INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Of the currently available polyene antifungals, amphotericin B is the only one used systemically. The reason is that these drugs are poorly water soluble and highly toxic. Efforts have been made to increase the maximum daily dose by reformulating amphotericin B in a variety of lipid preparations (1, 4, 11, 14, 18, 22). Three of these are now licensed in the United States. These include amphotericin B in liposomes (AmBisome; Nextar), amphotericin B colloidal dispersion (Amphotec; Sequus), and amphotericin B lipid complex (Abelcet; The Liposome Company). All of these formulations permit higher dosages of amphotericin B, 5 mg/kg of body weight per dose, with minimal to no nephrotoxicity, but they have not eliminated the infusion toxicities of amphotericin B. Alternatively, the polyene molecule has been esterified to increase solubility, but there has been severe central nervous system toxicity alleged to be due to some of these drugs (5, 8, 9). Another drug (BRL49574A) has been shown to be effective but is limited by chemical instability, tending to precipitate after long-term storage (10). Despite these failures, a water-soluble, nontoxic polyene remains a highly desirable alternative to amphotericin B. KY-62 is a candidate water-soluble polyene with a unique structure, shown in Fig. 1. In the present studies, KY-62 was compared with amphotericin B in vitro against Candida albicans and in vivo in mice infected intravenously (i.v.) with clinical isolates of C. albicans.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   Structures of KY-62 and amphotericin B.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Antifungal drugs. Amphotericin B deoxycholate was purchased from Adria Laboratories. KY-62 was synthesized by one of the investigators (S.L.R.) and made available in a powder form. KY-62 was prepared from amphotericin B as indicated in the studies of Yamashita et al. (25). In this reference, KY-62 is labelled amphotericin B conjugate 2. For parenteral administration, drugs were dissolved in sterile water to the desired dose. Flucytosine (ANCOBON; Roche Laboratories, Nutley, N.J.) was dissolved in distilled drinking water. The daily dose was calculated by estimating that 27- to 30-g mice each consumed 4 ml per day. In prior studies, we had determined that the rate of water consumption was relatively constant at 4 to 4.5 ml/mouse/day, up until the last 24 to 48 h of life.

In vitro assay. Five clinical isolates of C. albicans were used. All were clinical isolates obtained from the Fungus Testing Laboratory, University of Texas Health Science Center, San Antonio. They were stored frozen on Sabouraud dextrose agar until utilized for MIC testing. KY-62 was dissolved in water for testing. Testing of MICs at 24, 48, and 72 h was done by the National Committee for Clinical Laboratory Standards macrodilution method, with twofold dilutions downwards from 12.5 µg/ml (15, 17).

Measurement of KY-62 in serum. For each data point, groups of four to five uninfected mice were treated with KY-62 either intraperitoneally (i.p.) or subcutaneously (s.c.) at the indicated dose and bled by cardiac puncture. Their serum was pooled and assayed. Serum samples were frozen at 20°C until thawed for assay. High-pressure liquid chromatography analysis of KY-62 was performed with samples extracted in chloroform-methanol (8:2).

Infection model. Isolate 92-343 was used for infection. This isolate was maintained between studies at 4°C on Sabouraud dextrose agar. Before studies, the isolate was transferred to brain heart infusion broth and grown at 37°C overnight. Inoculum was washed three times in saline, and an aliquot was used for hemacytometer counting. The inoculum was adjusted to an 0.2-ml/mouse volume, and the count of viable organisms (as reported in the text and the tables) was determined by colony count dilutions. Groups of 7 to 10 ICR outbred mice of 25 g were used. Mice were infected i.v. with doses of C. albicans ranging from 2.8 × 10⁵ to 3.2 × 10⁶ CFU/mouse. One day later, treatment was initiated at 0.2 ml/mouse i.v. or i.p. For studies of survival, treatment was continued through day 10 after infection, and mice were observed to 30 days after infection. For studies of tissue burden, mice were treated from day 1 through 6 and were sacrificed 2 days after the last treatment dose for quantitative cultures of the spleens and kidneys. For survival studies, mice appearing moribund were sacrificed and their deaths were recorded as occurring on the next day. Quantitative tissue counts were done by homogenizing the tissue in 2 ml of sterile saline and performing serial 10-fold colony count dilutions.

Statistics. For survival, comparisons were made by the log rank test and Wilcoxon test of life tables. The rank sum test was used for comparison of tissue counts. Overall significance was determined at P < 0.05 for comparison for two groups, with adjustment of P values when more than two groups were compared.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

The in vitro susceptibilities of the five Candida isolates are presented in Table 1. All had a 48-h MIC of 3.03 µg/ml. Concentrations of KY-62 in serum (Table 2) did not reach the MIC after 10 days of 1 mg/kg/day i.p. or 5 mg/kg s.c. twice daily. After a single 30-mg/kg dose, concentrations were still well above the MIC by 8 h after the dose but had fallen below it by 24 h. We did not have enough data points to determine the initial distribution half-life, but the terminal half-life of KY-62 in the mouse was approximately 12 h.

For studies of efficacy, groups of 10 immunocompetent mice were infected i.v. with 7.3 × 10⁵ CFU of C. albicans per mouse and then treated with either KY-62 at 1, 5, 10, 20, or 30 mg/kg/day i.p. or amphotericin B given at 1 mg/kg/day i.v. or 5 mg/kg/day i.p. As shown in Table 3, by 30 days after infection mice treated with 1 mg of amphotericin B per kg per day had 90% survival, and those treated with 5 mg/kg/day had 100% survival. For KY-62, survival was similarly and significantly prolonged for all doses from 1 up through 30 mg/kg/day. The study was repeated with amphotericin B at 1 mg/kg/day and KY-62 at 30 mg/kg/day i.p., and results were similar. There was 90% survival of treated mice at day 30 versus 100% deaths by day 7 for controls (data not shown). A companion study, with treatment from day 1 to 6, and sacrifice for tissue counts on day 8, is presented in Table 4. Both KY-62 (30 mg/kg) and amphotericin B (1 mg/kg) sharply reduced spleen counts more than 2 logs beneath the lowest count for control mice. Kidney counts in both treatment groups were over 1 log less than that of controls.

In a subsequent study comparing routes of treatment, groups of 9 to 10 neutropenic mice were infected with 1.5 × 10⁶ CFU of C. albicans per mouse. Amphotericin B was given at 1 mg/kg/day i.v. or 5 mg/kg/day i.p. and compared with KY-62 given i.v. at 1, 5, or 10 mg/kg/day. Mean survival of control mice was 7.9 ± 2.6 days. All treated groups survived significantly longer, to between day 26 and day 30, the end of the observation period (data not shown).

Additionally, neutropenic mice were infected with 3 × 10⁵ CFU of C. albicans per mouse. Mice were treated with amphotericin B at 1 mg/kg i.v. or 5 mg/kg i.p. or KY-62 at 1, 5, 20, or 30 mg/kg i.v. or i.p. Results are presented in Table 5. Amphotericin B at 1-mg/kg/dose and KY-62 at 5 and 10 mg/kg/dose were the only groups to achieve significant protection. Two early deaths in the KY-62 group at 30-mg/kg/dose suggested acute toxicity. An additional study explored twice-daily dosing with KY-62. Neutropenic mice were infected with 2.8 × 10⁵ CFU of C. albicans per mouse and treated from day 1 through day 10. Survival of controls was 3.7 ± 0.4 days, survival of the group receiving amphotericin B at 1 mg/kg/day was 17.6 ± 4.5 days, survival of the group receiving KY-62 at 15 mg/kg twice daily was 6.7 ± 0.3 days, and that of the group receiving KY-62 at 30 mg/kg twice daily was 6.3 ± 0.3 days. Although all treatment groups lived significantly longer than controls, only the amphotericin B group had a meaningful extension of survival. Clearly, the benefit of KY-62 is best appreciated in immunocompetent mice.

Finally, the potential benefit of combining KY-62 with flucytosine was explored. Two studies are summarized in Table 6. Although mice receiving combined therapy survived longer than those receiving KY-62, the difference was not significant compared with those receiving flucytosine alone.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Amphotericin B is well known to be a highly effective but highly toxic antifungal polyene. The amphotericin B controls in our studies confirm these characteristics. Our appreciation for the complexity of action(s) of amphotericin B continues to increase. Amphotericin B was thought to act by preferential binding to ergosterol in fungal cell membranes (over cholesterol in mammalian cell membranes) and by destabilizing the integrity of the membrane, permitting uncontrolled leakage of potassium out of the cell and of sodium into the cell (2, 3). More recently appreciated have been the disruptive effects of amphotericin B on the oxidative mechanism within fungal cells (3). Amphotericin B may act by internalization in both plasma membranes and endosomal fractions of mammalian cells and may in high concentrations block fusion of endosomes and lysosomes (21). This may have the consequence of reducing cell protease activity and may have other metabolic consequences (21). Also, amphotericin B has immunological activity in augmenting macrophage production of interleukin-1 and tumor necrosis factor, at least partially augmenting their antifungal activity by enhancement of macrophage nitric oxide and superoxide production (19, 24).

These effects may contribute to both efficacy of amphotericin B against fungi and toxicity of amphotericin B against mammalian cells (12). For mice, the maximal nonlethal dose of amphotericin B is about 1 to 1.5 mg/kg when given i.v. Above this dose, mice die within minutes of injection. Amphotericin B can be given at higher doses i.p. without acute toxicity. As shown in our studies, 1 mg of amphotericin B per kg per day i.v. is effective in prolonging survival of normal and neutropenic mice. At 1 mg/kg/day, amphotericin B also sharply reduces the spleen and kidney counts of C. albicans.

KY-62 is a highly water-soluble analog of amphotericin B and is stable in powder at 4°C for more than 3 months. KY-62 causes no acute clinical toxicity in mice at doses up to 30 mg/kg twice daily. In one study, 30 mg/kg rapidly caused death in 2 of 10 mice. In immunocompetent mice given a lethal infection with C. albicans, KY-62 has potency similar to that amphotericin B and can be safely administered at much larger doses than amphotericin B. Protection is manifested both in prolongation of survival and in marked reduction of spleen and kidney fungal burden.

In addition to demonstrating the efficacy of KY-62, these results are of interest because they highlight the importance of host immune status rather than serum concentration in distinguishing the drugs. In studies of murine cryptococcosis and histoplasmosis, we have found that immunosuppressed (athymic) mice respond less well to antifungals than do immunocompetent mice (6, 7, 23). The present studies suggest that a decreased response to at least some antifungals may be present in neutropenic mice with candidiasis. For our immunocompetent mice, KY-62 is effective at doses as low as 1 mg/kg, even though the concentration in serum does not exceed the MIC of 3.03 µg/ml for C. albicans. However, in neutropenic mice, amphotericin B retains potency while KY-62 markedly decreases in potency. This occurs all the way up to 30 mg/kg, a dose at which the serum concentrations markedly exceed the MIC for the C. albicans infecting isolate for more than 12 h after the dose. Further, flucytosine, which may be additive in effect with amphotericin B, gave no significant increase in survival when combined with KY-62. The reason for decreased potency of KY-62 in immunosuppressed mice is unclear. It is possible that the higher water solubility of KY-62 may reduce both toxicity and penetration into tissue or cell sites of fungal infection, including intracellular cytoplasm. Alternatively, immunomodulatory effects of amphotericin B on macrophages may be stronger than those of KY-62 and may be preserved even in neutropenic mice.

Tumor necrosis factor, one of the cytokines increased by amphotericin B both in vitro and in vivo, is protective in murine candidiasis (13, 20). KY-62 has been designed specifically to reduce the damage and subsequent leakage of ions from mammalian cells, and a consequence of this may be reduced immunomodulatory effects on macrophages (25). Decreased mammalian cell toxicity has been observed for other analogs of amphotericin B, specifically its monomethyl ester derivative (16).

Therefore, the present studies emphasize that in vitro and in vivo activity of antifungal drugs do not always correlate and that immune status may be more important than measurement of levels in serum. At this point, it is unclear whether in vivo antifungal activity of KY-62 or similar drugs will have a similar dependence on neutrophil function in other animal species and in humans. The effect of KY-62 in other mycoses in immunocompetent and immunodeficient animals also requires definition before one may conclude that reduced activity in immunodeficient animals is specific for candidiasis or is more generalizable.