The methodology that was used in this study was first described by Mello, et al. .
We used 80 Swiss male mice (30-45 g), provided by the Central Vivarium of the Universidade do Oeste Paulista (UNOESTE), that were allocated in cages measuring 30x16x19 cm (5 animals per cage) and maintained in a room with a controlled temperature of 25 ± 2°C, a relative humidity of 50 ± 15%, and a normal photoperiod (12–12 h light-dark cycle).
The animals were randomly divided into the following four groups (n = 20):
- SG, saline group: exposed to nebulization of 10 ml of 0.9% sodium chloride solution.
- LCG (low 2,4-D concentration group): exposed to herbicide mist with 3.71 x 10-3 grams of active ingredient per hectare (g.i.a. / ha) (4.6 microliters of the pesticide was added to saline), corresponding to 187,17 mg/m3 of 2,4-D.
- MCG (middle 2,4-D concentration group): exposed to nebulization of the herbicide with 6.19 x 10-3 g.i.a./ha (7.7 microliters of pesticide was added to saline), corresponding to 313,31 mg/m3 of 2,4-D.
- HCG (high 2,4-D concentration group): exposed to nebulization of the herbicide with 9.28 x 10-3 g.i.a./ha (11.5 microliters of pesticide was added to saline), corresponding to 467,93 mg/m3 of 2,4-D.
The different concentrations of the herbicide 2,4-D were diluted in 10 ml of 0.9% sodium chloride to perform the nebulization. The solutions were prepared at the time of use.
The different concentrations of the 2,4-D herbicide were formulated based on the product label, which shows the different herbicide concentrations for each type of crop to be sprayed, and a dose-adjustment was made to the box area to simulate environmental occupational exposure.
2,4-D herbicide exposure protocol
The mice were exposed to the herbicide dichlorophenoxyacetic acid (Nortox SA, Arapongas, Paraná, Brazil), which had the following composition: (2,4-dichlorophenoxy) acetic acid (2,4-D) dimethylamine salt: 806 g / liter (80.6% w / v), acid equivalent of 2,4-D: 670 g / liter (67.0% w / v) and inert ingredients: 424 g / liter (42.4% w / v). Handling of the 2,4-D herbicide was performed with the following personal protective equipment: rubber gloves, goggles and filter masks for gases.
The nebulization protocol consisted of two boxes (32x24x32 cm), and each box was attached to a Pulmosonic Star® ultrasonic nebulizer (Soniclear Ind. Com. Imp. and Exp. Ltda., São Paulo, Brazil) . The exposure time was approximately 15 minutes.
Five animals from each group were exposed to nebulization at different times and were thus identified:
- 1D: one nebulization;
- 2D: two nebulizations on consecutive days with a 24 hours difference between each exposure;
- 3D: three nebulizations on consecutive days with a 24 hours difference between each exposure.
The animals were euthanized 24 hours after the last nebulization, and five of the animals from the group that was nebulized three times were euthanized eight days (8D) after the last nebulization (Figure 1).
Euthanasia was performed with an intraperitoneal injection of sodium thiopental at a dose of 100 mg / kg body weight . After euthanasia, the entire tongue was removed for the histopathological analysis.
The tongue was longitudinally sectioned and fixed in 10% buffered formalin (Kinetics Chemical Industry, São Paulo, Brazil) for 24 hours, and the samples were submitted for normal histological processing by embedding the tissues in paraffin (Dynamic Analytical Reagents, São Paulo, Brazil). Serial sections of 5 μm were obtained and stained by the hematoxylin-eosin (HE) method (Dolles, São Paulo, Brazil).
The investigator who conducted the histopathological analysis was blinded and evaluated the following types of tongue lesions: congestion (0 = absent, 1 = mild, 2 = moderate, 3 = marked); presence of hyperkeratosis (0 = absent, 1 = mild, 2 = moderate, 3 = marked); presence of parakeratosis (0 = absent, 1 = focal, 2 = diffuse); presence and intensity of polymorphonuclear and / or mononuclear inflammatory infiltrate (0 = absent, 1 = discrete, 2 = moderate, 3 = intense); presence of individual cell necrosis (0 = absent, 1 = present); preneoplastic or dysplastic lesions (0 = absent, 1 = mild, 2 = moderate, 3 = severe dysplasia); and benign neoplastic lesions or malignant neoplastic lesions .
The tongue sections were also stained by Toluidine blue (Merck, Darmstadt, Germany) for the identification of mast cells. The number of mast cells was counted in the lamina propria in 10 high power fields (HPF), corresponding to approximately 1 mm2 .
The measurements of tongue mucosa thickness were performed in two areas each  of the dorsal and ventral surfaces of the tongue in all animals at 200x magnification using ImageJ® software from the National Institutes of Health (NIH), available free of charge on the Internet (http: //rsbweb.nih.gov/ij/).
To count the NORs (nucleoli organizing regions), other deparaffinized sections of the tongue were stained by silver impregnation according to Ploton, et al. . The counting pattern of the NORs was performed considering 10 cells / HPF, repeating in 10 random microscopic fields to total the count in 100 cells per animal . The NORs were separately counted in the dorsal and ventral surfaces of the tongue.
For the qualitative variables, the likelihood ratio was calculated, and a two-way analysis of variance was used to analyze the quantitative variables. The correlation between the quantitative variables was also used. The level of significance was set at 5%, and SPSS V.22 software was used to perform the analyses.