Fluoroquinolone Resistance Mutations in the parC, parE, and gyrA Genes of Clinical Isolates of Viridans Group Streptococci

Abstract

The nucleotide sequences of the quinolone resistance-determining regions (QRDRs) of the parC and gyrA genes from seven ciprofloxacin-resistant (Cp^r) isolates of viridans group streptococci (two high-level Cp^r Streptococcus oralis and five low-level Cp^r Streptococcus mitis isolates) were determined and compared with those obtained from susceptible isolates. The nucleotide sequences of the QRDRs of the parE and gyrB genes from the five low-level Cp^r S. mitis isolates and from the NCTC 12261 type strain were also analyzed. Four of these low-level Cp^r isolates had changes affecting the subunits of DNA topoisomerase IV: three in Ser-79 (to Phe or Ile) of ParC and one in ParE at a position not previously described to be involved in quinolone resistance (Pro-424). One isolate did not show any mutation. The two high-level Cp^r S. oralis isolates showed mutations affecting equivalent residue positions of ParC and GyrA, namely, Ser-79 to Phe and Ser-81 to Phe or Tyr, respectively. The parC mutations were able to transform Streptococcus pneumoniae to ciprofloxacin resistance, while the gyrA mutations transformed S. pneumoniae only when mutations in parC were present. These results suggest that DNA topoisomerase IV is a primary target of ciprofloxacin in viridans group streptococci, DNA gyrase being a secondary target.

Viridans group streptococci (VGS) are recognized as a major cause of bacteremia in neutropenic cancer patients (NCP). In some institutions, resistance to penicillin and macrolide antibiotics among these microorganisms isolated from NCP is frequent (1, 2, 8). In addition, the development of bacteremia due to penicillin- and macrolide-resistant strains in patients receiving prophylaxis with quinolones and either penicillin or macrolides was recently described (5). At our institution (Hospital Princeps d’Espanya), 11 (3.2%) of 343 VGS isolates from blood from nonneutropenic cancer patients were resistant to ciprofloxacin, but the percentage increased to 15% (9 of 61) of VGS isolates from blood from NCP (unpublished data). Furthermore, in an in vitro study (39), 71% of 55 evaluated strains from NCP were resistant to pefloxacin. The prior administration of quinolones could be an important risk factor for quinolone-resistant strain selection, as was observed for respiratory infections caused by ciprofloxacin-resistant (Cp^r) Streptococcus pneumoniae (21). However, the mechanisms involved in fluoroquinolone resistance of the VGS have not been investigated.

DNA gyrase (gyrase) and DNA topoisomerase IV (topo IV), the bacterial type II DNA topoisomerases, are known to be the intracellular targets of fluoroquinolones. Both enzymes function by passing a DNA double helix through another by use of a transient double-strand break (23). Gyrase, an A₂B₂ complex encoded by the gyrA and gyrB genes, catalyzes ATP-dependent negative supercoiling of DNA and is involved in DNA replication, recombination, and transcription (40); topo IV, a C₂E₂ complex encoded by the parC and parE genes, is essential for chromosome partitioning (17). The deduced amino acid sequences of ParC and ParE are homologous to those of GyrA and GyrB, respectively. Previous studies of Escherichia coli identified quinolone resistance mutations in a discrete region of GyrA (43) and GyrB (42) termed the quinolone resistance-determining region (QRDR). Recent studies identified similar mutations in the analogous region of ParC. However, E. coli parC mutations are expressed only in the presence of gyrA mutations (14, 18), and purified E. coli topo IV is less sensitive to quinolones than E. coli gyrase (18). Similar observations have been reported for Neisseria gonorrhoeae (4) and Haemophilus influenzae (13). The data suggest that in these gram-negative bacteria, gyrase and topo IV are primary and secondary targets, respectively, for ciprofloxacin. The inverse is true for S. pneumoniae (16, 25, 27, 36), for Staphylococcus aureus (11), and possibly for Enterococcus faecalis (20). Thus, in S. aureus and S. pneumoniae, mutations altering amino acid residues of the QRDR of ParC confer low-level Cp^r, and mutations altering those of the QRDRs of both ParC and GyrA confer high-level resistance. It appears that topo IV is the primary target for ciprofloxacin in these gram-positive bacteria.

The aim of this study was the characterization of the QRDRs of parC and gyrA of several ciprofloxacin-susceptible and -resistant strains of VGS identified as Streptococcus oralis and Streptococcus mitis. In addition, the QRDRs of parE and gyrB from some S. mitis strains were also characterized. The final goal was to establish the correlation between the genotype and the phenotype of susceptibility of resistance to fluoroquinolones and to demonstrate that, as in S. pneumoniae, topo IV is the primary target for ciprofloxacin in the VGS, gyrase being a secondary target.

MATERIALS AND METHODS

Bacterial strains and plasmids.

The S. oralis strains were classified by use of the API 32 Strep identification system (Biomerieux). The six S. mitis strains were isolated consecutively from blood samples taken between January 1988 and December 1994 at the Hospital Princeps d’Espanya (Barcelona, Spain). These strains were recovered from adult NCP (<500 granulocytes per mm³) to whom norfloxacin had been administered orally (400 mg twice daily) as the only prophylaxis. Only one isolate per patient was tested. Blood culturing was performed with the Roche Septi-Check system (Hoffmann-La Roche Inc., Nutley, N.J.), or samples were inoculated into BACTEC bottles and tested on the BACTEC NR 860 instrument (Johnson Laboratories, Inc., Towson, Md.). VGS were identified by standard methods (10, 33). Colony morphology was evaluated, and pure cultures were tested for the production of acid from trehalose, sorbitol, lactose, mannitol, sucrose, inulin, raffinose, glycerol, arabinose, maltose, and sorbose. The isolates were additionally tested for reactions on esculin agar and bile esculin agar, growth in 6.5% sodium chloride broth, ammonia production from arginine, pyruvate utilization, sodium hippurate hydrolysis, and hydrolysis of starch.

Growth, susceptibility testing, and transformation of bacteria.

Streptococci were grown in Todd-Hewitt broth (Difco) supplemented with 0.5% yeast extract (Difco). MICs were determined by the agar dilution method as recommended by the National Committee for Clinical Laboratory Standards. The inoculum was prepared by suspending some colonies from an overnight blood agar culture in sterile 0.9% saline and adjusting the turbidity to a 0.5 McFarland standard (ca. 10⁸ CFU/ml). The suspension was further diluted to provide a final bacterial concentration of 10⁴ CFU per ml per spot, which was delivered with a Steers replicator onto Mueller-Hinton agar plates (Difco) supplemented with 5% defibrinated sheep blood and appropriate concentrations of the antibiotics. S. pneumoniae R6, E. faecalis ATCC 29212, and S. aureus ATCC 293 were used for quality control. The following fluoroquinolones were used: ciprofloxacin (Bayer, West Haven, Conn.), sparfloxacin (Rhône-Poulenc Rorer, Antony, France), and clinafloxacin (Parke-Davis, Ann Arbor, Mich.). The plates were incubated at 37°C in a 5% CO₂ atmosphere for 20 to 24 h.

S. pneumoniae M22 (32), used in the transformation experiments, was also grown in liquid C medium containing 0.08% yeast extract, and transformation was performed as described by Tomasz (37). Cultures containing 5 × 10⁶ CFU per ml were treated with DNA (prepared as described below) at 0.1 to 1 μg/ml for 40 min at 30°C and then for 90 min at 37°C before transfer to selective media. Transformants of S. pneumoniae were selected on plates containing ciprofloxacin at the required concentration.

DNA isolation and manipulation.

Chromosomal DNA was obtained from streptococci as described previously (41), except that cells were taken from a 45-ml stationary-phase culture and suspended in 567 μl of 10 mM Tris (pH 8.0)–0.1 mM EDTA. Treatment with 25 μg of RNase per ml for 30 min at 37°C and phenol extraction were performed. Gel electrophoresis of PCR products was carried out with agarose gels as described previously (34). DNA was recovered from gel slices by use of a Gene Clean II Kit (Bio 101).

PCR amplification and DNA sequence determination and analysis.

The oligonucleotide primers used for PCR amplification and for sequencing were synthesized with a Pharmacia LKB Gene Assembler Plus DNA synthesizer. PCR amplification was performed with 2 U of Taq polymerase (Pharmacia), 1 μg of chromosomal DNA, a 1 μM concentration of each synthetic oligonucleotide primer, a 200 μM concentration of each deoxynucleoside triphosphate, and 5 mM MgCl₂ in the buffer recommended by the manufacturer. Amplification was achieved with an initial cycle of 5 min of denaturation at 95°C, 15 min of annealing at 55°C (7 min before and 8 min after the addition of the enzyme), and 6 min of polymerase extension at 72°C. Then, 20 cycles of 1 min of denaturation at 95°C, 2 min of annealing at 55°C, and 2.5 min of polymerase extension at 72°C, with a final 20 min of extension at 72°C and slow cooling at 4°C, were carried out. For amplification of the pertinent gyrA, parC, parE, and gyrB regions, oligonucleotide primers designed from the published sequences of these genes from S. pneumoniae (3, 24, 25, 28) were used. For gyrA, we used gyrA44 (5′-CCGTCGCATTCTCTACGGAATGAATGAATT-3′), coding for RRILYGMNEL, and gyrA170 (5′-AGTTGCTCCATTAACCAAAAGGTTTGGAAA-3′), the complementary strand of the primer coding for FPNLLVNGAT. Oligonucleotides parC50 and parC152 and oligonucleotides gyrB376 and gyrB52 (25) were used for the amplification of parC and gyrB, respectively. Amplification of the parE region was done with oligonucleotides parE398 (5′-AAGGCGCGTGATGAGAGC-3′), coding for KARDES, and parE483 (5′-TCTGCTCCAACACCCGCA-3′), the complementary strand of the primer coding for AGVGAD. DNA sequencing was carried out with protocols and materials from the fmol DNA Sequencing System (Promega). All sequences shown in this study were determined with both DNA strands. DNA and protein sequence comparisons were done with Intelligenetics PC Gene 6.0 software.

Nucleotide sequence accession numbers.

The new DNA sequences reported in this paper have been assigned the following GenBank accession numbers: AF079187 to AF79197 (parC regions), AF079198 to AF079208 (gyrA regions), AF079209 to AF079214 (parE regions), and AF079215 to AF079220 (gyrB regions).

RESULTS

Clinical strains.

We selected Cp^r S. mitis and S. oralis strains because they belong to the same group of VGS (7) and are presumably very close genetically. The two S. oralis strains were taken from sputum samples and initially were identified as S. pneumoniae because of their optochin sensitivity. Further characterization showed that they were insoluble in deoxycholate, and they were classified as S. oralis by use of the API 32 Strep identification system. The six S. mitis strains were isolated consecutively from blood samples from adult NCP who had received norfloxacin prophylaxis. Susceptible control strains (ciprofloxacin MICs, ≤2 μg/ml) were S. oralis NCTC 11427 and ATCC 10557 (type strains), S. mitis NCTC 12261 (type strain), and an S. mitis clinical isolate (strain 888). The susceptibilities of these strains to ciprofloxacin, sparfloxacin, and clinafloxacin were determined and are shown in Table 1. A 4- to 8-fold increase in Cp^r was observed for the five low-level Cp^r S. mitis isolates (MICs, 8 to 16 μg/ml) compared to Cp^r isolate 888 (or 8- to 16-fold compared to NCTC 12261^T), while a 32- to 64-fold increase was observed for the two high-level Cp^r S. oralis isolates (MICs, ≥64 μg/ml) (Table 1). No significant increase in resistance to the other two fluoroquinolones was observed for the S. mitis Cp^r isolates, with the exception of isolates 718 and 720 and clinafloxacin (a fourfold increase compared to Cp^s isolate 888). The two high-level Cp^r S. oralis isolates showed significant increases in resistance to sparfloxacin (at least 16-fold compared to the type strains) and clinafloxacin (at least 8-fold).

TABLE 1.

Susceptibility of strains to selected fluoroquinolones and mutations in the parC, parE, gyrA, and gyrB genes

Strains	MIC (μg/ml)a of:			Amino acid substitutionb
	CIP	SPA	CLX	ParC	ParE	GyrA	GyrB
S. pneumoniae R6, M22	0.5	0.25	0.125	—	—	—	—
S. oralis NCTC 11427T	2	0.5	0.125	—	ND	—	ND
S. oralis ATCC 10557T	2	0.5	0.125	—	ND	—	ND
S. oralis 3180	128	16	2	Ser-79→Phe	ND	Ser-81→Tyr	ND
S. oralis 3870	64	8	1	Ser-79→Phe	ND	Ser-81→Phe	ND
S. mitis NCTC 12261T	1	0.25	0.06	—	—	—	—
S. mitis 14	8	0.5	0.125	Ser-79→Ile	—	—	—
S. mitis 112	8	0.5	0.125	—	Pro-424→Gln	—	—
S. mitis 197	8	0.5	0.125	Ser-79→Phe	—	—	Ser-494→Thr
S. mitis 718	16	1	0.5	Ser-79→Ile	—	—	—
S. mitis 720	8	1	0.5	—	—	—	—
S. mitis 888	2	0.5	0.125	—	ND	—	ND
R13180C	2	ND	ND	Ser-79→Phe	ND	ND	ND
R23429C/3180A	64	ND	ND	Ser-79→Tyr	ND	Ser-81→Tyr	ND
R23429C/3870A	32	ND	ND	Ser-79→Tyr	ND	Ser-81→Phe	ND

^aCIP, ciprofloxacin; SPA, sparfloxacin; CLX, clinafloxacin. ND, not determined. 

^bPositions of substitutions are according to the S. pneumoniae R6 coordinates (3, 24, 25, 28). —, no change. 

Sequencing of the parC, parE, gyrA, and gyrB QRDRs.

We recently cloned and sequenced the S. pneumoniae parC (25), parE (25), gyrA (3), and gyrB (24) genes. The data allowed us to design PCR primers derived from the pneumococcal genes and to use them to amplify the QRDRs of the corresponding VGS genes. Amplification with oligonucleotides parC50 and parC152 yielded PCR products of 311 bp that were sequenced with the same oligonucleotides as primers. The sequences of 231 nucleotides (nt) of these parC products are shown in Fig. 1. The 292-nt sequence of the gyrA gene was determined in a similar way after amplification of 382-bp PCR products with oligonucleotides gyrA44 and gyrA170 (Fig. 2). The parC and gyrA nucleotide sequences showed high similarity (more than 89 and 87% identities, respectively) among the different strains. Comparisons of the amino acid sequences of the VGS with that of S. pneumoniae R6 showed a change of GyrA at residue 114 to Ser in R6 and to Gly in all other strains. Also, a change of ParC Asn-91 (to Asn in R6 and to Asp in all other strains, with the exception of S. mitis 112 and 720) was observed. Among the five low-level Cp^r S. mitis isolates, three carried sense mutations affecting residue Ser-79 of ParC: Ser-79 to Phe (strain 197) and Ser-79 to Ile (strains 718 and 14). No changes in GyrA, with the exception of the Ala-117 change to Asp in S. mitis 197 (Fig. 2), were observed. The relevance of this change will be discussed later. The two high-level Cp^r S. oralis isolates had amino acid changes at Ser-81 of GyrA (Ser-81 to Tyr in strain 3180 and Ser-81 to Phe in strain 3870) and Ser-79 of ParC (Ser-79 to Phe).

FIG. 1.

Comparison of the nucleotide sequences of a region of parC containing the QRDRs from S. pneumoniae R6 (SPN R6) and various strains of S. oralis (SOR) and S. mitis (SMI). R1(3180C) is M22 transformed to the first level of ciprofloxacin resistance by use of the parC50-152 PCR product from strain 3180. Nucleotides and amino acids (italics) are numbered by taking the first parC nucleotide as nt 1 and the first ParC residue as number 1. The strand corresponding to the mRNA is shown. Only nucleotides different from those of S. pneumoniae R6 are indicated; dashes indicate nucleotide identity. Sense mutations and the corresponding amino acid changes are shown in bold. The deduced amino acid sequence of S. pneumoniae R6 is shown at the bottom, together with the amino acid changes observed in the indicated strains (in parentheses).

FIG. 2.

Comparison of the nucleotide sequences of a region of gyrA containing the QRDRs from S. pneumoniae R6 (SPN R6) and various strains of S. oralis (SOR) and S. mitis (SMI). R2(3429C/3180A) and R2(3429C/3870A) are R1^3429C (25) transformed to the second level of ciprofloxacin resistance by use of the gyrA46-172 PCR products from strains 3180 and 3870, respectively. Nucleotides and amino acids (italics) are numbered by taking the first gyrA nucleotide as nt 1 and the first GyrA residue as number 1. Symbols are as defined in the legend to Fig. 1.

Since mutations in the C terminus of the parE and gyrB products may contribute to fluoroquinolone resistance in S. pneumoniae (27), the regions corresponding to residue positions 376 to 512 of GyrB and 398 to 483 of ParE were amplified with the pneumococcus-specific oligonucleotides, and PCR products were sequenced with the same oligonucleotides. Amplification was carried out with DNAs from the five Cp^r S. mitis isolates and NCTC 1226^T. As shown in Fig. 3, we found one change in the amino acid sequence of GyrB of strain 197 (Ser-494 to Thr), and a Gln substitution was observed in ParE Pro-424 of strain 112.

FIG. 3.

Comparison of the nucleotide sequences of regions of parE (A) and gyrB (B) containing the QRDRs from S. pneumoniae R6 (SPN R6) and various strains of S. mitis (SMI). Nucleotides and amino acids (italics) are numbered by taking the first parE or gyrB nucleotide as nt 1 and the first ParE or GyrB residue as number 1. Symbols are as defined in the legend to Fig. 1.

Genetic exchange of ciprofloxacin resistance by genetic transformation.

Direct evidence indicating that the parC and gyrA mutations are responsible for ciprofloxacin resistance was obtained by genetic transformation. Since no system for the transformation of streptococci other than S. pneumoniae has been developed, we used competent pneumococcal M22 cells as recipients and both chromosomal DNAs and the parC50-152 and gyrA46-172 PCR products from the VGS as donor DNAs. Chromosomal DNAs obtained from the Cp^r isolates were able to transform competent S. pneumoniae M22 (Cp^s; MIC, 0.5 μg/ml), while DNA from Cp^s S. mitis 888 was not. The frequencies of transformation to Cp^r achieved were about 10⁵ transformants per ml when S. oralis chromosomal DNA was used (the same frequency was observed with chromosomal DNA from Cp^r S. pneumoniae 1244 [25]) and 10⁴ when S. mitis chromosomal DNAs were used and when selection was done at 1 μg/ml. However, when selection was done at 4 μg/ml, only chromosomal DNAs from S. oralis 3180 and 3870 were able to transform, although at frequencies about 100 times lower (the same frequency was observed with S. pneumoniae 1244 chromosomal DNA). The latter frequency was consistent with transformation by two unlinked markers.

Only the parC PCR products of S. oralis 3180 and 3870 transformed S. pneumoniae to Cp^r, at frequencies of 1 × 10³ to 8 × 10³, while the PCR products from the S. mitis strains did not. The PCR products from S. pneumoniae 1244 transformed at frequencies of 10⁵ transformants per ml. Transformants obtained with the parC PCR products from S. oralis 3180 and 3870 increased the ciprofloxacin MIC for recipient strain M22 to 2 to 8 μg/ml. One of these transformants (R1^3180C) was selected for sequencing and showed the Phe substitution present in Ser-79 of the donor strain S. oralis 3180 (Fig. 1). This mutation conferred a fourfold increase in the ciprofloxacin MIC for recipient strain M22. Although the gyrA46-172 PCR products from the high-level Cp^r strains 3180 and 3870 did not transform strain M22, they were able to transform an M22 mutant strain carrying a Ser-79 change to Tyr in ParC (strain R1^3429C [25]) at frequencies of 1 × 10³ to 5 × 10³. The MICs of ciprofloxacin for the transformants were 32 to 64 μg/ml (64- to 128-fold increase). The nucleotide sequences of two of these transformants are shown in Fig. 2.

DISCUSSION

In the present study, we sequenced pertinent regions of the genes encoding topo IV and gyrase from VGS included in the S. mitis group (9) in an effort to characterize the mechanism of fluoroquinolone resistance. We were successful in amplifying regions of the parC, parE, gyrA, and gyrB genes from both clinical isolates and type strains by using oligonucleotides derived from the S. pneumoniae R6 sequence (25). This result is an indication of both the high level of conservation of these genes and the genetic proximity between pneumococci and this group of VGS.

The high level of variation observed between the nucleotide sequences of identical genes from clinical isolates of the same species was striking. In a 292-nt portion of the gyrA gene from the five S. mitis strains, we observed between 1 and 38 nt differences (excluding mutations involved in Cp^r) (Fig. 2). This result is in clear contrast to data obtained from the sequences of 13 independent clinical isolates of S. pneumoniae, in which no differences (6 isolates) or 1 nt difference (7 isolates) was observed (unpublished data). For parC, S. mitis strains showed between 1 and 20 nt differences within the 231 nt sequenced (Fig. 1), while for S. pneumoniae we found 2 nt changes (2 isolates), 1 nt change (1 isolate), and no changes (10 isolates) in the 13 isolates analyzed. A similar large number of nucleotide differences were observed when the sequences of the S. mitis clinical isolates were compared with that of ATCC 12261^T: from 7 to 37 nt differences in gyrA and from 10 to 23 nt differences in parC. Likewise, S. oralis NCTC 11427^T and ATCC 10557^T, although quite similar in their gyrA gene sequences (6 changes), were different in their parC gene sequences (15 changes). This high level of sequence heterogeneity among the VGS strains studied may be an indication of the poor classification of the group, as has been suggested by several authors (19, 30, 38).

On the other hand, when we considered individual sequences, we found identity in a 231-nt portion of the parC gene of S. pneumoniae R6 and S. mitis 112 and near identity (only 1 nt difference) between S. pneumoniae R6 and S. mitis 720. However, when the 292-nt gyrA gene sequences were considered, R6 and 112 differed by 18 nt and R6 and 720 differed by 16 nt. These results may reflect an interchange of genetic material between the VGS and pneumococci; such an interchange has been proposed to be the origin of the mosaic structure of the genes encoding the penicillin-binding proteins (35). Also, the laboratory transformation of competent S. pneumoniae cells with total chromosomal DNA from the VGS suggested that this interchange of genetic material may be possible in nature.

We have found mutations in the parC and gyrA genes that produce amino acid substitutions in equivalent residues: Ser-79 of ParC and Ser-81 of GyrA. These residue positions have been found to be involved in ciprofloxacin resistance in pneumococci (25) as well as in other bacteria. We have also found two novel sense mutations, one affecting the topo IV ParE subunit (S. mitis 112) and the other affecting the gyrase GyrB subunit (S. mitis 197). The ParE substitution (Pro-424 to Gln) in S. mitis 112 (Fig. 3) is in a region that is highly conserved between the ParE and the GyrB subunits and that is thought to be a domain involved in the interactions between the GyrB and the GyrA subunits. A quinolone resistance mutation leading to a ParE substitution (Asp-435 to Asn) in S. pneumoniae has been described (29), and gyrB mutations leading to similar Asp-to-Asn changes have been described for several species (15). Other gyrB mutations (12), as well as a parE mutation found in E. coli (6), are also located in this highly conserved region. The ParE substitution (Asp-435 to Asn) found in S. pneumoniae (29) conferred a fourfold increase in ciprofloxacin resistance, the same increase as that observed for S. mitis 112 compared to Cp^s strain 888 (Table 1). This result suggests that the parE mutation found in S. mitis 112 may be involved in ciprofloxacin resistance. However, a comparison of the mutations in parC of S. mitis 14, 197, and 718 and the coincidence of their ciprofloxacin MICs (8 μg/ml) (Table 1) suggest that the amino acid changes at Ser-79 of ParC are responsible for their Cp^r phenotype and that the gyrB mutation present in S. mitis 197 is not involved.

Although we were able to transform S. pneumoniae to Cp^r by using PCR fragments containing parC and gyrA mutations from the S. oralis Cp^r strains, no transformation was observed with PCR fragments from the S. mitis strains. That lack of transformation may have depended on two factors. One is the location of the mutation in the donor DNA, markers in a central position being transferred more frequently than those near the ends (22). Another factor is the extent of similarity between donor and recipient DNAs: the longer the regions of identical nucleotides located at both ends of a marker, the higher the efficiency of transformation (31). These factors may explain the absence of transformation with the S. mitis PCR products as well as the differences between the transformation efficiencies of the parC and gyrA PCR products from the S. oralis strains (1 × 10³ to 8 × 10³ transformants/ml) and their corresponding chromosomal DNAs (1 × 10⁵).

Our results show that the VGS and pneumococci share the same mechanism of ciprofloxacin resistance. The presence of mutations in parC or parE in low-level Cp^r isolates and the presence of mutations in both parC and gyrA in high-level Cp^r isolates indicate that the primary target for ciprofloxacin is topo IV, gyrase being a secondary target. However, this conclusion may not be true for other fluoroquinolones (26), since the levels of susceptibility to sparfloxacin and clinafloxacin were slightly affected or were not affected by a mutation in either parC or parE (Table 1). An increase in the levels of resistance to sparfloxacin and clinafloxacin was observed for the two S. oralis strains that harbored mutations in both parC and gyrA.

One of the Cp^r S. mitis strains, 720, was devoid of mutations in the QRDRs of gyrA, gyrB, parC, and parE (Table 1). Other portions of these genes could be implicated in resistance. A more likely explanation for this finding, however, is that another resistance mechanism, possibly active efflux of the drug, is involved.