Silver nanoparticles induce apoptotic cell death in Candida albicans through the increase of hydroxyl radicals

Authors


D. G. Lee, School of Life Sciences and Biotechnology, College of Natural Sciences, Kyungpook National University, Daehak-ro 80, Buk-gu, Daegu 702-701, Korea
Fax: +82 53 955 5522
Tel: +82 53 950 5373
E-mail: dglee222@knu.ac.kr

Abstract

Silver nanoparticles have been shown to be detrimental to fungal cells although the mechanism(s) of action have not been clearly established. In this study, we used Candida albicans cells to show that silver nanoparticles exert their antifungal effect through apoptosis. Many studies have shown that the accumulation of reactive oxygen species induces and regulates the induction of apoptosis. Furthermore, hydroxyl radicals are considered an important component of cell death. Therefore, we assumed that hydroxyl radicals were related to apoptosis and the effect of thiourea as a hydroxyl radical scavenger was investigated. We measured the production of reactive oxygen species and investigated whether silver nanoparticles induced the accumulation of hydroxyl radicals. A reduction in the mitochondrial membrane potential shown by flow cytometry analysis and the release of cytochrome c from mitochondria were also verified. In addition, the apoptotic effects of silver nanoparticles were detected by fluorescence microscopy using other confirmed diagnostic markers of yeast apoptosis including phosphatidylserine externalization, DNA and nuclear fragmentation, and the activation of metacaspases. Cells exposed to silver nanoparticles showed increased reactive oxygen species and hydroxyl radical production. All other phenomena of mitochondrial dysfunction and apoptotic features also appeared. The results indicate that silver nanoparticles possess antifungal effects with apoptotic features and we suggest that the hydroxyl radicals generated by silver nanoparticles have a significant role in mitochondrial dysfunctional apoptosis.

Abbreviations
DHR-123

dihydrorhodamine

DiOC6(3)

3,3′-dihexyloxacarbocyanine iodide

FITC

fluorescein isothiocyanate

H2O2

hydrogen peroxide

HPF

2-[6-(4′-hydroxy) phenoxy-3H-xanthen-3-on-9-yl]-benzoic acid

nano-Ag

silver nanoparticles

MIC

minimum inhibitory concentration

PI

propidium iodide

ROS

reactive oxygen species

TUNEL

terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling

Introduction

It has been known since ancient times that silver and its compounds are effective antimicrobial agents [1,2]. In the 19th century, microbial infections were treated with 0.5% AgNO3, which was also used for the prevention of infections in burns. When the era of antibiotics began with the discovery of penicillin, the use of silver slowly declined [3]. Currently, due to the appearance of micro-organisms insensitive to conventional drugs, the use of silver for treating infections has once again gained importance. However, the use of silver ions has one major drawback; they are easily inactivated by complexation and precipitation and the use of silver ions has therefore been limited [4].

Here, silver nanoparticles (nano-Ag), which are not electrocharged, can be a valuable alternative to ionic silver [5]. Nano-Ag are clusters of silver atoms that range in diameter from 1 to 100 nm and are attracting interest as antibacterial and antimicrobial agents. In particular, because of recent advances in research on metal nanoparticles, nano-Ag have received special attention as a possible antimicrobial agent. Nano-Ag are known to be a nontoxic and safe antibacterial agents for the human body. In addition, nano-Ag have also been reported to possess antifungal activity [6], anti-inflammatory properties [7], antiviral activity [8] and anti-angiogenic activity [9]. Although the antimicrobial effects of nano-Ag are well known, their mechanisms of action have been addressed only sporadically in the literature. Recent studies have shown that nano-Ag interact with three main components of micro-organisms to produce the antimicrobial effect: the membrane or cell wall [6,10], DNA [11] and microbial proteins [10]. In addition, there is substantial evidence that nano-Ag produce reactive oxygen species (ROS) [12]. The accumulation of intracellular ROS is well known as an important regulator of apoptosis accumulating in the early apoptosis phase [13]. Subsequently, the level of intracellular ROS accumulation increases, which initiates mitochondrial fragmentation [14]. Some other studies have shown that hydroxyl radicals are linked to cell death [15]. Because apoptosis is one of the mechanisms of cell death, we investigated whether there are any connections between apoptosis and hydroxyl radicals.

Candida albicans is probably one of the most successful opportunistic pathogens in humans. Under conditions of a weakened immune system, colonizing C. albicans can become opportunistic, causing recurrent mucosal infections and life-threatening contagious infections with high mortality rates. Furthermore, the number of known multidrug resistant bacteria and fungi is increasing rapidly. Thus, the development of more effective antifungal therapies is of great importance. Understanding the mechanisms and decisions of cell death in fungi may provide new developments in the search for diverse novel antifungal nanoparticles.

According to previously reported studies, nano-Ag possess antifungal effects and cell-cycle analysis has shown significantly arrested cell cycles during the G2/M phase [6]. There are many studies showing G2/M-phase-mediated apoptosis [16]. For these reasons, we investigated whether nano-Ag could exert apoptotic cell death in C. albicans and found a relationship between mitochondrial dysfunction and hydroxyl radicals, which was induced by nano-Ag, during apoptotic cell death.

Results

Intracellular ROS accumulation

In a previous study, nano-Ag showed anticandidal activity against C. albicans (Fig. 1). This substance exhibited a minimum inhibitory concentration (MIC) value of 2 μg·mL−1, which was as efficient as that of 3 mm hydrogen peroxide (H2O2) on C. albicans (data not shown). We used H2O2 as a positive control to determine programmed cell death [16].

Figure 1.

 Transmission electron micrograph of the nano-Ag used in this work. The bar marker represents 20 nm.

ROS are continuously formed because of cellular oxygen metabolism. Recent studies have suggested that the accumulation of ROS induces and regulates the induction of apoptosis in metazoans and yeasts [17]. Therefore, to determine the production and accumulation of intracellular ROS induced by nano-Ag, we chose to use the ROS-sensitive dye dihydrorhodamine (DHR-123), which has been used previously as a general indicator of cellular ROS levels. Multiple ROS directly oxidize DHR-123 to the highly stable, fluorescent derivative rhodamine-123 in such a way that an increase in the fluorescent signal reflects ROS production [18]. Cells treated with nano-Ag exhibited high ROS levels compared with untreated cells. In the positive control, there was a significant increase in the amount of fluorescence when the cells were treated with H2O2 (Fig. 2).

Figure 2.

 Flow cytometric analysis of ROS accumulation in nano-Ag (blue) and H2O2 (red solid line) treated C. albicans cells stained with DHR-123.

First, we investigated the activity of nano-Ag for chemically generated ROS. The iron-catalyzed Haber–Weiss process is known to be a promoter of oxygen radicals under aerobic conditions. Ferritin, the iron storage protein, is the principal reservoir for iron within the cell [19]. For this reason, we used ferrous perchlorate as a positive control in solely aqueous solution. To detect hydroxyl radicals (OH) formed in the Fenton reaction, we used the fluorescent dye 2-[6-(4′-hydroxy)phenoxy-3H-xanthen-3-on-9-yl]-benzoic acid (HPF). The fluorescence intensity did not increase upon the addition of H2O2 alone, but did increase substantially upon the addition of nano-Ag or ferrous perchlorate in the presence of H2O2. The results clearly show that nano-Ag could transmute H2O2 into OH (Fig. 3A). We thought that nano-Ag induces apoptotic cell death through the formation of highly ROS such as OH.

Figure 3.

 (A) Detection of hydroxyl radicals in the Fenton reaction using HPF (final 5 μm; 0.1% dimethylformamide as a cosolvent). The fluorescence intensity was determined at 515 nm with excitation at 490 nm. Nano-Ag (lower solid line) and ferrous perchlorate (upper dotted line) were added at 40 s. (B) Flow cytometric analysis of the formation of hydroxyl radicals in C. albicans using the dye HPF. (a) Control, (b) cells exposed to nano-Ag, (c) cells exposed nano-Ag with thiourea, (d) cells exposed to H2O2.

We examined OH formation with HPF, which is oxidized by OH with high specificity, because hydroxyl radicals have been suggested to be a crucial component of apoptosis in many studies [20]. Consistent with the increase in intracellular ROS, the level of intracellular OH was markedly increased in nano-Ag-treated cells (Fig. 3B). These results indicate that ROS induced by nano-Ag accumulated in the interior of C. albicans cells, and most were converted into the strong oxidant OH, considered to be a significant factor in aging and apoptosis in yeast cells. To demonstrate that thiourea acts as an OH scavenger, we also treated cells exposed to nano-Ag with thiourea. Thiourea significantly reduced OH formation in nano-Ag treated cells (Fig. 3B,c). We used thiourea in subsequent experiments to show the effect of decreased hydroxyl radicals on mitochondria-mediated apoptotic cell death.

Measurement of mitochondrial membrane potential (ΔΨm)

In many systems, apoptosis is associated with loss of the mitochondrial inner membrane potential (ΔΨm), which may be regarded as a limiting factor in the apoptotic pathway. Reduction of ΔΨm is among the changes encountered during the early reversible stages of apoptosis and is preceded by cytochrome c release in several cell types [21,22].

To investigate whether nano-Ag decreased ΔΨm, we used the mitochondria-specific voltage-dependent dye 3,3′-dihexyloxacarbocyanine iodide, DiOC6(3), which aggregates inside healthy mitochondria and fluoresces green. When the mitochondrial membrane depolarizes, the dye no longer accumulates and is distributed throughout the cell, resulting in a decrease in green fluorescence. The results show that nano-Ag-treated cells had a decreased ΔΨm, which was in agreement with the pattern induced by H2O2 treatments as the positive control (Fig. 4A). However, cells that were treated with nano-Ag and thiourea did not undergo substantial changes (Fig. 4A,a).

Figure 4.

 (A) Loss of the mitochondrial inner membrane potential in C. albicans induced by treatment with nano-Ag (a), and H2O2 (b) for 1 h. In each panel, the untreated control is the black background peak and the red solid lines represent individual treatment with nano-Ag or H2O2 only. Nano-Ag treatment with thiourea is shown by the blue solid lines (a). Cells were stained with DiOC6 and the fluorescence was measured by flow cytometry. A decrease in fluorescent signal (shift to the left) corresponds with a loss in the mitochondrial membrane potential. (B) Quantitative mitochondrial membrane potential of C. albicans stained by JC-1 and measured by FACS. The area under the horizontal line displays cells with decreased membrane potential. (a) Control, (b) cells exposed to nano-Ag, (c) cells exposed to nano-Ag with thiourea, (d) cells exposed to H2O2. (C) Detection of cytochrome c released from C. albicans mitochondria following the incubation with nano-Ag. Cytosol was ultracentrifuged and the supernatants were subjected to SDS/PAGE and western blotting for released cytochrome c. The untreated control (lane a) or cells cultured in nano-Ag (lane b), nano-Ag treated with thiourea (lane c), and H2O2 (lane d).

We performed the mitochondrial ΔΨm assay with JC-1 to verify our results. JC-1 has advantages over other cationic dyes in that it can selectively enter the mitochondria and reversibly change color from red to green as the membrane potential decreases. In healthy cells with high mitochondrial ΔΨm, JC-1 spontaneously forms complexes known as J-aggregates with intense red fluorescence. However, in apoptotic or unhealthy cells with low ΔΨm, JC-1 remains in the monomeric form, which shows only green fluorescence [23]. The ratio of green to red fluorescence is dependent only on the membrane potential and not on other factors such as mitochondrial size, shape and density, which may influence single-component fluorescence signals. Flow cytometric analysis of JC-1 fluorescence is best performed using 2D green versus red fluorescence plots. As shown in Fig. 4B, both nano-Ag and H2O2 treatments induced a significant decrease in ΔΨm, whereas the combined treatment with nano-Ag and thiourea appeared to have only a slight effect. Therefore, the results suggest that nano-Ag induced the breakdown of ΔΨm, which is a critical step in cells undergoing apoptosis, and the loss of mitochondrial permeability. This result suggests that restriction of OH formation helps maintain the balance of the mitochondrial membrane.

Cytochrome c release

Translocation of cytochrome c from the mitochondria to the cytosol is a pivotal event in apoptotic cell death. Cytochrome c, a soluble protein electrostatically bound to the outer face of the inner mitochondrial membrane, is an essential component of the respiratory chain acting as an electron carrier between the cytochrome bc1 and cytochrome c oxidase complex [24]. We assumed that cytochrome c would be detected in the cytosol because of the results of the previous mitochondrial membrane potential assay. In this regard, we investigated whether nano-Ag-treated cells could induce cytochrome c release from the mitochondria. A large amount of cytochrome c was detected in the cytosolic buffer medium following the nano-Ag-treated cells, although cytochrome c rarely appeared in supernatants that were additionally treated with thiourea (Fig. 4C). These results show that nano-Ag induced the release of cytochrome c from the mitochondria and suggest that the mitochondria of nano-Ag-treated cells, which suppressed the formation of OH by thiourea, are not directly affected by the OH.

Annexin V–propidium iodide double staining

The early stages of apoptotic phenomenon can be detected with fluorescein isothiocyanate (FITC)–Annexin V staining, which binds to phosphatidylserine with high affinity in the presence of Ca2 + [25], combined with the membrane-impermeable dye propidium iodide (PI). Phosphatidylserine is only distributed in the inner leaflet of the lipid bilayer of the plasma membrane, which is maintained by the ATP-binding cassette transporters in C. albicans. To determine whether nano-Ag could induce apoptotic features, the FITC–Annexin V and PI double-staining method was used.

As shown in Fig. 5, the cell population in the lower right (LR) quadrant, which corresponds to the percentage of early apoptotic cells (Annexin V-positive and PI-negative), increased to 35.87% and 45.37% after treating the cells with nano-Ag and H2O2, respectively, for 1 h. Curiously, the percentage of nano-Ag treated with thiourea did not increase significantly as when treated solely with nano-Ag. To show the distinct difference, we drew a bar graph showing the percentage of apoptotic cells at the bottom. These results demonstrate that it is possible for nano-Ag to induce apoptotic cell death in C. albicans cells. Hence, it was confirmed that the generation and accumulation of intracellular ROS, specifically hydroxyl radicals, induced by nano-Ag was related to an apoptotic mechanism in C. albicans cells.

Figure 5.

 Effect of nano-Ag on the exposition of phosphatidylserine at the cytoplasmic membrane. C. albicans cells. Protoplasts were harvested, stained with FITC–Annexin V and PI, and observed with a FACS. The bottom bar graph shows the percentage of apoptotic cells. (A) Control, (B) cells exposed to nano-Ag, (C) cells exposed nano-Ag with thiourea, (D) cells exposed to H2O2.

Measurement of DNA damage

To further confirm the apoptotic features induced in nano-Ag-treated C. albicans cells, a terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay was conducted to detect apoptotic DNA fragmentation by labeling 3′-OH termini with modified nucleotides catalyzed by terminal deoxynucleotidyltransferase. The labeling of breaks in the DNA by TUNEL, a reliable method for the identification of apoptotic cells, is utilized to visualize the apoptotic phenotype of cells [26].

A strong blue fluorescence indicated a greater degree of typical apoptotic DNA condensation and fragmentation in the nuclei of C. albicans cells exposed to nano-Ag than in the intact nuclei of normal control cells. 4′-6-Diamidino-2-phenylindole staining of the nano-Ag-treated cells showed the distributed nuclear fragments (Fig. 6A). Similar results were obtained by TUNEL assay staining of the breaks in the DNA nuclear strands during the late stages of apoptosis. TUNEL-positive cells, which showed a strong green fluorescence or intense green fluorescent spots, were observed in the population treated with nano-Ag (Fig. 6B). In untreated cultures, the nucleus appeared as a single round spot in the cells (Fig. 6A,a) or did not show up well against the backgrounds (Fig. 6B,a). Candida albicans is known to activate programmed cell death with features reminiscent of apoptosis in response to a variety of environmental stimuli such as H2O2 [26–28]. For this reason, we used cells treated with H2O2 as a positive control. Supporting our observations, exposure of C. albicans cells to nano-Ag resulted in apoptotic DNA damage. Furthermore, we ascertained the oxidative stress-protecting effects of thiourea.

Figure 6.

 DNA and nuclear fragmentation were shown by 4′-6-diamidino-2-phenylindole (A) and TUNEL (B) staining. Effect of nano-Ag on the activity of metacaspase in C. albicans (C). Nano-Ag-treated cells were collected, stained and observed under a fluorescent microscope. (a) Control, (b) cells exposed to nano-Ag, (c) cells exposed nano-Ag with thiourea, (d) cells exposed to H2O2.

Measurement of metacaspase activation

Caspases are typically activated in the early stages of apoptosis and they play a central role in the apoptotic signaling network. Although caspases are not present in fungi, orthologs of caspases in animals, termed metacaspases, have been identified in fungi and plants, and their activity can be assessed using the same detection marker [29,30]. In order to confirm metacaspase activation, cells were incubated with the CaspACE™ FITC–VAD–FMK in situ marker that binds to the active site of metacaspases, and detected using a fluorescence microscope. Cells with intracellular active metacaspases stained fluorescent green, whereas nonapoptotic cells appeared unstained. Fluorescence analysis of the cells treated with nano-Ag showed a significant green fluorescence in the FITC–VAD–FMK-loaded cells that was consistent with the positive control treated with H2O2 (Fig. 6C). In addition, the number of activated metacaspases decreased, which also reduced OH formation in thiourea-treated cells (Fig. 6C,c), as expected. These results suggest that nano-Ag treatment did initially lead to significant generation of strong oxidant hydroxyl radicals, which are well-known to be important regulators of yeast apoptosis, and then the hydroxyl radicals activated the metacaspases.

Discussion

Apoptosis is a highly regulated cellular suicide program crucial for development and homeostasis in metazoan organisms, resulting in the removal of unwanted, mutated, damaged or simply dispensable cells without an inflammatory reaction occurring [31,32]. Apoptosis has been accepted as a process that is not exclusive to multicellular organisms, but rather is a universal mechanism of cell elimination operating according to a basic program, including in simpler and more ancient forms of single-celled eukaryotes. The full apoptotic program comprises two phases, one of which has necrotic features [33]. Therefore, we analyzed the more definitive signs of the apoptosis process in this study.

ROS, such as inline image, H2O2 and OH, are considered to be crucial regulators of aging, and their accumulation has been proven to play a key role in apoptosis [17]. We used DHR-123 to determine ROS accumulation during exposure to nano-Ag in C. albicans. Nano-Ag-treated cells displayed increased intracellular ROS levels compared with untreated cells (Fig. 2). In addition, ROS damaged iron–sulfur clusters, making ferrous iron available for oxidation by the Fenton reaction and these events appear to be mediated through the Tricarboxylic acid cycle and the transient depletion of NADH [34]. The Fenton reaction leads to OH formation, and OH damages DNA, proteins and lipids, resulting in cell death. OH is the neutral form of the hydroxide ion. OH is highly reactive and consequently causes damage to oxidative cells. The Haber–Weiss reaction generates OH from H2O2 and superoxide (inline image) [19]. This reaction can occur in cells and is therefore a possible source of oxidative stress. The reaction is very slow, but is catalyzed by iron. For this reason, we thought it possible that nano-Ag induces OH formation as an iron catalyst. As expected, the fluorescence intensity increased substantially upon the addition of nano-Ag in the presence of H2O2 (Fig. 3A). After that, we examined the intracellular levels of hydroxyl radicals treated with nano-Ag and tried to learn how the thiourea impacts OH accumulation in C. albicans cells treated with nano-Ag. We used thiourea as a scavenger of OH. Thiourea is a potent OH scavenger that has an established means of mitigating the effects of OH damage in both eukaryotes and prokaryotes [35–37]. The results showed that C. albicans cells treated with nano-Ag produced hydroxyl radicals, and thiourea was accompanied by a reduction in OH formation (Fig. 3B).

Several other studies have linked cytochrome c release, ROS formation and changes in the mitochondrial membrane potential to yeast apoptosis [38,39]. During apoptosis, the decrease in ΔΨm is caused by the opening of membrane pores that are located in the mitochondrial membrane. Consequently, the decrease in ΔΨm leads to the translocation and activation of various proapoptotic factors. Reduction of the mitochondrial inner membrane potential (ΔΨm) is among the changes encountered during the early reversible stages of apoptosis and is related to cytochrome c release [21,22]. Thus, we determined ΔΨm. The results showed that mitochondrial permeability in nano-Ag-treated cells was damaged by the breakdown of ΔΨm (Fig. 4A,B). By contrast, cells with hydroxyl radical accumulation inhibited by thiourea did not show substantial changes. The contents of cytochrome c released into the cytosol and mitochondrial membrane depolarization were measured to understand the influence of substances on the intrinsic pathway. Cytochrome c, which is located in the mitochondrial membrane, is released into the cytosol during the early phases of apoptosis and a caspase-cascade is then activated as a representative of the other apoptotic protease [40]. As a result of defects in the mitochondrial electron transport system, cytochrome c is reduced when it is released into the cytosol because of the loss of the cytochrome c oxidase activity. Upon the release of cytochrome c into the cytoplasm, the protein binds to apoptotic protease-activating factor [38]. The release of cytochrome c requires an increase in the permiability of the mitochondrial outer membrane. The increase in the mitochondrial transmembrane potential, which has been predicted to promote osmotic matrix swelling, is associated with one model for cytochrome c release from the mitochondria during apoptosis. Because the mitochondrial inner membrane, with its numerous cristae, has a considerably larger surface area than that of the outer membrane, expansion of the inner membrane upon matrix swelling can break the outer membrane, which would be expected to trigger the release of cytochrome c to the cytosol [41]. Treatment with nano-Ag enhanced the content of cytosolic cytochrome c in C. albicans cells (Fig. 4C), suggesting that nano-Ag may trigger cytochrome  c-mediated intrinsic apoptosis. As expected, the addition of thiourea to nano-Ag-treated cells, which do not produce hydroxyl radicals ordinarily, exhibited reduced cytochrome c release compared with those treated with only nano-Ag. Thus, we believe that nano-Ag induces apoptosis through the formation OH and that OH is important to the apoptotic process.

Furthermore, we investigated a series of normally apoptotic properties including the exposition of phosphatidylserine, DNA and nuclear fragmentation, and the activity of metacaspases finally.

To discriminate between apoptotic and necrotic cells, FITC–Annexin V and PI double staining were used [25]. Candida albicans cells exposed to nano-Ag stained Annexin V-positive and PI-negative, which was similar to the response to H2O2, an inducer of apoptosis in yeast cells (Fig. 5). However, cells exposed nano-Ag with thiourea showed decreasing apoptotic features, which seemed to be protected by the thiourea.

In addition, we treated cells with nano-Ag and monitored the proportion of cells positively stained for 4′-6-diamidino-2-phenylindole and TUNEL staining to study the development of the apoptotic phenotype, including DNA and nuclei change (Fig. 6A,B). Finally, cells exposed to nano-Ag exhibited metacaspase activity, but cells treated with nano-Ag and thiourea did not show any activity (Fig. 6C). These phenomena indicate that nano-Ag induces apoptosis in C. albicans and that highly reactive hydroxyl radicals are important to apoptosis triggered by nano-Ag.

In conclusion, this study demonstrated for the first time that nano-Ag promotes apoptosis in C. albicans through phosphatidylserine exposure, DNA damage and the activation of metacaspases. Ultimately, nano-Ag disrupts the mitochondrial integrity and induces cytochrome c release. Although the mechanisms of nano-Ag in mitochondria-dependent apoptosis in C. albicans have not been fully elucidated, this report supports that nano-Ag induces programmed cell death through ROS accumulation, especially OH. As shown in Fig. 3, nano-Ag had the ability to generate OH and cells treated with thiourea decreased OH production. Consequently, the reduction in OH accumulation contributed to diminished mitochondrial dysfunction-mediated apoptosis. We conclude that nano-Ag induce apoptotic cell death in C. albicans through OH generation, which deserves further study to provide elaboration on the apoptosis mechanisms of nano-Ag.

Materials and methods

Reagents and culture conditions

The H2O2 and thiourea used in this study were purchased from the Sigma Chemical Co. (St. Louis, MO, USA). Nano-Ag were stored at 4 °C. Candida albicans (ATCC 90028) cells were cultured in YPD broth (Difco, Franklin Lakes, NJ, USA) containing yeast extract, peptone and dextrose (50 g·L−1) with aeration at 28 °C.

Preparation of nano-Ag

One hundred grams of solid silver were dissolved in 100 mL of 100% nitric acid at 90 °C, and 1 L of distilled water was added. By adding sodium chloride to the silver solution, the Ag ions were precipitated and then clustered together to form monodispersed nanoparticles in an aqueous medium. The sizes and morphology of the nano-Ag were examined by TEM (H-7600; Hitachi Ltd, Tokyo, Japan). The results showed that nano-Ag was spherical in form and its average size was 3 nm (Fig. 1). Because the final concentration of colloidal silver was 60 000 p.p.m., this solution was diluted, and then used to investigate the apoptotic antifungal effects.

Intracellular ROS accumulation

Intracellular ROS production and the accumulation of hydroxyl radicals (OH) were measured using the fluorescent dye DHR-123 and HPF. In a previous study, nano-Ag showed significant antifungal activity at low concentrations, which was similar to the level of amphotericin B [6]. Since then, we have determined the most efficient concentration of H2O2 for the induction of apoptosis [16]. Cells (2 × 108·mL−1) were treated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 1 h at 28 °C, based on the MIC value as a criterion (data not shown). After incubation, the cells were washed with NaCl/Pi before being stained with 5 μg·mL−1 DHR-123 and analyzed using a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA).

The reactivity of nano-Ag for ROS was compared with ferrous perchlorate [Fe(ClO4)2], which was used as the Fenton reaction. We tried to detect OH formed in the Fenton reaction, using HPF. Five micromoles of HPF was added to sodium phosphate buffer (0.1 m, pH 7.4) containing 3 mm H2O2 and then 2 μg·mL−1 nano-Ag or 100 μm ferrous perchlorate was added. The OH formation was detected as an increase in HPF fluorescence by a Spectrofluorometer (Shimadzu RF-5301PC; Shimadzu, Japan) at 490 nm excitation and 515 nm emission wavelength.

The intracellular OH accumulation was measured by incubating the cells with 2 μg·mL−1 nano-Ag and 3 mm H2O2 in NaCl/Pi containing 5 μm using the dye HPF for 1 h at 28 °C. Subsequently, the cells were washed twice in NaCl/Pi and analyzed by flow cytometry [42]. For the OH quenching experiments, 150 mm of thiourea was added simultaneously with nano-Ag. Thiourea has been used at mm levels in vitro as a OH scavenger [43]. Thiourea was used for all subsequent tests.

Measurement of mitochondrial membrane potential (ΔΨm)

Fungal mitochondrial membrane depolarization was analyzed by DiOC6(3) staining. Cells (2 × 108·mL−1) were harvested and incubated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 1 h at 28 °C. Subsequently, the cells were washed with NaCl/Pi and incubated with 2 ng·mL−1 of DiOC6(3) for 30 min. Cells were analyzed by flow cytometer.

JC-1 (Molecular Probes, Carlsbad, CA, USA) is a mitochondrial dye (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine chloride) that stains mitochondria in living cells in a membrane potential-dependent fashion. JC-1 was also used to confirm the decrease in membrane potential and the number of mitochondria specifically. Cells (2  × 108·mL−1) were treated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 1 h at 28 °C. Treated cells were washed in NaCl/Pi, suspended in 200 μL staining solution containing 2 μg·mL−1 of JC-1 for 20 min at 37 °C. The cells were centrifuged at 500 g for 5 min and then the pellet was resuspended with 1 mL NaCl/Pi. Cells were then analyzed by flow cytometer.

Cytochrome c release

To investigate cytochrome c release from the mitochondria, isolations of mitochondria were prepared [44]. Candida albicans cells were cultured in 500 mL of YPD medium for 24 h at 30 °C, collected by centrifugation at 500 g, and washed twice with NaCl/Pi and once with 1 m sorbitol. These cells were treated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 2 h at 28 °C. The treated cells were lysed with lysis buffer (150 mm sodium chloride, 1% Triton X-100, 1 mm EDTA, 1 mm EGTA, 50 mm Tris, pH 8) and then centrifuged at 2000 g for 10 min to remove the cell debris and unbroken cells. The supernatants were collected and centrifuged at 40 000 g for 1 h. The supernatants were collected to assay for cytochrome c released from the mitochondira to the cytoplasm. The protein content of these supernatants was estimated using a NanoVue Plus Spectrophotometer (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Each sample equivalent to 50 μg of protein was resolved on 12% SDS/PAGE. Separated proteins were transferred to a nitrocellulose membrane and analyzed by western blotting with rabbit polyclonal anti-(yeast cytochrome c) [45]. Horseradish peroxidase-linked goat anti-(rabbit IgG) was used as the secondary antibody, and enhanced-chemiluminescence substrate was used for the detection of cytochrome c.

Annexin V–PI double staining

Protoplasts of C. albicans were stained with FITC-labeled Annexin V and PI using the FITC–Annexin V apoptosis detection kit. Cells (2 × 108·mL−1) were digested for 1 h at 28 °C in a potassium phosphate buffer (pH 6.0) containing 20 mg·mL−1 lysing enzyme and 1 m sorbitol. Protoplasts were incubated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 1 h at 28 °C, based on the MIC value as a criterion, and incubated for 20 min in an Annexin-binding buffer containing 5 μL FITC–Annexin V·mL−1 and PI. Protoplasts were then examined by a FACSCalibur flow cytometer (Becton Dickinson).

Measurement of DNA damages

DNA strand breaks in C. albicans cells were analyzed by TUNEL [46]. Cells (2 × 108·mL−1) treated for 2 h with 2 μg·mL−1 nano-Ag and 3 mm H2O2, were washed in NaCl/Pi, permeabilized for 2 min on ice and washed again with a NaCl/Pi. DNA ends were labeled with an in situ cell death detection kit for 1 h at 37 °C. The stained cells were observed with a fluorescence microscope.

Nuclear condensation and fragmentation were analyzed by 4′-6-diamidino-2-phenylindole staining [47]. Cells were treated with 2 μg·mL−1 nano-Ag and 3 mm H2O2 for 2 h and then collected. For nuclear staining, cells were washed twice with NaCl/Pi, permeabilized in a permeabilization solution (0.1% Triton X-100 and 0.1% sodium citrate) and incubated with 1 μg·mL−1 of 4′-6-diamidino-2-phenylindole in the dark for 20 min. Cells were then examined by a fluorescence microscope.

Measurement of metacaspase activation

Activated metacaspases in C. albicans were measured using the CaspACE™ FITC–VAD–FMK in situ marker (Promega). Briefly, each substance treated cell was washed in NaCl/Pi, suspended in 200 μL staining solution containing 10 μm of CaspACE™ FITC–VAD–FMK in situ marker and incubated for 30 min at room temperature in the dark. Cells were then washed once and suspended in NaCl/Pi. Sample analysis was performed with a fluorescence microscope, the Axio Imager A1, and Axio Cam MR5 (Carl Zeiss, Thornwood, NY, USA).

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2011-0000915) and by the Next-Generation BioGreen 21 Program (No. PJ008158), Rural Development Administration, Republic of Korea.

Ancillary