The influence of organic solvents on estimates of genotoxicity and antigenotoxicity in the SOS chromotest

In this work, the toxicity and genotoxicity of organic solvents (acetone, carbon tetrachloride, dichloromethane, dimethylsulfoxide, ethanol, ether and methanol) were studied using the SOS chromotest. The influence of these solvents on the direct genotoxicity induced by the mutagens mitomycin C (MMC) and 4-nitroquinoline-1-oxide (4-NQO) were also investigated. None of the solvents were genotoxic in Escherichia coli PQ37. However, based on the inhibition of protein synthesis assessed by constitutive alkaline phosphatase activity, some solvents (carbon tetrachloride, dimethylsulfoxide, ethanol and ether) were toxic and incompatible with the SOS chromotest. Solvents that were neither toxic nor genotoxic to E. coli (acetone, dichloromethane and methanol) significantly reduced the genotoxicity of MMC and 4-NQO. When these solvents were used to dissolve vitamin E they increased the antigenotoxic activity of this compound, possibly through additive or synergistic effects. The relevance of these results is discussed in relation to antigenotoxic studies. These data indicate the need for careful selection of an appropriate diluent for the SOS chromotest since some solvents can modulate genotoxicity and antigenotoxicity.


Introduction
Antimutagens are agents that reduce the number of spontaneous and/or induced mutations in cells. These agents are classified based on their mechanism of action (Wattenberg, 1985). A practical classification is one that separates antimutagens into des-mutagens and bio-antimutagens (Kada and Shimoi, 1987), with des-mutagens being compounds that inactivate mutagens before they can attack the DNA molecule whereas bio-antimutagens interfere with cellular processes that repair DNA damage, e.g., DNA replication and/or repair mechanisms (Kuroda and Inoue, 1988).
Antimutagens may help to prevent mutations involved in cancer and aging by regulating cellular mutability. Consequently, antimutagens can be classified into several categories and subcategories, depending on the stage in which they interfere with mutagenesis and carcinogenesis and on how they modulate host defense mechanisms; these points of interference allow the rational implementation of chemoprevention strategies for controlling cancer and de-generative diseases (De Flora, 1998;De Flora and Ramel, 1988;De Flora and Ferguson, 2005).
The assessment of natural antimutagens has involved mainly the use of short-term bacterial assays because their sensitivity, simplicity and flexibility to different experimental settings facilitates the study of antimutagenesis mechanisms (De Flora et al., 1992a,b). Among the in vitro assays used in antimutagen studies, the SOS chromotest (Quillardet et al., 1982) is one of the most widely used. This assay, which is based on a colorimetric assessment of SOS transcriptional gene fusion in Escherichia coli PQ37, allows one to estimate primary DNA damage produced by chemical and physical mutagens by measuring the expression of a reporter gene, b-galactosidase (Quillardet and Hofnung, 1993). Compared to other methods, the SOS chromotest allows greater throughput when screening natural compounds (Fuentes et al., 2006a).
The use of in vitro assays to study antimutagens/antigenotoxins can be limited by the toxicity, e.g., cytotoxic and cytostatic effects and inhibition of macromolecule synthesis, of the compounds being investigated (Zeiger, 2007). Secondary interactions between the solvent used to dissolve the test compounds or extracts and the mutagen may result in antagonistic effects in the cells, depending on the compounds or mixtures involved (Donnelly et al., 1998).
For example, dimethylsulfoxide (DMSO, a commonly used solvent for many test compounds) reduces the mutagenicity of benzo[a]pyrene, cisplatin derivatives, dimethylnitrosamine, dimethylcarbamylchlorides and diethyl-carbamylchlorides in the Salmonella/microsome reversion assay (Yahagi et al., 1977;Hermann et al., 1978;Maron et al., 1981;Uno and Morita, 1993), attenuates the genotoxicity of 2-aminoanthracene and 2-nitrofluorene in the microscreen prophage-induction assay (DeMarini et al., 1991) and attenuates cisplatin genotoxicity in the SOS chromotest (Gebel and Koenig, 1999). Other solvents, such as acetone and ethanol, also reduce benzo [a]pyrene mutagenicity in the Salmonella/microsome reversion assay (Maron et al., 1981) whereas methanol inhibits 2-aminoanthracene and 2-nitrofluorene genotoxicity in the microscreen prophageinduction assay (DeMarini et al., 1991). Since the inhibition of toxicity and genotoxicity by solvents can lead to overestimates of antigenotoxicity it is essential to determine which solvents are appropriate for screening antigenotoxins with the SOS chromotest.
We have previously shown that the antigenotoxicity of certain compounds is influenced by the solvent used to dissolve essential oils (Vicuña et al., 2010;López et al., 2011). We have observed that the use of DMSO leads to overestimation of the antigenotoxicity of essential oils when compared with results obtained using distilled water as the diluent (unpublished data). In the present study, we used a series of approaches to examine the extent to which organic solvents can affect estimates of antigenotoxicity in the SOS chromotest. First, we examined the direct toxicity and genotoxicity of several solvents (acetone, carbon tetrachloride, dichloromethane, DMSO, ether, ethanol, and methanol) in E. coli PQ-37. Second, solvents that were neither toxic nor genotoxic to E. coli (acetone, dichloromethane and methanol) were assayed for their ability to interfere with the genotoxic effect of the directly acting mutagens mitomycin C (MMC) and 4-nitro-quinoline-1-oxide (4-NQO). MMC is an alkylating agent that produces adducts and strand crosslinking in DNA (Tomasz and Palom, 1997) while 4-NQO produces mainly guanine adducts (Fronza et al., 1994). Both mutagens are powerful inducers of the SOS response in E. coli (Quillardet and Hofnung, 1993). Third, solvent concentrations that did or did not interfere (noninterfering) with direct genotoxic activities were assayed to assess whether they increased the antigenotoxicity of a-tocopherol (vitamin E). Vitamin E was used because this compound protects against oxidative mutagenesis (Odin, 1997), the primary mechanism by which MMC and 4-NQO damage DNA.

Bacterial strain and culture
Escherichia coli strain PQ37 [Fthr leu his-4 pyrD thi galE galK or galT lacDU169 srl300::Tn10 rpoB rpsL uvrA rfa trp::Muc + sfiA::Mud(Ap,lac)ts] which has been recommended for the detection of genotoxic carcinogens (Quillardet et al., 1982) was used in this work. This strain carries the sulA::lacZ fusion gene as a reporter gene for primary DNA damage induced during the SOS response. The cells were grown overnight at 37°C and shaken at 100 rpm in LB medium (10 g tryptone/L, 5 g yeast extract/L, 10 g sodium chloride/L, pH 7.4) supplemented with ampicillin (50 mg/mL) and tetracycline (17 mg/mL).

Genotoxicity assay
Genotoxicity assays were done using the SOS chromotest as described by Quillardet et al. (1982), with small modifications. Briefly, overnight cultures were grown in fresh LB medium to an optical density (OD 600nm ) of 0.4, diluted 10-fold in double strength LB medium (20 g tryptone/L, 10 g yeast extract/L, 20 g sodium chloride/L, pH 7.4) and mixed (v/v) with the test substances of interest (mutagens and solvents). A negative control (distilled water) was always included in each assay. The bacteria were exposed to different concentration ranges, depending on the test substance, for 30 min at 8°C and then cultured for 2 h at 37°C.
b-Galactosidase and alkaline phosphatase were assayed in 96-well plates (Brand GMBH, Germany). For b-galactosidase activity, cell membranes were disrupted by mixing 142 mL of Z buffer (60 mM Na 2 HPO 4 , 40 mM NaH 2 PO 4 , 10 mM KCl, 1 mM Mg 2 SO 4 , 0.1% SDS and 40 mM b-mercaptoethanol, pH 7.0) with 15 mL of cell culture for 20 min at room temperature. The reaction was started by adding 30 mL of ONPG (4 mg/mL in T buffer: 1 M Tris-HCl, pH 8.8). After 40 min, the enzymatic reaction was stopped by adding 50 mL of 1 M Na 2 CO 3 . After five min, 50 mL of 2 M Tris was added to restore the color. For alkaline phosphatase activity, cell membranes were disrupted by adding 130 mL of T buffer, 7 mL of chloroform and 5 mL of 0.1% sodium dodecyl sulfate (SDS) solution to 15 mL of cell culture followed by mixing vigorously for 20 min at room temperature. The enzyme reaction was started by adding 30 mL of PNPP solution (4 mg/mL in T buffer). After 40 min, the reaction was stopped by adding 50 mL of 2 M HCl. After 5 min, 50 mL of 2 M Tris was added to restore the color. The final absorbances of the b-galactosidase and alkaline phosphatase assays were measured at 420 nm using an iMark microplate reader (BioRad, Hercules, CA, USA). b-Galactosidase and alkaline phosphatase activities were calculated using the relationship: enzyme units = (1000 x A 420 )/t (Quillardet and Hofnung, 1985), where A 420 is the optical density at 420 nm and t is the length of incubation (min) with substrate (ONPG or PNPP). The ratio of b-galactosidase units to alkaline phosphatase units (R = (b -galactosidase units)/alkaline phosphatase units) reflects the induction of the sulA gene even when there is some inhibition of protein synthesis.
The genotoxicity criterion used was the SOS induction factor (I) that represents the normalized induction data of the sulA gene in each treatment (mutagen, solvent, etc.) and was therefore considered to be an indirect measure of the primary DNA damage (genotoxicity) induced by the treatments. This parameter was calculated as: I = R t /R nt , where t and nt are the treated and non-treated cells, respectively. A substance was classified as not genotoxic if I was < 1.5, inconclusive if I was between 1.5 and 2.0, and genotoxic if I was > 2.0 and a clear concentration-response relationship was observed.

Antigenotoxicity assay
Antigenotoxicity was assayed using a co-incubation procedure described by Fuentes et al. (2006b). The antigenotoxicity assay was done essentially as the genotoxicity protocol described above, except that the cells were cotreated with different concentrations of the test substances (solvents, vitamin E or both) and mutagens (0.187 mM MMC or 2.34 mM 4-NQO), simultaneously. Co-treatments with solvent and mutagens allowed us to determine the solvent concentrations that interfered with MMC and 4-NQO activities. To determine whether the solvents also affected the antigenotoxicity of vitamin E, a stock solution (8 mM) of this substance was prepared as indicated by Aiub et al. (2009). For the SOS chromotest, vitamin E (0.02-2 mM; antigenotoxic concentration range) was prepared by dissolving the compound in compatible solvents (acetone, dichloromethane and methanol) at a final solvent concentration that did or did not interfere with the genotoxic activities of MMC and 4NQO. The interfering solvent concentration in co-treatments was always 25% while the non-interfering solvent concentration varied as follows: MMC/acetone = 0.78%, 4-NQO/acetone = 1.56%, MMC/dichloromethane = 3.12%, 4-NQO/dichloromethane = 6.25%, MMC/methanol = 3.12% and 4-NQO/methanol = 3.12%. These non-interfering solvent concentrations were selected based on their insignificant effect in the solvent/mutagen co-treatments. A control assay in which vitamin E was dissolved in distilled water was included in all experiments for comparison.
Antigenotoxicity (the ability of the test substance to protect DNA against damage) was measured as a signifi-Organic solvents and the SOS chromotest 505 cant reduction in the induction factor (I) in the combined treatments (substance + mutagens) and expressed as a percentage of the genotoxicity inhibition (%GI): where I CO is the SOS induction factor in co-treated cells (substance + mutagen), I BASAL is the basal SOS induction factor and I MUT is the SOS induction factor in mutagentreated cells. Negative values of %GI were considered as zero, which meant that this parameter ranged from a minimum of 0% to a maximum of 100%.

Statistical analysis
The b-galactosidase and alkaline phosphatase activities and the induction factor (I) values were expressed as the mean±SEM. The data normality was tested using the Kolmogorov-Smirnov test and the homogeneity of variances was assessed by analysis of variance (ANOVA). Statistical comparisons were done using Student's t-test. Product-moment (Pearson) correlation analysis was used to examine the concentration-response relationship in the genotoxicity experiments. A value of p < 0.05 indicated significance in all cases. All data analyses were done using the STATISTICA software package v. 6 (StatSoft Inc., Tulsa, OK, USA).

Results
Induction kinetics of the sulA gene by standard mutagens Figure 1 shows the induction kinetics of the sulA gene in E. coli PQ37 following treatment with mutagens MMC and 4-NQO. The extent and kinetics of sulA gene induction by MMC and 4-NQO varied with the mutagen concentration, with the highest levels of induction occurring at 0.187 mM and 2.34 mM, respectively. At these concentrations, there was no significant reduction in protein synthesis as indicated by the constitutive alkaline phosphatase activity. These concentrations were therefore used for cotreatments in antigenotoxicity assays.

Solvent toxicity and genotoxicity in E. coli PQ37
Solvent toxicity was measured indirectly based on the inhibition of constitutive alkaline phosphatase activity in E. coli PQ37 (Table 2). Acetone, dichloromethane and methanol did not significantly reduce protein synthesis over the concentration range studied. In contrast, carbon tetrachloride, DMSO, ethanol and ether significantly reduced alkaline phosphatase activity at concentrations = 161, 438, 266 and 150 mM, respectively. The inhibition of protein synthesis by these solvents was also evidenced by the decrease in b-galactosidase activity (data not shown).
The genotoxicity of the organic solvents was assayed before their modulating effect was investigated (Table 2). Acetone, dichloromethane, DMSO, ethanol, ether and methanol did not significantly increase the I values in E. coli PQ37, indicating that these compounds do not induce the SOS response in E. coli. Carbon tetrachloride significantly increased the I values at the highest concentration (5158 mM), although correlation analysis revealed no concentration-response relationship for this solvent (data not shown). Only solvents that were neither toxic nor genotoxic to E. coli PQ37 (acetone, dichloromethane and methanol) were considered for further assays. 506 Quintero et al.   Table 3 shows the capacity of the solvents to interfere with mutagen genotoxicity. All of the solvents assayed (acetone, dichloromethane and methanol) interfered with MMC-and 4NQO -induced genotoxicity. Dichlorome-thane and methanol concentrations > 6.25% significantly reduced the I values in co-treated cells compared to the positive control whereas concentrations £ 3.12% did not. In contrast, acetone reduced the I values in co-treatment with MMC and 4NQO at concentrations ³ 1.56% and ³ 3.12%, respectively. 508 Quintero et al.

Influence of organic solvents on vitamin E antigenotoxicity
We hypothesized that the solvent concentrations that interfered with the genotoxic activities of MMC-and 4-NQO could increase the antigenotoxicity of vitamin E against these mutagens, resulting in overestimates of antigenotoxicity. To assess this hypothesis, we assayed the acetone, dichloromethane and methanol concentrations that did and did not interfere with the genotoxic activities of MMC and 4-NQO (Tables 4 and 5). A control assay in Organic solvents and the SOS chromotest 509 6.5 ± 1.6 4.9 ± 1.1 6.5 ± 1.9 Vitamin E (0.016 mM) + MMC 7.7 ± 1.9 (0) 0.5 ± 0.2 (100)* 6.6 ± 1.4 (0.0) Vitamin E (0.031 mM) + MMC 6.8 ± 2.1 (0) 0.4 ± 0.1 (100)* 5.9 ± 1.1 (11) Vitamin E (0.062 mM) + MMC 5.5 ± 0.9 (17)  which vitamin E was dissolved in distilled water was included in all experiments for comparison. Tables 4 and 5 show the antigenotoxicity of vitamin E against DNA damage induced by MMC and 4-NQO. When dissolved in distilled water (treatment A), vitamin E signifi-cantly inhibited the MMC and 4-NQO genotoxicities at concentrations ³ 0.125 mM. In contrast, when dissolved in solutions containing 25% solvent (treatment B), vitamin E significantly reduced the genotoxicities of MMC and 4-NQO genotoxicities at concentrations = 0.016 mM. These 510 Quintero et al.  findings support the hypothesis that organic solvents can enhance antigenotoxic activity, possibly through an additive effect. Surprisingly, the antigenotoxic effect of vitamin E against MMC and 4-NQO (treatment C) was also enhanced when the vitamin was dissolved in a non-interfering concentration of solvent. Only the co-treatment with MMC/vitamin E dissolved in 0.78% acetone showed similar antigenotoxicity to that of treatment A, although there was a significant reduction in the MMC genotoxicity at a lower vitamin concentration (0.062 mM). These results indicate that low, non-interfering concentrations of solvent can also influence the antigenotoxicity of vitamin E.

Discussion
In this study, we examined some of the factors that can influence the results of the SOS chromotest that is frequently used to screen for antimutagenic activity. Specifically, we investigated (1) the toxicity and genotoxicity of solvents, (2) the ability of solvents to interfere with mutagenesis, and (3) the contribution of organic solvents to overestimates of the antimutagenic activity of vitamin E (through additive or synergistic effects of the solvents).
Based on previous studies with the SOS chromotest (Von der Quillardet and Hofnung, 1993;Mersch-Sundermann et al., 1994), the solvents studied here were not genotoxic. However, some solvents such as carbon tetrachloride, DMSO, ethanol and ether were toxic to E. coli at high concentrations, as indicated by the inhibition of protein synthesis. In the standard procedure for the SOS chromotest (Quillardet and Hofnung, 1985) the maximum volume of solvent (DMSO or ethanol) used is 20 mL/assay (i.e., for 600 mL of cell suspension), which corresponds to a final solvent concentration of 3.3% in the reaction mixture.
Since there is very little information on the influence of solvents in the SOS chromotest, the data reported in Table 2 can be of practical use in choosing solvents compatible with the SOS chromotest when screening natural antigenotoxic compounds.
The ability of solvents to interfere with chemical mutagenesis has been documented, as indicated in the Introduction. For example, solvents such as DMSO can interfere with the genotoxicity of cisplatin in the SOS chromotest (Gebel and Koenig, 1999). As shown here, several solvents (including some not studied before) significantly reduced the genotoxicities of MMC and 4-NQO at high concentrations; this interference may limit the usefulness of the SOS Chromotest during antimutagenesis screening if the test substances are soluble only in organic solvents. Under these conditions, the assay can yield overestimates of antigenotoxicity for the tested compound. Acetone, dichloromethane and methanol increased the antigenotoxicity of vitamin E when the solvents were tested at a final concentration of 25%, which interfered with mutagenic activity. Surprisingly, these solvents also increased vitamin E anti-genotoxicity at a concentration that did not interfere with mutagen activity, which suggested a synergistic interaction between solvents and vitamin E under these conditions. Donnelly et al. (1998) have previously shown that this synergic and/or additive interaction is characteristic of complex mixtures that contain mutagens.
DMSO is a cytoprotective compound frequently used as a solvent in mutagenic and anti-mutagenic studies because it has been considered to be non-genotoxic (Yu and Quinn, 1994). Indeed, several studies have used DMSO as a solvent to study the antimutagenic properties of biological extracts and/or natural compounds in the SOS chromotest (Table 6). However, DMSO can act as a free radical scavenger and can penetrate cell membranes to affect cell membrane stability (Yu and Quinn, 1998) and this in turn can interfere with the activity of antimutagens. As shown by Gebel and Koenig (1999), DMSO can also interfere with the activity of different mutagens in the SOS chromotest. Based on these findings, a critical evaluation of the antigenotoxic potential of the antimutagens indicated in Table 6 is required.
We have previously proposed the SOS chromotest as a potentially useful assay for primary screening of medicinal plant extracts for antigenotoxic activity against g-rays (Fuentes et al., 2006a). With this test, we have observed that the use of DMSO as solvent resulted in overestimates of the antimutagenic activity of essential oils when compared with distilled water (unpublished data). The additive and synergistic effect of solvents on vitamin E antigenotoxicity shown here supports the use of distilled water as the ideal solvent for antimutagen screening with the SOS chromotest. When substances soluble only in organic solvents are tested it is necessary to establish adequate final solvent concentrations to avoid interference with the test results. Solvents such as acetone, dichloromethane and methanol that were neither toxic nor genotoxic to E. coli PQ37 cells appear to be compatible with the SOS chromotest and are potentially useful for studying antimutagens. However, their ability to interfere with the assayed mutagens needs to be evaluated in each case. Based on the results described here, we conclude that: (1) the solvents studied were not genotoxic in E. coli PQ37 cells, although some of them (carbon tetrachloride, DMSO, ethanol and ether) inhibited protein synthesis, depending on their concentration, (2) organic solvents can reduce the genotoxicity of MMC and 4-NQO in the SOS chromotest in a concentration-dependent manner; water is the best solvent when screening for antigenotoxicity in this assay, and (3) a solvent concentration that interferes directly with mutagen activity can significantly increase the antigenotoxicity of vitamin E, possibly through additive and synergistic effects; this interference can result in overestimates of antigenotoxicity. Overall, these findings indicate the need to carefully determine the appropriate concentration Organic solvents and the SOS chromotest 511