ln-vitro activity of newer quinolones against aerobic bacteria

Journal of Antimicrobial Chemotherapy (96) 7, Suppl. B, 9-39 ln-vitro activity of newer quinolones against aerobic bacteria R. Anckenthaler, M. Michea-Hamzehponr and J. C. Pectere Laboratoire Central de Bactiriologie, Hopital Cantonal Universitaire, -Geneve, Switzerland Nalidixic and five newer -quinolones,,, norfloxarin, and pcfloxacin were tested against S76 recent clinical aerobic bacterial isolates. The -quinolones were regularly active ( 9 < mg/) against the following bacteria: Staphylococcus aweus, S. epidermidis, S. saprophyticus, different Enterobacteriaceae, Haemophilus influenzae, Campylobacter jejuni, Pseudomonas aeruginosa, Agrobacter spp., Aeromonas spp., Plesiomonas spp., Neisseria meningitidis. Other bacteria were usually intermediately susceptible or resistant: different streptococci, Luteria monocytogenes, Nocardia asteroides, P. maltophilia, Achromobacter xylosoxydans and Akaligenes denitrificans. Cipr was the most potent compound, followed by and, and being less active. All the -quinolones were much more active than. The MBC/ ratios of the -quinolones were between and with a majority of strains, and between and 3 with Streptococcus agedactiae, Str. faeccdis and L. monocytogenes. A two- to eight-fold increase of was observed by increasing the inoculum,-fold with most of the strains tested. Susceptible bacterial population of Klebsiella pneumoniae, Enterobacter cloacae, Serratia marcescens and P. aeruginosa contained more clones resistant to ( to s at four times the ) than to -quinolones ( 5 to ' at four times the ). Supplementing the media with MgSO produced smaller inhibition zone diameters with a disc diffusion method than those obtained with non-supplemented agar, with all quinolone or strains. Less regular effect, or no effect was obtained after supplementation with ZnSO or Ca(NO 3 ). Introduction In contrast to and early derivatives such as cinoxacin, rosoxacin or flumequine, the activity of newer -quinolone compounds does not limit them to the oral therapy of urinary tract infections caused by Enterobacteriaceae. Recently developed -quinolone are characterized by a wider spectrum including Pseudomonas spp., Legionella spp. (Greenwood & Laverick, 93), Gram-positive bacteria and obligate intracellular organisms such as Mycobacteriwn spp. (Gay, DeYoung & Roberts, 9) or Chlamydia spp. (Heesen & Muytiens, 9; von Roosbroeck, Privinciael & Caekenberghe, 9) and Mycoplasma spp. (Ridgway et al., 9). In contrast, the activity against anaerobic bacteria is limited. The extremely low s of -quinolones against aerobic bacteria and their pharmacokinetic properties suggests their use in numerous clinical situations. In the present study we compare the in-vitro activity of,,, and against routine clinical isolates of aerobic Gram-positive and Gram-negative organisms. 35-753/6/7B9+./ 9 96 The British Society for Antimicrobial Chemotherapy

3 R. Anckenthaler et al Material and methods Antimicrobial agents. Standard powder of the following drugs were obtained from their manufacturers: Nalidixic acid (Winthrop, Switzerland), (Bayer AG, Germany), (Roger Bellon, France), (Merck Sharp and Dohme, Switzerland), (Roussel-Hoechst, France) and (Rhone Poulenc, France). solutions were prepared in water or in broth and used immediately. Bacterial strains. Five hundred and seventy-six clinical strains collected from patients hospitalized in the University Hospital of Geneva or, occasionally, in other Swiss hospitals, and seven ATCC control organisms were used. They were kept frozen in skim milk at 7 C. Before study, organisms were thawed, streaked onto sheep blood agar and incubated overnight at 35 C. Antimicrobial susceptibility tests. Antimicrobial activity was measured by the microdilution method in Mueller-Hinton broth (NCCLS, 93). Disposable material was used exclusively in order to avoid cross-contamination with quinolones. The final inoculum of 3-6 cfu/ml was prepared from a trypticase soy broth inoculated h before and controlled by counting the bacteria in a calibrated volume. The inoculum for Haemophilus influenzae was prepared and tested in Brain Heart Infusion broth supplemented with 5% nicotinamide diphosphate and % haemin. The minimal inhibitory concentration () was read after h of incubation at 35 C in air or at C in % CO atmosphere for Campylobacter jejuni. The minimal bactericidal concentration (MBC) was determined by subculturing ml on Mueller-Hinton agar and defined as 99-9% reduction in the initial inoculum. Nocardia asteroides was tested by agar dilution on Mueller-Hinton agar with * cfu/spot with 35 C incubation for three days. Effect of calcium and magnesium. Defined medium (Iso-sensitest agar, Oxoid) containing -3 mmol Ca +, - mmol Mg + and mmol Zn + respectively was supplemented in order to obtain a final concentration of - or - mmol/ Ca(NO3), - or 5 mmol/ MgSO, and 3 or 7 mmol/ ZnSO*. Inoculum effect. This effect was measured on Mueller-Hinton agar with or * cfu per inoculum. s of quinolones Results Against Gram-positive genera, had very poor activity as shown in Table I. In contrast 96 out of 97 staphylococci were susceptible to -quinolones; the exception was one methicillin resistant Staphylococcus aureus which was resistant to the quinolones tested. Activity of quinolones was similar against S. epidermidis and three groups of S. aureus classified according to their susceptibility to penicillin G and methicillin. S. saprophyticus was found to be one- or two-fold less susceptible than S. epidermidis. Streptococci were far less susceptible than staphylococci, most of the strains being in the intermediate susceptibility range. s of quinolones were similar against Streptococcus pneumoniae, Str. agalactiae and Str. faecalis. The majority of Listeria monocytogenes and N. asteroides were resistant to all quinolones tested except for a few sensitive strains. Quinolones were more active against the Enterobacteriaceae (Table II) than against Gram-positive bacteria. Most strains being inhibited by

In-vitro activity of newer qninolones 3 Table I. Activity of quinolones against Gram positive bacteria 3 (mg/) 9 (mg/) (mg/) Penicillin-sensitive S. aweus () Penicillin-resistant Oxacillin-sensitive S. aweus (9) Oxacillin-rcsistant S. aweus () S. epidermidis () 6 6 6 6-5 -5 > 3-> -5- ^ - -5- - 3- -5- ^* - -5- - 3-> -5- -3 O-5->3 5- - 6-> 5- - - -5- - S. saprophyticus () > > -> -5- ^* - - ^» Str. pneumoniae (3) > 6 3-> 6- -6-3 -5- - Str. agalactiae (9) Str.faecalis () > 3 > > 3 6 6 > > - -3-6 - -3 > - - -A - -

3 R. Acickenthaler et al Table I contd. 3 (mg/) 9 (mg/) (mg/) L. monocytogenes (6) N. asteroides (9) pcfloxacin > > 3 3 > 6 > 6 6 6 > - -6-6 - - 6-> -3-6 -6-6 Table IL Activity of quinolones against Entcrobacteriaceae 3 (mg/) 9 (mg/) (mg/) E. coli (3) Salmonella spp. () Shigella spp. (6) Y. enterocolitica (7) 5 5-6 6 6-3 5 5-6 6-5 -5 5 5-6 5 3 6 5 5 5 6-3 5-6 5 5-3 5-5 -> -> -- -5- -6- -5- -6- - - -5 5-.5 6-5 5- -6-.5 - -5- ---6-6- -6-.5-3- -6--5 l^t - -.3 5-6--5-3--5-5-

In-vitro activity of newer qtrfnotones 33 Table U contd. 3 (mg/) MIQ. (mg/) (mg/) Klebsiella spp. (9) Enterobacter spp. (7) Ser. mwcescens (3) Citrobacter freundii (3) Proteus spp. (3) Morganella morganii (6) Providencia rettgeri (6) cnoxacin cnoxacin -5-5 -5 5 5 5-5 - -6-5 5-5 5-5 -6 5-3 5-6 3 3 6 6 > 3-5 6-3 -3 5-5 -6-6 -- -6- -3- -5- -3- -3-6 ---6-5- -6- -3-5- -3-3 O-5-> -5- -5-5-> -5- -6 - - 5-5- -3- -6- -3 - -- -5- -3- -3-5- -6 - ---3 5- -3--5-6- 6- -> -3 -- 5-6 6-6 5-> -5-6

3 R. Anckenthaler et al. Table IIL Activity of quinolones against other aerobic and microaerophilic Gram-negative bacteria 5 (mg/), (mg/) (mg/) H. influenzae () C. jejuni () P. aeruginosa (9) P. maltophilia (7) Other Pseudomonas spp. (not aeruginosa) () Achromo. xylosoxydans and Alcalig. denitrificans (6) Acinetobacter spp. (5) cnoxacin cnoxacin - -5-6 -5-3 6 3-5 >6 6 > 5 6 6 >3 6 6-5 -6-3 -3 6 6 6 6 >6 3 > 6 3 > > >3 >3 3 6 6 6 - - -5-6- -3- -5- <5-l - - - - - - ^t 6-6-6-6- - - -O6-> - -6 >6 - -6 6->3 > - -6-6 3- -6-6^ -6- <km 6-3 > l-> 6->3 >3-3 -6 - - --5 -- -3-3 5- -3-

In-ritro actirity of newer qainoloues 35 Table VH contd. 5 (mg/) 9 (mg/) (mg/) Agrobacter spp. () Aeromonas and Plesiomonas spp. (9) N. meningitidis () najidixic acid cnoxacin 6-6 -5 - -6-3 -5 <-5 - -3-3 - -3 6-6 - 5-6 -5-6 - -6-3 -5-3 -3-3 -3--6 - -5- - -5-O-5 - -5- <O-O5-3--5 <<M)5-O-O6 ---3 <-5-O-O6 - O-O- <O5--6 < -5--3 ---5 <(H)5--6 < mg/. Some genera were rather more susceptible, such as Escherichia coli, Salmonella spp., Shigella spp. and Yersinia enterocolitica, while others were distinctly more resistant, such as Enterobacter spp., Serratia marcescens and Providencia rettgeri. Although not shown in Table II, ampicillin, carbenicillin, cephalothin, cefotaxime, gentamicin and amikacin were tested simultaneously with the quinolones. Many of the strains of enterobacteria used were resistant to penicillins, cephalothin and gentamicin. No relationship between resistance to these agents and to -quinolones could be found. However, several strains showed a high level of resistance to not seen with the -quinolones. As well as the Enterobacteriaceae, many other aerobic Gramnegative bacteria were susceptible to quinolones (Table III). These agents were extremely active against Aeromonas spp., Plesiomonas spp., Neisseria meningitidis and H. influenzae. Among the pseudomonas, P. maltophilia was more resistant than P. aeruginosa. The quinolones showed activity against C. jejuni and Agrobacter spp. comparable to that against the majority of enterobacteria, while Achromoxylosoxydans and Alcaligenes denitrificans were generally resistant. Acinetobacter spp. were susceptible to wide range concentrations of quinolones. Comparing the activity of the different quinolones tested on a weight basis, was the most potent compound, followed by and. Norfloxacin and were less active than. Differences between these compounds were more pronounced with Gram-negative bacteria than with Grampositive genera. All the -quinolones were more active than. MBCs of quinolones. The MBC/ ratios of, and were between and for the majority of strains, and between and 3 with Str.

36 R. Anckentfaaler et al Table IV. Mean MDC/ ratios for five quinolones Number Nalidixic Cipro- Nortested acid floxacin Enoxacin fioxacin Pcfloxacin PeniciUin-sensitive -7-5 -6-6 - S. aureus Penicillin-resistant -3 - -5 - -9 Methicillin-sensitive S. aureus Penicillin-resistant 6-7 -7-6 - -7 Methicillin-resistant S. aureus S. epidermidis 7-7 -9-7 -9 - S. saprophyticus inactive -6-7 - Str. agalactiae inactive - N.D. -6-5 Str. faecalis 6 inactive -6 3-5 -9 Str. pneumonias inactive -3-3 -3 - L. monocytogenes inactive - - -9 E. coli -3-6 - Shigella spp. - N.D. -3 Salmonella spp. 9 - -6 - -5 Enterobacter spp. 3-5 - -3 Klebsiella spp. 5 3-3 -3-7 Serratia spp. 3- N.D. - Proteus spp. - -6-5 - P. aeruginosa -7-5 -6 - -9 Pseudomonas spp. 5-7 -5 - -9 (not aeruginosa) * N.D. = not determined. agalactiae, Str. faecalis and L. monocytogenes (Table IV). In comparison with these three agents, bactericidal activity of was somewhat different, with higher MBC/ ratios against most staphylococci and Pseudomonas, but lower ratios against Str. agalactiae, Str. faecalis, and enterobacteria. Inoculum effect and mutational frequencies. Determinations of s with two inocula of and cfu showed an inoculum effect particularly obvious with Enterobacter cloacae and Ser. marcescens (Table V). The inoculum effect was most Table V. Fold increase of s with an inoculum of cfu per spot as compared to an inoculum of cfu per spot Strains Fold increase of s range with: Number nalidixic cipro- nortested acid floxacin floxacin E.coli K. pneumoniae E. cloacae Ser. marcescens P. aeruginosa S. aureus Str. faecalis ^* - -3-6 -6 - resistant - - -6-6 - - - - - -3-3 - - -6 - - - -6, - - - -6 - - - -

In-vitro activity of newer qninolodes 37 Table VI. Log values of the number of colonies growing on Mueller-Hinton Agar containing quinolone concentrations four or eight times the s (inoculum: 9 cfu)* Strains E. coli () K. pneumoniae () E. cloacae () Ser. marcescens () P. aeruginosa () S. aureus () of: Log of colonies growing at four or eight times the nalidixic acid ofloxatin x x x x x x x x x x x x - - - - 3-5 - -3 - - - - -3 - - - - - - - - - - No colonies grew at 6 times the with either compound. - - - - - - - - - - - -3 - - - - - - - - -3 - - - - - - marked with while the differences found between the -quinolones were not significant. Data of Table V correlated well with determination of mutation frequencies shown in Table VI. E. coli populations contained less than 9 mutants resistant to the quinolones tested. Within the other strains studied, Ser. marcescens and E. cloacae had the highest frequency of mutants, and we found more clones resistant to (* at a concentration of four times the s) than to -quinolones ( 5-9 at four times the s). All the -quinolones tested selected resistant mutants at similar frequencies. No colonies grew at 6 times the s with either compounds. Influence of cation supplementation. Addition of 3 or 6 mmol of ZnSO did not alter the size of the inhibition zone diameters produced by,, and gentamicin when testing six strains in a disc-diffusion method (Figure ). Supplementation with Ca(NO 3 ) reduced the inhibition zone diameters of, and gentamicin but not those of. Supplementation with MgSO produced significantly smaller inhibition zone diameters than those obtained with non-supplemented agar in all cases, whatever the antibiotic or the strain considered. Control CotNOjlj O- Co(NOj) - MqSQ, MgS 5O ZnS O3 ZnS O7 Gprofknocin -» I I i -I I Oflouxin Control Refkwocm Co(NOj) O- Co(NOj) - MqSOt MqS 5O i h ZnS OO3 ZnS O7 i 3 I Zone sue (mm) Gentomicin I 3 Figure. Influence of cation supplementation on the inhibition zone size obtained by disc diffusion with four antibiotics.

3 R. Auckenthaler et al. Discussion Newer -quinolones are characterized by a broader spectrum of activity when compared to. The data of this study confirm the similarity of the antibacterial spectrum for the -quinolone compounds tested (Wise, Andrews & Edwards, 93; Barry et al, 9; Bauerafeind & Petermuller, 9; van Caekenberghe & Pattyn, 9; Chin & Neu, 9; Hoogkamp, 9). -quinolones are active against all Enterobacteriaceae, including Enterobacter spp. or Ser. marcescens which are often resistant also to newer cephalosporins. In particular organisms causing diarrhoea including Campylobacter spp. are very sensitive and may therefore be useful in this clinical situation. Against P. aeruginosa the activity is modest and not related to concomittant resistance to carbenicillin or gentamicin. Other non-fermenters such as P. maltophilia, Achromobacter spp. or Alcaligenes spp. are more than ten times less sensitive and often resistant to quinolones. The spectrum of -quinolones also includes Gram-positive organisms including staphylococci and to a lesser degree streptococci and Listeria spp. The s of Nocardia spp. are close to the achievable peak serum levels of quinolones and therefore cannot be recommended for clinical trials. As found by other authors (Bauernfeind & Petermuller, 9; van Caekenberghe & Pattyn, 9) the activity on a weight basis is highest for, followed by and,, and the least active. However, this observation will have to be matched with pharmacokinetics and clinical results, because other examples have taught, that in-vitro activity is not a sufficient criterion for the evaluation of new antimicrobial agents. The bactericidal activity of quinolones is in general within --fold the. Only Str. agalactiae, Str. faecalis, and L. monocytogenes have MBC two to three times higher than the. Therefore the use of quinolones particularly for enterococcal endocarditis has first to be confirmed in the animal model. Cipr has higher MBC- ratios against P. aeruginosa when compared to the other -quinolones. Emergence of resistance during therapy has been noted with (Slack, 9), and newer -quinolones (Acar, Kitzis & Goldstein, 95; Lauwers et al., 95). In this study we have evaluated the frequency of resistant mutations by measuring the increase of by raising the inoculum, fold. The increase was two- to eightfold for most of the organisms tested. However, Enterobacter spp. and Ser. marcescens both well known for emergence of resistance during therapy with cephalosporin (Sanders & Sanders, 95), again have a higher increase of the up to 3 times and this could well explain the clinical failures. The mechanism of resistance is not known, but could be explained by a defect in bacterial penetration as the observation is common to all -quinolones tested (Smith, 9). The influence of ph, cations and various media has been mentioned by various authors. We have measured the influence of increased calcium, magnesium or zinc concentration on the zone size of a disc diffusion test. At the tested concentrations only magnesium reduced the zone size significantly for all compounds. The mechanism of impaired activity in presence of magnesium is unknown. Possibly, magnesium might interfere at least at two levels, either on the outer membrane as is the case with aminoglycosides, or at the level of DNA-gyrase-DNA interaction. The clinical significance of the magnesium effect is unknown. However, increased magnesium concentrations in the urine, together with a low ph, could be responsible for impaired therapeutic response in difficult to treat urinary tract infections. In addition, in testing

In-vitro activity of newer qninolooes 39 susceptibility of -quinolones, concentration of the magnesium in the medium should be standardized. References Acar, J., Kitzis, M.D. & Goldstein, F. (95). Frequency and mutation rate of new quinolones compared to. In th International Congress of Chemotherapy, Kyoto. Abstract, WS-7-6. Barry, A. L., Jones, R. N., Thorasberry, C, Ayers, L. W., Gerlach, E. H. & Sommers, H. M. (9). Antibacterial activities of,, oxolinic acid, craoxacin, and. Antimicrobial Agents and Chemotherapy 5, 633-7. Bauernfeind, A. & Petermuller, C. (9). Vergleich der antibacteriellen Profile von Cipr, Norfloxacin, Pefloxacin, Ofloxacin, Enoxacin, Pipemidsaure und Nalidixinsaure. Fortschritte antimikrobieller und antineoplastischer Chemotherapie 3-5, 53-. van Caekenberghe, D. L. & Pattyn, S. R. (9). In vitro activity of compared with those of other new fluorinated piperazinyl-substituted quinoline derivatives. Antimicrobial Agents and Chemotherapy 5, 5-. Chin, N. X. & Neu, H. C. (9). Cipr, a quinolone carboxylic acid compound active against aerobic and anaerobic bacteria. Antimicrobial Agents and Chemotherapy 5, 39-6. Greenwood, D. & Laverick, A. (93). Activities of newer quinolones against Legionella group organisms. Lancet ii, 79-. Gay, D., DeYoung, D. R. & Roberts, G. D. (9). In vitro activities of and against Mycobacternan tuberculosis, M. avium complex, M. chelonei, M. fortuitum, and M. kansasii. Antimicrobial Agents and Chemotherapy 6, 9-6. Heesen, F. W. A. & Muytiens, H. L. (9). In vitro activities of,,, cinoxacin, and against Chlamydia trachomatis. Antimicrobial Agents and Chemotherapy 5, 3-. Hoogkamp, J. A. (9). Comparative in vitro activity of five quinoline derivatives and five other antimicrobial agents used in oral therapy. European Journal of Clinical Microbiology 3, 333-. Lauwers, S., Vincken, W., Naessens, A. & Pierard, D. (95). The efficacy and safety of in the treatment of severe infections in patients hospitalised in intensive care units. In th International Congress of Chemotherapy, Kyoto. Abstract, WS-7-. National Committee for Clinical Laboratory Standards (93). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Tentative Standard M7-T. National Committee for Clinical Laboratory Standards, Villanova, Pa. Ridgway, G. L., Mumtaz, G., Gabriel, F. G. & Oriel, D. J. (9). The activity of and other -quinolones against Chlamydia trachomatis and Mycoplasma in vitro. European Journal of Clinical Microbiology 3, 3-6. van Roosbroeck, R. J., Privinciael, D. R. & van Caekenberghc, D. L. (9). Activity of the newer quinolones against Chlamydia trachomatis. British Journal of Veneral Disease 6, 35. Sanders, C. C. & Sanders, E. Jr. (95). Microbial resistance to newer generation betalactam antibiotics: clinical and laboratory implications. Journal of Infectious Diseases 5, 399-6. Slack, R. (9). Review of bacterial resistance, a challenge to the treatment of urinary infections. Journal of Antimicrobial Chemotherapy 3, Suppl. B, -7. Smith, J. T. (9). Chemistry and mode of action of -quinolone agents. Fortschritte antimikrobieller und antineoplastischer Chemotherapie 3-5, 93-5. Wise, R., Andrews, J. M. & Edwards, L. J. (93). In vitro activity of Bay 967, a new quinoline derivative, compared with those of other antimicrobial agents. Antimicrobial Agents and Chemotherapy 3, 559-6.