Zearalenone Does Not Show Genotoxic Effects in the Drosophila melanogaster Wing Spot Test, but It Induces Oxidative Imbalance, Development, and Fecundity Alterations

Zearalenone (ZEN) is a non-steroidal mycoestrogen produced by the Fusarium genus. ZEN and its metabolites compete with 17-beta estradiol for cytosolic estrogen receptors, causing reproductive alterations in vertebrates. ZEN has also been associated with toxic and genotoxic effects, as well as an increased risk for endometrial adenocarcinomas or hyperplasia, breast cancer, and oxidative damage, although the underlying mechanisms remain unclear. Previous studies have monitored cellular processes through levels of transcripts associated with Phase I Xenobiotic Metabolism (Cyp6g1 and Cyp6a2), oxidative stress (hsp60 and hsp70), apoptosis (hid, grim, and reaper), and DNA damage genes (Dmp53). In this study, we evaluated the survival and genotoxicity of ZEN, as well as its effects on emergence rate and fecundity in Drosophila melanogaster. Additionally, we determined levels of reactive oxygen species (ROS) using the D. melanogaster flare and Oregon R(R)-flare strains, which differ in levels of Cyp450 gene expression. Our results showed that ZEN toxicity did not increase mortality by more than 30%. We tested three ZEN concentrations (100, 200, and 400 μM) and found that none of the concentrations were genotoxic but were cytotoxic. Taking into account that it has previously been demonstrated that ZEN administration increased hsp60 expression levels and apoptosis gene transcripts in both strains, the data agree with an increase in ROS and development and fecundity alterations. Since Drosophila lacks homologous genes for mammalian estrogen receptors alpha and beta, the effects of this mycotoxin can be explained by a mechanism different from estrogenic activity.

. Mortality percentages of larvae fed with ZEN. Flare and Oregon R(R)-Flare third instar larvae were fed until pupation with medium supplemented with ZEN [0, 100, 200, 300, and 500 µM]. The dotted line indicates the mortality rate corresponding to 260 µM. The data were obtained from four independent experiments (with three replicates per treatment) and were analyzed using a two −way ANOVA with a significance level of p < 0.05.

Genotoxicity Test
As expected, the pro-mutagen Urethane (URE, bioactivation and positive control) resulted in higher spot frequencies in the HB cross compared to the ST cross, due to the former's high and inducible Cyp450s [20]. However, none of the concentrations of ZEN tested showed genotoxic effects, such as mutation, recombination, deletion, and aneuploidy [21], in both crosses (Table 1). Although we did not observe genotoxicity with this mycotoxin, statistical differences (p < 0.05) were found between the frequencies of small and total spots at 100 and 400 µM and between the small spot frequency in the 100 and 200 µM treatments in the ST cross with inducible Cyp450s. Notably, these frequencies decreased as the concentration increased (Table 1). Additionally, although we did not observe ZEN genotoxicity, the analysis of the accumulated mwh clone size class distribution of both crosses, using the Kolgomorov-Smirnov test (p < 0.05) [22], showed statistically significant differences between all treatments and the corresponding PBS dissolvent control, with a concentration-dependent decrease in the accumulated mwh clone size class distribution, but only in the ST cross (Table 2). The dotted line indicates the mortality rate corresponding to 260 µM. The data were obtained from four independent experiments (with three replicates per treatment) and were analyzed using a two −way ANOVA with a significance level of p < 0.05.

Genotoxicity Test
As expected, the pro-mutagen Urethane (URE, bioactivation and positive control) resulted in higher spot frequencies in the HB cross compared to the ST cross, due to the former's high and inducible Cyp450s [20]. However, none of the concentrations of ZEN tested showed genotoxic effects, such as mutation, recombination, deletion, and aneuploidy [21], in both crosses (Table 1). Although we did not observe genotoxicity with this mycotoxin, statistical differences (p < 0.05) were found between the frequencies of small and total spots at 100 and 400 µM and between the small spot frequency in the 100 and 200 µM treatments in the ST cross with inducible Cyp450s. Notably, these frequencies decreased as the concentration increased (Table 1). Additionally, although we did not observe ZEN genotoxicity, the analysis of the accumulated mwh clone size class distribution of both crosses, using the Kolgomorov-Smirnov test (p < 0.05) [22], showed statistically significant differences between all treatments and the corresponding PBS dissolvent control, with a concentration-dependent decrease in the accumulated mwh clone size class distribution, but only in the ST cross (Table 2).

ZEN Identification and Quantification by HPLC in the Flare and Oregon R(R)-Flare Strains
Based on the HPLC analysis of three extract replicates from the flare and Oregon R(R)-flare imagos, a peak corresponding to the ZEN standard was detected, indicating that ZEN was incorporated into the larvae. The area under the curve was measured to quantify the amount of ZEN incorporated, and no statistical differences were found between the two strains. The mean concentration of ZEN in the flare strain was 3.77 mg/kg, while it was 5.86 mg/kg for the Oregon R(R)-flare strain, with standard deviations of +/− 2.0 and +/− 2.3, respectively (according to Supplementary Figure S1).

Emergence
The percentage of emergence for each strain and treatment ( Figure 2) showed statistically significant differences for the following comparisons: strains vs. ZEN (260 µM), PBS (dissolvent control) vs. toluene (TOL) 50 mM (positive control) and PBS vs. ZEN. In contrast, no significant differences in the emergence were observed between TOL and ZEN. The emergence of imagos from the flare strain treated with ZEN decreased 20.8% with respect to the PBS dissolvent control, while, in the Oregon R(R)-flare strain, the emergence decreased 27.7% with respect to the control. On the other hand, the emergence of imagos from the flare and Oregon R(R)-flare larvae treated with TOL decreased to 17% and 16% only, respectively.

ZEN identification and quantification by HPLC in the flare and Oregon R(R)-flare strains
Based on the HPLC analysis of three extract replicates from the flare and Oregon R(R)-flare imagos, a peak corresponding to the ZEN standard was detected, indicating that ZEN was incorporated into the larvae. The area under the curve was measured to quantify the amount of ZEN incorporated, and no statistical differences were found between the two strains. The mean concentration of ZEN in the flare strain was 3.77 mg/kg, while it was 5.86 mg/kg for the Oregon R(R)-flare strain, with standard deviations of +/− 2.0 and +/− 2.3, respectively (according to Supplementary Figure S1).

Emergence
The percentage of emergence for each strain and treatment ( Figure 2) showed statistically significant differences for the following comparisons: strains vs. ZEN (260 µM), PBS (dissolvent control) vs. toluene (TOL) 50 mM (positive control) and PBS vs. ZEN. In contrast, no significant differences in the emergence were observed between TOL and ZEN. The emergence of imagos from the flare strain treated with ZEN decreased 20.8% with respect to the PBS dissolvent control, while, in the Oregon R(R)-flare strain, the emergence decreased 27.7% with respect to the control. On the other hand, the emergence of imagos from the flare and Oregon R(R)-flare larvae treated with TOL decreased to 17% and 16% only, respectively. The asterisks over the lines indicate statistically significant differences. The data were analyzed using a two-way ANOVA with a significance level of p < 0.05 *, p < 0.01 **.

Fecundity
The mean daily egg-laying rate of female flies was compared between PBS dissolvent control, TOL (50 mM), and ZEN (260 µM) dissolved in PBS. Both flare ( Figure 3A) and Oregon R(R)-flare ( Figure 3B) strains treated with TOL and ZEN showed an overall decreasing trend in mean daily egg-laying rates compared to the PBS control, with the ZEN treatment resulting in a lower rate over 10 days. The total number of eggs laid by The asterisks over the lines indicate statistically significant differences. The data were analyzed using a two-way ANOVA with a significance level of p < 0.05 *, p < 0.01 **.

Fecundity
The mean daily egg-laying rate of female flies was compared between PBS dissolvent control, TOL (50 mM), and ZEN (260 µM) dissolved in PBS. Both flare ( Figure 3A) and Oregon R(R)-flare ( Figure 3B) strains treated with TOL and ZEN showed an overall decreasing trend in mean daily egg-laying rates compared to the PBS control, with the ZEN treatment resulting in a lower rate over 10 days. The total number of eggs laid by female flies from each strain was also recorded and the mean percentage calculated (Figure 4). Significant differences were observed between treatments and strains compared to the PBS control. The flare strain treated with TOL showed only 40% fecundity, while the Oregon R(R)-flare strain showed 64% fecundity. In the ZEN treatment, the flare strain had a higher fecundity (61%) than the Oregon R(R)-flare strain (45%). female flies from each strain was also recorded and the mean percentage calculated ( Figure 4). Significant differences were observed between treatments and strains compared to the PBS control. The flare strain treated with TOL showed only 40% fecundity, while the Oregon R(R)-flare strain showed 64% fecundity. In the ZEN treatment, the flare strain had a higher fecundity (61%) than the Oregon R(R)-flare strain (45%).  The asterisks above the lines indicate statistically significant differences. The data were analyzed using a two-way ANOVA with a significance level of p < 0.05 *, 0.01 **, 0.001 ***.

Transcriptional Expression in Basal State
Prior to evaluating the expression of transcripts, we verified the size of the amplifications for each oligonucleotide pair ( Figure 5).Transcriptional expression of genes associated with XM, oxidative stress and apoptosis in the basal state revealed differences between the studied untreated strains. Our data indicated that the relative expressions of Cyp6g1 and Cyp6a2 genes were 10 and three times higher, respectively, in the larvae of the Oregon R(R)-flare strain compared to the flare strain (see Figure 6). Additionally, we observed contrasting expression levels of hsp60 and hsp70 genes, as the anti-and proapoptotic hsp60 levels decreased in the Oregon R(R)-flare strain, while the pro-apoptotic hsp70 levels increased approximately five times more than those in the flare strain. Similarly, the apoptosis genes transcripts showed contrasting expressions in the Oregon R(R)-flare strain, where the transcript abundance of grim was lower than in the flare strain, but the levels of hid and reaper were 3.74 and 2.81-fold higher than those in the flare strain exposed to similar conditions (see Figure 6A). female flies from each strain was also recorded and the mean percentage calculated ( Figure 4). Significant differences were observed between treatments and strains compared to the PBS control. The flare strain treated with TOL showed only 40% fecundity, while the Oregon R(R)-flare strain showed 64% fecundity. In the ZEN treatment, the flare strain had a higher fecundity (61%) than the Oregon R(R)-flare strain (45%).  The asterisks above the lines indicate statistically significant differences. The data were analyzed using a two-way ANOVA with a significance level of p < 0.05 *, 0.01 **, 0.001 ***.

Transcriptional Expression in Basal State
Prior to evaluating the expression of transcripts, we verified the size of the amplifications for each oligonucleotide pair ( Figure 5).Transcriptional expression of genes associated with XM, oxidative stress and apoptosis in the basal state revealed differences between the studied untreated strains. Our data indicated that the relative expressions of Cyp6g1 and Cyp6a2 genes were 10 and three times higher, respectively, in the larvae of the Oregon R(R)-flare strain compared to the flare strain (see Figure 6). Additionally, we observed contrasting expression levels of hsp60 and hsp70 genes, as the anti-and proapoptotic hsp60 levels decreased in the Oregon R(R)-flare strain, while the pro-apoptotic hsp70 levels increased approximately five times more than those in the flare strain. Similarly, the apoptosis genes transcripts showed contrasting expressions in the Oregon R(R)-flare strain, where the transcript abundance of grim was lower than in the flare strain, but the levels of hid and reaper were 3.74 and 2.81-fold higher than those in the flare strain exposed to similar conditions (see Figure 6A). and ZEN 260 µM dissolved in PBS. The asterisks above the lines indicate statistically significant differences. The data were analyzed using a two-way ANOVA with a significance level of p < 0.05 *, 0.01 **, 0.001 ***.

Transcriptional Expression in Basal State
Prior to evaluating the expression of transcripts, we verified the size of the amplifications for each oligonucleotide pair ( Figure 5). Transcriptional expression of genes associated with XM, oxidative stress and apoptosis in the basal state revealed differences between the studied untreated strains. Our data indicated that the relative expressions of Cyp6g1 and Cyp6a2 genes were 10 and three times higher, respectively, in the larvae of the Oregon R(R)-flare strain compared to the flare strain (see Figure 6). Additionally, we observed contrasting expression levels of hsp60 and hsp70 genes, as the anti-and pro-apoptotic hsp60 levels decreased in the Oregon R(R)-flare strain, while the pro-apoptotic hsp70 levels increased approximately five times more than those in the flare strain. Similarly, the apoptosis genes transcripts showed contrasting expressions in the Oregon R(R)-flare strain, where the transcript abundance of grim was lower than in the flare strain, but the levels of hid and reaper were 3.74 and 2.81-fold higher than those in the flare strain exposed to similar conditions (see Figure 6A).

Transcriptional Expression Associated with ZEN Exposure
The transcript levels of genes related to XM, oxidative stress, and apoptosis were analyzed in untreated and ZEN-treated larvae from the flare and Oregon R(R)-flare strains. Figure 6B presents the gene expression data for the flare strain following ZEN administration, compared to control conditions (PBS pH7). The ZEN treatment in flare larvae led to increased transcript levels of Cyp6g1, Cyp6a2, hsp60, hsp70, grim, reaper, and Dmp53 genes (3.39, 1.85, 2.9, 1.71, 2.96, 5.49, and 1.95-fold higher than under control conditions, respectively), except for hid, which showed downregulated expression (0.19) ( Figure 6). In contrast, ZEN administration to Oregon R(R)-flare larvae induced a different expression profile ( Figure 6B). While Cyp6g1 and Cyp6a2 transcripts were induced by ZEN, the amount of Cyp6a2 transcript was 16-fold higher than under the control condition. Interestingly, the oxidative stress genes showed a contrasting expression profile, ZEN increased the amount of hsp60 transcript (1.69) but the expression level of hsp70 decreased (0.5 times), compared to the control conditions. Finally, exposure to the mycotoxin increased the levels of grim, hid, and Dmp53 (1.99, 1.21, and 1.82-fold, respectively), while the levels did not changer with reaper, compared to the control conditions ( Figure 6B).
Thus, ZEN exposure modified the gene expression in D. melanogaster flare and Oregon R(R)-flare strains differently. The amounts of Cyp6g1 and Cyp6a2 transcripts were 4.5 and 3.0-fold higher in the Oregon R(R)-flare strain than in the flare strain when ZEN

Transcriptional Expression Associated with ZEN Exposure
The transcript levels of genes related to XM, oxidative stress, and apoptosis were analyzed in untreated and ZEN-treated larvae from the flare and Oregon R(R)-flare strains. Figure 6B presents the gene expression data for the flare strain following ZEN administration, compared to control conditions (PBS pH7). The ZEN treatment in flare larvae led to increased transcript levels of Cyp6g1, Cyp6a2, hsp60, hsp70, grim, reaper, and Dmp53 genes (3.39, 1.85, 2.9, 1.71, 2.96, 5.49, and 1.95-fold higher than under control conditions, respectively), except for hid, which showed downregulated expression (0.19) ( Figure 6). In contrast, ZEN administration to Oregon R(R)-flare larvae induced a different expression profile ( Figure 6B). While Cyp6g1 and Cyp6a2 transcripts were induced by ZEN, the amount of Cyp6a2 transcript was 16-fold higher than under the control condition. Interestingly, the oxidative stress genes showed a contrasting expression profile, ZEN increased the amount of hsp60 transcript (1.69) but the expression level of hsp70 decreased (0.5 times), compared to the control conditions. Finally, exposure to the mycotoxin increased the levels of grim, hid, and Dmp53 (1.99, 1.21, and 1.82-fold, respectively), while the levels did not changer with reaper, compared to the control conditions ( Figure 6B).
Thus, ZEN exposure modified the gene expression in D. melanogaster flare and Oregon R(R)-flare strains differently. The amounts of Cyp6g1 and Cyp6a2 transcripts were 4.5 and 3.0-fold higher in the Oregon R(R)-flare strain than in the flare strain when ZEN

Transcriptional Expression Associated with ZEN Exposure
The transcript levels of genes related to XM, oxidative stress, and apoptosis were analyzed in untreated and ZEN-treated larvae from the flare and Oregon R(R)-flare strains. Figure 6B presents the gene expression data for the flare strain following ZEN administration, compared to control conditions (PBS pH7). The ZEN treatment in flare larvae led to increased transcript levels of Cyp6g1, Cyp6a2, hsp60, hsp70, grim, reaper, and Dmp53 genes (3.39, 1.85, 2.9, 1.71, 2.96, 5.49, and 1.95-fold higher than under control conditions, respectively), except for hid, which showed downregulated expression (0.19) (Figure 6). In contrast, ZEN administration to Oregon R(R)-flare larvae induced a different expression profile ( Figure 6B). While Cyp6g1 and Cyp6a2 transcripts were induced by ZEN, the amount of Cyp6a2 transcript was 16-fold higher than under the control condition. Interestingly, the oxidative stress genes showed a contrasting expression profile, ZEN increased the amount of hsp60 transcript (1.69) but the expression level of hsp70 decreased (0.5 times), compared to the control conditions. Finally, exposure to the mycotoxin increased the levels of grim, hid, and Dmp53 (1.99, 1.21, and 1.82-fold, respectively), while the levels did not changer with reaper, compared to the control conditions ( Figure 6B Figure 6B). On the other hand, the relative expression levels of anti-apoptotic and pro-apoptotic hsp60 and pro-apoptotic hsp70 were similar (0.9 and 1.2, respectively) in both strains under the same conditions. Among the apoptosis and DNA damage markers the transcript levels of hid in the Oregon R(R)-flare strain were outstanding (19-fold higher than in the flare strain), while the expressions of the other genes were down-regulated (grim 0.4 and reaper 0.6) or similar (Dmp53 1.3) ( Figure 6B).

Reactive Oxygen Species Quantification
The redox state of larval midgut cells was directly measured by flow cytometry using 2 ,7 -dichlorodihydrofluorescein diacetate (DCF-DA). ZEN exposure induced a statistically significant increase (p < 0.0001) in the percentage of positive events (Figure 7) in both strains. Interestingly, the increase in the Oregon R(R)-flare strain (in relation to the control) was less than that induced in the flare strain. was added ( Figure 6B). On the other hand, the relative expression levels of anti-apoptotic and pro-apoptotic hsp60 and pro-apoptotic hsp70 were similar (0.9 and 1.2, respectively) in both strains under the same conditions. Among the apoptosis and DNA damage markers the transcript levels of hid in the Oregon R(R)-flare strain were outstanding (19fold higher than in the flare strain), while the expressions of the other genes were downregulated (grim 0.4 and reaper 0.6) or similar (Dmp53 1.3) ( Figure 6B).

Reactive Oxygen Species Quantification
The redox state of larval midgut cells was directly measured by flow cytometry using 2′,7′-dichlorodihydrofluorescein diacetate (DCF-DA). ZEN exposure induced a statistically significant increase (p < 0.0001) in the percentage of positive events (Figure 7) in both strains. Interestingly, the increase in the Oregon R(R)-flare strain (in relation to the control) was less than that induced in the flare strain. The data are presented as the mean ± SD of three independent experiments. Asterisks denote statistical significance, with p-values of <0.005 * and <0.0001 ****, as determined by a two-tailed ttest (indicated by the capped line).

Discussion
ZEN's toxicity was not demonstrated in the D. melanogaster flare and Oregon R(R)flare strains through the survival assay, as mortality rates were not directly proportional to concentrations. Nonetheless, mortality did not exceed 30%. Therefore, we used a sublethal concentration (260 µM), which caused mortality between 20-25% in both strains, for the assays described, except for the Drosophila wing spot test. D. melanogaster possesses a steroid dehydrogenase [23] that can reduce ZEN to α-ZOL or β-ZOL metabolites, and the Cyp1a2 and Cyp6g1 genes, homologs to human Cyp1A2 and Cyp3A4 genes, respectively, that yield OH-ZEN [24]. Since the Oregon R(R)-flare strain showed high levels of Cyp450s [20] and a more efficient XM [25], a lower mortality (20% vs. 23%) was induced by ZEN (260 µM), in contrast with the flare strain, which had inducible Cyp450s levels that could account for the polynomial curves observed in the mortality rates at different concentrations (Figure 1).
The results of the Drosophila wing spot test in the ST and HB crosses suggested that ZEN and its metabolites did not have genotoxic effects, such as mutations, deletions, aneuploidy, or recombination. This contrasted with the findings of Abbès et al. [26], who observed an increase in chromosome aberrations in mouse bone-marrow cells exposed to melanogaster flare and Oregon R(R)-flare larvae exposed to ZEN (260 µM) dissolved in PBS pH 7. The data are presented as the mean ± SD of three independent experiments. Asterisks denote statistical significance, with p-values of <0.005 * and <0.0001 ****, as determined by a two-tailed t-test (indicated by the capped line).

Discussion
ZEN's toxicity was not demonstrated in the D. melanogaster flare and Oregon R(R)flare strains through the survival assay, as mortality rates were not directly proportional to concentrations. Nonetheless, mortality did not exceed 30%. Therefore, we used a sublethal concentration (260 µM), which caused mortality between 20-25% in both strains, for the assays described, except for the Drosophila wing spot test. D. melanogaster possesses a steroid dehydrogenase [23] that can reduce ZEN to α-ZOL or β-ZOL metabolites, and the Cyp1a2 and Cyp6g1 genes, homologs to human Cyp1A2 and Cyp3A4 genes, respectively, that yield OH-ZEN [24]. Since the Oregon R(R)-flare strain showed high levels of Cyp450s [20] and a more efficient XM [25], a lower mortality (20% vs. 23%) was induced by ZEN (260 µM), in contrast with the flare strain, which had inducible Cyp450s levels that could account for the polynomial curves observed in the mortality rates at different concentrations ( Figure 1).
The results of the Drosophila wing spot test in the ST and HB crosses suggested that ZEN and its metabolites did not have genotoxic effects, such as mutations, deletions, aneuploidy, or recombination. This contrasted with the findings of Abbès et al. [26], who observed an increase in chromosome aberrations in mouse bone-marrow cells exposed to ZEN and its metabolites (α-ZOL and β-ZOL) at higher concentrations than those used in our study. However, the ST cross showed a significant decrease in the number of, and altered distribution of, mwh clones at 100 and 400 µM treatments ( Table 2). The mwh clones' distribution results from cell division of imaginal disc cells of the third instar larvae, so the alteration of this process might have been due to a cytotoxic effect or apoptosis induction. All ZEN treatments, when compared to the dissolvent control (PBS), showed statistical differences, with a tendency for there to be as decrease in the mwh clones only in the ST cross. The statistical analyses of spot frequencies/fly and "mean mwh clone size class" were performed with the computer program SMART PC-version 2.1. To compare the "mean mwh clone size class" distributions of treated and control series statistically, we performed a one-sided Kolmogorov-Smirnov test (p level < 0.05). The "accumulated mwh clone size class" results showed significant differences between treated and both control series, indicating that cell division was disrupted by the parental compound or its metabolites. Therefore, these results indicated cytotoxicity, according to Santos-Cruz et al., and Graf et al. [21,22].
To confirm that Drosophila flare and Oregon R(R)-flare strains can incorporate ZEN, an HPLC analysis was conducted on adults that emerged from the treated larvae, which revealed that larvae from both strains incorporated similar amounts of ZEN. However, toxic effects, on development and reproduction, were observed, as evidenced by a decrease in imago emergence and a significant decrease in fecundity in both strains. The Oregon R(R)flare strain exhibited higher sensitivity than the flare strain, with an emergence decrease of 27.7% and a 55% decrease in fecundity, compared to 20.8% and 39%, respectively, in the flare strain ( Figure 2). This higher sensitivity may be related to Cyp450s enzyme levels, which could generate more OH-ZEN metabolites, as demonstrated by Bravin et al. using microsomes, and, thereby, exert effects on development [24].
It is worth noting that ZEN and its metabolites are known to bind to estrogen receptors [27,28], but since D. melanogaster lacks an estrogen receptor, the effects of ZEN on this route cannot be related to this receptor. However, as there are no known reported ligands for the ERR of Drosophila, it is possible that the chemical structure of OH-ZEN metabolites could bind to the orphan nuclear receptors, although this union and its effects on carbohydrate metabolism or testicular development must be demonstrated in the future [29,30].
It has been shown in human cell lines that ZEN modifies DNA global methylation patterns, as well as the expression profiles of some genes involved in cell metabolism [31]. In Caenorhabditis elegans, ZEN modifies the expression of a set of genes involved in collagen synthesis, which could account for disruption in its viability and reproduction [32]. This mycotoxin affected the viability of Sertoli cells in rats through arrest of primary cells and the triggering of autophagy [33]. On the other hand, ZEN metabolites have been proven to be involved in oxidative stress [34]. Heat shock proteins are key cellular responses to toxicant exposure and have been used as an indicator of short-term stress and toxicity [35]. We determined the relative expression levels of oxidative stress taking into account previous data for anti-apoptotic and pro-apoptotic hsp60 and pro-apoptotic hsp70, apoptosis markers (hid, grim and reaper) and DNA damage (Dmp53) genes in the flare and Oregon R(R)-flare strains treated with ZEN [36][37][38]. Our results indicated that, even under control conditions, the genetic background of Oregon R(R)-flare included the overexpression of hsp70 (4.9-fold higher than in the flare strain under similar conditions), hid (3.74), reaper (2.81) and Dmp53 (1.37). The levels of Cyp6g1 and Cyp6a2, which had been previously shown [25,39] were confirmed to be 10.3-and 3-fold higher, respectively ( Figure 6).
Interestingly, our results demonstrated that ZEN treatment modified the expression profiles in both strains. In flare, the mycotoxin induced upregulation of all markers (Cyp6g1, Cyp6a2, hsp60, hsp70, grim, reaper and Dmp53) except hid, which was downregulated. The Oregon R(R)-flare strain expression profile was modified similarly, but, in this strain, hsp70 was downregulated and reaper expression levels remained unchanged, compared to the dissolvent control condition. Proteins Hsp60 and Hsp70 are involved in the anti-and pro-apoptotic processes in D. melanogaster, specifically the Hsp60 family of chaperones, which have anti-and pro-apoptotic properties in the cytosol and organelles, predominantly in the mitochondria. In larvae, overexpression of this family of chaperones is located only in the Malpighian tubules and is induced by cell stress [40].
On the other hand, the Hsp70 stress-inducible protein mainly acts as an anti-apoptotic protein that inhibits apoptosis induced by cytotoxicity through chaperone-dependent, as well as independent, activities. However, the activity of the apoptotic caspase-activated DNase (CAD) is necessary [36]. Therefore, our ROS quantification results could provide evidence of oxidative cellular stress caused by ZEN metabolism, which, in turn, could induce apoptosis, especially in the flare strain, which lacks efficient detoxification Phase I and II XM, compared to the Oregon R(R)-flare strain [25,41]. We believe this assumption could be associated with the alteration of cell division in the imaginal disc cells, observed in the SMART wing test, as evidenced by the mwh clone size cytotoxic results. To further investigate the cellular processes involved in the metabolism induced by ZEN treatment, we analyzed the levels of some apoptosis biomarkers. In Drosophila, genes reaper, grim and hid (located in H99 locus) code for pro-apoptotic proteins, and their ectopic expression can induce apoptosis [42].
The relative expression levels of grim and reaper were increased in the flare strain by 2.96 and 5.49, respectively, indicating that ZEN, or its metabolites, induced the expression of these pro-apoptotic genes ( Figure 6B). It has been reported that the transcription of reaper, grim and hid can be activated in response to many different signals, including steroid hormones, developmental signals, radiation, and various forms of cell stress or injury [43]. Therefore, our data support the proposal that cellular stress is induced by ZEN and/or its metabolites. Additionally, the increase in Dmp53 expression levels and apoptosis transcripts in both strains was in agreement with the oxidative stress data presented, indicating the induction of imbalance in oxidative radicals and the triggering of DNA repair or cell cycle death by ZEN and/or its metabolites =. Therefore, we propose that the toxicity of ZEN in this study was due to its oxidative damage, since Drosophila does not have an estrogen receptor [11].
In summary, the results of our study indicate that ZEN does not exhibit genotoxic activity in the Drosophila wing spot test, but it does affect cell division, as evidenced by the distribution of mwh clones in the flare strain, as well as emergency and fecundity in both strains. This effect cannot be attributed to ZEN or its metabolites interacting with the estrogen receptor, as Drosophila lacks a homologous receptor. Instead, other mechanisms not yet reported must be involved. Our study also showed that ZEN exposure increased the expression levels of anti-apoptotic and pro-apoptotic hsp60 in both strains and modulated the cell pathways of apoptosis gene transcripts (reaper in the flare strain and grim in both strains), suggesting an increase in ROS and supporting cytotoxic activity. Additionally, we demonstrated that ZEN exposure induced the expression of Cyp6a2 and Cyp6g1, indicating that these enzymes are involved in ZEN metabolism in this model organism and may explain the cytotoxicity, oxidative imbalance, and development and fecundity alterations observed in our study.

Survival Assays (LC 50 )
A Lethal media concentration assay (LC 50 ) assay was conducted for ZEN in Drosophila melanogaster flare and Oregon R(R)-flare strains, following the protocol described by Castañeda et al. [46]. The flies were reared in 250 mL flasks with a culture medium at 25 • C and 65% relative humidity [45]. Eggs were collected for 8 h at 25 • C, 65% RH, and in the dark, in flasks containing fermenting fresh baker's yeast supplemented with sucrose and water. The eggs were then incubated under the same conditions. Three days later, the third instar larvae (72 ± 4 h) were washed out of the bottles with tap water (25 • C) through a fine-meshed stainless-steel strainer. Subsequently, 10 larvae were transferred to vials containing 0.5 g of Drosophila Instant Medium (DIM), supplemented with 2 mL of ZEN solutions [100, 200, 300, and 500 µM] in potassium phosphate monobasic buffer 150 mM, pH 7 (PBS). The treatment vials were kept at 25 • C, 65% RH, and were conducted in the dark, to complete their development. The surviving adult flies were counted and collected from the vials 10 to 12 days after the eggs were collected. The LC50 data obtained from four independent experiments (with three replicates per treatment) were analyzed using a two-way ANOVA at a significance level of p < 0.05 (SPSS software vs. 2, IBM, New York, NY, USA) (Supplementary Information). Since a 100% mortality rate was not reached, ZEN was also tested at 260 µM, a sublethal concentration that corresponded to a 20-25% mortality rate in both strains.

Drosophila Wing Spot Test
Virgin females from the flare and Oregon R(R)-flare strains were mated with males from the multiple wing hair strain to perform standard (ST) and high bioactivation (HB) crosses, respectively [47]. Eggs were collected using the method described earlier, and third instar larvae (72 ± 4 h) were transferred to vials containing 0.5 g of DIM added with 2.0 mL of negative control (milliQ water), dissolvent control (PBS), pro-mutagen bioactivation positive control (urethane, 20 mM), or ZEN dissolved in PBS (100, 200 and 400 µM) as treatments. The vials were cultured under dark conditions at 25 • C and 65% relative humidity until imago emergence. Surviving flies were collected and stored in 70% EtOH.
To assess the mutation frequency, wild type wings (trans-heterozygous, mwh +/+ flr3) of flies from the two crosses were mounted on microscope slides using Entellan ® solution, and scored for unbiased observation under a microscope at 400× magnification, where areas with morphological mutations were visible as clone spots. The frequency of each type of spot (small and large single spots or twin spots) and the total spot frequency per treatment [21] were compared pairwise using unpublished SMART computer software [48], based on the Kastenbaum-Bowman test (p < 0.05) [49] and the Mann-Whitney U test, using the STAT Graphics version 6.0 software (Statgraphics Technologies, The Plains, Virginia, USA) (p < 0.05).
Furthermore, the Kolmogorov-Smirnov test was performed to statistically analyze the accumulated mwh clone size distribution in each treatment against the corresponding control (p < 0.05). This test indicated significant alteration of the mitotic division on imaginal wing cells, and, therefore, indicated cytotoxic effects [21,22].

High Performance Liquid Cromatography (HPLC) Analysis
An HPLC analysis was performed in three independent assays to detect and quantify the remaining ZEN in imagos obtained from flare and Oregon R(R)-flare strains that were treated with ZEN. For this purpose, third instar larvae (72 ± 4 h) from each strain were fed with medium supplemented with ZEN (260 µM), or not so supplemented, for approximately 48 h, and 15 imagos were ground in methanol (500 µL). The samples were centrifuged to remove cell debris, and the supernatant was collected for analysis. The HPLC was carried out in accordance with the method described by Visconti and Pascale [50], with an Agilent 1100 system (Hewlett Packard) equipped with the following: a manual injector (Agilent Technologies) with a 20 µL loop, a degasser (Agilent), a pump (Agilent), and a fluorescence detector LS50B (Perkin Elmer). The chromatographic conditions were as follows: the mobile phase consisted of a mixture of methanol: acetic acid 1% 62:38 (v/v), the flow rate was 1 mL/min, the Supelcosil LC-18 column (250 mm × 4.6 mm) was kept at room temperature, and the fluorescence detector was set at excitation and emission wavelengths of 280 and 460 nm, respectively. The detection limit of the method was 8.0902 ng and the limit of quantification was 22.608 ng.

Emergence
Third instar larvae (72 ± 4 h) from the flare and Oregon R(R)-flare strains were transferred separately to vials containing 0.5 g of DIM with 2 mL of PBS as a dissolvent control (10 organisms per vial and five vials per treatment). Another group of organisms was placed in vials containing 0.5 g of DIM with 2 mL of toluene (TOL, 50 mM) dissolved in PBS, serving as a positive control [51]. Finally, a third group of larvae was placed in vials with 0.5 g of DIM added to 2.0 mL of ZEN (260 µM) dissolved in PBS. All larvae were maintained in their respective treatments for~48 h and then transferred to vials containing 0.5 g of DIM added with 2 mL of milliQ water. These vials were monitored daily, and the imagos were collected and counted until all emerged. The total number of imagos was considered to be the corresponding emergence for each treatment. The data obtained from these experiments were analyzed with a two-way ANOVA at a significance level of p < 0.05, using the SPSS software version 2 (IBM, New York, NY, USA).

Fecundity
Fecundity was assessed by maintaining the surviving imagos from each strain and treating the previous emergence assay in vials containing 0.5 g DIM added with 2.0 mL milliQ water. Virgin female flies were selected, and ten individual pairs of females and males were placed in empty vials for 5-6 h. Each pair was then transferred to a modified vial with an upside-down tap, filled with pulverized DIM 0.5 g added to 2.0 mL milliQ water and food colorant. The number of eggs laid by each female was recorded daily for ten days, and the percentage of eggs laid per female was considered to be fecundity. The data obtained from these experiments were analyzed using a two-way ANOVA with a significance level of p < 0.05 (SPSS software version 2).

Total RNA Isolation and cDNA Synthesis
Batches of third instar larvae (72 ± 4 h) from the flare and Oregon R(R)-flare strains were fed for approximately 48 h with a mixture of 0.5 g of DIM and 2.0 mL of ZEN 260 µM dissolved in PBS. The larvae were immediately immersed in liquid nitrogen and transferred to an ultra-low freezer at −70 • C for future molecular assays. Total RNA was extracted using TRIzol reagent, chloroform, isopropanol, and ethanol, following the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA). The extracted RNA was diluted in 50 µL of RNase-free water and treated with DNase (Turbo DNA-free, Invitrogen) to ensure its integrity. The RNA was quantified in a spectrophotometer (Nano Drop 2000 UV-Vis, Thermo Scientific, San Diego, CA, USA) and stored at −70 • C for future use. The cDNA was synthesized from 2 µg of total RNA using reverse transcriptase reaction with oligo (dT)20 primer and a Revert