TEXT

Top Abstract Text References

Trovafloxacin is a new fluoroquinolone with potent activity against both gram-negative and gram-positive bacteria and anaerobes (7, 10). In this study, the activity of trovafloxacin was compared with those of three fluoroquinolones (sparfloxacin, ofloxacin, and ciprofloxacin) and with those of unrelated antimicrobial agents (doxycycline and erythromycin) against different mycoplasma species found in humans. We also studied the activities of the four fluoroquinolones cited above against quinolone-resistant mutants of Mycoplasma hominis and Ureaplasma urealyticum that have been genetically characterized.

A total of 112 fluoroquinolone-susceptible strains, including 32 strains of Mycoplasma pneumoniae (31 clinical respiratory isolates and 1 reference strain [strain FH]), 7 strains of Mycoplasma genitalium (5 clinical isolates and 2 reference strains [strains G37 and M30]), 20 M. hominis doxycycline-susceptible strains (19 clinical isolates and 1 reference strain [strain PG21]), 10 doxycycline-resistant clinical isolates of M. hominis, 11 strains of Mycoplasma fermentans (9 clinical strains and 2 reference strains [strains PG18 and K7]), 2 strains of Mycoplasma penetrans (1 urethral isolate and 1 reference strain [strain GTU-54]), 15 U. urealyticum doxycycline-susceptible strains (13 clinical isolates and 2 reference strains [strains of serovars 2 and 8, respectively]), and 15 U. urealyticum doxycycline-resistant strains (14 clinical isolates and 1 reference strain [a strain of serovar 9]) were studied. Among the fluoroquinolone-resistant isolates, 12 M. hominis strains previously described and genetically characterized, 5 clinical isolates from three different patients (6), and 7 in vitro mutants (2, 4) were tested. These 12 strains harbored different levels of resistance depending on the type and the number of mutations and on the quinolone tested. One clinical isolate of U. urealyticum that was resistant to fluoroquinolones and that is described in this report was also studied. This isolate, named UUa, was obtained from synovial fluid from the knee of a 33-year-old hypogammaglobulinemic patient.

Each of the following antimicrobial agents was provided by the manufacturer: trovafloxacin and doxycycline (Pfizer, Orsay, France), sparfloxacin (Rhône-Poulenc-Rorer, Vitry-sur-Seine, France), ofloxacin and erythromycin (Roussel Uclaf, Romainville, France), and ciprofloxacin (Bayer-Pharma, Puteaux, France). Susceptibility testing was carried out as described previously (1) by an agar dilution method with Hayflick modified agar for mycoplasmal strains and by a broth dilution method with Shepard medium for ureaplasmal strains. The minimal bactericidal concentrations (MBCs) of trovafloxacin and of the comparative compounds for a reference strain of each species were determined as reported previously (3).

Different mutations in DNA gyrase and topoisomerase IV have been shown to confer fluoroquinolone resistance (8). Chromosomal DNA of U. urealyticum strain UUa was used as a template in PCR to amplify the quinolone resistance-determining regions (QRDRs) of the gyrA, gyrB, parC, and parE genes. Primers gyrA-1 (5'-TTGCTGCTTTCGAAAACGG-3') and gyrA-2 (5'-CTGATGGTAAAACACTTGG-3'), primers gyrB-3 (5'-CCTGGTAAATTAGCTGACTG-3') and gyrB-4 (5'-TTCGAATATGACTGCCATC-3'), primers parC-5 (5'-ACGCAATGAGTGAATTAGG-3') and parC-6 (5'-CACTATCATCAAAGTTTGGAC-3'), and primers parE-7 (5'-ATGGGCGGAAAATTAACGC-3') and parE-8 (5'-CTTGGATGTGACTACCATCG-3') were used as described previously (2). PCR-amplified DNA was sequenced in both directions by previously described methods (6).

The comparative in vitro activities of trovafloxacin and the other antimicrobial agains against all mycoplasmal and ureaplasmal strains except fluoroquinolone-resistant mutants are shown in Table 1. The overall activity of trovafloxacin against all the strains tested was very good. A concentration of 0.25 µg of trovafloxacin per ml inhibited 90% of the strains tested except the fluoroquinolone-resistant mutants of M. hominis and U. urealyticum.

For M. pneumoniae and M. genitalium, the MICs of trovafloxacin (MICs at which 90% of isolates are inhibited [MIC₉₀s], 0.25 µg/ml; MIC range, 0.03 to 0.06 µg/ml) were comparable to those of sparfloxacin and doxycycline but 4- to 30-fold lower than those of ofloxacin and ciprofloxacin. Erythromycin showed the best activity, inhibiting all the M. pneumoniae and M. genitalium strains at a concentration 0.015 µg/ml. M. hominis and M. fermentans were highly susceptible to trovafloxacin (MIC₉₀s, 0.03 µg/ml) and sparfloxacin (MIC₉₀s, 0.03 and 0.06 µg/ml, respectively). As expected, M. hominis and M. fermentans were resistant to erythromycin. Against M. penetrans, trovafloxacin (MIC range, 0.015 to 0.03 µg/ml) was 2- to 4-fold more active than sparfloxacin and 8- to 16-fold more active than ofloxacin and ciprofloxacin.

The activity of trovafloxacin was compared with those of sparfloxacin, ofloxacin, and ciprofloxacin against 12 fluoroquinolone-resistant mutants of M. hominis selected in vitro and in vivo (Table 2). Trovafloxacin showed the best activity (MIC, 0.5 µg/ml) against all five clinical isolates (isolates MHa to MHc2) that harbored alterations in both GyrA plus ParC or GyrA plus ParE, with MICs 4- to 32-fold lower than those of the other fluoroquinolones tested. Furthermore, the trovafloxacin MIC of 0.5 µg/ml is below the breakpoint (1 µg/ml). Seven in vitro-selected mutants that have been genetically characterized (2, 4) were also tested. Trovafloxacin kept the same good activity (MIC, 0.1 µg/ml) against isolates with single mutations in both gyrase (gyrA) or topoisomerase IV (parC or parE). Double mutations (gyrA plus parE, gyrA plus parC, or gyrA plus gyrA) seem to be necessary for large increases in the trovafloxacin MIC. Thus, alterations in GyrA Ser-83 and ParE Asp-420 (mutant N7, selected in multiple steps on norfloxacin), GyrA Ser-83 and ParC Glu-84 (mutant IIS1; a second-step mutant selected on sparfloxacin), and GyrA Ser-83 and Ser-84 (mutant IIS3A; a second-step mutant selected on sparfloxacin) were associated with 33-, 66-, and 133-fold increased trovafloxacin MICs, respectively, over the breakpoints for strains IIS1 and IIS3A. The last mutant, IIIS3A1, a third-step sparfloxacin-resistant mutant with three alterations in quinolone targets (two gyrA and one parC), was the most resistant, with a 266-fold increase in the trovafloxacin MIC for the strain (Table 2). It should be noted that in strains with double mutations the increase in the trovafloxacin MIC seems to depend on the altered position. Thus, the association which gave the highest level of resistance to trovafloxacin was the double GyrA mutations Ser-83 plus Ser-84, followed by the association of GyrA Ser-83 plus ParC Ser-84.

Trovafloxacin harbored the best activity against all the ureaplasma strains studied (Table 1). Doxycycline-resistant strains were as susceptible to trovafloxacin as doxycycline-susceptible strains, as described previously for other fluoroquinolones (3, 5). Trovafloxacin had the lowest MIC (4 µg/ml) for the U. urealyticum clinical strain resistant to fluoroquinolones (Table 2) in comparison to those of the three other quinolones, which had 64- to 128-fold increased MICs for the clinical strain compared to those for the reference strain. However, the trovafloxacin MIC for strain UUa was over the breakpoint. Strain UUa has been characterized genetically and harbors four alterations in both GyrA and ParC QRDRs in comparison to the sequence of the U. urealyticum serovar 3 reference strain. In the GyrA QRDR, two amino acid substitutions (base changes indicated by the underscores), a Gln(CAA)-to-Arg(CGA) change at position 83 and a Asp(GAC)-to-Glu(GAA) change at position 95, have been found. The ParC subunit presented two alterations, a Thr122(ACA)-to-Ala(GCT) change and a Thr133(ACT)-to-Ala(GCC) change, outside the QRDR right end but very near the Tyr-120 active site. It is noteworthy that strain UUa was not resistant to doxycycline but had intermediate susceptibility to erythromycin.

For the reference strain of each mycoplasma and ureaplasma species, the MBC of trovafloxacin ranged from 0.015 to 1 µg per ml, that is, a value 16-fold higher than the MIC for the reference strain, depending on the species tested (Table 1).

In summary, trovafloxacin ranked among the best of the fluoroquinolones active against mycoplasmas. This study extends significantly the results of a previous study by other investigators (9). Furthermore, trovafloxacin had the best activity against fluoroquinolone-resistant mutants of M. hominis and U. urealyticum, with an MIC below the breakpoints for some strains. These results suggest that development of clinical resistance to trovafloxacin in mycoplasmas would probably require two or more mutations in fluoroquinolone targets.

(This work was presented in part at the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, Calif., 26 to 29 September 1999.)