Science of the Total Environment 497–498 (2014) 293–306

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

The impact of organochlorines and metals on wild fish living in a tropical hydroelectric reservoir: bioaccumulation and histopathological biomarkers Marcelo Gustavo Paulino a, Tayrine Paschoaletti Benze a, Helen Sadauskas-Henrique a, Marise Margareth Sakuragui a, João Batista Fernandes b, Marisa Narciso Fernandes ⁎,a a b

Physiological Sciences Department, Federal University of São Carlos, Rodovia Washington Luiz, km 235, 13565-905 São Carlos, São Paulo, Brazil Chemistry Department, Federal University of São Carlos, Rodovia Washington Luiz, km 235, 13565-905 São Carlos, São Paulo, Brazil

H I G H L I G H T S • • • • •

Multiple contaminants accumulation in wild fish causes histological changes. Metals and organochlorines have similar bioaccumulation pattern in gills and liver. Contaminant bioaccumulation was higher in the liver than in the gills. The histopathological indices in the liver were greater than those of gills. Gills and liver lesions are useful for the discrimination of contaminated field sites.

a r t i c l e

i n f o

Article history: Received 6 April 2014 Received in revised form 30 July 2014 Accepted 31 July 2014 Available online xxxx Editor: D. Barcelo Keywords: Astyanax fasciatus Gills Histopathological index Liver Mucous cells Pimelodus maculatus

a b s t r a c t This study evaluates the contaminants in water and their bioaccumulation in the gills and liver of two ecologically distinct fish species, Astyanax fasciatus and Pimelodus maculatus, living in the reservoir of the Furnas hydroelectric power station located in Minas Gerais in the southeastern Brazil. The histological alterations in these organs are also examined. Water and fish were collected in June and December from five sites (site 1: FU10, site 2: FU20, site 3: FU30, site 4: FU40 and site 5: FU50) in the reservoir, and agrochemicals and metals selected based on their use in the field crops surrounding the reservoir were analyzed in the water and in the fish gills and livers. The concentrations of the organochlorines aldrin/dieldrin, endosulfan and heptachlor/heptachlor epoxide as well as the metals copper, chromium, iron and zinc in the gills and livers of both fish species were higher in June than in December; the liver accumulated higher concentrations of contaminants than the gills. The organochlorine metolachlor was detected only in the liver. The histological pattern of changes was similar in both species with regard to contaminant accumulation in the gills and liver. Fish from FU10, the least contaminated site, exhibited normal gill structure and moderate to heavy liver damage. Fish collected at FU20 to FU50, which were contaminated with organochlorines and metals, showed slight to moderate gill damage in June and irreparable liver damage in the livers in June and December. The histological changes in the gills and liver were suitable to distinguishing contaminated field sites and are therefore useful biomarkers for environmental contamination representing a biological end-point of exposure. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Aquatic ecosystems and wildlife are continuously threatened by anthropogenic activities. In the reservoirs erected to generate hydroelectricity, the aquatic ecosystems are phenotypically intermediate between those of lakes and rivers with specific limnological characteristics due to the intermittent water flow (Straskraba and Tundisi, 1999). ⁎ Corresponding author. Tel.: +55 16 33518742; fax: +55 16 33518401. E-mail address: [email protected] (M.N. Fernandes).

http://dx.doi.org/10.1016/j.scitotenv.2014.07.122 0048-9697/© 2014 Elsevier B.V. All rights reserved.

Although the generation of electricity does not change the water quality, the inadequate use of the soil surrounding the reservoir may contribute to aquatic environmental degradation due to the domestic, industrial and agricultural effluents that are produced as a consequence of an increasing population and increased economic development. The Furnas hydroelectric power station (HPS) reservoir is one of largest in southeastern Brazil and was formed by the damming of two rivers: the Grande River and the Sapucai River. The reservoir has a 1440 km2 overflow area containing 21 million m3 of water with a perimeter of 3500 km, and it is bordered by 34 small to medium-sized

294

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

cities, most of which engage in intense agricultural and cattle farming activities. The limnological, physical and chemical variables of the reservoir water are within the range of the limits recommended by the Brazilian Environment National Council for water biota preservation (CONAMA 357, 2005) and exhibit oligotrophic characteristics (Heleno, 2004). However, the dendritic morphology of the reservoir favors horizontal variation in the water quality, and some regions are highly degraded and exhibit mesotrophic- and eutrophic characteristics (Sá Junior, 1994; Heleno, 2004; Negreiros et al., 2010). The presence of trace metals (aluminum, chromium, copper, iron and zinc) and organochlorines (aldrin/dieldrin, endosulfan, heptachlor/heptachlor epoxide and metolachlor) in the water and sediment of different regions of the reservoir highlights the dissimilarity between the Grande and the Sapucai rivers (Sadauskas-Henrique, 2008; Sadauskas-Henrique et al., 2011). Chronic exposure to multiple contaminants in water, even at low levels, affects the resident biota, including the fish. Contaminant accumulation in the tissues induces changes in the biochemistry and physiology of the cells (Boon et al., 2002; Fernandes et al., 2007). Previous studies of fish collected from the Furnas HPS reservoir reported changes in blood variables and in the induction of oxidative stress in the blood, gills and liver of fish (Sadauskas-Henrique et al., 2011; Sakuragui et al., 2013) as well as inhibition of Na+/K+-ATPase and alterations in the surface architecture of pavement and chloride cells in the gills (Fernandes et al., 2013). Such biochemical and physiological alterations may lead to histological changes in these organs. Histological biomarkers are an intermediate level of biological organization (between the molecular and individual levels), and they are useful tools for integrating the cumulative effects of biochemical and physiological alterations (Myers and Fournie, 2002). Chemical tissue analyses and histological biomarkers permit the identification and evaluation of the impact of contaminants present in aquatic ecosystems in addition to the degree of environmental pollution; thus, they provide important information for the generation of public policies for monitoring and conservation of biota (Van der Oost et al., 2003). In fish, the gills and liver are the main target organs used to evaluate water quality. The gills are the main sites for gas exchange as well as acid–base and ionic regulation and are the first organs to contact the contaminants. The large surface area of the gills that is in contact with water and the very thin diffusion distance between water and blood favor the uptake of contaminant molecules dissolved in water. The contaminant uptake from water into a fish by the gills depends on the movement of the water with dissolved contaminants through the gill lamella and the diffusion of the contaminants across the water and the gill epithelium and from the gill into the blood (McKim and Erickson, 1991). The physicochemical and biological mechanisms that influence the diffusion of contaminants across the biological membranes of gills include the concentration of the contaminant in the water, its molecular weight and volume, the charge on the molecule and its lipid solubility (Spacie and Hamelink, 1982) in addition to the flow of water and blood to and from the gills, which maintains the diffusion gradient. The liver has numerous functions: it is the main organ involved in the metabolism of proteins, lipids and carbohydrates; it plays an important role in the storage and distribution of compound reserves; and it has a key role in xenobiotic biotransformation, detoxification and excretion. In this context, the present study was designed to integrate the histological changes in gills and liver with the chemical analysis of these organs in two native fish species, the lambari, Astyanax fasciatus, and the mandi, Pimelodus maculatus, which inhabit the reservoir of the Furnas HPS. The histological changes permit evaluation of the effect of multiple contaminants on the structure of these organs and the possible harmful effects on fish health. Both fish species are omnivorous and have different feeding habits: A. fasciatus is a benthic–pelagic species, and P. maculatus is a benthic and bottom-feeding species (Esteves and Galetti, 1995; Ramos et al., 2011). These species are widely distributed throughout the reservoir and do not exhibit long distance migratory

behavior (Costa et al., 2013), making them useful for comparing different sites in the same aquatic system. 2. Materials and methods 2.1. Study area and water and fish sampling Water and specimens of A. fasciatus (n = 20/site, Body mass [MB] = 37.8 ± 2.6 g, total Length [Lt] = 14.3 ± 0.3 cm) and P. maculatus (n =15/ site, MB = 182.3 ± 32.9 g, Lt = 25.1 ± 1.4 cm) were collected from five sampling sites in the reservoir of the Furnas HPS, Minas Gerais, Brazil (Fig. 1) in June (winter, dry season) and December (summer, wet season) of 2006. Site 1: FU10–Turvo (the reference site) is located at the confluence of the Grande and Sapucaí Rivers (S20° 40′ 835″ W46° 13′ 232″); site 2: FU20–Guapé (S20° 44′ 331″ W 45° 55′ 800″) and site 3: FU50–Porto Fernandes (S20° 48′ 826″ W45° 40′ 567″) are located in the Grande River axis; while site 4: FU30–Barranco Alto (S21° 10′ 510″ W45° 57′ 061″) and site 5: FU40–Fama (S21° 24′ 074″ W45° 49′ 621″) are located in the Sapucaí River axis (Fig. 1). Collected fish were anesthetized with benzocaine (0.1 g L−1) and killed by medullar sectioning. The gills and liver were removed and systematically sampled. Tissue samples for chemical analyses were stored at − 80 °C, and samples for histological analyses (5 samples/organ/fish/site) were immediately fixed with 2.5% glutaraldehyde buffered at pH 7.3 in 0.1 M phosphate buffer solution or Bouin's solution. Water samples (3 L) were collected (0.20 m below the surface water) into pre-cleaned amber glass bottles (0.2–1.0 L) and preserved according to the analyses to be done: physicochemical variables, metals and agrochemicals. All water samples were stored at ~ 4 °C immediately after collection until laboratory analysis. The agrochemicals evaluated in the water and in the gills and liver of fish were those generally applied to the field crops cultivated around the reservoir, the organochlorines: the 2,4-dichlorophenoxyacetic acid, alachlor, atrazine, hexachlorobenzene, lindane (gamma-BHC), metolachlor, methoxychlor, permethrin, propanil, simazine, aldrin, dieldrin, chlordane, endosulfan, endrin, heptachlor and heptachlor epoxide, pentachlorophenol and dichlorodiphenyltrichloroethane (DDT isomers), and the non-organochlorines: molinate, pendimethalin, trifluralin and bentazon (non-organochlorine contaminants). 2.2. Chemicals All reagents for metal analysis were of analytical grade. n-Hexane, dichloromethane, diethyl ether (pesticide grade), sulfuric acid 95%– 97%, acetone and anhydrous sodium sulfate for analysis were from Sigma Aldrich (Milwaukee, WI, USA). The purity of the solvents was determined by gas chromatography coupled to electro capture detection (GC–ECD). The standard 2,4-dichlorophenoxyacetic acid, alachlor, atrazine, hexachlorobenzene, lindane (gamma-BHC), metolachlor, methoxychlor, permethrin, propanil, simazine, aldrin, dieldrin, chlordane, endosulfan, endrin, heptachlor and heptachlor epoxide, pentachlorophenol and dichlorodiphenyltrichloroethane (DDT isomers), molinate, pendimethalin, trifluralin and bentazon, and element standards were from Sigma Aldrich (Milwaukee, WI, USA). Florisil powder (Merck; 0.150–0.250 mm in diameter) was used for residue analysis quality assurance and was baked at 150 °C for 4 h to ensure dryness. 2.3. Water chemical analyses Dissolved oxygen (DO), conductivity, temperature and pH were measured in the field using a multi-parameter water analyzer (YSI, 600XL). Alkalinity was determined as described by Golterman et al. (1978), and hardness was determined following APHA (1992) methodologies. Concentrations of nitrogen in the forms of ammonium, nitrite and nitrate were determined using a colorimetric method (Mackereth et al., 1978). The concentrations of agrochemicals and metals were determined by the laboratory BioEng Ltda (São Carlos, SP, Brazil).

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

295

Fig. 1. Location of the Furnas Hydroelectric Power Station reservoir in Minas Gerais (MG), Brazil and map of the Furnas HPS reservoir showing the sites of water and fish collection: FU10 (Turvo), FU20 (Guapé), FU30 (Barranco Alto), FU40 (Fama), and FU50 (Porto Fernandes).

Metal concentrations were determined following standard methods SW84603050/3051 (US EPA, 1986) using an atomic absorption spectrometer (AA12/1475 Gemini, Varian, USA). The aluminum concentration was measured using Eryochrom cyanine R SM 3500-a1 B. Acidified sample water (1 L) from each site was filtered through a 0.45-μm nylon filter. Blank solutions containing only reagents were used to deduce the content of metals resulting from the reagents and laboratory instruments. In the metal analysis, the blanks indicated negligible contamination, and the measured metal concentrations were within the certified range in the reference sample (between 98% and 105%). Ultrapure water was used to prepare standard solutions, dilutions and blanks. Standard materials were used to verify the accuracy of the analytical measurements. The equipment stability was checked by measuring the calibration curve before and after the measurement of the samples. The repeatability of measurements was generally ≥ 97% based on comparisons of triplicate values. For the agrochemical analysis, the 1 L of collected water samples was filtered (0.7 μm pore size) with glass fiber filters on a vacuum filter system for the extraction of the suspended sediments. Thereafter, each water sample was dried over silica gel until it reached constant weight. A mixture of hexane:dichloromethane (1:1) was used as the extraction solvent. Extracts were evaporated at room temperature to near dryness using N2 flow in an exhaust chamber, spiked with mixture of internal standards at a concentration of 160 μg L−1 and reconstituted with hexane, and the final volume was adjusted to 250 μL into glass amber vials for gas chromatography analysis. The analysis of agrochemicals concentrations was performed following US EPA protocols (US EPA, 1995, 2007) using an HP 5980 gas chromatograph and an HP 5970 MSD mass spectrometer (Hewlett Packard, USA). The applied agrochemical standards were in a mixture, consisting of the analyzed agrochemicals. Chromatographic separation was performed on a fused silica capillary column DB-5 MS (Agilent Technologies, CA, USA) with a 30 m × 0.25 mm I.D. and 0.25 μm film thickness, using helium as the carrier gas. The mass spectrometer was operated in the electron ionization mode with an ionizing energy of 70 eV, and the extracts were injected in the splitless mode, keeping the split valve closed for 0.8 min. Oven temperatures were programmed following Navarro-Ortega et al. (2010). Each extract was injected three times in a specific GC/MS program for organochlorines and non-chlorine compounds with the following oven temperature program: from 60 °C

(holding time 1 min) to 130 °C at 10 °C min − 1 to 220 °C at 3 °C min− 1 and finally to 300 °C at 10 °C min − 1 (holding time 5 min). Injection was achieved in the splitless mode keeping the split valve closed for 0.8 min. Helium was used as the carrier gas at a flow of 1.2 mL min− 1. The injector, transfer and ion source temperatures were set at 280 °C, 250 °C and 200 °C respectively. Acquisition was achieved in time scheduled Selected Ion Monitoring (SIM) mode to increase sensitivity and selectivity. Each compound was separately identified and quantified using a five-point calibration of mixed standard solutions in the range from 50 to 1000 μg L− 1. Method detection limits ranged from 0.002 to 0.025 μg L− 1 and the recoveries ranged from 90% to 100% for all compounds. The GC injections were performed with an automatic injector to improve reproducibility. The repeatability of the measurements was generally ≥ 97% when values derived from triplicate measurements were compared. The stability of the equipment was checked by measuring the calibration curve before and after the measurement of the samples. 2.4. Gill and liver chemical analyses To determine the metal concentrations in the gills and liver, 0.8– 1.0 g samples of these organs were dried at 60 °C until a constant weight was reached and subsequently digested with a solution (1:1) of ultra-pure HNO 3 and H2O 2 (30%). After complete tissue destruction, the samples were brought to a final concentration of 0.1% HNO 3 and stored at 4 °C until metal analysis was completed. All samples were completely clear and were not filtered. Metal concentrations were determined as described in Section 2.3. The metal concentration was expressed as μg g− 1 wet weight. To obtain the agrochemicals from the gill and liver tissues, 5–7 g samples of each tissue were homogenized with anhydrous Na2SO4 until a fine powder was obtained. This mixture was introduced into cellulose cartridges and Soxhlet extracted with n-hexane/dichloromethane (4:1 vv) for 18 h. The extracts were submitted to solid phase extraction (SPE) using a Florisil cartridge with n-hexane and diethyl ether (US EPA, 1995, 2007). Lipid content was determined gravimetrically using 20% of the extract. Compound standards were added to the rest of the extract, which was subsequently rinsed with sulfuric acid (5 times). All n-hexane or diethyl ether solutions were combined and concentrated by vacuum rotatory evaporation (20 °C, 20 torr) to small volumes (ca. 500 μL),

296

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

further concentrated to near dryness under gentle nitrogen flow, and redissolved in 250 μL of n-hexane. The concentrations of agrochemicals were performed as described above in Section 2.3. 2.5. Histological analyses For histopathological analyses, the gill (5–10 gill filaments/sample, 5 samples/fish) and the liver (3 mm tissue/sampling, 5 samples/fish) that were fixed in glutaraldehyde solution were dehydrated in ethanol and embedded in Historesin® (Leica, Heidelberg, Germany), which has negligible section shrinkage (Cruz et al., 2009). Gill sections (3 μm in thickness) were stained with toluidine blue, and liver sections (2 μm in thickness) were stained with toluidine blue + basic fuchsin. Gill samples fixed with Bouin's solution were dehydrated in ethanol, cleared in xylene, and embedded in paraffin. Sections (5 μm in thickness) were stained with periodic acid (PAS) and Alcian blue (AB) at pH 2.5 or 1.0 to identify the mucous cell types. All stained sections were observed under a BX51 Olympus microscope (Olympus, Denmark), and the images were captured by using a JVC-TKC1380 digital video camera and analyzed using Motic Images Plus 2.0. The histological changes were quantified in the gills and livers (5 random microscope fields/section, 5 sections/sampling, and 5 samplings/fish) according to the tissue specificity. The incidence and distribution of the lesions were evaluated based on the following criteria: 0, absence of lesions (absence of lesions or lesions on up to 10% of the total analyzed tissue); 0 +, rarely present (occurrence of lesions on 11% to 25% of the total analyzed tissue); +, present (occurrence of lesions on 26% to 50% of the analyzed tissue); ++, frequent (occurrence of lesions from 51% to 75% of analyzed tissue) and +++, highly frequent lesions (occurrence of lesions on 76% to 100% of the analyzed tissue). The mean alteration value (MAV) for each animal was calculated according to Schwaiger et al. (1997) but was slightly modified to follow a respective numeric value: 0–1.0, no pathological alterations; 1.1–2.0, focal mild alterations; 2.1–3.0, moderately spread lesions; 3.1–4.0, frequent lesions and 4.0–5.0, widely distributed lesions. The histopathological index (HI) for the gills (HIG) was calculated according to Cerqueira and Fernandes (2002) as modified from Poleksic and Mitrovic-Tutundzic (1994). The HI for the liver (HIL) was calculated according to Camargo and Martinez (2007). The HI was calculated based on the type, location and severity of the lesion. The gill lesions were classified into four groups: hypertrophy and hyperplasia of gill epithelium and related changes; changes in mucous and/or chloride cells; blood vessel changes and fibrosis and necrosis. The liver lesions were also classified into four groups: lesions in the structure of liver parenchyma, including interstitial tissue; hepatocyte changes including cytoplasmic and nuclear changes; blood vessel changes and necrosis. The lesions were classified into three progressive stages (S) based on the possibility that the normal structure of the organ would be restored after environmental amelioration: SI, changes that are reversible in which the normal organ structure can be recovered; SII, changes that are more severe and affect the organ function but may be repairable after environmental improvement; and SIII, changes in which the restoration of organ structure is not possible even after environmental improvement. The HI was calculated from the sum of the lesion types within each of the three stages multiplied by the stage index using the following mathematical equation proposed by Poleksic and Mitrovic-Tutundzic (1994):

0

I ¼ 10

a X i¼1

1

ai þ 10

b X i¼1

2

bi þ 10

c X

ci

i¼1

where a = first stage alterations (SI), b = second stage alterations (SII) and c = third stage alterations (SIII). An average index for fish collected at each site was calculated from the index obtained for each individual fish. The HI value was divided into categories: 0–10 = structural normal organ, 11–20 = slight to moderate damaged to the organ, 21–50 =

moderate to heavy damage to the organ and N 100 = irreparably damaged organ. Mucous cells were classified by a positive reaction to PAS and AB at pH 2.5 or pH 1.0, which is slightly modified from the protocol of SabóiaMoraes et al. (1996). Mucous cells were quantified in 5 random microscope fields/section in the gill filament (n = 25) for each animal collected at each site. 2.6. Statistical analyses The data are presented as the means ± SEM. After the homogeneity of the data from each group was verified using Bartlett's test, the parametric analysis of variance (ANOVA) was applied to determine significant differences among the groups. Tukey's test with 95% confidence limits was applied to compare the means whenever significance was observed (GraphPad InStat Software, San Diego, CA). 3. Results 3.1. Validation of analytical methods [limit of detection (LOD), limit of quantification (LOQ)] The validation of analytical methods was done as described by Bashir et al. (2012). Briefly, the precision and accuracy of the applied analytical method were validated by accurate analysis of standard reference materials. All the runs were carried out in triplicate. The results obtained were in good agreement with the certified values for all metals and pesticides and the recovered values were between 82% and 110% of the certified value. Calibration curves produced good correlation coefficient presenting r2 of approximately 0.999. The limit of detection (LOD) was determined as thrice the standard deviation of ten reagent blanks and the limit of quantification (LOQ) was equal to 10 times the standard deviation of the results for the triplicates used to establish a reasonable boundary of detection. LOD/LOQ in water: aluminum 1/10 μg L−1, chromium 20/200 μg L−1, copper 10/100 μg L−1, iron 10/100 μg L−1, zinc 10/ 100 μg L−1, alkalinity as CaCO3 100/1000 μg L− 1, hardness as CaCO3 100/1000 μg L−1, N-ammonia 10/100 μg L−1, N-nitrite 1/10 μg L−1, nitrate 10/100 μg L−1, aldrin/dieldrin 0.16/1.60 ng L−1, endosulfan (a,b, sulfate) 6.7/67.0 ng L−1, Heptachlor/Hept. Epoxide 0.1/1.0 ng L−1, Metolachlor 100/1000 ng L− 1 and in gill and liver: chromium 0.1/ 1.0 μg g−1, copper 0.1–1.0 μg g−1, iron 0.1/1.0 μg g−1, zinc 0.02/0.2 μg g− 1, aldrin/dieldrin 0.25/2.50 ng g−1, Endosulfan (a,b,sulfate) 0.45/ 4.50 ng g− 1, Heptachlor/Hept. Epoxide 0.5/5.0 ng g−1, Metolachlor 0.25/2.5 ng g−1. 3.2. Water analysis The physical and chemical variables of the water from the Furnas HPS reservoir exhibited fluctuation between sites; however, there was no major variation in the variables between June and December, except for temperature, which was lower in June (21 ± 1 °C) than in December (24.9 ± 1 °C) (Table 1). All of these limnological characteristics were within the range of limits recommended for water by the Brazilian Environment National Council for biota preservation (CONAMA 357, 2005). Most of the concentrations of agrochemicals analyzed were less than the detection limits of the analytical methods; only the organochlorines aldrin/dieldrin, endosulfan, heptachlor epoxide and metolachlor were detected and quantified in the water at sites FU20, FU 30, FU40 and FU50. The concentrations of these organochlorines were greater than those recommended by CONAMA 357 (2005), which are 0.005, 0.056, 0.01 and 10 μg L− 1, respectively, for aldrin/dieldrin, endosulfan, heptachlor epoxide and metolachlor (Table 1). No agrochemicals were detected at FU10 (Table 1). Low levels of aluminum were detected at all sites, Fe was also detected at all sites except FU30, and Zn was detected only at FU30 and FU40 (Table 1).

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

297

Table 1 Water variables (mean ± SD) of the reservoir of Furnas HPS, MG, Brazil, in June and December 2006. Variables

Furnas HPS Reservoir FU10

pH Temperature (°C) Dissolved oxygen (mg/L) Conductivity (μS cm−1) Alkalinity (mg L−1 as CaCO3) Hardness (mg L−1 as CaCO3) N-ammonia (mg L−1 N) N-nitrite (mg L−1 N) N-nitrate (mg L−1 N) Aldrin and Dieldrin (μg L−1) Endosulfan (μg L−1) Heptachlor epoxide (μg L−1) Metolachlor (μg L−1) Aluminum (μg L−1) Iron (μg L−1) Zinc (μg L−1)

FU20

FU30

FU40

FU50

June

December

June

December

June

December

June

December

June

December

7.2 22.6 7.2 33.0 29.3 24.6 0.20 0.04 0.25 nd nd nd nd 2.1⁎ 30.5⁎ nd

7.3 24.5 7.6 36.5 24.1 20.0 0.10 0.01 nd nd nd nd nd 5.0⁎ 30.2⁎ nd

7.7 21.8 7.2 30.0 13.8 11.7 0.10 nd 0,16 1.10 0.80 0.45 36.0 6.2⁎ 280.5 nd

7.2 25.0 7.0 36.5 22.5 21.0 0.20 0.01 nd 0.86 0.74 0.40 18.0 4.8⁎ 302.1 nd

7.4 20.8 8.2 32.0 15.4 13.1 1.00 0.10 0.39 0.50 1.00 0.80 10.0 5.0⁎ nd 10.0⁎

7.3 25.4 7.8 43.0 18.0 19.0 0.10 0.01 nd nd nd 0.31 8.1 2.8⁎ nd 10.0⁎

7.6 20.9 7.3 33.0 12.7 9.0 0.49 0.10 0.05 0.003 1.0 nd nd 3.5⁎ 640.4 20.0⁎

6.9 24.8 7.7 38.6 20.0 23.0 0.05 0.03 0.10 nd nd nd nd 3.1⁎ 710.1 20.0⁎

7.4 21.6 7.4 30.0 13.0 11.3 0.10 0.10 0.29 1.03 0.28 0.32 29.0 2.0⁎ 250.2 nd

7.3 24.6 7.4 35.4 15.0 18.0 0.02 0.01 nd 1.00 0.10 0,30 14.0 1.0⁎ 220.2 nd

FU10: Turvo; FU20: Guapé, FU30: Barranco Alto, FU40: Fama, FU50: Porto Fernandes. nd, not detected. ⁎ below the quantification limit.

3.3. Gill and liver chemical analysis The chemical analysis of the gills and liver revealed the presence of metals and organochlorines in both organs at varying concentrations

among the sites, periods of the year and fish species (Table 2). Copper, iron and zinc were detected in the gills and liver of both fish species from all sites with the exception of Zn, which was not detected in the liver of P. maculatus from FU10, in June. Chromium was detected in

Table 2 Organochlorines and metals in the gills and liver of Astyanax fasciatus and Pimelodus maculatus collected in the reservoir of Furnas HPS, MG, Brazil, in June and December 2006. Variables

Furnas HPS Reservoir FU10

GILLS Astyanax fasciatus Aldrin/dieldrin (ng g−1) Endosulfan (a,b,sulfate) (ng g−1) Heptachlor/Hept. Epoxide (ng g−1) Copper (μg g−1) Chromium (μg g−1) Iron (μg g−1) Zinc (μg g−1) Pimelodus maculatus Aldrin/dieldrin (ng g−1) Endosulfan (a,b,sulfate) (ng g−1) Heptachlor/Hept. Epoxide (ng g−1) Copper (μg g−1) Chromium (μg g−1) Iron (μg g−1) Zinc (μg g−1) LIVER Astyanax fasciatus Aldrin/dieldrin (ng g−1) Endosulfan (a,b,sulfate) (ng g−1) Heptachlor/Hept. Epoxide (ng g−1) Metolachlor (ng g−1) Copper (μg g−1) Chromium (μg g−1) Iron (μg g−1) Zinc (μg g−1) Pimelodus maculatus Aldrin/dieldrin (ng g−1) Endosulfan (a,b,sulfate) (ng g−1) Heptachlor/Hept. Epoxide (ng g−1) Metolachlor (ng g−1) Copper (μg g−1) Chromium (μg g−1) Iron (μg g−1) Zinc (μg g−1)

FU20

FU30

FU40

FU50

June

December

June

December

June

December

June

December

June

December

nd nd nd 0.3⁎ nd 8.7 5.25

nd nd nd 0.3⁎ nd 6.2 5.02

80.50 51.02 30.40 0.8⁎ 0.4⁎ 8.2 7.85

80.17 nd nd 1.1 0.6⁎ 8.5 9.45

29.77 49.25 30.50 1.2 1.3 9.8 8.00

nd nd nd 1.0 1.0 9.8 5.46

nd 39.90 nd 0.8⁎ 0.3⁎ 9.6 8.32

nd nd nd 0.4⁎ 0.3⁎ 7.6 3.40

50.23 nd 19.82 1.2 1.1 8.7 8.94

nd nd nd 0.9⁎ 0.8⁎ 6.8 6.15

nd nd nd 0.9⁎ nd 6.4 7.32

nd nd nd 0.8⁎ nd 6.1 5.45

89.72 60.12 39.95 0.6⁎ 0.4⁎ 19.7 8.85

80.31 50.52 nd 0.3⁎ 0.6⁎ 19.7 8.84

89.09 81.01 30.44 0.6⁎ 1.3 14.3 8.20

nd nd nd 0.9⁎ 0.9⁎ 8.0 8.80

nd 49.84 nd 1.2 1.1 9.8 8.08

nd nd nd 1.1 0.8⁎ 8.4 10.01

90.15 nd 9.84 1.5 0.3⁎ 12.4 12.83

30.52 nd 9.93 1.2 1.0 9.8 17.25

nd nd nd nd 0.7⁎ nd 2.8 0.92

nd nd nd nd 0.9⁎ nd 2.2 0.50

142.11 30.02 21.0 nd 0.9⁎ 0.6⁎ 10.6 4.45

89.65 9.80 nd nd 0.9⁎ 0.4⁎ 10.0 4.25

58.92 9.60 9.6 0.60⁎ 5.6 1.0 12.6 6.21

nd nd 10.3 nd nd 0.6⁎ 14.2 2.42

nd nd nd nd 10.2 1.0 23.8 9.45

nd nd nd nd 8.6 0.6⁎ 18.2 9.09

9.94 nd nd nd 0.8⁎ 0.2⁎ 10.0 3.66

70.42 nd nd nd 0.8⁎ 0.1⁎ 9.1 2.60

nd nd nd nd 1.4 nd 4.9 nd

nd nd nd nd 3.4 0.1⁎ 2.1 0.02⁎

420.40 4.04⁎ 709.9 0.98⁎ 1.1 0.7⁎ 18.9 8.9

149.59 nd 560.4 0.86⁎ 0.8⁎ 1.0 8.7 6.2

59.81 1.00⁎ 118.9 0.90⁎ 8.5 0.9⁎ 29.4 10.2

30.05 nd nd nd 14.2 1.0 34.8 17.6

nd nd nd nd 10.8 0.6⁎ 12.9 2.2

nd nd nd nd 8.5 0.9⁎ 13.8 2.0

10.30 nd nd nd 1.0 0.5⁎ 12.1 6.8

3.00 nd 6.0 nd 0.9⁎ nd 16.5 nd

FU10: Turvo; FU20: Guapé, FU30: Barranco Alto, FU40: Fama, FU50: Porto Fernandes. nd, not detected. ⁎ below the quantification limit.

298

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

the gills and liver of both species from FU20 and FU50 in June and December, and it was also detected in the liver of P. maculatus from FU10 in December. The organochlorines aldrin/dieldrin, endosulfan and heptachlor/heptachlor epoxide were detected in the gills and liver of fish collected at all sites, except those collected at FU10 (Table 2). The concentrations of these organochlorines in the gills of fish from FU20, FU30 and FU50 were similar for both species in June, and aldrin/dieldrin and endosulfan were also detected in P. maculatus in December at FU20 and FU50. In the liver, the concentration of aldrin/dieldrin was higher than the other organochlorines in A. fasciatus from FU20, and the concentrations of aldrin/dieldrin and heptachlor/heptachlor epoxide were high in P. maculatus collected at the same site, in June and December (Table 2). Metolachlor was detected in the liver of both species (Table 2). 3.4. Histological changes in the gills The general gill structures of A. fasciatus and P. maculatus are similar to other teleost fish (Figs. 2A, 3A, B). The filament epithelium is stratified (4–6) and the lamellar epithelium contains two epithelial cell layers.

Pavement cells (PVCs) are the most abundant cells in the filament and lamellar epithelium. Chloride cells (CCs) are distributed in the interlamellar regions of the filament epithelium and are dispersed in the lamellar epithelium. Mucous cells (MCs) are localized in the leading and trailing edges of the gill filaments and were more numerous in P. maculatus (Fig. 4). Five MC types were identified in both fish species based on positive reactions to PAS and AB: type 1 cells contained neutral glycoproteins and sialic acids, type 2 cells contained acid mucosubstances with carboxyl groups, type 3 cells contained acid mucosubstances with sulfate esters, type 4 cells contained hexoses and acid sulfate esters, and type 5 cells contained all types of mucosubstances (Table 4, Fig. 4). MC type 1 was the most abundant cell type, and type 3 cells were less abundant in both species, with P. maculatus generally exhibiting a greater abundance of MC than A. fasciatus. All fish exhibited histological changes. Figs. 2B–K and 3C–L show the most frequent histopathological findings in the gills of both fish species, and Table 3 summarizes the frequency and the degree of severity of each histopathological finding. Epithelial hypertrophy and hyperplasia in the filament and lamellar epithelia were the most common changes (Figs. 2, 3) causing incomplete or total lamellar fusion (Figs. 2E, F, 3D,

Fig. 2. Representative histopathologies in the gills of Astyanax fasciatus collected in the Furnas HPS reservoir, MG, Brazil. A. Normal gill. B. Lamellar epithelial lifting (arrow); C. Lamellar epithelial hypertrophy (arrow); D. Pavement (arrow) and chloride (double arrow) cell hypertrophy; E. Cell hypertrophy and total fusion of several lamellae (arrow); F. Epithelial hyperplasia and total fusion of several lamellae (arrow); G. Mucous cell proliferation (arrow); H. Lamellar chloride cell hyperplasia and hypertrophy (arrow); I. Lamellar telangiectasis (*); J. Lamellar aneurysm (*); K. Lamellar telangiectasis (*) and hemorrhages (arrow). F, filament; L, lamella. Scale bar in μm.

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

299

Fig. 3. Representative histopathologies in the gills of Pimelodus maculatus collected in Furnas HPS reservoir, MG, Brazil. A. Normal gill. B. High magnification of lamella. Note flattened epithelium (arrow) in the lamella and 5 to 6 cell layers in the filament epithelium. C. Lamellar epithelial hypertrophy (arrow) and filament epithelial hyperplasia (*); D. Epithelial hyperplasia and total fusion of several lamellae (*); E. Lamellar epithelial lifting (arrow) and epithelial necrosis (double arrow); F. Lamellar epithelial lifting (arrow) and filament hypertrophy (*); G. Mucous cell proliferation (arrow); H. Chloride cell hypertrophy (arrows); I. Lamellar congestion (arrow); J. Lamellar telangiectasis (*); K and L. Lamellar aneurysm (*) and hemorrhages (arrow). F, filament; L, lamella. Scale bar in μm.

F, H). The abundance of CCs decreased in the gills of fish from sites FU20, FU30 and FU50 in June, but these cells showed hypertrophy, particularly those of P. maculatus gills (Table 3, Figs. 2D, H, 3H). MC hyperplasia occurred in the interlamellar region of the filament epithelium of specimens of both species collected in June (Table 3, Figs. 2G, 3G). In A. fasciatus, the type 1 and 2 MC abundances were greater in fish collected from FU20, FU30 and FU50, and in P. maculatus, the abundances of these MC types were greater in fish collected from FU20, FU30, FU40 and FU50 than in those collected from FU10 in June and December (Table 4). MC type 3 did not vary significantly among the sites or during the periods of the year (Table 4). Lamellar blood congestion, lamellar telangiectasis, aneurysms and punctual rupture of epithelium with hemorrhages characterized the main gill vascular pathologies (Figs. 2I–K, 3I–L). Tissue fibrosis and gill parasites were absent from both species. The MAV calculated for the gills (MAVG) of both species, for June and December varied from 0.7 to 1.6 in fish collected from FU10, indicating

mild focal alterations. In June, MAVG varied from 1.8 to 3.4 in fish collected from FU20, FU30, FU40 and FU50, indicating moderate to frequent spread of lesions (Fig. 5A, E). In December, MAVG decreased in P. maculatus collected from FU20 and FU30 compared with June. HIG varied from 5.2 to 10.0 in both species collected from FU10, indicating normal gill structure. In fish collected from FU20 to FU50, HIG varied from 9.4 to 14.8 in June, indicating slight to moderate damage, and decreased in December, indicating normal gill structure (Fig. 5B, F). 3.5. Histological changes in the liver The liver parenchyma comprises hepatocytes arranged in a cylindrical tubular pattern in which the apices surround the biliary lumen and the base faces the sinusoid vessels, forming a cord-like structure accompanied by the sinusoid vessels. Typical hepatocytes have a round nucleus with a prominent nucleolus. In A. fasciatus and P. maculatus

300

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

same sites varied from 3.8 to 4.4, indicating a wide distribution of lesions throughout the organ (Fig. 5C, G). In December, the MAVL values decreased, indicating the presence of moderately spread lesions for A. fasciatus and frequent lesions for P. maculatus. The HIL of both species collected from FU10 varied from 32.8 to 46. 9, indicating the presence of moderately spread lesions; these values did not change between June and December (Fig. 5D, H). In fish collected from FU20, FU30, FU40, and FU50, the HIL values of both species were N100, indicating the presence of irreparably damaged organs in both June and December.

4. Discussion

Fig. 4. A and B. High magnifications of representative mucous cells containing neutral glycoproteins and sialic acids (arrow) in A and mucous cells containing acid mucosubstances with sulfate esters (arrow) and those containing all types of mucosubstances (double arrow) in B; C–F. Distribution of mucous cells in the trailing and leading edges of the gill filaments (cross section) of Astyanax fasciatus (C and D) and Pimelodus maculatus (E and F) from the Furnas HPS reservoir, MG, Brazil. Scale bar in μm.

numerous alterations were found in the parenchyma and in the vascular and biliary systems. In general, the hepatocytes in the fish livers collected from FU10 showed the typical cord-like arrangement surrounded by sinusoids (Fig. 6A); however, some histological alterations were observed (Table 5). Fish collected from FU20, FU30, FU40 and FU50 showed numerous alterations in the nucleus and the cytoplasm of the hepatocytes and in the sinusoid and biliary systems (Fig. 6B–L). Table 5 summarizes the frequency of the histopathological findings in the liver of A. fasciatus and P. maculatus and indicates the degrees of severity of each. In general, large liver areas exhibited disorganization of the tubular cord-like structure (Fig. 6B), large melanomacrophage aggregates close to the blood vessels (Fig. 6C) and sinusoid dilatation characterized by blood congestion with consequent liver hyperemia (Fig. 6D). Focal necrosis and changes in the epithelial cells of the bile duct were also frequent occurrences in both species (Table 5, Fig. 6E, F). The most frequent alterations observed in the parenchyma were characterized by hepatocyte hypertrophy, the presence of cytoplasmic granules and vacuolization (Fig. 6G–L). Loss of the cell outline (Fig. 5H, K) and numerous nuclear alterations, such as pyknosis, vacuoles, high concentrations of peripheral heterochromatin, and irregular nuclear contours (Fig. 6B–L), were more frequent in fish collected from FU20, F30 and FU50 (Table 5, Fig. 6). The livers of P. maculatus exhibited a greater percentage of alterations than those of A. fasciatus (Table 5). The MAV indexes for the liver (MAVL) in fish collected from FU10 varied from 1.8 to 2.6 in both species, indicating the presence of moderately spread lesions (Fig. 5C, G). MAVL for A. fasciatus collected from FU20 to FU50 in June varied from 2.6 to 4.0, indicating moderately spread to frequent lesions. MAVL for P. maculatus collected from the

The limnological variables of the Furnas HPS reservoir were within the ranges recommended for aquatic biota by the CONAMA 357 (2005); however, the organochlorines aldrin/dieldrin, endosulfan, heptachlor/heptachlor epoxide and metolachlor and the metals detected in the water and their presence or accumulation in the gills and liver of A. fasciatus and P. maculatus confirmed the contamination of reservoir. Different concentrations of aldrin/dieldrin, endosulfan, heptachlor/heptachlor epoxide and metolachlor and the metals, aluminum and iron, in the Furnas HPS reservoir were reported by SadauskasHenrique et al. (2011), and endosulfan, heptachlor/heptachlor epoxide were also reported in the water in different areas of Furnas HPS reservoir (FUPAI, 2013). The use of aldrin/dieldrin, endosulfan and heptachlor/heptachlor epoxide in agriculture was officially discontinued in Brazil, in 1985, with some exceptions, such as the control of ants and termites in reforesting areas and agricultural emergencies. The production of aldrin/dieldrin was stopped in 1998, and heptachlor was completely banned in 2004 and endosulfan in 2014 (ANVISA, 2014). Aldrin/dieldrin, heptachlor and endosulfan have also been detected in the water and sediment of many other aquatic ecosystems in Brazil (Rissato et al., 2006; Silva et al., 2008; Figueiredo et al., 2013). Presently, there is no restriction on metolachlor application on field crops (ANVISA, 2014). Thus, the organochlorine concentrations in the water may be due to their accumulation in the sediment leading to bioavailability in the water or to recent inputs into aquatic ecosystem due to the periodicity of agrochemical application on field crops and in farm management. With the exception of Al, all of the metals detected in the water and fish – Cu, Cr, Fe and Zn – are essential elements for most living organisms, but they can be toxic at high concentrations (Heath, 1995). In the case of Cr, trivalent Cr is an essential element, but at high concentrations or in different forms, such as hexavalent Cr, it is toxic even at low concentrations. The accumulation of Cu and Zn in the gills and liver of both species at different levels in June and December demonstrates that these metals are present in the sediment of all sites and that their bioavailability during different periods of the year is most likely due to the dynamics of the metals in the water-sediment of the reservoir. Chromium accumulation in fish collected from all sites, except FU10, may be associated with bovine leather tanning, which is a common activity in the small cities located near the reservoir. The diffusion of organic compounds such as aldrin/dieldrin, endosulfan, heptachlor and metolachlor through the biological membranes is related to their lipid solubility, but some physicochemical restrictions, such as the water layer on the gill epithelium, may limit their absorption (McKim and Erickson, 1991). Conversely, the lipid membrane is a barrier for organic and inorganic ions; the diffusion of free ions such as metals from water into the gills depends on their speciation, which is influenced by water pH, hardness and alkalinity, and generally occurs through the ionic channels in the cell membrane and/or cellular junctions (Tao et al., 2001). The compounds and metals binding proteins and lipids affect the diffusion gradient across the gill epithelium, and they are transported through the blood (McKim and Erickson, 1991). Then, the bioaccumulation of organochlorines and metals in the gills and liver is the result of numerous physicochemical factors and also of

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

301

Table 3 Histopathology in the gills of Astyanax fasciatus and Pimelodus maculatus collected in the Furnas HPS reservoir, MG, Brazil in June and December 2006. Variables

Stage

Furnas HPS Reservoir FU10

FU20

FU30

FU40

FU50

Astyanax fasciatus June General alterations in the gill epithelium Lamellar epithelial hypertrophy Irregular lamellar epithelial hyperplasia Lamellar epithelial lifting Partial fusion of several lamellae Total fusion of several lamellae Lamellar epithelial rupture Focal Necrosis Mucous cell hypertrophy Mucous cell hyperplasia Chloride cell hypertrophy Chloride cell hyperplasia Blood vessel changes Pillar cell system constriction Pillar cell system enlargement Lamellar blood congestion Lamellar telangiectasis Lamellar aneurysm Epithelial rupture and hemorrhage December General alterations in the gill epithelium Lamellar epithelial hypertrophy Irregular lamellar epithelial hyperplasia Lamellar epithelial lifting Partial fusion of several lamellae Total fusion of several lamellae Lamellar epithelial rupture Focal Necrosis Mucous cell hypertrophy Mucous cell hyperplasia Chloride cell hypertrophy Chlorice cell hyperplasia Blood vessel changes Pillar cell system constriction Pillar cell system enlargement Lamellar blood congestion Lamellar telangiectasis Lamellar aneurysm Epithelial rupture and hemorrhage

FU10

FU20

FU30

FU40

FU50

Pimelodus maculatus

I I I I II II III I I I I

0+ 0+ + 0 0 0 0 0 0 0 +

+++ ++ + ++ 0+ 0+ 0+ 0 0+ + 0+

+++ ++ ++ ++ 0 0+ + 0 0+ + 0+

+ 0 ++ + 0 0 0 0+ 0 0+ +

+++ 0+ + ++ 0 + 0 0 0 + 0+

+ + 0 0+ 0 0 0 0 0 0+ +

+++ +++ + +++ +++ + 0+ 0+ + ++ 0+

+++ ++ ++ +++ 0+ + 0+ 0+ + ++ 0+

++ 0+ 0+ ++ 0 0 0 0+ 0+ 0+ +

+++ ++ + +++ 0+ 0+ 0 0+ 0+ ++ 0+

I I I I II II

0 0 0 0+ 0 0

0 0 + + + +

0 0 + + + +

0 0 0+ 0 0 0+

0+ 0+ + + + +

0 0 0 0+ 0 0

0+ ++ + + 0+ +

0+ 0 ++ + + +

0+ 0 0+ 0 0 0+

0+ 0+ 0+ 0+ 0+ 0

I I I I II II III I I I I

0 0+ 0 0+ 0 0 0 0 0+ 0 +

0+ 0+ ++ + 0+ 0+ 0 0 0 + 0+

0+ + + + 0+ 0 0 0 0 + +

0 0+ 0 0+ 0 0 0 0+ 0+ 0+ +

0 0+ 0+ 0+ 0+ 0+ 0 0 0+ + +

0 0+ 0+ 0+ 0 0 0 0 0 0+ ++

++ ++ ++ ++ + 0+ 0+ 0+ + ++ 0+

+ + + 0+ 0+ 0+ 0 0+ 0 ++ ++

0+ 0 0 0 0 0+ 0 0+ 0+ + ++

+ 0+ 0+ 0+ 0+ 0+ 0+ + + + 0+

I I I I II II

0 0 0 0+ 0+ 0

0 0 0+ 0+ + +

0 0 0+ 0+ 0+ +

0 0 0+ 0+ 0 +

0 0+ 0+ 0+ 0+ +

0 0 0 0+ 0 0

+ 0+ + + + 0+

0+ 0 0+ 0 0+ 0+

0 0 + 0 0+ 0+

0+ 0+ 0+ + + 0+

FU10: Turvo; FU20: Guapé, FU30: Barranco Alto, FU40: Fama, FU50: Porto Fernandes. 0 absent; 0+ rarely present; + present; ++ frequent; +++ highly frequent.

Table 4 Mucous cells (.103 mm−3) in the gills of Astyanax fasciatus and Pimelodus maculatus collected in the Furnas HPS reservoir, MG, Brazil in June and December 2006. Mucous cells

Furnas HPS Reservoir FU10

FU20

FU30

FU40

FU50

FU10

Astyanax fasciatus June Type 1 Type 2 Type 3 Type 4 Type 5

54 19 17 38 10

December Type 1 Type 2 Type 3 Type 4 Type 5

60 ± 6 36 ± 5⁎ 12 ± 1 36 ± 4 6 ± 1⁎

± ± ± ± ±

4 4 2 4 1

60 55 15 32 15

FU20

FU30

FU40

FU50

Pimelodus maculatus ± ± ± ± ±

4a 6a 3 1 1

55 ± 2 44 ± 4 17 ± 2 32 ± 1 8 ± 1⁎

4 6a 3 1 2

42 ± 1 25 ± 2 13 ± 4 36 ± 6 9±1

50 49 17 34 12

6 8a 3 2 2

71 ± 4 56 ± 4 8±2 46 ± 6 6±0

126 ± 8a 112 ± 4a 6±1 62 ± 6a 8±3

112 ± 8a 116 ± 10a 12 ± 1 72 ± 2a 7±1

130 ± 6a 126 ± 8a 6±5 40 ± 2 6±4

134 ± 9a 127 ± 12a 8±2 65 ± 6a 6±2

56 ± 6 48 ± 2a 16 ± 2 28 ± 3 8±2

52 ± 4 25 ± 4 14 ± 1 32 ± 4 5±2

46 ± 4 48 ± 2a 16 ± 3 24 ± 6 6±2

64 ± 4 56 ± 6 8±1 42 ± 5 8±2

92 ± 4a⁎ 105 ± 8a 4±2 63 ± 4ae 10 ± 2

90 ± 2a 111 ± 11a 4±2 64 ± 4a 6±1

80 ± 6a⁎ 112 ± 10a 8±2 70 ± 6ae 8±2

110 ± 7ad 103 ± 8a 6±3 58 ± 6 6±2

56 68 20 28 13

± ± ± ± ±

± ± ± ± ±

Type 1, containing hexoses and sialic acids; type 2, containing acid mucosubstances with carboxyl groups; type 3, containing acid mucosubstances with sulfate esters; type 4, containing hexoses and acid mucosubstances with carboxyl groups and type 5, containing hexoses and acid mucosubstances with sulfate esters. a, b, c, d and e indicate significant differences from FU10, FU20, FU30, FU40 and FU50, respectively. ⁎ indicates significant differences between June and December.

302

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

Fig. 5. Mean alteration values (MAV) and histopathological alteration index (HI) of gills (G) and livers (L) of Astyanax fasciatus and Pimelodus maculatus collected in the Furnas HPS reservoir, MG, Brazil. a, b, c, d and e indicate significant differences at FU10, FU20, FU30, FU40 and FU50, respectively. *indicates significant differences between June and December.

the capability of gill and liver cells to metabolize (Sakuragui et al., 2013) and eliminate these compounds. Gill and liver histological changes in both fish species clearly indicated cellular reactions to contaminant accumulation in these organs; the liver damage suggests that this organ was more adversely affected by the contaminant accumulation. The transfer of contaminants from the gills to the blood and their preferential accumulation in the liver, the main organ responsible for biodegradation, was already postulated

by Jin et al. (2010) and Paulino et al. (2012a). In general, the overall effect of contaminants is described by the histopathological indexes (MAV and HI); the lower MAV and HI in the gills and liver of fish from FU10 than those from other sites are due to the lower metal accumulation and the absence of organochlorines at FU10. In the gills, despite the presence of mild focal alterations in the organ, the normal gill structure was maintained. In contrast, the metal accumulation in the liver resulted in scattered lesions and moderate to heavy damage to the liver

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

303

Fig. 6. Representative histopathologies in the livers of Astyanax fasciatus and Pimelodus maculatus collected Furnas HPS reservoir, MG, Brazil. A. Normal liver. Note the tubular arrangement of hepatocytes (H) in cross section showing a central biliary lumen (*) surrounded by sinusoids (arrow); B. Liver exhibiting disorganization of hepatocytes cords (arrow); C. Melanomacrophage center (*); D. Hyperemia and sinusoid dilatation (double arrow), with central vein (v) and hepatocytes arranged in cord-like structure (arrow); E. Necrosis focus (*), note cellular disorganization, karyolysis (arrows), karyorrhexis (double arrow); F. Necrosis focus (*) showing alterations in the epithelial cells of the bile duct (arrow); P, Pancreatic cells; G. Hepatocyte hypertrophy (arrow), with cytoplasm inclusions (double arrow); H. Hepatocytes containing acidophil granules (arrow); I. Cytoplasm degeneration (arrow); J. Cytoplasm vacuolization (arrows); pyknotic nuclei (double arrow); K. Pyknotic nuclei (arrow), karyorrhexis (double arrow), karyolysis (triple arrow), and nuclei absence (*). Scale bar in μm.

structure. The greater metal accumulation in the gills and liver of fish from FU40 than of fish collected from FU10 and the endosulfan accumulation (June) resulted in greater damage in the liver. Both species from FU40 maintained the normal gill structure but, their livers exhibited more frequent and severe lesions resulting in moderate to heavy damage in the organ. The chromium detected in the liver together with the accumulation of other metals in fish from FU40 is likely related to an increase in the number of lesions observed in this organ. Chromium

induces alterations in the genetic, physiological, biochemical and morphological aspects of gills and liver in fish and these alterations are dependent upon the concentration and time of exposure (Velma et al., 2009). In fish collected from FU20, FU30 and FU50, the accumulation of metals and organochlorines resulted in more frequent lesions and slight to moderate damage to the gills, whereas in the liver, the lesions were widely distributed throughout the organ, with numerous regions of

304

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

Table 5 Histopathological changes in the liver of Astyanax fasciatus and Pimelodus maculatus collected in the Furnas HPS reservoir, MG, Brazil, in June and December 2006. Variables

Furnas HPS Reservoir Stage

FU10

FU20

FU30

FU40

FU50

Astyanax fasciatus June General changes in the liver Hepatic cell cord disorganization Melano-macrophage aggregates Hyperemia Focal necrosis Hepatocyte changes Nuclear hypertrophy Nuclear deformation Pyknotic nucleus Nucleolus absence Nuclear vacuolization (degeneration) Nucleus absence Cytoplasmic vacuolization Cytoplasmic granules Cytoplasmic degeneration Cytoplasmic lipid droplets Cellular hypertrophy December General changes in the liver Hepatic cell cord disorganization Melano-macrophage aggregates Hyperemia Focal necrosis Hepatocyte changes Nuclear hypertrophy Nuclear deformation Pyknotic nucleus Nucleolus absence Nuclear vacuolization (degeneration) Nucleus absence Cytoplasmic vacuolization Cytoplasmic granules Cytoplasmic degeneration Irregular cellular limits Cellular hypertrophy

FU10

FU20

FU30

FU40

FU50

Pimelodus maculatus

I I I III

0 0+ 0 0

+ ++ + +++

+ ++ 0+ ++

0+ 0+ 0+ ++

+ + 0+ ++

0 0+ 0 0+

0+ ++ +++ +++

0+ + ++ +++

0+ 0+ ++ ++

0+ + ++ +++

I I II II II II I I II I I

0+ 0+ 0 0+ + 0 0+ + 0+ 0+ 0

++ ++ ++ ++ ++ ++ ++ +++ ++ + ++

++ + + ++ ++ 0+ ++ ++ + + ++

+ 0+ + + + 0+ + ++ + 0+ +

++ +++ ++ ++ ++ + ++ +++ ++ ++ ++

0 0+ 0+ 0 0+ 0+ 0 0+ 0 0+ 0

+++ +++ ++ +++ ++ + +++ +++ +++ +++ ++

++ +++ ++ ++ ++ 0+ ++ +++ +++ +++ ++

++ + + + + + ++ ++ ++ ++ +++

+++ +++ ++ + ++ ++ ++ +++ +++ +++ ++

I I I III

0 0 0 0+

0+ 0+ + 0+

0 0 0+ +

0 0 0 0+

0+ 0 0 ++

0 0 0 0+

+ + + +

0+ 0+ 0+ 0+

0+ 0+ 0+ +

+ + + 0+

I I II II II II I I II I I

0+ 0 0 0+ 0 0 0+ 0+ + 0+ +

+ + + + + 0+ + + ++ ++ ++

+ + + + + 0 + + + + ++

0+ 0+ + 0+ 0+ 0 0+ + + 0+ +

+ + + + + 0 0+ + ++ ++ ++

0 0 0+ 0+ 0+ 0 + + 0 0+ +

++ ++ + ++ ++ 0+ ++ ++ ++ ++ ++

++ + + 0+ ++ 0 + + + ++ +

0+ + + + + 0 + + + ++ 0+

++ + + ++ ++ 0+ ++ ++ + ++ ++

FU10: Turvo; FU20: Guapé, FU30: Barranco Alto, FU40: Fama, FU50: Porto Fernandes. 0 absent; 0+ rarely present; + present; ++ frequent; +++ highly frequent.

necrosis, which resulted in heavy and irreparable liver damage. Lesions classified as stages I and II are repairable with water quality amelioration, but they may progress to stage III if the level of contamination is maintained or increased. Lesions classified as stage III cannot be restored (Poleksic and Mitrovic-Tutundzic, 1994). The morphological changes observed in the gills and liver of A. fasciatus and P. maculatus living in the Furnas HPS reservoir can be attributed to the physiological and biochemical changes and the degree of lipid peroxidation in the cells (Sakuragui et al., 2013). The detailed analysis of the lesions in the gills of fish from FU20, FU30 and FU50 revealed changes in the integrity of the lamellae, which may cause gas exchange dysfunction. Epithelial hypertrophy and hyperplasia resulted in partial or complete fusion of several lamellae, reducing the respiratory surface area and the interlamellar space through which water flows during the respiratory cycle. Epithelial lifting, which is the morphological expression of edema in the lamellar epithelium as a consequence of an inflammatory processes or increased permeability in the pillar cell system, increases the water-blood diffusion distance for gas exchange. All of these changes are defense responses that help to prevent the uptake of contaminants from water (Mallat, 1985; Cerqueira and Fernandes, 2002); however, they may result in internal hypoxia. These gill alterations are common responses to a large class of contaminants, including organochlorines and metals, and have been previously reported for fish collected from polluted environments (Ribeiro et al., 2005; Capkin et al., 2006, 2009; Fernandes et al., 2008; Triebskorn et al., 2008; Yasser and Naser, 2011; Souza

et al., 2013). Tissue necrosis, which was rarely present in the gills of A. fasciatus and P. maculatus, leads to a loss of lamellae, depending on the extent of the necrosis, and contributes to drastic reduction in the respiratory surface area. Blood vessel changes, such as telangiectasis and aneurysms, occur due to the direct action of the contaminants on the cellular adhesion of the pillar cell system via changes in the pressure of the afferent/ efferent blood vessels of the gills to optimize the gill function or due to pillar cell necrosis. Telangiectasis and aneurysms, although reversible, may progress under chronic exposure and result in epithelial rupture and hemorrhages (Poleksic and Mitrovic-Tutundzic (1994), as observed in both species collected from the Furnas HPS reservoir. These lesions also contribute to decreased gas exchange rates in the gills. The changes in the CCs are directly related to ionic regulation (Evans et al., 2005; Paulino et al., 2012b). The hypertrophy of CCs in fish from all sites except FU10 indicates an increase in cellular activity to maintain ionic uptake; in fish collected from FU40, this increasing was accompanied by CC hyperplasia, suggesting a compensatory response to maintaining ionic homeostasis. The induction of CC proliferation in fish collected from FU40 could be related to metals, which were the main contaminants accumulated in the gills. Metals displace Ca2+ from cellular junctions, contributing to increased water and ion permeability through the gill epithelium and indirect activation of the hypothalamus–pituitary–interrenal (HPI) axis by stimulating an increase in cortisol in the blood plasma (Wendelaar Bonga, 1997). Furthermore, Cu inhibits Na+/K+-ATPase activity (Li et al., 1996), and Cr inhibits

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

total ATPase (Praveena et al., 2013). Cortisol increases the CC density and the density of Na+/K+-ATPase in membrane of these cells (Dang et al., 2000) to increase NaCl uptake. In contrast, the CC reduction of CC in the gill epithelia of fish from FU20 and FU50 suggests a direct effect of contaminants on these cells by causing cellular necrosis, most likely due to organochlorine accumulation. Aldrin/dieldrin and endosulfan were related to CC density and to the fractional CC apical area reduction in the gill epithelium and the inhibition of Na+/K+-ATPase activity in fish collected from Furnas HPS, particularly in P. maculatus (Fernandes et al., 2013). The results of the present study corroborate laboratory studies with endosulfan and some metals that resulted in changes in the CC density and inhibition of Na+/K+-ATPase activity (Mazon et al., 2002; Glover et al., 2007; Camargo et al., 2009; Fernandes et al., 2013). These changes cause ionic imbalance in fish. The MC of fish may present different mucus chemical compositions related to their habitat (Calabro et al., 2005). P. maculatus, benthic fish, has higher abundances of type 1 and type 2 MCs and lower abundance of type 3 MCs than A. fasciatus, which is predominantly pelagic. Mucus-containing neutral glycoproteins have low viscosity, whereas the presence of acid glycoproteins containing sialic acid and/or sulfate esters leads to an increase in viscosity (Tibbetts, 1997; Fiertak and Kilarski, 2002). The increasing MC abundance in the gills and the alterations in MC types are mechanisms that reduce the contaminant uptake, protecting the gill epithelium (Tkatcheva et al., 2004; Banerjee and Chandra, 2005; Banerjee, 2007; Paulino et al., 2012b). The increased type 1 and type 2 MC abundances in both species collected from FU20, FU30, FU40 and FU50 suggest a mechanism that favors the washing of contaminants away from the gill surface, as these cell types produce low-viscosity mucosubstances. In the liver, which exhibited scattered (FU10 and FU40) to widely distributed lesions (FU20, FU30 and FU50), the cytoplasmic vacuolization was the most frequent hepatocyte change and can be a signal of degenerative processes involving metabolic damage (Pacheco and Santos, 2002). Numerous other alterations, such as nuclear deformation, the presence of pyknotic nuclei and the absence of nucleoli, reinforced the observations that hepatocyte degeneration occurred mainly in fish collected from FU20, FU30 and FU50 and more frequently in P. maculatus than in A. fasciatus. Numerous areas of focal necrosis found in fish from Furnas HPS reservoir, except those collected from FU10, clearly indicated that liver toxicity occurred. Hepatocellular necrosis has been frequently observed in fish from contaminated ecosystems in response to exposure to Cu and Zn (Fernandes et al., 2008), polycyclic aromatic hydrocarbons (Osório et al., 2013), and a mixture of contaminants (Camargo and Martinez, 2007; Brito et al., 2012). Melanomacrophage aggregates play an important role in the sequestration of the products of cellular degradation and in potentially toxic tissue materials, such as free radicals and catabolic breakdown products (Agius and Roberts, 2003). The increase in melanomacrophage aggregates in the liver of A. fasciatus and P. maculatus from the Furnas HPS reservoir seems to be related to liver toxicity at the more contaminated sites. Liver hyperemia was related to metabolic detoxification processes that occur in these fish. In general, the hepatocytes alterations and the high HI indices in the liver were related to organochlorine and metal contaminants present in Furnas HPS reservoir. Organochlorine and metals activate the antioxidant systems of both species from the Furnas HPS reservoir; however, the increased lipid peroxidation in the liver causes oxidative stress and the organochlorine aldrin/dieldrin is the pollutant most strongly associated with oxidative stress in the liver (Sakuragui et al., 2013). 5. Conclusions In conclusion, the concentrations of contaminants in water, their bioaccumulation in the gills and liver of fish, and the correlations with the histological changes observed in the organs and the histopathological indices at each site demonstrated that the histological changes

305

were suitable for determining the most contaminated sites in the Furnas HPS. Although other contaminants not detected or analyzed in the present study could also contribute to histological changes in the gills and liver of these species, histological biomarkers were useful biomarkers of environmental contamination, providing a definite biological endpoint of exposure. Furthermore, such biomarkers are an important tool to assess the health of fish in contaminated environments as the changes in the gills and liver affect organ function. The discrete differences in the histological changes of the gills and liver of A. fasciatus and P. maculatus collected from the Furnas HPS reservoir may be related to differences in the competence for metabolism and excretion of contaminants and/or the sensitivity to them between the two species. Acknowledgments This study was part of the P&D ANEEL PROGRAM of the Furnas Centrais Elétricas S.A. (Proc. 0394-097-2003) and was developed by the Federal University of São Carlos in conjunction with the Hydrobiology and Aquaculture Station of the FURNAS HPS. The authors participated in the CNPq/INCT in Aquatic Toxicology (Proc. 573949/2008-5). The authors are thankful to Dirceu Marzulo Ribeiro and Maria das Neves Lima Ferreira for their assistance during this study. M.M. Sakuragui, H. Sadauskas Henrique, M.G. Paulino and T.P. Benzy acknowledge fellowships from the FURNAS Centrais Elétricas S.A. and CAPES. Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.scitotenv.2014.07.122. These data include Google map of the most important areas described in this article. References Agius C, Roberts RJ. Melano-macrophage centres and their role in fish pathology. J Fish Dis 2003;26:499–509. ANVISA. Agência Nacional de Vigilância Sanitária. Agrotóxicos/Toxicologia. Disponível em: http://www.anvisa.gov.br, 2014. [Accessed May/2014]. APHA. Standard methods for the examination of water and wastewater. 18th ed. Washington: American Public Health Association; 1992. Banerjee TK. Histopathology of respiratory organs of certain air-breathing fishes of India. Fish Physiol Biochem 2007;33:441–54. Banerjee TK, Chandra S. Estimation of toxicity of zinc chloride by histopathological analysis of the respiratory organs of the air breathing ‘murrel’ Channa striata (Bloch, 1797) (Channiformes, Pisces). Vet Archiv 2005;75:253–63. Bashir FA, Shuhaimi-Othman M, Mazlan AG. Evaluation of trace metal levels in tissues of two commercial fish species in Kapar and Mersing coastal waters, peninsular Malaysia. J Environ Public Health 2012:1–10. [Article ID 352309]. Boon JP, Lewis WE, Choy MR, Allchin CR, Law RJ, De Boer J. Levels of polybrominated diphenyl ether (PBDE) flame retardants in animals representing different trophic levels of the North Sea food web. Environ Sci Technol 2002;36:4025–32. Brito IA, Freire CA, Yamamoto FY, Assis HCS, Souza-Bastos LR, Cestari MM, et al. Monitoring water quality in reservoirs for human supply through multi-biomarker evaluation in tropical fish. J Environ Monit 2012;14:615–25. Calabro C, Albanese MP, Lauriano ER, Martella S, Licata A. Morphological, histochemical and immunohistochemical study of the gill epithelium in the abyssal teleost fish Coelorhynchus coelorhynchus. Folia Histochem Cytobiol 2005;43:51–6. Camargo MMP, Martinez CBR. Histopathology of gills, kidney and liver of a neotropical fish caged in an urban stream. Neotrop Ichthyol 2007;5:327–36. Camargo MMP, Fernandes MN, Martinez CBR. How aluminium exposure promotes osmoregulatory disturbances in the Neotropical freshwater fish Prochilus lineatus. Aquat Toxicol 2009;94:40–6. Capkin E, Altinok I, Karahan S. Water quality and fish size affect toxicity of endosulfan, an organochlorine pesticide, to rainbow trout. Chemosphere 2006;64:1793–800. Capkin E, Birincioglu S, Altinok I. Histopathological changes in rainbow trout (Oncorhynchus mykiss) after exposure to sublethal composite nitrogen fertilizers. Ecotoxicol Environ Saf 2009;72:1999–2004. Cerqueira CCC, Fernandes MN. Gill tissue recovery after copper exposure and blood parameter responses in the tropical fish, Prochilodus scrofa. Ecotoxicol Environ Saf 2002;52:83–91. CONAMA Conselho Nacional do Meio Ambiente/Ministerio do Meio Ambiente. Resolucao No. 357 de 17 de marco de 2005. Disponivel em http://www.mma.gov.br/port/ conama/legiano1.cfm?codlegitipo=3&ano=2005, 2005. [accessed in October 2013]. Costa MR, Mattos TM, Borges JL, Araujo FG. Habitat preferences of common native fishes in a tropical river in Southeastern Brazil. Neotrop Ichthyol 2013;11:871–80.

306

M.G. Paulino et al. / Science of the Total Environment 497–498 (2014) 293–306

Cruz AL, Fernandes MN, Perry SF. Effect of histological processing and methacrylate sectioning on the area of gill tissue in teleost. Braz J Biol 2009;69:385–7. Dang ZC, Balm PHM, Flick G, Wendelaar Bonga SE. Cortisol increases NA+/K+-ATPase density in plasma membranes of gill chloride cells in the freshwater tilapia Oreochromis mossambicus. J Exp Biol 2000;203:2349–55. Esteves KE, Galetti Jr PM. Food partioning among some characids of a small Brazilian foodplain Lake from the Paraná River basin. Environ Biol Fishes 1995;42:357–89. Evans DH, Piermarini PM, Choe CP. The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid–base regulation, and excretion of nitrogenous waste. Physiol Rev 2005;85:97–177. Fernandes MN, Moron SE, Sakuragui MM. Gill morphological adjustments to environment and the gas exchange function. In: Fernandes MN, Glass ML, Rantin FT, Kapoor BG, editors. Fish respiration and environment. Enfield: Science Publishers; 2007. p. 93–120. Fernandes C, Fontaínhas-Fernandes A, Rocha E, Salgado MA. Monitoring pollution in Esmoriz–Paramos lagoon, Portugal: liver histological and biochemical effects in Liza saliens. Environ Monit Assess 2008;145:315–22. Fernandes MN, Paulino MG, Sakuragui MM, Ramos CA, Pereira CDS, Sadauskas-Henrique H. Organochlorines and metals induce changes in the mitochondria-rich cells of fish gills: an integrative field study involving chemical, biochemical and morphological analyses. Aquat Toxicol 2013;126:180–90. Fiertak A, Kilarski WM. Glycoconjugates of the intestinal goblet cells of four cyprinids. Cell Mol Life Sci 2002;59:1724–33. Figueiredo L, Chiavelli L, Costa W. Determination of concentration levels of organochlorine pesticides in water of Mandacaru stream in Maringá-Paraná-Brazil employing gas chromatography–mass spectrometry. Anal Lett 2013;46:1597–606. FUPAI - Fundação de Pesquisa e Assessoramento à Indústria de Itajubá. Plano diretor de recursos hídricos da bacia hidrográfica do entorno do Lago de Furnas. Diagnóstico da bacia hidrográfica. Relatório parcial 1; 2013. [305 pp.]. Glover CN, Petri D, Tollefsen K-E, Jørum N, Handy RD, Berntssen MHG. Assessing the sensitivity of Atlantic salmon (Salmo salar) to dietary endosulfan exposure using tissue biochemistry and histology. Aquat Toxicol 2007;84:346–55. Golterman HL, Clymo RS, Ohnstad MAM. Methods for physical and chemical analysis of freshwaters. IBP handbook, vol. 8. Oxford: Blackwell Scientific Publications; 1978. Heath AG. Water pollution and fish physiology. 2nd ed. Florida: Lewis Publishers; 1995. p. 644. Heleno GRV. Relatório UHE de Furnas Centrais Elétricas S.A. 1993–2003. Belo Horizonte: Plancton Consultoria Ltda; 2004. p. 110. Jin Y, Zhang X, Shu L, Chen L, Sun L, Qian H, et al. Oxidative stress response and gene expression with atrazine exposure in adult female zebrafish (Danio rerio). Chemosphere 2010;78:846–52. Li J, Lock RAC, Klaren PHM, Swarts HGP, Schuurmans Stekhoven FMAH, Wendelaar Bonga SE. Kinetics of copper inhibtion of Na+/K + −ATPase. Toxicol Lett 1996;87:31–8. Mackereth FJH, Heron J, Talling JF. Water analysis: some revised methods for limnologists. Cumbria: Scientific Publication; 1978. Mallat J. Fish gill structural changes induced by toxicants and other irritants: a statistical review. Can J Fish Aquat Sci 1985;42:630–48. Mazon AF, Cerqueira CCC, Fernandes MN. Gill cellular changes induced by copper exposure in the South American tropical freshwater fish, Prochilodus scrofa. Environ Res 2002;88A:52–63. McKim JM, Erickson RJ. Environmental impacts on the physiological mechanisms controlling xenobiotic transfer across fish gills. Physiol Zool 1991;64:39–67. Myers MS, Fournie JW. Histopatholohgical biomarkers as integrators of anthropogenic and environmental stressors. In: Adams SM, editor. Biological indicators of aquatic ecosystem stress. Bestheda: American Fisheries Soc.; 2002. p. 221–87. Navarro-Ortega A, Tauler R, Lacorte S, Barceló D. Occurrence and transport of PAHs, pesticides and alkylphenols in sediment samples along the Ebro River Basin. J Hydrol 2010;383:5–17. Negreiros NF, Santos-Wisniewski MJ, Santos RM, Rocha O. The influence of environmental factors on the seasonal dynamics and composition of the Rotifera in the Sapucaí River arm of Furnas Reservoir, MG, Brazil. Biota Neotrop 2010;10:173–82. Osório FHT, Silva LFO, Piancini LAS, Azevedo ACB, Liebel S, Yamamoto FY, et al. Water quality assessment of the Tubarão River through chemical analysis and biomarkers in the Neotropical fish Geophagus brasiliensis. Environ Sci Pollut Res 2013. http://dx.doi.org/10.1007/s11356-013-1512-5. Pacheco M, Santos MA. Biotransformation, genotoxic and histopathological effects of environmental contaminants in European eel (Anguilla anguilla L.). Ecotoxicol Environ Saf 2002;53:331–47. Paulino MG, Sakuragui MM, Fernandes MN. Effects of atrazine on the gill cells and ionic balance in a neotropical fish, Prochilodus lineatus. Chemosphere 2012a;86:1–7. Paulino MG, Souza NES, Fernandes MN. Subchronic exposure to atrazine induces biochemical and histopathological changes in the gills of a Neotropical freshwater fish, Prochilodus lineatus. Ecotoxicol Environ Saf 2012b;80:6–13.

Poleksic V, Mitrovic-Tutundzic V. Fish gills as a monitor of sublethal and chronic effects of pollution. In: Muller R, Lloyd R, editors. Sublethal and chronic effects of pollutants on freshwater fish. Oxford: Fishing News Books; 1994. p. 339–52. Praveena M, Kavitha N, Jayantha Rao K. Chromium induced alterations on total ATPases in different tissues of a fresh water fish, Labeo Rohita. Indian J Appl Res 2013;3:50–1. Ramos IP, Vidotto-Magnoni AP, Brandão H, David GS, Carvalho ED. Feeding, reproduction and population structure of Pimelodus maculatus Lacépéde, 1803 (Teleostei, Siluriformes, Pimelodiae) from Paraná basin: a review. Bol Assoc Bras Limnol 2011; 39:1–15. Ribeiro CAO, Vollaire Y, Sanchez-Chardi A, Roche H. Bioaccumulation and the effects of organochlorine pesticides, PAH and heavy metals in the eel (Anguilla anguilla) at the Camargue Nature Reserve, France. Aquat Toxicol 2005;4:53–69. Rissato SR, Galhiane MS, Ximenes VF, Andrade RMB, Talamoni JLB, Libânio M, et al. Organochlorine pesticides and polychlorinated biphenyls in soil and water samples in the Northeastern part of São Paulo State, Brazil. Chemosphere 2006;65:1949–58. Sá Junior WP. Production of planktonic biomass for feed of alevins ar the Furnas hydrobiology and hatchery station. In: Pinto-Coelho RM, Giani A, Sperling E, editors. Ecology and human impact on lakes and reservoirs in Minas Gerais with special reference to future development and management strategies. Belo Horizonte: SEGRAC; 1994. p. 133–40. Sabóia-Moraes SMT, Hernandez-Blazquez FJ, Mota DL, Bittencourt AM. Mucous cell types in the branchial epithelium of the euryhaline fish Poecilia vivipara. J Fish Biol 1996;49: 545–8. Sadauskas-Henrique H. Avaliação morfofuncional das brânquias e variáveis hematológicas de duas espécies de teleósteos (Astyanax fasciatus and Pimelodus maculatus) do reservatório da Usina Hidroelétrica de Furnas. Aspectos Fisioecológicos [Ms Thesis] São Carlos, SP, Brazil: Universidade Federal de São Carlos; 2008 [145 pp.]. Sadauskas-Henrique H, Sakuragui MM, Paulino MG, Fernandes MN. Using condition factor and blood variable biomarkers in fish to assess water quality. Environ Monit Assess 2011;181:29–42. Sakuragui MM, Paulino MG, Pereira CDS, Carvalho CS, Sadauskas-Henrique H, Fernandes MN. Integrated use of antioxidant enzymes and oxidative damage in two fish species to assess pollution in man-made hydroelectric reservoirs. Environ Pollut 2013;178: 41–51. Schwaiger J, Wanke R, Adam S, Pawert M, Wolfgang H, Triebskorn R. The use of histopathological indicators to evaluate contaminant-related stress in fish. J Aquat Ecosyst Stress Recov 1997;6:75–86. Silva DML, Camargo PB, Martinelli LA, Lanças FM, Pinto JSS, Avelar WEP. Organochlorine pesticides in Piracicaba River basin (São Paulo/Brazil): a survey of sediment, bivalve and fish. Quim Nova 2008;31:214–9. Souza IC, Duarte ID, Pimentel NQ, Rocha LD, Morozesk M, Bonomo MM, et al. Matching metal pollution with bioavailability, bioaccumulation and biomarkers response in fish (Centropomus parallelus) resident in neotropical estuaries. Environ Pollut 2013; 180:136–44. Spacie A, Hamelink JL. Alternative models for describing the bioconcentration of organics in fish. Environ Toxicol Chem 1982;1:309–20. Straskraba M, Tundisi JG. Reservoir ecosystem functioning: theory and application. In: Tundisi JG, Straskraba M, editors. Theoretical reservoir ecology and its applications. International Institute of Ecology. São Carlos, SP: Backhuys Publishers; 1999. p. 565–83. Tao S, Wen Y, Long A, Dawson R, Cao J, Xu F. Simulation of acid–base condition and copper speciation in fish gill microenvironment. Comput Chem 2001;25:215–22. Tibbetts R. The distribution and function of mucous cells and their secretions in the alimentary tract of Arrhamphus sclerolepis krefftii. J Fish Biol 1997;50:809–20. Tkatcheva V, Hyvärinen H, Kukkonen J, Ryzhkov LP, Holopainen IJ. Toxic effects of mining effluents on fish gills in a subarctic lake system in NW Russia. Ecotoxicol Environ Saf 2004;57:278–89. Triebskorn R, Telcean I, Casper H, Farkas A, Sandu C, Stan G, et al. Monitoring pollution in River Mureş, Romania, part II: metal accumulation and histopathology in fish. Environ Monit Assess 2008;141:177–88. US EPA. Test method for evaluating solid waste. Report, No. SW846, Washington, DC; 1986. US EPA. Guidance for risk characterization. Washington, DC; 1995. US EPA. Method 3620C, Florisil cleanup, Washington, DC; 2007. Van der Oost R, Beyer J, Vermeulen NPE. Fish bioaccumulation and biomarkers in environmental risk assessment: a review. Environ Toxicol Pharmacol 2003;13:57–149. Velma V, Vutukuru SS, Tchounwou PB. Ecotoxicology of hexavalent chromium in freshwater fish: a critical review. Rev Environ Health 2009;24:129–45. Wendelaar Bonga SE. The stress response in fish. Physiol Rev 1997;77:591–625. Yasser AG, Naser MD. Impact of pollutants on fish collected from different parts of Shatt Al-Arab River: a histopathological study. Environ Monit Assess 2011;181:175–82.