizer HP. (2003) Efflux as a mechanism of resistance to antimicrobials in Pseudomonas aeruginosa and related bacteria: unanswered questions. Genet Mol Res 2:48–62.[Medline]
32
Nishino K and Yamaguchi A. (2001) Analysis of a complete library of putative drug transporter genes in Escherichia coli. J Bacteriol 183:5803–12.
33
Sulavik MC, Houseweart C, Cramer C, et al. (2001) Antibiotic susceptibility profiles of Escherichia coli strains lacking multidrug efflux pump genes. Antimicrob Agents Chemother 45:1126–36.
34 Morita Y, Kimura N, Mima T, et al. (2001) Roles of MexXY- and MexAB-multidrug efflux pumps in intrinsic multidrug resistance of Pseudomonas aeruginosa PAO1. J Gen Appl Microbiol 47:27–32.
35 Morita Y, Komori Y, Mima T, et al. (2001) Construction of a series of mutants lacking all of the four major mex operons for multidrug efflux pumps or possessing each one of the operons from Pseudomonas aeruginosa PAO1: MexCD-OprJ is an inducible pump. FEMS Microbiol Lett 202:139–43.[CrossRef][ISI][Medline]
36
Li XZ, Barre N, Poole K. (2000) Influence of the MexA-MexB-oprM multidrug efflux system on expression of the MexC-MexD-oprJ and MexE-MexF-oprN multidrug efflux systems in Pseudomonas aeruginosa. J Antimicrob Chemother 46:885–93.
37
Lee A, Mao W, Warren MS, et al. (2000) Interplay between efflux pumps may provide either additive or multiplicative effects on drug resistance. J Bacteriol 182:3142–50.
38
Eaves DJ, Ricci V, Piddock LJ. (2004) Expression of acrB, acrF, acrD, marA, and soxS in Salmonella enterica serovar Typhimurium: role in multiple antibiotic resistance. Antimicrob Agents Chemother 48:1145–50.
39
Li XZ, Zhang L, Poole K. (2000) Interplay between the MexA-MexB-OprM multidrug efflux system and the outer membrane barrier in the multiple antibiotic resistance of Pseudomonas aeruginosa. J Antimicrob Chemother 45:433–6.
40 Warren MS, Lee JC, Lee A, et al. An inhibitor of the Mex pumps from Pseudomonas aeruginosa is also a substrate of these pumps. Programs and Abstracts of the Fortieth Interscience Conference on Antimicrobial Agents and Chemotherapy2003Toronto, Canada(American Society for Microbiology, Washington, DC, USA) pp. 207 Abstract 1495.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Pumbwe, C. A. Skilbeck, and H. M. Wexler Induction of multiple antibiotic resistance in Bacteroides fragilis by benzene and benzene-derived active compounds of commonly used analgesics, antiseptics and cleaning agents J. Antimicrob. Chemother., December 1, 2007; 60(6): 1288 - 1297. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mima, S. Joshi, M. Gomez-Escalada, and H. P. Schweizer Identification and Characterization of TriABC-OpmH, a Triclosan Efflux Pump of Pseudomonas aeruginosa Requiring Two Membrane Fusion Proteins J. Bacteriol., November 1, 2007; 189(21): 7600 - 7609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Wexler Bacteroides: the Good, the Bad, and the Nitty-Gritty Clin. Microbiol. Rev., October 1, 2007; 20(4): 593 - 621. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Pumbwe, D. Glass, and H. M. Wexler Efflux Pump Overexpression in Multiple-Antibiotic-Resistant Mutants of Bacteroides fragilis. Antimicrob. Agents Chemother., September 1, 2006; 50(9): 3150 - 3153. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||