Biochemical and Genotoxic Effects of Iron and Manganese in Oreochromis Niloticus (Teleostei: Cichlidae)

The Doce River, southeastern Brazil, in 2015 received iron mining tailings after the Fundão (MG) dam burst, which resulted in the leakage of about 50 million cubic meters of tailings mud, which have iron (Fe) and manganese (Mn) as main components, causing much damage to health to aquatic organisms, including death. Since exposure of aquatic organisms to metals can cause genotoxic damage and induce the generation of reactive oxygen species, causing oxidative damage to biomolecules, the present study aimed to evaluate the toxicity of the association between Fe and Mn in Oreochromis niloticus through genotoxic (micronucleus test and comet assay), and biochemical (CAT and GST enzymes) assays. The tested treatments were T1 = control group, T2 = 3.81 mg/L of Fe + 0.5 mg/L of Mn, and T3 = 7.62 mg/L of Fe + 5.23 mg/L of Mn, during 96-h bioassays. All animals exposed to the metals showed a signicant increase in erythrocyte micronucleus frequency and DNA damage. The hepatic GST activity increased two times in animals exposed to T3 compared to control group. The results indicate that Fe + Mn caused genotoxic and biochemical changes in exposed sh. Therefore, excess metals in ecosystems, even those essential for organisms, can be dangerous for the local biota due to the risk associated with high concentrations of these metals.


Introduction
In November 2015, the Doce River (southeastern Brazil The major metals that compose iron ore are iron (Fe) and manganese (Mn) (Veronez et al. 2018). The increase in these metal concentrations in aquatic environments can impact the biota, causing imbalances in these ecosystems and associated organisms (Zhang et al. 2018).
Iron and Manganese are essential for living organisms and their metabolic functions. However, at high concentrations, they can become toxic and cause damage. Manganese is involved in several biological processes, such as the metabolism of carbohydrates, lipids, and proteins, and as an enzymatic cofactor (Keen 1984). However, Mn overload can cause damages, such as changes in the immune responses and de ciency in calcium absorption (Hernroth et al. 2004; Gunter et al. 2006). Iron participates in oxygen transport, DNA synthesis, and electron transport associated with cellular respiration (Crichton 1991). Iron overload can damage organs, tissues, and cells, cause histopathological changes and decrease the number of relevant cells to the immune system (Sousa et al. 2020).
The use of biomarkers has great relevance in environmental monitoring. These are molecular, cellular, or systemic markers of great importance to the evaluation of the organisms' response to the effects caused by a pollutant. Metals can also react with genetic material, producing genotoxic damage detected by the micronucleus test and the comet assay, respectively. Analyses like micronucleus test and comet assays are capable of detecting anomalies caused by contaminants in the animals' DNA and of measuring physical and biochemical changes in the blood (Nussey et al. 1995). Among the various biochemical biomarkers studied, there are two very important ones, the enzymes glutathione-S-transferase (GST) and catalase (CAT), in the liver and gills. Glutathione-S-transferase is an enzyme of the phase II of biotransformation metabolism, conjugating xenobiotic to polar molecules (Gao et al. 2020 (Vasylkiv et al. 2011). The gills are extensively studied because are constantly exposed to environmental changes due to respiratory processes and the liver for being the main organ for detoxifying xenobiotic. A good biological model is indicated to effectively evaluate an organism's response to the studied biomarkers. Bioindicator organisms are usually indicated because their condition re ects the environmental conditions. Fishes are good bioindicators, as they are constantly exposed to environmental variations and can metabolize, concentrate, and accumulate pollutants, in addition to being sensitive to biochemical and genotoxic analyses (Collins et al. 2004). Oreochromis niloticus (Linnaeus, 1758) is a sh species belonging to the Cichlidae family, native to the African continent, but widely distributed in reservoirs and rivers of tropical regions, including the Doce River. Furthermore, this species responds promptly to environmental changes caused by contaminants (Almeida et al. 2002).
Therefore, due to the importance of understanding the synergistic effect of these metals, mostly due to the composition of iron ore, the objective of this study was to evaluate the toxicity of associated Fe + Mn using biochemical (CAT and GST enzymes) and genotoxic (micronucleus test and comet assay) biomarkers. Proposing the hypothesis that these metals in an association are toxic to O. niloticus even at low concentrations, causing enzymatic changes and damage to the genetic material.

Acclimatization
Fifty juvenile individuals of O. niloticus (62.6 ± 4.9 g and 15.7 ± 1.33 cm) were acquired from the Aquamais sh farm located in Guarapari, Espírito Santo State, Brazil. They were rapidly transported to the Applied Ichthyology Laboratory at the University of Vila Velha (LabPeixe/UVV). There, they were maintained in 500-L polyethylene tanks with ltered water and constant aeration for four weeks for acclimatization. They were fed daily with a protein-rich (55%) feed.

Experimental design
After the acclimatization period, 24 specimens of O. niloticus (54.16 ± 0.76 g and 15.45 ± 0.64 cm) were individually allocated to three treatments (n = 8 individuals per treatment). The following predetermined After the experimental period, the specimens were sedated with a Benzocaine solution (0.1 g/L), submitted to weight and dimension measurements, and had their blood collected by puncturing the caudal vein to perform the genotoxic assays. They were euthanized by cervical section and the tissues (liver and gills) were stored at -80°C (Ultra Freezer CL 120 − 80 V) until the biochemical analyses.

Genotoxic analyses 2.3.1 Micronucleus Test
The micronucleus test was performed according to Grisolia et al. (2005). After blood was drawn via a caudal puncture, a blood smear was made on microscopic slides. After drying, the slides were xed with methanol and stained with 5% Giemsa for 40 min. The material was observed under a microscope to count the micronucleus in red blood cells (1000 cells per slide), with two slides per individual. The micronucleus identi cation was carried out according to the criteria proposed by Fenech et al. (2003) and the average micronucleus frequency (‰) in each treatment was calculated.

Alkaline Comet Assay
The comet assay was performed under alkaline conditions and stained with silver nitrate according to Tice et al. (2000). The slides were previously coated with 1.5% agarose. The blood samples were diluted in an RPMI solution and mixed with low melting agarose. The slides passed through the electrophoresis phase, in an electrophoretic buffer, followed by a 15-min electrophoretic run at 25 V and 300 mA. After the run, the slides were neutralized with TRIS buffer and placed in a xative solution. Finally, they were hydrated and stained with silver nitrate. The comets' identi cation followed the criteria of Collins (2004) according to the shape and size of the tail (where the size of the tail is proportional to the number of DNA fragments). The cells were observed under an optical microscope at 40-fold magni cation, and 100 cells per slide (two slides for each individual) were counted, classifying comets in classes from 0 (undamaged) to 4 (maximum damage).

Biochemical analyses
The branchial and hepatic tissues were homogenized with phosphate buffer (pH 7.0) and centrifuged (3030 g) for 30 min at 4°C to obtain the supernatants for the biochemical analyses.
Glutathione S-transferase activity (E.C. 2.5.1.18) was determined using phosphate buffer (pH 7.0), 1 mM 1-chloro-2,4-dinitrobenzene (CDNB) and 1 mM glutathione (GSH) as substrate. Its activity was calculated by the absorbance reading obtained in a microplate reader at 340 nm. The absolute activity was estimated using the CDNB extinction coe cient (Habig and Jakoby 1981). The results obtained were expressed in g fresh tissue/min. Catalase activity (E.C. 1.11.1.6) was assessed by observing the continuous decrease in hydrogen peroxide (H 2 O 2 ) concentration according to Aebi (1984). Buffer was used as a reaction medium with 10 mM H 2 O 2 and TE buffer (1 M Tris-HCl and 5 mM EDTA). The samples were read in a spectrophotometer at a wavelength of 240 nm. The results were expressed in µmol H 2 O 2 metabolized/min/g of fresh tissue.

Concentrations of Iron and Manganese in the water
Water samples were collected in each experimental aquarium after contamination (0 h) and at the end of the experimental period (96 h). The samples were acidi ed with 10% of the total volume of the samples with nitric acid (65%) for dissolved Fe and Mn analyses. The samples were then ltered through precleaned, non-sterile 13-mm lters, with 0.45-µm pores (Analytical). They were read on an Atomic Absorption Spectrophotometer (AAS) operating in ame mode (Thermo Fisher Scienti c ICE3500, Waltham, MA, USA). For Fe, the limit of detection (LOD) was 3.12 µg/L, and the limit of quanti cation (LOQ) was 9.47 µg/L. The Mn presented LOD of 5.36 µg/L and LOQ of 16.25 µg/L.

Statistical analyses
Data normality was analyzed by a Shapiro-Wilk test. The catalase enzyme activity data were normalized by log10-transformation. The results for biochemical analyses, comet assay, and water physicochemical parameters were compared between treatments, and their differences detected by One-Way Analysis of Variance (ANOVA), followed by Tukey post-hoc test for multiple comparisons. The micronucleus test was analyzed by Dunnet's post-test. All statistical analyses were performed on the SigmaPlot 12.5 software and statistical signi cance was considered when p < 0.05.

Metal concentration and water physicochemical parameters
At the beginning of the experiment, the actual dissolved Mn concentrations were 0.12 mg/L for T1 (control group), 0.22 mg/L for T2, and 3.49 mg/L for T3. For dissolved Fe, the concentrations were 0.45 mg/L for T1, 2.60 mg/L for T2, and 4.40 mg/L for T3. At the end of the experimental period, Mn concentrations were 0.03 mg/L for T1, 0.07 mg/L for T2, and 1.23 mg/L for T3. For Fe, concentrations were 0.54 mg/L for T1, 2.26 mg/L for T2, and 3.71 mg/L for T3.
Throughout the experimental period, dissolved oxygen, temperature, pH, hardness, alkalinity, nitrite, and ammonia remained constant, ensuring good water quality. Electrical conductivity presented a higher value in T3 when compared to T1 and T2 (Table 1).

Genotoxic analyses
Exposure to the association of Fe + Mn in O. niloticus individuals induced a signi cant increase in the frequency of micronucleated erythrocytes (p ≤ 0.001). There was an increase in micronucleus incidence by 11 times, for T2 compared to the control group, and 20 times for T3 compared to the control group ( Fig. 1). The DNA damage index (DI) of the shes' erythrocytes was high in the two treatments exposed to Fe + Mn. There was an increase of about 8 times for T2, and 22 times for T3, both concerning the control group (p ≤ 0.001). T3 individuals were mainly classi ed as class 3 (severe damage) or 4 (very severe damage) in the comet assays (Fig. 2).

Biochemical analyses
The CAT activity had no signi cant differences in the gills (p = 0.26) (Fig. 3A) or the liver (p = 0.75) (Fig. 3B) of the shes between treatments. On the other hand, the GST activity showed a signi cant increase in the liver samples (p = 0.03) in specimens exposed to the highest concentrations tested (T3) (Fig. 3D). The GST activity in the gills did not present signi cant differences (p = 0.43) (Fig. 3C).

Discussion
In the present study, there was an increase in the hepatic GST activity at the highest concentrations of metals. Furthermore, genotoxic damages were also detected in the two tested concentrations of Fe + Mn (T2 and T3). The exposure to Fe + Mn induced a signi cant increase in the frequency of micronucleated erythrocytes.
The formation of micronuclei in the exposed tilapia re ects structural problems or chromosomal changes during mitosis; therefore, it is possible to identify the genotoxic potential of chemicals such as metals, even those essential to metabolism (Kample et al. 2018). The same result was found in acute exposure to Mn at concentrations of 3.88 and 7.52 mg/L in the sh species C. auratus (Valbona et al. 2018), as well as in a study with exposure to iron oxide (0.3 mg/L) in guppy sh (Poecilia reticulate) (Qualhato et al. 2017). Both studies detected the genotoxic potential of isolated metals, and our research, in this way, has been complementing the effects of these metals together.
We detected high levels of DNA damage, with the formation of class 3 and 4 comets in treatments T2 and T3, which are the highest DNA damage levels that can be found. The T3 shes were the most affected, differing signi cantly from the other groups. Coppo et al. (2018) found that isolated Mn harms the replication of genetic material in O. niloticus. A study on the exposure of a guppy sh to iron oxide identi ed comet formation from short experimental exposures (3 and 7 days) to longer ones (14 and 21 days) (Qualhato et al. 2017). With these extents of damage, consequently, there is interference in the accuracy of the genetic material replication in the sh organism. Therefore, the comet assay is an important biomarker for checking acute DNA changes in the presence of Fe and Mn (Hariri et al. 2020). Both analyzes, micronucleus test and comet assay, proved to be e cient to evaluate the effects of these two metals together in O. niloticus, showing good biomarkers for this purpose.
GST is an enzyme of fundamental importance in protecting organisms from environmental stressors. Its activity may increase or decrease when exposed to metals, depending on the concentration and the period of exposure (Guilherme et al. 2008). The increase in the hepatic GST activity may indicate the onset of the organism's detoxi cation process against the metal, since, this enzyme participates in the biotransformation and conjugation of xenobiotic, and the liver plays an important role in metabolizing contaminants (Landi 2000;Moniruzzaman et al. 2020). Studies corroborate our results, through the detection of changes in aquatic organisms exposed to metals (i.e., Fe and Mn). Valbona et al. (2018), found a signi cant increase in GST activity in specimens of Carassius auratus, as well as Veronez et al. (2018) reported an increase in GST in liver tissues of tadpoles exposed to Fe, Mn, and iron ore. Thus, GST is a good biomarker to assess the degree of impact and the effects caused by Fe + Mn in sh and may contribute to the understanding of the mechanisms of action of these compounds in the face of environmental variation. On the other hand, CAT activity did not change in any treatment for both organs and hence cannot be considered a contamination biomarker for these associated metals in the gills and liver of O. niloticus. Other metabolic routes may have been activated (Pandey et al. 2003).
Even at low concentrations, associated Fe + Mn was potentially dangerous to sh specimens in the present study. Despite the importance of studying the effect of Fe and Mn together, especially due to the composition of iron ore, few studies portray their synergistic effects on sh. The rupture of the iron mining tailings dam in Mariana, Brazil released several toxic elements in the environment, such as Fe and Mn. According to Queiroz et al. (2018), seven days after the Fundão dam burst (in 2015), the concentration of Fe and Mn found in the sediment of the Doce River estuary were 34,900 and 586 mg/kg, respectively. In 2018, three years after the disaster, metal concentrations remained high, with 26,450 mg of Fe/kg and 1075 mg of Mn/kg ). Thus, with resuspension events in the sediment, which are frequent in rivers, the metals associated with the sediment particles can become bioavailable again in the water column, contaminating the biota present in the river (Queiroz et al. 2018). Hence, understanding how these metals work together is extremely necessary. In general, mining activities are very damaging to ecosystems and the biota present, and in an accident, there is a very high risk of altering the food chain, with persistent damage to local biodiversity in the long term (Espindola et al. 2016).

Conclusion
In conclusion, associated Fe + Mn was toxic to the analyzed sh (O. niloticus) even at low concentrations. There was micronucleus formation in erythrocytes and damage in the genetic material, in addition to an increase in hepatic GST activity. These metals constitute the iron ores and remain present at high concentrations in the Doce River after the disaster that occurred with the rupture of the tailings mud dam. Therefore, organisms that are present in ecosystems contaminated by these metals can suffer deleterious damage to their genetic material, cells, and systems.