104 RBCIAMB | n.42 | dez 2016 | 104-112 André Favaro Environmental Engineer by the Federal University of Triângulo Mineiro (UFTM) – Uberaba (MG), Brazil. Daniele Trenas Granados Environmental Engineer by the Federal University of Triângulo Mineiro (UFTM) – Uberaba (MG), Brazil. Alex Garcez Utsumi Master in Cartographic Sciences by the State University of São Paulo Júlio de Mesquita Filho (Unesp) – Presidente Prudente (SP). Professor at the Department of Environmental Engineering, Institute of Technology and Exact Sciences, Universidade Federal do Triângulo Mineiro (UFTM) – Uberaba (MG), Brazil. Ana Carolina Borella Marfil Anhê PhD in Health Sciences by the Oswaldo Cruz Foundation (FIOCRUZ) – Belo Horizonte (MG). Professor at the Department of Environmental Engineering, Institute of Technology and Exact Sciences, UFTM – Uberaba (MG), Brazil. Ana Paula Milla dos Santos Senhuk PhD in Sciences by the University of São Pauo (USP) – Ribeirão Preto (SP). Professor at the Department of Environmental Engineering, Institute of Technology and Exact Sciences, UFTM – Uberaba (MG), Brazil. Corresponding address: Ana Paula Milla dos Santos Senhuk – Universidade Federal do Triângulo Mineiro – Avenida Dr. Randolfo Borges Júnior, 1.250 – 38064‑200 – Uberaba (MG), Brazil. E‑mail: ana.santos@icte.uftm.edu.br; anapmilla@yahoo.com.br ABSTRACT Water toxicity was biomonitoried by Tradescantia pallida, relating the results to physicochemical and qualitative analysis. Two clusters were formed by hierarchical analysis of micronuclei percentage (MN), dissolved oxygen (DO), electrical conductivity (EC) and turbidity. The first cluster corresponded to a stream that receives urban untreated sewage, where the highest MN (6.1%) corroborated the physicochemical results (1.90 mg.L‑1 DO, 575.0 μS.cm‑1 and 115.6 NTU). The other cluster includes sampling sites with pollutants from agriculture, livestock and industry. Some of them presented high MN even with acceptable values of physicochemical parameters, according to the Brazilian legislation. So for a better quality information on water pollution monitoring it is essential to recognize the variables that affect the biotic components, in a holistic approach of aquatic ecosystem. In this sense the bioassay with T. pallida becomes a useful tool that can be applied in many tropical and subtropical regions, because of its widely distribution. Keywords: environmental health; water quality; bioindicator; Trad‑MN; bioassay. RESUMO A toxicidade da água foi biomonitorada com Tradescantia pallida, relacionando os resultados com análises físico‑químicas e qualitativa. Dois grupos foram formados por meio de análise hierárquica da porcentagem de micronúcleos (MN), oxigênio dissolvido (OD), condutividade elétrica e turbidez. O primeiro grupo corresponde a um córrego que recebe efluente urbano não tratado, onde a maior porcentagem de MN (6,1%) corroborou os resultados físico‑químicos (1,90 mg.L‑1 de OD, 575,0 μS.cm‑1 e 115,6 UNT). O outro grupo inclui córregos que recebem poluentes da agricultura, pecuária e indústria. Alguns deles apresentaram alta porcentagem de MN mesmo com valores aceitáveis de parâmetros físico‑químicos, de acordo com a legislação brasileira. Assim, para monitoramento da poluição da água é essencial reconhecer as variáveis que afetam os componentes bióticos, numa abordagem holística do ecossistema aquático. Neste sentido, o bioensaio com T. pallida torna‑se uma ferramenta útil que pode ser aplicada em muitas regiões tropicais e subtropicais, devido à sua ampla distribuição geográfica. Palavras‑chave: saúde ambiental; qualidade da água; bioindicador; Trad‑MN; bioensaio. DOI: 10.5327/Z2176-947820160130 BIOMONITORING OF SURFACE WATER TOXICITY RELATED TO URBAN AND INDUSTRIAL WASTEWATER RELEASE BIOMONITORAMENTO DA TOXICIDADE DE ÁGUAS SUPERFICIAIS RELACIONADA AO LANÇAMENTO DE EFLUENTES URBANO E INDUSTRIAL Biomonitoring of surface water toxicity related to urban and industrial wastewater release 105 RBCIAMB | n.42 | dez 2016 | 104-112 INTRODUCTION Water quality issues are a major challenge that humani‑ ty is facing in this century, especially in developing coun‑ tries, where it is urgent to improve sanitation and en‑ vironmental conservation, in a current water shortages scenario (SCHWARZENBACH et al., 2010; LOYOLA & BINI, 2015). Without adequate treatment of wastewater, the health of the water body receptor is impacted, chang‑ ing the interactions between biotic and abiotic compo‑ nents of the aquatic ecosystem and also affecting other water users. The lack of proper sanitation is still a great problem of freshwater quality in Brazil. Over 70% of cit‑ ies have no municipal basic sanitation policy and 62% of the urban population is covered with collecting sewage, with only 30% of sewage being treated (IBGE, 2010). About 6,000 deaths per year could be prevented if all Bra‑ zilian regions had equal adequate sanitation (FUNASA, 2010). Constant urbanization and pressure leads to a significant increase in environmental preventable dis‑ eases, when there is no efficiency in water use prac‑ tices, appropriate land use or pollution control. Con‑ sequently, there is an extra cost to the public coffers, especially with hospitalizations for health problems related to poor water quality (WHO, 2015). The development of efficient methods for diagnosing environmental quality is an important advance in the search for solutions to social and environmental im‑ pacts caused by inadequate water management (BUSS; BAPTISTA; NESSIMIAN, 2003). It is observed the need for a more detailed diagnosis that considers interaction of management and planning instruments of water re‑ sources in sanitation (ANA, 2013). In this sense, bio‑ monitoring has been proven effective in the integrat‑ ed analysis of water quality, considering the biological system as a whole, in a complementary manner to the physicochemical parameters traditionally used. Due to the increasing restriction of the use of laborato‑ ry animals, tests with bacteria, plants and cell cultures have been gaining prominence. Among these, plant biomonitoring stands out for providing information on the bioavailability of contaminants and can be used to evaluate the effects of contaminants at low concentra‑ tions, such as heavy metals (MIELLI; SALDIVA; UMBU‑ ZEIRO, 2009). These bioindicators can also be used for determining contamination patterns in large areas and over time, with more representative results for the ex‑ posed population. The genus Tradescantia belongs to the Commelinace‑ ae family and comprises about 500 species, which are found mainly in subtropical and tropical areas (MIŠÍK et al., 2011). Tradescantia pallida (Rose) D.R.Hunt is an ornamental herbaceous species, which has been used in outdoor air quality studies (SANTOS et al., 2015), genotoxicity of sewage (THEWES; ENDRES JR.; DROSTEI, 2011) and exposure to ozone, naphthalene and X‑rays (SUYAMA et al., 2002; ALVES et al., 2008; LIMA; SOUZA; DOMINGOS, 2009). As well as others species of the same gender, T. pallida has also proven to be an efficient pollution bioindicator, for its easy propagation and high sensitivity to genotoxic agents. By its favorable genetic characteristics, it is pos‑ sible to analyze the environmental toxicity by Trad‑MN bioassay that assesses mutation in pollen grain mother cells (tetrads) (MA, 1981; RODRIGUES et al., 1997). As a complement of biomonitoring studies, the Rapid Assessment Protocols (RAP) of rivers provide environ‑ mental diagnosis in a wide view, and have been used by several researchers in order to facilitate the under‑ standing of the holistic approach of aquatic ecosystems. These protocols are a useful environmental manage‑ ment tool to compare various areas of the same hydro‑ graphic basin, characterizing them in a qualitative way (CALLISTO et al., 2002; MACHADO et al., 2015). In this context, this study aimed to evaluate the toxicity of a watershed polluted by urban and industrial waste‑ water, using T. pallida as a bioindicator and relating the results to physicochemical and qualitative analyses. MATERIAL AND METHODS Study area The study area is located at Uberaba city, Triângu‑ lo Mineiro, in Minas Gerais State, with more than 318,000 inhabitants, and about 96% of them living in urban areas (IBGE, 2015). The research was conducted Favaro, A. et al. 106 RBCIAMB | n.42 | dez 2016 | 104-112 at the Conquistinha River watershed, which receives about 22% of untreated urban and industrial waste‑ waters. The main urban expansion area of Uberaba is located in this watershed. Conquistinha River is a direct tributary of the Grande River, and can be a source of water supply for human consumption in the future. Data collection We collected water samples in eight sites in Conquis‑ tinha river watershed, more precisely in the streams: Desbarrancado (A), Três Córregos (B), Sucuri (C and D) and Conquistinha river (E‑H) (Figure 1). Data collection campaigns were made in April 2015, at the end of the rainy season. The following water physicochemical parameters were analyzed for each sampling site: temperature, Figure 1 – Sampling sites at the Conquistinha River watershed, Uberaba, Minas Gerais, Brazil. 187500 192000 196500 201000 205500 78 11 00 0 78 04 50 0 77 98 00 0 77 91 50 0 C. Desbarrancado C. Três Córregos R. Conquistinha R. Conquistinha C. Sucuri A B C E F D G H W N E S 0 1.5 3 6 km Datum: SIRGAS 2000 Projeção: UTM Zona: 23 S Biomonitoring of surface water toxicity related to urban and industrial wastewater release 107 RBCIAMB | n.42 | dez 2016 | 104-112 dissolved oxygen (DO), pH, electrical conductivity and turbidity. They were analyzed in field through Vernier sensors coupled to a LabQuest datalogger. Flower stems of T. pallida, with approximately 10cm length, were placed in tap water for 24 hours for adaptation. Then, they were exposed for over six hours to the negative control (tap water), positive control (0.2% formaldehyde) and water samples from the eight sites at the Conquistinha River watershed. After expo‑ sure, the inflorescences were placed again in tap water for a period of 24 hours to recover, then fixed in a solution of acetic acid and ethanol (1:3) and stored in 70% ethanol. Slides for micronuclei analysis were prepared as pro‑ posed by Ma (1981). The micronucleus count was con‑ ducted through an optical microscope at 400x magni‑ fication in a random group of 300 tetrads. At least five slides per sample site were analyzed. The frequency of micronuclei was expressed in percentage. The environmental assessment of the surroundings of each sample site was done by applying a RAP, modified from Cal‑ listo et al. (2002) (Chart 1). A score between 0 and 4 was attributed to each parameter, which corresponds to the environmental integrity. The analyzed stretches were clas‑ sified according to the environmental quality by adding the scores of each parameter in four condition categories: good (31 to 40 pts), regular (21 to 30 pts) and poor (0 a 20 pts). The stretches were also evaluated for human activities on site by their presence or absence — observed during data collection — in a qualitative analysis. Statistical analysis Micronuclei (MN) data were submitted to an analysis of variance (ANOVA), in order to determine differenc‑ es between the sampling sites. When the level of sig‑ nificance was reached (p<0.05), we applied the Tukey test for multiple comparisons. The results of the phys‑ icochemical parameters and biomonitoring were ana‑ lyzed by the Pearson correlation coefficient. A cluster analysis (CA) was also used in order to de‑ tect the similarity groups between the sampling sites, based on MN and physicochemical data. We calculated Parameters/ score Good (4) Regular (2) Poor (0) 1. Occupation of the banks of the water body (main activity) Natural vegetation Pasture/ agriculture Residential/Commercial/ Industrial 2. Erosion on the banks of the water body and siltation in its bed Absent Moderate Severe 3. Anthropogenic changes Absent Changes from domestic sources (sewage, garbage) Changes from industrial sources (factories, steel mills, channeling the river course…) 4. Full coverage in water bed Partial Total Absent 5. Water odor Absent Sewage Industrial/ oil 6. Water oily Absent Moderate Abundant 7. Water transparency Transparent Blurred Opaque or colored 8. Sediment odor Absent Sewage Industrial/ oil 9. Sediment oily Absent Moderate Abundant 10. Sediment types Stone/ gravel Mud/ sand Cement/ canalized Chart 1 – Rapid assessment protocol of habitat diversity in watershed stretches. Favaro, A. et al. 108 RBCIAMB | n.42 | dez 2016 | 104-112 hierarchical agglomerative CA on the normalized data set by means of the Ward’s method, using squared Eu‑ clidean distances as a measure of similarity. Results of CA were represented with a dendrogram. RESULTS AND DISCUSSION Results of the genotox test with T. pallida physicochemi‑ cal and qualitative analyses are shown in Table 1, as well as the physicochemical and qualitative analyses. All sam‑ pling sites showed greater toxicity, by MN percentages, than the negative control (1.1%±0.4). In our results, MN ranged from 1.5 to 6.1%, similar to those found in sam‑ ples collected in rivers of southern and southeastern Brazil (UMBUZEIRO et al., 2007; COSTA et al., 2014). Trad‑MN bioassays are more sensitive to environmen‑ tal contaminants than other plant bioassays, as root tip cells and somatic stamen hair analysis, used in paralel experiments (MIŠÍK & MICIETA, 2002). Most studies were conducted with a hybrid between T. hirsutiflora and T. subacaulis, the clone 4430; but recently, sever‑ al Brazilian research groups have been using T. pallida for environmental monitoring, obtaining satisfactory results (MIŠÍK et al., 2011). This species is broadly dis‑ tributed, more adapted to tropical climates and more resistant to deseases than the clone. Differences in the percentage of MN indicate some spa‑ tial variation for water pollution at the watershed studied. Water collected in sampling site A presented the highest MN (6.1%±1.0). This value was more than the double of any other sampling site and 25% higher than the positive control. The second highest MN (5.1%±1.0) was found in the last sampling site (H). Water samples from C, D and E had the lowest MN (2.1% on average). The A sampling site had the highest electric conductiv‑ ity (575.0 μS.cm‑1) and turbidity (NTU 115.6), and the lowest DO concentration (1.90 mg.L‑1). With the ex‑ ception of that site, DO concentrations ranged from 6.2 to 8.1 mg.L‑1 with an average of 7.5 mg.L‑1 (Chart 1). Only B and E sites presented low values of electrical conductivity, with less than 50 μS.cm‑1. On average, pH was 7.3 and temperature was 22.4°C. Two clusters were formed by the hierarchical analysis of MN, DO, electrical conductivity and turbity (Figure 2). The first one corresponds only to sampling site A. By cal‑ culating the Pearson coefficient, MN had a strong cor‑ relation with the DO concentration (p<‑0.79) and moder‑ ate correlation with electrical conductivity and turbidity (p=0.57 and p=0.63, respectively). The moderate correlation of Trad‑MN bioassay results with electrical conductivity and turbidity indicates that this biomonitoring cannot exclude traditional analyses, Sampling site MN (%) EC (µS.cm‑1) Turbidity (NTU) DO (mg.L‑1) pH T (°C) RAP (score) A 6.1±1.0a 575 115.60 1.90 7.20 21.60 Poor (18) B 4.1±0.9b 36 22.40 6.22 7.56 25.28 Regular (28) C 2.3±0.3c 226 17.50 7.95 7.10 21.50 Good (32) D 1.5±0.4c 172 50.20 8.05 7.10 22.60 Good (32) E 2.5±0.7c 30 20.20 7.80 7.80 27.70 Good (32) F 3.0±0.7bc 250 32.90 6.86 7.30 20.20 Regular (24) G 2.8±0.6bc 206 56.20 8.10 7.10 21.10 Good (38) H 5.1±1.0ab 219 47.20 7.47 7.00 19.60 Poor (22) Table 1 – Environmental diagnosis of the Conquistinha River watershed, Uberaba, Minas Gerais, Brazil. MN: micronuclei frequencies ± standard deviation; sifferent lowercase indicate significant differences by Tukey test (p<0.05); EC: electrical con‑ ductivity; DO: dissolved oxygen; T: temperature; RAP: Rapid Assessment Protocol. Source: modified from Callisto et al., 2002. Biomonitoring of surface water toxicity related to urban and industrial wastewater release 109 RBCIAMB | n.42 | dez 2016 | 104-112 but complement them. It is noteworthy that the Trad‑MN bioassay indicates the toxicity of water, but should be used with caution, because of its low discriminatory po‑ wer, requiring further analysis to infer the source of pol‑ lution. Nevertheless, positive MN results are important in genotoxic risk assessments, indicating the need for fur‑ ther testing with some more relevant species bioindicator (MIELLI; SALDIVA; UMBUZEIRO, 2009). The percentage of MN found in sampling site A was similar to Cristais River (6.2%), downstream of a textile industry, at southeastern Brazil (UMBUZEIRO et al., 2007). This site is located at Desbarrancado Stream, one of the first tribu‑ taries of Conquistinha River, with the nascent and most of its length within the urban perimeter, polluted especially by urban untreated sewage. The A site could be viewed separately in one of the clusters of Figure 2, since the poor water quality inferred by biomonitoring results corrobo‑ rated the physicochemical parameters analyzed. The A site does not fit into any of the four freshwater ratings established by CONAMA Resolution n° 357/2005 (CONAMA, 2005), according to DO concentration and turbidity results. Values above the threshold of electri‑ cal conductivity (100 μS.cm‑1) were observed in most of the sampling sites, especially in site A (CETESB, 2014). High electrical conductivity of the water may be explained by diffuse pollution with the flow of sediments and rural effluents, which is intensified in the absence of riparian vegetation, as evidenced in the watershed studied. The other cluster of Figure 2 includes sampling sites that receive pollutants from agriculture, livestock and industrial effluents. Water samples collected in some of these sites, as B and H, caused high toxicity to T. pallida, even with acceptable physicochemical conditions according to the current Brazilian legislation. The B site presented micronu‑ clei about 40% higher than the E site, also considered no impacted, probably due to its location close to the urban area and a few meters from an agricultural substrate fac‑ tory. The highest percentage of micronuclei observed in H site may be related to contamination by industrial effluent, since this site is located in one of the industrial districts of the city, 1.5 km from of the Conquistinha River mouth, in Grande River, an important source of water supply. After about 3 km from the confluence with the Des‑ barrancado stream (A site), the Conquistinha River is considered impacted, with an increase of up to 700% of the electrical conductivity (CETESB, 2014). However, regarding to turbidity and DO concentration, all sites fit between freshwater classes I and II, which can be intended for human consumption supply after conven‑ tional treatment (CONAMA, 2005). Figure 2 – Dendrogram showing clustering of sampling sites according to water quality characteristics. 100% 80% 60% 40% 20% A B E H C F D G 0 Favaro, A. et al. 110 RBCIAMB | n.42 | dez 2016 | 104-112 According to the environmental assessment given by the application of the RAP in three tributaries of Conquistinha River, the A site was severely impacted (18 pts), the B site was moderately changed (28 pts) and C and D sites were in better conservation condition (32 pts). The four sam‑ pling stretches at Conquistinha River (E‑H) ranged from regular to good environmental conditions (22 to 32 pts). In general, almost 70% of the evaluated stretches had margins with high degree of deforestation, soil erosion process and siltation of the stream bed, with the use of land surrounding mainly for sugarcane cultivation (44%) and pasture (33%). Domestic sewage release was observed in sampling site A, and industrial sewage, probably in the last sampling site (H). Bad sewage smell was also evident in A, E and H sites. Besides the lack of sanitation evident in the watershed studied, the high degree of deforestation in the analyzed stretches compromises the future availability of water re‑ sources. Currently, there are extremely few water resourc‑ es in good conditions of conservation outside protected areas, even with the obligation to preserve the riparian vegetation in Permanent Preservation Areas tracks (PPAs). Deforestation has been demonstrated to cause several changes in environment biodiversity, and influence lit‑ ter accumulation in the remnant patches (MACHADO et al., 2015). Among various impacts of deforestation, there is the reduction of water infiltration into the soil and supply aquifers, increasing runoff and soil loss. Thus, the conservation of watersheds riparian vegeta‑ tion offers more and better water quality. CONCLUSIONS Before the occurrence of extreme events of prolonged droughts in a current context of climate changes, ag‑ riculture and urban expansion, coupled with the ab‑ sence of conservation measures, greatly increase the vulnerability of aquatic ecosystems. River basins can have their hydrological behavior altered, which leads to the urgent need to improve the water infrastructure in Brazil, facing the actual water shortage. Genotox test with T. pallida was sensitive to the complex mixture of water pollutants in the watershed studied, since the percentage of MN found in the sampling sites were, on average, 2.2 times higher than the negative con‑ trol. It was also possible to detect spatial variation in water pollution, even with acceptable values of physicochemical parameters, according to the Brazilian legislation. So for better quality information on water pollution monitoring, it is essential to recognize the variables that affect the biotic components, in a holistic ap‑ proach of the aquatic ecosystem. In this sense the bio‑ assay with T. pallida becomes a useful tool, which may be applied in many tropical and subtropical regions, due to its widely distribution. ACKNOWLEDGEMENTS The two first authors received grants from Minas Gerais State Foundation‑FAPEMIG. REFERENCES AGÊNCIA NACIONAL DE ÁGUAS – ANA. Report of water resources brings balance of the situation and the management of freshwater in Brazil. Brasília: ANA, 2013. Disponible at: . Acesso em: 12 dez. 2015. ALVES, E. S.; SOUZA, S. R.; PEDROSO, A. N.; DOMINGOS, M. Potential of the Trad‑MCN assay applied with inflorescences of Tradescantia pallida “Purpurea” for evaluating air contamination by naphthalene. Ecotoxicology and Environmental Safety, v. 71, p. 717‑721, 2008. BUSS, D. F.; BAPTISTA, D. F.; NESSIMIAN, J. L. Conceptual basis for the application of biomonitoring on stream water quality programs. Reports in Public Health, v. 19, n. 2, p. 465‑473, 2003. Biomonitoring of surface water toxicity related to urban and industrial wastewater release 111 RBCIAMB | n.42 | dez 2016 | 104-112 CALLISTO, M.; FERREIRA, W. R.; MORENO, P.; GOULART, M.; PETRUCIO, M. Aplicação de um protocolo de avaliação rápida da diversidade de habitats em atividades de ensino e pesquisa (MG‑RJ). Acta Limnologica Brasiliensia, v. 14, n. 1, p. 91‑98, 2002. COMPANHIA AMBIENTAL DO ESTADO DE SÃO PAULO – CETESB. Surface water quality report of São Paulo state. São Paulo: CETESB, 2014. Disponible at: . Acesso em: 12 dez. 2015. CONSELHO NACIONAL DO MEIO AMBIENTE – CONAMA. Resolution n.° 357/2005. Brasília: Ministry of Environment, 2005. p. 58‑63. COSTA, G. M.; CASSANEGO, M. B. B.; PETRY, C. T.; BENVENUTI, T.; RUBIO, M. A. K.; RODRIGUES, M. A. S.; DROSTE, A. Chemical and genotoxic potential monitoring for diagnosis of water bodies quality. Revista Brasileira de Ciências Ambientais, v. 32, p. 65‑74, 2014. FUNDAÇÃO NACIONAL DE SAÚDE – FUNASA. Impacts on health and the SUS due to diseases related to inadequate environmental sanitation. Brasília: Health Ministry/FUNASA, 2010. 248p. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA – IBGE. Cities. 2015. Disponível em: . Acesso em: 12 Dec. 2015. ______. National survey on basic sanitation. Rio de Janeiro: IBGE, 2010. LIMA, E. S.; SOUZA, S. R.; DOMINGOS, M. Sensitivity of Tradescantia pallida (Rose) Hunt. “Purpurea” Boom to genotoxicity induced by ozone. Mutation Research, v. 675, p. 41‑45, 2009. LOYOLA, R.; BINI, L. M. Water shortage: a glimpse into the future. Natureza & Conservação, v. 13, p. 1‑2, 2015. MA, T. H. Tradescantia micronucleus bioassay and pollen tube chromatic aberration test for in situ monitoring and mutagens screening. Environmental Health Perspectives, v. 37, p. 85‑90, 1981. MACHADO, C. S.; ALVES, R. I.; FREGONESI, B. M.; BEDA, C. F.; SUZUKI, M. N.; TREVILATO, R. B.; NADAL, M.; DOMINGO, J. L.; SEGURA‑MUÑOZ, S. I. Integrating three tools for the environmental assessment of the Pardo River, Brazil. Environmental Monitoring and Assessment, v. 187, p. 569, 2015. MIELLI, A. C.; SALDIVA, P. H. N.; UMBUZEIRO, G. A. Comparação entre as Respostas do Clone 4430 e Tradescantia pallida no Teste de Micronúcleos (Trad‑MN). Journal of the Brazilian Society of Ecotoxicology, v. 4, n. 1‑3, p. 49‑54, 2009. MIŠÍK, M.; MA, T. H.; NERSESYAN, A.; MONARCA, S.; KIM, J. K.; KNASMUELLER, S. Micronucleus assays with Tradescantia pollen tetrads: an update. Mutagenesis, v. 26, n. 1, p. 215‑221, 2011. MIŠÍK, M.; MICIETA, K. Tradescantia micronucleus and Vicia chromosome anatelophase assays in monitoring of genotoxicity of urban soil. Water, Air, and Soil Pollution, v. 141, p. 181‑187, 2002. RODRIGUES, G. S.; MA, T.‑H.; PIMENTEL, D.; WEINSTEIN, L. H.; ICHIKAWA, S. Tradescantia bioassays as monitoring systems for environmental mutagenesis: a review. Critical Reviews in Plant Sciences, v. 16, n. 4, p. 325‑359, 1997. SANTOS, A. P. M.; SEGURA‑MUÑOZ, S. I.; NADAL, N.; SCHUHMACHER, M.; DOMINGO, J. L.; MARTINEZ, C. A.; MAGOSSO TAKAYANAGUI, A. M. Traffic‑related air pollution biomonitoring with Tradescantia pallida (Rose) Hunt. cv. purpurea Boom in Brazil. Environmental Monitoring and Assessment, v. 187, n. 2, 2015. SCHWARZENBACH, R. P.; EGLI, T.; HOFSTETTER, T. B.; GUNTEN, U. V.; WEHRLI, B. Global Water Pollution and Human Health. Annual Review Environmental Resources, v. 35, p. 109‑136, 2010. Favaro, A. et al. 112 RBCIAMB | n.42 | dez 2016 | 104-112 SUYAMA, F.; GUIMARÃES, E. T.; LOBO, D.‑‑J.A.; RODRIGUES, G. S.; DOMINGOS, M.; ALVES, E. S.; CARVALHO, H. A.; SALDIVA, P. H. N. Pollen mother cells of Tradescantia clone 4430 and Tradescantia pallida var. purpurea are equally sensitive to the clastogenic effects of X‑rays. Brazilian Journal of Medical and Biological Research, v. 35, p. 127‑129, 2002. THEWES, M. R.; ENDRES JR., D.; DROSTEI, A. Genotoxicity biomonitoring of sewage in two municipal wastewater treatment plants using the Tradescantia pallida var. purpurea bioassay. Genetics and Molecular Biology, v. 34, n. 4, p. 689‑693, 2011. UMBUZEIRO, G. A.; COIMBRÃO, C. A.; KUMMROW, F.; LOBO, D. J. A.; SALDIVA, P. H. N. Mutagenic activity assessment of Cristais River, São Paulo, Brazil, using the blue rayon/Salmonella microsome and the Tradescantia pallida micronuclei assays. Journal of Brazilian Society of Ecotoxicology, v. 2, p. 163‑171, 2007. WORLD HEALTH ORGANIZATION – WHO. Water Sanitation Health. 2015. Disponível em: < http://www.who.int/water_ sanitation_health/diseases‑risks/diseases/en/>. Acesso em: 12 dez. 2015.