Heterotrophic Bacteria Associated with Cyanobacteria in Recreational and Drinking Water

Katri Berg

Department of Applied Chemistry and Microbiology Division of Microbiology Faculty of Agriculture and Forestry University of Helsinki

Academic Dissertation in Microbiology

To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public criticism in Walter hall (2089) at Agnes Sjöbergin katu 2, EE building on December 11th at 12 o’clock noon.

Helsinki 2009 Supervisors: Docent Jarkko Rapala National Supervisory Authority for Welfare and Health, Finland

Docent Christina Lyra Department of Applied Chemistry and Microbiology University of Helsinki, Finland

Academy Professor Kaarina Sivonen Department of Applied Chemistry and Microbiology University of Helsinki, Finland

Reviewers: Professor Marja-Liisa Hänninen Department of Food and Environmental Hygiene University of Helsinki, Finland

Docent Stefan Bertilsson Department of Ecology and Evolution Uppsala University, Sweden

Opponent: Professor Agneta Andersson Department of Ecology and Environmental Sciences Umeå University, Sweden

Printed Yliopistopaino, Helsinki, 2009 Layout Timo Päivärinta

ISSN 1795-7079 ISBN 978-952-10-5892-9 (paperback) ISBN 978-952-10-5893-6 (PDF) e-mail katri.berg@helsinki.ﬁ

Front cover picture DAPI stained cells of Anabaena cyanobacterium and Paucibacter toxinivorans strain 2C20.

CONTENTS

LIST OF ORIGINAL PAPERS THE AUTHOR’S CONTRIBUTION ABBREVIATIONS ABSTRACT TIIVISTELMÄ (ABSTRACT IN FINNISH) 1 INTRODUCTION ...... 1 1.1 Cyanobacteria ...... 1 1.2 Cyanobacterial mass occurrences ...... 1 1.2.1 Bloom forming cyanobacteria ...... 1 1.3 Cyanobacterial toxins ...... 2 1.3.1 Hepatotoxins ...... 2 1.3.2 Neurotoxins ...... 3 1.3.3 Other toxins of cyanobacteria ...... 4 1.3.4 Toxins analyses ...... 5 1.3.5 Persistence of cyanobacterial toxins ...... 6 1.4 Aquatic cyanobacteria and heterotrophic bacteria ...... 6 1.4.1 Aquatic bacteria ...... 7 1.4.2 Interactions between heterotrophic bacteria and cyanobacteria ...... 7 1.5 Bacterial characterisation ...... 8 1.5.1 Isolation of bacteria ...... 9 1.5.2 Phenotypic analyses ...... 9 1.5.3 Molecular analyses in taxonomic identifi cation of bacteria ...... 10 1.5.4 Genetic analyses in assessment of physiological features ...... 11 1.5.5 Statistical analyses of multivariate data ...... 11 1.6 Problems associated with cyanobacterial water blooms ...... 12 1.6.1 Adverse human health effects associated with cyanobacterial water blooms ...... 12 1.6.2 Cyanobacterial water blooms and drinking water ...... 14 1.6.3 Cyanobacterial toxins and the drinking water treatment ...... 14 1.6.4 Combined effects of cyanobacterial water blooms and heterotrophic bacteria on the drinking water quality ...... 15 2 AIMS OF THE STUDY ...... 18 3 MATERIALS AND METHODS ...... 19 3.1 Water samples and isolated bacterial strains ...... 19 3.2 Methods ...... 19 4 RESULTS ...... 21 4.1 Heterotrophic bacteria associated with cyanobacteria ...... 21 4.1.1 Taxonomic assignment of the isolated bacterial strains ...... 21 4.1.2 Bacteria degrading cyanobacterial hepatotoxins: Paucibacter toxinivorans gen. nov. sp. nov...... 21 4.1.3 Virulence potential of the heterotrophic bacteria associated with cyanobacteria ...... 24 4.1.4 Effects of the heterotrophic bacteria on the growth of cyanobacteria ...... 25 4.2 Selectivity of the used growth media ...... 26 4.3 Drinking water treatment effi ciency ...... 28 4.3.1 Removal of toxins by the water treatment ...... 28 4.3.2 Removal of planktonic cells by the water treatment ...... 29 4.3.3 Taxonomy of the heterotrophic bacteria isolated from drinking water ...... 30 5 DISCUSSION ...... 31 5.1 Characteristics of the isolated heterotrophic bacteria ...... 31 5.2 Drinking water treatment effi ciency ...... 34 6 CONCLUSIONS ...... 36 7 ACKNOWLEDGEMENTS ...... 38 8 REFERENCES ...... 39 APPENDIX 1 The bacterial strains analysed in this study, water type of their origin, isolation medium used, length of the acquired 16S rRNA gene sequence and its accession number, and the taxonomic assignment of the strains ...... 52 LIST OF ORIGINAL PAPERS

I K. A. Berg, C. Lyra, K. Sivonen, L. Paulin, S. Suomalainen, P. Tuomi, and J. Rapala. 2009. High diversity of cultivable heterotrophic bacteria in association with cyanobacterial water blooms. ISME J. 3: 314–325.

II K. A. Berg, C. Lyra, R. M. Niemi, B. Heens, K. Hoppu, K. Erkomaa, K. Sivonen, and J. Rapala. Virulence genes in Aeromonas strains isolated from cyanobacterial water bloom samples associated with human health symptoms. Manuscript, submitted to ISME J.

III J. Rapala, K. A. Berg, C. Lyra, R. M. Niemi, W. Manz, S. Suomalainen, L. Paulin, and K. Lahti. 2005. Paucibacter toxinivorans gen. nov., sp. nov., a bacterium that degrades cyclic cyanobacterial hepatotoxins microcystins and nodularin. Int. J. Syst. Evol. Microbiol. 55: 1563–1568.

IV J. Rapala, M. Niemelä, K. A. Berg, L. Lepistö, and K. Lahti. 2006. Removal of cyanobacteria, cyanotoxins, heterotrophic bacteria and endotoxins at an operating surface water treatment plant. Wat. Sci. Technol. 54(3): 23–28.

The original papers were reproduced with the kind permission of the copyright holders.

THE AUTHOR’S CONTRIBUTION

I Katri Berg participated in designing of the experiment, isolated the majority of the bacterial strains and performed their taxonomic identiﬁ cation. She interpreted the results and wrote the paper.

II Katri Berg participated in designing of the experiment and isolation of the bacterial strains. She also performed taxonomic identification of the bacteria and the statistical analyses of the data, interpreted the results and wrote the paper.

III Katri Berg participated in designing of the experiment, studied the presence of ﬂ agellae and bacteriochlorophyll of the strains, performed the toxin degradation tests and interpreted the results of the toxin analyses. She wrote a part of the paper.

IV Katri Berg participated in the isolation of the bacterial strains, the toxin analysis and the taxonomic identiﬁ cation of the strains. ABBREVIATIONS

Adda (2S,3S,8S,9S)-3-amino-9-methoxy-2,6,8-trimethyl-10- phenyldeca-4,6-dienoic acid AH Aeromonas hydrophila BA Blood agar BLAST Basic local alignment search tool CAP Program for canonical analysis of principal coordinates CFU Colony forming unit DISTLMf Program for distance-based multivariate analysis for a linear model using forward selection ELISA Enzyme-linked immunosorbent assay EU Endotoxin unit GLC Gas-liquid chromatography HPLC High-performance liquid chromatography

LD50 Lethal dose, 50% Microcystin-LR Microcystin with leusine as aminoacid 2 and arginine as amino acid 4 Microcystin-YR Microcystin with tyrosine as aminoacid 2 and arginine as amino acid 4 NCBI National Center for Biotechnology Information PAUP Phylogenetic analysis using parsimony PCO Program for principal coordinate analysis PCR Polymerase chain reaction PHYLIP Phylogeny inference package R2A Reasoner’s 2A medium SAA Starch ampicillin agar TSA Trypticase soy agar UPGMA Unweighted pair group method with arithmetic averages WHO World Health Organization

ABSTRACT

Cyanobacterial mass occurrences, also known as water blooms, have been associated with adverse health effects of both humans and animals. They can also be a burden to drinking water treatment facilities. Risk assessments of the blooms have generally focused on the cyanobacteria themselves and their toxins. However, heterotrophic bacteria thriving among cyanobacteria may also be responsible for many of the adverse health effects, but their role as the etiological agents of these health problems is poorly known. In addition, studies on the water purifi cation effi ciency of operating water treatment plants during cyanobacterial mass occurrences in their water sources are rare. In the present study, over 600 heterotrophic bacterial strains were isolated from natural freshwater, brackish water or from treated drinking water. The sampling sites were selected as having frequent cyanobacterial occurrences in the water bodies or in the water sources of the drinking water treatment plants. In addition, samples were taken from sites where cyanobacterial water blooms were surmised to have caused human health problems. The isolated strains represented bacteria from 57 different genera of the Gamma-, Alpha- or Betaproteobacteria, Actinobacteria, Flavobacteria, Sphingobacteria, Bacilli and Deinococci classes, based on their partial 16S rRNA sequences. Several isolates had no close relatives among previously isolated bacteria or cloned 16S rRNA genes of uncultivated bacteria. The results show that water blooms are associated with a diverse community of cultivable heterotrophic bacteria. Chosen subsets of the isolated strains were analysed for features such as their virulence gene content and possible effect on cyanobacterial growth. Of the putatively pathogenic haemolytic strains isolated in the study, the majority represented the genus Aeromonas. Therefore, the Aeromonas spp. strains isolated from water samples associated with adverse health effects were screened for the virulence gene types encoding for enterotoxins (ast, alt and act/aerA/hlyA), fl agellin subunits (fl aA/fl aB), lipase (lip/pla/ lipH3/alp-1) and elastase (ahyB) by PCR. The majority (90%) of the Aeromonas strains included one or more of the six screened Aeromonas virulence gene types. The most common gene type was act, which was present in 77% of the strains. The fl a, ahyB and lip genes were present in 30–37% of the strains. The prevalence of the virulence genes implies that the Aeromonas may be a factor in some of the cyanobacterial associated health problems. Of the 183 isolated bacterial strains that were studied for possible effects on cyanobacterial growth, the majority (60%) either enhanced or inhibited growth of cyanobacteria. In most cases, they enhanced the growth, which implies mutualistic interactions. The results indicate that the heterotrophic bacteria have a role in the rise and fall of the cyanobacterial water blooms. The genetic and phenotypic characteristics and the ability to degrade cyanobacterial hepatotoxins of 13 previously isolated Betaproteobacteria strains, were also studied. The strains originated from Finnish lakes with frequent cyanobacterial occurrence. Tested strains degraded microcystins -LR and -YR and nodularin. The strains could not be assigned to any described bacterial genus or species based on their genetic or phenotypic features. On the basis of their characteristics a new genus and species Paucibacter toxinivorans was proposed for them. The water purifi cation effi ciency of the drinking water treatment processes during cyanobacterial water bloom in water source was assessed at an operating surface water treatment plant. Large phytoplankton, cyanobacterial hepatotoxins, endotoxins and cultivable heterotrophic bacteria were effi ciently reduced to low concentrations, often below the detection limits. In contrast, small planktonic cells, including also possible bacterial cells, regularly passed though the water treatment. The passing cells may contribute to biofi lm formation within the water distribution system, and therefore lower the obtained drinking water quality. The bacterial strains of this study offer a rich source of isolated strains for examining interactions between cyanobacteria and the heterotrophic bacteria associated with them. The degraders of cyanobacterial hepatotoxins could perhaps be utilized to assist the removal of the hepatotoxins during water treatment, whereas inhibitors of cyanobacterial growth might be useful in controlling cyanobacterial water blooms. The putative pathogenicity of the strains suggests that the health risk assessment of the cyanobacterial blooms should also cover the heterotrophic bacteria.

TIIVISTELMÄ (ABSTRACT IN FINNISH)

Syanobakteerien massaesiintymät, sinileväkukinnat, on yhdistetty monenlaisiin haitallisiin terveysvaikutuksiin, kuten pahoinvointiin, kuumeeseen, päänsärkyyn ja iho- sekä vatsaoireisiin. Kukinnat voivat myös aiheuttaa ylimääräistä kuormitusta juomaveden puhdistuslaitoksille ja heikentää tuotetun juomaveden laatua. Syanobakteerikukintojen aiheuttamien terveysriskien arviointi on yleensä keskittynyt itse syanobakteereihin ja niiden tuottamiin toksiineihin vaikka oireiden mahdollisina aiheuttajina voivat toimia myös kukintojen yhteydessä elävät heterotrofi set bakteerit. Niiden osuus kukintojen aiheuttamien oireiden synnyssä on kuitenkin pitkälti tuntematon. Myös tutkimukset juomaveden puhdistuslaitosten vedenpuhdistuksen tehokkuudesta raakavedessä esiintyvien kukintojen aikana ovat harvinaisia. Tässä työssä eristettiin yli 600 heterotrofi sta bakteerikantaa makea- ja murtovetisistä luonnonvesistöistä sekä puhdistetusta juomavedestä. Vesinäytteiden keräyskohteet oli valittu vesistöissä tai juomaveden raakavedessä usein toistuvien syanobakteeriesiintymien perusteella. Näytteitä otettiin myös vesiympäristöistä, joissa syanobakteerikukintojen epäiltiin aiheuttaneen terveysoireita ihmisille. Eristetyt bakteerikannat edustivat 16S rRNA -geenisekvenssiensä perusteella 57 eri bakteerisukua jotka kuuluivat Gamma-, Alpha- ja Betaproteobacteria, Actinobacteria, Flavobacteria, Sphingobacteria, Bacilli ja Deinococci -luokkiin. Useilla kannoilla ei ollut läheisiä sukulaisia aiemmin eristettyjen bakteerien tai viljelystä riippumattomilla menetelmillä kerättyjen 16S rRNA -geenisekvenssien joukossa. Tulosten perusteella syanobakteerikukintojen yhteydessä esiintyy monipuolinen viljeltävissä olevien heterotrofi sten bakteerien yhteisö. Eristetyistä heterotrofi sista bakteerikannoista analysoitiin ominaisuuksia kuten virulenssigeenien esiintymistä ja mahdollisia vaikutuksia syanobakteerien kasvuun. Tutkituista hemolyyttisistä ja siten mahdollisesti virulenteista kannoista valtaosa edusti Aeromonas-sukua. Tulosten perusteella terveysoireisiin liittyneistä vesinäytteistä eristetyistä Aeromonas-kannoista etsittiin kuutta Aeromonas virulenssigeenityyppiä PCR menetelmällä. Nämä geenityypit koodaavat enterotoksiineja (ast, alt ja act/aerA/hlyA geenit), fl agelliini-alayksikköjä (fl aA/falB), lipaasia (lip/pla/lipH3/alp-1) ja elastaasia (ahyB). Valtaosa (90 %) tutkituista Aeromonas-kannoista sisälsi yhden tai useamman geenityypeistä. Niistä yleisin oli act, joka havaittiin 77 % kannoista. Myös fl a, ahyB ja lip -geenejä havaittiin 30–37%:ssa kannoista. Virulenssigeenien yleisyys viittaa siihen, että Aeromonas-bakteerit saattavat olla yksi syanobakteerikukintoihin yhdistettyjen oireiden aiheuttajista. Niistä yli 180 eristetystä bakteerikannasta, joiden kykyä vaikuttaa syanobakteerien kasvuun tutkittiin, valtaosa (60 %) joko kiihdytti tai esti syanobakteerien kasvua. Useimmissa tapauksissa vaikutus oli kasvua kiihdyttävä, mikä viittaa mutualistiseen vuorovaikutukseen. Tulosten perusteella heterotrofi set bakteerit saattavat vaikuttaa syanobakteerien massaesiintymien muodostumiseen ja kestoon. Työssä tutkittiin myös kolmentoista aiemmin eristetyn Betaproteobacteria-kannan geneettisiä ja fenotyyppisiä ominaisuuksia, kuten 16S rRNA geenisekvenssiä ja entsyymiaktiivisuuksia, ja kykyä hajottaa syanobakteerien tuottamia maksatoksiineita, mikrokystiineitä ja nodulariinia. Kannat eivät geneettisten ja fenotyyppisten piirteidensä perusteella kuuluneet aiemmin kuvattuihin bakteerisukuihin tai -lajeihin. Kaikki tutkitut kannat hajottivat maksatoksiineita. Kantojen geneettisten- ja fenotyyppisiin ominaisuuksiin perustuen kuvattiin uusi bakteerisuku ja -laji, Paucibacter toxinivorans. Lisäksi työssä arvioitiin toiminnassa olevan juomavedenveden puhdistuslaitoksen vedenpuhdistustehokkuutta syanobakteerikukinnan aikana. Suurten kasviplanktonsolujen, syanobakteerien maksatoksiinien, endotoksiinien ja viljeltävien heterotrofi sten bakteerien määrät vähenivät vedenkäsittelyn aikana tehokkaasti. Sen sijaan pienet planktonsolut läpäisivät vedenkäsittelyketjun säännöllisesti. Puhdistusketjun läpäisevät solut voivat edistää biofi lmien muodostumista vedenjakeluverkostossa ja siten heikentää käsitellyn veden laatua. Työssä käsitellyt heterotrofi set bakteerikannat tarjoavat monipuolisen kokoelman bakteerikantoja syanobakteerien ja niiden yhteydessä esiintyvien heterotrofisten bakteerien välisten vuorovaikutusten tutkimiseksi. Syanobakteerien tuottamia maksatoksiineita hajottavia kantoja voitaisiin mahdollisesti hyödyntää maksatoksiinien poistamisessa vedenpuhdistuksessa ja syanobakteerien kasvua estäviä kantoja kukintojen kontrolloinnissa. Kantojen mahdollinen patogeenisyys viittaa siihen, että myös kukintojen yhteydessä esiintyvät heterotrofi set bakteerit pitäisi ottaa huomioon syanobakteerikukintoihin liittyvien terveysriskien arvioinnissa.

Introduction

1. INTRODUCTION

1.1 Cyanobacteria salinity and moisture. They can be found Cyanobacteria are fascinating organisms in hot springs and in salt lakes, on desert that posses features familiar to both crusts or symbionts with lichens and are prokaryotic and eukaryotic branches an important part of both freshwater and of the tree of life. Their cell structure marine ecosystems (Whitton and Potts and composition is similar to that of a 2000, Oren 2000). We live beside the prokaryotic cell in that they lack the cell cyanobacteria and exploit them for various nucleus and distinctive organelles of purposes ranging from oxygen sources eukaryotic cells. However, in contrast to rice fi eld fertilizers (Oren 2000, van to typical prokaryotes they are able Hove and Lejeune 2002). Unfortunately to perform oxygenic photosynthesis this coexistence does not always occur (Castenholtz 2001). Evidence of fossilised without problems, which has given the cyanobacteria-like organisms date back for cyanobacteria a somewhat infamous more than 2500 million years and that of reputation. versatile cyanobacterial biota for around 2000 million years (Olson 2006, Knoll 1.2 Cyanobacterial mass 2008). Cyanobacteria are believed to have occurrences had a considerable effect on the formation Perhaps the most familiar association that of the oxygen rich gas composition of people have with cyanobacteria is the mass Earth’s atmosphere (Dismukes 2001, Paul occurrences of planktonic cyanobacteria, 2008). This in turn had a profound effect also known as the cyanobacterial water on the planet’s environmental conditions blooms. During the mass occurrences such as the availability of nutrients and the cyanobacterial cells can be spread out formation of the ozone layer. In addition, into a large water mass and be quite they are also considered to be the origin of imperceptible (Mur et al. 1999, Oliver and eukaryotic chloroplasts (Raven and Allen Ganf 2000, Huisman and Hulot 2005). In 2003). calm weather the cells can then rapidly During their long evolutionary concentrate onto the surface layer and history cyanobacteria have evolved a wide form a dense, fl oating cell matt. Mild wind spectrum of forms varying from unicellular and currents can transport the water bloom to fi lamentous morphology (Mur et al. to shore. This increases the possibility of 1999, Castenholtz 2001). Many of them humans and animals to come in contact produce specialised cells called heterocytes with the cyanobacterial mass. In addition, for the assimilation of molecular nitrogen cyanobacteria such as Planktothrix and or resting cells called acinetes. In addition, Cylindrospermopsis can form mass many cyanobacteria produce motile occurrences several meters below the fi laments called hormogonia that function water surface, which can hinder their as pioneers for starting new colonies away detection (Mur et al. 1999, Havens 2008, from the main cyanobacterial biomass Sivonen and Börner 2008). (Castenholtz 2001). Cyanobacteria are known to withstand 1.2.1 Bloom forming cyanobacteria a wide range of environmental conditions The water blooms can be produced by in terms of temperature, light conditions, various cyanobacterial species within

1 Introduction genera such as Anabaena, Aphanizomenon, are known to include several kinds of Cylindrospermopsis, Planktothrix, compounds which are toxic to humans, Gloeotrichia, Microcystis and Nodularia animals and plants. The toxin production (Chorus et al. 2000, Oliver and Ganf of aquatic, planktonic and benthic 2000). In the tropics the blooms can prevail cyanobacteria is the most studied and all year round, whereas in the temperate the majority of the cyanobacterial water zones these water blooms usually occur blooms are shown to be toxic (Sivonen during late summer and early autumn, and Börner 2008). However, terrestrial subsiding during the cold and dark winter cyanobacteria are also known to produce period (Sivonen and Jones 1999, Oliver toxins (Oksanen et al. 2004, Sivonen and and Ganf 2000). However, cyanobacterial Börner 2008). water blooms can persist through winter also in temperate zones, as happened in 1.3.1 Hepatotoxins Finland during the early 1980’s on Lake The best characterised cyanobacterial Vesijärvi and in the winter of 2008–2009 toxins are the hepatotoxic microcystins on Lake Saimaa (Keto and Tallberg 2000, and nodularins. They are the most Kaakkois-Suomen ympäristökeskus frequently found toxins in association 2009). with cyanobacterial water blooms Cyanobacteria have various features (Sivonen and Börner 2008). Microcystins that help them to overcome other are cyclic heptapeptides and nodularins phytoplankton species and form dense cell cyclic pentapeptides. Both microcystins masses in aquatic environments. Perhaps and nodularins contain an unusual Adda the most important of these features amino acid ((2S,3S,8S,9S)-3-amino-9- are the gas vesicles possessed by many methoxy-2,6,8-trimethyl-10-phenyldeca- planktonic cyanobacterial species (Paerl 4,6-dienoic acid), that has an important 2000, Castenholtz 2001). These vesicles role in their toxicity together with their can be used by cyanobacteria to regulate cyclic structure (Craig and Holmes 2000, their buoyancy to position themselves at Sivonen and Börner 2008). a depth favourable in terms of nutrients More than 80 microcystins variants and light. Another important feature in are known, whereas known variants of aquatic environments where nitrogen nodularin are relatively rare (Sivonen is often the growth limiting factor, is 2009). Microcystins and nodularins are the capability of many cyanobacteria produced nonribosomally by peptide to fix atmospheric molecular nitrogen synthetase and polyketide synthase (Castenholtz 2001). Other features thought complexes (Börner and Dittmann 2005, to enhance cyanobacterial survival and Welker and von Döhren 2006). During competition status include features such the active growth phase the toxins as siderochrome excretion for sequestering are intracellular (Sivonen and Jones iron, and the ability to utilize low photon 1999). During cell death and lysis the ﬂ uxes for photosynthesis. toxins are released into the surrounding water. Microcystins are produced by 1.3 Cyanobacterial toxins cyanobacteria which belong to the Cyanobacteria produce a wide variety of genera such as Microcystis, Planktothrix/ secondary metabolites, many of them with Oscillatoria, Anabaena, Anabaenopsis, so far unknown functions. The metabolites Radiocystis, Synechococcus and Nostoc

2 Introduction

(Sivonen and Börner 2008). The such as the proposed tumour promotion, nodularins are produced by Nodularia. by long term exposure to low toxin doses The amounts and types of produced toxins (Kuiper-Goodman et al. 1999, Falconer vary between different cyanobacterial 2008, Humpage 2008). genera and species. In addition, the phase of the cyanobacterial bloom and the size 1.3.2 Neurotoxins of the cyanobacterial colonies may affect Both planktonic and benthic forms toxin production (Kardinaal and Visser of cyanobacteria produce several 2005). kinds of neurotoxins. These toxins Microcystins and nodularins include anatoxin-a and its homologue are inhibitors of eukaryotic protein homoanatoxin-a, anatoxin-a(S), saxitoxins phosphatases 1 and 2A (Sivonen and and β-N-methylamino-L-alanine (Sivonen Börner 2008). They are water soluble and Börner 2008). Anatoxin-a is a and are transported into cells mostly secondary amine, which is produced by by bile acid type transporters present in species within the genera Anabaena, hepatocytes and in the enterocytes of Cylindrospermum, Aphanizomenon, the small intestine (Kuiper-Goodman et Planktothrix, Oscillatoria, Phormidium al. 1999, Sivonen and Jones 1999). The and Arthrospira (Sivonen and Börner speciﬁ c uptake system largely restricts 2008). It is an alkaloid that binds to the effect of the toxins in mammals to the neuronal nicotinic acetyl cholinesterase liver, which gives them the classiﬁ cation receptors and causes nerve depolarisation of hepatotoxins (Sivonen and Jones and muscle overstimulation (Kuiper- 1999). Cyanobacterial hepatotoxins cause Goodman et al. 1999, Humpage 2008). disruption of intracellular architecture This can lead to health effects such as and a loss of cell-to-cell adhesion, which staggering, gasping, convulsions and leads to haemorrhage (Craig and Holmes respiratory failure (Humpage 2008). 2000). Other acute effects attributed to the Its homoanalogue homoanatoxin-a, is cyanobacterial hepatotoxins may include produced by Oscillatoria/Planktothrix the generation of reactive oxygen species, and Raphidiopsis (Humpage 2008, alteration of mitochondrial membrane Sivonen and Börner 2008). Anatoxin- permeability and induction of cell a(S) is an organophosphate produced by apoptosis (Humpage 2008). Anabaena (Sivonen and Börner 2008), that The lethal dose of cyanobacterial inhibits the enzyme acetyl cholinesterase hepatotoxins varies depending on the route (Falconer 2008, Humpage 2008). The of exposure as well as the hepatotoxin resulting health effects are similar to those variant and the exposed organism. caused by anatoxin-a, with the addition of

In mice the lethal dose, 50% (LD50) lacrimation, ataxia, diarrhoea, tremors and of microcystin-LR by intraperitoneal hypersalivation, for which the (S) in the injection varies from around 25 to 150 μg toxin name stands for. -1 kg body weight whereas the LD50 dose Saxitoxins are alkaloid neurotoxins by the oral route is 5000 μg kg-1 body also known as paralytic shellﬁ sh poisons weight (Kuiper-Goodman et al. 1999, de (Sivonen and Börner 2008). They include Figuiredo et al. 2004). In addition to acute nonsulphated (saxitoxin and neosaxitoxin), intoxication, cyanobacterial hepatotoxins singly sulphated (gonyautoxins) and can cause adverse chronic health effects doubly sulphated (C-toxins) compounds.

3 Introduction

Of these compounds, saxitoxin is the which then may function as transporters most toxic. Saxitoxins were originally between organisms (Murch et al. 2004). In identiﬁ ed in marine dinoﬂ agellates. Of addition, they can act as a reservoir of the the cyanobacteria they are produced toxin, releasing it slowly during protein by Aphanizomenon, Anabaena, metabolism. Cylindrospermopsis, Lyngbya and Planktothrix (Sivonen and Börner 1.3.3 Other toxins of cyanobacteria 2008). Saxitoxins affect the nerve axon Examples of other types of toxins or membranes by blocking sodium ion toxic compounds of the cyanobacteria channels, which causes changes in nerve are the cylindrospermopsin and impulse generation (Kuiper-Goodman et lipopolysaccharides (Sivonen and Jones al. 1999). This can lead to symptoms such 1999). Cytotoxic cylindrospermopsin as lip numbness, paralysis and respiratory is a cyclic guanide alkaloid that is failure. produced by species within the genera

The acute LD50 dose of cyanobacterial Cylindrospermopsis, Umezakia, neurotoxins by intraperitoneal injection in Aphanizomenon, Anabaena and mice tests varies from about 10 μg kg-1 Raphidiopsis (Sivonen and Börner 2008). body weight for saxitoxin, to 200−250 μg It inhibits protein synthesis in general, kg-1 body weight for anatoxin-a and for causing tissue injury in the liver, kidneys, homoanatoxin-a (Smith 2000). Saxitoxins lungs, heart, adrenal glands, spleen and may accumulate in shellfish that form thymus (Kuiper-Goodman et al. 1999, an additional route of oral exposure Humpage 2008). Cylindrospermopsin through ingestion and cases of acute has also been shown to have genotoxic intoxication due saxitoxins in people effects, and in vivo studies indicate it to be consuming shellﬁ sh have been described carcinogenic (Humpage 2008). (Kuiper-Goodman et al. 1999, Humpage In contrast to the above

2008). However, the LD50 values by oral mentioned cyanobacterial toxins, exposure are several times higher that the lipopolysaccharides are cell wall of the intraperitoneal route. The chronic components of the cyanobacteria and effects of the cyanobacterial neurotoxins of Gram-negative bacteria in general. are largely unknown (Kuiper-Goodman The cyanobacterial lipopolysaccharides et al. 1999, Falconer 2008, Humpage seem have a lower toxicity than the 2008). Some ﬁ ndings even indicate that lipopolysaccharides of the Gram-negative pre-exposure to saxitoxins may lower the bacteria (Stewart et al. 2006a). Their susceptibility to adverse health effects on role as toxins is under study but they subsequent exposure to these toxins. are suspected to act as contact irritants Quite recently, another neurotoxin, and a possible cause of gastro-enteritis the β-N-methylamino-L-alanine, originally and adverse respiratory effects (Kuiper- found in cycad seeds was shown to be Goodman et al. 1999, Anderson et al. produced by various cyanobacterial 2002, Stewart et al. 2006a). The typical genera (Cox et al. 2005, Humpage 2008). exposure routes of humans to external The β-N-methylamino-L-alanine has endotoxins include direct physical contact, been associated with neurodegenerative oral ingestion of endotoxin containing disorders (Humpage 2008). It can be matter, inhalation of endotoxic aerosols bound by the proteins within the body, and in some cases intravenous exposure.

4 Introduction

The most signifi cant route of exposure in hepatotoxins, saxitoxins, anatoxin-a and association with cyanobacterial blooms cylindrospermopsin (Spoof 2005, Metcalf is probably by inhalation (Stewart et al. and Codd 2005a, b and c, Lawton and 2006a, Anderson et al. 2007). Edwards 2008). The enzyme-linked immunosorbent (ELISA) and protein 1.3.4 Toxin analyses phosphatase inhibition assays are used The methods used to detect cyanobacterial as relatively inexpensive and rapid toxins vary depending on the toxin type methods for analyses of cyanobacterial and whether the objective of the study hepatotoxins, whereas the acetylcholine is in examining the toxin types, toxin esterase inhibition assay is used for concentration or toxicity of the sample. screening of the putative presence of In the early studies, the mouse bioassay anatoxin-a(S) (Ellman et al. 1961, was often used but its use has since Chu 2000, Rapala et al. 2002a, Spoof declined due to its non specifi city, the 2005, Lawton and Edwards 2008). development of new detection methods Immunoassays, such as RIDASCREEN® and ethical issues concerning animal (R–Biopharm AG, Germany) and testing (Chu 2000, Sivonen 2000, Spoof Jellet Rapid Test (Jellett Rapid Testing 2005). Other bioassays, such as the desert Ltd., Canada), are also available for locust (Schistocerca gregaria) test for saxitoxins (Jellet et al. 2002, Lawton and saxitoxins and brine shrimp (Artemia Edwards 2008). The lipopolysaccharide salina) test for neurotoxins, hepatotoxins concentrations can be measured using the and cylindrospermopsin have also been chromogenic Limulus amoebocyte lysate used (Chu 2000, Sivonen 2000, Metcalf et assay (Metcalf and Codd 2005b). al. 2002). The chemical, immunological and The bioassays have mostly been enzymatic assays outcompete the bioassays replaced by a wide variety of chemical, in accuracy, specifi city and having a lesser enzymatic and immunochemical assays need for the test materials. However, they (Chu 2000, Sivonen 2000, Lawton have their own disadvantages such as a and Edwards 2008). The chemical high cost of the equipments, the need for assays include methods such as high- expertise, problems in sensitivity due to performance liquid chromatography structural variations within the studied (HPLC), mass spectrometry, thin-layer toxin types and the lack of available toxin chromatography and combinations standards (Chu 2000, Sivonen 2000, of these. The immunochemical Lawton and Edwards 2008). Therefore, assays include methods such as bioassays such as the Artemia salina test radioimmunoassay and modifications can still be useful as a relatively cheap of the enzyme-linked immunosorbent and easy option for cyanobacterial toxin assay (ELISA). The enzymatic assays analyses, especially to screen samples for utilize protein phosphatase inhibition the presence of different cyanobacterial assays for microcystins and nodularin toxins types (Lahti et al. 1995, Sivonen or acetylcholine esterase inhibition for 2000, McElhiney and Lawton 2005). preliminary identification of anatoxin- Bioassays are also needed in assessing the a(S). The high-performance liquid acute and chronic toxicity of compounds chromatography (HPLC) approach is often produced by cyanobacterial on organisms used for the analyses of cyanobacterial such as mammals.

5 Introduction

1.3.5 Persistence of cyanobacterial in the elimination of other cyanobacterial toxins toxin types, as could be seen in the The persistence of the cyanobacterial study of Rapala et al. (1994) in which toxins varies depending on the toxin the anatoxin-a degradation depended on type and whether the toxins are located biodegradation by microorganisms from within or outwith the cyanobacterial the sediments. cells (Sivonen and Jones 1999, Chorus The majority of the biodegradation et al. 2000). In addition, the chemical, studies on cyanobacterial toxins physiological and biological circumstances have concentrated on hepatotoxins. affect the persistence of cyanobacterial Biodegradation of the cyanobacterial toxins. In general, the cyanobacterial hepatotoxins have been detected in lake toxins are more stable in the dark than and river water, lake sediments, sediments in sunlight. The hydrolysing effect of of a water recharge facilities and in the sunlight can be potentiated by the sewage efﬂ uent waters (Sivonen and Jones presence of cell pigments (Tsuji et al. 1999, Holst 2003). So far, the majority of 1994, Robertson et al. 1999). Other the isolated bacterial strains capable of chemical or physical factors that affect degrading cyanobacterial hepatotoxins the toxin breakdown include temperature, have belonged to the genus Sphingomonas pH, and the presence of oxidizing agents and to unidentified Betaproteobacteria such as ozone and chlorine (Sivonen and (Jones et al. 1994, Bourne et al. 1996 and Jones 1999, Chu 2000). However, the 2001, Lahti et al. 1998, Park et al. 2001, most important method of removing the Saito et al. 2003, Saitou et al. 2003, Ishii cyanobacterial toxins in the environment et al. 2004, Maryama et al. 2006, Ho et al. is probably biodegradation by microbes. 2007). In general, the cyanobacterial The hepatotoxin degradation hepatotoxins microcystins and nodularins mechanism of Sphingomonas has been are chemically more stable than the studied and it seems to include protein neurotoxins or cylindrospermopsin. In the products encoded by the genes mlrA, dark they can remain toxic for months or mlrB, mlrC and mlrD (Bourne et al. years and their toxicity is not destroyed 1996 and 2001, Saito et al. 2003, Ho even by boiling (Sivonen and Jones 1999). et al. 2007). Together they break down The persistence of cyanobacterial toxins the cyclic microcystin-LR structure into can be further assisted through protection a linear toxin by breaking the Adda- by the cyanobacterial cells. For example, Arginine peptide bond and by breaking the hepatotoxin containing cyanobacterial cell linear toxin into tetrapeptides and further scum found on shores can release toxins into smaller peptides and single amino into water (Jones et al. 1995). They can acids (Bourne et al. 1996 and 2001). The also form another route of exposure to hepatotoxin degradation mechanism in the the toxins through aerosols from toxic unidentified Betaproteobacteria strains cell matter. Therefore, the importance of (Lahti et al. 1998) is unknown. biodegradation of both cyanobacterial cell mass and toxins is emphasized in the 1.4 Aquatic cyanobacteria and removal of microcystins and nodularins heterotrophic bacteria from the environment (Sivonen and Jones Cyanobacteria and heterotrophic 1999). Biodegradation is also important bacteria are an important part of aquatic

6 Introduction ecosystems. They have an invaluable group of water bodies (Sivonen et al. role in nutrient cycling in freshwater 2007). and marine environments. At times planktonic cyanobacteria can form the 1.4.2 Interactions between majority of the phytoplankton biomass heterotrophic bacteria and and are responsible for most of the cyanobacteria plankton primary production (Paerl 2000, The abundance and taxonomic distribution Giovannoni and Stingl 2005, Eiler et al. of the heterotrophic bacterioplankton 2006b). They are associated with various associated with cyanobacterial mass kinds of heterotrophic bacteria (Eiler and occurrences can be affected by Bertilsson 2004, Kolmonen et al. 2004, several chemical, physiological and Eiler et al. 2006b, Tuomainen et al. 2006), biological factors such as nutrient, which use the carbon ﬁ xed by the primary light and temperature conditions, pH, producers and form an important part of salinity and grazing. The presence and the aquatic nutrient web. metabolic activities of large amounts of cyanobacterial cells can affect many 1.4.1 Aquatic bacteria of these chemical, physiological and Based on cultivation-independent studies, biological factors (Paerl 2000, Wilson et the microbial plankton populations seem al. 2006, Havens 2008). The metabolic to form habitat-specific clusters that activities such as photosynthesis differ between marine and freshwater performed by cyanobacterial cells can environments (Hagström et al. 2002, affect the oxygen and pH conditions of Zwart et al. 2002, Giovannoni and Stingl the surrounding water and change the 2005). Cyanobacteria are detected from amounts of available organic compounds freshwater and marine environments and nutrients as well as abundances in varying proportions. The major of antibacterial compounds (Paerl and bacterial divisions of freshwater Pinckney 1996, Østensvik et al. 1998, heterotrophic bacteria include Alpha- Casamatta and Wickstrom 2000, Oliver and Betaproteobacteria, Actinobacteria, and Ganf 2000, Kirkwood et al. 2006). For Cytophaga-Flavobacterium-Bacteroidetes example, many planktonic cyanobacterial and Verrucomicrobia (Glöckner et al. species can ﬁ x atmospheric nitrogen and 2000, Zwart et al. 2002). In contrast, the convert it into forms accessible to other Betaproteobacteria form only a small organisms (Oliver and Ganf 2000, Paerl proportion of the marine heterotrophic 2000). The cyanobacterial cell mass also bacterial plankton, whereas the Gamma- offers additional particle surfaces to which and Alphaproteobacteria form a majority the heterotrophic bacteria can attach to of the detected bacterial plankton of and provide protection by means of the the marine environments (Hagström et nutrient-rich polysaccharide sheaths of the al. 2002, Zwart et al. 2002, Rusch et al. cyanobacteria (Paerl and Pinckney 1996, 2007). Interfaces of these two aquatic Islam et al. 1999, Paerl 2000, Maryama et environment types such as estuaries and al. 2003, Salomon et al. 2003, Eiler et al. coastal seas may contain bacterial groups 2006b). typical of both fresh and marine water In considering these factors, it is no type. As a brackish water body the Baltic surprise that the cyanobacterial water Sea also largely falls into this intermediate blooms seem to affect the amounts

7 Introduction and composition of the planktonic formation and decline of cyanobacterial heterotrophic bacteria. The observed cell mass occurrences. The known bacteria concentrations of heterotrophic bacteria that are able to affect the cyanobacterial can be substantially higher during and growth include species such as Cytophaga, immediately after cyanobacterial water Shewanella, Streptomyces and Vibrio blooms than in their absence (Bouvy et al. strains (Yamamoto et al. 1998, Yoshikawa 2001, Eiler and Bertilsson 2004 and 2007). et al. 2000, Rashidan and Bird 2001, In the studies by Islam et al. (1999, 2002 Manage et al. 2000, Salomon et al. 2003). and 2004) the presence of cyanobacteria They can inhibit the cyanobacterial growth was found to prolong the cultivability of by means such as competition for limiting Vibrio cholerae and the viable Vibrio cells nutrients and by production of diffusible could be seen attached to cyanobacterial lytic compounds (Yamamoto et al. 1998, cells long after they could no longer be Rashidan and Bird 2001). The growth can cultivated. In addition, in a study by Eiler be enhanced by means such as production et al. (2007), dissolved organic matter of CO2, which cyanobacteria can use as from the cyanobacterium Nodularia a carbon source or by the stimulation of spumigena increased the measured Vibrio cyanobacterial photosynthetic and nitrogen cholerae and Vibrio vulnifi cus abundances. fixation activity through the reduction Specifi c attachment to the heterocytes, of photosynthetic oxygen tension (Paerl cyanobacterial nitrogen fixing cells, and Kellar 1978, Mouget et al. 1995, by heterotrophic bacteria has also been Paerl and Pinckney 1996, Paerl 2000). detected, with indications that the attached The cyanobacteria may also benefi t from bacteria utilize amino acids produced by the vitamins and remineralized nutrients the cyanobacteria (Paerl 1976, Paerl and produced by the heterorophic bacteria Kellar 1978). Furthermore, the study by (Paerl and Pickney 1996, Simon et al. Kangatharalingam et al. (1991) showed 2002) However, the exact mechanism a positive relationship between the behind the cyanobacterial growth chemotaxis of Aeromonas hydrophila inhibition or enhancement by heterotrophic to cyanobacteria and the cyanobacterial bacteria is often unknown. biomass. Positive chemotaxis to cyanobacterial exudates has also been 1.5 Bacterial characterisation seen in other bacteria co-occuring with Interpretation of the interactions between cyanobacteria (Casamatta and Wickstrom bacteria and their role in the environment 2000). These studies indicate that requires knowledge on their characteristics heterotrophic bacteria often benefi t from such as enzymatic and metabolic activities. the presence of cyanobacteria. The fact For studying the bacterial characteristics, that many cyanobacterial bloom species cultivable isolated bacterial strains have not successfully been grown as provide a valuable means compared with axenic cultures but seem to prefer the the cultivation-independent methods presence of other bacteria indicates that (Palleroni 1997, Zinder and Salyers 2001, the benefi t is in several cases mutual (Paerl Roselló-Mora and Amann 2001). Thus, 1996 and 2000). the cultivation methods are an invaluable The heterotrophic bacteria, in turn, asset of the microbiological and ecological can inhibit or enhance the growth of studies, even though they are laborious cyanobacteria and therefore affect the and the detected bacteria are restricted to

8 Introduction those supported by the chosen cultivation for describing a new taxon (Palleroni conditions. 1997, Roselló-Mora and Amann 2001). In addition, informativity of the taxonomic 1.5.1 Isolation of bacteria assignment of new bacterial isolates or The isolated bacterial strains can vary 16S rRNA gene clones is largely based greatly depending on factors such as on the known characteristics of the close the condition of the cells, incubation relatives of the studied bacteria (Palleroni temperature, incubation time, pH, the 1997). surrounding oxygen levels and whether The analyses of cell morphology a solid or liquid growth medium is used include microscopy for studying cell (Zinder and Salyers 2001, Simu and features such as size and shape, Gram Hagström 2004, Cardenas and Tiedje staining and presence of flagellas 2008). In addition, the used growth media (Roselló-Mora and Amann 2001). In may lack substances essential to the addition, the morphology includes the desired bacteria or alternatively contain characters such as colour and size of the them at toxic concentrations. For example, bacterial colonies. Of the biochemical rich growth media may contain nutrients components, characteristics such as fatty in concentrations toxic to aquatic bacteria acid composition and protein profile that are usually adapted to quite nutrient of the cells and cell wall composition poor conditions (Zinder and Salyers 2001). of the Gram-positive bacteria are often For these bacteria a growth medium with investigated. low nutrient content, such as sterilized Analyses for enzymatic activities and natural water or artifi cial medium such growth requirements are usually performed as Reasoner’s 2A (R2A), would be more using commercial test kits, that contain suitable (Massa 1998, Zinder and Salyers several tests in miniature form. These 2001). However, nutrient rich growth include kits such as the widely used API media may help to identify important tests (bioMèrieux, Marcy-l’Etoile, France) bacteria that might not be detected by and the Biolog (Biolog, Inc., Hayward, using only nutrient poor growth media. CA, USA) for identifi cation of bacteria of Therefore, combinations of different specifi c metabolic types. For example, the growth media and incubation conditions API 20 NE test kit for the identifi cation should be used when endeavouring to of non-fastidious, non-enteric, Gram- access as many of the cultivable bacteria negative rods, such as Pseudomonas, as possible. Flavobacterium and Vibrio, and Biolog tests versions for both Gram-negative 1.5.2 Phenotypic analyses and Gram-positive bacteria are available Before the era of molecular methods, (Biolog 2001a and b, bioMèrieux 2009, bacterial classification was based on Stager and Davis 1992, Truu et al. 1999, required growth conditions, morphology Awong-Talylor et al. 2008). The API ZYM and physiological features of the bacteria test kit and the ZymProfi ler® fl uorogenic (Roselló-Mora and Amann 2001). analysis designed by Vepsäläinen et al. Analyses of these characteristics still (2001) can be used for the screening of continue to play an important role in the enzymatic activities of bacteria from bacterial identifi cation and information on purifi ed or complex samples (bioMèrieux phenotypic characteristics is always needed 2009, Tharagonnet et al. 1977, Allen et

9 Introduction al. 2002, Niemi et al. 2004). In addition, A wide variety of methods included noncommercial methods such as the in both free and commercial software Aeromonas hydrophila (AH)-medium tube packages, such as the commonly used test by Kaper et al. (1979) can be used for phylogenetic analyses using parsimony examining several physiological features (PAUP, D. L. Swofford, Sinauer simultaneously. The information obtained Associates, Sunderland, MA, USA) and from cultivated, pure bacterial strains offer phylogeny inference package (PHYLIP, new reference points for the large scale Felsenstein 2005), are also available for studies, which relay on the data obtained analyses of the phylogenetic relationships. by molecular methods, and help to clarify The phylogenetic relationships between the roles of different bacterial groups in bacteria are often visualized by using the environment. phylogenetic trees generated by methods such as neighbour-joining, unweighted 1.5.3 Molecular analyses in taxonomic pair group method with arithmetic identifi cation of bacteria averages (UPGMA), maximum parsimony The use of DNA based methods in or maximum likelihood (Nei and Kumar taxonomic identification of bacteria 2000b, c and d, Felsenstein 2004b). began with the analyses of genomic The neighbour-joining method and guanine-cytosine content and DNA-DNA unweighted pair group method with hybridization of the whole genomic DNA arithmetic averages (UPGMA) are based (Roselló-Mora and Amann 2001). These on distance matrixes calculated from methods offer a valuable asset in closer sequence distances using models such as taxonomic assessment of bacteria and are the Kimura or the F84 model (Nei and still required for describing a new taxon. Kumar 2000b, Felsenstein 2004b). In However, the routine use of polymerase contrast, the maximum parsimony and chain reaction (PCR) and DNA sequencing maximum likelihood methods are based on have made the analyses of the rRNA gene the actual sequence data (Nei and Kumar sequences an inseparable part of bacterial 2000c and d). In addition to the tree taxonomic analyses. Sequence analyses construction strategies, the methods differ of the universal and conserved 16S rRNA in the required computer time. Therefore, gene are especially used as relatively the choice of method has often been the cheap and quick methods of taxonomic computationally efficient neighbour- identifi cation of bacteria (Roselló-Mora joining, especially when analysing large and Amann 2001, Clarridge 2004). The datasets. In the maximum parsimony popularity of the16S rRNA analyses is method, the needed computer time can be based partly on the rapidly growing 16S limited by choosing the heuristic search rRNA sequence databases and a wide option instead of exhaustive search (Nei variety of quick search tools provided and Kumar 2000d). The reliability of freely on the internet. These include tools the obtained trees can be assessed by a such as the basic local alignment search confi dence test such as bootstrapping, and tool (BLAST) provided by the National by comparing the trees obtained by using Center for Biotechnology Information different tree building methods (Nei and (NCBI) and the taxonomic classifi er and Kumar 2000a, Felsenstein 2004a). sequence match tools provided by the Ribosomal Database Project II.

10 Introduction

1.5.4 Genetic analyses in assessment of The suitability of the statistical methods physiological features for analysing the measured variables Since the physiological features of the differs according to several factors such bacteria are ultimately based on their gene as whether or not a quantitative form and content, molecular methods such as PCR normal distribution of data are required. with gene specifi c primers can also be used Therefore, for example the chosen distance to assess putative physiological features or dissimilatory measure can considerably of the bacteria. For example, the virulence affect the results of the analysis. potential of Aeromonas strains has been Many of the available methods are assessed by screening for virulence genes based on a specific distance measure, (e.g. Kingombe et al. 1999, Albert et al. which can limit their usefulness for 2000, Chacón et al. 2003, Sen and Rodgers analysing the studied data (Anderson and 2004). The virulence of the Aeromonas Willis 2003). In contrast, the programs has been suggested to be multifactorial for the principal coordinate analysis (also and to depend on several different factors called metric multidimensional scaling), such as the production of cytotoxic and canonical discriminant and correlation cytotonic enterotoxins, haemolysin, lipase analyses or the multiple regression with and elastase and the ability to move by forward selection analysis by Anderson fl agella (Janda 1991, Sen and Rodgers (2003a and b, 2004) allow the distance 2004, Galindo et al. 2006). PCR primers or dissimilarity measure of the analysis have been designed for many of the genes to be chosen by the user. This enables the that encode the virulence features and the analyses included in the programs to be presence of such genes has been screened adjusted for a wide range of variable sets for in both clinical and environmental ranging from DNA sequence data to often Aeromonas strains (e.g. Chacón et al. zero infl ated species abundance data. In 2003, Sen and Rodgers 2004). However, addition, the methods are suitable for data to writer’s knowledge, studies about in which the number of observations far Aeromonas strains associated with exceeds the number of variables. cyanobacterial water blooms have not been The analyses included in the principal performed and the virulence potential of coordinate analysis program (PCO) such Aeromonas in general is still largely can be used to visualize the ordination unknown. of data on the basis of their calculated distances such as grouping of bacterial 1.5.5 Statistical analyses of multivariate strains according to their 16S rRNA data sequence distances (Anderson 2003b). Possible patterns and associations of the The canonical discriminant and correlation studied variables, such as physiological analyses included in the canonical analysis characteristics, genetic composition or of principal coordinates program (CAP), the isolation source of the bacteria, can be can be used to uncover patterns in the examined and visualized using numerical multivariate data and to test a priori or statistical analyses. These include hypotheses of differences among groups methods such as the principal coordinate (Anderson and Robinson 2003, Anderson analysis, discriminant, regression or and Willis 2003). The regression analyses correlation analyses (Legendre and included in the distance-based multivariate Legendre 1998a and b, Zuur et al. 2007). analysis for a linear model using forward

11 Introduction selection program (DISTLMf) can be used 1.6.1 Adverse human health effects to analyse hypotheses on the relationship associated with cyanobacterial water between chosen response variables and blooms explanatory variables (McArdle and One of the problems associated with Anderson 2001, Anderson 2003a). The the cyanobacterial water blooms is the methods can also be combined, in order adverse human health effects they can to select the most notable variables for the cause. The health effects associated with canonical discriminant analysis. cyanobacterial mass occurrences vary from mild to fatal and from chronic to acute 1.6 Problems associated with (Kuiper-Goodman et al. 1999, Stewart cyanobacterial water blooms 2006a and b). Typical routes of exposure Eutrofication of aquatic systems has are direct physical contact with blooms, lead to the worldwide enforcement of ingestion of water containing bloom matter cyanobacterial mass occurrences and or by exposure to the aerosols emanating the increasing problems caused by from the bloom. Possible consequences of them (Bartram et al. 1999, Paerl 2008). exposure to a cyanobacterial water bloom Cyanobacterial water blooms can cause can be quite varied and include effects such various kinds of nuisances. As a relatively as skin irritation, fever, headache, nausea, mild consequence the cyanobacterial cell vomiting, muscular pains, gastrointestinal, mass can produce unpleasant odours and neurological and liver symptoms, and taste to water but their effects can be much symptoms of eyes, nose, ears and throat. more profound and damaging (Falconer The various kinds of toxins et al. 1999). Large cyanobacterial masses cyanobacteria produce as secondary can change the physical and chemical metabolites are often suspected, and in features of a water body, such as lighting some cases proven to be the source of and the availability of growth limiting the adverse health effects. The severity nutrients, thus causing changes to the of the effects depends on many factors aquatic species composition and nutrient such as the magnitude and length of the web (Sivonen 2000, Fournie et al. 2008). exposure, the type and concentration of Adverse health effects in humans and the cyanobacteria and their toxins, the type poisonings of cattle and wildlife such as of contact, the age and the health status of ﬁ sh, birds and muskrats associated with the patient (Kuiper-Goodman et al. 1999, toxic cyanobacterial blooms in natural Salmela et al. 2001, Hoppu et al. 2002, water bodies are also often reported Stewart 2006b). However, the specific (Kuiper-Goodman et al. 1999, Stewart cause of the health effects often remains 2006b, Stewart et al. 2008, Sivonen 2009). unclear, due to the lack of sufficient In addition, insufﬁ cient drinking water sample data and epidemiological studies treatment of source water contaminated on the reported cases (Bartram et al. 1999, by cyanobacterial mass occurrence has Stewart 2006b, Pilotto 2008). caused adverse human health effects Despite evidence of the close (Kuiper-Goodman et al. 1999, Falconer interactions between cyanobacteria and Humpage 2005). and heterotrophic bacteria, the possible influence of these associations have rarely been studied when assessing the health risks posed by the blooms (Chorus

12 Introduction et al. 2000, Stewart et al. 2006b). For infections in risk groups in humans such example, the assessment of human health as immunocompromized persons, children risks associated with cyanobacterial and the elderly can be high, 25−75 % water blooms has been strongly focused (Janda and Abbott 1998). Underlying on cyanobacteria and their toxins, even diseases of malignancy, hepatobiliary though many of the detected human health disease and diabetes can also increase effects, such as skin- and gastrointestinal the susceptibility to Aeromonas infection problems, could also have been caused by and severity of the obtained illness heterotrophic bacteria. (Figueras 2005). Many of the Aeromonas Members of the genera such as species are resistant to several antibiotics Aeromonas and Vibrio have been and infections caused by them can associated with adverse health effects and progress quite rapidly and with serious are indigenous to aquatic environments consequences, even among persons with worldwide (Thompson et al. 2004, Martin- no underlying diseases (Figueras 2005, Carnahan and Joseph 2005). For example, Martin-Carnahan and Joseph 2005). A. caviae, A. eucrenophila, A. hydrophila, Vibrio cholerae is best known for the A. jandaei, A. media, A. popoffii, A. cholera epidemics and pandemics that the sobria and A. veronii are found in aquatic strains of the serogroups O1 and O139 environments (Hänninen and Siitonen can cause (Faruque et al.1998, Leclerc 1995, Borrell et al. 1998, Martin-Carnahan et al. 2002). However, other V. cholerae and Joseph 2005). The highest Aeromonas strains and strains of V. parahaemolyticus concentration in water environments are and V. vulniﬁ cus are also associated with usually reached during warm seasons human infections such as gastroenteritis, when the temperatures are also optimal skin infections, wound infections and for growth of most cyanobacteria, and septicaemia (Thompson et al. 2004, recreational use of water bodies is most Lukinmaa et al. 2006). common (Mur et al. 1999, Castenholz In addition, cyanobacteria together 2001, Farmer et al. 2006, Martin-Carnahan with heterotrophic bacteria may represent and Joseph 2005). an increased health risk. The infection In clinical samples the A. caviae, susceptibility of persons exposed to A. veronii and A. hydrophila species are cyanobacterial water blooms could the most prevalent Aeromonas detected be lowered by the adverse effects (Janda and Abbot 1998, Figueras 2005, of the cyanobacterial toxins. The Lamy et al. 2009). Other Aeromonas lipopolysaccharides of the Gram-negative found in human clinical samples include bacteria, in turn, inhibit glutathione species such as A. media, A. jandaei, S-transferase activity, which is important mesophilic A. salmonicida, A. sobria, in detoxifying of microcystins, and A. allosaccharophila, A. popofﬁ i and A. may increase the risk posed by the eucrenophila. The adverse human health cyanobacterial hepatotoxins (Best et al. effects associated with Aeromonas include 2002, Rapala et al. 2002b). Therefore, conditions such as diarrhoea, wound studies on the heterotrophic bacteria infection, bacteraemia, urinary tract associated with planktonic cyanobacteria infection, pneumonia and fever (Janda could provide valuable information for the 1991, Janda and Abbot 1998, Figueras risk assessment of these blooms. 2005). The mortality to Aeromonas

13 Introduction

1.6.2 Cyanobacterial water blooms and life situation during a cyanobacterial water drinking water bloom in the source water can be hard to The key to microbiologically, chemically assess. and radiologically safe drinking water lies in the use of the combination of 1.6.3 Cyanobacterial toxins and the different treatment and monitoring steps, drinking water treatment to ensure suffi cient water quality despite If surface water is used, the water the normal performance fl uctuations of treatment steps usually include individual stages (LeChevallier and Au coagulation, fl occulation and clarifi cation 2004b, WHO 2008c and f). This multiple (by sedimentation or fl otation) followed barrier approach begins from source water by fi ltration and disinfection (LeChevallier protection and ends at the protection of the and Au 2004d). The effectiveness of distribution system (LeChevallier and Au different treatment procedures on the 2004b). removal of cyanobacterial toxins has Cyanobacterial blooms can cause been assessed but the available data are problems for drinking water treatment still quite limited (Lawton and Robertson all the way from the water source to the 1999, Hitzfeld et al. 2000, Svrcek and drinking water distribution (Hrudey Smith 2004). Most of the studies have et al. 1999, Svrcek and Smith 2004). concentrated on the removal of the Toxic cyanobacteria are listed among chemically stable hepatotoxins. However, emerging pathogens that may contribute to different cyanobacterial toxins are waterborne diseases and the World Health removed with varying degree of effi ciency Organization (WHO) has proposed a and the effi cient removal of hepatotoxins lifetime safe consumption level of 1 μg l-1 does not guarantee removal of other for the microcystin-LR in drinking water types of cyanobacterial toxins (Hitzfeld (Falconer et al. 1999, Hunter et al. 2003). et al. 2000, Svrcek and Smith 2004). For Studies on the effects of cyanobacteria example, ozonation of water can effi ciently on the drinking water quality have remove cyanobacterial hepatotoxins and concentrated mainly on the removal and anatoxin-a but is quite ineffective against inactivation of cyanobacterial toxins, saxitoxins (Lawton and Robertson 1999, in most cases microcystins, by specifi c Hitzfeld et al. 2000, Rositano et al. 2001, treatment procedures (Kuiper-Goodman Svrcek and Smith 2004). et al. 1999, Lawton and Robertson 1999, The common factor affecting the Hitzfeld et al. 2000, Svrcek and Smith effi ciency of removal of the cyanobacterial 2004). Studies on operational water toxins in many of the treatment steps is treatment plants are also largely centred whether the toxins remain inside or are on microcystins, and investigations on released outside the cyanobacterial cells other factors such as the removal of (Hitzfeld et al. 2000, Svrcek and Smith cyanobacterial neurotoxins, phytoplankton 2004). The coagulation, clarifi cation and cells and endotoxins are few (Lepistö et fi ltration steps are used mainly to remove al. 1994, Lambert et al. 1996, Lahti et particulate matter such as microbial cells al. 2001, Karner et al. 2001, Rapala et al. (LeChevallier and Au 2004d, WHO 2002b, Hoeger et al. 2005). Therefore, the 2008b). By detaining the cyanobacterial overall water purifi cation effi ciency of cells, these water treatment steps also different treatment combinations in a real effi ciently remove cyanobacterial toxins

14 Introduction within those cells (Hitzfeld et al. 2000, toxins are the main target of the toxin Drikas et al. 2001, Svrcek and Smith removal, so that the chosen treatment steps 2004). In contrast, coagulation and will not counteract with each other (Lawton clarifi cation are quite ineffi cient against and Robertson 1999, Svrcek and Smith toxins already released from the cells into 2004). Various methods and application are water. Filtration, especially with slow sand available for each of the usual purifi cation fi lters that contain a biological layer, may steps. For example, in the fi ltration step also help to remove the dissolved toxins rapid gravity and pressure fi lters or slow by biological degradation (Svrcek and sand fi lters can be used (Svrcek and Smith Smith 2004, LeChevallier and Au 2004d, 2004, LeChevallier and Au 2004d, WHO Ho et al. 2006, WHO 2008b). 2008b). Additional steps such as pre- Disinfection procedures such as treatment of the water with bank fi ltration chlorination and ozonation destroy or precipitation of calcium and magnesium cyanobacterial toxins outside the cells by adjusting the pH of water with lime, more effi ciently than when the toxins are can also be used as a part of the treatment protected by intact cell structures (Hitzfeld process (LeChevallier and Au 2004d, et al. 2000, Svrcek and Smith 2004). The WHO 2008b). The removal of organic primary aim of disinfection is to inactivate compounds, such as the cyanobacterial viable microbes and the concentrations toxins, can be further improved by using and contact times used in disinfection are granular-activated carbon or powdered optimized for this purpose (LeChevallier activated carbon to adsorb them (Cook and Au 2004a, WHO 2008b). In addition, and Newcombe 2002, Logsdon et al. compounds such as ozone can be used 2002, Svrcek and Smith 2004). Effi cient as preoxidants to aid the coagulation and biodegradation of microcystins has also fi ltration processes (Widrig et al. 1996, been detected during granular activated Yukselen et al. 2006, Chen et al. 2009, Li carbon filtration, indicating that both et al 2009). Ozonation within the range adsorption and biodegradation contribute of normal disinfectant concentrations and to the removal and inactivation of the contact times used in water treatment is toxins (Wang et al. 2007). quite effi cient against most toxin types (Hrudey et al. 1999, Lawton and Robertson 1.6.4 Combined effects of 1999, Hitzfeld et al. 2000, Rositano et al. cyanobacterial water blooms and 2001, Svrcek and Smith 2004, Westrick heterotrophic bacteria on the drinking 2008). The efficiency of chlorination water quality depends on the chlorine compound used In addition to the produced toxins, and higher concentrations and longer cyanobacterial water blooms in general contact times than normally used are often can cause problems for the water treatment required for effective toxin inactivation. and water quality by the changing physical Since the source water can contain and chemical conditions that affect the both cell retained toxins and extracellular overall functioning of the water treatment toxins, the water treatment process processes. They also produce compounds, should be able to cope with both of them. such as geosmin, that can cause unpleasant In addition, when designing the water taste and odour to the water (Peterson treatment train, it should be considered et al. 1995, Chow et al. 1998, Falconer whether the cell retained or extracellular et al. 1999). Cyanobacterial water

15 Introduction blooms in the source water can cause Legionella, Mycobacterium, Aeromonas substantial increases in the amounts of or Pseudomonas (Percival and Walker organic particles and overall amounts of 1999, Theron and Cloete 2002, Berry et organic compounds that enter the water al. 2006, WHO 2008c and e). The biofi lms treatment system (Hrudey et al. 1999). offer protection from the effects of the This additional organic matter can cause disinfectants to the organisms within them problems for the water quality, if the (Percival and Walker 1999, Theron and treatment process is not suffi ciently quickly Cloete 2002, Berry et al. 2006,WHO 2008a adjusted to match the new situation. The and e). In addition, the interaction between adjustment can be made, for example, by biofi lms and disinfectants can lead to the changing the source water intake depth, by formation of harmful compounds such optimizing the coagulation process, or by as nitrite. The interactions between the utilizing activated carbon for the removal materials of the distribution system and the of organic substances or by increasing bacterial cells as well as the compounds, the amounts of disinfectant used (Bursill such as acids, produced by the microbial 2001, LeChevallier and Au 2004a and d, metabolism can lead to substantial Svrcek and smith 2004, Westric 2008). corrosion of the distribution system

In addition, the CO2 consumption by (Percival and Walker 1999). This causes the photosynthetic reactions within the problems to both the water quality and cyanobacterial bloom can elevate the raw the condition of the distribution system, water pH, which in turn can affect several which may have significant economic water treatment steps, especially those consequences. The lipopolysaccharides of of coagulation and ozonation (Chow et the Gram-negative heterotrophic bacteria al. 1998, Hitzfeld et al. 2000, Briley and may also increase the risk posed by the Knappe 2002, WHO 2008b, Havens 2008, cyanobacterial hepatotoxins that may pass Westrick 2008). through the water treatment process (Best During cyanobacterial mass et al. 2002, Rapala et al. 2002b). Therefore occurrences, when a treatment system is the assessment of risks posed by the struggling at its uppermost limit, higher cyanobacterial water blooms to drinking than normal amounts of heterotrophic water safety should cover the possible bacterial and phytoplankton cells may pass combined effects of cyanobacteria and through the treatment system (Lepistö et heterotrophic bacteria. al. 1994). This may increase direct health As a counterbalance to their negative risks to the water consumers (WHO effects, the heterotrophic bacteria can 2008d). The viable bacteria and organic also provide a means to control the matter passing through the water treatment risks posed by cyanobacterial water system may also cause problems to the blooms to the drinking water safety and water quality by biofi lm formation in the quality. In a study by Ho et al. (2006) distribution system, which can affect the microcystins were shown to be removed water quality both directly and indirectly from water through a process of biological (Rusin 1997, Percival and Walker 1999, degradation in sand fi lters. Furthermore, WHO 2008a, c and e). Biofi lms can affect a Sphingopyxis strain that degraded the taste and odour as well the colour of microcystins was isolated from such a the water and in some occasions function fi lter (Ho et al. 2007). The studies showed as a shelter for putative pathogens, such as that the heterotrophic bacteria adapted to

16 Introduction metabolize the cyanobacterial hepatotoxins and Bird 2000). Therefore, the overall can also contribute to the removal of risks of cyanobacterial water blooms to the toxins from water during the water the drinking water quality could perhaps treatment process. Such bacteria might be be reduced by utilizing such bacteria in utilized in the removal of the toxins in the controlling cyanobacterial blooms in drinking water treatment, for example as drinking water reservoirs. However, more a part of the biological layer of the slow information on the characteristics of such sand ﬁ lters. In addition, cyanobacterial bacteria and on their interactions with growth inhibiting bacteria have been cyanobacteria are needed for safe and proposed to be used as biological control efﬁ cient use of such applications. agents of cyanobacterial blooms (Rashidan

17 Aims of the Study

2. AIMS OF THE STUDY

Cyanobacterial mass occurrences, the water blooms, frequently cause problems for recreational and potable use of water. The factors that affect the health risks associated with cyanobacterial water blooms are in many respects still unclear. In addition, the studies that assess the overall water treatment efﬁ ciency of operating treatment plants encumbered with cyanobacterial mass occurrences are rare. The risk assessment concerning the cyanobacterial blooms has largely been focused on the cyanobacteria and their toxins. However, the water blooms are associated with diverse communities of heterotrophic bacteria. These bacteria may, for their part, affect the development and decline of the water blooms and also contribute to the health problems caused by the blooms, but little is known about them. The studies concerning these bacteria have mainly been based on cultivation-independent molecular methods that provide no bacterial isolates for closer study.

The aims of this study were:

- to examine the community of cultivable heterotrophic bacteria associated with the cyanobacterial mass occurrences using taxonomic and phenotypic characterisation of bacterial isolates (I).

- to assess the virulence potential of Aeromonas spp. strains isolated from cyanobacterial water blooms, by studying the presence of six virulence gene types specific to Aeromonas (II).

- to study the genetic and phenotypic characteristics of previously isolated bacterial strains shown to be able to degrade cyanobacterial hepatotoxins (III).

- to assess the water puriﬁ cation efﬁ ciency of a modern drinking water treatment process at an operating water treatment plant, during mass occurrence of cyanobacteria in the source water (IV).

18 Materials and Methods

3. MATERIALS AND METHODS

3.1 Water samples and isolated bacterial strains Water samples were collected from Finnish lakes, rivers or ponds and from the Baltic Sea (I, II, IV) and from treated drinking water of water treatment plants (I, IV). The sample sources were selected based on increased cyanobacterial occurrence in the water bodies or in the water source utilized for producing the drinking water. Samples were also collected from sites where a cyanobacterial water bloom was suspected to have caused human health symptoms (II). The water samples and bacterial strains isolated from the samples are identiﬁ ed in the original articles (I, II, IV). In addition, 13 previously isolated hepatotoxin degrading Betaproteobacteria strains were studied (III).

3.2 Methods The methods used in this study are listed in table 1. A more detailed description of the methods are given in the original papers (I−IV)

Table 1: Methods used in the study Method Paper Sampling and strain isolation I, II, IVa Telephone interviews for collecting data on adverse human health effects II Phytoplankton identifi cation and biomass analysis IV Gram staining I, II, III, IV Flagella staining III API NE and API ZYM tests III ZymProfi ler® test III GN MicroPlate test III Fatty acid analysis by gas-liquid chromatography (GLC) III Bacterioclorophyll a analysis by spectrophotometry III Oxidase and catalase tests II, III Aeromonas hydrophila (AH) medium tests II Motility on casitone-yeast extract agar III Haemolytic activity on blood agar I Testing of the effects of the isolates on cyanobacterial growth I Hepatotoxins by enzyme-linked immunosorbent assay (ELISA) II, IV Hepatotoxins by protein phosphatase inhibition assay III, IV Endotoxins by kinetic chromogenic Limulus amoebocyte lysate assay II, IV Overall toxicity by Artemia salina bioassay IV Hepatotoxin degradation test III Anatoxin-a by high-performance liquid chromatography (HPLC) II, IVb Saxitoxins by Jellett Rapid paralytic shellfi sh poisons test strips II, IVc Putative presence of anatoxin-a(S) by acetylcholine esterase inhibition test II, IVd DNA extraction from bacterial strains I, II, III, IVe 16S rRNA gene amplifi cation and sequencing I, II, III, IVf Genomic guanine-cytosine content by high-performance liquid chromatography III (HPLC) Taxonomic classifi cation by basic local alignment search tool (BLAST) I, III, IV

19 Materials and Methods

Table 1 cont.

Method Paper Taxonomic classifi cation by taxonomic classifi er and sequence match tools I, II 16S rRNA gene sequence similarity analysis by BioEdit II, III Phylogenetic analysis with neighbour-joining and maximum parsimony criteria III Visualization of a priori grouping by principal coordinate analysis II Screening of Aeromonas virulence genes by PCR II Multivariate multiple regression analysis II Canonical discriminant analysis II aStrain isolation was done as described in I. bThe presence of anatoxin-a, homoanatoxin-a, and their epoxy and dihydro degradation products were analysed as described by James et al. (1997). cThe presence of saxitoxins was analysed using the commercial Jellett Rapid paralytic shellfi sh poisons test strips (Jellett et al. 2002). dThe putative presence of anatoxin-a(S) was analysed as described by Ellman et al. (1961). eThe DNA extraction was executed as described in III. fSequencing was carried out at the Institute of Biotechnology of the University of Helsinki.

20 Results

4. RESULTS the confidence threshold (80%) that was set as the limit for the taxonomic 4.1 Heterotrophic bacteria assignment by the Ribosomal Database associated with cyanobacteria Project II classiﬁ er (Fig 1, Appendix 1; I: In order to study the heterotrophic bacteria Table 1, supplementary data; II: Table 3, in association with cyanobacteria, over supplementary data). The isolated strains 600 bacterial strains were isolated. also included several of those that had These strains originated from freshwater, no close relatives (maximum identity brackish water or treated drinking water values <98.7% by basic local alignment samples, which were obtained from sites search tool (BLAST) analysis) among with increased cyanobacterial occurrence data base sequences of either the 16S in the water bodies or in the water source rRNA genes of cultivated bacterial strains of the drinking water. Chosen subsets of or of uncultivated bacteria (I: Table 2). the strains were studied for characteristics Therefore, the strains represented a diverse such as a set of virulence genes in 176 group of heterotrophic bacteria, including Aeromonas spp. strains, or for their effect previously undescribed bacterial species. on cyanobacterial growth in 183 strains of various bacterial genera. In addition, 4.1.2 Bacteria degrading cyano- 13 cyanobacterial hepatotoxin degrading bacterial hepatotoxins: Paucibacter bacteria strains previously isolated from toxinivorans gen. nov. sp. nov. Finnish lakes with frequent cyanobacterial The previously isolated 13 Betaproteo- occurrence were characterised. bacteria strains were chosen for characterization due to shown ability 4.1.1 Taxonomic assignment of the to degrade cyanobacterial hepatotoxins isolated bacterial strains microcystins and nodularin, which had The isolated strains represented bacteria been tested mainly using one microcystin from ﬁ ve different phyla, eight different variant and with two of the strains also classes, and 57 different genera based using nodularin (III). Five of the strains on analyses of their partial (397–905 were tested for the ability to degrade of bp) 16S rRNA gene sequences (Fig. 1, nodularin of the Nodularia 790 strain and Appendix 1; I; II; IV). The majority (82%) microcystins LR and YR (leucine and of the strains were assigned to Gamma-, arginine (LR) or tyrosine and arginine Alpha- and Betaproteobacteria classes, (YR) as the aminoacids 2 and 4) of whereas Actinobacteria, Flavobacteria, the Microcystis 764 strain (Appendix Sphingobacteria and Bacilli formed 18% 1; III). All five strains degraded both of the isolated strains. The Alpha- and hepatotoxin types, as measured by the Betaproteobacteria were both represented protein phosphatase inhibition analysis. by 13 different genera, and Actinobacteria, Nodularin was degraded at the range of Flavobacteria, Sphingobacteria, Bacilli maximum rates from 3 to 560 fg cell-1 h-1 and Gammaproteobacteria by 3–9 genera. and the microcystins LR and YR at the The one strain representing the class maximum rates from 2 to 30 fg cell-1 h-1. Deinococci belonged to Deinococcus. The results implied that the strains have In addition, 21 of the isolated strains a stronger ability to metabolize nodularin could be assigned only at family level than microcystins. and 3 strains only to class level above

21 Results

The taxonomy and physiological leucine arylamidase (III: Supplementary characteristics of the 13 strains were table S1). In addition, all were positive studied using molecular and biochemical for acid phosphatase, naphthol-AS-BI- analyses. The closest relatives of the phosphohydrolase, phosphodiesterase, strains, based on nearly complete (1384– esterase, lipase and lipase esterase. The 1444 bp) 16S rRNA gene sequences, were enzyme reactions varied from weak to the Betaproteobacteria strains R-8875 strong. Most of the 13 strains were also (AJ440994, 97.0–97.1% similarity) and positive for phosphomonoesterase and CD12 (AF479324, 96.5–96.8%), Mitsuaria cysteine and valine arylamidases. The chitosanitabida (AB006851, 96.1–96.8%, cells grew heterotrophically under aerobic formerly Matsuebacter chitosanotabidus) conditions. All strains utilized Tween 40 as and Roseateles depolymerans (AB003623 a carbon source but only few other carbon and AB003625, 95.7–96.3%; III: Fig. 1). sources were utilized (III: Supplementary Based on the 16S rRNA gene sequence table S2). The results indicated that the similarities, the 13 strains could be strains had a high potential to mineralize assigned only at the family Incertae cedis organic phosphorus compounds and 5 of the order Burkholderiales, implying that they utilized amino acid residues. that they represented a new bacterial In contrast, they had only limited genus. potential in using common environmental The strains were oxidase positive macromolecules for carbon and sulphur and weakly catalase positive. All strains sources. were strongly or clearly positive for The cells of the strains were Gram- alkaline phosphatase, chymotrypsin and negative, rod-shaped (0.5–0.7×1.3–5.0

A Number of the strains

0 40 80 120 160 200 240 280 320 360 Actinobacteria 5% Flavobacteria 7% Sphingobacteria 3% Deinococci 02% Bacilli Class 3% Alphaproteobacteria 16% Betaproteobacteria 9% Gammaproteobacteria 57%

TOX CYA R2A Z8 BA SAA

Figure 1: Taxonomic distribution of the isolated heterotrophic bacterial strains in total and in relation to the used isolation media. A: at the class level. The percentage of the strains representing each class, of the total number of the isolated strains, is shown together with the class bars. B: at the genus level. TOX: medium with demethyl variants of microcystins. CYA: medium with frozen and thawed culture of non-microcystin-producing Anabaena. R2A: Reasoner’s 2A medium. Z8: nutrient poor medium. BA: blood agar. SAA: starch ampicillin agar.

22 Results

B Number of the strains 0 10 20 30 33121 282 Arthrobacter Kocuria Microbacterium Micrococcus 1 Mycobacterium Nocardia Nocardioides Rhodococcus Streptomyces Assigned at family level Chryseobacterium 2 Flavobacterium Riemerella Assigned at family level

Arcicella 1 Actinobacteria Chitinophaga 6 3 Flectobacillus Alphaproteobacteria Pedobacter Spirosoma 4 Deinococcus Bacillus Caryophanon 2 5 Lactococcus Flavobacteria Paenibacillus Staphylococcus 7

Azospirillum Betaproteobacteria Blastomonas Bosea

Brevundimonas 3

Caulobacter Sphingobacteria Methylobacterium Novosphingobium 6 Paracoccus Phyllobacterium Porphyrobacter Class (1–8) and genus

Sphingobium 8

Sphingomonas Gammaproteobacteria Sphingopyxis

Assigned at family level 4

Assigned at class level Deinococci Aquabacterium Aquitalea Burkholderia Chitinimonas

Chromobacterium 5

Duganella Bacilli Herbaspirillum 7 Iodobacter Janthinobacterium Pelomonas Rhodoferax Roseateles Vogesella Assigned at family level Assigned at class level Acinetobacter Aeromonas Erwinia Pseudomonas 8 Rheinheimera Serratia Stenotrophomonas Vibrio Assignedat family level

TOXTOX CYA R2A R2A Z8 Z8 BA BA SAA SAA

23 Results

μm) and motile by a single polar ﬂ agellum 4.1.3 Virulence potential of the (III). They formed greyish colonies on the heterotrophic bacteria associated with relatively oligotrophic R2A medium, but cyanobacteria did not grow on rich growth media such Haemolytic strains representing 11 genera as trypticase soy agar (TSA; III). The were isolated by choosing haemolytic diameter of the colonies on the R2A was colonies from the blood agar (Fig. 2; I: ≤1 mm after incubation for 65 h at 20±2 supplementary data). The most dominant ºC. The cells grew well at 20–30 ºC but genera among the haemolytic strains were not at 37 ºC. The main fatty acids were Aeromonas spp. (69%), Acinetobacter spp.

C16:0, C16:1ω7c; and ω7c, ω9c and ω12t (8%), Vibrio spp. (8%) and Pseudomonas forms of C18:1 (III: Supplementary table spp. (5%). The haemolytic characteristic S3). The guanine-cytosine content of the of the strains implied their potential genomic DNA was 66.1–68.0 mol% (III: virulence. Supplementary table S2). On the basis Due to the high proportion of the of these characteristics a new genus and haemolytic Aeromonas spp. strains this species Paucibacter toxinivorans was genus was targeted for a closer study. The described, with strain 2C20 as the type majority (90%) of the 176 Aeromonas strain. strains isolated from the fresh and brackish

Pseudomonas; 6; 5% Rheinheimera; 2; 2% Vibrio; 11; 8% Vibrionaceae; 1; 1%

Aeromonas; 92; 69% Flavobacteriaceae; 1; 1% Bacillus; 2; 2% Staphylococcus; 1; 1% Brevundimonas; 1; 1% Aquitalea; 1; 1% Burkholderia; 1; 1% Chromobacterium; 4; 3%

Acinetobacter; 10; 8%

Bacteroidetes: Flavobacteria Firmicutes: Bacilli Proteobacteria: Alphaproteobacteria Betaproteobacteria Gammaproteobacteria

Figure 2: Taxonomic distribution of the isolated haemolytic bacterial strains. The number of the strains assigned at each species or family level and their percentage of the haemolytic strains (n=133) are shown together with their taxon.

24 Results water samples that were associated with including also 73% of the strains in various adverse health effects in humans genogroup I that contained the tentative A. contained at least one out of the six veronii and A. jandaei strains (II: Fig. 2). screened Aeromonas virulence gene types, The results suggested a higher virulence based on the PCR analysis (II: Fig. 2). The potential of the strains in genogroups II most common virulence gene type was the and III compared to those of genogroup I. cytotoxic enterotoxin and haemolytic gene The water samples, from which the products encoding act (act/aerA/hlyA) that screened Aeromonas strains were isolated, was present in 77% of the strains studied were also analysed for cyanobacterial (II: Fig. 2). The fl agellin subunits encoding hepatotoxins, neurotoxins and bacterial fl a (fl aA/fl aB), elastase encoding ahyB and endotoxins (II). The concentrations of the lipase encoding lip (lip/pla/lipH3/alp-1) hepatotoxins microcystins and nodularin gene types were present in 30–37% of varied from below the detection limit of the strains. The counts of the presumptive 0.1, to 12.3 μg l-1 in the majority (97%) Aeromonas colonies measured on the of the analysed samples (II: Table 2). The starch ampicillin agar medium from the median of the concentrations was below water samples varied between 5−4.5×104 the detection limit of 0.1 μg l-1. Of the colony forming units (CFU) ml-1 with neurotoxins, low concentrations (1.9–10.3 a median of 68 CFU ml-1 (II). The most μg l-1) of anatoxin-a, homoanatoxin-a common health effects reported by or their degradation products were the affected persons were fever and detected in 4 of the 22 analysed samples gastrointestinal disturbances (II: Fig. 1). and acetylcholinesterase inhibition that The common occurrence of the virulence suggested the presence of anatoxin-a(S) genes among the studied Aeromonas in 1 of the 31 analysed samples. The strains indicated a virulence potential of presence of saxitoxins was not detected the strains and suggested that they may be in the 27 samples analysed. The measured the source of some of the reported adverse concentrations of bacterial endotoxin health effects. varied from 10 to 300 endotoxin units The Aeromonas strains that were (EU) ml-1 in the majority (89%) of the screened for the virulence genes formed samples, with the median of 58 EU three genogroups (II: Table 3). The ml-1. The estimated endotoxin aerosol cytotonic enterotoxin encoding ast concentrations of the majority (89%) of gene was strongly associated with the the samples varied between 0.3–60.4 genogroup III that contained the tentative EU m-3 (II). Based on the resuts, the A. hydrophila and A. media strains, analysed toxins were either present in low identifi ed on the basis of their partial 16S concentrations or undetectable from the rRNA gene sequences (II: Fig. 3). The lip, samples. fla, ahyB and the cytotonic enterotoxin encoding alt genes were associated 4.1.4 Effects of the heterotrophic most strongly with the genogroup II that bacteria on the growth of contained the tentative A. salmonicida, cyanobacteria A. popoffii and A. sobria strains. The The majority (60%) of the 183 associations were studied using canonical heterotrophic bacterial strains analysed discriminant analyses. The most abundant either enhanced or inhibited the growth of virulence gene type, act, was detected the microcystins-producing Microcystis in strains of all the three genogroups, PCC 7941 or the non-microcystin-

25 Results producing Anabaena PCC 7122 4.2 Selectivity of the used growth strain when they were incubated with media cyanobacterial cells on agar plates (Fig. 3; The strains isolated using the TOX I: Table 3). The growth of one or both of medium that contained demethyl variants the cyanobacterial strains was enhanced by of microcystins, represented all 8 classes 48% of the heterotrophic bacterial strains and 51% of the total 57 genera that were and inhibited by 10% of the strains. Two identified among the bacterial strains of the strains (1%) enhanced the growth isolated in this study (Fig. 1, Appendix of one of the test cyanobacterial strains 1; I: Fig 2). The strains from the CYA and inhibited the growth of the other. medium, which contained frozen and Most notably, the cyanobacterial growth thawed culture of non-microcystin- was enhanced by over 50% of the tested producing Anabaena and compounds for Alphaproteobacteria, Betaproteobacteria allowing the isolation of slow growing and Actinobacteria strains. Overall, the bacteria, represented 7 classes and 47% isolated strains representing all eight of the genera. Those growth media had bacterial classes and 30 different genera been formulated to yield putative utilisers enhanced or inhibited the cyanobacterial of cyanobacterial biomass and toxins. The growth. The results implied that the ability strains from the R2A medium, chosen to to affect cyanobacterial growth is common yield high numbers of different bacteria, among the cyanobacteria associated represented 6 classes and 46% of the heterotrophic bacteria. genera. The strains from the nutrient poor

A Number of the strains 0 5 10 15 20 25 30 35 40 45 50 55 Actinobacteria 29% 54% 17% Flavobacteria 59% 22% 19% Sphingobacteria 40% 35% 25% Deinococci 100% Class Bacilli 25% 50% 25% Alphaproteobacteria 35% 60% 4% 2% Betaproteobacteria 36% 58% 3%3% Gammaproteobacteria 55% 45%

No effect Enhanced Inhibited Enhanced/inhibited

Figure 3: Effect of the tested heterotrophic bacterial strains on the growth of the microcystins- producing Microcystis PCC 7941 and non-microcystin-producing Anabaena PCC 7122 strains. A: at the class level. The percentages of strains with speciﬁ c effect on cyanobacterial growth, of the total number of tested strains within each class, are shown together with the class bars. B: at the genus level. The strains identiﬁ ed only at family level are underlined. No effect: the strain did not affect the cyanobacterial growth. Enhanced: the strain enhanced the growth of one or both of the cyanobacterial strains. Inhibited: the strain inhibited the growth of one or both of the cyanobacterial strains. Enhanced/inhibited: the strain enhanced the growth of one of the cyanobacterial strain and inhibited the growth of the other.

26 Results

B Number of the strains

0 5 10 15 20 25 30 Arthrobacter Kocuria Microbacterium Micrococcus 1 Nocardioides Rhodococcus Streptomyces Microbacteriaceae

Chryseobacterium 2 Flavobacterium Riemerella 1 Actinobacteria Arcicella 6

Chitinophaga Alphaproteobacteria 3 Flectobacillus Pedobacter Spirosoma

4 Deinococcus 2 Flavobacteria

5 Bacillus Staphylococcus 7 Betaproteobacteria Betaproteobacteria Azospirillum

Bosea 3

Brevundimonas Sphingobacteria Caulobacter Methylobacterium 6 Novosphingobium Paracoccus 8

Sphingobium Gammaproteobacteria

Class (1–8) genusand family or Sphingomonas 4

Sphingopyxis Deinococci Acetobacteraceae

Duganella Herbaspirillum 5

Iodobacter Bacilli Janthinobacterium Pelomonas 7 Rhodoferax Roseateles Vogesella Incertae sedis 5 Oxalobacteraceae

Acinetobacter Aeromonas 8 Pseudomonas Rheinheimera Stenotrophomonas

No effect Enhanced Inhibited Enhanced/inhibited

27 Results

Z8 medium represented four classes and plant (IV). The assessed factors were the 11% of the genera. The strains from the removal of planktonic cells, cyanobacterial blood agar (BA) medium, from which hepatotoxins, cultivable heterotrophic haemolytic colonies were primarily bacteria and endotoxins. The water chosen, represented 6 classes and 28% treatment process consisted of coagulation of the genera. The Z8 media was chosen with aluminium salts, clarification by for yielding the most oligotrophic bacteria sedimentation, sand fi ltration, ozonation, and the BA medium chosen for yielding slow sand fi ltration and chlorination. In putative pathogenic bacteria. addition, the taxonomic distribution of the Altogether, the strains isolated using heterotrophic bacterial genera originating TOX, CYA and R2A media represented a from treated drinking water (included majority (84%) of the 57 isolated genera in 4.1.1) of the above mentioned water (Fig. 1, Appendix 1; I: Fig 2). However, treatment plant and of six other plants was genera such as Rhodococcus, Aeromonas assessed (I, IV). and Pseudomonas were represented only or mostly by strains isolated using the 4.3.1 Removal of toxins by the water BA or Z8 media. Based on these results, treatment combination of TOX, CYA and R2A media The microcystin concentrations in the was suitable for isolating a high variety of treated drinking water varied from <0.02 heterotrophic bacteria co-occurring with to 0.03μg l-1, based on the enzyme-linked cyanobacteria but the BA and Z8 media immunosorbent assay (ELISA) and protein provided a notable addition of putative phosphatase inhibition analyses (IV: Fig 2). pathogens and oligotrophic bacteria to the The concentrations of microcystins in the total number of isolated strains. incoming water varied from <0.02 up to 10 Of the presumptive Aeromonas μg l-1. The maximum concentration in the strains obtained from the starch treated water was detected during ozonator ampicillin agar (SAA) medium, 95% malfunction. During normal operation, the were verified as Aeromonas, whereas coagulation, clarifi cation, sand fi ltration the remaining 5% were assigned to the and ozonation steps usually reduced the genera Chryseobacterium and Erwinia, microcystin concentrations below the or the family Enterobacteriaceae (Fig. 1, detection limit of 0.02 μg l-1, indicating Appendix 1; II: Table 3). The SAA medium efficient removal of the microcystins had been chosen as a selective growth from water in the fi rst phases of the water medium for Aeromonas spp. isolation. The treatment also during cyanobacterial mass results indicated that the SAA medium was occurrences. selective for the isolation of Aeromonas The endotoxin activities in the treated even from samples as complex as the drinking water were low, 4–38 EU ml-1, cyanobacterial water blooms. based on the kinetic chromogenic Limulus amoebocyte lysate analysis (IV: Table 3). 4.3 Drinking water treatment In contrast, endotoxin concentrations of effi ciency up to 3,300 EU ml-1 were measured in the The efficiency of the drinking water source water. Coagulation, clarification treatment process during cyanobacterial and sand fi ltration reduced the endotoxins mass occurrence in the source water was by 0.72–2.03 log10 units, and ozonation by assessed at an operating water treatment 0.04–0.19 log10 units. The effi ciency of the

28 Results slow sand fi ltration varied from ineffective the treated water was 2.80×104 counting -1 to a maximum reduction of 1.35 log10 units units l and the maximum wet weight and the chlorination step was ineffective 14.13 μg l-1. In contrast, phytoplankton against endotoxins. However, assessment wet weight concentrations of >1.0×103 of the efficiency of the last treatment μg l-1 were detected on several occasions steps was hindered by the low endotoxin from the incoming source water (IV: Fig. concentrations encountered after the 1). This indicated that the water treatment preceding water treatment steps. The process was efficient at removing large results indicated that water treatment, phytoplankton. especially the coagulation-sand fi ltration Picoplankton (size class ≤2 μm3, steps, removed endotoxins effi ciently. including possible bacterial cells) and Signs of neurotoxic agent were monad cells (size class 6–65 μm3) detected by the Artemia salina were detected in the treated drinking bioassay, which was used to assess water 25 to 22 times out of 27 sampling the overall toxicity, in samples taken dates, respectively (IV: Table 2). The from cyanobacterial water blooms that picoplankton cells were present up to occurred at or upwards of the water 1.73×106 counting units l-1 (3.46 μg l-1) and intake of the waterworks (IV: Table 1). the monads up to 2.27×105 counting units The neurotoxin remained unidentified, l-1 (5.12 μg l-1). The results indicate that even though the samples were tested the picoplankton and monad cells could for anatoxin-a, homoanatoxin-a and pass the treatment system quite regularly. their epoxy and dihydro degradation Oscillatoriales spp. filaments (size products by high-performance liquid class 314 μm3) were present in low (up to chromatography (HPLC), anatoxin- 2.80×104 counting units l-1) concentrations a(S) by acetylcholinesterase inhibition in the treated water on 20 occasions out assay, and saxitoxins by Jellett rapid of 27 sampling dates (IV: Table 2). The paralytic shellfi sh poisons test (IV). The fi laments were not detected in the source most dominant cyanobacterial species in water, which implied that they originated seven out of nine samples, and second inside the water treatment facilities. most dominant in one, of the neurotoxic The heterotrophic bacterial counts bloom samples was Anabaena cf. smithii, measured in the treated drinking water implicating it as the most probable origin were mostly low, usually less than 1 CFU of the neurotoxic agent. ml-1, based on the colony counts on R2A plates (IV). The chlorination reduced 4.3.2 Removal of planktonic cells by the the cultivable heterotrophic bacterial water treatment counts by >3 log10 units, coagulation-sand

Large phytoplankton, belonging mainly filtration by 0.6–1.0 log10 units and the 3 to the size class of >65 μm were rarely slow sand fi ltration step by 0.5–1.0 log10 (1–5 times out of 27 sampling dates) units. In contrast the ozonation step was detected in the treated drinking water ineffective. The total range of bacterial based on the microscopic analyses, with removal was 2.0–5.0 log10 units, varying the exception of Oscillatoriales spp. generally between 4.0–5.0 log10 units. The filaments (IV: Table 2). The maximum results indicated effi cient removal of the count of cells, colonies or fi laments (100 cultivable heterotrophic bacteria during μm) of large phytoplankton detected in the water treatment process.

29 Results

4.3.3 Taxonomy of the heterotrophic genera) and Gammaproteobacteria (2 bacteria isolated from drinking water genera) were isolated from the treated The majority (54%) of the 61 bacterial drinking waters (Fig. 4, Appendix 1; I: strains isolated from the treated drinking Supplementary data; IV). In addition, water, belonged to the Alphaproteobacteria two strains could only be assigned at (Fig. 4, Appendix 1; I; IV). Strains of this the family level as Sphingomonadaceae class assigned to the genera Sphingomonas and two others as Oxalobacteraceae. All and Brevundimonas were isolated from the but four of the genera isolated from the treated drinking water of ﬁ ve of the seven treated drinking waters were also isolated drinking water treatment plants studied, from the freshwater samples (Fig. 4, I: regardless of the used treatment system Supplementary data). This implied that the (I, IV). In total, strains from 19 different freshwater bacteria are the most probable genera of the classes Actinobacteria (3 source of the cultivable bacteria passing genera), Flavobacteria (2 genera), Bacilli through water treatment processes. (3 genera), Alpha- (7 genera), Beta- (2

Micrococcus: 1/ES, 1/RA Mycobacterium: 1/ES Pseudomonas: 3/RN, 4/SL, 1/VA Nocardia: 1/KK Chryseobacterium: 1/VA Acinetobacter: 1/SL Flavobacterium: 2/KK Bacillus: 2/ES, 1/RN Oxalobacteraceae: Paenibacillus: 1/ES 1/KK, 1/RN Staphylococcus: 1/SL Herbaspirillum: 1/KK, 2/RN Blastomonas: 2/RN

Duganella: 2/KK Bosea: Sphingomonadaceae: 1/KK, 1/RA, 1/RN 1/RA, 1/RN Sphingopyxis: 1/RA Brevundimonas: 1/HV, 1/KK, 1/RA, 3/SL, 1/VA Sphingomonas: 4/HV, 2/RA, 1/RN, Methylobacterium: 4/ES 2/SL, 2/VA Phyllobacterium: 2/HV, 1/VA

Actinobacteria: Actinobacteria Bacteroidetes: Flavobacteria Firmicutes: Bacilli Proteobacteria: Alphaproteobacteria Betaproteobacteria Gammaproteobacteria

Figure 4: Taxonomic distribution of the heterotrophic bacterial strains (n=61) isolated from treated drinking water. The number of the strains and the treatment plant (ES, HV, KK, RA, RN, SL or VA) they originated from are given together with the taxon. The four genera which in this study were only isolated from treated drinking water are underlined.

30 Discussion

5. DISCUSSION

5.1 Characteristics of the isolated Among the isolated heterotrophic heterotrophic bacteria bacterial strains were more than 100 The heterotrophic bacterial strains isolated haemolytic strains, which represented from natural water or treated drinking 11 different genera, predominantly the water with increased cyanobacterial genus Aeromonas (I). Their taxonomic occurrence in the water bodies or in the assignment and haemolytic properties source water for drinking water treatment implied virulence potential. Furthermore, represented a wide variety of bacterial the majority (90%) of the tested taxa (I, II, IV). They also included several Aeromonas strains contained at least one strains of potential new bacterial taxa from of the six Aeromonas virulence gene types species to order levels. The majority of the screened for in this study, suggesting isolated bacterial strains isolated in this a virulence capacity of the strains (II). study were assigned to the Alpha-, Beta- The most common of the detected gene and Gammaproteobacteria. Bacteria of types was the cytotoxic enterotoxin and these classes have typically been found haemolytic gene products encoding act in freshwater, and in association with (act/aerA/hlyA). The tested strains were cyanobacterial blooms, in studies that isolated from aquatic samples that were used cultivation-independent molecular associated with adverse health effects methods (Zwart et al. 2002, Eiler and in humans, most commonly fever or Bertilsson 2004, Kolmonen et al. 2004). gastrointestinal disturbances, surmised In contrast, bacteria of taxa such as the to be caused by cyanobacterial water Verrucomicrobia that have been detected blooms. However, health effects such as previously by cultivation-independent fever and gastrointestinal disturbances are methods (e.g. Eiler and Bertilsson 2004, also associated with Aeromonas (Janda Kolmonen et al. 2004), were not detected and Abbott 1998, Stewart et al. 2006b). among the strains isolated in this study. The cytotoxic enterotoxin encoded by Cultivation methods can be laborious and the act gene induces an inflammatory the diffi culties in fi nding suitable growth response in host cells (Galindo et al. conditions for any particular species 2006) that can contribute to systemic narrow the bacterial taxons detected using effects such as fever. The presence of these methods (Hugenholtz et al. 1998, act has also been connected to bloody Zinder and Salyers 2001). Nevertheless, diarrhoea, and in a mouse test, a mutant the studies with isolated strains provide Aeromonas with an act deletion induced useful reference points of bacteria with 64% lower fl uid secretion than the wild known characteristics also for the analyses type Aeromonas (Albert et al. 2000, done by the cultivation-independent Sha et al. 2002). The haemolytic toxins molecular methods (Zinder and Salyers encoding aerA and hlyA genes have also 2001). Therefore, both cultivation and been shown to contribute to the virulence cultivation-independent molecular of A. hydrophila in studies using aerA methods are needed to improve our and hlyA inactivated mutants and mice knowledge on the bacterial communities tests (Wong et al. 1998). The common and their roles in the environment. presence of act and the other virulence

31 Discussion genes suggests that the Aeromonas strains V. cholerae of the serogroups O1 and may be the source of the detected human O139 (Faruque et al.1998, Farmer et al. health effects, especially those of fever 2005). In addition, adverse health effects and gastrointestinal disturbances. causing pathogenic strains exist also in In the toxin analysis of the water other Vibrio species and among non- samples from which the Aeromonas strains cholera V. cholerae strains (Thompson et screened for the virulence genes originated al. 2004, Lukinmaa et al. 2006). In a study from, the bacterial endotoxins were by Eiler et al. (2006a) members of this detected in relatively low concentrations genus were regularly detected in the Baltic in the majority of the samples (II). Sea, including also populations containing The cyanobacterial hepatotoxins and mostly V. cholerae. Their study indicated neurotoxins concentrations of the majority that vibrios may potentially be endemic of the samples were below the detection pathogens of the cold and brackish waters limits. Little information about the effects of the Baltic Sea (Eiler et al. 2006a), from of endotoxin exposure by ingestion is which the Vibrio strains of the present available. However, for aerosol exposure study also originated (I). The association a quideline of 10 ng m-3 (~100 EU with cyanobacterial cells has been shown m-3) for no observed effects of airway to prolong cultivable and viable state of inﬂ ammations and 100 ng m-3 (~1000 EU vibrios, and cyanobacterial blooms have m-3) for no observed systemic effects has been proposed to function as a natural been proposed (Anderson et al. 2002 and reservoir for Vibrio (Islam et al. 1999, 2007). The estimated endotoxin aerosol 2002 and 2004). Therefore, the presence concentrations (<1–60 EU m-3) of a of bacteria such as Vibrio further supports majority of the samples is this study were the importance of assessing the health below 100 EU m-3 guideline. Hepatotoxin risks posed by the cyanobacterial blooms concentrations of up to 10 μg l-1 of as a whole. microcystins are considered to present a In contrast to the putative pathogens, relatively low probability of adverse health the 13 previously isolated cyanobacterial effects (Falconer et al. 1999). Furthermore, hepatotoxins degrading Betaproteobacteria in a study by Fawell et al. (1999) using that were further characterised in this study animal models the no observed adverse may reduce the health risks associated effect level dose of orally given anatoxin-a with cyanobacterial water blooms. These was 98 μg kg-1 body weight. Therefore, Betaproteobacteria strains that could the detected toxins by themselves are not be assigned to any described genus unlikely to be the cause of the reported or species previously, were described gastrointestinal disturbances or fever. as a new genus and species Paucibacter In addition to Aeromonas, haemolytic toxinivorans (III). Tested Paucibacter strains of various other genera such as toxinivorans strains degraded nodularin, Vibrio, Pseudomonas and Acinetobacter with a considerably higher degradation rate were isolated (I). These genera are than that of the microcystins. These strains known to contain both human and animal were isolated by Lahti et al. (1998) from pathogens (Farmer et al. 2005, Palleroni sediments of Finnish lakes with frequent 2005, Peleg et al. 2008). For example, cyanobacterial occurrence. The majority Vibrio cholerae is a well known pathogen of the previously isolated bacterial strains for cholera disease, which is caused by that are able to degrade cyanobacterial

32 Discussion hepatotoxins have belonged to the family or inhibit the cyanobacterial growth is Sphingomonadaceae, most commonly to often unknown. In many cases the studied the genus Sphingomonas (e.g. Jones et al. effect has been growth inhibiting, and lytic 1994, Park et al. 2001, Saito et al. 2003, bacteria have been proposed as a short Saitou et al. 2003, Maryama et al. 2006, term means for reducing problems caused Ho et al. 2007). Only few isolated bacterial by cyanobacterial water blooms (Rashidan strains capable of degrading cyanobacterial and Bird 2001). However, long term microcystins have also been shown to be control would require better understanding able to degrade nodularin. These include also on the effects supporting the two Paucibacter toxinivorans strains tested cyanobacterial water bloom formation. previously for nodularin degradation, The strains isolated in this study provide a and an unidentiﬁ ed Sphingomonas strain diverse group of bacteria for studying the isolated by Lahti et al. (1998) as well mechanisms behind cyanobacterial growth as two Sphingomonas strains, the 7CY inhibition and enhancement by individual isolated by Ishii et al. (2004) and the B9 bacterial strains or strain combinations. (Imanishi et al. 2005) isolated by Harada The isolated strains of genera et al. (2004). The Paucibacter toxinivorans such as Streptomyces, Flavobacterium, strains widened the taxonomic distribution Sphingomonas and Acinetobacter also of isolated bacteria known to degrade provide a wide range of putative producers cyanobacterial hepatotoxins, especially of bioactive compounds or degraders nodularin and showed that the ability to of recalcitrant organic compounds (I). degrade the hepatotoxins is not restricted Members of Actinobacteria, most notably to Sphingomonadaceae. bacteria of the genus Streptomyces, are The majority (60%) of the isolated known to produce a wide variety of heterotrophic bacterial strains that bioactive compounds, including over half were tested for their ability to affect of the compounds listed in the Antibiotic cyanobacterial growth either enhanced or Literature Database (Lazzarini et al. inhibited the growth (I). They represented 2000). Members of the Flavobacterium all eight bacterial classes embodied and Sphingomonas genera are known among the strains isolated in this study to degrade various kinds of complex and 30 different genera in total. The fact organic or synthetic compounds (Balkwill that the majority (81%, 29 genera) of the et al. 2006, Bernardet and Bowman strains that affected the cyanobacterial 2006) such as crude oil or the endocrine- growth enhanced the growth indicated disruptive phthalate esters (Baraniecki a mutualistic interaction between these et al. 2002, Rahman et al. 2002, Liang bacteria and the cyanobacteria. Previously et al. 2008). Production of bioactive known cyanobacterial growth affecting compounds and versatile degradation of bacteria have represented relatively few organic compounds could explain the taxa such as Alcaligenes, Streptomyces, effects that many of these strains have on Cytophaga, Pseudomonas and Shewanella cyanobacterial growth and the wide range (Yamamoto et al. 1998, Dakhama et al. of bacteria associated with cyanobacteria. 1993, Manage et al. 2000, Rashidan and However, these potential biosynthetic or Bird 2001, Salomon et al. 2003). The exact degradation abilities were not investigated mechanisms, by which the bacteria enhance in this study.

33 Discussion

5.2 Drinking water treatment among the strains isolated from the treated effi ciency drinking water (I, IV). Bacteria of these The drinking water treatment steps genera are known carotenoid producers of coagulation, clarification and sand (Masetto et al. 2001, Asker et al. 2007). filtration were the most efficient as a The production of pigments such as whole in reducing both hepatotoxin carotenoids and flavonoids may protect and endotoxin concentrations as well the bacteria from the harmful effects of the as heterotrophic bacterial counts (IV). ozonation (LeChevallier and Au 2004a). The results were in accordance with Therefore the pigment production may previous studies by Lepistö et al. (1994) allow bacteria, such as Sphingomonas and and Rapala et al. (2002b). These steps Brevundimonas, pass through the water are used mainly to remove particulate treatment system in a viable state. matter, such as microbial cells and with Picoplankton and monad cells them their endotoxins (WHO 2008b, regularly passed through the water LeChevallier and Au 2004d). As the treatment process (IV). In contrast, majority of the hepatotoxins produced by the larger phytoplankton particles the cyanobacteria remain inside the cells rarely passed through the treatment until the cells die, the coagulation-fi ltration process, implying particle size to have a steps also reduce the hepatotoxins as long determining role in plankton removal. as the cyanobacterial cells remain intact Attachment to particles protects bacteria (Hitzfeld et al. 2000, Svrcek and Smith from disinfection processes such as 2004). The hepatotoxin concentrations may chlorination (LeChevallier and Au be further reduced due to biodegradation 2004a). The oxidation of the cell matter by microbial communities within the sand also consumes ozone (Svrcek and Smith filters as was shown in the study of Ho 2004). Therefore, the presence of the et al. (2006 and 2007). The importance phytoplankton cells may help viable of these combined treatment steps in heterotrophic bacteria to pass through the effective removal of cells and their toxins water treatment process and also might was supported by the results of this study. hinder the destruction of the bacteria The efficiencies of the ozonation, by secondary disinfection during water slow sand fi ltration and chlorination steps distribution. on removal of microcystins, endotoxins The Oscillatoriales fi laments detected and heterotrophic bacteria varied in the drinking water were not detected in depending on the measured agent (IV). the untreated source water, which indicates The most variable results were obtained that they originated from inside the water with the ozonation step, which was treatment facilities (IV). The sand fi lters, effi cient in the removal of microcystins. especially the slow sand filters have In contrast, it only slightly reduced the biological layer as a functional component endotoxin amounts and was inefficient within which photosynthetic organisms against the heterotrophic bacteria. The may function as important carbon inputs ineffi ciency of the ozonation step in the (Campos et al. 2002, LeChevallier and Au removal of heterotrophic bacteria and 2004d, WHO 2008b). Therefore, the most endotoxins may be partly explained probable source of the Oscillatoriales by the prevalence of bacteria such as filaments was the slow sand filter. the Sphingomonas and Brevundimonas Consequently, the Oscillatoriales

34 Discussion concentrations in the treated water might distribution system in a study by Tokajian be reduced by covering the slow sand fi lter et al. (2005). These fi ndings indicate the to prevent photoautotrophic functions. ability of Alphaproteobacteria to survive in The heterotrophic bacterial strains water supply systems. Microbes, including isolated from treated drinking water bacterial genera such as Pseudomonas and of the seven drinking water treatment Mycobacterium, produce biofi lms on the plants with differring water treatment surfaces of water distribution systems, procedures, were dominated by bacteria which can lower the obtained drinking of the Alphaproteobacteria class (I, water quality and cause corrosion of the IV). Genera such as Mycobacterium, distribution system (Percival and Walker Flavobacterium and Pseudomonas, of 1999, Norton and LeChervallier 2000, the classes Actinobacteria, Flavobacteria LeChevallier and Au 2004c, Payment and and Gammaproteobacteria, respectively, Robertson 2004, Berry et al. 2006). They were also present. Bacteria of the can also protect other bacteria, including Alphaproteobacteria class were also pathogenic strains, from removal by found to predominate in the treated disinfectants and water fl ows. Therefore, distribution water in a distribution system detected viable heterotrophic bacteria may simulator study by Williams et al. (2004). cause problems within water distribution In addition, they were one of the major systems. groups of bacteria isolated from a water

35 Conclusions

6. CONCLUSIONS

This study provided information on bacteria and cyanobacteria. The common the heterotrophic bacterial community ability to affect cyanobacterial growth associated with the cyanobacterial water indicates that the heterotrophic bacteria blooms and on the capability of modern may have a role in the formation and drinking water treatment processes to decline of the cyanobacterial water cope with the extra burden caused by such blooms. water blooms. The taxonomic assignment and The removal of phytoplankton, haemolytic features of many of the isolates cyanobacterial hepatotoxins, endotoxins indicated that the blooms harbour a wide and heterotrophic bacteria by the water variety of putative pathogenic bacteria. treatment was for the most part highly The most prominent group of such bacteria efficient. However, the cells passing were the Aeromonas strains, which were through the treatment process may in isolated from both fresh and brackish the long term cause problems such as water samples. The majority of the tested microbial proliferation in biofilms, Aeromonas strains included one or more corrosion of water conducting pipes, a of the screened Aeromonas virulence deterioration in the water distribution gene types, most commonly that of the system and a reduction in the drinking cytotoxic enterotoxin and haemolytic water quality. The water treatment steps products encoding act. The prevalence of varied in their efficiency in reducing the screened virulence genes within the the detected toxins and organisms. This Aeromonas strains further implies that supports the importance of combining they have virulence potential for both multiple treatment steps to ensure overall humans and animals. Therefore, they may good quality of drinking water. pose health risks and economic losses to The heterotrophic bacterial strains the recreational and potable use of water. isolated from treated drinking water The unidentiﬁ ed Betaproteobacteria predominantly represented genera strains previously isolated from lakes that were also isolated from untreated with frequently occurring cyanobacterial freshwater, which implies the freshwater water blooms degraded the cyanobacterial bodies as being the original source. hepatotoxins, microcystins and nodularin. Overall, the isolated bacterial strains Based on their genetic and phenotypic were taxonomically widely distributed, characteristics they were described as which showed that the cyanobacterial a new genus and species Paucibacter water blooms are associated with a diverse toxinivorans. These strains widen community of cultivable heterotrophic the spectrum of known degraders of bacteria. The majority of the tested strains cyanobacterial hepatotoxins, especially either enhanced or inhibited the growth nodularins, and thus give support for the of the cyanobacterial strains Microcystis role of microbial biodegradation in the PCC 7941 and Anabaena PCC 7122. In removal of the chemically stable toxins most cases they enhanced the growth of from the environment. In addition, the cyanobacteria, which implies mutualistic isolated bacterial strains representing interactions between the heterotrophic genera such as Sphingomonas and

36 Conclusions

Flavobacterium may include new the heterotrophic bacterial community degraders of cyanobacterial hepatotoxins associated with them. However, the and other complex organic compounds. cyanobacterial water blooms also seem The heterotrophic bacteria of this to harbour a wide variety of putative study offer a rich source of strains for producers of bioactive compounds or examination of both mutual and biased degraders of recalcitrant compounds. interactions between cyanobacteria and Therefore, both cyanobacteria and their the heterotrophic bacteria associated heterotrophic associates should be taken with them. The putative pathogenicity of into consideration, when studying the many of the strains support the need to growth dynamics and risks as well as widen the health risk assessment of the opportunities posed by the cyanobacterial cyanobacterial water blooms to cover also water blooms.

37 Acknowledgements

7. ACKNOWLEDGEMENTS

This work was carried out at the Finnish Environment Institute and the Department of Applied Chemistry and Microbiology, Division of Microbiology. The Academy of Finland (Microbes and Man (MICMAN) Research Programme, grant 201356 awarded to Jarkko Rapala and the Centres of Excellence Programmes, grants 53305 and 118637 awarded to Kaarina Sivonen), the Finnish Environment Institute and the University of Helsinki are gratefully thanked for their ﬁ nancial support. I would like to express my special gratitude to my supervisors, Docent Jarkko Rapala, Docent Christina Lyra and Academy Professor Kaarina Sivonen for all the time and effort they have invested in this work. Without Jarkko’s ideas and trust in me I probably would never even have began this expedition and without his and Christina’s advice and support I would still be wandering in the wilderness. Kaarina kindly took me under her wing just when it seemed that the journey would be cut short. Her care for the wellbeing of her group members and even strays like me has greatly helped in restoring my faith in human kind. The three of you will always have a special place in my memories. Professors Mirja Salkinoja-Salonen and Kristina Lindström are thanked for their help and for instructions they have given as well as for the inspiring example they have set. Professor Marja-Liisa Hänninen and Docent Stefan Bertilsson are sincerely thanked for reviewing the thesis manuscript and for their most valuable comments on improving it. I heartily thank all the co-workers at the Finnish Environment Institute for their assistance and friendliness. Especially Kaisa Heinonen, Minna Madsen, Liisa Lepistö, Maija Niemelä and Reija Jokipii were invaluable for the implementation of this work. However, all of you, from the old timers like Maarit Niemi, Tuula Ollinkangas and Sinikka Pahkala to summer trainees valuably aided this work, whether in the form of work effort and advice in the laboratory or friendly conversations over a cup of coffee. I would also like to acknowledge the members of the cyano group for their friendly welcome and their generous help. A special thanks goes to Anne Ylinen, Kaisa Rantasärkkä and Hanna Sinkko for their advice and plentiful support they have given since the day I arrived. All the co-authors are warmly thanked for their contributions to the articles. Jarkko, Christina and Kaarina again deserve my special gratitude for the kind schooling they gave me on the many aspects of writing a scientiﬁ c article. I could not have had BETTER mentors. Partners in cooperation are acknowledged for their many contributions in collecting and analysing the studied material. This work would have been quite skeletal without you. I am also deeply grateful to my relatives and friends of all the encouragement they have given over the years. You are the backbone that supports me. My men Mikko and Leo then, there are no words strong enough to tell how much you mean to me. As captivating as this experience has been, without you it would in many ways have been an empty journey through a barren land.

To my valiant princes

Helsinki, November 2009

38 References

8. REFERENCES

Albert, M. J., Ansaruzzaman, M., Talukder, K. A., Chopra, A. K., Kuhn, I., Rahman, M., Faruque, A. S. G., Islam, M. S., Sack, R. B., and Mollby, R. 2000. Prevalence of enterotoxin genes in Aeromonas spp. isolated from children with diarrhea, healthy controls, and the environment. J. Clin. Microbiol. 38: 3785–3790. Allen, S. D., Emery, C. L., and Siders, J. A. 2002. Anaerobic bacteriology, pp. 50–83. In Truant, A. L. (ed.). Manual of commercial methods in clinical microbiology. American Society for Microbiology, Washington D.C., USA. Anderson, M. J. 2003a. DISTLM forward: a FORTRAN computer program to calculate a distance-based multivariate analysis for a linear model using forward selection. Department of Statistics, University of Auckland, New Zealand. http://www.stat.auckland.ac.nz/ ~mja. Anderson, M. J. 2003b. PCO: a FORTRAN computer program for principal coordinate analysis. Department of Statistics, University of Auckland, New Zealand. http://www.stat.auckland. ac.nz/ ~mja. Anderson, M. J. 2004. CAP: a FORTRAN computer program for canonical analysis of principal coordinates. Department of Statistics, University of Auckland, New Zealand. http://www.stat. auckland.ac.nz/ ~mja. Anderson, M. J., and Robinson, J. 2003. Generalized discriminant analysis based on distances. Aus. N. Z. J. Stat. 45: 301–318. Anderson, M. J., and Willis, T. J. 2003. Canonical analysis of principal coordinates: a useful method of constrained ordination for ecology. Ecology 84: 511–525. Anderson, W. B., Slawson, R. M., and Mayfi eld, C. I. 2002. A review of drinking-water- associated endotoxin, including potential routes of human exposure. Can. J. Microbiol. 48: 567–587. Anderson, W. B., Dixon, D. G., and Mayfi eld, C. 2007. Estimation of endotoxin inhalation from shower and humidifi er exposure reveals potential risk to human health. J. Water Health 5: 553–572. Asker, D., Beppu, T., and Ueda, K. 2007. Unique diversity of carotenoid-producing bacteria isolated from Misasa, a radioactive site in Japan. Appl. Microbiol. Biotechnol. 77: 383–392. Awong-Taylor, J., Craven, K. S., Griffi ths, L., Bass, C., and Muscarella, M. 2008. Comparison of biochemical and molecular methods for the identifi cation of bacterial isolates associated with failed loggerhead sea turtle eggs. J. Appl. Microbiol. 104: 1244–1251. Balkwill, D. L., Fredrickson, J. K., and Romine, M. F. 2006. Sphingomonas and related genera, pp. 605–629. In Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., and Stackebrandt, E. (ed.). The prokaryotes, a handbook on the biology of bacteria. Volume 7: Proteobacteria: Delta and epsilon subclasses. Deeply rooting bacteria. 3rd edition. Springer, New York, USA. Baraniecki, C. A., Aislabie, J., and Foght, J. M. 2002. Characterization of Sphingomonas sp. Ant 17, an aromatic hydrocarbon-degrading bacterium isolated from Antarctic soil. Microb. Ecol. 43: 44–54. Bartram, J., Carmichael, W. W., Chorus, I., Jones, G., and Skulberg, O. M. 1999. Introduction, pp. 1–14. In Chorus, I., and Bartram, J. (ed.). Toxic cyanobacteria in water: A guide to their public health consequences, monitoring and management. E & FN Spon, London, UK. Bernardet, J.-F., and Bowman, J. P. 2006. The genus Flavobacterium, pp. 481–531. In Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., and Stackebrandt, E. (ed.). The prokaryotes, a handbook on the biology of bacteria. Volume 7: Proteobacteria: Delta and epsilon subclasses. Deeply rooting bacteria. 3rd edition. Springer, New York, USA. Berry, D., Xi, C., and Raskin, L. 2006. Microbial ecology of drinking water distribution systems. Curr. Opin. Biotechnol. 17: 297–302.

39 References

Best, J. H., Pfl ugmacher, S., Wiegand, C., Eddy, F. B., Metcalf, J. S., and Codd, G. A. 2002. Effects of enteric bacterial and cyanobacterial lipopolysaccharides, and of microcystin-LR, on glutathione S-transferase activities in zebra fi sh (Danio rerio). Aquat. Toxicol. 60: 223–231. Biolog. 2001a. Gram negative identification test panel. http://www.biolog.com/pdf/GN2b_ Brochure.pdf Biolog. 2001b. Gram positive identification test panel. http://www.biolog.com/pdf/GP2b_ Brochure. pdf bioMèrieux. 2009. API. http://www.biomerieux-usa.com/servlet/srt/bio/usa/dynPage?doc =USA_ PRD_LST_G_PRD_USA_5 Borrell, N., Figueras, M. J., and Guarro, J. 1998. Phenotypic identifi cation of Aeromonas genomospecies from clinical and environmental sources. Can. J. Microbiol. 44: 103–108. Bourne, D. G., Jones, G. J., Blakeley, R. L., Jones, A., Negri, A. P., and Riddles, P. 1996. Enzymatic pathway for the bacterial degradation of the cyanobacterial cyclic peptide toxin microcystin LR. Appl. Environ. Microbiol. 62: 4086–4094. Bourne, D. G., Riddles, P., Jones, G. J., Smith, W., and Blakeley, R. L. 2001. Characterisation of a gene cluster involved in bacterial degradation of the cyanobacterial toxin microcystins LR. Environ. Toxicol. 16: 523–534. Bouvy, M., Pagano, M., and Troussellier, M. 2001. Effects of a cyanobacterial bloom (Cylindrospermopsis raciborskii) on bacteria and zooplankton communities in Ingazeira reservoir (northeast Brazil). Aquat. Microb. Ecol. 25: 215–227. Briley, D. S., and Knappe, D. R. U. 2002. Optimizing ferric sulfate coagulation of algae with streaming current measurements. J. Am. Water Works Assoc. 94: 80–90. Bursill, D. 2001. Drinking water treatment - understanding the processes and meeting the challenges. Wat. Sci. Technol. 1: 1–7. Börner, T., and Dittmann, E. 2005. Molecular biology of cyanobacterial toxins. Genetic basis of microcystin production, pp. 25–40. In Huisman, J., Matthijs, H. C. P., and Visser, P. M (ed.). Harmful cyanobacteria. Springer, Dordrecht, the Netherlands. Campos, L. C., Su, M. F. J., Graham, N. J. D., and Smith, S. R. 2002. Biomass development in slow sand fi lters. Water Res. 36: 4543–4551. Cardenas, E., and Tiedje, J. M. 2008. New tools for discovering and characterizing microbial diversity. Curr. Opin. Biotechnol. 19: 544–549. Casamatta, D. A., and Wickstrom, C. E. 2000. Sensitivity of two disjunct bacterioplankton communities to exudates from the cyanobacterium Microcystis aeruginosa Kützing. Microb. Ecol. 41: 64–73. Castenholtz, R. W. 2001. General characteristics of the cyanobacteria, pp. 474–487. In Garrity, G. M., Boone, D. R., and Castenholtz, R. W. (ed.). Bergey’s manual of systematic bacteriology. Volume 1: The Archaea and the deeply branching and phototrophic Bacteria. 2nd edition. Springer, New York, USA. Chacón, M. R., Fiqueras, M. J., Castro-Escarpulli, G., Soler, L., and Guarro, J. 2003. Distribution of virulence genes in clinical and environmental isolates of Aeromonas spp. Anton. Leeuwenh. 84: 269–278. Chen, J.-J., Yeh, H.-H., and Tseng, I.-C. 2009. Effect of ozone and permanganate on algae coagulation removal - Pilot and bench scale tests. Chemosphere 74: 840–846. Chorus, I., Falconer, I. R., Salas, H. J., and Bartram, J. 2000. Health risks caused by freshwater cyanobacteria in recreational waters. J. Toxicol. Environ. Health Part B. 3: 323–347. Chow, C. W. K., House, J., Velzeboer, R. M. A., Drikas, M., Burch, M. D., and Steffensen, D. A. 1998. The effect of ferric chloride fl occulation on cyanobacterial cells. Water Res. 32: 808–814.

40 References

Chu, F. S. 2000. Freshwater hepatotoxins: Chemistry and detection, pp. 613–642. In Botana, L.M. (ed). Seafood and freshwater toxins: pharmacology, physiology, and detection. Marcel Dekker Inc., New York, USA. Clarridge, J. E. 2004. Impact of 16S rRNA gene sequence analysis for identifi cation of bacteria on clinical microbiology and infectious diseases. Clin. Microbiol. Rev. 17: 840–862. Cook, D., and Newcombe, G. 2002. Removal of microcystin variants with powdered activated carbon. Water Sci. Technol. 2: 201–207. Cox, P. A., Banack, S. A., Murch, S. J., Rasmussen, U., Tien, G., Bidigare, R. R., Metcalf, J. S., Morrison, L. F., Codd, G. A., and Bergman, B. 2005. Diverse taxa of cyanobacteria produce β-N-methylamino-L-alanine, a neurotoxic amino acid. Proc. Nat. Acad. Sci. USA. 102: 5074–5078. Craig, M., and Holmes, C. F. B. 2000. Freshwater hepatotoxins: Microcystin and nodularin, mechanisms of toxicity and effects on health, pp. 643–671. In Botana, L.M. (ed). Seafood and freshwater toxins: pharmacology, physiology, and detection. Marcel Dekker Inc., New York, USA. Dakhama, A., de la Noüe, J., and Lavoie, M. C. 1993. Isolation and identifi cation of antialgal substances produced by Pseudomonas aeruginosa. J. Appl. Phycol. 5: 297–306. deFigueiredo, D. R., Azeiteiro, U. M., Esteves, S. M., Gonçalves, F. J. M., and Pereira, M. J. 2004. Microcystin-producing blooms—a serious global public health issue. Ecotoxicol. Environ. Safety 59: 151–163. Dismukes, G. C., Klimov, V. V., Baranov, S. V., Kozlov, Y. N., DasGupta, J., and Tyryshkin, A. 2001. The origin of atmospheric oxygen on Earth: The innovation of oxygenic photosynthesis. Proc. Nat. Acad. Sci. USA. 98: 2170–2175. Drikas, M., Chow, C. W. K., House, J., and Burch, M. D. 2001. Using coagulation, fl occulation, and settling to remove toxic cyanobacteria. J. Am. Water Works Assoc. 93: 100–111. Eiler, A., and Bertilsson, S. 2004. Composition of freshwater bacterial communities associated with cyanobacterial blooms in four Swedish lakes. Environ. Microbiol. 6: 1228–1243. Eiler, A., and Bertilsson, S. 2007. Flavobacteria blooms in four eutrophic lakes: linking population dynamics of freshwater bacterioplankton to resource availability. Appl. Environ. Microbiol. 73: 3511–3518. Eiler, A., Johansson, M., and Bertilsson, S. 2006a. Environmental influences on Vibrio populations in northern temperate and boreal coastal waters (Baltic and Skagerrak seas). Appl. Environ. Microbiol. 72: 6004–6011. Eiler, A., Olsson, J. A., and Bertilsson, S. 2006b. Diurnal variations in the auto- and heterotrophic activity of cyanobacterial phycospheres (Gloeotrichia echinulata) and the identity of attached bacteria. Freshwat. Biol. 51: 298–311. Eiler, A., Gonzalez-Rey, C., Allen, S., and Bertilsson, S. 2007. Growth response of Vibrio cholerae and other Vibrio spp. to cyanobacterial dissolved organic matter and temperature in brackish water. FEMS Microbiol. Ecol. 60: 411–418. Ellman, G. L., Courtney, K. D., Andres, V., and Featherstone, R. M. 1961. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7: 88–95. Falconer, I. R. 2008. Health effects associated with controlled exposures to cyanobacterial toxins, pp. 607–612. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Falconer, I., Bartram, J., Chorus, I., Kuiper-Goodman, T., Utkilen, H., Burch, M., and Codd, G. A. 1999. Safe levels and safe practices, pp. 155–178. In Chorus, I., and Bartram, J. (ed.). Toxic cyanobacteria in water: A Guide to their public health consequences, monitoring and management. E & FN Spon, London, UK.

41 References

Falconer, I. R., and Humpage, A. R. 2005. Health risk assessment of cyanobacterial (blue-green algal) toxins in drinking water. Int. J. Environ. Res. Public Health 2: 43–50. Farmer III, J. J., Janda, J. M., Brenner, F. W., Cameron, D. N., and Birkhead, K. M. 2005. Genus I. Vibrio, pp. 494–546. In Garrity, G. M., Brenner, D. J., Krieg, N. R., and Staley, J. T (ed.). Bergey’s manual of systematic bacteriology. Volume 2: The Proteobacteria. 2nd edition. Springer, New York, USA. Farmer III, J. J, Arduino, M. J., and Hickman-Brenner, F. W. 2006. The Genera Aeromonas and Plesiomonas, pp. 564–596. In Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., and Stackebrandt, E. (ed.). The prokaryotes, a handbook on the biology of bacteria. Volume 6: Gamma subclass. 3rd edition. Springer, New York, USA. Faruque, S. M., Albert, M. J., Mekalanos, J. J. 1998. Epidemiology, genetics, and ecology of toxigenic Vibrio cholera. Microbiol. Mol. Biol. Rev. 62: 1301–1314. Fawell, J. K., Mitchell, R. E., Hill, R. E., and Everett, D. J. 1999. The toxicity of cyanobacterial toxins in the mouse: II Anatoxin-a. Hum. Exp. Toxicol. 18: 168–173. Felsenstein, J. 2004a. Bootstrap, jacknife, and permutation tests, pp. 335–363. In Felsenstein, J. (aut.). Inferring phylogenies. Sinauer associates Inc., Massachusetts, USA. Felsenstein, J. 2004b. Models of DNA evolution, pp. 196–221. In Felsenstein, J. (aut.). Inferring phylogenies. Sinauer associates Inc., Massachusetts, USA. Felsenstein, J. 2005. PHYLIP (Phylogeny Inference Package) version 3.6. Distributed by the author. Department of Genome Sciences, University of Washington, Seattle, USA. Figueras, M. J. 2005. Clinical relevance of Aeromonas sM503. Rev. Med. Microbiol. 16: 145– 153. Fournie, J. W., Hilborn, E. D., Codd, G. A., Coveney, M., Dyble, J., Havens, K., Ibelings, B. W., Landsberg, J., and Litaker, W. 2008. Ecosystem effects workgroup report, pp. 655–674. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Galindo, C. L., Sha, J., Fadl, A. A., Pillai, L. L., and Chopra, A. K. 2006. Host immune responses to Aeromonas virulence factors. Curr. Immunol. Rev. 2: 13–26. Giovannoni, S. J., and Stingl, U. 2005. Molecular diversity and ecology of microbial plankton. Nature 437: 343–348. Glöckner, F. O., Zaichikov, E., Belkova, N., Denissova, L., Pernthaler, J., Pernthaler, A., and Amann, R. 2000. Comparative 16S rRNA analysis of lake bacterioplankton reveals globally distributed phylogenetic clusters including an abundant group of Actinobacteria. Appl. Environ. Microbiol. 66: 5053–5065. Hagström, Å., Pommier, T., Rohwer, F., Simu, K., Stolte, W., Svensson, D., and Zweifel, U. L. 2002. Use of 16S ribosomal DNA for delineation of marine bacterioplankton species. Appl. Environ. Microbiol. 68: 3628–3633. Harada, K.-i., Imanishi, S., Kato, H., Mizuno, M., Ito, E., and Tsuji, K. 2004. Isolation of Adda from microcystin-LR by microbial degradation. Toxicon 44: 107–109. Havens, K. E. 2008. Cyanobacterial blooms: effects on aquatic ecosystems, pp. 733–747. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Hitzfeld, B. C., Höger, S. J., and Dietrich, D. R. 2000. Cyanobacterial toxins: removal during drinking water treatment, and human health assessment. Environ. Health Perspect. 108: 113– 122. Ho, L., Meyn, T., Keegan, A., Hoefel, D., Brookes, J., Saint, C. P., and Newcombe, G. 2006. Bacterial degradation of microcystin toxins within a biologically active sand fi lter. Water Res. 40: 768–774. Ho, L., Hoefer, D., Saint, C. P., and Newcombe, G. 2007. Isolation and identifi cation of a novel microcystin-degrading bacterium from a biological sand fi lter. Water Res. 41: 4685–4695.

42 References

Hoeger, S. J., Hitzfeldt, B. C., and Dietrich, D. R. 2005. Occurrence and elimination of cyanobacterial toxins in drinking water treatment plants. Toxicol. Appl. Pharmacol. 203: 231– 242. Holst, T., Jørgensen, N. O. G., Jørgensen, C., and Johansen, A. 2003. Degradation of microcystins in sediments at oxic and anoxic, denitrifying conditions. Water Res. 37: 4748– 4760. Hoppu, K., Salmela, J., Lahti, K. 2002. High risk for symptoms from use of water contaminated with cyanobacteriae in sauna (Abstract). J. Toxicol - Clin. Toxicol. 40: 309–310. Hrudey, S., Burch, M., Drikas, M., and Gregory, R. 1999. Remedial measures, pp. 275–312. In Chorus, I., and Bartram, J. (ed.). Toxic cyanobacteria in water: A guide to their public health consequences, monitoring and management. E & FN Spon, London, UK. Hugenholtz, P., Goebel, B. M., and Pace, N. R. 1998. Impact of culture-independent studies on the emerging phylogenetic view of bacterial diversity. J. Bacteriol. 180: 4765–4774. Huisman, J., and Hulot, F. D. 2005. Population dynamics of harmful cyanobacteria, factors affecting species composition, pp. 143–176. In Huisman, J., Matthijs, H. C. P., and Visser, P. M. (ed.). Harmful cyanobacteria. Springer, Dordrecht, the Netherlands. Humpage, A. 2008. Toxin types, toxicokinetics and toxicodynamics, pp. 383–415. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Hunter, P. R., Payment, P., Ashbolt, N., and Bartram, J. 2003. Assessment of risk, pp. 79–109. In Dufour, A., Snozzi, M., Koster, W., Bartram, J., Ronchi, E., and Fewtrell, L. (ed.). Assessing microbial safety of drinking water, improving approaches and methods. IWA Publishing, London, UK. Hänninen, M.-L., and Siitonen, A. 1995. Distribution of Aeromonas phenospecies and genospecies among strains isolated from water, foods or from human clinical samples. Epidemiol. Infect. 115: 39–50. Imanishi, S., Kato, H., Mizuno, M., Tsuji, K., and Harada, K.-i. 2005. Bacterial degradation of microcystins and nodularin. Chem. Res. Toxicol. 18: 591–598. Ishii, H., Nishijima, M., and Abe, T. 2004. Characterization of degradation process of cyanobacterial hepatotoxins by a gram-negative aerobic bacterium. Water Res. 38: 2667– 2676. Islam, M. S., Rahim, Z., Alam, M. J., Begum, S., Moniruzzaman, S. M., Umeda, A., Amako, K., Albert, M. J., Sack, R. B., Huq, A., and Colwell, R. R. 1999. Association of Vibrio cholerae O1 with the cyanobacterium, Anabaena sp., elucidated by polymerase chain reaction and transmission electron microscopy. Trans. Roy. Soc. Trop. Med. Hyg. 93: 36–40. Islam, M. S., Goldar, M. M., Morshed, M. G., Khan, M. N. H., Islam, M. R., and Sack, R. B. 2002. Involvement of the hap gene (mucinase) in the survival of Vibrio cholerae O1 in association with the blue-green alga, Anabaena sp.. Can. J. Microbiol. 48: 793–800. Islam, M. S., Mahmuda, S., Morshed, M. G., Bakht, H. B. M., Khan, M. N. H., Sack, R. B., and Sack, D. A. 2004. Role of cyanobacteria in the persistence of Vibrio cholerae O139 in saline microcosms. Can. J. Microbiol. 50: 127–131. James, K. J., Sherlock, I. R., and Stack, M. A. 1997. Anatoxin-a in Irish freshwater and cyanobacteria, determined using a new ﬂ uorimetric liquid chromatographic method. Toxicon 35: 963–971. Janda, J. M. 1991. Recent advances in the study of the taxonomy, pathogenicity, and infectious syndromes associated with the genus Aeromonas. Clin. Microbiol. Rev. 4: 397–410. Janda, J. M., and Abbott, S. L. 1998. Evolving concepts regarding the genus Aeromonas: An expanding panorama of species, disease presentations, and unanswered questions. Clinic. Infect. Dis. 27: 332–344.

43 References

Jellett, J. F., Roberts, R. L., Laycock, M. V., Quilliam, M. A., and Barrett, R. E. 2002. Detection of paralytic shellfi sh poisoning (PSP) toxins in shellfi sh tissue using MIST Alert, a new rapid test, in parallel with the regulatory AOAC mouse bioassay. Toxicon 40: 1407– 1425. Jones, G. J., Bourne, D. G., Blakeley, R. L., and Doelle, H. 1994. Degradation of the cyanobacterial hepatotoxin microcystins by aquatic bacteria. Nat. Toxins 2: 228–235. Jones, G. J., Falconer, I. R., and Wilkins, R. M. 1995. Persistence of cyclic peptide toxins in dried Microcystis aeruginosa crusts from Lake Mokoan, Australia. Environ. Toxicol. Water Qual. 10: 19–24. Kaakkois-Suomen ympäristökeskus. 2009. Poikkeuksellinen tilanne Pien-Saimaalla – sinilevä edelleen voimissaan jään alla. http://www.ymparisto.fi /default.asp?contentid=310778&lan=fi Kangatharalingam, N., Wang, L., and Priscu, J. 1991. Evidence for bacterial chemotaxis to cyanobacteria from a radioassay technique. Appl. Environ. Microbiol. 57: 2395–2398. Kaper, J., Seidler, R. J., Lockman, H., and Colwell, R. R. 1979. Medium for the presumptive identifi cation of Aeromonas hydrophila and Enterobacteriaceae. Appl. Environ. Microbiol. 28: 1023–1026. Kardinaal, W. E. A., and Visser, P. M. 2005. Dynamics of cyanobacterial toxins, Sources of variability in microcystin concentrations, pp. 41–63. In Huisman, J., Matthijs, H. C. P., and Visser, P. M (ed.), Harmful cyanobacteria. Springer, Dordrecht, the Netherlands. Karner, D. A., Standridge, J. H., Harrington, G. W., and Bartnum, R. P. 2001. Microcystin algal toxins: In source and fi nished drinking water. J. Am. Water Works Assoc. 93: 72–81. Keto, J., and Tallberg, P. 2000. The recovery of Vesijärvi, a lake in southern Finland: water quality and phytoplankton interpretations. Bor. Environ. Res. 5: 15–26. Kingombe, C. I. B, Huys, G., Tonolla, M., Albert, M. J., Swings, J., Peduzzi, R., and Jemmi, T. 1999. PCR detection, characterization, and distribution of virulence genes in Aeromonas spp. Appl. Environ. Microbiol. 65: 5293–5302. Kirkwood, A. E., Nalewajko, C., and Fulthorpe, R. R. 2006. The effects of cyanobacterial exudates on bacterial growth and biodegradation of organic contaminants. Microb. Ecol. 51: 4–12. Knoll, A. H. 2008. Cyanobacteria and Earth history, pp. 1–19. In Herrero, A., and Flores, E. (ed.), The cyanobacteria, molecular biology, genomics and evolution. Caister Academic Press, Norfolk, UK. Kolmonen, E., Sivonen, K., Rapala, J., and Haukka, K. 2004. Diversity of cyanobacteria and heterotrophic bacteria in cyanobacterial blooms in Lake Joutikas, Finland. Aquat. Microb. Ecol. 36: 201–211. Kuiper-Goodman, T., Falconer, I., and Fitzgerald, J. 1999. Human health aspects, pp. 113– 153. In Chorus, I., and Bartram, J. (ed.), Toxic cyanobacteria in water: A guide to their public health consequences, monitoring and management. E & FN Spon, London, UK. Lahti, K., Ahtiainen, J., Rapala, J., Sivonen, K., and Niemelä, S. I. 1995. Assessment of rapid bioassays for detecting cyanobacterial toxicity. Lett. Appl. Microbiol. 21: 109–114. Lahti, K., Niemi, M. N., Rapala, J., and Sivonen, K. 1998. Biodegradation of cyanobacterial hepatotoxins- Characterization of toxin degrading bacteria, pp. 363–365. In Reguera, B., Blanco, J., Fernández, M. L., and Wyatt, T. (ed.), Harmful algae, proceedings of the VIII international conference of harmful algae. Lahti, K., Rapala, J., Kivimäki, A.-L., Kukkonen, J., Niemelä, M., and Sivonen, K. 2001. Occurrence of microcystins in raw water sources and treated drinking water of Finnish waterworks. Wat. Sci. Technol. 43: 225–228. Lambert, T. W., Holmes, C. F. B., and Hrudey, S. E. 1996. Adsorbtion of microcystin-LR by activated carbon and removal in full scale water treatment. Water Res. 30: 1411–1422.

44 References

Lamy, B., Kodjo, A., colBVH Study group, and Laurent, F. 2009. Prospective nationwide study of Aeromonas infections in France. J. Clin. Microbiol. 47: 1234–1237. Lawton, L. A., and Edwards, C. 2008. Conventional laboratory methods for cyanotoxins, pp. 513–537. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Lawton, L. A., and Robertson, P. K. J. 1999. Physico-chemical treatment methods for the removal of microcystins (cyanobacterial hepatotoxins) from potable waters. Chem. Soc. Rev. 28: 217–224. Lazzarini, A., Cavaletti, L., Toppo, G., and Marinelli, F. 2000. Rare genera of actinomycetes as potential producers of new antibiotics. Anton Leeuwenh. 78: 399–405. LeChevallier, M. W., and Au, K.-K. 2004a. Inactivation (disinfection) processes, pp. 41–65. In LeChevallier, M.W., and Au, K.-K. (aut.). Water treatment and pathogen control, process effi ciency in achieving safe drinking water. IWA Publishing, London, UK. LeChevallier, M. W., and Au, K.-K. 2004b. Introduction, pp. 1–4. In LeChevallier, M.W., and Au, K.-K. (aut.). Water treatment and pathogen control, process effi ciency in achieving safe drinking water. IWA Publishing, London, UK. LeChevallier, M. W., and Au, K.-K. 2004c. Process control, pp. 81–92. In LeChevallier, M.W., and Au, K.-K. (aut.). Water treatment and pathogen control, process effi ciency in achieving safe drinking water. IWA Publishing, London, UK. LeChevallier, M. W., and Au, K.-K. 2004d. Removal processes, pp. 5–40. In LeChevallier, M.W., and Au, K.-K. (aut.). Water treatment and pathogen control, process effi ciency in achieving safe drinking water. IWA Publishing, London, UK. Leclerc, H., Schwartzbrod, L., and Dei-Cas, E. 2002. Microbial agents associated with waterborne diseases. Crit. Rev. Microbiol. 28: 371–409. Legendre, P., and Legendre, L. 1998a. Canonical analysis, pp. 575–635. In Legendre, P., and Legendre, L. (aut.). Numerical ecology. 2nd edition. Elsevier Science, Amsterdam, the Netherlands. Legendre, P., and Legendre, L. 1998b. Interpretation of ecological structures, pp. 481–574. In Legendre, P., and Legendre, L. (aut.). Numerical ecology. 2nd edition. Elsevier Science, Amsterdam, the Netherlands. Lepistö, L., Lahti, K., Niemi, J., and Färdig, M. 1994. Removal of cyanobacteria and other phytoplankton in four Finnish waterworks. Algol. Stud. 75: 167–181. Li, T., Yan, X., Wang, D., and Wang, F. 2009. Impact of preozonation on the performance of coagulated fl ocs. Chemosphere 75: 187–192. Liang, D.-W., Zhang, T., Fang, H. H. P., and He, J. 2008. Phthalates biodegradation in the environment. Appl. Microbiol. Biotechnol. 80: 183–198. Logsdon, G. S., Kohne, R., Abel, S., and LaBonde, S. 2002. Slow sand fi ltration for small water systems. J. Environ. Eng. Sci. 1: 339–348. Lukinmaa, S., Mattila, K., Lehtinen, V., Häkkinen, M., Koskela, M., and Siitonen, A. 2006. Territorial waters of the Baltic Sea as a source of infections caused by Vibrio cholerae non-O1, non-O139: report of 3 hospitalized cases. Diagn. Microbiol. Infect. Dis. 54: 1–6. Manage, P. M., Kawabata, Z., and Nakano, S.-i. 2000. Algicidal effect of the bacterium Alcaligenes denitrifi cans on Microcystis spp. Aquat. Microb. Ecol. 22: 111–117. Martin-Carnahan, A., and Joseph, S. W. 2005. Genus I. Aeromonas, pp. 556–578. In Garrity, G. M., Brenner, D. J., Krieg, N. R., and Staley, J. T (ed.). Bergey’s manual of systematic bacteriology. Volume 2: The Proteobacteria. 2nd edition. Springer, New York, USA. Maruyama, T., Kato, K., Yokoyama, A., Tanaka, T., Hiraishi, A., and Park, H.-D. 2003. Dynamics of microcystin-degrading bacteria in mucilage of Microcystis. Microb. Ecol. 46: 279–288.

45 References

Maruyama, T., Park, H.-D., Ozawa, K., Tanaka, Y., Sumino, T., Hamana, K., Hiraishi, A., and Kato, K. 2006. Sphingosinicella microcystinivorans gen. nov., sp. nov., a microcystins- degrading bacterium. Int. J. Syst. Evol. Microbiol. 56: 85–89. Masetto, A., Flores-Cotera, L. B., Diaz, C., Langley, E., and Sanchez, S. 2001. Application of a complete factorial design for the production of zeaxanthin by Flavobacterium sp.. J. Biosci. Bioeng. 92: 55–58. Massa, S., Caruso, M., Trovatelli, F., and Tosques, M. 1998. Comparison of plate count agar and R2A medium for enumeration of heterotrophic bacteria in natural mineral water. World. J. Microbiol. Biotechnol. 14: 727–730. McArdle, B. H., and Anderson, M. J. 2001. Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology 82: 290–297. McElhiney, J., and Lawton, L. A. 2005. Detection of the cyanobacterial hepatotoxins microcystins. Toxicol. Appl. Pharmacol. 203: 219–230. Metcalf, J. S., and Codd, G. A. 2005a. Anatoxin-a, pp. 41–42. In Meriluoto, J., and Codd, G. A. (ed.). Toxic: Cyanobacterial monitoring and cyanotoxin analysis. Åbo Akademi University Press, Åbo, Finland. Metcalf, J. S., and Codd, G. A. 2005b. Analysis of other cyanobacterial toxins, pp. 47–48. In Meriluoto, J., and Codd, G. A. (ed.). Toxic: Cyanobacterial monitoring and cyanotoxin analysis. Åbo Akademi University Press, Åbo, Finland. Metcalf, J. S., and Codd, G. A. 2005c. Cylindrospermopsin, pp. 43–45. In Meriluoto, J., and Codd, G. A. (ed.). Toxic: Cyanobacterial monitoring and cyanotoxin analysis. Åbo Akademi University Press, Åbo, Finland. Metcalf, J. S., Lindsay, J., Beattie, K. A., Birmingham, S., Saker, M. L., Törökńe, A. K., and Codd, G. A. 2002. Toxicity of cylindrospermopsin to the brine shrimp Artemia salina: comparison with protein synthesis inhibitors and microcystins. Toxicon 40: 1115–1120. Mouget, J.-L., Dakhama, A., Lavoie, M. C., and de la Noüe, J. 1995. Algal growth enhancement by bacteria: Is consumption of photosynthetic oxygen involved? FEMS Microbiol. Ecol. 18: 35–44. Mur, L. R., Skulberg, O. M., and Utkilen, H. 1999. Cyanobacteria in the environment, pp. 1540. In Chorus, I., and Bartram, J. (ed.). Toxic cyanobacteria in water: A guide to their public health consequences, monitoring and management. E & FN Spon, London, UK. Murch, S. J., Cox, P. A., and Banack, S. A. 2004. A mechanism for slow release of biomagnifi ed cyanobacterial neurotoxins and neurodegenerative disease in Guam. Proc. Nat. Acad. Sci. USA. 101: 12228–12231. Nei, M., and Kumar, S. 2000a. Accuracies and statistical tests of phylogenetic trees, pp. 165–186. In Nei, M., and Kumar, S. (aut.). Molecular evolution and phylogenetics. Oxford University Press, New York, USA. Nei, M., and Kumar, S. 2000b. Phylogenetic inference: distance method, pp. 87–113. In Nei, M., and Kumar, S. (aut.). Molecular evolution and phylogenetics. Oxford University Press, New York, USA. Nei, M., and Kumar, S. 2000c. Phylogenetic inference: maximum likelihood methods, pp. 147–164. In Nei, M., and Kumar, S. (aut.). Molecular evolution and phylogenetics. Oxford University Press, New York, USA. Nei, M., and Kumar, S. 2000d. Phylogenetic inference: maximum parsimony method, pp. 115–145. In Nei, M., and Kumar, S. (aut.). Molecular evolution and phylogenetics. Oxford University Press, New York, USA. Niemi, M., Vepsäläinen, M., Simpanen, S., Erkomaa, K., and Münster, U. 2004. ZymProfi ler® testisarjan kehittäminen ja soveltaminen ympäristönäytteiden entsyymiaktiivisuuksien mittaamiseen. Suomen Ympäristökeskus, Helsinki, Finland. ISBN 952-11-1650-1. www. ymparisto. fi /julkaisut.

46 References

Norton, C., and LeChevallier, M. W. 2000. A pilot study of bacteriological population changes through potable water treatment and distribution. Appl. Environ. Microbiol. 66: 268–276. Oksanen, I., Jokela, J., Fewer, D. P., Wahlsten, M., Rikkinen, J., and Sivonen, K. 2004. Discovery of rare and highly toxic microcystins from lichen-associated cyanobacterium Nostoc sp. strain IO-102-I. Appl. Environ. Microbiol. 70: 5756–5763. Oliver, R. L., and Ganf, G. G. 2000. Freshwater blooms, pp. 149194. In Whitton, B. A. and Potts, M. (ed.). The ecology of cyanobacteria: Their diversity in time and space. Kluwer Academic Publishers, Dordrecht, the Netherlands. Olson, J. M. 2006. Photosynthesis in the Archean Era. Photosynth. Res. 88: 109–117. Oren, A. 2000. Salts and brines, pp. 281–306. In Whitton, B. A., and Potts, M. (ed.). The ecology of cyanobacteria: Their diversity in time and space. Kluwer Academic Publishers, Dordrecht, the Netherlands. Paerl, H. W. 1976. Speciﬁ c association of the blue green algae Anabaena and Aphanizomenon with bacteria in freshwater blooms. J. Phycol. 12: 431–435. Paerl, H. W. 1996. Microscale physiological and ecological studies of aquatic cyanobacteria: marcosacle implications. Microsc. Res. Tech. 33: 47–72. Paerl, H. W. 2000. Marine plankton, pp. 121148. In Whitton, B. A., and Potts, M. (ed.). The ecology of cyanobacteria: Their diversity in time and space. Kluwer Academic Publishers, Dordrecht, the Netherlands. Paerl, H. W. 2008. Nutrient and other environmental controls of harmful cyanobacterial blooms along the freshwater-marine continuum, pp. 217–237. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Paerl, H. W., and Kellar, P. E. 1978. Signiﬁ cance of bacterial-Anabaena (Cyanophyceae)

association with respect to N2 fi xation in freshwater. J. Phycol. 14: 254–260. Paerl, H. W., and Pinckney, J. L. 1996. A mini-review of microbial consortia: Their roles in aquatic production and biochemical cycling. Microb. Ecol. 31: 225–247. Palleroni, N. J. 1997. Prokaryotic diversity and importance of culturing. Anton. Leeuwenh. 72: 3–19. Palleroni, N. J. 2005. Pseudomonas, pp. 323–379. In Brenner, D. J., Krieg, N. R., Staley, J. T., and Garrity, G. M. (ed.). Bergey’s manual of systematic bacteriology. Volume 2: The Proteobacteria. 2nd edition. Springer, New York, USA. Park, H.-D., Sasaki, Y., Maruyama, T., Yanagisawa, E., Hiraishi, A., and Kato, K. 2001. Degradation of the cyanobacterial hepatotoxin microcystin by a new bacterium isolated from hypertrofi c lake. Environ. Toxicol. 16: 337–343. Paul, V. J. 2008. Global warming and cyanobacterial harmful algal blooms, pp. 239–257. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Payment, P., and Robertson, W. 2004 The microbiology of piped distribution systems and public health, pp. 1–18. In Ainsworth, R. (ed.). Safe piped water: managing microbial water quality in piped distribution systems. IWA Publishing, London, UK. Peleg, A. Y., Seifert, H., and Peterson, D. L. 2008. Acinetobacter baumannii: emergence of a successful pathogen. Clin. Microbiol. Rev. 21: 538–582. Percival, S. L., and Walker, J. T. 1999. Potable water and biofi lms: a rewiev of the public health implications. Biofouling 14: 99–115. Peterson, H. G., Hrudey, S. E., Cantin, I. A., Perley, T. R., and Kenefick, S. L. 1995. Physiological toxicity, cell membrane damage and the release of dissolved organic carbon and geosmin by Aphanizomenon fl os-aquae after exposure to water treatment chemicals. Water Res. 29: 1515–1523.

47 References

Pilotto, L. S. 2008. Epidemiology of cyanobacteria and their toxins, pp. 639–649. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Rahman, K. S. M., Thahira-Rahman, J., Lakshmanaperumalsamy, P., and Banat, I. M. 2002. Towards effi cient crude oil degradation by a mixed bacterial consortium. Bioresource Technol. 85: 257–261. Rapala, J., Lahti, K., Sivonen, K., and Niemelä, S. I. 1994. Biodegradability and adsorption on lake sediments of cyanobacterial hepatotoxins and anatoxin-a. Lett. Appl. Microbiol. 19: 423–428. Rapala, J., Erkomaa, K., Kukkonen, J., Sivonen, K., and Lahti, K. 2002a. Detection of microcystins with protein phosphatase inhibition assay, high-performance liquid chromatography–UV detection and enzyme-linked immunosorbent assay: Comparison of methods. Anal. Chim. Acta 466: 213–231. Rapala, J., Lahti, K., Räsänen, L. A., Esala, A.-L., Niemelä, S. I., and Sivonen, K. 2002b. Endotoxins associated with cyanobacteria and their removal during drinking water treatment. Water Res. 36: 2627–2635. Rashidan, K. K., and Bird, D. F. 2001. Role of predatory bacteria in the termination of a cyanobacterial bloom. Microb. Ecol. 41: 97–105. Raven, J. A., and Allen, J. F. 2003. Genomics and chloroplast evolution: what did cyanobacteria do for plants? Gen. Biol. 4: 209. Robertson, P. K. J., Lawton, L. A., and Cornish, B. J. P. A. 1999. The involvement of phycocyanin pigment in the photodecomposition of the cyanobacterial toxin, microcystin-LR. J. Porphyr. Phthalocya. 3: 544–551. Roselló-Mora, R., and Amann, R. 2001. The species concept for prokaryotes. FEMS Micobiol. Rev. 25: 39–67. Rositano, J., Newcombe, G., Nicholson, B., and Sztajnbok, P. 2001. Ozonation of NOM and algal toxins in four treated waters. Water Res. 35: 23–32. Rusch, D. B., Halpern, A. L., Sutton, G., Heidelberg, K. B., Williamson, S., Yooseph, S., Wu, D., Eisen, J. A., Hoffman, J. M., Remington, K., Beeson, K., Tran, B., Smith, H., Baden-Tillson, H., Stewart, C., Thorpe, J., Freeman, J., Andrews-Pfannkoch, C., Venter, J. E., Li, K., Kravitz, S., Heidelberg, J. F., Utterback, T., Rogers, Y.-H., Falcón, L. I., Souza, V., Bonilla-Rosso, G., Eguiarte, L. E., Karl, D. M., Sathyendranath, S., Platt, T., Bermingham, E., Gallardo, V., Tamayo-Castillo, G., Ferrari, M. R., Strausberg, R. L., Nealson, K., Friedman, R., Frazier, M., and Venter, J. C. 2007. The Sorcerer II Global Ocean Sampling expedition: northwest Atlantic through eastern tropical Pacifi c. PLoS Biol. 5: e77. Rusin, P. A., Rose, J. B., Haas, C. N., and Gerba, C. P. 1997. Risk assessment of opportunistic bacterial pathogens in drinking water. Rev. Environ. Contam. Toxicol. 152: 57–83. Saito, T., Okano, K., Park, H.-D., Itayama, T., Inamori, Y., Neilan, B. A., Burns, B. P., and Sugiura, N. 2003. Detection and sequencing of the microcystin LR-degrading gene, mlrA, from new bacteria isolated from Japanese lakes. FEMS Microbiol. Lett. 229: 271–276. Saitou, T., Sugiura, N., Itayama, T., Inamori, Y., and Matsumura, M. 2003. Degradation characteristics of microcystins by isolated bacteria from Lake Kasumigaura. J. Water Supply Res. Technol. Aqua 52: 13–18. Salmela J, Lahti K., and Hoppu K. 2001. Sinileväpitoinen saunavesi voi aiheuttaa oireita ihmisille. Suomen Lääkärilehti 56: 2891–2895. Salomon, P. S., Janson, S., and Granéli, E. 2003. Molecular identifi cation of bacteria associated with fi laments of Nodularia spumigena and their effect on the cyanobacterial growth. Harm. Alg. 2: 261–272.

48 References

Sen, K., and Rodgers, M. 2004. Distribution of six virulence factors in Aeromonas species isolated from US drinking water utilities: a PCR identifi cation. J. Appl. Microbiol. 97: 1077– 1086. Sha, J., Kozlova, E. V., and Chopra, A. K. 2002. Role of various enterotoxins in Aeromonas hydrophila-induced gastroenteritis: Generation of enterotoxin gene-defi cient mutants and evaluation of their enterotoxic activity. Infect. Immun. 70: 1924–1935. Simon, M., Grossart, H.-P., Schweitzer, B., and Ploug, H. 2002. Microbial ecology of organic aggregates in aquatic ecosystems. Aquat. Microb. Ecol. 28: 175–211. Simu, K., and Hagström, Å. 2004. Oligotrophic bacterioplankton with a novel single-cell life strategy. Appl. Environ. Microbiol. 70: 2445–2451. Sivonen, K. 2000. Freshwater cyanobacterial neurotoxins: Ecobiology, chemistry, and detection, pp. 567–581. In Botana, L.M. (ed.). Seafood and freshwater toxins: pharmacology, physiology, and detection. Marcel Dekker, Inc. New york, USA. Sivonen, K. 2009. Cyanobacterial toxins, pp. 290–307. In Schaechter, M. (ed.). Encyclopedia of microbiology. Elsevier, Oxford, UK. Sivonen, K., and Börner, T. 2008. Bioactive compounds produced by cyanobacteria, pp. 159- 197. In Herrero, A., and Flores, E. (ed.). The cyanobacteria, molecular biology, genomics and evolution. Caister Academic Press, Norfolk, UK. Sivonen, K., Halinen, K., Sihvonen, L. M., Koskenniemi, K., Sinkko, H., Rantasärkkä, K., Moisander, P. H., and Lyra, C. 2007. Bacterial diversity and function in the Baltic Sea with an emphasis on cyanobacteria. Ambio 36: 180–185. Sivonen, K., and Jones, G. 1999. Cyanobacterial toxins, pp. 41–111. In Chorus, I., and Bartram, J. (ed.), Toxic cyanobacteria in water: A guide to their public health consequences, monitoring and management. E & FN Spon, London, UK. Smith, P. T. 2000. Freshwater neurotoxins: Mechanisms of action, pharmacology, toxicology, and impacts an aquaculture, pp. 583–602. In Botana, L.M. (ed.). Seafood and freshwater toxins: pharmacology, physiology, and detection. Marcel Dekker, Inc. New York, USA. Spoof, L. 2005. Microcystins and nodularins, pp. 15–39. In Meriluoto, J., and Codd, G. A. (ed.). Toxic: Cyanobacterial monitoring and cyanotoxin analysis. Åbo Akademi University Press, Åbo, Finland. Stager, C. E., and Davis, J. R. 1992. Automated systems for identifi cation of microorganisms. Clin. Microbiol. Rev. 5: 302–327. Stewart, I., Schluter, P. J., and Shaw, G. R. 2006a. Cyanobacterial lipopolysaccharides and human health- a review. Environ. Health 5: 7. Stewart, I., Webb, P. M., Schluter, P. J., and Shaw, G. R. 2006b. Recreational and occupational field exposure to freshwater cyanobacteria – a review of anecdotal and case reports, epidemiological studies and the challenges for epidemiologic assessment. Environ. Health 5: 6. Stewart, I., Seawright, A. A., and Shaw, G. R. 2008. Cyanobacterial poisoning in livestock, wild mammals and birds - an overview, pp. 613–637. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Svrcek, C., and Smith, D. W. 2004. Cyanobacteria toxins and the current state of knowledge on water treatment options: a review. J. Environ. Eng. Sci. 3: 155–185. Tharagonnet, D., Sisson, P. R., Roxby, C. M., Ingham, H. R., and Selkon, J. B. 1977. The API ZYM system in the identifi cation of Gram-negative anaerobes. J. Clin. Path. 30: 505–509. Theron, J., and Cloete, T. E. 2002. Emerging waterborne infections: Contributing factors, agents, and detection tools. Crit. Rev. Microbiol. 28: 1–26. Thompson, F. L., Iida, T., and Swings, J. 2004. Biodiversity of vibrios. Microbiol. Mol. Biol. Rev. 68: 403–431.

49 References

Tokajian, S. T., Hashwa, F. A., Hancock, I. C., and Zallou, P. A. 2005. Phylogenetic assessment of heterotrophic bacteria from a water distribution system using 16S rDNA sequencing. Can. J. Microbiol. 51: 325–335. Truu, J., Talpsep, E., Heinaru, E., Stottmeister, U., Wand, H., and Heinaru, A. 1999. Comparison of API 20NE and Biolog GN identifi cation systems assessed by techniques of multivariate analyses. J. Microbiol. Meth. 36: 193–201. Tsuji, K., Naito, S., Kondo, F., Ishikawa, N., Watanabe, M. F., Suzuki, M., and Harada, K.-i. 1994. Stability of microcystins from cyanobacteria: Effect of light on decomposition and isomerization. Environ. Sci. Technol. 28: 173–177. Tuomainen, J., Hietanen, S., Kuparinen, J., Martikainen, P. J., and Servomaa, K. 2006. Community structure of the bacteria associated with Nodularia sp. (cyanobacteria) aggregates in the Baltic Sea. Microb. Ecol. 53: 513–522. van Hove, C., and Lejeune, A. 2002. The Azolla-Anabaena symbiosis. Biol. Environ. 102B: 23–26. Vepsäläinen, M., Kukkonen, S., Vestberg, M., Sirviö, H., and Niemi, R. M. 2001. Application of soil enzymatic activity test kit in a fi eld experiment. Soil Biol. Biochem. 33: 1665–1672. Wang, H., Ho, L., Lewis, D. M., Brookes, J. D., and Newcombe, G. 2007. Discriminating and assessing adsorption and biodegradation removal mechanisms during granular activated carbon fi ltration of microcystin toxins. Water Res. 41: 4262–4270. Welker, M., and von Döhren, H. 2006. Cyanobacterial peptides – Nature’s own combinatorial biosynthesis. FEMS Microbiol. Rev. 30: 530–563. Westrick, J. A. 2008. Cyanobacterial toxin removal in drinking water treatment processes and recreational waters, pp. 275–290. In Hudnell, H. K. (ed.). Cyanobacterial harmful algal blooms: State of science and research needs. Springer, New York, USA. Whitton, B. A., and Potts, M. 2000. Introduction to cyanobacteria, pp. 1–11. In Whitton, B. A., and Potts, M. (ed.), The ecology of cyanobacteria: Their diversity in time and space. Kluwer Academic Publishers, Dordrecht, the Netherlands. WHO. 2008a. Application of the guidelines in specifi c circumstances, pp. 99–120. In Sheffer, M. (ed.). Guidelines for Drinking-water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). WHO. 2008b. Chemical aspects, pp. 145–196. In Sheffer, M. (ed.). Guidelines for drinking- water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). WHO. 2008c. Introduction, pp. 1–21. In Sheffer, M. (ed.). Guidelines for drinking-water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). WHO. 2008d. Microbial aspects, pp. 121–144. In Sheffer, M. (ed.). Guidelines for drinking- water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). WHO. 2008e. Microbial fact sheet, pp. 221–295. In Sheffer, M. (ed.). Guidelines for drinking- water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). WHO. 2008f. Water safety plans, pp. 48–83. In Sheffer, M. (ed.). Guidelines for drinking- water quality. Volume 1, Recommendations. 3rd edition. World Health Organization, Geneva, Switzerland. ISBN 978 92 4 154761 1 (WEB version). Widrig, D. L., Gray, K. A., and McAuliffe, K. S. 1996. Removal of algal-derived organic matter by preozonation and coagulation: Monitoring changes in organic quality by pyrolysis- GS-MS. Water Res. 30: 2621–2632.

50 References

Williams, M. M., Domingo, J. W. S., Meckes, M. C., Kelty, C. A., and Rochon, H. S. 2004. Phylogenic diversity of drinking water bacteria in distribution system simulator. J. Appl. Microbiol. 96: 954–964. Wilson, A. E., Sarnelle, O., and Tillmans, A. R. 2006. Effects of cyanobacterial toxicity and morphology on the population growth of freshwater zooplankton: Meta-analyses of laboratory experiments. Limnol. Oceanogr. 51: 1915–1924. Wong, C. Y. F., Heuzenroeder, M. W., and Flower, R. L. P. 1998. Inactivation of two haemolytic toxin genes in Aeromonas hydrophila attenuates virulence in a suckling mouse model. Microbiol. 144: 291–298. Yamamoto, Y., Kouchiwa, T., Hodoki, Y., Hotta, K., Uchida, H., and Harada, K.-I. 1998. Distribution and identifi cation of actinomycetes lysing cyanobacteria in a eutrophic lake. J. Appl. Phycol. 10: 391–397. Yoshikawa, K., Adachi, K., Nishijima, M., Takadera, T., Tamaki, S., Harada, K.-I., Mochida, K., and Sano, H. 2000. β-Cyanoalanine production by marine bacteria on cyanide-free medium and its specifi c inhibitory activity toward cyanobacteria. Appl. Environ. Microbiol. 66: 718– 722. Yukselen, M. A., O’Halloran, K., and Gregory, J. 2006. Effect of ozone on the formation and breakup of fl ocs in raw waters with high algae content. J. Environ. Sci. Health A. 41: 1173–1182. Zinder, S. H., and Salyers, A. A. 2001. Microbial ecology-New directions, new importance, pp. 101–107. In Garrity, G. M., Boone, D. R., and Castenholtz, R. W. (ed.). Bergey’s manual of systematic bacteriology. Volume 1: The Archaea and the deeply branching and phototrophic Bacteria. 2nd edition. Springer Verlag, New York, USA. Zuur, A. F., Ieno, E. N., and Smith, G. M. 2007. Principal coordinate analysis and non-metric multidimensional scaling, pp. 259–264. In Zuur, A. F., Ieno, E. N., and Smith, G. M.(aut.). Analysing ecological data. Springer Verlag, New York, USA. Zwart, G., Crump, B. C., Kamst-van Agterveld, M. P., Hagen, F., and Han, S. K. 2002. Typical freshwater bacteria: an analysis of available 16S rRNA gene sequences from plankton of lakes and rivers. Aquat. Microb. Ecol. 28: 141–155. Østensvik, Ø., Skulberg, O. M., Underdal, B., and Hormazabal, V. 1998. Antibacterial properties of extracts from selected planktonic freshwater cyanobacteria – a comparative study of bacterial bioassays. J. Appl. Microbiol. 84: 1117–1124.

51 Appendix 1

Appendix 1

The bacterial strains analysed in this study, water type of their origin, isolation medium used, length of the acquired 16S rRNA gene sequence and its accession number, and the taxonomic assignment of the strains.

Water type: FW refers to freshwater sample taken from Finnish lake, river or pond water, TW to sample of treated drinking water, and BW to brackish water sample taken from the Baltic Sea. Isolation medium: The BA refers to blood agar, the CYA to growth medium with frozen and thawed culture of non-microcystin-producing Anabaena strain, the R2A to Reason- er’s 2A medium, the SAA to starch ampicillin agar, the TOX to medium with demethyl variants of microcystins and the Z8 to nutrient poor medium. Bacterial strain type Water Isolation medium gene 16S rRNA Acquired sequence (bp) Gene bank accession number assignment Taxonomic (phylum; class) Actinobacteria; Actinobacteria Classiﬁ ed at genus level JO92 FW CYA 872 AM988866 Arthrobacter RU7 FW CYA 791 AM988867 Kocuria ET4 FW CYA 838 AM988868 Microbacterium HE90 FW BA 830 AM988870 Micrococcus HE95 FW BA 832 AM988871 Micrococcus LO10 FW CYA 740 AM988872 Micrococcus ES1 TW R2A 836 AM988869 Micrococcus RA1 TW R2A 833 AM988873 Micrococcus TA6A FW CYA 833 AM988874 Mycobacterium ES10 TW R2A 829 AM988875 Mycobacterium KK5 TW R2A 833 AM988876 Nocardia LO12 FW CYA 778 AM988877 Nocardioides JO69 FW Z8 868 AM988878 Rhodococcus JO70 FW Z8 868 AM988879 Rhodococcus JO72 FW Z8 868 AM988880 Rhodococcus JO75 FW Z8 868 AM988881 Rhodococcus EN4 FW CYA 741 AM988882 Streptomyces ET11 FW CYA 889 AM988883 Streptomyces LO2 FW CYA 823 AM988885 Streptomyces LO4 FW CYA 734 AM988886 Streptomyces LO5 FW CYA 826 AM988887 Streptomyces LO6 FW CYA 819 AM988888 Streptomyces

52 Appendix 1

LO13 FW CYA 701 AM988889 Streptomyces LO14 FW CYA 823 AM988890 Streptomyces PU7 FW CYA 821 AM988891 Streptomyces TE6 FW CYA 830 AM988892 Streptomyces TE13 FW CYA 831 AM988893 Streptomyces TE14 FW CYA 869 AM988894 Streptomyces JO77 FW R2A 869 AM988884 Streptomyces TE37 FW TOX 825 AM988895 Streptomyces TE41 FW TOX 833 AM988896 Streptomyces Classiﬁ ed at family level RU14 FW CYA 829 AM988897 Microbacteriaceae Bacteroidetes; Flavobacteria Classiﬁ ed at genus level HE85 FW BA 839 AM988898 Chryseobacterium HE92 FW BA 849 AM988899 Chryseobacterium AE103 FW SAA 456 EU724056 Chryseobacterium AE107 FW SAA 456 EU724051 Chryseobacterium AE109 FW SAA 456 EU724052 Chryseobacterium AE110 FW SAA 456 EU724053 Chryseobacterium AE118 FW SAA 456 EU724054 Chryseobacterium AE120 FW SAA 456 EU724055 Chryseobacterium JO2 FW TOX 881 AM988900 Chryseobacterium JO7 FW TOX 882 AM988901 Chryseobacterium VA6 TW R2A 843 AM988902 Chryseobacterium PU5 FW CYA 835 AM988924 Flavobacterium KU12 FW R2A 841 AM988923 Flavobacterium RU8 FW R2A 838 AM988927 Flavobacterium RU9 FW R2A 831 AM988928 Flavobacterium HA4 FW TOX 837 AM988903 Flavobacterium JO5 FW TOX 838 AM988904 Flavobacterium JO9 FW TOX 851 AM988905 Flavobacterium JO11 FW TOX 791 AM988906 Flavobacterium JO12 FW TOX 821 AM988907 Flavobacterium JO15 FW TOX 841 AM988908 Flavobacterium JO16 FW TOX 879 AM988909 Flavobacterium JO23 FW TOX 834 AM988910 Flavobacterium JO24 FW TOX 878 AM988911 Flavobacterium JO25 FW TOX 838 AM988912 Flavobacterium JO26 FW TOX 840 AM988913 Flavobacterium JO27 FW TOX 882 AM988914 Flavobacterium JO28 FW TOX 835 AM988915 Flavobacterium JO31 FW TOX 879 AM988916 Flavobacterium JO32 FW TOX 889 AM988917 Flavobacterium JO36 FW TOX 838 AM988918 Flavobacterium JO41 FW TOX 879 AM988919 Flavobacterium JO104 FW TOX 878 AM988920 Flavobacterium RJ1 FW TOX 832 AM988925 Flavobacterium RJ5 FW TOX 834 AM988926 Flavobacterium TE19 FW TOX 827 AM988929 Flavobacterium TE25 FW TOX 826 AM988930 Flavobacterium TE26 FW TOX 841 AM988931 Flavobacterium

53 Appendix 1

TE28 FW TOX 840 AM988932 Flavobacterium TE31 FW TOX 835 AM988933 Flavobacterium TE40 FW TOX 824 AM988934 Flavobacterium KK2 TW R2A 841 AM988921 Flavobacterium KK3 TW R2A 840 AM988922 Flavobacterium TE18 FW TOX 845 AM988935 Riemerella Classifi ed at family level HE91 FW BA 870 AM988936 Flavobacteriaceae Bacteroidetes; Sphingobacteria Classifi ed at genus level OT2 FW CYA 792 AM988938 Arcicella TE7 FW CYA 831 AM988939 Arcicella JO107 FW TOX 869 AM988937 Arcicella TE29 FW TOX 803 AM988940 Arcicella RU12 FW CYA 842 AM988943 Chitinophaga JO14 FW TOX 888 AM988941 Chitinophaga JO108 FW TOX 884 AM988942 Chitinophaga LI1 FW CYA 809 AM988944 Flectobacillus JO45 FW CYA 840 AM988947 Pedobacter PU10 FW CYA 837 AM988949 Pedobacter TA2 FW CYA 837 AM988951 Pedobacter TE12 FW CYA 884 AM988953 Pedobacter IJ6 FW TOX 781 AM988945 Pedobacter JO4 FW TOX 884 AM988946 Pedobacter PU13 FW TOX 794 AM988950 Pedobacter TE5 FW TOX 886 AM988952 Pedobacter TE20 FW TOX 833 AM988954 Pedobacter TE20B FW TOX 837 AM988955 Pedobacter TE23 FW TOX 839 AM988956 Pedobacter JO56 FW Z8 882 AM988948 Pedobacter JO115 FW TOX 862 AM988957 Spirosoma Deinococcus-Thermus; Deinococci Classifi ed at genus level RU1 FW TOX 778 AM988958 Deinococcus Firmicutes; Bacilli Classifi ed at genus level HE105 BW BA 496 AM988965 Bacillus 4285C FW BA 496 AM988959 Bacillus 4414 FW BA 496 AM988960 Bacillus HE99 FW BA 496 AM988964 Bacillus ET1 FW R2A 858 AM988963 Bacillus RJ4 FW TOX 816 AM988966 Bacillus TE36 FW TOX 903 AM988968 Bacillus ES2 TW R2A 863 AM988961 Bacillus ES8 TW R2A 859 AM988962 Bacillus RN13 TW R2A 849 AM988967 Bacillus 4361A BW BA 495 AM988969 Caryophanon HE53 FW BA 861 AM988970 Lactococcus ES4 TW R2A 833 AM988971 Paenibacillus 4337D FW BA 497 AM988972 Staphylococcus

54 Appendix 1

4346A FW BA 497 AM988973 Staphylococcus JO48 FW CYA 903 AM988975 Staphylococcus SL5 TW R2A 844 AM988974 Staphylococcus Proteobacteria; Alphaproteobacteria Classiﬁ ed at genus level KU2 FW R2A 805 AM988977 Azospirillum TE21 FW TOX 781 AM988978 Azospirillum JO50 FW Z8 836 AM988976 Azospirillum RN8 TW R2A 840 AM988979 Blastomonas RN9 TW R2A 744 AM988980 Blastomonas HA3 FW CYA 779 AM988981 Bosea LO7 FW CYA 790 AM988983 Bosea RU13 FW CYA 774 AM988986 Bosea TE8 FW CYA 797 AM988987 Bosea TE10 FW CYA 795 AM988988 Bosea TE11 FW CYA 800 AM988989 Bosea TE34 FW TOX 795 AM988990 Bosea KK9 TW R2A 791 AM988982 Bosea RA3 TW R2A 799 AM988984 Bosea RN17 TW R2A 794 AM988985 Bosea 4363A FW BA 442 AM988991 Brevundimonas JO43 FW CYA 837 AM988994 Brevundimonas JO46 FW CYA 839 AM988995 Brevundimonas PU6 FW CYA 790 AM989000 Brevundimonas RU6 FW CYA 808 AM989002 Brevundimonas KU11 FW R2A 779 AM988999 Brevundimonas JO6 FW TOX 836 AM988993 Brevundimonas JO103 FW TOX 837 AM988997 Brevundimonas JO62 FW Z8 840 AM988996 Brevundimonas HV3 TW R2A 838 AM988992 Brevundimonas KK4 TW R2A 780 AM988998 Brevundimonas RA10 TW R2A 771 AM989001 Brevundimonas SL1 TW R2A 778 AM989003 Brevundimonas SL2A TW R2A 779 AM989004 Brevundimonas SL2B TW R2A 779 AM989005 Brevundimonas VA7 TW R2A 836 AM989006 Brevundimonas ET2 FW CYA 793 AM989007 Caulobacter JO49 FW CYA 838 AM989008 Caulobacter JO91 FW CYA 839 AM989011 Caulobacter JO95 FW CYA 837 AM989012 Caulobacter PU8 FW CYA 799 AM989014 Caulobacter TE1 FW CYA 803 AM989015 Caulobacter JO76 FW R2A 840 AM989010 Caulobacter JO100 FW TOX 840 AM989013 Caulobacter JO51 FW Z8 838 AM989009 Caulobacter EN3 FW CYA 787 AM989016 Methylobacterium EN5 FW CYA 790 AM989017 Methylobacterium ET7 FW CYA 750 AM989022 Methylobacterium ET10 FW CYA 849 AM989023 Methylobacterium LO11 FW CYA 788 AM989025 Methylobacterium OT1 FW CYA 792 AM989026 Methylobacterium

55 Appendix 1

OT4 FW CYA 771 AM989027 Methylobacterium OT5 FW CYA 742 AM989028 Methylobacterium OT7 FW CYA 750 AM989029 Methylobacterium PU2 FW CYA 757 AM989030 Methylobacterium PU3 FW CYA 757 AM989031 Methylobacterium TA3 FW CYA 737 AM989032 Methylobacterium TA5 FW CYA 736 AM989033 Methylobacterium KU9 FW R2A 875 AM989024 Methylobacterium ES3 TW R2A 794 AM989018 Methylobacterium ES5 TW R2A 799 AM989019 Methylobacterium ES6 TW R2A 800 AM989020 Methylobacterium ES11 TW R2A 799 AM989021 Methylobacterium ET5 FW CYA 795 AM989034 Novosphingobium TA4A FW CYA 736 AM989036 Novosphingobium TA1 FW R2A 788 AM989035 Novosphingobium LO3 FW CYA 809 AM989037 Paracoccus HV6 TW R2A 837 AM989038 Phyllobacterium HV9 TW R2A 776 AM989039 Phyllobacterium VA5 TW R2A 835 AM989040 Phyllobacterium KU5 BW R2A 748 AM989041 Porphyrobacter IJ2 FW CYA 792 AM989042 Sphingobium IJ5 FW TOX 802 AM989043 Sphingobium ET6 FW CYA 750 AM989044 Sphingomonas JO47 FW CYA 840 AM989051 Sphingomonas JO78 FW CYA 839 AM989052 Sphingomonas JO79 FW CYA 792 AM989053 Sphingomonas JO89 FW CYA 794 AM989054 Sphingomonas LI2 FW CYA 840 AM989055 Sphingomonas LO8 FW CYA 805 AM989056 Sphingomonas TA2A FW CYA 788 AM989062 Sphingomonas TU6A FW CYA 798 AM989064 Sphingomonas TU3 FW R2A 788 AM989063 Sphingomonas JO3 FW TOX 840 AM989049 Sphingomonas JO8 FW TOX 817 AM989050 Sphingomonas HV2 TW R2A 841 AM989045 Sphingomonas HV4 TW R2A 786 AM989046 Sphingomonas HV7 TW R2A 832 AM989047 Sphingomonas HV10 TW R2A 839 AM989048 Sphingomonas RA5 TW R2A 788 AM989057 Sphingomonas RA7 TW R2A 789 AM989058 Sphingomonas RN2 TW R2A 788 AM989059 Sphingomonas SL7 TW R2A 788 AM989060 Sphingomonas SL9 TW R2A 803 AM989061 Sphingomonas VA4 TW R2A 796 AM989065 Sphingomonas VA8 TW R2A 841 AM989066 Sphingomonas KU13 FW R2A 787 AM989068 Sphingopyxis JO106 FW TOX 841 AM989067 Sphingopyxis RA4 TW R2A 789 AM989069 Sphingopyxis Classiﬁ ed at family level TA7 FW CYA 803 AM989070 Acetobacteraceae TA7A FW CYA 788 AM989073 Sphingomonadaceae RA2 TW R2A 788 AM989071 Sphingomonadaceae RN19 TW R2A 842 AM989072 Sphingomonadaceae

56 Appendix 1

Classifi ed at class level KU10 FW R2A 793 AM989074 Alphaproteobacteria TA3A FW CYA 800 AM989075 Alphaproteobacteria Proteobacteria; Betaproteobacteria Classifi ed at species level 15Ba TOX 1422 AY515379 Paucibacter toxinivorans 2a TOX 1423 AY515380 Paucibacter toxinivorans 2B15a TOX 1431 AY515381 Paucibacter toxinivorans 2B17a TOX 1428 AY515382 Paucibacter toxinivorans 2B3a TOX 1421 AY515383 Paucibacter toxinivorans S1030 a,b TOX 1428 AY515384 Paucibacter toxinivorans G44a TOX 1434 AY515385 Paucibacter toxinivorans F39a TOX 1425 AY515386 Paucibacter toxinivorans F36 a,b TOX 1430 AY515387 Paucibacter toxinivorans 7Aa TOX 1414 AY515388 Paucibacter toxinivorans 2C23a,b TOX 1444 AY515389 Paucibacter toxinivorans 2C20T a,b TOX 1429 AY515390 Paucibacter toxinivorans 2B5a TOX 1384 AY515391 Paucibacter toxinivorans Classifi ed at genus level KU6 BW R2A 837 AM989076 Aquabacterium HE147 FW BA 483 AM989077 Aquitalea 4398A BW BA 413 AM989078 Burkholderia TA13 FW TOX 886 AM989079 Chitinimonas 4288A FW BA 483 AM989080 Chromobacterium 4336A FW BA 483 AM989081 Chromobacterium HE106 FW BA 483 AM989082 Chromobacterium HE127 FW BA 474 AM989083 Chromobacterium HE160 FW BA 485 AM989084 Chromobacterium HA1 FW CYA 850 AM989085 Duganella LO1 FW CYA 839 AM989090 Duganella TA4 FW CYA 903 AM989091 Duganella TA11 FW CYA 848 AM989092 Duganella JO22 FW TOX 807 AM989086 Duganella JO73 FW TOX 871 AM989087 Duganella KK1 TW R2A 823 AM989088 Duganella KK11 TW R2A 823 AM989089 Duganella EN2 FW CYA 851 AM989093 Herbaspirillum JO80 FW CYA 891 AM989096 Herbaspirillum JO83 FW CYA 893 AM989097 Herbaspirillum LO9 FW CYA 846 AM989100 Herbaspirillum PU4 FW CYA 843 AM989101 Herbaspirillum HA5 FW TOX 848 AM989094 Herbaspirillum JO114 FW TOX 888 AM989098 Herbaspirillum TE24 FW TOX 848 AM989102 Herbaspirillum JO59 FW Z8 855 AM989095 Herbaspirillum KK10 TW R2A 839 AM989099 Herbaspirillum RN14 TW R2A 693 Herbaspirillum RN18 TW R2A 695 Herbaspirillum PU12 FW TOX 802 AM989103 Iodobacter JO1 FW TOX 890 AM989104 Janthinobacterium

57 Appendix 1

ET8 FW CYA 847 AM989105 Pelomonas ET9 FW CYA 886 AM989106 Pelomonas HA2 FW CYA 833 AM989107 Pelomonas JO84 FW CYA 830 AM989108 Pelomonas JO86 FW CYA 835 AM989109 Pelomonas JO99 FW CYA 880 AM989110 Pelomonas TE4 FW TOX 843 AM989111 Rhodoferax IJ3 FW CYA 839 AM989112 Roseateles JO87 FW CYA 832 AM989113 Roseateles JO90 FW CYA 881 AM989114 Roseateles JO96 FW CYA 829 AM989115 Roseateles JO97 FW CYA 854 AM989116 Roseateles KU4 FW R2A 835 AM989117 Roseateles KU7 FW R2A 817 AM989118 Roseateles TE15 FW TOX 832 AM989119 Vogesella TE17 FW TOX 849 AM989120 Vogesella TE22 FW TOX 851 AM989121 Vogesella Classifi ed at family level KU8 FW R2A 838 AM989122 Alcaligenaceae JO111 FW TOX 880 AM989123 Incertae sedis 5 KU1 FW R2A 851 AM989128 Oxalobacteraceae JO21 FW TOX 797 AM989124 Oxalobacteraceae JO113 FW TOX 891 AM989126 Oxalobacteraceae JO53 FW Z8 889 AM989125 Oxalobacteraceae KK7 TW R2A 847 AM989127 Oxalobacteraceae RN12 TW R2A 806 AM989129 Oxalobacteraceae Classifi ed at class level KU14 FW R2A 805 AM989130 Betaproteobacteria Proteobacteria; Gammaproteobacteria Classifi ed at genus level 4286C FW BA 485 AM989131 Acinetobacter 4300B FW BA 485 AM989132 Acinetobacter 4362A FW BA 483 AM989133 Acinetobacter 4367I FW BA 485 AM989134 Acinetobacter HE21 FW BA 856 AM989137 Acinetobacter HE55 FW BA 853 AM989138 Acinetobacter HE120 FW BA 485 AM989139 Acinetobacter HE124 FW BA 485 AM989140 Acinetobacter HE128 FW BA 485 AM989141 Acinetobacter HE138 FW BA 488 AM989142 Acinetobacter HE139 FW BA 483 AM989143 Acinetobacter HE140 FW BA 483 AM989144 Acinetobacter HE144 FW BA 487 AM989145 Acinetobacter HE145 FW BA 487 AM989146 Acinetobacter HE146 FW BA 487 AM989147 Acinetobacter AU2 FW R2A 849 AM989135 Acinetobacter AU3 FW R2A 847 AM989136 Acinetobacter JO101 FW TOX 895 AM989148 Acinetobacter RU5 FW TOX 853 AM989149 Acinetobacter TE16 FW TOX 850 AM989151 Acinetobacter SL10 TW R2A 851 AM989150 Acinetobacter

58 Appendix 1

4218B BW BA 492 AM989159 Aeromonas 4218E BW BA 492 AM989160 Aeromonas 4282D BW BA 492 AM989161 Aeromonas 4282J BW BA 492 AM989162 Aeromonas 4294H BW BA 492 AM989166 Aeromonas 4295A BW BA 492 AM989167 Aeromonas 4295F BW BA 491 AM989168 Aeromonas 4298D BW BA 491 AM989169 Aeromonas 4329A BW BA 492 AM989182 Aeromonas 4330A BW BA 489 AM989183 Aeromonas 4330D BW BA 492 AM989184 Aeromonas 4330E BW BA 492 AM989185 Aeromonas 4330G BW BA 494 AM989186 Aeromonas 4348A BW BA 491 AM989188 Aeromonas 4348D BW BA 494 AM989189 Aeromonas 4359B BW BA 492 AM989190 Aeromonas 4368D BW BA 492 AM989193 Aeromonas 4387A BW BA 492 AM989194 Aeromonas 4387B BW BA 492 AM989195 Aeromonas HE1 BW BA 867 AM989198 Aeromonas HE6 BW BA 861 AM989201 Aeromonas HE7 BW BA 864 AM989202 Aeromonas HE34 BW BA 857 AM989218 Aeromonas HE35 BW BA 857 AM989219 Aeromonas HE94 BW BA 856 AM989250 Aeromonas HE112 BW BA 492 AM989253 Aeromonas HE125 BW BA 492 AM989257 Aeromonas HE130 BW BA 489 AM989258 Aeromonas HE132 BW BA 489 AM989259 Aeromonas HE136 BW BA 492 AM989260 Aeromonas HE149 BW BA 492 AM989261 Aeromonas HE154 BW BA 492 AM989263 Aeromonas HE163 BW BA 492 AM989268 Aeromonas AE26 BW SAA 468 EU724006 Aeromonas AE27 BW SAA 452 EU724007 Aeromonas AE28 BW SAA 468 EU723962 Aeromonas AE30 BW SAA 453 EU723973 Aeromonas AE31 BW SAA 468 EU723952 Aeromonas AE32 BW SAA 444 EU723963 Aeromonas AE33 BW SAA 468 EU723902 Aeromonas AE34 BW SAA 468 EU723977 Aeromonas AE35 BW SAA 468 EU723956 Aeromonas AE36 BW SAA 468 EU724009 Aeromonas AE37 BW SAA 468 EU724010 Aeromonas AE39 BW SAA 468 EU723964 Aeromonas AE40 BW SAA 468 EU723967 Aeromonas AE41 BW SAA 468 EU723958 Aeromonas AE42 BW SAA 454 EU724011 Aeromonas AE43 BW SAA 468 EU724012 Aeromonas AE44 BW SAA 468 EU723968 Aeromonas AE45 BW SAA 468 EU723972 Aeromonas AE59 BW SAA 468 EU723978 Aeromonas

59 Appendix 1

AE60 BW SAA 468 EU723979 Aeromonas AE61 BW SAA 468 EU724013 Aeromonas AE62 BW SAA 468 EU724014 Aeromonas AE63 BW SAA 468 EU724015 Aeromonas AE64 BW SAA 468 EU723989 Aeromonas AE65 BW SAA 468 EU723994 Aeromonas AE67 BW SAA 468 EU724016 Aeromonas AE72 BW SAA 468 EU724017 Aeromonas AE73 BW SAA 468 EU724018 Aeromonas AE74 BW SAA 431 EU723912 Aeromonas AE81 BW SAA 468 EU724021 Aeromonas AE83 BW SAA 468 EU723991 Aeromonas AE84 BW SAA 468 EU724022 Aeromonas AE85 BW SAA 468 EU723993 Aeromonas AE86 BW SAA 468 EU724023 Aeromonas AE87 BW SAA 468 EU724024 Aeromonas AE88 BW SAA 468 EU724025 Aeromonas AE89 BW SAA 468 EU724026 Aeromonas AE90 BW SAA 468 EU724027 Aeromonas AE91 BW SAA 468 EU723999 Aeromonas AE95 BW SAA 468 EU724028 Aeromonas AE131 BW SAA 468 EU723948 Aeromonas AE137 BW SAA 468 EU723981 Aeromonas AE138 BW SAA 468 EU723982 Aeromonas AE139 BW SAA 468 EU724045 Aeromonas AE141 BW SAA 468 EU724031 Aeromonas AE158 BW SAA 468 EU724046 Aeromonas AE169 BW SAA 468 EU723965 Aeromonas AE170 BW SAA 468 EU723983 Aeromonas AE171 BW SAA 468 EU723960 Aeromonas AE172 BW SAA 468 EU723974 Aeromonas AE174 BW SAA 452 EU723908 Aeromonas AE175 BW SAA 468 EU723893 Aeromonas AE176 BW SAA 468 EU723900 Aeromonas AE188 BW SAA 468 EU723995 Aeromonas AE189 BW SAA 468 EU723961 Aeromonas AE191 BW SAA 468 EU724035 Aeromonas AE192 BW SAA 468 EU723992 Aeromonas AE246 BW SAA 431 EU723913 Aeromonas AE248 BW SAA 468 EU723996 Aeromonas AE249 BW SAA 468 EU724044 Aeromonas AE250 BW SAA 468 EU723987 Aeromonas AE251 BW SAA 468 EU723997 Aeromonas 4192A FW BA 491 AM989152 Aeromonas 4201B FW BA 492 AM989153 Aeromonas 4201D FW BA 492 AM989154 Aeromonas 4201G FW BA 492 AM989155 Aeromonas 4207A FW BA 492 AM989156 Aeromonas 4207C FW BA 492 AM989157 Aeromonas 4207D FW BA 492 AM989158 Aeromonas 4286A FW BA 492 AM989163 Aeromonas 4286B FW BA 492 AM989164 Aeromonas

60 Appendix 1

4287D FW BA 492 AM989165 Aeromonas 4300D FW BA 491 AM989170 Aeromonas 4300E FW BA 492 AM989171 Aeromonas 4301B FW BA 492 AM989172 Aeromonas 4301C FW BA 492 AM989173 Aeromonas 4301D FW BA 492 AM989174 Aeromonas 4302G FW BA 492 AM989175 Aeromonas 4316E FW BA 492 AM989176 Aeromonas 4316F FW BA 492 AM989177 Aeromonas 4318A FW BA 492 AM989178 Aeromonas 4318B FW BA 905 AM989179 Aeromonas 4318C FW BA 492 AM989180 Aeromonas 4326G FW BA 492 AM989181 Aeromonas 4337B FW BA 492 AM989187 Aeromonas 4365A FW BA 492 AM989191 Aeromonas 4366A FW BA 492 AM989192 Aeromonas 4400A FW BA 492 AM989196 Aeromonas 4402C FW BA 492 AM989197 Aeromonas HE3 FW BA 856 AM989199 Aeromonas HE4 FW BA 855 AM989200 Aeromonas HE9 FW BA 864 AM989203 Aeromonas HE10 FW BA 861 AM989204 Aeromonas HE11 FW BA 859 AM989205 Aeromonas HE12 FW BA 863 AM989206 Aeromonas HE13 FW BA 842 AM989207 Aeromonas HE14 FW BA 863 AM989208 Aeromonas HE15 FW BA 857 AM989209 Aeromonas HE16 FW BA 864 AM989210 Aeromonas HE17 FW BA 865 AM989211 Aeromonas HE18 FW BA 855 AM989212 Aeromonas HE19 FW BA 856 AM989213 Aeromonas HE20 FW BA 863 AM989214 Aeromonas HE22 FW BA 864 AM989215 Aeromonas HE28 FW BA 860 AM989216 Aeromonas HE29 FW BA 856 AM989217 Aeromonas HE37 FW BA 856 AM989220 Aeromonas HE38 FW BA 856 AM989221 Aeromonas HE39 FW BA 858 AM989222 Aeromonas HE40 FW BA 851 AM989223 Aeromonas HE44 FW BA 855 AM989224 Aeromonas HE45 FW BA 851 AM989225 Aeromonas HE46 FW BA 855 AM989226 Aeromonas HE47 FW BA 857 AM989227 Aeromonas HE48 FW BA 864 AM989228 Aeromonas HE51 FW BA 859 AM989229 Aeromonas HE56 FW BA 855 AM989230 Aeromonas HE57 FW BA 858 AM989231 Aeromonas HE60 FW BA 855 AM989232 Aeromonas HE61 FW BA 858 AM989233 Aeromonas HE62 FW BA 856 AM989234 Aeromonas HE63 FW BA 855 AM989235 Aeromonas HE64 FW BA 863 AM989236 Aeromonas

61 Appendix 1

HE65 FW BA 864 AM989237 Aeromonas HE73 FW BA 859 AM989238 Aeromonas HE74 FW BA 866 AM989239 Aeromonas HE75 FW BA 856 AM989240 Aeromonas HE76 FW BA 859 AM989241 Aeromonas HE77 FW BA 858 AM989242 Aeromonas HE78 FW BA 856 AM989243 Aeromonas HE79 FW BA 857 AM989244 Aeromonas HE80 FW BA 863 AM989245 Aeromonas HE82 FW BA 864 AM989246 Aeromonas HE83 FW BA 858 AM989247 Aeromonas HE86 FW BA 858 AM989248 Aeromonas HE89 FW BA 858 AM989249 Aeromonas HE101 FW BA 491 AM989251 Aeromonas HE110 FW BA 492 AM989252 Aeromonas HE113 FW BA 492 AM989254 Aeromonas HE114 FW BA 491 AM989255 Aeromonas HE115 FW BA 492 AM989256 Aeromonas HE150 FW BA 492 AM989262 Aeromonas HE155 FW BA 492 AM989264 Aeromonas HE157 FW BA 492 AM989265 Aeromonas HE158 FW BA 492 AM989266 Aeromonas HE159 FW BA 492 AM989267 Aeromonas HE165 FW BA 492 AM989269 Aeromonas TU4 FW CYA 850 AM989271 Aeromonas TU6 FW CYA 807 AM989272 Aeromonas AE1 FW SAA 481 EU724000 Aeromonas AE2 FW SAA 468 EU723949 Aeromonas AE3 FW SAA 443 EU723887 Aeromonas AE6 FW SAA 468 EU724001 Aeromonas AE7 FW SAA 468 EU724048 Aeromonas AE9 FW SAA 468 EU723888 Aeromonas AE10 FW SAA 468 EU724002 Aeromonas AE12 FW SAA 468 EU724003 Aeromonas AE13 FW SAA 468 EU724004 Aeromonas AE14 FW SAA 468 EU723889 Aeromonas AE15 FW SAA 451 EU724008 Aeromonas AE16 FW SAA 468 EU723985 Aeromonas AE17 FW SAA 443 EU723907 Aeromonas AE18 FW SAA 468 EU723953 Aeromonas AE19 FW SAA 452 EU724005 Aeromonas AE20 FW SAA 468 EU723905 Aeromonas AE21 FW SAA 468 EU723998 Aeromonas AE23 FW SAA 468 EU723918 Aeromonas AE25 FW SAA 468 EU723906 Aeromonas AE48 FW SAA 448 EU723947 Aeromonas AE53 FW SAA 468 EU723927 Aeromonas AE71 FW SAA 468 EU723954 Aeromonas AE75 FW SAA 468 EU724019 Aeromonas AE78 FW SAA 468 EU724020 Aeromonas AE92 FW SAA 468 EU723970 Aeromonas AE93 FW SAA 445 EU723915 Aeromonas

62 Appendix 1

AE98 FW SAA 468 EU723932 Aeromonas AE99 FW SAA 468 EU723933 Aeromonas AE101 FW SAA 468 EU723934 Aeromonas AE102 FW SAA 468 EU723935 Aeromonas AE104 FW SAA 468 EU723896 Aeromonas AE105 FW SAA 468 EU723938 Aeromonas AE106 FW SAA 468 EU724029 Aeromonas AE111 FW SAA 468 EU724030 Aeromonas AE117 FW SAA 452 EU723984 Aeromonas AE119 FW SAA 433 EU723916 Aeromonas AE121 FW SAA 468 EU723890 Aeromonas AE122 FW SAA 468 EU723919 Aeromonas AE123 FW SAA 443 EU723895 Aeromonas AE125 FW SAA 443 EU723897 Aeromonas AE126 FW SAA 443 EU723910 Aeromonas AE127 FW SAA 468 EU723898 Aeromonas AE128 FW SAA 443 EU723899 Aeromonas AE132 FW SAA 443 EU723891 Aeromonas AE133 FW SAA 443 EU723892 Aeromonas AE143 FW SAA 468 EU723988 Aeromonas AE145 FW SAA 468 EU723955 Aeromonas AE147 FW SAA 468 EU723959 Aeromonas AE150 FW SAA 450 EU723931 Aeromonas AE152 FW SAA 468 EU723942 Aeromonas AE153 FW SAA 468 EU724032 Aeromonas AE154 FW SAA 468 EU723929 Aeromonas AE161 FW SAA 468 EU723950 Aeromonas AE163 FW SAA 468 EU723951 Aeromonas AE165 FW SAA 468 EU723980 Aeromonas AE167 FW SAA 467 EU724033 Aeromonas AE168 FW SAA 468 EU724034 Aeromonas AE180 FW SAA 468 EU723925 Aeromonas AE181 FW SAA 468 EU723976 Aeromonas AE194 FW SAA 468 EU724036 Aeromonas AE195 FW SAA 468 EU723901 Aeromonas AE196 FW SAA 468 EU723920 Aeromonas AE197 FW SAA 468 EU724037 Aeromonas AE198 FW SAA 446 EU723917 Aeromonas AE199 FW SAA 468 EU724038 Aeromonas AE200 FW SAA 468 EU724039 Aeromonas AE201 FW SAA 443 EU723909 Aeromonas AE202 FW SAA 468 EU723936 Aeromonas AE203 FW SAA 468 EU723969 Aeromonas AE204 FW SAA 468 EU723921 Aeromonas AE205 FW SAA 468 EU723975 Aeromonas AE206 FW SAA 468 EU723894 Aeromonas AE207 FW SAA 468 EU723923 Aeromonas AE208 FW SAA 468 EU724040 Aeromonas AE209 FW SAA 468 EU724041 Aeromonas AE210 FW SAA 468 EU723926 Aeromonas AE213 FW SAA 468 EU724042 Aeromonas AE214 FW SAA 468 EU723914 Aeromonas

63 Appendix 1

AE215 FW SAA 468 EU723928 Aeromonas AE216 FW SAA 468 EU723924 Aeromonas AE217 FW SAA 445 EU723903 Aeromonas AE218 FW SAA 433 EU723930 Aeromonas AE219 FW SAA 468 EU723937 Aeromonas AE222 FW SAA 468 EU723990 Aeromonas AE223 FW SAA 468 EU723940 Aeromonas AE230 FW SAA 468 EU724047 Aeromonas AE232 FW SAA 468 EU724049 Aeromonas AE233 FW SAA 468 EU724043 Aeromonas AE234 FW SAA 468 EU723939 Aeromonas AE235 FW SAA 445 EU723943 Aeromonas AE237 FW SAA 468 EU723922 Aeromonas AE238 FW SAA 468 EU723946 Aeromonas AE239 FW SAA 468 EU723986 Aeromonas AE240 FW SAA 468 EU723945 Aeromonas AE243 FW SAA 468 EU723966 Aeromonas AE256 FW SAA 443 EU723904 Aeromonas AE258 FW SAA 468 EU723971 Aeromonas AE259 FW SAA 443 EU723911 Aeromonas AE260 FW SAA 468 EU723957 Aeromonas JO110 FW TOX 896 AM989270 Aeromonas AE162 FW SAA 441 EU724057 Erwinia 4329C BW BA 482 AM989273 Pseudomonas HE5 BW BA 856 AM989275 Pseudomonas HE49 BW BA 851 AM989280 Pseudomonas HE54 BW BA 847 AM989281 Pseudomonas HE108 BW BA 484 AM989290 Pseudomonas HE116 BW BA 482 AM989291 Pseudomonas HE2 FW BA 858 AM989274 Pseudomonas HE8 FW BA 856 AM989276 Pseudomonas HE23 FW BA 857 AM989277 Pseudomonas HE24 FW BA 857 AM989278 Pseudomonas HE27 FW BA 848 AM989279 Pseudomonas HE66 FW BA 850 AM989282 Pseudomonas HE67 FW BA 851 AM989283 Pseudomonas HE68 FW BA 848 AM989284 Pseudomonas HE69 FW BA 849 AM989285 Pseudomonas HE70 FW BA 848 AM989286 Pseudomonas HE71 FW BA 849 AM989287 Pseudomonas HE93 FW BA 847 AM989288 Pseudomonas HE103 FW BA 484 AM989289 Pseudomonas RU10 FW CYA 813 AM989297 Pseudomonas KU3 FW R2A 852 AM989293 Pseudomonas JO33 FW TOX 888 AM989292 Pseudomonas TE27 FW TOX 848 AM989302 Pseudomonas RN1 TW R2A 888 AM989294 Pseudomonas RN5 TW R2A 888 AM989295 Pseudomonas RN21 TW R2A 889 AM989296 Pseudomonas SL3 TW R2A 850 AM989298 Pseudomonas SL4 TW R2A 834 AM989299 Pseudomonas SL8 TW R2A 852 AM989300 Pseudomonas

64 Appendix 1

SL11 TW R2A 850 AM989301 Pseudomonas VA2 TW R2A 889 AM989303 Pseudomonas 4218A BW BA 482 AM989304 Rheinheimera 4218G BW BA 483 AM989305 Rheinheimera TE32 FW TOX 838 AM989306 Rheinheimera HE59 FW BA 857 AM989307 Serratia TE38 FW TOX 855 AM989308 Stenotrophomonas HE30 BW BA 855 AM989309 Vibrio HE32 BW BA 864 AM989310 Vibrio HE33 BW BA 864 AM989311 Vibrio HE36 BW BA 855 AM989312 Vibrio HE42 BW BA 854 AM989313 Vibrio HE43 BW BA 852 AM989314 Vibrio HE50 BW BA 854 AM989315 Vibrio HE87 BW BA 853 AM989316 Vibrio HE88 BW BA 853 AM989317 Vibrio HE111 BW BA 487 AM989318 Vibrio HE161 BW BA 490 AM989319 Vibrio HE162 BW BA 489 AM989320 Vibrio Classiﬁ ed at family level HE25 FW BA 854 AM989321 Enterobacteriaceae HE26 FW BA 855 AM989322 Enterobacteriaceae HE58 FW BA 860 AM989323 Enterobacteriaceae HE72 FW BA 852 AM989324 Enterobacteriaceae AE51 FW SAA 397 EU724058 Enterobacteriaceae AE164 FW SAA 452 EU724059 Enterobacteriaceae 4218F BW BA 492 AM989325 Vibrionaceae aPaucibacter toxinivorans strains were originally isolated by Lahti et al. (1998) and were shown to degrade microcystins in their preliminary studies, using mainly one demethyl variant of microcystin-RR. The strains G44 and F36 were shown to degrade also nodularin in preliminary studies. bThe strain was found to degrade microcystins LR and YR as well as nodularin in this study.