Medical Mycology October 2004, 42, 427 /432

Chemiluminescent visualization of superoxide generated by Candida albicans

S. MASUI*, T. MAJIMA*, K. NAKAMURA$, S. ITO-KUWA$, K. TAKEO% & S. AOKI$ *Pharmaceuticals Division, POLA Chemical Industries, Yokohama, $Advanced Research Center, Nippon Dental University, Niigata and %Research Center for Pathogenic Fungi and Microbial Toxicoses, Chiba University, Chiba, Japan

Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

The high toxicity of reactive oxygen species (ROS) suggested a possible role in the pathogenicity of human pathogenic fungi. We previously reported a chemiluminescence method for measuring ROS generation in Candida albicans . In the present ( ), study, we attempted to visualize the ROS, superoxide anion radical (O2 generated by paraquat (PQ)-stimulated C. albicans using methyl-Cypridina luciferin analog (MCLA) as a chemiluminescence probe. Colonies of a wild-type C. albicans parent strain and its respiration-deficient mutant grown on agar plates were overlaid with a mixture of PQ and MCLA solutions. MCLA-dependent light emission from the colonies was recorded with a Hamamatsu ultralow-light-imaging apparatus with a CCD camera in a light-tight box. In the wild-type strain, marginal regions of growing colonies were strongly illuminated. The light emission from the colonies was extinguished by superoxide dismutase (SOD), proving that the light emission was strictly due to the superoxide anion. However, colonies of the respiration-deficient mutant poorly generated superoxide. Chemiluminecence measurements by a luminometer showed vigorous superoxide generation by the exponential phase cells of the parent strain but weak generation by the stationary phase cells. In the mutant, superoxide generation was weak compared with the parent strain. These results indicate that expansion of the colonies was due to the actively respiring cells located in the marginal regions. To our knowledge, the present report is the first chemiluminescent visualization of ROS including superoxide generated by C. albicans . Keywords nescence Candida albicans , superoxide, ultralow light imager, chemilumi-

Introduction
It is well established that reactive oxygen species (ROS), ( such as the superoxide radical (O2 ), hydrogen peroxide + (H2O2), hydroxyl radical ( OH) and singlet oxygen (1O2), are produced during oxidative metabolisms in aerobic cells. The ROS are proposed as a putative cause of certain diseases because of their high potential to induce a variety of molecular and cellular damage [1,2]. However, ROS produced by phagocytes are important

Received 3 June 2003; Accepted 29 September 2003 Correspondence: S. Aoki, Advanced Research Center, Nippon Dental University, 1-8 Hamaura-cho, Niigata 951-8580, Japan. Fax: '/81 25 267 1134; E-mail: mylab@ngt.ndu.ac.jp

for killing invading microbial pathogens in the host defense process. In contrast to a great number of studies on ROS production in phagocyte cells, there is only limited information on its production in pathogenic fungi [3 /7]. Schro ter et al . [4] first succeeded in measuring ROS generated in Candida albicans using lucigenin as a chemiluminescence probe and they suggested a relationship between the ability to generate ROS and virulence. Aoki et al . [7] recently developed a chemiluminescence method for measuring superoxide generated by C. albicans cells using methyl-Cypridina luciferin analog (MCLA) as a probe. The results obtained by comparison between a wild-type strain and a respiration-deficient mutant showed that superoxide produced in candidal cells was efficiently
DOI: 10.1080/13693780310001644716

2004 ISHAM

428

Masui et al.

dismutated under normal conditions. However, the superoxide-generating herbicide paraquat (PQ) induced respiration-dependent superoxide generation beyond the maximal ability to dismutate superoxide [7]. In a previous study, superoxide generation was measured with suspended candidal cells using a chemiluminescence reader [7]. On the basis of these results, we attempted to visualize superoxide generation by C. albicans colonies grown on agar plates in the present study. An ultralow-light-imaging apparatus equipped with a CCD camera was used to detect weak MCLAdependent chemiluminescence due to superoxide.

MCLA mixture at 40 units/ml. The MCLA-dependent chemiluminescence due to ROS generated by the colonies was recorded for 5 min in a light-tight box. The measured chemiluminescence intensities were processed by the ARGUS software and displayed in pseudo-color images.

Chemiluminescence measurements
Quantitative measurements of superoxide production by the candidal cells were carried out with a chemiluminescence reader using MCLA as a chemiluminescence probe, according to the previously reported method with slight modification [7]. Cells precultured in PYG broth were transferred to fresh PYG broth in flasks at an initial OD at 550 nm of 0.05 and grown at 378C with shaking. In the parent strain, exponential phase and stationary phase cells were respectively harvested at 4.5 and 22 h growth. Growth of the mutant KRD-19 is very slow [7,9]. Therefore, exponential phase cells were harvested at 15.5 h and stationary phase cells at 39.5 h of growth. Cells harvested at both growth phases were centrifugally washed in distilled water. The washed cells were suspended at 5 )/106 cells/ml of 20 mmol/l Hepes buffer (pH 7.5) containing 10 mmol/l glucose in test tubes. The tubes were set in an Aloka chemiluminescence reader BLR-301 (Tokyo) at 378C, and MCLA and PQ were sequentially added to the tubes to give final concentrations of 10 mmol/l and 1 mmol/l, respectively. The chemiluminescence intensity was expressed as counts per min (c.p.m.).

Materials and methods

Fungal strains
Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

The wild-type parent strain (K) of C. albicans was an oral isolate [8], and a respiration-deficient mutant (KRD-19) was derived from strain K by treatment with a chemical mutagen [9].

Chemicals
Paraquat and MCLA were products of Nacalai Tesque (Kyoto, Japan) and Tokyo Kasei Kogyo (Tokyo, Japan), respectively. The reagents were dissolved in sterile distilled water at a concentration of 1 mol/l and 0.5 mol/l, respectively, and stored at (/308C in the dark. The stock solutions were appropriately diluted with distilled water before use. Superoxide dismutase (SOD) from bovine erythrocytes (Sigma Chemicals, St Louis, MO, USA) was dissolved in 50 mmol/l phosphate buffer (pH 7.8), 0.1 mmol/l EDTA at a concentration of 875 units per 50 ml and stored at (/308C. Other chemicals and ingredients of culture media were obtained from Wako Pure Chemical Industries (Osaka, Japan).

Results
Figure 1 shows emission of MCLA-dependent chemiluminescence due to ROS generated by colonies of the wild-type strain, K, during growth on PYG agar plates at 378C. Light emission was observed in nearly all parts of the young colonies grown for 1 day although the emission was not vigorous. In parallel with the increase in the colony size after incubation for 3 and 5 days, the marginal regions of the colonies were very bright, showing localization of actively respiring cells in the marginal regions. As seen in Fig. 2, growth of the respiratory mutant was slow and light emission from the colonies was weak compared with the parent strain. The photon emission from the wild-type colonies almost completely vanished after the addition of the SOD enzyme (Fig. 3B). This result reconfirms that light emission is surely due to the superoxide anion, as reported previously [7]. The antioxidant, L-cysteine, at concentrations of 10 mmol/l or more also extinguished
2004 ISHAM, Medical Mycology, 42, 427 /432

Chemiluminescence images
The strains were precultured overnight in liquid PYG medium (2% polypepton, 1% yeast extract, 2% glucose) at 378C with shaking. The cultures were diluted and 0.1 ml of the dilutions containing about 50 cells was spread on PYG agar plates. After incubation at 378C for 1 /6 days, the plates were observed under an ultralow light image analyzer (ARGUS-50, Hamamatsu Photonics, Hamamatsu, Japan) equipped with a photon-counting CCD camera (C2400-30H). After taking photographs of colonies under light, a mixture of 0.1 mol/l PQ and 0.05 mmol/l MCLA (1:1) was gently dropped onto the colonies. To examine the effects of SOD, the enzyme was added to the PQ-

Visualization of superoxide generated by C. albicans

429

Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

Fig. 1 Colonies of the parent Candida albicans strain K (A, B and C) and their methyl-Cypridina -luciferin analog (MCLA)-dependent chemiluminescence images (D, E and F). Growth times in days are 1 for A, 3 for B and 5 for C.

Fig. 2 Colonies of the Candida albicans respiration-decient mutant KRD-19 (A and B) and their methyl-Cypridina -luciferin analog (MCLA)-dependent chemiluminescence images (C and D). Growth times in days are 2 for A and 6 for B.

dimensional images by the Hamamatsu image analyzer. The square area used for constructing the threedimensional images shown in Fig. 3C corresponds to 120 )/120 0/14 400 pixels under the measured conditions. The total photon counts of the colonies numbered 1, 2 and 3 in Fig. 3B were calculated as 1.67 )/ 103, 18.8 )/103 and 24.6 )/103 per 14 400 pixels, respectively. Figure 4 shows chemiluminescence measurements of superoxide generation by cells grown on liquid PYG medium at 378C. In the parent strain, superoxide generation was more vigorous in exponential phase cells but very poor in stationary phase cells. Compared with the parent strain, superoxide generation by the mutant cells was weak. These results are consistent with the chemiluminescence images shown in Fig. 1. The slight increase in chemiluminescence observed after addition of MCLA was due to auto-oxidation of MCLA and not due to superoxide generated by candidal cells, as reported previously [7].

Discussion
Active respiration supported by a sufficient oxygen supply is required for PQ-induced superoxide generation [7]. Chemiluminescence measurements showed extensive PQ-induced superoxide generation in exponentially growing cells of the parent strain (Fig. 4). Thus, the results shown in Fig. 1 indicate that (i) small premature colonies mainly consist of actively respiring young cells, and (ii) in maturing colonies, young cells in

the photon emission from the colonies (data not shown). Photon emission was clear in colonies treated with a mixture of MCLA and PQ. However, weak photon emission was also observed in colonies treated with MCLA alone, suggesting endogenous superoxide generation without stimulation by PQ. The intensity of superoxide generation from Candida colonies could be quantitatively expressed as three 2004 ISHAM, Medical Mycology, 42, 427 /432

430

Masui et al.

Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

Fig. 3 Effect of superoxide dismutase (SOD) on photon emission by colonies of the parent Candida albicans strain K grown for 3 days. (A) Three pairs of colonies before photon measurement. (B) Each pair of the colonies was treated with methyl-Cypridina -luciferin analog (MCLA) plus Paraquat (PQ) (lower pair), MCLA alone (middle) or MCLA'/PQ'/SOD (upper) and photon emission was measured for 5 min. (C) Threedimensional images of photon emission from the colonies numbered 1, 2 and 3 in Fig. 3B. The unit of chemiluminescence intensity in the pseudo color scale is arbitrary.

the marginal regions multiply actively and expand the colonies by leaving aged cells in the central regions. This distribution pattern of younger and older cells in single growing colonies is acceptable for considering growth physiology of the fungal colony. The light emission from colonies of the parent strain was effectively extinguished by the addition of SOD (Fig. 3). This result confirms that the ROS responsible for photon emission from PQ-stimulated candidal colonies is the superoxide anion. In the previously reported measurements with a chemiluminescence reader, superoxide generation by candidal cells could not be detected without stimulation by PQ [7]. However, weak photon emission was observed from colonies

treated with MCLA alone with the ultralow-light imager (Fig. 3). This indicated that the CCD camera of the Hamamatsu ultralow-light imager was extremely more sensitive than the chemiluminescence reader. The previous results [7] and those presented in Fig. 4 showed photon emission due to auto-oxidation of MCLA. Thus, it is necessary to know influences of the auto-oxidation on photon emission images of colonies. One drop of the mixture of PQ and MCLA was overlaid onto a non-inoculated PYG agar plate and photon emission was monitored. Photon emission from the area exposed to the PQ-MCLA mixture was negligible and not different from that observed in the control, non-exposed area (data not shown).
2004 ISHAM, Medical Mycology, 42, 427 /432

Visualization of superoxide generated by C. albicans

431

components of ROS generation in phagocytic cells [11]. Similarly, the intracellular compartments responsible for ROS generation in single fungal cells may be visualized using a microscope coupled with an ultralow-light-imaging apparatus. It has been well documented that ROS levels change in response to physiological stimuli and ROS participate in mediation of signal transduction in mammalian cells [12]. Interestingly, it has been reported that the formation of endogenous ROS is essential for exhibition of antifungal effects of the human salivary peptide histatin 5 [13] and miconazole [14] in C. albicans . These results have encouraged us to further investigate the roles of ROS in control mechanisms in fungal cells.

Acknowledgements
Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

Fig. 4 Superoxide generation by cells of the parent Candida albicans strain K (A) and the mutant KRD-19 (B) measured by a chemiluminescence method. Methyl-Cypridina -luciferin analog (MCLA) and Paraquat (PQ) were added at the time indicted by arrows. k, exponential phase cells; m, stationary phase cells.

This study was supported in part by a Grant-in-Aid from the Ministry of Education, Science, Sports, Culture and Technology of Japan (12671788) and by the Cooperative Research Program of the Research Center for Pathogenic Fungi and Microbial Toxicoses, Chiba University (2001 /9 and 2002 /20). We thank Professor Libero Ajello for critically reading and improving the manuscript.

References
1 Fridovich I. The biology of oxygen radicals. Science 1978; 201: 875 /880. 2 Cadenas E. Biochemistry of oxygen toxicity. Annu Rev Biochem 1989; 58: 79 /110. 3 Hipler UC, Wollina U, Mayser P. Chemiluminescence measurements of reactive oxygen species (ROS) generated by different stimulated Trichosporon strains. In: Roda A, Pazzagli M, Kricka LJ, Stanley PE (eds). Bioluminescence and Chemiluminescence. Perspectives for the 21st Century. Chichester: John Wiley & Sons, 1999: 307 /310. 4 Schro ter C, Hipler UC, Wilmer A, Ku nkel W, Wollina U. Generation of reactive oxygen species by Candida albicans in relation to morphogenesis. Arch Dermatol Res 2000; 292: 260 / 264. 5 Sander CS, Hipler UC, Wollina U, Elsner P. Inhibitory effect of terbinane on reactive oxygen species (ROS) generation by Candida albicans . Mycoses 2002; 45: 152 /155. 6 Hipler UC, Wollina U, Denning D, Hipler B. Fluorescence analysis of reactive oxygen species (ROS) generated by six isolates of Aspergillus fumingatus . In: Case JF, Herring PJ, Robinson BH, Haddock SHD, Kricka LJ, Stanley PE (eds). Bioluminescence and Chemiluminescence 2000 . Singapore: World Scientic, 2001: 411 / 414. 7 Aoki S, Ito-Kuwa S, Nakamura K, Nakamura Y, Vidotto V, Takeo K. Chemiluminescence of superoxide generated by Candida albicans : differential effects of the superoxide generator paraquat on a wild-type strain and a respiratory mutant. Med Mycol 2002; 40: 13 /19. 8 Aoki S, Ito-Kuwa S. Respiration of Candida albicans in relation to its morphogenesis. Plant Cell Physiol 1982; 23: 721 /726.

As to chemiluminescence imaging of ROS production, Yasui and Sakurai [10] succeeded in visualizing ROS generated in live mouse skin exposed to UVA light using a low-light-imaging apparatus that was similar to that used in our study. To our knowledge, our report is the first showing chemiluminescence images of ROS generation by C. albicans . The present study has demonstrated a useful technique with which to investigate ROS in pathogenic fungi. First, in our experimental conditions, active superoxide generation from C. albicans cells is clearly observable when stimulated by the superoxide generator PQ [7]. However, generation of ROS has been demonstrated in Trichosporon strains [3], C. albicans [4,5] and Aspergillus fumingatus [6] without oxidative stimuli. In the present study, superoxide production in C. albicans was confirmed without the action of PQ, though the production was not extensive. Thus, it would be very interesting to examine using the methods described here whether pathogenic fungi, such as dermatophytes, produce ROS in the process of infection. Second, there are a number of cytochemical studies on the cellular
2004 ISHAM, Medical Mycology, 42, 427 /432

432

Masui et al.

9 Aoki S, Ito-Kuwa S. Induction of petite mutation with acriavine and elevated temperature in Candida albicans . J Med Vet Mycol 1987; 25: 269 /277. 10 Yasui H, Sakurai H. Chemilunimescent detection and imaging of reactive oxygen species in live mouse skin exposed to UVA. Biochem Biophys Res Commun 2000; 269: 131 /136. 11 Karnovsky MJ. Cytochemistry and reactive oxygen species: a retrospective. Histochem 1994; 102: 15 /27. 12 Wolin MS, Mohazzab-H KM. Mediation of signal transduction by oxidants. In: Scandalios JG (ed.). Oxidative Stress and the

Molecular Biology of Antioxidant Defenses . New York: Cold Spring Harbor Laboratories, 1997: 21 /48. 13 Helmerhorst EJ, Troxler RF, Oppenhaim FG. The human salivary peptide histatin 5 exerts its antifungal activity through the formation of reaction oxygen species. Proc Natl Acad Sci USA 2001; 98: 14637 /14642. 14 Kobayashi D, Kondo K, Uehara N, Tsuji N, Yagihashi A, Watanabe N. Endogenous reactive oxygen species is an important mediator of miconazole antifungal effect. Antimicrob Agents Chemother 2002; 46: 3113 /31173.

Med Mycol Downloaded from informahealthcare.com by HINARI on 05/07/13 For personal use only.

2004 ISHAM, Medical Mycology, 42, 427 /432