2.1. Area of study
The Pirajibú River is the primary tributary of the right bank of the Sorocaba River and is part of the Sorocaba and Médio Tietê hydrographic basin. Its source is near the border between the municipalities of Sorocaba and Alumínio (São Paulo, Brazil), at an altitude close to 1,000 meters. The river spans approximately 43 kilometers and flows into the Sorocaba River at an altitude of 540 meters downstream of the industrial zone of Sorocaba. Covering an area of approximately 41,400 hectares, the watercourses within this basin extend for about 540 kilometers (Manfredini et al. 2019).
The predominant land use in this basin is residential and industrial, although agricultural areas and regions with exposed soil are also present. Vegetation in the region is sparse, with fragments concentrated near the less anthropized source region. As the Pirajibú River approaches its mouth, the density of vegetation decreases, with the remaining fragments consisting mainly of riparian forests surrounding the sources and watercourses (Canabarro et al. 2008).
2.2. Animal acquisition and acclimatization in the laboratory
Bullfrog tadpoles (N = 90), A. catesbeiana (2.09 ± 0.20 g) (Shaw 1802), at stage 25 of development (Gosner 1960), were sourced from the Santa Rosa frog farm in Santa Bárbara d'Oeste/SP, Brazil (22° 46′53.0″S, 47°24′17.7″W). The animals were transported in 80-liter plastic bags and promptly conveyed to the laboratory. The tadpoles underwent a nine-day acclimatization period in a 50-liter aquarium filled with dechlorinated supply water. The animals were subjected to a 12-hour light and 12-hour dark cycle and fed once a day ad libitum with commercial trout feed, containing 35% to 40% crude protein.
During the acclimatization phase, the aquarium water was maintained at controlled parameters as follows: temperature 25 ± 1ºC, constant aeration (dissolved oxygen levels between 7.0 mg.L-1 and 7.5 mg.L‑1), water hardness (CaCO3)between 50 and 58 mg.L-1, pH between 7.2 and 7.6, conductivity between 56 and 97 ± 0.02 µS.cm-1, and ammonia concentration lower than 1 mg.L-1.
2.3. Water sampling and analyses
Surface water samples were collected in September 2020 from the Pirajibú River at two distinct points, namely upstream (23°30'53.43"S 47°18'35.34"W) and downstream (23°24'57.23"S 47°26'18.75"W) of the industrial zone (Figure 1). The sampling procedures followed the recommendations outlined in the National Guide for Sample Collection and Preservation – Water, Sediment, Aquatic Communities, and Liquid Effluents (CETESB 2011). The samples were preserved for water characterization by freezing at -22ºC until the tests were conducted. However, the water used during exposure experiments was collected and immediately used.
All water samples were evaluated for concentrations of Aluminum (Al), Barium (Ba), Cadmium (Cd), Cobalt (Co), Chromium (Cr), Copper (Cu), Potassium (K), Manganese (Mn), Molybdenum (Mo), Nickel (Ni), Strontium (Sr), and Zinc (Zn). Psychochemical parameters, including pH, Nitrogen dioxide (NO2), Nitrate (NO3), Ammonium ion (NH4+), Phosphate (PO4), total dissolved phosphorus (TDP), and Chemical Oxygen Demand (COD) were also assessed.
The samples underwent acidification to determine metal concentrations using 65% nitric acid (HNO3) diluted at 1:3 (v/v) until a pH lower than 2.0 was achieved. Following acidification, the samples were stored at 4°C ± 2°C until further use. Acid digestion was done by mixing 100 mL of the water sample at room temperature with 5 mL of concentrated nitric acid (65% w/w). The mixture was heated on a hotplate at 120°C until the volume was reduced to less than 10 mL. After cooling, the samples were transferred to 25 mL flasks for subsequent metal quantification through microwave-induced plasma atomic emission spectrometry (MP–AES) using Agilent Model MP-AES-4200 spectrophotometer (Marzola et al. 2019; APHA 2005).
In order to analyze the concentration of ammonium ion (NH4+), water samples were frozen at -22°C and subsequently analyzed according to the procedure described by Koroleff (1976). Other parameters such as NO2, NO3, PO4, total dissolved phosphorus (TDP), and COD were analyzed according to the methodology outlined by the American Public Health Association (APHA 2005) as follows: method 5220 D was utilized for COD analysis, method 4500 NO2 B for nitrogen dioxide, method 4500 NO3 B for nitrate, method 4500 P E for phosphate, and methods 4500 P B and 4500 P E for the analysis of total dissolved phosphorus.
2.4. Exposure of animals to river water
The bullfrog tadpoles were exposed to river water for 96 hours, following the procedure recommended by the American Society for Testing and Materials – ASTM (2000) for acute toxicity tests in fish, macroinvertebrates, and amphibians. The animals were divided into three groups: a) the Control group, exposed to dechlorinated tap water; b) the Upstream group, exposed to water sampled from upstream of the industrial zone; and c) the Downstream group, exposed to water sampled from downstream of the industrial zone. Experimental groups were tested in duplicate, each group comprising 15 animals exposed in a static system.
The aquariums housing the specimens had a capacity of 16 liters, and the water parameters were the same as the acclimatization conditions. Throughout the exposure experiments, the behavior of the animals was regularly observed. After 96 hours, the animals were anesthetized with 0.01% benzocaine and euthanized by sectioning the medulla, following the guidelines of the American Veterinary Medical Association (AVMA 2001). Subsequently, the livers of the animals were promptly removed and frozen at -80°C for biochemical analyses.
The development of this study received approval from the Committee of Ethics in Animal Use of the Federal University of São Carlos (CEUA – UFSCar) under protocol number 2775051219/2019.
2.5. Biomarkers and biochemical analyses
The ecotoxicological effects on bullfrog tadpoles were evaluated by analyzing biochemical biomarkers in A. catesbeiana (Shaw 1802). Tissue samples from the liver were individually homogenized with a Potter-Elvehjem Teflon tissue homogenizer in three volumes of 100 mM Tris-HCl buffer (pH 7.4) containing 1 mM dithiothreitol (DTT) and ethylenediaminetetraacetic (EDTA) acid 1 mM at 4°C. All biochemical analyses were performed in a microplate reader – Synergy HTX Multi-Mode Reader spectrofluorometer, BioTek. The homogenate was centrifuged at 10,000 g in a HERMLE Z 323 K centrifuge for 30 minutes at 4°C, and the supernatant was fractionated into aliquots as necessary.
The reduced glutathione (GSH) was determined according to the methodology described by Beutler (1963) using Elmann's reagent (5,5'-dithio di-2-nitrobenzoic acid). The results were expressed in nmol. Glutathione peroxidase (GPx) activity was determined according to the method of Flohé and Günzler (1984) by oxidation of reduced β-Nicotinamide adenine dinucleotide phosphate (NADPH) catalyzed by glutathione reductase at 340 nm. Enzyme activity was expressed as µmol of NADPH used per minute per mg of protein, using the NADPH molar extinction coefficient of 6.22 µmol.cm-1.
Glutathione S-transferase (GST) activity was determined according to the method described by Keen et al. (1976) using 2 mM 1-chloro-2,4-dinitrobenzene as substrate at 340 nm. Enzyme activity was expressed in nmol thioether formed per minute per mg of protein. Superoxide dismutase (SOD) activity was determined by the method of Mccord and Fridovich (1969). The reading was performed at a wavelength of 550 nm, and the SOD values were expressed in units per mg of protein. Catalase (CAT) activity was determined by the method of Aebi (1974) by the decomposition of hydrogen peroxide (H2O2) at 240 nm. The results were expressed as μmol of decomposed H2O2 per mg of protein.
Total proteins were determined according to a standard curve of bovine serum albumin (Bradford 1976) in a microplate at 595 nm and expressed in mg per mL. Glucose was determined using the Labtest Kit, Liquiform nº 133, and triglycerides were determined using the Labtest Liquiform Kit nº 87. The concentration of these metabolites was expressed as mg per mL, and their determinations were based on the equation obtained from a standard curve at 505 nm. Lipoperoxidation (LPO) was determined according to the method described by Jiang (1991) at 560 nm, by the FOX method (ferrous oxidation-xylenol orange). Hydroperoxide cement was used as a standard, and the results were expressed in nmol of hydroperoxide per mg of protein.
Protein carbonyl (PCO) was determined according to the method described by Levine et al. (1994) using 2,4-dinitrophenylhydrazine at 370 nm and expressed in nmol of carbonyls per mg of protein. Metallothionein (MT) levels were determined by the concentration of SH groups according to the methodology described by Viarengo et al. (1997) using Ellman's reagent (DTNB-5,5-dithiobis-2-nitrobenzoic acid) at 412 nm. MT was expressed as nmol-SH (sulfhydryl or thiol group) per mg of protein.
2.6. Integrated Biomarker Response (IBR)
The results of all biomarkers were applied to an Integrated Biomarkers Response (IBR) analysis based on the procedure of Beliaeff and Burgeot (2002), modified by Sanchez et al. (2013). Initially, the individual mean of each biomarker (Xi) was divided by the mean of the respective control (Xo), followed by transformation to log [Yi=log (Xi/Xo)]. The mean (µ) and standard deviation (s) of Yi were calculated, followed by the normalization of Yi [Zi=(Yi-µ)/s]. The standardized response of the biomarker (Zi) and the standardized response of the control (Zo) are used to define the index deviation of the biomarker (S) [S=Zi-Zo], used to create the star graph. The S value is the response value for each biomarker. S > 0 represents induction, and S < 0 represents inhibition. The sum of each route's S (A) absolute value represents the IBR value.
2.7. Statistical analyses
The concentration of metals within water samples and biomarkers were expressed as mean ± standard deviation. The average values obtained for each parameter (physicochemical parameters and biochemical biomarkers) in each treatment (control, upstream, and downstream) were compared using the t-test or Mann-Whitney test, depending on the data distribution (normality and homogeneity of variance). All tests used SigmaStat 3.5 for Windows (Copyright © 2006, Systat Software, Inc.). Values of p ˂ 0.05 were considered statistically significant.