Journal of Global Antimicrobial Resistance 1 (2013) 109–113
Contents lists available at SciVerse ScienceDirect
Journal of Global Antimicrobial Resistance journal homepage: www.elsevier.com/locate/jgar
Short communication
Vancomycin resistance in Streptomyces coelicolor is phosphate-dependent but is not mediated by the PhoP regulator Fernando Santos-Beneit a, Juan F. Martı´n b,* a b
Instituto de Biotecnologı´a de Leo´n, INBIOTEC, Parque Cientı´fico de Leo´n, Av. Real 1, 24006 Leo´n, Spain A´rea de Microbiologı´a, Fac. CC. Biolo´gicas y Ambientales, Universidad de Leo´n, Campus de Vegazana, s/n, 24071 Leo´n, Spain
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 December 2012 Received in revised form 11 February 2013 Accepted 5 March 2013
Vancomycin is an essential antibiotic to treat infections caused by multidrug-resistant bacteria. Several bacteria show resistance to vancomycin, including the model actinomycete Streptomyces coelicolor. In this study, vancomycin disk diffusion tests were performed to determine vancomycin resistance in S. coelicolor M145 under rich (TSA medium) or defined (MMCGT medium) growth conditions. A vancomycin-susceptible phenotype was observed when the TSA rich medium was used, whereas a resistant phenotype was obtained when the low-phosphate MMCGT medium was used. To identify which component was responsible for the vancomycin-resistant phenotype, all the components of the MMCGT medium were added individually to the TSA medium, and vice versa. Addition of phosphate to the MMCGT medium (the phosphate concentration is much higher in TSA than in MMCGT) produced a vancomycin-susceptible phenotype in MMCGT. Phosphate regulation of vancomycin resistance is not PhoP-dependent since the same minimum inhibitory concentrations were obtained in S. coelicolor parental and DphoP mutant strains. This phosphate regulation was not observed in the vancomycinproducer Amycolatopsis orientalis NRRL 2452, which was always resistant both in TSA and MMCGT (with or without phosphate addition) media. On the other hand, other Streptomyces spp. were susceptible to vancomycin in all conditions tested, including Streptomyces toyocaensis, the producer of a glycopeptide antibiotic different from vancomycin. In conclusion, the phosphate concentration clearly affects the resistance of S. coelicolor to vancomycin. ß 2013 International Society for Chemotherapy of Infection and Cancer. Published by Elsevier Ltd. All rights reserved.
Keywords: Vancomycin Phosphate Streptomyces coelicolor PhoP
1. Introduction The increasing number of antibiotic-resistant bacteria and the decrease in new antibiotic discovery is a handicap for medicine. It is therefore essential to understand how bacterial resistance mechanisms develop and how they are influenced by growth conditions. The glycopeptide antibiotic vancomycin is currently reserved in the clinic for the last-resort treatment of enterococcal infections and meticillin-resistant Staphylococcus aureus (MRSA), a major killer in hospital-acquired infections. Vancomycin acts by binding to the terminal D-alanyl-D-alanine (D-Ala-D-Ala) moieties of peptidoglycan precursors and hampering the transpeptidation and transglycosylation steps of the cell wall assembly process [1]. Resistance to vancomycin first appeared clinically in enterococcal species and later spread to other bacteria, including MRSA [2,3]. Hong et al. [4] reported in Streptomyces coelicolor the first example of vancomycin resistance in a non-pathogenic and
* Corresponding author. Tel.: +34 987 291 505; fax: +34 987 210 388. E-mail address:
[email protected] (J.F. Martı´n).
non-glycopeptide-producing bacterium. Most of the Streptomyces spp. are non-pathogenic, although some of them are pathogens for humans and plants [5]. Some of them are also producers of glycopeptide antibiotics, such as Streptomyces toyocaensis [6]. The importance of the Streptomyces genus resides in its ability to produce a wide variety of secondary metabolites, including twothirds of all commercially important antibiotics [5]. The vancomycin resistance mechanism in S. coelicolor, as well as in Enterococcus faecalis and S. aureus, involves conversion of the dipeptide D-Ala-D-Ala to D-Ala-D-lactate coupled with a peptidasemediated elimination of D-Ala-D-Ala dipeptides from the cell wall [2,4]. The number of genes present in the resistance cluster can vary, but the ‘core’ cluster consists of three genes (vanHAX). In S. coelicolor, the vancomycin resistance cluster is composed of seven genes (vanSRJKHAX) divided into four transcription units: vanRS; vanJ; vanK; and vanHAX [4]. These transcription units are regulated by the VanR–VanS two-component system, which consists of a receptor histidine kinase (VanS) that binds vancomycin and activates its partner response regulator, VanR [4,7,8]. Understanding the mechanism of resistance to vancomycin in S. coelicolor may provide information on the origin of vancomycin resistance that
2213-7165/$ – see front matter ß 2013 International Society for Chemotherapy of Infection and Cancer. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jgar.2013.03.003
F. Santos-Beneit, J.F. Martı´n / Journal of Global Antimicrobial Resistance 1 (2013) 109–113
110
might be associated with the natural resistance in the vancomycinproducing actinomycete. S. coelicolor has a high level of resistance to vancomycin. Hong et al. determined the minimum inhibitory concentration (MIC) of vancomycin for S. coelicolor M600 and several mutants on defined MMCGT medium. The MIC for the parental strain was >200 mg/mL, whilst it was <10 mg/mL in vanK, vanR, vanH, vanA, vanX and vanHAX null mutants [4,7]. Using the same medium, Novotna et al. [9] performed vancomycin disk diffusion tests with these S. coelicolor parental and mutant strains. In contrast to the parental strain, clear inhibition halos were observed in vanRS and vanHAX mutants when 30 mg vancomycin disks were applied. Interestingly, almost all studies on vancomycin resistance in S. coelicolor have been performed in MMCGT medium [4,7,9,10]. MMCGT consists of minimal medium [11] supplemented with 0.6% (w/v) Difco casamino acids, 0.75% (v/v) Tiger milk [11] and 0.5% (w/v) glucose. This medium has a relatively low phosphate concentration (2.87 mM) especially when compared with rich media such as TSA (phosphate concentration > 15 mM), which is normally used for robust growth of Streptomyces [11]. Phosphate is an important modulator of the antibiotic biosynthesis and probably also of the antibiotic resistance mechanisms [12]. Interestingly, in early studies in TSA medium, we observed that S. coelicolor was susceptible to vancomycin, in apparent contradiction to the results of Hong et al. [4]. To clarify this finding, in this study disk diffusion tests were performed using commercial vancomycin disks and two different growth media, i.e. MMCGT and TSA. Moreover, the vancomycin resistance of several streptomycetes as well as the vancomycin-producer Amycolatopsis orientalis was also tested. 2. Materials and methods 2.1. Strains The species used in this work were: S. coelicolor M145 [13]; Streptomyces lividans 1326 [11]; Streptomyces avidinii NRRL 3077; Streptomyces clavuligerus NRRL 3585; Streptomyces hygroscopicus NRRL 3602; Streptomyces griseus NRRL 3851; Streptomyces tsukubaensis NRRL 18488; Streptomyces avermitilis NRRL 8165; Streptomyces natalensis NRRL 2651; and S. toyocaensis NRRL 15009. The vancomycin-producer A. orientalis NRRL 2452 was used as a positive control. The S. coelicolor M145 DphoP mutant was obtained as before [13]. 2.2. Vancomycin bioassays For the vancomycin disk diffusion assay, commercial antibiotic disks containing 30 mg of the antibiotic were purchased from BD BBLTM (Heidelberg, Germany) and were placed onto freshly [(Fig._1)TD$IG]inoculated TSA medium (3% of tryptone soya broth (TSB), pH 7.2
[11]) or onto MMCGT medium [10]. MMCGT is made of minimal medium [11] plus 0.6% (w/v) Difco casamino acids, 0.75% (v/v) Tiger milk (11) and 0.5% (w/v) glucose, adjusted to pH 7.2. For the assay, Streptomyces spores were homogenized into 5 mL of warm TSA or MMCGT (containing 0.7% of agar) and poured as a thin layer over a solid layer (20 mL) of either TSA or MMCGT media (containing 1.4% of agar). After solidification of the added layer, vancomycin disks were placed on the plates. In all cases, the same amount of spores (108) was used. The results were evaluated after 2 days of incubation at 30 8C. 3. Results 3.1. Streptomyces coelicolor is susceptible or resistant to vancomycin depending on the culture medium Previously it has been shown that S. coelicolor is resistant to vancomycin (disks of 30 mg) when MMCGT medium is used [4,9]. However, possible nutrient regulation of vancomycin resistance has never been tested in S. coelicolor. As a first approach, disk diffusion assays were performed on TSA and MMCGT media using commercial vancomycin disks, containing 30 mg of antibiotic, as described in Section 2. TSA is a rich medium that is able to support the abundant growth of many Streptomyces spp. Interestingly, the S. coelicolor TSA cultures showed a vancomycin-susceptible phenotype, which contrasts with the resistant phenotype of the S. coelicolor MMCGT cultures (Fig. 1). This result demonstrates that vancomycin resistance in S. coelicolor is culture medium-dependent. 3.2. Vancomycin resistance is phosphate-dependent in Streptomyces coelicolor As a second approach, we decided to identify which component of the medium was responsible for the different sensitivity of S. coelicolor to vancomycin in TSA and MMCGT media. To achieve this aim, first each nutrient of the MMCGT defined medium was added to the TSA rich medium. None of the additions (0.5% glucose, 0.05% K2HPO4, 0.02% MgSO47H2O, 0.001% FeSO47H2O, 0.05% L-asparagine, 0.6% Difco casamino acids and 0.75% Tiger milk) or the combination of some of them, such as glucose and phosphate or glucose and Tiger milk, changed the susceptible phenotype of S. coelicolor on TSA. On the other hand, addition of 3% TSB to the MMCGT changed the S. coelicolor resistant phenotype to a clear susceptible phenotype (Fig. 1). The composition of the TSB has 1.7% pancreatic digest of casein, 0.3% enzymatic digest of soya bean, 0.25% glucose, 0.5% NaCl and 0.25% K2HPO4. As the concentration of glucose in the TSA medium is lower than in MMCGT (0.25% vs. 0.5%), addition of glucose to the MMCGT medium was excluded, but the addition of NaCl (0.5%) and K2HPO4 (0.20%) was tested. Addition of NaCl did not change the resistant phenotype of S. coelicolor in
Fig. 1. Vancomycin sensitivity tests of Streptomyces coelicolor M145 in TSA, MMCGT and MMCGT plus 3% of tryptone soya broth (TSB).
[(Fig._2)TD$IG]
F. Santos-Beneit, J.F. Martı´n / Journal of Global Antimicrobial Resistance 1 (2013) 109–113
111
Fig. 2. Vancomycin sensitivity tests of Streptomyces coelicolor M145 and DphoP mutant in MMCGT medium with addition of different concentrations of phosphate.
MMCGT but, interestingly, vancomycin disks formed clear inhibition halos of growth after addition of phosphate to MMCGT. Following this result, different concentrations of K2HPO4 (0%, 0.25%, 0.5% and 1%) were added to the MMCGT medium and the size of the inhibition halos was observed. As shown in Fig. 2, the size of the halos increased in parallel with the increment in the phosphate concentration. Phosphate control of metabolism in Streptomyces, as in other bacteria, is mediated by the two-component system PhoR–PhoP. Under phosphate limitation, the phosphorylated active form of PhoP binds to specific sequences and triggers expression of PhoPdependent genes [12]. To check whether the PhoR–PhoP system could be implicated in this regulation, the same phosphategradient study was carried out with a S. coelicolor M145 DphoP mutant previously obtained in our laboratory [13]. The inhibition halos also increased as the phosphate concentration was raised (Fig. 2). The size of the inhibition halos appeared a bit larger in the DphoP mutant; therefore, agar dilution analyses were performed to determine the MIC of the antibiotic in both S. coelicolor strains under phosphate-limited and phosphate-replete conditions. The MIC for both strains on MMCGT was 80 mg/mL. On MMCGT plus 1% phosphate, the MIC in both strains was decreased to 20 mg/mL. In summary, in concordance with the disk diffusion assays, vancomycin resistance of S. coelicolor decreases when phosphate is added to the MMCGT medium. In addition, MICs are the same for S. coelicolor M145 and DphoP strains either in high or low phosphate concentrations. Therefore, the effect of phosphate on vancomycin resistance is not mediated by the PhoP regulator. 3.3. Vancomycin resistance of glycopeptide-producing and nonproducing actinomycetes To test whether this medium-dependent regulation of vancomycin resistance is also exerted in other species of the Streptomyces genus, spores of nine different streptomycetes, including the producer of the glycopeptide antibiotic A47934 S. toyocaensis [6] and S. lividans, a very close relative of S. coelicolor,
were tested for vancomycin resistance both in TSA and MMCGT media. All of the strains were susceptible to vancomycin when they were assayed on TSA medium (data not shown). When MMCGT was used, all of the species except S. lividans showed a vancomycin-susceptible phenotype, although the diameters of the inhibition zones varied among them (Fig. 3). Similar inhibition halos were observed on the plates when 1% K2HPO4 was added to the MMCGT, with the exception of S. lividans that, in the same way as S. coelicolor, became susceptible to vancomycin in phosphatesupplemented medium (Fig. 3). On the other hand, the vancomycin-producer A. orientalis, included in this study as a positive control, was resistant in all media and at all phosphate concentrations used (Fig. 3). In summary, the ten Streptomyces spp. tested in this study were susceptible to vancomycin when they were assayed either on TSA medium or MMCGT supplemented with phosphate. On the other hand, only S. coelicolor and S. lividans were resistant to vancomycin in MMCGT without phosphate addition. This result is not strange since all of the S. coelicolor vancomycin resistance genes (van) are also present in S. lividans [4]. 4. Discussion Understanding how bacteria can become resistant or remain sensitive to a certain antibiotic is of great importance not only for knowledge of bacterial biology but also for medicine. This knowledge is more important for human health when the antibiotic is used for last-resort treatments in chemotherapy, as is the case for vancomycin. The origin of many antibiotic resistance determinants may be related to the antibiotic resistance genes that occur naturally in the antibiotic-producing actinomycetes. During the past decades, several bacteria have been reported to have acquired resistance to vancomycin, including the model actinomycete S. coelicolor [2–4]. In this study, we have shown that S. coelicolor shows vancomycin-susceptible or -resistant phenotypes depending on
[(Fig._3)TD$IG]
112
F. Santos-Beneit, J.F. Martı´n / Journal of Global Antimicrobial Resistance 1 (2013) 109–113
Fig. 3. Vancomycin sensitivity tests in MMCGT (without and with 1% K2HPO4 addition) of Amycolatopsis orientalis NRRL 2452 and different species of Streptomyces: S. lividans 1326; S. avidinii NRRL 3077; S. clavuligerus NRRL 3585; S. hygroscopicus NRRL 3602; S. griseus NRRL 3851; S. tsukubaensis NRRL 18488; S. avermitilis NRRL 8165; S. natalensis NRRL 2651; and S. toyocaensis NRRL 15009. Note that for A. orientalis only the condition of MMCGT with 1% K2HPO4 is shown.
the phosphate concentration of the culture medium. This phosphate regulation is not PhoP-dependent since the same parental phenotype was obtained with a S. coelicolor DphoP mutant. The phosphate-dependent regulation of vancomycin resistance is conserved in S. lividans, a close relative of S. coelicolor, but not in the vancomycin-producer A. orientalis. A. orientalis was resistant both in phosphate-replete and -limited conditions, therefore the phosphate-dependent vancomycin-resistant phenotype of Streptomyces is not extended to this Amycolatopsis sp. On the other hand, the following eight Streptomyces spp. were susceptible to vancomycin in all media conditions tested in the study: S. avidinii; S. clavuligerus; S. hygroscopicus; S. griseus; S. tsukubaensis; S. avermitilis; S. natalensis; and S. toyocaensis. S. toyocaensis produces the glycopeptide antibiotic A47934 [6]. The gene cluster responsible for the biosynthesis of A47934 possesses S. coelicolor van homologues genes. However, as previously reported by other authors, and also shown in this study, vancomycin resistance is not induced in this bacterium [4,7]. It is not clear how phosphate reduces vancomycin resistance in S. coelicolor. In previous transcriptomic studies, we have not observed regulation of the vancomycin resistance genes in phosphate-replete or -limited conditions, although those studies were performed in the absence of vancomycin [12]. In recent transcriptomic analyses with TSA cultures and vancomycin induction, we have observed van transcriptional upregulation in a S. coelicolor mutant highly resistant to vancomycin but not in the wild-type strain (unpublished data). Therefore, it seems that transcriptional activation of S. coelicolor van genes depends on the presence of vancomycin accompanied with a low phosphate concentration in the medium. This is not strange since in work with Streptomyces rimosus, coordinated regulation of oxytetracycline production and resistance to the antibiotic was evidenced under conditions of low phosphate
concentrations. This regulatory mechanism may ensure that resistance to the antibiotic increases in proportion to its production [14]. S. coelicolor does not produce vancomycin but it could regulate the vancomycin resistance genes with a similar mechanism. Nevertheless, whether phosphate controls directly or indirectly the expression of the vancomycin resistance genes or regulates other mechanisms that play a role in vancomycin resistance will be a subject for future work. All together, these findings highlight the importance of growth conditions in bacterial antibiotic resistance. Funding This research was supported by grant BIO2010-16094 from the Ministry of Science and Innovation (Spain). Competing interests None declared. Ethical approval Not required. Acknowledgments The authors thank Stephano Donadio and Paolo Monciardini for providing the A. orientalis NRRL 2452 strain. References [1] Barna JCJ, Williams DH. The structure and mode of action of glycopeptide antibiotics of the vancomycin group. Annual Review of Microbiology 1984;38:339–57.
F. Santos-Beneit, J.F. Martı´n / Journal of Global Antimicrobial Resistance 1 (2013) 109–113 [2] Arthur M, Reynolds P, Courvalin P. Glycopeptide resistance in enterococci. Trends in Microbiology 1996;4:401–7. [3] Chang S, Sievert DM, Hageman JC, Boulton ML, Tenover FC, Downes FP, et al. Infection with vancomycin-resistant Staphylococcus aureus containing the vanA resistance gene. New England Journal of Medicine 2003;348: 1342–7. [4] Hong HJ, Hutchings MI, Neu JM, Wright GD, Paget MSB, Buttner MJ. Characterization of an inducible vancomycin resistance system in Streptomyces coelicolor reveals a novel gene (vanK) required for drug resistance. Molecular Microbiology 2004;52:1107–21. [5] Hopwood DA. Streptomyces in nature and medicine. New York, NY: Oxford University Press; 2007. [6] Boeck LD, Mertz FP. A47934, a novel glycopeptide-aglycone antibiotic produced by a strain of Streptomyces toyocaensis. Taxonomy and fermentation studies. Journal of Antibiotics (Tokyo) 1986;39:1533–40. [7] Hutchings MI, Hong HJ, Buttner MJ. The vancomycin resistance VanRS twocomponent signal transduction system of Streptomyces coelicolor. Molecular Microbiology 2006;59:923–35. [8] Koteva K, Hong HJ, Wang XD, Nazi I, Hughes D, Naldrett MJ, et al. A vancomycin photoprobe identifies the histidine kinase VanSsc as a vancomycin receptor. Nature Chemistry & Biology 2010;6:327–9.
113
[9] Novotna G, Hill C, Vincent K, Liu C, Hong HJ. A novel membrane protein, VanJ, conferring resistance to teicoplanin. Antimicrobial Agents and Chemotherapy 2012;56:1784–96. [10] Hong HJ, Paget MS, Buttner MJ. A signal transduction system in Streptomyces coelicolor that activates the expression of a putative cell wall glycan operon in response to vancomycin and other cell wall-specific antibiotics. Molecular Microbiology 2002;44:1199–211. [11] Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA. Practical Streptomyces genetics. Norwich, UK: The John Innes Foundation; 2000. [12] Martı´n JF, Santos-Beneit F, Rodrı´guez-Garcı´a A, Sola-Landa A, Smith MC, Ellingsen TE, et al. Transcriptomic studies of phosphate control of primary and secondary metabolism in Streptomyces coelicolor. Applied Microbiology and Biotechnology 2012;95:61–75. [13] Santos-Beneit F, Rodrı´guez-Garcı´a A, Sola-Landa A, Martı´n JF. Cross-talk between two global regulators in Streptomyces: PhoP and AfsR interact in the control of afsS, pstS and phoRP transcription. Molecular Microbiology 2009;72:53–68. [14] McDowall KJ, Thamchaipenet A, Hunter IS. Phosphate control of oxytetracycline production by Streptomyces rimosus is at the level of transcription from promoters overlapped by tandem repeats similar to those of the DNA-binding sites of the OmpR family. Journal of Bacteriology 1999;181:3025–32.