Relationship between MIC and Minimum Sterol 14{alpha}-Demethylation-Inhibitory Concentration as a Factor in Evaluating Activities of Azoles against Various Fungal Species

The minimum growth-inhibitory concentrations (MICs) of azoleantifungals were compared to their minimum sterol 14 ${alpha}$ -demethylation-inhibitoryconcentrations (MDICs) for clinical fungal isolates. The ascomycetousCandida yeasts tested were clearly divided into two groups:group I, consisting of C. albicans, C. tropicalis, and C. lusitaniae,had MICs that were much higher than the MDICs, whereas groupII, comprising C. glabrata, C. parapsilosis, C. guilliermondii,and C. krusei, had MICs that were approximately equal to theMDICs. In the ascomycetous fungi Aspergillus fumigatus and Sporothrixschenckii, the MICs were indistinguishable from the MDICs. Inthe basidiomycetous fungi Cryptococcus (Filobasidiella) neoformans,C. curvatus, and Trichosporon asahii, the MICs and the MDICswere practically identical. These results support the notionthat there are two distinct classes of fungi differing in theirdegree of tolerance to sterol 14 ${alpha}$ -demethylation deficiency. Thesefindings have significant implications for both fungal physiologyand antifungal chemotherapy.

INTRODUCTION

Top

Introduction

Azole antifungal agents are known to inhibit the activity ofcytochrome P450_14DM, which catalyzes the 14 ${alpha}$ -demethylation ofsterols in ergosterol biosynthesis (21). As a consequence, cellstreated with those drugs accumulate 14 ${alpha}$ -methylated sterols intheir membranes in place of ergosterol. A notable feature ofthis change in sterol composition is that it may or may notinhibit cell growth depending on the fungus involved. Thus,our previous work has demonstrated that under conditions of14 ${alpha}$ -demethylation inhibition, cells of Candida albicans, C. tropicalis,C. guilliermondii, and C. kefyr can proliferate in vitro, whilethose of C. krusei, C. glabrata, and C. parapsilosis cannot(18). Unfortunately, however, it has remained unclear whetherthis characteristic varies within each species due to the smallnumber of strains examined.

In vivo data, however, indicate that azoles are therapeuticallyeffective in C. albicans infection due to the inhibition of14 ${alpha}$ -demethylation (11). This could be because 14 ${alpha}$ -demethylation-deficientcells are more vulnerable than normal cells to killing by reactiveoxygen species (16) or phagocytes (3, 4, 5, 7, 20).

The most important implication of the above-mentioned findingsis that the minimum sterol 14 ${alpha}$ -demethylation-inhibitory concentration(MDIC), rather than the minimum growth-inhibitory concentration(MIC), may be of primary importance for the clinical use ofazole drugs. Theoretically at least, if the MIC of a drug ishigher than its MDIC for a certain fungus, the drug may suppressinfections it causes even at a serum concentration below theMIC (but above the MDIC) as long as host defenses are unimpaired.It is therefore necessary to collect more information aboutthe relationship between MIC and MDIC. In the present study,we provide such data for a number of clinical isolates, includingnot only Candida species but also other ascomycetous and basidiomycetousfungi.

MATERIALS AND METHODS

Top

Materials and Methods

Fungal strains. Ninety-five clinical isolates of yeast-like fungi, recoveredfrom the bloodstream of patients with fungal infections, consistedof 26 strains of C. albicans, 17 strains of C. tropicalis, 8strains of C. glabrata, 10 strains of C. parapsilosis, 21 strainsof C. guilliermondii, 9 strains of C. krusei (of which 4 arelaboratory strains), 2 strains of C. lusitaniae (1), 2 strainsof Cryptococcus (Filobasidiella) neoformans, 2 strains of Cryptococcuscurvatus (8), and 2 strains of Trichosporon asahii (19). Theseisolates were identified presumptively by colony morphologyon CHROMagar Candida (Becton Dickinson), chlamydospore formationon cornmeal agar, and sugar assimilation patterns with the APIID32C system (BioMerieux SA, Marcy-l'Etoile, France). Speciesidentification was confirmed by direct sequencing of the D1-D2region of 26S rRNA (6). Five strains of Aspergillus fumigatusisolated from animals were provided by Rui Kano of the Schoolof Veterinary Medicine, Nippon University. Sporothrix schenckiiIFM41598, originally isolated from a patient with sporothrichosis,was provided by Kazuko Nishimura of Chiba University.

MIC and MDIC determination. We have previously devised a simple method capable of measuringthe MDIC of an azole drug for those fungi whose viability isunaffected by sterol 14 ${alpha}$ -demethylation deficiency, e.g., C. albicans(17, 18). The method is based on the fact that 14 ${alpha}$ -demethylationdeficiency, whether caused by an azole or another mechanism,makes the fungal cells susceptible to growth inhibition by acetateadded to the growth medium, probably due to an increase in permeabilityof the cell membrane (17). Thus, for fungi of this class theMIC of an azole as determined in acetate-supplemented medium(MIC_Ac) is lower than its MIC measured in acetate-free mediumand is similar to its MDIC, as estimated by sterol profilingwith thin-layer chromatography; hence, the relationship wasMIC > MIC_Ac = MDIC (18). In contrast, those fungi which areunable to tolerate 14 ${alpha}$ -demethylation deficiency, e.g., C. krusei,are characterized by the formula MIC = MIC_Ac = MDIC (18). Incidentally,acetate has no effect on sterol demethylation or growth rateat the concentration used. In practice, the determination ofMIC and MDIC was carried out as follows. Twofold dilutions ofthe test drug were made in yeast extract-peptone-glucose (YEPG)medium consisting of 1% yeast extract, 2% polypeptone, and 2%glucose (for MIC^Y, which means MIC measured in YEPG) or YEPGsupplemented with 0.24 M sodium acetate (YEPG-Ac) (for MDIC).A 0.1-ml portion of each dilution was mixed in a microplatewell with 0.1 ml of a cell suspension in the same medium containingapproximately 10⁵ CFU. The plates were incubated at 37°Cfor 2 days with shaking. The lowest drug concentration thatgave a prominent decrease in turbidity was taken to representthe MIC^Y or MDIC of the drug. It should be noted that the MIC^Yobtained in this way is not equivalent to the MIC determinedby the routine method described by the National Committee forClinical Laboratory Standards (9). In the case of S. schenckii,the high susceptibility of this fungus to acetate made the useof the above method for MDIC determination inadequate. Hence,direct estimation of MDIC by sterol profiling with thin-layerchromatography was carried out as previously described (17).

RESULTS

Top

Results

Candida species. Our previous study demonstrated that the strains of Candidaspecies examined were divided into two groups with respect tothe relationship between MIC and MDIC (18). However, the smallnumber of test strains made it uncertain whether this characteristicwas species specific. In the present work, we studied this dichotomywith a much larger number of strains (Table 1). For strainsof C. albicans and C. tropicalis, the MIC^Y (>200 µg/mlfor fluconazole) was much higher than the MDIC (0.2 to 3.1 µg/ml),as previously reported. In addition, strains of C. lusitaniaewere found to belong to this group. On the other hand, the MIC^Ywas essentially equal to the MDIC for strains of C. glabrata,C. parapsilosis, and C. krusei, in agreement with the previousresults (18). Only C. guilliermondii exhibited an intraspecificheterogeneity: strain IFO0454, used in the previous studies,is a member of the former group, while the 21 strains testedin this work were found to belong to the latter group. Theseresults strongly suggest that intraspecies variation in thisMIC-MDIC characteristic is rare. In other words, this characteristiccan be regarded as species specific for Candida species.

View this table:
[in this window]
[in a new window]
TABLE 1. MIC^Ys and MDICs of azoles for various fungi^a

Basidiomycetous yeast species. For all strains of the basidiomycetous yeasts C. neoformans,C. curvatus, and Trichosporon asahii tested, the MIC^Y was foundto be comparable to the MDIC (Table 1).

Filamentous fungi. We successfully applied the YEPG-Ac method for MDIC determinationto five strains of the ascomycetous mold A. fumigatus, and wefound for all of them that the MIC^Y was comparable to the MDIC(Table 1). The ascomycetous fungus S. schenckii is dimorphicand grows either as filaments at 25°C or as yeasts at 37°C(10). Because the growth of this fungus turned out to be severelyinhibited by acetate even in the absence of azoles, the useof YEPG-Ac medium for MDIC determination was considered inadequate.We therefore studied growth and sterol composition of cellsof this fungus cultured aerobically in YEPG medium at 25°Cin the presence of various concentrations of ketoconazole. Theazole drug inhibited by more than 80% both cell growth and sterol14 ${alpha}$ -demethylation at a concentration of 1.25 µg/ml andhigher compared to the drug-free control cells (Fig. 1). Thesame results were obtained with yeast cells grown at 37°C(data not shown). Thus, the MIC^Y and MDIC of ketoconazole wereshown to be similar for this organism. In addition, the MIC^Yestimated by optical density was compatible with morphologicalobservations. While cells grew in typically filamentous formwithout ketoconazole, the presence of the drug at 1.25 µg/mlcompletely inhibited their growth.

View larger version (111K):
[in this window]
[in a new window]
FIG. 1. Thin-layer chromatography profiles of sterols from S. schenckii cells grown in the presence of various concentrations of ketoconazole. Cells were grown with shaking in YEPG (see Materials and Methods) at 25°C for 3 days. The cell yields were assessed by optical density, and the cellular lipids were extracted by the method of Bligh and Dyer (2). Numbers below and above each lane are ketoconazole concentration (in micrograms/milliliter) and relative cell yields (in percent), respectively. Thin-layer chromatography analysis was carried out as described previously (12, 14) with the solvent system of diethyl ether/acetic acid/petroleum ether (100:3:97). Identification of sterols: a, 4,4',14-methylated sterols; b, 4,14-methylated sterols; c, ergosterol; d, phospholipids; e, polar sterol (14).

DISCUSSION

Top

Discussion

The results of the present study are consistent with the notionthat fungi can be divided into two types based on their toleranceto deficiency in sterol 14 ${alpha}$ -demethylation. Most, if not all,species appear to be homogeneous with respect to this trait,and the species for which the drug MIC and MDIC are comparableto each other appear to be more common than those for whichthe in vitro MIC is higher than the MDIC. The mechanism underlyingthe difference in the degree of tolerance to 14 ${alpha}$ -methylated sterolsis unknown at present. One possibility is that different degreesof tolerance may be due to the effect of the 14 ${alpha}$ -methyl groupon the function of one or more membrane proteins that interactwith sterols in membranes.

C. albicans and C. tropicalis, for which the MIC^Y is higherthan the MDIC, are both important opportunistic pathogens, frequentlycausing systemic mycoses in compromised hosts. Our previousstudies have shown that C. albicans cells exhibit increasedsensitivity to various antifungal chemicals when their membranescontain 14 ${alpha}$ -methylated sterols in place of ergosterol, probablydue to enhanced membrane permeability (13, 15). This observationimmediately suggests the possibility that a combination of anazole and an antifungal from another class could act synergisticallyon fungal cells at least of this type. It is hoped that thismay open up a new avenue of antifungal chemotherapy.

The values of MDICs for Candida species are similar to MICsobtained by the standard NCCLS RPMI method (9, 18) and correlatewith the effectiveness of antifungal agents in vivo. We aretherefore currently investigating the wider application of thesimple MDIC methodology to the azole susceptibility testingof filamentous fungi, such as Aspergillus spp.

ACKNOWLEDGMENTS

This work was supported by a grant from the Japan Health SciencesFoundation to O.S. and M.N. and Health Science Research Grantsfor Research on Emerging and Re-emerging Infectious Diseases,Ministry of Health, Labor and Welfare of Japan to M.N. and K.K.

Gifts of azole drugs from Pfizer Pharmaceuticals and JanssenResearch Foundation are gratefully acknowledged. We thank KazukoNishimura of Chiba University and Rui Kano of Nippon Universityfor providing us with fungal strains. We would also like tothank Hiroaki Nakayama, emeritus professor of Kyushu University,and Richard D. Cannon of Otago University for scientific discussionand critical reading of the manuscript.

FOOTNOTES

* Corresponding author. Present address: Department of Microbiology, Nihon Pharmaceutical University, 10281 Komuro, Ina-cho, Kita-adachi-gun, Saitama 326-0806, Japan. Phone: 81-48-721-1155. Fax: 81-48-721-6718. E-mail: simo{at}nichiyaku.ac.jp.

REFERENCES

Top