Abstract
BAY 10-8888, a cyclic β-amino acid, exerts its antifungal activity by inhibition of isoleucyl-tRNA synthetase activity after accumulation to a millimolar concentration inside the cell. We have selected and characterized BAY 10-8888-resistant Candida albicans mutants. Reduced BAY 10-8888 accumulation as well as increased isoleucyl-tRNA synthetase activity was observed in these mutants. Some of the mutants were cross-resistant to cispentacin, a structurally related β-amino acid, while sensitivities to 5-fluorocytosine and fluconazole remained unchanged in all mutants. All except two in vitro-resistant mutants were pathogenic in a murine candidiasis model, and BAY 10-8888 failed to cure the infection. Furthermore, we have characterized BAY 10-8888 transport and isoleucyl-tRNA synthetase activity in several Candida tropicalis strains which showed MICs higher than those of other Candida strains. An analysis of the C. tropicalis strains revealed that intracellular concentrations of BAY 10-8888 were in the millimolar range, comparable to those for C. albicans. However, these isolates expressed isoleucyl-tRNA synthetase activities about fourfold higher than those for C. albicans. To test the possibility of resistance modeling, we determined the correlations between the intracellular concentration of BAY 10-8888, the specific activity of isoleucyl-tRNA synthetase, the number of free, i.e., noninhibited, isoleucyl-tRNA synthetase molecules/cell, and growth, assuming a linear relation. We found significant correlations between growth and the intracellular concentration of BAY 10-8888 and between growth and the number of free isoleucyl-tRNA synthetase molecules/cell, but not between growth and the specific activity of isoleucyl-tRNA synthetase.
Drug-resistant mutants of susceptible fungi are important tools for unraveling the mode of action of antifungals and for the characterization of possible resistance mechanisms (2, 6, 7, 9, 12–14, 18, 20, 22, 23). For example, resistant Saccharomyces cerevisiae mutants were used to clone a subunit of the target protein for echinocandins, the β-glucan synthase (6), and the gene encoding the target for aureobasidin (12, 13, 23). Characterization of a Candida albicans mutant resistant to nikkomycin suggested that its uptake by a dipeptide permease is an important step for nikkomycin action (22). Furthermore, 5-fluorocytosine-resistant clinical isolates of C. albicans displayed a low level of activity of UMP pyrophosphorylase, an enzyme which is part of the biochemical pathway converting 5-fluorocytosine into a DNA and RNA synthesis inhibitor (34).
BAY 10-8888 is a synthetic derivative of the naturally occurring β-amino acid cispentacin (15, 17, 24, 25) with potent anti-Candida activity (4). The structures of BAY 10-8888 and cispentacin are shown in Fig. 1. Recently, we have unraveled the mode of action of BAY 10-8888. BAY 10-8888 is accumulated about 200-fold by Candida spp. and S. cerevisiae. Inside the cell, BAY 10-8888 inhibits isoleucyl-tRNA synthetase, which results in the inhibition of protein biosynthesis and cell growth (37).
FIG. 1.
Structures of BAY 10-8888 [(-)-(1R,2S)-2-amino-4-methylene-cyclopentane-1-carboxylic acid] and cispentacin [(-)-(1R,2S)-2-amino-cyclopentane-1-carboxylic acid]).
The characterization of BAY 10-8888-resistant C. albicans mutants, presented here, supports our current view of its mode of action. Furthermore, our studies explain the reduced sensitivities of some Candida tropicalis isolates.
MATERIALS AND METHODS
Radiolabelled compounds.
[14C]BAY 10-8888 (12 mCi/mmol) was provided by M. Radtke, Institute for Pharmacokinetics, Bayer AG. The radiochemical purity was >96%.
Materials.
The β-amino acids cispentacin and BAY 10-8888 and fluconazole were synthesized at Bayer AG. 5-Fluorocytosine, β-chloro-dl-alanine, dl-thiaisoleucine, cycloleucine, and cycloheximide were obtained from Sigma, Deisenhofen, Germany.
Organisms.
All C. albicans and C. tropicalis strains represent clinical isolates from various sources and are maintained in the Bayer strain collection mycology (BSM). The following yeasts were used: C. albicans BSMY 212 (ATCC 200498), C. tropicalis BSMY 601, C. tropicalis BSMY 605, C. tropicalis BSMY 610, C. tropicalis BSMY 611, C. tropicalis BSMY 612, C. tropicalis BSMY 616, and C. tropicalis BSMY 620. Strains were maintained and grown as described previously (37). C. albicans CAI4 was a gift from W. Fonzi (8).
Mutant selection.
To select for C. albicans mutants resistant to BAY 10-8888, 108 log-phase C. albicans cells grown in YNG medium (0.67% Bacto yeast nitrogen base [Difco], 1.0% glucose [pH 7.0]) at 30°C were plated onto YNG medium-agar plates (1.5% agarose [Serva, Mannheim, Germany]) containing 100 or 500 μg of BAY 10-8888 per ml. After incubation for 48 h at 37°C the number of resistant colonies was determined. Susceptibility testing was performed as described previously with YNG medium (see above) (37). MIC90 was defined as the lowest compound concentration at which no visible growth occurred. Methods for measuring BAY 10-8888 uptake and isoleucyl-tRNA synthetase activity have been described previously (37).
Generation of resistant mutants by adaptation.
C. albicans cells were inoculated at a cell density of 105/ml into YNG medium (10 ml) containing 1 μg of BAY 10-8888 per ml. Cells were incubated at 37°C with shaking until visible growth occurred. Then, 200 μl was transferred to fresh YNG medium containing 2 μg of BAY 10-8888 per ml, and incubation was continued at 37°C. By successive twofold increases in concentration over a period of about 1 month mutants resistant to 640 and 1,280 μg of BAY 10-8888 per ml were obtained. Clones were isolated by streaking out an aliquot on YNG-agar plates containing appropriate concentrations of BAY 10-8888.
In vivo pathogenicity and sensitivity.
We used the mouse candidiasis model described in reference 26 to test C. albicans mutants for pathogenicity and BAY 10-8888 sensitivity. Briefly, mutant yeast variants were grown for 24 h at 28°C in Nervina Agar (0.5% [vol/vol] glycerol, 0.5% Bacto Peptone, 0.5% sodium chloride, 4% malt extract, 2% Bacto Agar; pH 7.0) tubes and rinsed off with phosphate-buffered saline (PBS). Mice were infected by injection of 106 cells in 0.2 ml of PBS into the caudal vein. Mice were treated orally (p.o.) with 10 mg of BAY 10-8888 per kg of body weight twice daily for 4 days starting immediately after infection. Nontreated controls were included. Survival was monitored for 7 days. Mice severely ill on day 7 were recorded as dead. Survival curves were calculated according to the Kaplan-Meier method with the Prism program (GraphPad Software Inc., San Diego, Calif.) for microcomputers and compared by the log rank test. A P value <0.05 was considered significant.
Calculation of the number of tRNA-synthetase molecules.
The number of isoleucyl-tRNA synthetase molecules per cell was calculated based on data from a procedure for the purification of isoleucyl-tRNA synthetase from baker’s yeast (32). Van der Haar isolated 160 A280 units (which is equivalent to 160 mg) of isoleucyl-tRNA synthetase from 6,000 g of baker’s yeast. Assuming a wet weight of 60 pg/cell (30), this corresponds to 1014 cells. The purity of the isoleucyl-tRNA synthetase preparation was 20 to 60%, and the yield (based on total activity) was 70.5%. The amount of isoleucyl-tRNA synthetase per cell can be calculated as follows: 0.16 g × 0.2 × 1.41/1014 cells = 4.5 × 10−16 g/cell and 0.16 g × 0.6 × 1.41/1014 cells = 1.35 × 10−15 g/cell, assuming 20 and 60% purity, respectively. The molecular weight of isoleucyl-tRNA synthetase from S. cerevisiae is 124,000 (3). Thus, the number of isoleucyl-tRNA synthetase molecules per cell can be calculated as follows: 4.5 × 10−16 g/cell × 6.022 × 1023 molecules/mol/124,000 g/mol = 2,185 molecules/cell and 1.35 × 10−15 g/cell × 6.022 × 1023 molecules/mol/124,000 g/mol = 6,556 molecules/cell, again assuming 20 and 60% purity, respectively. The specific activity of isoleucyl-tRNA synthetase (0.11 to 0.13 U/mg of protein) in crude cell extract of C. albicans BMSY 212 was similar to the specific activity (0.12 U/mg of protein) in crude cell extract of baker’s yeast (32). Assuming that C. albicans and S. cerevisiae contain the same amount of protein per cell, the number of isoleucyl-tRNA synthetase molecules per cell expressed by C. albicans BMSY 212 under the growth conditions used here is similar to the number we calculated for S. cerevisiae. The concentration of free (i.e., not inhibited by BAY 10-8888) isoleucyl-tRNA synthetase [Efree] was calculated according to the following equation: [Efree] = Ki × [Et]/[Et + I], which was derived from the dissociation equation Ki = [E] × [I]/[EI]. [Et] is the total concentration of isoleucyl-tRNA synthetase inside the cell. Ki is the dissociation constant of BAY 10-8888 (37), and [I] is the intracellular concentration of BAY 10-8888. [Et] was calculated by the equation [Et] M = number of isoleucyl-tRNA synthetase molecules/cell/(6.022 × 1023 molecules/mol × 5.7 × 10−14 liters/cell) where 5.7 × 10−14 liters is the volume of a C. albicans BSMY 212 cell (37). Therefore, [Et] is 6.4 × 10−8 M for 2,200 molecules/cell and 2 × 10−7 M for 6,600 molecules/cell.
RESULTS
Generation and characterization of BAY 10-8888-resistant C. albicans mutants.
C. albicans mutants resistant in vitro to BAY 10-8888 were selected by two methods. To isolate C. albicans mutants that reveal a high level of resistance, susceptible C. albicans BSMY 212 cells were plated on YNG-agarose containing 100 or 500 μg of BAY 10-8888 per ml. After incubation for 48 h at 37°C, BAY 10-8888-resistant colonies appeared at a frequency of 1.2 × 10−7 and 3.8 × 10−7, respectively. Several colonies were picked from plates containing 100 μg of BAY 10-8888 per ml (mutants S13 to S17) and 500 μg of BAY 10-8888 per ml (mutants S18 to S22) and tested for their sensitivities to BAY 10-8888, the related β-amino acid cispentacin, and the antifungals 5-fluorocytosine and fluconazole (Table 1). For mutants selected on 100 μg of BAY 10-8888 per ml the MIC90s for BAY 10-8888 were increased 250- to 8,000-fold, while all mutants selected on agar plates containing 500 μg of BAY 10-8888 per ml showed about an 8,000-fold decrease in susceptibility to BAY 10-8888. Susceptibilities to cispentacin decreased (MIC90s, 64 to 128 μg/ml), except for that of mutant S15. Susceptibilities to 5-fluorocytosine and fluconazole remained unchanged. Sensitivity to BAY 10-8888 did not increase after repeated passage of the mutants in BAY 10-8888-free medium, indicating that the resistant phenotype was stable. Mutants S13 to S22 grew at rates comparable to that of the wild type (data not shown).
TABLE 1.
MIC90s of different compounds for wild-type C. albicans BSMY 212 and derived mutants
| Wild type or mutant | MIC90 of:
|
|||
|---|---|---|---|---|
| BAY 10-8888 | Cispentacin | 5-Fluorocytosine | Fluconazole | |
| Wild type | 2 | 2 | 0.063 | 4 |
| 640/1 | >16,400 | >1,024 | 0.125 | 4 |
| 640/2 | 1,024 | 128 | 0.032 | 2 |
| 1280/1 | >16,400 | >1,024 | 0.063 | 4 |
| S13 | 8,200 | 64 | 0.063 | 2 |
| S14 | 16,400 | 64 | 0.125 | 4 |
| S15 | 512 | 16 | 0.063 | 4 |
| S16 | 16,400 | 64 | 0.125 | 2 |
| S17 | >16,400 | 64 | 0.063 | 2 |
| S18 | >16,400 | 128 | 0.125 | 4 |
| S19 | >16,400 | 64 | 0.063 | 2 |
| S20 | >16,400 | 128 | 0.125 | 4 |
| S21 | >16,400 | 128 | 0.125 | 2 |
| S22 | >16,400 | 64 | 0.125 | 2 |
In a second approach, sensitive C. albicans BSMY 212 cells were adapted stepwise to higher BAY 10-8888 concentrations in liquid culture over a period of about 1 month. The final concentrations of BAY 10-8888 in the growth medium were 640 and 1,280 μg/ml. Three clones, 640/1 and 640/2 from the culture containing 640 μg of BAY 10-8888 per ml and 1280/1 from the culture containing 1,280 μg of BAY 10-8888 per ml, were isolated and tested for their sensitivities to BAY 10-8888, cispentacin, 5-fluorocytosine, and fluconazole (Table 1). Sensitivities to 5-fluorocytosine and fluconazole remained unchanged for all three mutants. The MIC90s of BAY 10-8888 and cispentacin for mutants 640/1 and 1280/1 were >16,400 μg/ml and >1,024 μg/ml, respectively. Mutant 640/2 was more sensitive to cispentacin than to BAY 10-8888 (MIC90, 128 versus 1,024 μg/ml; Table 1). The sensitivity of mutant 1280/1 to other amino acid analogs was tested. The MIC90s of β-chloro-dl-alanine and dl-thiaisoleucine increased from 31 (level for wild type) to >1,000 μg/ml; the MIC90 of cycloleucine increased from 25 to >100 μg/ml. In contrast, the sensitivity of this mutant to cycloheximide remained unchanged. Mutants 640/1 and 1280/1 grew as fast as the wild type, whereas mutant 640/2 grew more slowly (data not shown).
To test whether resistance was due to alterations in the target molecules of BAY 10-8888, we investigated BAY 10-8888 accumulation under steady-state conditions and isoleucyl-tRNA synthetase activities in selected mutants. As shown previously, BAY 10-8888 inhibits isoleucyl-tRNA synthetase activity after accumulation to millimolar concentrations within the cell (37). The results are shown in Table 2. All mutants showed reduced BAY 10-8888 uptake after incubation for 30 min compared to the wild type. Except for mutant S15, BAY 10-8888 accumulation was below 10% of that of the wild type. The MIC90 for mutant S15, which showed the highest residual activity (14.1% of that of the wild type), was lower than that for any other mutant (Table 2).
TABLE 2.
Characterization of BAY 10-8888-resistant C. albicans BSMY 212 mutants
| Wild type or mutant | MIC90 of BAY 10-8888 (μg/ml) | Relative accumulation/mg of protein ± SD (% of wild type)a | Sp actb of isoleucyl-tRNA synthetase (U/mg of protein) ± SD |
|---|---|---|---|
| Wild type | 2 | 100 ± 4 | 0.13 ± 0.026 |
| 640/1 | >16,400 | 1.1 ± 0.03 | 0.33 ± 0.072 |
| 640/2 | 1,024 | 5.7 ± 0.083 | 0.25 ± 0.04 |
| 1280/1 | >16,400 | 3.1 ± 0.02 | 0.08 ± 0.015 |
| S13 | 8,200 | 5.7 ± 0.079 | 0.13 ± 0.021 |
| S14 | 16,400 | 6.4 ± 0.13 | 0.115 ± 0.019 |
| S15 | 512 | 14.1 ± 0.31 | n.d.c |
| S17 | >16,400 | 8.0 ± 0.4 | 0.125 ± 0.018 |
| S18 | >16,400 | 4.0 ± 0.08 | 0.12 ± 0.03 |
| S20 | >16,400 | 2.2 ± 0.12 | n.d. |
| S22 | >16,400 | 2.2 ± 0.04 | n.d. |
Values were measured after 30 min of accumulation.
In crude extract.
n.d., not determined.
Isoleucyl-tRNA synthetase activity was determined in crude extracts of selected mutants (Table 2). The specific activity was identical to wild-type activity for mutants selected on BAY 10-8888-containing agar plates (S13, S14, S17, and S18). In contrast, mutants 640/1 and 640/2, which were adapted to growth in liquid medium containing 640 μg of BAY 10-8888 per ml, showed increased isoleucyl-tRNA synthetase activity. Specific activity was increased 2.5-fold for mutant 640/1 and 2-fold for mutant 640/2.
The pathogenicities of C. albicans BSMY 212 mutants selected for decreased in vitro sensitivity to BAY 10-8888 (S13 to S22) were tested in a murine candidiasis model. Groups of mice were infected with 106 cells, and survival was recorded over a period of 7 days postinfection. A sample from the cell suspension used to infect mice was used for in vitro sensitivity testing against BAY 10-8888. Only wild-type C. albicans BSMY 212, not the derived mutants, was sensitive to BAY 10-8888. The survival curves of mice infected with mutants were individually compared with the survival curve of mice infected with wild-type C. albicans BSMY 212. Infection with C. albicans BSMY 212 resulted in 80% death within 7 days. Groups of mice infected with mutants S14 (P = 0.0066) and S17 (P = 0.035) showed prolonged survival (see Fig. 2A). All other mutants (S13, S15, S16, and S18 to S22) showed the same pathogenicity as wild-type C. albicans BSMY 212 (see Fig. 2A for a representative example). Therefore, our selection procedure for decreased in vitro sensitivity against BAY 10-8888 did not result in the loss of pathogenicity in a systemic infection in mice for most of the mutants. To test for in vivo sensitivity to BAY 10-8888, treatment groups received 10 mg of BAY 10-8888 per kg p.o. twice daily for 4 days starting immediately after infection. Survival was recorded for 7 days. Survival curves from treatment groups were compared with those from their respective untreated control groups. Treatment of mice infected with wild-type C. albicans BSMY 212 increased survival from 20 to 93% (P < 0.0001). No significant difference in the survival of mice infected with mutants S13 to S22 between treatment and control groups could be detected (see Fig. 2B for a representative example). This indicated that the loss of BAY 10-8888 accumulation also resulted in a decreased in vivo sensitivity to BAY 10-8888 in these mutants.
FIG. 2.
(A) Kaplan-Meier plot of groups of mice (n = 15) infected with wild-type C. albicans BSMY 212 and derived mutants S13, S14, and S17. (B) Kaplan-Meier plot of groups of mice (n = 15) infected with wild-type C. albicans BSMY 212 and derived mutant S13. Treatment groups (indicated by t) received 10 mg of BAY 10-8888 per kg p.o. twice daily for 4 days starting immediately after the infection.
Characterization of BAY 10-8888-resistant C. tropicalis strains.
We determined the accumulations of BAY 10-8888 and the specific isoleucyl-tRNA synthetase activities for seven C. tropicalis isolates less sensitive to BAY 10-8888 (MIC90s, ≥64 μg/ml) in vitro (Table 3). Relative to C. albicans BSMY 212 all C. tropicalis isolates accumulated 52 to 82% BAY 10-8888 per mg of protein after 30 min. The specific activities of isoleucyl-tRNA synthetase ranged from about 2.5- to 4-fold higher compared to that of C. albicans BSMY 212 (Table 3).
TABLE 3.
Characterization of Candida tropicalis isolates
| Isolate | MIC90 of BAY 10-8888 (μg/ml) | Relative accumulation/mg protein (%) ± SDa | Sp actb of isoleucyl-tRNA synthetase (U/mg protein) ± SD |
|---|---|---|---|
| C. albicans BSMY 212 | 2 | 100 ± 18 | 0.11 ± 0.019 |
| C. tropicalis BSMY 601 | 128 | 77 ± 20 | 0.47 ± 0.005 |
| C. tropicalis BSMY 605 | 64 | 65 ± 11 | 0.44 ± 0.013 |
| C. tropicalis BSMY 610 | >128 | 69 ± 17 | 0.36 ± 0.043 |
| C. tropicalis BSMY 611 | >128 | 62 ± 0.5 | 0.41 ± 0.024 |
| C. tropicalis BSMY 612 | >128 | 82 ± 8 | 0.47 ± 0.028 |
| C. tropicalis BSMY 616 | >128 | 52 ± 10 | 0.42 ± 0.066 |
| C. tropicalis BSMY 620 | >128 | 58 ± 6 | 0.28 ± 0.013 |
Values were measured after 30 min of accumulation. The accumulation for C. albicans BSMY 212 was taken as 100%.
In crude extract.
Prediction of growth in the presence of BAY 10-8888.
As shown above several C. albicans mutants and C. tropicalis isolates, less sensitive to BAY 10-8888 in vitro, showed reduced active accumulation of BAY 10-8888 or increased isoleucyl-tRNA synthetase activity or both. We tested whether changes in one of these two parameters or in the parameter resulting from the combination of the two, namely, the number of noninhibited isoleucyl-tRNA synthetase molecules per cell, can be correlated with growth at 8 μg of BAY 10-8888 per ml (56 μM). Mutants 640/1, 640/2, and 1280/1 were not included, since they were selected to grow at very high concentrations of BAY 10-8888 and may, therefore, have multiple mutations.
To calculate the number of isoleucyl-tRNA synthetase molecules per cell we used data from the closely related baker’s yeast S. cerevisiae, because the corresponding data are not available for C. albicans. The total number of isoleucyl-tRNA synthetase molecules (Et) was calculated to range between 2,200 and 6,600 molecules/cell. Then, by using the equations described in Materials and Methods the number of free isoleucyl-tRNA synthetase molecules, i.e., those not inhibited in the presence of 8 μg of BAY 10-8888 per ml in the medium, per cell was calculated (Table 4). Next, we calculated the correlation coefficients assuming a linear relation between growth and the intracellular concentration of BAY 10-8888, the specific activity of isoleucyl-tRNA synthetase, and the number of free isoleucyl-tRNA synthetase molecules per cell. The correlation coefficients, r, were −0.74 (P = 0.004) for the correlation between relative growth and the intracellular concentration of BAY 10-8888 (Fig. 3A), −0.27 (P = 0.37) for the correlation of the specific activity of isoleucyl-tRNA synthetase with relative growth (Fig. 3B), and 0.75 (P = 0.0029) for the correlation between relative growth and the number of free isoleucyl-tRNA synthetase molecules per cell (Fig. 3C). This indicated that there was no linear correlation between growth and the specific activity of isoleucyl-tRNA synthetase, while weak linear correlation was observed between growth and the intracellular concentration of BAY 10-8888 and between growth and the number of free isoleucyl-tRNA synthetase molecules/cell. More detailed inspection of the plot shown in Fig. 3A revealed that C. albicans CAI4 and C. tropicalis isolates 605 and 620 were outliers. While C. tropicalis isolates 605 and 620 remained outliers when the number of free isoleucyl-tRNA synthetase molecules per cell versus growth was plotted (Fig. 3C), C. albicans CAI4, due to the high level of specific activity of isoleucyl-tRNA synthetase, was no longer an outlier. This suggested that C. tropicalis isolate 620 was resistant due to other mechanisms and that the sensitivity of isolate 605 cannot be explained by the relatively high number of free isoleucyl-tRNA synthetase molecules.
TABLE 4.
Calculation of the number of free isoleucyl-tRNA synthetase molecules per cell
| Mutant or isolate | Sp act (U/mg of protein) | Relative activitya | Intracellular concn (M) | Relative accumulationb (%) | BAY 10-8888 concn (M) | Efree concnc (M) | No. of free molecules/celld | Growth (% of control)e |
|---|---|---|---|---|---|---|---|---|
| BSMY 212 | 0.11 | 1.00 | 6.4 × 10−8 | 100.0 | 0.01 | 5.8 × 10−9 | 200 | 0 |
| S13 | 0.13 | 1.19 | 7.6 × 10−8 | 5.7 | 0.0006 | 4.8 × 10−8 | 1,667 | 88 |
| S14 | 0.12 | 1.06 | 6.8 × 10−8 | 6.4 | 0.0006 | 4.2 × 10−8 | 1,428 | 100 |
| S17 | 0.13 | 1.15 | 7.4 × 10−8 | 8.0 | 0.0008 | 4.1 × 10−8 | 1,403 | 88 |
| S18 | 0.12 | 1.13 | 7.2 × 10−8 | 4.0 | 0.0004 | 5.1 × 10−8 | 1,768 | 82 |
| CAI4 | 0.21 | 1.91 | 1.2 × 10−8 | 248 | 0.0248 | 4.7 × 10−9 | 163 | 15 |
| BSMY 601 | 0.47 | 4.30 | 2.7 × 10−7 | 76.7 | 0.0077 | 3.2 × 10−8 | 1,089 | 45 |
| BSMY 605 | 0.44 | 4.06 | 2.6 × 10−7 | 65.1 | 0.0065 | 3.5 × 10−8 | 1,190 | 30 |
| BSMY 610 | 0.36 | 3.32 | 2.1 × 10−7 | 69.1 | 0.0069 | 2.7 × 10−8 | 924 | 55 |
| BSMY 611 | 0.41 | 3.80 | 2.4 × 10−7 | 62.2 | 0.0062 | 3.4 × 10−8 | 1,159 | 49 |
| BSMY 612 | 0.47 | 4.31 | 2.8 × 10−7 | 81.6 | 0.0082 | 3.0 × 10−8 | 1,037 | 62 |
| BSMY 616 | 0.42 | 3.90 | 2.5 × 10−7 | 52.4 | 0.0052 | 4.0 × 10−8 | 1,375 | 53 |
| BSMY 620 | 0.28 | 2.62 | 1.7 × 10−7 | 58.3 | 0.0058 | 2.5 × 10−8 | 844 | 90 |
Compared to that of C. albicans BSMY 212.
Per milligram of protein. These values were used to calculate the intracellular concentrations of BAY 10-8888.
Calculated as described in Materials and Methods.
Derived from [Efree] assuming 2,200 molecules/cell for C. albicans BSMY 212.
Relative growth at 8 μg of BAY 10-8888 per ml, compared to the untreated control.
FIG. 3.
Correlation of growth after 24 h in YNG medium containing 8 μg of BAY 10-8888 per ml with the intracellular concentration of BAY 10-8888 (A), the specific activity of isoleucyl-tRNA synthetase (B), and the number of free isoleucyl-tRNA (Ile-tRNA) synthetase molecules per cell (C). Shown are the data for wild-type C. albicans (Wildtype), derived spontaneous mutants S13, S14, S17, and S18 (Sp. Mutants), and C. tropicalis isolates (C. tropicalis).
DISCUSSION
We have selected and characterized C. albicans mutants resistant to BAY 10-8888, an antifungal cyclic β-amino acid. Mutants spontaneously resistant to high concentrations of BAY 10-8888 showed decreased accumulation of BAY 10-8888. This provides additional evidence that accumulation is a prerequisite for the in vitro antifungal activity of BAY 10-8888 (37). The mutants were only partially cross-resistant to cispentacin. This supports the finding that cispentacin is accumulated by carriers specific for other amino acids (5, 16). Except for two, all resistant mutants were still pathogenic for mice in an intravenous challenge model. BAY 10-8888 did not cure infections with these mutants, indicating that the accumulation of BAY 10-8888 is also important for its in vivo mode of action. More detailed functional characterization of the carrier transporting BAY 10-8888 for in vivo efficacy, viability, and pathogenicity involves the cloning of the carrier and the generation of C. albicans knockout mutants (8, 10, 11). From our results, we cannot predict whether the kind of mutation we characterized here occurs under in vivo conditions and, if so, at what frequency. The reduced uptake of BAY 10-8888 in resistant mutants could also be due to the increased expression of efflux pumps as, for example, observed for azole-resistant strains (1, 27–29). Because of the unchanged sensitivities of all mutants to the azole fluconazole, we can exclude resistance due to overexpression of the multidrug resistance transporter CDR1, but not other export pumps (28, 29).
Mutants 640/1 and 1280/1, stepwise selected to grow in medium containing 640 and 1,280 μg of BAY 10-8888 per ml, respectively, were cross-resistant to cispentacin as well as to other toxic amino acid analogs. One can speculate that multiple permeases are affected, as described, for example, for the S. cerevisiae SHR3 mutation (21). In addition to reduced BAY 10-8888 accumulation, 640/1 and 640/2 had elevated levels of isoleucyl-tRNA synthetase, the target enzyme of BAY 10-8888. This was not observed in mutants spontaneously resistant to high concentrations of BAY 10-8888 and suggests that the mutation frequency for accumulation is higher than those for the subsequent resistance mechanisms. Mutant 1280/1 seems to have additional resistance mechanisms, e.g., mutation of the target enzyme.
We also characterized seven C. tropicalis isolates, which showed reduced in vitro sensitivity to BAY 10-8888. Despite intracellular accumulation of BAY 10-8888 to concentrations in the millimolar range, these isolates grew in the presence of 8 μg of BAY 10-8888 per ml, a concentration that is fourfold higher than the MIC90 for C. albicans BSMY 212. Interestingly, all isolates showed a higher level of specific activity of isoleucyl-tRNA synthetase. Since we have shown that overexpression of endogenous isoleucyl-tRNA synthetase in S. cerevisiae decreases BAY 10-8888 sensitivity (37), this may be responsible for the decreased sensitivity.
C. albicans and other fungi possess many resistance mechanisms to antifungals (1, 9, 27–29, 31, 33, 35, 36), which cannot all be monitored at the same time for practical reasons. Furthermore, it is always difficult to decide whether an observed change in, for example, the expression of the gene encoding the target, a mutation, or an increased expression of an efflux pump is indeed responsible for the resistance phenotype or represents only an epiphenomenon. Therefore, we tested whether one of the observed characteristics of BAY 10-8888-resistant C. albicans mutants and C. tropicalis isolates, decreased accumulation of BAY 10-8888 or increased specific activity of isoleucyl-tRNA synthetase, or whether, as a result of both effects, the number of free, i.e., noninhibited, isoleucyl-tRNA synthetase molecules per cell can be linearly correlated with growth at 8 μg of BAY 10-8888 per ml. Due to many assumptions and unknown variables the number of free isoleucyl-tRNA synthetase molecules per cell is more a rough estimate than an exact calculation. Nevertheless, it proved to be useful to explain resistance to BAY 10-8888 (see below). While the specific activity of isoleucyl-tRNA synthetase could not be correlated with growth, the intracellular concentration of BAY 10-8888 and the number of free isoleucyl-tRNA synthetase molecules per cell showed significant but weak correlation. An examination of the graphs shown in Fig. 3A and C revealed that the two outliers, C. tropicalis 605 and 620, are, most likely, sensitive and resistant, respectively, to BAY 10-8888 because of other mechanisms. In contrast, C. albicans CAI4 only represents an outlier when the intracellular concentration of BAY 10-8888 is correlated with growth, not if the number of free isoleucyl-tRNA synthetase molecules per cell is correlated with growth. This suggests that the number of free isoleucyl-tRNA synthetase molecules per cell is a better parameter to explain the sensitivity of this strain. The number of free isoleucyl-tRNA synthetase molecules per cell is also sufficient to explain the sensitivity or resistance of the remaining isolates. However, this needs to be further validated, as we tested only a limited number of C. albicans strains. Furthermore, from a physiological point of view, one has to take into account the fact that the linear correlation we assumed occurs only within certain limits. For example, for unlimited growth a certain number of active isoleucyl-tRNA synthetase molecules is required. Further increase in the molecule number will not increase growth rate, as other reactions or metabolites will become growth limiting. On the other hand, a minimal number of active isoleucyl-tRNA synthetase molecules may be required for any growth to occur.
The resistance mechanisms observed for BAY 10-8888 are different from those for other antifungals. Resistance against 5-fluorocytosine can result from mutation of any of the enzymes (cytosine permease, cytosine deaminase, uridine 5′-monophosphate pyrophosphate-phosphoribosyltransferase) involved in its conversion and incorporation in RNA (31). For azoles several resistance mechanisms have been described. These include overexpression or mutation of the targets enzyme lanosterol 14α demethylase (19, 35, 36). The major resistance mechanism, however, is increased expression of ABC transporters or major facilitator pumps (1, 27–29, 33, 35). None of the C. albicans mutants showed any cross-resistance to fluconazole or 5-fluorocytosine, indicating that resistance mechanisms against BAY 10-8888 are different from those against fluconazole and 5-fluorocytosine.
In summary, our data as well as our theoretical considerations fully support the two-step mode of action for the antifungal β-amino acid BAY 10-8888, namely, inhibition of isoleucyl-tRNA synthetase after active accumulation inside the cell.
ACKNOWLEDGMENTS
I thank W. Schönfeld for critically reading the manuscript and S. Badock and A. Ludwig for excellent technical assistance.
REFERENCES
- 1.Albertson G D, Niimi M, Cannon R D, Jenkinson H F. Multiple efflux mechanisms are involved in Candida albicans fluconazole resistance. Antimicrob Agents Chemother. 1996;40:2835–2841. doi: 10.1128/aac.40.12.2835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Angiolella L, Bromuro C, Simonetti N, Cassone A. Glucan synthesis and its inhibition by cilofungin in susceptible and resistant strains of Candida albicans. J Met Vet Mycol. 1992;30:369–376. doi: 10.1080/02681219280000481. [DOI] [PubMed] [Google Scholar]
- 3.Bhanot O S, Kucan Z, Aoyagi S, Lee F C, Chambers R W. Purification of tyrosine transfer RNA ligase, valine transfer RNA ligase, alanine transfer ligase and isoleucine transfer ligase from Saccharomyces-cerevisiae alpha S-288C. Methods Enzymol. 1974;29:547–576. [Google Scholar]
- 4.Bremm, K., R. Endermann, F. Kunisch, M. Matzke, K.-G. Metzger, H. Militzer, J. Mittendorf, and M. Plempel. May 1993. EP patent 571,870.
- 5.Capobianco J, Zakula D Z, Coen M L, Goldman R C. Anti-Candida activity of cispentacin: the active transport of amino acid permeases and its possible mechanism of action. Biochem Biophys Res Commun. 1993;190:1037–1044. doi: 10.1006/bbrc.1993.1153. [DOI] [PubMed] [Google Scholar]
- 6.Douglas C M, Foor F, Marrinan J A, Morin N, Nielsen J B, Dahl A M, Mazur P, Baginsky W, Li W L, El-Sherbeini M, Clemas J A, Mandala S M, Frommer B R, Kurtz M B. The Saccharomyces cerevisiae FKS1 (ETG1) gene encodes an integral membrane-protein which is a subunit of 1,3-beta-d-glucan synthase. Proc Natl Acad Sci USA. 1994;91:12907–12911. doi: 10.1073/pnas.91.26.12907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Douglas C M, Marrinan J A, Li W, Kurtz M B. A Saccharomyces cerevisiae mutant with echinocandin-resistant 1,3-β-d-glucan synthase. J Bacteriol. 1994;176:5686–5696. doi: 10.1128/jb.176.18.5686-5696.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fonzi W A, Irwin M Y. Isogenic strain construction and gene mapping in Candida albicans. Genetics. 1993;134:717–728. doi: 10.1093/genetics/134.3.717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Frost D J, Knapp M, Brandt K, Shadron A, Goldman R C. Characterization of a lipopeptide-resistant strain of Candida albicans. Can J Microbiol. 1997;43:122–128. doi: 10.1139/m97-016. [DOI] [PubMed] [Google Scholar]
- 10.Goldway M, Teff D, Schmidt R, Oppenheim A B, Koltin Y. Multidrug resistance in Candida albicans: disruption of the BENr gene. Antimicrob Agents Chemother. 1995;39:422–426. doi: 10.1128/aac.39.2.422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gow N A R, Robbins P W, Lester J W, Brown A J P, Fonzi W A, Chapman T, Kinsman O S. A hyphal-specific chitin synthase gene (CHS2) is not essential for growth, dimorphism, or virulence of Candida albicans. Proc Natl Acad Sci USA. 1994;91:6216–6220. doi: 10.1073/pnas.91.13.6216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hashidaokado T, Ogawa A, Endo M, Yasumoto R, Takesako K, Kato I. AUR1, a novel gene conferring aureobasidin resistance on Saccharomyces cerevisiae—a study of defective morphologies in AUR1P-depleted cells. Mol Gen Genet. 1996;251:236–244. doi: 10.1007/BF02172923. [DOI] [PubMed] [Google Scholar]
- 13.Heidler S A, Radding J A. The AUR1 gene in Saccharomyces cerevisiae encodes dominant resistance to the antifungal agent auroeobasidin A ( LY295337) Antimicrob Agents Chemother. 1995;39:2765–2769. doi: 10.1128/aac.39.12.2765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hitchcock C A, Russell N J, Barrett B K. Sterols in Candida albicans mutants resistant to polyene or azole antifungals, and of a double mutant C. albicans 6.4. Crit Rev Microbiol. 1987;15:111–115. doi: 10.3109/10408418709104454. [DOI] [PubMed] [Google Scholar]
- 15.Iwamoto T, Tsujii E, Ezaki M, Fujie A, Hashimoto S, Okuhara M, Kohsaka M, Imanaka H. FR109615, a new antifungal antibiotic from Streptomyces setonii: taxonomy, fermentation, isolation, physico-chemical properties and biological activity. J Antibiot. 1990;43:1–7. doi: 10.7164/antibiotics.43.1. [DOI] [PubMed] [Google Scholar]
- 16.Jethwaney D, Hofer M, Khaware R K, Prasad R. Functional reconstitution of a purified proline permease from Candida albicans: interaction with the antifungal cispentacin. Microbiology. 1997;143:397–404. doi: 10.1099/00221287-143-2-397. [DOI] [PubMed] [Google Scholar]
- 17.Konishi M, Nishio M, Saitoh K, Miyaki T, Oki T, Kawaguchi K. Cispentacin, a new antifungal antibiotic. I. Production, isolation, physico-chemical properties and structure. J Antibiot. 1989;42:1749–1755. doi: 10.7164/antibiotics.42.1749. [DOI] [PubMed] [Google Scholar]
- 18.Kurtz M B, Abruzzo G, Flatlery A, Bartizal K, Marrinan J A, Li W, Milligan J, Nollstadt K, Douglas C M. Characterization of echinocandin-resistant mutants of Candida albicans: genetic, biochemical, and virulence studies. Infect Immun. 1996;64:3244–3251. doi: 10.1128/iai.64.8.3244-3251.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lamb D C, Kelly D E, Schunck W H, Shyadehi A Z, Akhtar M, Lowe D J, Baldwin B C, Kelly S L. The mutation T315A in Candida albicans sterol 14 alpha-demethylase causes reduced enzyme activity and fluconazole resistance through reduced affinity. J Biol Chem. 1997;272:5682–5688. doi: 10.1074/jbc.272.9.5682. [DOI] [PubMed] [Google Scholar]
- 20.Lees N D, Broughton M C, Sanglard D, Bard M. Azole susceptibility and hyphal formation in a cytochrome P-450-deficient mutant of Candida albicans. Antimicrob Agents Chemother. 1990;34:831–836. doi: 10.1128/aac.34.5.831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ljungdahl P O, Gimeno C, Styles C A, Fink G R. A novel component of the secretory pathway specifically required for localization of amino acid permeases in yeast. Cell. 1992;71:463–478. doi: 10.1016/0092-8674(92)90515-e. [DOI] [PubMed] [Google Scholar]
- 22.McCarthy P J, Troke P F, Gull K. Mechanism of action of nikkomycin and the peptide transport system of Candida albicans. J Gen Microbiol. 1985;131:775–780. doi: 10.1099/00221287-131-4-775. [DOI] [PubMed] [Google Scholar]
- 23.Nagiec M M, Nagiec E E, Baltisberger J A, Wells G B, Lester R L, Dickson R C. Sphingolipid synthesis as a target for antifungal drugs—complementation of the inositol phosphoceramide synthase defect in a strain of Saccharomyces cerevisiae by the AUR1 gene. J Biol Chem. 1997;272:9809–9817. doi: 10.1074/jbc.272.15.9809. [DOI] [PubMed] [Google Scholar]
- 24.Ohki H, Inamoto Y, Kawabata K, Kamimura T, Sakane K. Synthesis and antifungal activity of FR109615 analogs. J Antibiot. 1991;44:546–549. doi: 10.7164/antibiotics.44.546. [DOI] [PubMed] [Google Scholar]
- 25.Oki T, Hirano M, Tomatsu K, Numata K, Kamei H. Cispentacin, a new antifungal antibiotic. II. In vitro and in vivo antifungal activities. J Antibiot. 1989;42:1756–1762. doi: 10.7164/antibiotics.42.1756. [DOI] [PubMed] [Google Scholar]
- 26.Plempel M. Antimycotic activity of Bay N 7133 in animal experiments. J Antimicrob Chemother. 1984;13:447–463. doi: 10.1093/jac/13.5.447. [DOI] [PubMed] [Google Scholar]
- 27.Sanglard D, Ischer F, Monod M, Bille J. Cloning of Candida albicans genes conferring resistance to azole antifungal agents: characterization of CDR2, a new multidrug ABC transporter gene. Microbiology. 1997;143:405–416. doi: 10.1099/00221287-143-2-405. [DOI] [PubMed] [Google Scholar]
- 28.Sanglard D, Ischer F, Monod M, Bille J. Susceptibilities of Candida albicans multidrug transporter mutants to various antifungal agents and other metabolic inhibitors. Antimicrob Agents Chemother. 1996;40:2300–2305. doi: 10.1128/aac.40.10.2300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sanglard D, Kuchler K, Ischer F, Pagani J-L, Monod M, Bille J. Mechanisms of resistance to azole antifungal agents in Candida albicans isolates from AIDS patients involve specific multidrug transporters. Antimicrob Agents Chemother. 1995;39:2378–2386. doi: 10.1128/aac.39.11.2378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Sherman F. Getting started with yeast. Methods Enzymol. 1991;194:3–21. doi: 10.1016/0076-6879(91)94004-v. [DOI] [PubMed] [Google Scholar]
- 31.Vanden Bosche H, Marchial P, Odds F C. Molecular mechanisms of drug resistance in fungi. Trends Microbiol. 1994;2:393–400. doi: 10.1016/0966-842x(94)90618-1. [DOI] [PubMed] [Google Scholar]
- 32.Van der Haar F. Purification of aminoacyl-tRNA synthetases. Methods Enzymol. 1979;59:257–267. doi: 10.1016/0076-6879(79)59088-0. [DOI] [PubMed] [Google Scholar]
- 33.Walsh T J, Kasai M, Franscesconi A, Landsman D, Chanock S J. New evidence that Candida albicans possesses additional ATP-binding cassette MDR-like genes: implications for antifungal azole resistance. J Med Vet Mycol. 1997;35:133–137. [PubMed] [Google Scholar]
- 34.Whelan W L, Kerridge D. Decreased activity of UMP pyrophosphorylase associated with resistance to 5-fluorocytosine in Candida albicans. Antimicrob Agents Chemother. 1984;26:570–574. doi: 10.1128/aac.26.4.570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.White T C. Increased mRNA levels of ERG16, CDR, and MDR1 correlate with increases in azole resistance in Candida albicans isolates from a patient infected with human immunodeficiency virus. Antimicrob Agents Chemother. 1997;41:1482–1487. doi: 10.1128/aac.41.7.1482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.White T C. The presence of an R467K amino acid substitution and loss of allelic variation correlate with an azole-resistant lanosterol 14α demethylase in Candida albicans. Antimicrob Agents Chemother. 1997;41:1488–1494. doi: 10.1128/aac.41.7.1488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ziegelbauer, K., P. Babczinski, and W. Schönfeld. Submitted for publication.