- Email: [email protected]
Allicin enhances the oxidative damage effect of amphotericin B against Candida albicans
International Journal of Antimicrobial Agents 33 (2009) 258–263
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Allicin enhances the oxidative damage effect of amphotericin B against Candida albicans MaoMao An a , Hui Shen b , YongBing Cao c , JunDong Zhang c , Yun Cai a , Rui Wang a,∗ , YuanYing Jiang c,∗∗ a
Department of Clinical Pharmacology, Chinese People’s Liberation Army General Hospital, 28 Fuxing Road, Beijing 100853, PR China Animal Experiment Center, Fu Wai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, 167 Beilishi Road, Beijing 100037, PR China c Department of Pharmacology, School of Pharmacy, Second Military Medical University, 325 Guo He Road, Shanghai 200433, PR China b
a r t i c l e
i n f o
Article history: Received 25 June 2008 Accepted 18 September 2008 Keywords: Candida albicans Amphotericin B Allicin Reactive oxygen species
a b s t r a c t Amphotericin B (AmB) is the gold standard of antifungal treatment for the most severe invasive mycoses. In addition to the interaction of AmB with ergosterol in the fungi cell membrane, several studies have demonstrated oxidative damage involved in the fungicidal activity of AmB. In this study, allicin, an allyl sulphur compound from garlic, was shown to enhance significantly the effect of AmB against Candida albicans in vitro and in vivo, although allicin did not exert a fungicidal effect. Further study first demonstrated that allicin-mediated oxidative damage, such as phospholipid peroxidation in the plasma membrane, via influencing the defence of C. albicans against oxidative damage may be the cause of the synergistic interaction between allicin and AmB. We envision that a combination of AmB with allicin may prove to be a promising strategy for the therapy of disseminated candidiasis. © 2008 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Candida albicans, the major opportunistic fungal pathogen of humans, is associated with clinical conditions ranging from irritating superficial infections of the oral and vaginal mucosa to life-threatening systemic disease in immunocompromised patients [1–3]. Treatment for candidiasis includes antifungal drugs, mainly amphotericin B (AmB) and the azoles. The polyene macrolide antibiotic AmB is widely used in the treatment of many systemic mycoses. However, owing to its poor permeability across membranes, an increased amount of AmB must be administered to patients in clinical situations, often resulting in severe side effects such as renal damage [4–6]. To lessen the severity of the side effects, AmB is often combined with other antifungal drugs such as the azoles [7–9]; however, coincident with the increased use of antifungal azole derivatives, the incidence of drug resistance has recently been increasing [10,11]. Thus, an investigation of reducing the AmB dose by combining it with a new compound is necessary. It is generally accepted that AmB binds to the ergosterol of the fungi cell membrane and thus causes a change in the permeability of the membrane, eventually killing the cells [12,13]. In addition,
∗ Corresponding author. Tel.: +86 10 6693 7909; fax: +86 10 8821 4425. ∗∗ Co-corresponding author. Tel.: +86 21 2507 0371; fax: +86 21 6549 6501. E-mail addresses: [email protected] (R. Wang), [email protected] (Y. Jiang).
several studies have demonstrated the involvement of oxidative damage induced by reactive oxygen species (ROS) in AmB antifungal activity [12,14–18]. The promotive effect of AmB on the generation of ROS such as hydroxyl radicals induces fungal cell damage via oxidation of proteins, cleavage of DNA and RNA, or phospholipid peroxidation resulting in plasma membrane disruption. Allicin (diallyl thiosulfinate) (Fig. 1) is the main biologically active component of freshly crushed garlic extract [19,20]. Allicin exerts various biological activities such as antimicrobial and anticancer activities in addition to the capacity to lower serum lipid levels, particularly cholesterol levels, and ocular pressure [19–21]. Previous published data indicated that allicin had certain in vitro antifungal activity, but the minimal inhibitory concentration (MIC) was relatively high, limiting its clinical utility. Researchers demonstrated that Cu2+ exerted fungicidal activity by promoting endogenous ROS production, and recent studies found that allicin could enhance the fungicidal activity of Cu2+ [22,23]. As ROS overproduction is also involved in the antifungal activity of AmB, it is suggested that allicin may also be able to enhance the fungicidal activity of AmB via oxidative damage. The present study describes a synergic effect of allicin with AmB against C. albicans in vivo and in vitro as well as suggesting the possible mechanism. The results suggest a role for oxidative damage via promoting the phospholipid peroxidation reaction induced by endogenous ROS in the fungicidal activity of AmB. We first report that allicin-mediated oxidative damage may contribute to the synergistic interaction of allicin and AmB, although further research is needed.
0924-8579/$ – see front matter © 2008 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2008.09.014
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
259
more active agent alone, whilst a change of <2 log10 CFU/mL was considered indifferent [28]. 2.4. Animal studies Fig. 1. Structure of allicin.
2. Materials and methods 2.1. Strains and agents Forty clinical isolates of C. albicans were used in the study; C. albicans ATCC 90028 was used as a quality control strain. One of the clinical isolates (C. albicans 0506108) was used for time–kill curve studies and inducing disseminated candidiasis in mice. Drugs prepared in dimethyl sulphoxide (DMSO) included AmB (Sigma–Aldrich, St Louis, MO) and allicin (Lukangchenxin Pharmaceuticals, Shandong, China). Stock solutions of allicin (2 mg/mL) were stored in the dark at 4 ◦ C for <48 h. Before each experiment, the concentration of the allicin stock solutions was confirmed by a spectrophotometric assay as described by Miron et al., as well as the absence of breakdown products by high-performance liquid chromatography analysis [24,25]. Strains were cultured at 30 ◦ C under constant shaking (200 rpm) in a liquid complete medium (YPD) consisting of 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) dextrose. 2.2. Antifungal susceptibility testing The in vitro MICs of the compounds against all 40 clinical isolates of C. albicans were determined by the microbroth dilution method according to the Clinical and Laboratory Standards Institute (formerly the National Committee for Clinical Laboratory Standards) [26]. The initial concentration of the fungal suspension in RPMI 1640 medium was 103 colony-forming units (CFU)/mL and the final drug concentrations were 1–1024 g/mL for allicin and 0.0117–1.5 g/mL for AmB. Plates were incubated at 35 ◦ C for 24 h. Optical density was measured at 630 nm and background optical densities were subtracted from that of each well. Each isolate was tested in triplicate. The MIC90 and MIC50 values were determined as the lowest concentration of the drugs (alone or in combination) that inhibited growth by 90% or 50%, respectively, compared with that of drug-free wells. The fractional inhibitory concentration index (FICI) is defined as the sum of the MIC of each drug when used in combination divided by the MIC of the drug used alone. Synergy and antagonism were defined by FICIs of ≤0.5 and >4, respectively. A FICI result of >0.5 but ≤4 was considered indifferent [27]. 2.3. Time–kill curve studies Candida albicans 0506108 was grown in YPD medium with vigorous shaking at 30 ◦ C. After diluting an overnight culture with distilled water to 107 cells/mL and 105 cells/mL, cell suspensions were incubated with AmB (0.5 g/mL), allicin (2 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL). For growth determination, the cell suspensions were incubated under a normal oxygen concentration (20%) and a hypoxic condition (1%) at 35 ◦ C. Portions of cell suspensions were withdrawn at predetermined time points (0, 2, 4, 6 and 8 h after incubation with agitation) and streaked on a Sabouraud dextrose agar (SDA) plate. CFUs were determined after incubation for 48 h at 30 ◦ C. The experiment was performed in triplicate. Synergism and antagonism were defined as a respective increase or decrease of ≥2 log10 CFU/mL in antifungal activity produced by the combination compared with that of the
Based on the in vitro data, the synergic effect was determined in an animal model. BALB/c female mice weighing 18–20 g (Center of Experimental Animals, Chinese People’s Liberation Army General Hospital, Beijing, China) were used in the study. Susceptibility to AmB, allicin and the combination of AmB + allicin of C. albicans 0506108 was tested in a murine model of disseminated candidiasis by inoculating 1 × 105 cells per mouse in the lateral tail vein. AmB therapy with a dose of 0.5 mg/kg daily, allicin therapy with a dose of 1 mg/kg daily and combination therapy (AmB 0.5 mg/kg and allicin 1 mg/kg daily) by lateral tail vein injection were initiated at 24 h post challenge. Control mice were given the same volume of sterile saline. After 5 days of treatment, mice were sacrificed and the kidneys were excised by a sterile technique, weighed and homogenised in 2 mL of sterile 0.9% saline. The homogenates were diluted 10-fold in sterile saline and then 0.1 mL of each dilution and the undiluted homogenate were cultured in triplicate on SDA. Culture plates were incubated for 48 h at 30 ◦ C and the number of CFU/g of tissue was calculated. 2.5. Measurement of reactive oxygen species production Intracellular ROS production was measured by a method dependent on intracellular deacylation and the oxidation of 2 ,7 -dichlorodihydrofluorescein diacetate to the corresponding fluorescent compound [29]. Following pre-incubation of C. albicans 0506108 (107 cells/mL) in YPD medium with 40 M 2 ,7 dichlorodihydrofluorescein diacetate at 30 ◦ C for 60 min, the cells were collected by centrifugation and suspended in an equal volume of distilled water. The cell suspensions (1.0 mL) were further treated with each chemical for 4 h and 8 h, respectively, and then washed and re-suspended in 100 L of phosphate-buffered saline. The fluorescence intensity of a cell suspension (100 L) containing 107 cells was measured using a Cytoflow 2300 fluorescence spectrometer (Millipore Co., Billerica, MA) with excitation at 480 nm and emission at 530 nm. The arbitrary units were based directly on fluorescence intensity. 2.6. Measurement of phospholipid peroxidation Phospholipid peroxidation of membranes can be followed by measuring the levels of malondialdehyde (MDA), which is mainly produced by decomposition of lipid hydroperoxides caused by ROS. In the present study, the amount of MDA produced by C. albicans in the presence of either compound was measured by the thiobarbituric acid (TBA)-reactive substances [30]. Overnight cultures of C. albicans 0506108 were grown in a shaker at 30 ◦ C in YPD medium. The cells were then washed twice with distilled deionised sterilised water and then re-suspended in minimal medium to a density of ca. 4 × 107 cells/mL. After 15 min equilibration, AmB (0.5 g/mL), allicin (2 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL) were added. After 4 h or 8 h of incubation at 30 ◦ C, 0.5 mL samples of supernatant were removed. Then, 0.4 mL of 15% (w/v) trichloroacetic acid and 0.8 mL of 0.67% (w/v) TBA in 0.3 N NaOH were added to the supernatant. Samples were heated in a boiling bath for 20 min and, after cooling and adding 1.5 mL of 1-butanol, were centrifuged at 4 ◦ C at 2000 rpm for 5 min to remove cell debris. Fluorescence was measured with EX = 515 nm and EM = 555 nm using a Fluoroskan Ascent FL spectrofluorimeter (Labsystems). Values were referred to
260
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
Table 1 Interaction of amphotericin B (AmB) and allicin against 40 clinical isolates of Candida albicans by the checkerboard microdilution assay.a . MIC50 (g/mL)
AmB Allicin AmB + allicinb FICI
MIC90 (g/mL)
Median
Range
Median
Range
0.375 64 0.0234/1 0.169
0.188–0.75 32–80 <0.0117–0.0938/1–4 0.045–0.893
0.75 128 0.0469/1 0.067
0.375–1.5 64–128 <0.0234–0.0938/1–8 0.023–0.325
MIC50/90 , minimal inhibitory concentration inhibiting growth by 90% or 50%, respectively, compared with that of drug-free wells; FICI, fractional inhibitory concentration index. a When analysed with MIC50 values, synergism was observed in 34 (85%) of 40 isolates and indifference was observed in 6 (15%); at MIC90 values, synergy was observed in all 40 of the isolates. b MICs in combination are expressed as [AmB]/[allicin].
a standard curve using 1,1,3,3-tetramethoxypropane. The value of EM = 555 nm was normalised by C. albicans protein concentration. Candida albicans protein extraction was performed as described and the concentration was determined by the bicinchoninic acid protein assay [31]. The percent ratio of the treated samples to the untreated control was calculated from the results obtained as nmol equivalent of MDA per mg protein. 2.7. Statistical criteria MIC data are reported as the median and range of the concentration of each antifungal agent inhibiting 50% (MIC50 ) and 90% (MIC90 ) of the isolates tested. Other data are expressed as mean ± standard deviation. For statistical analysis, analysis of variance (ANOVA) was used. P < 0.01 was considered statistically significant. 3. Results 3.1. Antifungal activity of amphotericin B in the absence or presence of allicin in vitro The susceptibility of C. albicans to AmB in the absence or presence of allicin was confirmed by determining the lowest drug concentrations that gave 50% (MIC50 ) or 90% (MIC90 ) inhibition. The results of the checkerboard analysis are summarised in Table 1. The
AmB + allicin combination markedly reduced MICs, especially the MIC90 values of either individual agent. Synergism was observed in all 40 isolates (100%) in terms of MIC90 . The corresponding median FICI was 0.067 (range 0.023–0.325). When analysed with MIC50 values instead of MIC90 values, the FICI was generally higher (median 0.169; range 0.045–0.893). Of the 40 isolates, synergism was observed in 34 (85%) and indifference was observed in 6 (15%). Regardless of MIC endpoints, antagonism was not observed with the combination. The synergic effect of allicin with AmB was also confirmed in time–kill curves (Fig. 2). Allicin did not affect the time–kill curve at 2 g/mL after 8 h regardless of the initial inoculum, and the fungistatic activity of AmB was dramatically enhanced by addition of allicin. Given an initial inoculum of 105 CFU/mL, the combination yielded a 2.41 log10 CFU/mL decrease compared with 0.5 g/mL AmB alone at 8 h (Fig. 2A). Under a starting inoculum of 107 CFU/mL, the AmB + allicin combination produced a 3.04 log10 CFU/mL decrease compared with AmB alone at 8 h (Fig. 2B). 3.2. Allicin synergy with amphotericin B against disseminated candidiasis in mice To assess the antifungal activity of AmB combined with allicin, initial experiments were performed by challenging mice (n = 10) with 1 × 105 CFU per mouse and then treating mice with allicin
Fig. 2. Representative time–kill curves of Candida albicans 0506108 obtained using initial inocula of (A) 105 colony-forming units (CFU)/mL and (B) 107 CFU/mL. Candida albicans 0506108 was incubated in distilled water with allicin (2 g/mL) or with amphotericin B (AmB) (0.5 g/mL) in the absence or presence of allicin (2 g/mL), respectively. (♦) Growth control, () allicin, (×) AmB, () AmB + allicin. Data are expressed as the mean of triplicate assays.
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
261
hypoxic conditions. Given an initial inoculum of 105 CFU/mL or 107 CFU/mL, AmB (0.5 g/mL) treatment yielded 0.87 log10 CFU/mL and 0.99 log10 CFU/mL decrease under the normal oxygen and hypoxic conditions, respectively, at 8 h (Figs. 2 and 4), suggesting that AmB activity against C. albicans was partially reduced under hypoxic conditions. Furthermore, the effect of AmB was also not enhanced by addition of allicin at an initial inoculum of either 105 CFU/mL or 107 CFU/mL under hypoxic conditions (Fig. 4), indicating that oxidative damage may be involved in the synergistic interaction of allicin and AmB. 3.4. Allicin-enhanced phospholipid peroxidation of Candida albicans induced by amphotericin B
Fig. 3. Fungal burden in kidney tissues of mice infected with Candida albicans 0506108 and treated with saline control, allicin (1 mg/kg) or with AmB (0.5 mg/kg) in the absence or presence of allicin (1 mg/kg). Data are the mean ± standard deviation of 10 mice. *P < 0.01 compared with the group treated with saline, AmB or allicin. CFU, Colony-forming units.
or AmB alone or the combination of AmB with allicin by intravenous injection. After 5 days of treatment, mice treated with saline control, allicin, AmB alone and AmB + allicin all survived. Fig. 3 presents the fungal burdens in kidneys of mice treated by the above compounds. The combination of AmB (0.5 mg/kg) and allicin (1 mg/kg) significantly reduced the number of CFU/g of kidneys in mice infected by C. albicans 0506108 compared with control mice without treatment, but either AmB at the dose of 0.5 mg/kg daily or allicin at the dose of 1 mg/kg daily was ineffective (P < 0.05) (Fig. 3). 3.3. Lack of a synergistic effect of allicin with amphotericin B against C. albicans under hypoxic conditions To determine the role of oxidative damage in the fungicidal activity of AmB and whether it is involved in the synergistic effect of allicin with AmB, time–kill curves of C. albicans cells to AmB (0.5 g/mL) with or without allicin (2 g/mL) were studied under
Overproduction of ROS in the antifungal activity of AmB has been demonstrated and plasma membrane phospholipids are a major target of ROS [12,14–18,32]. In the present study, ROS and phospholipid peroxidation levels of C. albicans treated with AmB (0.5 g/mL) in the absence and presence of allicin (2 g/mL) were studied. Consistent with previous studies, AmB promoted endogenous ROS production (Fig. 5A) and thus resulted in C. albicans membrane phospholipid peroxidation, as indicated by the MDA level (Fig. 5B). Although the MDA level, representing the phospholipid peroxidation reaction, was significantly elevated in C. albicans treated by AmB in the presence of allicin, ROS production did not change evidently (Fig. 5), indicating that allicin could not accelerate the endogenous ROS production of C. albicans induced by AmB. 4. Discussion AmB is known to complex with sterols in the fungal cell membrane and to induce structural changes that allow small ions to follow across the cell membrane resulting in fungal cell death [12,13]. In addition, the involvement of oxidative damage induced by endogenous ROS has been reported in the activity of AmB against C. albicans [12,14–18]. Sokol-Anderson et al. [15–17] reported that increased levels of intracellular or extracellular catalase, as well as incubation under hypoxic conditions, reduced AmB toxicity toward C. albicans. They also demonstrated the involvement of amplified defence against oxidative damage in the diminished response to
Fig. 4. Representative time–kill curves of Candida albicans 0506108 obtained using initial inocula of (A) 105 colony-forming units (CFU)/mL and (B) 107 CFU/mL under hypoxic conditions (1% oxygen). Candida albicans 0506108 was incubated in distilled water with allicin (2 g/mL) or with amphotericin B (AmB) (0.5 g/mL) in the absence or presence of allicin (2 g/mL) under hypoxic conditions. (♦) Growth control, () allicin, (×) AmB, () AmB + allicin. Data are expressed as the mean of triplicate assays.
262
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
Fig. 5. (A) Endogenous reactive oxygen species (ROS) production and (B) malondialdehyde (MDA) levels of Candida albicans 0506108 after 4 h and 8 h incubated with allicin (2 g/mL), AmB (0.5 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL). All results are the mean ± standard deviation of three independent measurements. Endogenous ROS production of C. albicans 0506108 treated with AmB did not change significantly by the addition of allicin, but the MDA level, which represents the phospholipid peroxidation reaction, was significantly elevated in the presence of allicin.
AmB-induced lethality by using C. albicans mutants lacking membrane ergosterol. In the present study, we also found a diminished fungicidal activity of AmB under hypoxic conditions compared with the normal oxygen concentration (Figs. 2 and 4). Although AmB is the gold standard of antifungal treatment for the most severe invasive mycoses, adverse effects are common, with nephrotoxicity being the most serious, occurring early in the course of treatment [4–6,33,20]. The severe side effects restrict AmB clinical application and this has increased the interest in using drug combinations to enhance its efficacy in order to reduce the dose of AmB required for therapy and thus lessen the severity of side effects. In the present study, we reported that allicin, an allyl sulphur compound from garlic, enhanced the fungicidal activity of AmB against C. albicans in vitro and in vivo (Table 1; Figs. 2 and 3). Although previous studies have indicated the synergistic antifungal activities of this compound [34,35], the present study first demonstrated the synergistic effect in clinical isolates in vitro and in disseminated candidiasis in vivo, and we first reported that oxidative damage may be involved in the synergistic effect between allicin and AmB. The checkerboard microdilution assay was used to examine the relationship between AmB and allicin in terms of growth-inhibitory activities against C. albicans. As shown in Table 1, all the clinical isolates were susceptible to AmB, and allicin also had certain inhibitory activity to the growth of C. albicans although the MIC was relatively
high. The FICI indicated that there was a significantly synergistic interaction between AmB and allicin (Table 1). The time–kill curve studies not only confirmed the above checkerboard microdilution assay result but also suggested that allicin could not exert a lethal effect on C. albicans, as expected from its growth-inhibitory effect in YPD medium (Fig. 2). A mouse model of systemic candidiasis also demonstrated the significant synergistic interaction of allicin and AmB in vivo, which was consistent with the results in vitro (Fig. 3). Both the in vitro and in vivo results indicated that although the anticandidal activity of allicin alone was not as efficient as AmB, allicin strikingly enhanced the killing effect of AmB against C. albicans. Allicin, one of the main effective antibacterial ingredients isolated from garlic, was demonstrated to have weak activity against fungi with little adverse reaction [36]. A previous study reported that allicin could enhance the fungicidal activity of Cu2+ , exerting fungicidal activity by accelerating an endogenous generation of ROS in fungal cells at a lethal concentration [22,37]. We therefore hypothesised that oxidative damage may be also involved in the synergistic interaction, confirmed by the result that no synergistic interaction between AmB and allicin existed when incubating C. albicans under hypoxic conditions (Fig. 4). The target of oxidative damage included proteins, sugars, DNA, RNA and phospholipid. Plasma membrane phospholipids are a major target of oxidative stress and their disruption can be achieved by the peroxidation of fatty acid components to generate lipoperoxide, as seen from the cytotoxic effect of H2 O2 [32]. Phospholipid peroxidation can initiate and spread free radical reactions leading to cell death. Our results demonstrated that allicin significantly enhanced the phospholipid peroxidation reaction induced by AmB (Fig. 5B), also showing that oxidative damage is involved in the synergistic interaction of allicin and AmB. Interestingly, we found that allicin could not accelerated the ROS production of C. albicans induced by AmB (Fig. 5A), indicating that allicin may enhance the fungicidal activity of AmB by defending against oxidative damage of C. albicans. The antimicrobial activities of allicin are considered to depend on its inhibitory effect on certain thiol-containing enzymes via SH-modifying properties, as reflected by the production of S-allylmercaptocysteine from l-cysteine [20,38]. Allicin could decrease cellular glutathione content as a result of its direct interaction with the corresponding SH group and this appears to be a cause of the enhancement of fungal sensitivity to AmB toxicity [19,21]. However, allicin rather exhibits oxidative activity because S-allylmercaptoglutathione, the product of allicin in the reaction with glutathione, still possesses antioxidative activity in addition to SH-modifying activity [21]. As deduced from the cysteine-rich structure of metallothionein, allicin is likely to inactivate the protein by directly interacting with these cysteine residues, resulting in the enhancement of AmB activity against C. albicans. However, metallothionein is not the possible target of allicin, since a previous study indicated that the fungicidal activity of Cd2+ , which exerted an effect via metallothionein, was partially attenuated in the presence of this allyl sulphur compound [22]. We also found that the addition of either N-acetyl cysteine or dithiothreitol could significantly suppress the combined fungicidal activity of allicin and AmB but not the lethal effect of AmB itself (data not shown). Therefore, allicin still likely inhibits a certain thiol-containing molecule other than glutathione or metallothionein that functions in defence against oxidative damage in C. albicans. A previous study suggested that allicin could interfere with the function of alkyl hydroperoxide reductase 1 (AHP1), a thiol-containing molecule, which functions as a defence against plasma membrane phospholipid peroxidation in the fungal cell [22]. Whether AHP1 is a target of allicin in enhancing the fungicidal activity of AmB as well as other possible targets still needed to be further researched.
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
The present study shows that allicin has a synergistic effect with AmB against C. albicans via enhancing the phospholipid peroxidation reaction in vitro and in vivo. The results suggested that allicin was a promising and safe agent to combine with AmB for anticandidal efficacy and to reduce the AmB dose to lessen side effects, although the molecular mechanisms need further study. Funding: This study was supported by the Chinese People’s Liberation Army General Hospital, Beijing, China. Competing interests: None declared. Ethical approval: All animals received humane care in compliance with the institutional animal care guidelines, approved by the Local Institutional Committee. References [1] Ellis M, Richardson M, De Pauw B. Epidemiology. Hosp Med 2000;61:605–9. [2] Garber G. An overview of fungal infections. Drugs 2001;61:1–12. [3] Maertens J, Vrebos M, Boogaerts M. Assessing risk factors for systemic fungal infections. Eur J Cancer Care 2001;10:56–62. [4] Matsuoka S, Murata M. Cholesterol markedly reduces ion permeability induced by membrane-bound amphotericin B. Biochim Biophys Acta 2002;1564:429–34. [5] Fanos V, Cataldi L. Amphotericin B-induced nephrotoxicity: a review. J Chemother 2000;12:463–70. [6] Mayer J, Doubek M, Doubek J, Horky´ D, Scheer P, Stepánek M. Reduced nephrotoxicity of conventional amphotericin B therapy after minimal nephroprotective measures: animal experiments and clinical study. J Infect Dis 2002;186:379–88. [7] Sugar AM, Hitchcock CA, Troke PF, Picard M. Combination therapy of murine invasive candidiasis with fluconazole and amphotericin B. Antimicrob Agents Chemother 1995;39:598–601. [8] Vazquez JA, Arganoza MT, Vaishampayan JK, Akins RA. In vitro interaction between amphotericin B and azoles in Candida albicans. Antimicrob Agents Chemother 1996;40:2511–6. [9] Lewis RE, Lund BC, Klepser ME, Ernst EJ, Pfaller MA. Assessment of antifungal activities of fluconazole and amphotericin B administered alone and in combination against Candida albicans by using a dynamic in vitro mycotic infection model. Antimicrob Agents Chemother 1998;42:1382–6. [10] Cowen LE, Anderson JB, Kohn LM. Evolution of drug resistance in Candida albicans. Annu Rev Microbiol 2002;56:139–65. [11] Perea S, López-Ribot JL, Kirkpatrick WR, McAtee RK, Santillán RA, Martínez M, et al. Prevalence of molecular mechanisms of resistance to azole antifungal agents in Candida albicans strains displaying high-level fluconazole resistance isolated from human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2001;45:2676–84. [12] Brajtburg J, Powderly WG, Kobayashi GS, Medoff G. Amphotericin B: current understanding of mechanisms of action. Antimicrob Agents Chemother 1990;34:183–8. [13] Sperry PJ, Cua DJ, Wetzel SA, Adler-Moore JP. Antimicrobial activity of AmBisome and non-liposomal amphotericin B following uptake of Candida glabrata by murine epidermal Langerhans cells. Med Mycol 1998;36:135–41. [14] Hamilton AJ, Holdom MD. Antioxidant systems in the pathogenic fungi of man and their role in virulence. Med Mycol 1999;37:375–89. [15] Sokol-Anderson ML, Brajtburg J, Medoff G. Amphotericin B-induced oxidative damage and killing of Candida albicans. J Infect Dis 1986;154:76–83. [16] Sokol-Anderson ML, Brajtburg J, Medoff G. Sensitivity of Candida albicans to amphotericin B administered as single or fractionated doses. Antimicrob Agents Chemother 1986;29:701–2.
263
[17] Sokol-Anderson M, Sligh Jr JE, Elberg S, Brajtburg J, Kobayashi GS, Medoff G. Role of cell defense against oxidative damage in the resistance of Candida albicans to the killing effect of amphotericin B. Antimicrob Agents Chemother 1988;32:702–5. [18] Okamoto Y, Aoki S, Mataga I. Enhancement of amphotericin B activity against Candida albicans by superoxide radical. Mycopathologia 2004;158:9–15. [19] Miron T, Rabinkov A, Mirelman D, Wilchek M, Weiner L. The mode of action of allicin: its ready permeability through phospholipid membranes may contribute to its biological activity. Biochim Biophys Acta 2000;1463:20–30. [20] Ankri S, Mirelman D. Antimicrobial properties of allicin from garlic. Microbes Infect 1999;1:125–9. [21] Rabinkov A, Miron T, Mirelman D, Wilchek M, Glozman S, Yavin E, et al. S-Allylmercaptoglutathione: the reaction product of allicin with glutathione possesses SH-modifying and antioxidant properties. Biochim Biophys Acta 2000;1499:144–53. [22] Ogita A, Hirooka K, Yamamoto Y, Tsutsui N, Fujita K, Taniguchi M, et al. Synergistic fungicidal activity of Cu2+ and allicin, an allyl sulfur compound from garlic, and its relation to the role of alkyl hydroperoxide reductase 1 as a cell surface defense in Saccharomyces cerevisiae. Toxicology 2005;215:205–13. [23] Ogita A, Fujita K, Taniguchi M, Tanaka T. Dependence of synergistic fungicidal activity of Cu2+ and allicin, an allyl sulfur compound from garlic, on selective accumulation of the ion in the plasma membrane fraction via allicin-mediated phospholipid peroxidation. Planta Med 2006;72:875–80. [24] Rosen RT, Hiserodt RD, Fukuda EK, Ruiz RJ, Zhou Z, Lech J, et al. Determination of allicin, S-allylcysteine and volatile metabolites of garlic in breath, plasma or simulated gastric fluids. J Nutr 2001;131:968S–71S. [25] Miron T, Rabinkov A, Mirelman D, Weiner L, Wilchek M. A spectrophotometric assay for allicin and alliinase (alliin lyase) activity: reaction of 2-nitro-5thiobenzoate with thiosulfinates. Anal Biochem 1998;265:317–25. [26] National Committee for Clinical Laboratory Standards. Reference method for broth dilution antifungal susceptibility testing of yeasts. Approved standard. Document M27-A2. Wayne, PA NCCLS; 2002. [27] Odds FC. Synergy, antagonism, and what the chequerboard puts between them. J Antimicrob Chemother 2003;52:1. [28] Eliopoulos GM, Moellering RC. Antimicrobial combinations. In: Lorian V, editor. Antibiotics in laboratory medicine. 4th ed. Baltimore, MD: Williams & Wilkins Co.; 1996. p. 330–96. [29] Machida K, Tanaka T, Taniguchi M. Depletion of glutathione as a cause of the promotive effects of polygodial, a sesquiterpene on the production of reactive oxygen species in Saccharomyces cerevisiae. J Biosci Bioeng 1999;88:526–30. [30] Aust SD. Thiobarbituric acid assay reactants. Methods Toxicol 1994;1B:367–74. [31] Horvath A, Riezman H. Rapid protein extraction from Saccharomyces cerevisiae. Yeast 1994;10:1305–10. [32] Gallis HA, Drew RH, Pickard WW. Amphotericin B: 30 years of clinical experience. Rev Infect Dis 1990;12:308–29. [33] Hartsel SC, Benz SK, Ayenew W, Bolard J. Na+ , K+ and Cl– selectivity of the permeability pathways induced through sterol-containing membrane vesicles by amphotericin B and other polyene antibiotics. Eur Biophys J 1994;23: 125–32. [34] Gadd GM. Interactions of fungi with toxic metals. New Phytol 1993;124:25–60. [35] Ogita A, Fujita K, Taniguchi M, Tanaka T. Enhancement of the fungicidal activity of amphotericin B by allicin, an allyl-sulfur compound from garlic, against the yeast Saccharomyces cerevisiae as a model system. Planta Med 2006;72:1247–50. [36] Davis LE, Shen J, Royer RE. In vitro synergism of concentrated Allium sativum extract and amphotericin B against Cryptococcus neoformans. Planta Med 1994;60:546–9. [37] Janero DR, Hreniuk D, Sharif HM. Hydrogen peroxide-induced oxidative stress to the mammalian heart-muscle cell (cardiomyocyte): lethal peroxidative membrane injury. J Cell Physiol 1991;149:347–64. [38] Oommen S, Anto RJ, Srinivas G, Karunagaran D. Allicin (from garlic) induces caspase-mediated apoptosis in cancer cells. Eur J Pharmacol 2004;485:97–103.
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Allicin enhances the oxidative damage effect of amphotericin B against Candida albicans MaoMao An a , Hui Shen b , YongBing Cao c , JunDong Zhang c , Yun Cai a , Rui Wang a,∗ , YuanYing Jiang c,∗∗ a
Department of Clinical Pharmacology, Chinese People’s Liberation Army General Hospital, 28 Fuxing Road, Beijing 100853, PR China Animal Experiment Center, Fu Wai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, 167 Beilishi Road, Beijing 100037, PR China c Department of Pharmacology, School of Pharmacy, Second Military Medical University, 325 Guo He Road, Shanghai 200433, PR China b
a r t i c l e
i n f o
Article history: Received 25 June 2008 Accepted 18 September 2008 Keywords: Candida albicans Amphotericin B Allicin Reactive oxygen species
a b s t r a c t Amphotericin B (AmB) is the gold standard of antifungal treatment for the most severe invasive mycoses. In addition to the interaction of AmB with ergosterol in the fungi cell membrane, several studies have demonstrated oxidative damage involved in the fungicidal activity of AmB. In this study, allicin, an allyl sulphur compound from garlic, was shown to enhance significantly the effect of AmB against Candida albicans in vitro and in vivo, although allicin did not exert a fungicidal effect. Further study first demonstrated that allicin-mediated oxidative damage, such as phospholipid peroxidation in the plasma membrane, via influencing the defence of C. albicans against oxidative damage may be the cause of the synergistic interaction between allicin and AmB. We envision that a combination of AmB with allicin may prove to be a promising strategy for the therapy of disseminated candidiasis. © 2008 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Candida albicans, the major opportunistic fungal pathogen of humans, is associated with clinical conditions ranging from irritating superficial infections of the oral and vaginal mucosa to life-threatening systemic disease in immunocompromised patients [1–3]. Treatment for candidiasis includes antifungal drugs, mainly amphotericin B (AmB) and the azoles. The polyene macrolide antibiotic AmB is widely used in the treatment of many systemic mycoses. However, owing to its poor permeability across membranes, an increased amount of AmB must be administered to patients in clinical situations, often resulting in severe side effects such as renal damage [4–6]. To lessen the severity of the side effects, AmB is often combined with other antifungal drugs such as the azoles [7–9]; however, coincident with the increased use of antifungal azole derivatives, the incidence of drug resistance has recently been increasing [10,11]. Thus, an investigation of reducing the AmB dose by combining it with a new compound is necessary. It is generally accepted that AmB binds to the ergosterol of the fungi cell membrane and thus causes a change in the permeability of the membrane, eventually killing the cells [12,13]. In addition,
∗ Corresponding author. Tel.: +86 10 6693 7909; fax: +86 10 8821 4425. ∗∗ Co-corresponding author. Tel.: +86 21 2507 0371; fax: +86 21 6549 6501. E-mail addresses: [email protected] (R. Wang), [email protected] (Y. Jiang).
several studies have demonstrated the involvement of oxidative damage induced by reactive oxygen species (ROS) in AmB antifungal activity [12,14–18]. The promotive effect of AmB on the generation of ROS such as hydroxyl radicals induces fungal cell damage via oxidation of proteins, cleavage of DNA and RNA, or phospholipid peroxidation resulting in plasma membrane disruption. Allicin (diallyl thiosulfinate) (Fig. 1) is the main biologically active component of freshly crushed garlic extract [19,20]. Allicin exerts various biological activities such as antimicrobial and anticancer activities in addition to the capacity to lower serum lipid levels, particularly cholesterol levels, and ocular pressure [19–21]. Previous published data indicated that allicin had certain in vitro antifungal activity, but the minimal inhibitory concentration (MIC) was relatively high, limiting its clinical utility. Researchers demonstrated that Cu2+ exerted fungicidal activity by promoting endogenous ROS production, and recent studies found that allicin could enhance the fungicidal activity of Cu2+ [22,23]. As ROS overproduction is also involved in the antifungal activity of AmB, it is suggested that allicin may also be able to enhance the fungicidal activity of AmB via oxidative damage. The present study describes a synergic effect of allicin with AmB against C. albicans in vivo and in vitro as well as suggesting the possible mechanism. The results suggest a role for oxidative damage via promoting the phospholipid peroxidation reaction induced by endogenous ROS in the fungicidal activity of AmB. We first report that allicin-mediated oxidative damage may contribute to the synergistic interaction of allicin and AmB, although further research is needed.
0924-8579/$ – see front matter © 2008 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2008.09.014
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
259
more active agent alone, whilst a change of <2 log10 CFU/mL was considered indifferent [28]. 2.4. Animal studies Fig. 1. Structure of allicin.
2. Materials and methods 2.1. Strains and agents Forty clinical isolates of C. albicans were used in the study; C. albicans ATCC 90028 was used as a quality control strain. One of the clinical isolates (C. albicans 0506108) was used for time–kill curve studies and inducing disseminated candidiasis in mice. Drugs prepared in dimethyl sulphoxide (DMSO) included AmB (Sigma–Aldrich, St Louis, MO) and allicin (Lukangchenxin Pharmaceuticals, Shandong, China). Stock solutions of allicin (2 mg/mL) were stored in the dark at 4 ◦ C for <48 h. Before each experiment, the concentration of the allicin stock solutions was confirmed by a spectrophotometric assay as described by Miron et al., as well as the absence of breakdown products by high-performance liquid chromatography analysis [24,25]. Strains were cultured at 30 ◦ C under constant shaking (200 rpm) in a liquid complete medium (YPD) consisting of 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) dextrose. 2.2. Antifungal susceptibility testing The in vitro MICs of the compounds against all 40 clinical isolates of C. albicans were determined by the microbroth dilution method according to the Clinical and Laboratory Standards Institute (formerly the National Committee for Clinical Laboratory Standards) [26]. The initial concentration of the fungal suspension in RPMI 1640 medium was 103 colony-forming units (CFU)/mL and the final drug concentrations were 1–1024 g/mL for allicin and 0.0117–1.5 g/mL for AmB. Plates were incubated at 35 ◦ C for 24 h. Optical density was measured at 630 nm and background optical densities were subtracted from that of each well. Each isolate was tested in triplicate. The MIC90 and MIC50 values were determined as the lowest concentration of the drugs (alone or in combination) that inhibited growth by 90% or 50%, respectively, compared with that of drug-free wells. The fractional inhibitory concentration index (FICI) is defined as the sum of the MIC of each drug when used in combination divided by the MIC of the drug used alone. Synergy and antagonism were defined by FICIs of ≤0.5 and >4, respectively. A FICI result of >0.5 but ≤4 was considered indifferent [27]. 2.3. Time–kill curve studies Candida albicans 0506108 was grown in YPD medium with vigorous shaking at 30 ◦ C. After diluting an overnight culture with distilled water to 107 cells/mL and 105 cells/mL, cell suspensions were incubated with AmB (0.5 g/mL), allicin (2 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL). For growth determination, the cell suspensions were incubated under a normal oxygen concentration (20%) and a hypoxic condition (1%) at 35 ◦ C. Portions of cell suspensions were withdrawn at predetermined time points (0, 2, 4, 6 and 8 h after incubation with agitation) and streaked on a Sabouraud dextrose agar (SDA) plate. CFUs were determined after incubation for 48 h at 30 ◦ C. The experiment was performed in triplicate. Synergism and antagonism were defined as a respective increase or decrease of ≥2 log10 CFU/mL in antifungal activity produced by the combination compared with that of the
Based on the in vitro data, the synergic effect was determined in an animal model. BALB/c female mice weighing 18–20 g (Center of Experimental Animals, Chinese People’s Liberation Army General Hospital, Beijing, China) were used in the study. Susceptibility to AmB, allicin and the combination of AmB + allicin of C. albicans 0506108 was tested in a murine model of disseminated candidiasis by inoculating 1 × 105 cells per mouse in the lateral tail vein. AmB therapy with a dose of 0.5 mg/kg daily, allicin therapy with a dose of 1 mg/kg daily and combination therapy (AmB 0.5 mg/kg and allicin 1 mg/kg daily) by lateral tail vein injection were initiated at 24 h post challenge. Control mice were given the same volume of sterile saline. After 5 days of treatment, mice were sacrificed and the kidneys were excised by a sterile technique, weighed and homogenised in 2 mL of sterile 0.9% saline. The homogenates were diluted 10-fold in sterile saline and then 0.1 mL of each dilution and the undiluted homogenate were cultured in triplicate on SDA. Culture plates were incubated for 48 h at 30 ◦ C and the number of CFU/g of tissue was calculated. 2.5. Measurement of reactive oxygen species production Intracellular ROS production was measured by a method dependent on intracellular deacylation and the oxidation of 2 ,7 -dichlorodihydrofluorescein diacetate to the corresponding fluorescent compound [29]. Following pre-incubation of C. albicans 0506108 (107 cells/mL) in YPD medium with 40 M 2 ,7 dichlorodihydrofluorescein diacetate at 30 ◦ C for 60 min, the cells were collected by centrifugation and suspended in an equal volume of distilled water. The cell suspensions (1.0 mL) were further treated with each chemical for 4 h and 8 h, respectively, and then washed and re-suspended in 100 L of phosphate-buffered saline. The fluorescence intensity of a cell suspension (100 L) containing 107 cells was measured using a Cytoflow 2300 fluorescence spectrometer (Millipore Co., Billerica, MA) with excitation at 480 nm and emission at 530 nm. The arbitrary units were based directly on fluorescence intensity. 2.6. Measurement of phospholipid peroxidation Phospholipid peroxidation of membranes can be followed by measuring the levels of malondialdehyde (MDA), which is mainly produced by decomposition of lipid hydroperoxides caused by ROS. In the present study, the amount of MDA produced by C. albicans in the presence of either compound was measured by the thiobarbituric acid (TBA)-reactive substances [30]. Overnight cultures of C. albicans 0506108 were grown in a shaker at 30 ◦ C in YPD medium. The cells were then washed twice with distilled deionised sterilised water and then re-suspended in minimal medium to a density of ca. 4 × 107 cells/mL. After 15 min equilibration, AmB (0.5 g/mL), allicin (2 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL) were added. After 4 h or 8 h of incubation at 30 ◦ C, 0.5 mL samples of supernatant were removed. Then, 0.4 mL of 15% (w/v) trichloroacetic acid and 0.8 mL of 0.67% (w/v) TBA in 0.3 N NaOH were added to the supernatant. Samples were heated in a boiling bath for 20 min and, after cooling and adding 1.5 mL of 1-butanol, were centrifuged at 4 ◦ C at 2000 rpm for 5 min to remove cell debris. Fluorescence was measured with EX = 515 nm and EM = 555 nm using a Fluoroskan Ascent FL spectrofluorimeter (Labsystems). Values were referred to
260
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
Table 1 Interaction of amphotericin B (AmB) and allicin against 40 clinical isolates of Candida albicans by the checkerboard microdilution assay.a . MIC50 (g/mL)
AmB Allicin AmB + allicinb FICI
MIC90 (g/mL)
Median
Range
Median
Range
0.375 64 0.0234/1 0.169
0.188–0.75 32–80 <0.0117–0.0938/1–4 0.045–0.893
0.75 128 0.0469/1 0.067
0.375–1.5 64–128 <0.0234–0.0938/1–8 0.023–0.325
MIC50/90 , minimal inhibitory concentration inhibiting growth by 90% or 50%, respectively, compared with that of drug-free wells; FICI, fractional inhibitory concentration index. a When analysed with MIC50 values, synergism was observed in 34 (85%) of 40 isolates and indifference was observed in 6 (15%); at MIC90 values, synergy was observed in all 40 of the isolates. b MICs in combination are expressed as [AmB]/[allicin].
a standard curve using 1,1,3,3-tetramethoxypropane. The value of EM = 555 nm was normalised by C. albicans protein concentration. Candida albicans protein extraction was performed as described and the concentration was determined by the bicinchoninic acid protein assay [31]. The percent ratio of the treated samples to the untreated control was calculated from the results obtained as nmol equivalent of MDA per mg protein. 2.7. Statistical criteria MIC data are reported as the median and range of the concentration of each antifungal agent inhibiting 50% (MIC50 ) and 90% (MIC90 ) of the isolates tested. Other data are expressed as mean ± standard deviation. For statistical analysis, analysis of variance (ANOVA) was used. P < 0.01 was considered statistically significant. 3. Results 3.1. Antifungal activity of amphotericin B in the absence or presence of allicin in vitro The susceptibility of C. albicans to AmB in the absence or presence of allicin was confirmed by determining the lowest drug concentrations that gave 50% (MIC50 ) or 90% (MIC90 ) inhibition. The results of the checkerboard analysis are summarised in Table 1. The
AmB + allicin combination markedly reduced MICs, especially the MIC90 values of either individual agent. Synergism was observed in all 40 isolates (100%) in terms of MIC90 . The corresponding median FICI was 0.067 (range 0.023–0.325). When analysed with MIC50 values instead of MIC90 values, the FICI was generally higher (median 0.169; range 0.045–0.893). Of the 40 isolates, synergism was observed in 34 (85%) and indifference was observed in 6 (15%). Regardless of MIC endpoints, antagonism was not observed with the combination. The synergic effect of allicin with AmB was also confirmed in time–kill curves (Fig. 2). Allicin did not affect the time–kill curve at 2 g/mL after 8 h regardless of the initial inoculum, and the fungistatic activity of AmB was dramatically enhanced by addition of allicin. Given an initial inoculum of 105 CFU/mL, the combination yielded a 2.41 log10 CFU/mL decrease compared with 0.5 g/mL AmB alone at 8 h (Fig. 2A). Under a starting inoculum of 107 CFU/mL, the AmB + allicin combination produced a 3.04 log10 CFU/mL decrease compared with AmB alone at 8 h (Fig. 2B). 3.2. Allicin synergy with amphotericin B against disseminated candidiasis in mice To assess the antifungal activity of AmB combined with allicin, initial experiments were performed by challenging mice (n = 10) with 1 × 105 CFU per mouse and then treating mice with allicin
Fig. 2. Representative time–kill curves of Candida albicans 0506108 obtained using initial inocula of (A) 105 colony-forming units (CFU)/mL and (B) 107 CFU/mL. Candida albicans 0506108 was incubated in distilled water with allicin (2 g/mL) or with amphotericin B (AmB) (0.5 g/mL) in the absence or presence of allicin (2 g/mL), respectively. (♦) Growth control, () allicin, (×) AmB, () AmB + allicin. Data are expressed as the mean of triplicate assays.
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
261
hypoxic conditions. Given an initial inoculum of 105 CFU/mL or 107 CFU/mL, AmB (0.5 g/mL) treatment yielded 0.87 log10 CFU/mL and 0.99 log10 CFU/mL decrease under the normal oxygen and hypoxic conditions, respectively, at 8 h (Figs. 2 and 4), suggesting that AmB activity against C. albicans was partially reduced under hypoxic conditions. Furthermore, the effect of AmB was also not enhanced by addition of allicin at an initial inoculum of either 105 CFU/mL or 107 CFU/mL under hypoxic conditions (Fig. 4), indicating that oxidative damage may be involved in the synergistic interaction of allicin and AmB. 3.4. Allicin-enhanced phospholipid peroxidation of Candida albicans induced by amphotericin B
Fig. 3. Fungal burden in kidney tissues of mice infected with Candida albicans 0506108 and treated with saline control, allicin (1 mg/kg) or with AmB (0.5 mg/kg) in the absence or presence of allicin (1 mg/kg). Data are the mean ± standard deviation of 10 mice. *P < 0.01 compared with the group treated with saline, AmB or allicin. CFU, Colony-forming units.
or AmB alone or the combination of AmB with allicin by intravenous injection. After 5 days of treatment, mice treated with saline control, allicin, AmB alone and AmB + allicin all survived. Fig. 3 presents the fungal burdens in kidneys of mice treated by the above compounds. The combination of AmB (0.5 mg/kg) and allicin (1 mg/kg) significantly reduced the number of CFU/g of kidneys in mice infected by C. albicans 0506108 compared with control mice without treatment, but either AmB at the dose of 0.5 mg/kg daily or allicin at the dose of 1 mg/kg daily was ineffective (P < 0.05) (Fig. 3). 3.3. Lack of a synergistic effect of allicin with amphotericin B against C. albicans under hypoxic conditions To determine the role of oxidative damage in the fungicidal activity of AmB and whether it is involved in the synergistic effect of allicin with AmB, time–kill curves of C. albicans cells to AmB (0.5 g/mL) with or without allicin (2 g/mL) were studied under
Overproduction of ROS in the antifungal activity of AmB has been demonstrated and plasma membrane phospholipids are a major target of ROS [12,14–18,32]. In the present study, ROS and phospholipid peroxidation levels of C. albicans treated with AmB (0.5 g/mL) in the absence and presence of allicin (2 g/mL) were studied. Consistent with previous studies, AmB promoted endogenous ROS production (Fig. 5A) and thus resulted in C. albicans membrane phospholipid peroxidation, as indicated by the MDA level (Fig. 5B). Although the MDA level, representing the phospholipid peroxidation reaction, was significantly elevated in C. albicans treated by AmB in the presence of allicin, ROS production did not change evidently (Fig. 5), indicating that allicin could not accelerate the endogenous ROS production of C. albicans induced by AmB. 4. Discussion AmB is known to complex with sterols in the fungal cell membrane and to induce structural changes that allow small ions to follow across the cell membrane resulting in fungal cell death [12,13]. In addition, the involvement of oxidative damage induced by endogenous ROS has been reported in the activity of AmB against C. albicans [12,14–18]. Sokol-Anderson et al. [15–17] reported that increased levels of intracellular or extracellular catalase, as well as incubation under hypoxic conditions, reduced AmB toxicity toward C. albicans. They also demonstrated the involvement of amplified defence against oxidative damage in the diminished response to
Fig. 4. Representative time–kill curves of Candida albicans 0506108 obtained using initial inocula of (A) 105 colony-forming units (CFU)/mL and (B) 107 CFU/mL under hypoxic conditions (1% oxygen). Candida albicans 0506108 was incubated in distilled water with allicin (2 g/mL) or with amphotericin B (AmB) (0.5 g/mL) in the absence or presence of allicin (2 g/mL) under hypoxic conditions. (♦) Growth control, () allicin, (×) AmB, () AmB + allicin. Data are expressed as the mean of triplicate assays.
262
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
Fig. 5. (A) Endogenous reactive oxygen species (ROS) production and (B) malondialdehyde (MDA) levels of Candida albicans 0506108 after 4 h and 8 h incubated with allicin (2 g/mL), AmB (0.5 g/mL) or the combination of AmB (0.5 g/mL) and allicin (2 g/mL). All results are the mean ± standard deviation of three independent measurements. Endogenous ROS production of C. albicans 0506108 treated with AmB did not change significantly by the addition of allicin, but the MDA level, which represents the phospholipid peroxidation reaction, was significantly elevated in the presence of allicin.
AmB-induced lethality by using C. albicans mutants lacking membrane ergosterol. In the present study, we also found a diminished fungicidal activity of AmB under hypoxic conditions compared with the normal oxygen concentration (Figs. 2 and 4). Although AmB is the gold standard of antifungal treatment for the most severe invasive mycoses, adverse effects are common, with nephrotoxicity being the most serious, occurring early in the course of treatment [4–6,33,20]. The severe side effects restrict AmB clinical application and this has increased the interest in using drug combinations to enhance its efficacy in order to reduce the dose of AmB required for therapy and thus lessen the severity of side effects. In the present study, we reported that allicin, an allyl sulphur compound from garlic, enhanced the fungicidal activity of AmB against C. albicans in vitro and in vivo (Table 1; Figs. 2 and 3). Although previous studies have indicated the synergistic antifungal activities of this compound [34,35], the present study first demonstrated the synergistic effect in clinical isolates in vitro and in disseminated candidiasis in vivo, and we first reported that oxidative damage may be involved in the synergistic effect between allicin and AmB. The checkerboard microdilution assay was used to examine the relationship between AmB and allicin in terms of growth-inhibitory activities against C. albicans. As shown in Table 1, all the clinical isolates were susceptible to AmB, and allicin also had certain inhibitory activity to the growth of C. albicans although the MIC was relatively
high. The FICI indicated that there was a significantly synergistic interaction between AmB and allicin (Table 1). The time–kill curve studies not only confirmed the above checkerboard microdilution assay result but also suggested that allicin could not exert a lethal effect on C. albicans, as expected from its growth-inhibitory effect in YPD medium (Fig. 2). A mouse model of systemic candidiasis also demonstrated the significant synergistic interaction of allicin and AmB in vivo, which was consistent with the results in vitro (Fig. 3). Both the in vitro and in vivo results indicated that although the anticandidal activity of allicin alone was not as efficient as AmB, allicin strikingly enhanced the killing effect of AmB against C. albicans. Allicin, one of the main effective antibacterial ingredients isolated from garlic, was demonstrated to have weak activity against fungi with little adverse reaction [36]. A previous study reported that allicin could enhance the fungicidal activity of Cu2+ , exerting fungicidal activity by accelerating an endogenous generation of ROS in fungal cells at a lethal concentration [22,37]. We therefore hypothesised that oxidative damage may be also involved in the synergistic interaction, confirmed by the result that no synergistic interaction between AmB and allicin existed when incubating C. albicans under hypoxic conditions (Fig. 4). The target of oxidative damage included proteins, sugars, DNA, RNA and phospholipid. Plasma membrane phospholipids are a major target of oxidative stress and their disruption can be achieved by the peroxidation of fatty acid components to generate lipoperoxide, as seen from the cytotoxic effect of H2 O2 [32]. Phospholipid peroxidation can initiate and spread free radical reactions leading to cell death. Our results demonstrated that allicin significantly enhanced the phospholipid peroxidation reaction induced by AmB (Fig. 5B), also showing that oxidative damage is involved in the synergistic interaction of allicin and AmB. Interestingly, we found that allicin could not accelerated the ROS production of C. albicans induced by AmB (Fig. 5A), indicating that allicin may enhance the fungicidal activity of AmB by defending against oxidative damage of C. albicans. The antimicrobial activities of allicin are considered to depend on its inhibitory effect on certain thiol-containing enzymes via SH-modifying properties, as reflected by the production of S-allylmercaptocysteine from l-cysteine [20,38]. Allicin could decrease cellular glutathione content as a result of its direct interaction with the corresponding SH group and this appears to be a cause of the enhancement of fungal sensitivity to AmB toxicity [19,21]. However, allicin rather exhibits oxidative activity because S-allylmercaptoglutathione, the product of allicin in the reaction with glutathione, still possesses antioxidative activity in addition to SH-modifying activity [21]. As deduced from the cysteine-rich structure of metallothionein, allicin is likely to inactivate the protein by directly interacting with these cysteine residues, resulting in the enhancement of AmB activity against C. albicans. However, metallothionein is not the possible target of allicin, since a previous study indicated that the fungicidal activity of Cd2+ , which exerted an effect via metallothionein, was partially attenuated in the presence of this allyl sulphur compound [22]. We also found that the addition of either N-acetyl cysteine or dithiothreitol could significantly suppress the combined fungicidal activity of allicin and AmB but not the lethal effect of AmB itself (data not shown). Therefore, allicin still likely inhibits a certain thiol-containing molecule other than glutathione or metallothionein that functions in defence against oxidative damage in C. albicans. A previous study suggested that allicin could interfere with the function of alkyl hydroperoxide reductase 1 (AHP1), a thiol-containing molecule, which functions as a defence against plasma membrane phospholipid peroxidation in the fungal cell [22]. Whether AHP1 is a target of allicin in enhancing the fungicidal activity of AmB as well as other possible targets still needed to be further researched.
M. An et al. / International Journal of Antimicrobial Agents 33 (2009) 258–263
The present study shows that allicin has a synergistic effect with AmB against C. albicans via enhancing the phospholipid peroxidation reaction in vitro and in vivo. The results suggested that allicin was a promising and safe agent to combine with AmB for anticandidal efficacy and to reduce the AmB dose to lessen side effects, although the molecular mechanisms need further study. Funding: This study was supported by the Chinese People’s Liberation Army General Hospital, Beijing, China. Competing interests: None declared. Ethical approval: All animals received humane care in compliance with the institutional animal care guidelines, approved by the Local Institutional Committee. References [1] Ellis M, Richardson M, De Pauw B. Epidemiology. Hosp Med 2000;61:605–9. [2] Garber G. An overview of fungal infections. Drugs 2001;61:1–12. [3] Maertens J, Vrebos M, Boogaerts M. Assessing risk factors for systemic fungal infections. Eur J Cancer Care 2001;10:56–62. [4] Matsuoka S, Murata M. Cholesterol markedly reduces ion permeability induced by membrane-bound amphotericin B. Biochim Biophys Acta 2002;1564:429–34. [5] Fanos V, Cataldi L. Amphotericin B-induced nephrotoxicity: a review. J Chemother 2000;12:463–70. [6] Mayer J, Doubek M, Doubek J, Horky´ D, Scheer P, Stepánek M. Reduced nephrotoxicity of conventional amphotericin B therapy after minimal nephroprotective measures: animal experiments and clinical study. J Infect Dis 2002;186:379–88. [7] Sugar AM, Hitchcock CA, Troke PF, Picard M. Combination therapy of murine invasive candidiasis with fluconazole and amphotericin B. Antimicrob Agents Chemother 1995;39:598–601. [8] Vazquez JA, Arganoza MT, Vaishampayan JK, Akins RA. In vitro interaction between amphotericin B and azoles in Candida albicans. Antimicrob Agents Chemother 1996;40:2511–6. [9] Lewis RE, Lund BC, Klepser ME, Ernst EJ, Pfaller MA. Assessment of antifungal activities of fluconazole and amphotericin B administered alone and in combination against Candida albicans by using a dynamic in vitro mycotic infection model. Antimicrob Agents Chemother 1998;42:1382–6. [10] Cowen LE, Anderson JB, Kohn LM. Evolution of drug resistance in Candida albicans. Annu Rev Microbiol 2002;56:139–65. [11] Perea S, López-Ribot JL, Kirkpatrick WR, McAtee RK, Santillán RA, Martínez M, et al. Prevalence of molecular mechanisms of resistance to azole antifungal agents in Candida albicans strains displaying high-level fluconazole resistance isolated from human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2001;45:2676–84. [12] Brajtburg J, Powderly WG, Kobayashi GS, Medoff G. Amphotericin B: current understanding of mechanisms of action. Antimicrob Agents Chemother 1990;34:183–8. [13] Sperry PJ, Cua DJ, Wetzel SA, Adler-Moore JP. Antimicrobial activity of AmBisome and non-liposomal amphotericin B following uptake of Candida glabrata by murine epidermal Langerhans cells. Med Mycol 1998;36:135–41. [14] Hamilton AJ, Holdom MD. Antioxidant systems in the pathogenic fungi of man and their role in virulence. Med Mycol 1999;37:375–89. [15] Sokol-Anderson ML, Brajtburg J, Medoff G. Amphotericin B-induced oxidative damage and killing of Candida albicans. J Infect Dis 1986;154:76–83. [16] Sokol-Anderson ML, Brajtburg J, Medoff G. Sensitivity of Candida albicans to amphotericin B administered as single or fractionated doses. Antimicrob Agents Chemother 1986;29:701–2.
263
[17] Sokol-Anderson M, Sligh Jr JE, Elberg S, Brajtburg J, Kobayashi GS, Medoff G. Role of cell defense against oxidative damage in the resistance of Candida albicans to the killing effect of amphotericin B. Antimicrob Agents Chemother 1988;32:702–5. [18] Okamoto Y, Aoki S, Mataga I. Enhancement of amphotericin B activity against Candida albicans by superoxide radical. Mycopathologia 2004;158:9–15. [19] Miron T, Rabinkov A, Mirelman D, Wilchek M, Weiner L. The mode of action of allicin: its ready permeability through phospholipid membranes may contribute to its biological activity. Biochim Biophys Acta 2000;1463:20–30. [20] Ankri S, Mirelman D. Antimicrobial properties of allicin from garlic. Microbes Infect 1999;1:125–9. [21] Rabinkov A, Miron T, Mirelman D, Wilchek M, Glozman S, Yavin E, et al. S-Allylmercaptoglutathione: the reaction product of allicin with glutathione possesses SH-modifying and antioxidant properties. Biochim Biophys Acta 2000;1499:144–53. [22] Ogita A, Hirooka K, Yamamoto Y, Tsutsui N, Fujita K, Taniguchi M, et al. Synergistic fungicidal activity of Cu2+ and allicin, an allyl sulfur compound from garlic, and its relation to the role of alkyl hydroperoxide reductase 1 as a cell surface defense in Saccharomyces cerevisiae. Toxicology 2005;215:205–13. [23] Ogita A, Fujita K, Taniguchi M, Tanaka T. Dependence of synergistic fungicidal activity of Cu2+ and allicin, an allyl sulfur compound from garlic, on selective accumulation of the ion in the plasma membrane fraction via allicin-mediated phospholipid peroxidation. Planta Med 2006;72:875–80. [24] Rosen RT, Hiserodt RD, Fukuda EK, Ruiz RJ, Zhou Z, Lech J, et al. Determination of allicin, S-allylcysteine and volatile metabolites of garlic in breath, plasma or simulated gastric fluids. J Nutr 2001;131:968S–71S. [25] Miron T, Rabinkov A, Mirelman D, Weiner L, Wilchek M. A spectrophotometric assay for allicin and alliinase (alliin lyase) activity: reaction of 2-nitro-5thiobenzoate with thiosulfinates. Anal Biochem 1998;265:317–25. [26] National Committee for Clinical Laboratory Standards. Reference method for broth dilution antifungal susceptibility testing of yeasts. Approved standard. Document M27-A2. Wayne, PA NCCLS; 2002. [27] Odds FC. Synergy, antagonism, and what the chequerboard puts between them. J Antimicrob Chemother 2003;52:1. [28] Eliopoulos GM, Moellering RC. Antimicrobial combinations. In: Lorian V, editor. Antibiotics in laboratory medicine. 4th ed. Baltimore, MD: Williams & Wilkins Co.; 1996. p. 330–96. [29] Machida K, Tanaka T, Taniguchi M. Depletion of glutathione as a cause of the promotive effects of polygodial, a sesquiterpene on the production of reactive oxygen species in Saccharomyces cerevisiae. J Biosci Bioeng 1999;88:526–30. [30] Aust SD. Thiobarbituric acid assay reactants. Methods Toxicol 1994;1B:367–74. [31] Horvath A, Riezman H. Rapid protein extraction from Saccharomyces cerevisiae. Yeast 1994;10:1305–10. [32] Gallis HA, Drew RH, Pickard WW. Amphotericin B: 30 years of clinical experience. Rev Infect Dis 1990;12:308–29. [33] Hartsel SC, Benz SK, Ayenew W, Bolard J. Na+ , K+ and Cl– selectivity of the permeability pathways induced through sterol-containing membrane vesicles by amphotericin B and other polyene antibiotics. Eur Biophys J 1994;23: 125–32. [34] Gadd GM. Interactions of fungi with toxic metals. New Phytol 1993;124:25–60. [35] Ogita A, Fujita K, Taniguchi M, Tanaka T. Enhancement of the fungicidal activity of amphotericin B by allicin, an allyl-sulfur compound from garlic, against the yeast Saccharomyces cerevisiae as a model system. Planta Med 2006;72:1247–50. [36] Davis LE, Shen J, Royer RE. In vitro synergism of concentrated Allium sativum extract and amphotericin B against Cryptococcus neoformans. Planta Med 1994;60:546–9. [37] Janero DR, Hreniuk D, Sharif HM. Hydrogen peroxide-induced oxidative stress to the mammalian heart-muscle cell (cardiomyocyte): lethal peroxidative membrane injury. J Cell Physiol 1991;149:347–64. [38] Oommen S, Anto RJ, Srinivas G, Karunagaran D. Allicin (from garlic) induces caspase-mediated apoptosis in cancer cells. Eur J Pharmacol 2004;485:97–103.