Caspofungin Kills Candida albicans by Causing both Cellular Apoptosis and Necrosis

Abstract

Caspofungin exerts candidacidal activity by inhibiting cell wall (1,3)-β-d-glucan synthesis. We investigated the physiologic mechanisms of caspofungin-induced Candida albicans cell death. Apoptosis (programmed cell death) and necrosis were studied after C. albicans SC5314 cells were exposed to caspofungin at 0.06, 0.125, and 0.5 μg/ml (0.5×, 1×, and 4× the MIC, respectively) for 3 h. Caspofungin at 0.125 and 0.5 μg/ml reduced cellular viability by >50%, as measured by colony counts and methylene blue exclusion. Apoptosis and necrosis were demonstrated by annexin V and propidium iodide staining for phosphatidylserine externalization and loss of membrane integrity, respectively. At all concentrations of caspofungin, 20 to 25% and 5 to 7% of C. albicans cells exhibited early apoptosis and late apoptosis/necrosis, respectively (P value was not significant [NS]). Necrosis, on the other hand, was significantly greater at 0.125 (43%) and 0.5 (48%) μg/ml than at 0.06 μg/ml (26%) (P values of 0.003 and 0.003, respectively). The induction of apoptosis at concentrations less than or equal to the MIC was corroborated by dihydrorhodamine 123 (DHR-123) and dihydroethidium (DHE) staining (reactive oxygen species production), JC-1 staining (mitochondrial membrane potential dissipation), and terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) and 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining (DNA damage and nuclear fragmentation). Moreover, electron microscopy of cells exposed to 0.125 μg/ml of caspofungin showed hallmark apoptotic features like chromatin margination and condensation and nuclear blebs. Apoptosis was associated with metacaspase 1 activation, as demonstrated by D2R staining. Caspofungin exerts activity against C. albicans by directly killing cells (resulting in necrosis) and causing others to undergo programmed cell death (apoptosis). Apoptosis is initiated at subinhibitory concentrations, suggesting that strategies to target this process may augment the benefits of antifungal agents.

INTRODUCTION

Caspofungin and other agents in the echinocandin class of antifungals have assumed an increasingly important role in the therapy of invasive candidiasis (1). These agents are nontoxic and exert potent fungicidal activity against Candida albicans and other Candida spp. Their antifungal activity is achieved through inhibition of (1,3)-β-d-glucan synthase (2), an enzyme that synthesizes a major constituent of the fungal cell wall. Although the mechanism of activity for the echinocandins is known, the physiological mechanisms by which they cause cell death are not defined. At least two types of mammalian cell death, necrosis and apoptosis, have been described (3). Necrosis is death resulting from direct cellular injury, which is best defined by cell and organelle swelling and lysis (4). Apoptosis, on the other hand, is programmed cell death, the principal morphological feature of which is shrinkage of the cell and its nucleus (3, 4).

Over the last decade, there have been a number of reports on apoptosis in yeasts and filamentous fungi (5). Indeed, apoptosis can be induced in C. albicans by oxidative stress (6), intracellular acidification, and the antifungal agent amphotericin B (7). Notably, C. albicans cells exhibit apoptotic markers that are similar to those of mammalian cells, including phosphatidylserine externalization, reactive oxygen species (ROS) accumulation, mitochondrial membrane potential dissipation, and DNA condensation and fragmentation (8). In this study, we evaluated the mechanisms of C. albicans cell death caused by caspofungin. We demonstrated that caspofungin causes both apoptosis and necrosis of C. albicans cells.

MATERIALS AND METHODS

C. albicans strain and growth conditions.

C. albicans SC5314 was grown in synthetic dextrose complete (SDC) medium (6.7 g of yeast nitrogen base and 20 g of glucose in 1 liter) at 30°C (9). Caspofungin powder was purchased from the University of Pittsburgh Medical Center pharmacy. Media and chemicals were purchased from Becton, Dickinson and Company and Fisher Scientific, respectively, unless specifically stated otherwise. The caspofungin MIC was determined by the broth microdilution method (10). For all assays described below, C. albicans cells in exponential phase in SDC medium were incubated with various concentrations of caspofungin (0, 0.06, 0.125, and 0.5 μg/ml). At specific time points, aliquots were obtained for the respective assays. Viability of C. albicans cells was determined by a colony count determination, and vitality was determined by a methylene blue exclusion assay (11).

Annexin V and PI staining.

C. albicans cells exposed to caspofungin were washed in phosphate-buffered saline (PBS) and incubated at 30°C for 10 min in 0.02 mg/ml Zymolyase 20T in 0.1 M potassium phosphate buffer (PPB; 0.5 ml of 50 mM K₂HPO₄, 5 mM EDTA, 50 mM dithiothreitol [DTT], 50 mM KH₂PO₄, 40 mM 2-mercaptoethanol) with sorbitol at a final concentration of 2.4 M and at pH 7.2 (7, 12). Thereafter, 100 μl of permeabilization solution (0.1 M sodium citrate [pH 6.0] with 0.1% Triton X-100) was added to the washed protoplasts, which were placed on ice for 2 min and washed again. Protoplasts were fixed with 70% ethanol at 30°C for 20 min and subsequently washed with Annexin-V-Fluos (Roche Applied Science) incubation buffer. Annexin V/propidium iodide (PI) binding assays were performed according to the staining kit protocol, using 10% annexin reagent, 10% PI reagent, and 1 mg/ml of RNase A at 37°C for 30 min. Cell analysis for these and other assays in the study was performed using an Olympus FluoView FV500/IX laser scanning confocal microscope. All assays were performed at least in triplicate and repeated at least three times.

ROS production.

C. albicans cell suspensions were incubated for 10 min with 5 mg/ml of dihydrorhodamine 123 (DHR-123) (Sigma-Aldrich). Samples were quantitatively analyzed by flow cytometry (13). ROS production was also determined following incubation for 45 min with 10 μM dihydroethidium (DHE) (MGT Inc.) (14).

Mitochondrial membrane potential (ΔΨ_m).

C. albicans cell suspensions were transferred into sterile amber tubes on ice, and 5 μM JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide) (JC-1 mitochondrial membrane potential detection kit; Biotium Inc.) was added to each sample prior to incubation at 30°C for 15 min (15, 16). Fluorescence was observed by laser scanning confocal microscopy, and samples were quantitatively analyzed by flow cytometry. An untreated sample and a 50 μM carbonyl cyanide 4-(trifluoromethoxy) phenyl hydrazone (FCCP)-treated (Tocris Inc.) sample were used as negative and positive controls, respectively.

DNA damage and nuclear fragmentation.

We utilized terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) and 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining. For TUNEL assays, protoplasts were prepared as described above (7, 17). TUNEL reactions were carried out by using the in situ cell death detection kit (Roche Applied Science) (18). The basic protocol for DAPI staining of nuclei was carried out as previously described (14) using 1 μg of DAPI/ml (Molecular Probes).

Electron microscopy.

C. albicans cells (1 × 10⁶/ml) in SDC medium were incubated with 0.125 μg/ml of caspofungin at 30°C, with shaking (200 rpm) for 1 h. The cells were then fixed with a modified Karnovsky fixative (19). Transmission electron microscopy imaging was performed by the University of Pittsburgh Center of Biologic Imaging Core Laboratories.

Intracellular metacaspase activation.

Caspase activity was detected by staining C. albicans cells exposed to caspofungin with D2R (CaspScreen flow cytometric apoptosis detection kit; BioVision) (9, 20). The cells were observed by laser scanning confocal microscopy and quantitated using flow cytometry.

Statistical analysis.

Results are presented as means and standard deviations. For nonnormally distributed values, the data were log transformed prior to statistical analyses.

Differences in results at two caspofungin concentrations were assessed using Student's t test. Comparisons of results at more than two caspofungin concentrations were made using analysis of variance (ANOVA). P values <0.05 were considered significant.

RESULTS

Caspofungin effects on C. albicans SC5314 viability.

Our primary goal in this study was to determine if caspofungin induced both apoptosis and necrosis as mechanisms of cell death. We focused on caspofungin effects within the first 3 h of exposure for three reasons. First, a time period of 3 h was sufficient to ensure the replication of control C. albicans SC5314 cells, which doubled in concentration by this time point. Second, exposures longer than 3 h are associated with greater caspofungin kills and higher necrotic yeast populations, traits which may obscure the extent of apoptosis (7). Third, apoptotic markers, such as ROS production, may be consequences of cellular stress at later time points rather than causes of apoptosis. To begin the study, the effects of caspofungin on C. albicans SC5314 were measured by CFU enumeration following incubation of mid-exponential-phase cells in various drug concentrations (Fig. 1A). At subinhibitory caspofungin concentrations (0.06 μg/ml) for up to 3 h, growth was indistinguishable from that of the controls. At concentrations of 0.125 (MIC) and 0.5 μg/ml (4× the MIC), significant reductions from the starting inocula were first evident at 2 h. By 3 h, ∼50% reductions from starting inocula were observed (P < 0.0001); there were no significant differences between the two concentrations (P value was not significant [NS]).

Fig 1.

Effects of caspofungin on C. albicans SC5314 viability. (A) Viability as determined by colony counts. C. albicans SC5314 cells were exposed to various concentrations of caspofungin. Data at the respective time points are mean percentages of the starting inoculum ± standard deviations. *, significant differences for 0.125 μg/ml caspofungin versus the no-drug control and 0.06 μg/ml caspofungin (P values of 0.04 and 0.01, respectively) and for 0.5 μg/ml caspofungin versus the no-drug control and 0.06 μg/ml caspofungin (P values of 0.006 and 0.0001, respectively); **, significant differences for 0.125 μg/ml caspofungin versus the no-drug control and 0.06 μg/ml caspofungin and for 0.5 μg/ml caspofungin versus the no-drug control and 0.06 μg/ml caspofungin with a P value of <0.0001. (B) Cell vitality as determined by methylene blue exclusion after exposure to caspofungin for 3 h. Data are mean cell counts (per ml) ± standard deviations at 10⁻⁴ dilutions for each caspofungin concentration. The numbers of cells that were methylene blue negative (i.e., healthy and viable) and methylene blue positive (i.e., nonviable or damaged) are presented as white and gray bars, respectively. The percentages (numbers) of cells with particular staining appear within the appropriate bars. There were significant differences in the percentage of methylene blue-positive cells between cells exposed to caspofungin and control cells (all P values < 0.05) as well as between cells exposed to caspofungin at 0.06 μg/ml and cells exposed to 0.125 μg/ml and 0.5 μg/ml caspofungin.

To further assess the effects of caspofungin on C. albicans SC5314, we used a methylene blue assay to measure cell vitality (Fig. 1B). In this assay, healthy cells exclude methylene blue whereas nonviable or damaged cells accumulate the stain. The percentages of cells that were methylene blue positive at 3 h were significantly increased following exposure to caspofungin at 0.06, 0.125, and 0.5 μg/ml compared to the no-drug control (P < 0.0001; ANOVA). Of note, there were no differences in methylene blue staining between mother and daughter cells at any of the caspofungin concentrations (data not shown).

Induction of C. albicans apoptosis and necrosis by caspofungin.

To assess if killing by caspofungin occurred through induction of apoptosis and/or necrosis, we performed annexin V and PI staining. Annexin V stains phosphatidylserine, which is a negatively charged phospholipid that is translocated from the inner leaflet of the plasma membrane to the outer leaflet during early apoptosis (21). PI staining identifies necrotic cells, as it does not permeate cells with intact membranes. Therefore, staining patterns discriminate between live cells (annexin V negative [annexin V−]/PI−) and early apoptosis (annexin V+/PI−), necrosis (annexin V−/PI+), and late apoptosis/necrosis (annexin V+/PI+) (Fig. 2).

Fig 2.

Annexin V and propidium iodide stains of C. albicans SC5314 cells exposed to caspofungin (0.125 μg/ml) for 3 h (left) and phase image (right). Note apoptotic cells that stained only green (A), late-apoptotic/necrotic cells that stained green and red (B), and necrotic cells that stained only red (C).

Annexin V and PI assays were performed on yeast spheroplasts that were prepared using Zymolyase, a cell wall-active lytic enzyme that hydrolyzes (1,3)-β-d-glucan linkages. We first performed pilot experiments to ensure that Zymolyase did not cause false-positive annexin V or PI staining. Indeed, there was no evidence for apoptosis or necrosis by these assays until Zymolyase concentrations were ≥0.1 mg/ml (data not shown). Thereafter, assays in the presence of caspofungin were performed using 0.02 mg/ml of Zymolyase. As summarized in Table 1, caspofungin at 0.06, 0.125, and 0.5 μg/ml induced early apoptosis in 20 to 25% of cells after 3 h; there were no significant differences between the percentages of early-apoptotic cells at the three concentrations (P = 0.19; ANOVA). Necrosis, on the other hand, was significantly greater in response to 0.125 and 0.5 μg/ml of caspofungin than to 0.06 μg/ml (P values of 0.003 and 0.003, respectively). The percentages of late-apoptotic/necrotic cells were 5 to 7% at the three concentrations of caspofungin (P value was NS; ANOVA).

Table 1.

Effects of caspofungin on early apoptosis, late apoptosis, and necrosis, as determined by annexin V and propidium iodide staining

Caspofungin concn (μg/ml) or control	% of cells with each result after each time perioda
	Annexin V+/PI− (early apoptosis)			Annexin V+/PI+ (late apoptosis/necrosis)			Annexin V−/PI+ (necrosis)
	1 h	2 h	3 h	1 h	2 h	3 h	1 h	2 h	3 h
Control	1.7% ± 0.4%	1.8% ± 0.4%	1.5% ± 0.2%	0.9% ± 0.2%	1.0% ± 0.3%	0.8% ± 0.3%	2.4% ± 0.4%	2.2% ± 0.2%	2.1% ± 0.2%
0.06	6.1% ± 0.2%c	12.6% ± 0.7%c	24.6% ± 1.1%c	1.9% ± 0.2%b	5.8% ± 1.1%b	5.0% ± 0.4%c	5.0% ± 0.6%c	16.9% ± 1.1%c	25.7% ± 1.3%c
0.125	8.8% ± 0.2%c	14.1% ± 0.2%c	20.4% ± 2.3%c	4.8% ± 0.3%c	5.2% ± 0.7%c	4.7% ± 0.2%c	13.5% ± 1.4%c	26.9% ± 0.8%c	43.5% ± 2.2%c
0.5	8.8% ± 0.2%c	14.8% ± 0.7%c	19.7% ± 3.0%c	3.4% ± 0.4%c	7.4% ± 0.1%c	5.9% ± 0.2%c	8.8% ± 0.3%c	31.5% ± 2.1%c	47.9% ± 6.0%c

^aThe data are mean percentages ± standard deviations.

^bSignificant difference in positive staining between cells exposed to caspofungin and control cells at a P value of <0.05.

^cSignificant difference in positive staining between cells exposed to caspofungin and control cells at a P value of <0.01.

Caspofungin effects on other markers of apoptosis. (i) ROS production.

To examine apoptotic effects of caspofungin in greater detail, we first assayed ROS production by DHR-123 staining. ROS production has been implicated in the induction and regulation of the apoptotic pathway in yeast (8, 18). DHR-123 is oxidized by intracellular ROS to the fluorescent chromophore rhodamine 123. By DHR-123 staining, ROS generation by C. albicans SC5314 started at ∼2 h of exposure to caspofungin at 0.125 and 0.5 μg/ml (Fig. 3A). Control cells and cells exposed to 0.06 μg/ml of caspofungin showed little ROS accumulation. To corroborate that caspofungin caused ROS accumulation, we also performed dihydroethidium (DHE) staining on C. albicans SC5314 cells exposed to 0.125 μg/ml of caspofungin for 3 h. DHE is oxidized by ROS to ethidium, which intercalates within nucleic acids and stains cells with a bright red fluorescence. As expected, a significantly higher percentage of caspofungin-exposed cells than of the controls were positive by DHE staining (25.1% ± 4.4% versus 8.1% ± 3.3%; P < 0.0001).

Fig 3.

Effects of caspofungin on various markers of apoptosis. (A) ROS production as determined by DHR-123 staining. There were significantly more C. albicans cells producing ROS following exposure to 0.125 and 0.5 μg/ml of caspofungin than following exposure to 0.06 μg/ml caspofungin or in the absence of the drug (control). *, significant difference in the percentage of DHR-123-positive cells between cells exposed to caspofungin and control cells; §, significant difference in the percentage of DHR-positive cells between cells exposed to caspofungin at 0.06 μg/ml and cells exposed to caspofungin at 0.125 μg/ml and 0.5 μg/ml (all P values < 0.05). (B) Mitochondrial membrane potential as determined by JC-1 staining. Beginning at 1.5 h, the percentage of cells with a loss of mitochondrial membrane potential was significantly greater following exposure to caspofungin than among all control cells (*; P < 0.0001). By 3 h, the percentage of cells with a loss of mitochondrial membrane potential was greater at concentrations of 0.125 and 0.5 μg/ml than at a concentration of 0.06 μg/ml (§; P < 0.0002). (C) DNA damage as determined by TUNEL staining. Significantly higher percentages of C. albicans cells exposed to caspofungin at all concentrations than of control cells sustained DNA fragmentation. (D) Nuclear fragmentation as demonstrated by DAPI staining. C. albicans SC5314 cells are shown following exposure to caspofungin at 0.125 μg/ml for 3 h (left) and in the absence of caspofungin (controls; right). Cells exposed to caspofungin exhibit tear-drop-shaped tubular DNA (long white arrow) and irregular, fragmented DNA (short arrow), which are typical of nuclear abnormalities associated with DNA damage during apoptosis (33, 38). In control cells, on the other hand, DAPI staining revealed a single, bright, round nucleus and peripheral cell spots corresponding to stained mitochondria.

Mitochondrial membrane potential.

The dissipation of mitochondrial membrane potential (ΔΨ_m) is a key cellular event during early apoptosis (22), which leads to the opening of the transition pores of the mitochondrial membrane and the release of apoptogenic factors into the cytosol (23). We used the vital mitochondrial dye JC-1 to investigate mitochondrial function. JC-1 selectively enters the mitochondria. In healthy cells with high ΔΨ_m, it spontaneously forms complexes known as J-aggregates that exhibit intense red fluorescence. In apoptotic or unhealthy cells with low ΔΨ_m, JC-1 remains in the monomeric form, which shows green fluorescence. As shown in Fig. 3B, significant loss of ΔΨ_m was observed by 1.5 h, even at the subinhibitory caspofungin concentration of 0.06 μg/ml.

DNA damage and nuclear fragmentation.

During the late stages of apoptosis, chromatin is damaged because of the proteolysis of nuclear proteins, a process which results in DNA damage and chromatin condensation (24). To investigate the features of late apoptosis in response to caspofungin, we utilized TUNEL staining. The TUNEL assay measures DNA fragmentation at the single-cell level by incorporating biotylated or fluorescent dUTP at sites of free 3′ -OH in DNA (25). The percentage of C. albicans cells that exhibited TUNEL-positive nuclei following exposure to caspofungin for 3 h was significantly greater than the percentage of untreated controls exhibiting TUNEL-positive nuclei (Fig. 3C); there were no differences between cells exposed to the various concentrations of caspofungin (P value was NS; ANOVA). We corroborated these findings with those of DAPI staining (Fig. 3D). DAPI binds to AT sites within the minor groove of DNA, where its fluorescence can be assessed (8). During apoptosis, the permeability of the dye is increased, resulting in deep blue fluorescence that highlights nuclei with abnormal margins and condensed chromatin. C. albicans cells exposed to caspofungin at the different concentrations for 3 h had evidence of nuclear fragmentation associated with DNA damage, including typical tubular staining patterns (13, 21). As with TUNEL staining, there were no differences between cells exposed to the different caspofungin concentrations.

Electron microscopy.

To visualize changes in intracellular morphology consistent with apoptosis, we performed electron microscopy on C. albicans SC5314 cells exposed to 0.125 μg/ml of caspofungin (Fig. 4). Images revealed nuclear blebs and chromatin margination and condensation in the nucleus after 1 h, findings that are hallmarks of cellular apoptosis.

Fig 4.

Electron micrograph of C. albicans cells exposed to caspofungin (0.125 μg/ml) for 1 h (left) and no-drug control cells (right). Note the chromatin margination and condensation along the nucleus (large black dots indicated by a long white arrow) and blebs from the nucleus (indicated by short white arrows), which are hallmark ultrastructural signs of apoptosis, among cells exposed to caspofungin. The cells in the two panels are shown at comparable magnifications. Upon exposure to caspofungin, C. albicans cells become enlarged prior to shrinking during later stages of apoptosis.

Caspofungin effects on metacaspase activation.

In C. albicans, a putative caspase (metacaspase 1) has been shown to be involved in apoptosis (26). We investigated caspase activity in response to caspofungin exposure by staining cells with D2R, which releases green fluorescent rhodamine upon cleavage by caspases. The percentage of C. albicans cells stained by D2R was significantly higher for cells exposed to all concentrations of caspofungin than for control cells at 3 h (Fig. 5). There were no significant differences in D2R staining at the different concentrations (ANOVA).

Fig 5.

Effect of caspofungin on intracellular metacaspase activation, as determined by D2R staining. C. albicans cells exposed to caspofungin (including a subinhibitory concentration) exhibited increased metacaspase activation at 3 h compared to that of control cells.

DISCUSSION

To our knowledge, this is the first study to assess the physiologic mechanisms of C. albicans cell death induced by caspofungin, an echinocandin that inhibits (1,3)-β-d-glucan synthesis in the cell wall. We showed that both cellular apoptosis and necrosis accounted for caspofungin-induced cell death. Our findings suggest that caspofungin and other echinocandins exert their potent fungicidal activity against Candida species by inhibiting cell wall integrity (thereby causing necrosis) in some cells and by inducing others to initiate programmed cell death (i.e., apoptosis).

We used multiple key markers of early and late apoptosis to conclusively implicate this process as a mechanism of caspofungin-induced cell death. Our primary screening assay for apoptosis was annexin V staining, which targets a fundamental apoptotic event: the plasma membrane externalization of phosphatidylserine. Apoptosis was induced within the first hour of caspofungin exposure, and early apoptosis and late apoptosis/necrosis were apparent in 20 to 25% and 5 to 7% of C. albicans cells, respectively, following 3 h exposure. The rate of early apoptosis was corroborated by assays for ROS production (DHR-123 and DHE staining) and mitochondrial membrane potential and the rate of late apoptosis by assays for DNA damage and nuclear fragmentation (TUNEL and DAPI staining). In each of the assays, apoptosis was evident at caspofungin concentrations that were less than or equal to the MIC (0.125 μg/ml). Moreover, electron microscopy of C. albicans cells exposed to 0.125 μg/ml of caspofungin clearly showed hallmark features of apoptosis. Taken together, therefore, the data suggest that caspofungin at concentrations below the MIC causes C. albicans cells to undergo apoptosis but that increasing caspofungin concentrations above the MIC does not further induce programmed cell death responses.

Previous studies have shown that several stimuli, including acetic acid (7), hydrogen peroxide (27), 1,10-phenanthroline metal complexes (28), silver-coumarin complexes (17), diallyl disulfide (29), farsenol (26), lactoferrin (30), defensins (31), and the antifungal agent amphotericin B (9), cause C. albicans cells to undergo apoptosis. Our data demonstrate that cell wall stress stemming from inhibition of (1,3)-β-d-glucan, as caused by caspofungin and Zymolyase (at concentrations ≥0.1 mg/ml), is also a trigger for apoptosis. It is well established that nonlethal exposures to caspofungin rapidly induce compensatory responses by C. albicans, including upregulation of CHS chitin synthase genes and activation of the protein kinase C (PKC) cell wall integrity signaling pathway (32, 33). PKC signaling facilitates cell survival in part through the mitogen-activated protein kinase (MAPK) cascade and cross talk with calcineurin signaling. Therefore, it appears that individual C. albicans cells may take one of the following three paths in response to caspofungin: (i) direct toxicity due to loss of cell wall integrity and cell lysis, (ii) compensatory cell wall remodeling and survival, or (iii) induction of programmed cell death. It is enticing to speculate that C. albicans has evolved mechanisms of programmed cell death (also referred to as “cellular suicide”) to eliminate cells whose vitality is impaired under given environmental conditions as a strategy to conserve resources for healthier cells that are more likely to survive and replicate efficiently (34).

Interestingly, some yeast cells exposed to amphotericin B demonstrate a capacity for resuscitation, although they cannot replicate (35). Moreover, some C. albicans cells with apoptotic markers are able to recover and form colonies on solid medium (36), indicating that early apoptosis is potentially reversible. Along these lines, it is notable that C. albicans cells in our study clearly underwent early apoptosis at a subinhibitory caspofungin concentration (0.06 μg/ml) despite the fact that there was no loss of viability by CFU determinations. In fact, cells exposed to 0.06 μg/ml of caspofungin had evidence of cell wall or plasma membrane damage, as shown by methylene blue and PI staining (40% and 31% of cells at 3 h, respectively). We conclude that C. albicans cells exposed to 0.06 μg/ml of caspofungin retained their viability (as evidenced by the CFU determinations) but had marked impairments of health and vitality (37). Indeed, it has been suggested that measurements of vitality, which assess not only live and dead organisms but also the health status of “vigorous, frail, and injured” cells, may provide a better understanding of antimicrobial activity than colony count assays (35). Of note, percentages of mother and daughter cells that exhibited methylene blue staining at given caspofungin concentrations were similar. The inhibition of newly budding daughter cells is expected, as cell walls are being rapidly synthesized, but the data for mother cells suggest that more-mature cell walls undergo continuous remodeling.

There is accumulating evidence in the baker's yeast Saccharomyces cerevisiae and other eukaryotes that different stimuli induce different apoptotic pathways (38, 39). Caspases are classes of cysteine-aspartic acid proteases regulated at the posttranslational level which, when cleaved, convey a signal in a proteolytic cascade that induces apoptosis and leads to cell death (40). In C. albicans, a putative caspase encoded by metacaspase 1 (CaMCA1) has been shown to be involved in apoptosis (26). In this study, we used D2R, which is cleaved by caspases to release green fluorescence, to show that C. albicans caspases were activated in response to caspofungin exposure. The data imply that apoptosis of C. albicans cells proceeds, at least in part, through caspase-dependent pathways, although further study is needed to demonstrate how metacaspase 1 or other proteases contribute to this process. In addition to caspase-dependent pathways, S. cerevisiae has several caspase-like proteases, such as Kex1 and Esp1, that participate in caspase Yca1p-independent apoptotic cell death. Future studies are indicated to determine if caspase-independent pathways are also involved in regulating apoptosis upon exposure of C. albicans cells to caspofungin.

In conclusion, we demonstrated that caspofungin promoted apoptosis of C. albicans cells at concentrations below the MIC. At the same time, caspofungin also killed C. albicans cells by causing necrosis, which was the predominant physiological mechanism of cell death at concentrations greater than or equal to the MIC. The components of the apoptotic pathway(s) induced by caspofungin await further clarification, as do the roles of apoptosis and necrosis in determining the therapeutic efficacy of the echinocandins. In addition, future studies of caspofungin-resistant C. albicans fks mutant strains may identify off-target effects of the echinocandins that induce apoptosis. Our findings have relevance for understanding the mechanisms by which echinocandins exert anticandidal activity and suggest that the benefits of antifungal agents may be augmented by strategies that induce apoptosis or apoptosis-like cell death.