2-Amino-nonyl-6-methoxyl-tetralin muriate activity against Candida albicans augments endogenous reactive oxygen species production – a microarray analysis study Rong Mei Liang1,2, Xiao Lan Yong2, Yun Ping Jiang2, Yong Hong Tan3, Bao Di Dai1, Shi Hua Wang3, Ting Ting Hu2, Xi Chen2, Nan Li2, Zhao Hui Dong2, Xiao Chun Huang2, Jun Chen2, Yong Bing Cao1 and Yuan Ying Jiang1 Drug Development Center, School of Pharmacy, Second Military Medical University, Shanghai, China Department of Clinical Pharmacy, General Hospital of Chengdu Military Command Region, Chengdu, China Department of Pharmacy, General Hospital of Chengdu Military Command Region, Chengdu, China Keywords 10b; action mechanism; Candida albicans; microarray analysis; ROS Correspondence Y B Cao and Y Y Jiang, Drug Development Center, School of Pharmacy, Second Military Medical University, 325 Guohe Road, Shanghai 200433, China Fax: +86 021 6549 0641 Tel: +86 021 8187 1357; +86 021 8187 1275 E-mail: jiangyysmmu@163.com; ybcao@vip.sina.com (Received 21 October 2010, revised 14 December 2010, accepted 18 January 2011) doi:10.1111/j.1742-4658.2011.08021.x Candida infections have become an increasingly significant problem, mainly because of the widespread nature of Candida and drug resistance There is an urgent need to develop new classes of drugs for the treatment of opportunistic Candida infections, especially in medically complex patients Previous studies have confirmed that 2-amino-nonyl-6-methoxyl-tetralin muriate (10b) possesses powerful antifungal activity in vitro against Candia albicans To clarify the underlying action mechanism, an oligonucleotide microarray study was performed in C albicans SC5314 without and with 10b treatment The analytical results showed that energy metabolism-related genes, including glycolysis-related genes (PFK1, CDC19 and HXK2), fermentation-related genes (PDC11, ALD5 and ADH1) and respiratory electron transport chain-related genes (CBP3, COR1 and QCR8), were downregulated significantly Functional analysis revealed that 10b treatment increased the generation of endogenous reactive oxygen species, and decreased mitochondrial membrane potential, ubiquinone–cytochrome c reductase (complex III) activity and intracellular ATP levels in C albicans SC5314 Also, addition of the antioxidant ascorbic acid reduced the antifungal activity of 10b significantly These results suggest that mitochondrial aerobic respiration shift and endogenous reactive oxygen species augmentation might contribute to the antifungal activity of 10b against C albicans This information may prove to be useful for the development of new strategies to treat Candida infections Introduction Candida infections have become a serious medical problem, because of the high incidence and mortality in AIDS patients, transplant recipients and other immunosuppressed individuals [1–4] Despite continuous expansion of the arsenal of antifungal drugs, the antifungal drugs available cannot meet the increas- ing requirements for managing infections in medically complex patients 2-Amino-nonyl-6-methoxyl-tetralin muriate (10b), a 2-aminotetralin derivative, was synthesized as an antifungal agent with a novel chemical structural (Fig 1) [5] Tetralin derivatives are known to be potentially Abbreviations AA, ascorbic acid; CFU, colony-forming unit; DCFH-DA, 2,7-dichlorofluorescein diacetate; FI, fluorescence intensity; MCZ, miconazole; ROS, reactive oxygen species; DWm, mitochondrial membrane potential; 10b, 2-amino-nonyl-6-methoxyl-tetralin muriate FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS 1075 10b and endogenous reactive oxygen species R M Liang et al Results CH3O Response of gene expression to 10b exposure NH(CH2)8CH3.HCl Fig Chemical structure of 10b applicable to psychiatry, and dialkylated tetralin derivatives are the most effective [6] Aminotetralins, including 8-hydroxy-2-(di-n-propylamino)-tetralin and 7-hydroxy-2-(di-n-propylamino)-tetralin, behave as preferential agonists at serotonin (5-hydroxytryptamine)1A and dopamine D3 and D2 receptors [7] The former affects intracranial self-stimulation, and the latter possess anxiolytic properties However, there are few reports from other laboratories describing the antifungal activities of 2-aminotetralin derivatives Our previous studies [5,8] indicated that 10b possessed high antifungal activity In an oophorectomized female SD rat model of Candida albicans infections, 10b consistently exhibited better antifungal activity than itraconazole [5] Also, 10b significantly reduced the ergostrol content by inhibiting the activity of sterol C-14 reductase, encoded by ERG24 in the ergosterol biosynthetic pathway [8] However, C albicans ERG24 is not required for growth The erg24 mutant of C albicans is capable of growth under normal aerobic conditions on standard defined and enriched media There is a suggestion that the significant level of ergosta-8,14,22trienol accumulated by C albicans erg24 mutants may be the element that allows growth under normal conditions [9], implying that the major action mechanism of 10b against C albicans is not correlated with inhibition of the activity of sterol C-14 reductase in the ergosterol biosynthetic pathway The present study was a continuation of our previous studies, in an attempt to clarify the mechanism of action of 10b against C albicans through analyzing the gene expression profiles of C albicans SC5314 without and with 10b treatment, using oligonucleotide microarray assays and real-time RT-PCR assays It was found that a series of differentially expressed genes were involved in energy metabolism, oxidoreduction and other biological functions In addition, measurements of endogenous reactive oxygen species (ROS) generation, mitochondrial membrane potential (DWm), intracellular ATP concentration, respiratory electron transport chain complex III (ubiquinone–cytochrome c reductase) activity and the effect of antioxidant ascorbic acid on the antifungal activity of 10b suggested that the antifungal activity of 10b against C albicans might be related to mitochondrial aerobic respiration shift and endogenous ROS augmentation 1076 A total of 957 differentially expressed genes were found upon exposure to 10b; the expression of 457 genes was decreased, and the expression of 500 genes was increased Forty-five downregulated genes and nine upregulated genes were found to be related to energy metabolism Of the 45 downregulated genes, 34 were involved in glycolysis (e.g PFK1, CDC19 and HXK2), fermentation (e.g., PDC11, ALD5 and ADH1), the respiratory electron transport chain (e.g CBP3, COR1 and QCR8) and ROS scavenging (e.g GPX2) Of the nine upregulated genes, five were related to fermentation (e.g ADH3 and ADH5) and ROS scavenging (e.g GPX1, SOD5 and SOD6) An additional 29 upregulated genes and 15 downregulated genes were concerned with lipid metabolism Of the 29 upregulated genes, nine were directly linked to ergosterol biosynthesis In addition, 93 (20.35%) of the 457 downregulated genes were related to translation, whereas only two genes in this category were upregulated, suggesting that the translation level might be lower in SC5314 cells exposed to 10b than in controls (Tables S1 and S2) Validation of microarray data by real-time RT-PCR Knowing that augmentation of endogenous ROS production was directly related to the antifungal activity of some antifungal drugs [10–12], we were particularly interested in energy metabolism-related genes Therefore, real-time RT-PCR analysis was designed to detect genes related to energy metabolism Real-time RTPCR reactions were performed in triplicate with independent RNA isolations As shown in Fig 2, the expression levels of glycosis-related genes, PFK1, PFK2, HXK2 and CDC19, decreased significantly by 30.30-fold, 43.48-fold, 20.83-fold and 20.41-fold, respectively; the expression levels of fermentationrelated genes, ALD5, ADH1 and PDC11, decreased significantly by 15.63-fold, 2.20-fold and 83.33-fold respectively; and the expression levels of genes coding for mitochondrial respiratory chain complex III, CBP3 (333.33-fold), COR1 (9.62-fold), QCR2 (8.70-fold), QCR8 (5.43-fold), CYT1 (11.24-fold) and RIP1 (12.5fold), were also markedly decreased Also, the expression level of GPX2 decreased significantly, by 12.05-fold, whereas the expression levels of SOD5 and GPX1 increased by 38.686-fold and 5.433-fold, respectively In addition, the expression levels of ADH3 and ADH5, two fermentation-related genes, also increased FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS R M Liang et al Log2 ratio –1 –2 –3 –4 –5 –6 –7 –8 –9 –10 A D H A LD PD C C D C 19 H XK PF K PF K C B P3 C O R R IP C YT Q C R Q C R C O X4 G PX SO D G PX A D H A D H Fig Changes in gene expression levels of 19 energy metabolism-related genes in 10b-treated C albicans SC5314 The concentration of 10b was lgỈmL)1 All genes were examined by real-time RT-PCR with gene-specific primers Relative fold change was calculated with the CT value (see details in Experimental procedures) Results are the mean ± standard deviations for three independent experiments 10b and endogenous reactive oxygen species significantly, by 10.382-fold and 3.212-fold, respectively 25 000 Fluorescence intensity µg·mL–1 ROS production in C albicans treated with 10b Before measurement of the level of ROS production, drug concentrations inhibiting growth to 80% of control levels were estimated by interpolation, and this concentration for 10b was 0.5 lgỈmL)1 The level of endogenous ROS production induced by 10b was measured with concentrations of 1, and lgỈmL)1 Because miconazole (MCZ) is known to be an excellent endogenous ROS inducer [10], it was selected as the positive control to investigate the effect of 10b on endogenous ROS production in C albicans SC5314 As shown in Fig 3, after treatment with lgỈmL)1 10b, the level of ROS production increased by 22.7fold in a dose-dependent manner MCZ treatment also augmented ROS production, but to a lesser extent Effects on DWm, complex III activity and ATP content in mitochondria of C albicans after 10b treatment Treatment with 10b caused DWm degradation in a dose-dependent manner (Fig 4A), which was opposite to the result obtained for endogenous ROS production Although the generation of ROS is exponentially dependent on DWm [13], dysfunction of proton pumps can disrupt the positive correlation between endogenous ROS production and DWm [14] Therefore, we determined the activity of two important proton pumps, complex III and complex I (NADH–ubiquinone reductase), which are the main sources of ROS in mitochondria As was expected, complex III activity decreased in a dose-dependent manner after 10b treat- 20 000 * µg·mL–1 µg·mL–1 15 000 10 000 * 5000 ** Control MCZ * * ** 10b Fig Endogenous ROS generation in C albicans SC5314 cells without and with 10b and MCZ The concentrations of 10b and MCZ were 1, and lgỈmL)1 ROS levels represent the mean ± standard deviations for three independent experiments Statistically significant differences (as determined by Student’s t-test, as compared with control): *P < 0.01; **P < 0.05 ment (Fig 4B) The inhibitory efficiencies of 10b at lgỈmL)1 and lgỈmL)1 were 25.43% and 57.26%, respectively, after h of exposure However, no significant difference in complex I activity was observed between the control group and the 10b group (data not shown) It was therefore assumed that 10b treatment inhibited complex III activity, resulting in proton pump inactivation and a decrease in DWm Because intracellular ATP generation is known to be positively correlated with DWm in C albicans under normal culture conditions, the intracellular ATP concentration in cells without and with 10b was also measured The results revealed a dose-dependent decrease in intracellular ATP generation, which was consistent with the result obtained for DWm (Fig 4C) FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS 1077 120 100 B 120 100 ** 80 * 60 * 40 20 R M Liang et al Complex III activity (% of control) A Ratio (red fluorescence/ green fluorescence) 10b and endogenous reactive oxygen species ** 80 * 60 40 20 Control (µg·mL–1) C 10b concentration (µg·mL–1) 600 ATP concentration (nM) 10b concentration 500 ** Fig Mitochondrial functional analysis in C albicans SC5314 treated or untreated with 10b: (A) DWm (B) Complex III activity (C) Intracellular ATP level The concentrations of 10b were 1, and lgỈmL)1 DWm, complex III activity and ATP levels represent the mean ± standard deviations for three independent experiments Statistically significant differences (as determined by Student’s t-test, as compared with control): *P < 0.01; **P < 0.05 400 * * 300 200 100 Control 10b concentration (µg·mL–1) Effects of ROS on the antifungal activity of 10b against C albicans 25 000 * To determine whether ROS production was directly involved in the antifungal effect of 10b, the effect of an antioxidant on the net level of ROS production and antifungal activity was observed in 10b-treated cells The net ROS production in cells induced by 10b treatment was inhibited by addition of the antioxidant ascorbic acid (AA) in a dose-dependent manner, with complete inhibition occurring at 10 mm (Fig 5) We then examined whether AA treatment interfered with the antifungal effect of 10b In a colony formation assay, lgỈmL)1 10b caused a cytostatic effect (approximately 72% inhibition) at 48 h AA treatment prevented the colony-inhibitory effect induced by 10b in a dose-dependent manner (Fig 6) Discussion To further investigate the mechanism of action of 10b at a molecular level, an oligonucleotide microarray study was performed in C albicans SC5314 without and with 10b treatment The results showed that differentially expressed genes were involved in multiple biochemical functions Many experiments have confirmed that ROS play a central role in yeast signaling and 1078 ROS production (FI) 20 000 15 000 * 10 000 5000 ** –5000 10b (9 µg·mL–1) – + + + + AA (mM) 0 2.5 10 Fig Effect of AA on ROS production in 10b-treated C albicans The concentrations of AA were 2.5, and 10 mM, and that of 10b was lgỈmL)1 Data represent the mean ± standard deviations for three independent samples Statistically significant differences (as determined by Student’s t-test, as compared with control): *P < 0.01; **P < 0.05 apoptotic death [15–19] In addition, they can damage a wide range of molecules, including nucleic acids, proteins and lipids, that are involved in a variety of key events leading to cell death [20–22] Damage to FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS 10b and endogenous reactive oxygen species 120 Mitochondrial oxidative phosphorylation is a major ATP synthetic pathway in eukaryotes, where electrons liberated from reducing substrates are delivered to O2 via a chain of respiratory proton pumps These pumps (complexes I, III and IV) establish a proton electrochemical gradient (proton concentration gradient and DWm) across the inner mitochondrial membrane to store energy for the production of ATP Endogenous ROS are derived from mitochondrial respiratory chain electron leakage The main source of ROS in mitochondria is the ubisemiquinone radical intermediate (QHỈ), which is formed during the ubiquinone cycle at the Qo site of complex III [28–30] Complex I is also a source of ROS, although the mechanism of generation is less clear than that of complex III In vitro, electrons entering complex II can flow backwards through complex I to make ROS [31] Experimentally, a large increase in ROS formation is often seen in the condition known as reverse electron flow [32] In this study, we found that 10b treatment promoted ROS generation and decreased DWm and ATP production in mitochondria of C albicans The results of our further investigation showed that 10b inhibited complex III activity It is therefore presumed that 10b inhibited mitochondrial complex III activity, causing a reverse flow of electrons from complex II to complex I, resulting in ROS augmentation Simultaneously, 10b blocked electron transport in the mitochondrial respiratory chain and decreased DWm and ATP production (Fig 8) AA is a known antioxidant, and interactions between AA and ROS may attenuate the oxidant effect of ROS and alleviate ROS-induced damage to the organism [33] In this study, addition of AA significantly reduced the antifungal activity of 10b against C albicans, indicating that the antioxidant could alleviate the oxidative damage caused to the organism by endogenous ROS, allowing C albicans to survive 10b treatment These results also imply that endogenous ROS augmentation might be a major mechanism of the activity of 10b against C albicans The results of the present study demonstrate that 10b treatment could augment the production of endogenous ROS via three different mechanisms in C albicans: (a) providing more electrons for the mitochondrial respiratory chain by enhancing the tricarboxylic acid cycle; (b) attenuating ROS scavenging; and (c) enhancing the reverse flow of electrons from complex II to complex I by inhibiting complex III activity Increased ROS production contributes to the antifungal effect by means of strong oxidative damage to the organism This biochemical process might be involved in the mechanism of action of 10b against C albicans Cell survival (% of 10buntreated cell) R M Liang et al * 100 * 80 ** 60 40 20 0 2.5 10 AA (mM) Fig Effect of AA on 10b-induced colony inhibition in C albicans The concentrations of AA were 2.5, and 10 mM, and that of 10b was lgỈmL)1 Cells were incubated at 30 °C under constant shaking (200 r.p.m.) for 48 h, and the colonies were counted The rate of cell survival is represented as a percentage of the survival rate for cells not treated with 10b Data represent the mean ± standard deviations for three independent experiments Statistically significant differences (as determined by Student’s t-test, as compared with 10b treatment alone): *P < 0.01; **P < 0.05 mitochondrial macromolecules may lead to increased ROS production and further damage to mitochondrial components, thus causing a ‘vicious downward spiral’ in terms of ROS production and damage accumulation [23] The antifungal activities of several compounds, including ketoconazole, polygodial, histain 5, UK-2A and UK-3A, are achieved by inhibiting the respiratory activity of mitochondria [24–27] Therefore, we were particularly interested in the striking changes in the expression levels of energy metabolism-related genes: glycolysis-related genes, fermentation-related genes, respiratory electron transport chain-related genes, and ROS scavenging-related genes According to the Pasteur effect, by which aerobic oxidation can inhibit glycolysis (alcohol-facient fermentation) under aerobic circumstances, the tricarboxylic acid cycling process might be enhanced following exposure to 10b, because of the marked decreases in the expression levels of the genes participating in glycolysis, including those coding for rate-limiting enzymes (HXK2, PFK1, PFK2 and CDC19) Also, significantly downregulated genes were also involved in fermentation (e.g ADH1, ALD5 and PDC11) and ROS scavenging (e.g GPX2) Together, these changes mean that endogenous ROS generation might be strongly augmented, as shown in Fig Interestingly, we also found that the expression levels of two ROS scavenging-related genes (SOD5 and GPX1) and two fermentation-related genes (ADH3 and ADH5) were notably increased This might be the result of feedback control in response to high ROS levels FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS 1079 10b and endogenous reactive oxygen species R M Liang et al Glucose HXK2 GPX2 Glc6P PGI1 Glycolysis H2O or ROH + H 2O GSH Fru6P Cyt c PFK1 PFK2 TPI1 GrnP GraP e– e– Fru(1,6)P FBA1 GSSG ROS Gri(1,3)P LAC Inner membrane CX Pyruvate GA3P NADH e– e– Q e– CX CX CX GPM2 GPM1 PEP ENO1 Ethanol Fermentation FADH2 Aldehyde y Aldehyde TCA cycle ALD5 GA2P O2 Acetic acid Acetic acid Mitochondrion Cytoplasm Fig Central carbon metabolism in C albicans SC5314 The gray rectangles indicate low expression genes coding for metabolic enzymes of participated in energy metabolic process after 10b treatment The black ellipse or circle indicate augmented the tricarboxylic acid (TCA) cycling process and endogenous ROS generation after 10b treatment Glc6P, glucose 6-phosphate; Fru6P, fructose 6-phosphate; Fru(1,6)P2, fructose 1,6-bisphosphate; GraP, glyceraldehyde 3-phosphate; GrnP, dihydroxyacetone phosphate; Gri(1,3)P2, 1,3-bisphosphoglycerate; GA3P, 3-phosphoglyceric acid; GA2P, 2-phosphoglyceric acid; PEP, phosphoenolpyruvate; LAC, lactic acid; CX I–V, complexes I–V; Q, ubiquinone cycle; Cyt c, cytochrome c Experimental procedures Strains and culture C albicans SC5314 was used throughout this study The antifungal reagents used in the present study included 10b (Department of Medicinal Chemistry, School of Pharmacy of the Second Military Medical University, Shanghai, China), MCZ (Pfizer-Roerig Pharmaceuticals, New York, NY, USA) and stock solutions of various concentrations in dimethylsulfoxide C albicans cells were propagated in yeast–peptone–dextrose) medium [1% (w ⁄ v) yeast extract, 2% (w ⁄ v) peptone, and 2% (w ⁄ v) dextrose] RNA isolation and microarray hybridization The number of cells was adjusted to 1.0 · 106 colony-forming units (CFUs)ỈmL)1 in yeast–peptone–dextrose medium, divided into two parts: one was exposed to lgỈmL)1 10b, and the other was used as the control The cells were then grown at 30 °C under constant shaking (200 r.p.m.) for an additional h for the control and h for the 10b group, 1080 until they reached 0.6–0.9 · 107 CFUsỈmL)1, as described by Hughes [34] Cells were then collected by centrifugation at 3000 g for at room temperature, and frozen in liquid nitrogen We chose this 10b concentration because it had an obvious inhibitory effect on C albicans and allowed for recovery of a sufficient cellular mass for RNA extraction Total RNA was isolated by the hot phenol method, and purified with a NecleoSpin Extract II kit (Machery-Nagel Corp., Duren, Germany) [35] A 7925 C albicans genome ă 70-mer oligonucleotide microarray was obtained from CapitalBio Corporation (Beijing, China) A 1-lg sample of total RNA was used for preparing fluorescent dye-labeled cDNA by linear mRNA amplification [36] A DNDNA hybridization protocol was used to replace RNDNA hybridization, to reduce cross-hybridization [37] The labeled cDNAs were dissolved in 80 lL of hybridization solution [3 · SSC, 0.2% (w ⁄ v) SDS, · Denhardt’s solution, 25% (v ⁄ v) formamide], and denatured at 95 °C for before hybridization A sample of the mixed hybridization buffer was placed onto a microarray slide and covered with a glass coverslip Hybridization was performed with a BioMixer II FEBS Journal 278 (2011) 1075–1085 ª 2011 The Authors Journal compilation ª 2011 FEBS R M Liang et al 10b and endogenous reactive oxygen species ΔΨm Intermembrane space/cytoplasm H+ H+ H+ ROS C CX e– e– Q QH eQ QH2 NAD+ NADH TCA cycle e– + – e– Inner membrane Cyt c CX C CX e- F0 – CX V CX e– F1 O2 10b + ATP Fig Proposed mechanism of action of ROS augmentation induced by 10b Stimulation of the tricarboxylic acid cycle (TCA) by 10b treatment enhances electron flow into the mitochondrial respiratory chain, and generates a reverse flow of electrons from complex II to complex I by inhibiting complex III, which inhibits electron transport and causes a collapse of the proton gradient across the mitochondrial inner membrane These events enhance ROS generation and decrease DWm and ATP production Dashed lines indicate the subdued metabolic process CX I–V, complexes I–V; Q, ubiquinone cycle; Cyt c, cytochrome c (CapitalBio Corp.) After hybridization, the slides were washed with washing solution (2 · SSC, 0.2% SDS) and then with washing solution (2 · SSC) at 42 °C for Self-hybridization of the control sample was used to evaluate the system noise Microarray data processing The microarrays were scanned with a LuxScan 10KAscanner (CapitalBio Corp.) at two wavelengths to detect emissions from both Cy3 and Cy5 The images obtained were analyzed with luxscan 3.0 software (CapitalBio Corp.) Normalization was performed on the basis of a Lowess program The Cy5 ⁄ Cy3 ratio was calculated for each location on each microarray To minimize artefacts arising from low expression values, only genes with raw intensity values of > 800 counts for both Cy3 and Cy5 were chosen for analysis Significance Analysis of Microarrays software (sam) was used to identify significantly differentially expressed genes Genes with a false discovery rate < 5%, a q-value