Enfumafungin Derivative MK-3118 Shows Increased In Vitro Potency against Clinical Echinocandin-Resistant Candida Species and Aspergillus Species Isolates

TEXT

The echinocandins are first-line agents for treating severe invasive fungal infections (IFIs) (1), being fungicidal against yeast and fungistatic against molds. They alter the integrity of the fungal cell wall via the inhibition of the synthesis of the β-(1,3)-glucan, its major component (2). Specifically, echinocandins target the catalytic subunit of the enzymatic complex β-(1,3)-glucan synthase, encoded by the FKS genes. Reduced susceptibility to echinocandins is associated with mutations in two specific regions in the FKS genes known as hot spots (HS) 1 and 2 that lead to clinical failure or poor response to the therapy (3). The three echinocandins approved by the Food and Drug Administration (FDA) for the treatment of IFIs (caspofungin, anidulafungin, and micafungin) are available only in intravenous formulation, which limits their use in the treatment of less-severe infections or as oral step-down agents. Enfumafungin is one among several new fungal triterpenoid glycosides isolated from the fermentation of Hormonema sp. (4) that present potent in vitro antifungal activity by inhibiting the β-(1,3)-glucan synthase (5). Recently, a semisynthetic derivative of enfumafungin, MK-3118 (Fig. 1), which is being evaluated as an oral therapy for fungal infections, was described (6). This new compound showed MIC values of ≤1 μg/ml and ≤0.015 μg/ml against 160 strains of 7 Candida spp. and 40 Aspergillus spp., respectively (7). Moreover, MK-3118 showed promising in vivo efficacy in murine models of candidiasis and aspergillosis (7, 8). To better understand the antifungal efficacy of MK-3118, we evaluated this new compound against a well-characterized panel of echinocandin-resistant (ER) fks mutants derived from patients who failed echinocandin therapy.

• Open in new tab
• Download powerpoint

FIG 1

Structures of enfumafungin and MK-3118, a semisynthetic enfumafungin derivative.

Antifungal susceptibility testing was performed in triplicate for a collection of 95 Candida strains (20 C. albicans, 20 C. glabrata, 2 C. dubliniensis, 15 C. krusei, 19 C. parapsilosis, and 19 C. tropicalis) that included 30 isolates showing an echinocandin resistance (ER) phenotype (caspofungin [CAS] MIC ≥ 0.5 μg/ml) and for a panel of 40 Aspergillus strains (14 A. fumigatus, 10 A. flavus, 10 A. niger, and 6 A. terreus) that included 1 isolate showing an ER phenotype (9) in accordance with the guidelines described in CLSI documents M27-A3 and M38-A2 (10, 11). In the case of the Candida isolates, MICs were also determined in the presence of 50% human serum (Sigma-Aldrich) (from human male blood, type AB) or mouse serum (Millipore) for C. glabrata isolates. C. parapsilosis ATCC 22019 and C. krusei ATCC 6258 were used as quality control strains. Caspofungin and MK-3118 were obtained as standard powders from their manufacturer (Merck & Co. Inc., Rahway, NJ), and stock solutions were prepared by dissolving the compounds in water or 100% dimethyl sulfoxide (MK-3118).

The MIC distributions of the Candida isolates after 48 h of growth at 35°C for CAS and MK-3118 are shown in Table 1. MK-3118 did not show significant differences in MIC values for the wild-type (WT) isolate population, although overall it presented enhanced in vitro efficacy compared to that of CAS for nearly all echinocandin-resistant isolates, especially among the C. albicans and C. glabrata isolates, where the MIC values decreased by 1- to 8-fold and 4- to 32-fold, respectively. C. tropicalis isolates showed a 4-fold decrease in MIC, while the fold change for C. krusei ER isolates was 2 to 4 times lower. Specifically, 50% of the ER isolates of C. albicans showed MIC values for CAS of ≥2 mg/liter whereas 70% of the ER isolates showed an MIC value of ≤0.5 mg/liter for MK-3118 after 48 h of growth (4- to 8-fold change). Moreover, only 30% of the ER strains showed MIC values of ≤0.25 mg/liter for CAS, while 60% were below this level for MK-3118. The decrease in the MIC values was genotype dependent. Thus, prominent mutations conferring modification of Ser 645 within hot spot (HS) 1 of Fks1p for C. albicans showed a 4- to 16-fold reduction in MIC values whereas strains containing modifications at Phe 641 showed results for CAS and MK-3118 that were similar. In addition, 64% of C. glabrata ER strains showed MIC values of ≤0.5 mg/liter for MK-3118 after 48 h of growth whereas all ER isolates showed MIC values of ≥1 mg/liter for CAS. The decrease in the MIC values was not genotype dependent in C. glabrata, as mutations in either the FKS1 gene or the FKS2 gene showed comparable results (8- to 32-fold reduction in both cases) (Table 2). Similar results were described by Pfaller and collaborators (12), who found that, in a cohort of wild-type clinical isolates (without FKS mutations), there was little or no difference in MIC values between MK-3118 and CAS by broth microdilution for C. albicans, C. krusei, C. parapsilosis, and C. tropicalis. The only exception was C. glabrata, where MK-3118 was 8-fold more potent than CAS. Moreover, 71% of clinical isolates harboring mutations in the FKS gene(s) were inhibited by MK-3118 at ≤1 mg/liter (12), which correlates well with our data.

The echinocandins are highly bound to serum proteins, with a rate of 98% reported for caspofungin (13), which alters its antifungal properties. In fact, it has been reported that the addition of 50% of human serum increased caspofungin MICs an average of 2-fold with a range of 1- to 16-fold (14). In order to ascertain if the relative in vitro potency of MK-3118 was affected by serum, 50% (wt/vol) human serum was added to the MIC plates, as previously described (13). In the case of C. glabrata isolates, 50% mouse serum was used because human serum can inhibit the growth of this organism (15). The addition of serum to the plates increased the MIC of MK-3118 an average of 16-fold with a range of 8- to 64-fold, four times higher than the values obtained for CAS (Table 1). The reduced antifungal properties of this compound in the presence of serum suggested that protein binding was having a direct effect on the drug, perhaps by altering its ability to inhibit glucan synthase, as was observed previously for the echinocandins (14, 16).

Abnormal growth morphology was used to establish a minimum effective concentration (MEC) for Aspergillus spp. after 24 h and 48 h of growth at 35°C. MK-3118 and caspofungin were quite active against the four species of filamentous fungi analyzed, including eight A. fumigatus strains with an azole-resistant phenotype. Interestingly, the growth of all Aspergillus isolates was completely inhibited by treatment with high concentrations (8 to 16 μg/ml) of MK-3118. These data are in accord with those of Pfaller et al. (17), who showed that MK-3118 was active against 71 Aspergillus isolates, including 8 itraconazole-resistant isolates (MIC ≥ 4 μg/ml). Since echinocandin-resistant isolates from Aspergillus spp. have rarely been observed, MK-3118 was tested against the only ER strain of A. fumigatus available to date, which presents the amino acid substitution S678P, equivalent to that of the S654P of C. albicans (9). In this isolate, MK-3118 showed prominent increased potency with an MIC that was 133 times less than that of CAS after 24 h of growth (Table 3).

To better assess direct inhibition of MK-3118 on glucan synthase, the kinetic inhibition parameter IC₅₀ (half-maximal inhibitory concentration) was determined for glucan synthases from wild-type and fks-containing strains. Product-entrapped 1,3-β-d-glucan synthase complexes (GS) were extracted from wild-type and ER strains containing fks mutations from C. albicans (1 WT and 3 fks mutant strains), C. glabrata (1 WT and 2 fks mutant strains), and A. fumigatus (1 WT strain and 1 fks mutant strain), as described previously (18, 19). As expected, evaluation of kinetic inhibition of product-entrapped enzymes isolated from ER strains yielded lower IC₅₀s for MK-3118 than for CAS in Candida albicans (3- to 5.5-fold) and Candida glabrata (3.5- to 62-fold) (Fig. 2A and Table 4). An exception was observed for glucan synthase harboring the mutation F641S, as inhibition of GS activity did not exhibit any variation in the percentage of incorporation even after exposure to high doses of the drugs (10 μg/ml); it was not possible to obtain an in-range IC₅₀ for MK-3118 (Fig. 2A). In the case of A. fumigatus, a decrease of 28-fold was detected in the IC₅₀ of a prominent ER strain (Fks1p-S678P) compared to that of the WT for MK-3118 (Fig. 2B and Table 4), indicating a potential advantage of MK-3118 over echinocandin drugs for certain echinocandin-resistant strains.

• Open in new tab
• Download powerpoint

FIG 2

Antifungal inhibition profiles for product-entrapped 1,3-β-glucan synthase enzyme complexes (GS) for caspofungin (CAS) and MK-3118 for wild-type and ER clinical isolates. GS inhibition was assessed by the incorporation of [³H]glucose into radiolabeled product. (A) Candida albicans. (B) Aspergillus fumigatus.

In summary, MK-3118 was highly active on most fks-mediated echinocandin-resistant strains, especially those from C. albicans and C. glabrata. It was also active on Aspergillus spp. at high concentrations of the drug and was active against a highly ER strain. As observed previously with echinocandin drugs, serum shifted the relative efficacy of the new compound MK-3118, which was most effective against echinocandin-resistant strains.