Effect of media composition on the susceptibility of Xanthomonas maltophilia to pmactam antibiotics

Transcription

1 Journal of Antimicrobial Chemotherapy (99) 2, 37-S42 Effect of media composition on the susceptibility of Xanthomonas maltophilia to pmactam antibiotics G. Bonfiglio* and D. M. Lfrennoret Department of Medical Microbiology, The London Hospital Medical College, Turner Street, London El 2AD, UK The susceptibility of Xanthomonas maltophtiia strains to 0-lactams was shown to depend on the concentrations at which individual media were prepared. MIC and disc susceptibility tests were performed on solidified Mueller-Hinton and Iso-Sensitest media prepared at 0,,, and 3 x the concentrations recommended by the manufacturers. Nine of ten X. maltophilia strains tested showed increasing sensitivity to meropenem, cefotaxime, cefoperazone, pipcracillin and latamoxef as the nutrient concentrations of the two media were increased. The opposite pattern was found with one strain (NCTC 0257) on Mueller-Hinton agar. This strain behaved inconsistently on Iso-Sensitest agar. Previous studies have shown that medium-dependent susceptibility in X. maltophilia is not related to /J-lactamase expression. The present study demonstrated that 22 and 25 kda outer membrane proteins were induced on media with higher nutrient concentrations. The possible relationship of these proteins to sensitivity is considered. Introduction Xanthomonas maltophilia is a free-living organism with a wide geographic distribution. In clinical specimens it is often a contaminant, but has been associated also with primary pneumonia and opportunistic infections. It is resistant to many /f-lactam antibiotics, including carbapenems (Rolston, Anaissie & Bodey, 97). This resistance is due to two factors: production of two different chromosomally-mediated /Mactamases, L-l and L-2, and low outer membrane permeability (Saino et al., 92; Mett et al., 9). It has been noted that the MICs of 0-lactam antibiotics for X. maltophilia isolates vary according to the test medium used. The MICs on Diagnostic Sensitivity Test agar (DSTA) are lower than those on Mueller-Hinton agar (MHA), while those on Iso-Sensitest agar (STA) are intermediate (Edwards et al., 99; King, Boothman & Phillips, 99). Our previous studies (Akova, Bonfiglio & Livermore, 99) have shown that this behaviour does not depend on varying efficiencies of /Mactamase induction. The aim of the present study was to determine whether alteration of the nutrient concentration in the individual media also changed the resistance of X. maltophilia strains to /Mactam antibiotics. Downloaded from at Pennsylvania State University on September, 206 'Present address: Istituto di Microbiologia Medica, UnivenhA di Milano, via Pascal n. 34, 2000 Milano, Italia, t Corresponding author /9/ J02.00/0 99 The British Society for Antimicrobial Chemotherapy

2 3 G. Boofiglio and D. M. Lfonnore Strains Materials and methods X. maltophilia strains NCTC 0257, 025, 0499 and seven clinical isolates were used. Identification of the latter organisms was by the API 20NE system (API System, La Balme Les Grottes, Montalieu-Vercieu, France). Antibiotics Imipenem was obtained from Merck Sharp and Dohme, meropenem from ICI Pharmaceuticals, cefotaxime from Roussel Laboratories, cefoperazone from Pfizer, piperacillin from Lederle Laboratories, and latamoxef from Eli Lilly. All other chemical reagents were obtained from Sigma. Media and susceptibility Mueller-Hinton broth (Difco Laboratories) and Iso-Sensitest broth (Oxoid Ltd) were prepared at 0-,, and 3x the concentrations recommended by their manufacturers. These media were then solidified by addition of -6% Bacteriological Agar No. (Oxoid). MICs were determined by incorporating doubling dilutions of antibiotics into the above media and applying inocula (prepared by growing strains overnight in Oxoid nutrient broth) of about 0 4 cfu to the agar surface. The MIC was defined as the lowest antibiotic concentration that prevented visible growth of the bacteria after incubation at 37 C for -24 h. Disc susceptibility tests also were performed on the various media. The discs, imipenem (0 /ig), cefotaxime (30 pg), cefoperazone (30 /xg), piperacillin (75 /*g) and latamoxef (30 fig), were purchased from Oxoid, while meropenem (0 fig) discs were prepared by ourselves. The inocula (about 0 5 cfu) were sufficient to give semiconfluent growth. The size of the inhibition zone was measured with callipers after incubation at 37 C for -24 h. Outer membrane preparation The method for preparation of outer membranes was that of Anwar et al. (93), but with some modifications. Lawns of bacteria were grown overnight on the various solid media. The cells from the surfaces of ten plates of each medium were washed into 30 mm Tris-HCl buffer, ph -0, harvested by centrifugation, and then resuspended in 20 ml of the same buffer containing 20% (w/v) sucrose, 0- mg deoxyribonuclease I (Sigma)/mL, and 0- mg ribonuclease A (Sigma)/mL. The mixtures were cooled to 4 C, and all subsequent operations were done at this temperature. The cells were disrupted by passing twice through a French Pressure Cell (AMINCO; SLM Instruments Inc, Urbana, Illinois, USA) at 2,000 lb/sq. in. Lysozyme (Sigma) was added (0-4 mg/l) to the mixture, which was then kept for 30 min at 4 C. The preparation was diluted with 20 ml 30 mm Tris-HCl buffer, ph -0, and unbroken cells were removed by centrifugation at 3000 g for 0 min. The supernatant was retained and centrifuged at 00,000 g for 30 min to pellet the membranes, which subsequently were resuspended in 2 ml 20% w/v sucrose in 30 mm Tris-HCl buffer, ph -0. This sample was layered on top of Downloaded from at Pennsylvania State University on September, 206

3 Media and sagllitity in X. mahopmba 39 a two-step sucrose gradient. The lower step contained 9 ml 70% w/v sucrose in 30 nut Tris-HG buffer, ph -0, while the upper step contained 9 ml 60% w/v sucrose in the same buffer. The gradient was centrifuged at 00,000 g overnight, thereby separating the outer and inner membrane vesicles, which formed two distinct bands. The lower band, containing the outer membrane vesicles, was recovered, diluted with 20 ml 30 mm Tris-HCl buffer, ph -0, and then centrifuged at 00,000 g for 60 min to harvest the membranes. The resulting pellet was resuspended in ml 30 mm Tris-HCl buffer, ph 0, and stored at 20 C. The protein content was estimated by the method of Lowry et a/. (95), with bovine serum albumin as the standard, and adjusted to a final concentration of 2 mg/ml by addition of 30 HIM Tris-HCl buffer, ph -0. Samples of the preparations (00 fj.l) were mixed with equal amounts of gel sample buffer (20% v/v glycerol; 4% w/v sodium dodecyl sulphate; 0% v/v /?-mercaptoethanol; 0-005% w/v bromophenol blue in 0-25 M Tris-HCl, ph 6-). Each sample was heated to 95 C for 0 min, and then separated by SDS-PAGE. Electrophoresis was performed at a constant current of 50 ma/gel with the method described by Lugtenberg et al. (975) with a running gel containing 4-5% w/v acrylamide and 0-255% w/v N,N',-metnylene- s-acrylamide. After electrophoresis, proteins were visualized by staining with a solution of 0-% Brilliant Blue R in mcthanol: acetic acid:distilled water (50:0:40), followed by background destaining in mcthanol: acetic acid: distilled water (5:0:5). Sensitivity tests Results Figure shows the inhibition zone diameters around the antibiotic discs for X. maltophilia strains NCTC 0257 and 0499 grown on 0-,, or 3 x MHA and 0-, or x ISTA. No growth was observed on 3 x ISTA. With all antibiotics except imipenem, the sensitivity of strain NCTC 0499 increased as the medium concentration was increased. For strain NCTC 0257, the opposite pattern was observed on MHA: i.e. sensitivity decreased as the medium concentration was increased. The behaviour of this strain on ISTA was inconsistent: sensitivity to meropenem and cefotaxime increased as the medium concentration was increased, whereas sensitivity to cefoperazone, piperacillin and latamoxef was unaffected or decreased marginally. Imipenem gave no zone on any medium for either strain. The MIC results (data not shown) agreed well with those of disc tests. Of the eight other strains tested, all behaved similarly to NCTC 0499, with their sensitivity increasing as the medium concentration was increased. None showed the same behaviour as NCTC The Table illustrates this behaviour for all the strains in response to cefotaxime only. Downloaded from at Pennsylvania State University on September, 206 Outer membrane protein profiles The electrophoretic profiles of outer membrane proteins from X. maltophilia strains NCTC 0257 and 0499 are shown in Figure 2. With both strains, the intensity of staining of the 22 and 25 kda proteins was more apparent when the organisms had been grown on media with the higher nutrient concentrations.

4 40 G. Bonfiglio and D. M. Lfonnore Medium concentration 0- Figure. Disc susceptibility test resulu for X. maltophilia strain NCTC 0499 on MHA (a) and on ISTA (b), and for strain NCTC 0257 on MHA (c) and ISTA (d). Data are for discs containing imipenem (0 /ig) O, meropenem (0 /ig) #, cefotaxime (30 jig) D. cefoperazone (30 /jg), piperacillin (75 /jg) A, latamoxef (30/ig)A. Strain GIL GIS Table. Susceptibility to cefotaxime of X. maltophilia strains grown on different media Inhibition zone diameters (mm) to cefotaxime (30 Mlj) discs MHA ISTA I I I MICs of cefotaxime (mg/l) 6 MHA ISTA Downloaded from at Pennsylvania State University on September, 206 No zone.

5 Media and sensitivity in X. mabopuua 4 Figure 2. Outer membrane protein profiles of (a) X. maltophilia NCTC 0499 and (b) X. maltophilia NCTC For both strains the growth media were as follow* track,0- x MHA; 2, x MHA; 3, x MHA; 4, 3 x MHA; 6, 0- x ISTA; 7, x ISTA;, x ISTA. Track 5 contains marker protein* as follows: phosphorylase b, 97-4 kda; bovine serum albumin, 66-2 kda; ovalbumin, 45-0 kda; carbonic anhydrase, 3-0 kda; soybean trypsin inhibitor, 2-5 kda; lyiozyme, 4-4 kda. The 22 and 25 kda proteins induced in the more concentrated media are arrowed. Discussion It has been noted previously that X. maltophilia strains are commonly more resistant to 0-lactams on MHA than on ISTA or DSTA (Edwards et al., 99; King et al., 99). The present findings show that sensitivity varies in line with the nutrient concentration of both MHA and ISTA, rather than being contingent on the medium per se. In nine of ten strains examined, the sensitivity increased with the nutrient concentration of the medium; the one exceptional strain, NCTC 0257, showed the opposite behaviour on MHA and behaved inconsistently on ISTA. It might be thought that these observations could relate to induction of /Mactamase. X. maltophilia possesses two different inducible /Mactamases, L-l and L-2 (Saino et al., 92; Mett et al., 9). If induction of these enzymes varied between different media, this might account for the variation in susceptibility. However, the effects of the media on sensitivity are independent of /Mactamase expression since they occur with /Mactamase basal and high level constitutive mutants as well as with /Mactamase inducible strains (Akova et al., 99). In addition, the level of /Mactamase induced by cefotaxime is independent of the growth medium. Thus /Mactamase was not responsible for the medium-dependent susceptibility of the strains examined. Downloaded from at Pennsylvania State University on September, 206

6 42 G. BonfigHo and D. M. Lfrermore Another mechanism whereby the medium might influence sensitivity is by affecting permeability, perhaps by modulating expression of porin(s). When the outer membrane protein profiles of X. maltophilia strains grown on the different media were examined, the results showed that the expression of 22 and 25 kda proteins increased when the organisms were grown on media containing the higher nutrient concentrations. It would be tempting to speculate that these proteins played a role in the uptake of /Mactams. However, two points argue against this hypothesis: (i) known porins vary from kda in mass; (ii) induction of these proteins occurred in strain NCTC 0257 grown on MHA, despite the fact that this strain showed increasing resistance when the nutrient concentration of this medium was increased. Thus, it is not yet possible to give a full explanation of the behaviour observed. Nevertheless, it is apparent that considerable medium dependency does arise when testing the sensitivity of X. maltophilia to /Mactams, and that this depends on the nutrient concentrations in media rather than on the specific medium itself. We are unaware of any other species which behaves in this way with /Mactams. It has been noted previously that temperature affects the susceptibility of X. maltophilia to aminoglycosides and polymyxin B, with strains appearing more resistant at 30 C than at 37 C. Temperature did not, however, affect susceptibility to piperacillin or ceftazidime (Wheat, Winstanley & Spencer, 95). Taken together, these observations emphasize the severe difficulty in predicting accurately the susceptibility of X. maltophilia from clinical laboratory tests. We suggest that the clinician should interpret the results of such tests with the utmost caution. References Akova, M., Bonfiglio, G. & Livermore, D. M. (99). Susceptibility to 0-lactam antibiotics of Xmthomonas maltophilia mutants with high- and low-level constitutive expression of L-l and L-2 0-lactamases. Journal of Medical Microbiology, in press. Anwar, H., Brown, M. R. W., Cozens, R. M. & Lambert, P. A. (93). Isolation and characterization of the outer and cytoplasmic membranes of Pseudomonas cepacia. Journal of General Microbiology 29, Edwards, J. R., Turner, P. J., Wannop, C, WithneU, E. S., Grindey, A. J. & Nairn, K. (99). In vitro antibacterial activity of SM-733, a carbapenem antibiotic with stability to dehydropeptidase I. Antimicrobial Agents and Chemotherapy 33, King, A., Boothman, C. & Phillips, I. (99). Comparative in-vitro activity of mcropencm on clinical isolates from the United Kingdom. Journal of Antimicrobial Chemotherapy 24, Suppl. A, Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (95). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 93, Lugtenbcrg, B., Meijers, J., Peters, R., Van Der Hoech, P. & Van Alphen, L. (975). Electrophoretic resolution of the "major outer membrane protein" of Escherichia coli K2 into four bands. FEBS Letters 5, Mett, H., Rosta, S., Schacher, B. & Frci, R. (9). Outer membrane permeability and /Mactamase content in Pseudomonas maltophilia clinical isolates and laboratory mutants. Reviews of Infectious Diseases 0, Rolston, K. V. I., Anaissie, E. A. & Bodey, G. P. (97). In-vitro susceptibility of Pseudomonas species to fifteen antimicrobial agents. Journal of Antimicrobial Chemotherapy 9, Saino, Y., Kobayashi, F., Inoue, M. & Mitsuhashi, S. (92). Purification and properties of inducible penicillin /Mactamase isolated from Pseudomonas maltophilia. Antimicrobial Agents and Chemotherapy 22, Wheat, P. F., Winstanley, T. G. & Spencer, R. C. (95). Effect of temperature on antimicrobial susceptibilities of Pseudomonas maltophilia. Journal of Clinical Pathology 3, {Received 24 May 99; accepted 9 July 99) Downloaded from at Pennsylvania State University on September, 206