Abstract
Invasive aspergillosis remains a potentially life-threatening infection, the incidence of which is increasing. Current methods used to determine the susceptibilities of Aspergillus strains to antifungal drugs are often unreliable. Nuclear magnetic resonance (NMR) spectroscopy can identify the metabolic complement of microorganisms while monitoring nutrient utilization from the incubation medium. We used 600-MHz 1H NMR spectroscopy to monitor the metabolic responses of five Aspergillus species cultured in RPMI 1640-2% glucose-morpholinepropanesulfonate buffer to various concentrations of the antifungal drugs amphotericin B (AMB) and caspofungin. The metabolic endpoint (MEP) was determined from nutrient and metabolite resonances, measured as a function of the drug concentration, and was defined as a ≥50% reduction in nutrient consumption or metabolite production. MICs were evaluated by a modification of Clinical and Laboratory Standards Institute broth microdilution method M27-A, and minimal effective concentrations (MECs) were determined by microscopic examination of fungal hyphae. For AMB, the MEPs coincided with the MICs. For caspofungin, the MEPs agreed with the MECs for several Aspergillus strains, but the effect of drug pressure was more complex for others. Expansion of the MEP definition to include any significant changes in metabolite production resulted in agreement with the MEC in most cases. Paradoxical metabolic responses were observed for several Aspergillus strains at either high or low caspofungin concentrations and for one Aspergillus terreus strain with AMB. NMR spectroscopy proved to be a powerful tool for detecting the subtle effects of drug pressure on fungal metabolism and has the potential to provide an alternative method for determining the susceptibilities of Aspergillus species to antifungal drugs.
Invasive aspergillosis is the most common filamentous fungal disease in severely immunocompromised patients with hematological malignancies. Mortality rates among all patients with invasive aspergillosis range from 30 to 90% (7, 8, 11). Aspergillus fumigatus accounts for over 90% of the cases of invasive pulmonary aspergillosis, the most common form of the infection, while A. terreus, A. niger, and A. flavus account for the majority of the remainder of the cases (12).
The triazole drug voriconazole has replaced amphotericin B (AMB) as the drug of choice for the treatment of invasive aspergillosis. It has a number of advantages over AMB, that is, fewer toxic side effects (14), the decreased need for other licensed antifungal therapies subsequent to primary treatment (24), a wider spectrum of activity within Aspergillus spp. (A. terreus and A. flavus may be resistant to AMB), and oral availability; but it retains class-specific disadvantages, primarily significant drug interactions with other cytochrome P450 binding agents (12).
Caspofungin is a lipopeptide of the echinocandin group of antifungal agents. It is fungistatic against Aspergillus spp. (10), inhibiting the synthesis of β-(1,3)-d-glucan, an important component of the cell wall, at the growing hyphal tips. Caspofungin has been introduced as salvage therapy for invasive aspergillosis (10). Its selectivity for fungi (there is no known mammalian cell target for this drug) is a significant advantage, explaining its excellent safety profile compared with those of AMB and the triazole drugs.
Drug-induced inhibition of fungal growth is used in the diagnostic laboratory to predict the therapeutic efficacies of antifungal agents. However, existing in vitro assays for determination of the susceptibility of Aspergillus have problems with accuracy and reproducibility (21). Furthermore, the results of in vitro antifungal susceptibility testing may correlate poorly with the therapeutic outcome (31).
Numerous methods have been developed for antifungal susceptibility testing (25, 34). The reproducibility and endpoint determination by the standard broth microdilution susceptibility method for testing of the activity of caspofungin against Aspergillus species are poor (2, 5, 21). This is largely because the activities of lipopeptide drugs are concentrated at the growing hyphal tips, making endpoints difficult to determine by assessment of turbidity or optical density. Kurtz and coworkers overcame this problem to some extent when they observed by microscopic examination that treatment of Aspergillus spp. with two pneumocandin lipopeptides effected profound dose-dependent morphological changes in the fungal hyphae (16). They proposed that a more meaningful susceptibility indicator would be a “minimal effective concentration” (MEC), defined as the drug concentration at which severe growth abnormality occurred. The MEC was found to greatly improve the reproducibilities of susceptibility tests (2, 21). With the use of spectrophotometry and further refinement of the MEC definition as the drug concentration that caused a certain percentage of growth reduction, the subjectivity inherent in determining the MEC can be reduced (5). However, the time required to achieve a result is still dependent on determination of growth and, hence, may take up to 72 to 96 h.
In this study, the in vitro susceptibilities of five Aspergillus species (A. fumigatus, A. terreus, A. flavus, A. niger, and A. nidulans) to caspofungin and AMB were investigated by an alternative approach in which the effects of different concentrations of caspofungin on the production of fungal metabolites and on nutrient utilization from the growth medium were simultaneously and rapidly identified by nuclear magnetic resonance (NMR) spectroscopy.
MATERIALS AND METHODS
Fungal strains.
The provenances of the fungal species used in this study are summarized in Table 1.
TABLE 1.
Provenances of Aspergillus strains used in this study
| Species | Strain | Source |
|---|---|---|
| A. fumigatus | 06-067-1639 | Clinical isolate, Westmead Hospital, Sydney |
| ATCC 204305 | ||
| 04-200167 | Clinical isolate, Westmead Hospital, Sydney | |
| 272-4007 | Clinical isolate, Westmead Hospital, Sydney | |
| A. terreus | 06-059-2428 | Clinical isolate, Westmead Hospital, Sydney |
| 04-188-2548 | Clinical isolate, Westmead Hospital, Sydney | |
| A. niger | 05-290-2479 | Clinical isolate, Westmead Hospital, Sydney |
| 05-112-3701 | Clinical isolate, Westmead Hospital, Sydney | |
| A. nidulans | 05-326-1785 | Clinical isolate, Westmead Hospital, Sydney |
| 00-202661 | Clinical isolate, Women's and Children's Hospital, Adelaide | |
| A. flavus | ATCC 204304 | |
| 05-314-1100 | Clinical isolate, Westmead Hospital, Sydney |
Antifungal agents.
Caspofungin pure powder (Merck Research Laboratories, Rahway, NJ) was dissolved in sterile water. Stock solutions of AMB powder (Apothecon; Bristol-Myers Squibb, Princeton, NJ) were made up in dimethyl sulfoxide (Sigma-Aldrich, St. Louis, MO). Standard RPMI 1640 (Sigma-Aldrich) was supplemented with l-glutamine (0.3 g/liter; Sigma-Aldrich) and glucose (20 g/liter; Merck) and buffered to pH 7 with 3-(N-morpholino)propanesulfonic acid (MOPS; 0.165 M; Sigma-Aldrich). Serial dilutions of drug (0.03125 to 16 μg/ml) in supplemented RPMI 1640 were dispensed in aliquots (100 μl) into 96-well, flat-bottom microtiter plates (Linbro; ICN Biomedicals and Reagents, Sydney, NSW, Australia). The plates were stored at −70°C for a maximum of 3 weeks before use.
Broth microdilution MICs.
Isolates were recovered from stock and cultured by the protocol outlined in the M38-A document (19) of the Clinical and Laboratory Standards Institute (formerly the National Committee for Clinical Laboratory Standards), except that the RPMI 1640 medium was supplemented with 2% glucose rather than 0.2% glucose. The experiments were performed in duplicate. The MIC was defined as the lowest concentration of drug that completely inhibited visible growth. For the caspofungin studies, the MEC was determined at 48 h as the lowest drug concentration resulting in aberrant hyphal growth by examination with an inverted microscope (1).
Broth microdilution samples and NMR spectroscopy.
After incubation for 48 h (see above), the contents of each microtiter well (200 μl) and phosphate-buffered saline-D2O (300 μl; ICN Biomedicals, Aurora, OH) were transferred into 5-mm NMR tubes. Spectra were acquired at a 1H observation frequency of 600.13 MHz and a temperature of 298 K, with D2O as a field-frequency lock. For each sample 32 transients were collected over a spectral width of 6,410 Hz with a relaxation delay of 3 s and an acquisition time of 3 s. A line-broadening factor of 1.0 Hz was applied to all free induction decays prior to Fourier transformation. Chemical shifts were referenced to the α-glucose anomeric proton resonance (δ 5.233) (20). Spectra were phase and baseline corrected prior to analysis. Signal assignment was facilitated by acquisition of a suite of two-dimensional heteronuclear NMR spectra, namely, 1H-13C heteronuclear single-quantum coherence and 1H-13C heteronuclear multibond correlation experiments.
Determination of MEP.
Stack plots of NMR spectra were constructed to show the effect of drug pressure on the utilization of nutrients from the medium and the production of fungal metabolites. Preliminary experiments were conducted on a 400-MHz spectrometer to determine the effect of the fungal inoculum, the glucose concentration (0.2%, 0.5%, and 2%), and the time of incubation (16 h, 24 h, 48 h, and 72 h) by using strain ATCC 204305 and two additional clinical isolates of A. fumigatus (isolates 04-200167 and 272-4007). Reproducible results demonstrating a drug effect on metabolite production were found by 24 h and were more marked at 48 h (data not shown). We chose to determine the metabolic endpoint (MEP) after incubation for 48 h in RPMI 1640 containing 2% glucose. MEPs were determined from spectral peaks as the lowest concentration of drug at which nutrient utilization or the production of certain fungal metabolites was inhibited by ≥50%. Under these conditions, the MEPs for the three A. fumigatus isolates incubated with caspofungin were 0.5 μg ml−1 (ATCC 204305), 0.5 μg ml−1 (04-200167), and 1 μg ml−1 (272-4007).
In the definitive experiments performed on the 600-MHz spectrometer, MEPs were determined by inspection of stack plots of spectral peaks and were the lowest concentration of drug at which nutrient utilization or fungal metabolite production by a given Aspergillus species suppressed the peaks by ≥50%. To obtain a quantifiable endpoint, integrals for selected metabolites for each species were also measured. Clearly resolved resonances surrounded by flat baselines were selected for integration. Integrals were measured for acetate and arginine β-CH2 (1.89 to 1.94 ppm); citrate (2.70 to 2.88 ppm); ethanol CH3 (1.15 to 1.22 ppm); fumarate (6.51 to 6.53 ppm); α-glucose H-1 (5.21 to 5.26 ppm); β-glucose H-2 (3.21 to 3.28 ppm); glutamine γ-CH2 (2.42 to 2.49 ppm); glycine betaine CH3 (3.32 to 3.35 ppm); isoleucine, leucine, and valine overlapping CH3 resonances (0.92 to 1.06 ppm); malate C-H (4.28 to 4.33 ppm); phenylalanine C6H5 (7.32 to 7.35 ppm); succinate (2.39 to 2.42 ppm); and tyrosine H-5,5′ (7.18 to 7.22 ppm).
Analyses were conducted in duplicate, and MEP results were reported as ranges, if they were not identical.
RESULTS
NMR spectrum from A. fumigatus.
A typical NMR spectrum, obtained from a culture of A. fumigatus (ATCC 204305) in RPMI 1640 containing 2% glucose, is shown in Fig. 1. Resonances from components of the growth medium include those from glucose, amino acids (isoleucine, leucine, phenylalanine, tyrosine, and valine), and MOPS buffer. Fungal metabolites include ethanol, fumarate, glycine betaine (N,N,N-trimethylglycine), malate, and succinate. The major metabolites produced and the nutrients consumed by two different isolates of each of the five most common pathogenic species of Aspergillus are summarized in Table 2. A. nidulans grew more slowly than the other Aspergillus species; hence, the spectra were recorded after incubation for 72 h rather than 48 h. The spectra from each strain were reproducible.
FIG. 1.
Overview of a typical 1H NMR spectrum obtained from a culture of A fumigatus (ATCC 204305) showing major nutrients and metabolites. The peaks labeled 1, 2, and 3 are due to the MOPS buffer. The nutrient and metabolite resonances are reported in the Materials and Methods.
TABLE 2.
Nutrient consumption and metabolite production obtained from NMR peak area integrations for each of the Aspergillus strains used in this study
| Species | Strain | Major nutrients consumed | Major metabolites produceda |
|---|---|---|---|
| A. fumigatus | 06-067-1639 | Arg, Gln, Ile, Leu, Phe, Tyr, Val | B, F, M, S |
| ATCC 204305 | Arg, Gln, Ile, Leu, Phe, Tyr, Val | B, F, M, S | |
| A. terreus | 06-059-2428 | Nutrient consumption marginal | E, F, M |
| 04-188-2548 | Nutrient consumption marginal | E, F, M | |
| A. niger | 05-290-2479 | Glc | C, E |
| 05-112-3701 | Glc | C, E | |
| A. nidulans | 05-326-1785 | Nutrient consumption marginal | E, F, M |
| 00-202661 | Nutrient consumption marginal | F, M | |
| A. flavus | ATCC 204304 | Arg, Glc, Gln, Ile, Leu, Phe, Tyr, Val | A, F, M, S |
| 05-314-1100 | Arg, Glc, Gln, Ile, Leu, Phe, Tyr, Val | A, F, M, S |
A, acetate; B, glycine betaine; C, citrate; E, ethanol; F, fumarate; Glc, glucose; M, malate; S, succinate.
Effect of drug pressure on metabolite production and nutrient utilization.
Changes in metabolite production were the most obvious indicator of drug pressure and varied with the different Aspergillus spp. Only the rapidly growing species A. niger consumed large quantities of glucose from the medium. The effect of drug pressure on fungal metabolite production is readily seen in the representative stack plots in Fig. 2, which compare the effects of AMB and caspofungin on A. fumigatus (Fig. 2a and b) and the effects of caspofungin on A. niger (Fig. 2c) and A. flavus (Fig. 2d). The MEPs (indicated by arrows) are clearly defined: 0.5 μg ml−1 for AMB against A. fumigatus (result for the duplicate sample, 0.25 μg ml−1), 0.5 μg ml−1 for caspofungin against A. fumigatus and A. niger, and >16 μg ml−1 for caspofungin against A. flavus. A paradoxical effect on metabolite production at high concentrations of caspofungin is clearly seen in the spectra from A. fumigatus (fumarate, malate, and glycine betaine) and A. niger (glucose and citrate). This effect was also noted with A. fumigatus in the preliminary experiments (data not shown).
FIG. 2.
NMR stack plots (expanded spectral regions) showing changes in fungal metabolite production and nutrient consumption at different concentrations of AMB or caspofungin. Arrows indicate the MEPs for (a) A. fumigatus ATCC 204305 incubated with AMB, (b) A. fumigatus ATCC 204305 incubated with caspofungin and showing a paradoxical increase in metabolite production at higher caspofungin concentrations, (c) A. niger 05-112-3701 incubated with caspofungin (both glucose consumption and citrate production are observed to increase at higher drug concentrations), and (d) A. flavus ATCC 204304 incubated with caspofungin (large increases in fumarate production occur at low caspofungin concentrations, but little effect on malate production is observed). GC, growth control.
Susceptibility curves derived by plotting the drug concentration against peak integrals for selected metabolites produced or nutrients utilized by the 10 Aspergillus strains used in this study (see Materials and Methods) are shown in Fig. 3. A sudden, marked reduction in the production of betaine and fumarate by A. fumigatus strain ATCC 204305 is seen with AMB (MEP, 0.5 μg ml−1) and caspofungin (MEP, 0.5 μg ml−1) at 48 h of incubation (Fig. 3a and b). In 72-h cultures with both drugs, glucose consumption was also evident (data not shown). For A. fumigatus and AMB, the levels of metabolite production were similar in the 48- and 72-h cultures and the MEP was determined to be 1 μg ml−1 (data not shown). Determination of the MEP for caspofungin from the 72-h cultures was less clear-cut than that at 48 h (0.03 μg ml−1 by using malate, but the MEP was not calculable by using the predefined reduction in the peak area of at least 50% for glycine betaine and fumarate [Fig. 3c]). Notably, a paradoxical increase in metabolite production (glycine betaine and fumarate [Fig. 3b and c] and succinate [data not shown]) was seen at caspofungin concentrations of ≥0.5 μg ml−1 (48 h; Fig. 3b) and 1 μg ml−1 (72 h; Fig. 3c), respectively. The effect of AMB on the clinical isolate of A. fumigatus (06-067-1639) was similar to that on the ATCC strain (Fig. 3a). However, with caspofungin, neither glycine betaine (the levels of which paradoxically doubled at caspofungin concentrations ≥1 μg ml−1) nor malate and fumarate (20% reduction at 2 to 4 μg ml−1) were significantly inhibited, suggesting that this isolate is caspofungin resistant. Alternatively, use of the sudden increase in the glycine betaine peak integral to define the MEP resulted in a value of 1 μg ml−1. Spectra from repeat experiments were identical for this strain.
FIG. 3.
Plots of metabolite or nutrient peak areas as a function of the AMB or caspofungin concentration for the 10 Aspergillus strains used in this study. In the case of metabolites, the peak areas are expressed relative to that for the growth control (GC); for nutrients, the peak areas are relative to that for the medium (sterile control). Incubation conditions were 48 h at 35°C unless stated otherwise. A. fumigatus was incubated with AMB (a) or caspofungin (b) for 48 h and caspofungin (c) for 72 h. The effects on fumarate, glycine betaine (a, b, and c) and, in addition, malate (c) are shown. A. terreus was incubated with AMB (d) or caspofungin (e) for 48 h, with effects on the production of ethanol and fumarate (insets) also shown. A. niger was incubated with AMB (f), which shows the complete utilization of glucose (Glc) in the growth control and in the presence of a low concentration of drug and a clear-cut MEP at 0.125 mg ml−1. The effect of caspofungin on glucose utilization and citrate production (inset) is shown (g). A. nidulans incubated with AMB (h) and caspofungin (h, inset) exhibited reduced fumarate production, whereas complex effects of AMB and caspofungin were seen in A. flavus (i).
AMB (Fig. 3d) and caspofungin (Fig. 3e) predominantly affected ethanol production in both isolates of A. terreus (MEPs, 0.06 μg ml−1 and 1 to 2 μg ml−1, respectively, for AMB, and 0.06 μg ml−1 and 0.5 μg ml−1, respectively, for caspofungin). One of the two A. terreus isolates showed a paradoxical increase in ethanol at high caspofungin concentrations.
Glucose consumption was complete by 48 h in cultures of A. niger and provided clear-cut MEPs for both AMB (0.25 to 0.5 and 0.125 μg ml−1 for the respective strains; Fig. 3f) and caspofungin (0.5 to 1 μg ml−1 for both strains; Fig. 3g). High concentrations of caspofungin caused a paradoxical increase in glucose consumption and citrate production in one strain of A. niger (Fig. 3g).
Fumarate was the most useful indicator of the effects of the drugs on A. nidulans (AMB MEPs, 0.5 to 1 μg ml−1 for both isolates; caspofungin MEPs, 4 to 8 μg ml−1 for strain 05-326-1785 and 2 to 4 μg ml−1 for strain 00-202661; Fig. 3h) and A. flavus (Fig. 3i). Complex spectra were obtained with A. flavus, whereby at low concentrations of AMB and caspofungin (clinical strain 05-314-1100) there was an initial twofold increase in fumarate production, which subsequently fell rapidly. With the other strain (ATCC 204304) there was an expected fall in fumarate correlating with an AMB MEP of 2 μg ml−1, whereas with caspofungin there was an initial rise (at 0.03 μg ml−1) which plateaued and remained high at 16 μg ml−1.
Effect of glucose concentration on drug-dependent spectral changes in cultures of A. fumigatus.
The effect of caspofungin was also tested on A. fumigatus ATCC 204305 incubated for 48 h and 72 h in RPMI 1640 medium containing 0.2% glucose instead of 2% glucose. The nutrients consumed and the metabolites produced were identical, except that glucose consumption could be clearly visualized. Glucose peak areas did not change as a function of the caspofungin concentration, in agreement with the marginal decreases in nutrient consumption observed for all caspofungin experiments with this strain. The MEP of the 72-h sample was 0.03 μg ml−1 caspofungin, and for the 48-h sample it was 0.06 μg ml−1 caspofungin (based on malate). The metabolite peak areas for both the 48-h and the 72-h samples increased with increasing drug concentration (data not shown, but they are very similar to those for the 72-h 2.0% glucose sample; Fig. 3c).
Comparison of MEPs with standard MICs.
Table 3 compares the MEP results obtained from analysis of the NMR spectra with the MIC or the MEC results for each of the Aspergillus species. In general, there was a correlation between the MEPs and the MICs or MECs, with the MEPs being two to four times lower for AMB and similar or lower for caspofungin. With three isolates (A. fumigatus clinical strain 06-067-1639, A. flavus ATCC 204304, and A. flavus clinical strain 05-314-1100) the caspofungin MEPs resembled the broth dilution MECs only when the MEP was based on a >50% increase (rather than the predefined decrease) in fumarate production.
TABLE 3.
Comparison of MEPs with MICs for AMB and caspofungin based on NMR peak integrals calculated for specified nutrients and metabolites
| Drug and species | Strain | Time (h) | MEPa (μg/ml) | MIC or MECb (μg/ml) |
|---|---|---|---|---|
| AMB | ||||
| A. fumigatus | 06-067-1639 | 48 | 0.5-1 (B, F, M, Phe, Tyr) | 1 |
| A. fumigatus | ATCC 204305 | 48 | 0.25-0.5 (B, F, M, Phe, Tyr) | 1 |
| A. fumigatus | ATCC 204305 | 72 | 1 | 1 |
| A. terreus | 06-059-2428 | 48 | 0.06 (E) | 1 |
| A. terreus | 04-188-2548 | 48 | 2 (E, F, M) | 2 |
| A. niger | 05-290-2479 | 48 | 0.125 (Glc) | 0.25-0.5 |
| A. niger | 05-112-3701 | 48 | 0.25-0.5 (C, Glc) | 0.5 |
| A. nidulans | 05-326-1785 | 72 | 0.5-1 (F) | 0.5 |
| A. nidulans | 00-202661 | 72 | 0.25 (F) | 1 |
| A. flavus | ATCC 204304 | 48 | 2 (F, M) | 8-16 |
| A. flavus | 05-314-1100 | 48 | 2 (F, M) 0.06 (Fc) | 4-8 |
| Caspofungin | ||||
| A. fumigatus | 06-067-1639 | 48 | 1 (Bc) | 0.03 |
| A. fumigatus | ATCC 204305 | 48 | 0.5 (B, F, M) | 0.03-0.06 |
| A. fumigatus | ATCC 204305 | 72 | 0.03 (M) | 0.03-0.06 |
| A. fumigatus | ATCC 204305 | 72 | 0.03 (M) | 0.03-0.06 |
| A. fumigatus | ATCC 204305 | 48 | 0.06 (M) | 0.03-0.06 |
| A. terreus | 06-059-2428 | 48 | 0.06 (E) | 0.03 |
| A. terreus | 04-188-2548 | 48 | 0.5 (E, M) | 0.03-0.06 |
| A. niger | 05-290-2479 | 48 | 0.25 (C, Glc) | 0.5-1 |
| A. niger | 05-112-3701 | 48 | 0.5-1 (C, Glc) | 0.5-1 |
| A. nidulans | 05-326-1785 | 72 | 4-8 (F) | 4-8 |
| A. nidulans | 00-202661 | 72 | 2-4 (F) | 2-4 |
| A. flavus | ATCC 204304 | 48 | >16 (F, M) 0.03 (Fc) | 0.03 |
| A. flavus | 05-314-1100 | 48 | 4-8 (F, M) 0.03 (Fc) | 0.03 |
A, acetate; B, glycine betaine; C, citrate; E, ethanol; F, fumarate; Glc, glucose; M, malate; Phe, phenylalanine; S, succinate; Tyr, tyrosine.
MICs are for AMB and MECs are for caspofungin. MICs were determined by the CLSI (2% glucose) modified method. The MEC was the lowest concentration resulting in aberrant hyphal growth.
The MEP was based on a >50% increase in metabolite production.
DISCUSSION
The results obtained from this study indicate that NMR spectroscopy generally provides an objective measure of the effects of drugs on a range of medically significant Aspergillus species. Clear interpretations of the effects of drug pressure on metabolite production were nevertheless not possible for all Aspergillus strains studied. The MEP values obtained for AMB were typically 1 to 2 twofold dilutions lower than the MICs for all Aspergillus species, suggesting that NMR spectroscopy can detect metabolic changes earlier than visualization of growth inhibition in broth culture. With caspofungin, agreement between the MEP and the MEC was observed for several Aspergillus strains when the MEP was used as defined (a ≥50% reduction in nutrient consumption or metabolite production). However, for some strains (e.g., A. fumigatus ATCC 204305, A. terreus 04-188-2548, and A. flavus 05-314-1100) the MEPs were 2 to 8 twofold dilutions higher than the MECs. Furthermore, MEP values >16 μg ml−1, indicating probable caspofungin resistance, were observed for A. flavus ATCC 204304 (Table 3).
These discrepancies between the MEP and MICs/MECs may have resulted from differences in the method, the culture medium, and the incubation time. Microdilution MICs have consistently been higher than MECs for a range of Aspergillus species (2, 15, 26). Increasing the incubation times from 24 h to 48 to 72 h increased the MIC (broth dilution) 2 to 10 times relative to the MEC in cultures of A. fumigatus (2, 3). Additionally, growth in 2% glucose-containing RPMI 1640 medium resulted in MICs for Aspergillus species greater than those in 0.2% glucose-containing medium (2). In this study, the MEP for A fumigatus ATCC 204305 was not changed by growth in 2% glucose versus 0.2% glucose or for 48 h versus 72 h and was almost identical to the MEC. Further refinement of the MEP determined by NMR is required, along with testing of other growth media; better growth would reduce the incubation times. The use of RPMI 1640 medium with 2% glucose did not provide a significant growth advantage over the use of RPMI 1640 with 0.2% glucose and prevented the monitoring of glucose consumption in many Aspergillus species in this study.
The definition of the MEP is critical to the nature of the correlation between the MEP and the MEC or MIC. In some strains, an increase in metabolite production was detected at low drug concentrations; e.g., A. flavus strains ATCC 204304 and 05-314-1100 showed increases in the fumarate peak area at 0.03 μg ml−1 caspofungin, which is identical to the MEC. The MEPs were >16 and 4 to 8 μg ml−1, respectively, by using the MEP as it was initially defined (a 50% change in metabolite production or nutrient utilization). The increase in fumarate production represents the response of the fungus to drug pressure, and if the MEP definition is modified to specify any measurable reproducible change in metabolite production, MEP and MEC values coincide in the majority of cases (80%) (Table 3).
Paradoxical increases in metabolite production and nutrient consumption at high drug concentrations.
A paradoxical increase in the production of ethanol, but not that of fumarate, was reproducibly seen at higher concentrations of AMB with one clinical isolate of A. terreus, an observation that has not been reported previously for AMB. Similar effects on this isolate were seen with caspofungin (see below). None of the other nine Aspergillus strains used in this study showed this paradoxical effect with increasing concentrations of AMB.
However, the paradoxical increase in metabolite production and/or nutrient consumption was observed at higher caspofungin concentrations in several Aspergillus species. The production of fumarate, glycine betaine, malate, and succinate by A. fumigatus all increased at higher caspofungin concentrations. In the case of strain 06-067-1639, the glycine betaine peak area doubled at caspofungin concentrations ≥1 μg ml−1. Glycine betaine was not detected in any samples of the other Aspergillus species used in this study.
Glycine betaine functions as an efficient alternative to homocysteine as a methyl donor in methionine biosynthesis; the reaction is catalyzed by betaine homocysteine methyltransferase (4, 18). It also acts as an osmoprotectant in plants, fungi, and bacteria (23, 30). Its accumulation may reflect the response of A. fumigatus to the cell wall damage caused by caspofungin, which, by inhibition of fungal β-(1,3)-d-glucan synthase, adversely affects cell wall integrity and osmotic stability (10). Repression of glycine betaine catabolism results in its accumulation, which does not occur in the absence of osmotic stress (4). Glycine betaine can also act as a reserve form of choline, as was found in sulfur-deficient cultures of the fungus Penicillium fellutanum; choline O-sulfate accumulated in medium containing adequate sulfate (22). Thus, it is possible that the sulfate requirements are increased for the A. fumigatus strains in the presence of caspofungin.
The tricarboxylic acid (TCA) cycle intermediates fumarate and, to a lesser extent, malate increased in A. fumigatus and A. terreus at higher concentrations of caspofungin and in A. flavus at low concentrations of caspofungin. A. flavus has been reported to produce high levels of l-malic acid but not fumaric acid (13). Organic acid intermediates of the oxidative TCA cycle can accumulate to high concentrations in culture media for filamentous fungi such as Aspergillus spp. and Rhizopus spp. under specific stress conditions (13). Fumaric and l-malic acid accumulation occurs via the reductive TCA pathway, which is localized in the cytosol rather than the mitochondria (13). Fumarase catalyzes the reversible hydration of fumaric acid to l-malic acid and must be involved in the accumulation of fumarate. In an attempt to explain the accumulation of fumaric acid in the fungus Rhizoctonia oryza, Goldberg et al. (13) suggested that two genes encoding two different fumarases may exist. One fumarase is present in mitochondria and catalyzes the conversion of fumaric acid to l-malic acid, and another is present in the cytosol and catalyzes the reverse reaction. The cytosolic fumarase is induced under specific stress conditions and leads to the buildup of fumaric acid. Caspofungin would act as a stressor under the conditions used in our study and would cause a similar effect.
Both strains of A. niger showed an increase in glucose consumption and citrate production at high caspofungin concentrations, and a paradoxical increase in ethanol production was observed in A. terreus 06-059-2428. It is of particular interest that similar paradoxical increases in ethanol production were observed in A. terreus 06-059-2428 with both AMB and caspofungin. Although the clinical implication of this observation is not known, A. terreus is less susceptible to AMB than other species, and clinical failures have been reported (17, 29).
Paradoxical growth increases at high caspofungin concentrations have been reported in vitro and in vivo for yeasts and filamentous fungi (9, 27, 32) and have been detected in Candida species by NMR spectroscopy (9). The effect is partly medium dependent; but it is not due to a resistant subpopulation of organisms or to drug decomposition, and it is not specific to echinocandins, nor is it characteristic of echinocandins, since it was not observed with micafungin (27). It has been suggested that calcineurin-mediated and protein kinase C-mediated signaling pathways may be affected by high caspofungin concentrations and may play a role in regulating functionally redundant cellular events important for resistance to caspofungin (33). Recently, Stevens et al. (28) demonstrated that when β-(1,3)-d-glucan polymer synthesis is inhibited by the presence of caspofungin in Candida species, there is a compensatory increase in chitin synthesis. The authors argued that these results provide an explanation for the paradoxical effect. β-Glucans are believed to be covalently linked to chitin, contributing to the structural rigidity of the cell wall (6). If a compensatory increase in chitin synthesis occurs in response to the inhibitory effect of caspofungin on β-(1,3)-d-glucan synthesis, the strength of the resulting cell wall may be reduced.
In our study, there appeared to be two different effects in response to drug pressure. In the case of fumarate, for example, in some cases the peak area increased rapidly at a low drug concentration and decreased just as rapidly at a high drug concentration (e.g., for A. flavus 05-314-1100 and A. terreus 04-188-2548); in the presence of AMB the rapid decrease coincided with the MEP, but with caspofungin other metabolites were relatively unaffected. In other cases, the level of fumarate increased rapidly and reached a plateau or continued to increase. Various susceptibilities to caspofungin may explain these differences. In those strains in which metabolite production plateaus or continues to increase with increasing drug concentration, sufficient β-(1,3)-d-glucan may still be present in the cell wall to maintain its integrity.
In conclusion, NMR spectroscopy is a powerful tool for detecting subtle effects on fungal metabolism as a result of drug pressure. It can be automated and therefore has great potential for use in high-throughput screening of antifungal activity in new drug development. Its further investigation as a platform for determining antifungal susceptibility in reference mycology laboratories is warranted.
Acknowledgments
This work was supported by a Medical School grant from Merck Sharpe & Dohme Pty. Ltd. and a Centre of Clinical Research Excellence program grant (grant no. 264625) from the National Health and Medical Research Council of Australia. The work of P.W.K. is supported by a Discovery grant from the Australian Research Council.
Footnotes
Published ahead of print on 4 September 2007.
REFERENCES
- 1.Abdou, R. F., S. E. Megalla, A. M. Moharram, K. M. Abdel-Gawad, T. H. Sherif, A. L. el-Syed Mahmood, and A. E. Lottfy. 1989. Cytological effects of fungal metabolites produced by fungi isolated from Egyptian poultry feedstuffs. J. Basic Microbiol. 29:131-139. [DOI] [PubMed] [Google Scholar]
- 2.Arikan, S., M. Lozano-Chiu, V. Paetznick, and J. H. Rex. 2001. In vitro susceptibility testing methods for caspofungin against Aspergillus and Fusarium isolates. Antimicrob. Agents Chemother. 45:327-330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Arikan, S., P. Yurdakul, and G. Hascelik. 2003. Comparison of two methods and three end points in determination of in vitro activity of micafungin against Aspergillus spp. Antimicrob. Agents Chemother. 47:2640-2643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barra, L., C. Fontenelle, G. Ermel, A. Trautwetter, G. C. Walker, and C. Blanco. 2006. Interrelations between glycine betaine catabolism and methionine biosynthesis in Sinorhizobium meliloti strain 102F34. J. Bacteriol. 188:7195-7204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bartizal, C., and F. C. Odds. 2003. Influences of methodological variables on susceptibility testing of caspofungin against Candida species and Aspergillus fumigatus. Antimicrob. Agents Chemother. 47:2100-2107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Borgia, P. T., and C. L. Dodge. 1992. Characterization of Aspergillus nidulans mutants deficient in cell wall chitin or glucan. J. Bacteriol. 174:377-383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brakhage, A. A. 2005. Systemic fungal infections caused by Aspergillus species: epidemiology, infection process and virulence determinants. Curr. Drug Targets 6:875-886. [DOI] [PubMed] [Google Scholar]
- 8.Brakhage, A. A., and K. Langfelder. 2002. Menacing mold: the molecular biology of Aspergillus fumigatus. Annu. Rev. Microbiol. 56:433-455. [DOI] [PubMed] [Google Scholar]
- 9.Coen, M., J. Bodkin, D. Power, W. A. Bubb, U. Himmelreich, P. W. Kuchel, and T. C. Sorrell. 2006. Antifungal effects on metabolite profiles of medically important yeast species measured by nuclear magnetic resonance spectroscopy. Antimicrob. Agents Chemother. 50:4018-4026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Denning, D. W. 2003. Echinocandin antifungal drugs. Lancet 362:1142-1151. [DOI] [PubMed] [Google Scholar]
- 11.Denning, D. W. 1998. Invasive aspergillosis. Clin. Infect. Dis. 26:781-805. [DOI] [PubMed] [Google Scholar]
- 12.Doffman, S. R., A. G. Agrawal, and J. S. Brown. 2005. Invasive pulmonary aspergillosis. Expert Rev. Anti-Infective Ther. 3:613-627. [DOI] [PubMed] [Google Scholar]
- 13.Goldberg, I., J. S. Rokem, and O. Pines. 2006. Organic acids: old metabolites, new themes. J. Chem. Technol. Biotechnol. 81:1601-1611. [Google Scholar]
- 14.Herbrecht, R., D. W. Denning, T. F. Patterson, J. E. Bennett, R. E. Greene, J.-W. Oestmann, W. V. Kern, K. A. Marr, P. Ribaud, O. Lortholary, R. Sylvester, R. H. Rubin, J. R. Wingard, P. Stark, C. Durand, D. Caillot, E. Thiel, P. H. Chandrasekar, M. R. Hodges, H. T. Schlamm, P. F. Troke, B. de Pauw, R. Allen, M. Aoun, C. Aul, M. Bjorkholm, K. L. Blanchard, M. Boogaerts, E. Bouza, E. J. Bow, H. R. Brodt, J. Brown, D. Buchheidt, J. Y. Cahn, A. Calmaggi, J. M. Cisneros, C. Cordonnier, J. Daly, C. A. Da Cunha, R. De Bock, A. Del Favero, J. Diaz Mediavilla, M. C. Dignani, C. Doyen, J. S. Dummer, B. Dupont, M. Egyed, D. Engelhard, G. Faetkenheuer, R. Feld, D. Fiere, G. Fioritoni, G. Garber, Z. Gasztonyi, K. Godder, D. Graham, A. Gratwohl, R. Greenberg, K. High, F. Jacobs, V. Kremery, P. Kumar, W. Langer, M. Laverdiere, P. Ljungman, H. Lode, A. Louie, D. Maki, J. P. Marie, D. J. E. Marriott, D. S. McKinsey, R. Mertelsmann, M. K. Nair, N. Milpied, A. Nagler, D. Niederwieser, L. Pagano, P. Pappas, J. Perfect, J. Pottage, V. Raina, J. Reinhardt, S. Richardson, L. Rickman, M. Ruhnke, I. Salit, W. M. Scheld, S. Schuler, M. Schuster, R. Schwerdtfeger, S. D. Shafran, B. Simmons, M. Slavin, M. Sokol-Anderson, P. Tebas, C. Tsoukas, A. Ullmann, J. Van Burik, J. W. Van't Wout, E. C. Vinaya Kumar, et al. 2002. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N. Engl. J. Med. 347:408-415. [DOI] [PubMed] [Google Scholar]
- 15.Imhof, A., S. A. Balajee, and K. A. Marr. 2003. New methods to assess susceptibilities of Aspergillus isolates to caspofungin. J. Clin. Microbiol. 41:5683-5688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kurtz, M. B., I. B. Heath, J. Marrinan, S. Dreikorn, J. Onishi, and C. Douglas. 1994. Morphological effects of lipopeptides against Aspergillus fumigatus correlate with activities against (1,3)-β-d-glucan synthase. Antimicrob. Agents Chemother. 38:1480-1489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lass-Florl, C., K. Griff, A. Mayr, A. Petzer, G. Gastl, H. Bonatti, M. Freund, G. Kropshofer, M. P. Dierich, and D. Nachbaur. 2005. Epidemiology and outcome of infections due to Aspergillus terreus: 10 year single centre experience. Br. J. Haematol. 131:201-207. [DOI] [PubMed] [Google Scholar]
- 18.Lee, M. B., J. W. Blunt, M. Lever, and P. M. George. 2004. A nuclear-magnetic-resonance-based assay for betaine-homocysteine methyltransferase activity. Anal. Biochem. 330:199-205. [DOI] [PubMed] [Google Scholar]
- 19.National Committee for Clinical Laboratory Standards. 2002. Reference method for broth dilution antifungal susceptibility testing of conidium-forming filamentous fungi. Accepted standard. NCCLS document M38-A. National Committee for Clinical Laboratory Standards, Wayne, PA.
- 20.Nicholson, J. K., P. J. D. Foxall, M. Spraul, R. D. Farrant, and J. C. Lindon. 1995. 750 MHz 1H and 1H-13C NMR spectroscopy of human blood plasma. Anal. Chem. 67:793-811. [DOI] [PubMed] [Google Scholar]
- 21.Odds, F. C., M. Motyl, R. Andrade, J. Bille, E. Canton, M. Cuenca-Estrella, A. Davidson, C. Durussel, D. Ellis, E. Foraker, A. W. Fothergill, M. A. Ghannoum, R. A. Giacobbe, M. Gobernado, R. Handke, M. Laverdiere, W. Lee-Yang, W. G. Merz, L. Ostrosky-Zeichner, J. Peman, S. Perea, J. R. Perfect, M. A. Pfaller, L. Proia, J. H. Rex, M. G. Rinaldi, J.-L. Rodriguez-Tudela, W. A. Schell, C. Shields, D. A. Sutton, P. E. Verweij, and D. W. Warnock. 2004. Interlaboratory comparison of results of susceptibility testing with caspofungin against Candida and Aspergillus species. J. Clin. Microbiol. 42:3475-3482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Park, Y., M. L. Buszko, and J. E. Gander. 1999. Glycine betaine: reserve form of choline in Penicillium fellutanum in low-sulfate medium. Appl. Environ. Microbiol. 65:1340-1342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Park, Y., and J. E. Gander. 1998. Choline derivatives involved in osmotolerance of Penicillium fellutanum. Appl. Environ. Microbiol. 64:273-278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Patterson, T. F., H. W. Boucher, R. Herbrecht, D. W. Denning, O. Lortholary, P. Ribaud, R. H. Rubin, J. R. Wingard, B. De Pauw, H. T. Schlamm, P. Troke, and J. E. Bennett. 2005. Strategy of following voriconazole versus amphotericin B therapy with other licensed antifungal therapy for primary treatment of invasive aspergillosis: impact of other therapies on outcome. Clin. Infect. Dis. 41:1448-1452. [DOI] [PubMed] [Google Scholar]
- 25.Rex, J. H., M. A. Pfaller, T. J. Walsh, V. Chaturvedi, A. Espinel-Ingroff, M. A. Ghannoum, L. L. Gosey, F. C. Odds, M. G. Rinaldi, D. J. Sheehan, and D. W. Warnock. 2001. Antifungal susceptibility testing: practical aspects and current challenges. Clin. Microbiol. Rev. 14:643-658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Shalit, I., Y. Shadkchan, Z. Samra, and N. Osherov. 2003. In vitro synergy of caspofungin and itraconazole against Aspergillus spp.: MIC versus minimal effective concentration end points. Antimicrob. Agents Chemother. 47:1416-1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Stevens, D. A., M. Espiritu, and R. Parmar. 2004. Paradoxical effect of caspofungin: reduced activity against Candida albicans at high drug concentrations. Antimicrob. Agents Chemother. 48:3407-3411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Stevens, D. A., M. Ichinomiya, Y. Koshi, and H. Horiuchi. 2006. Escape of Candida from caspofungin inhibition at concentrations above the MIC (paradoxical effect) accomplished by increased cell wall chitin: evidence for β-1,6-glucan synthesis inhibition by caspofungin. Antimicrob. Agents Chemother. 50:3160-3161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sutton, D. A., S. E. Sanche, S. G. Revankar, A. W. Fothergill, and M. G. Rinaldi. 1999. In vitro amphotericin B resistance in clinical isolates of Aspergillus terreus, with a head-to-head comparison to voriconazole. J. Clin. Microbiol. 37:2343-2345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Talibart, R., M. Jebbar, K. Gouffi, V. Pechereau, G. Gouesbet, C. Blanco, T. Bernard, and J. Pocard. 1997. Transient accumulation of glycine betaine and dynamics of endogenous osmolytes in salt-stressed cultures of Sinorhizobium meliloti. Appl. Environ. Microbiol. 63:4657-4663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.van Vianen, W., S. de Marie, M. T. ten Kate, R. A. A. Mathot, and I. A. J. M. Bakker-Woudenberg. 2006. Caspofungin: antifungal activity in vitro, pharmacokinetics, and effects on fungal load and animal survival in neutropenic rats with invasive pulmonary aspergillosis. J. Antimicrob. Chemother. 57:732-740. [DOI] [PubMed] [Google Scholar]
- 32.Weiderhold, N. P., D. P. Kontoyiannis, J. Chi, R. A. Prince, V. H. Tam, and R. E. Lewis. 2004. Pharmacodynamics of caspofungin in a murine model of invasive pulmonary aspergillosis: evidence of concentration-dependent activity. J. Infect. Dis. 190:1464-1471. [DOI] [PubMed] [Google Scholar]
- 33.Weiderhold, N. P., D. P. Kontoyiannis, R. A. Prince, and R. E. Lewis. 2005. Attenuation of the activity of caspofungin at high concentrations against Candida albicans: possible role of cell wall integrity and calcineurin pathways. Antimicrob. Agents Chemother. 49:5146-5148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Wetter, T. J., K. C. Hazen, and J. E. Cutler. 2003. Modification of rapid susceptibility assay for antifungal susceptibility testing of Aspergillus fumigatus. J. Clin. Microbiol. 41:4252-4258. [DOI] [PMC free article] [PubMed] [Google Scholar]



