ANTIMICROBIAL RESISTANCE in Neisseria gonorrhoeae occurs as chromosomally mediated resistance to a variety of antimicrobial agents and plasmid-mediated, high-level resistance to penicillin and tetracycline. In the United States, because of the emergence and spread of high-level, plasmid-mediated resistance to penicillin and tetracycline, broad-spectrum cephalosporins and fluoroquinolones (500-mg ciprofloxacin or 400-mg ofloxacin) have been recommended for the primary treatment of uncomplicated gonorrhea.1 In developing countries, such antibiotics often are unavailable or prohibitively expensive, and agents such as penicillins are still used, and the fluoroquinolones are increasingly available for the treatment of gonorrhea.
The failure of gonococcal infections to respond to 500 mg of ciprofloxacin or 400 mg of ofloxacin has been reported from Australia, the United Kingdom, and the United States.2–4 Gonococcal strains with minimum inhibitory concentrations (MICs) of ≥1.0 μg/ml of ciprofloxacin and ≥2.0 μg/ml of ofloxacin are considered to be clinically resistant to treatment with the Centers for Disease Control-recommended doses of these agents.5 The increased isolation, in Australia, Canada, and the United States,3,4,6 of fluoroquinolone-resistant strains of N. gonorrhoeae from persons with infections contracted in the western Pacific or the Far East has increased interest in the frequency, geographic distribution, and diversity of strains from that region. A recent report described a high prevalence of decreased susceptibility or higher level resistance to fluoroquinolones among a high proportion of isolates from commercial sex workers in two Philippine cities.7
In this study, these strains were characterized by plasmid content, auxotype, serovar, and antimicrobial susceptibilities to assess the current diversity of antimicrobial-resistant strains of N. gonorrhoeae in this patient population. In addition, based on a previous observation that some strains classified as susceptible to ciprofloxacin by criteria recommended by the National Committee for Clinical Laboratory Standards were resistant to nalidixic acid (i.e., exhibited a genetically based decreased susceptibility to fluoroquinolones),5,8,9 strains also were screened for resistance to nalidixic acid to determine accurately the proportion of strains with a genetically based, decreased susceptibility to fluoroquinolones.
Materials and Methods
Between July 1994 and October 1994, N. gonorrhoeae was isolated from 32 (11.7%) of 273 and 68 (22.7%) of 300 female sex workers attending social hygiene clinics in Manila and Cebu City, Republic of the Philippines (RP), respectively. Specimens were inoculated on modified Thayer-Martin medium and incubated at 35 °C in a carbon dioxide-enriched atmosphere for 24 to 48 hours. Isolates were identified presumptively by gram stain and oxidase reaction and stored in trypticase soy broth containing 20% glycerol at −70 °C until transported to the Neisseria Reference Laboratory, University of Washington, Seattle, Washington. At that laboratory, 92 (92%) of 100 strains were viable and were confirmed as N. gonorrhoeae by biochemical tests; these strains included 28 isolates from Manila and 64 isolates from Cebu City.
Isolates were characterized by auxotype, serovar, and plasmid content as described previously.10–12 Susceptibilities to penicillin G, tetracycline, ceftriaxone, ciprofloxacin, ofloxacin, norfloxacin, erythromycin, and spectinomycin were determined by agar dilution tests. Susceptibilities were determined on GC II agar medium (Becton Dickinson, Cockeysville, MD) supplemented with 1% IsoVitaleX (Becton Dickinson).8 Quality control strains N. gonorrhoeae strain American Type Culture Collection 49226, F-28 (spectinomycin-resistant), and F-45 (chromosomally mediated-resistant N. gonorrhoeae [CMRNG]) were included in each susceptibility test run. The susceptibilities of strains to penicillin G, tetracycline, ceftriaxone, and spectinomycin were interpreted as recommended by the National Committee for Clinical Laboratory Standards (NCCLS)8; interpretive criteria were not available for norfloxacin and erythromycin. The NCCLS-recommended criteria were used to interpret susceptibility to ciprofloxacin and ofloxacin8; criteria proposed by the Centers for Disease Control were used to interpret resistance and intermediate resistance to ciprofloxacin, ofloxacin, and norfloxacin.5 Isolates with chromosomal resistance to both penicillin and tetracycline (MIC ≥ 2.0 μg/ml) and including two isolates with MICs of ≥1.0 μg/ml of both penicillin and tetracycline were defined as CMRNG13; an isolate with an MIC of ≥2.0 μg/ml of tetracycline and an MIC of <1.0 μg/ml of penicillin was defined as tetracycline-resistant.
Disk diffusion susceptibilities of isolates with MICs ≥ 0.03 μg/ml of ciprofloxacin were determined to nalidixic acid (30 μg, Becton Dickinson). Susceptibilities were determined, in duplicate, on supplemented GC II medium by using the method recommended by the NCCLS.9 Susceptibility tests were controlled with three N. gonorrhoeae control strains: American Type Culture Collection 49226, the results of which met the acceptable control ranges recommended by the NCCLS,9 and with N. gonorrhoeae control strains F-28 and F-45. Interpretive criteria were not available for nalidixic acid9; strains with zone inhibition diameters ≤ 9 mm were interpreted as resistant to nalidixic acid.5
Means were compared by analysis of variance or by Kruskal-Wallis one-way analysis of variance.14 Categoric data were compared by the chi-square test.
A total of 92 isolates of N. gonorrhoeae from female sex workers in Manila and Cebu City were characterized by plasmid content, antimicrobial susceptibility profile, auxotype, and serovar. Isolates belonged to a total of 26 auxotype/serovar (A/S) classes. The distribution and frequency of penicillin/tetracycline resistance phenotypes in Manila and Cebu City is shown, by A/S class, in Figure 1. Penicillinase-producing strains lacking the TetM determinant (penicillinase-producing N. gonorrhoeae [PPNG]) accounted for 65 (70.7%) of the 92 isolates tested (Table 1). Strains possessing the 3.2-Mda β-lactamase plasmid were isolated more frequently in Cebu City (37/64; 57.8%) than in Manila (8/28; 28.6%) (P = 0.02). One of the PPNG strains possessed a previously undescribed 3.9-Mda β-lactamase plasmid; this strain also possessed a 24.5-Mda conjugative plasmid. Only 14 (15.2%) of the 92 strains were susceptible to both penicillin and tetracycline. The distribution and frequency, by penicillin/tetracycline resistance phenotype and A/S class, are shown in Figure 1 for strains isolated in Manila and Cebu City. The susceptibilities (MIC range, geometric mean) of strains are listed, by penicillin-tetracycline resistance phenotype, in Table 2; susceptibilities of isolates from Manila and Cebu City that belonged to the same resistance phenotype were similar (data not shown). Strains possessing the 3.2-Mda plasmid had significantly lower MICs to penicillin, tetracycline, ceftriaxone, ciprofloxacin, and erythromycin (P < 0.01 for each comparison) than did strains possessing the 3.05-, 3.9-, or 4.4-Mda plasmid (Table 2). The susceptibilities of strains to ofloxacin and norfloxacin (data not shown) were 1 to 3 concentrations higher than those for ciprofloxacin.
Based on their resistance to nalidixic acid (inhibition zone diameters, ≤9 mm), 50 (54.3%) of 92 strains exhibited decreased susceptibility to fluoroquinolones. The distribution of these strains, by penicillin/tetracycline resistance phenotype and ciprofloxacin MIC, is listed in Table 3. When the NCCLS-recommended criterion for the interpretation of susceptibility to ciprofloxacin (MIC, ≤0.06 μg/ml) was used to classify strains,8 8 (16%) of the 50 strains were susceptible to ciprofloxacin; 42 (84%) of these 50 strains exhibited decreased susceptibilities to ciprofloxacin (MIC, ≥0.125 μg/ml). The PPNG strains accounted for 40 (80%) of 50 nalidixic acid-resistant strains and 34 (81%) of 42 strains with MICs of ≥0.125 μg/ml of ciprofloxacin. The PPNG strains possessing the 3.2-Mda β-lactamase plasmid accounted for 21 (50%) of 42 strains with MICs ≥ 0.125 μg/ml of ciprofloxacin and 2 (18.2%) of 11 strains with MICs ≥ 1.0 μg/ml of ciprofloxacin. The PPNG strains possessing the 3.05-Mda β-lactamase plasmid accounted for 4 (9.5%) of 42 strains with MICs ≥ 0.125 μg/ml of ciprofloxacin; none had MICs ≥ 1.0 μg/ml of ciprofloxacin. The PPNG strains possessing the 4.4-Mda β-lactamase plasmid accounted for 10 (23.8%) of 42 strains with ciprofloxacin MICs ≥ 0.125 μg/ml and 6 (54.5%) of 11 strains with MICs ≥ 1.0 μg/ml of ciprofloxacin. Strains with ciprofloxacin MICs of ≥0.125 μg/ml and ≥1.0 μg/ml also included CMRNG and tetracycline-resistant strains; susceptible strains had MICs 0.125 to 0.5 μg/ml, but not ≥1.0 μg/ml, of ciprofloxacin.
In this study, the population of gonococcal strains was marked by considerable diversity in A/S classes, penicillin/tetracycline resistance phenotypes, and β-lactamase plasmid profiles. Approximately 85% of strains were resistant to penicillin or tetracycline. Approximately 80% of all strains exhibited plasmid-mediated resistance to penicillin or tetracycline. These high rates of resistance to penicillin and tetracycline occurred despite the fact that neither of these agents is recommended for the treatment of gonorrhea in the RP. However, the finding of high rates of resistant strains may be consistent with apparent widespread use of over-the-counter antibiotics. Among these workers, 38% acknowledged use of an antibiotic within the 7 days before testing; the most frequently used antibiotics included penicillin (27%) and tetracycline (8%) (unpublished data, 1994). Such patterns of antimicrobial usage may have provided an advantage, particularly for strains with high-level, plasmid-mediated resistance.
Four different β-lactamase plasmids, including a previously undescribed 3.9-Mda plasmid, were identified among these strains; the PPNG strains possessing the 3.2-Mda β-lactamase plasmid accounted for approximately 70% of these strains. More than 50% of all strains were resistant to nalidixic acid, an indicator of decreased susceptibility to fluoroquinolones. Eighty percent of strains resistant to fluoroquinolones (ciprofloxacin MIC ≥ 1.0 μg/ml) were PPNG. Despite resistance to penicillin and tetracycline and decreased susceptibility to fluoroquinolones, all strains were susceptible to ceftriaxone.
In 1988, PPNG accounted for 41 (37.6%) of 109 strains isolated from women attending the Social Hygiene Clinic in Olongapo, a town on Subic Bay, not far from Manila (Figure 1); no data are available on the plasmid content of these strains.15 Of 55 PPNG strains isolated from commercial sex workers in 1989 in Angeles (Figure 1), 40 (72.7%) of 55, 12 (21.8%) of 55, and 3 (5.5%) of 55 strains possessed the 4.4-, 3.2-, and 3.05-Mda β-lactamase plasmids, respectively.16 Of these strains, 15 (37.5%) of 40 strains that possessed the 4.4-Mda plasmid and 9 (75%) of 12 strains that possessed the 3.2-Mda plasmid also possessed the 24.5-Mda conjugative plasmid. All strains that possessed the 3.05-Mda plasmid also possessed a 24.5-Mda conjugative plasmid. Thus, it appears that the proportion of PPNG strains that possess the 3.2-Mda β-lactamase plasmid may have increased in frequency in the RP since 1989. The proportion of isolates possessing the 3.05-Mda β-lactamase plasmid apparently has doubled since 1989. In this study, a strain possessing a new 3.9-Mda β-lactamase plasmid and the 24.5-Mda conjugative plasmid was identified. The 3.9-Mda plasmid could be transferred between gonococcal isolates by conjugation (S. R. Johnson; personal communication, 1996). This strain exhibited antimicrobial susceptibilities similar to strains possessing the 4.4-Mda β-lactamase plasmid. The fate of this plasmid will depend on whether its host strain is spread successfully and the plasmid itself is transferred to other strains that also are spread successfully.
Recent studies of susceptibilities of strains of N. gonorrhoeae isolated in the RP to antimicrobial agents have focused on susceptibilities to quinolones.7,15 Susceptibilities of gonococcal isolates in this survey can be compared, by penicillin/tetracycline susceptibility phenotype, with those of strains from uncomplicated infections in persons from various other geographic sources, including isolates from women in the United States diagnosed with pelvic inflammatory disease.13,17 With the exception that strains in the RP belonging to the susceptible phenotype do not exhibit the highly susceptible MICs of penicillin (0.015–0.03 μg/ml), the range of susceptibilities, MIC50, MIC90 values for penicillin G, tetracycline, ceftriaxone, and cefoxitin for susceptible, CMRNG, tetracycline-resistant N. gonorrhoeae (TRNG), and PPNG strains are similar to those determined recently in the United States.13,18 The CMRNG strains were slightly less susceptible to erythromycin than those tested in the United States. The PPNG strains possessing 3.2-, 4.4-, or 3.05-Mda plasmids exhibited MICs of penicillin and ceftriaxone similar to those measured previously in the United States.13 Only susceptibilities to fluoroquinolones were distinctly higher than those measured previously.13,18 Comparison of susceptibilities (geometric means) of PPNG strains showed that susceptibilities of strains possessing the 4.4-, 3.05-, and 3.9-Mda plasmids were similar, but markedly higher than those of strains possessing the 3.2-Mda β-lactamase plasmid. These data suggest that PPNG strains possessing the 3.2-Mda plasmid could have been imported into the RP and were not derived de novo in the Far East from PPNG strains possessing the 4.4-Mda plasmid.
Susceptibilities to fluoroquinolones of strains of N. gonorrhoeae from the RP were documented in 1988.15 At that time, strains exhibited a range of MICs of ciprofloxacin of 0.001 to 0.5 μg/ml (MIC50, 0.125 μg/ml; MIC90, 0.25 μg/ml), indicating that many strains exhibited decreased susceptibilities to the fluoroquinolones; no data are available on the diversity of such strains with respect to antimicrobial resistance phenotype or strain type.15 A recent survey of PPNG isolates, which exhibited decreased susceptibilities to ciprofloxacin (MICs ≥ 0.125 μg/ml), was from persons in Hawaii who apparently acquired infections in the RP and showed that strains with MICs of 2.0 μg/ml of ciprofloxacin possessed the 3.2-Mda β-lactamase plasmid, whereas strains with MICs of 0.06 to 0.125 μg/ml of ciprofloxacin possessed either the 3.2-, the 3.05-, or the 4.4-Mda β-lactamase plasmid.6
Strains classified as susceptible to ciprofloxacin by NCCLS-recommended criteria may exhibit genetically based, decreased susceptibility to fluoroquinolones as indicated by their resistance to nalidixic acid5,8; in this study, an additional eight strains were identified as exhibiting such decreased susceptibility to fluoroquinolones by this method. It has been shown that, in vitro, N. gonorrhoeae strains developed resistance to fluoroquinolones in a stepwise fashion.17 Thus, strains may be screened for resistance to nalidixic acid to accurately determine the proportion of strains that may constitute a reservoir from which strains exhibiting clinically significant resistance could emerge.
In summary, with the exception of susceptibilities to fluoroquinolones and the absence of strains highly susceptible to penicillin and tetracycline, N. gonorrhoeae strains isolated recently in the RP belong to a variety of penicillin/tetracycline resistance phenotypes with susceptibilities to other antimicrobial agents similar to those of strains belonging to the same resistance phenotypes isolated recently in the United States.13,18 Thus, differences in susceptibilities of the gonococcal strain population as a whole in the RP to various antimicrobial agents reflect differences in the proportion of strains belonging to various penicillin-tetracycline resistance phenotypes rather than differences in the susceptibilities of strains within individual resistance phenotypes. The gonococcal strain populations in the RP were distinguished by the high proportion of PPNG strains and the frequency of strains exhibiting decreased susceptibility to fluoroquinolones. These data illustrate the importance of periodic surveillance for resistance in N. gonorrhoeae to antimicrobial agents used for the primary therapy of gonorrhea to revise the agents of choice to effectively control gonorrhea.
1. Centers for Disease Control and Prevention. 1993 Sexually transmitted diseases treatment guidelines. MMWR 1993; 42(RR-14):57–59.
2. Corkill JE, Percival A, Lind M. Reduced uptake of ciprofloxacin in a resistant strain of Neisseria gonorrhoeae
and transformation of resistance to other strains. J Antimicrob Chemother 1991; 28:601–604.
3. Tapsall JW, Lovett R, Munro R. Failure of 500 mg ciprofloxacin therapy in male urethral gonorrhoea. Med J Aust 1992; 156:143.
4. Centers for Disease Control and Prevention. Fluoroquinolone resistance in Neisseria gonorrhoeae
-Colorado and Washington, 1995. MMWR 1995; 44:761–764.
5. Knapp JS, Hale JA, Wintersheid K, Neal SW, Whittington WL. Proposed criteria for interpretation of susceptibilities of strains of Neisseria gonorrhoeae
to ciprofloxacin, ofloxacin, enoxacin, lomefloxacin, and norfloxacin. Antimicrob Agents Chemother 1995; 39:2442–2445.
6. Knapp JS, Neal SW, Parekh MC, Ohye R, Higa H, Rice RJ. Emerging in vitro resistance to quinolones in penicillinase-producing Neisseria gonorrhoeae
strains in Honolulu, Hawaii. Antimicrob Agents Chemother 1994; 38:2200–2203.
7. Manalastas R, Abellanosa IP, Melosa VP, et al. Fluoroquinolone resistance in Neisseria gonorrhoeae
in the Republic of the Philippines. IUVDT World STD/AIDS Congress 1995; 136. Abstract.
8. National Committee for Clinical Laboratory Standards. Approved standard M7-A3. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Villanova, PA: National Committee for Clinical Laboratory Standards, 1993.
9. National Committee for Clinical Laboratory Standards. Approved standard M2-A5. Performance standards for antimicrobial disk susceptibility tests. 5th ed. Villanova, PA: National Committee for Clinical Laboratory Standards, 1993.
10. Short HB, Ploscowe VB, Weiss JS, Young FE. Rapid method for auxotyping multiple strains of Neisseria gonorrhoeae.
J Clin Microbiol 1977; 6:244–248.
11. Knapp JS, Tam MR, Nowinski RC, Holmes KK, Sandström EG. Serological classification of Neisseria gonorrhoeae
with use of monoclonal antibodies to gonococcal outer membrane protein I. J Infect Dis 1984; 150:44–48.
12. Meyers JA, Sanchez D, Elwell S, Falkow S. Simple agarose gel electrophoretic method for the identification and characterization of plasmid deoxyribonucleic acid. J Bacterial 1976; 127:1529–1537.
13. Rice RJ, Knapp JS. Antimicrobial susceptibilities of Neisseria gonorrhoeae
strain representing five distinct resistant prototypes. Antimicrobial Agents Chemother 1994; 38:155–158.
14. Siegel S. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, 1956.
15. Joyce MP, Aying BB, Vaughan GH, et al. In vitro sensitivity of Neisseria gonorrhoeae
to quinolone antibiotics in the Republic of the Philippines. Sixth International Pathogenic Neisseria Conference; 1988. Abstract E19.
16. Sarafian SK, Chu M-L, Kojima H, Sng EH, Joyce MP, Knapp JS. Distribution of the 3.05-Mdal “Toronto” β-lactamase plasmid among penicillinase-producing isolates of Neisseria gonorrhoeae
in the Far East. Sex Transm Dis 1991; 18:201–204.
17. Belland RJ, Morrison SG, Ison C, Huang WM. Neisseria gonorrhoeae
acquires mutations in analogous regions of gyrA
in fluoroquinolone-resistant isolates. Mol Microbiol 1994; 14:371–380.
18. Rice RJ, Knapp JS. Susceptibility of Neisseria gonorrhoeae
associated with pelvic inflammatory disease to cefoxitin, ceftriaxone, clindamycin, gentamicin, doxycycline, azithromycin, and other antimicrobial agents. Antimicrob Agents Chemother 1994; 38:1688–1691.