JAC Increase in glutamine-non-amidated muropeptides in the peptidoglycan of vancomycin-resistant Staphylococcus aureus strain Mu50

Transcription

1 Journal of Antimicrobial Chemotherapy (1998) 42, JAC Increase in glutamine-non-amidated muropeptides in the peptidoglycan of vancomycin-resistant Staphylococcus aureus strain Mu50 H. Hanaki a, H. Labischinski b, Y. Inaba a, N. Kondo a, H. Murakami and K. Hiramatsu a a Department of Bacteriology, Juntendo University, Hongo, Bunkyo-ku, Tokyo, Japan; b Bayer AG, Germany The peptidoglycan compositions of vancomycin-resistant Staphylococcus aureus (VRSA) strain Mu50 (MIC 8 mg/l) and hetero-vrsa strain Mu3 (MIC 3 mg/l) were compared in order to understand the mechanism of vancomycin resistance. As compared with Mu3, the cell wall of Mu50 had increased amounts of glutamine-non-amidated muropeptides and decreased crosslinking of peptidoglycan with a greatly decreased dimer/monomer ratio of muropeptides. In agreement with this observation, the peptidoglycan of Mu50 bound 1.4 times more vancomycin than that of Mu3. The increase in non-amidated muropeptides and the reduced crosslinking of the cell-wall peptidoglycan may contribute to the vancomycin resistance by increasing the consumption of vancomycin by the pre-existing cell wall of Mu50 and reducing the amount of vancomycin reaching the cytoplasmic membrane where the vital targets of the antibiotic are situated. Introduction Vancomycin-resistant Staphylococcus aureus (VRSA) strains are defined as having vancomycin MICs of 8 mg/l. 1 A methicillin-resistant VRSA strain, Mu50, was isolated from a 4 month old male with a methicillinresistant S. aureus (MRSA) wound infection in Before that, a hetero-vrsa strain (i.e. one in which only some cells at least 1 in 10 6 are vancomycin-resistant), Mu3, was isolated from an MRSA pneumonia patient who responded poorly to vancomycin therapy. 1 Although Mu3 contains a small population of cells with vancomycin resistance, the vancomycin MIC of Mu3 is 3 mg/l. Thus, Mu3 is judged as susceptible according to the criteria of The British Society for Antimicrobial Chemotherapy 3 and those of the National Committee for Clinical Laboratory Standards (NCCLS) in the USA. 4 Despite its apparent vancomycin susceptibility, Mu3 contrasts sharply with other staphylococcal strains in the frequency with which vancomycin resistance can be generated: resistant mutants are easily obtained from Mu3 by one-step vancomycin selection (at a frequency of 1 in 10 6 ), whereas they are extremely difficult to obtain from vancomycin-susceptible coagulase-negative staphylococci, 5 S. aureus 6 and from a teicoplanin-resistant S. aureus clinical strain. 7 Therefore, it is thought that hetero-vrsa Mu3 must have already undergone a certain change or changes that are prerequisites to achieving the level of vancomycin resistance found in Mu50. 1 Activation of cell wall synthesis, which is found in both Mu50 and Mu3 but not in vancomycinsusceptible strains, may be one such prerequisite. 8 The final step of development of vancomycin resistance in Mu50 may be associated with cell wall thickening, as this strain has a cell wall twice as thick as that of Mu3. 8 This study was designed to investigate further the phenotype associated with the final step of resistance development by comparing the structure of purified peptidoglycan of Mu3 and Mu50. Materials and methods Bacterial strains and growth conditions Mu50, a VRSA strain (MIC 8 mg/l) and Mu3, a hetero- VRSA strain (MIC 3 mg/l) are methicillin-resistant clinical strains isolated from patients who responded poorly to vancomycin therapy. 1,2 Vancomycin-susceptible MRSA strains H1 (MIC 1 mg/l; isolated from a MRSA pneumonia patient who responded favourably to vancomycin therapy) and BB270 (MIC 1 mg/l) have been described previously. 1,9 All strains were grown in brain heart infusion (BHI) broth (Difco, Detroit, MI, USA) at *Corresponding author. Tel: ; Fax: ; hiram@med.juntendo.ac.jp 1998 The British Society for Antimicrobial Chemotherapy 315

2 H. Hanaki et al. 37 C with aeration. For each experiment, an overnight culture was diluted 100-fold into prewarmed BHI broth to ensure exponential growth conditions. Growth was monitored by measuring the optical density at 578 nm (OD 578 ) with a spectrophotometer (Pharmacia LKB Biotechnology, Inc., Uppsala, Sweden). Analysis of peptidoglycan Preparation of peptidoglycan. Peptidoglycan was prepared and analysed essentially as described previously. 10 Briefly, bacteria were grown in BHI broth and harvested in mid-exponential growth phase. Cells were disrupted with glass beads and washed with 0.5% sodium dodecyl sulphate (SDS). Protein A was removed by incubation with trypsin, and then teichoic acids were removed by hydrofluoric acid treatment. Purified peptidoglycan was washed with water and lyophilized. The preparation was used for both composition analysis and vancomycin binding assays. Digestion and reduction of peptidoglycan. The lyophilized peptidoglycan was digested with mutanolysin (Sigma, St Louis, MO, USA). Samples were adjusted to ph 4.0 with phosphoric acid, and incubated at 100 C for 5 min. Borohydride was added to reduce the digested peptidoglycan (muropeptides). The muropeptides were then adjusted to ph 12 with NaOH to hydrolyse remaining O-acetyl groups present attached to the muramic acid residues. The muropeptide solution was adjusted to ph 2.5 with HCl and filtered to remove insoluble material. Fractionation of muropeptides by HPLC. Muropeptides were separated by reversed-phase high performance liquid chromatography (HPLC). The column (3 m Hypersil ODS, 4.0 by 250 nm CompAx-peek; Knauer, Berlin, Germany) was eluted with a linear gradient from 5% methanol in 50 mm sodium phosphate (ph 2.5) containing % sodium azide to 30% methanol in 50 mm sodium phosphate (ph 2.8) within 210 min. Muropeptide peaks were detected by absorption at 206 nm. of vancomycin remaining in the supernatant and washing solution was quantified by HPLC as described previously. 11 The number of vancomycin molecules bound to 1 mg of the peptidoglycan was calculated from the formula: (T 1 T 2 )/ A, where T 1 is the amount of vancomycin added ( g in this study), T 2 is the amount of vancomycin recovered in the supernatant and SDS-washing solution (in grams), is the molecular weight of vancomycin and A is Avogadro s number. Antagonism of vancomycin action by synthetic peptides Peptides, both amidated (H-D-Glu[L-Lys-D-Ala-D-Ala- OH]-NH 2 ) and non-amidated (H-D-Glu[L-Lys-D-Ala-D- Ala-OH]-OH), were custom synthesized by Bachem (Bubendorf, Switzerland). Paper discs (5 mm diameter) impregnated with 100 g of either of the peptides and varying amounts of vancomycin (3 10 g) were placed on BHI agar containing about 10 6 cfu/ml of Bacillus subtilis strain DSM402. After 24 h incubation at 37 C, the diameter of the inhibition zone was determined from duplicate experiments. Results Increased vancomycin binding to Mu3 and Mu50 peptidoglycan Figure 1 shows the number of vancomycin molecules bound to the peptidoglycan (1 mg dry weight) prepared from Mu50, Mu3 and control strains. The peptidoglycan of Mu50 bound molecules of vancomycin, i.e. about 2.4 times the number bound to S. aureus type strain Amino acid composition analysis. Some muropeptide peaks were fractionated, lyophilized and then hydrolysed in 6 N HCl at 166 C for 1 h. Samples were analysed using an amino acid analyser (Biotronic LC5000). Binding assay of vancomycin to peptidoglycan A total of 250 g of lyophilized peptidoglycan was suspended in 1 ml of 10 mm potassium phosphate buffer, ph 7.2. A total of 300 g of vancomycin was added to the suspension and the suspension was incubated at 37 C for 15 min. Then, the peptidoglycan was pelleted by centrifugation at 12,000g for 3 min. The pellet was washed twice with 1 ml of water containing 4% SDS. The total amount Figure 1. The amount of vancomycin bound to purified peptidoglycan of Mu50, Mu3 and control strains of S. aureus. 316

3 Peptidoglycan of VRSA FDA209P ( molecules/mg peptidoglycan). Mu3 bound molecules, i.e. 1.8 and 1.1 times the number bound by FDA209P and by the vancomycinsusceptible MRSA strain H1 ( /mg), respectively, but 0.7 times the number bound by Mu50. Comparison of peptidoglycan composition in Mu50 and Mu3 Profiles of the mutanolysin-digested peptidoglycan of Mu50 and Mu3 are compared with that of BB270 in Figure 2a. Besides the normal monomeric and dimeric muropeptide peaks (labelled M4 and D3, respectively), additional peaks were found in the Mu50 profile (Figure 2b); they were designated M9, D5 and D9 by comparing their retention times with those in a previous study. 12 These peaks were also found in lesser amounts in Mu3 (Figure 2c). These additional peaks of Mu50 showed a strong shift to the lower retention times when a less acidic (ph 3.5) buffer was used for elution (Figure 2d). The reported muropeptide peaks M9, D5 and D9 of femc mutant strain BB589, 13 corresponding to the monomer of glutamine non-amidated muropeptide (M9), the heterodimer of amidated and non-amidated muropeptides (D5), and the homodimer of non-amidated muropeptides (D9), respectively, were eluted simultaneously by HPLC with the three peaks of Mu50 in both acidic and less acidic conditions (data not shown). The Table shows the results of amino acid composition analysis of the muropeptide peaks M4, M9, D3, D5 and D9 of Mu50 peptidoglycan. The data were compatible with the interpretation that peak M9 corresponded to monomers of glutamine non-amidated muropeptide, D5 to a heterodimer of amidated and non-amidated muro- Table. Amino acid composition of the muropeptide of Mu50 separated in Figure 2 Amount of amino acid b Peak a Glx Lys Ala Gly M M D D D a Muropeptide peak numbers as given in Figure 2. b Molar ratios normalized to Glx 1.0. (d) Figure 2. The peptidoglycan composition profile of Mu50 and Mu3 in comparison with that of BB270. (a) BB270; (b) and (d) Mu50; (c) Mu3. In panels (a) (c) muropeptides were eluted with a linear gradient of buffer with ph range of In panel (d) elution was performed with a buffer ph of 3.5. Peaks M4 and D3 correspond to the monomer and dimer of normal (amidated) muropeptide subunit, respectively. 317

4 H. Hanaki et al. peptides, and D9 to a homodimer of non-amidated muropeptides (D9). Figure 2 also shows that, compared with BB270, Mu50 peptidoglycan is much less cross-linked and contains a much lower ratio of dimeric/monomeric muropeptide components (Figure 2b). Mu3 also had a slightly decreased dimer/monomer ratio and decreased cross-linking (Figure 2c), but not to the extent seen in Mu50. Antagonism of vancomycin action by synthetic peptides containing amidated vs non-amidated glutamic acid Figure 3 shows the diameter of the inhibition zone plotted against the amount of vancomycin when either amidated or the non-amidated peptide is present. No inhibition zone was observed with discs containing either peptide alone (data not shown). Vancomycin alone generated clear inhibition zones (Figure 3). Both peptides decreased the activity of vancomycin, presumably by binding to it but, in the presence of non-amidated peptide, inhibition zones were smaller than those in the presence of the amidated peptide, indicating that the non-amidated peptide has a greater binding affinity to vancomycin than the amidated peptide. Figure 3. Antagonistic effect of synthetic peptide on the antimicrobial activity of vancomycin. Diminution of the diameter of the inhibition zone generated by vancomycin against a Bacillus subtilis strain in the presence of synthetic peptides. Diffusion susceptibility tests were performed with discs impregnated with vancomycin alone ( ), vancomycin plus the synthetic peptide H-D-Glu[L-Lys-D-Ala-D-Ala-OH]-NH 2 ( ), and vancomycin plus the synthetic peptide H-D-Glu[L-Lys-D- Ala-D-Ala-OH]-OH ( ). The peptide with a non-amidated glutamic acid was more potent in the inhibition of vancomycin activity. Discussion The mechanism of vancomycin resistance in enterococci has been clearly explained by the replacement of D- alanyl-d-alanine dipeptide termini of the cell-wall muropeptide subunits by D-alanyl-D-lactate depsipeptide. 14 It is not known whether the same mechanism applies to glycopeptide-resistant staphylococci. 15 VRSA show a lower level of resistance than vancomycin-resistant enterococci (VRE), and lack vana, vanb and vanc1 3 genes as well as D-lactate-terminated murein monomer precursor in the strains, indicating that these two genera have different mechanisms. 1,8,16 In Mu3 and Mu50, vancomycin resistance seems to be associated with activated cell-wall synthesis, as evidenced by increased incorporation of N-acetylglucosamine, an increased pool size of U D P -N-acetyl-muramylpentapeptides, and increased production of penicillin-binding proteins (PBPs) 2 and 2. 8 However, there must be some difference between Mu3 and Mu50 that will account for the difference in the level of vancomycin resistance between the two strains. 1 Since vancomycin acts on bacterial cell wall synthesis, cannot penetrate the bacterial cell membrane and is not destroyed by any known bacterial enzyme, it has been assumed that an alteration in cell wall structure was involved. So far, the most impressive phenotypic difference observed between Mu3 and Mu50 has been the pronounced cell wall thickness in untreated Mu50 as observed in transmission electron microscopy. 8 In this study, it was demonstrated that the peptidoglycan of Mu50 had a 1.4-fold higher vancomycin binding capacity than Mu3. There was also a lower degree of cross-linking in the Mu50 peptidoglycan than in the Mu3 peptidoglycan. A lower degree of crosslinking of peptidoglycan signifies that there are increased numbers of D-alanyl-D-alanine termini of muropeptide subunits in the peptidoglycan to which vancomycin binds. Because of this and the thickened cell wall as compared with that of Mu3, a single Mu50 cell is estimated to bind 2.8 (1.4 2) times more vancomycin molecules than a single Mu3 cell. Since vancomycin-binding target sites in the pre-existing peptidoglycan are not the vital targets of vancomycin, this increased number of false binding sites may contribute to the raised resistance to vancomycin by capturing more vancomycin molecules within the cell wall and reducing the number of the molecules that reach the cytoplasmic membrane where the real vital targets of vancomycin are located. 17 An increase in the glutamine-non-amidated muropeptide components in the peptidoglycan may also contribute directly to an increase in vancomycin resistance in Mu50 by efficient consumption of vancomycin molecules within the cell wall. The experiment on the mixture of vancomycin with two model peptides in a paper disc suggested that the non-amidated murein monomer may have a higher affinity of binding to vancomycin than amidated murein monomer. These non-amidated muro- 318

5 Peptidoglycan of VRSA peptides have previously been noticed as minor components of S. aureus peptidoglycan. 18 They constitute a pronounced proportion of muropeptides within the peptidoglycan of femc mutant strains. 19 In these strains, expression of the glna gene, encoding glutamine synthetase, is drastically depressed by the polar effect of the TN551 transposon inserted upstream of the gene. 13 The decrease in the glutamine synthetase activity in the mutant cell causes a decrease in the intracellular concentration of glutamine, which serves as a donor of NH 4 to the isoglutamate residue of murein monomer precursor. 20 This causes increased incorporation of non-amidated muropeptide subunits in the cell wall peptidoglycan. Since nonamidated muropeptide precursor is known to be an inefficient substrate for transpeptidase, 21 cross-linking of the peptidoglycan decreases when the proportion of nonamidated muropeptides rises. 19 The mechanism by which glutamine non-amidated muropeptides are produced in Mu50 remains to be elucidated. Reduction of cell wall cross-linking has also been observed with an in-vitro-trained vancomycin-resistant S. aureus strain. 22,23 These authors reported that the strain VM produced a large amount of cell wall material with reduced cross-linking, which was considered to sequestrate vancomycin molecules in the culture medium. However, these alterations, including grossly thickened cell walls, became evident only when strain VM was grown in the presence of vancomycin, while the alterations described in our studies in Mu50 occurred in the absence of the drug. Another difference between VM and Mu50 was noticed in the peptidoglycan composition; reduction of cross-linking is striking in the oligomer fractions of muropeptides in VM, and there is no evident decrease in the dimer/monomer ratio of muropeptides which was characteristically observed in the peptidoglycan of Mu Therefore, although a drastic decrease in the cross-linking of peptidoglycan and resultant increase in the vancomycin-binding false targets (D-alanyl-D-alanine residues) in the pre-existing cell wall are common to both strains, the mechanism(s) causing the alteration of peptidoglycan cross-linking is apparently different. In conclusion, Mu50 peptidoglycan showed reduced cross-linking and had an increased proportion of glutamine non-amidated muropeptides. Increased vancomycinbinding capacity of the peptidoglycan, and possible increased binding affinity of the non-amidated muropeptide components to vancomycin, may at least partially account for the raised level of vancomycin resistance in VRSA strain Mu50. Acknowledgements We greatly acknowledge the technical assistance of K. Servan and A. Polaczyk during peptidoglycan preparation and HPLC analysis. This work was supported by the Japanese Ministry of Education, Science, Sports and Culture (Grant-in-Aid for Scientific Research No ). References 1. Hiramatsu, K., Aritaka, N., Hanaki, H., Kawasaki, S., Hosoda, Y., Hori, S. et al. (1997). Dissemination in Japanese Hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin. Lancet 350, Hiramatsu, K., Hanaki, H., Ino, T., Yabuta, K., Oguri, T. & Tenover, F. C. (1997). Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. Journal of Antimicrobial Chemotherapy 40, Working Party of the British Society for Antimicrobial Chemotherapy. (1991). A guide to sensitivity testing. Journal of Antimicrobial Chemotherapy 27, Suppl. D, National Committee for Clinical Laboratory Standards. (1997). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically Fourth Edition: Approved Standard M7 A4. NCCLS, Villanova, PA. 5. Watanakunakorn, C. (1988). In-vitro induction of resistance in coagulase-negative staphylococci to vancomycin and teicoplanin. Journal of Antimicrobial Chemotherapy 22, Watanakunakorn, C. (1990). In-vitro selection of resistance of Staphylococcus aureus to teicoplanin and vancomycin. Journal of Antimicrobial Chemotherapy 25, Shlaes, D. M. & Shlaes, J. H. (1995). Teicoplanin selects for Staphylococcus aureus that is resistant to vancomycin. Clinical Infectious Diseases 20, Hanaki, H., Kuwahara-Arai, K., Boyle-Vavra, S., Daum, R. S., Labischinski, H. & Hiramatsu, K. (1998). Activated cell-wall synthesis is associated with vancomycin resistance in methicillinresistant Staphylococcus aureus clinical strains Mu3 and Mu50. Journal of Antimicrobial Chemotherapy 42, Ryffel, C., Strassle, A., Kayser, F. H. & Berger-Bachi, B. (1994). Mechanisms of heteroresistance in methicillin-resistant Staphylococcus aureus. Antimicrobial Agents and Chemotherapy38, Kopp, U., Roos, M., Wecke, J. & Labischinski, H. (1996). Staphylococcal peptidoglycan interpeptide bridge biosynthesis: a novel antistaphylococcal target? Microbal Drug Resistance 2, Hosotsubo, H. (1989). Rapid and specific method for the determination of vancomycin in plasma by high-performance liquid chromatography on an aminopropyl column. Journal of Chromatography 487, Stranden, A. M., Ehlert, K., Labischinski, H. & Berger-Bachi, B. (1997). Cell wall monoglycine cross-bridges and methicillin hypersusceptibility in a femab null mutant of methicillin-resistant Staphylococcus aureus. Journal of Bacteriology 179, Gustafson, J., Strassle, A., Hachler, H., Kayser, F. H. & Berger-Bachi, B. (1994). The femc locus of Staphylococcus aureus required for methicillin resistance includes the glutamine synthetase operon. Journal of Bacteriology 176, Arthur, M. & Courvalin, P. (1993). Genetics and mechanisms of glycopeptide resistance in enterococci. Antimicrobial Agents and Chemotherapy 37,

6 H. Hanaki et al. 15. Dutka-Malen, S., Leclercq, R., Coutant, V., Duval, J. & Courvalin, P. (1990). Phenotypic and genotypic heterogeneity of glycopeptide resistance determinants in Gram-positive bacteria. Antimicrobial Agents and Chemotherapy 34, Tenover, F. C., Lancaster, M. V., Hill, B. C., Steward, C. D., Stocker, S. A., Hancock, G. A. et al. (1998). Characterization of staphylococci with reduced susceptibility to vancomycin and other glycopeptides. Journal of Clinical Microbiology 36, Reynolds, P. E. (1989). Structure, biochemistry and mechanism of action of glycopeptide antibiotics. European Journal of Clinical Microbiology and Infectious Diseases 8, Roos, M., Pittenauer, E., Schmid, E., Beyer, M., Reinike, B., Allmaier, G. et al. (1998). Improved high-performance liquid chromatographic separation of peptidoglycan isolated from various Staphylococcus aureus strains for mass spectrometric characterization. Journal of Chromatography B, Biomedical Sciences and Applications 705, Stranden, A. M., Roos, M. & Berger-Bachi, B. (1996). Glutamine synthetase and heteroresistance in methicillin-resistant Staphylococcus aureus. Microbial Drug Resistance 2, Siewert, G. & Strominger, J. L. (1968). Biosynthesis of the peptidoglycan of bacterial cell walls. XI. Formation of the isoglutamine amide group in the cell walls of Staphylococcus aureus. Journal of Biological Chemistry 243, Nakel, M., Ghuysen, J.-M. & Kandler, O. (1971). Wall peptidoglycan in Aerococcus viridans strains 201 Evans and ATCC and in Gaffkya homari strain ATCC Biochemistry 10, Sieradzki, K. & Tomasz, A. (1996). A highly vancomycinresistant laboratory mutant of Staphylococcus aureus. FEMS Microbiological Letters 142, Sieradzki, K. & Tomasz, A. (1997). Inhibition of cell wall turnover and autolysis by vancomycin in a highly vancomycinresistant mutant of Staphylococcus aureus. Journal of Bacteriology 179, Received 17 February 1998; returned 18 March 1998; revised 7 May 1998; accepted 17 June