- Email: [email protected]
Molecular characterisation of ABC-type multidrug efflux systems in Bifidobacterium longum
Anaerobe 32 (2015) 63e69
Contents lists available at ScienceDirect
Anaerobe journal homepage: www.elsevier.com/locate/anaerobe
Molecular biology, genetics and biotechnology
Molecular characterisation of ABC-type multidrug efflux systems in Bifidobacterium longum Clinton Moodley, Sharon J. Reid, Valerie R. Abratt* Department of Molecular and Cell Biology, University of Cape Town, Cape Town, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 August 2014 Received in revised form 25 November 2014 Accepted 17 December 2014 Available online 18 December 2014
Administration of probiotic bacteria such as Bifidobacterium spp. can prevent antibiotic associated diarrhoea since they can survive the often harsh conditions of the gut. In Bifidobacterium longum subsp. longumT NCIMB 702259, two gene clusters, with homology to the ATP-binding cassette (ABC) family of efflux transporters, were identified and studied to assess their functional contribution to antibiotic resistance. Both gene clusters contained two genes encoding putative efflux transporters and a regulator gene, upstream of the structural genes. Reverse transcriptase analysis indicated that the genes in each cluster were transcribed as operons, one where all three genes, including a putative MarR-type regulator were transcribed together (BLLJ_1496/1495/1494), and the other where the two ABC-type transporter genes (BLLJ_1837/1836) were co-transcribed, but excluded the putative regulator (BLLJ_1838). Heterologous expression of the cloned BLLJ_1837/1836 transporter genes in Lactococcus lactis conferred resistance to erythromycin and tetracycline by increasing the minimum inhibitory concentration between 1.5 and 3 fold. The presence of these genes also allowed a 16% increase in the efflux of Hoechst 33342 from L. lactis cells containing the two transporter genes, BLLJ_1837-6. In B. longum, an increase in the levels of transcription of 3.3 fold was observed for BLLJ_1837 in the presence of erythromycin, as measured by multiplex quantitative PCR. In contrast to this, the expression of the genes of the BLLJ_1495/1494 operon in L. lactis did not show significant drug resistance functionality. Gel shift experiments showed that in the BLLJ_1495/1494 operon, the putative MarR-type regulator protein (BLLJ_1496) bound with high affinity to the DNA sequence upstream of the operon in which it was located but this was not erythromycin dependent. This study demonstrated the occurrence of a drug inducible, ABC-type transporter system (BLLJ_1837/1836) in B. longum as well as a putative MarR-type DNA binding protein (BLLJ_1496). © 2014 Elsevier Ltd. All rights reserved.
Keywords: Bifidobacterium longum ABC-Transporter Drug efflux
1. Introduction The gastrointestinal tract (GIT) of animals accommodates a complex bacterial micro-ecosystem which varies significantly between humans in a host specific manner and is influenced by numerous internal and external factors such as age, diet and health [1e3]. Gut commensals such as Bifidobacterium and Lactobacillus species, and other lactic acid bacteria, have been shown to confer health benefits on the host [4,5]. The use of antibiotics to treat bacterial infections causes significant changes in the gut microbiota of humans since beneficial bacteria are simultaneously eliminated, increasing the chances of secondary infection by opportunistic
* Corresponding author. E-mail addresses: [email protected] (C. Moodley), [email protected] (S.J. Reid), [email protected] (V.R. Abratt). http://dx.doi.org/10.1016/j.anaerobe.2014.12.004 1075-9964/© 2014 Elsevier Ltd. All rights reserved.
bacteria, which would otherwise not be able to bind to and colonise the gut because of competition with the gut commensals. Certain bacteria are innately resistant to antibiotics due to the presence of transporter proteins which allow them to remove toxic molecules before they cause cell damage. These transporters may specifically target only one molecule or several structurally unrelated molecules (multidrug transporters) [6]. Members of the ATPbinding cassette type family of transporters (ABC-type) are able to actively extrude toxic molecules, such as antibiotics, into the surrounding environment, with the concomitant hydrolysis of ATP. This is facilitated by the conserved Walker A, B and signature motif domains, which bind to and hydrolyse two ATP molecules to allow for active transport through the channel formed by twelve transmembrane segments [7,8]. Elucidation of the crystal structure of certain ABC-type transporters, such as Sav1866 from Staphylococcus aureus, has confirmed the arrangement of these membrane proteins and the joining of two protein molecules to form one
64
C. Moodley et al. / Anaerobe 32 (2015) 63e69
complete transporter [6]. The efflux mechanism of the human ABCtype efflux pump, P-glycoprotein (ABCB1), has been shown to involve alternating access to the faces of the efflux protein during ATP hydroysis [9]. Probiotic bacteria, such as certain Bifidobacterium and Lactobacillus species, are intrinsically resistant to certain antimicrobial compounds through ABC-type efflux transporters that are not readily transferred to other bacteria [10e12]. This feature would be beneficial when selecting commercial strains for probiotic supplements, allowing them to survive both bile compounds or antibiotic therapy and continue providing beneficial properties in the GIT without the risk of antibiotic resistance gene transfer to other bacteria. ABC-type multidrug efflux systems have been characterised in several probiotic bacteria including the lactococcal lmrA and lmrCD genes [13] and the horA system in Lactobacillus [14,15]. B. longum, however, has not yet been characterised in this context. In this study, B. longum NCIMB 702259 ABC-type transporters and regulator proteins, which may play a role in conferring resistance to erythromycin and tetracycline, were identified and characterised. 2. Methods 2.1. Bacterial strains and plasmids B. longum subsp. longumT NCIMB 702259 (NC_015067.1) [16] cells (NCIMB, UK) were grown anaerobically at 37 C for 24e48 h in BYG medium [17] as described by Magwira et al. [18]. Lactococcus lactis subsp. lactis NZ9000 and NZ9700 [19] and the expression vectors pNZ8048 and pNZE8048 [20] were grown aerobically for 24 h at 30 C without agitation in M17 medium (Sigma Aldrich) supplemented with 1% Glucose (Sigma) (GM17) or 1% Glucose and 1% Sucrose (GSM17 medium). Escherichia coli BL21 (pLysS) (Novagen) cells and E. coli DH5a cells were grown aerobically at 37 C for 24 h in LuriaeBertani (LB) broth [21]. 2.2. Genomic analysis Putative ABC-type transporter gene-clusters and their transcriptional regulators were identified in the published B. longum NCC2705 (NC_004307) genome (Transport DB tool) [22] and the homologous genes BLLJ_1838-37-36 (YP_004221594-93-92) and BLLJ_1496-95-94 (YP_004221255-54-53) identified in B. longum NCIMB 702259 [16]. The Walker A, B and Signature motifs [23,24], were confirmed (CDD algorithm) (www.ncbi.nlm.nih.gov), and transmembrane domains identified using hydropathy plots [25] and TopPred 0.01 software. Multiple protein-sequence alignments were constructed using ClustalW [26]. 2.3. DNA extraction B. longum NCIMB 702259 genomic DNA was isolated from cultures using the method of Magwira et al. [18]. Plasmids were isolated from Lactococcus lactis subsp. lactis NZ9000 (pNZ8048) and E. coli BL21 (pLysS) using the alkaline lysis method [27]. 2.4. Reverse transcriptase PCR analysis of expression of the gene clusters Total cellular RNA was extracted from B. longum cells grown in the presence or absence of 1.0 mg/ml Em [28] and residual genomic DNA was removed using DNaseI (Roche). Duplicate cDNA conversions were performed (ImProm-II Reverse Transcription System (A3800, Promega) with 1 mg RNA as template and random hexamer primers, and the products pooled. Oligonucleotide primers
(Table S1) were designed using the published genome sequence of B. longum NCC2705 (NC_004307) to target the gene homologues in B. longum NCIMB 702259. PCR reactions were carried out in a GeneAmp 9700 thermocycler (Applied Biosystems) using the GoTaq PCR kit (Promega) in 25 ml reactions containing 0.4 mM dNTP mix (Promega), 1.5 mM MgCl2, 0.5 mM of each oligonucleotide primer, 0.5 U GoTaq, and 2 ml cDNA as template. Genomic DNA and RNA were used for positive and negative controls, respectively, and PCR products were visualised by ethidium bromide stained agarose gel electrophoresis. 2.5. Quantitative real time PCR (qRT-PCR) B. longum NCIMB 702259 cells were grown in BYG broth to midlog phase (OD600nm z 0.3) before exposure to 1.0 mg/ml Em, with unexposed cells as the control. RNA was extracted after 2 h of growth [28], with the addition of a final purification step, using the RNeasy kit (Roche). Experiments were conducted in biological triplicate. Duplicate cDNA reactions were performed as described before, using 2 mg RNA. cDNA production was confirmed using primers Bif 164F and Bif 662R PCR [29] (Table S1), before pooling the reactions. Expression of the gene(s) of interest (GOI) in the absence and presence Em was analysed using Taqman based multiplex qRT-PCR [30]. Oligonucleotide primers and probes (Table S1) were designed using the Beacon Designer 7.21 software (PREMIER Biosoft International) and targeted internal fragments of BLLJ_1495 and BLLJ_1837, as well as the 16S rRNA gene. Each TaqMan multiplex qRT-PCR reaction was performed in a 12.5 ml reaction mixture containing 1SensiMix, 200 nM of each oligonucleotide primer (16S rRNA and GOI), 66.67 nM of each fluorescently labelled TaqMan probe (16S rRNA and GOI) and 1 ml of cDNA from uninduced or Em induced cells. RNA (20 ng) and 1 ml sterile dH2O were used as the RNA or no template controls (NTC), respectively (Table S1). For standard curve construction, equal volumes of uninduced and induced cDNA were pooled and then serially diluted from 100 to 104 in sterile dH2O. All measurements were conducted in technical triplicate and each replicate experiment was measured in the same run using the Rotor-Gene 6000 Series (Corbett Life Science) and analysed using the supplied software (Version 1.7.87). The relative fold change in gene expression of the GOI, compared to the 16S rRNA gene was normalised under the same conditions, and calculated using the Pfaffl method [31,32]. To infer statistical significance, data sets were normalized by calculating the Log10 equivalent of each value and a regression analysis was carried out and t-tests performed [33]. 2.6. Cloning and functional expression of genes The BLLJ_1837e1836 operon was PCR amplified using the High Fidelity PCR Enzyme Mix (Fermentas), with B. longum NCIMB 702259 genomic DNA and the oligonucleotide primers BLLJ_1837LF and BLLJ_1836LR (Table S1). The BLLJ_1837-6 PCR fragment and pNZ8048 were digested with PscI and HindIII (Fermentas), and ligated using T4 DNA ligase (Fermentas) to generate plasmid pNZ1837-6. To construct plasmid pNZ1837D6, a deletion of pNZ1837-6 was made by digesting it with KspAI and HindIII, gel purifying the 2354 bp fragment, blunt-ending with T4 DNA polymerase (Fermentas) and self-ligating the fragment using T4 DNA ligase (Fermentas). The BLLJ_1495-4 operon was cloned using primers BLLJ_1495FF (50 AATATATTTAAATATGAGCGATACCGCAGAGGC 30 ) and 0 0 BLLJ_1494FR (5 CAATTCAAATGTAAGCTTTTCACC 3 ). The PCR product and pTZ57 R/T (InsTAclone kit, Fermentas) were digested with SmaI and HindIII, ligated and transformed into competent
C. Moodley et al. / Anaerobe 32 (2015) 63e69
65
E. coli JM109 cells [32]. The PCR conditions were: 95 Ce5 min, 30 cycles 95 C e 30 s, 56 C e 30 s, 72 C e 4 min, polish 72 C e 10 min. The insert was released by digestion with NotI, followed by S1 nuclease treatment to generate a blunt 50 end, and then digestion with HindIII before ligating into prepared pNZ8048, to generate pNZ1495-4. The BLLJ_1495 gene was PCR cloned using the primers BLLJ_1495LF and BLLJ_1495LR (Table S1) and Supertherm Taq (Southern BioCross) followed by gel purification and ligation into pTZ57 R/T. Recombinant plasmid DNA was extracted and linearized by SpeI restriction digestion. After partial digestion with NcoI, a 1996 bp band was excised from an agarose gel and purified using the Gel extraction kit (Biospin). The vector pNZ8048 was digested with NcoI and SpeI, purified (Biospin) and ligated with the BLLJ_1495 gene using T4 DNA ligase. The recombinant clone was designated pNZ1495. Electrocompetent L. lactis NZ9000 cells were transformed with 1 mg of plasmid DNA [34] resuspended in 5 ml pre-warmed GSM17 broth containing 0.1% nisin supernatant from the nisin producing strain L. lactis NZ9700, and incubated at 30 C for 1 h. Chl (5 mg/ml) was added, the culture incubated for another hour at 30 C, before cells were centrifuged and plated on GSM17 agar amended with 5 mg/ml Chl and 0.1% nisin supernatant. After incubation at 30 C for 24e48 h, antibiotic resistant recombinants were identified by target gene specific PCR. E-tests for erythromycin, tetracycline and streptomycin (AB Biodisk, Solna, Sweden) were performed on the L. lactis cells expressing the heterologous genes [35].
Target DNA fragments (Fig 1a) were PCR amplified using the relevant primer pairs (Table S1), excised from agarose gels, purified using the Gel Extraction Kit (BioSpin), and the DNA sequence confirmed. The DNA was 30 labelled and the DNA-protein binding reactions were prepared on ice using the poly-(d[I-C]) and poly-Llysine solutions (1 mg/ml), 3.1 fmol labelled DNA and a range of purified protein concentrations (DIG Gel Shift Kit, 2nd Generation, Roche). For binding inhibitions assays, varying concentrations of Em (1, 5, 10, 20, 50, 100 mM) and Na-salicylate (0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 1.2 M) (Sigma) were added to the reaction mix. The DNA-protein binding reactions were mixed with 2.5 ml of loading buffer (0.25 TBE buffer [22.25 mM Tris-borate, 0.5 mM EDTA], 60% [w/v] glycerol, 0.2% [w/v] bromophenol blue) and analyzed using a native, 6% (w/v) polyacrylamide gel (30%, 29:1 acrylamide mix (SigmaeAldrich); 0.25 TBE buffer; 0.1% APS; 0.1% TEMED), at 100 V (Mini-PROTEAN Electrophoresis System, BioRad), at 4 C. Free DNA and DNA-protein complexes were transferred to a Hybond-Nþ nylon membrane (Amersham Biosciences) via electroblotting, UV-crosslinked and visualised using the chemiluminescent CSPD substrate (Roche) (DIG Application Manual for Filter Hybridisation).
2.7. Efflux assays
3. Results and discussion
The efflux assays were performed using whole L. lactis cells [36]. Cultures were grown to OD600 of 0.6e0.8 at 30 C and nisin from L. lactis NZ9700 was added to induce transcription. No nisin was added to the control. After 90 min growth at 30 C, the cells were harvested by centrifugation and the pellets resuspended in 10 ml 50 mM Potassium-HEPES (pH 7.0) þ 5 mM MgSO4 (Sigma). To deenergise the cells, 50 ml of a 100 mM 2,4-dinitrophenol solution (0.5 mM) (Sigma) was added and the cells were incubated for 30 min at 30 C followed by four washes with 5 ml of 50 mM Potassium-HEPES (pH7.0) and 5 mM MgSO4. To measure the rate of efflux, the washed cells were diluted to an OD660 of z0.5 in 15 ml of HEPES buffer and 22.48 ml of Bisbenzimide Hoechst 33342 (1 mM) (Sigma) was added. Aliquots (200 ml) of each strain were added to the wells in black microtiter plates (Amersham). Excitation (365 nm) and emission (457 nm) measurements were recorded at 2 min intervals for a total of 20 min using the UV optical kit in the Modulus Microplate Fluorometer 9300 (Turner BioSystems). The cells were re-energised with 10 ml of a 25 mM glucose solution and dye efflux was measured over 6 min after which 1.05 ml of 100 mM ortho-vanadate (0.5 mM final) (Sigma) was added and efflux measured over a further 20 min.
3.1. Genomic analysis
2.8. Regulator gene expression and protein purification The putative regulator proteins BLLJ_1838 and BLLJ_1496 were PCR amplified from B. longum NCIMB 702259 genomic DNA using HiFi Taq (Fermentas) (Table S1). The resulting amplicons were RE digested (Table S1) and ligated into similarly digested pET22b (þ) (Novagen). Recombinant putative regulator proteins, BLLJ_1838 or BLLJ_1496, were over-expressed in E. coli BL21 (pLysS) (pET System Manual, Novagen). The protein was isolated [21], purified using the HIS-Select nickel affinity agarose column (SigmaeAldrich) and quantified [37]. Proteins were separated using SDS-polyacrylamide gel electrophoresis (PAGE) in the Mini-PROTEAN Electrophoresis System (Bio-Rad). Hexahistidine-tagged proteins were detected by standard Western blot using an anti-6-His rabbit antibody together
with the chromogenic TMB membrane peroxidase substrate (GeneTex) [21]. 2.9. Electrophoretic mobility shift assays (EMSA)
Putative ABC-type transporter gene-clusters, BL0162-1063 and BL1766-1767 from B. longum NCC2705, were selected from the list of possible transporters on the TransportDB website [22]. Bioinformatic analysis of these genes in B. longum NCIMB 702259 confirmed the presence of homologous gene clusters, each containing two putative ABC-type transporter genes (BLLJ_1837-6 and BLLJ_1495-4), and a putative transcriptional regulator upstream of each gene cluster (BLLJ_1838 and BLLJ_1496) (Fig. 1a). All four of the deduced ABC-type transporter proteins possessed the Walker A, B and Signature motifs, as well as the 6 transmembrane domains found in all ABC-type transporters [38,39]. Bioinformatic analysis of the regulator proteins indicated that both BLLJ_1496 and BLLJ_1838 had strong homology to proven MarR regulator proteins and possessed the characteristic MarR DNA-binding domain [40,41] (Fig. S1). BLLJ_1838, however, lacked the MarR ‘winged’ domain which was present in BLLJ_1496. These results suggested that BLLJ_1496 may belong to the MarR family of regulators. A study conducted by Margolles et al. [42] in Bifidobacterium breve UCC2003 showed the presence of a similarly clustered set of ABCtype efflux transporter genes (bbmA and bbmB), which conferred multidrug resistance on the host. Amino acid sequence comparison of the translated sequences showed that the B. longum proteins had only 24e54% amino acid sequence identity to BbmA and BbmB from B. breve. B. breve UCC2003 does contain homologues of the B. longum BLLJ_1496-5-4 and BLLJ_1838-7-6 gene clusters, which show 95e98% and 87e88% amino acid sequence identity with the translated proteins from B. longum. However, these newly identified gene clusters have not been functionally characterised in B. breve. 3.2. Transcriptional analysis of gene arrangement The arrangement of functionally related genes grouped together with a putative regulator is indicative of a possible operon-like
66
C. Moodley et al. / Anaerobe 32 (2015) 63e69
Fig. 1. RT-PCR of intergenic regions of selected gene clusters. (a) Regions targeted for RT-PCR. Numbered arrows above the block arrows indicate the DNA regions targeted for RTPCR analysis. Lettered regions below the block arrows indicate the regions targeted for EMSA. (b) RT-PCR results. Lanes 1, 2 amplicons from target region 1; lanes 3, 4 amplicons from target region 2; lanes 5, 6 amplicons from target region 3; lanes 7, 8 amplicons from target region 4. (c) The () sign below the image indicates the results obtained when cDNA from uninduced cells was used as template. The (þ) sign below the image indicates the results obtained when cDNA from erythromycin induced cells was used as template.
structure. Reverse transcriptase PCR was used to determine whether the genes of interest in each gene cluster were expressed on the same mRNA transcript. PCR reactions using cDNA as template and primers (Table S1) targeting intergenic regions 1 and 2 (Fig. 1a) produced amplicons, which indicated that the gene cluster BLLJ_1496-5-4 was transcribed as a three gene operon, comprising the two ABC-type structural genes and the putative transcriptional regulator, BLLJ_1496 (Fig. 1b). In the second cluster, an amplicon was generated from the BLLJ_1837-6 intergenic region 4 (Fig. 1a), but not from region 3, which indicated that BLLJ_1837-6 was transcribed as a two gene operon, with the regulator, BLLJ_1838, transcribed separately (Fig. 1b). Research to-date has showed that ABC transporters usually function as homodimers, comprised of two identical subunits [6,13], however, the gene arrangement within this three gene operon in Bifidobacterium longum suggests that the two subunits may function as a heterodimer. There was a slight increase in the transcription levels of the operons when the cells were grown in the presence of Em (Fig. 1b, lanes 2, 4 and 8), compared to the uninduced condition (Fig. 1b, lanes 1, 3 and 7). This semi-quantitative observation was further examined using qRT-PCR.
3.3. Quantitation of gene transcription levels using quantitative real time PCR Previous experiments on B. longum NCIMB 702259 showed that pre-exposure to sub-lethal concentrations of Em led to a significant increase in the Em MIC on subsequent challenge [43]. In the current study, TaqMan based qRT-PCR [30] was used to quantify RNA transcription levels in B. longum cells exposed to 1 mg/ml Em (sublethal concentration) for two hours. The transcription levels of the first structural gene of each operon (BLLJ_1495 and BLLJ_1837) were examined by multiplex qRT-PCR in relation to the 16S rRNA reference gene [44,45]. There was a statistically significant 3.3 fold increase in transcription of BLLJ_1837 following cell exposure to 1 mg/ml Em (p ¼ 0.018). BLLJ_1495 showed an increase in transcription (1.8 fold) after Em exposure, but this was not significant. There were no significant changes in the transcription levels of the 16S rRNA gene under these conditions. This data correlates well with other studies which have shown that bacteria exposed to sub-
inhibitory concentrations of antibiotics can exhibit increased levels of gene transcription, including those involving multidrug efflux systems [46e48]. 3.4. Gene cloning and functional expression The Nisin Controlled Expression (NICE) system in Lactococcus lactis NZ9000 [49] was used to assess the functional role of the selected ABC-type transporter genes in this study. The genes BLLJ_1837-6, BLLJ_1837D6 and BLLJ_1495 were expressed as translational fusions using the pNZ8048 expression vector, under control of the nisin-inducible promoter. No stable clone of BLLJ_1495-4 could be generated despite numerous strategies. Although recombinant plasmids (pNZ1495-4) were initially observed, they did not survive subculture in the presence of selective pressure, possibly due to instability of these two genes when expressed in a heterologous host without the regulator gene. The Etest results indicated that the heterologously expressed transporter genes, BLLJ_1837-6, did confer increased resistance to both Em and Tet on the L. lactis host cells, increasing the levels of resistance approximately 3-fold compared to L. lactis carrying the empty vector (Table 1). The plasmid constructs containing only the first structural gene of each operon showed only a 1.5 e 2-fold increase when compared to the empty vector, indicating some residual activity for these efflux subunits, possibly acting as homodimers [14,50]. To confirm that the above-mentioned antibiotic resistance was due to increased efflux of the antibiotics, efflux studies were carried out. The transport of Hoechst 33342 was investigated in Lactococcus lactis cells containing plasmids with the different constructs in the presence or absence of nisin. Hoechst 33342 fluoresces when partitioned in the membrane, and is pumped out of the cells by
Table 1 MIC and fold increase results of heterologously expressed B. longum efflux genes in L. lactis pNZ9000. Recombinant plasmids expressed Em MIC in L. lactis NZ900 (mg/ml)
Em fold increase
Tet MIC (mg/ml)
Tet fold increase
pNZ8048 vector (no insert) pNZ1837-6 pNZ1837D6 pNZ1495
e 3.04 1.52 1.52
0.125 0.38 0.25 0.25
e 3.04 2.00 2.00
0.125 0.38 0.19 0.19
C. Moodley et al. / Anaerobe 32 (2015) 63e69
functional transporters when glucose is provided, resulting in a decrease in fluorescence. The rates of Hoechst transport were determined from the slope of each graph during the energised efflux phase when uninduced or induced with nisin. The rates of active efflux during the glucose re-energising phase showed a notable increase in efflux for cells harbouring the exogenous ABCtype transporters when nisin was present, with the plasmid bearing BLLJ_1837-6 showing a 16.4% increase in efflux activity. When the BLLJ_1836 gene was deleted from this construct, only a 7.1% increase was observed, and the plasmid containing the BLLJ_1495 gene only caused a marginal increase of 2.8%. 3.5. Regulator studies using EMSA EMSA was used to determine whether regulation of the two operons under investigation was controlled by the proteins encoded by the genes upstream of the transporters. The BLLJ_1838 and BLLJ_1496 putative MarR transcriptional regulator genes were cloned and their proteins expressed using the pET expression system. The purified proteins were used in in vitro binding experiments to possible DNA target sequences (Fig. 2). Based on the bioinformatic and transcriptional analysis of the operons, DNA regions consisting of approximately 250e300 bp upstream of each regulator gene, as well as the region between the regulators and the structural genes in each operon, were selected as possible regulator binding sites [50] (Fig. 1a, targets A e D). These DNA sequences were PCR amplified, gel excised and DIG-labelled for EMSA experiments. When the purified BLLJ_1496 protein from the 3-gene operon was added to target region A DNA, a notable, concentrationdependent shift in the migration of the bound DNA was observed on a native PAGE gel (Fig. 2). At high concentrations (2e3 mg) of added protein, a “super-shift” was observed (Fig. 2, lanes 7 and 8) indicating the presence of a potential secondary binding site in this region, which is commonly observed for MarR-type, regulator-DNA binding reactions [51]. The ability of BLLJ_1496 to bind to and regulate other cognate ABC-type transporter genes was also tested, since it has been shown that MarR-type regulators in Escherichia coli and S. typhimurium may recognise similar binding sites for other promoters in their regulon [52]. An EMSA, conducted using the target DNA fragments from regions B e D of the second efflux operon (Fig. 1a), showed no binding to the target region (data not shown). The BLLJ_1496 protein also did not bind to the proven functional promoter sequence of the a-galactosidase gene from
67
Bifidobacterium longum NCC2705 [53] (data not shown). This confirmed that the BLLJ_1496 protein did not recognise general promoter sequences and is likely to be the specific transcriptional regulator of the BLLJ_1496-5-4 operon. Sodium salicylate has been previously shown to disrupt the binding of MarR-type transcriptional regulators to their cognate DNA sequences, thus inducing gene expression [50e52]. However, in B. longum, sodium salicylate appeared to have very little effect on regulator binding, causing only a minor, reduced band-shift when very high concentrations (1 and 1.2 M) were added (data not shown). It has been shown that the exposure of B. longum NCIMB 702259 to sub-lethal concentrations of erythromycin resulted in an increase in the transcription levels of genes downstream of the MarR-type transcriptional regulator BLLJ_1496 (Fig. 1). The EMSA was therefore conducted in the presence of different concentrations of Em using the target A DNA fragment with 3 mg of the purified BLLJ_1496 protein. The results showed no reduction in binding efficiency in the presence of erythromycin as an effector molecule (data not shown). Bioinformatic and transcriptional analysis of the second operon, BLLJ_1837-6, suggested several candidate DNA regions for possible regulator binding (Fig. 1a, targets B e D). EMSA was carried out using the purified BLLJ_1838 putative transcriptional regulator protein with these target regions upstream of each ORF. No BLLJ_1838 protein binding was observed with any of the DNA targets using the in vitro conditions tested. The role of this protein in regulating the two gene ABC-type transporter operon, therefore, cannot be confirmed. The data presented here using the BLLJ_1496 protein showed a gene-specific, high affinity binding to the putative promoter region upstream of the structural genes of the operon, giving a strong indication of a regulatory function for this protein. However, erythromycin was not a direct effector molecule of the BLLJ_1496 protein, and this is supported by the finding that transcription of the 3-gene efflux operon was not significantly up-regulated by sublethal concentration of this antibiotic. MarR-type regulators are known to bind to un-conserved, direct-repeat sequences called ‘marboxes’, upstream of the resistance genes they regulate [50,52,54]. However, no similar sequence could be identified in the upstream regions of the B. longum efflux genes. The mar regulon in bacteria is known to lead to the up-regulation of several multidrug resistance genes through the use of MarA, the mar activator protein which is part of the marRAB locus. MarA acts as the transcriptional activator of the mar regulon by de-repressing binding of marR repressor proteins to their specific DNA sequences [50,52,54]. No
Fig. 2. Native PAGE gel of EMSA of BLLJ_1497-6 Target A DNA fragment with purified BLLJ_1496 protein. Lane 1, BLLJ_1497-96 DIG labelled DNA, lanes 2-8 labelled DNA with increasing amounts of purified BLLJ_1496 protein. Lanes 9 and 10, same as 6 and 7 except with 100 ng unlabelled DNA added as competitor (Comp).
68
C. Moodley et al. / Anaerobe 32 (2015) 63e69
marA gene homolog has been identified in B. longum NCIMB 702259 and further analysis is required to identify other transcriptional activators which may act as up-regulators of the ABCtype transporter genes identified in this study. 4. Conclusions Probiotic bacteria, such as Bifidobacterium species, are a major constituent of the gut microbiota of both humans and animals and their presence may confer health benefits on the host. However, their ability to survive in the gastrointestinal tract requires that they can withstand various antimicrobial substances such as bile and antibiotics. In this study, it was confirmed that, in addition to the Ctr transporter which is involved in the transport of bile compounds [43], B. longum harbours at least two clusters of genes encoding ABC-type efflux systems. Reverse transcriptase PCR experiments showed that the two gene clusters investigated here were transcribed as operons, one comprising three genes including a MarR transcriptional regulator, and the other comprising two genes, with an adjacent transcriptional regulator. qRT-PCR showed that the transcription of the genes of one of these operons, BLLJ_1837-6, was up-regulated when Em, a known substrate of ABC-type transporters, was present. The ABC-type transporter genes from this operon were cloned and heterologously expressed in Lactococcus lactis cells, where they conferred the host with increased resistance to erythromycin and tetracycline. The presence of these genes also allowed a 16% increase in the efflux rate of Hoechst 33342 from L. lactis cells containing the two transporter genes, BLLJ_1837-6. On the basis of this data, we propose that these genes BLLJ_1837 and BLLJ_1836 be named blmA (Bifidobacterium longum multidrug transporter) and blmB, respectively. RegulatorDNA binding studies showed that, for the three gene operon BLLJ_1496-5-4, the transcriptional regulator protein BLLJ_1494 was able to bind specifically and with high affinity to target DNA upstream of the operon, however, the expression of this operon in L. lactis did not show significant functionality in drug efflux, and the cloned genes were not stably maintained on a plasmid. This is the first experimental evidence pointing to a functional MarR-type DNA-binding protein in B. longum, which may regulate its cognate ABC transporter gene cluster. This study describes the occurrence of ABC-type transporter genes in a probiotic bacterium, which is widely used in the food industry as a dietary supplement. The function of these genes may have a protective function against cytotoxic compounds, such as bile, antibiotics and other antimicrobial substances found in the natural habitat of the bacterium or in the mammalian gut. Acknowledgements We would like to thank Dr. A. Margolles (Spain) for providing us with the L. lactis NICE gene expression system. We acknowledge The Woolworths Trust, the South African National Research Foundation and the University of Capetown for contributing to the funding of this work. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.anaerobe.2014.12.004. References [1] E.G. Zoetendal, A.D.L. Akkermans, W.M. De Vos, Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal samples reveals stable and host-specific communities of active bacteria, Appl. Environ. Microbiol. 64
(1998) 3854e3859. [2] H. Brüssow, Microbiota and healthy ageing: observational and nutritional intervention studies, Microb. Biotechnol. 6 (2013) 326e334. [3] M.J. Claesson, I.B. Jeffery, S. Conde, S.E. Power, E.M. O/'Connor, S. Cusack, et al., Gut microbiota composition correlates with diet and health in the elderly, Nature 488 (2012) 178e184. [4] W.H. Holzapfel, U. Schillinger, Introduction to pre- and probiotics, Food Res. Int. 35 (2002) 109e116. [5] C. Picard, J. Fioramonti, A. Francois, T. Robinson, F. Neant, C. Matuchansky, Bifidobacteria as probiotic agents e physiological effects and clinical benefits, Aliment. Pharmacol. Ther. 22 (2005) 495e512. [6] R.J.P. Dawson, K.P. Locher, Structure of a bacterial multidrug ABC transporter, Nature 443 (2006) 180e185. [7] C.F. Higgins, ABC transporters: physiology, structure and mechanism e an overview, Res. Microbiol. 152 (2001) 205e210. [8] A.L. Davidson, E. Dassa, C. Orelle, J. Chen, Structure, function, and evolution of bacterial ATP-binding cassette systems, Microbiol. Mol. Biol. Rev. 72 (2008) 317e364. [9] J.H. van Wonderen, R.M. McMahon, M.L. O'Mara, C.A. McDevitt, A.J. Thomson, I.D. Kerr, et al., The central cavity of ABCB1 undergoes alternating access during ATP hydrolysis, Fed. Eur. Biochem. Soc. 281 (2014) 2190e2201. [10] L. Masco, K. Van Hoorde, E. De Brandt, J. Swings, G. Huys, Antimicrobial susceptibility of Bifidobacterium strains from humans, animals and probiotic products, J. Antimicrob. Chemother. 58 (2006) 85e94. [11] M.R. D'Aimmo, M. Modesto, B. Biavati, Antibiotic resistance of lactic acid bacteria and Bifidobacterium spp. isolated from dairy and pharmaceutical products, Int. J. Food Microbiol. 115 (2007) 35e42. nchez, C.G. de los Reyes-Gavila n, A. Margolles, Antibiotic [12] M. Gueimonde, B. Sa resistance in probiotic bacteria, Front. Microbiol. 4 (2013). [13] K. Pleban, A. Macchiarulo, G. Costantino, R. Pellicciari, P. Chiba, G.F. Ecker, Homology model of the multidrug transporter LmrA from Lactococcus lactis, Bioorg. Med. Chem. Lett. 14 (2004) 5823e5826. [14] J. Lubelski, W.N. Konings, A.J.M. Driessen, Distribution and physiology of ABCtype transporters contributing to multidrug resistance in bacteria, Microbiol. Mol. Biol. Rev. 71 (2007) 463e476. [15] K. Sakamoto, A. Margolles, H.W. van Veen, W.N. Konings, Hop resistance in the beer spoilage bacterium Lactobacillus brevis is mediated by the ATP-binding cassette multidrug transporter HorA, J. Bacteriol. 183 (2001) 5371e5375. [16] S. Fukuda, H. Toh, K. Hase, K. Oshima, Y. Nakanishi, K. Yoshimura, et al., Bifidobacteria can protect from enteropathogenic infection through production of acetate, Nature 469 (2011) 543e547. [17] B.A. Degnan, G.T. Macfarlane, Transport and metabolism of glucose and arabinose in Bifidobacterium breve, Arch. Microbiol. 160 (1993) 144e151. [18] C.A. Magwira, B. Kullin, S. Lewandowski, A. Rodgers, S.J. Reid, V.R. Abratt, Diversity of faecal oxalate-degrading bacteria in black and white South African study groups: insights into understanding the rarity of urolithiasis in the black group, J. Appl. Microbiol. 113 (2012) 418e428. [19] O.P. Kuipers, P.G.G.A. de Ruyter, M. Kleerebezem, W.M. de Vos, Quorum sensing-controlled gene expression in lactic acid bacteria, J. Biotechnol. 64 (1998) 15e21. [20] P.G. de Ruyter, O.P. Kuipers, W.M. de Vos, Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin, Appl. Environ. Microbiol. 62 (1996) 3662e3667. [21] J. Sambrook, D. Russell, Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory Press, 2001. [22] Q. Ren, K.H. Kang, I.T. Paulsen, Transport DB: a relational database of cellular membrane transport systems, Nucleic Acids Res. 32 (D284-D8) (2004). [23] A. Marchler-Bauer, J.B. Anderson, F. Chitsaz, M.K. Derbyshire, C. DeWeeseScott, J.H. Fong, et al., CDD: specific functional annotation with the Conserved Domain Database, Nucleic Acids Res. 37 (D205-D10) (2009). [24] A. Marchler-Bauer, S. Lu, J.B. Anderson, F. Chitsaz, M.K. Derbyshire, C. DeWeese-Scott, et al., CDD: a Conserved Domain Database for the functional annotation of proteins, Nucleic Acids Res. 39 (2011). D225-D9. [25] J. Kyte, R.F. Doolittle, A simple method for displaying the hydropathic character of a protein, J. Mol. Biol. 157 (1982) 105e132. [26] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673e4680. [27] D. Ish-Horowicz, J.F. Burke, Rapid and efficient cosmid cloning, Nucleic Acids Res. 9 (1981) 2989e2998. [28] C. Magni, P. Marini, D. De Mendoza, Extraction of RNA from Gram-positive bacteria, Biotechniques 19 (1995) 880e884. [29] P.S. Langendijk, F. Schut, G.J. Jansen, G.C. Raangs, G.R. Kamphuis, M.H. Wilkinson, et al., Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its application in fecal samples, Appl. Environ. Microbiol. 61 (1995) 3069e3075. [30] P.M. Holland, R.D. Abramson, R. Watson, D.H. Gelfand, Detection of specific polymerase chain reaction product by utilizing the 50 e–30 exonuclease activity of Thermus aquaticus DNA polymerase, Proc. Natl. Acad. Sci. 88 (1991) 7276e7280. [31] S.A. Bustin, V. Benes, J.A. Garson, J. Hellemans, J. Huggett, M. Kubista, et al., The MIQE guidelines: minimum information for publication of quantitative realtime PCR experiments, Clin. Chem. 55 (2009) 611e622. [32] M.W. Pfaffl, A new mathematical model for relative quantification in real-time
C. Moodley et al. / Anaerobe 32 (2015) 63e69 RTePCR, Nucleic Acids Res. 29 (2001) e45. [33] S. Fleige, V. Walf, S. Huch, C. Prgomet, J. Sehm, M.W. Pfaffl, Comparison of relative mRNA quantification models and the impact of RNA integrity in quantitative real-time RT-PCR, Biotechnol. Lett. 28 (2006) 1601e1613. [34] M. Papagianni, N. Avramidis, G. Filioussis, High efficiency electrotransformation of Lactococcus lactis spp. lactis cells pretreated with lithium acetate and dithiothreitol, Biomed. Cent. Biotechnol. 7 (2007) 15. n, Mac[35] A. Margolles, J.A. Moreno, D. van Sinderen, C.G. de los Reyes-Gavila rolide resistance mediated by a Bifidobacterium breve membrane protein, Antimicrob. Agents Chemother. 49 (2005) 4379e4381. [36] Van Den Berg, H.B. Van Saparoea, J. Lubelski, R. Van Merkerk, P.S. Mazurkiewicz, A.J.M. Driessen, Proton motive force-dependent Hoechst 33342 transport by the ABC transporter LmrA of Lactococcus lactis, Biochemistry 44 (2005) 16931e16938. [37] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248e254. [38] S.C. Hyde, P. Emsley, M.J. Hartshorn, M.M. Mimmack, U. Gileadi, S.R. Pearce, et al., Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport, Nature 346 (1990) 362e365. [39] J.E. Walker, M. Saraste, M.J. Runswick, N.J. Gay, Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATPrequiring enzymes and a common nucleotide binding fold, Eur. Mol. Biol. Organ. J. 1 (1982) 945e951. [40] C.O. Pabo, R.T. Sauer, Transcription factors: structural families and principles of DNA recognition, Annu. Rev. Biochem. 61 (1992) 1053e1095. [41] M.N. Alekshun, S.B. Levy, T.R. Mealy, B.A. Seaton, J.F. Head, The crystal structure of MarR, a regulator of multiple antibiotic resistance, at 2.3 A resolution, Nat. Struct. Biol. 8 (2001) 710e714. rez, J.A. Moreno, D. van Sinderen, C.G. de los Reyes[42] A. Margolles, A.B. Flo Gavil an, Two membrane proteins from Bifidobacterium breve UCC2003 constitute an ABC-type multidrug transporter, Microbiology 152 (2006) 3497e3505. [43] C.E. Price, S.J. Reid, A.J.M. Driessen, V.R. Abratt, The Bifidobacterium longum NCIMB 702259T ctr gene codes for a novel cholate transporter, Appl. Environ.
69
Microbiol. 72 (2006) 923e926. [44] I. Nazarenko, B. Lowe, M. Darfler, P. Ikonomi, D. Schuster, A. Rashtchian, Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore, Nucleic Acids Res. 30 (2002) e37. [45] W. Li, B. Xi, W. Yang, M. Hawkins, U.K. Schubart, Complex DNA melting profiles of small PCR products revealed using SYBR Green I, Biotechniques 35 (2003) 702e704, 6. [46] A.A. Huet, J.L. Raygada, K. Mendiratta, S.M. Seo, G.W. Kaatz, Multidrug efflux pump overexpression in Staphylococcus aureus after single and multiple in vitro exposures to biocides and dyes, Microbiology 154 (2008) 3144e3153. [47] J.T. Lin, M.B. Connelly, C. Amolo, S. Otani, D.S. Yaver, Global transcriptional response of Bacillus subtilis to treatment with subinhibitory concentrations of antibiotics that inhibit protein synthesis, Antimicrob. Agents Chemother. 49 (2005) 1915e1926. [48] G.E. Louw, R.M. Warren, N.C. Gey van Pittius, R. Leon, A. Jimenez, R. Hernandez-Pando, et al., Rifampicin reduces susceptibility to ofloxacin in rifampicin-resistant Mycobacterium tuberculosis through efflux, Am. J. Respir. Crit. Care Med. 184 (2011) 269e276. [49] I. Mierau, M. Kleerebezem, 10 years of the nisin-controlled gene expression system (NICE) in Lactococcus lactis, Appl. Microbiol. Biotechnol. 68 (2005) 705e717. [50] R.G. Martin, J.L. Rosner, Binding of purified multiple antibiotic-resistance repressor protein (MarR) to mar operator sequences, Proc. Natl. Acad. Sci. 92 (1995) 5456e5460. [51] R.G. Martin, P.S. Nyantakyi, J.L. Rosner, Regulation of the multiple antibiotic resistance (mar) regulon by marORA sequences in Escherichia coli, J. Bacteriol. 177 (1995) 4176e4178. [52] M.N. Alekshun, S.B. Levy, Regulation of chromosomally mediated multiple antibiotic resistance: the Mar regulon, Antimicrobial. Agents Chemother. 41 (1997) 2067e2075. [53] A. Klijn, D. Moine, M. Delley, A. Mercenier, F. Arigoni, R.D. Pridmore, Construction of a reporter vector for the analysis of Bifidobacterium longum promoters, Appl. Environ. Microbiol. 72 (2006) 7401e7405. [54] M.N. Alekshun, S.B. Levy, The mar regulon: multiple resistance to antibiotics and other toxic chemicals, Trends Microbiol. 7 (1999) 410e413.
Contents lists available at ScienceDirect
Anaerobe journal homepage: www.elsevier.com/locate/anaerobe
Molecular biology, genetics and biotechnology
Molecular characterisation of ABC-type multidrug efflux systems in Bifidobacterium longum Clinton Moodley, Sharon J. Reid, Valerie R. Abratt* Department of Molecular and Cell Biology, University of Cape Town, Cape Town, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 August 2014 Received in revised form 25 November 2014 Accepted 17 December 2014 Available online 18 December 2014
Administration of probiotic bacteria such as Bifidobacterium spp. can prevent antibiotic associated diarrhoea since they can survive the often harsh conditions of the gut. In Bifidobacterium longum subsp. longumT NCIMB 702259, two gene clusters, with homology to the ATP-binding cassette (ABC) family of efflux transporters, were identified and studied to assess their functional contribution to antibiotic resistance. Both gene clusters contained two genes encoding putative efflux transporters and a regulator gene, upstream of the structural genes. Reverse transcriptase analysis indicated that the genes in each cluster were transcribed as operons, one where all three genes, including a putative MarR-type regulator were transcribed together (BLLJ_1496/1495/1494), and the other where the two ABC-type transporter genes (BLLJ_1837/1836) were co-transcribed, but excluded the putative regulator (BLLJ_1838). Heterologous expression of the cloned BLLJ_1837/1836 transporter genes in Lactococcus lactis conferred resistance to erythromycin and tetracycline by increasing the minimum inhibitory concentration between 1.5 and 3 fold. The presence of these genes also allowed a 16% increase in the efflux of Hoechst 33342 from L. lactis cells containing the two transporter genes, BLLJ_1837-6. In B. longum, an increase in the levels of transcription of 3.3 fold was observed for BLLJ_1837 in the presence of erythromycin, as measured by multiplex quantitative PCR. In contrast to this, the expression of the genes of the BLLJ_1495/1494 operon in L. lactis did not show significant drug resistance functionality. Gel shift experiments showed that in the BLLJ_1495/1494 operon, the putative MarR-type regulator protein (BLLJ_1496) bound with high affinity to the DNA sequence upstream of the operon in which it was located but this was not erythromycin dependent. This study demonstrated the occurrence of a drug inducible, ABC-type transporter system (BLLJ_1837/1836) in B. longum as well as a putative MarR-type DNA binding protein (BLLJ_1496). © 2014 Elsevier Ltd. All rights reserved.
Keywords: Bifidobacterium longum ABC-Transporter Drug efflux
1. Introduction The gastrointestinal tract (GIT) of animals accommodates a complex bacterial micro-ecosystem which varies significantly between humans in a host specific manner and is influenced by numerous internal and external factors such as age, diet and health [1e3]. Gut commensals such as Bifidobacterium and Lactobacillus species, and other lactic acid bacteria, have been shown to confer health benefits on the host [4,5]. The use of antibiotics to treat bacterial infections causes significant changes in the gut microbiota of humans since beneficial bacteria are simultaneously eliminated, increasing the chances of secondary infection by opportunistic
* Corresponding author. E-mail addresses: [email protected] (C. Moodley), [email protected] (S.J. Reid), [email protected] (V.R. Abratt). http://dx.doi.org/10.1016/j.anaerobe.2014.12.004 1075-9964/© 2014 Elsevier Ltd. All rights reserved.
bacteria, which would otherwise not be able to bind to and colonise the gut because of competition with the gut commensals. Certain bacteria are innately resistant to antibiotics due to the presence of transporter proteins which allow them to remove toxic molecules before they cause cell damage. These transporters may specifically target only one molecule or several structurally unrelated molecules (multidrug transporters) [6]. Members of the ATPbinding cassette type family of transporters (ABC-type) are able to actively extrude toxic molecules, such as antibiotics, into the surrounding environment, with the concomitant hydrolysis of ATP. This is facilitated by the conserved Walker A, B and signature motif domains, which bind to and hydrolyse two ATP molecules to allow for active transport through the channel formed by twelve transmembrane segments [7,8]. Elucidation of the crystal structure of certain ABC-type transporters, such as Sav1866 from Staphylococcus aureus, has confirmed the arrangement of these membrane proteins and the joining of two protein molecules to form one
64
C. Moodley et al. / Anaerobe 32 (2015) 63e69
complete transporter [6]. The efflux mechanism of the human ABCtype efflux pump, P-glycoprotein (ABCB1), has been shown to involve alternating access to the faces of the efflux protein during ATP hydroysis [9]. Probiotic bacteria, such as certain Bifidobacterium and Lactobacillus species, are intrinsically resistant to certain antimicrobial compounds through ABC-type efflux transporters that are not readily transferred to other bacteria [10e12]. This feature would be beneficial when selecting commercial strains for probiotic supplements, allowing them to survive both bile compounds or antibiotic therapy and continue providing beneficial properties in the GIT without the risk of antibiotic resistance gene transfer to other bacteria. ABC-type multidrug efflux systems have been characterised in several probiotic bacteria including the lactococcal lmrA and lmrCD genes [13] and the horA system in Lactobacillus [14,15]. B. longum, however, has not yet been characterised in this context. In this study, B. longum NCIMB 702259 ABC-type transporters and regulator proteins, which may play a role in conferring resistance to erythromycin and tetracycline, were identified and characterised. 2. Methods 2.1. Bacterial strains and plasmids B. longum subsp. longumT NCIMB 702259 (NC_015067.1) [16] cells (NCIMB, UK) were grown anaerobically at 37 C for 24e48 h in BYG medium [17] as described by Magwira et al. [18]. Lactococcus lactis subsp. lactis NZ9000 and NZ9700 [19] and the expression vectors pNZ8048 and pNZE8048 [20] were grown aerobically for 24 h at 30 C without agitation in M17 medium (Sigma Aldrich) supplemented with 1% Glucose (Sigma) (GM17) or 1% Glucose and 1% Sucrose (GSM17 medium). Escherichia coli BL21 (pLysS) (Novagen) cells and E. coli DH5a cells were grown aerobically at 37 C for 24 h in LuriaeBertani (LB) broth [21]. 2.2. Genomic analysis Putative ABC-type transporter gene-clusters and their transcriptional regulators were identified in the published B. longum NCC2705 (NC_004307) genome (Transport DB tool) [22] and the homologous genes BLLJ_1838-37-36 (YP_004221594-93-92) and BLLJ_1496-95-94 (YP_004221255-54-53) identified in B. longum NCIMB 702259 [16]. The Walker A, B and Signature motifs [23,24], were confirmed (CDD algorithm) (www.ncbi.nlm.nih.gov), and transmembrane domains identified using hydropathy plots [25] and TopPred 0.01 software. Multiple protein-sequence alignments were constructed using ClustalW [26]. 2.3. DNA extraction B. longum NCIMB 702259 genomic DNA was isolated from cultures using the method of Magwira et al. [18]. Plasmids were isolated from Lactococcus lactis subsp. lactis NZ9000 (pNZ8048) and E. coli BL21 (pLysS) using the alkaline lysis method [27]. 2.4. Reverse transcriptase PCR analysis of expression of the gene clusters Total cellular RNA was extracted from B. longum cells grown in the presence or absence of 1.0 mg/ml Em [28] and residual genomic DNA was removed using DNaseI (Roche). Duplicate cDNA conversions were performed (ImProm-II Reverse Transcription System (A3800, Promega) with 1 mg RNA as template and random hexamer primers, and the products pooled. Oligonucleotide primers
(Table S1) were designed using the published genome sequence of B. longum NCC2705 (NC_004307) to target the gene homologues in B. longum NCIMB 702259. PCR reactions were carried out in a GeneAmp 9700 thermocycler (Applied Biosystems) using the GoTaq PCR kit (Promega) in 25 ml reactions containing 0.4 mM dNTP mix (Promega), 1.5 mM MgCl2, 0.5 mM of each oligonucleotide primer, 0.5 U GoTaq, and 2 ml cDNA as template. Genomic DNA and RNA were used for positive and negative controls, respectively, and PCR products were visualised by ethidium bromide stained agarose gel electrophoresis. 2.5. Quantitative real time PCR (qRT-PCR) B. longum NCIMB 702259 cells were grown in BYG broth to midlog phase (OD600nm z 0.3) before exposure to 1.0 mg/ml Em, with unexposed cells as the control. RNA was extracted after 2 h of growth [28], with the addition of a final purification step, using the RNeasy kit (Roche). Experiments were conducted in biological triplicate. Duplicate cDNA reactions were performed as described before, using 2 mg RNA. cDNA production was confirmed using primers Bif 164F and Bif 662R PCR [29] (Table S1), before pooling the reactions. Expression of the gene(s) of interest (GOI) in the absence and presence Em was analysed using Taqman based multiplex qRT-PCR [30]. Oligonucleotide primers and probes (Table S1) were designed using the Beacon Designer 7.21 software (PREMIER Biosoft International) and targeted internal fragments of BLLJ_1495 and BLLJ_1837, as well as the 16S rRNA gene. Each TaqMan multiplex qRT-PCR reaction was performed in a 12.5 ml reaction mixture containing 1SensiMix, 200 nM of each oligonucleotide primer (16S rRNA and GOI), 66.67 nM of each fluorescently labelled TaqMan probe (16S rRNA and GOI) and 1 ml of cDNA from uninduced or Em induced cells. RNA (20 ng) and 1 ml sterile dH2O were used as the RNA or no template controls (NTC), respectively (Table S1). For standard curve construction, equal volumes of uninduced and induced cDNA were pooled and then serially diluted from 100 to 104 in sterile dH2O. All measurements were conducted in technical triplicate and each replicate experiment was measured in the same run using the Rotor-Gene 6000 Series (Corbett Life Science) and analysed using the supplied software (Version 1.7.87). The relative fold change in gene expression of the GOI, compared to the 16S rRNA gene was normalised under the same conditions, and calculated using the Pfaffl method [31,32]. To infer statistical significance, data sets were normalized by calculating the Log10 equivalent of each value and a regression analysis was carried out and t-tests performed [33]. 2.6. Cloning and functional expression of genes The BLLJ_1837e1836 operon was PCR amplified using the High Fidelity PCR Enzyme Mix (Fermentas), with B. longum NCIMB 702259 genomic DNA and the oligonucleotide primers BLLJ_1837LF and BLLJ_1836LR (Table S1). The BLLJ_1837-6 PCR fragment and pNZ8048 were digested with PscI and HindIII (Fermentas), and ligated using T4 DNA ligase (Fermentas) to generate plasmid pNZ1837-6. To construct plasmid pNZ1837D6, a deletion of pNZ1837-6 was made by digesting it with KspAI and HindIII, gel purifying the 2354 bp fragment, blunt-ending with T4 DNA polymerase (Fermentas) and self-ligating the fragment using T4 DNA ligase (Fermentas). The BLLJ_1495-4 operon was cloned using primers BLLJ_1495FF (50 AATATATTTAAATATGAGCGATACCGCAGAGGC 30 ) and 0 0 BLLJ_1494FR (5 CAATTCAAATGTAAGCTTTTCACC 3 ). The PCR product and pTZ57 R/T (InsTAclone kit, Fermentas) were digested with SmaI and HindIII, ligated and transformed into competent
C. Moodley et al. / Anaerobe 32 (2015) 63e69
65
E. coli JM109 cells [32]. The PCR conditions were: 95 Ce5 min, 30 cycles 95 C e 30 s, 56 C e 30 s, 72 C e 4 min, polish 72 C e 10 min. The insert was released by digestion with NotI, followed by S1 nuclease treatment to generate a blunt 50 end, and then digestion with HindIII before ligating into prepared pNZ8048, to generate pNZ1495-4. The BLLJ_1495 gene was PCR cloned using the primers BLLJ_1495LF and BLLJ_1495LR (Table S1) and Supertherm Taq (Southern BioCross) followed by gel purification and ligation into pTZ57 R/T. Recombinant plasmid DNA was extracted and linearized by SpeI restriction digestion. After partial digestion with NcoI, a 1996 bp band was excised from an agarose gel and purified using the Gel extraction kit (Biospin). The vector pNZ8048 was digested with NcoI and SpeI, purified (Biospin) and ligated with the BLLJ_1495 gene using T4 DNA ligase. The recombinant clone was designated pNZ1495. Electrocompetent L. lactis NZ9000 cells were transformed with 1 mg of plasmid DNA [34] resuspended in 5 ml pre-warmed GSM17 broth containing 0.1% nisin supernatant from the nisin producing strain L. lactis NZ9700, and incubated at 30 C for 1 h. Chl (5 mg/ml) was added, the culture incubated for another hour at 30 C, before cells were centrifuged and plated on GSM17 agar amended with 5 mg/ml Chl and 0.1% nisin supernatant. After incubation at 30 C for 24e48 h, antibiotic resistant recombinants were identified by target gene specific PCR. E-tests for erythromycin, tetracycline and streptomycin (AB Biodisk, Solna, Sweden) were performed on the L. lactis cells expressing the heterologous genes [35].
Target DNA fragments (Fig 1a) were PCR amplified using the relevant primer pairs (Table S1), excised from agarose gels, purified using the Gel Extraction Kit (BioSpin), and the DNA sequence confirmed. The DNA was 30 labelled and the DNA-protein binding reactions were prepared on ice using the poly-(d[I-C]) and poly-Llysine solutions (1 mg/ml), 3.1 fmol labelled DNA and a range of purified protein concentrations (DIG Gel Shift Kit, 2nd Generation, Roche). For binding inhibitions assays, varying concentrations of Em (1, 5, 10, 20, 50, 100 mM) and Na-salicylate (0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 1.2 M) (Sigma) were added to the reaction mix. The DNA-protein binding reactions were mixed with 2.5 ml of loading buffer (0.25 TBE buffer [22.25 mM Tris-borate, 0.5 mM EDTA], 60% [w/v] glycerol, 0.2% [w/v] bromophenol blue) and analyzed using a native, 6% (w/v) polyacrylamide gel (30%, 29:1 acrylamide mix (SigmaeAldrich); 0.25 TBE buffer; 0.1% APS; 0.1% TEMED), at 100 V (Mini-PROTEAN Electrophoresis System, BioRad), at 4 C. Free DNA and DNA-protein complexes were transferred to a Hybond-Nþ nylon membrane (Amersham Biosciences) via electroblotting, UV-crosslinked and visualised using the chemiluminescent CSPD substrate (Roche) (DIG Application Manual for Filter Hybridisation).
2.7. Efflux assays
3. Results and discussion
The efflux assays were performed using whole L. lactis cells [36]. Cultures were grown to OD600 of 0.6e0.8 at 30 C and nisin from L. lactis NZ9700 was added to induce transcription. No nisin was added to the control. After 90 min growth at 30 C, the cells were harvested by centrifugation and the pellets resuspended in 10 ml 50 mM Potassium-HEPES (pH 7.0) þ 5 mM MgSO4 (Sigma). To deenergise the cells, 50 ml of a 100 mM 2,4-dinitrophenol solution (0.5 mM) (Sigma) was added and the cells were incubated for 30 min at 30 C followed by four washes with 5 ml of 50 mM Potassium-HEPES (pH7.0) and 5 mM MgSO4. To measure the rate of efflux, the washed cells were diluted to an OD660 of z0.5 in 15 ml of HEPES buffer and 22.48 ml of Bisbenzimide Hoechst 33342 (1 mM) (Sigma) was added. Aliquots (200 ml) of each strain were added to the wells in black microtiter plates (Amersham). Excitation (365 nm) and emission (457 nm) measurements were recorded at 2 min intervals for a total of 20 min using the UV optical kit in the Modulus Microplate Fluorometer 9300 (Turner BioSystems). The cells were re-energised with 10 ml of a 25 mM glucose solution and dye efflux was measured over 6 min after which 1.05 ml of 100 mM ortho-vanadate (0.5 mM final) (Sigma) was added and efflux measured over a further 20 min.
3.1. Genomic analysis
2.8. Regulator gene expression and protein purification The putative regulator proteins BLLJ_1838 and BLLJ_1496 were PCR amplified from B. longum NCIMB 702259 genomic DNA using HiFi Taq (Fermentas) (Table S1). The resulting amplicons were RE digested (Table S1) and ligated into similarly digested pET22b (þ) (Novagen). Recombinant putative regulator proteins, BLLJ_1838 or BLLJ_1496, were over-expressed in E. coli BL21 (pLysS) (pET System Manual, Novagen). The protein was isolated [21], purified using the HIS-Select nickel affinity agarose column (SigmaeAldrich) and quantified [37]. Proteins were separated using SDS-polyacrylamide gel electrophoresis (PAGE) in the Mini-PROTEAN Electrophoresis System (Bio-Rad). Hexahistidine-tagged proteins were detected by standard Western blot using an anti-6-His rabbit antibody together
with the chromogenic TMB membrane peroxidase substrate (GeneTex) [21]. 2.9. Electrophoretic mobility shift assays (EMSA)
Putative ABC-type transporter gene-clusters, BL0162-1063 and BL1766-1767 from B. longum NCC2705, were selected from the list of possible transporters on the TransportDB website [22]. Bioinformatic analysis of these genes in B. longum NCIMB 702259 confirmed the presence of homologous gene clusters, each containing two putative ABC-type transporter genes (BLLJ_1837-6 and BLLJ_1495-4), and a putative transcriptional regulator upstream of each gene cluster (BLLJ_1838 and BLLJ_1496) (Fig. 1a). All four of the deduced ABC-type transporter proteins possessed the Walker A, B and Signature motifs, as well as the 6 transmembrane domains found in all ABC-type transporters [38,39]. Bioinformatic analysis of the regulator proteins indicated that both BLLJ_1496 and BLLJ_1838 had strong homology to proven MarR regulator proteins and possessed the characteristic MarR DNA-binding domain [40,41] (Fig. S1). BLLJ_1838, however, lacked the MarR ‘winged’ domain which was present in BLLJ_1496. These results suggested that BLLJ_1496 may belong to the MarR family of regulators. A study conducted by Margolles et al. [42] in Bifidobacterium breve UCC2003 showed the presence of a similarly clustered set of ABCtype efflux transporter genes (bbmA and bbmB), which conferred multidrug resistance on the host. Amino acid sequence comparison of the translated sequences showed that the B. longum proteins had only 24e54% amino acid sequence identity to BbmA and BbmB from B. breve. B. breve UCC2003 does contain homologues of the B. longum BLLJ_1496-5-4 and BLLJ_1838-7-6 gene clusters, which show 95e98% and 87e88% amino acid sequence identity with the translated proteins from B. longum. However, these newly identified gene clusters have not been functionally characterised in B. breve. 3.2. Transcriptional analysis of gene arrangement The arrangement of functionally related genes grouped together with a putative regulator is indicative of a possible operon-like
66
C. Moodley et al. / Anaerobe 32 (2015) 63e69
Fig. 1. RT-PCR of intergenic regions of selected gene clusters. (a) Regions targeted for RT-PCR. Numbered arrows above the block arrows indicate the DNA regions targeted for RTPCR analysis. Lettered regions below the block arrows indicate the regions targeted for EMSA. (b) RT-PCR results. Lanes 1, 2 amplicons from target region 1; lanes 3, 4 amplicons from target region 2; lanes 5, 6 amplicons from target region 3; lanes 7, 8 amplicons from target region 4. (c) The () sign below the image indicates the results obtained when cDNA from uninduced cells was used as template. The (þ) sign below the image indicates the results obtained when cDNA from erythromycin induced cells was used as template.
structure. Reverse transcriptase PCR was used to determine whether the genes of interest in each gene cluster were expressed on the same mRNA transcript. PCR reactions using cDNA as template and primers (Table S1) targeting intergenic regions 1 and 2 (Fig. 1a) produced amplicons, which indicated that the gene cluster BLLJ_1496-5-4 was transcribed as a three gene operon, comprising the two ABC-type structural genes and the putative transcriptional regulator, BLLJ_1496 (Fig. 1b). In the second cluster, an amplicon was generated from the BLLJ_1837-6 intergenic region 4 (Fig. 1a), but not from region 3, which indicated that BLLJ_1837-6 was transcribed as a two gene operon, with the regulator, BLLJ_1838, transcribed separately (Fig. 1b). Research to-date has showed that ABC transporters usually function as homodimers, comprised of two identical subunits [6,13], however, the gene arrangement within this three gene operon in Bifidobacterium longum suggests that the two subunits may function as a heterodimer. There was a slight increase in the transcription levels of the operons when the cells were grown in the presence of Em (Fig. 1b, lanes 2, 4 and 8), compared to the uninduced condition (Fig. 1b, lanes 1, 3 and 7). This semi-quantitative observation was further examined using qRT-PCR.
3.3. Quantitation of gene transcription levels using quantitative real time PCR Previous experiments on B. longum NCIMB 702259 showed that pre-exposure to sub-lethal concentrations of Em led to a significant increase in the Em MIC on subsequent challenge [43]. In the current study, TaqMan based qRT-PCR [30] was used to quantify RNA transcription levels in B. longum cells exposed to 1 mg/ml Em (sublethal concentration) for two hours. The transcription levels of the first structural gene of each operon (BLLJ_1495 and BLLJ_1837) were examined by multiplex qRT-PCR in relation to the 16S rRNA reference gene [44,45]. There was a statistically significant 3.3 fold increase in transcription of BLLJ_1837 following cell exposure to 1 mg/ml Em (p ¼ 0.018). BLLJ_1495 showed an increase in transcription (1.8 fold) after Em exposure, but this was not significant. There were no significant changes in the transcription levels of the 16S rRNA gene under these conditions. This data correlates well with other studies which have shown that bacteria exposed to sub-
inhibitory concentrations of antibiotics can exhibit increased levels of gene transcription, including those involving multidrug efflux systems [46e48]. 3.4. Gene cloning and functional expression The Nisin Controlled Expression (NICE) system in Lactococcus lactis NZ9000 [49] was used to assess the functional role of the selected ABC-type transporter genes in this study. The genes BLLJ_1837-6, BLLJ_1837D6 and BLLJ_1495 were expressed as translational fusions using the pNZ8048 expression vector, under control of the nisin-inducible promoter. No stable clone of BLLJ_1495-4 could be generated despite numerous strategies. Although recombinant plasmids (pNZ1495-4) were initially observed, they did not survive subculture in the presence of selective pressure, possibly due to instability of these two genes when expressed in a heterologous host without the regulator gene. The Etest results indicated that the heterologously expressed transporter genes, BLLJ_1837-6, did confer increased resistance to both Em and Tet on the L. lactis host cells, increasing the levels of resistance approximately 3-fold compared to L. lactis carrying the empty vector (Table 1). The plasmid constructs containing only the first structural gene of each operon showed only a 1.5 e 2-fold increase when compared to the empty vector, indicating some residual activity for these efflux subunits, possibly acting as homodimers [14,50]. To confirm that the above-mentioned antibiotic resistance was due to increased efflux of the antibiotics, efflux studies were carried out. The transport of Hoechst 33342 was investigated in Lactococcus lactis cells containing plasmids with the different constructs in the presence or absence of nisin. Hoechst 33342 fluoresces when partitioned in the membrane, and is pumped out of the cells by
Table 1 MIC and fold increase results of heterologously expressed B. longum efflux genes in L. lactis pNZ9000. Recombinant plasmids expressed Em MIC in L. lactis NZ900 (mg/ml)
Em fold increase
Tet MIC (mg/ml)
Tet fold increase
pNZ8048 vector (no insert) pNZ1837-6 pNZ1837D6 pNZ1495
e 3.04 1.52 1.52
0.125 0.38 0.25 0.25
e 3.04 2.00 2.00
0.125 0.38 0.19 0.19
C. Moodley et al. / Anaerobe 32 (2015) 63e69
functional transporters when glucose is provided, resulting in a decrease in fluorescence. The rates of Hoechst transport were determined from the slope of each graph during the energised efflux phase when uninduced or induced with nisin. The rates of active efflux during the glucose re-energising phase showed a notable increase in efflux for cells harbouring the exogenous ABCtype transporters when nisin was present, with the plasmid bearing BLLJ_1837-6 showing a 16.4% increase in efflux activity. When the BLLJ_1836 gene was deleted from this construct, only a 7.1% increase was observed, and the plasmid containing the BLLJ_1495 gene only caused a marginal increase of 2.8%. 3.5. Regulator studies using EMSA EMSA was used to determine whether regulation of the two operons under investigation was controlled by the proteins encoded by the genes upstream of the transporters. The BLLJ_1838 and BLLJ_1496 putative MarR transcriptional regulator genes were cloned and their proteins expressed using the pET expression system. The purified proteins were used in in vitro binding experiments to possible DNA target sequences (Fig. 2). Based on the bioinformatic and transcriptional analysis of the operons, DNA regions consisting of approximately 250e300 bp upstream of each regulator gene, as well as the region between the regulators and the structural genes in each operon, were selected as possible regulator binding sites [50] (Fig. 1a, targets A e D). These DNA sequences were PCR amplified, gel excised and DIG-labelled for EMSA experiments. When the purified BLLJ_1496 protein from the 3-gene operon was added to target region A DNA, a notable, concentrationdependent shift in the migration of the bound DNA was observed on a native PAGE gel (Fig. 2). At high concentrations (2e3 mg) of added protein, a “super-shift” was observed (Fig. 2, lanes 7 and 8) indicating the presence of a potential secondary binding site in this region, which is commonly observed for MarR-type, regulator-DNA binding reactions [51]. The ability of BLLJ_1496 to bind to and regulate other cognate ABC-type transporter genes was also tested, since it has been shown that MarR-type regulators in Escherichia coli and S. typhimurium may recognise similar binding sites for other promoters in their regulon [52]. An EMSA, conducted using the target DNA fragments from regions B e D of the second efflux operon (Fig. 1a), showed no binding to the target region (data not shown). The BLLJ_1496 protein also did not bind to the proven functional promoter sequence of the a-galactosidase gene from
67
Bifidobacterium longum NCC2705 [53] (data not shown). This confirmed that the BLLJ_1496 protein did not recognise general promoter sequences and is likely to be the specific transcriptional regulator of the BLLJ_1496-5-4 operon. Sodium salicylate has been previously shown to disrupt the binding of MarR-type transcriptional regulators to their cognate DNA sequences, thus inducing gene expression [50e52]. However, in B. longum, sodium salicylate appeared to have very little effect on regulator binding, causing only a minor, reduced band-shift when very high concentrations (1 and 1.2 M) were added (data not shown). It has been shown that the exposure of B. longum NCIMB 702259 to sub-lethal concentrations of erythromycin resulted in an increase in the transcription levels of genes downstream of the MarR-type transcriptional regulator BLLJ_1496 (Fig. 1). The EMSA was therefore conducted in the presence of different concentrations of Em using the target A DNA fragment with 3 mg of the purified BLLJ_1496 protein. The results showed no reduction in binding efficiency in the presence of erythromycin as an effector molecule (data not shown). Bioinformatic and transcriptional analysis of the second operon, BLLJ_1837-6, suggested several candidate DNA regions for possible regulator binding (Fig. 1a, targets B e D). EMSA was carried out using the purified BLLJ_1838 putative transcriptional regulator protein with these target regions upstream of each ORF. No BLLJ_1838 protein binding was observed with any of the DNA targets using the in vitro conditions tested. The role of this protein in regulating the two gene ABC-type transporter operon, therefore, cannot be confirmed. The data presented here using the BLLJ_1496 protein showed a gene-specific, high affinity binding to the putative promoter region upstream of the structural genes of the operon, giving a strong indication of a regulatory function for this protein. However, erythromycin was not a direct effector molecule of the BLLJ_1496 protein, and this is supported by the finding that transcription of the 3-gene efflux operon was not significantly up-regulated by sublethal concentration of this antibiotic. MarR-type regulators are known to bind to un-conserved, direct-repeat sequences called ‘marboxes’, upstream of the resistance genes they regulate [50,52,54]. However, no similar sequence could be identified in the upstream regions of the B. longum efflux genes. The mar regulon in bacteria is known to lead to the up-regulation of several multidrug resistance genes through the use of MarA, the mar activator protein which is part of the marRAB locus. MarA acts as the transcriptional activator of the mar regulon by de-repressing binding of marR repressor proteins to their specific DNA sequences [50,52,54]. No
Fig. 2. Native PAGE gel of EMSA of BLLJ_1497-6 Target A DNA fragment with purified BLLJ_1496 protein. Lane 1, BLLJ_1497-96 DIG labelled DNA, lanes 2-8 labelled DNA with increasing amounts of purified BLLJ_1496 protein. Lanes 9 and 10, same as 6 and 7 except with 100 ng unlabelled DNA added as competitor (Comp).
68
C. Moodley et al. / Anaerobe 32 (2015) 63e69
marA gene homolog has been identified in B. longum NCIMB 702259 and further analysis is required to identify other transcriptional activators which may act as up-regulators of the ABCtype transporter genes identified in this study. 4. Conclusions Probiotic bacteria, such as Bifidobacterium species, are a major constituent of the gut microbiota of both humans and animals and their presence may confer health benefits on the host. However, their ability to survive in the gastrointestinal tract requires that they can withstand various antimicrobial substances such as bile and antibiotics. In this study, it was confirmed that, in addition to the Ctr transporter which is involved in the transport of bile compounds [43], B. longum harbours at least two clusters of genes encoding ABC-type efflux systems. Reverse transcriptase PCR experiments showed that the two gene clusters investigated here were transcribed as operons, one comprising three genes including a MarR transcriptional regulator, and the other comprising two genes, with an adjacent transcriptional regulator. qRT-PCR showed that the transcription of the genes of one of these operons, BLLJ_1837-6, was up-regulated when Em, a known substrate of ABC-type transporters, was present. The ABC-type transporter genes from this operon were cloned and heterologously expressed in Lactococcus lactis cells, where they conferred the host with increased resistance to erythromycin and tetracycline. The presence of these genes also allowed a 16% increase in the efflux rate of Hoechst 33342 from L. lactis cells containing the two transporter genes, BLLJ_1837-6. On the basis of this data, we propose that these genes BLLJ_1837 and BLLJ_1836 be named blmA (Bifidobacterium longum multidrug transporter) and blmB, respectively. RegulatorDNA binding studies showed that, for the three gene operon BLLJ_1496-5-4, the transcriptional regulator protein BLLJ_1494 was able to bind specifically and with high affinity to target DNA upstream of the operon, however, the expression of this operon in L. lactis did not show significant functionality in drug efflux, and the cloned genes were not stably maintained on a plasmid. This is the first experimental evidence pointing to a functional MarR-type DNA-binding protein in B. longum, which may regulate its cognate ABC transporter gene cluster. This study describes the occurrence of ABC-type transporter genes in a probiotic bacterium, which is widely used in the food industry as a dietary supplement. The function of these genes may have a protective function against cytotoxic compounds, such as bile, antibiotics and other antimicrobial substances found in the natural habitat of the bacterium or in the mammalian gut. Acknowledgements We would like to thank Dr. A. Margolles (Spain) for providing us with the L. lactis NICE gene expression system. We acknowledge The Woolworths Trust, the South African National Research Foundation and the University of Capetown for contributing to the funding of this work. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.anaerobe.2014.12.004. References [1] E.G. Zoetendal, A.D.L. Akkermans, W.M. De Vos, Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal samples reveals stable and host-specific communities of active bacteria, Appl. Environ. Microbiol. 64
(1998) 3854e3859. [2] H. Brüssow, Microbiota and healthy ageing: observational and nutritional intervention studies, Microb. Biotechnol. 6 (2013) 326e334. [3] M.J. Claesson, I.B. Jeffery, S. Conde, S.E. Power, E.M. O/'Connor, S. Cusack, et al., Gut microbiota composition correlates with diet and health in the elderly, Nature 488 (2012) 178e184. [4] W.H. Holzapfel, U. Schillinger, Introduction to pre- and probiotics, Food Res. Int. 35 (2002) 109e116. [5] C. Picard, J. Fioramonti, A. Francois, T. Robinson, F. Neant, C. Matuchansky, Bifidobacteria as probiotic agents e physiological effects and clinical benefits, Aliment. Pharmacol. Ther. 22 (2005) 495e512. [6] R.J.P. Dawson, K.P. Locher, Structure of a bacterial multidrug ABC transporter, Nature 443 (2006) 180e185. [7] C.F. Higgins, ABC transporters: physiology, structure and mechanism e an overview, Res. Microbiol. 152 (2001) 205e210. [8] A.L. Davidson, E. Dassa, C. Orelle, J. Chen, Structure, function, and evolution of bacterial ATP-binding cassette systems, Microbiol. Mol. Biol. Rev. 72 (2008) 317e364. [9] J.H. van Wonderen, R.M. McMahon, M.L. O'Mara, C.A. McDevitt, A.J. Thomson, I.D. Kerr, et al., The central cavity of ABCB1 undergoes alternating access during ATP hydrolysis, Fed. Eur. Biochem. Soc. 281 (2014) 2190e2201. [10] L. Masco, K. Van Hoorde, E. De Brandt, J. Swings, G. Huys, Antimicrobial susceptibility of Bifidobacterium strains from humans, animals and probiotic products, J. Antimicrob. Chemother. 58 (2006) 85e94. [11] M.R. D'Aimmo, M. Modesto, B. Biavati, Antibiotic resistance of lactic acid bacteria and Bifidobacterium spp. isolated from dairy and pharmaceutical products, Int. J. Food Microbiol. 115 (2007) 35e42. nchez, C.G. de los Reyes-Gavila n, A. Margolles, Antibiotic [12] M. Gueimonde, B. Sa resistance in probiotic bacteria, Front. Microbiol. 4 (2013). [13] K. Pleban, A. Macchiarulo, G. Costantino, R. Pellicciari, P. Chiba, G.F. Ecker, Homology model of the multidrug transporter LmrA from Lactococcus lactis, Bioorg. Med. Chem. Lett. 14 (2004) 5823e5826. [14] J. Lubelski, W.N. Konings, A.J.M. Driessen, Distribution and physiology of ABCtype transporters contributing to multidrug resistance in bacteria, Microbiol. Mol. Biol. Rev. 71 (2007) 463e476. [15] K. Sakamoto, A. Margolles, H.W. van Veen, W.N. Konings, Hop resistance in the beer spoilage bacterium Lactobacillus brevis is mediated by the ATP-binding cassette multidrug transporter HorA, J. Bacteriol. 183 (2001) 5371e5375. [16] S. Fukuda, H. Toh, K. Hase, K. Oshima, Y. Nakanishi, K. Yoshimura, et al., Bifidobacteria can protect from enteropathogenic infection through production of acetate, Nature 469 (2011) 543e547. [17] B.A. Degnan, G.T. Macfarlane, Transport and metabolism of glucose and arabinose in Bifidobacterium breve, Arch. Microbiol. 160 (1993) 144e151. [18] C.A. Magwira, B. Kullin, S. Lewandowski, A. Rodgers, S.J. Reid, V.R. Abratt, Diversity of faecal oxalate-degrading bacteria in black and white South African study groups: insights into understanding the rarity of urolithiasis in the black group, J. Appl. Microbiol. 113 (2012) 418e428. [19] O.P. Kuipers, P.G.G.A. de Ruyter, M. Kleerebezem, W.M. de Vos, Quorum sensing-controlled gene expression in lactic acid bacteria, J. Biotechnol. 64 (1998) 15e21. [20] P.G. de Ruyter, O.P. Kuipers, W.M. de Vos, Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin, Appl. Environ. Microbiol. 62 (1996) 3662e3667. [21] J. Sambrook, D. Russell, Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory Press, 2001. [22] Q. Ren, K.H. Kang, I.T. Paulsen, Transport DB: a relational database of cellular membrane transport systems, Nucleic Acids Res. 32 (D284-D8) (2004). [23] A. Marchler-Bauer, J.B. Anderson, F. Chitsaz, M.K. Derbyshire, C. DeWeeseScott, J.H. Fong, et al., CDD: specific functional annotation with the Conserved Domain Database, Nucleic Acids Res. 37 (D205-D10) (2009). [24] A. Marchler-Bauer, S. Lu, J.B. Anderson, F. Chitsaz, M.K. Derbyshire, C. DeWeese-Scott, et al., CDD: a Conserved Domain Database for the functional annotation of proteins, Nucleic Acids Res. 39 (2011). D225-D9. [25] J. Kyte, R.F. Doolittle, A simple method for displaying the hydropathic character of a protein, J. Mol. Biol. 157 (1982) 105e132. [26] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673e4680. [27] D. Ish-Horowicz, J.F. Burke, Rapid and efficient cosmid cloning, Nucleic Acids Res. 9 (1981) 2989e2998. [28] C. Magni, P. Marini, D. De Mendoza, Extraction of RNA from Gram-positive bacteria, Biotechniques 19 (1995) 880e884. [29] P.S. Langendijk, F. Schut, G.J. Jansen, G.C. Raangs, G.R. Kamphuis, M.H. Wilkinson, et al., Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its application in fecal samples, Appl. Environ. Microbiol. 61 (1995) 3069e3075. [30] P.M. Holland, R.D. Abramson, R. Watson, D.H. Gelfand, Detection of specific polymerase chain reaction product by utilizing the 50 e–30 exonuclease activity of Thermus aquaticus DNA polymerase, Proc. Natl. Acad. Sci. 88 (1991) 7276e7280. [31] S.A. Bustin, V. Benes, J.A. Garson, J. Hellemans, J. Huggett, M. Kubista, et al., The MIQE guidelines: minimum information for publication of quantitative realtime PCR experiments, Clin. Chem. 55 (2009) 611e622. [32] M.W. Pfaffl, A new mathematical model for relative quantification in real-time
C. Moodley et al. / Anaerobe 32 (2015) 63e69 RTePCR, Nucleic Acids Res. 29 (2001) e45. [33] S. Fleige, V. Walf, S. Huch, C. Prgomet, J. Sehm, M.W. Pfaffl, Comparison of relative mRNA quantification models and the impact of RNA integrity in quantitative real-time RT-PCR, Biotechnol. Lett. 28 (2006) 1601e1613. [34] M. Papagianni, N. Avramidis, G. Filioussis, High efficiency electrotransformation of Lactococcus lactis spp. lactis cells pretreated with lithium acetate and dithiothreitol, Biomed. Cent. Biotechnol. 7 (2007) 15. n, Mac[35] A. Margolles, J.A. Moreno, D. van Sinderen, C.G. de los Reyes-Gavila rolide resistance mediated by a Bifidobacterium breve membrane protein, Antimicrob. Agents Chemother. 49 (2005) 4379e4381. [36] Van Den Berg, H.B. Van Saparoea, J. Lubelski, R. Van Merkerk, P.S. Mazurkiewicz, A.J.M. Driessen, Proton motive force-dependent Hoechst 33342 transport by the ABC transporter LmrA of Lactococcus lactis, Biochemistry 44 (2005) 16931e16938. [37] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248e254. [38] S.C. Hyde, P. Emsley, M.J. Hartshorn, M.M. Mimmack, U. Gileadi, S.R. Pearce, et al., Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport, Nature 346 (1990) 362e365. [39] J.E. Walker, M. Saraste, M.J. Runswick, N.J. Gay, Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATPrequiring enzymes and a common nucleotide binding fold, Eur. Mol. Biol. Organ. J. 1 (1982) 945e951. [40] C.O. Pabo, R.T. Sauer, Transcription factors: structural families and principles of DNA recognition, Annu. Rev. Biochem. 61 (1992) 1053e1095. [41] M.N. Alekshun, S.B. Levy, T.R. Mealy, B.A. Seaton, J.F. Head, The crystal structure of MarR, a regulator of multiple antibiotic resistance, at 2.3 A resolution, Nat. Struct. Biol. 8 (2001) 710e714. rez, J.A. Moreno, D. van Sinderen, C.G. de los Reyes[42] A. Margolles, A.B. Flo Gavil an, Two membrane proteins from Bifidobacterium breve UCC2003 constitute an ABC-type multidrug transporter, Microbiology 152 (2006) 3497e3505. [43] C.E. Price, S.J. Reid, A.J.M. Driessen, V.R. Abratt, The Bifidobacterium longum NCIMB 702259T ctr gene codes for a novel cholate transporter, Appl. Environ.
69
Microbiol. 72 (2006) 923e926. [44] I. Nazarenko, B. Lowe, M. Darfler, P. Ikonomi, D. Schuster, A. Rashtchian, Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore, Nucleic Acids Res. 30 (2002) e37. [45] W. Li, B. Xi, W. Yang, M. Hawkins, U.K. Schubart, Complex DNA melting profiles of small PCR products revealed using SYBR Green I, Biotechniques 35 (2003) 702e704, 6. [46] A.A. Huet, J.L. Raygada, K. Mendiratta, S.M. Seo, G.W. Kaatz, Multidrug efflux pump overexpression in Staphylococcus aureus after single and multiple in vitro exposures to biocides and dyes, Microbiology 154 (2008) 3144e3153. [47] J.T. Lin, M.B. Connelly, C. Amolo, S. Otani, D.S. Yaver, Global transcriptional response of Bacillus subtilis to treatment with subinhibitory concentrations of antibiotics that inhibit protein synthesis, Antimicrob. Agents Chemother. 49 (2005) 1915e1926. [48] G.E. Louw, R.M. Warren, N.C. Gey van Pittius, R. Leon, A. Jimenez, R. Hernandez-Pando, et al., Rifampicin reduces susceptibility to ofloxacin in rifampicin-resistant Mycobacterium tuberculosis through efflux, Am. J. Respir. Crit. Care Med. 184 (2011) 269e276. [49] I. Mierau, M. Kleerebezem, 10 years of the nisin-controlled gene expression system (NICE) in Lactococcus lactis, Appl. Microbiol. Biotechnol. 68 (2005) 705e717. [50] R.G. Martin, J.L. Rosner, Binding of purified multiple antibiotic-resistance repressor protein (MarR) to mar operator sequences, Proc. Natl. Acad. Sci. 92 (1995) 5456e5460. [51] R.G. Martin, P.S. Nyantakyi, J.L. Rosner, Regulation of the multiple antibiotic resistance (mar) regulon by marORA sequences in Escherichia coli, J. Bacteriol. 177 (1995) 4176e4178. [52] M.N. Alekshun, S.B. Levy, Regulation of chromosomally mediated multiple antibiotic resistance: the Mar regulon, Antimicrobial. Agents Chemother. 41 (1997) 2067e2075. [53] A. Klijn, D. Moine, M. Delley, A. Mercenier, F. Arigoni, R.D. Pridmore, Construction of a reporter vector for the analysis of Bifidobacterium longum promoters, Appl. Environ. Microbiol. 72 (2006) 7401e7405. [54] M.N. Alekshun, S.B. Levy, The mar regulon: multiple resistance to antibiotics and other toxic chemicals, Trends Microbiol. 7 (1999) 410e413.