Submitted by Isaac Francisco de Souza Dias (isaac.souzadias@ufpe.br) on 2015-05-26T14:55:46Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO Marcus Vinicius Alves dos Santos.pdf: 2498912 bytes, checksum: b27022556e6dc4f6fed3d4cfc7f0653e (MD5) / Made available in DSpace on 2015-05-26T14:55:46Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO Marcus Vinicius Alves dos Santos.pdf: 2498912 bytes, checksum: b27022556e6dc4f6fed3d4cfc7f0653e (MD5) Previous issue date: 2014-11-17 / CAPES / Sistemas de tratamento contendo lagoas de estabilização, maturação ou polimento podem produzir elevadas densidades de fitoplancton contendo cianobactérias, que são lançadas nos corpos d’água receptores. Muitas espécies de cianobactérias produzem cianotoxinas que podem acarretar problemas de saúde pública. A aplicação de filtro de pedra após essas lagoas de poderia ser uma alternativa atrativa para a remoção de sólidos em suspensão e, por conseguinte, o fitoplancton. Entretanto, pouco se sabe sobre a remoção de cianobactérias em filtro de pedra e nem a potencialidade de produção de toxicidade das mesmas cultivadas nesse ambiente. No presente trabalho, avaliou-se a eficiência de filtro de pedra de fluxo horizontal na remoção de sólidos e de cianobactérias e da matéria orgânica residual, através da determinação de parâmetros físico-químicos. Alem disso foi verificado, por meio e técnicas de biologia molecular, a presença do gene mcyE, indicador de toxinas produzidas pelo gênero Microcystis sp. Para isso foram monitorados dois filtros em paralelo durante 230 dias de experimento com coleta a cada 15 dias. Os resultados revelam que os filtros de pedra eram semelhantes (teste T) e foram bastante efetivos na remoção da DBO residual, alcançando no efluente os valores de 5,0±1,4mg.L-1 para ambos os filtros. A eficiência média de remoção de SST foi 65 % e 66% correspondendo a valores de 33±17,4 e 32±15,3 mg.L-1para os efluentes dos filtros 1 e 2, respectivamente. Já a remoção média do número total de células do fitoplâncton foi de 98% para ambos os filtros, com densidade média de cianobactérias no efluente do filtro 1 de 2.464±766 cel.mL-1 e de 2.448±728 cel.mL-1 para o filtro 2, ficando muito abaixo dos limites preconizados na resolução Conama 357/05 para as águas doces superficiais. Não foram encontrados os gêneros específicos do gene mcyE, indicando a ausência de toxinas produzidas porMicrocystis sp., em todas as amostras analisadas. Pode-se concluir que os filtros de pedra são eficientes na remoção de sólidos em suspensão e cianobactérias e que, nas condições estudadas, não favorecem o desenvolvimento de organismos com potencialidade de produção de toxinas.

Dans cette étude de trois lacs sujets aux efflorescences de cyanobactéries, nous avons examiné la diversité des bactéries diazotrophes et des cyanobactéries toxiques. Nous avons tenté de définir les facteurs environnementaux influençant la composition des communautés phytoplanctoniques, la concentration ainsi que la composition des microcystines (MCs). Nous avons émis l’hypothèse que l’azote jouerait un rôle majeur dans le façonnement des communautés cyanobactériennes et influencerait la concentration et composition des MCs. Des concentrations de cette toxine ainsi que le gène mcyE codant pour l’enzyme microcystine synthétase ont été détectés à chaque échantillonnage dans tous les lacs. L’azote, particulièrement sous sa forme organique dissoute (AOD) ainsi que la température de l’eau étaient les facteurs environnementaux expliquant le mieux les concentrations des MCs, tandis que la biomasse de Microcystis spp. était globalement le meilleur prédicteur. Le gène nifH codant pour l’enzyme nitrogénase (fixation d’azote) a aussi été détecté dans chaque échantillon. Malgré les concentrations faibles en azote inorganique dissous (AID) et les densités importantes d’hétérocystes, aucun transcrits du gène n’a été détecté par réverse-transcription (RT-PCR), indiquant que la fixation d’azote n’avait pas lieu à des niveaux détectables au moment de l’échantillonnage. De plus, le pyroséquençage révèle que les séquences des gènes nifH et mcyE correspondaient à différents taxons, donc que les cyanobactéries n’avaient pas la capacité d’effectuer les deux fonctions simultanément. / In this study of three eutrophic lakes prone to cyanobacterial blooms, we examined the diversity of diazotrophic bacteria and toxic cyanobacteria. We evaluated the environmental factors effects on the community composition, the concentration and composition of the family of toxins microcystins (MCs). Since the assimilation of nitrogen and the synthesis of MCs in cyanobacteria are thought to be under the same control of the NtcA transcriptor, we hypothesised that nitrogen played a major role in shaping cyanobacterial communities and influenced indirectly the concentration and composition of MCs. The mcyE gene coding for the microcystin synthetase enzyme and MCs concentrations were detected at each sampling date in all lakes. Nitrogen, particularly under its organic dissolved form (DON) as well as water temperature were the environmental factors explaining the most variation in MC concentration although Microcystis spp. biomass was overall the best predictor. The nifH gene coding for the nitrogenase enzyme (N2-fixation) was also detected at all times. Even though the concentrations of dissolved inorganic nitrogen were relatively low, and that heterocysts were present in high densities, no nifH transcripts were detected by RT-PCR, indicating that no N2-fixation was going on at detectable levels at the time of sampling. Moreover, pyrosequencing revealed that sequences of the genes nifH and mcyE corresponded to different taxa, thus cyanobacteria did not have the capacity to perform both functions simultaneously.

The western embayment of Lake Ontario, Hamilton Harbour, is one of the most polluted sites in the Laurentian Great Lakes and in recent years has seen a reoccurrence of cyanobacterial blooms. This study uses a multidisciplinary approach to examine the presences of toxic Cyanobacteria in the harbour in order to gain insight into these recurrent blooms. Microscopic analyses of phytoplankton samples collected during the 2009 summer-fall sampling season from two locations within the harbour showed the spatial and seasonal diversity of the contemporary cyanobacterial community. Microcystis colonies relative abundances in relation to total algal numbers were estimated. The lowest and highest relative abundances of Microcystis in the phytoplankton population were 0.6% and 9.7%, respectively, and showed seasonal variability between stations. Fourteen cyanobacterial genera comprising six families and three orders were identified and for which the most abundant filamentous genera during the summer-fall sampling season were Planktothrix, Aphanizomenon and Limnothrix. Potential microcystin producers Microcystis, Planktothrix, Aphanizomenon and Dolichospermum were also present and during the sampling period Microcystis was recorded at both stations on all dates, however, its relative abundance was below 10 % throughout the study period. The composition and abundance of filamentous cyanobacteria were observed to be positively statistically correlated to water quality environmental parameters dissolved nitrates (NO3/NO2), dissolved inorganic carbon (DIC), and conductivity. Redundancy analysis (RDA) found that 53.35% total variance of Aphanizomenon was correlated to low water column NO3/NO2 and conductivity, and higher water column DIC. 58.13% of the relative abundance of Planktothrix was correlated to high concentrations of dissolved nitrates, while 51.69% of total variance of Limnothrix was correlated to higher DIC and lower water column dissolved nitrate concentrations. Information about past cyanobacterial communities was obtained from the sediment core analysis, using paleolimnological and modern molecular methods. The age of the 100.5 cm long sediment core retrieved from the deepest part of Hamilton Harbour was established to be 140 years (1869-2009), using the Constant Rate of Supply (CRS) 210Pb age model. This age was not sufficient to provide information of harbour’s environmental conditions, presence of the blooms, and triggers for their occurrence before European settlement in the area. Results of the HPLC analysis of fossil pigments indicated that the dominant members of the algal community have not changed over the 140 years and that cyanobacteria were regular members of the phytoplankton community. The composition of the major chlorophyll pigments indicated high presence of Chlorophyta and Bacillariophyta in the harbour at all times. The main algal groups identified on the basis of marker pigments presence, besides the Chlorophyta and Bacillariophyta, were the Dinophyta and the Cryptophyta. The presence of a scytonemin derivative, compound B, indicated that cyanobacterial blooms were occurring in past, before the first officially recorded blooms in the 1960s. Cyanobacterial pigments presence indicated that Cyanobacteria have been a regular but not dominant feature of Hamilton Harbour phytoplankton in the past. To our knowledge, this study is the first one examining fossil pigments from Hamilton Harbour. Results of the PCR-DGGE molecular analysis of 16S rRNA-V3 gene fragments from sedimentary DNA revealed the presence of thirteen cyanobacterial genotypes. The temporal change in the cyanobacterial community composition was indicated by the increasing number of species over time, from the oldest to the most recent sediment layers. The deepest sediment strata showed the lowest number (two bands) and intensity of bands. The most recent sediment layer had the greatest numbers (11) and intensity of bands. This increased diversity indicated changing environmental conditions in the harbour, primarily nutrient pollution and worsening water quality. Results of the PCR-DGGE molecular analysis of mcyE-AMT gene fragments showed that Microcystis aeruginosa and Planktothrix rubescens were two microcystin producers present in Hamilton Harbour over the last 80 years. The persistent presence and resilience of these two genera indicated a more serious and longer-term issue of toxic blooms than previously recognized. Historical records show that noticeable anthropogenic impact on Lake Ontario environment has been measurable since the 1780s, the first dramatic impact on the Lake Ontario watershed was evident from the mid1880s, the earliest evidence of eutrophication in the lake occurred between 1820 and 1850, while human induced environmental changes in Hamilton Harbour date back ca. 350 years. In the 1960s, cyanobacterial blooms were first officially recognized in the harbour and the lower Great Lakes. The present research is the first report of the mcyE module and AMT domain of microcystin genes being amplified from sediment of North American lakes, and showed that toxic Cyanobacterial have been regular members of Hamilton Harbour phytoplankton community for almost a century. This research considerably deepened the knowledge of the past toxic cyanobacterial blooms in Hamilton Harbour and their possible causes. It also showed that in the absence of historical records, both the PCR-DGGE method and the mcyE-AMT gene may be used for reconstruction of the past toxic blooms not only in the Laurentian Great Lakes, but also in other aquatic regions of the world impacted by toxic cyanobacterial blooms. Also, it demonstrated the utility of the combined molecular and paleolimnological analyses, which might become a useful tool in the determination of the bloom causes factors and in the mitigation of the future production of toxic blooms.

Dans cette étude de trois lacs sujets aux efflorescences de cyanobactéries, nous avons examiné la diversité des bactéries diazotrophes et des cyanobactéries toxiques. Nous avons tenté de définir les facteurs environnementaux influençant la composition des communautés phytoplanctoniques, la concentration ainsi que la composition des microcystines (MCs). Nous avons émis l’hypothèse que l’azote jouerait un rôle majeur dans le façonnement des communautés cyanobactériennes et influencerait la concentration et composition des MCs. Des concentrations de cette toxine ainsi que le gène mcyE codant pour l’enzyme microcystine synthétase ont été détectés à chaque échantillonnage dans tous les lacs. L’azote, particulièrement sous sa forme organique dissoute (AOD) ainsi que la température de l’eau étaient les facteurs environnementaux expliquant le mieux les concentrations des MCs, tandis que la biomasse de Microcystis spp. était globalement le meilleur prédicteur. Le gène nifH codant pour l’enzyme nitrogénase (fixation d’azote) a aussi été détecté dans chaque échantillon. Malgré les concentrations faibles en azote inorganique dissous (AID) et les densités importantes d’hétérocystes, aucun transcrits du gène n’a été détecté par réverse-transcription (RT-PCR), indiquant que la fixation d’azote n’avait pas lieu à des niveaux détectables au moment de l’échantillonnage. De plus, le pyroséquençage révèle que les séquences des gènes nifH et mcyE correspondaient à différents taxons, donc que les cyanobactéries n’avaient pas la capacité d’effectuer les deux fonctions simultanément. / In this study of three eutrophic lakes prone to cyanobacterial blooms, we examined the diversity of diazotrophic bacteria and toxic cyanobacteria. We evaluated the environmental factors effects on the community composition, the concentration and composition of the family of toxins microcystins (MCs). Since the assimilation of nitrogen and the synthesis of MCs in cyanobacteria are thought to be under the same control of the NtcA transcriptor, we hypothesised that nitrogen played a major role in shaping cyanobacterial communities and influenced indirectly the concentration and composition of MCs. The mcyE gene coding for the microcystin synthetase enzyme and MCs concentrations were detected at each sampling date in all lakes. Nitrogen, particularly under its organic dissolved form (DON) as well as water temperature were the environmental factors explaining the most variation in MC concentration although Microcystis spp. biomass was overall the best predictor. The nifH gene coding for the nitrogenase enzyme (N2-fixation) was also detected at all times. Even though the concentrations of dissolved inorganic nitrogen were relatively low, and that heterocysts were present in high densities, no nifH transcripts were detected by RT-PCR, indicating that no N2-fixation was going on at detectable levels at the time of sampling. Moreover, pyrosequencing revealed that sequences of the genes nifH and mcyE corresponded to different taxa, thus cyanobacteria did not have the capacity to perform both functions simultaneously.