INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The types and frequencies of life-threatening fungal infections have increased dramatically in the past several years as a direct result of the increase in susceptible patients, which in turn is attributable to human immunodeficiency virus infection and iatrogenic immunosuppression. Treatment of deep-seated mycoses is complicated by the limited choice and spectra of existing antifungal agents as well as the toxicity often associated with antifungal chemotherapy.

One of the targets in the development of antifungal agents is cell wall chitin, a polysaccharide that is found in most medically important fungi. Since chitin does not exist in mammalian cells, compounds that inhibit chitin biosynthesis are considered potential antifungal drugs of low toxicity for humans. Nikkomycins are nucleoside-peptide antibiotics that inhibit chitin synthesis in fungi and insects (8, 9, 21). Like polyoxins, they act as competitive analogs of the substrate UDP-N-acetylglucosamine for chitin synthase (5, 19). Lack of chitin in the cell wall eventually leads to osmotic lysis (2).

The antifungal activity of nikkomycin Z, one of the naturally derived nikkomycins, has been described by many investigators (3, 5, 7, 11, 12, 19). In an animal model, nikkomycin Z was reported to prolong the survival of mice infected with Coccidioides immitis and Blastomyces dermatitidis (17). However, the antifungal efficacy of nikkomycin Z in experimental candidiasis was low (8), in spite of inhibition of chitin synthesis in vivo (6).

To increase the therapeutic index of antifungal drugs, one approach is to use these agents in combinations for synergistic interactions. Synergistic interactions of nikkomycin Z with azoles and with other compounds against Candida albicans have been reported by many investigators (15, 16, 20, 28). Combination therapy is of particular importance in infections involving immunocompromised patients where the infecting organisms are often resistant to antifungal therapy.

To explore possible applications of nikkomycin Z for treatment of mycoses, we conducted in vitro susceptibility testing of nikkomycin Z alone and in combination with fluconazole and itraconazole against a wide array of medically important fungi. Additive and synergistic interactions between nikkomycin Z and either fluconazole or itraconazole were observed for C. albicans, Candida parapsilosis, Cryptococcus neoformans, and Coccidioides immitis. Synergism was also observed between nikkomycin Z and itraconazole against Aspergillus fumigatus and Aspergillus flavus.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Organisms and cultural conditions. A total of 110 isolates, listed in Table 1, were tested. All fungi were clinical isolates obtained from the Fungus Testing Laboratory in the University of Texas Health Science Center at San Antonio and from the Fungal Reference Laboratory in the Audie L. Murphy Memorial Veterans Hospital. Yeast isolates were maintained and grown on Sabouraud dextrose agar at 37°C. Filamentous fungi were grown on potato dextrose agar slants at room temperature.

Drugs. Nikkomycin Z was obtained as a powder from Shaman Pharmaceuticals, South San Francisco, Calif. Fluconazole was received as a powder from Pfizer Inc., New York, N.Y.; itraconazole was received from Janssen, Piscataway, N.J. Stocks of nikkomycin Z and fluconazole and subsequent dilutions were prepared in sterile distilled water. A stock of itraconazole was prepared in polyethylene glycol (PEG) 400; subsequent dilutions were made in 50% PEG 400. Drug preparation and subsequent inoculation of yeast isolates were performed according to the macrobroth dilution protocol recommended by the National Committee for Clinical Laboratory Standards, M27-T (22). Aliquots of 50 µl of each drug at a concentration of 20 times the final concentration were dispersed in polystyrene tubes with caps (12 by 75 mm; Becton Dickinson, Lincoln Park, N.J.). In the checkerboard scheme, each tube contained different proportions of each drug tested. With single drug controls, 50 µl of the drug was mixed with 50 µl of sterile water. Drug preparations were stored at 70°C until used.

Antifungal susceptibility testing. Susceptibility tests were carried out in RPMI 1640 medium with L-glutamine, without sodium bicarbonate, and buffered with 0.165 M morpholinepropanesulfonic acid (MOPS). The pH of the medium was adjusted to 6 with 1 M HCl, as nikkomycin Z is more stable under acidic conditions. Yeast inocula (0.9 ml) were standardized spectrophotometrically and further diluted as described previously (22). For most filamentous fungi, hemacytometer conidia counts were used for inoculum standardization. Inocula of Coccidioides immitis were adjusted with a spectrophotometer to 95% optical transmission at 530 nm and further diluted 1/1,000 in medium. The final inoculum was approximately 0.5 × 10³ to 2.5 × 10³ cells/ml for yeast isolates and 1 × 10⁴ conidia/ml for filamentous fungi. Final concentrations of fluconazole after inoculation ranged from 0.125 to 16 µg/ml, those of itraconazole ranged from 0.03 to 4 µg/ml (contained 2.5% PEG 400), and those for nikkomycin Z ranged from 0.5 to 64 µg/ml in twofold increments. Controls included drug-free inocula, blank media, and media containing 2.5% PEG 400 for itraconazole activity.

Definitions. Drug interactions were classified as synergistic, additive, indifferent, or antagonistic. The interaction was defined as synergistic if the MIC of each drug was at least fourfold lower when the drug was used in combination than when it was used alone. Additive interaction was defined as at least a twofold decrease in the MIC of each drug when two drugs were used in combination. The interaction was defined as indifferent when the MIC(s) of one drug (or both drugs) used in combination remained the same as when the drug was used alone. Antagonism was defined as a higher MIC for either drug when it was used in combination than when it was used alone.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

When tested alone, nikkomycin Z MIC₈₀ values ranged from 0.5 to 32 µg/ml for C. albicans (Table 1) and 1 to 4 µg/ml for C. parapsilosis, including azole-resistant isolates of both species. Other Candida species, including strains of C. tropicalis, C. krusei, and C. glabrata, were resistant to nikkomycin Z at the highest concentration tested (64 µg/ml) (Table 1). Two of 15 isolates of Cryptococcus neoformans were susceptible to nikkomycin Z (MIC₈₀s, 0.5 and 32 µg/ml); for the other 13 isolates MIC₈₀s were higher than 64 µg/ml. When nikkomycin Z was used alone against mycelial-phase Coccidioides immitis, its MIC₈₀ was between 1 and 16 µg/ml (Table 1).

In the combinations of nikkomycin Z with azoles, synergistic or additive interaction was observed with most strains of C. albicans (11 of 15 for the nikkomycin Z and fluconazole combination and 15 of 15 for the nikkomycin Z and itraconazole combination based on MIC₁₀₀ results) (Fig. 1A) and all isolates of C. parapsilosis (Fig. 1B). Synergistic or additive interactions of nikkomycin Z and azoles were also observed with some isolates of Cryptococcus neoformans (Table 2). The azole MIC₈₀ for some isolates of C. albicans, C. parapsilosis, and Cryptococcus neoformans was at the lowest concentration tested. Therefore, the interaction (additive or synergistic) could not be determined based on MIC₈₀ results. Consequently, drug interactions with these species were determined based on the MIC₁₀₀ results.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   MICs of nikkomycin Z for 15 C. albicans (A), 4 C. parapsilosis (B), and 10 Coccidioides immitis (C) isolates when nikkomycin Z was used alone ( and ) and in combination with fluconazole () and itraconazole (). Values in parentheses are numbers of isolates. Note that MIC80s and MIC100s were used for Coccidioides immitis and Candida species, respectively.

A striking synergistic interaction was observed between nikkomycin Z and either fluconazole or itraconazole against essentially all isolates of Coccidioidis immitis tested (Table 2 and Fig. 1C). Against Aspergillus species, nikkomycin Z by itself was ineffective in growth inhibition (MIC > 64). However, nikkomycin Z in combination with itraconazole showed marked synergy when the drugs were tested against virtually all isolates of A. fumigatus and A. flavus (Tables 1 and 2). This synergistic interaction of nikkomycin Z and itraconazole was observed in growth inhibition (lowered MIC₈₀s and MIC₁₀₀s) as well as in fungicidal effect (lowered MLCs) (Fig. 2). Similar effects were not observed against Aspergillus niger, Aspergillus versicolor, and Aspergillus nidulans, nor were they observed for the nikkomycin Z and fluconazole combination.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Itraconazole MLCs for 13 isolates of A. fumigatus (A) and 13 isolates of A. flavus (B) when itraconazole was used alone () and in combination with nikkomycin Z (). Values in parentheses are numbers of isolates.

Isolates of Scedosporium, Paecilomyces, Trichoderma, Penicillium, and Fusarium species were resistant to nikkomycin Z (Table 1). The combination of nikkomycin Z and itraconazole resulted in moderate growth inhibition (Table 2). Among the dematiacious fungi tested (Exophiala, Bipolaris, Wangiella, Ochroconi, Phaeoannellomyces, and Cladophialophora spp.), MICs of nikkomycin Z alone were higher than 64 µg/ml except for the one isolate of Ochroconi humicola (Table 1). Moderate additive or synergistic interactions based on MIC₈₀s were recorded when nikkomycin Z and fluconazole were used in combination against one isolate each of Exophiala jeanselmia and Bipolaris sp. An additive interaction was also observed between nikkomycin Z and itraconazole against one isolate of Wangiella dermatitidis (Table 2).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Chitin and glucans are major components of the fungal cell wall and play important roles in maintaining the structural integrity of fungal cells (13, 26). Compounds which interfere with synthesis of the fungal cell wall may result in growth inhibition, analogous to the effect of beta-lactams on bacteria. Nikkomycins have long been known to inhibit chitin synthesis in fungi and insects (3, 9, 11). Among the medically important fungi, Coccidioidis immitis and B. dermatitidis are susceptible to nikkomycin Z both in vitro and in vivo (17, 23). However, other fungi such as Cryptococcus neoformans, Aspergillus spp., and Fusarium spp. have been reported to be resistant to nikkomycins (18). Although isolates of C. albicans are moderately susceptible (6, 15), other Candida species are resistant to nikkomycin Z in vitro (14). The mechanisms for nikkomycin resistance are not clear.

The clinically useful aspect of nikkomycin may come from its additive or synergistic interactions with other antifungal drugs (15, 16, 20, 28). In this study, nikkomycin Z worked synergistically with fluconazole and itraconazole against C. albicans, C. parapsilosis, Cryptococcus neoformans, and Coccidioides immitis. Synergism was also observed between nikkomycin Z and itraconazole against A. fumigatus and A. flavus. Although the mechanism for this positive interaction between nikkomycin and azoles is not known, it is possible that the loss of membrane integrity caused by azoles facilitates the cellular uptake of nikkomycin Z. Alternatively, azoles themselves may interrupt chitin synthesis or precursor transport to the cell wall (20, 24, 27). Since chitin synthase is located at the plasma membrane, it is reasonable to speculate that changes in the cell membrane affect enzyme activity (1).

Nikkomycin Z inhibits the growth of C. albicans in our assay system at concentrations from 0.5 to 32 µg/ml. It is also fungicidal for most (12 of 15) C. albicans isolates tested, with MLCs between 4 and 64 µg/ml (data not shown). Previous reports showed low efficacy of nikkomycin Z for candidiasis in animal models (6, 8), possibly due to low intracellular concentrations of the drug or the low percentage of chitin in the cell wall of C. albicans. Our data confirm the observation made by others of the synergistic effect of nikkomycin Z and azoles in inhibition of C. albicans in vitro and in vivo (16, 20). We also showed a similar effect of the drug combination against C. parapsilosis. The fungicidal nature of nikkomycin Z and its synergistic interaction with azoles indicate possible usefulness of nikkomycin Z in combination therapy for candidiasis.

The efficacy of nikkomycin Z against Coccidioides immitis has been reported (18, 23). In a microdilution assay, the MIC of nikkomycin Z was 0.125 µg/ml for the spherule-endospore phase (17). Our results indicated a higher MIC₈₀ (4.9 µg/ml) for the mycelial phase, although a different method was employed (macrobroth dilution). Combinations of nikkomycin Z and azoles showed synergism in our study, based on growth inhibition of the mycelial phase (Fig. 2C). Given the high content of chitin (5 to 22%) in mycelial Coccidioides immitis (29, 30), the growth inhibition may have been due to blockage of septum formation as a result of reduced chitin synthesis. However, under the conditions of these assays, nikkomycin Z by itself is not fungicidal for most mycelial isolates of Coccidioides immitis at the concentrations tested and it does not readily affect the MLCs of azoles for most isolates.

As previously reported, the MIC of nikkomycin Z for Aspergillus species was greater than 64 µg/ml (18). When it was used in combination with itraconazole, synergistic interaction was observed for most strains of A. fumigatus and A. flavus, resulting in at least a fourfold decrease in the MIC₈₀s of both drugs (Table 1). Nikkomycin Z also dramatically reduced the MLCs of itraconazole for these species (Fig. 2). It is not clear why similar effects were not observed with the fluconazole-nikkomycin Z combination or against other species such as A. nidulans, as chitin and glucans are essential polymers of the cell wall of A. nidulans (4, 31). There may be chemical-structural differences between the plasma membranes and cell walls of species of Aspergillus that are responsible for the observed differences in drug susceptibility.

It is well known that the determination of meaningful MICs of antifungal compounds is difficult, especially for filamentous fungi for which inoculum size is hard to standardize. The interpretation of MICs is further complicated by the frequent lack of in vitro-in vivo correlations (10, 25). While many host factors need to be considered for clinical efficacy of a particular compound, in vitro susceptibility testing remains an invaluable tool for antifungal drug development. Results of our experiments indicate that the combinations of nikkomycin Z and azoles are inhibitory or fungicidal for several clinically important fungi. These drug combinations should be further studied in animal models for in vivo efficacy and toxicity.