- Email: [email protected]
Hydroxyl radical generation by red tide algae
ARCHIVES
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 294, No. 1, April, pp. 38-43, 1992
Hydroxyl Radical Generation
by Red Tide Algae’
Tatsuya Oda,* Takaaki Akaike,f‘ Keizo Sate,? Atsushi Ishimatsu,$ Satoshi Takeshita,* Tsuyoshi Muramatsu,* and Hiroshi Maedat*2 *Division of Biochemistry, Faculty of Fisheries, Nagasaki University, Nagasaki 852, Japan; tDepartm.ent of Microbiology, Kumamoto University Medical School, Kumumoto 860, Japan; and $Nomo Fisheries Station, Nagasaki University, Nomozaki, Nagasaki 851-05, Japan
Received August 23,1991, and in revised form November
22,199l
Hydroxyl radical induced photochemically plays an important role in atmospheric chemistry (1). Its generation by photodynamic reaction in the atmosphere and in the sea is documented relatively recently (1, 2). However, little information is available for the generation of oxygen radicals in the marine biosphere. There has recently been a global increase in the frequency, magnitude, and geographical extent of red tides. This increase seems to be well correlated with the degree of coastal pollution or the utilization of coastal water for aquaculture (3). Several species of phytoplankton have been identified as the causative organism of red tide (3). Chuttonella marina (Raphidophyceae) is one of the most noxious red tide phytoplankton and causes severe damage to fish farming. Although several hypotheses have been proposed about the cause of fish death by Chattondlu sp. (4,5), the precise mechanism of the toxic action remains unclear. It has been reported that edema formation in gill lamellae of fish was induced by Chattondu sp. as a typical histological change and that the gill damage resulted in fish death because of oxygen deficiency (6). Our preliminary experiments indicated that C. marina exhibited the active oxygen-mediated toxic effect against Vibrio alginolyticus, which was isolated from a seaweed (7). Thus, our study was undertaken to confirm the generation of oxygen radicals by C. marina by using various methods and to identify the radical species by ESR spin trapping.
unicellular marine phytoplankton Chattonella marina is known to have toxic effects against various living marine organisms, especially fishes. However, details of the mechanism of the toxicity of this plankton remain obscure. Here we demonstrate the generation of superoxide and hydroxyl radicals from a red tide unicellular organism, C. marina, by using ESR spectroscopy with the spin traps 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) and N-t-butyl-cr-phenylnitrone (PBN), and by using the luminol-enhanced chemiluminescence response. The spin-trapping assay revealed productions of spin adduct of superoxide anion (0;) (DMPO-OOH) and that of hydroxyl radical (. OH) (DMPO-OH) in the algal suspension, which was not observed in the ultrasonic-ruptured suspension. The addition of superoxide dismutase (500 U/ml) almost completely inhibited the formation of both DMPO-OOH and DMPO-OH, and carbon-centered radicals were generated with the disappearance of DMPO-OH after addition of 5% dimethyl sulfoxide (MesSO) and 5% ethanol. Furthermore, the generation of methyl and methoxyl radicals, which are thought to be produced by the reaction of hydroxyl radical and Mess0 under aerobic condition, was identified using spin trapping with a combination of PBN and MesSO. Luminol-enhanced chemiluminescence assay also supported the above observations. These results clearly indicate and releases the superoxide that C. marina generates radical followed by the production of hydroxyl radical to the surrounding environment. The velocity of superoxide generation by C. marina was about 100 times faster than that by mammalian phagocytes per cell basis. The generation of oxygen radical is suggested to be a pathogenic principle in the toxication of red tide to susceptible aquaculture fishes and may be directly correlated with the coastal pollution by red tide. Q 1~82 Academic
1 Supported in part by Grants-in-Aid for Scientific Research from Monbusho (the Ministry of Education, Science, and Culture) of Japan,
Press, Inc.
1990-1991.
The
’ To whom correspondence
should be addressed.
38 Copyright
ooo3-9S61/92 $3.00 0 1992 by Academic Press, Inc.
All rigbta of reproduction in any
form
reserved
HYDROXYL
EXPERIMENTAL
RADICAL
PRODUCTION
PROCEDURES
Materials We used 5,5-dimethyl-1-pyrroline-N-oxide (DMP0)3 and N-t-butyla-phenylnitrone (PBN) throughout the experiments as spin traps. DMPO was synthesized and purified by Mitaui Toatsu Chemical Co., Ltd., Mobara, Japan, or Dojin Chemical Laboratories Co., Ltd., Kumamoto, Japan. PBN, 2,2,6,6-tetramethyl-4-hydroxypiperidine (Tempol), luminol, horse heart ferricytochrome c (type VI), and catalase were purchased from Sigma Chemical Co., St. Louis, Missouri, and human recombinant Cu, Zn-superoxide dismutase (SOD) was a gift from Nippon Kayaku Co., Ltd., Tokyo, Japan. Xanthine oxidase from cow milk was a product of Boehringer-Mannheim GmbH, Mannheim, Germany. Other chemicals were of the highest grade commercially available.
Methods Culture for phytoplankton. C. marina was generously provided by Kagoshima Prefectural Fisheries Experimental Station, Japan, and was cultured at 26’C in Erd-Schreiber modified (ESM) medium (pH 8.0) under 3000 lx illumination with a cycle of 12 h light and 12 h dark (4). ESM medium was prepared according to the composition reported originally without soil extract, i.e., 120 mg NaN03, 5 mg K,HPO,, 0.1 mg vitamin B1 , 0.01 mg vitamin B,z, 0.001 mg biotin, 0.26 mg EDTA-Fe3+, 0.33 mg EDTA-Mn’+, and 1 g tris(hydroxymethyl)aminomethane were dissolved into 1 liter of seawater and the pH was adjusted to 8.0, followed by filtration with a 0.25-pm (pore size) filter and autoclaving (121°C, 30 min). All cultivation was carried out with using sterilized instruments. In addition, microscopic examination at Xl000 magnification of wellgrown algal suspension revealed neither distinctive proliferative bacteria or fungi other than C. marina in the medium. In the case of spin trapping with DMPO, 140 ESR spin trapping. gl of algal suspension (final concentration, 1.5-2.5 X lo4 cells/ml in ESM medium) was mixed with DMPO in the presence of 500 pM of diethylenetriaminepentaacetic acid (DTPA) (final volume, 200 ~1). DMPO appears to be toxic to the algae at a concentration higher than 90 mM. Therefore, we usually used 90 mM of DMPO for measurements of oxygen radicals, although radical capturing efficiency, especially against superoxide radical, became reduced to about lo-20% of the saturated level of DMPO-OOH, which was obtained at a concentration higher than 0.67 M of DMPO, as reported by Mitsuta et al. (8). The mixture was immediately transferred to the ESR quartz flat cell (inner size, 60 X 10 X 0.31 mm), and the cell was placed in a JEOL JES-RElX ESR spectrometer (JEOL Co., Ltd., Akishima, Japan). ESR spectra were recorded at room temperature under the following conditions: modulation frequency, 100 kHz; modulation amplitude, 0.079 mT, scanning field, 336.2 + 5 mT; receiver gain, 10 X 100, response time, 0.1 s; sweep time, 2 min; microwave power, 40 mW; and microwave frequency, 9.421 GHz. To confirm the generation of hydroxyl radical in plankton suspension, we used the spin trapping with PBN and dimethyl sulfoxide (Me&SO) designed by Britigan et al. (9). Namely ESR experiments were performed in the same manner as spin trapping with DMPO except that reaction mixtures contained 50 mM of PBN and 10% Me,SO instead of DMPO. Quuntitation of DMPO-OOH by ESR spectroscopy. After recording the ESR spectra as described above, the signal intensity of DMPOOOH was normalized as a relative height against the standard signal intensity of the manganese oxide marker (MnO). An absolute concen-
3 Abbreviations used: DMPO, 5,5-dimethyl-1-pyrroline-N-oxide; PBN, N-t-butyl-a-phenylnitrone; Tempol, 2,2,6,6-tetramethyl-4-hydroxypiperidine; SOD, superoxide dismutase; MexSO, dimethyl sulfoxide; ESM medium, Erd-Schreiber modified medium; DTPA, diethylenetriaminepentaacetic acid; HX/XO, hypoxanthine/xanthine oxidase.
BY SEA ALGAE
39
tration of DMPO-OOH was determined by a double-integration of the ESR spectrum, in which 1.0 pM of Tempo1 solution was used for a primary standard of ESR absorption. Chemiluminescence assay. Chemiluminescence amplified by luminol was used for kinetic assessment of superoxide production by C. marina. After the addition of 100 ~1 of luminol solution (final concentration, 1.13 pM) to 900 al of algal suspension (2-3 X 10’ cells/ml in ESM medium), the chemiluminescence response was continuously recorded at 26°C during 5 min with a six-channel luminometer (Model LB 9505, Laboratorium Berthold AG, Wildbad, Germany) with use of transparent polystyrene vials. Cytochrome c reduction assay. After the addition of cytochrome c (final concentration, 50 MM) to the plankton suspension in ESM medium, the reduction of cytochrome c was measured as an increase in the optical density at 550-540 nm with a spectrophotometer (Beckman DU-40, Beckman Instruments, Inc., Fullerton, CA) in the presence or absence of 100 U/ml of SOD at 26°C. Optical density reading was converted to nanomoles cytochrome c reduced by using the molar absorbance coefficient of 19.1 mM-’ cm-’ (10) after subtracting the 0 time value in the presence of 100 U/ml of SOD as a background value.
RESULTS AND DISCUSSION
The generation of oxygen radicals in C. marina suspension was measured by ESR spin trapping. The superoxide and hydroxyl radicals react with a spin trap DMPO to yield a DMPO-OOH signal (aN = 1.43 mT, a& = 1.17 mT, and a& = 0.13 mT) and a DMPO-OH signal (aN = au = 1.48 mT), respectively (11,12). The hyperfine splitting constants for the adducts of these radicals and the half-life of the adducts formed were quite different; thus the signal of DMPO-OOH can be distinguished easily from that of DMPO-OH. Our ESR study showed that both DMPO-OOH adducts with a 12-lined signal (an = 1.42 mT, a& = 1.12 mT, and a& = 0.13 mT) and DMPO-OH adducts with a 4-lined signal (aN = an = 1.49 mT) were produced in the suspension of intact C. marina (13, 14) (Fig. 1A). Analysis of the time course of the spin trap signals showed that the DMPO-OOH signal gradually changed to the DMPOOH signal. To characterize these generated radicals, we examined the effect of SOD. As shown in Fig. lB, the addition of 500 U/ml of SOD resulted in the disappearance of the DMPO-OOH signal and also a decrease in the DMPO-OH signal to a negligible level. Furthermore, the generation of oxygen radicals from C. marina depended on the viability of the plankton cells; no spin adduct signal was observed when plankton cells were ruptured by sonication (Fig. 1E). Heating the algal cells at 56°C for 30 min also results in complete inhibition of the generation of superoxide and hydroxyl radical spin adducts of DMPO (data not shown). Detection of DMPO-OH does not necessarily mean that * OH has been trapped. In fact, it was reported that DMPO-OH could be formed via the breakdown of DMPO-OOH adduct, which is formed by the reaction of superoxide with DMPO (13). However, in our experiments, the generation of DMPO-OH was completely inhibited by the addition of Me$SO or ethanol, with simul-
40
ODA
ET AL.
presence of SOD or catalase (Figs. 2D and 2E). No nitroxide was detected in the absence of Me$SO (data not shown), which indicates that spin trapping of lipid radicals such as alkyl and alkoxyl radical (14) originating from planktons or ESM medium was not responsible for generation of these radicals. Using ESR spin trapping with DMPO, the amount of superoxide radical could be quantitated by a double integration of the ESR spectrum (8), in which the hypoxanthine/xanthine oxidase (HX/XO) system was used for the reference standard of superoxide generation. Spin trapping integration (amount of DMPO-OOH) exhibited a good correlation with that determined by the SOD-inhibitable cytochrome c reduction (8) (Fig. 3). In the case of C. marina, it was found that superoxide generation checked by the cytochrome c reduction assay showed a linear increment up to 2.0 min (Fig. 5) (7) similar to the enzymatic generation of 0; by the HX/XO system (8). Therefore, the data of spin signal integration for FIG. 1. ESR spectra of DMPO spin adducts obtained witb suspensions of intact (A, B, C, and D) or ultrasonic-treated (E) C. marina. (A) The spectra were recorded at 1, 3.5, and 6 min after addition of DMPO to the algal suspension. (B) The spectrum at 1 min after addition of DMPO in the presence of SOD (500 U/ml). (C) The spectrum at 3.5 min after addition of DMPO in the presence of 5% MerSO. A sextet signal marked with closed circles belongs to the adducts of methyl radical. (D) Same as (C) with 5% ethanol instead of MesSO. Sextet lines marked with open circles belong to the adduct of a-hydroxyethyl radical. (E) The spectrum of the ultrasonic-treated algal suspension at 1 min. In this last case, plankton cells were completely ruptured during 60-s sonication. See text for details of ESR measurement.
taneous generation of sextet signals as shown in Figs. 1C and 1D. Computer simulation of these spectra demonstrated that the hyperfine splitting constant of each component signal was aN = 1.64 mT and an = 2.34 mT (Fig. 1C) or aN = 1.58 mT and an = 2.28 mT (Fig. lD), indicative of the trapped methyl radical or a-hydroxyethyi radical, respectively (14). Furthermore, we assessed the generation of hydroxyl radical by using an independent spin trapping with a combination of PBN and MezSO as designed by Britigan et al. (9). In this assay system, hydroxyl radical generated in tested samples readily reacts with Me&SO, resulting in production of methoxyl radical (. OCHB) in addition to methyl radical (. CHs) under aerobic conditions. As shown in Fig. 2A, we observed the generation of two species of radicals, for which hyperfine splitting constants were assigned by computer simulation (Fig. 2B) to be aN = 1.508 mT and an = 0.351 mT, and aN = 1.651 mT and on = 0.370 mT, respectively. These hyperfine splitting constants were quite compatible with those reported by Britigan et al. (9) for PBN-OCH, and for PBN-CHB . In addition, these radicals in suspensions of C. marinu were inhibited to negligible levels in the
C
D
Ei MnO
1- mT
-I
MnO
FIG. 2. ESR spectra of PBN adducts observed by the spin trapping with PBN and Me&SO. ESR experiments were performed in the same manner as spin trapping with DMPO as designed by Britigan et al. (9) except that reaction mixtures contained 50 mM of PBN and 10% Me&SO instead of DMPO. (A) The spectrum was recorded at 8.5 min after addition of PBN to algal suspension in the presence of Me&l0 and DTPA. (B) The computer simulation of experimental spectrum consisting of two components: a signal with an = 1.508 mT and on = 0.351 mT; line width, 0.15 mT; weight, 0.7; and a signal with on = 1.651 mT and an = 0.370 mT, line width, 0.12 mT, weight, 0.3. (C) The spectrum recorded at 8.5 min after addition of PBN to the same mixture as (A) containing hypoxanthine (100 pM) and xanthine oxidase (0.02 U/ml) instead of C. marinu (D) The spectrum recorded with the same mixture as (A) with SOD (560 U/ml). (E) Same as (D) with 1009 U/ml of cat&se instead of SOD. The conditions of ESR measurements were the same as those in Fig. 1.
HYDROXYL
RADICAL
PRODUCTION
0
FIG. 3. Quantitation of superoxide radical generation by C. marina using ESR spin trapping. The standard curve of the ESR spin trapping assay for superoxide generated by C. marina was prepared by using the HX/XO system, a quantitative system generating superoxide. The amount of superoxide radical in a given aliquot was determined by SODinhibitable cytochrome c reduction, and simultaneously ESR spin trapping was employed to measure the amount of DMPO-OOH formed in each HX/XO system in ESM medium (see text for details). The value of 0; generated was plotted against that of DMPO-OOH in each assay mixture (0). A linear and well-correlated line was observed between the amount of superoxide by cytochrome c reduction assay and DMPOOOH (correlation efficiency: r2 = 0.998). Based upon this standard curve, the amount of superoxide radical produced by C. marina was calculated from the values determined by seven experiments with different cell density (1.5-2.5 X 10’ cells/ml) using ESR spin trapping (0). Mean values of the amount of superoxide radical and DMPO-OOH generated in algal suspension of 1.0 X 10’ cells were indicated with dashed lines, and SE for each mean value was expressed with two crossbars.
DMPO-OOH in Fig. 3 by ESR spectroscopy were calculated at 1.0 min after addition of DMPO to the algal suspension, from which it was possible to deduce the amount of superoxide generated by C. marina. As shown in Fig. 3, we estimated that the amount of superoxide produced by C. marina was 2.48 f 0.34 ~M/min/lo4 cells. This value is about 100 times higher than that of mammalian phagocytes per cell during respiratory burst (15-18). These results indicate that the superoxide and hydroxyl radicals were indeed generated in this system, in which the hydroxyl radical could be produced by the reaction of superoxide radical and hydrogen peroxide. The participation of transition metals, such as Fe’+ and Cul+, acting as reducing agents in the formation of hydroxyl radical through the Fenton reaction and the Haber-Weiss cycle has been proposed previously [Eqs. [l] and [2] (19, 20)]. 0; + Me”+ + O2 + Mecn-‘)+ HzOz + Me@-‘)+ +
- OH + OH- + Men+
PI PI
Appropriate amounts of iron are known to be required for efficient growth of phytoplankton (21), and ESM me-
41
BY SEA ALGAE
fFi+-+b, 0.2
0.4 Incubation
0.6 time
0.8
1.0
(min)
FIG. 4. Luminol-dependent chemiluminescence responses of intact C. marina in the absence (A) or presence (B) of SOD (10 U/ml) and response of a suspension of sonicated plankton (C). Immediately after the addition of luminol to plankton suspensions, the chemiluminescence response was recorded with a multichannel analyzer at 26°C. (D) shows luminol alone (the stable background of photon counting in ESM medium without plankton). See text for details.
dium contains EDTA-Fe3+ as a constituent at the concentration of 0.5 PM as well as trace amounts of other transition metal ions. Therefore, it is reasonable to assume that the hydroxyl radicals will be generated via the transition metal (iron)-catalyzed Fenton-type Haber-Weiss reaction as described above. This is substantiated by the almost complete inhibition of hydroxyl radical in algal suspension with either SOD or catalase as shown in Figs. 2D and 2E. To elaborate the catalytic mechanism of iron or trace amounts of other transition metals which exist in ESM medium, we checked the generation of hydroxyl
4-
0
20
10
Time
30
(mid
FIG. 5. Time course of the reduction of cytochrome c in C. marina suspensions in the absence (0) or presence (0) of SOD (106 U/ml). Each point represents the mean of duplicate determinations. SE is less than 10% of the value of each point.
42
ODA ET AL.
radical by adding the HX/XO system to the ESM medium without C. marina. The results showed the remarkable production of hydroxyl radical by HX/XO in this medium as judged by the spin trapping with DMPO (not shown). ESR analysis using PBN and Me,SO, demonstrated in Fig. 2C, also indicated the generation of hydroxyl radical in this system. Because the iron is found in seawater in the 0.01-l pM range (22), we speculate that the abovementioned formation of hydroxyl radical via the generation of superoxide radical by the plankton in the seawater is feasible and that this is what we have observed in Figs. 1 and 2. C. marina kept in ESM medium was assayed for chemiluminescence immediately after the addition of luminol. Luminol alone in ESM medium without any plankton gave only a negligible light emission. In the absence of luminol, a very weak chemiluminescence response was observed with the C. marina suspension (data not shown). As shown in Fig. 4, luminol addition caused a rapid increase in chemiluminescence without a lag time: peak activity of 1.62 X lo6 cpm was reached within 10 s, followed by a rapid decline to the background level. In addition, the experiments using the cytochrome c reduction assay revealed a somewhat slower reaction velocity, with no further reduction of cytochrome c after 10 min (Fig. 5). In the presence of 100 U/ml of SOD, the reduction of cytochrome c was inhibited up to 60%; i.e., the inhibition was incomplete. This incomplete inhibition may be attributed to the drawbacks of this method for quantitation of superoxide radical, with which cytochrome c may be reduced by agents other than superoxide radical, such as cytochrome c reducing enzymes and other reducing agents. Several active oxygens species, such as superoxide, hydrogen peroxide, and singlet oxygen, and hydroxyl radical have been considered to be responsible for the luminolenhanced chemiluminescence (23-26). In the case of C. marina, the addition of 10 U/ml (2.5 pg/ml) of SOD caused more than 90% inhibition of chemiluminescence (Fig. 4), but no appreciable inhibition of chemiluminescence was obtained with scavengers for hydroxyl radical, such as MezSO (data not shown). Therefore, the main cause of luminol-dependent chemiluminescence is the superoxide radical. No significant chemiluminescence was observed in ultrasonic-ruptured (Fig. 4) or heat-treated plankton (data not shown), similar to the results of ESR. An important point is that C. marina generates the superoxide radical under physiologically permissible conditions without addition of specific stimulants or triggers to the algal suspension. This observation is significantly different from superoxide generation by phagocytes in mammals, which require specific stimulants (15-18) for the respiratory burst. However, it may be difficult to eliminate the possibility that the algae are responding to some reagents in the assay mixtures. We checked the ef-
fect of DMPO on the superoxide generation by the algae, which is measured by using cytochrome c reduction assay. In this assay system, almost all superoxide radical generated from the algae could react with cytochrome c (50 PM) even in the presence of DMPO (90 mM) because of a higher rate constant (3.0 X 10 M-l se1 at pH 8.0) (27) for the reaction of superoxide radical with cytochrome c compared with that (15.7 M-’ se1 at pH 8.0) (28) of superoxide with DMPO. The results showed that neither enhancement nor inhibition of superoxide generation by the algae was observed by the addition of DMPO at the concentration of 90 mM or below to the algal suspension (data not shown). Therefore, it seems that DMPO itself could not be a stimulant for the generation of superoxide radical by C. marina in ESR spin trapping with DMPO (Fig. 1). With regard to the mechanism of superoxide generation by C. marina, it may be speculated that C. marina has the metabolic or enzymatic systems that are responsible for superoxide production, similar to higher plant cells (29) rather than animal cells. However, the photodynamic reaction of the plankton may not be involved in this radical generation, because our measurements were done mostly in the dark. It is now generally accepted that the most deleterious effects of oxygen radicals in biological systems are caused by hydroxyl radical, which is a strong oxidant which will destroy various important biomolecules (19, 20). On the contrary, it has recently been reported that the hydroxyl radical produced photochemically on the surface of ocean plays an important role in catabolism in marine processes in ecosystem, i.e., bioutilization for organisms of carbon sources from degradated products by reacting with organic materials (2). Therefore, the production of hydroxyl radical via superoxide radical generation by the red tide organism would cause a significant effect in some aquatic environments in addition to the toxic effects to susceptible aquaculture fishes, which is frequently observed in Japanese fish farming. We do not know what mechanism triggers the rapid propagation of the red tide phytoplankton. However, once superoxide radical is generated it can be converted to hydroxyl radical via Fenton-type Haber-Weiss reaction in the presence of transition metals as described above (Figs. 1 and 2) (Eqs. [l] and [2]). In addition to the known effect of iron in growth triggering (21), the hydroxyl radical thus generated may be beneficial for the plankton itself for bioutilization of various organic compounds in seawater. This assumption may be valid only in the wild environment under oligotrophic conditions but not so in the coastal polluted red tide water with high nutritional elements. Conversely, hydroxyl radical becomes more noxious in the aquaculture at the high density of C. marina In conclusion, by using spin-trapping ESR techniques in addition to chemiluminescence and cytochrome c re-
HYDROXYL
RADICAL
PRODUCTION
duction assays, we report here that the unicellular marine microorganism C. marina generates superoxide radical and hence hydroxyl radical under a physiologically permissible condition. The present findings will provide a clue in understanding the ecology of red tide and in the prevention of red tide toxication in aquaculture. ACKNOWLEDGMENTS We thank Ms. Sumiko Ijiri (K.U.M.S.) for technical assistance in our experiments, Dr. Masahiro Kohno (ESR Application Laboratory, JEOL, Co., Ltd., Akishima, Japan) for stimulating discussion, and Ms. Judith B. Gandy for editorial work.
REFERENCES 1. Atkinson,
R. (1985) Chem. Rev. 85,69-201.
2. Mopper, K., and Zhou, X. (1990) Science 250,661~664. 3. Anderson, D. M. (1989) in Red Tides: Biology, Environmental
4.
5.
6. 7. 8. 9. 10.
Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 11-16, Elsevier, New York. Onoue, Y., and Nozawa, K. (1989) in Red Tides: Biology, Environmental Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 371-374, Elsevier, New York. Shimada, M., Shimono, R., Murakami, T. H., Yoshimatsu, S., and Ono, C. (1989) in Red Tides: Biology, Environmental Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 443-446, Elsevier, New York. Toyoshima, T., Ozaki, H. S., Shimada, M., Okaichi, T., and Murakami, T. H. (1985) Mar. Biol. 88, 101-108. Oda, T., Ishimatau, A., Shimada, M., Takeshita, S., and Muramatau, T. (1992) Mar. Bill., in press. Mitsuta, K., Mizuta, Y., Kohno, M., Hiramatsu, M., and Mori, A. (1996) Bull. Chem. Sot. Jpn. 63, 187-191. Britigan, B. E., Coffman, T. J., and Buettner, G. R. (1990) J. Biol. Chem. 265,2650-2656. Kakinuma, K., and Minakami, S. (1978) B&him. Biophys. Acta
638,50-59.
43
BY SEA ALGAE
11. Harbour,
J. R., Chow, V., and Boiton,
J. R. (1974) Can. J. Chem.
52,3549-3553. 12. Janzen, E. G., Nutter, D. E., Jr., Davis, E. R., Blackburn, B. J., Poyer, J. L., and McCay, P. B. (1978) Can. J. Chem. 56,2237-2242. 13. Finkelstein, Pharmacol. 14. Buettner, 15. Nathan,
E., Rosen, G. M., and Rauckman, 2 1,262-265.
E. J. (1982) Mol.
G. R. (1987) Free Radical Biol. Med. 3,259-304. C. F., and Root, R. K. (1977) J. Exp. Med. 146, 1648-
1662. 16. Oda, T., Morinaga, Med. 181,9-17.
T., and Maeda, H. (1986) Proc. Sot. Exp. Biol.
17. Badway, J. A., and Karnovsky,
M. L. (1980) Annu. Reu. Biochem.
49,695-726. 18. Nakashima, H., Ando, M., Sugimoto, M., Suga, M., Soda, K., and Araki, S. (1987) Am. Rev. Respir. Dis. 136,310-315. 19. Halliwell, 14.
B., and Gutteridge,
J. M. C. (1984) Biochem. J. 219, l-
20. Simpson, J. A., Cheeseman, K. H., Smith, S. E., and Dean, R. T. (1988) Biochem. J. 254,519-523.
21. Sunda, W. G., Swift, D. G., and Huntsman, S. A. (1991) Nature 351,55-57. 22. Brewer, P. G. (1975) in Chemical Oceanography: Minor Elements in Sea Water (Riley, J. P., and Skirrow, Academic Press, New York.
G., Eds.), pp. 416-490,
23. Allen, R. C., Stjernholm,
R. L., and Steele, R. H. (1972) Biochem. Biophys. Rex Commun. 47,679-684.
24. Rosen, H., and Klebanoff, S. J. (1976) J. Clin. Znuest. 58, 50-60. 25. Cheson, B. E., Christensen, R. L., Sperling, R., Kohler, B. E., and Babior, B. M. (1976) J. Clin. Invest. 58, 789-796.
26. Ushijima, Y., and Nakano, M. (1980) J. Appl. Biochem. 2,138-151. 27. Butler, J., Jayson, G. G., and Swallow, A. J. (1975) B&him. Biophys. Acta 408,215-222.
28. Finkelsmin,
E., Rosen, G. M., and Rauckman, Chem. Sot. 102,4994-4999.
29. Asada, K., Kiso, K., and Yoshikawa, 2175-2181.
E. J. (1980) J. Am.
K. (1974) J. Biol. Chem. 249,
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 294, No. 1, April, pp. 38-43, 1992
Hydroxyl Radical Generation
by Red Tide Algae’
Tatsuya Oda,* Takaaki Akaike,f‘ Keizo Sate,? Atsushi Ishimatsu,$ Satoshi Takeshita,* Tsuyoshi Muramatsu,* and Hiroshi Maedat*2 *Division of Biochemistry, Faculty of Fisheries, Nagasaki University, Nagasaki 852, Japan; tDepartm.ent of Microbiology, Kumamoto University Medical School, Kumumoto 860, Japan; and $Nomo Fisheries Station, Nagasaki University, Nomozaki, Nagasaki 851-05, Japan
Received August 23,1991, and in revised form November
22,199l
Hydroxyl radical induced photochemically plays an important role in atmospheric chemistry (1). Its generation by photodynamic reaction in the atmosphere and in the sea is documented relatively recently (1, 2). However, little information is available for the generation of oxygen radicals in the marine biosphere. There has recently been a global increase in the frequency, magnitude, and geographical extent of red tides. This increase seems to be well correlated with the degree of coastal pollution or the utilization of coastal water for aquaculture (3). Several species of phytoplankton have been identified as the causative organism of red tide (3). Chuttonella marina (Raphidophyceae) is one of the most noxious red tide phytoplankton and causes severe damage to fish farming. Although several hypotheses have been proposed about the cause of fish death by Chattondlu sp. (4,5), the precise mechanism of the toxic action remains unclear. It has been reported that edema formation in gill lamellae of fish was induced by Chattondu sp. as a typical histological change and that the gill damage resulted in fish death because of oxygen deficiency (6). Our preliminary experiments indicated that C. marina exhibited the active oxygen-mediated toxic effect against Vibrio alginolyticus, which was isolated from a seaweed (7). Thus, our study was undertaken to confirm the generation of oxygen radicals by C. marina by using various methods and to identify the radical species by ESR spin trapping.
unicellular marine phytoplankton Chattonella marina is known to have toxic effects against various living marine organisms, especially fishes. However, details of the mechanism of the toxicity of this plankton remain obscure. Here we demonstrate the generation of superoxide and hydroxyl radicals from a red tide unicellular organism, C. marina, by using ESR spectroscopy with the spin traps 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) and N-t-butyl-cr-phenylnitrone (PBN), and by using the luminol-enhanced chemiluminescence response. The spin-trapping assay revealed productions of spin adduct of superoxide anion (0;) (DMPO-OOH) and that of hydroxyl radical (. OH) (DMPO-OH) in the algal suspension, which was not observed in the ultrasonic-ruptured suspension. The addition of superoxide dismutase (500 U/ml) almost completely inhibited the formation of both DMPO-OOH and DMPO-OH, and carbon-centered radicals were generated with the disappearance of DMPO-OH after addition of 5% dimethyl sulfoxide (MesSO) and 5% ethanol. Furthermore, the generation of methyl and methoxyl radicals, which are thought to be produced by the reaction of hydroxyl radical and Mess0 under aerobic condition, was identified using spin trapping with a combination of PBN and MesSO. Luminol-enhanced chemiluminescence assay also supported the above observations. These results clearly indicate and releases the superoxide that C. marina generates radical followed by the production of hydroxyl radical to the surrounding environment. The velocity of superoxide generation by C. marina was about 100 times faster than that by mammalian phagocytes per cell basis. The generation of oxygen radical is suggested to be a pathogenic principle in the toxication of red tide to susceptible aquaculture fishes and may be directly correlated with the coastal pollution by red tide. Q 1~82 Academic
1 Supported in part by Grants-in-Aid for Scientific Research from Monbusho (the Ministry of Education, Science, and Culture) of Japan,
Press, Inc.
1990-1991.
The
’ To whom correspondence
should be addressed.
38 Copyright
ooo3-9S61/92 $3.00 0 1992 by Academic Press, Inc.
All rigbta of reproduction in any
form
reserved
HYDROXYL
EXPERIMENTAL
RADICAL
PRODUCTION
PROCEDURES
Materials We used 5,5-dimethyl-1-pyrroline-N-oxide (DMP0)3 and N-t-butyla-phenylnitrone (PBN) throughout the experiments as spin traps. DMPO was synthesized and purified by Mitaui Toatsu Chemical Co., Ltd., Mobara, Japan, or Dojin Chemical Laboratories Co., Ltd., Kumamoto, Japan. PBN, 2,2,6,6-tetramethyl-4-hydroxypiperidine (Tempol), luminol, horse heart ferricytochrome c (type VI), and catalase were purchased from Sigma Chemical Co., St. Louis, Missouri, and human recombinant Cu, Zn-superoxide dismutase (SOD) was a gift from Nippon Kayaku Co., Ltd., Tokyo, Japan. Xanthine oxidase from cow milk was a product of Boehringer-Mannheim GmbH, Mannheim, Germany. Other chemicals were of the highest grade commercially available.
Methods Culture for phytoplankton. C. marina was generously provided by Kagoshima Prefectural Fisheries Experimental Station, Japan, and was cultured at 26’C in Erd-Schreiber modified (ESM) medium (pH 8.0) under 3000 lx illumination with a cycle of 12 h light and 12 h dark (4). ESM medium was prepared according to the composition reported originally without soil extract, i.e., 120 mg NaN03, 5 mg K,HPO,, 0.1 mg vitamin B1 , 0.01 mg vitamin B,z, 0.001 mg biotin, 0.26 mg EDTA-Fe3+, 0.33 mg EDTA-Mn’+, and 1 g tris(hydroxymethyl)aminomethane were dissolved into 1 liter of seawater and the pH was adjusted to 8.0, followed by filtration with a 0.25-pm (pore size) filter and autoclaving (121°C, 30 min). All cultivation was carried out with using sterilized instruments. In addition, microscopic examination at Xl000 magnification of wellgrown algal suspension revealed neither distinctive proliferative bacteria or fungi other than C. marina in the medium. In the case of spin trapping with DMPO, 140 ESR spin trapping. gl of algal suspension (final concentration, 1.5-2.5 X lo4 cells/ml in ESM medium) was mixed with DMPO in the presence of 500 pM of diethylenetriaminepentaacetic acid (DTPA) (final volume, 200 ~1). DMPO appears to be toxic to the algae at a concentration higher than 90 mM. Therefore, we usually used 90 mM of DMPO for measurements of oxygen radicals, although radical capturing efficiency, especially against superoxide radical, became reduced to about lo-20% of the saturated level of DMPO-OOH, which was obtained at a concentration higher than 0.67 M of DMPO, as reported by Mitsuta et al. (8). The mixture was immediately transferred to the ESR quartz flat cell (inner size, 60 X 10 X 0.31 mm), and the cell was placed in a JEOL JES-RElX ESR spectrometer (JEOL Co., Ltd., Akishima, Japan). ESR spectra were recorded at room temperature under the following conditions: modulation frequency, 100 kHz; modulation amplitude, 0.079 mT, scanning field, 336.2 + 5 mT; receiver gain, 10 X 100, response time, 0.1 s; sweep time, 2 min; microwave power, 40 mW; and microwave frequency, 9.421 GHz. To confirm the generation of hydroxyl radical in plankton suspension, we used the spin trapping with PBN and dimethyl sulfoxide (Me&SO) designed by Britigan et al. (9). Namely ESR experiments were performed in the same manner as spin trapping with DMPO except that reaction mixtures contained 50 mM of PBN and 10% Me,SO instead of DMPO. Quuntitation of DMPO-OOH by ESR spectroscopy. After recording the ESR spectra as described above, the signal intensity of DMPOOOH was normalized as a relative height against the standard signal intensity of the manganese oxide marker (MnO). An absolute concen-
3 Abbreviations used: DMPO, 5,5-dimethyl-1-pyrroline-N-oxide; PBN, N-t-butyl-a-phenylnitrone; Tempol, 2,2,6,6-tetramethyl-4-hydroxypiperidine; SOD, superoxide dismutase; MexSO, dimethyl sulfoxide; ESM medium, Erd-Schreiber modified medium; DTPA, diethylenetriaminepentaacetic acid; HX/XO, hypoxanthine/xanthine oxidase.
BY SEA ALGAE
39
tration of DMPO-OOH was determined by a double-integration of the ESR spectrum, in which 1.0 pM of Tempo1 solution was used for a primary standard of ESR absorption. Chemiluminescence assay. Chemiluminescence amplified by luminol was used for kinetic assessment of superoxide production by C. marina. After the addition of 100 ~1 of luminol solution (final concentration, 1.13 pM) to 900 al of algal suspension (2-3 X 10’ cells/ml in ESM medium), the chemiluminescence response was continuously recorded at 26°C during 5 min with a six-channel luminometer (Model LB 9505, Laboratorium Berthold AG, Wildbad, Germany) with use of transparent polystyrene vials. Cytochrome c reduction assay. After the addition of cytochrome c (final concentration, 50 MM) to the plankton suspension in ESM medium, the reduction of cytochrome c was measured as an increase in the optical density at 550-540 nm with a spectrophotometer (Beckman DU-40, Beckman Instruments, Inc., Fullerton, CA) in the presence or absence of 100 U/ml of SOD at 26°C. Optical density reading was converted to nanomoles cytochrome c reduced by using the molar absorbance coefficient of 19.1 mM-’ cm-’ (10) after subtracting the 0 time value in the presence of 100 U/ml of SOD as a background value.
RESULTS AND DISCUSSION
The generation of oxygen radicals in C. marina suspension was measured by ESR spin trapping. The superoxide and hydroxyl radicals react with a spin trap DMPO to yield a DMPO-OOH signal (aN = 1.43 mT, a& = 1.17 mT, and a& = 0.13 mT) and a DMPO-OH signal (aN = au = 1.48 mT), respectively (11,12). The hyperfine splitting constants for the adducts of these radicals and the half-life of the adducts formed were quite different; thus the signal of DMPO-OOH can be distinguished easily from that of DMPO-OH. Our ESR study showed that both DMPO-OOH adducts with a 12-lined signal (an = 1.42 mT, a& = 1.12 mT, and a& = 0.13 mT) and DMPO-OH adducts with a 4-lined signal (aN = an = 1.49 mT) were produced in the suspension of intact C. marina (13, 14) (Fig. 1A). Analysis of the time course of the spin trap signals showed that the DMPO-OOH signal gradually changed to the DMPOOH signal. To characterize these generated radicals, we examined the effect of SOD. As shown in Fig. lB, the addition of 500 U/ml of SOD resulted in the disappearance of the DMPO-OOH signal and also a decrease in the DMPO-OH signal to a negligible level. Furthermore, the generation of oxygen radicals from C. marina depended on the viability of the plankton cells; no spin adduct signal was observed when plankton cells were ruptured by sonication (Fig. 1E). Heating the algal cells at 56°C for 30 min also results in complete inhibition of the generation of superoxide and hydroxyl radical spin adducts of DMPO (data not shown). Detection of DMPO-OH does not necessarily mean that * OH has been trapped. In fact, it was reported that DMPO-OH could be formed via the breakdown of DMPO-OOH adduct, which is formed by the reaction of superoxide with DMPO (13). However, in our experiments, the generation of DMPO-OH was completely inhibited by the addition of Me$SO or ethanol, with simul-
40
ODA
ET AL.
presence of SOD or catalase (Figs. 2D and 2E). No nitroxide was detected in the absence of Me$SO (data not shown), which indicates that spin trapping of lipid radicals such as alkyl and alkoxyl radical (14) originating from planktons or ESM medium was not responsible for generation of these radicals. Using ESR spin trapping with DMPO, the amount of superoxide radical could be quantitated by a double integration of the ESR spectrum (8), in which the hypoxanthine/xanthine oxidase (HX/XO) system was used for the reference standard of superoxide generation. Spin trapping integration (amount of DMPO-OOH) exhibited a good correlation with that determined by the SOD-inhibitable cytochrome c reduction (8) (Fig. 3). In the case of C. marina, it was found that superoxide generation checked by the cytochrome c reduction assay showed a linear increment up to 2.0 min (Fig. 5) (7) similar to the enzymatic generation of 0; by the HX/XO system (8). Therefore, the data of spin signal integration for FIG. 1. ESR spectra of DMPO spin adducts obtained witb suspensions of intact (A, B, C, and D) or ultrasonic-treated (E) C. marina. (A) The spectra were recorded at 1, 3.5, and 6 min after addition of DMPO to the algal suspension. (B) The spectrum at 1 min after addition of DMPO in the presence of SOD (500 U/ml). (C) The spectrum at 3.5 min after addition of DMPO in the presence of 5% MerSO. A sextet signal marked with closed circles belongs to the adducts of methyl radical. (D) Same as (C) with 5% ethanol instead of MesSO. Sextet lines marked with open circles belong to the adduct of a-hydroxyethyl radical. (E) The spectrum of the ultrasonic-treated algal suspension at 1 min. In this last case, plankton cells were completely ruptured during 60-s sonication. See text for details of ESR measurement.
taneous generation of sextet signals as shown in Figs. 1C and 1D. Computer simulation of these spectra demonstrated that the hyperfine splitting constant of each component signal was aN = 1.64 mT and an = 2.34 mT (Fig. 1C) or aN = 1.58 mT and an = 2.28 mT (Fig. lD), indicative of the trapped methyl radical or a-hydroxyethyi radical, respectively (14). Furthermore, we assessed the generation of hydroxyl radical by using an independent spin trapping with a combination of PBN and MezSO as designed by Britigan et al. (9). In this assay system, hydroxyl radical generated in tested samples readily reacts with Me&SO, resulting in production of methoxyl radical (. OCHB) in addition to methyl radical (. CHs) under aerobic conditions. As shown in Fig. 2A, we observed the generation of two species of radicals, for which hyperfine splitting constants were assigned by computer simulation (Fig. 2B) to be aN = 1.508 mT and an = 0.351 mT, and aN = 1.651 mT and on = 0.370 mT, respectively. These hyperfine splitting constants were quite compatible with those reported by Britigan et al. (9) for PBN-OCH, and for PBN-CHB . In addition, these radicals in suspensions of C. marinu were inhibited to negligible levels in the
C
D
Ei MnO
1- mT
-I
MnO
FIG. 2. ESR spectra of PBN adducts observed by the spin trapping with PBN and Me&SO. ESR experiments were performed in the same manner as spin trapping with DMPO as designed by Britigan et al. (9) except that reaction mixtures contained 50 mM of PBN and 10% Me&SO instead of DMPO. (A) The spectrum was recorded at 8.5 min after addition of PBN to algal suspension in the presence of Me&l0 and DTPA. (B) The computer simulation of experimental spectrum consisting of two components: a signal with an = 1.508 mT and on = 0.351 mT; line width, 0.15 mT; weight, 0.7; and a signal with on = 1.651 mT and an = 0.370 mT, line width, 0.12 mT, weight, 0.3. (C) The spectrum recorded at 8.5 min after addition of PBN to the same mixture as (A) containing hypoxanthine (100 pM) and xanthine oxidase (0.02 U/ml) instead of C. marinu (D) The spectrum recorded with the same mixture as (A) with SOD (560 U/ml). (E) Same as (D) with 1009 U/ml of cat&se instead of SOD. The conditions of ESR measurements were the same as those in Fig. 1.
HYDROXYL
RADICAL
PRODUCTION
0
FIG. 3. Quantitation of superoxide radical generation by C. marina using ESR spin trapping. The standard curve of the ESR spin trapping assay for superoxide generated by C. marina was prepared by using the HX/XO system, a quantitative system generating superoxide. The amount of superoxide radical in a given aliquot was determined by SODinhibitable cytochrome c reduction, and simultaneously ESR spin trapping was employed to measure the amount of DMPO-OOH formed in each HX/XO system in ESM medium (see text for details). The value of 0; generated was plotted against that of DMPO-OOH in each assay mixture (0). A linear and well-correlated line was observed between the amount of superoxide by cytochrome c reduction assay and DMPOOOH (correlation efficiency: r2 = 0.998). Based upon this standard curve, the amount of superoxide radical produced by C. marina was calculated from the values determined by seven experiments with different cell density (1.5-2.5 X 10’ cells/ml) using ESR spin trapping (0). Mean values of the amount of superoxide radical and DMPO-OOH generated in algal suspension of 1.0 X 10’ cells were indicated with dashed lines, and SE for each mean value was expressed with two crossbars.
DMPO-OOH in Fig. 3 by ESR spectroscopy were calculated at 1.0 min after addition of DMPO to the algal suspension, from which it was possible to deduce the amount of superoxide generated by C. marina. As shown in Fig. 3, we estimated that the amount of superoxide produced by C. marina was 2.48 f 0.34 ~M/min/lo4 cells. This value is about 100 times higher than that of mammalian phagocytes per cell during respiratory burst (15-18). These results indicate that the superoxide and hydroxyl radicals were indeed generated in this system, in which the hydroxyl radical could be produced by the reaction of superoxide radical and hydrogen peroxide. The participation of transition metals, such as Fe’+ and Cul+, acting as reducing agents in the formation of hydroxyl radical through the Fenton reaction and the Haber-Weiss cycle has been proposed previously [Eqs. [l] and [2] (19, 20)]. 0; + Me”+ + O2 + Mecn-‘)+ HzOz + Me@-‘)+ +
- OH + OH- + Men+
PI PI
Appropriate amounts of iron are known to be required for efficient growth of phytoplankton (21), and ESM me-
41
BY SEA ALGAE
fFi+-+b, 0.2
0.4 Incubation
0.6 time
0.8
1.0
(min)
FIG. 4. Luminol-dependent chemiluminescence responses of intact C. marina in the absence (A) or presence (B) of SOD (10 U/ml) and response of a suspension of sonicated plankton (C). Immediately after the addition of luminol to plankton suspensions, the chemiluminescence response was recorded with a multichannel analyzer at 26°C. (D) shows luminol alone (the stable background of photon counting in ESM medium without plankton). See text for details.
dium contains EDTA-Fe3+ as a constituent at the concentration of 0.5 PM as well as trace amounts of other transition metal ions. Therefore, it is reasonable to assume that the hydroxyl radicals will be generated via the transition metal (iron)-catalyzed Fenton-type Haber-Weiss reaction as described above. This is substantiated by the almost complete inhibition of hydroxyl radical in algal suspension with either SOD or catalase as shown in Figs. 2D and 2E. To elaborate the catalytic mechanism of iron or trace amounts of other transition metals which exist in ESM medium, we checked the generation of hydroxyl
4-
0
20
10
Time
30
(mid
FIG. 5. Time course of the reduction of cytochrome c in C. marina suspensions in the absence (0) or presence (0) of SOD (106 U/ml). Each point represents the mean of duplicate determinations. SE is less than 10% of the value of each point.
42
ODA ET AL.
radical by adding the HX/XO system to the ESM medium without C. marina. The results showed the remarkable production of hydroxyl radical by HX/XO in this medium as judged by the spin trapping with DMPO (not shown). ESR analysis using PBN and Me,SO, demonstrated in Fig. 2C, also indicated the generation of hydroxyl radical in this system. Because the iron is found in seawater in the 0.01-l pM range (22), we speculate that the abovementioned formation of hydroxyl radical via the generation of superoxide radical by the plankton in the seawater is feasible and that this is what we have observed in Figs. 1 and 2. C. marina kept in ESM medium was assayed for chemiluminescence immediately after the addition of luminol. Luminol alone in ESM medium without any plankton gave only a negligible light emission. In the absence of luminol, a very weak chemiluminescence response was observed with the C. marina suspension (data not shown). As shown in Fig. 4, luminol addition caused a rapid increase in chemiluminescence without a lag time: peak activity of 1.62 X lo6 cpm was reached within 10 s, followed by a rapid decline to the background level. In addition, the experiments using the cytochrome c reduction assay revealed a somewhat slower reaction velocity, with no further reduction of cytochrome c after 10 min (Fig. 5). In the presence of 100 U/ml of SOD, the reduction of cytochrome c was inhibited up to 60%; i.e., the inhibition was incomplete. This incomplete inhibition may be attributed to the drawbacks of this method for quantitation of superoxide radical, with which cytochrome c may be reduced by agents other than superoxide radical, such as cytochrome c reducing enzymes and other reducing agents. Several active oxygens species, such as superoxide, hydrogen peroxide, and singlet oxygen, and hydroxyl radical have been considered to be responsible for the luminolenhanced chemiluminescence (23-26). In the case of C. marina, the addition of 10 U/ml (2.5 pg/ml) of SOD caused more than 90% inhibition of chemiluminescence (Fig. 4), but no appreciable inhibition of chemiluminescence was obtained with scavengers for hydroxyl radical, such as MezSO (data not shown). Therefore, the main cause of luminol-dependent chemiluminescence is the superoxide radical. No significant chemiluminescence was observed in ultrasonic-ruptured (Fig. 4) or heat-treated plankton (data not shown), similar to the results of ESR. An important point is that C. marina generates the superoxide radical under physiologically permissible conditions without addition of specific stimulants or triggers to the algal suspension. This observation is significantly different from superoxide generation by phagocytes in mammals, which require specific stimulants (15-18) for the respiratory burst. However, it may be difficult to eliminate the possibility that the algae are responding to some reagents in the assay mixtures. We checked the ef-
fect of DMPO on the superoxide generation by the algae, which is measured by using cytochrome c reduction assay. In this assay system, almost all superoxide radical generated from the algae could react with cytochrome c (50 PM) even in the presence of DMPO (90 mM) because of a higher rate constant (3.0 X 10 M-l se1 at pH 8.0) (27) for the reaction of superoxide radical with cytochrome c compared with that (15.7 M-’ se1 at pH 8.0) (28) of superoxide with DMPO. The results showed that neither enhancement nor inhibition of superoxide generation by the algae was observed by the addition of DMPO at the concentration of 90 mM or below to the algal suspension (data not shown). Therefore, it seems that DMPO itself could not be a stimulant for the generation of superoxide radical by C. marina in ESR spin trapping with DMPO (Fig. 1). With regard to the mechanism of superoxide generation by C. marina, it may be speculated that C. marina has the metabolic or enzymatic systems that are responsible for superoxide production, similar to higher plant cells (29) rather than animal cells. However, the photodynamic reaction of the plankton may not be involved in this radical generation, because our measurements were done mostly in the dark. It is now generally accepted that the most deleterious effects of oxygen radicals in biological systems are caused by hydroxyl radical, which is a strong oxidant which will destroy various important biomolecules (19, 20). On the contrary, it has recently been reported that the hydroxyl radical produced photochemically on the surface of ocean plays an important role in catabolism in marine processes in ecosystem, i.e., bioutilization for organisms of carbon sources from degradated products by reacting with organic materials (2). Therefore, the production of hydroxyl radical via superoxide radical generation by the red tide organism would cause a significant effect in some aquatic environments in addition to the toxic effects to susceptible aquaculture fishes, which is frequently observed in Japanese fish farming. We do not know what mechanism triggers the rapid propagation of the red tide phytoplankton. However, once superoxide radical is generated it can be converted to hydroxyl radical via Fenton-type Haber-Weiss reaction in the presence of transition metals as described above (Figs. 1 and 2) (Eqs. [l] and [2]). In addition to the known effect of iron in growth triggering (21), the hydroxyl radical thus generated may be beneficial for the plankton itself for bioutilization of various organic compounds in seawater. This assumption may be valid only in the wild environment under oligotrophic conditions but not so in the coastal polluted red tide water with high nutritional elements. Conversely, hydroxyl radical becomes more noxious in the aquaculture at the high density of C. marina In conclusion, by using spin-trapping ESR techniques in addition to chemiluminescence and cytochrome c re-
HYDROXYL
RADICAL
PRODUCTION
duction assays, we report here that the unicellular marine microorganism C. marina generates superoxide radical and hence hydroxyl radical under a physiologically permissible condition. The present findings will provide a clue in understanding the ecology of red tide and in the prevention of red tide toxication in aquaculture. ACKNOWLEDGMENTS We thank Ms. Sumiko Ijiri (K.U.M.S.) for technical assistance in our experiments, Dr. Masahiro Kohno (ESR Application Laboratory, JEOL, Co., Ltd., Akishima, Japan) for stimulating discussion, and Ms. Judith B. Gandy for editorial work.
REFERENCES 1. Atkinson,
R. (1985) Chem. Rev. 85,69-201.
2. Mopper, K., and Zhou, X. (1990) Science 250,661~664. 3. Anderson, D. M. (1989) in Red Tides: Biology, Environmental
4.
5.
6. 7. 8. 9. 10.
Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 11-16, Elsevier, New York. Onoue, Y., and Nozawa, K. (1989) in Red Tides: Biology, Environmental Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 371-374, Elsevier, New York. Shimada, M., Shimono, R., Murakami, T. H., Yoshimatsu, S., and Ono, C. (1989) in Red Tides: Biology, Environmental Science, and Toxicology (Okaichi, T., Anderson, D. M., and Nemoto, T., Eds.), pp. 443-446, Elsevier, New York. Toyoshima, T., Ozaki, H. S., Shimada, M., Okaichi, T., and Murakami, T. H. (1985) Mar. Biol. 88, 101-108. Oda, T., Ishimatau, A., Shimada, M., Takeshita, S., and Muramatau, T. (1992) Mar. Bill., in press. Mitsuta, K., Mizuta, Y., Kohno, M., Hiramatsu, M., and Mori, A. (1996) Bull. Chem. Sot. Jpn. 63, 187-191. Britigan, B. E., Coffman, T. J., and Buettner, G. R. (1990) J. Biol. Chem. 265,2650-2656. Kakinuma, K., and Minakami, S. (1978) B&him. Biophys. Acta
638,50-59.
43
BY SEA ALGAE
11. Harbour,
J. R., Chow, V., and Boiton,
J. R. (1974) Can. J. Chem.
52,3549-3553. 12. Janzen, E. G., Nutter, D. E., Jr., Davis, E. R., Blackburn, B. J., Poyer, J. L., and McCay, P. B. (1978) Can. J. Chem. 56,2237-2242. 13. Finkelstein, Pharmacol. 14. Buettner, 15. Nathan,
E., Rosen, G. M., and Rauckman, 2 1,262-265.
E. J. (1982) Mol.
G. R. (1987) Free Radical Biol. Med. 3,259-304. C. F., and Root, R. K. (1977) J. Exp. Med. 146, 1648-
1662. 16. Oda, T., Morinaga, Med. 181,9-17.
T., and Maeda, H. (1986) Proc. Sot. Exp. Biol.
17. Badway, J. A., and Karnovsky,
M. L. (1980) Annu. Reu. Biochem.
49,695-726. 18. Nakashima, H., Ando, M., Sugimoto, M., Suga, M., Soda, K., and Araki, S. (1987) Am. Rev. Respir. Dis. 136,310-315. 19. Halliwell, 14.
B., and Gutteridge,
J. M. C. (1984) Biochem. J. 219, l-
20. Simpson, J. A., Cheeseman, K. H., Smith, S. E., and Dean, R. T. (1988) Biochem. J. 254,519-523.
21. Sunda, W. G., Swift, D. G., and Huntsman, S. A. (1991) Nature 351,55-57. 22. Brewer, P. G. (1975) in Chemical Oceanography: Minor Elements in Sea Water (Riley, J. P., and Skirrow, Academic Press, New York.
G., Eds.), pp. 416-490,
23. Allen, R. C., Stjernholm,
R. L., and Steele, R. H. (1972) Biochem. Biophys. Rex Commun. 47,679-684.
24. Rosen, H., and Klebanoff, S. J. (1976) J. Clin. Znuest. 58, 50-60. 25. Cheson, B. E., Christensen, R. L., Sperling, R., Kohler, B. E., and Babior, B. M. (1976) J. Clin. Invest. 58, 789-796.
26. Ushijima, Y., and Nakano, M. (1980) J. Appl. Biochem. 2,138-151. 27. Butler, J., Jayson, G. G., and Swallow, A. J. (1975) B&him. Biophys. Acta 408,215-222.
28. Finkelsmin,
E., Rosen, G. M., and Rauckman, Chem. Sot. 102,4994-4999.
29. Asada, K., Kiso, K., and Yoshikawa, 2175-2181.
E. J. (1980) J. Am.
K. (1974) J. Biol. Chem. 249,