Journal of Antimicrobial Chemotherapy (1992) 29, 115-120
Proposed mechanism for metronidazole resistance in Helicobacter pylori
'Department of Clinical Microbiology, Central Hospital, Vaxjo;bDepartment of Medical Microbiology, University of Lund, Sweden Metronidazole MIC values were determined for ten isolates of Helicobacter pylori. UndeT microaerobic conditions four were inhibited by < 0-25 mg/L, two strains required 20 mg/L and four strains had MIC > 32 mg/L. The strains were tested for their susceptibility to metronidazole after varying lengths of anaerobic exposure prior to or during microaerobic incubation. The susceptibility was determined by E-test and traditional agar dilution technique. The level of susceptibility of primarily susceptible strains was not significantly affected by a 2 to 24 h period of anaerobic incubation. However, the effect on resistant strains was dramatic. These strains were increasingly susceptible when a 2 to 12 h period of anaerobicity was provided anytime during the first 48 h and fully susceptible (MIC < 0.1 mg/L) if 24 h of anaerobicity was provided. When tested again under microaerobic conditions the strains exhibited their original MIC values. The composition of the medium did not influence the results. We propose that "metronidazole resistance" in H. pylori is due to a decreased ability of these strains to achieve a sufficiently low redox potential under microaerobic conditions for the necessary reduction of metronidazole, and that these strains during short periods of anaerobicity manage to reduce and "store" sufficient amounts of metronidazole to appear fully susceptible after subsequent microaerobic incubation.
Introduction
Helicobacter pylori is now well established in the pathogencsis of type B gastritis (Rathbone, Wyatt & Heatly, 1986). Numerous clinical studies have been conducted with different antibiotics aimed at eradicating the micro-organisms. There is now general agreement that at least two antibiotics, preferably in combination with a bismuth preparation (Goodwin et al., 1988), should be included in the treatment regimen. Metronidazole combined with amoxycillin (Borsch, Mai & Opferkuch, 1988) or tetracycline (Rauws et al., 1988) have been successfully used. However, there are now several reports on the rapid in-vitro (Glupczynski et al., 1988; Goodwin et al., 1988) and in-vivo (Goodwin et al., 1985, 1988; Hirshl et al., 1988) development of resistance to metronidazole. To our knowledge the mechanism of resistance has not hitherto been explained. Metronidazole is primarily active against certain parasites and anaerobic bacteria, and is also active against H. pylori (Warren & Marshall, 1983). Strains of H. pylori grow best in a microaerobic (microaerophilic) atmosphere of 5% O2 with 5-10% CO2 (Goodwin et al., 1985) and will not grow in the complete absence of oxygen unless an appropriate respiratory electron acceptor, such as fumarate is added (Stouthamer, De Vries & Niekus, 1976; Morris, 1983). Susceptibility testing of 115 0305-7453/92/020115+06 $02.00/0
© 1992 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014
Gunflla Cederbranf, Gannar Kahlmeter* and Asa Ljungh'
116
G. Cedertmurt et aL
H. pylori has therefore been performed under microaerobic conditions and this study was performed to investigate the mechanism of in-vitro metronidazole resistance in H. pylori. Methods and materiab Strains
Inoculum Vials stored at — 80°C were inoculated on GAB—camp plates and incubated microaerobically (CampyPac, BBL) for 48 h. Colonies were suspended in 5 ml phosphate buffer saline (PBS) pH 7-2 to a density of 0-5 McFarland (approximately 10* cfu/mL) and further diluted in GB-broth (see below). Media MIC determinations were made on GAB-camp plates (Soltesz, Schalen & Mardh, 1987) consisting of 25 g/L GC Agar Base (BBL) and 2-9 g/L of Bacto Agar (Difco), supplemented with 05 g/L of cystein hydrochloride (Merck), 8-5% (v/v) of defibrinated horse blood, haematinized for 20 min at +80°C, 10% of inactivated horse serum and 0-35% of Isovitalex (BBL). E-tests were obtained from AB Biodisk, Sweden. The E-test is a recently introduced concept for performing MIC determinations. It is based on diffusion of a pre-formed antibiotic gradient from a plastic strip (Bolmstrom et al., 1988) and has recently been evaluated for Enterobacteriaceae and several Gram-positive bacteria (Brown & Brown, 1991) and for H. pylori (Cederbrant, Kahlmeter & Ljungh, 1991). The E-tests were performed on GAB-camp plates, GAB plates (Mardh, Martensson & Soltesz, 1978) and blood agar plates. GAB-camp plates were prepared as described above. Blood agar plates were prepared from blood agar base (Oxoid) with 4% human blood. GB-broth with 5% horse serum, used for suspending H. pylori, was prepared from 16 g/L Proteose Peptone N. 3 (Difco), 3 g/L Liver Digest neutralized (Oxoid), 0-5 g/L Cystein H2O HCL (Merck), 0-8 g/L Na alginate (Kebo, Sweden), 5 g/L NaCl, 1-6 g/L Na2HPO4 2H2O (Kebo, Sweden). This was enriched with 1% (v/v) of Isovitalex, 0-9% of haemin (0-2%), 0-6% of Yestolate (Difco) (10%) and 5% of inactivated horse serum. Antibiotics Metronidazole was obtained as standard powder and as E-test through the courtesy of AB Biodisk, Sweden. Metronidazole was dissolved in PBS pH 7-2 to a standard stock solution of 1600 mg/L, and further diluted in PBS.
Downloaded from http://jac.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014
Ten distinct clinical isolates of H. pylori were studied. They were identified by accepted criteria (Gram's stain, oxidase, catalasc and urease test). Four of these were reported as resistant to metronidazole (MIC > 32 mg/L), two exhibited intermediate MIC values (2 mg/L) and four strains exhibited MIC values < 0-25 mg/L. The strains were stored in glass vials at — 80°C in Tryptic soy broth (TSB) containing 15% (v/v) glycerol.
Mechanism for metronid azote
in H. pylori
117
MIC and E-test determination
Results Figure 1 shows the MIC- and IC-values of 10 H. pylori strains following the three different modes of incubation. As measured by MIC- and IC-values the susceptibility of the less susceptible strains increased dramatically if the total incubation period included a period of anaerobic incubation. Resistant strains appeared fully susceptible. It did not seem to matter greatly if the microaerobic incubation was preceded by a 24 h-period of anaerobic incubation or if the anaerobic period was between two microaerobic incubation periods. In both instances there was no visible growth after the first 48 h. Strain A B C
1
•
'
i
Proposed mechanism for metronidazole resistance in Helicobacter pylori
'Department of Clinical Microbiology, Central Hospital, Vaxjo;bDepartment of Medical Microbiology, University of Lund, Sweden Metronidazole MIC values were determined for ten isolates of Helicobacter pylori. UndeT microaerobic conditions four were inhibited by < 0-25 mg/L, two strains required 20 mg/L and four strains had MIC > 32 mg/L. The strains were tested for their susceptibility to metronidazole after varying lengths of anaerobic exposure prior to or during microaerobic incubation. The susceptibility was determined by E-test and traditional agar dilution technique. The level of susceptibility of primarily susceptible strains was not significantly affected by a 2 to 24 h period of anaerobic incubation. However, the effect on resistant strains was dramatic. These strains were increasingly susceptible when a 2 to 12 h period of anaerobicity was provided anytime during the first 48 h and fully susceptible (MIC < 0.1 mg/L) if 24 h of anaerobicity was provided. When tested again under microaerobic conditions the strains exhibited their original MIC values. The composition of the medium did not influence the results. We propose that "metronidazole resistance" in H. pylori is due to a decreased ability of these strains to achieve a sufficiently low redox potential under microaerobic conditions for the necessary reduction of metronidazole, and that these strains during short periods of anaerobicity manage to reduce and "store" sufficient amounts of metronidazole to appear fully susceptible after subsequent microaerobic incubation.
Introduction
Helicobacter pylori is now well established in the pathogencsis of type B gastritis (Rathbone, Wyatt & Heatly, 1986). Numerous clinical studies have been conducted with different antibiotics aimed at eradicating the micro-organisms. There is now general agreement that at least two antibiotics, preferably in combination with a bismuth preparation (Goodwin et al., 1988), should be included in the treatment regimen. Metronidazole combined with amoxycillin (Borsch, Mai & Opferkuch, 1988) or tetracycline (Rauws et al., 1988) have been successfully used. However, there are now several reports on the rapid in-vitro (Glupczynski et al., 1988; Goodwin et al., 1988) and in-vivo (Goodwin et al., 1985, 1988; Hirshl et al., 1988) development of resistance to metronidazole. To our knowledge the mechanism of resistance has not hitherto been explained. Metronidazole is primarily active against certain parasites and anaerobic bacteria, and is also active against H. pylori (Warren & Marshall, 1983). Strains of H. pylori grow best in a microaerobic (microaerophilic) atmosphere of 5% O2 with 5-10% CO2 (Goodwin et al., 1985) and will not grow in the complete absence of oxygen unless an appropriate respiratory electron acceptor, such as fumarate is added (Stouthamer, De Vries & Niekus, 1976; Morris, 1983). Susceptibility testing of 115 0305-7453/92/020115+06 $02.00/0
© 1992 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014
Gunflla Cederbranf, Gannar Kahlmeter* and Asa Ljungh'
116
G. Cedertmurt et aL
H. pylori has therefore been performed under microaerobic conditions and this study was performed to investigate the mechanism of in-vitro metronidazole resistance in H. pylori. Methods and materiab Strains
Inoculum Vials stored at — 80°C were inoculated on GAB—camp plates and incubated microaerobically (CampyPac, BBL) for 48 h. Colonies were suspended in 5 ml phosphate buffer saline (PBS) pH 7-2 to a density of 0-5 McFarland (approximately 10* cfu/mL) and further diluted in GB-broth (see below). Media MIC determinations were made on GAB-camp plates (Soltesz, Schalen & Mardh, 1987) consisting of 25 g/L GC Agar Base (BBL) and 2-9 g/L of Bacto Agar (Difco), supplemented with 05 g/L of cystein hydrochloride (Merck), 8-5% (v/v) of defibrinated horse blood, haematinized for 20 min at +80°C, 10% of inactivated horse serum and 0-35% of Isovitalex (BBL). E-tests were obtained from AB Biodisk, Sweden. The E-test is a recently introduced concept for performing MIC determinations. It is based on diffusion of a pre-formed antibiotic gradient from a plastic strip (Bolmstrom et al., 1988) and has recently been evaluated for Enterobacteriaceae and several Gram-positive bacteria (Brown & Brown, 1991) and for H. pylori (Cederbrant, Kahlmeter & Ljungh, 1991). The E-tests were performed on GAB-camp plates, GAB plates (Mardh, Martensson & Soltesz, 1978) and blood agar plates. GAB-camp plates were prepared as described above. Blood agar plates were prepared from blood agar base (Oxoid) with 4% human blood. GB-broth with 5% horse serum, used for suspending H. pylori, was prepared from 16 g/L Proteose Peptone N. 3 (Difco), 3 g/L Liver Digest neutralized (Oxoid), 0-5 g/L Cystein H2O HCL (Merck), 0-8 g/L Na alginate (Kebo, Sweden), 5 g/L NaCl, 1-6 g/L Na2HPO4 2H2O (Kebo, Sweden). This was enriched with 1% (v/v) of Isovitalex, 0-9% of haemin (0-2%), 0-6% of Yestolate (Difco) (10%) and 5% of inactivated horse serum. Antibiotics Metronidazole was obtained as standard powder and as E-test through the courtesy of AB Biodisk, Sweden. Metronidazole was dissolved in PBS pH 7-2 to a standard stock solution of 1600 mg/L, and further diluted in PBS.
Downloaded from http://jac.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014
Ten distinct clinical isolates of H. pylori were studied. They were identified by accepted criteria (Gram's stain, oxidase, catalasc and urease test). Four of these were reported as resistant to metronidazole (MIC > 32 mg/L), two exhibited intermediate MIC values (2 mg/L) and four strains exhibited MIC values < 0-25 mg/L. The strains were stored in glass vials at — 80°C in Tryptic soy broth (TSB) containing 15% (v/v) glycerol.
Mechanism for metronid azote
in H. pylori
117
MIC and E-test determination
Results Figure 1 shows the MIC- and IC-values of 10 H. pylori strains following the three different modes of incubation. As measured by MIC- and IC-values the susceptibility of the less susceptible strains increased dramatically if the total incubation period included a period of anaerobic incubation. Resistant strains appeared fully susceptible. It did not seem to matter greatly if the microaerobic incubation was preceded by a 24 h-period of anaerobic incubation or if the anaerobic period was between two microaerobic incubation periods. In both instances there was no visible growth after the first 48 h. Strain A B C
1
•
'
i
