- Email: [email protected]
Toxicity of Organic and Inorganic Mercury to Saccharomyces cerevisiae
Ecotoxicology and Environmental Safety 43, 149}155 (1999) Environmental Research, Section B Article ID eesa.1999.1767, available online at http://www.idealibrary.com on
Toxicity of Organic and Inorganic Mercury to Saccharomyces cerevisiae Athanassios Kungolos, Isao Aoyama, and S. Muramoto Laboratory of Ecological Chemistry, Research Institute for Bioresources, Okayama University, 2-20-1, Chu-o, Kurashiki 710, Japan Received April 7, 1998
In this study the e4ect if six di4erent forms of mercury on the growth of the yeast Saccharomyces cerevisiae is presented. Five kinds of strains of S. cerevisiae were used. They were a wild type, a mercury-resistant type, and three mutants: mutation repairde5cient mutant, excision repair-de5cient mutant, and recombination repair-de5cient mutant. In terms of EC50 toward the wild-type strain, the toxicity order for the inorganic forms was Hg(NO3)2 > HgSO4 > HgCl2 . Monovalent nitrate mercury Hg(NO3)2 was more toxic than bivalent Hg(NO3)2. The toxicity of organic mercury CH3HgCl on cell growth was two orders of magnitude higher than that of inorganic HgCl2. Between the two organic forms, CH3HgCl was more toxic than CH3HgOH. The survival rate in the presence of a certain concentration of CH3HgCl was about one-hundredth of the survival in presence of the same concentration of HgCl2. On the other hand, the concentration of CH3HgCl in the cell was about 170 times that of HgCl2. The addition of chelating agents, EDTA and methylpenicillamine, to the medium did not reduce the toxicity of mercury. Among the three mutants tested, the one de5cient in recombination repair systems was the most sensitive to mercury. 1999 Academic Press
Key Words: mercury; toxicity; bioconcentration factor; Saccharomyces cerevisiae.
INTRODUCTION
The high toxicity of organic mercury has been known for a long time. There have been some reports on the di!erence of toxicity between organic and inorganic mercury (Roderer, 1983; Tsai and Olson, 1990). Not much research has been done on the di!erence of toxicity between various forms of organic mercury or inorganic mercury (Farrel et al., 1990, 1993). Although the e!ects of mercury on a lot of organisms have been evaluated (Lee et al., 1992; Mattheis Present address: Chemical Process Engineering Research Institute, P.O. Box 361, Thermi, 57001 Thessaloniki, Greece. To whom all correspondence should be addressed.
et al., 1994; McCrary and Heagler, 1997; Schulz and Newman, 1997), there are not many reports on the e!ect of mercury on Saccharomyces cerevisiae (Kungolos and Aoyama, 1992). Only a few researchers have used S. cerevisiae in toxicity assessments. (Bitton et al., 1984; Kwashiewska and Kaiser, 1984; Eckardt and Siede, 1985). The authors of this article have advocated the use of S. cerevisiae as a eukaryotic organism, for the development of a quick automated method for toxicity screening of chemicals (Kungolos and Aoyama, 1992). Increased mercury contamination in aquatic systems has become a major concern. The rise in mercury concentrations in water and sediments leads to increased bioaccumulation by organisms (Camusso et al., 1995; Eisemann et al., 1997). The relationship between bioaccumulation and toxicity has been established with a number of organisms, such as Daphnia magna (Kungolos and Aoyama, 1993) and Chlorella ellipsoidea (Aoyama et al., 1987). However, because the exact mechanisms of toxicity are not fully understood, there is a need for more data, especially in the case of species of the same type with a big di!erence in toxicity. In this research, the e!ect of two organic forms of mercury, CH HgOH and CH HgCl, and four inorganic ones, HgCl , HgSO , Hg(NO ) , and Hg (NO ) , is evaluated. Then, the bioaccumulation of mercury in the cells of S. cerevisiae, exposed to CH HgCl or to HgCl is compared. Finally, the e!ect of two complexing agents in the toxicity of mercury to S. cerevisiae is investigated. MATERIALS AND METHODS
The S. cerevisiae strains used in this research were obtained from the National Institute of Radiological Science as well as from the Department of Pharmacy, Okayama University. The composition of the YPD medium, used for the the culture of S. cerevisiae, was as follows: Ten grams of bacto-yeast extract, 20 g of bacto-peptone, and 20 g of
149 0147-6513/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
150
KUNGOLOS, AOYAMA, AND MURAMOTO
D-glucose were dissolved in 1 liter of distilled water. The medium and all the laboratory equipment that would come in touch with S. cerevisiae culture were autoclaved. The pH of the YPD medium was 5.7}5.9. Before starting the experiment, the cell density was measured with a cell counter (Celltac Nihon Koden Ltd.) and the initial cell density was adjusted at about 5;10 cells/ml. The culture of S. cerevisiae was done in cuvettes that were placed inside a Bioscreen-C analyzer system (developed by Labsystems) and the light absorbance of the culture medium was automatically measured. The relationship between the light absorbance and the cell density of S. cerevisiae is given by >"aX@,
(1)
where >"cell density, and X"light absorbance. The values of a and b remain relatively stable for two di!erent ranges of cell densities (the high-density range and the low-density range). These values have been experimentally evaluated and with the help of a computer program by which the growth of S. cerevisiae can be followed for the duration of the experiment at intervals of 15 min. The incubation temperature was 30$13C and the culture was done without continuous shaking except for automatic shaking for about 10 s just before the light absorbance was measured every 15 min. The maximum duration of the experiment was set to be 72 h although in most cases results were able to be taken within 12 h. In this research the maximum speci"c growth rate (k ) was mainly used as a toxicity index, but the area
under the light absorbance curve upto the turning point of the control (A) was also used (Kungolos and Aoyama, 1992). The Probit model was used for the analysis of the experimental results and calculation of EC values (Finney, 1952). The following strains of S. cerevisiae were used in this research: * AOY1, a wild-type strain possessing the three DNA repair systems, excision repair, mutation repair, and recombination repair, provided by Dr. B. Ono from the Department of Pharmacy, Okayama University. * ONO726, mutant resistant to Hg, also provided by Dr. B. Ono. The following three strains were produced using backcrossing "ve times in the laboratory and they are considered to be statistically isogenic with AOY1. The original strains were provided by the National Institute of Radiological Science of Japan: * CA11, mutation repair-de"cient mutant (rad18). * CA13, excision repair-de"cient mutant (rad2). * CA15, recombination repair-de"cient mutant (rad52).
FIG. 1. Inhibitions caused by two organic Hg compounds on the maximum speci"c growth rate (k ) of strain AOY1.
RESULTS
Toxicities of Various Chemical Forms of Mercury The toxicities of two organic forms of Hg, CH HgOH and CH HgCl, to S. cerevisiae were investigated. Nine dif ferent concentrations were tested for each chemical as follows: 0.02, 0.04, 0.08 , 0.2. 0.4, 0.8, 2, 4, and 8 lmol/liter. The results are provided in Fig. 1 for the strain AOY1. At low concentrations (up to 0.08 lmol/liter) the inhibition caused by both chemicals was equally low and within 5% of control. For the middle range of concentrations the inhibition caused by CH HgCl was higher S. cerevisiae was not at all able to grow at CH HgCl concentrations higher than 2 lmol/liter or at CH HgOH concentrations higher than 4 lmol/liter. The EC values for both compounds based on k are
presented on Table 1. The EC50 for CH HgOH toward the strain AOY1 was 1.77 lmol/liter, about "ve times higher than that of CH HgOH (0.35 lmol/liter). Four di!erent forms of inorganic mercury were tested: HgCl , HgSO , Hg(NO ) , and Hg (NO ) . Three of these compounds contain Hg(II), the bivalent form of mercury
TABLE 1 EC50 Values for Various Forms of Hg for Four Di4erent Strains? Strain
HgSO
HgCl
Hg(NO )
CH HgOH
CH HgCl
AOY1 CA15 CA13 CA11
71.6 47.3 *@ 33.5
192 56.9 * *
68.5 91.2 * 30
1.77 1.6 * *
0.35 2.7 5.4 2.8
?Expressed as lmol/liter. @Not tested.
MERCURY TOXICITY TO Saccharomyces cerevisiae
151
FIG. 2. Inhibitions caused by three inorganic Hg compounds on the k of strain AOY1.
FIG. 4. Inhibitions caused by CH HgOH on the growth of strains AOY1 and ONO726 (using A and k as toxicity indices).
and only one Hg (NO ) , contains the monovalent form. Nine di!erent concentrations of the inorganic mercury compounds were tested: 2, 4, 8, 20, 40, 200, 400, and 800 lmol/liter. The inhibition rates caused by the three bivalent compounds in the k of AOY1 are provided in
Fig. 2. It indicates that the inhibitions are about the same for the sulfate and the nitrate form, but lower for chloride. The inhibition becomes 100% at 200 lmol/liter for nitrate and sulfate and at 800 lmol/liter for chloride. In terms of EC values toward AOY1, the sequence of toxicity is as follows: Hg(NO ) 'HgSO 'HgCL . Figure 3 presents the inhibition on A (area up to the turning point) for Hg (NO ) and Hg(NO ) . It indicates that the monovalent form is more toxic to the strain AOY1 than the bivalent form.
Figure 4 provides the inhibitions caused by CH HgOH on strains AOY1 and ONO726, using k and A as toxicity
indices. It indicates that the inhibitions at the same concentrations are much lower for ONO726, for both indices A and k . Strain ONO726 is a mutant resistant to Hg.
Strain ONO726 exhibited resistance also to the other Hg compounds used in this study. Figure 4 indicates that the inhibition also depends on the index, with A being more sensitive than k .
Strains CA11, CA13, and CA15 were produced in the laboratory with "ve times backcrossing and are considered to be statistically isogenic with AOY1. Each of them is de"cient in one separate DNA repair system. Figure 5 demonstrates the inhibitions in the k of the
three strains caused by HgCl. The sensitivity of the three
FIG. 3. Di!erence in inhibition caused by monovalent and bivalent Hg nitrates on the growth of strain AOY1, using A (area) as a toxicity index.
FIG. 5. Inhibitions caused by HgCl on the growth of three di!erent DNA repair-de"cient strains.
152
KUNGOLOS, AOYAMA, AND MURAMOTO
strains to HgCl was in the order CA15'CA11'CA13. On the other hand, all three strains were more sensitive to Hg than wild-type strain AOY1.
Bioaccumulation of Hg in the Cells of S. cerevisiae In order to explain the di!erence in toxicity between organic and inorganic mercury, the amount of mercury absorbed by the cells of S. cerevisiae was evaluated. In this part of the research HgCl and CH HgCl were used. For this experiment S. cerevisiae was cultured in 500-ml #asks containing 300 ml of medium. At 8 and 24 h after the beginning of the experiment, the S. cerevisiae cells were collected from the medium by centrifuging and washing the residue three times. Then the cells were digested by wet digestion. The technique for measuring small amounts of mercury with atomic absorption has been described by Hatch and Welland (1968). Figure 6 presents the relationship between survival rate and mercury concentration in water. It indicates that for the same survival rate the concentration of inorganic mercury is more than one hundred times higher than that of inorganic. Figures 7 and 8 present the concentrations in cells for di!erent concentrations in water for the two compounds. Both "gures demonstrate that the concentration after 8 h is higher than that after 24 h. It is assumed that mercury is quickly absorbed by S. cerevisiae and then, probably after the death of the cell, is excreted. By comparing the two "gures one can see that the concentration in the cells of S. cerevisiae exposed to a relatively low concentration of organic mercury is about the same or even higher than the respective concentration in the cells exposed to a concentration of inorganic mercury one hundred times higher. These
FIG. 6. Relationship between concentration of inorganic and organic mercury in water and survival rate after 8 and 24 h.
FIG. 7. Relationship between concentration in cell and concentration in water for HgCl , after 8 and 24 h.
results indicate the big di!erence in bioaccumulation between organic and inorganic mercury. Figure 9 presents the relationship between the concentration in cells and survival rate. It demonstrates that in all cases concentration in cells is higher for low survival rates. It can be concluded that increased bioaccumulation leads to the death of yeast. Figures 10 and 11 provide the bioconcentration factors for the two compounds. The "rst thing that can be noted is that the bioconcentration factors are much higher for organic mercury. For CH HgCl the bioconcentration factor goes as high as 250,000, while for HgCl it does not exceed 1500. Also, one can see that the bioconcentration factor changes with water concentration for the organic form, but it stays almost constant for HgCl .
FIG. 8. Relationship between concentration in cell and concentration in water for CH HgCl, after 8 and 24 h.
MERCURY TOXICITY TO Saccharomyces cerevisiae
FIG. 9. Relationship between concentration of inorganic and organic mercury in cells and survival rate after 8 and 24 h (In.Hg, inorganic Hg; Or.Hg, organic Hg).
Ewect of Complexing and Chelating Agents on the Toxicity of Mercury to S. cerevisiae EDTA is the most common complexing agent for metals and it has been used successfully in detoxi"cation research (McLeese and Ray, 1986). Figure 12 presents the e!ect of EDTA on the toxicity of HgCl to S. cerevisiae. The growth curves the in presence of 4;10\ mol/liter and 8; 10\ lmol/liter HgCl2 are compared with the respective ones when 20% excess EDTA is added to the culture. The cell densities were compared at di!erent times, using Student's t test. There was no signi"cant di!erence between the cell density in the presence and absence of EDTA. Figure 13 provides the cell density curves for S. cerevisiae exposed to CH HgCl, in the presence or absence of EDTA. The CH HgCl concentrations in the culture solution were
FIG. 10. Bioconcentration factors for di!erent HgCl concentrations in water after 8 and 24 h.
153
FIG. 11. Bioconcentration factors for di!erent CH HgCl concentra tions in water after 8 and 24 h.
4;10\ and 8;10\ mol/liter. At the concentration 8;10\ mol/liter, EDTA seemed to decrease slightly the toxicity of organic mercury. This decrease in toxicity still did not prove to be signi"cant, according to Student's t test, probably due to a high standard deviation of the results. The results of these experiments revealed that EDTA was not a good complexing agent for either HgCl or CH HgCl. It has been reported (Jones and Harbison, 1974) that EDTA was not a good chelator for mercurials. Methylpencillamine has been reported to be a strong chelator for mercurials. Figure 14 indicates the growth of S. cerevisiae exposed to CH HgCl, in the presence or absence of N-t BOC-s-methyl -D-penicillamine. In this experiment as well, the chelating substance was not able to decrease the e!ect of organic mercury on S. cerevisiae.
FIG. 12. Growth of S. cerevisiae exposed to HgCl , in presence or absence of EDTA (4E-5, for 4;10\ mol/liter).
154
KUNGOLOS, AOYAMA, AND MURAMOTO
FIG. 13. Growth of S. cerevisiae exposed to CH HgCl, in presence or absence of EDTA (4E-7, 4;10\ mol/liter).
DISCUSSION
It has long been known that the toxicity of organic mercury is higher than that of inorganic. On the other hand, the di!erence in toxicity, in terms of EC values, between inorganic chloride and organic chloride was higher than two orders of magnitude (see Table 1), which is higher than values found in other literature (Tsai and Olson, 1990). It was also revealed that the toxicity of mercury not only depends on it being organic or inorganic but also strongly depends on its speciation, with CH HgCl being more toxic than CH HgOH and Hg (NO ) more toxic than Hg(NO ) .
FIG. 14. Growth of S. cerevisiae exposed to CH HgCl, in presence or absence of N-¹-BOC-s-methyl-D-penicillamine (4E-7,4;10\ lmol/ liter).
The toxicity of mercury to S. cerevisiae depends not only on the chemical form of mercury but also on the speci"c strain of S. cerevisiae used in the experiment. It was demonstrated that strain ONO726 was able to grow at mercury concentrations that caused high inhibition in the growth of the wild-type strain. The resistance to mercury by strain ONO726 has been explained by Ono et al. (1991) On the other hand, strains de"cient in one DNA repair system were more a!ected by mercury than the wild-type strain. The fact that strains that are de"cient in one DNA repair system exhibit higher sensitivity to Hg than the wild-type strain, which possesses all three DNA repair system possessed by S. cerevisiae, implies that Hg damages the DNA of yeast. The DNA damaging capacity of Hg is discussed further in another article (Kungolos and Aoyama, 1999a). The di!erence in sensitivity between the three strains CA11, CA13, and CA15 also implies a di!erent DNA repair capacity for each of the three DNA repair systems. It is clear that organic mercury is much more readily absorbed by the cells of S. cerevisiae than the inorganic form and this big di!erence in bioaccumulation explains the big (two orders of magnitude) di!erence in toxicity between the two compounds. Organic mercury is very readily absorbed by S. cerevisiae cells. A very high bioconcentration factor, close to 250,000, was estimated for methylmercury chloride. This bioconcentration factor is one of the highest reported in the literature (for any organism). The results of the bioaccumulation experiments indicate that the big di!erence in toxicity between organic and inorganic Hg is due to the di!erence in bioaccumulation of the two species. It was demonstrated that the bioaccumulation of Hg in the cells of S. cerevisiae exposed to a concentration of organic Hg in water of about 0.1 lmol/liter is about the same or even greater than the bioaccumulation of Hg in the organism cell exposed to a concentration of inorganic Hg in water 100 times higher, that is, about 10 lmol/liter. On the other hand, if the toxicities of the organic and inorganic form per amount of Hg absorbed are compared, these toxicities are in the same order of magnitude. These results imply that the big di!erence in toxicity between organic and inorganic Hg is due to di!erent bioconcentration factors, or, in other words, to the ease with which organic Hg is absorbed by the cell, and not to a di!erent mechanism of toxicity between the two species. EDTA was not able to signi"cantly reduce the e!ect of Hg on S. cerevisiae. These results are in contrast with results for other metals, such as cadmium or nickel. EDTA was able to reduce the toxicity of cadmium and nickel to S. cerevisiae by at least one order of magnitude (Kungolos and Aoyama, 1999b). Even methyl-penicillamine, which has been reported to be a good chelator for mercurials, was not successful in reducing Hg toxicity to S. cerevisiae. These results indicate that mercury is readily absorbed by the cells of S. cerevisiae and not easily removed.
MERCURY TOXICITY TO Saccharomyces cerevisiae
CONCLUSIONS
The toxicity of mercury to S. cerevisiae strongly depends on Hg speciation. The toxicity of organic mercury to S. cerevisiae, as expressed by EC values, is more than one hundred times higher than that of inorganic mercury. This big di!erence in toxicity is due to di!erences in bioaccumulation of the two species. The bioconcentration factor for methylmercury chloride was more than two orders of magnitude higher than that for inorganic mercury chloride. On the other hand, the toxicities of organic and inorganic mercury to S. cerevisiae per amount of Hg absorbed were in the same order of magnitude. Di!erent strains of S. cerevisiae had di!ering sensitivities to Hg. Strains not possessing all three DNA repair systems possessed by S. cerevisiae were more sensitive to Hg than strains possessing all three systems. These results demonstrate that Hg damages the DNA of S. cerevisiae. It was also proved that the DNA repair capacity of the three DNA repair systems was di!erent. The e!ect of complexing and chelating agents on the toxicity of mercury to S. cerevisiae was almost negligible. EDTA and methyl-penicillamine were only a little successful in reducing the toxicity of Hg to S. cerevisiae.
REFERENCES Aoyama, I., Okamura, H., and Yagi, M. (1987). The interacttion e!ects of toxic chemical combinations on Chlorella ellipsoidea. Toxic. Assessment: Int. Q. 2, 341}355. Bitton, G., Koopman, B., and Wang, H. D. (1984). Baker's yeast assay procedure for testing heavy metal toxicity. Bull. Enviorn. Contam. Toxicol. 32, 80}84. Camusso, M., Vigano, L., and Balestrini, R. (1995). Bioconcentrations of trace metals in rainbow trout: A "eld study. Ecotoxicol. Envion. Saf. 31, 133}141. Eckardt, F., and Siede, W. (1985). Mutagen testing with yeast. Basic Life Sci. 34, 305}322. Eisemann, J. D., Beyer, W. N., Benetts, R. E., and Morton, A. (1997). Mercury residues in South Florida apple snails (Pomacea paludosa). Bull. Environ. Contam. Toxicol. 58, 739}743. Farrell, R. D., Germida, J. J., and Huang, P. M. (1990). Biotoxicity of mercury as in#uenced by mercury (II) speciation. Appl. Environ. Microbiol. 56, 3006}3016.
155
Farrell R. D., Germida, J. J., and Huang, P. M. (1993). E!ects of chemical speciation in growth media on the toxicity of mercury (II). Appl. Environ. Microbiol. 59, 1507}1514. Finney, D. J. (1952). Probit Analysis, 2nd ed. University Press, Cambridge. Hatch, W. R., and Welland, L. O. (1968). Determination of sub-microgram quantities of mercury by atomic absorption spectrophotometry. Anal. Chem. 40, 2085}2087. Jones M., and Harbison, D. (1974). Phthalytetrathioacetic acid (PTTA), a new mercury chelating agent. Res. Commun. Chem. Pathol. Pharmacol. 7, 389}398. Kungolos, A., and Aoyama I. (1992). Using Saccharomyces cerevisiae for toxicity assessment including interacting e!ects and DNA damage. Wat. Sci. Tech. 25, 309}316. Kungolos, A., and Aoyama, I. (1993). Interaction e!ect, food e!ect and bioaccumulation of cadmium and chromium for the system Daphnia magna}Chlorella ellipsoidea. Environ. Toxic. Water Qual. 8, 351}369. Kungolos, A., and Aoyama, I. (1999a). Testing DNA damage caused by chemicals with Saccharomyces cerevisiae. J. of Environ. Sci. Health A, in press. Kungolos, A., and Aoyama, I. (1999b). Interactive e!ects of heavy metals on Sacharomyces cerevisiae. Environmental ¹oxicology, in press. Kwasniewska, K., and Kaiser, K. (1984). Toxicities of selected chloroanilines to four strains of yeast. In QSAR in Environmental Toxicology (K. L. E. Kaiser, Ed.), pp. 223}233. Reidel, Dordrecht. Lee, H. L., Lustigman, B., Schwinge, V., Chiu, I. Y., and Hsu, S. (1992). E!ect of mercury and cadmium on the growth of Anacystis nidulans. Bull. Environ. Contam. Toxicol. 49, 272}278. Mattheis, T., Pietrock, M., and Kruger, R. (1994). Residues of total mercury in "sh from two small lakes in the biosphere reserve of Schorfheide-Chorin in Bradenburg, East Germany. Environ. Toxicol. Water Qual. 9, 299}302. McCrary, J. E., and Heagler, M. G. (1997). The use of a simultaneous species acute toxicity test to compare the relative sensitivities of aquatic organisms to mercury. J. Environ. Sci. Health A 32, 73}81. McLeese, D. W., and Ray, S. (1986). Toxicity of CdCl , CdEDTA, CuCl , and CuEDTA to marine Invertebrates. Bull. Eniron. Contam. Toxicol. 36, 749}755. Ono, B., Ishii, N., Fujino, S., and Aoyama, I. (1991). Role of hydrosul"de ions (HS\) in methylmercury resistance in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 57, 3183}3186. Roderer, G. (1983). Di!erential toxic e!ects of mercuric chloride and methylmercuric chloride on the freshwater alga Poterioochromonas malhamensis. Aquat. Toxicol. 3, 23}34. Schultz, I. R., and Newman, M. C. (1997). Methyl mercury toxicokinetics in channel cat"sh (Ictalurus punctatus) and largemouth bass (Micropterus salmoides) after intravascular administration. Environ. Toxicol. Chem. 16, 990}996. Tsai, Y., and Olson, B. (1990). E!ects of Hg>, CH\Hg>, and temperature on the expression of mercury resistance genes in environmental bacteria. Appl. Environ. Microbiol. 56, 3266}3272.
Toxicity of Organic and Inorganic Mercury to Saccharomyces cerevisiae Athanassios Kungolos, Isao Aoyama, and S. Muramoto Laboratory of Ecological Chemistry, Research Institute for Bioresources, Okayama University, 2-20-1, Chu-o, Kurashiki 710, Japan Received April 7, 1998
In this study the e4ect if six di4erent forms of mercury on the growth of the yeast Saccharomyces cerevisiae is presented. Five kinds of strains of S. cerevisiae were used. They were a wild type, a mercury-resistant type, and three mutants: mutation repairde5cient mutant, excision repair-de5cient mutant, and recombination repair-de5cient mutant. In terms of EC50 toward the wild-type strain, the toxicity order for the inorganic forms was Hg(NO3)2 > HgSO4 > HgCl2 . Monovalent nitrate mercury Hg(NO3)2 was more toxic than bivalent Hg(NO3)2. The toxicity of organic mercury CH3HgCl on cell growth was two orders of magnitude higher than that of inorganic HgCl2. Between the two organic forms, CH3HgCl was more toxic than CH3HgOH. The survival rate in the presence of a certain concentration of CH3HgCl was about one-hundredth of the survival in presence of the same concentration of HgCl2. On the other hand, the concentration of CH3HgCl in the cell was about 170 times that of HgCl2. The addition of chelating agents, EDTA and methylpenicillamine, to the medium did not reduce the toxicity of mercury. Among the three mutants tested, the one de5cient in recombination repair systems was the most sensitive to mercury. 1999 Academic Press
Key Words: mercury; toxicity; bioconcentration factor; Saccharomyces cerevisiae.
INTRODUCTION
The high toxicity of organic mercury has been known for a long time. There have been some reports on the di!erence of toxicity between organic and inorganic mercury (Roderer, 1983; Tsai and Olson, 1990). Not much research has been done on the di!erence of toxicity between various forms of organic mercury or inorganic mercury (Farrel et al., 1990, 1993). Although the e!ects of mercury on a lot of organisms have been evaluated (Lee et al., 1992; Mattheis Present address: Chemical Process Engineering Research Institute, P.O. Box 361, Thermi, 57001 Thessaloniki, Greece. To whom all correspondence should be addressed.
et al., 1994; McCrary and Heagler, 1997; Schulz and Newman, 1997), there are not many reports on the e!ect of mercury on Saccharomyces cerevisiae (Kungolos and Aoyama, 1992). Only a few researchers have used S. cerevisiae in toxicity assessments. (Bitton et al., 1984; Kwashiewska and Kaiser, 1984; Eckardt and Siede, 1985). The authors of this article have advocated the use of S. cerevisiae as a eukaryotic organism, for the development of a quick automated method for toxicity screening of chemicals (Kungolos and Aoyama, 1992). Increased mercury contamination in aquatic systems has become a major concern. The rise in mercury concentrations in water and sediments leads to increased bioaccumulation by organisms (Camusso et al., 1995; Eisemann et al., 1997). The relationship between bioaccumulation and toxicity has been established with a number of organisms, such as Daphnia magna (Kungolos and Aoyama, 1993) and Chlorella ellipsoidea (Aoyama et al., 1987). However, because the exact mechanisms of toxicity are not fully understood, there is a need for more data, especially in the case of species of the same type with a big di!erence in toxicity. In this research, the e!ect of two organic forms of mercury, CH HgOH and CH HgCl, and four inorganic ones, HgCl , HgSO , Hg(NO ) , and Hg (NO ) , is evaluated. Then, the bioaccumulation of mercury in the cells of S. cerevisiae, exposed to CH HgCl or to HgCl is compared. Finally, the e!ect of two complexing agents in the toxicity of mercury to S. cerevisiae is investigated. MATERIALS AND METHODS
The S. cerevisiae strains used in this research were obtained from the National Institute of Radiological Science as well as from the Department of Pharmacy, Okayama University. The composition of the YPD medium, used for the the culture of S. cerevisiae, was as follows: Ten grams of bacto-yeast extract, 20 g of bacto-peptone, and 20 g of
149 0147-6513/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
150
KUNGOLOS, AOYAMA, AND MURAMOTO
D-glucose were dissolved in 1 liter of distilled water. The medium and all the laboratory equipment that would come in touch with S. cerevisiae culture were autoclaved. The pH of the YPD medium was 5.7}5.9. Before starting the experiment, the cell density was measured with a cell counter (Celltac Nihon Koden Ltd.) and the initial cell density was adjusted at about 5;10 cells/ml. The culture of S. cerevisiae was done in cuvettes that were placed inside a Bioscreen-C analyzer system (developed by Labsystems) and the light absorbance of the culture medium was automatically measured. The relationship between the light absorbance and the cell density of S. cerevisiae is given by >"aX@,
(1)
where >"cell density, and X"light absorbance. The values of a and b remain relatively stable for two di!erent ranges of cell densities (the high-density range and the low-density range). These values have been experimentally evaluated and with the help of a computer program by which the growth of S. cerevisiae can be followed for the duration of the experiment at intervals of 15 min. The incubation temperature was 30$13C and the culture was done without continuous shaking except for automatic shaking for about 10 s just before the light absorbance was measured every 15 min. The maximum duration of the experiment was set to be 72 h although in most cases results were able to be taken within 12 h. In this research the maximum speci"c growth rate (k ) was mainly used as a toxicity index, but the area
under the light absorbance curve upto the turning point of the control (A) was also used (Kungolos and Aoyama, 1992). The Probit model was used for the analysis of the experimental results and calculation of EC values (Finney, 1952). The following strains of S. cerevisiae were used in this research: * AOY1, a wild-type strain possessing the three DNA repair systems, excision repair, mutation repair, and recombination repair, provided by Dr. B. Ono from the Department of Pharmacy, Okayama University. * ONO726, mutant resistant to Hg, also provided by Dr. B. Ono. The following three strains were produced using backcrossing "ve times in the laboratory and they are considered to be statistically isogenic with AOY1. The original strains were provided by the National Institute of Radiological Science of Japan: * CA11, mutation repair-de"cient mutant (rad18). * CA13, excision repair-de"cient mutant (rad2). * CA15, recombination repair-de"cient mutant (rad52).
FIG. 1. Inhibitions caused by two organic Hg compounds on the maximum speci"c growth rate (k ) of strain AOY1.
RESULTS
Toxicities of Various Chemical Forms of Mercury The toxicities of two organic forms of Hg, CH HgOH and CH HgCl, to S. cerevisiae were investigated. Nine dif ferent concentrations were tested for each chemical as follows: 0.02, 0.04, 0.08 , 0.2. 0.4, 0.8, 2, 4, and 8 lmol/liter. The results are provided in Fig. 1 for the strain AOY1. At low concentrations (up to 0.08 lmol/liter) the inhibition caused by both chemicals was equally low and within 5% of control. For the middle range of concentrations the inhibition caused by CH HgCl was higher S. cerevisiae was not at all able to grow at CH HgCl concentrations higher than 2 lmol/liter or at CH HgOH concentrations higher than 4 lmol/liter. The EC values for both compounds based on k are
presented on Table 1. The EC50 for CH HgOH toward the strain AOY1 was 1.77 lmol/liter, about "ve times higher than that of CH HgOH (0.35 lmol/liter). Four di!erent forms of inorganic mercury were tested: HgCl , HgSO , Hg(NO ) , and Hg (NO ) . Three of these compounds contain Hg(II), the bivalent form of mercury
TABLE 1 EC50 Values for Various Forms of Hg for Four Di4erent Strains? Strain
HgSO
HgCl
Hg(NO )
CH HgOH
CH HgCl
AOY1 CA15 CA13 CA11
71.6 47.3 *@ 33.5
192 56.9 * *
68.5 91.2 * 30
1.77 1.6 * *
0.35 2.7 5.4 2.8
?Expressed as lmol/liter. @Not tested.
MERCURY TOXICITY TO Saccharomyces cerevisiae
151
FIG. 2. Inhibitions caused by three inorganic Hg compounds on the k of strain AOY1.
FIG. 4. Inhibitions caused by CH HgOH on the growth of strains AOY1 and ONO726 (using A and k as toxicity indices).
and only one Hg (NO ) , contains the monovalent form. Nine di!erent concentrations of the inorganic mercury compounds were tested: 2, 4, 8, 20, 40, 200, 400, and 800 lmol/liter. The inhibition rates caused by the three bivalent compounds in the k of AOY1 are provided in
Fig. 2. It indicates that the inhibitions are about the same for the sulfate and the nitrate form, but lower for chloride. The inhibition becomes 100% at 200 lmol/liter for nitrate and sulfate and at 800 lmol/liter for chloride. In terms of EC values toward AOY1, the sequence of toxicity is as follows: Hg(NO ) 'HgSO 'HgCL . Figure 3 presents the inhibition on A (area up to the turning point) for Hg (NO ) and Hg(NO ) . It indicates that the monovalent form is more toxic to the strain AOY1 than the bivalent form.
Figure 4 provides the inhibitions caused by CH HgOH on strains AOY1 and ONO726, using k and A as toxicity
indices. It indicates that the inhibitions at the same concentrations are much lower for ONO726, for both indices A and k . Strain ONO726 is a mutant resistant to Hg.
Strain ONO726 exhibited resistance also to the other Hg compounds used in this study. Figure 4 indicates that the inhibition also depends on the index, with A being more sensitive than k .
Strains CA11, CA13, and CA15 were produced in the laboratory with "ve times backcrossing and are considered to be statistically isogenic with AOY1. Each of them is de"cient in one separate DNA repair system. Figure 5 demonstrates the inhibitions in the k of the
three strains caused by HgCl. The sensitivity of the three
FIG. 3. Di!erence in inhibition caused by monovalent and bivalent Hg nitrates on the growth of strain AOY1, using A (area) as a toxicity index.
FIG. 5. Inhibitions caused by HgCl on the growth of three di!erent DNA repair-de"cient strains.
152
KUNGOLOS, AOYAMA, AND MURAMOTO
strains to HgCl was in the order CA15'CA11'CA13. On the other hand, all three strains were more sensitive to Hg than wild-type strain AOY1.
Bioaccumulation of Hg in the Cells of S. cerevisiae In order to explain the di!erence in toxicity between organic and inorganic mercury, the amount of mercury absorbed by the cells of S. cerevisiae was evaluated. In this part of the research HgCl and CH HgCl were used. For this experiment S. cerevisiae was cultured in 500-ml #asks containing 300 ml of medium. At 8 and 24 h after the beginning of the experiment, the S. cerevisiae cells were collected from the medium by centrifuging and washing the residue three times. Then the cells were digested by wet digestion. The technique for measuring small amounts of mercury with atomic absorption has been described by Hatch and Welland (1968). Figure 6 presents the relationship between survival rate and mercury concentration in water. It indicates that for the same survival rate the concentration of inorganic mercury is more than one hundred times higher than that of inorganic. Figures 7 and 8 present the concentrations in cells for di!erent concentrations in water for the two compounds. Both "gures demonstrate that the concentration after 8 h is higher than that after 24 h. It is assumed that mercury is quickly absorbed by S. cerevisiae and then, probably after the death of the cell, is excreted. By comparing the two "gures one can see that the concentration in the cells of S. cerevisiae exposed to a relatively low concentration of organic mercury is about the same or even higher than the respective concentration in the cells exposed to a concentration of inorganic mercury one hundred times higher. These
FIG. 6. Relationship between concentration of inorganic and organic mercury in water and survival rate after 8 and 24 h.
FIG. 7. Relationship between concentration in cell and concentration in water for HgCl , after 8 and 24 h.
results indicate the big di!erence in bioaccumulation between organic and inorganic mercury. Figure 9 presents the relationship between the concentration in cells and survival rate. It demonstrates that in all cases concentration in cells is higher for low survival rates. It can be concluded that increased bioaccumulation leads to the death of yeast. Figures 10 and 11 provide the bioconcentration factors for the two compounds. The "rst thing that can be noted is that the bioconcentration factors are much higher for organic mercury. For CH HgCl the bioconcentration factor goes as high as 250,000, while for HgCl it does not exceed 1500. Also, one can see that the bioconcentration factor changes with water concentration for the organic form, but it stays almost constant for HgCl .
FIG. 8. Relationship between concentration in cell and concentration in water for CH HgCl, after 8 and 24 h.
MERCURY TOXICITY TO Saccharomyces cerevisiae
FIG. 9. Relationship between concentration of inorganic and organic mercury in cells and survival rate after 8 and 24 h (In.Hg, inorganic Hg; Or.Hg, organic Hg).
Ewect of Complexing and Chelating Agents on the Toxicity of Mercury to S. cerevisiae EDTA is the most common complexing agent for metals and it has been used successfully in detoxi"cation research (McLeese and Ray, 1986). Figure 12 presents the e!ect of EDTA on the toxicity of HgCl to S. cerevisiae. The growth curves the in presence of 4;10\ mol/liter and 8; 10\ lmol/liter HgCl2 are compared with the respective ones when 20% excess EDTA is added to the culture. The cell densities were compared at di!erent times, using Student's t test. There was no signi"cant di!erence between the cell density in the presence and absence of EDTA. Figure 13 provides the cell density curves for S. cerevisiae exposed to CH HgCl, in the presence or absence of EDTA. The CH HgCl concentrations in the culture solution were
FIG. 10. Bioconcentration factors for di!erent HgCl concentrations in water after 8 and 24 h.
153
FIG. 11. Bioconcentration factors for di!erent CH HgCl concentra tions in water after 8 and 24 h.
4;10\ and 8;10\ mol/liter. At the concentration 8;10\ mol/liter, EDTA seemed to decrease slightly the toxicity of organic mercury. This decrease in toxicity still did not prove to be signi"cant, according to Student's t test, probably due to a high standard deviation of the results. The results of these experiments revealed that EDTA was not a good complexing agent for either HgCl or CH HgCl. It has been reported (Jones and Harbison, 1974) that EDTA was not a good chelator for mercurials. Methylpencillamine has been reported to be a strong chelator for mercurials. Figure 14 indicates the growth of S. cerevisiae exposed to CH HgCl, in the presence or absence of N-t BOC-s-methyl -D-penicillamine. In this experiment as well, the chelating substance was not able to decrease the e!ect of organic mercury on S. cerevisiae.
FIG. 12. Growth of S. cerevisiae exposed to HgCl , in presence or absence of EDTA (4E-5, for 4;10\ mol/liter).
154
KUNGOLOS, AOYAMA, AND MURAMOTO
FIG. 13. Growth of S. cerevisiae exposed to CH HgCl, in presence or absence of EDTA (4E-7, 4;10\ mol/liter).
DISCUSSION
It has long been known that the toxicity of organic mercury is higher than that of inorganic. On the other hand, the di!erence in toxicity, in terms of EC values, between inorganic chloride and organic chloride was higher than two orders of magnitude (see Table 1), which is higher than values found in other literature (Tsai and Olson, 1990). It was also revealed that the toxicity of mercury not only depends on it being organic or inorganic but also strongly depends on its speciation, with CH HgCl being more toxic than CH HgOH and Hg (NO ) more toxic than Hg(NO ) .
FIG. 14. Growth of S. cerevisiae exposed to CH HgCl, in presence or absence of N-¹-BOC-s-methyl-D-penicillamine (4E-7,4;10\ lmol/ liter).
The toxicity of mercury to S. cerevisiae depends not only on the chemical form of mercury but also on the speci"c strain of S. cerevisiae used in the experiment. It was demonstrated that strain ONO726 was able to grow at mercury concentrations that caused high inhibition in the growth of the wild-type strain. The resistance to mercury by strain ONO726 has been explained by Ono et al. (1991) On the other hand, strains de"cient in one DNA repair system were more a!ected by mercury than the wild-type strain. The fact that strains that are de"cient in one DNA repair system exhibit higher sensitivity to Hg than the wild-type strain, which possesses all three DNA repair system possessed by S. cerevisiae, implies that Hg damages the DNA of yeast. The DNA damaging capacity of Hg is discussed further in another article (Kungolos and Aoyama, 1999a). The di!erence in sensitivity between the three strains CA11, CA13, and CA15 also implies a di!erent DNA repair capacity for each of the three DNA repair systems. It is clear that organic mercury is much more readily absorbed by the cells of S. cerevisiae than the inorganic form and this big di!erence in bioaccumulation explains the big (two orders of magnitude) di!erence in toxicity between the two compounds. Organic mercury is very readily absorbed by S. cerevisiae cells. A very high bioconcentration factor, close to 250,000, was estimated for methylmercury chloride. This bioconcentration factor is one of the highest reported in the literature (for any organism). The results of the bioaccumulation experiments indicate that the big di!erence in toxicity between organic and inorganic Hg is due to the di!erence in bioaccumulation of the two species. It was demonstrated that the bioaccumulation of Hg in the cells of S. cerevisiae exposed to a concentration of organic Hg in water of about 0.1 lmol/liter is about the same or even greater than the bioaccumulation of Hg in the organism cell exposed to a concentration of inorganic Hg in water 100 times higher, that is, about 10 lmol/liter. On the other hand, if the toxicities of the organic and inorganic form per amount of Hg absorbed are compared, these toxicities are in the same order of magnitude. These results imply that the big di!erence in toxicity between organic and inorganic Hg is due to di!erent bioconcentration factors, or, in other words, to the ease with which organic Hg is absorbed by the cell, and not to a di!erent mechanism of toxicity between the two species. EDTA was not able to signi"cantly reduce the e!ect of Hg on S. cerevisiae. These results are in contrast with results for other metals, such as cadmium or nickel. EDTA was able to reduce the toxicity of cadmium and nickel to S. cerevisiae by at least one order of magnitude (Kungolos and Aoyama, 1999b). Even methyl-penicillamine, which has been reported to be a good chelator for mercurials, was not successful in reducing Hg toxicity to S. cerevisiae. These results indicate that mercury is readily absorbed by the cells of S. cerevisiae and not easily removed.
MERCURY TOXICITY TO Saccharomyces cerevisiae
CONCLUSIONS
The toxicity of mercury to S. cerevisiae strongly depends on Hg speciation. The toxicity of organic mercury to S. cerevisiae, as expressed by EC values, is more than one hundred times higher than that of inorganic mercury. This big di!erence in toxicity is due to di!erences in bioaccumulation of the two species. The bioconcentration factor for methylmercury chloride was more than two orders of magnitude higher than that for inorganic mercury chloride. On the other hand, the toxicities of organic and inorganic mercury to S. cerevisiae per amount of Hg absorbed were in the same order of magnitude. Di!erent strains of S. cerevisiae had di!ering sensitivities to Hg. Strains not possessing all three DNA repair systems possessed by S. cerevisiae were more sensitive to Hg than strains possessing all three systems. These results demonstrate that Hg damages the DNA of S. cerevisiae. It was also proved that the DNA repair capacity of the three DNA repair systems was di!erent. The e!ect of complexing and chelating agents on the toxicity of mercury to S. cerevisiae was almost negligible. EDTA and methyl-penicillamine were only a little successful in reducing the toxicity of Hg to S. cerevisiae.
REFERENCES Aoyama, I., Okamura, H., and Yagi, M. (1987). The interacttion e!ects of toxic chemical combinations on Chlorella ellipsoidea. Toxic. Assessment: Int. Q. 2, 341}355. Bitton, G., Koopman, B., and Wang, H. D. (1984). Baker's yeast assay procedure for testing heavy metal toxicity. Bull. Enviorn. Contam. Toxicol. 32, 80}84. Camusso, M., Vigano, L., and Balestrini, R. (1995). Bioconcentrations of trace metals in rainbow trout: A "eld study. Ecotoxicol. Envion. Saf. 31, 133}141. Eckardt, F., and Siede, W. (1985). Mutagen testing with yeast. Basic Life Sci. 34, 305}322. Eisemann, J. D., Beyer, W. N., Benetts, R. E., and Morton, A. (1997). Mercury residues in South Florida apple snails (Pomacea paludosa). Bull. Environ. Contam. Toxicol. 58, 739}743. Farrell, R. D., Germida, J. J., and Huang, P. M. (1990). Biotoxicity of mercury as in#uenced by mercury (II) speciation. Appl. Environ. Microbiol. 56, 3006}3016.
155
Farrell R. D., Germida, J. J., and Huang, P. M. (1993). E!ects of chemical speciation in growth media on the toxicity of mercury (II). Appl. Environ. Microbiol. 59, 1507}1514. Finney, D. J. (1952). Probit Analysis, 2nd ed. University Press, Cambridge. Hatch, W. R., and Welland, L. O. (1968). Determination of sub-microgram quantities of mercury by atomic absorption spectrophotometry. Anal. Chem. 40, 2085}2087. Jones M., and Harbison, D. (1974). Phthalytetrathioacetic acid (PTTA), a new mercury chelating agent. Res. Commun. Chem. Pathol. Pharmacol. 7, 389}398. Kungolos, A., and Aoyama I. (1992). Using Saccharomyces cerevisiae for toxicity assessment including interacting e!ects and DNA damage. Wat. Sci. Tech. 25, 309}316. Kungolos, A., and Aoyama, I. (1993). Interaction e!ect, food e!ect and bioaccumulation of cadmium and chromium for the system Daphnia magna}Chlorella ellipsoidea. Environ. Toxic. Water Qual. 8, 351}369. Kungolos, A., and Aoyama, I. (1999a). Testing DNA damage caused by chemicals with Saccharomyces cerevisiae. J. of Environ. Sci. Health A, in press. Kungolos, A., and Aoyama, I. (1999b). Interactive e!ects of heavy metals on Sacharomyces cerevisiae. Environmental ¹oxicology, in press. Kwasniewska, K., and Kaiser, K. (1984). Toxicities of selected chloroanilines to four strains of yeast. In QSAR in Environmental Toxicology (K. L. E. Kaiser, Ed.), pp. 223}233. Reidel, Dordrecht. Lee, H. L., Lustigman, B., Schwinge, V., Chiu, I. Y., and Hsu, S. (1992). E!ect of mercury and cadmium on the growth of Anacystis nidulans. Bull. Environ. Contam. Toxicol. 49, 272}278. Mattheis, T., Pietrock, M., and Kruger, R. (1994). Residues of total mercury in "sh from two small lakes in the biosphere reserve of Schorfheide-Chorin in Bradenburg, East Germany. Environ. Toxicol. Water Qual. 9, 299}302. McCrary, J. E., and Heagler, M. G. (1997). The use of a simultaneous species acute toxicity test to compare the relative sensitivities of aquatic organisms to mercury. J. Environ. Sci. Health A 32, 73}81. McLeese, D. W., and Ray, S. (1986). Toxicity of CdCl , CdEDTA, CuCl , and CuEDTA to marine Invertebrates. Bull. Eniron. Contam. Toxicol. 36, 749}755. Ono, B., Ishii, N., Fujino, S., and Aoyama, I. (1991). Role of hydrosul"de ions (HS\) in methylmercury resistance in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 57, 3183}3186. Roderer, G. (1983). Di!erential toxic e!ects of mercuric chloride and methylmercuric chloride on the freshwater alga Poterioochromonas malhamensis. Aquat. Toxicol. 3, 23}34. Schultz, I. R., and Newman, M. C. (1997). Methyl mercury toxicokinetics in channel cat"sh (Ictalurus punctatus) and largemouth bass (Micropterus salmoides) after intravascular administration. Environ. Toxicol. Chem. 16, 990}996. Tsai, Y., and Olson, B. (1990). E!ects of Hg>, CH\Hg>, and temperature on the expression of mercury resistance genes in environmental bacteria. Appl. Environ. Microbiol. 56, 3266}3272.