Keisuke Nakamura, New Industry Creation Hatchery Center, Tohoku University, Aoba 6-6-10, Aramaki, Aoba-ku, Sendai, 980-8579, Japan. Tel: +81 22 795 3976; fax: +81 22 795 4110; email address: firstname.lastname@example.org
The purpose of the present study was to evaluate the mechanism of microbial resistance to oxidative stress induced by photolysis of hydrogen peroxide (H2O2) in relation to microbial catalase activity. In microbicidal tests, Staphylococcus aureus and Candida albicans were killed and this was accompanied by production of hydroxyl radicals. C. albicans was more resistant to hydroxyl radicals generated by photolysis of H2O2 than was S. aureus. A catalase activity assay demonstrated that C. albicans had stronger catalase activity; accordingly, catalase activity could be one of the reasons for the resistance of the fungus to photolysis of H2O2. Indeed, it was demonstrated that C. albicans with strong catalase activity was more resistant to photolysis of H2O2 than that with weak catalase activity. Kinetic analysis using a modified Lineweaver-Burk plot also demonstrated that the microorganisms reacted directly with hydroxyl radicals and that this was accompanied by decomposition of H2O2. The results of the present study suggest that the microbicidal effects of hydroxyl radicals generated by photolysis of H2O2 can be alleviated by decomposition of H2O2 by catalase in microorganisms.
A novel disinfection method whereby hydroxyl radicals, a ROS, are artificially generated by photolysis of H2O2 has been developed recently. In a previous study, we demonstrated that hydroxyl radicals generated by laser irradiation of H2O2 could kill pathogens of oral infectious diseases (1). Although laser irradiation of bacterial suspensions in 1 M H2O2 resulted in a >99.99% reduction in the numbers of bacteria within 3 min, the sensitivity of the bacteria to this disinfection system varied somewhat according to the species. This finding suggests that bacterial species have varying degrees of resistance to this disinfection system.
It is well known that some species of microorganisms are equipped with enzymatic defense mechanisms that suppress oxidative stress caused by ROS. As for defense mechanisms against H2O2, catalase, one such enzyme, is mainly responsible for these defense mechanisms, catalyzing a reaction in which H2O2 decomposes into water and oxygen (2–4). Hence, the catalase activity of microorganisms might affect the microbicidal effect of photolysis of H2O2. We have recently established an analytical method for evaluating the catalase activity of microorganisms at a cellular level (5). The assay enables us to measure the catalase activity of cell suspensions rather than cell lysates. Analysis of microbial resistance to photolysis of H2O2 associated with catalase activity would probably provide important information for characterizing this disinfection system.
To evaluate whether microorganisms have resistance to this disinfection system in relation to catalase activity, we applied kinetic analysis to the reactions between microorganisms and hydroxyl radicals generated by photolysis of H2O2. The Lineweaver-Burk plot has often been used for analysis of the kinetics of enzyme reactions with an inhibitor (6). In addition to enzyme reactions, Niwano et al. demonstrated that this analytical method can be used for kinetic analysis of the reactions between ROS and antioxidants using an ESR spin trapping technique (7). According to this theory, it is possible to determine whether the antioxidant directly scavenges ROS or suppresses its generation (7, 8). Thus, the same analytic method, using microorganisms instead of antioxidants, could clarify whether the microorganisms react directly with hydroxyl radicals or interrupt photolysis of H2O2.
The purpose of the present study was, therefore, to evaluate the mechanism of microbial resistance to oxidative stress induced by photolysis of H2O2 in relation to microbial catalase activity.
MATERIALS AND METHODS
Reagents were purchased from the following sources: catalase, n-decane, n-undecane, n-dodecane, and n-tridecane from Wako Pure Chemical Industries (Osaka, Japan); TEMPOL from Sigma Aldrich (St. Louis, MO, USA); DMPO from Labotec (Tokyo, Japan); and H2O2 from Santoku Chemical Industries (Tokyo, Japan). All other reagents used were of analytical grade.
An experimental light emitting diode device
In the present study, hydroxyl radicals were generated by irradiation of LED with a wavelength of 400 ± 20 nm (NHH105UV, JW-System, Tokyo, Japan) to H2O2. An experimental device equipped with two LEDs with an output power of 300 mW per LED was constructed. The energy density of the LED measured by a portable illuminance/luminance/irradiance gauge (Delta OHM, Caselle di Selvazzano, Italy) was 80 mW/cm2.
S. aureus or C. albicans ATCC 25923 was obtained from the American Type Culture Collection (Manassas, VA, USA) and Candida albicans. JCM 9061 from the Japan Collection of Microorganisms, RIKEN BioResource Center (Wako, Japan). S. aureus was cultured on brain heart infusion agar (Becton Dickinson Labware, Franklin Lakes, NJ, USA) and C. albicans on Sabouraud glucose agar (1% peptone, 4% glucose, and 1.5% agar). Both microorganisms were aerobically cultured at 37°C. Suspensions of each microorganism were prepared in sterile physiological saline and the number of the cells was adjusted to the required concentrations for each experiment.
Bactericidal and fungicidal testing
The bacterial and fungal suspensions were adjusted to 2 × 106 CFU/mL. Then, 250 μL of the suspension was mixed with 250 μL of 500 mM H2O2 to make final concentrations of 1 × 106 CFU/mL for the microorganisms and 250 mM for H2O2. Immediately after mixing, the test tubes were set in the LED device and irradiated with the LEDs for 1, 5 and 10 min. After LED irradiation, 50 μL of the sample was mixed with 50 μL of 5000 U/mL sterilized catalase solution in order to terminate the bactericidal and fungicidal effects of any remaining H2O2. Ten-fold serial dilutions of the mixture were prepared and then 10 μL of each dilution was plated on agar medium. The agar medium was cultured for 48 hr at 37°C and the number of CFU/mL of suspension was determined for each microorganism. The bactericidal and fungicidal effects of hydroxyl radicals generated by LED irradiation of H2O2 (expressed as LED(+)H2O2(+) was compared with the effects of (1) LED irradiation alone, LED(+)H2O2(−); (2) H2O2 alone, LED(−)H2O2(+); and (3) no treatment, LED(−)H2O2(−). For the condition of LED(−), the samples were kept in a light shielding box. For the condition of H2O2(−), sterile saline was added to the reaction system instead of H2O2. All the tests were performed in triplicate.
Catalase activity of the microorganisms
The catalase activity of S. aureus and C. albicans was evaluated by an ESR oximetry technique. The assay used in this study was identical to that described in our previous paper (5). In brief, TEMPOL and catalase were mixed, and then H2O2 was added to make final concentrations of 200 μM for TEMPOL, 0.625 to 5 U/mL for catalase, and 250 mM for H2O2. One unit of catalase was defined as the amount that catalyzes the decomposition of 1 μM of H2O2 per min at 25°C. The mixture was then transferred to a quartz cell for ESR spectrometry 30 s after the addition of H2O2 and the ESR spectrum was recorded on an X-band ESR spectrometer (JES-FA-100, Jeol, Tokyo, Japan). The measurement conditions for ESR were as follows: field sweep, 330.50–340.50 mT; field modulation frequency, 100 kHz; field modulation width, 0.1 mT; amplitude, 80; sweep time, 2 min; time constant, 0.03 s; microwave frequency, 9.420 GHz; and microwave power, 4 mW. The peak widths of the ESR signals for TEMPOL broadened by oxygen from the reaction between H2O2 and catalase were plotted against the concentration of catalase to construct a calibration curve. The same procedure was performed using different concentrations of the bacterial and fungal suspensions. These suspensions were added to the reaction system instead of catalase. The catalase activity of each microorganism was evaluated using the calibration curve.
Susceptibility of Candida albicans to photolysis of hydrogen peroxide in relation to catalase activity
Candida spp grown in a medium containing n-alkanes show stronger catalase activity than when grown in a medium containing glucose (9–11). Thus, we compared the bactericidal effect of the photolysis of H2O2 on C. albicans with high catalase activity to that on the fungus with normal catalase activity. A medium composed of 5% NH4H2PO4, 2.5% KH2PO4, 1% MgSO4·7H2O, 0.02% FeCl3·6H2O, 1% corn steep liquor, 0.5% Tween 80, and either 1%n-alkane mixture or 1.65% glucose as a carbon source was prepared according to a previous study (9). The n-alkane mixture consisted of 23.9%n-decane, 46.8%n-undecane, 26.3%n-dodecane and 3.0%n-tridecane. The fungal suspension prepared as described above was inoculated in the medium and further incubated at 37°C for 18 hr. The cells harvested by centrifugation at 5000 × g for 5 min were washed and resuspended in saline. The catalase activities of C. albicans grown in the media containing n-alkane mixture or glucose were determined as described above. The catalase activities of C. albicans grown in the different media were statistically analyzed by Student's t-test (P < 0.05 being considered significant). Next, the fungicidal tests were performed under LED(+)H2O2(+) and the LED(−)H2O2(−) conditions with a treatment time of 5 min. All the tests were performed in sextuplicate.
Kinetic analysis of the reactions between microorganisms and hydroxyl radicals
For the ESR spin trapping analysis, in which DMPO was used as a spin trap, the yield of DMPO-OH, a spin adduct of hydroxyl radical, was analyzed. To evaluate the inhibition of DMPO-OH formation caused by the microorganisms, quantitative analysis of DMPO-OH was performed according to our previous study (12). Each microorganism, H2O2 and DMPO were mixed to make final concentrations of 106–109 CFU/mL for the microorganisms, 250 mM for H2O2, and 300 mM for DMPO. The mixtures were set in the LED device immediately after the addition of H2O2 and irradiated for 1 min. After irradiation, the ESR spectra were recorded under the conditions described above. To calculate the concentrations of DMPO-OH using Digital Data Processing (JEOL, Tokyo, Japan), 20 μM TEMPOL was used as a standard and the ESR signal of manganese was used as an internal standard. The IC50 of the microorganisms against the yield of DMPO-OH were calculated. Similarly, the inhibition curve and IC50 of catalase against the yield of DMPO-OH were obtained as controls.
In ESR spin trapping analysis of photolysis of H2O2, a hydroxyl radical (HO•) is generated first and then DMPO-OH is formed as shown below:
Therefore, there are three possibilities for inhibition of DMPO-OH formation, namely, inhibition of generation of hydroxyl radicals, direct scavenging of hydroxyl radicals, or both. To investigate whether the microorganisms inhibit the generation of hydroxyl radicals or react directly with them, kinetic analysis was performed using a modified Lineweaver-Burk plot with different concentrations of DMPO. The reagents and the microorganisms were mixed according to the following protocol: 200 μL of each microorganism, 50 μL of DMPO and 250 μL of H2O2 were mixed to make final concentrations of 106 to 108 CFU/mL for the microorganisms, 25 to 300 mM for DMPO, and 250 mM for H2O2. The mixtures were then irradiated with the LEDs and the ESR spectra were recorded and analyzed as described above. The reciprocal values of the concentrations of DMPO-OH were plotted against those of DMPO for each concentration of the microorganisms. The double reciprocal plots constructed for the microorganisms were then compared with that for catalase, which decomposes H2O2 but not hydroxyl radicals.
Bactericidal and fungicidal tests
S. aureus was killed with four or more logarithmic reductions within 5 min under the condition of LED(+)H2O2(+) (Fig. 1a). LED(+)H2O2(−) and LED(−)H2O2(+) showed slight bactericidal effects with approximately one logarithmic reduction in 10 min. C. albicans was also killed with four or more logarithmic reduction within 10 min under the condition of LED(+)H2O2(+) (Fig. 1b). However, LED(+)H2O2(−) and LED(−)H2O2(+) killed hardly any C. albicans in 10 min.
Catalase activity of the microorganisms
The catalase activity (U/mL) of each microorganism was calculated using the calibration curve and plotted against the number of cells (Fig. 2). The slope values of the equation shown in Figure 2 denote the catalase activity per cell (U/mL). Thus, the catalase activity was calculated to be 3.1 × 10−8 U/cell and 5.3 × 10−8 U/cell for S. aureus and C. albicans, respectively.
Susceptibility of Candida albicans to photolysis of hydrogen peroxide in relation to catalase activity
C. albicans grown in the medium containing n-alkane mixture showed significantly stronger catalase activity than that grown in the medium containing glucose (Fig. 3a). In the fungicidal test, the fungus with strong catalase activity induced by the n-alkane mixture showed higher resistance to photolysis of H2O2 (Fig. 3b).
Kinetic analysis of the reaction between microorganisms and hydroxyl radicals
We observed that S. aureus and C. albicans inhibited DMPO-OH formation in a concentration dependent manner, as did catalase (Fig. 4a, b, c). According to analysis of the inhibition curves of S. aureus, C. albicans and catalase against DMPO-OH formation, the IC50s were 5.9 × 107 CFU/mL, 1.8 × 107 CFU/mL, and 10 U/mL, respectively, when DMPO was used at 300 mM.
The modified Lineweaver-Burk plots of the microorganisms and catalase are shown in Fig. 5a, b, c. In the case of the microorganisms, we observed linear patterns with the intersections shifted to the negative side of the x axis. This pattern indicates that DMPO-OH formation was inhibited not only by decomposition of H2O2, but also by reactions between the microorganisms and hydroxyl radicals (mixed inhibition). On the other hand, the modified Lineweaver-Burk plot of catalase showed a parallel linear pattern indicating that inhibition of DMPO-OH formation was due to the decomposition of H2O2 (noncompetitive inhibition).
Both S. aureus and C. albicans are frequently detected in the oral cavity and sometimes cause serious infectious diseases such as pneumonia, toxic shock syndrome and septicemia, especially in elderly people (15–17). Furthermore, since both microbial species contain catalase to protect themselves from oxidative stress, we selected them as the subjects for a study on the assumption that they might be resistant to a disinfection system utilizing photolysis of H2O2. Indeed, 250 mM H2O2 alone hardly killed any microorganisms. We believe that the concentration of 250 mM was too low to exert fungicidal and bactericidal effects because 3% H2O2, corresponding to approximately 890 mM, is usually used as a disinfectant. Meanwhile, when H2O2 was photolyzed by LED irradiation, sufficient bactericidal and fungicidal effects resulted, demonstrating that these effects are attributable to hydroxyl radicals generated by photolysis of H2O2. The bactericidal and the fungicidal effects were enhanced by longer LED irradiation times, suggesting that the microbicidal effect depends on the yield of hydroxyl radicals, as shown in our previous study (1). Even though the concentration of H2O2 used in the present study was lower than that used in our previous study (1), photolysis of H2O2 generated enough hydroxyl radicals to kill S. aureus and C. albicans with four or more logarithmic reduction within 5 and 10 min, respectively (Fig. 1a, b).
Since S. aureus and C. albicans contain catalase, they could be resistant to the disinfection system by decomposing H2O2. We measured the catalase activity of the microorganisms to evaluate how much their catalase contributed to their defense mechanisms. In a previous study, we proposed a new analytical method for determining catalase activity at a cellular level (5). The glutathione peroxidase system is one possible candidate for decomposing H2O2. However, as demonstrated in our previous study, the system not at all or only slightly affected the assay for catalase activity because it did not reduce the concentration of H2O2 under the condition applied in this catalase assay (5). Thus, we suggest that this assay specifically determines catalase activity. Analysis of catalase activity by the method described demonstrated that C. albicans has stronger catalase activity than S. aurues. The catalase activities of C. albicans and S. aureus in this study were comparable with those of a different stock strain of C. albicans and other aerobes used in our previous study (5).
To evaluate the contribution of catalase activity to the microbial defense mechanisms against photolysis of H2O2, C. albicans with different catalase activities, grown in the same medium except for the carbon source, were prepared. Analysis of catalase activity showed that the fungus grown in the medium containing n-alkane mixture had significantly stronger catalase activity than that grown in the medium containing glucose, as reported previously (9). Furthermore, the fungus with stronger catalase activity showed greater resistance to photolysis of H2O2. This suggests that catalase of C. albicans plays an important role in defense against photolysis of H2O2.
Analysis of the inhibition curve showed that DMPO-OH formation decreased with increasing concentrations of microorganisms. The IC50 of C. albicans (1.8 × 107 CFU/mL) was smaller than that of S. aureus (5.9 × 107 CFU/mL). This finding indicates that each C. albicans cell suppresses the formation of DMPO-OH more strongly than does each S. aureus cell. Because C. albicans cells are larger than S. aureus cells, each C. albicans cell is obviously exposed to larger amounts of hydroxyl radicals than each S. aureus cell. Nevertheless, C. albicans was more resistant to hydroxyl radicals than S. aureus. One possible explanation for the resistance of C. albicans is that it decomposes H2O2 more effectively than does S. aureus. Indeed, C. albicans showed stronger catalase activity per cell than did S. aureus, as discussed above. In addition, the differences between eukaryotic fungal cells and prokaryotic bacterial cells in cellular structure as well as gene expression associated with resistance to oxidative stress might allow C. albicans to tolerate larger amounts of hydroxyl radicals.
In the ESR spin trapping analyses, detection of DMPO-OH depends on two reactions, namely generation of hydroxyl radicals and formation of DMPO-OH, as shown in equations 1 and 2. Under the condition used in the present study, the formation of DMPO-OH was inhibited by hydroxyl radical scavengers, dimethyl sulfoxide and mannitol, but not by the superoxide radical scavenger, superoxide dismutase (data not shown). This finding suggests that DMPO-OH is formed by the reaction between DMPO and the hydroxyl radical as defined in equations 1 and 2. When DMPO-OH formation from the reaction between DMPO and hydroxyl radicals was inhibited, analysis using the modified Lineweaver-Burk plot enabled us to evaluate how that inhibition occurs. The theory is that a linear pattern with an intersection on the y axis means competitive inhibition, a parallel linear pattern means noncompetitive inhibition, and a linear pattern with an intersection on the negative side of x axis means mixed inhibition (7, 6, 18). Analysis using different concentrations of the microorganisms and catalase clearly demonstrated that there were different kinetic patterns between the microorganisms and catalase against DMPO-OH formation. The linear patterns of the double reciprocal plots for microorganisms showed different slopes and intersections shifted to the negative side of the x axis, whereas the linear patterns for catalase showed parallel lines without intersection. These findings suggest that the microorganisms function as mixed inhibitors whereas catalase functions as a noncompetitive inhibitor. This means that the microorganisms react directly to the hydroxyl radicals and simultaneously decompose H2O2 with their catalase, the latter reaction probably conferring some resistance to the disinfection system. Hydroxyl radicals are also generated by a Fenton-like reaction, in which hydrogen peroxide formed by dismutation of superoxide anion is decomposed in the presence of transition metals. Catalase probably contributes to reduced formation of such hydroxyl radicals in microorganisms.
Indeed, it has been demonstrated that the bactericidal effect of several hydroxyl radical generation systems is potentiated by an increased yield of hydroxyl radicals (19–21), and suppressed by the presence of hydroxyl radical scavengers (22, 23). Similarly, the kinetic analysis suggests that microbial catalase suppresses generation of hydroxyl radicals in the disinfection system. Therefore, it is possible that inactivation of catalase potentiates the bactericidal and the fungicidal effects of the disinfection system.
Hydroxyl radicals generated by photolysis of H2O2 using LED could potentially be adopted as a novel disinfection system. Besides having strong bactericidal and fungicidal effects, disinfection by ROS probably would not lead to development of bacterial and fungal resistance to these agents because ROS interact directly with several cell structures and different metabolic pathways (24, 25). In particular, hydroxyl radicals and singlet oxygen are thought to be free from induction of resistance because no defense mechanisms against these two ROS have been reported in living cells. From a clinical point of view, this is a major advantage of this disinfection system.
This research was supported by Ministry of Economy, Trade and Industry, Japan, Grant-in-Aid for “Regional Innovation Creation R&D Programs”, 21R2007C, 2010.
None of the authors has any conflicts of interest associated with this study.