Leibovitz’s L-15 and Ham’s F-12 media were obtained from Vitrocell Embriolife (Campinas, SP, Brazil). Dulbecco’s Modified Eagle’s Medium (DMEM) with high glucose content was obtained from Inlab Diagnóstica (Diadema, SP, Brazil). Defibrinated sheep blood was purchased from Laborclin (Pinhais, PR, Brazil). Trypsin/EDTA solution, Fetal Bovine Serum (FBS), and 3(4,5-dimethyl)-2 bromide-5 diphenyl tetrazolium (MTT) were obtained from Gibco (Gibco, Carlsbad, CA, United States). All other reagents and solvents used were of analytical or pharmaceutical grade. The IMZT sample was kindly provided by the Brazilian Federal Police (PF), after a seizure at the Brazil/Uruguay border, in the city of Jaguarão (Brazil), according to Berneira et al. [15] .

90 mg of the agrochemical formulation was extracted with 25 mL of acetonitrile. The material was centrifuged for 5 min, and 1 µL of the solvent fraction was injected into a GC-MS QP2010SE (Shimadzu^®) in split mode (1:25). Helium was used as the mobile phase and the Rtx-5MS was used as the capillary column. The temperature was set at 260˚C for the injector and 280˚C for the ion source. The initial temperature of the column was 200˚C for 12 min, with a 5˚C/min heat ramp until 280˚C, then holding for 5 min. MS ran in scan mode from 50 to 550 m/z [15] .

The thermal stability of the IMZT herbicide formulation was determined using the TGA Q5000 equipment (TA Instruments Inc., USA). For this analysis, the heating rate was 10˚C/min using an inert N₂ atmosphere (25 mL/min). The equipment was calibrated with CaC₂O₄·H₂O (99.9%). The mass of the sample was 4.473 mg. Data were processed using TA Universal Analysis 2000 Software, version 4.5 (TA Instruments Inc., USA).

Modulated Temperature Differential Scanning Calorimetry (MT-DSC) was performed on DSC Q2000 equipment (TA Instruments, USA) with MTDSC option, with RCS cooling and N₂ as purge gas (50 mL/min), which was used to determine DSC. The heating rate was 5˚C/min. The instrument was initially calibrated in the DSC standard form, using indium (99.99%). The sample mass (11.384 mg) was weighed with an accuracy of (±0.001 mg). Data were processed using TA Universal Analysis 2000 Software, version 4.5 (TA Instruments Inc., USA).

The cytotoxicity of the herbicide was initially evaluated in sheep erythrocytes and then using ZF-L cells.

The hemolytic activity (HA) induced by IMZT was evaluated spectrophotometrically in a hemoglobin release assay, using defibrinated sheep’s blood (Laborclin^®), as described by Cerveira et al. [29] . Briefly, we used a suspension of 4% (v/v) of red blood cells in 0.9% NaCl. This solution was then incubated with IMZT at dilutions of 4.82, 9.64, 19.29, 38.58, 77.16, 154.32, and 308.64 µg/mL at 37˚C for 1 h and centrifuged for 10 min at 2500 rpm. The supernatants were transferred to a 96-well plate and the absorbance was read at 419 nm. For negative and positive control, saline solution, and Triton-X 100 (0.1%) were used, respectively. The percentage of hemolysis was calculated as (AT − AC)/(AX − AC) × 100, where AT is the absorbance of the treated supernatant, AC is the absorbance of the negative control and AX is the absorbance of the positive control.

The Zebrafish-Liver (ZF-L) cell line was acquired from the Rio de Janeiro Cell Bank (BCRJ – Brazil). Cells were grown in culture flasks and maintained in 50% L-15 (Vitrocell^®, Brazil), 35% DMEM high glucose (INLAB^®, Brazil), 15% Ham’s F-12 (Vitrocell^®, Brazil), supplemented with 0.15 g/L NaHCO₃, 15 mM HEPES, 0.01 mg/mL insulin, 7% (v/v) heat-inactivated FBS (Gibco^®, Brazil), 50 µg/mL enrofloxacin and 2.5 µg/mL amphotericin B. The cells were cultured and maintained in a dry oven at 28˚C.

For cytotoxicity and oxidative stress assays, ZF-L cells were seeded in 96-well plates at a density of 3 × 10⁴ cells/well and in 6-well plates at a density of 6 × 10⁵ cells/well, respectively. The plates were incubated for 48 hours for complete cell adhesion. The formulation containing the IMZT herbicide was diluted in a complete medium without the addition of FBS immediately before each experiment. For cytotoxicity assays, concentrations were defined based on the results determined in the IC₅₀ of the HA, and they were set at 0.44, 0.88, 1.75, 3.51, 7.01, 14.03, and 28.06 µg/mL. To determine oxidative stress in ZF-L cells, concentrations of 1/2×, 1×, and 2× corresponding to the median IC₅₀ of the cytotoxicity assays were used. Triton X-100 at 5% was used as the positive control (PC) for the LDH assay, and H₂O₂ at 0.5% for MTT, NR assays, and oxidative stress tests. Cells treated only with culture medium were used as the negative control (NC) for all assays. The treated cells were incubated under the same conditions for 24 hours.

Mitochondrial viability was determined according to the method described by Mosmann [30] using MTT. This assay is based on the ability of mitochondria to reduce MTT (yellow) to blue formazan crystals. After the cell incubation period at the tested concentrations, the supernatant was removed, and the cells were washed with PBS. The MTT solution (1 mg/mL) was added 50 μL/well and incubated in an oven at 28˚C, for 3 h. The medium was removed, and 100 μL of dimethyl sulfoxide (DMSO) was added to dissolve the formazan crystals. The absorbance was recorded at 540 nm using a microplate reader (SpectraMax M3). The percentage of viability was calculated as AT/AC × 100; where AT and AC are the absorbances of the treated and control cells, respectively.

Lysosome integrity was measured using the neutral red (NR) (3-amino-7-dimethylamino-2-methylphenazine hydrochloride) uptake assay according to the method described by Borenfreund and Puermer [31] , with modifications. This assay evaluates the ability of the lysosome of viable cells to incorporate and retain neutral red. After the cell incubation period at the tested concentrations, the cells were washed with PBS, and 200 µL/well of neutral red solution (40 µg/mL) was added and incubated in an oven at 28˚C for 3 h. After the incubation period, the cells were washed twice with PBS again to remove the dye that was not incorporated into the lysosomes. Then, we added 150 μL/well of the acid-alcohol solution (50% ethanol, 49% distilled water, and 1% glacial acetic acid). Absorbance at 540 nm was determined using a microplate reader (SpectraMax M3). The percentage of viability was calculated using the formula AT/AC × 100; where AT and AC are the absorbances of the treated and control cells, respectively.

The enzyme release assay induced by the action of IMZT on the plasma membrane of ZF-L cells was determined using the lactate dehydrogenase (LDH) assay, as described by Vaucher et al. [32] with modifications. After the incubation period of the cells at the tested concentrations, the supernatants were collected, and the release of LDH. This test was performed through an enzymatic kinetics assay using a commercially available LDH (UV) kit (Bioclin^®– Quibasa Ltda, Belo Horizonte, MG, Brazil). Absorbance at 340 nm was determined using a Cobas MIRA^® automated analyzer (Roche Diagnostics, Basel, Switzerland), following the manufacturer’s instructions. The percentage of LDH release was calculated following the equation: (AT − AC)/(AX − AC) × 100, where AT is the absorbance of treated cells, AC is the control absorbance of untreated cells and AX is the absorbance of cells lysed with Triton X-100.

After the incubation period of the cells at concentrations of 1/2× (1.42 µg/mL), 1× (2.84 µg/mL) and 2× (5.68 µg/mL) corresponding to the median IC₅₀ value, the cells were washed twice with PBS. The lysate was prepared using a mechanical scraper. We then centrifuged at 1000 rpm for 10 min at 4˚C. The pellet was discarded, and the supernatant was used to evaluate oxidative stress parameters (SOD, CAT and –SH). Protein levels were measured using bovine serum albumin as a standard, as described by Lowry et al. [33] .

Intracellular ROS generation in intact cells is based on the oxidation of dichlorohydrofluorescein 2’-7’-diacetate (DCFH-DA) to fluorescent 2’,7’-dichlorofluorescein (DCF) and was determined by the DCFH assay as described by Ali et al. [34] . Briefly, after the incubation period of cells at concentrations of 1/2× (1.42 µg/mL), 1× (2.84 µg/mL), and 2× (5.68 µg/mL) corresponding to the median IC₅₀ value, in 96-well plates at a density of 3 × 10⁴ cells/well, cells were incubated with 1 μM DCFH-DA for 30 min. Fluorescence was measured with excitation/emission at 488/525 nm in a microplate reader (SpectraMax M3). ROS production was expressed as a percentage of control (NC).

Total sulfhydryl content in cell lysates was determined according to Aksenov and Markesbery [35] . This test is based on the reduction of 5,5’-dithio-bis(2-nitrobenzoic acid) (DTNB) by thiols. This reaction forms an oxidized disulfide generating a yellow derivative (TNB). The reaction was initiated by the addition of 5,5’-dithiobis (2-nitrobenzoic acid) (DTNB). Absorbance was measured at 412 nm in a microplate reader (SpectraMax M3) and results were expressed as a percentage of control (NC).

SOD activity in cell lysates was assessed as described by Misra and Fridovich [36] . Catalase (10 μM), glycine buffer (50 mM, pH 10.2), and adrenaline (60 mM) were added to the samples. Absorbance was measured at 480 nm in a microplate reader (SpectraMax M3) and the results were expressed as a percentage of the control (NC).

CAT activity in cell lysates was quantified by the method described by Aebi [37] . This process is based on the decomposition of 30 mM hydrogen peroxide (H₂O₂) in 50 mM potassium phosphate buffer (pH 7.0). The reaction was continuously monitored at 240 nm for 180 s at 37˚C. Reading was performed on a microplate reader (SpectraMax M3) and the results were expressed as a percentage of control (NC).

Data analysis was performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA), for the HA, MTT, NR, and LDH assays. Data were expressed as the mean ± standard deviation for triplicates using one-way analysis of variance (ANOVA) followed by Dunnett’s post-hoc test. IC₅₀ was calculated by non-linear regression analysis for HA, MTT, NR, and LDH assays. For oxidative stress parameters, data were subjected to one-way ANOVA, followed by Tukey’s post-hoc test for multiple comparisons. Differences between mean values were considered significant when p < 0.05.

The herbicide IMZT was identified through GC-MS, and its chromatogram is shown below in Figure 1 . The peak obtained for IMZT in the chromatogram was confirmed by its mass spectra (data not shown). All the other peaks are components of the formulation.

The results of thermogravimetric analysis (TGA) and the derivative curve (DTG), as well as the differential scanning calorimetry (DSC) of IMZT, are presented in Figure 2 .

In the TGA and DTG analysis, the peaks indicate the thermal events occurring in the TGA curve, confirmed by the DTG curve of mass loss concerning temperature. As observed in Figure 2(a) , this herbicide exhibited four mass loss events. The first event involved a mass loss of 52.57% at a midpoint temperature of 58.90˚C. The second event showed a loss of 21.82% at a midpoint of 189.12˚C. The third event had a decrease of approximately 4% at a midpoint of 284.10˚C. The fourth event involved a loss of 11.24% at a midpoint temperature of 348.08˚C. Finally, after 400˚C, no appreciable mass loss was observed. These mass losses could be attributed to thermal decomposition reactions such as decarboxylation and the bond breakage between the two heterocycles in the IMZT molecule.

The result of the DSC of herbicide is presented in Figure 2(b) . Physical-chemical transitions of the sample were observed, as indicated by energy variations concerning temperature. Three events occurred: the first, at a midpoint temperature of 101.31˚C with an enthalpy variation of 1317 J/g at 75.46˚C; the second at a midpoint temperature of 188.85˚C, with an enthalpy variation of 64.90 J/g at 161.92˚C; and finally, an enthalpy variation of 2.943 J/g at 254.79˚C occurred at a peak temperature of 258.87˚C ( Figure 2(b) ).

The effect of the IMZT herbicide formulation on hemolytic activity is illustrated in Figure 3 . The activity was determined as the percentage of lysis of defibrinated sheep blood erythrocytes. According to the data obtained in this study, it was observed that IMZT demonstrated the ability to induce hemolysis above 50% at concentrations of 19.29 to 308.64 µg/mL (70.44%, 99.23%, 99.74%, 100.18%, and 100.51%, respectively). At concentrations of 38.58 and 308.64 µg/mL, hemolysis similar to PC (100%) was observed. The IC₅₀ in this assay was 12.75 µg/mL.

Evaluation of cytotoxic activity on mitochondria was performed using the MTT assay. The values are presented in Figure 4(a) . ZF-L cells exposed to IMZT exhibited reduced mitochondrial activity after 24 h of exposure at all concentrations used (0.44 - 28.06 µg/mL) in 28.52%, 49.45%, 52.33%, 53.03%, 56.78%, 58.79%, and 60.16% when compared to the control (NC). The obtained IC₅₀ was 3.01 µg/mL.

Results of the evaluation of cytotoxic activity on lysosomes are expressed in Figure 4(b) . Treatment with IMZT reduced the lysosomal activity of ZF-L cells after 24 h of exposure at concentrations of 0.88 to 28.06 µg/mL in 18.77%, 60.45%, 67.29%, 69.80%, 72.56%, and 75.66% respectively, when compared to the control (NC). The obtained IC₅₀ was 2.67 µg/mL

LDH release assay was performed using the supernatant medium of cells exposed to IMZT for 24 hours. It was observed that the herbicide affected the membrane integrity, and cellular leakage a significant increase at concentrations of 3.51 to 28.06 µg/mL in 57.36%, 83.81%, 97.81%, and 101.31% respectively. At concentrations of 14.03 and 28.06 µg/mL, cytotoxicity similar to PC (100%) was observed. The IC₅₀ was 1.61 µg/mL ( Figure 4(c) ).

The global IC₅₀ values for the various assays are presented in Table 1 . The median IC₅₀ for the herbicide was calculated, obtaining a value of 2.84 ± 0.17 µg/mL.

⁽¹⁾The half-maximal inhibitory concentration.

The changes in oxidative stress parameters in ZF-L cells produced after exposure to the formulation containing the IMZT herbicide for 24 hours were evaluated and are described in Figure 5 . Three different concentrations were tested, corresponding to the median IC₅₀ values in Table 1 . The concentrations used were: 1/2× (1.42 µg/mL), 1× (2.84 µg/mL), and 2× (5.68 µg/mL).

In this study, we observed an increase in ROS production in ZF-L cells induced by IMZT at all evaluated concentrations. When compared to NC, percentage values of 128.22%, 292.47%, and 769.43% were observed, respectively ( Figure 5(a) ). Additionally, was observed a 61.82% reduction in antioxidant activity non-enzymatic, as determined by the total content of sulfhydryls ( Figure 5(b) ), at the 2× concentration (5.68 µg/mL) when compared to NC.

Interestingly, regarding the enzymatic activity of antioxidants (SOD and CAT), it was noted that IMZT reduced the activity of these enzymes at all tested concentrations compared to NC ( Figure 5(c) and Figure 5(d) ). Specifically, SOD activity exhibited reductions of 28.11%, 48.32%, and 64.11%, while CAT activity exhibited reductions of 71.44%, 74.04%, and 91.06%, respectively, at the tested concentrations of 1/2× (1.42 µg/mL), 1× (2.84 µg/mL), and 2× (5.68 µg/mL).