Identification of Novel Mutation Patterns in the parC Gene of Ciprofloxacin-Resistant Isolates of Neisseria gonorrhoeae

Abstract

Of 65 ciprofloxacin-resistant, clinical isolates of Neisseria gonorrhoeae, 5 isolates exhibited ParC mutations previously undescribed in the gonococcus. For isolates containing two ParC mutations (the Ser-87→Ile and Glu-91→Gly mutations and the Gly-85→Cys and Arg116→Leu mutations) the MICs of ciprofloxacin (8.0 to 64.0 μg/ml) were higher than those for the isolate containing the single ParC mutation (Arg-116→Leu; MIC, 1.0 μg/ml).

Fluoroquinolones are frequently used in the treatment of gonorrhea, with ciprofloxacin and ofloxacin used as primary treatment regimens in a number of countries. Recently, a number of gonococcal strains with decreased susceptibility or clinically significant resistance to the Centers for Disease Control and Prevention (CDC)-recommended doses of fluoroquinolones have been isolated (2, 11). A number of studies have been performed to determine the mechanism of fluoroquinolone resistance in Neisseria gonorrhoeae (1, 3, 4, 6, 18, 19). The most prevalent mechanisms contributing to fluoroquinolone resistance in the gonococcus involve mutations in the quinolone resistance-determining region of gyrA and the analogous region of the parC locus on the chromosome. These mutations result in altered GyrA and ParC proteins (1, 4, 6, 18). These altered proteins can no longer be bound by the fluoroquinolone, and therefore, the drug is unable to inhibit DNA replication and the bacterium becomes less susceptible or resistant. The level of susceptibility appears to correlate to the location and number of mutations present (6). This mechanism is analogous to those observed in Escherichia coli and other bacteria (9, 20). However, unlike E. coli, mutations in the N. gonorrhoeae gyrB gene do not appear to have a significant impact on fluoroquinolone resistance (7), and no parE homolog has been described.

Deguchi et al. (4–8) have studied the effects that these mutations have on the level of fluoroquinolone resistance seen in various clinical gonococcal isolates. The mutations reported to date have occurred in amino acids Ser-91 and Asp-95 of GyrA and amino acid Asp-86, Ser-87, Ser-88, or Glu-91 of ParC. The order in which mutations occur and result in increased fluoroquinolone MICs for gonococcal strains follow a fairly predictable order: mutations in Ser-91 and/or Asp-95 of GyrA result in decreased susceptibility to ciprofloxacin (ciprofloxacin MICs, 0.125 to 0.5 μg/ml). On the other hand, clinically significant resistance to CDC-recommended fluoroquinolone regimens (MIC, ≥1.0 μg of ciprofloxacin per ml) appears to require one or more mutations in the ParC gene, in addition to both gyrA mutations. However, this order of mutations is not absolute and a number of gyrA and parC mutation patterns have been described. In this paper, we report the characterization of gonococcal isolates exhibiting clinically significant resistance to ciprofloxacin (MICs, 1.0 to 64.0 μg/ml) in which the parC mutations differ either in number or in location from those described previously.

Eighty-three clinical isolates of N. gonorrhoeae were obtained from commercial sex workers during the course of a study on fluoroquinolone resistance in the Republic of the Philippines in 1996. All isolates exhibited decreased susceptibility to ciprofloxacin (MICs, ≥0.125 μg/ml) and were characterized by auxotype, serovar, and antibiotic sensitivities as described previously (10, 12–14, 17). Of these isolates, 65 were ciprofloxacin resistant (MIC range, 1.0 to 64.0 μg/ml). The presence of gyrA and parC mutations was determined by PCR and restriction endonuclease reaction analysis as described by Deguchi et al. (5, 8). Regions of gyrA and parC were amplified by PCR after extraction of the chromosomal DNA of each gonococcal strain. PCR amplification consisted of 35 cycles of denaturation at 94°C for 60 s, annealing at 52°C for 50 s, and extension at 72°C for 50 s in reaction aliquots of 50 μl. Each PCR mixture contained 5 μl of 10× PCR buffer (Boehringer Mannheim, Indianapolis, Ind.), 1 μl of a deoxynucleoside triphosphate stock solution (containing dATP, dCTP, dGTP, and dTTP each at a concentration of 10 mM; Boehringer Mannheim), 0.5 μl of Taq DNA polymerase (5 units/μl; Boehringer Mannheim), 2.5 μl of each of the appropriate primers (1 mM), 28.5 μl of sterile water, and 10 μl of template DNA. PCR primers specific for gyrA and parC were the same as those described by Deguchi et al. (8) and were synthesized at the National Center for Infectious Diseases core facility at CDC in Atlanta, Ga. For each isolate 10 μl of PCR product was restricted with endonuclease HinfI for gyrA or EcoRI, PstI, SalI, or HinfI for parC (Boehringer Mannheim) and was visualized after electrophoresis on a 6.5% polyacrylamide gel.

Of the 83 isolates characterized, five isolates (Table 1) exhibited mutation patterns not described previously (6). PCR and restriction endonuclease analysis showed that all five isolates contained GyrA mutations at both Ser-91 and Asp-95. Three isolates (isolates RP96-70, RP96-c1-003, and RP96-c2-001) contained two ParC mutations at Ser-87 and Glu-91. Two other isolates (isolates RP96-46 and RP96-61) contained no mutations in ParC detectable by PCR and restriction endonuclease analysis. All five isolates were further characterized by sequencing of the region of the ParC gene known to be involved in fluoroquinolone resistance. Primers for sequencing were designed to produce a 300-bp fragment, representing amino acids 60 through 139, by PCR. The primers used were a forward-synthesis 21-base oligonucleotide with the sequence 5′-GTT TCA GAC GGC CAA AAG CCC-3′ and a reverse-synthesis 21-base oligonucleotide with the sequence 5′-GGA CAA CAG CAA TTC CGC AAT-3′. PCR products were purified for sequencing with the High Pure PCR Product Purification Kit (Boehringer Mannheim). Sequencing reactions were performed in 20-μl volumes with ABI Prism dRhodamine Terminator Cycle Sequencing Ready Reaction Kit reagents (Applied Biosystems Division of the Perkin-Elmer Corporation, Foster City, Calif.) containing 8.0 μl of Terminator Ready Reaction Mix, 2 μl of PCR product template (approximately 75 ng), 3.2 μl of 1 μM primer, and 6.8 μl of sterile, deionized water. Reactions were cycle sequenced on a GeneAmp PCR System 2400 (Perkin-Elmer) in 25 cycles of denaturation at 96°C for 10 s, annealing at 50°C for 5 s, and extension at 60°C for 4 min. Spin column purification with Centrisep Columns (Princeton Separations, Adelphia, N.J.) was used to remove excess dye terminators and to purify the reaction mixtures before electrophoresis. Electrophoresis of the extension reaction mixtures was performed on the ABI Prism 377 DNA Sequencer (Applied Biosystems Division of Perkin-Elmer Corporation) with a 4% polyacrylamide gel (19:1 acrylamide:bis acrylamide; Bio-Rad Laboratories, Hercules, Calif.).

TABLE 1.

Characteristics of six clinical isolates of N. gonorrhoeae containing novel mutations in the ParC gene^a

Strain	Cip MIC (μg/ml)	ParCb		A/S class	Pen/Tet type
		Wild-type	Mutation
RP96-70	16.0	Ser-87, Glu-91	Ile-87, Gly-91	PA/IB-1	CMRx
		AGT, GAG	AAT, AAG
RP96-1-003	8.0	Ser-87, Glu-91	Ile-87, Gly-91	PA/IB-5	PPNG
		AGT, GAG	AAT, AAG
RP96-2-001	64.0	Ser-87, Glu-91	Ile-87, Gly-91	PA/IB-1	PPNG
		AGT, GAG	AAT, AAG
RP96-46	1.0	Arg-116	Leu-116	Proto/IB-1	PPNG
		CGC	CTC
RP96-61	16.0	Gly-85, Arg-116	Cys-85, Leu-116	Proto/IB-1	PPNG
		GGC, CGC	TGC, CTC

^aAll isolates contained GyrA mutations at Ser-91 and Asp-95. Abbreviations: Cip, ciprofloxacin; A/S, auxotype/serovar; Ser, serine; Glu, Glutamine; Gly, glycine; Arg, arginine; P, proline-requiring; A, arginine-requiring; Proto, no requirement for proline, arginine, hypoxanthine, uracil, or methionine; CMRx, strain for which the penicillin MIC was 2 μg/ml and the tetracycline MIC was 1 μg/ml (15, 16); PPNG, penicillinase-producing N. gonorrhoeae; Pen, penicillin; Tet, tetracycline. 

^bWild-type amino acid(s) followed by the amino acid(s) resulting from the mutation. The corresponding nucleotide codon sequence is directly below each amino acid. 

Sequence analysis confirmed the presence of two mutations in the ParC genes of isolates RP96-70, RP96-c1-003, and RP96-c2-001 that corresponded to amino acid changes at Ser-87→Ile and Glu-91→Gly. Sequence analysis of isolates RP96-46 and RP96-61 demonstrated the presence of mutations in parC that had not been described previously. Isolate RP96-46 contained a mutation which resulted in an amino acid change at position 116 from arginine to leucine. Isolate RP96-61 contained not only the mutation at position Arg-116 but also an additional mutation which caused an amino acid change at position 85 from glycine to cysteine.

This is the first report of the presence of double ParC mutations identified in clinical isolates of N. gonorrhoeae. A double mutation was identified in a laboratory-induced resistant strain(1); the locations of the ParC mutations identified in this strain (Ser-88 and Gly-91) were different from the locations of the double mutations that we describe here. Additional examination of the three isolates containing the Ser-87→Ile and Glu-91→Gly ParC mutations suggest that they are not separate isolations of the same clone (Table 1). RP96-70 was auxotype/serovar class PA/IB-1 and exhibited decreased susceptibility or resistance to penicillin and tetracycline (CMRx), while RP96-c1-003 was PA/IB-32 (penicillinase-producing N. gonorrhoeae [PPNG]) and RP96-c2-001 was PA/IB-1 (PPNG). The other two isolates in this study present a different scenario. Isolates RP96-46 and RP96-61 belonged to auxotype/serovar class Proto/IB-1 and were PPNG. RP96-46, for which the ciprofloxacin MIC was 1.0 μg/ml, contained only the mutation Arg-116→Leu. In contrast, RP96-61 contained the Gly-85→Cys mutation in addition to the Arg-116→Leu mutation, and the ciprofloxacin MIC for this strain was 16.0 μg/ml. Acknowledging that transformation studies need to be performed for confirmation, these results suggest that the Arg-116→Leu mutation alone may be sufficient to produce clinically significant resistance (ciprofloxacin MIC, 1.0 μg/ml) but that the addition of the Gly-85→Cys mutation is necessary for the high-level MICs for RP96-61 (ciprofloxacin MIC, 16.0 μg/ml).

In summary, we have identified five ciprofloxacin-resistant isolates of N. gonorrhoeae that contain novel mutation patterns in parC, including two double mutations, the Ser-87→Ile and Glu-91→Gly mutations and the Gly-85→Cys and Arg-116→Leu mutations. Of these mutations, the Gly-85→Cys and Arg-116→Leu mutations in parC have not been previously observed in the gonococcus. The importance of the double parC mutations appears to be the high-level ciprofloxacin resistance exhibited by the isolates which contain them. All four isolates which possess parC double mutations have high-level resistance to ciprofloxacin (MICs, 8.0 to 64.0 μg/ml), whereas for the isolate with a single parC mutation the ciprofloxacin MIC was 1.0 μg/ml. That the three strains that contain the Ser-87→Ile and Glu-91→Gly double mutations have different levels of ciprofloxacin resistance is not unexpected. For strains which contain the same mutation pattern with a single ParC mutation, a similar wide, although lower, range of MICs has been demonstrated (unpublished data). We speculate that the presence of double ParC mutations facilitates the high ciprofloxacin MICs for a strain, but the MIC may also be influenced by other characteristics of the strain.