Main content area

A screening test based on hematological and histological biomarkers to evaluate the environmental impacts in tambaqui (Colossoma macropomum) from a protected area in Maranhão, Brazilian Amazon

Pinheiro-Sousa, Débora Batista, Torres Junior, Audalio Rebelo, Silva, Dilson, Santos, Ricardo Luvizotto, Fortes Carvalho Neta, Raimunda Nonata
Chemosphere 2019 v.214 pp. 445-451
Colossoma macropomum, acridine orange, biomarkers, blood sampling, conservation areas, environmental impact, environmental monitoring, eosin, erythrocytes, fish, hematologic tests, histology, histopathology, mathematical models, necrosis, physical chemistry, prediction, probability, rivers, tissues, water pollution, Amazonia, Brazil
Nowadays biomonitoring programs can benefit with mathematical models able to correlate biomarkers to monitor water pollution. The aim of this study was to develop a screening test based on hematological parameters and histological lesions in tambaqui (Colossoma macropomum), to allow the assessment of environmental impacts on fish inhabiting a protected area in Maranhão inside of Brazilian Amazon. Samples collected during three years (2012, 2013 and 2014) were grouped by season (dry and rainy) Water samples were also collected for physical chemistry analysis. Blood samples were stained with Acridine Orange to detect micronuclei and erythrocyte abnormalities. Gill tissues were stained with hematoxylin and counterstained with alcoholic eosin, and histopathological lesions were scored on a scale of 1–3, being 1 = minimal pathological importance, 2 = moderate pathological importance and 3 = marked pathological importance. A screening test for evaluating environmental impact was developed by fitting the measured data (necrosis, erythrocyte abnormalities, number of micronuclei) from tambaqui. A three-dimensional surface was fit to the empirical data. Our proposed model predicted the probability of necrosis (observed in euthanized animals) based on the numbers of micronuclei and abnormal erythrocytes (observed in blood samples from live animals) (correlation coefficient R = 0.89). The methodology could be applied for predicting contamination histories (chronic pollution that induces branchial lesions) in rivers using the micronucleus and erythrocyte abnormalities of the fishes (with a simple blood sample).