- Email: [email protected]
The toxic potential of an industrial effluent determined with the Saccharomyces cerevisiae-based assay
ARTICLE IN PRESS
Water Research 39 (2005) 3211–3218 www.elsevier.com/locate/watres
The toxic potential of an industrial effluent determined with the Saccharomyces cerevisiae-based assay Marcel Schmitta, Georg Gellertb, Hella Lichtenberg-Frate´a, a
Institut fu¨r zellula¨re und molekulare Botanik, AG Molekulare Bioenergetik, Universita¨t Bonn, Kirschallee 1, 53115 Bonn, Germany b Staatliches Umweltamt Siegen, Unteres Schloss, 57 072 Siegen, Germany Received 17 September 2004; received in revised form 28 February 2005; accepted 25 May 2005 Available online 6 July 2005
Abstract Increasing levels of environmental pollution and the continuous monitoring of water quality both request specific and sensitive methods for the detection of detrimental water contents. On a regulatory basis genotoxicity is assessed by the standard umu-test (ISO 13829) that responds to DNA damage induced by chemicals. The focus of this study was the examination of the toxic potential of samples taken from the wastewater treatment plant of a refinery factory to explore the applicability of the Saccharomyces cerevisiae (bakers yeast) test for the detection of bio-available genotoxic activity in complex matrices. The toxic potential of samples without pre-treatment and following centrifugation was determined with the eukaryotic Saccharomyces cerevisiae bioassay based on the transcriptional activation of the green fluorescent protein (gfp) fused to the DNA damage inducible RAD54 promoter and general growth inhibition. Primary effluent samples were taken as qualified sterile spot samples from the final effluent of the purification plant. The Saccharomyces cerevisiae assay yielded geno- and cytotoxic responses in all complex untreated and centrifuged samples with high reproducibility. The obtained results suggest that the yeast assay is suited as a screening tool to monitor genotoxic potential of wastewater. r 2005 Elsevier Ltd. All rights reserved. Keywords: Saccharomyces cerevisiae; Wastewater; Cytotoxicity; Genotoxicity; Reporter
1. Introduction The increased population growth, industrialization and the development of new chemical substances cause an increase in the pollutants in the environment and the demand for reliable and sensitive short-term bioassays for the detection of specific toxic effects. Bioassays are essential for the evaluation of toxic effects in complex matrices like industrial discharges that might interact, Corresponding author. Tel.: +49 228 73 55 18;
fax: +49 228 73 55 04. E-mail address: [email protected] (H. Lichtenberg-Frate´).
e.g. with the DNA of aquatic organisms. Currently, standardized prokaryotic genotoxicity procedures include the Ames test (Maron and Ames, 1983; Gee et al., 1994), the umu-test (Oda et al., 1985; Reifferscheid et al., 1991) and the SOS Chromotest (Quillardet and Hofnung, 1985) which are based on genetically engineered Salmonella typhimurium strains. In Germany, the governmental survey of wastewater (monitoring of plant effluents) comprises the umu-test (ISO 13829) as the relevant biological parameter. This assay utilizes prokaryotic Salmonella typhimurium TA 1535 pSK1002 cells; the detected genetic damage represents DNA damage from point mutations. Recently, Saccharomyces cerevisiae tester strains have been developed for the
0043-1354/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2005.05.034
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3212
detection of genotoxic (Walmsley et al., 1997; Afanassiev et al., 2000) and combined geno- and cytotoxic potential (Schmitt et al., 2002; Lichtenberg-Frate´ et al., 2003). The objectives of this survey were to determine the toxic potential of wastewater samples using the Saccharomyces cerevisiae reporter assay with sample handling according to the standard quality guidelines. The Saccharomyces cerevisiae reporter assay is based on transcriptional activation of the yeast optimized version of the green fluorescent protein (gfp) of Aequorea victoria (Cormack et al., 1996) by the DNA damage inducible RAD54 promoter in a sensitized yeast strain devoid of endogenous efflux transporters. Like the rfa mutation in the Salmonella typhimurium-based umutest—the rfa gene-cluster encodes the Lipopolysaccharides (LPS) core oligosaccharide biosynthesis—deletions of ABC-type transporters in the Saccharomyces cerevisiae strain aim to enhance test sensitivity. In yeast, the activities of these ABC-type extrusion systems are involved in drug extrusion and the continuous removal of the relevant compounds from the cell can compromise the sensitivity of any assay (Kolaczkowska and Goffeau, 1999). This study focused on cyto- and genotoxicity because the latter parameter is under continuous observation within the environmental survey of treated wastewater from refinery plants. Such discharges are known to occasionally contain chemicals that interact with DNA. Wastewater effluents from 3 different discharges of a refinery factory were assayed as both complex untreated and centrifuged samples as a blind study. The Saccharomyces cerevisiae cyto- and genotoxicity test was performed without metabolic activation by microsomal (S9) preparations.
2. Material and methods 2.1. Sample handling Three samples were taken by the Federal Environmental Agency of North-Rhine Westfalia (Germany) as
qualified sterile spot samples from the final effluent of a wastewater treatment plant of a refinery factory and split in half for a total of six samples. Thus sample 1 was identical to sample 5, sample 2 identical to 4 and sample 3 to 6. Sample handling was performed according to the standard quality guidelines with initial storage at 18 1C. At the test period the origin of the samples was unknown. Following thawing to room temperature sample aliquots were tested as follows: (i) agitation to resuspend potential sediments, or (ii) centrifugation at 4000g to avoid potential interferences by suspended solids. Tests were performed by serial dilutions of the samples (after adjustment of the pH to 6.4) with Millipore water to determine the threshold at which samples lose their toxic effects. The lowest possible dilution factor was 1.25 and the highest dilution factor was 16 (Table 1). River Rhine and tap water were tested in identical dilution series. 2.2. Assay conditions, yeast growth and fluorescence monitoring For quantitative assessment of growth phenotypes and fluorescence development with the Saccharomyces cerevisiae assay stationary cells were diluted to a start OD600 of 0.5 (Pharmacia Ultraspec 2000 Spectrophotometer) in 5-fold concentrated yeast nitrogen base (YNB) medium, (YNB, 8.5 g L1; AA Drop-Out-Mix, 2.5 g L1, citrate buffer, 52.5 g L1 corresponding to 250 mM, glucose, 2.5%; pH 6.4). The start OD (A600) was finally reduced to 0.1 (corresponding to 3 106 cells/ml) by addition of the test material or its serial dilution. The cultures were incubated at 30 1C and continuously agitated at a frequency of 950 rpm for 8 h. Read-outs were performed in the corresponding temperature adjusted reader. Before incubation the plates were covered with standard lids and sealed with an adhesive film to prevent evaporation of volatile organic sample contents. Three replicates of each sample, agitated or centrifuged, were analyzed on different days. Each experiment
Table 1 Composition of the test cultures Dilution factor G
1.25 2 3 4 6 8 12 16
Waste water fraction in %
80 50 33.33 25 16.7 12.5 8.33 6.25
Volume (ml) Sample
Growth medium and inoculum
Clean water
Total
160 100 66.7 50 33.3 25 16.7 12.5
40 40 40 40 40 40 40 40
0 60 93.3 110 126.7 135 143.3 147.5
200 200 200 200 200 200 200 200
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3213
consisted of 1 control and 8 serial dilutions in 6-fold repetitions on 1 plate. Growth was estimated as turbidity measured at A600 as a single endpoint measurement using a microplate reader (Tecan, Spectrofluor Plus). Tests were considered valid when the turbidity of the control cultures increased at least 5 times during the incubation period. For fluorescence development read-outs, the excitation wavelength was adjusted to 485 nm and emission was observed at 535 nm (25 nm bandwidth) as 8 h endpoint (Tecan, Spectrofluor Plus).
with the non-parametric ranking Spearman Rank Correlation Test. Pairwise comparisons of tests carried out with untreated (only agitated) and centrifuged samples (to indicate significant differences linked to suspended solids) were performed with the Student ttest. The multiple Dunn test for associations was applied to prove differences in the genotoxicity of all tested samples.
2.3. Data analysis and presentation
The Saccharomyces cerevisiae test results for the agitated as well as for the centrifuged samples 1–6, River Rhine and tap water are diagrammed in Figs. 1 and 2, respectively. The sample dilution factor was plotted versus the growth inhibition, representing cytotoxicity and the calculated IR, representing genotoxicity. The dilution factor (G) represents a serial dilution step along the German convention. The respective actual wastewater fraction is depicted in Table 1. The threshold of 1 for the induction ratio represents the baseline fluorescence of uninduced cells and also corresponds to the 20% growth inhibition threshold (dotted line). All results were obtained without S9 mediated metabolic activation. Potential genotoxic properties of a sample were detected by induction of the RAD54 controlled gfp. The highest IR level for all tested samples was obtained for the lowest dilution factor of 1.25. The IR values for the agitated, slightly turbid samples 1, 2 and 3 (Fig. 1A–C) were 1.384, 1.368 and 1.253 for this dilution factor. Samples 4, 5 and 6 (Fig. 1D–F) showed similar genotoxic values (IR 1.368, 1.293 and 1.271) when tested 4 weeks later. The analysis of supernatants following particle centrifugation revealed similar results with IRs of 1.378, 1.415 and 1.285 for samples 1, 2 and 3 (Fig. 2A–C) and 1.489, 1.320 and 1.268 for samples 4, 5 and 6 (Fig. 2D–F). With increasing sample dilution the IR declined towards the fluorescence base line of not induced cells. A dilution factor was evaluated as genotoxic if the fluorescence induction signal differed significantly from that of the negative control (t-test, 0.95 significance level). All data are summarized in Table 2. Maximal growth inhibition and therewith highest cytotoxicity was also recorded at the dilution factor of 1.25. The mean inhibition rate at this dilution factor oscillated between 44% and 56% for the samples 1, 2 and 3 (Fig. 1A–C). The follow up of the experiments, with samples 4, 5 and 6 (identical to samples 2, 1 and 3) which were performed 4 weeks later (Fig. 1D–F), yielded only 2/3 of the cytotoxic effects (31–36% growth inhibition). This is also reflected in the dilution factor, which was about 3 for samples 1, 2 and 3 and that decreased 1 dilution factor for samples 4, 5 and 6.
Cytotoxic effects were calculated by quantification of the extent of growth inhibition in the test cultures in correlation to the corresponding control cultures. Cytotoxicity was determined as that dilution factor (G) at which growth inhibition did not exceed 20%. The threshold of 20% growth inhibition is in accordance with the German convention classifying the first dilution factor within a rising serial dilution below this threshold as non-toxic. Genotoxic effects were assessed using the transcriptional activation of the gfp fused to the RAD54 DNA damage inducible promoter and responses were quantified as the induction ratio according to the German standard method DIN 38415-3 procedure for the umutest (ISO/FDIS 13829:1999). The induction ratio (IR) was calculated as follows: IR ¼ 1=G i C i =C n , where Ci is the mean fluorescence (corrected for blank) of the test concentration (i) at the end of the incubation period. Cn is the mean fluorescence (corrected for blank) for the negative controls at the end of the incubation period. Gi represents the mean inhibition of growth. The factor 1/Gi serves to correct the green fluorescence induction for the growth inhibition in the test cultures in comparison to control cultures. A positive genotoxic signal is identified if the emission of the test cultures differs significantly from the fluorescence of the control cultures (t-test; significance level: 0.05). Toxicity data are presented as the first dilution factor (G) at which no cyto- and genotoxic effects were observed. The dilution scale started with the lowest possible dilution factor of 1.25. Subsequent steps were 2, 3, 6, 8, 12 and 16, the latter as the most diluted sample. A dilution factor of 1.25 indicates no poisonous effects of the sample. The higher the dilution factor was, the stronger the yeast responded to the respective sample. 2.4. Statistical evaluation The ranking of the wastewater samples by the obtained cyto- and genotoxicity results was compared
3. Results
ARTICLE IN PRESS
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0
0.8
0
12
4 6 8 dilution factor
12
16
0.8
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
2
3
4 6 8 dilution factor
12
growth inhibition in %
70
induction ratio AU
1.5
(C)
16
0.8 1.25
2
3
(D)
4 6 8 dilution factor
12
16
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
(E)
4 6 8 dilution factor
12
growth inhibition in %
70
16
0.8 1.25
2
3
(F)
4 6 8 dilution factor
12
16
70
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
growth inhibition in %
growth inhibition in %
3
70
1.25
growth inhibition in %
2
(B)
0
0.8
0 1.25
(G)
1.25
16
induction ratio AU
growth inhibition in %
6 4 8 dilution factor
2
3
4
6
8
dilution factor
12
0.8
0
16
1.25
(H)
induction ratio AU
3
induction ratio AU
2
2
3
4
6
8
dilution factor
12
16
induction ratio AU
1.25
(A)
growth inhibition in %
70
induction ratio AU
1.5
induction ratio AU
growth inhibition in %
70
induction ratio AU
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3214
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
Almost identical results were obtained for the centrifuged samples (Fig. 2A–F). The cytotoxic data of all samples, obtained with the Saccharomyces cerevisiae test, are presented in Table 3. In control experiments, water from the river Rhine and tap water were tested as untreated and centrifuged samples (Figs. 1G, H and 2G, H). None of these exposures exerted cyto- or genotoxic effects in the Saccharomyces cerevisiae test. From the statistical analysis, the Spearman correlation coefficients showed no relationship between genoand cytotoxicity (two sided; significance level: 0.95). The Student t-test revealed that the centrifugation of the samples did not alter their cyto- and genotoxic properties significantly (two sided; significance level: 0.95). Visibly turbid suspensions did not exert toxicity. All samples were genotoxic in a similar order of magnitude without significant differences as indicated by the Dunn test for associations (significance level: 0.95).
4. Discussion Earlier investigations with the Saccharomyces cerevisiae cyto- and genotoxicity assay using monosubstances provided reliable and sensitive results compared to other test systems (Schmitt et al., 2004; Lichtenberg-Frate´ et al., 2003). In this study, the assay was challenged with complex samples taken from a wastewater treatment plant of a refinery factory in order to explore its applicability for biomonitoring. By partial automation and adaptation to the microtiter scale it was possible to test 8 sample dilution steps on one 96-well plate, including 6-fold repetitions of 1 dilution step and thus 2 samples as complex agitated and centrifuged aliquots per week. For all 6 samples a close correlation between the dilution factor and the degree of toxicity was observed. The growth inhibition responded to a considerably weaker extent (dilution factors between 2 and 4) to the industrial wastewater samples than the fluorescence induction (dilution factors between 12 and 16) indicating
3215
a primarily genotoxic charge among the samples. Since the induction rates of the agitated and the centrifuged samples fairly matched (Table 2) it is concluded that the genotoxic charge was within the water phase and not particle bound. The reproducibility of both the growth inhibition and the fluorescent signal was high. Observed variances in dilution factors result from decreasing differences in the total waste water fraction between 2 consecutive dilution factors, i.e. at higher dilutions the absolute sample fractions differ less than at low dilutions (Table 1). Between G factors 12 and 16 the sample volume difference was 4.2 ml (2.08%), whereas between factors 3 and 4 the difference was 16.7 ml (8.33%), supporting the notion of considerable sensitivity of the test system. As expected, experiments involving river Rhine and tap water revealed no toxicity response. The possibility of false positive results can thus comfortably be excluded. With a robust test system the statistical spread of obtained data in percent should not exceed 20%. The mean distribution of single data compared to the arithmetic mean was 10.7% for the parameter growth inhibition and 4.0% for the induction ratio. In the Saccharomyces cerevisiae assay the genotoxic threshold is not a fixed value but derived from the variability of the actual 6 test wells per dilution step and statistically evaluated as a significant difference to the respective negative control cultures by the t-test (significance level: 0.05). Thus it is possible that in case of homogeneous fluorescence values induction ratios even below 1.1 represent a genotoxic threshold. This is in contrast to the umu-test which requires an IR higher than 1.5 as threshold for the interpretation of a genotoxic signal. All results were obtained without S9 activation as it is applied in the umu-test. Since yeast contains 3 endogenous P450 enzymes their activities may to some extent account for activation of promutagens. The cytotoxicity data are summarized in Table 3. The most obvious effect was the time-dependent decrease of cytotoxicity observed for the samples 4, 5 and 6. These
Fig. 1. Diagrams of the complex untreated wastewater samples test results obtained with the S. cerevisiae test. Samples 1 (A), 2 (B) and 3 (C) are identical to samples 5 (E), 4 (D) and 6 (F). Initially, the original samples were split into half and the latter samples tested approximately 4 weeks later. The sample dilution factor was plotted against the growth inhibition in % representing cytotoxicity and displayed as columns and the calculated induction ratio (IR), representing genotoxicity displayed as solid line. Data points are shown as means+SD for growth inhibition and 7SD for the induction ratio (n ¼ 3 experiments, each with 6 analyzed cavities). The dotted line indicates both the induction ratio 1 representing the baseline fluorescence of uninduced cells as well as the 20% growth inhibition threshold for determination of the corresponding dilution factor G. The genotoxicity read-outs in arbitrary fluorescence brightness units were normalized to the growth of the cells (after subtraction of the corresponding blank control data) thus taking into account the endogenous background fluorescence of a given cell population. The results of the control experiments with river Rhine and tap water are shown as (G) and (H), respectively. The outlier in (H) at dilution factor 3 based on one single read-out influencing the mean data presentation.
ARTICLE IN PRESS 1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
3
12
16
4 6 8 dilution factor
12
16
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
2
3
4 6 8 dilution factor
12
16
growth inhibition in %
70
induction ratio AU
1.5
(C)
0.8 1.25
2
3
(D)
4 6 8 dilution factor
12
16
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
(E)
4 6 8 dilution factor
12
growth inhibition in %
70
16
0.8 1.25
2
3
(F)
4 6 8 dilution factor
12
16
70
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
4 6 8 dilution factor
12
growth inhibition in %
growth inhibition in %
3
70
1.25
growth inhibition in %
2
(B)
0
(G)
0.8 1.25
induction ratio AU
growth inhibition in %
(A)
4 8 6 dilution factor
16
0.8 1.25
(H)
induction ratio AU
2
induction ratio AU
1.25
induction ratio AU
0
growth inhibition in %
70
induction ratio AU
1.5
induction ratio AU
growth inhibition in %
70
induction ratio AU
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3216
2
3
4 6 8 dilution factor
12
16
Fig. 2. Diagrams of the centrifuged wastewater samples test results obtained with the S. cerevisiae test. See Fig. 1 for details.
Table 2 Genotoxic characterization of the refinery wastewater treatment samples by the S. cerevisiae test Sample
Dec. Dec. Dec. Jan. Jan. Jan. Jan. Jan.
After centrifugation
Dilution factor G
a
Induction ratio at dilution factor 1.25
Dilution factor Ga
Induction ratio at dilution factor 1.25
Day 1
Day 2
Day 3
Day 1
Day 2
Day 3
Ø
Day 1
Day 2
Day 3
Day 1
Day 2
Day 3
Ø
12 8 12 12 12 12 1.25 1.25
16 12 12 6 8 12 1.25 1.25
12 12 16 12 16 12 1.25 1.25
1.439 1.291 1.229 1.428 1.229 1.281 0.962 0.998
1.425 1.344 1.221 1.470 1.373 1.228 0.974 0.984
1.289 1.468 1.310 1.207 1.278 1.303 0.965 1.019
1.384 1.368 1.253 1.368 1.293 1.271 0.967 1.000
16 8 16 8 12 6 1.25 1.25
16 12 12 4 8 12 1.25 1.25
8 12 12 12 8 12 1.25 1.25
1.375 1.385 1.331 1.533 1.265 1.279 0.995 0.977
1.466 1.369 1.237 1.661 1.397 1.248 0.969 0.992
1.293 1.490 1.287 1.273 1.299 1.278 0.997 0.994
1.378 1.415 1.285 1.489 1.320 1.268 0.987 0.964
G ¼ (for genotoxicity): smallest dilution factor of the test batch at which a significant induction ratio was not obtained.
Table 3 Cytotoxic characterization of the tested samples from a refinery wastewater treatment plant with the S. cerevisiae test Sample
Date of test After agitation
After centrifugation
Dilution factor Ga
1 2 3 4 5 6 River Rhine Tap water a
10–12 15–17 21–23 06–08 12–14 15–19 28–30 20–22
Dec. Dec. Dec. Jan. Jan. Jan. Jan. Jan.
Growth inhibition in % at dilution factor 1.25
Dilution factor Ga
Growth inhibition in % at dilution factor 1.25
Day 1 Day 2 Day 3 Day 1
Day 2
Day 3
Ø
Day 1 Day 2 Day 3 Day 1
Day 2
Day 3
Ø
4 4 3 2 3 2 1.25 1.25
54.9 52.5 40.0 37.7 29.5 31.8 0 0.1
37.4 59.2 40.8 29.7 32.6 33.3 0 0
50.6 55.9 43.6 35.6 32.7 31.4 0 0
4 4 3 2 3 2 1.25 1.25
52.6 49.5 37.0 38.6 28.0 31.9 0 0
37.8 59.7 43.2 27.5 33.2 33.0 0 1.4
50.1 54.7 43.2 34.5 33.3 31.8 0 0
4 4 3 2 2 2 1.25 1.25
3 4 3 2 2 2 1.25 1.25
59.6 55.9 50.0 39.5 36.1 29.1 0 0
4 3 3 2 2 2 1.25 1.25
2 4 3 2 2 2 1.25 1.25
60.0 55.0 49.5 37.3 38.6 30.5 0.4 0
G ¼ (for cytotoxicity): smallest dilution factor of the test batch at which for the first time the growth inhibition did not exceed 20 % in relation to the negative control.
ARTICLE IN PRESS
a
10–12 15–17 21–23 06–08 12–14 15–19 28–30 20–22
After agitation
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
1 2 3 4 5 6 River Rhine Tap Water
Date of test
3217
ARTICLE IN PRESS 3218
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
samples were identical to 2, 1 and 3 but each was tested approximately 4 weeks later. Such a decrease of toxicity, even under deep freezing storage conditions of the samples may be due to either the freezing process itself or to the evaporation of organic compounds as it was reported in other studies. For frozen (2673 1C) samples from industrial plants known for their organic level content, Naudin et al. (1995) noted precipitation phenomena and a considerable toxicity decrease in some cases. The freezing process can reduce the sample toxicity to a certain extent, but the reduction seem to be limited, as results obtained by Naudin et al. (1995) with the long-term toxicity tests using Ceriodaphnia (Crustacea) and Brachydanio (Fish) indicated. Also, Schuytema et al. (1989) reported a toxicity decrease when sediments were frozen for 14 days, probably due to a release of soluble organic carbon from the sediments during the freezing process. The observed decreased cytotoxicity for samples 4, 5 and 6 analyzed later supports the notion of reliable sensitivity of the yeast test. The presented results suggest that the yeast test is quite sensitive to screen for the presence of soluble DNA-damaging agents and particularly suited to monitor for cytotoxicity in wastewater samples.
5. Conclusions In the yeast assay all samples revealed genotoxic properties without S9 mediated metabolic activation. The DNA damage induced transcriptional activation of the gfp in the yeast assay is a suitable marker for genotoxicity. The yeast assay, with gfp as directly measurable genotoxicity reporter and growth inhibition as cytotoxicity reporter can be performed in 96-well plates within 8 h and provides a high specificity by the separate analysis of 2 independent parameters. The obtained results indicate the reproducibility of the yeast assay with small variations. These properties suggest that the Saccharomyces cerevisiae geno- and cytotoxicity assay is suitable for medium or high throughput screening system for toxic effects and as biomonitor for complex matrices. Even the aging of frozen samples, containing organics, can be detected.
References Afanassiev, V., Sefton, M., Anantachaiyong, T., Barker, G., Walmsley, R., Wo¨lfl, S., 2000. Application of yeast cells transformed with GFP expression constructs containing the RAD54 or RNR2 promoter as a test for the genotoxic potential of chemical substances. Mutat. Res. 464, 297–308. Cormack, B.P., Valdivia, R.H., Falkow, S., 1996. FACSoptimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. Gee, P., Maron, D.M., Ames, B.N., 1994. Detection and classification of mutagens: a set of base-specific Salmonella tester strains. Proc. Natl. Acad. Sci. 91, 11606–11610. Kolaczkowska, A., Goffeau, A., 1999. Regulation of pleiotropic drug resistance in yeast. Drug Resist. Updat. 2, 403–414. Lichtenberg-Frate´, H., Schmitt, M., Gellert, G., Ludwig, J., 2003. A yeast based method for the detection of cyto- and genotoxicity. Toxicol. in Vitro 17, 709–716. Maron, D.M., Ames, B.N., 1983. Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Naudin, S., Garric, J., Vindimian, E., Bray, M., Migeon, B., Vollat, B., Lenon, G., 1995. Influence of the sample preservation mode to assess the chronic toxicity of an industrial effluent. Ecotoxicol. Environ. Saf. 30, 54–62. Oda, Y., Nakamura, S., Oki, I., Kato, T., Shinagawa, H., 1985. Evaluation of the new system (umu-test) for the detection of environmental mutagens and carcinogens. Mutat. Res. 147, 219–229. Quillardet, P., Hofnung, M., 1985. The SOS Chromotest, a colorimetric bacterial assay for genotoxins: procedures. Mutat. Res. 147, 65–78. Reifferscheid, G., Heil, J., Oda, Y., Zahn, R.K., 1991. A microplate version of the SOS/umu-test for rapid detection of genotoxins and genotoxic potentials of environmental samples. Mutat. Res. 253, 215–222. Schmitt, M., Gellert, G., Kirberg, B., Ludwig, J., LichtenbergFrate, H., 2002. CyGene: Eine neue hefebasierte Methode zur Bestimmung des zytotoxischen Potenzials von Umweltgiften im Wasserbereich. Vom Wasser 99, 1111–1118. Schmitt, M., Gellert, G., Ludwig, J., Lichtenberg-Frate´, H., 2004. Phenotypic yeast growth analysis for chronic toxicity testing. Ecotox. Environ. Saf. 59, 142–150. Schuytema, G.S., Nebeker, A.V., Griffis, W.L., Miller, C.E., 1989. Effects of freezing on toxicity of sediments contaminated with DDT and endrin. Environ. Toxicol. Chem. 8, 150–155. Walmsley, R.M., Billinton, N., Heyer, W.-D., 1997. Green fluorescent protein as a reporter for the DNA damageinduced gene RAD54 in Saccharomyces cerevisiae. Yeast 13, 1535–1545.
Water Research 39 (2005) 3211–3218 www.elsevier.com/locate/watres
The toxic potential of an industrial effluent determined with the Saccharomyces cerevisiae-based assay Marcel Schmitta, Georg Gellertb, Hella Lichtenberg-Frate´a, a
Institut fu¨r zellula¨re und molekulare Botanik, AG Molekulare Bioenergetik, Universita¨t Bonn, Kirschallee 1, 53115 Bonn, Germany b Staatliches Umweltamt Siegen, Unteres Schloss, 57 072 Siegen, Germany Received 17 September 2004; received in revised form 28 February 2005; accepted 25 May 2005 Available online 6 July 2005
Abstract Increasing levels of environmental pollution and the continuous monitoring of water quality both request specific and sensitive methods for the detection of detrimental water contents. On a regulatory basis genotoxicity is assessed by the standard umu-test (ISO 13829) that responds to DNA damage induced by chemicals. The focus of this study was the examination of the toxic potential of samples taken from the wastewater treatment plant of a refinery factory to explore the applicability of the Saccharomyces cerevisiae (bakers yeast) test for the detection of bio-available genotoxic activity in complex matrices. The toxic potential of samples without pre-treatment and following centrifugation was determined with the eukaryotic Saccharomyces cerevisiae bioassay based on the transcriptional activation of the green fluorescent protein (gfp) fused to the DNA damage inducible RAD54 promoter and general growth inhibition. Primary effluent samples were taken as qualified sterile spot samples from the final effluent of the purification plant. The Saccharomyces cerevisiae assay yielded geno- and cytotoxic responses in all complex untreated and centrifuged samples with high reproducibility. The obtained results suggest that the yeast assay is suited as a screening tool to monitor genotoxic potential of wastewater. r 2005 Elsevier Ltd. All rights reserved. Keywords: Saccharomyces cerevisiae; Wastewater; Cytotoxicity; Genotoxicity; Reporter
1. Introduction The increased population growth, industrialization and the development of new chemical substances cause an increase in the pollutants in the environment and the demand for reliable and sensitive short-term bioassays for the detection of specific toxic effects. Bioassays are essential for the evaluation of toxic effects in complex matrices like industrial discharges that might interact, Corresponding author. Tel.: +49 228 73 55 18;
fax: +49 228 73 55 04. E-mail address: [email protected] (H. Lichtenberg-Frate´).
e.g. with the DNA of aquatic organisms. Currently, standardized prokaryotic genotoxicity procedures include the Ames test (Maron and Ames, 1983; Gee et al., 1994), the umu-test (Oda et al., 1985; Reifferscheid et al., 1991) and the SOS Chromotest (Quillardet and Hofnung, 1985) which are based on genetically engineered Salmonella typhimurium strains. In Germany, the governmental survey of wastewater (monitoring of plant effluents) comprises the umu-test (ISO 13829) as the relevant biological parameter. This assay utilizes prokaryotic Salmonella typhimurium TA 1535 pSK1002 cells; the detected genetic damage represents DNA damage from point mutations. Recently, Saccharomyces cerevisiae tester strains have been developed for the
0043-1354/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2005.05.034
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3212
detection of genotoxic (Walmsley et al., 1997; Afanassiev et al., 2000) and combined geno- and cytotoxic potential (Schmitt et al., 2002; Lichtenberg-Frate´ et al., 2003). The objectives of this survey were to determine the toxic potential of wastewater samples using the Saccharomyces cerevisiae reporter assay with sample handling according to the standard quality guidelines. The Saccharomyces cerevisiae reporter assay is based on transcriptional activation of the yeast optimized version of the green fluorescent protein (gfp) of Aequorea victoria (Cormack et al., 1996) by the DNA damage inducible RAD54 promoter in a sensitized yeast strain devoid of endogenous efflux transporters. Like the rfa mutation in the Salmonella typhimurium-based umutest—the rfa gene-cluster encodes the Lipopolysaccharides (LPS) core oligosaccharide biosynthesis—deletions of ABC-type transporters in the Saccharomyces cerevisiae strain aim to enhance test sensitivity. In yeast, the activities of these ABC-type extrusion systems are involved in drug extrusion and the continuous removal of the relevant compounds from the cell can compromise the sensitivity of any assay (Kolaczkowska and Goffeau, 1999). This study focused on cyto- and genotoxicity because the latter parameter is under continuous observation within the environmental survey of treated wastewater from refinery plants. Such discharges are known to occasionally contain chemicals that interact with DNA. Wastewater effluents from 3 different discharges of a refinery factory were assayed as both complex untreated and centrifuged samples as a blind study. The Saccharomyces cerevisiae cyto- and genotoxicity test was performed without metabolic activation by microsomal (S9) preparations.
2. Material and methods 2.1. Sample handling Three samples were taken by the Federal Environmental Agency of North-Rhine Westfalia (Germany) as
qualified sterile spot samples from the final effluent of a wastewater treatment plant of a refinery factory and split in half for a total of six samples. Thus sample 1 was identical to sample 5, sample 2 identical to 4 and sample 3 to 6. Sample handling was performed according to the standard quality guidelines with initial storage at 18 1C. At the test period the origin of the samples was unknown. Following thawing to room temperature sample aliquots were tested as follows: (i) agitation to resuspend potential sediments, or (ii) centrifugation at 4000g to avoid potential interferences by suspended solids. Tests were performed by serial dilutions of the samples (after adjustment of the pH to 6.4) with Millipore water to determine the threshold at which samples lose their toxic effects. The lowest possible dilution factor was 1.25 and the highest dilution factor was 16 (Table 1). River Rhine and tap water were tested in identical dilution series. 2.2. Assay conditions, yeast growth and fluorescence monitoring For quantitative assessment of growth phenotypes and fluorescence development with the Saccharomyces cerevisiae assay stationary cells were diluted to a start OD600 of 0.5 (Pharmacia Ultraspec 2000 Spectrophotometer) in 5-fold concentrated yeast nitrogen base (YNB) medium, (YNB, 8.5 g L1; AA Drop-Out-Mix, 2.5 g L1, citrate buffer, 52.5 g L1 corresponding to 250 mM, glucose, 2.5%; pH 6.4). The start OD (A600) was finally reduced to 0.1 (corresponding to 3 106 cells/ml) by addition of the test material or its serial dilution. The cultures were incubated at 30 1C and continuously agitated at a frequency of 950 rpm for 8 h. Read-outs were performed in the corresponding temperature adjusted reader. Before incubation the plates were covered with standard lids and sealed with an adhesive film to prevent evaporation of volatile organic sample contents. Three replicates of each sample, agitated or centrifuged, were analyzed on different days. Each experiment
Table 1 Composition of the test cultures Dilution factor G
1.25 2 3 4 6 8 12 16
Waste water fraction in %
80 50 33.33 25 16.7 12.5 8.33 6.25
Volume (ml) Sample
Growth medium and inoculum
Clean water
Total
160 100 66.7 50 33.3 25 16.7 12.5
40 40 40 40 40 40 40 40
0 60 93.3 110 126.7 135 143.3 147.5
200 200 200 200 200 200 200 200
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3213
consisted of 1 control and 8 serial dilutions in 6-fold repetitions on 1 plate. Growth was estimated as turbidity measured at A600 as a single endpoint measurement using a microplate reader (Tecan, Spectrofluor Plus). Tests were considered valid when the turbidity of the control cultures increased at least 5 times during the incubation period. For fluorescence development read-outs, the excitation wavelength was adjusted to 485 nm and emission was observed at 535 nm (25 nm bandwidth) as 8 h endpoint (Tecan, Spectrofluor Plus).
with the non-parametric ranking Spearman Rank Correlation Test. Pairwise comparisons of tests carried out with untreated (only agitated) and centrifuged samples (to indicate significant differences linked to suspended solids) were performed with the Student ttest. The multiple Dunn test for associations was applied to prove differences in the genotoxicity of all tested samples.
2.3. Data analysis and presentation
The Saccharomyces cerevisiae test results for the agitated as well as for the centrifuged samples 1–6, River Rhine and tap water are diagrammed in Figs. 1 and 2, respectively. The sample dilution factor was plotted versus the growth inhibition, representing cytotoxicity and the calculated IR, representing genotoxicity. The dilution factor (G) represents a serial dilution step along the German convention. The respective actual wastewater fraction is depicted in Table 1. The threshold of 1 for the induction ratio represents the baseline fluorescence of uninduced cells and also corresponds to the 20% growth inhibition threshold (dotted line). All results were obtained without S9 mediated metabolic activation. Potential genotoxic properties of a sample were detected by induction of the RAD54 controlled gfp. The highest IR level for all tested samples was obtained for the lowest dilution factor of 1.25. The IR values for the agitated, slightly turbid samples 1, 2 and 3 (Fig. 1A–C) were 1.384, 1.368 and 1.253 for this dilution factor. Samples 4, 5 and 6 (Fig. 1D–F) showed similar genotoxic values (IR 1.368, 1.293 and 1.271) when tested 4 weeks later. The analysis of supernatants following particle centrifugation revealed similar results with IRs of 1.378, 1.415 and 1.285 for samples 1, 2 and 3 (Fig. 2A–C) and 1.489, 1.320 and 1.268 for samples 4, 5 and 6 (Fig. 2D–F). With increasing sample dilution the IR declined towards the fluorescence base line of not induced cells. A dilution factor was evaluated as genotoxic if the fluorescence induction signal differed significantly from that of the negative control (t-test, 0.95 significance level). All data are summarized in Table 2. Maximal growth inhibition and therewith highest cytotoxicity was also recorded at the dilution factor of 1.25. The mean inhibition rate at this dilution factor oscillated between 44% and 56% for the samples 1, 2 and 3 (Fig. 1A–C). The follow up of the experiments, with samples 4, 5 and 6 (identical to samples 2, 1 and 3) which were performed 4 weeks later (Fig. 1D–F), yielded only 2/3 of the cytotoxic effects (31–36% growth inhibition). This is also reflected in the dilution factor, which was about 3 for samples 1, 2 and 3 and that decreased 1 dilution factor for samples 4, 5 and 6.
Cytotoxic effects were calculated by quantification of the extent of growth inhibition in the test cultures in correlation to the corresponding control cultures. Cytotoxicity was determined as that dilution factor (G) at which growth inhibition did not exceed 20%. The threshold of 20% growth inhibition is in accordance with the German convention classifying the first dilution factor within a rising serial dilution below this threshold as non-toxic. Genotoxic effects were assessed using the transcriptional activation of the gfp fused to the RAD54 DNA damage inducible promoter and responses were quantified as the induction ratio according to the German standard method DIN 38415-3 procedure for the umutest (ISO/FDIS 13829:1999). The induction ratio (IR) was calculated as follows: IR ¼ 1=G i C i =C n , where Ci is the mean fluorescence (corrected for blank) of the test concentration (i) at the end of the incubation period. Cn is the mean fluorescence (corrected for blank) for the negative controls at the end of the incubation period. Gi represents the mean inhibition of growth. The factor 1/Gi serves to correct the green fluorescence induction for the growth inhibition in the test cultures in comparison to control cultures. A positive genotoxic signal is identified if the emission of the test cultures differs significantly from the fluorescence of the control cultures (t-test; significance level: 0.05). Toxicity data are presented as the first dilution factor (G) at which no cyto- and genotoxic effects were observed. The dilution scale started with the lowest possible dilution factor of 1.25. Subsequent steps were 2, 3, 6, 8, 12 and 16, the latter as the most diluted sample. A dilution factor of 1.25 indicates no poisonous effects of the sample. The higher the dilution factor was, the stronger the yeast responded to the respective sample. 2.4. Statistical evaluation The ranking of the wastewater samples by the obtained cyto- and genotoxicity results was compared
3. Results
ARTICLE IN PRESS
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0
0.8
0
12
4 6 8 dilution factor
12
16
0.8
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
2
3
4 6 8 dilution factor
12
growth inhibition in %
70
induction ratio AU
1.5
(C)
16
0.8 1.25
2
3
(D)
4 6 8 dilution factor
12
16
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
(E)
4 6 8 dilution factor
12
growth inhibition in %
70
16
0.8 1.25
2
3
(F)
4 6 8 dilution factor
12
16
70
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
growth inhibition in %
growth inhibition in %
3
70
1.25
growth inhibition in %
2
(B)
0
0.8
0 1.25
(G)
1.25
16
induction ratio AU
growth inhibition in %
6 4 8 dilution factor
2
3
4
6
8
dilution factor
12
0.8
0
16
1.25
(H)
induction ratio AU
3
induction ratio AU
2
2
3
4
6
8
dilution factor
12
16
induction ratio AU
1.25
(A)
growth inhibition in %
70
induction ratio AU
1.5
induction ratio AU
growth inhibition in %
70
induction ratio AU
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3214
ARTICLE IN PRESS M. Schmitt et al. / Water Research 39 (2005) 3211–3218
Almost identical results were obtained for the centrifuged samples (Fig. 2A–F). The cytotoxic data of all samples, obtained with the Saccharomyces cerevisiae test, are presented in Table 3. In control experiments, water from the river Rhine and tap water were tested as untreated and centrifuged samples (Figs. 1G, H and 2G, H). None of these exposures exerted cyto- or genotoxic effects in the Saccharomyces cerevisiae test. From the statistical analysis, the Spearman correlation coefficients showed no relationship between genoand cytotoxicity (two sided; significance level: 0.95). The Student t-test revealed that the centrifugation of the samples did not alter their cyto- and genotoxic properties significantly (two sided; significance level: 0.95). Visibly turbid suspensions did not exert toxicity. All samples were genotoxic in a similar order of magnitude without significant differences as indicated by the Dunn test for associations (significance level: 0.95).
4. Discussion Earlier investigations with the Saccharomyces cerevisiae cyto- and genotoxicity assay using monosubstances provided reliable and sensitive results compared to other test systems (Schmitt et al., 2004; Lichtenberg-Frate´ et al., 2003). In this study, the assay was challenged with complex samples taken from a wastewater treatment plant of a refinery factory in order to explore its applicability for biomonitoring. By partial automation and adaptation to the microtiter scale it was possible to test 8 sample dilution steps on one 96-well plate, including 6-fold repetitions of 1 dilution step and thus 2 samples as complex agitated and centrifuged aliquots per week. For all 6 samples a close correlation between the dilution factor and the degree of toxicity was observed. The growth inhibition responded to a considerably weaker extent (dilution factors between 2 and 4) to the industrial wastewater samples than the fluorescence induction (dilution factors between 12 and 16) indicating
3215
a primarily genotoxic charge among the samples. Since the induction rates of the agitated and the centrifuged samples fairly matched (Table 2) it is concluded that the genotoxic charge was within the water phase and not particle bound. The reproducibility of both the growth inhibition and the fluorescent signal was high. Observed variances in dilution factors result from decreasing differences in the total waste water fraction between 2 consecutive dilution factors, i.e. at higher dilutions the absolute sample fractions differ less than at low dilutions (Table 1). Between G factors 12 and 16 the sample volume difference was 4.2 ml (2.08%), whereas between factors 3 and 4 the difference was 16.7 ml (8.33%), supporting the notion of considerable sensitivity of the test system. As expected, experiments involving river Rhine and tap water revealed no toxicity response. The possibility of false positive results can thus comfortably be excluded. With a robust test system the statistical spread of obtained data in percent should not exceed 20%. The mean distribution of single data compared to the arithmetic mean was 10.7% for the parameter growth inhibition and 4.0% for the induction ratio. In the Saccharomyces cerevisiae assay the genotoxic threshold is not a fixed value but derived from the variability of the actual 6 test wells per dilution step and statistically evaluated as a significant difference to the respective negative control cultures by the t-test (significance level: 0.05). Thus it is possible that in case of homogeneous fluorescence values induction ratios even below 1.1 represent a genotoxic threshold. This is in contrast to the umu-test which requires an IR higher than 1.5 as threshold for the interpretation of a genotoxic signal. All results were obtained without S9 activation as it is applied in the umu-test. Since yeast contains 3 endogenous P450 enzymes their activities may to some extent account for activation of promutagens. The cytotoxicity data are summarized in Table 3. The most obvious effect was the time-dependent decrease of cytotoxicity observed for the samples 4, 5 and 6. These
Fig. 1. Diagrams of the complex untreated wastewater samples test results obtained with the S. cerevisiae test. Samples 1 (A), 2 (B) and 3 (C) are identical to samples 5 (E), 4 (D) and 6 (F). Initially, the original samples were split into half and the latter samples tested approximately 4 weeks later. The sample dilution factor was plotted against the growth inhibition in % representing cytotoxicity and displayed as columns and the calculated induction ratio (IR), representing genotoxicity displayed as solid line. Data points are shown as means+SD for growth inhibition and 7SD for the induction ratio (n ¼ 3 experiments, each with 6 analyzed cavities). The dotted line indicates both the induction ratio 1 representing the baseline fluorescence of uninduced cells as well as the 20% growth inhibition threshold for determination of the corresponding dilution factor G. The genotoxicity read-outs in arbitrary fluorescence brightness units were normalized to the growth of the cells (after subtraction of the corresponding blank control data) thus taking into account the endogenous background fluorescence of a given cell population. The results of the control experiments with river Rhine and tap water are shown as (G) and (H), respectively. The outlier in (H) at dilution factor 3 based on one single read-out influencing the mean data presentation.
ARTICLE IN PRESS 1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
3
12
16
4 6 8 dilution factor
12
16
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
2
3
4 6 8 dilution factor
12
16
growth inhibition in %
70
induction ratio AU
1.5
(C)
0.8 1.25
2
3
(D)
4 6 8 dilution factor
12
16
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
(E)
4 6 8 dilution factor
12
growth inhibition in %
70
16
0.8 1.25
2
3
(F)
4 6 8 dilution factor
12
16
70
1.5
70
1.5
60
1.4
60
1.4
50
1.3
50
1.3
40
1.2
40
1.2
30
1.1
30
1.1
20
1.0
20
1.0
10
0.9
10
0.9
0.8
0
0 1.25
2
3
4 6 8 dilution factor
12
growth inhibition in %
growth inhibition in %
3
70
1.25
growth inhibition in %
2
(B)
0
(G)
0.8 1.25
induction ratio AU
growth inhibition in %
(A)
4 8 6 dilution factor
16
0.8 1.25
(H)
induction ratio AU
2
induction ratio AU
1.25
induction ratio AU
0
growth inhibition in %
70
induction ratio AU
1.5
induction ratio AU
growth inhibition in %
70
induction ratio AU
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
3216
2
3
4 6 8 dilution factor
12
16
Fig. 2. Diagrams of the centrifuged wastewater samples test results obtained with the S. cerevisiae test. See Fig. 1 for details.
Table 2 Genotoxic characterization of the refinery wastewater treatment samples by the S. cerevisiae test Sample
Dec. Dec. Dec. Jan. Jan. Jan. Jan. Jan.
After centrifugation
Dilution factor G
a
Induction ratio at dilution factor 1.25
Dilution factor Ga
Induction ratio at dilution factor 1.25
Day 1
Day 2
Day 3
Day 1
Day 2
Day 3
Ø
Day 1
Day 2
Day 3
Day 1
Day 2
Day 3
Ø
12 8 12 12 12 12 1.25 1.25
16 12 12 6 8 12 1.25 1.25
12 12 16 12 16 12 1.25 1.25
1.439 1.291 1.229 1.428 1.229 1.281 0.962 0.998
1.425 1.344 1.221 1.470 1.373 1.228 0.974 0.984
1.289 1.468 1.310 1.207 1.278 1.303 0.965 1.019
1.384 1.368 1.253 1.368 1.293 1.271 0.967 1.000
16 8 16 8 12 6 1.25 1.25
16 12 12 4 8 12 1.25 1.25
8 12 12 12 8 12 1.25 1.25
1.375 1.385 1.331 1.533 1.265 1.279 0.995 0.977
1.466 1.369 1.237 1.661 1.397 1.248 0.969 0.992
1.293 1.490 1.287 1.273 1.299 1.278 0.997 0.994
1.378 1.415 1.285 1.489 1.320 1.268 0.987 0.964
G ¼ (for genotoxicity): smallest dilution factor of the test batch at which a significant induction ratio was not obtained.
Table 3 Cytotoxic characterization of the tested samples from a refinery wastewater treatment plant with the S. cerevisiae test Sample
Date of test After agitation
After centrifugation
Dilution factor Ga
1 2 3 4 5 6 River Rhine Tap water a
10–12 15–17 21–23 06–08 12–14 15–19 28–30 20–22
Dec. Dec. Dec. Jan. Jan. Jan. Jan. Jan.
Growth inhibition in % at dilution factor 1.25
Dilution factor Ga
Growth inhibition in % at dilution factor 1.25
Day 1 Day 2 Day 3 Day 1
Day 2
Day 3
Ø
Day 1 Day 2 Day 3 Day 1
Day 2
Day 3
Ø
4 4 3 2 3 2 1.25 1.25
54.9 52.5 40.0 37.7 29.5 31.8 0 0.1
37.4 59.2 40.8 29.7 32.6 33.3 0 0
50.6 55.9 43.6 35.6 32.7 31.4 0 0
4 4 3 2 3 2 1.25 1.25
52.6 49.5 37.0 38.6 28.0 31.9 0 0
37.8 59.7 43.2 27.5 33.2 33.0 0 1.4
50.1 54.7 43.2 34.5 33.3 31.8 0 0
4 4 3 2 2 2 1.25 1.25
3 4 3 2 2 2 1.25 1.25
59.6 55.9 50.0 39.5 36.1 29.1 0 0
4 3 3 2 2 2 1.25 1.25
2 4 3 2 2 2 1.25 1.25
60.0 55.0 49.5 37.3 38.6 30.5 0.4 0
G ¼ (for cytotoxicity): smallest dilution factor of the test batch at which for the first time the growth inhibition did not exceed 20 % in relation to the negative control.
ARTICLE IN PRESS
a
10–12 15–17 21–23 06–08 12–14 15–19 28–30 20–22
After agitation
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
1 2 3 4 5 6 River Rhine Tap Water
Date of test
3217
ARTICLE IN PRESS 3218
M. Schmitt et al. / Water Research 39 (2005) 3211–3218
samples were identical to 2, 1 and 3 but each was tested approximately 4 weeks later. Such a decrease of toxicity, even under deep freezing storage conditions of the samples may be due to either the freezing process itself or to the evaporation of organic compounds as it was reported in other studies. For frozen (2673 1C) samples from industrial plants known for their organic level content, Naudin et al. (1995) noted precipitation phenomena and a considerable toxicity decrease in some cases. The freezing process can reduce the sample toxicity to a certain extent, but the reduction seem to be limited, as results obtained by Naudin et al. (1995) with the long-term toxicity tests using Ceriodaphnia (Crustacea) and Brachydanio (Fish) indicated. Also, Schuytema et al. (1989) reported a toxicity decrease when sediments were frozen for 14 days, probably due to a release of soluble organic carbon from the sediments during the freezing process. The observed decreased cytotoxicity for samples 4, 5 and 6 analyzed later supports the notion of reliable sensitivity of the yeast test. The presented results suggest that the yeast test is quite sensitive to screen for the presence of soluble DNA-damaging agents and particularly suited to monitor for cytotoxicity in wastewater samples.
5. Conclusions In the yeast assay all samples revealed genotoxic properties without S9 mediated metabolic activation. The DNA damage induced transcriptional activation of the gfp in the yeast assay is a suitable marker for genotoxicity. The yeast assay, with gfp as directly measurable genotoxicity reporter and growth inhibition as cytotoxicity reporter can be performed in 96-well plates within 8 h and provides a high specificity by the separate analysis of 2 independent parameters. The obtained results indicate the reproducibility of the yeast assay with small variations. These properties suggest that the Saccharomyces cerevisiae geno- and cytotoxicity assay is suitable for medium or high throughput screening system for toxic effects and as biomonitor for complex matrices. Even the aging of frozen samples, containing organics, can be detected.
References Afanassiev, V., Sefton, M., Anantachaiyong, T., Barker, G., Walmsley, R., Wo¨lfl, S., 2000. Application of yeast cells transformed with GFP expression constructs containing the RAD54 or RNR2 promoter as a test for the genotoxic potential of chemical substances. Mutat. Res. 464, 297–308. Cormack, B.P., Valdivia, R.H., Falkow, S., 1996. FACSoptimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. Gee, P., Maron, D.M., Ames, B.N., 1994. Detection and classification of mutagens: a set of base-specific Salmonella tester strains. Proc. Natl. Acad. Sci. 91, 11606–11610. Kolaczkowska, A., Goffeau, A., 1999. Regulation of pleiotropic drug resistance in yeast. Drug Resist. Updat. 2, 403–414. Lichtenberg-Frate´, H., Schmitt, M., Gellert, G., Ludwig, J., 2003. A yeast based method for the detection of cyto- and genotoxicity. Toxicol. in Vitro 17, 709–716. Maron, D.M., Ames, B.N., 1983. Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Naudin, S., Garric, J., Vindimian, E., Bray, M., Migeon, B., Vollat, B., Lenon, G., 1995. Influence of the sample preservation mode to assess the chronic toxicity of an industrial effluent. Ecotoxicol. Environ. Saf. 30, 54–62. Oda, Y., Nakamura, S., Oki, I., Kato, T., Shinagawa, H., 1985. Evaluation of the new system (umu-test) for the detection of environmental mutagens and carcinogens. Mutat. Res. 147, 219–229. Quillardet, P., Hofnung, M., 1985. The SOS Chromotest, a colorimetric bacterial assay for genotoxins: procedures. Mutat. Res. 147, 65–78. Reifferscheid, G., Heil, J., Oda, Y., Zahn, R.K., 1991. A microplate version of the SOS/umu-test for rapid detection of genotoxins and genotoxic potentials of environmental samples. Mutat. Res. 253, 215–222. Schmitt, M., Gellert, G., Kirberg, B., Ludwig, J., LichtenbergFrate, H., 2002. CyGene: Eine neue hefebasierte Methode zur Bestimmung des zytotoxischen Potenzials von Umweltgiften im Wasserbereich. Vom Wasser 99, 1111–1118. Schmitt, M., Gellert, G., Ludwig, J., Lichtenberg-Frate´, H., 2004. Phenotypic yeast growth analysis for chronic toxicity testing. Ecotox. Environ. Saf. 59, 142–150. Schuytema, G.S., Nebeker, A.V., Griffis, W.L., Miller, C.E., 1989. Effects of freezing on toxicity of sediments contaminated with DDT and endrin. Environ. Toxicol. Chem. 8, 150–155. Walmsley, R.M., Billinton, N., Heyer, W.-D., 1997. Green fluorescent protein as a reporter for the DNA damageinduced gene RAD54 in Saccharomyces cerevisiae. Yeast 13, 1535–1545.