Genotoxic and Mutagenic Effects of the Antifouling Biocide Chlorothalonil on the Estuarine Fish Micropogonias Furnieri, Desmarest, 1823

Chlorothalonil is a widely used fungicide in agriculture, and as biocide in antifouling paints. Although it causes toxic effects on non-target organisms and can bioaccumulate in sh tissues, little is known about its sublethal effects. Thus, we evaluated the genotoxic and mutagenic effects of chlorothalonil in Micropogonias furnieri, an estuarine and commercially important sh with potential as a test organism for ecotoxicology assays. We showed that chlorothalonil exerts genotoxic (DNA damage) and mutagenic (micronuclei and nuclear abnormalities) in a dose-dependent manner (0.35 and 3.5 μg g -1 ). As the genomic instability may lead to carcinogenesis, our data can assist decision-makers with evidence for banning this compound since any benet to portuary activities and maritime navigation is outweighed by the cost to aquatic ecosystems and to human health.


Introduction
Chlorothalonil (2,4,6,6-tetrachloroisophthalonitrile) (CHLT) is a broad-spectrum non-systemic fungicide from the isophthalonitrile group widely used in agriculture (Wu et al. 2012). Moreover, it has been used as biocide in the manufacture of antifouling paints, increasing its presence in the aquatic environment (Gallo and Tosti 2015). As a consequence, this biocide is found in high concentrations in port regions The antifouling boosters prevent the establishment and growth of aquatic organisms -bacteria, microalgae, and invertebrates -on submerged or semi-submerged surfaces (Yebra et al. 2004). Its usage is increasing since fouling organisms cause damage to the vessel's structure, loss of speed due to hull irregularities, increase in fuel consumption, and longer mooring period (Brito et al. 2014). However, these biocides can affect non-target species from marine and estuarine environments in addition to fouling organisms (Sakkas et al. 2002), such as sh (Caux et al. 2015).
CHLT has a moderated Log K ow of 2.64, low solubility in water (0.6 mg L -1 ) and is extremely susceptible to photodegradation due to a half-life up to four weeks in seawater (Davies 1987) and eight to nine days in estuarine water (Caux et al. 2015). The high toxicity of chlorothalonil is related to the multiple reactive electrophilic centers in the molecule which depletes the levels of glutathione (GSH), a major player in organisms defenses against xenobiotics (Long and Siegel 1975).
Due to its toxic potential, this biocide has been banned from the United Kingdom (Thomas and Brooks 2010), and recently from the European Union and Switzerland due to its carcinogenic potential (Kiefer et al. 2020). In Brazil, ANVISA (Brazilian Health Regulatory Agency) created a model for reevaluating pesticides based on risk criteria for human health, and ranked Chlorothalonil as a potential carcinogen (ANVISA, 2019). However, little is known about the genotoxic and mutagenic potential (i.e., sublethal or chronic effects) of CHLT to non-target aquatic organisms. Such studies can be used for Health Surveillance given that their predictive character as indicator of carcinogenic effects might be useful to de ne which substances and concentrations are allowed to be used for each purpose.
Biomarkers are valuable tools to study the effects of toxicants on organisms. The comet assay, micronucleus test, and nuclear abnormalities are the most used ones to assess the potential risk of chemicals to cause DNA damage and mutation. Fishes are among the most widely used organisms for detecting harmful effects of xenobiotics, even when exposed to low concentrations, due to their high sensitivity and capacity to bioaccumulate (Yancheva et al., 2015). Micropogonias furnieri (whitemouth croaker) is a native estuarine sh that has been used for biomonitoring, since it is abundant and commercially relevant along the Brazilian coast (Marcovecchio, 2004;Amado et al., 2006). Thus, we used M. furnieri as ecotoxicological test organism to assess the potential of genotoxic and/or mutagenic effects associated with exposure to sublethal concentrations of CHLT.

Methodology
Test organism M. furnieri was collected in estuarine streams in the town of Raposa (Maranhão State, Northeast Brazil: 2º25'19.8''S 44º05'29.4''W), using a 30 mm mesh cast-net (SisBio-IBAMA License N. 55187-1). Juvenile specimens (total length < 250 mm; Juras, 1984) were pre-selected in situ (n= 80) and transported in a recipient with local water and constant aeration to the laboratory. At the laboratory, they were measured, weighed (length < 250 mm; weight = 18.3 g ± 4 g), and forwarded to a pre-treatment following Blaylock et al (2005). Brie y, they rst received a freshwater bath for 3-5 minutes and then were acclimatized for 11 days (25 o C ± 1 o C, 12C/12D photoperiod, salinity = 20 g/kg and constant aeration) in 310 L PVC tanks with canisters lters at a density of 1 g L -1 . Fishes were fed ad libitum with shrimp twice a day.
Intraperitoneal injections were used to avoid the generation of toxic biocide residues and to achieve biocide accumulation in tissues. According to Kinkel et al. (2010), intraperitoneal injections are more effective for delivering a de ned quantity of a chemical to each sh based on weight, particularly in experiments related to metabolic studies. The CHLT lowest dose was based on the bioconcentration factor of 18 of the marine sh Galaxias sauratus after 14 days of exposure to 20 μg L -1 of CHLT dissolved in water (Davies 1988). The lowest concentration was considered as the non-lethal one that accumulates in wet tissues; the highest one -one order of magnitude higher -was considered as the sub-lethal. The specimens were weighed and received an intraperitoneal dosage of 1 μL g -1 according to each treatment. After injections, shes were kept for 96 h (without feeding) in the same conditions of acclimatization and then anesthetized with eugenol (100 mg L -1 ). A sample of peripheral blood was taken from the brachial artery (Research ethics committee authorization: CEUA-UFMA N.23115.008075/2016-12) to assess the potential of genotoxic and/or mutagenic effects.

Genotoxic and mutagenic assays
The comet assay was performed according to Cestari  The micronucleus test followed the technique described by Hooftman and Raat (1982). A blood sample was smeared on a slide, xed using methanol for 10 minutes, and left drying at room temperature. Slides were stained with the Panoptic hematological kit (Laborclin, Brazil) for 20 minutes, washed in running water, and dried at room temperature. The frequency of micronuclei in erythrocytes was determined by analyzing 1,000 cells per sample, considering only red blood cells with intact nuclear and cytoplasmic membranes. Micronuclei were classi ed according to Hooftman and Raat (1982) and the nuclear abnormalities (nuclear bud, apoptotic fragments, bilobed cells, and binucleated cells) according to Barsiene et al. (2006) Data analysis The normal distribution and homogeneity of the data sets were evaluated by the Shapiro Wilk and Levene tests, respectively. Since the data sets did not t a normal distribution, we used the Kruskal-Wallis test, followed by the Dunnet test (p<0.05), to evaluate the statistical signi cance of the differences for each biomarker among treatments.

Results
M. furnieri showed to be resistant to manipulation (capture and transportation) and acclimation to laboratory conditions as no mortality or incidence of disease during these periods were observed. The sh was also sensitive to CHLT and cyclophosphamide (positive control).
Both concentrations of CHLT caused DNA damage (p <0.05) in M. furnieri with a 2.8-fold increase in the damage scores (genotoxic effects) compared to the negative control and at a similar magnitude as the positive control (Fig. 1A). The micronucleus frequency (Fig. 1B)  We also observed statistically signi cant effects in the formation of nuclear buds compared to the negative control, with a 2.8-fold increase for the positive control in the lowest concentration (0.35 μg g -1 ) and a 1.9-fold increase in the highest one (3.5 μg g -1 ), but no statistical differences between dosages and the positive control ( Fig. 2A). The positive control and CHLT (both dosages) also induced the formation of apoptotic fragments, causing an increase of 14.5, 6.0 and 13.8-fold higher than the negative control (Fig.  2B). The number of bilobed cells was also signi cantly higher for the positive control and CHLT exposures than the negative control, with an increase of 2.9, 3.0 and 6.9-fold, respectively (Fig. 2C). On the other hand, despite increased frequency of binucleated cells, no statistical (p > 0.05) effects were observed for both the positive control and CHLT treatments (Fig. 2D). For nuclear abnormalities, only the frequencies of apoptotic fragments and bilobed cells were dose-dependent (p < 0.05).

Discussion
M. furnieri is suitable as a test organism species for ecotoxicological assays. Although it has been used as a biomonitor for aquatic pollution ( e.g., Kehrig 2015). Plus, as a native and commercial species, the results can be translated to their wild counterparts.
The biomarkers show that CHLT cause DNA damage (comet assay) and mutation (micronucleus test and nuclear abnormalities, except for binucleated cells) in a dose-dependent manner, sometimes similar to the positive control. These effects may be associated to an increase in oxidative stress -increasing the production of reactive oxygen species (ROS) -and/or to a decrease in the antioxidant defenses, such as glutathione (Pompella et al. 2003 The mechanism of chlorothalonil action resembles reactions involving both low and high molecularweight thiols, and its toxicity resides in inhibiting thiol-dependent enzymes (Tillman et al. 1973). Arvanites and Boerth (2001), working with fungi con rmed that CHTL toxicity was associated with the rapid conjugation of cellular thiols derivatives with CHTL, speci cally with thiol-rich enzymes, such as GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and GSH, depleting cellular GSH reserves. It is known that GSH is an important protein in the cellular defense, being able to act also against toxic xenobiotics such as drugs, pollutants, and carcinogens compounds. It is important to note that exposure to CHLT also causes an increase in enzyme gluthatione stransferase (GST) activity in sh (Davies 1985; Lopes et al. 2019) reducing GSH availability, and decreasing its protective role in the cell's antioxidant defense, especially, in ROS neutralization.
Recent studies with different kinds of biomarkers have been used to identify CHLT potential for toxicity on different kinds of organisms (Table 1)   Decrease: expression of genes related to immunity, reproduction, and xenobiotic clearance.
Zhang et al., Ascidians Deleterious effects on gametes and fertilization; Interference in embryonic development as induction in larval malformation; Teratogenic effects. Gallo and Tosti (2015) Amphibians Increase: numbers of liver granulocytes and melanomacrophages, corticosterone, immune cell levels, and liver damage.

Rainbow trout
Increase: phagocytic leukocytes, respiratory burst, and phagocytic cells.  In addition to be potentially genotoxic and mutagenic, CHLT is also a potential carcinogen. The United States Environmental Protection Agency (US EPA) ranks CHLT as a suspected carcinogen to humans based on studies with mice (US EPA, 1999). In Brazil, the National Health Surveillance Agency created a model for reevaluating pesticides based on risk criteria for human health. In this assessment, Chlorothalonil is ranked as a potential carcinogen for humans (ANVISA, 2019). Other studies have shown clearly the carcinogenic effects of chlorothalonil in sh (Lopes et al., 2019, Garayzar et al., 2016and Gallo and Tosti, 2015. Our results also suggest that CHLT may be involved with carcinogenesis because it causes genomic instability and mutation, which might overwhelm the DNA repair system. Banning CHLT as occurred in the United Kingdon, European Union and Switzerland (Kiefer et al 2020) should be in sight of the Brazilian health authorities. Any bene ts to portuary activities and maritime navigation provided by CHLT-based antifouling paints is outweighed by the risks of its biomagni cation through the food chain -eventually reaching humans -the carcinogenic potential and the damage to aquatic ecosystems.

Conclusion
The estuarine sh M. furnieri has the potential for use as a test organism in ecotoxicological assays. We also show that chlorothalonil causes DNA damage and mutation, probably, by depleting GSH availability, a major player in the cell's antioxidant defense. As the genomic instability and mutation may lead to carcinogenesis, our data can assist decision-makers with evidence for banning this compound from