Mycopathologia 108: 173-178, 1989. 9 1989KluwerAcademic Publishers. Printedin Belgium.

173

Fungispecificity of fluconazole against Candida albicans Robert H. Liss & Richard J. Letourneau Department of Surgery, Children's Hospital Medical Center, 300 Longwood Avenue, Boston, MA 02115 and Chemical and Life Sciences Section, Arthur D. Little, Inc., Cambridge, MA 02140, USA Accepted 16 April 1989

Key words: Candida albicans, candidiasis, clotrimazole, fluconazole, mycotic infection

Abstract

Candida albicans is an opportunistic pathogen of human mucosal surfaces. Colonization of oral and vaginal mucosa by this yeast is antagonized by the resident normal bacterial population. However, antibacterial therapy can alter the normal flora to allow fungal cells to attach, grow and invade host tissues. We studied the antimicrobic activity of fluconazole against clinical isolates of oral and vaginal bacteria and Candida albieans in vitro and in vivo by scanning and transmission electron microscopy; we also compared the bactericidal activity of fluconazole with clotrimazole in vitro by microbiologic assay. Fluconazole lysed fungi but did not change the ultrastructure of bacteria. Clotrimazole, but not fluconazole, was bactericidal against lactobacillus and streptococcus, the principal species of the oral and vaginal cavities. We conclude that Candida albicans, but not oral and vaginal bacteria, is susceptible to fluconazole. These observations help explain the antimycotic specificity of fluconazole and its efficacy against candidiasis in humans.

Introduction

The normal flora of the oral and vaginal cavities protect against colonization and subsequent infection of the mucosa by virulent pathogenic species [ 1, 2]. When the resident microbial population is altered, disease-producing organisms can attach to susceptible host tissue, grow, and become invasive or produce toxins [3, 4, 5]. In opportunistic mycotic infections of the oral and vaginal mucosa, the most common fungal pathogen is Candida albieans [6]. The increasing incidence of both oral and vaginal candidiasis is attributed to a number of factors including inadequate chemotherapy of mycotic infections and

widespread antimicrobial treatment with broad spectrum antibiotics [7, 8]. Suboptimal tissue concentrations of antifungal drugs fail to eradicate all yeast in a mycotic infection; broad spectrum antibacterial agents suppress also the normal, nonvirulent, protective resident bacterial population of the oral and vaginal mucosa. In either event, recurrent opportunistic mycoses or bacterial infections often are a consequence [9]. Thus, the ideal antimycotic agent should produce potent activity specific for fungal pathogens and without significant antimicrobic effect against the normal bacterial population protective of the muco s a. Fluconazole, a new antifungal triazole deriva-

174 tive, has been reported to be effective in treating candidiasis [10, 11, 12, 13]. To evaluate the antimycotic specificity of fluconazole in vivo, we used scanning and transmission electron microscopy to study clinical scrapings from patients with oral and vaginal candidiasis treated with fluconazole. We also compared the antimicrobic effects of fluconazole and clotrimazole by microbiologic assay and by electron microscopy in an in vitro model of oral and vaginal candidiasis.

Materials and methods

Microorgan&ms Microorganisms used in the study were a female genital tract isolate of Candida albicans, and clinical isolates of oral and vaginal bacteria including peptostreptococci, peptococci, veillonella, bifidobacterium, lactobacillus, bacteroides, fusobacterium and eubacterium obtained from the Maxwell Finland Laboratory for Infectious Diseases, Boston City Hospital, Growth media For microscopy studies, Candida albicans was cultured in synthetic amino acid medium [ 14] and in Sabouraud's dextrose broth (Difco, Detroit, Michigan). The bacteria were maintained in trypticase soy broth or Muller-Hinton medium. Antimicrobial agents Fluconazole (UK 49898; Lot R17) was supplied by Pfizer Central Research, Sandwich, England, UK. Clotrimazole (Sigma Lot Number 27F-0393) was purchased from Sigma Chemical Company, St. Louis, Missouri. Both drugs were used at a concentration of 100 micrograms per milliliter. Assay of clotrimazole and fluconazole for bactericidal activity Clinical isolates of Streptococcus agalactiae of female genital tract origin (Boston City Hospital), and Lactobacillus acidophilus (ATCC Number 1283) were grown in Eugonic Broth (BBL Number 97424), subcultured onto sheep blood agar (Scott, Number 3000-1200) and incubated over-

night at 35 ~ in 5 ~ CO 2. Replicate cultures were inoculated at 10 3 and 10s colony forming units per milliliter and bacterial viability assayed by standard minimum bactericidal concentration methods (99~o kill) at 0, 18 and 36 hours postinoculation for streptococcus, and at 0, 48, and 96 hours post-inoculation for lactobacillus. Selection of minimum antimycotic concentration Selection of a minimum antimycotic concentration of fluconazole and clotrimazole for the clinical isolate of Candida albicans used in this study was based on the method of Hoeprich and Finn [15], reports in the literature on drug concentration and incubation time required to achieve fungicidal activity with ketoconazole [16], and preliminary range finding experiments in our laboratory. These data led to selection of a minimum antimycotic concentration for both compounds of 100 micrograms per milliliter at an initial inoculum of 10 3 fungal cells per milliliter and an incubation time of 6 hours. Transmission electron microscopy Oral and vaginal scrapings acquired from patients before and after treatment with fluconazole were fixed in 2~o glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.2 at 4 ~ Thereafter, these were postflxed in 2~o buffered osmium tetroxide, washed, dehydrated, infiltrated with epoxy resin and prepared for observation in a Philips EM 201 transmission electron microscope. Scanning electron microscopy To model in vitro the in vivo disposition of the drug-treated polymicrobic flora of the oral and vaginal mucosa, we incubated clinical isolates of Candida and representative bacterial strains with a minimum inhibitory concentration of fluconazole, fixed the cells thereafter in glutaraldehyde and dispersed them in vitro on oral and vaginal tissues fixed in glutaraldehyde. For scanning electron microscopy, oral and vulvovaginal tissue scrapings were coated with Candida albicans and bacteria incubated with fluconazole (100 mcg/ml) for 6 and 48 hours, or with untreated controls.

175 Thereafter, specimens fixed in glutaraldehyde were critical point dried, coated with carbon and gold-palladium, and observed in an AMR 1000A scanning electron microscope. The approximate initial concentrations for the fungal and bacterial cells were 103 and 105 cells/milliliter, respectively.

Results

Transmission electron microscopy of Candida albicans, and oral and vaginal bacteriafrom patients treated withfluconazole Pretreatment transmission electron photomicrographs of human vaginal and oral scrapings revealed oval cells of Candida and a population of diverse bacteria adherent at the mucosal surfaces

(Figs. 1A and 1C). Typical blastoconidial morphology of Candida was seen in the uniform central density and intact cell wall. Exponential growth by the several bacterial strains was evidenced in cross-sections of rods and cocci undergoing division. In contrast to control specimens, Candida in scrapings from patients treated with fluconazole exhibited alterations in cell structure. Lysis and disruption of cell contents were seen in fluconazole-treated specimens (Figs. 1B and 1D). However, the ultrastructure of adjacent bacteria in these specimens was unaffected by fluconazole. Scanning electron microscopy of Candida albicans and oral and vaginal bacteria treated with fluconazole in vitro. Control specimens of Candida albicans and the mixed bacterial population adhered uniformly to

Fig. 1. Transmission electron photomicrographs ofCandida albicans and bacteria growing in vivo in vaginal mucosa pre- (A) and post-treatment (B) and oral mucosa pre- (C) and post-treatment (D). Marked alterations are seen in vivo in the cellular structure offluconazole-treated Candida (B, D, arrows) but not to the adjacent resident bacterial population (*). Magnification: x 10 000.

176

Fig. 2. Scanningelectron photomicrographs of Candida albicans and bacteria in vitro on oral mucosa pre- (A), x 5000) and 4 hrs post-treatment (C), x 7500), and on vaginal mucosa pre- (B), x 5000) and 24 hrs post-treatment (D), x 9000) with fluconazole (100 mcg/ml). Normal candidal blastoconidia, mycelia and bacteria on oral and vaginal mucosa contrast with disrupted C. albicans (arrows) after fluconazole. Bacterial ultrastructure is unchanged.

the oral and vaginal mucosa (Figs. 2A and 2B). Clumps of blastoconidia and mycelial components grew in colonies over the mucosal surface. Germ tubes and long mycelial forms in irregular patterns were observed extended across the surface. Distinct domains of actively growing bacterial cocii and rods were adherent to the mucosal surface adjacent to colonies of Candida. The disposition of Candida and the mixed bacterial population depicted adherence of these microbes to oral and vaginal mucosa in vivo [ 16]. At four hours following incubation with fluconazole (100 mcg/ml), the drug's antimycotic effect was evident (Fig. 2C). The cell walls of most candidal cells in any field of view exhibited

alterations and distortions. These ranged from an initial ruffling of the normal smooth cell wall surface to depressions, cavitation and collapse. In contrast, the mixed bacterial population exposed to fluconazole under identical conditions continued to grow; both cocci and rods showed division figures indicative of logarithmic growth. Twenty-four hours after incubation with fluconazole, only a few crenated ghosts of candidal cells were seen in any field of view (Fig. 2D). In contrast, varieties of actively growing and dividing cocci and rods were evident across the mucosal surface.

177 EXPOSURE TO FLUCONAZOLE AND CLOTRIMAZOLE (lO0,ug/ml each )

Streptococcus aga/achoe

7 %

%

I 0

I ~8

I

36

% Lacfoboc/Y/us ac/doph//us

L3 % ..j

,IL

i 0

I 48

916

HOURS AFTER EXPOSURE

Fig. 3. Exposure to fluconazole and clotrimazole (100 mcg/

ml each).

Assay of clotrimazole and fluconazole for bactericidal activity The killing curves in Fig. 3 demonstrate that clotrimazole was bactericidal against Lactobacillus acidophilus and Streptococcus agalactiae, the principal bacterial species of the oral and vaginal cavities. In contrast, fluconazole had no antimicrobic activity against either bacterial strain.

Discussion

Opportunistic pathogens like Candida albicans are a threat to adhere to, colonize and invade oral and vaginal mucosa in patients whose normal microbiologic flora have been altered by chemotherapy [4, 18, 19, 13]. We observed penetrating lesions of cell walls

and destruction of internal structure in Candida from tissue specimens of human subjects treated with fluconazole; such changes were not seen in pretreatment specimens [14]. Significantly, the ultrastructure and growth of bacteria adjacent to candidal cells in polymicrobic infections of the human oral and vaginal mucosa were not altered by fluconazole. These observations of fluconazole, both in vivo and under controlled conditions in vitro, demonstrate the drug's potent antimycotic activity and correlate with other reports of its effectiveness against fungal infections in humans [11, 20]. The antimycotic specificity offluconazole (i.e., fungispecificity) demonstrated in this study is meaningful in vivo because, under normal physiologic conditions, Candida albicans and the several bacterial species in the flora of the oral and vaginal cavities are in dynamic, stable balance [21]. Any drug-induced change in composition of the bacterial flora can favor opportunistic mycotic infection and disease [22]. Thus, fungispecificity insures targeted therapy against the causative pathogen and does not alter or diminish protective microbial species. Episodes of recurrent candidiasis in humans correlate with the use of broad-spectrum antibiotics. These agents reduce the numbers of normal protective resident bacteria, especially lactobacilli and streptococci. Lactobacilli have been identified as the dominant host defense factor in preventing the establishment of symptomatic vulvovaginal candidiasis and vaginitis by virulent bacterial species [ 1]. These aerobic, grampositive bacilli are the most prevalent microbial species in the vagina of women without vaginitis [23 ]. In our study, dominant facultative gram-positive cocci were modeled by Streptococcus agalactiae. A consequence of reduced numbers of non-virulent bacteria is a decrease in the host's natural resistance to colonization. Our in vitro microbiologic assay demonstrated that fluconazole did not effect the growth of Lactobacillus acidophilus or Streptococcus agalactiae. On the contrary, clotrimazole, another azole used in clinical therapy of fungal infections, was bactericidal against both species. Thus, effective clinical therapy of candidiasis by

178

fluconazole with no concomitant effect on the bacterial population should mitigate against infection. The results presented here demonstrate that the activity of fluconazole is specific for Candida albicans and does not effect the normal protective bacteria of the oral and vaginal cavities. References 1. Cohen MS, Black JR, Proctor RA, Sparling PF. Host defences and the vaginal mucosa. Scan J Urol Nephrol (suppl) 1984; 86: 13-22. 2. Larsen B, Galask RP. Vaginal microbial flora: practical and theoretical relevance. Obstet and Gynecol 1980; 55(5): 100S-113S. 3. Hill GB, Eschenbach DA, Holmes KK. Bacteriology of the vagina. Stand J Urol Nephrol (suppl) 1984; 86: 23-39. 4. Lee JC, King RD. Characterization of Candida albicans adherence to human vaginal epithelial cells in vitro. Infect Immun 1983; 41: 1024-30. 5. Lehre N, Sega Barr-Nea L. In vitro and in vivo adherence of Candida albicans to mucosal surfaces. Ann Microbiol (Paris) 1983; 134B: 293-306. 6. Troke PF, Andrews RJ, Brammer KW, Marriott MS, Richardson, K. Efficacy of UK-49,858 (Fluconazole) against Candida albicans experimental infections. Antimicrob Agents Chemother 1985; 8(6): 815-8. 7. Larsen B, Galask RP. Vaginal microbial flora: composition and influence of host physiology. Annals Int Med 1982; 96(2): 926-30. 8. Hill LVH, Embil JAE. Vaginitis: current microbiologic and clinical concepts. Can Med Assoc J 1986; 134: 321-31. 9. Sobel JD. Epidemiology and pathogenesis of recurrent vulvo-vaginal candidiasis. Am J Obstet Gynecol 1985; 152(7): 924-935. 10. Humphrey MJ, Jevons S, Tarbit MH. Pharmacokinetic evaluation of UK-49,858, a metabolically stable triazole antifungal drug in humans and animals. Antimicrob Agents Chemother 1985; 28(5); 648-53.

11. Saag MS, Dismukes R. Azole antifungal agents: emphasis on new triazoles. Antimicrob Agents Chemother 1988; 32(1): 1-8. 12. Richardson K, Brammer KW, Marriott MS, Troke PF. Activity of UK 49,858, a bis-triazole derivative, against experimental infections with Candida albicans and Trichophyton mentagrophytes. Antimicrob Agents Chemother 1985; 27: 832-835. 13. Lyman CA, Sugar AM, Diamond RD. Comparative activities of UK-49,858 and amphotericin B against Blastomyces dermatitidis infections in mice. Antimicrob Agents Chemother 1986; 29: 161-162. 14. WilbornWH, Montes LF. Scanning electron microscopy of oral lesions in chronic mucocutaneous candidiasis. J Am Med Assoc 1980; 244: 2294-7. 15. Hoeprich PD, Finn PD. Obfuscation of the activity of antifungal antibiotics by culture media. J Inf Dis 1972; 126: 353-61. 16. Bauer TM, Kronsteiner W, Bassler M, Daschner FD. Sensitivitytesting with ketoconazole in an assay containing Candida albicans, human polymorphonuclear leukocytes and serum. Eur J Clin Microbiol 1986; 5: 665-8. 17. Howlett JA, Squier CA. Candida albicans ultrastructure: colonization and invasion of the oral epithelium. Infect Immun 1980; 29: 252-60. 18. King RD, Lee JC, Morris AL. Adherence of Candida albicans and other species to mucosal epithelial cells. Infect Immun 1980; 27: 667-674. 19. Samaranayake LP, McFarlane TW. The adhesion of the yeast Candida albicans to epithelial cells of human origin in vitro. Arch Oral Biol 1981; 26: 815-20. 20. Rogers TE, JN Galgiani. Activity of fluconazole (UK 49,858) and ketoconazole against Candida albicans in vitro and in vivo. Antimicrob Agents Chemother 1986; 30: 418-422. 21. Odds FC. Ecology and epidemiology of Candida species. Zbl Bakt Hyg A 1984; 257: 207-12. 22. Sutter V. Frequency of occurrence and antimicrobial susceptibility of bacterial isolates from the intestinal and female genital tracts. Rev Inf Dis 1983; 5: $84-$89. 23. Eschenbach DA. Vaginal infection. Clin Obstet and Gynecol 1983; 26(1): 186-202.