1142
Hydrogen Peroxide Production by Lactobacillus Species: Correlation with Susceptibility to the Spermicidal Compound Nonoxynol-9 Jacqueline A. McGroarty, Lisa Tomeczek, Donald G. Pond, Gregor Reid, and Andrew W. Bruce
Departments of Surgery (Division of Urology). Microbiology. and Obstetrics and Gynaecology. University of Toronto and Toronto General Hospital; Department ofMicrobiology and Immunology, University of Western Ontario, London, Canada
Lactobacilli are believed to play a protective role in the urogenital tract, guarding against infection by pathogens by a combination of competitive exclusion and inhibitor production [I]. Lactobacilli sterically hinder the adhesion of pathogens to the urogenital epithelia. Lactobacilli attached to cells block the adhesion of pathogens. Lactobacilli also produce a number of antimicrobial substances such as lactic acid, which is produced from carbohydrates and helps to maintain a low vaginal pH. In addition, many strains produce hydrogen peroxide (HzOz) and antibiotic-like substances that can inhibit a range of pathogens including Escherichia coli [2], enterococci [3], and gardnerellae [4). Women with a history of urinary tract infection have a urogenital flora dominated by pathogens, whereas lactobacilli predominate in healthy, premenopausal women [5]. Women using diaphragms and spermicide as a means of contraception are at an increased risk of acquiring urinary tract infections [6-8] and bacterial vaginosis [8). In addition, E. coli is recovered twice as frequently from the vaginas of diaphragm users, whose vaginal pH is higher, than from women using other forms of contraception [7-9). It has been suggested that this is due, in part, to spermicidal compounds altering the vaginal flora. Nonoxynol-9 is the active compo-
Received 17 April 1991; revised 30 December 1991. Informed consent was obtained from all subjects. Grant support: Medical Research Council ofCanada. Ontario Ministry of Health. and Sir Jules Thorn Charitable Trust. Reprints or correspondence (present address): Dr. J. A. McGroarty. Department of Microbiology, University of Glasgow. Glasgow G 12 8QQ, UK. The Journal of Infectious Diseases 1992;165:1142-4 © 1992 by The University of Chicago. All rights reserved. 0022-1899/92/6506-0027$01.00
nent of many spermicidal preparations. It is generally used at 5% concentration in creams and at 12.5% in spermidical foams and would be present at concentrations of 1.6%-2.7% after sexual intercourse. It has been shown to inhibit most lactobacilli at concentrations of 0.1%-1.0% (wt/vol) in vitro, but it has no effect on the growth of uropathogenic bacteria, including E. coli [10). It is feasible that the presence of nonoxynol-9 could upset the ecologic balance of the vagina by inhibiting the protective lactobacilli and allowing overgrowth ofpathogenic microorganisms. However, an examination ofthe vaginal bacterial flora by Watt et al. [9] found that the frequency of isolation of lactobacilli was similar in diaphragm users and nonusers. They did not quantity the lactobacilli present. Since vaginal pH is higher in spermicide users [7), it is possible that lactobacilli numbers are reduced. Alternatively, it may be that only non protective lactobacilli survive the presence of the spermicide. Eschenbach et al. [II) found HzOz-producing facultative Lactobacillus species in the vaginas of 96%of normal women but in only 6%of women with bacterial vaginosis. They concluded that the production of HzOz by Lactobacillus species may represent a nonspecific antimicrobial defense mechanism of normal vaginal ecosystems. We sought to determine the frequency of isolation of HzOz-producing lactobacilli in normal, healthy women and to examine the lactobacilli for a correlation between HzOz production and susceptibility to nonoxynol-9.
Materials and Methods
Patient population. Vaginal swabs were obtained from 88 nonpregnant, healthy, premenopausal, sexually active women
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
Facultative anaerobic lactobacilli were recovered from the vaginas of 96.8% of 63 nonpregnant, healthy, premenopausal women. The predominant specieswere Lactobacillusjensen ii, Lactobacillus acidophilus, and Lactobacillus casei. Of the women, 74.6% had hydrogen peroxideproducing lactobacilli, 22.2% had non-hydrogen peroxide-producing lactobacilli, and 3.2% had no lactobacilli. None of the 68 isolates had catalase activity. Some 68.2% of the isolates were inhibited by concentrations of ~ 1% (wt/vol) of nonoxynol-9 (bactericidal for 73.3% of isolates, bacteriostatic for 26.7%).The remaining 31.8% could grow in all concentrations to 25% (wt/vol) of nonoxynol-9. All of the lactobacilli that were sensitive to nonoxynol-9 produced hydrogen peroxide whereas only 3 of 21 resistant strains were hydrogen peroxide producers. A significant correlation (P < .001, x 2 test) was found between hydrogen peroxide production and sensitivity to nonoxynol-9. It is suggested that the vaginal flora of spermicide users could be depleted of hydrogen peroxide-producing lactobacilli, possibly increasing susceptibility to urogenital infection.
JID 1992;165 (June)
Concise Communications
Catalase assay. The lactobacilli were tested for catalase activity by running a 3%(vol/vol) solution ofH 202 down an MRS agar slant. The agar slants containing lactobacilli growth were observed for bubbles of oxygen that are liberated upon the splitting of H 20 2 by peroxidases such as catalase. MIC of nonoxynol-9. Dilutions of nonoxynol-9 were made in MRS broth (pH 6.5) from 0.1 % to 25% (wt/vol]. The pH of the nonoxynol-9-supplemented medium was adjusted to the same level as un supplemented medium when necessary. Lactobacillus isolates were subcultured three times before testing in the nonoxynol-9 broths. The cultures were washed three times in PBS, pH 7.1, and resuspended to a concentration of 107 cells/ ml. Aliquots of 50 JLI were added to test tubes, in triplicate, containing 3 ml of either medium alone (control) or medium plus nonoxynol-9, mixed thoroughly, and incubated for 18 h in an atmosphere of 5% CO 2 at 37°C. The tubes were scored for growth and the MIC was recorded as the lowest concentration demonstrating no growth. Aliquots (50 JLI) were transferred from nonoxynol-9-MRS tubes showing no growth to 3 ml of fresh MRS, incubated for 18 h at 37°C, and examined visually for turbidity. Statisticalanalysis. All testing for significance was done by X2 analysis using 2 X 2 contingency tables with I dj Results
Lactobacillus species identification. Only facultative anaerobic lactobacilli were isolated. Species are shown in table I. Sixty-eight lactobacilli were recovered from swabs from 63 patients. Two distinct lactobacilli were coisolated from five swabs. No lactobacilli were recovered from 3.2% of women. Of the lactobacilli, L. jensenii was the most frequently recovered followed by L. casei and L. acidophilus. L. gasseri, L. plantarum, and L. brevis were also isolated. H 202 production by lactobacilli. H 20 2 production by lactobacilli is shown in table 1. A total of 74.6% of women harbored H 2 0 2-producing lactobacilli as the predominant lactobacillus in their vaginal flora. Some 22.2% ofthe women had non-H 2 0 2-producing lactobacilli. Both H 2 0 2-producing and -nonproducing lactobacilli were present in 6.3% of the women.
Table 1. Lactobacillus species identified from swabs of 63 women. No. (%) women with species
Species L. jensenii L. acidophilus L. casei L. acidophilus and L. jensenii L. casei and L. jensenii L. gasseri and L. jensenii L. gasseri L. plantarum L. brevis None
NOTE.
-t-, positive; -, negative.
22 (34.9) 13 (20.6) 15 (23.8) 3 (4.8) I (1.6) I (1.6) 3 (4.8) 2 (3.2) I (1.6) 2 (3.2)
100 77
40 33 (67% + and-) + and+ and100 50
o
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
(mean age, 31 years) attending an outpatient clinic for a routine gynecological checkup. Twenty-five patients were excluded from the study because of vaginitis or positive cervical cultures. Of the remaining 63,24% took birth control pills and 76%used no form ofcontraception and had no signs or symptoms ofbacterial, yeast, or trichomonas vaginitis. A clean, unlubricated speculum was inserted into the vagina and the posterior fornix was swabbed using three separate sterile, cotton-tipped swabs. Two of the swabs were placed into Amies clear transport medium. A wet mount was made with the remaining swab and examined immediately under a light microscope. Neither Trichomonas vaginalis nor yeast or clue cells indicative of Gardnerella vaginalis were observed on the wet mount examination or swabs processed by the Toronto Hospital Microbiology Laboratory. All were negative for pathogenic organisms, including T. vaginalis, yeast, G. vaginalis, and E. coli. Endocervical material was obtained with separate swabs for the isolation of Neisseria gonorrhoeae and Chlamydia trachomatis. Women were excluded from the study if they had received systemic antibiotic therapy or local vaginal antimicrobial therapy within the preceding month and if they had positive cultures for any of the pathogens listed. Isolation oflactobacilli. Swabs were processed within I h of collection. Smears from the swabs were stained (Gram's), examined for the presence oflactobacilli, and plated onto lactobacilli de Mann, Rogosa, Sharpe (MRS) agar. The plates were incubated at 37°C for 18 h in 5%CO 2 , The predominant presumptive lactobacilli, recovered from the plates, were identified and speciated according to published techniques [12]. The lactobacilli were stored at -70°C in MRS broth plus 20%(wt/vol) glycerol. H202 production. H 20 2 production was determined using a modified version of the method described by Eschenbach et al. [II]. The lactobacilli were plated onto MRS agar containing 0.25 mg/rnl tetramethylbenzidine (TMB; Sigma, St. Louis) and 0.01 mg/ml horseradish peroxidase (Sigma). The plates were incubated under anaerobic conditions at 37°C for 2-3 days, after which the isolates were exposed to ambient air. The horseradish peroxidase in the medium oxidizes the TMB in the presence of H 20 2 to form a blue pigment in and surrounding an H 20 2-producing colony. Because peroxidase is unstable, the plates were used within 3 days of preparation.
1143
1144
Concise Communications
Catalase production.
Discussion This study concentrated on facultatively anaerobic Lactobacillus species, which predominate in the female genital tract. Strictly anaerobic lactobacilli, which account for only 4% of the lactobacilli isolated from normal, healthy women [11], were not examined. L. jensenii was isolated most frequently, followed by L. acidophilus and L. casei. These results are similar to those reported by Eschenbach et al. [11], who found L. jensenii and L. acidophi/us to be the predominant species present in the vagina of normal, healthy women, and by Giorgi et al. [13], who found L. jensenii and L. gasseri (previously incorporated taxonomically with acidophilus). Eschenbach et al. [11] found that 96% of the facultative lactobacilli isolated from healthy women produced H 20 2 , compared with 74.6% in this study. However, we examined the predominant lactobacilli isolated, and.it is possible that H 20 2-producing lactobacilli were present in more of the women examined but in very low numbers. The women examined in our study were also on average older than those studied by Eschenbach et al. Non-H 20 2 producers predominated in 22.2% of the women. None of the lactobacilli isolated had pseudocatalase activity and would not be expected to break down any H 20 2 generated, producing false-negative results. H 20 2 could be envisaged to build up in excess of the ability ofother organisms to degrade it, leaving it available to inhibit or kill other susceptible vaginal organisms, particularly those lacking catalase. The absence of H 20 2-producing lactobacilli in a group of pregnant women, without bacterial vaginosis, was found to be a significant risk factor for the development of bacterial vaginosis later in the pregnancy [11]. A similar situation may exist for urinary tract infections. It has been suggested that the absence or presence in low numbers oflactobacilli in the
introitus is a predisposing factor in the development of urinary tract infection (14]. This study shows a good correlation between the sensitivity of lactobacilli to the action of nonoxynol-9 and production of H 20 2 . The lactobacilli, which are extremely susceptible to low concentrations of nonoxynol-9, would be expected to be killed in spermicide users thus removing the potentially protective H20 2 producers. The nonoxynol-9-resistant lactobacilli would remain but might not afford protection against urogenital infections because they do not produce H 20 2 •
Acknowledgment We thank Jack Clark of R. W. Johnson Pharmaceutical Research Institute for providing nonoxynol-9. References I. Reid G. Bruce AW. McGroarty lA, Cheng KJ, Costerton lW. A review of bacterial interference in medical and veterinary applications. Clin Microbiol Rev 1990;3:335-44. 2. McGroarty lA, Reid G. Detection of a lactobacillus substance which inhibits Escherichia coli. Can 1 Microbiol 1988;34:974-8. 3. McGroarty lA, Reid G. Inhibition of enterococci by Lactobacillus species in vitro. Microb Ecol Health Dis 1988;1:215-9. 4. Skarin A, Sylwan 1. Vaginal lactobacilli inhibiting growth of Gardnerella vaginalis. Mobi/uncus and other bacterial species cultured from vaginal content of women with bacterial vaginosis. Acta Pathol Microbiol Immunol Scand [B) 1986;94:399-403. 5. Pfau A, Sacks T. The bacterial flora of the vaginal vestibule, urethra and vagina in premenopausal women with recurrent urinary tract infections. 1 Clin MicrobioI1981;126:630-4. 6. Vessey M, Doll R, Peto R. lohnson B. Wiggins P. A long-term followup of women using different methods of contraception-an interim report. 1 Biosoc Sci 1976;8:373-427. 7. Fihn SO. Latham RH. Roberts P. Running K. Stamm WE. Association between diaphragm use and urinary tract infection. lAMA 1985;254:240-5. 8. Hooton TM. Fihn SO, Johnson C, Roberts PL. Stamm WE. Association between bacterial vaginosis and acute cystitis in women using diaphragms. Arch Intern Med 1989;149:1932-6. 9. Watt B. Goldacre Ml, Loudon N. Annat 01, Harris RI. Vessey MP. Prevalence of bacteria in the vagina of normal young women. Br 1 Obstet Gynaecol 1981;88:588-95. 10. McGroarty lA, Chong S, Reid G, Bruce AW. Influence of the spermicidal compound nonoxynol-9 on the growth and adhesion ofurogenital bacteria in vitro. Curr MicrobioI1990;21:219-23. I I. Eschenbach DA, Davick PR, Williams BL. et al. Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis. 1 Clin MicrobioI1989;27:251-6. 12. Holdemand LV, Cato EP. Moore WEe, eds. Anaerobe laboratory manual. Blacksburg, VA: Virgina Polytechnic Institute and State University. 1974. 13. Giorgi A, Torriani S, Dellaglio F, Bo G, Stoia E, Bernuzzi L. Identification of vaginal lactobacilli from asymptomatic women. Microbiologica 1987;10:377-84. 14. Stamey TA, Timothy MM. Studies of introital colonization in women with recurrent urinary infections. I. The role of vaginal pH. 1 Urol 1975;114:261-3.
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
None of the 68 lactobacilli isolated produced any catalase, pseudocatalase, or peroxidase activity. Nonoxynol-9 susceptibility of lactobacilli. A total of 66 lactobacilli were tested; 2 of the original isolates lost viability and their susceptibility to nonoxynol-9 could not be determined. Of 66 isolates, 45 (68.2%) were inhibited by concentrations
Hydrogen Peroxide Production by Lactobacillus Species: Correlation with Susceptibility to the Spermicidal Compound Nonoxynol-9 Jacqueline A. McGroarty, Lisa Tomeczek, Donald G. Pond, Gregor Reid, and Andrew W. Bruce
Departments of Surgery (Division of Urology). Microbiology. and Obstetrics and Gynaecology. University of Toronto and Toronto General Hospital; Department ofMicrobiology and Immunology, University of Western Ontario, London, Canada
Lactobacilli are believed to play a protective role in the urogenital tract, guarding against infection by pathogens by a combination of competitive exclusion and inhibitor production [I]. Lactobacilli sterically hinder the adhesion of pathogens to the urogenital epithelia. Lactobacilli attached to cells block the adhesion of pathogens. Lactobacilli also produce a number of antimicrobial substances such as lactic acid, which is produced from carbohydrates and helps to maintain a low vaginal pH. In addition, many strains produce hydrogen peroxide (HzOz) and antibiotic-like substances that can inhibit a range of pathogens including Escherichia coli [2], enterococci [3], and gardnerellae [4). Women with a history of urinary tract infection have a urogenital flora dominated by pathogens, whereas lactobacilli predominate in healthy, premenopausal women [5]. Women using diaphragms and spermicide as a means of contraception are at an increased risk of acquiring urinary tract infections [6-8] and bacterial vaginosis [8). In addition, E. coli is recovered twice as frequently from the vaginas of diaphragm users, whose vaginal pH is higher, than from women using other forms of contraception [7-9). It has been suggested that this is due, in part, to spermicidal compounds altering the vaginal flora. Nonoxynol-9 is the active compo-
Received 17 April 1991; revised 30 December 1991. Informed consent was obtained from all subjects. Grant support: Medical Research Council ofCanada. Ontario Ministry of Health. and Sir Jules Thorn Charitable Trust. Reprints or correspondence (present address): Dr. J. A. McGroarty. Department of Microbiology, University of Glasgow. Glasgow G 12 8QQ, UK. The Journal of Infectious Diseases 1992;165:1142-4 © 1992 by The University of Chicago. All rights reserved. 0022-1899/92/6506-0027$01.00
nent of many spermicidal preparations. It is generally used at 5% concentration in creams and at 12.5% in spermidical foams and would be present at concentrations of 1.6%-2.7% after sexual intercourse. It has been shown to inhibit most lactobacilli at concentrations of 0.1%-1.0% (wt/vol) in vitro, but it has no effect on the growth of uropathogenic bacteria, including E. coli [10). It is feasible that the presence of nonoxynol-9 could upset the ecologic balance of the vagina by inhibiting the protective lactobacilli and allowing overgrowth ofpathogenic microorganisms. However, an examination ofthe vaginal bacterial flora by Watt et al. [9] found that the frequency of isolation of lactobacilli was similar in diaphragm users and nonusers. They did not quantity the lactobacilli present. Since vaginal pH is higher in spermicide users [7), it is possible that lactobacilli numbers are reduced. Alternatively, it may be that only non protective lactobacilli survive the presence of the spermicide. Eschenbach et al. [II) found HzOz-producing facultative Lactobacillus species in the vaginas of 96%of normal women but in only 6%of women with bacterial vaginosis. They concluded that the production of HzOz by Lactobacillus species may represent a nonspecific antimicrobial defense mechanism of normal vaginal ecosystems. We sought to determine the frequency of isolation of HzOz-producing lactobacilli in normal, healthy women and to examine the lactobacilli for a correlation between HzOz production and susceptibility to nonoxynol-9.
Materials and Methods
Patient population. Vaginal swabs were obtained from 88 nonpregnant, healthy, premenopausal, sexually active women
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
Facultative anaerobic lactobacilli were recovered from the vaginas of 96.8% of 63 nonpregnant, healthy, premenopausal women. The predominant specieswere Lactobacillusjensen ii, Lactobacillus acidophilus, and Lactobacillus casei. Of the women, 74.6% had hydrogen peroxideproducing lactobacilli, 22.2% had non-hydrogen peroxide-producing lactobacilli, and 3.2% had no lactobacilli. None of the 68 isolates had catalase activity. Some 68.2% of the isolates were inhibited by concentrations of ~ 1% (wt/vol) of nonoxynol-9 (bactericidal for 73.3% of isolates, bacteriostatic for 26.7%).The remaining 31.8% could grow in all concentrations to 25% (wt/vol) of nonoxynol-9. All of the lactobacilli that were sensitive to nonoxynol-9 produced hydrogen peroxide whereas only 3 of 21 resistant strains were hydrogen peroxide producers. A significant correlation (P < .001, x 2 test) was found between hydrogen peroxide production and sensitivity to nonoxynol-9. It is suggested that the vaginal flora of spermicide users could be depleted of hydrogen peroxide-producing lactobacilli, possibly increasing susceptibility to urogenital infection.
JID 1992;165 (June)
Concise Communications
Catalase assay. The lactobacilli were tested for catalase activity by running a 3%(vol/vol) solution ofH 202 down an MRS agar slant. The agar slants containing lactobacilli growth were observed for bubbles of oxygen that are liberated upon the splitting of H 20 2 by peroxidases such as catalase. MIC of nonoxynol-9. Dilutions of nonoxynol-9 were made in MRS broth (pH 6.5) from 0.1 % to 25% (wt/vol]. The pH of the nonoxynol-9-supplemented medium was adjusted to the same level as un supplemented medium when necessary. Lactobacillus isolates were subcultured three times before testing in the nonoxynol-9 broths. The cultures were washed three times in PBS, pH 7.1, and resuspended to a concentration of 107 cells/ ml. Aliquots of 50 JLI were added to test tubes, in triplicate, containing 3 ml of either medium alone (control) or medium plus nonoxynol-9, mixed thoroughly, and incubated for 18 h in an atmosphere of 5% CO 2 at 37°C. The tubes were scored for growth and the MIC was recorded as the lowest concentration demonstrating no growth. Aliquots (50 JLI) were transferred from nonoxynol-9-MRS tubes showing no growth to 3 ml of fresh MRS, incubated for 18 h at 37°C, and examined visually for turbidity. Statisticalanalysis. All testing for significance was done by X2 analysis using 2 X 2 contingency tables with I dj Results
Lactobacillus species identification. Only facultative anaerobic lactobacilli were isolated. Species are shown in table I. Sixty-eight lactobacilli were recovered from swabs from 63 patients. Two distinct lactobacilli were coisolated from five swabs. No lactobacilli were recovered from 3.2% of women. Of the lactobacilli, L. jensenii was the most frequently recovered followed by L. casei and L. acidophilus. L. gasseri, L. plantarum, and L. brevis were also isolated. H 202 production by lactobacilli. H 20 2 production by lactobacilli is shown in table 1. A total of 74.6% of women harbored H 2 0 2-producing lactobacilli as the predominant lactobacillus in their vaginal flora. Some 22.2% ofthe women had non-H 2 0 2-producing lactobacilli. Both H 2 0 2-producing and -nonproducing lactobacilli were present in 6.3% of the women.
Table 1. Lactobacillus species identified from swabs of 63 women. No. (%) women with species
Species L. jensenii L. acidophilus L. casei L. acidophilus and L. jensenii L. casei and L. jensenii L. gasseri and L. jensenii L. gasseri L. plantarum L. brevis None
NOTE.
-t-, positive; -, negative.
22 (34.9) 13 (20.6) 15 (23.8) 3 (4.8) I (1.6) I (1.6) 3 (4.8) 2 (3.2) I (1.6) 2 (3.2)
100 77
40 33 (67% + and-) + and+ and100 50
o
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
(mean age, 31 years) attending an outpatient clinic for a routine gynecological checkup. Twenty-five patients were excluded from the study because of vaginitis or positive cervical cultures. Of the remaining 63,24% took birth control pills and 76%used no form ofcontraception and had no signs or symptoms ofbacterial, yeast, or trichomonas vaginitis. A clean, unlubricated speculum was inserted into the vagina and the posterior fornix was swabbed using three separate sterile, cotton-tipped swabs. Two of the swabs were placed into Amies clear transport medium. A wet mount was made with the remaining swab and examined immediately under a light microscope. Neither Trichomonas vaginalis nor yeast or clue cells indicative of Gardnerella vaginalis were observed on the wet mount examination or swabs processed by the Toronto Hospital Microbiology Laboratory. All were negative for pathogenic organisms, including T. vaginalis, yeast, G. vaginalis, and E. coli. Endocervical material was obtained with separate swabs for the isolation of Neisseria gonorrhoeae and Chlamydia trachomatis. Women were excluded from the study if they had received systemic antibiotic therapy or local vaginal antimicrobial therapy within the preceding month and if they had positive cultures for any of the pathogens listed. Isolation oflactobacilli. Swabs were processed within I h of collection. Smears from the swabs were stained (Gram's), examined for the presence oflactobacilli, and plated onto lactobacilli de Mann, Rogosa, Sharpe (MRS) agar. The plates were incubated at 37°C for 18 h in 5%CO 2 , The predominant presumptive lactobacilli, recovered from the plates, were identified and speciated according to published techniques [12]. The lactobacilli were stored at -70°C in MRS broth plus 20%(wt/vol) glycerol. H202 production. H 20 2 production was determined using a modified version of the method described by Eschenbach et al. [II]. The lactobacilli were plated onto MRS agar containing 0.25 mg/rnl tetramethylbenzidine (TMB; Sigma, St. Louis) and 0.01 mg/ml horseradish peroxidase (Sigma). The plates were incubated under anaerobic conditions at 37°C for 2-3 days, after which the isolates were exposed to ambient air. The horseradish peroxidase in the medium oxidizes the TMB in the presence of H 20 2 to form a blue pigment in and surrounding an H 20 2-producing colony. Because peroxidase is unstable, the plates were used within 3 days of preparation.
1143
1144
Concise Communications
Catalase production.
Discussion This study concentrated on facultatively anaerobic Lactobacillus species, which predominate in the female genital tract. Strictly anaerobic lactobacilli, which account for only 4% of the lactobacilli isolated from normal, healthy women [11], were not examined. L. jensenii was isolated most frequently, followed by L. acidophilus and L. casei. These results are similar to those reported by Eschenbach et al. [11], who found L. jensenii and L. acidophi/us to be the predominant species present in the vagina of normal, healthy women, and by Giorgi et al. [13], who found L. jensenii and L. gasseri (previously incorporated taxonomically with acidophilus). Eschenbach et al. [11] found that 96% of the facultative lactobacilli isolated from healthy women produced H 20 2 , compared with 74.6% in this study. However, we examined the predominant lactobacilli isolated, and.it is possible that H 20 2-producing lactobacilli were present in more of the women examined but in very low numbers. The women examined in our study were also on average older than those studied by Eschenbach et al. Non-H 20 2 producers predominated in 22.2% of the women. None of the lactobacilli isolated had pseudocatalase activity and would not be expected to break down any H 20 2 generated, producing false-negative results. H 20 2 could be envisaged to build up in excess of the ability ofother organisms to degrade it, leaving it available to inhibit or kill other susceptible vaginal organisms, particularly those lacking catalase. The absence of H 20 2-producing lactobacilli in a group of pregnant women, without bacterial vaginosis, was found to be a significant risk factor for the development of bacterial vaginosis later in the pregnancy [11]. A similar situation may exist for urinary tract infections. It has been suggested that the absence or presence in low numbers oflactobacilli in the
introitus is a predisposing factor in the development of urinary tract infection (14]. This study shows a good correlation between the sensitivity of lactobacilli to the action of nonoxynol-9 and production of H 20 2 . The lactobacilli, which are extremely susceptible to low concentrations of nonoxynol-9, would be expected to be killed in spermicide users thus removing the potentially protective H20 2 producers. The nonoxynol-9-resistant lactobacilli would remain but might not afford protection against urogenital infections because they do not produce H 20 2 •
Acknowledgment We thank Jack Clark of R. W. Johnson Pharmaceutical Research Institute for providing nonoxynol-9. References I. Reid G. Bruce AW. McGroarty lA, Cheng KJ, Costerton lW. A review of bacterial interference in medical and veterinary applications. Clin Microbiol Rev 1990;3:335-44. 2. McGroarty lA, Reid G. Detection of a lactobacillus substance which inhibits Escherichia coli. Can 1 Microbiol 1988;34:974-8. 3. McGroarty lA, Reid G. Inhibition of enterococci by Lactobacillus species in vitro. Microb Ecol Health Dis 1988;1:215-9. 4. Skarin A, Sylwan 1. Vaginal lactobacilli inhibiting growth of Gardnerella vaginalis. Mobi/uncus and other bacterial species cultured from vaginal content of women with bacterial vaginosis. Acta Pathol Microbiol Immunol Scand [B) 1986;94:399-403. 5. Pfau A, Sacks T. The bacterial flora of the vaginal vestibule, urethra and vagina in premenopausal women with recurrent urinary tract infections. 1 Clin MicrobioI1981;126:630-4. 6. Vessey M, Doll R, Peto R. lohnson B. Wiggins P. A long-term followup of women using different methods of contraception-an interim report. 1 Biosoc Sci 1976;8:373-427. 7. Fihn SO. Latham RH. Roberts P. Running K. Stamm WE. Association between diaphragm use and urinary tract infection. lAMA 1985;254:240-5. 8. Hooton TM. Fihn SO, Johnson C, Roberts PL. Stamm WE. Association between bacterial vaginosis and acute cystitis in women using diaphragms. Arch Intern Med 1989;149:1932-6. 9. Watt B. Goldacre Ml, Loudon N. Annat 01, Harris RI. Vessey MP. Prevalence of bacteria in the vagina of normal young women. Br 1 Obstet Gynaecol 1981;88:588-95. 10. McGroarty lA, Chong S, Reid G, Bruce AW. Influence of the spermicidal compound nonoxynol-9 on the growth and adhesion ofurogenital bacteria in vitro. Curr MicrobioI1990;21:219-23. I I. Eschenbach DA, Davick PR, Williams BL. et al. Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis. 1 Clin MicrobioI1989;27:251-6. 12. Holdemand LV, Cato EP. Moore WEe, eds. Anaerobe laboratory manual. Blacksburg, VA: Virgina Polytechnic Institute and State University. 1974. 13. Giorgi A, Torriani S, Dellaglio F, Bo G, Stoia E, Bernuzzi L. Identification of vaginal lactobacilli from asymptomatic women. Microbiologica 1987;10:377-84. 14. Stamey TA, Timothy MM. Studies of introital colonization in women with recurrent urinary infections. I. The role of vaginal pH. 1 Urol 1975;114:261-3.
Downloaded from http://jid.oxfordjournals.org/ at Indiana University Library on June 24, 2014
None of the 68 lactobacilli isolated produced any catalase, pseudocatalase, or peroxidase activity. Nonoxynol-9 susceptibility of lactobacilli. A total of 66 lactobacilli were tested; 2 of the original isolates lost viability and their susceptibility to nonoxynol-9 could not be determined. Of 66 isolates, 45 (68.2%) were inhibited by concentrations
