Dynamics of Toxic Cyanobacteria in Lakes and Artificial Water Reservoirs

Dynamics of toxic cyanobacteria in lakes and artificial water reservoirs

DISSERTATION

zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr.rer.nat.)

vorgelegt von Barbara Weisbrod an der

Mathematisch-naturwissenschaftliche Sektion

Fachbereich Biologie

Konstanz, 2020

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-2-2xwjal4f9pd41

Tag der mündlichen Prüfung: 20.01.2021

1. Referent: Prof. Dr. Daniel R. Dietrich

2. Referent: Prof. Dr. Dominik Martin-Creuzburg

Blue, green, red and brown

Cyanobacteria

Pretty, but toxic

(Haiku on Cyanobacteria, Stefan Altaner 2020)

DANKSAGUNG

I. DANKSAGUNG

Diese Arbeit wäre nicht realisierbar gewesen ohne die Hilfe und Unterstützung zahlreicher Personen, Einrichtungen und Organisationen denen ich an dieser Stelle ausführlich danken möchte. Zuallererst, vielen Dank an Daniel Dietrich für die Bereitstellung des Themas, die Unterstützung während der gesamten Arbeit und dafür, dass er mir den Forschungsaufenthalt in Neuseeland ermöglicht hat. Ebenso danke ich Dominik Martin- Creuzburg für seine Arbeit als Zweitgutachter und seine Mitarbeit und Unterstützung bei mehreren Manuskripten.

Besonderer Dank gilt dem Ministerium für Wissenschaft, Forschung und Kunst Baden- Württemberg durch dessen Finanzierung (Wassernetzwerk Baden-Württemberg "Challenges of Reservoir Management - Meeting Environmental and Social Requirements"; 814/15) mein Forschungsprojekt erst ermöglicht wurde. Außerdem stellte die International Max Planck Research School of Organismal Biology ein Reisestipendium bereit. Vielen Dank auch an das gesamt CHARM Team für die Zusammenarbeit und den Austausch während der letzten 4.5 Jahre und im speziellen an Jorge Encinas für seine Hilfe bei vielen Feldkampagnen.

Ich danke den Mitarbeitern des Cawthron Institute in Nelson, Neuseeland für inspirierende zwei Monate und eine tolle Zusammenarbeit. Insbesondere danke ich Susie Wood, Jonathan Puddick und Konstanze Steiner, die mich beim Schreiben und im Labor unterstützt haben. Weiterer Dank gilt Julia Kleinteich, die mein Vorbild in der Forschung ist. Danke für deine Expertise in Sachen Cyanos und NGS Methodik und für das Korrekturlesen der Einleitung. Außerdem möchte ich der International Max-Planck Research School of Organismal Biology danken. Durch meine Mitgliedschaft habe ich viele inspirierende Persönlichkeiten getroffen, an interessanten Workshops teilgenommen und einige gute Freunde gefunden.

Vielen vielen vielen Dank an die gesamte AG Dietrich! Dank euch ist Konstanz schnell zu meinem neuen zu Hause geworden. Allen voran Heinke Bastek und Stefan Altaner, die mich von Tag 1 in der AG so herzlich aufgenommen haben! Danke Heinke für deine geduldige Unterstützung in den molekularbiologischen Methoden. Danke Stef, für deine geballte Microcystin Kompetenz, für das Korrekturlesen und die Formatierungshilfe in den Endzügen der Dissertation – und natürlich für zahlreiche Laufrunden! Ein großes Dankeschön an die aktuelle AG Besetzung: Eva, Feli, Goran, Nadja und Regina. Es gab Zucker, Sport, ein offenes Ohr, Computer-Nothilfe und Sekt wann immer ich eines davon brauchte. Ihr seid die Besten! Danke auch an Karin, Pia und Sascha für die Organisation und Unterstützung im

III

DANKSAGUNG

Dokumentenkrieg und an Sabine für ihre stetige Hilfe im Labor, besonders in Zeiten des Western-Blots. Viele der Daten sind durch die Betreuung von Studenten und deren Abschlussarbeiten entstanden. Besonders Helena, Isabel und Japhet haben mich längere Zeit begleitet und tolle Arbeit geleistet. Danke dafür!

Mein herzlichster und größter Dank geht an meine Freunde und Familie. Danke Laura dafür, dass du mich seit 30 Jahren begleitest und in allem unterstützt was ich so tue. Ebenso danke ich der Flerella-Baylin-Kombi, für eine lange gemeinsame Zeit und dafür, dass ihr die weiten Wege auf euch nehmt um mich zu besuchen, egal wo es mich hin verschlägt. Und danke Aylin, für deine Korrekturarbeit. Danke Mama und Papa, dass ihr mich immer unterstützt habt, finanziell und emotional. Ihr habt mir das Studium erst ermöglicht und mich immer meinen Weg gehen lassen, auch wenn das bisweilen sicher nicht einfach war. Mein allergrößter Dank geht an Simon, mein Herz, mein bester Freund. Danke, dass du seit fast zehn Jahren an meiner Seite bist, mich in allem unterstützt, aufbaust und mehr an mich glaubst als ich es selbst tue. Danke für deine unendliche Geduld und dafür, dass du mich frei sein lässt und zu dem Menschen machst der ich sein möchte.

TABLE OF CONTENTS

II. TABLE OF CONTENTS

I. DANKSAGUNG ...... III

II. TABLE OF CONTENTS ...... V

III. RECORD OF ACHIEVEMENTS ...... IX

IV. SUMMARY ...... XI

V. DEUTSCHE ZUSAMMENFASSUNG ...... XIII

VI. TABLE OF FIGURES...... XV

VII. TABLE OF TABLES ...... XVI

VIII. TABLE OF SUPPLEMENTARY FIGURES ...... XVI

IX. TABLE OF SUPPLEMENTARY TABLES...... XVII

X. ABBREVIATIONS ...... XIX

1. GENERAL INTRODUCTION ...... 1

1.1. Cyanobacteria ...... 1

1.2. Diversity and taxonomy of cyanobacteria ...... 2 1.2.1. Microcystis ...... 3 1.2.2. Dolichospermum ...... 3

1.3. Cyanobacterial harmful algal blooms ...... 4 1.3.1. Rise of cyanobacterial blooms: Global warming and eutrophication ...... 5 1.3.2. Traits that facilitate bloom formation ...... 8

1.4. Cyanobacterial toxins ...... 10 1.4.1. Microcystins: Occurrence, structure and toxicity ...... 11 1.4.2. Anabaenopeptins: Occurrence, structure and toxicity ...... 12 1.4.3. Other cyanotoxins ...... 13 1.4.4. Biosynthesis, genetics and evolution of cyanotoxins ...... 14 1.4.5. Regulatory factors and biological functions of cyanotoxins ...... 15

1.5. Bloom dynamics ...... 18 1.5.1. Short-term species and toxin dynamics ...... 18 1.5.2. Consequences of cyanobacterial blooms ...... 21

1.6. Aim of the thesis ...... 25

2. MANUSCRIPT 1 ...... 27

TABLE OF CONTENTS

2.1. Abstract ...... 27

2.2. Introduction ...... 28

2.3. Material & Methods ...... 30 2.3.1. Study site ...... 30 2.3.2. Sediment sampling ...... 31 2.3.3. DNA extraction and inhibition test ...... 32 2.3.4. Quantitative PCR for Microcystis and mcyE enumeration in surface sediments ...... 32 2.3.5. Droplet digital PCR for mcyE enumeration in sediment cores ...... 33 2.3.6. High Throughput Sequencing ...... 33 2.3.7. Graphics and data analysis of High Throughput-Sequences ...... 34 2.3.8. Toxin analysis ...... 35 2.3.9. Sediment chronology of mid-lake sediment deep core ...... 35

2.4. Results ...... 36 2.4.1. Physicochemical parameters ...... 36 2.4.2. Surface sediments ...... 36 2.4.3. Sediment cores ...... 39 2.4.4. Sediment chronology ...... 41

2.5. Discussion ...... 41 2.5.1. Bacterial community and Microcystis abundance in surface lake sediment ..41 2.5.2. Historic Microcystis populations ...... 43 2.5.3. Why is Microcystis absent in the surface sediments? ...... 43

2.6. Acknowledgments ...... 45

2.7. Supplemental Information ...... 46

3. MANUSCRIPT 2 ...... 48

3.1. Abstract ...... 48

3.2. Introduction ...... 49

3.3. Material & Methods ...... 50 3.3.1. Test organisms and culture condition...... 50 3.3.2. Chemostat experiments ...... 51 3.3.3. Cell counting ...... 52 3.3.4. Growth curve analysis ...... 52 3.3.5. Statistics and data analysis ...... 53 3.3.6. Nutrient analysis ...... 54 3.3.7. Protein extraction and Western Blot ...... 54

TABLE OF CONTENTS

3.4. Results ...... 55 3.4.1. Mono-cultures ...... 55 3.4.2. Co-cultures ...... 56 3.4.3. Western Blot ...... 57 3.4.4. Nutrient concentrations ...... 57

3.5. Discussion ...... 58 3.5.1. Growth competition between Dolichospermum and Chlorella ...... 59

3.6. Acknowledgements ...... 61

3.7. Supplemental Information ...... 62

4. MANUSCRIPT 3 ...... 64

4.1. Abstract ...... 64

4.2. Introduction ...... 65

4.3. Material & Methods ...... 66 4.3.1. Sampling location ...... 66 4.3.2. Field sampling ...... 67 4.3.3. Dolichospermum quantification ...... 67 4.3.4. Genomic DNA extraction ...... 67 4.3.5. High Throughput Sequencing ...... 68 4.3.6. Detection of toxin genes ...... 69 4.3.7. Toxin quantification ...... 69 4.3.8. Fungal infection intensity ...... 71 4.3.9. Physical and chemical parameters ...... 71 4.3.10. Data analysis ...... 71

4.4. Results ...... 72 4.4.1. Dolichospermum quantification and correlation to abiotic parameters ...... 72 4.4.2. Bloom phylotype dynamics ...... 74 4.4.3. Toxin and toxins genes ...... 76 4.4.4. Chytrid parasitism ...... 77

4.5. Discussion ...... 79 4.5.1. Morphotype, phylotype and toxin dynamics ...... 79 4.5.2. Cyanobacterial toxins as potential defence compounds against fungal infections ...... 80 4.5.3. Morphotype dynamics in the context of chytrid infection ...... 81 4.5.4. Selective parasitism as a potential driver of seasonal bloom dynamics ...... 82

4.6. Acknowledgements ...... 83

VII

TABLE OF CONTENTS

4.7. Supplemental Information ...... 83

5. MANUSCRIPT 4 ...... 91

5.1. Abstract ...... 91

5.2. Introduction ...... 92

5.3. Sediments ...... 92

5.4. Biostabilization of fine sediments ...... 94

5.5. Cyanobacteria ...... 95 5.5.1. Which species form blooms? ...... 96 5.5.2. Management tools ...... 96

5.6. Greenhouse gas emission ...... 97

5.7. Societal implications ...... 98

5.8. Occupancy of water reservoirs and their direct and indirect implications ..99

5.9. Review tool for management ...... 100

5.10. Summary and conclusions ...... 100

5.11. Availability of data ...... 101

5.12. Acknowledgements ...... 101

6. GENERAL DISCUSSION ...... 102

6.1. Using lake sediment to reconstruct cyanobacterial bloom history (Manuscript 1) ...... 102

6.2. Fungal parasitism and its influence on bloom toxicity (Manuscript 3) ...... 104

6.3. Nitrogen load control as bloom mitigation strategy (Manuscript 2 & 4) ... 106

6.4. Risk assessment and short-term bloom dynamics ...... 108

7. CONCLUSIONS AND OUTLOOK ...... 112

8. RECORD OF CONTRIBUTION ...... 113

9. REFERENCES ...... 115

VIII

RECORD OF ACHIEVEMENTS

III. RECORD OF ACHIEVEMENTS

Journal publications

"Is a central sediment sample sufficient? Exploring spatial and temporal microbial diversity in a small lake"; B. Weisbrod, S.A. Wood, K. Steiner, R. Whyte-Wilding, J. Puddick, O. Laroche, D.R. Dietrich. Toxins 2020, 12(9), 580; doi: 10.3390/toxins12090580.

"Can toxin warfare against fungal parasitism influence seasonal Dolichospermum bloom dynamics? – A field observation"; B. Weisbrod, E. Riehle, M. Helmer, D. Martin- Creuzburg, D.R. Dietrich. Harmful Algae 2020, 99, 101915; doi:10.1016/j.hal.2020.101915.

"Competition for nitrogen: N2 fixation allows cyanobacteria to outcompete green algae"; B. Weisbrod, J. Breiholz, D. Martin-Creuzburg, D.R. Dietrich. Manuscript ready for submission.

"Interdisciplinary Water Reservoir Management"; M. Daus, K. Koberger, K. Koca, F. Beckers, J. Encinas-Fernandes, B. Weisbrod, D.R. Dietrich, S. Gerbersdorf, R. Glaser, S. Haun, H. Hofmann, D. Martin-Creuzburg, F. Peeters, S. Wieprecht. Submitted to Water Resources Management.

Talks

"Spatial and temporal distribution of Microcystis in lake sediment"; 11th International conference of toxic cyanobacteria (ICTC), Krakow, Poland 2019.

"Spatial distribution of Microcystis and microcystins in lake sediment"; 18th International conference of harmful algae (ICHA), Nantes, France 2018.

"Prozesse im Kleinen - Auswirkungen im Großen"; Colloquium Stauraummanagement Freiburg, Germany 2017.

RECORD OF ACHIEVEMENTS

Poster presentations

"Anabaena bloom dynamics in a hydropower reservoir"; B. Weisbrod, M. Helmer, D. Martin-Creuzburg, D.R. Dietrich, 11th International conference of toxic cyanobacteria (ICTC), Krakow, Poland 2019.

"Toxin warfare: Does fungal parasitism influence success and decline of cyanobacterial blooms?"; B. Weisbrod, E. Riehle, M. Helmer, D. Martin-Creuzburg, D.R. Dietrich, Powerful women in science, Stuttgart, Germany 2020.

Public outreach

Protagonist TV documentary "Betrifft – Klimawandel: Wie verändert sich der Südwesten", Südwestrundfunk 2017.

Scholarship and funding

Funding was provided by the by the Ministry of Science, Research and the Arts of the Federal State Baden-Württemberg, Germany (Water Research Network project ‘Challenges of Reservoir Management - Meeting Environmental and Social Requirements’; 814/15). The grant included funding for B. Weisbrod, consumables and experimental and travel costs.

Further travel funding was provided by the International Max Planck Research School (IMPRS) of Organismal Biology. B. Weisbrod was a member of the IMPRS during the whole time of her PhD work.

SUMMARY

IV. SUMMARY

Cyanobacteria are photosynthetic prokaryotes occurring worldwide in almost every ecosystem. They are most common in marine and freshwater habitats where they are an integral part of the phytoplankton community, and account for a large share of the global primary production. Under certain environmental conditions, cyanobacteria can form dense agglomerations in the water body, so-called cyanobacterial blooms, which are characterized by sudden growth and high cell densities. Cyanobacterial bloom incidents have been recognized worldwide with a significant increase over the last decades. Scientists agree that anthropogenic eutrophication of global surface waters is the primary cause for the proliferation of cyanobacteria, but also that climatic changes may play a significant role. Cyanobacteria can produce secondary metabolites known as cyanotoxins, which are toxic to humans and wildlife. Cyanobacterial blooms, especially if formed by cyanotoxin-producing species, can have harmful effects on human, wildlife and ecosystem health and can cause tremendous financial costs. The present work deals with several aspects regarding the temporal and spatial dynamics of cyanobacterial blooms in lakes and artificial water reservoirs. The few studies providing quantitative evidence for the historical increase of cyanobacterial blooms in lakes employ single sediment cores as natural bloom archive. The present work verified the sampling procedure of these paleolimnological studies by investigating spatial and historical distribution of bloom proxies in sediments of Lake Rotorua, a small eutrophic lake in New Zealand. The results from bacterial community (16S rRNA metabarcoding) sequencing, cyanotoxin analysis and radionuclides measurements showed that potentially toxic blooms increased since the 1950’s. Furthermore, the single core from a central site captured dominant microbial communities of the whole lake. Nutrient load reduction is the most promising bloom mitigation strategy. While the efficiency of phosphorus (P) load reduction is widely accepted, there is an uncertainty regarding the efficiency of nitrogen (N) reduction. Some cyanobacteria are diazotroph, meaning they can fix atmospheric nitrogen (N2). Therefore, N load reduction is suspected to cause a shift from green algae to diazotrophic cyanobacteria. Competitive growth experiments were performed, using the N2-fixing cyanobacterium Dolichospermum sp. and the green alga Chlorella, under N-limiting and N-saturating condition. Results indicated that Dolichospermum can fully compensate N limitation by N2 fixation. It was also shown that Dolichospermum competitively excluded Chlorella under N-limiting condition. N load reduction should therefore be considered carefully on a case-by-case basis and be combined with simultaneous P load reduction.

SUMMARY

Cyanotoxin concentrations of blooms can differ tremendously within days or even hours. Understanding these dynamics and their causes is of great importance for the cyanotoxin risk assessment of recreational and drinking waters. This study monitored cyanotoxin and species dynamics of a natural Dolichospermum bloom and its abiotic and abiotic triggers. Results indicated that concentrations of the toxins anabaenopeptins and microcystins were correlated to varying abundances of different Dolichospermum phylotypes, rather than environmental parameters. Furthermore, selective fungal parasite pressure was identified as a potential factor steering Dolichospermum phenotype and toxin dynamics. In summary, the presented work helps to understand and reverse environmental changes promoting cyanobacterial blooms; it improves future bloom monitoring and risk assessment in bathing waters and drinking water supplies; and it provides an information tool for water managements, which helps to decide upon appropriate bloom mitigation strategies.

XII

DEUTSCHE ZUSAMMENFASSUNG

V. DEUTSCHE ZUSAMMENFASSUNG

Cyanobakterien sind fotosynthetische Prokaryoten die in nahezu allen Ökosystemen weltweit vorkommen. Als wichtiger Bestandteil des Phytoplanktons sind sie vor allem in marinen Lebensräumen, sowie Süßwasser zu finden und sind für einen großen Teil der globalen Primärproduktion verantwortlich. Unter bestimmten Umweltbedingungen können Cyanobakterien dichte Ansammlungen in Gewässern bilden. Diese so genannten "Cyanobakterienblüten" zeichnen sich durch ein schnelles Wachstum und eine hohe Zelldichte aus. Zahlreiche Publikationen berichten über einen signifikanten Anstieg von Cyanobakterienblüten weltweit in den vergangenen Jahrzehnten. Es ist wissenschaftlicher Konsens, dass die von Menschen verursachte Eutrophierung von Oberflächengewässern Hauptgrund für diesen Anstieg ist. Außerdem gibt es Hinweise darauf, dass der Klimawandel einen fördernden Einfluss auf die Bildung von Blüten haben könnte. Manche Cyanobakterien produzieren Sekundärmetabolite, so genannte Cyanotoxine die toxisch für Vertebraten und einige Invertebraten sind. Cyanbakterienblüten, vor allem solche in denen Toxine gebildet werden, können schädliche Folgen für Mensch und Umwelt haben und immensen finanziellen Schaden verursachen. Die vorliegende Arbeit beschäftigt sich mit unterschiedlichen Aspekten der zeitlichen und räumlichen Dynamik von Cyanobakterienblüten in natürlichen Gewässern und künstlichen Reservoiren. Es gibt nur wenige Publikationen, die den historischen Anstieg von Cyanobakterienblüten mit quantitativen Daten belegen. Diese Studien basieren auf dem paleolimnologischen Ansatz: einzelne Sedimentkerne werden als natürliches Archiv für das Auftreten von Blüten genutzt. In der vorliegenden Arbeit wurde das in diesen Studien verwendete Verfahren zur Probenentnahme dieser Studien verifiziert, indem die räumliche und historische Verteilung von verschiedenen Blütenproxies in Sedimenten von Lake Rotorua, ein eutropher See in Neuseeland, analysiert wurden. Mittels Sequenzierung der Bakteriengemeinschaft (16S rRNA metabarcoding), Cyanotoxinanalytik (UPLC-MS/MS) und der Analyse von Radionukliden konnte gezeigt werden, dass das Vorkommen potentiell toxischer Microcystis Blüten seit den 1950ern ansteigt. Es stellte sich heraus, dass ein einzelner Sedimentkern aus der Seemitte die allgemeine Bakteriengemeinschaft im See widerspiegeln kann. Als die effizienteste und nachhaltigste Methode um das Aufkommen von Cyanobakterien in einem Gewässer langfristig zu verringern, gilt die Reduzierung von Nährstoffeinträgen. Während die Wirksamkeit von Phosphor(P)-reduzierung bereits vielfach gezeigt wurde, ist die Wirksamkeit einer Reduzierung von Stickstoff (N) umstritten. Da manche

Cyanobakterien in der Lage sind Stickstoff aus der Atmosphäre (N2) zu fixieren wird vermutet,

XIII

DEUTSCHE ZUSAMMENFASSUNG dass eine N-Reduzierung in Gewässern, eher zu einer Verschiebung von Grünalgen zu N2- fixierenden Cyanobakterien führt, statt die absolute Biomasse von Cyanobakterien zu reduzieren. Um dieser Frage nachzugehen, wurde in der vorliegenden Arbeit kompetitive

Wachstumsexperimente mit dem N2-fixierenden Cyanobakterium Dolichospermum und der Grünalge Chlorella, unter N-limitierenden und N-gesättigten Bedingungen durchgeführt.

Dolichospermum konnte die N-Limitierung mittels N2-Fixierung kompensieren und als konkurrenzstärkere Art Chlorella verdrängen. Die Reduzierung von Stickstoff kann daher nur bedingt empfohlen werden und sollte in Einzelfällen und in Kombination mit P-Reduzierung erwogen werden. Toxinkonzentrationen in Cyanobakterienblüten können innerhalb von Tagen oder sogar Stunden enorm variieren. Um eine sichere Risikoabschätzung von Cyanbakterienblüten in Badegewässern und Trinkwasserreservoiren zu gewährleisten, ist es wichtig, diese zeitliche Dynamik und ihre Ursachen zu verstehen. Im Rahmen dieser Arbeit wurde eine Dolichospermum-Blüte im saisonalen Verlauf überwacht. Es wurde gezeigt, dass die Konzentration der Cyanotoxine Microcystin und Anabaenopeptin vor allem mit der Häufigkeit verschiedener Dolichospermum Phylotypen korreliert und weniger mit Umweltfaktoren. Darüber hinaus wurde parasitärer Pilzbefall von Dolichospermum als möglicher biologischer Einflussfaktor auf die Toxin- und Artendynamik identifiziert. Zusammenfassend hilft die vorliegende Arbeit durch die erlangten Ergebnisse und die entwickelten Methoden, blütenverursachende Umweltveränderung besser zu verstehen und ihnen entgegenzuwirken; sie ermöglicht eine verbesserte Überwachung und Risikoabschätzung von Blüten in Badegewässern und Trinkwasserreservoiren; und sie stellt ein Informationswerkzeug für Betreiber von Wasserreservoiren bereit.

XIV

VI.

VII. TABLE OF FIGURES

Figure 1-1 Cyanobacterial bloom at the Altmühlsee (Bavaria, Germany)...... 5

Figure 1-2 Summary of abiotic and biotic factors controlling cyanobacterial boom dynamics...... 20

Figure 2-1 Map of Lake Rotorua (Kaikoura, New Zealand)...... 31

Figure 2-2 Abundance of Microcystis specific, (A) 16S rRNA gene, and (B) mcyE gene in surface sediment samples (S01-S12)...... 36

Figure 2-3 Venn diagram...... 37

Figure 2-4 Bacterial community of surface sediment samples (S01-S12) based on 16S rRNA sequences...... 38

Figure 2-5. Two-dimensional non-metric multidimensional scaling (NMDS) ordination of microbial communities in the surface sediment samples (S01-S12)...... 38

Figure 2-6 Bacterial community of the sediment core layers (cm) from the mid-lake (C1). ....39

Figure 2-7 Two-dimensional non-metric multidimensional scaling (NMDS) ordination of bacterial community of sediment cores...... 40

Figure 2-8 Radionuclide profile of 137Cs and 210Pb of the sediment core from the mid-lake (C1)...... 41

Figure 3-1 Growth curves...... 55

Figure 3-2 Reaction norms of nitrogen-limited (N-) and nitrogen-saturated (N+) growth expressed as log-transformed endpoints of the chemostat experiments...... 56

Figure 3-3 Western-blot analysis of nitrogenase (NifH, ~35 kDa)...... 57

Figure 3-4 Nutrients measured in chemostat experiments...... 58

Figure 4-1 Dolichospermum morphotypes and nutrient concentrations in the Schwarzenbach reservoir during the bloom season 2018...... 73

TABLE OF TABLES

Figure 4-2 The two different Dolichospermum morphotypes found in the Schwarzenbach reservoir during the bloom season 2018...... 73

Figure 4-3 Bacterial community composition in the Schwarzenbach reservoir during the bloom season 2018...... 75

Figure 4-4 Quantification of cell-bound cyanobacterial toxins from the Schwarzenbach reservoir during the bloom season 2018...... 76

Figure 4-5 Fungal community composition from the Schwarzenbach reservoir during the bloom season 2018...... 77

Figure 4-6 Chytrid (R. crassum) life cycle in the Schwarzenbach reservoir...... 78

Figure 4-7 Asymmetric infection of straight Dolichospermum morphotype by the chytrid R. crassum...... 79

Figure 6-1 Seasonal (winter through autumn) biomass patterns of phytoplankton in a eutrophic (left) and an oligotrophic (right) water body...... 105

VIII. TABLE OF TABLES

Table 1-1 Examples of cyanotoxin-producing genera...... 2

Table 1-2 Trophic state index of lakes...... 7

Table 2-1 Sediment samples taken in this study...... 31

Table 3-1 Initial cell densities in mono- and co-culture chemostat experiments...... 51

Table 6-1 World Health Organization guideline values for the health risk assessment of cyanobacteria in recreational water and recommended actions...... 109

IX. TABLE OF SUPPLEMENTARY FIGURES

Figure S 2-1 Venn diagram...... 46

Figure S 2-2 Bacterial community of the sediment core layers (cm) from the mid-lake (C2). 46

XVI

TABLE OF SUPPLEMENTARY TABLES

Figure S 2-3 Biofilm growing on the surface sediment of Lake Rotorua ...... 47

Figure S 3-1 Nutrient ratios in chemostat experiments...... 62

Figure S 4-1 Total phosphorous (TP), total nitrogen (TN) and TN to TP ratio in the Schwarzenbach reservoir from 1990 to 2018 and the observed periods with cyanobacterial blooms...... 88

Figure S 4-2 Mean daily precipitation (blue) (mm ≡ L m-2), water temperature (red) and Chlorophyll a concentrations (green) at the Schwarzenbach reservoir 2018...... 88

Figure S 4-3 Particulate bound P (PPart) and N (NPart) concentrations in water surface samples of the Schwarzenbach reservoir in 2018...... 89

Figure S 4-4 Infection of Dolichospermum filaments: Single (A) and multiple (B) attached zoospores...... 89

Figure S 4-5 Water depth (solid line) daily water level fluctuation (dashed line) in the Schwarzenbach reservoir 2018...... 89

Figure S 4-6 Results of the unpaired Student’s t-test of mean total microcystin (MC) and anbaenopeptin (AP) concentrations between the first and the second bloom phase; * p < 0.05...... 90

X. TABLE OF SUPPLEMENTARY TABLES

Table S 2-1 Abiotic parameters of surface sediment sampling sites...... 47

Table S 2-2 Recovery rates of microcystin (MC) congeners...... 47

Table S 3-1 Endpoints of Dolichospermum and Chlorella mono-cultures...... 62

Table S 3-2 Endpoints of Dolichospermum and Chlorella co-cultures...... 63

Table S 4-1 Primer pairs used in the study...... 83

Table S 4-2 Spearman’s rank correlation coefficients between abiotic and biotic factors...... 84

Table S 4-3 Nucleotide sequences (407 bp) of the three Dolichospermum ASVs...... 84

XVII

TABLE OF SUPPLEMENTARY TABLES

Table S 4-4 Microcystin (MC) congener concentrations in analytes determined using via UPLC- MS/MS...... 85

Table S 4-5 Alignments of chytrid protein phosphatase (PP) 1 and 2A with human and mouse PP1 and 2A...... 86

XVIII

ABBREVIATIONS

XI. ABBREVIATIONS

°C degrees Celsius µ micro (10‐6) or growth rate 3‐amino‐9‐methoxy‐2,6,8 Adda trimethyl‐10‐phenyldeca‐4,6 dienoic acid ANOVA analysis of variance anaC anatoxin synthetase gene C ASVs amplicon sequence variants BLAST basic local alignment search tool bp base pairs BSA bovine serum albumin cHAB cyanobacterial harmful algal bloom CRS constant rate of supply Da Dalton(s) ddPCR droplet digital PCR DEU Deutschland DIN dissolved inorganic nitrogen DNA deoxyribonucleic acid dNTPs deoxyribonucleotides dt doubling time EDTA ethylenediaminetetraacetic acid ELISA Enzyme‐linked immunosorbant assay EU European Union g gram(s) Gt giga (109) tonne h hour(s)

H2O water HPLC highperformance liquid chromatography ITS intergenic spacer L liter LOD limit of detection log logarithm m milli (10‐3) or meter(s) M molar or moles per liter MC microcystin mcy microcystin gene cluster mcyE microcystin synthetase gene E MeOH methanol mg milligram

MgCl2 magnesium chloride min minute(s) MS mass spectrometry

XIX

ABBREVIATIONS

N nitrogen n sample size or nano (10‐9) N- nitrogen-limiting N+ nitrogen-saturating NCBI National Center for Biotechnology Information nd not detected NMDS non-metric-multidimensional-scaling

NPart organic particulate bound nitrogen NRPS non‐ribosomal peptide synthetase NZ New Zealand P phosphorus p pico (10-12) PCR polymerase chain reaction PERMANOVA permutational analysis of variance pH hydrogen ion concentration PKS polyketide synthase PP protein phosphatase

PPart particulate bound phosphorus ppm parts per million qPCR quantitative PCR R arginine RNA ribonucleic acid rRNA ribosomal ribonucleic acid RuBisCO Ribulose-1,5-bisphosphat-carboxylase/-oxygenase s second(s) SD standard deviation SDS Sodium dodecylsulfate ser serine SRP soluble reactive phosphorus TAE Tris base, acetic acid and EDTA thr threonine Tm annealing temperature TN total nitrogen TP total phosphorus UBA Umweltbundesamt (German Environment Agency) U units ultra performance liquid chromatography-tandem UPLC-MS/MS mass spectrometry UV ultra violet v/v volume per volume ratio W tryptophan w/v weight per volume ratio WHO World Health Organization

GENERAL INTRODUCTION

1. GENERAL INTRODUCTION

1.1. Cyanobacteria

Cyanobacteria are a phylum of gram negative prokaryotes which occur in almost every biotope and climate zone on this planet. They are the only oxygenic photosynthetic prokaryotes, meaning they can convert energy from sunlight to chemical energy by reducing water to fix CO2 for the synthesis of sugar under the emission of O2. Fossil records of cyanobacteria in form of stromatolites date back their origin around 3 to 3.5 billion years ago (Nutman et al., 2016; Planavsky et al., 2014; Schopf, 2002). Their ancestors were the first organisms to develop oxygenic photosynthesis. This ultimately led to the oxygenation of the atmosphere around 2.3 billion years ago (Schirrmeister et al., 2015), and thus provided the foundation for more complex life forms on earth.

Cyanobacteria are true cosmopolitans and belong to the most abundant organisms on earth with an estimate of a thousand million tons (1015 g) of wet weight biomass (Garcia-Pichel et al., 2003). Thanks to their long evolutionary history, they have developed adaptions to some of the harshest environments on this planet, including the polar and alpine regions, hot springs and deserts (Jungblut et al., 2010; Kleinteich et al., 2014; Lacap-Bugler et al., 2017; Ward and Castenholz, 2002). Cyanobacteria can occur in terrestrial environments like moist soils, lithic substrate, ice and even as endosymbionts in plants or fungi (Lacap-Bugler et al., 2017; Rai et al., 2000). However, the majority of cyanobacteria occurs in aquatic habitats, either as benthic or pelagic forms (Garcia-Pichel et al., 2003). As an integral part of the marine and freshwater phytoplankton, they account for a major share of the global primary production. Conservative estimates range up to 25 % of the oceans primary production, which corresponds to ~12.5 % of the global primary production (Flombaum et al., 2013).

Certain environmental factors can cause mass developments of cyanobacteria that form dense accumulations in the water body, so-called cyanobacterial blooms. These blooms are characterized by sudden growth and high cell concentrations. Some cyanobacteria produce secondary metabolites that are known to have toxic effects on other bacteria, plants, invertebrates and vertebrates, including humans (Table 1-1). Blooms, especially if formed by toxin-producing species, often cause a decrease of water quality and can be harmful for the ecosystem, wildlife, humans and economy. Even though cyanobacteria have long been present in earth history, several studies indicate that bloom occurrences have increased globally over the last decades, due to global warming and ongoing eutrophication of freshwater ecosystems (O’Neil et al., 2012; Paerl and Huisman, 2009).

GENERAL INTRODUCTION

Table 1-1 Examples of cyanotoxin-producing genera.

Genera (examples) Cyanotoxin Class Primary target organ Toxic effect Mode of action Variants Microcystis, Planktothrix, Microcystins 271 Dolichospermum, Nostoc Liver & kidney damage, Inhibition of ser/thr Liver tumour promotion specific PPs Nodularia Nodularins Cyclic peptides ≥10 inhibition of Dolichospermum, Anabaenopeptins Liver Liver & kidney damage carboxypeptidase A & 96 Aphanizomenon and ser/thr specific PP1

Raphidiopsis, Damage of multiple Inhibition of protein‐ and Dolichospermum, Cylindrospermopsins Liver & kidney organs, gastroenteritis, glutathione‐ synthesis, 5 Aphanizomenon genotoxicity DNA damage

Lyngbya, Schizothrix, Aplysiatoxins Skin Tumour promotion 2 Oscillatoria Inflammatory agent, protein kinase C aktivator Lyngbya Lyngbyatoxin-a Skin Tumour promotion 1 Alkaloids

Cylindrospermopsis, Paralysis, respiratory + Dolichospermum, Lyngbya; Saxitoxins Nervous sytem Blockage of Na - 57 failure channels also Dinoflagellates

Dolichospermum, Aphanizomenon, Binding to nicotinic Anatoxin-a Nervous sytem Respiratory failure 2 Microcystis, Planktothrix, acetylcholin receptor Raphidiopsis, Oscillatoria

Alkaloid Salivation and Inhibits acetylcholine Dolichospermum Anatoxin-a(S) Nervous sytem 1 (organophosphate) respiratory failure esterases

BMAA (beta - Amino acid Neurodegenerative Binding to Glu‐receptors, Many Nervous system Methylamino-L-alanine) (nonproteinogenic) disease protein integration

Lipopolysaccharide Skin irritation, fever, All LPS Skin, Mucosa Inflammatory (Cell wall component) gastrointestinal upset

1.2. Diversity and taxonomy of cyanobacteria

Cyanobacteria show a high taxonomic diversity as well as diverse morphology and organization structures. Traditionally, their taxonomic classification was based on morphological features, like cell dimension, shape, type of branching and color (Palinska and Surosz, 2014). This classification system has been challenged with the rise of molecular sequencing techniques, as many morphological features used for classification turned out to be polyphyletic in nature (Dvorak et al., 2017). Until today, the taxonomy system is continuously being revised and modified (Komárek, 2016, 2010). The discussion about their taxonomy is further complicated by the fact that there is, in general, no consensus regarding the prokaryotic species concept (Gevers et al., 2005). Modern molecular taxonomy of cyanobacteria, similar to other prokaryotes, is based on DNA sequences of different marker genes, most frequently used is the gene of the small subunit of the ribosomal RNA (16S rRNA) (Rajendhran and Gunasekaran, 2011). A prokaryotic species is considered to be a group of strains that are characterized by >97 % of 16S rRNA gene-sequence similarity (Vandamme et al., 1996). But also this concept is still under revision as the 16S rRNA comes to its limits to

GENERAL INTRODUCTION define the smallest taxonomic category of cyanobacteria (Palinska and Surosz, 2014). Based on molecular taxonomy, 4,734 species of cyanobacteria have been described (Guiry and Guiry, 2020) but estimates expect a total number of 8,000 species (Guiry, 2012).

Cyanobacteria occur as single cells (unicellular) or filaments (also known as trichomes). Filaments are either straight or coiled and some genera show a branching morphology. Filaments may consist of undifferentiated cells (homocytous) or differentiated cells (heterocytous). Both, single cells and filaments, are able to form colonies and can be found floating freely in the aquatic habitats, dispersed along the photic zone of the water column, but are also found as benthic mats.

1.2.1. Microcystis

The unicellular Microcystis is the most common bloom-forming cyanobacterial genus, occurring worldwide in water temperatures ranging from 12 to 30°C, from little ponds to large rivers, lakes and in the estuarine environment (Al-Tebrineh et al., 2012; Chorus et al., 2000; Visser et al., 2005). Species of this genus can form thick surface blooms often producing cyanobacterial toxins including microcystins (MCs), nodularins, anatoxin-a, aeruginosin, β‐ methylamino‐L‐alanine (BMAA), cyanopeptolin (Fristachi et al., 2008; Paerl and Otten, 2013). Almost all Microcystis blooms found in nature are toxic (Carmichael, 1995). Due to its relevance, Microcystis is also the best studied genus of the cyanobacteria, serving as model organism in numerous publications. Most commonly described and mentioned is the species Microcystis aeruginosa (M. aeruginosa). Microcystis usually occurs as free floating coccal cells, unicellular or as colonies embedded in mucilage matrix, but can also be found attached to substrate. Microcystis blooms are mainly found in eutrophic, nitrogen rich water (Dolman et al., 2012).

1.2.2. Dolichospermum

Dolichospermum (previously known as Anabaena) belongs to the order of Nostocales and is one of the most frequently occurring, bloom-forming, toxic and filamentous cyanobacteria genera. Filaments of Dolichospermum possess a high morphological diversity and can be found in coiled as well as straight forms (Komarek and Zapomelova, 2008, 2007). Furthermore, Dolichospermum has the ability to differentiate vegetative cells into specialized cells known as akinetes and heterocysts (Kumar et al., 2010). Akinetes are thick-walled resting cells that can survive adverse environmental condition like desiccation and cold. Heterocysts

GENERAL INTRODUCTION are non-photosynthetic cells that are able to fix atmospheric nitrogen (N2) (Thiel, 2005). Due to this additional N source, Dolichospermum can grow in N poorer waters compared to non-N2 fixing cyanobacteria, e.g. Microcystis (Dolman et al., 2012). The main bloom-forming species typically reported in the scientific literature are Dolichospermum circinalis (D. circinalis), D. flos- aquae and D. lemmermannii (O’Neil et al., 2012). Reported toxins produced by this genus are MCs, anabaenopeptins, anatoxin-a and anatoxin-a(S), cylindrospermopsins and saxitoxins (Harms et al., 2016; Sivonen, 2009).

1.3. Cyanobacterial harmful algal blooms

Cyanobacteria can form excessive accumulation of cells known as cyanobacterial blooms (also known as cyanobacterial harmful algal blooms (cHABs); Table 1-1). These blooms are marked as surface scums and visible discoloration of the water and are often accompanied by a putrid stench (Figure 1-1). Common bloom-forming genera include Microcystis, Dolichospermum, Aphanizomenon, Raphidiopsis, Nodularia, Planktothrix and Trichodesmium. Cyanobacterial blooms can grow at the surface or a specific depth in the water column, corresponding to their preferred light condition, and may occur in the marine environment, in estuaries, lakes and rivers, but also in artificial water reservoirs. Cyanobacterial blooms are considered harmful, causing a major decrease of water quality (Chorus and Bartram, 1999; Paerl and Otten, 2013). The harmful effects include taste and odor compounds produced by cyanobacteria that may interfere with the recreational function of lakes and the use of reservoirs for drinking water. Furthermore, blooms may cause an increase of water turbidity and decrease of sunlight availability in the deeper water column, thereby suppressing growth of other phytoplankton as well as aquatic macrophytes. Oxygen depletion during the microbial degradation of senescent blooms may lead to hypoxia and anoxia, causing the death of fish and benthic invertebrates. Toxins in the worst-case scenario, may lead to a complete shutdown of water supply or even intoxication of humans. blooms are often associated with high economic loss caused by increased water treatment costs, the logistics of alternative drinking water resources and the loss of tourism revenues. Environmental factors and physiological traits linked to bloom formation as well as consequences of harmful cyanobacterial blooms will be discussed in the following paragraphs.

GENERAL INTRODUCTION

Figure 1-1 Cyanobacterial bloom at the Altmühlsee (Bavaria, Germany). Photo: Luftbild Beringer.

1.3.1. Rise of cyanobacterial blooms: Global warming and eutrophication

Cyanobacteria in low amounts are usually an integral part of the phytoplankton community. The question is, what causes their excessive proliferation and bloom formation? As a matter of fact, not only cyanobacteria but also other phytoplankton species can form blooms, including dinoflagellates, diatoms and green algae (Paerl et al., 2001). Cyanobacterial blooms are not a recent phenomenon. Mammal mass mortalities in Pleistocene and Eocene lakes have been attributed to toxin-producing cyanobacterial blooms (Braun and Pfeiffer, 2002; Huisman et al., 2018). The first detailed scientific documentation of a toxin-producing cyanobacterial bloom, responsible for mass mortalities of livestock dates back to a Nature publication in 1878 (Francis, 1878). However, numerous studies state that cyanobacterial blooms have increased in global expansion and persistence over the last few decades (Huisman et al., 2018; O’Neil et al., 2012; Paerl and Huisman, 2008). Quantitative data for this comes from paleolimnological studies (Legrand et al., 2017; Taranu et al., 2015). Those studies analyze cyanobacterial bloom related proxies (e.g. cyanotoxin coding genes, cyanotoxins, resting cells) in sediment cores to reconstruct the bloom history of the respective lake or reservoir. Furthermore, they can provide information regarding changes of the trophic state of the system as well as environmental changes in the catchment area over time. These studies show that cyanobacteria have increased substantially during the 1900s in North

GENERAL INTRODUCTION

American (Taranu et al., 2015) and European (Legrand et al., 2017) lakes, and that cyanobacterial abundance has increased disproportionally relative to other phytoplankton more rapidly since 1945. There is a broad agreement that this worldwide expansion of cyanobacterial blooms is mainly promoted by the synergistic effect of two anthropogenic induced alterations: Global warming and nutrient over-enrichment (eutrophication, for definition of trophic states see Table 1-2 according to Carlson, 1977) of aquatic ecosystems (Rigosi et al., 2014). Since the onset of the industrial revolution around 1800 and especially since the 1950’s, humans have become a global geophysical force (Steffen et al., 2007; Vitousek et al., 1997). Urbanization, intensified agriculture (pasture and cropland) and industrial development caused a significant increase of key nutrient loads (N and P) in surface freshwaters of reservoirs, lakes, and rivers worldwide (Carpenter et al., 2011). Seitzinger et al. (2010) estimated that global riverine loads of N and P increased by approximately 30 % between 1970 and 2000. A multitude of studies, including the pioneering whole-lake experiment in by David Schindler (1977), show how this increased eutrophication promotes development, persistence and higher biovolumes of cyanobacterial blooms (Dolman et al., 2012; Downing et al., 2001; Harpole et al., 2011; Heisler et al., 2008; Lewis and Wurtsbaugh, 2008). Schindler demonstrated how adding a combination of P and N could quickly turn a pristine lake into a massive cyanobacterial bloom. The resulting bloom was dominated by N2-fixing Dolichospermum and Aphanizomenon (see chapter 1.3.2). Schindler concluded that reducing N loads in water would not decrease the growth of the blooms, as the involved species could compensate for the limited N, and therefore P alone should be the nutrient of concern. Other studies also associated cyanobacterial dominance primarily with P-enrichment, resulting in low nitrogen-to-phosphorus ratios (N:P ratios; Havens et al., 2003; Smith, 1983; Vrede et al.,

2009). According to this, N2-fixing taxa (Dolichospermum, Aphanizomenon, Raphidiopsis,

Nodularia, Trichodesmium) are described to dominate at low N:P ratios (<22:1), and non-N2- fixing cyanobacteria (Microcystis, Planktothrix, Oscillatoria) at N:P ratios <29:1. Green algae are expected to dominate at N:P ratios >29:1 (Paerl et al., 2001; Smith, 1983). However, the concept of N:P ratio is highly controversial (Downing et al., 2001) and still actively debated

(Schindler, 2012). More recent studies indicate that N2-fixation of diazotrophic cyanobacteria cannot compensate completely for depletion of dissolved N sources (Shatwell and Köhler, 2019; Willis et al., 2016). Consistent with this, numerous studies emphasize above average joint N and P concentrations as major trigger for higher total biovolumes of all cyanobacterial taxa, although relative biovolumes of some N2-fixing species might be higher in lakes with low N:P ratios (Dolman et al., 2012; Downing et al., 2001; Posch et al., 2012). Some consider N concentrations in lakes as even more important than P concentrations in driving cyanobacteria production (Shatwell and Köhler, 2019). The discussion regarding bloom promoting effects of

GENERAL INTRODUCTION nutrients is further complicated by the observation that cyanobacterial blooms may also occur in oligotrophic lakes or may even be sustained by oligotrophication (Ernst et al., 2009; Posch et al., 2012). Toxin-producing species forming blooms in oligotrophic systems belong to the

N2-fixing genera Dolichospermum and Aphanizomenon and to the non-N2-fixing genera Planktothrix and Gleotrichia (Callieri et al., 2014; Carey et al., 2008; Posch et al., 2012; Salmaso, 2010; Salmaso and Mosello, 2010). Very little is known about the mechanisms behind these ‘‘oligotrophic blooms’’ (Salmaso, 2000).

Table 1-2 Trophic state index of lakes. Relationships between trophic state, secchi depth, phosphate, nitrate, chlorophyll and O2. Based on Carlson, 1977.

Parameter oligotrophic mesotrophic eutrophic hypertrophic

Secchi depth (m) 5-10, max.15-20 1-2, max. 5-10 max. 2-3 <1

4-10 (0-4 = Phosphate (µg L-1) 10-35 35-100 >100 ultraoligotrophic) Nitrate and Ammonium ≤1 ≤1 >2 >2 in fall (mg N L-1) Chlorophyll annual <3.5 ≤7.0 <11 >11 mean (mg/m³)

-1 O2 (mg L ) mehr als 8 6-8 2-4

Simultaneously to eutrophication, atmospheric CO2 has risen from a preindustrial value of 275 ppm to about 380 ppm causing an increase of the average global temperate of 1.5°C (IPCC,

2018). Rising atmospheric CO2 levels are directly correlated to increased influx of CO2 in the water, enhancing the dissolved C supply for primary producers (Verspagen et al., 2014; Visser et al., 2016). Dense, surface bloom-forming cyanobacteria such as Microcystis exhibit a strong demand for inorganic carbon (C), to the extent that the CO2 supply can limit the rate of biomass production (Paerl and Huisman, 2009; Verspagen et al., 2014; Visser et al., 2016). Rising atmospheric CO2 levels are therefore expected to stimulate the proliferation of surface-dwelling cyanobacteria (Visser et al., 2016). Global warming also directly influences water surface temperatures and increases the stability of thermal stratification of surface waters (Carpenter et al., 2011). Higher water temperatures and prolonged stratification periods of surface waters were both shown to give cyanobacteria a competitive advantage over other phytoplankton such as diatoms and green algae (Jöhnk et al., 2008; Kosten et al., 2012; Paerl and Huisman, 2008). Cyanobacteria reach their maximal growth rates at relatively high temperatures; often in waters above of 25°C (Butterwick et al., 2005; Foy et al., 1976). At these elevated

GENERAL INTRODUCTION temperatures, cyanobacteria routinely outcompete eukaryotic algae (Elliott, 2010; Jöhnk et al., 2008; Paerl et al., 2011). Warmer surface temperatures also decrease the physiological costs of N2-fixation and might therefore stimulate growth and N2-fixation taxa (Brauer et al., 2013). Increased thermal stratification of the water column suppresses turbulent mixing. This is favorable for buoyant cyanobacteria because weak mixing allows them to float to the upper water layers, where they have better access to light, higher temperatures, CO2 and N2 while shading the non-buoyant phytoplankton below (Visser et al., 2016). Decreased mixing may also favor Planktothrix blooms as this allows Planktothrix to accumulate within the depth of optimal irradiance (Posch et al., 2012). Furthermore, warmer climate promotes the geographical distribution of tropical and subtropical species in more temperate regions as shown for the subtropical species Raphidiopsis raciborskii (Sinha et al., 2012; Sukenik et al., 2012; Wiedner et al., 2007), Anabaena bergii and Aphanizomenon aphanizomenoides (Mehnert et al., 2010; Stüken et al., 2006).

Despite ongoing research on bloom promoting factors, predicting their occurrence and identifying primary environmental factors remains challenging due to the complex interaction among physical, chemical, and biological processes (Figure 1-2; Paerl et al., 2001). Furthermore, bloom development might be entirely different between colony forming species like Microcystis, N2-fixing species, well mixed species (Raphidiopsis) and stratifying species (Planktothrix) (Dokulil and Teubner, 2000). Moreover, climate change scenarios predict bloom frequency and magnitude to worsen in the future (Chapra et al., 2017; O’Neil et al., 2012; Paerl and Huisman, 2009). For most surface waters, long-term, quantitative data regarding cyanobacteria assemblage changes over time do not exist. Expanding paleolimnological studies to a bigger variety of lakes and reservoirs might help to identify bloom triggers and to improve prediction models and management strategies of harmful cyanobacterial blooms.

1.3.2. Traits that facilitate bloom formation

During their evolutionary history, cyanobacteria have developed an impressive set of physiological traits, which differ in presence and manifestation among the various species and strains (Dokulil and Teubner, 2000). These traits help them to survive in diverse environments, but also to compete with and dominate over eukaryotic phytoplankton, ultimately leading to cyanobacterial bloom formation (Huisman et al., 2018).

One of the most significant traits is the capability of several genera to fix N2, which provides them with an additional N source – also known as diazotrophy (Herrero et al., 2001). While N2

GENERAL INTRODUCTION fixation is a common trait among prokaryotes, cyanobacteria are the only known photoautotrophs capable of N2 fixation. Diazotroph organisms use the enzyme nitrogenase to catalyze the conversion of N2 to ammonia (NH3). Since nitrogenase is strongly inhibited by oxygen, cyanobacteria have developed a variety of cell morphologies and metabolic strategies that allow spatial and temporal separation of N2 fixation from photosynthesis (Esteves-Ferreira et al., 2017). The fixation process and the nitrogenase enzyme itself differ among the diazotroph taxa such as unicellular Synechococcus, the non-heterocystous genera Phormidium and Trichodesmium and the heterocytous filamentous genera Dolichospermum, Fischerella, Chlorogloeopsis, Scytonema, Calothrix, Raphidiopsis, Aphanizomenon, Lyngbya, Nodularia and Gloeotrichia (Esteves-Ferreira et al., 2017). In heterocytous genera, e.g. Dolichospermum, the nitrogenase complex is confined to specialized cells called heterocysts. Heterocysts have a thick glycolipid cell walls which serve as a gas diffusion barrier to decrease the diffusion of O2 into the heterocyst, but allow a sufficient influx of N2 (Walsby, 1994). The fixation of N2 is energetically a very costly reaction (Gallon, 1992). Therefore, diazotrophic cyanobacteria require high light intensities for the fixation (Paerl, 2017; Staal et al., 2002) (Paerl, 2017; Staal et al., 2002) and thus are supposed to switch to this alternative only if dissolved N sources become depleted (Holl and Montoya, 2005). In addition to diazotrophy, cyanobacteria are capable of taking up P in access of cellular growth requirements and storing it in polyphosphate bodies for P-limiting environmental conditions (Chorus and Bartram, 1999; Paerl et al., 2001; Pettersson et al., 1993). Internally stored P is sufficient to perform 2-4 cell divisions, which corresponds to a 4-32 fold increase in biomass (Chorus and Bartram, 1999).

Like every other primary producer, cyanobacteria perform carbon fixation using the enzyme

RuBisCo. Similar to eukaryotic algae they have evolved CO2-concentrating mechanisms to overcome the low affinity of RuBisCo to CO2 (Visser et al., 2016). In cyanobacteria, five different inorganic C uptake systems have been identified, three for the uptake of bicarbonate and two for the conversion of CO2 (Visser et al., 2016). These uptake systems differ in their affinity to the respective C source as well as their flux rates. This enables cyanobacteria to respond effectively to changes in availability of inorganic C and to compensate for a 20-fold difference between the CO2 concentration in their environment and the Michaelis-Menten constant of their RuBisCo (Kaplan et al., 1991).

Cyanobacteria synthesize intracellular, hollow protein structures known as gas vesicles which act as flotation devices in aqueous environments (Walsby, 1994). Buoyant cells have a major competitive advantage, because they gain access to vertically separated resources such as nutrients, sunlight and atmospheric CO2, and may even shade non-buoyant phytoplankton deeper in the water column (Huisman et al., 2004; Mitrovic et al., 2001; Walsby, 2007).

GENERAL INTRODUCTION

Buoyancy is primarily regulated through an interplay of synthesis and respiration of carbohydrate ballast, controlling the cells sinking rate (Visser et al., 1995). Buoyant cyanobacteria can migrate between several meters or only a few centimeters within a few hours through the water column, depending on their cell or colony size (Visser et al., 1997). An especially impressive example for fine-tuned buoyancy control is the genus Planktothrix which forms stratified blooms in the metalimnion, in 10-14 m water depth (Halstvedt et al., 2007).

Like eukaryotic algae, cyanobacteria possess chlorophyll as major photopigment, but they also contain other pigments such as the phycobiliproteins, which include allophycocyanin (blue), phycocyanin (blue) and phycoerythrine (red). These pigments harvest light in the green, yellow and orange part of the spectrum (500-650 nm), which is hardly used by other phytoplankton species (Chorus and Bartram, 1999). They also cause the different coloring of the individual species. The variety of photopigments also allows cyanobacteria to adapt to a variety of light conditions and to use light more efficiently (Stal, 2015). Additionally, some cyanobacteria are capable of complementary chromatic adaptation (Mullineaux, 2001), meaning they can alter the composition of photopigments in response to the prevalent wavelengths of light in the environment (Grossman et al., 1993). Finally, some cyanobacteria form resting stages when environmental conditions become unfavorable. Those cells can resist adverse environmental conditions including temperature extremes and desiccation (Olsson-Francis et al., 2009). They survive and overwinter at the sediment surface, as conditions turn more favorable, they hatch to the water phase as a substantial inoculum for the bloom (Borges et al., 2016; Ståhl-Delbanco et al., 2003).

1.4. Cyanobacterial toxins

Cyanobacteria are known to produce a variety of secondary metabolites, some of which show harmful effects on a wide variety of organisms including bacteria, protozoans, zooplankton, fish, birds, and mammals (Chorus et al., 2000; Chorus and Bartram, 1999; Sivonen, 2009). Those compounds are summarized under the term cyanobacterial toxins (hereinafter cyanotoxins). Despite this shared terminology, cyanotoxins show a vast structural and functional diversity and are unevenly distributed among a variety of cyanobacteria genera and geographical locations (Bláha et al., 2009; Sivonen, 2009; Welker and Von Döhren, 2006). The most common and extensively studied cyanotoxins are: MCs, nodularins, saxitoxin (also known as paralytic shellfish poison), anatoxins (anatoxin-a, anatoxin-a(S), homoanatoxin), cylindrospermopsin, lipopolysaccharides (LPS), BMAA, lyngbyatoxin and aplysiatoxin.

GENERAL INTRODUCTION

Cyanotoxins can be classified by their toxicity to vertebrates, i.e. hepatotoxins (MCs, nodularin), neurotoxins (anatoxin-a, -a(S), BMAA), cytotoxins (cylindrospermopsins), dermatotoxins (lyngbya-, aplysiatoxins) and irritant toxins (LPS), or by their chemical structure into cyclic peptides, alkaloids, amino acids and LPS (Table 1-1) (Dittmann and Wiegand, 2006). Almost all cyanotoxins exist in multiple structural variants (Table 1-1), whereby MCs have by far the most with at least 271 structural variants recorded, differing in amino acid composition and side chain modification (Huisman et al., 2018; Meriluoto et al., 2017).

Peptides account for the majority of cyanotoxins with 600 cyanobacterial peptides being discovered and described over the past 30 years (Welker and Von Döhren, 2006). Besides MCs, these include also lesser studied cyanotoxins such as aeruginosins (Murakami et al., 1994), cyanopeptolins (Martin et al., 1993), anabaenopeptins (Harada et al., 1995), microginins (Okino et al., 1993) and microviridins (Ishitsuka et al., 1990). These peptides are also frequently found in cyanobacterial blooms, along with other not yet identified peptides (Welker and Von Döhren, 2006), but they receive much less attention compared to MCs (Janssen, 2019). Cyanotoxins include very potent toxins that can cause death within minutes (neurotoxins) or within hours (hepatotoxins) of animals in case of acute toxicity. Case studies and reports of animal and human poisonings caused by cyanotoxin exist worldwide (Buratti et al., 2017), and range back until 1878 (Sivonen, 2009) and 1931 respectively (Chorus et al., 2000). Due to the vast amount of cyanotoxins only the properties of most frequently detected ones will be discussed briefly in the following paragraph, including a more comprehensive description of MCs and anabaenopeptins.

1.4.1. Microcystins: Occurrence, structure and toxicity

MCs are named after Microcystis aeruginosa, the cyanobacterium in which the toxin was first isolated and described. Besides Microcystis, they are also produced by strains of the genera Dolichospermum, Nostoc, Planktothrix and Phormidium (Chorus, 2012a; Chorus and Bartram, 1999; Sivonen, 2009). Globally MCs are the most frequently found cyanotoxins (Chorus and Bartram, 1999). They are cyclic peptides that comprise seven amino acids and the general structure of cyclo-[D-Ala1]-[L-X2]-[β-D-MeAsp3]-[L-Z4]-[Adda5]-[γ-D-Glu6]- [Mdha7]. Their structural diversity is based on two variable amino acids X and Z in positions 2 and 4 respectively, as well as side chain modifications such as methylations. The one-letter code for the amino acids in the X and Z positions is used for the nomenclature of MCs: A MC with leucine (L) and arginine (R) in the X and Z positions respectively, is called MC-LR, while

GENERAL INTRODUCTION leucine and phenylalanine (F) will result in MC-LF. The congeners can exhibit different toxicities (Fischer et al., 2010), but MC-LR is regarded the most toxic one (Carmichael, 1995; Dittmann and Wiegand, 2006; Watanabe et al., 1988). MCs are very potent hepatotoxins and cause acute toxicity of animals due to liver dysfunction and hemorrhage (Dietrich et al., 2008). Evidence from several animal experiments also suggest tumor promotion properties of MCs as a result of chronic exposure (Dietrich and Hoeger, 2005; Humpage and Falconer, 1999; Nishiwaki-Matsushima et al., 1992). The toxicity of MCs is primarily caused by the amino acid on position 5 which is (2S,3S,8S,9S,4E,6E)-3-amino-9-methoxy-2,6,8-trimethyl-10-phenyl-4,6- decadienoic acid, also called Adda. This unique amino acid is only found in MCs and the related nodularins. Adda inhibits the catalytic subunit the serine/threonine (ser/thr) protein phosphatases (PP) 1 and 2A, thereby irreversible inhibiting the enzyme activity. Covalent binding of the methyldehydroalanine (Mdha) of MCs by PPs cysteine provides additional stability of the complex, but is not required for inhibition of the PP. MCs are produced in the cytoplasm and remain primarily intracellular in the thylakoid area until cell lysis (Rohrlack and Hyenstrand, 2007; Young et al., 2005). They are considered as hydrophilic molecules and therefore require active transport via organic anion transporting polypeptides (OATPs) to pass the bi-lipid cell membrane and thus reach the inside of the target cells (Fischer et al., 2005). Organ toxicity in higher animals including humans is therefore primarily based on the presence of these transporters. Nodularins also inhibit ser/thr PP1 and 2A but in contrast to MCs, they do not bind covalently to PP1 or PP2A (Bagu et al., 1997).

1.4.2. Anabaenopeptins: Occurrence, structure and toxicity

Anabaenopeptins are a diverse family of cyclic hexapeptides, named after the cyanobacterium Anabaena flos-aquae (now Dolichospermum flos-aquae) from which the first anabaenopeptins were isolated (Harada et al., 1995). They are also produced by strains of the genera Aphanizomenon, Microcystis, Planktothrix, Plectonema, Nodularia and Schizothrix and were reported for terrestrial, freshwater and estuarine habitats. Anabaenopeptins are comprised of a ring of five amino acid residues connected to an exocyclic residue through an ureido linkage. The general structure is L-X1-CO-cyclo-[D-Lys]-[L-X3]-[L-X4]-[L-MeX5]-[L-X6], where X1 and X3 to X6 are variable amino acids (Welker and Von Döhren, 2006). The ring is formed by an N-6-peptide bond between Lys and the carboxy group of the amino acid in position 6. The structure contains both protein and non-proteinogenic amino acids. To date, at least 96 structural variants have been reported (Spoof et al., 2015). Similar to MCs, anabaenopeptins inhibit ser/thr PP1 and 2A (Gkelis et al., 2006; Sano et al., 2001). In addition, they inhibit zinc-containing metalloexopeptidases such as carboxypeptidase A and B

GENERAL INTRODUCTION

(Murakami et al., 2000) as well as carboxypeptidase TAFIa, the thrombin activatable fibrinolysis inhibitor (Halland et al., 2015). The latter plays an important role in the balanced and coupled coagulation−fibrinolysis system, and its inhibition increases plasmin formation and endogenous fibrinolysis, leading to an antithrombotic effect. It is hypothesized that substitutions of the amino acid side-chain attached to the ureido group are closely linked to the inhibitory activity and enzyme specificity (Harms et al., 2016; Murakami et al., 2000).

1.4.3. Other cyanotoxins

Cyanobacterial neurotoxins are not as common as hepatotoxins but are more potent (Cox et al., 2003). There are three types known – anatoxin-a, anatoxin-a(S), and saxitoxins. In addition, the mild neurotoxin BMAA, a non-proteinogenic amino acid, has been found in a variety of cyanobacteria (Cox et al., 2003). Known producers of anatoxin-a belong to Dolichospermum, Aphanizomenon, Microcystis, Planktothrix, Raphidiopsis (previously Cylindrospermopsis) and Oscillatoria genera, while Anatoxin-a(S) is found in Dolichospermum only. Anatoxin-a as a nicotinic agonist binds irreversibly to nicotinic acetylcholine receptors in neurons and neuromuscular junctions. Its binding causes the opening and ultimately the blockage of neuronal transmission. In acute cases anatoxin-a causes loss of coordination, hyperactivity, muscular twitching and rapid death by respiratory arrest. Anatoxin-a(S) is an anticholinesterase with no structural variants. It irreversibly inhibits peripheral acetylcholine esterases, resulting in the blockage of nerve fluxes. Saxitoxins are a group of carbamate alkaloids produced in freshwater by Dolichospermum, Raphidiopsis, Lyngbya and Trichodesmium (Humpage et al., 1994; Sivonen, 2009). They block the neuronal Na+ and Ca2+ channels causing numbness and paralysis or even death by respiratory arrest (Wiese et al., 2010).

The alkaloid cylindrospermopsin was first isolated from the cyanobacterium Cylindrospermopsis raciborskii (now Raphidiopsis raciborskii) but is also produced by Aphanizomenon and Dolichospermum (Griffiths and Saker, 2003). Occurrences are mainly reported, though not exclusively, for tropical and subtropical countries (Bláha et al., 2009). Five structural variants are known. The primary toxic function of cylindrospermopsins is the inhibition of protein synthesis, causing cell death, with its main target being the liver. However, it can also affect other organs and metabolic pathways (Humpage et al., 2005). Acute toxicity can lead to renal and hepatic failure.

GENERAL INTRODUCTION

1.4.4. Biosynthesis, genetics and evolution of cyanotoxins

A major part of cyanotoxins (MCs, nodularins, anabaenopeptins, anatoxins, saxitoxins, cylindrospermopsins) are synthesized non-ribosomally by large, multi-enzyme complexes, known as non-ribosomal peptide synthetases (NRPS) and polyketide synthase (PKS) (Neilan et al., 2008). Synthesis can be performed by NRPS or PKS (Saxitoxin, (Kellmann et al., 2008) alone or by hybrid pathways including both (MCs, nodularins, anatoxins, cylindrospermopsins), though biosynthesis of many other cyanotoxins remains unsolved (Neilan et al., 2008).

The non-ribosomal synthesis of cyanotoxins is quite complex and energy demanding, e.g. the synthesis of one MC molecule requires 7 ATPs and 48 sequential catalytic reactions of six large multienzyme synthases/synthetases (McyA-E, G) and four monofunctional proteins (McyF/I/H/J) (Tillett et al., 2000). The NRPS and PKS enzymes are encoded by a large gene cluster (mcyS) which may constitute more than 5 % of the genome (Welker and Von Döhren, 2006), and has been described in detail for several cyanobacteria including Microcystis (Tillett et al., 2000), Dolichospermum (Rouhiainen et al., 2004) and Planktothrix (Christiansen et al., 2003). It spans over 55 kb and is composed of 10 genes (mcyA-J), arranged in 2 putative operons. The smaller operon (mcyA-C) encodes three NRPS (McyA-C). The larger one (mcyD- J), encodes a modular PKS (McyD), two hybrid enzymes comprising NRPS and PKS modules (McyE and McyG) and enzymes putatively involved in the tailoring (McyJ, F, and I) of the toxin. A third operon (mcyH) has been identified as an ATP-binding, ABC transporter‐like protein, potentially responsible for the active export of MCs across the cell membrane (Pearson et al., 2004; Rouhiainen et al., 2004; Tillett et al., 2000).

Anabaenopeptins are also produced by the NRPS pathway. The 32 kb Anabaenopeptins gene cluster (apt) consists of five genes (aptA1, aptA2, aptB, aptC, and aptD) which are organized in two operons, coding for the five respective enzymes (AptA1, AptA2, AptB, AptC, and AptD) (Rouhiainen et al., 2010). Two additional genes aptE and aptF, are also present in the apt gene cluster but do not code for NRPS enzymes. AptE is suggested to belong to a hydroxymethylglutaryl-coenzyme A lyase-like protein family, while AptF belongs to a family of ABC transporter-like proteins (Rouhiainen et al., 2010).

It is crucial to understand that not all cyanobacteria produce cyanotoxins. Genera with toxin- producing strains also contain related strains that lack the ability to produce this toxin. The difference between toxin-producing and non-producing strains is primarily owed to the presence or absence of the respective toxin synthetase gene cluster. Phylogenetic studies on the evolution of cyanotoxins compared the phylogenetic trees of housekeeping genes to MC,

GENERAL INTRODUCTION nodularin and saxitoxin synthetase gene trees (Dittmann et al., 2013; Rantala et al., 2004). The origin of the MC and nodularin gene clusters dates back some 1,600-2,000 million years ago (Rantala et al., 2004). The saxitoxins gene cluster also seems to have an ancient origin at least 2,100 million years ago (Murray et al., 2011). The ancient origin and the sporadic distribution of MC synthetase genes in modern cyanobacteria strongly suggests that the present nontoxic strains have lost the MC synthetase genes and the ability of MC production (Rantala et al., 2004). The situation for the other toxin families is still unclear and needs further investigations (Dittmann et al., 2013).

1.4.5. Regulatory factors and biological functions of cyanotoxins

The question why cyanobacteria produce toxins has been widely discussed but not entirely resolved yet. Despite the impact of cyanotoxins for human and wildlife health, the intoxication of vertebrates is most certainly not the selective force for their evolution. Cyanotoxins are classified as secondary metabolites, which means they rarely have a role in primary metabolism, growth, or reproduction but have evolved to somehow benefit the producing organism (Neilan et al., 2008). The high conservation of the cyanotoxin gene clusters over several million years until today, despite the energetic costs linked to their synthesis, indicates that they still have beneficial properties (Holland and Kinnear, 2013). Closely linked to the discussion about the biological function of cyanotoxins, is the investigation of physical, chemical and biological factors that regulate toxin production. There are numerous studies on the influence of abiotic factors, such as nutrients (N, P), trace metals (iron), growth temperature, light, salinity and pH and biotic factors like the presence of a competitor or predator on cyanotoxin production (Chorus and Bartram, 1999; Kaebernick and Neilan, 2001; Neilan et al., 2013). Often these studies have contradictory results or cannot be readily compared with each other, making it difficult to draw general conclusions from them (Chorus and Bartram, 1999; Neilan et al., 2013). This is owed to various reasons, such as different strains and species with varying growth optima being used, no standardized culture conditions or different methods used for toxin analysis. Nevertheless, several hypothesized functions of cyanotoxins emerged from these studies, which can be summarized under three main aspects: Physiological aid, chemical defence and infochemical (quorum sensing) (Holland and Kinnear, 2013; Kaplan et al., 2012; Leflaive and Ten‐ Hage, 2007).

Cyanotoxins have been speculated to have multiple physiological roles, such as cell signaling, nutrient uptake, iron scavenging, maintenance of homeostasis, and protection against oxidative stress (Kaplan et al., 2012; Neilan et al., 2013). MCs bind to a number of proteins

GENERAL INTRODUCTION apart from PPs. The covalent binding of the MdhA amino acid to thiol‐groups of several enzymes of the calvin cycle, phycobiliproteins and NADPH-dependent reductases is increased under high light conditions and oxidative stress (Zilliges et al., 2011). The study therefore suggests that MCs may play an important role in the protection of intracellular proteins that are sensitive to redox changes under conditions involving oxidative stress (Zilliges et al., 2011). Furthermore, MCs have been discussed to act as iron scavenging molecules, allowing toxic cells to take up or store iron more efficiently (Alexova et al., 2011a; Lukač and Aegerter, 1993; Sevilla et al., 2008; Utkilen and Gjølme, 1995). As a support of this theory, iron-deficient media was shown to cause an increase in transcription of the mcyD gene in Microcystis (Alexova et al., 2011a). The presence of MCs appears to give an advantage in the early stages of exposure to severe iron stress and may protect the cell from reactive oxygen species-induced damage (Alexova et al., 2011a). Other cyanotoxins may also cause an improved uptake of other nutrients, including P, N and C. This has especially been demonstrated for cylindrospermopsins produced by Aphanizomenon ovalisporum (Bar-Yosef et al., 2010). Under inorganic phosphate limitation, extracellular cylindrospermopsin concentrations increase, inducing alkaline phosphatase secretion by other phytoplankton species, thereby enhancing inorganic P supply for Aphanizomenon (Bar-Yosef et al., 2010). Similar to this, several studies have shown that also MC concentrations in Nostoc and Microcystis increase when cells are exposed to P-limiting condition (Kurmayer, 2011; Oh et al., 2000), while other studies report higher MC concentrations with increasing P (Rapala et al., 1997). MCs seem to be linked to the carbon-nitrogen metabolism (Alexova et al., 2011; Jähnichen et al., 2007; Waal et al., 2009). A comparative proteome study of six toxic and non-toxic strains of Microcystis aeruginosa found that nine proteins, all linked to carbon-nitrogen metabolism and redox maintenance, were differentially expressed amongst the toxic and non-toxin strains (Alexova et al., 2011).

For a long time, cyanotoxins, especially MCs, have been hypothesized to function as chemical defense against zooplankton grazers such as copepods and cladocerans (DeMott et al., 1991; Gilbert, 1996). For example, direct and indirect exposure of M. aeruginosa to several zooplankton species and their culture media was shown to cause an increased production of MC (Jang et al., 2003). Nevertheless, with the recently discovered ancient origin of MCs and saxitoxins, predating cyanotoxins long before metazoan life, zooplankton grazing pressure can be ruled out as primary reason for the evolution of cyanotoxins (Murray et al., 2011). However, pressure might also be provided by protozoan grazers (rhizopods, rotifers, ciliates) or by parasites such as fungi, lytic phages and heterotrophic bacteria with a much older evolutionary history (Gerphagnon et al., 2015; Holland and Kinnear, 2013). Especially oligopeptides, like

GENERAL INTRODUCTION

MCs and anabaenopeptins have been discussed as putative antiparasite defensive system but experimental data are very scarce (Rohrlack et al., 2013). In one of the few existing studies, Rohrlack et al. (2013) investigated the virulence of several parasitic fungi of the phylum Chytridiomycota (commonly referred to as chytrids) to a toxic Planktothrix strain producing MCs, microviridins, and anabaenopeptins, and to its non-toxic knock-out mutants. All chytrid strains were less virulent to the wild type than to at least one of its oligopeptide knock-out mutants, suggesting oligopeptide production in Planktothrix to be part of a defensive mechanism against chytrid parasitism.

Toxin-producing cyanobacteria constantly compete against other bacteria and eukaryotic algae for light, nutrients and space. Thus, an ecological role for cyanotoxins as allelopathic chemicals against those competitors has been suggested (Berry et al., 2008), and several studies have provided evidence for some cyanotoxins, including MCs (Kearns and Hunter, 2001; Singh et al., 2005) and anatoxin-a (Kearns and Hunter, 2001) action against eukaryotic alga Chlamydomonas or other cyanobacteria. A common problem of most of these studies is that they use purified or crude extracts of the toxins at concentrations exceeding environmental relevance and, in the case of crude extracts, may contain a variety of compounds besides the toxin of interest. Studies employing pure MC-LR have shown inhibitory effects on growth, morphology, and photosynthesis of some aquatic plants at ecological relevant concentrations (Pflugmacher, 2004; Romanowska‐ Duda and Tarczyńska, 2002). In contrast, other studies report no allelopathic effects of MC-LR and cylindrospermopsins on a variety of phytoplankton species (Pinheiro et al., 2013).

While the discovery of the hypothetical ABC‐transporter for MCs (mcyH) and anabaenopeptins (aptF) promotes the discussion about extracellular functions of those toxins, the active export has not been proven yet (Pearson et al., 2004). Cyanotoxins occur primarily intracellular, and are found extracellular only in small amounts until cell lysis. In this context, MCs have been discussed as signaling molecules for the intra-species communication in a quorum sensing‐ like manner (Kaplan et al., 2012; Neilan et al., 2013). Their release to the media by lysing cells due to various stressors is assumed to be sensed by the rest of the Microcystis cells which upregulate mcy transcription (Schatz et al., 2007). Furthermore, extracellular MC concentration seems to be positively correlated with Microcystis colony size (Gan et al., 2012; Wood et al., 2012).

In summary it can be concluded that despite extensive research, the variables regulating cyanotoxin production remain unclear. The primary function(s) of cyanotoxins are most likely unrelated to their toxic properties to invertebrates (Ouellette and Wilhelm, 2003). It is possible

GENERAL INTRODUCTION that the ancient primary function of cyanotoxins has shifted to several other intra- and extracellular functions (Holland and Kinnear, 2013; Kaplan et al., 2012). Considering this, in addition to the diverse chemical structures and mode of actions of cyanotoxins it seems unlikely that they have developed for a single purpose and should therefore be interpreted and discussed on an individual basis.

1.5. Bloom dynamics

Cyanobacterial blooms are usually classified into being toxic or non-toxic and by the predominant genus forming the bloom, e.g. a “toxic Microcystis bloom” or a “non-toxic Dolichospermum bloom”. This is a rather simplified version of a highly diverse and dynamic system. Species composition as well as toxicity of a cyanobacterial bloom are no static phenomena, but can vary both spatially and temporally within a short time-scale, i.e. within weeks, days or even hours (Berry et al., 2017; Steiner et al., 2017; Woodhouse et al., 2016). Understanding the short-term species succession and dynamics of toxin production is crucial to ensure reliable and safe bloom monitoring approaches, risk assessment and mitigation strategies.

1.5.1. Short-term species and toxin dynamics

It is well known that during the course of a cyanobacterial bloom and along with changing environmental parameters (Figure 1-2), succession from one predominant taxon to another can occur (Chia et al., 2018; Fernández et al., 2015). Since the rise of molecular sequencing tools, several publications have investigated the seasonal genetic diversity of cyanobacterial bloom populations in higher temporal resolution (Berry et al., 2017; Briand et al., 2009; Kardinaal et al., 2007; Woodhouse et al., 2016). These studies revealed that cyanobacterial blooms contain a plethora of different cyanobacterial species and strains co- existing and dynamically varying in space and time. For example, Berry et al. (2017) identified, based on the 16S rRNA sequence, 11 distinct genotypes (synonyme phylotypes) of four cyanobacterial genera during the bloom season. Even blooms which are mono specific at the first glace reveale a large variety of different genotypes (Kardinaal et al., 2007). Furthermore, cyanotoxin concentrations vary considerably not just from one bloom to another, but during the course of a single bloom (Buratti et al., 2017). Intracellular MC content varied up to 40-fold within several weeks during the same bloom, from 10 to 420 fg cell-1 in Planktothrix rubescens (P. rubescens) (Manganelli et al., 2016) and 40-fold from 100 to 4000 fg cell-1 in M. aeruginosa

GENERAL INTRODUCTION

(Sabart et al., 2010). Wood et al. (2011) showed that significant changes (28-fold) in MC cell quota can occur on a time scale of 2-6 h, resulting in an increase of total MC from ~100 to >2000 µg L-1. This variability can only partly be explained with changes in the levels of MC production by the individual MC-producing cells (Orr et al., 2018; Orr and Jones, 1998; Wood et al., 2011). As a matter of fact, laboratory studies with pure strains of cyanobacteria investigating correlations between MC cell quota and environmental factors, showed that induced changes in MC quotas are usually only three or four-fold (Sivonen and Jones, 1999). A much higher part of variability might be explained by the range of toxin content measured between individual genotypes of the same species. Kosol et al. (2009) analyzed MC and anabaenopeptin cell quota of 18 P. agardhii and 31 P. rubescens strains and found 14-fold and 12-fold differences respectively. Similar intraspecific variations were found for Cylindrospermopsis raciborskii (Davis et al., 2014; Willis et al., 2016) and M. aeruginosa strains (Wilson et al., 2006). This supports the assumption that much, if not most, of the variation in toxicity of natural blooms is the varying relative abundance of genotypes of the same species, but with varying individual toxin quotas (Buratti et al., 2017; Chorus and Bartram, 1999; Wood et al., 2017). In fact, co-occurrence and seasonal succession of toxic and non-toxic genotypes was shown in natural blooms for Microcystis by the presence or absence of the mcy gene cluster (Briand et al., 2009; Kurmayer et al., 2004; Kurmayer and Kutzenberger, 2003), or by correlating MC concentration to the presence of different 16S rRNA genotypes (Kardinaal et al., 2007). Nevertheless, environmental factors might still drive changes in bloom toxicity not through any stimulatory or trigger effect on the toxin production pathway itself, but through their selective effect on cell division rates and growth of different genotypes (Orr et al., 2018). A seasonal succession of different genotypes might often be a key mechanism determining cyanotoxin concentrations in lakes. Data from natural blooms exploring how environmental factors drive the succession of toxic and non-toxic genotypes are scarce and deserve further study.

GENERAL INTRODUCTION

Figure 1-2 Summary of abiotic and biotic factors controlling cyanobacterial boom dynamics. Adapted from Paerl 2018.

This is especially true for biotic factors, such as grazing, allelopathic and symbiotic interactions with heterotrophic bacteria and parasitism by viruses (Gons et al., 2002), pathogenic bacteria, protists and fungi (Gerphagnon et al., 2015), which have only recently come into focus as bloom-influencing forces (Berry et al., 2017; Sommer et al., 2012; Steiner et al., 2017; Woodhouse et al., 2016). Viral control of phytoplankton dynamics is well described for the marine environment (Brussaard, 2004) but less for the freshwater environment.

Chytrids are true fungi characterized by free-swimming motile stages called zoospores that actively find and infect new hosts. Chytrids infect a huge variety of phytoplankton species including filamentous cyanobacteria (Gerphagnon et al., 2015; Rasconi et al., 2012; Sønstebø and Rohrlack, 2011), and are highly host specific at the same time (Ibelings et al., 2004). Since several cyanotoxins are discussed as putative antiparasite compounds (see chapter 1.4.5), and parasites in general are highly hostspecific, it seems reasonable that chytrid parasite pressure might cause shifts between non-toxic and toxic genotypes in natural bloom communities. However, the extent to which chytrid parasitism can affect cyanobacterial bloom dynamics, or even result in the disintegration of blooms, is currently unresolved (Frenken et al., 2017; Gleason et al., 2015).

GENERAL INTRODUCTION

1.5.2. Consequences of cyanobacterial blooms

Cyanobacterial blooms can cause a multitude of water quality concerns, affecting human, wildlife and ecosystem health as well as the economy. The extent and severity of bloom impacts depend on the geographic location of the affected water body, size and toxicity of the bloom, prompt identification of the bloom, purpose of the water body and applied water treatment measures. Unlike natural lakes, artificial water reservoirs are formed or modified by human activity for specific purposes, in order to provide a reliable and controllable resource, including the production of drinking water, irrigation for agriculture, flood control, recreational services or generating hydroelectricity (hydropower dams), and often serve multiple of these purposes at the same time. Therefore, cyanobacterial blooms occurring in artificial waters reservoirs present a great potential to negatively affect human health and economy. At the same time, reservoirs are expected to become more important as reliable and sustainable water sources in the future, as water resources become more vulnerable to weather extremes linked to climate change (e.g. increased frequency and intensity of heat waves and droughts, changes in precipitation frequency and intensity) (Ehsani et al., 2017; Zarfl et al., 2015).

Human intoxication with cyanotoxins via drinking water or other pathways is considered the most severe impact of harmful cyanobacterial blooms. Although drinking water treatment processes can reduce cyanotoxins, the removal efficiency can be as low as 60 %, potentially leaving water quality compromised (Chapra et al., 2017; EPA, 2015; Zamyadi et al., 2012). As a result, several cases of human intoxication are linked to cyanobacterial blooms that occurred in drinking water supplies. The most fatal incident took place in a hemodialysis unit in Caruaru, Brazil in 1996. More than 120 dialysis patients were exposed intravenously to cyanotoxins, via poorly treated water from a contaminated drinking water reservoir. More than 74 patients died of liver failure as a consequence of intoxication with cyanobacterial toxins, now known to be mainly MCs (Azevedo et al., 2002; Pouria et al., 1998). The hemodialysis water originated in the Tabocas reservoir, which had experienced a massive growth of cyanobacteria after a drought, including species of the genera Microcystis, Dolichospermum and Raphidiopsis. The

Caruaru incident is the only confirmed case of acute lethal human toxicity from MCs. In a similar, but non-lethal incident in 2001, 44 patients in Rio de Janeiro were exposed to sublethal doses of MCs during hemodialysis (Soares et al., 2006). This time, the water originated from the Funil reservoir and the Guandu River, both drinking water supplies. Subsequent investigations revealed Microcystis and Dolichospermum cell concentrations up to 2 × 106 cells mL-1. Another serious case known as the "Palm Island mystery disease" occurred in 1979, Queensland Australia (Griffiths and Saker, 2003). More than 140 people had to be hospitalized after showing various symptoms of gastroenteritis. The sudden outbreak occurred

GENERAL INTRODUCTION after the Solomon dam, the local drinking water supply, was treated with copper sulfate to control a Raphidiopsis raciborskii bloom. The treatment caused cell lysis and thereby the release of intracellular cylindrospermopsin into the surrounding water (Griffiths and Saker, 2003). This example also highlights the importance of careful choices when deciding upon bloom mitigation measures.

Accidental ingestion of cyanotoxins might also occur during recreational activities like swimming or other water sports (Hilborn et al., 2014; Lévesque et al., 2014). Lévesque et al. (2014) found that gastrointestinal symptoms (diarrhoea, abdominal pain, nausea, vomiting, fever or abdominal cramps) were associated with recreational exposure to water bodies contaminated with cyanobacteria. Exposure to cyanobacteria and their toxins during recreational water use arises through three routes of exposure: direct contact of exposed parts of the body, including sensitive areas such as the ears, eyes, mouth and throat, and the areas covered by a bathing suit (which may collect cell material); accidental uptake of water containing cells by swallowing; and uptake of water containing cells by inhalation (WHO, 2003). However, MC concentrations in the aerosols produced by wind most likely would not reach critical concentrations (Wood and Dietrich, 2011). So far there are no known cases of intoxication with cyanotoxins via inhalation.

Cyanobacterial blooms occurring in water reservoirs used only for irrigation purposes, might still pose a threat to human health via food consumption. There are several indications that crops and vegetables spray-irrigated with contaminated water can accumulate cyanotoxins (Codd et al., 1999; Crush et al., 2008; Hereman and Bittencourt‐ Oliveira, 2012). Hereman and Bittencourt‐Oliveira (2012) showed that lettuce leaves sprayed with MCs at concentrations of 0.62-12.5 mg L-1 accumulated approximately 8-177 mg MC kg-1. This finding indicates that plants can accumulate MCs to levels that exceed the tolerable daily intake recommended by the World Health Organization (WHO) for drinking water (0.04 mg kg-1 body weight) (WHO, 2017). Via this way, cyanotoxins can potentially move into farm animal and human food chains. A rather new exposure route is the consumption of blue-green algae dietary supplements. Many of these products contain Aphanizomenon flos-aquae from upper Lake Klamath in Oregon, USA (Carmichael et al., 2000; Gilroy et al., 2000). In the same lake, blooms of toxin- producing M. aeruginosa can occur (Gilroy et al., 2000). The consumption of those supplements can pose a threat to humans, as they can contain MCs, sometimes above the guidance value of 1 μg g-1 dry weight given by the Oregon Health department (Gilroy et al., 2000) and 10 μg g-1 dry weight, respectively (Dietrich and Hoeger, 2005). Apart from cyanotoxins, the production of taste and odor compounds by cyanobacteria can be a major concern in lakes, rivers and reservoirs, that are used for drinking water supplies or

GENERAL INTRODUCTION recreation (Izaguirre et al., 1982; Li et al., 2012). Presence of taste and odor in drinking water may result in decreased consumer trust. The major taste and odor-causing compound in drinking water obtained from surface water is geosmin (-2-methyl isoborneol) and is produced by filamentous cyanobacteria (Srinivasan and Sorial, 2011). Geosmin causes an earthy-musty taste and odor and has an extremely low taste threshold in drinking-water of a few ng per liter (WHO, 2017). Due to its persistence to elimination in a conventional water treatment process such as coagulation, sedimentation, filtration and chlorination, geosmin removal causes additional water treatments expenses (Bruce et al., 2002; Srinivasan and Sorial, 2011).

Cyanobacterial blooms and toxins can cause substantial financial impacts on local economy. Financial losses are related to intensified water treatment and bloom monitoring, health consequences, restriction of recreational activities and lowered property values (Carmichael and Boyer, 2016; Hamilton et al., 2013; Merel et al., 2013). A famous example is Lake Erie, one of the five great lakes in North America providing drinking water for >11 million people around the city of Toledo. Lake Erie experiences recurring massive cyanobacterial blooms of Microcystis since decades (Rinta-Kanto et al., 2005; Steffen et al., 2014). In 2015, half a million residents lost their access to drinking water for several weeks after MCs were detected in the treated drinking water. Economic costs of intensified water treatment but also tourism loss in the Lake Erie region have been estimated to impose annual costs of $272 million over a 30- year period if left unsolved (Smith et al., 2019). Another example is the Darling River in Australia. During a neurotoxic Dolichospermum bloom in 1991, losses to the tourism industry were estimated around $1.5 million (Steffensen, 2008). More recently in 2009, a massive bloom impacted over 1,000 km of the Murray River in the same area, and prompted major management actions by public authorities (Al-Tebrineh et al., 2012), as the river serves as a major source for drinking water supply and livestock watering. Furthermore, cyanobacterial blooms have been associated with several fish kills in fish ponds and can therefore cause reduced yields in fish farming (Albay et al., 2003; Bürgi and Stadelmann, 2002; Drobac et al., 2016). Further costs are caused by expensive monitoring and cyanotoxin analysis which have to be performed multiple times per week once a bloom is observed in a water body, sometimes over several months (EPA, 2015).

Besides humans, also wildlife, livestock and pets can be exposed to cyanotoxins through toxin- contaminated drinking water or by swimming or playing in affected water bodies (Puschner et al., 1998; Stewart et al., 2008). Most recently, between March and June 2020, cyanobacteria were linked to the death off 330 elephants in the Okavango Delta in Botswana (Benza, 2020). According to the wildlife authorities blood tests of the carcasses identified cyanobacterial neurotoxins to be the cause of their death. This was consistent with their finding of neurotoxin-

GENERAL INTRODUCTION producing cyanobacteria in the waterholes. Though detailed and official information by the authorities regarding species and toxins involved are pending at the time of writing. Besides this extraordinary and severe event, dog poisonings are covered by local media every summer and are caused by dogs eating cyanobacterial mats or by cleaning their fur after playing in contaminated waters (Walker et al., 2008). Blooms may also harm the whole aquatic ecosystem, as they can affect physical and chemical characteristics of the water bodies, resulting in the loss of phytoplankton and zooplankton biodiversity. Common effects include the decrease of water transparency and dissolved CO2, the increase of pH, and the alteration of biogeochemical cycles (Sukenik et al., 2015). For example, surface blooms can block the sunlight from reaching the benthos, leading to changes in attached plant communities (Scheffer et al., 1997) and increased re–suspension of nutrients from the sediments. The reduction of the euphotic zone and the increase in the ratio between the mixing depth and the euphotic depth is a detrimental factor for most phytoplankters, but not for cyanobacteria actively controlling their vertical migration (Mitrovic et al., 2001). The high photosynthetic activity during the bloom may produce high amounts of oxygen, accompanied by depletion of

CO2 and increasing pH in the surface waters (Verspagen et al., 2014). If the pH drops down below 5 or rises above 10, most fish may become stressed and die (Wurts and Durborow, 1992). Sub-lethal or lethal high pH values can locally be reached during intense cyanobacterial blooms and high photosynthetic activity. The decomposition of the biomass during the

-1 declining bloom can lead to localized hypoxia (<2 mg O2 L ) or anoxia (Scavia et al., 2014), affecting reproductive capacity, behavior and physiology of fish at sub-lethal level and ultimately causing fish kills (Holzner et al., 2009; Richards et al., 2009). MCs were shown to be taken up by fish via ingestion and to accumulate in several organs (Ernst et al., 2006; Xie et al., 2005). Effects of cyanotoxin exposure of fish (e.g. zebra fish, tilapia, sea trout) were described for MCs (Malbrouck and Kestemont, 2006), nodularins and cylindrospermopsins (Sotton et al., 2014) and range from decrease in survival and growth rate of juveniles to multiple physiological and histopathological effects in adult fish. Cyanobacterial blooms may also affect invertebrates. During a bloom, cyanobacteria may account for 99 % of the total phytoplankton biomass and may therefore interfere with zooplankton food web processes. Especially large cladocerans like Daphnia are susceptible to interference from cyanobacterial filaments (de Bernardi and Giussani, 1990). Cyanobacteria tend to be of low food quality compared with eukaryotic phytoplankton due to morphological properties, toxicity, and the absence of only low amounts of polyunsaturated fatty acids and sterols (DeMott et al., 2001; Martin-Creuzburg et al., 2008; Müller-Navarra et al., 2004). Effects of cyanotoxins on invertebrates include acute effects (reduction in survivorship, feeding inhibition, paralysis), chronic effects (reduction in growth and fecundity), as well as biochemical and behavioral alterations (Ferrão-Filho and

GENERAL INTRODUCTION

Kozlowsky-Suzuki, 2011; Rohrlack et al., 2005; Sukenik et al., 2015). Bioaccumulation of cyanotoxins in invertebrates and vertebrates also has the potential of transferring these toxins through the food chain (Ferrão-Filho and Kozlowsky-Suzuki, 2011; Ibelings and Havens, 2008) and to ultimately become an issue for food safety.

1.6. Aim of the thesis

Cyanobacterial blooms can have harmful effects on human, wildlife and ecosystem health as well as severe consequences on the local economy. They can occur in natural lakes and rivers, but also in artificial water reservoirs. Furthermore, they show a distinct variability in genotype composition and toxicity. Due to climate change and ongoing eutrophication of freshwater ecosystems, global prevalence of harmful cyanobacterial blooms has increased in the last decades. Climate change scenarios predict bloom frequency and magnitude to even worsen in the future. At the same time, reliable and sustainable water resources including artificial water reservoirs will become more important. Consequently, there is a need for a better understanding of parameters causing cyanobacterial bloom occurrences as well as understanding factors that influence bloom species and toxin dynamics.

This thesis is divided into four manuscripts that improve our knowledge on individual aspects of bloom occurrences and dynamics in natural lakes and artificial water reservoirs. Manuscript 1 investigates the historical and spatial distribution of cyanobacterial bloom related proxies in lake sediment. This study improves the interpretation of paleolimnological studies based on sediment cores and helps to reconstruct historical changes in lake ecosystems which triggered bloom occurrences. Manuscript 2 investigates competitive growth of N2-fixing Dolichospermum with non-fixing green alga Chlorella under N-limiting and non-limiting growth condition. This study explores to what extent Dolichospermum can compensate N depletion using N2 fixation. Furthermore, it aims to identify nutrient scenarios that promote dominance of

N2 fixing cyanobacteria. This data may help to decide if N reduction can be used as bloom mitigation strategy. Manuscript 3 describes the intra-annual species and cyanotoxin dynamics of a natural Dolichospermum bloom in an oligotrophic reservoir. Furthermore, it monitors fungal parasitism of Dolichospermum filaments and discusses potential correlation between parasitism and bloom species and toxin dynamics. This study helps to understand abiotic and biotic factors influencing blooms dynamics in reservoirs. Manuscript 4 is a comprehensive and integrative information tool for water reservoir managements. It provides an overview of five different challenges of water reservoir management, namely: sedimentation, biostabilization of fine sediments, cyanobacterial blooms, greenhouse gas emissions, and

GENERAL INTRODUCTION social contestation or approval. The tool is attended to assist in the different steps of operation, to ensure a smooth and conflict-free operation of the reservoir system and the connected catchment area. The tool aims to positively contribute to future demands regarding managing water reservoirs.

MANUSCRIPT 1

2. MANUSCRIPT 1

Is a central sediment sample sufficient? Exploring spatial and temporal microbial diversity in a small lake

B. Weisbrod1, S.A. Wood2, K. Steiner2, R. Whyte-Wilding2, J. Puddick2, O. Laroche2, D.R. Dietrich1

1Human and Environmental Toxicology, University of Konstanz, Universitätsstrasse 10, 78457 Konstanz, Germany 2Cawthron Institute, 98 Halifax street East Nelson 7010, New Zealand

Published in Toxins 2020, 12(9), 580; doi: 10.3390/toxins12090580.

2.1. Abstract

Paleolimnological studies applying molecular techniques to sediment cores can explore long-term changes in lake ecology, including historical occurrences of harmful cyanobacterial blooms. Most studies are based on single cores from the center or deepest point of the lake, assuming this is representative of the whole lake. Data on small-scale spatial variability (tens to hundreds of meters) on microbial communities in lake sediment are scarce. In this study, surface sediments (top 0.5 cm) were collected from 12 sites (representing embayment, littoral and mid-lake sites; n = 36), and two sediment cores (littoral (32 cm depth) and mid-lake (40 cm depth)) in Lake Rotorua, a small eutrophic lake (New Zealand). Bacterial community (16S rRNA metabarcoding), Microcystis specific 16S rRNA, microcystin synthetase gene E (mcyE) and microcystins (MCs) were assessed. Measurements of radionuclides (210Pb, 137Cs) were used to date sediment core layers. Bacterial community of surface sediments differed significantly between sites (p < 0.001). The majority of amplicon sequence variants (88.8 %) were shared between samples, suggesting that a single core from a central site will capture dominant microbial communities. Despite intense toxin-producing Microcystis blooms in the past decades, no Microcystis specific 16S rRNA, mcyE and MCs were found in surface sediments. Microcystis specific 16S rRNA and mcyE occurred in sediment cores, but only between 4 to 24 cm depth (approximately 1950’s). 210Pb measurements showed a disturbed non-linear depth profile, similar to patterns previously observed in sediment cores, as a result of increased sediment fluxes after earthquakes. Our results indicate that toxin-producing Microcystis blooms are a relatively recent phenomenon in Lake Rotorua. We posit that the absence of Microcystis from the surface sediments is a consequence of increased deposition

MANUSCRIPT 1 of erosion material and sediment mixing stemming from the immense Kaikoura earthquake two years prior to our sampling.

2.2. Introduction

Lake ecosystems worldwide are experiencing dramatic changes in water quality, nutrient concentrations and phytoplankton community composition due to global warming, eutrophication and introductions of non-native species (Carpenter et al., 2011; Vitousek et al., 1997). Consequently, the occurrence of harmful cyanobacterial blooms has increased significantly (Huisman et al., 2018; Taranu et al., 2015). Many bloom-forming species produce cyanotoxins, which have neurotoxic, cytotoxic, and hepatotoxic effects on organisms (Sivonen, 2009). Contact with or ingestion of water contaminated with cyanotoxins can lead to poisoning, or death of humans, pets, stock, and wildlife (Sivonen, 2009). Amongst the cyanobacterial toxins, the cyclic heptapeptide microcystins (MCs) are the most frequently found (Legrand et al., 2017; Monchamp et al., 2016; Pilon et al., 2019; Zastepa et al., 2017, 2015).

MCs are potent inhibitors of eukaryotic protein phosphatases resulting in a wide range of organ toxicities, but primarily causing liver damage (Dietrich et al., 2008). Cyanobacteria genera known to produce MCs include Planktothrix, Oscillatoria, Dolichospermum and most prominently Microcystis (Sivonen, 2009). At least 271 structural variants of MCs have been described (Meriluoto et al., 2017). MCs are synthesized non-ribosomally by a large multifunctional enzyme complex which contains both non-ribosomal peptide synthetase and polyketide synthase domains (Tillett et al., 2000). The mcy gene cluster (mcyA-J) encoding these biosynthetic enzymes has been employed as the target region for the molecular detection of potentially toxin-producing cyanobacteria strains in water and sediment (Chorus, 2012a; Legrand et al., 2017; Medlin and Orozco, 2017).

Considerable effort has been invested in developing management and restoration plans for water bodies experiencing cyanobacterial blooms (Ibelings et al. 2016; Paerl 2014; Rastogi et al. 2015). Beyond the need for effective bloom mitigation strategies, there is an increasing demand for understanding how lake ecosystems and their catchments have changed historically and how this might correlate with increasing prevalence of cyanobacterial blooms. Unfortunately, long-term monitoring data are scarce. An emerging technique that can contribute historic data is the application of molecular biological methods to sediment core analysis (Giguet-Covex et al., 2019). Employing molecular techniques, e.g. metabarcoding and droplet digital PCR (ddPCR), allows reconstruction of historical cyanobacteria communities

MANUSCRIPT 1 using phylogenetic marker genes like the 16S ribosomal RNA (16S rRNA) but also functional genes (such as the mcy gene cluster) identified in sediment samples (Legrand et al., 2017; Monchamp et al., 2016). Accumulation of cyanobacterial DNA and toxins in sediments occurs via natural cell sedimentation during and after blooms (Wörmer et al., 2011). This is especially true for Microcystis sp. that overwinter in surface sediments (Borges et al., 2016; Preston et al., 1980). Both cyanobacterial DNA and MCs are thought to be well preserved in sediments, and both have been detected in sediment layers dating back over 200 years (Legrand et al., 2017; Monchamp et al., 2016; Pilon et al., 2019; Zastepa et al., 2017, 2015). Most paleolimnological studies on cyanobacteria (Legrand et al., 2017) but also eukaryotes (Capo et al., 2017; Rühland et al., 2003), collect sediment cores from a single central site or deepest point in a lake, assuming a relatively even distribution across the lake. However, during a bloom cell densities of buoyant cyanobacteria, e.g., Microcystis spp., are highly dependent on wind and current drift, hence there is often high spatial variability across a lake surface (Hunter et al., 2008; Vander Woude et al., 2019; Wang et al., 2016). Cell and toxin concentrations at one site can change dramatically within short time periods. Given the spatial patchiness of cyanobacterial blooms, and the accumulation of buoyant scums along shorelines, it is likely that cyanobacterial cells and toxins may also be unevenly distributed on lake sediment. In addition, MC sediment concentration resulting from surface blooms are likely influenced by a complex interplay of microbial degradation, thermal decomposition and photolysis via UV radiation which can all vary within different sites of the same lake (Christoffersen et al., 2002; Holst et al., 2003). Therefore, there is an uncertainty as to whether single core samples are representative of the whole lake and provide an accurate record of bloom history. Most research regarding spatial distribution of sediment microbial communities has been undertaken on intermediate (different lakes) (Monchamp et al., 2019) to global (thousands of kilometers) scales (Gucht et al., 2007; Lindström and Langenheder, 2012; Vargas et al., 2015). Hence, a better understanding of small-scale (tens to hundreds of metres) distribution patterns within a lake is required to ensure a robust interpretation of paleolimnological results.

This study focuses on Lake Rotorua, a small hypertrophic lake in the northeast of the South Island of New Zealand (Flint, 1975). The lake has experienced cyanobacterial blooms since at least the 1970’s with Dolichospermum, Aphanizomenon and MC-producing Microcystis present in high abundances. However, there is uncertainty as to whether these species have always been present in the lake and when blooms begun. The aims of this study were to: (i) determine the spatial variability in bacterial and cyanobacterial species in the surface sediment of Lake Rotorua, with a focus on the abundance of Microcystis, MC synthetase gene E (mcyE)

MANUSCRIPT 1 copy numbers and MCs; and (ii) explore the historical bacterial (in particular cyanobacteria) composition in two sediment cores.

2.3. Material & Methods

2.3.1. Study site

Lake Rotorua (42°24’05”S, 173°34’57”E) is a small (0.55 km2), shallow (max. depth 3 m), hypertrophic lake in the northeast of the South Island of New Zealand (Flint, 1975, Figure 2-1). The average temperature in the nearby town of Kaikoura is 12.7°C and precipitation averages 881 mm per year. In the early 1900s the native podocarp-mixed hardwood forest that dominated the catchment was removed. The catchment now has a mixed land use of low intensity grazing, native scrub and bushland, and exotic weeds and trees. Inflow comes largely from surface run-off during rainfall events. There is a single small outflow at the southern end of the lake, and the lake has an estimated residence of 0.86 y-1 (Kelly et al., 2013). During a two year monitoring study, Wood et al., (2017) showed that Lake Rotorua experiences annual cyanobacterial blooms between spring and autumn, with seasonal succession from nitrogen fixers (Dolichospermum and Aphanizomenon) to Microcystis later in summer. During Microcystis blooms, peak concentrations of >2,000 µg L-1 total MC and 4×106 cells mL-1 have been measured (Wood et al., 2017, 2011).

MANUSCRIPT 1

Figure 2-1 Map of Lake Rotorua (Kaikoura, New Zealand). Position of 14 sediment sampling sites (12 surface sediments (S01-S12) and two sediment cores (C1 & C2)). Inset: Map of New Zealand showing site of the lake.

2.3.2. Sediment sampling

Surface sediment samples were collected at 12 sites in January 2018 using a sediment core sampler (UWITEC, Mondsee, Austria; Figure 2-1). Surface samples were collected in embayments (S01 and S02), mid-lake (S03-05, S11) and littoral zones (S06-10, S12) from the top layer (0.5 cm) of the core (Table 2-1). At each site, three separate surface samples were taken. At two sites, sediment cores were retrieved; (i) C1, collected mid-lake (40 cm length), and (ii) C2, collected in the littoral zone (32 cm length). Sediment cores were sectioned into 1 cm increments. Sediment subsamples for DNA extraction were stored at -20°C until further processing. Subsamples for toxin extraction and age dating were sampled into 50 mL Falcon tubes and stored at -20°C until analysis.

Table 2-1 Sediment samples taken in this study.

MANUSCRIPT 1

At each sampling site the water temperature, pH, salinity and dissolved oxygen were measured using a YSI water quality sonde (YSI, USA).

2.3.3. DNA extraction and inhibition test

Subsamples of sediment (0.2-0.4 g) were weighed into the first tube of a PowerSoil DNA Isolation Kit (Qiagen, CA, USA) and DNA extracted according to the protocol supplied by the manufacturer. A random selection of DNA samples (n = 10) were screened in duplicates for inhibition using an internal control quantitative PCR (qPCR) assay (Haugland et al., 2005).

2.3.4. Quantitative PCR for Microcystis and mcyE enumeration in surface sediments

The number of toxin-producing Microcystis cells present was determined by amplifying the mcyE gene (a single copy gene) and quantifying the number of amplicons, using a standard curve generated from MC-producing Microcystis sp. strain CAWBG617 (Wood et al., 2017). The qPCR was performed in triplicate for each sample in a 12.5 μL of reaction mix containing 6.25 µL KAPA Probe Fast qPCR Kit Master Mix (2×), 1 µL of primers targeting a region within the mcyE open reading frame (0.4 µM, McyE-F2 and MicMcyE-R8) (Vaitomaa et al., 2003), 0.2 µL of mcyE probe (Rueckert and Cary, 2009) and 1 µL of template DNA per sample. The standard curve was constructed using a purified (AxyPrep PCR Clean-up Kit, Axygen Biosciences, USA) PCR product generated with the primers described above. The number of copies in the PCR product used for the standard curves was determined using: (A × 6.022 × 1023) / (B × 1×109 × 650), with A being the concentration of the PCR product, 6.022 × 1023 (Avogadro's number), B being the length of the PCR product, 1×109 used to convert to ng, and 650 the average molecular weight per base pair (bp). Each point of the standard curve was analyzed in triplicate for each qPCR run conducted. The standard curve generated was linear (R2 > 0.99) and PCR efficiency was > 0.9. The results of copies per μL DNA were converted to copies per gram of sediment using: (D × 80) / E, with D being the number of copies per μL, × 80 since 1 μL DNA of the elution volume (80 μL) was used, and E being the weight of sediment (g) used for DNA extraction.

To detect and quantify Microcystis sp., a region of the 16S rRNA gene was amplified using Microcystis sp. specific primers MICR184F (Neilan et al., 1997) and MICR431R (Rinta-Kanto et al., 2005). qPCR and standard curve construction was performed as described for mcyE enumeration.

MANUSCRIPT 1

2.3.5. Droplet digital PCR for mcyE enumeration in sediment cores

Quantification of the mcyE copy numbers in sediment cores was carried out with ddPCR using a BioRad QX200 system. Each ddPCR reaction included 2 μL of 900 nM of each primer and 0.55 μL of 250 nM of primer probe (primers and probe as described for qPCR), 11 μL of BioRad ddPCR Supermix for probes, 1.5 μL DNA, and 4.95 μL of sterile water. The reaction mixture was divided into nanodroplets by joining 20 μL of the reaction mixture with 70 μL of BioRad droplet oil. After processing, the nanodroplet volume resulted in a total of 40 μL, which was transferred to a PCR plate for amplification at 96°C for 10 min, followed by 40 cycles at 94°C for 30 s, 60°C for 60 s and a final extension step of 96°C for 10 min. The ddPCR was run on one plate, with four negative controls (containing all reagents plus DNA/RNA-free water) and four positive controls (genomic DNA extracted from a sample known to contain the mcyE gene) included. The results were converted to copies per gram of sediment using: (D × 66) / E, with D being the number of copies per μL, × 66 since 1.5 μL DNA of the elution volume (100 μL) was used, and E being the weight of sediment (g) used for DNA extraction.

2.3.6. High Throughput Sequencing

16S rRNA PCR Surface and core sediments were analyzed via Illumina™ sequencing. A region of the 16S rRNA gene approx. 400 bp long was amplified using bacteria-specific primers 341F: 5’- CCT ACG GGN GGC WGC AG-3’ and 805R: 5’-GAC TAC HVG GGT ATC TAA TCC-3’, modified to include Illumina™ adapters (Herlemann et al., 2011; Klindworth et al., 2013). PCR reactions were performed in 50 µL volumes containing 25 μL of AmpliTaq Gold® 360 master mix (Life Technologies, CA, USA), 12.5 μL GC inhibitor (Life Technologies), 2 μL of each primer (10 μM, Integrated DNA Technologies, IA, USA), 2 μL template DNA (between 60 and 220 ng) and 6.5 μL Milli-Q water. PCR cycling conditions were: 95°C for 10 min, followed by 27 cycles of 95°C for 30 s, 50°C for 30 s, 72°C for 45 s, and a final extension of 72°C for 7 min. PCR products were visualized with 1.5 % agarose gel electrophoresis with Red Safe DNA Loading Dye (iNtRON Biotechnology Inc, Kyungki-Do, Korea) and UV illumination to ensure amplification of a single 400 bp product. PCR products were purified (Agencourt® AMPure® XP Kit; Beckman Coulter, CA, USA), quantified (Qubit® 20 Fluorometer, Invitrogen), diluted to 10 ng µL-1 and submitted to New Zealand Genomics Limited (Auckland, New Zealand) for library preparation. Sequencing adapters and sample-specific indices were added to each amplicon via a second round of PCR using the NexteraTM Index kit (IlluminaTM). Amplicons

MANUSCRIPT 1 were pooled into a single library and paired-end sequences (2 × 250) were generated on a MiSeq instrument using the TruSeqTM SBS kit (IlluminaTM). Sequence data were automatically demultiplexed using MiSeq Reporter (v2), and forward and reverse reads were assigned to samples.

Quality control and taxonomic assignment of Illumina sequences Analysis of sequencing data was performed using the QIIME 2 2019.1 (Bolyen et al., 2018). Raw sequence reads were demultiplexed and quality filtered using q2-demux plugin. Denoising of sequences was followed by chimera identification and removal using DADA2 (Callahan et al., 2016). To account for differential sequencing depth among samples, the number of reads per sample was rarefied to 28,700 (surface sediment samples) and 26,177 reads (sediment core samples). Five samples were excluded from further analysis because of low read number; two surface sediments (S01 and S02), two subsamples from the mid-lake sediment core (9-10 cm and 17-18 cm) and one subsample from the littoral core (29-30 cm). The implemented q2-feature-classifier-plugin (a scikit-learn naive Bayes machine-learning classifier) (Pedregosa et al., 2012) was used for classification and pre-trained on the SILVA bacteria 16S rRNA gene reference database (version 132) (Quast et al., 2013), trimmed with the primer sequences prior to taxonomic assignment. Quality filtered and taxa assigned amplicon sequence variants (ASVs) were further pre-processed and filtered using R (version 3.6.1) (R Core Team, 2020) and the phyloseq package (McMurdie and Holmes, 2013). Chloroplast, rare ASVs accounting for less than 0.1 % of the total bacterial community, and ASVs which could not be assigned further than kingdom level were excluded from further analysis.

2.3.7. Graphics and data analysis of High Throughput-Sequences

To visualize bacterial community dissimilarities between sampling sites and depth increments, non-metric-multidimensional-scaling (NMDS) was performed, using the R vegan package (Oksanen et al., 2019). Filtered and pre-processed ASVs were previously fourth root transformed and a Bray-Curtis dissimilarity matrix was calculated. To test for significance of observed dissimilarities in bacterial community between sites, a one-way distance-based permutational analysis of variance (PERMANOVA, Anderson, 2001) was performed using PRIMER (version 7) with the PERMANOVA+ add on (Clarke and Gorley, 2015). Venn diagrams were designed using the Venny tool version 2.1 (Oliveros, 2015). GraphPad prism (version 5) for Windows was used for all other figure preparations.

MANUSCRIPT 1

2.3.8. Toxin analysis

Toxin extraction from surface sediments Surface sediment samples (6-10 g wet weight) were freeze-dried and extracted with three consecutive extractions using 100 % MeOH (5 mL each). During each extraction step, samples were vortexed for 1 min, sonicated for 15 min and centrifuged (3200×g, 10 min, 10°C). The extracts were pooled and dried at 40°C under a stream of nitrogen gas and dissolved in 1 mL 100 % MeOH. Extracts were stored at -20°C until measurement.

Toxin analysis of surface sediments Toxin extracts were analyzed by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) as described previously (Puddick et al., 2016). To assess the extraction method and check for matrix effects, the recovery rate of MC was determined by fortification with a semi-purified extract of Microcystis CAWBG11 which produces a range of MC congeners (Puddick et al., 2014). The MC-enriched extract (dissolved in 80 % MeOH) was added to sediment samples and incubated at 4°C for 30 min. Samples were extracted as described previously and analyzed via UPLC-MS/MS.

2.3.9. Sediment chronology of mid-lake sediment deep core

For age-dating of the mid-lake sediment core, subsamples representing six increments (4-5, 6-7, 10-11, 14-15, 24-25 and 32-33 cm) were analyzed using gamma spectrometry of the long-lived radionuclides 210Pb and 137Cs at the Institute of Environmental Science and Research (Christchurch, New Zealand). Activities were quantified using a gamma counter with a high-purity germanium well detector. Activities are reported in Bq per kg. Uncertainties are based on the combined standard uncertainty multiplied by a coverage factor (k) = 2, (providing a level of confidence of 95 %) as described before (FAO/IAEA. 2017, Sikorski, 2019). Core chronologies from 210Pb and 137Cs measurements were calculated using the constant rate of supply (CRS) model in R by means of the mudata2 package (Dunnington and Spooner, 2018). Assuming that the supply of unsupported 210Pb to the sediment is the same for each time interval, the CRS model is calculated as follows:

-1 -1 tz = λ -1 ln(A0 Az ) Equation (1) Where (t) is the age of any interval (z) (Appleby and Oldfield, 1978).

MANUSCRIPT 1

2.4. Results

2.4.1. Physicochemical parameters

Temperature, pH and salinity did not differ markedly between sampling sites (Table S 2-1). Dissolved oxygen concentration varied between 0.4 and 7.6 mg L-1 and was highest in the embayment (S01, 7.63 mg L-1;S02, 6.3 mg L-1) and at one littoral site (S08, 6.6 mg L-1; Table S 2-1).

2.4.2. Surface sediments

Detection of Microcystis sp. specific 16S rRNA and mcyE All samples, besides the littoral sample S06, contained low levels of the Microcystis sp. 16S rRNA gene (Figure 2-2A). The littoral samples S07, S09 and S12 contained more than 106 gene copy numbers g-1 sediment (Figure 2-2A). The mcyE gene copy numbers were relatively high in the three mid-lake samples (S03, S04, S05) (69,443-1.3 × 106 gene copy numbers g-1 sediment), low (≤15,565 gene copy numbers g-1 sediment) in samples S01-02, S06-07, S10 and S12, and no copies were detected in samples S08, S09 and S11 (Figure 2-2B). As a single copy gene, mcyE gene copy numbers should be present at a lower concentration compared to the Microcystis 16S rRNA copy numbers. This was observed for samples S01-03, and S06-12, while higher mcyE gene copy numbers than corresponding 16S rRNA copy numbers were observed for samples S04 and S05.

Figure 2-2 Abundance of Microcystis specific, (A) 16S rRNA gene, and (B) mcyE gene in surface sediment samples (S01-S12). Sample sites are indicated by color (embayment = green, mid-lake = blue, littoral = yellow). Data are log(x+1)- transformed.

MANUSCRIPT 1

Bacterial community The total number of unique 16S rRNA ASVs after quality-filtering and pre-processing was 1011. Negative controls contained 372 reads assigned to 12 ASVs. These were excluded when ASV that accounted for >0.01 % of reads were removed. A total of 546 ASVs (54 %) were shared between embayment, littoral and mid-lake sampling sites (Figure 2-3) and 65 ASVs (36 %) were shared between mid-lake and littoral sites (in total 90.1 % shared ASVs). Mid-lake and embayment sites shared three ASVs, and littoral and embayment sites shared 55 ASVs (Figure 2-3). The most common seven bacterial phyla were Proteobacteria, Bacteroidetes, Acidobacteria, Fibrobacteres, Verrucomicrobia and Planctomycetes. The relative abundance of cyanobacteria in surface sediments was low, accounting for 0.14 - 0.59 % of the total bacterial community (Figure 2-4A). Nine unique cyanobacteria ASVs were detected, representing four genera; Snowella, Planktothrix, Dolichospermum and Cyanobium (Figure 2-4B). Contrary to expectations based on Microcystis sp. specific 16S rRNA gene analyses, no Microcystis ASVs were detected (Figure 2-4B). Further integration of the raw data showed that Microcystis ASVs were present in very low numbers, but because they represented <0.01 % of all ASVs, they were discarded in the bioinformatic pipeline.

Figure 2-3 Venn diagram. Unique and shared amplicon sequence variant between the three sampling sites of surface sediments.

MANUSCRIPT 1

Figure 2-4 Bacterial community of surface sediment samples (S01-S12) based on 16S rRNA sequences. Relative abundance of; (A) cyanobacteria and the seven most abundant bacteria phyla the in the total bacterial community, and (B) cyanobacteria genera in the cyanobacteria portion of the data. Sample sites are indicated by color (embayment = green, mid-lake = blue, littoral = yellow).

The NMDS ordination showed clear grouping of the total bacterial community by sampling site (Figure 2-5). The communities in the embayment sites (S01 and S02) were significantly different from littoral and mid-lake sites (S03-S12), (PERMANOVA, emb.-littoral: t = 3.142, p = 0.0001; emb.-mid.: t = 4.011 p = 0.0007; mid.-littoral: t = 2.285, p = 0.0001; Figure 2-5A). Cyanobacterial communities were also significantly different between sites (PERMANOVA, emb.-littoral: t = 2.153, p = 0.001; emb.-mid.: t = 5.577, p = 0.0006; mid.-littoral: t = 3.026, p = 0.0001; Figure 2-5B).

Figure 2-5. Two-dimensional non-metric multidimensional scaling (NMDS) ordination of microbial communities in the surface sediment samples (S01-S12). Plots are based on Bray-Curtis dissimilarities of 16S rRNA sequences of the; (A) total bacterial community, and (B) cyanobacterial community. Sampling site is indicated by shape (circle, triangle, square). Each point represents a sample. Replicates form a polygon. Stress value indicates fit of ordination, values ≤0.1 are considered fair, values ≤0.05 indicate good fit.

Toxin analysis MCs were not detected (<0.02 ng mL-1 or <2 µg g-1 wet weight in a 10 g sample) in any of the surface sediments collected. Recovery rates of the spiked MC congeners ranged between

MANUSCRIPT 1

25 % for MC-RR, 73 % for MC-LR and 126 % for MC-LA (average result for the four spiked samples), indicating absorption to the sample matrix for certain congeners (Table S 2-2). Recovery rates also differed between the four samples fortified with MCs, reinforcing this observation.

2.4.3. Sediment cores

Bacterial community and detection of Microcystis sp. specific 16S rRNA High throughput sequencing of 16S rRNA amplicons of sediment core samples resulted in a total of 1,712 unique ASVs of which 1,521 (89 %) were shared among both cores (Figure S 2-1). An additional 159 ASVs occurred only in the mid-lake core, and 32 ASVs were unique to the littoral core (Figure S 2-1). NMDS ordination of the total bacterial community shifted with increasing depth in both cores, indicating an increasing dissimilarity between communities with increasing depth (Figure 2-7). In total, nine ASVs were assigned to cyanobacteria representing four genera Microcystis, Aphanizomenon, Dolichospermum and Cyanobium (Figure 2-6B). In the mid-lake core, the cyanobacterial ASVs occurred mainly in the upper 17 cm (0.4-16.5 % relative abundance) and in the littoral core, in the upper 12 cm (0.3-10.4 % relative abundance), showing in both cores a decreasing trend in relative abundance with increasing sediment depth (Figure 2-6A, Figure S 2-2A). In deeper sediment core layers, cyanobacterial ASV abundance was below 0.1 % of total bacterial community (Figure 2-6A, Figure S 2-2A).

Figure 2-6 Bacterial community of the sediment core layers (cm) from the mid-lake (C1). Relative abundance of; (A) Microcystis and other cyanobacteria in the total bacterial community, and (B) Cyanobacteria genera in the total cyanobacteria community based on 16S rRNA sequences. mcyE gene copies (g- 1 sediment) were detected with droplet digital PCR. Missing bars indicate that cyanobacterial taxa accounted for <0.1 % of the total bacterial community in these samples.

MANUSCRIPT 1

Figure 2-7 Two-dimensional non-metric multidimensional scaling (NMDS) ordination of bacterial community of sediment cores. Plots are based on Bray-Curtis similarities of 16S rRNA sequences of the total bacterial community. Color indicates depth layer, sampling site is indicated by shape (circles = C1 (mid-lake), triangles = C2 (littoral)). Each point represents a sample.

In the mid-lake core Microcystis was detected only in traces in the uppermost layer (0-1 cm, 0.03 %, Figure 2-6B), but was not detected in the upper 1-3 cm. No Microcystis was detected in the upper 0-3 cm of the littoral core (Figure S 2-2). Cyanobacterial taxa found in these upper layers belonged to the genera of Dolichospermum, Cyanobium and Aphanizomenon (Figure 2-6B, Figure S 2-2). In the mid-lake core, Microcystis was first detected at 4 cm depth and was present to a depth of 25 cm (Figure 2-6A). Microcystis abundance ranged between <1 % to 48 % of the cyanobacterial community accounting for 0-0.18 % of the total bacterial community in the sediment layers (Figure 2-6). A peak in the relative abundance for Microcystis was observed at 22-25 cm depth, accounting for 9 %, 12 % and 48 % of the cyanobacterial community, respectively. In the littoral core, Microcystis was first detected at 5 cm. The relative abundance of Microcystis in the cyanobacterial community was much lower compared to the mid-lake core accounting for ≤14 % (Figure S 2-2).

Detection of Microcystis specific mcyE The vertical profile of mcyE copy numbers in the mid-lake core showed a similar pattern to the sequencing results (Figure 2-6B). Copy numbers of mcyE were detected in lower numbers in the upper 3 cm of the core (1,089-1,361 copies g-1 sediment) but increased from 4 to 17 cm. A peak concentration of 18,217 mcyE copies g-1 sediment was observed at 10-11 cm. Beyond 17 cm, mcyE copy numbers were not detected (<100 copies g-1 sediment).

MANUSCRIPT 1

2.4.4. Sediment chronology

The 137Cs chronomarker profile showed a typical increase in activity over sediment depth. The increase was from 4 cm sediment depth down to a depth of 14-15 cm and was below background level for the two deeper layers (24-25 cm & 32-33 cm; Figure 2-8). The 1963-1964 (nuclear weapon testing) 137Cs activity peak was not directly measured. However, the peak was assumed to be located between the highest 137Cs activity (14 cm depth) and the first below background measurement (24 cm; Figure 2-8). Due to the low temporal resolution, no dating model was applicable to this 137Cs profile. 210Pb activity showed a non-linear activity pattern with no clear increasing or decreasing trend. Therefore, the pattern could not be interpreted as a function of radioactive decay and no 210Pb dating model could be applied.

Figure 2-8 Radionuclide profile of 137Cs and 210Pb of the sediment core from the mid-lake (C1). Supported 210Pb (226Rd) has been subtracted from plotted 210Pb data. Measured value consistent with the background measurement are indicated with (≤bg). Error bars indicate uncertainty based on the combined standard uncertainty multiplied by a coverage factor providing a level of confidence of 95 %. “1963-1964 peak” refers to the presumed marker position of the 137Cs maximum from 1963-1964.

2.5. Discussion

2.5.1. Bacterial community and Microcystis abundance in surface lake sediment

In this study, we initially set out to explore the spatial variability of Microcystis and other cyanobacteria in surface sediments to improve knowledge of fine-scale distribution patterns needed for paleolimnological research. Initially, we planned to explore only the cyanobacterial

MANUSCRIPT 1 component of the metabarcoding data but given the unexpectedly low abundance of cyanobacteria, we expanded our analysis to the entire bacterial community. This also gives these results greater relevance to paleolimnological studies which are now incorporating molecular microbial analysis into their suite of techniques (Capo et al., 2017; Monchamp et al., 2018; Rühland et al., 2003).

There are only a few studies available which focus on spatial variability of sediment DNA on a fine scale. O’Donnell et al. (2017) assessed a near-shore marine habitat and observed a decrease in similarity of sediment microbial communities with increasing distance between sampling sites. To our knowledge there is just one other paleolimnological study focusing on historic cyanobacteria communities which sampled several cores in the same lake (Savichtcheva et al., 2011). However, the latter study focused on comparing the quality and integrity of a preserved DNA fragments between cores, rather than characterising the cyanobacterial community.

The results of the present study demonstrate that surface sediment samples taken from a central point in the lake capture approximately 90 % of the microbial diversity in the mid-lake and littoral zones. This is especially important as most paleolimnological studies use single sediment cores from the deepest point in a lake. In contrast, the bacterial communities in the embayment sites were significantly different to the littoral and mid-lake sites and varied markedly among each other. This is most likely due to differences between abiotic parameters compared to littoral and mid-lake sites as reflected by lower dissolved oxygen and water depth. In addition, the different physical and chemical sediment characteristics might have an impact on DNA preservation between the embayment sites (Giguet-Covex et al., 2019). However, correlating differences in environmental parameters to differences of bacterial community structure was not the focus of this study and requires a more comprehensive sampling approach. It is important to recognize that Lake Rotorua is a small lake. In larger lakes, a higher spatial diversity of abiotic parameters and bacterial communities might be expected.

In the present study we expected to observe a high number of Microcystis 16S rRNA and mcyE gene copies and MCs in the surface sediment samples. This assumption was based on the occurrence of intense Microcystis blooms recorded within the last decade (Borges et al., 2016; Wood et al., 2017) and the occurrence of high numbers of Microcystis cells on surface sediments during a study undertaken in 2014 (Borges et al., 2016). Microcystis can overwinter on the sediment surface (Fallon and Brock, 1981; Preston et al., 1980; Reynolds et al., 1981). The overwintering cells can resist harsh environmental conditions and form an inoculum for new spring blooms (Brunberg and Blomqvist, 2003; Kitchens et al., 2018). The abundance of

MANUSCRIPT 1 overwintering cells has been shown to vary spatially, with higher abundance generally occurring at greater depths (Borges et al., 2016; Brunberg and Blomqvist, 2003). However, in the present study, only traces of Microcystis 16S rRNA gene copies and none of the eleven MCs analysed were detected in the sediment surface layers. The absence of Microcystis and MCs in the surface sediment, despite the recent reports of MC-producing Microcystis in the lake, prompted us to take sediment cores to ensure that our detection methods were viable and that DNA and toxins were preserved in the sediment of Lake Rotorua.

2.5.2. Historic Microcystis populations

In agreement with the results from surface sediments, Microcystis 16S rRNA and mcyE gene copies were absent in the upper 3 cm of the sediment cores. However, Microcystis 16S rRNA and mcyE gene copies were detected in greater depths indicating that toxin-producing strains of this species have been present in the lake since around the 1950’s. The reasons why there was no Microcystis and a generally low abundance of cyanobacteria before the 1950’s cannot be readily explained with the data presented here. It might be linked to changes in land use of the lake catchment. The land around the lake was once podocarp forest, but this was burnt and replaced by pasture. Together with the rising usage of fertilizers, this has caused significant nutrient enrichments in the region’s waterways (Ministry for the Environment & Stats NZ, 2019). An increase of cyanobacteria during the 1900s was also observed in other lake sediment studies of North America (Taranu et al., 2015) and Europe (Legrand et al., 2017) and was linked to nutrient enrichment as a result of intensified agriculture.

Differences in historic bacterial communities between the littoral and mid-lake cores are probably due to water level fluctuations of Lake Rotorua. We assume that the current littoral area for the lake may have been swamp or wetland for some time or even for reoccurring time periods. This was supported by the fact that we were unable to obtain a longer core as we hit what we presume was thick historic vegetation and root. The mid-lake core is therefore more representative for looking at the historic bacterial communities of Lake Rotorua.

2.5.3. Why is Microcystis absent in the surface sediments?

The absence of Microcystis 16S rRNA and mcyE gene copies and MCs in the uppermost layers of all sediment samples of Lake Rotorua, corresponds with the observation that since 2016 (July & October 2017, January 2018) no Microcystis cells have been recorded in water column samples (Environment Canterbury, unpublished monitoring data). In

MANUSCRIPT 1

November 2016, an earthquake of magnitude 7.8 struck the Kaikoura region. This was the second biggest earthquake recorded in New Zealand since European settlement. The earthquake triggered an estimated 100,000 landslides in the region, although no major ones were visible directly around the lake (Dellow et al., 2017; Massey et al., 2018). It is possible that this event resulted in a significant flux of sediment into the lake. Lake sediments have previously been used to reconstruct the chronology of natural disasters including floods (Schillereff et al., 2014) and earthquakes (Moernaut et al., 2014; Wilhelm et al., 2016). Earthquakes are linked to increased sediment fluxes to rivers and basins as a consequence of seismically induced landslides which often leads to a short-term increase of local and distant erosion rates (Avşar et al., 2016). These sediment fluxes can be identified as lake deposits (turbidites), which differ in structure from sediments originating from regular sedimentation processes, and can help to reconstruct long-term earthquake occurrences as previously shown for the New Zealand alpine fault (Howarth et al., 2014).

While the 137Cs profile of the sediment core was as expected, 210Pb did not display its typical linear decay profile known from undisturbed sediments. Non-linear 210Pb profiles where previously shown for Lake Antern in the French Alps (Arnaud et al., 2002) and the Chilean lake district (Arnaud et al., 2006) as a result of earthquake-induced sediment changes. These changes can result from direct seismological mixing of old sediment layers containing 210Pb depleted material (Arnaud et al., 2002), or from landslides triggered by earthquakes, which increase erosion rates and sediment fluxes to rivers and basins (Avşar et al., 2016), thereby overlaying lake sediments. The thickness of the earthquake-induced mass movement deposits varies between lakes and depends on the magnitude of the seismic event, site and size of triggered landslides as well as on the amount and quality of the erodible material in the catchment area. Of the few studies that have linked lake sediment deposits to the intensity of historic earthquake events, seismic event-linked deposits in lake sediment cores from the European Alps ranged between 1 to >20 cm thickness (Arnaud et al., 2002; Wilhelm et al., 2016), in a Chilean lake between 0.5 to 7 cm (Moernaut et al., 2014) and in alpine lakes of New Zealand between 0.2 to 20 cm (Howarth et al., 2014). This demonstrates that each lake needs to be analyzed individually to allow any deduction of the impact of the seismic event on lake sediments.

The 2016 Kaikoura earthquake had a magnitude of 7.8, which was higher or equivalent to the seismic event related changes in the sediments described for the European Alps, Chile and earlier seismic events in New Zealand. Thus, it is likely that the disturbed upper 3-4 cm in Lake Rotorua resulted from a mixing effect as well an increased deposition of erosion material (Arnaud et al., 2002). The finding that other cyanobacterial taxa were detected in the upper 3

MANUSCRIPT 1 cm and intact biofilms were found at the sediment surface (Figure S 2-3) may appear contradictory to the disturbed sediment hypothesis. However, Cyanobium and Snowella are common freshwater picocyanobacteria (Callieri et al., 2013). They are known to be very adaptive to changing environmental conditions (Callieri, 2010; Callieri et al., 2012) and may represent the pioneering species occupying new niches after major environmental changes induced by increased sediment fluxes. Aphanizomenon and Dolichospermum both produce akinetes, thick walled cells that remain dormant until environmental conditions are favorable. It is possible that these were more resilient to sediment mixing than Microcystis vegetative cells or survived in shoreline terrestrial habitats, allowing these genera to rapidly recolonize following the sediment disturbance.

2.6. Acknowledgments

The work was financially supported by the Ministry of Science, Research and the Arts of the Federal State Baden-Württemberg, Germany (Water Research Network project "Challenges of Reservoir Management - Meeting Environmental and Social Requirements"; 814/15), the Ministry of Business, Innovation and Employment, New Zealand (Programme "Our lake’s health; past, present and future"; C05X1707) and the Marsden Fund of the Royal Society of New Zealand (Blooming Buddies; CAW1601). The International Max Planck Research School for Organismal Biology provided a travel grant. The authors thank Sean Waters (Cawthron Institute) for his support during field sampling.

MANUSCRIPT 1

2.7. Supplemental Information

Figure S 2-1 Venn diagram. Unique and shared amplicon sequence variant (ASV) between the two sediment cores from the mid-lake (C1) and the littoral (C2).

Figure S 2-2 Bacterial community of the sediment core layers (cm) from the littoral (C2). Relative abundance of; (A) Microcystis and other cyanobacteria in the total bacterial community, and (B) Cyanobacteria genera in the total cyanobacteria community based on 16S rRNA sequences. Missing bars indicate that cyanobacterial taxa accounted for <0.1 % of the total bacterial community in these samples.

MANUSCRIPT 1

Figure S 2-3 Biofilm growing on the surface sediment of Lake Rotorua.

Table S 2-1 Abiotic parameters of surface sediment sampling sites.

Table S 2-2 Recovery rates of microcystin (MC) congeners.

LC-MS (ng/mL) MC-RR dmMC-RR MC-YR MC-LR dmMC-LR MC-FR MC-WR MC-RA MC-LA MC-FA MC-WA Total MCs MC-RRs MC-XRs MC-RZs MC-XAs Spike Control - A 9.98 0.51 2.09 41.6 9.24 22.5 8.33 3.73 16.09 5.3 1.03 120.4 10.49 83.76 3.73 22.42 Spike Control - B 7.38 0.39 1.7 30.79 6.3 17.6 5.99 2.7 11.81 4.05 0.77 89.48 7.77 62.38 2.7 16.63 Ave 8.68 0.45 1.895 36.195 7.77 20.05 7.16 3.215 13.95 4.675 0.9 104.94 9.13 73.07 3.215 19.525 0 0 0 0 0 Sediment Spike - 4c 2.36 0.12 1.81 39.2 9.02 18.82 5.98 3.63 18.57 5.5 1.07 106.08 2.48 74.83 3.63 25.14 Sediment Spike - 8b 6.35 0.54 2.4 54.6 12.5 27.16 9.76 4.65 21.27 6.7 1.33 147.26 6.89 106.42 4.65 29.3 Sediment Spike - 9b 0.19 4.29 0.72 2.33 0.64 0.53 14.25 4.01 0.73 27.69 0 8.17 0.53 18.99 Sediment Spike - 10c 0.25 7.09 1.25 3.42 0.89 0.77 16.34 4.67 0.84 35.52 0 12.9 0.77 21.85

% of Expected MC-RR dmMC-RR MC-YR MC-LR dmMC-LR MC-FR MC-WR MC-RA MC-LA MC-FA MC-WA Total MCs MC-RRs MC-XRs MC-RZs MC-XAs Sediment Spike - 4c 27% 27% 96% 108% 116% 94% 84% 113% 133% 118% 119% 101% 27% 102% 113% 129% Sediment Spike - 8b 73% 120% 127% 151% 161% 135% 136% 145% 152% 143% 148% 140% 75% 146% 145% 150% Sediment Spike - 9b 0% 0% 10% 12% 9% 12% 9% 16% 102% 86% 81% 26% 0% 11% 16% 97% Sediment Spike - 10c 0% 0% 13% 20% 16% 17% 12% 24% 117% 100% 93% 34% 0% 18% 24% 112% Ave 25% 37% 61% 73% 76% 65% 60% 74% 126% 112% 110% 75% 26% 69% 74% 122% SD 35% 57% 59% 68% 75% 60% 61% 64% 22% 25% 30% 55% 36% 66% 64% 23% Min 0% 0% 10% 12% 9% 12% 9% 16% 102% 86% 81% 26% 0% 11% 16% 97% Max 73% 120% 127% 151% 161% 135% 136% 145% 152% 143% 148% 140% 75% 146% 145% 150%

MANUSCRIPT 2

3. MANUSCRIPT 2

Competition for nitrogen: N2 fixation allows cyanobacteria to outcompete green algae

B. Weisbrod1, J. Breiholz1, D. Martin-Creuzburg2, D.R. Dietrich1

1Human and Environmental Toxicology, University of Konstanz, Universitätsstrasse 10, 78464 Konstanz, Germany

2Limnological Institute, University of Konstanz, Mainaustrasse 252, 78464 Konstanz, Germany

3.1. Abstract

Diazotrophic cyanobacteria are able to fix atmospheric nitrogen (N2) using the enzyme nitrogenase. This allows those taxa to grow under low nitrogen supply. Given the physiological costs of N2 fixation, there is an ongoing discussion about its efficiency, and if N2 fixation can completely compensate for other N sources, without decreasing growth rates of diazotrophic taxa. This is especially relevant, considering the effect of reducing N loads in surface waters as cyanobacterial bloom mitigation strategy, and if this may cause a shift towards N2 fixing taxa without decreasing the total cyanobacterial biomass. Here we investigated the growth and nitrogenase expression of diazotrophic Dolichospermum in chemostats under nitrogen-limiting (N-) and nitrogen-saturating (N+) conditions, both in mono-culture and in co-culture with the green alga Chlorella. Dolichospermum growth was not affected by N treatment in mono- cultures and nitrogenase expression was higher under N- condition, indicating that

Dolichospermum can fully compensate for N limitation using N2 fixation. In the co-cultures, Chlorella outcompeted Dolichospermum in the N+ treatment, while Dolichospermum outcompeted Chlorella in the N- treatment. Nitrogenase expression of Dolichospermum was higher in all N- treatments compared to the same day of the N+ treatment and showed a clear temporal increase over the course of the experiment. Our results provide experimental evidence that N-only reduction in eutrophic surface water may cause a shift from eukaryotic green algae towards diazetroph cyanobacterial species, e.g. Dolichospermum. Therefore, N- only input reductions should be considered carefully as bloom mitigation strategy.

MANUSCRIPT 2

3.2. Introduction

During their long evolutionary history (∼3.5 billion years), cyanobacteria have developed a variety of physiological adaptations, allowing them to inhabit almost every ecosystem on this planet, and to establish dense populations, called cyanobacterial blooms, with harmful effects on ecosystems, wildlife and human health (Huisman et al., 2018). Some genera have developed the ability to fix atmospheric nitrogen (N2), also known as diazotrophy

(Herrero et al., 2001). While N2 fixation is a common trait among the bacterial domain, cyanobacteria are the only known photoautotrophic prokaryotes. To catalyze the conversion of N2 to ammonia, diazotrophic organisms use the enzyme nitrogenase. The high sensitivity of this enzyme to oxygen requires spatial separation from oxygenic photosynthesis. In the filamentous, bloom-forming genus Dolichospermum, N2 fixation is confined to specialized cells, known as heterocysts (Kumar et al., 2010; Thiel, 2005). Theoretically, diazotrophic species, like Dolichospermum, have inexhaustible access to nitrogen, making them superior to non- diazotrophic phytoplankton at low dissolved inorganic nitrogen (DIN) concentrations. However, the fixation of N2 is energetically a very costly reaction, requiring 16 ATPs for every N2 molecule

- that is fixed (N2 + 8 H + 8 e + 16 ATP = 2 NH3 + H2 + 16 ADP/Pi; (Gallon, 1992). Therefore, diazotrophic cyanobacteria rely on high light intensities to compensate for the high costs associated with N2 fixation (Paerl, 2017; Staal et al., 2002), and thus are supposed to switch to this alternative pathway only at low DIN concentrations (Holl and Montoya, 2005). Given the physiological costs of N2 fixation, there is an ongoing discussion about its efficiency, and if N2 fixation can completely compensate for other nitrogen sources, without decreasing growth rates of diazotrophic taxa. This is especially relevant, considering the effect of reducing nitrogen loads in surface waters as cyanobacterial bloom mitigation strategy. For decades, reduction of phosphorus loads was recommended to reduce the overall dominance of cyanobacteria in phytoplankton communities, as a reduction of the nitrogen load without a proportional reduction of the phosphorus load was considered to solely cause a shift from non- fixing to N2-fixing species (Schindler, 1977; Schindler et al., 2008; Smith, 1983; Vrede et al.,

2009). However, more recent studies suggest that N2 fixation cannot compensate completely for depletion of DIN sources (Shatwell and Köhler, 2019; Willis et al., 2016) and therefore both nitrogen and phosphorus should be controlled simultaneously to mitigate cyanobacterial blooms (Dolman et al., 2012; Harpole et al., 2011; Lewis and Wurtsbaugh, 2008; Paerl et al., 2016). Some studies even suggest that reduction of nitrogen alone might be the key to bloom control (Shatwell and Köhler, 2019). Although cyanobacterial growth characteristics have been extensively studied in chemostat systems, surprisingly few studies have focused on diazotrophic species (Agawin et al., 2007; Brauer et al., 2015; De Nobel et al., 1998a, 1997;

MANUSCRIPT 2

Holl and Montoya, 2005; Layzell et al., 1985). To our knowledge there are just four chemostat studies on growth competition between diazotrophic cyanobacteria and green algae (Agawin et al., 2007; Huisman et al., 1999; Hyenstrand et al., 2000; Takeya et al., 2004), though none of these studies focused on the common bloom-forming genus Dolichospermum.

Here, we investigated the growth and nitrogenase expression of diazotrophic Dolichospermum in chemostats under nitrogen-limiting (N-) and nitrogen-saturating (N+) conditions, both in mono-culture and in co-culture with the green alga Chlorella. We tested the following hypotheses: (1) In mono-cultures, Dolichospermum growth rate is reduced under N- conditions

+ as compared to N conditions, due to the higher energy costs associated with N2 fixation. (2) Nitrogenase expression increases in Dolichospermum under N- conditions. (3) In co-cultures,

- N2 fixation allows Dolichospermum to outcompete Chlorella under N conditions, whereas (4) Chlorella competitively excludes Dolichospermum under N+ conditions.

3.3. Material & Methods

3.3.1. Test organisms and culture condition

For all experiments, we used the freshwater cyanobacterium Dolichospermum sp. 90 (previously called Anabaena sp. 90, hereinafter Dolichospermum). This strain was originally isolated from lake Vesijärvi, Finland in 1986 and was identified as Dolichospermum circinalis (Sivonen et al., 1992). Its genome has been fully sequenced (Wang et al., 2012). Dolichospermum sp. 90 is known to form heterocysts under nitrogen limitation and to produce microcystins, anabaenopeptins and anabaenopeptilides (Tonk et al., 2009). As eukaryotic competitor, we used the coccal, unicellular chlorophyte Chlorella cf. minutissima (hereinafter Chlorella), which was originally isolated from lake Pupuke, Auckland, New Zealand (Punčochářová, 1990). Small cells are considered to be stronger competitors for nutrients because of their higher surface-to-volume-ratio (Edwards et al., 2011). Chlorella was selected because its cell diameter of about ~5 µm is comparable to cell diameter of vegetative Dolichospermum cells. Both genera, Dolichospermum and Chlorella, are widely distributed in freshwater ecosystem around the globe and thus represent a suitable model system to study the competition for nitrogen between diazotrophic and non-diazotrophic phytoplankton species.

MANUSCRIPT 2

3.3.2. Chemostat experiments

Experiments were performed in chemostats, consisting of 2 L glass vessels and a peristaltic pump providing the chemostats with new media. Cultures were run with a volume of 1.75 L of modified Woods-Hole MBL (Guillard and Lorenzen, 1972), a dilution rate of 0.091 day-1 and an estimated initial cell density of 105 cells mL-1 per species. Actual cell densities did deviate from this (Table 3-1). Homogenous mixing of culture as well as sufficient supply of CO2 and N2 were ensured by aerating the culture vessel with sterile-filtered air. All experiments were run at 24±1 °C with a light intensity of 55 μmol photons m-2 s-1 (16:8 h light:dark cycle). Overflow from the chemostats was collected in sterile glass containers and culture material harvested over a 24 h sampling period was used for subsequent analyses, i.e. cell counting, Western Blot and nutrient analysis. Experiments were run as mono-cultures of Chlorella and Dolichospermum as well as co-cultures of both species. All chemostats were provided with the same phosphorus concentration (50 µM). Treatments were defined by two different nitrogen concentrations: The N+ treatment was provided with a concentration of 1000 µM N (N:P = 20) and the N- treatment with a concentration of 25 µM N (N:P = 0.5).

In total 12 chemostat runs were conducted (Table 3-1): Four mono-cultures of Dolichospermum under N+ (n = 2) and N- (n = 2) conditions and six co-cultures of Dolichospermum and Chlorella under N+ (n = 3) and N- (n = 3) conditions. Unfortunately, one of the N- replicates of the co-cultures failed and thus was excluded from further analysis. As Dolichospermum was the main focus of this study, Chlorella mono-cultures were run only once per nutrient treatment.

Table 3-1 Initial cell densities in mono- and co-culture chemostat experiments. Conducted at nitrogen-limiting (N-) and -saturating (N+) conditions.

Initial cell density (cells mL-1) Experiment Treatment Dolichospermum Chlorella Mono-cultures Chlorella N+ - 1.00×105 Chlorella N- - 5.75×104 Dolichospermum N+ 2.90×104 - Dolichospermum N+ 4.23×105 - Dolichospermum N- 3.77×104 - Dolichospermum N- 1.25×105 - Co-Cultures Dolichospermum × Chlorella N+ 3.94×104 6.00×104 Dolichospermum × Chlorella N+ 3.90×105 1.58×105 Dolichospermum × Chlorella N+ 5.12×105 1.43×105 Dolichospermum × Chlorella N- 4.21×105 8.38×104 - 5 5 Dolichospermum × Chlorella N 3.53×10 1.43×10

MANUSCRIPT 2

3.3.3. Cell counting

Samples for cell counting were taken daily throughout all experiments. 4-11 mL were immediately fixed in Lugol’s iodine solution and stored at 4 °C until counting. Chlorella cells were counted with a haemocytometer (Neubauer-Chamber, Brand, Germany).

Cell densities of Dolichospermum were estimated from average filament lengths and a previously established regression between filament lengths and cell numbers per filament using the equation:

D = A × B × 0.213

D = Cell densities mL-1 A = Filament mL-1 B = Average filament length, determined for each sample 0.213 = Linear regression between filament length and cell numbers per filament

Dolichospermum filaments per mL were first counted in an Utermöhl sedimentation chamber using an inverted microscope (Zeiss, Axiovert 135). Therefore, 10 mL were settled in the chamber for at least 12 h. Subsequently, filaments of five diameter transects were counted. In addition, the average filament length was calculated for every sample by measuring 20 random filaments.

3.3.4. Growth curve analysis

To test the effect of nitrogen treatments on population growth, we analyzed several endpoints of the growth curves. Specific growth rates (µ) during the exponential growth phase were assessed by fitting a four-parameter logistic equation to the individual growth curves and the resulting slope factor was used as µ.

MANUSCRIPT 2

The doubling time (dt) of the population during the exponential growth phase was determined using the following equation:

dt = ln (2) µ

Cell densities at steady state were determined on day 29 of the experiments. As not all chemostats ran until day 29, theoretical cell densities at day 29 were estimated using a modified, non-linear model based on the Gompertz function (equation 3; (Tsoularis and Wallace, 2002; YIN et al., 2003). The model assumes a sigmoidal growth curve characterized by a lag phase, followed by an exponentially growth phase and a steady state phase of constant cell density:

y0 ln( ) × e-µ × t y(t) = yM × e yM

y0 = cell density of initial population yM = maximum population size (carrying capacity) µ = specific growth rate t = time (days)

The total number of cells grown during each of the individual experiments was estimated from the area under each of the growth curves until day 29, which was determined using the fitted cell density data from the Gompertz function.

3.3.5. Statistics and data analysis

All statistical analyses were performed using GraphPad prism (version 5 for Windows). Unpaired two-tailed Student’s t-tests were used to assess the effect of nitrogen treatment on Dolichospermum growth in mono-cultures. Two-way ANOVAs were performed to test if the effect of nitrogen treatment differed between Dolichospermum and Chlorella in co-cultures. Endpoints were previously log-transformed to fit assumption of Gaussian distribution. For the endpoints ‘total cells’ and ‘cell density at steady state’, we used fitted values from the Gompertz function. In all statistical analyses, the level of significance was set to p < 0.05.

Densitometry analysis of western blot bands was performed using ImageJ version 1.52a. For all bands, the number of pixels in an equally sized area was determined. As we performed western blot only once per day and treatment, no further statistical analysis was performed on the densitometry data.

MANUSCRIPT 2

3.3.6. Nutrient analysis

Samples for nutrient analysis were taken on day 1, 7, 10, 13, 16, 20 and 29. Samples were filtered through a glass fibre filter (GFF; Whatman™, GE Healthcare Life Science, Chicago, USA) to separate the particulate nutrient fraction from the dissolved fraction. Both fractions were immediately stored at -20 °C until further analysis. Filters were used for particulate nitrogen (NPart) and carbon (CPart) analysis, using a Euro EA CHNSO Elemental Analyser (HEKAtech, Wegberg, Germany). Dissolved inorganic nitrogen (DIN) and soluble reactive phosphorus (SRP) were determined in an aliquot (2.7 mL) of the filtrate using an autoanalyzer (AutoAnalyzer 3, Seal Analytical, Southampton, UK).

3.3.7. Protein extraction and Western Blot

50 to 154 mL of sample was immediately processed for protein extraction. Samples were centrifuged (4000 × g, 10 min at 4°C) and the supernatant was discarded. The cell pellet was resuspended in 5 mL RIPA-Buffer (500 mM NaCl, 1 % (v/v) Triton-x-100, 50 mM Tris-HCl, 0.1 % (w/v) SDS, 1 % (w/v) Na-deoxycholat). PMSF (1 mM) was added to prevent any protease activity. The samples were shock frozen in liquid nitrogen and lysed by pulsed sonication at an amplitude of 25 % for 2 min (Brandson Digital Sonifier 450, Marshall Scientific, New Hampshire, USA). Cell extracts were centrifuged to remove any remaining cell debris (4000 × g, 10 min at 4°C) and supernatant was stored at -80°C until further analysis. Protein concentration was quantified using Pierce™ BCA Protein Assay Kit (Thermo Scientific, Massachusetts, USA). DTT (1 mM) was added to the protein extract to reduce disulfide bonds. The extract was heated for 10 min at 90°C prior to gel loading. Proteins were separated on a 12 % SDS-PAGE and transferred to a nitrocellulose membrane. The blot was then treated with the primary antibody binding to the nitrogenase enzyme. We used an anti-NifH antibody (0.68 µg mL-1, AS01 021A, Agrisera, Stockholm, Sweden) solution incubating over-night at 4°C. Incubation with the secondary antibody was performed using a rabbit anti-Chicken IgY (H&L) antibody (HRP conjugated, Agrisera, AS10 1489) at a 1:3000 dilution for one hour at room temperature. Blots were then treated with 200 μL Lumigen ECL Ultra (TMA-6) solution (Lumigen, Michigan, USA) and imaged using ImageQuant™ LAS 4000 (GE Healthcare Life Science, Chicago, USA).

MANUSCRIPT 2

3.4. Results

3.4.1. Mono-cultures

Cell growth of Chlorella in mono-cultures was strongly affected by nitrogen treatment (Figure 3-1A). Under N+ conditions, the cell density at steady state (2.71×107 cells mL-1) was 19-fold higher and the total number of cells (2.7×108 cells mL-1) 7.9-fold higher than under N- conditions (1.38×106 and 3.42×107 cells mL-1, respectively) (Table S 3-1). Despite the higher cell density at steady state, the growth rate of Chlorella was lower and the doubling time higher under N+ than N- treatment.

In contrast, the endpoints of the Dolichospermum growth curves in mono-cultures did not differ significantly between the N- and N+ treatment (Student’s t-test, p > 0.05, Table S 3-1, Figure 3-1B). Irrespective of nitrogen treatment, Dolichospermum cell densities at steady state and total cell numbers were lower than those of Chlorella grown under N+ conditions (steady state: 8.5-fold (N+) and 11-fold (N-) lower; total cells: 3.6-fold (N+) and 4.4-fold (N-) lower, respectively; (Figure 3-2A, B, Table S 3-1). In both N treatments, Dolichospermum reached the steady state phase much earlier (~ day 10) than Chlorella in the N+ treatment (~ day 30). Under N- conditions, Dolichospermum performed better compared to Chlorella, reaching a 1.8-fold higher steady state cell density and total cell number (Figure 3-2A, Table S 3-1). Growth rate and doubling time did not differ between Dolichospermum and Chlorella (Table S 3-1).

Figure 3-1 Growth curves. Chlorella (A) and Dolichospermum (B) in mono-cultures and co-cultures of both species (C) under nitrogen-limiting (N-) and nitrogen-saturating (N+) conditions (A: n = 1; B: n = 2; C: n = 3 for N+ and n = 2 for N-).

MANUSCRIPT 2

3.4.2. Co-cultures

The competitive outcome of the co-cultures clearly differed between the nitrogen treatments. In both N treatments, Dolichospermum and Chlorella cell densities initially increased (Figure 3-1C). From the fifth day on, competitive exclusion of one species started.

In the N+ treatment, Chlorella outcompeted Dolichospermum (Figure 3-1C), reaching an 8-fold higher steady state cell density and 4.7-fold higher total cell number compared to Dolichospermum (Figure 3-2B). Counterintuitive to this, growth rate of Chlorella was lower and doubling time higher compared to Dolichospermum (Figure 3-2B, Table S 3-2). However, the exponential growth phase of Dolichospermum was extremely short in the N+ treatment (Figure 3-1C), and is therefore not representative for the growth curve. In contrast, Dolichospermum outcompeted Chlorella in the N- treatment (Figure 3-1C), reaching a 5.4-fold higher cell density at steady state and a 2.7-fold higher total cell number, respectively (Figure 3-2B, Table S 3-2). Growth rate of Dolichospermum was lower in the N- treatment and doubling time higher compared to Chlorella (Figure 3-2B, Table S 3-2), but similar to Dolichospermum grown in the N+ treatment, exponential growth phase of Chlorella was negligible under N- treatment. A two- way ANOVA applied to assess the effects of ‘species’ and ‘N treatment’ on the growth endpoints in the co-cultures revealed a significant interaction between ‘species’ and ‘N treatment’ on all endpoints (growth rate: F1,6 = 26.27, p = 0.002; doubling time: F1,6 = 26.24, p = 0.002; cell density at steady state: F1,6 = 29.45, p = 0.002; total cell number: F1,6 = 10.05, p = 0.019).

Figure 3-2 Reaction norms of nitrogen-limited (N-) and nitrogen-saturated (N+) growth expressed as log-transformed endpoints of the chemostat experiments. (A 1-4) mono-cultures of Chlorella and Dolichospermum and (B 1-4) co-cultures of both species. Endpoints are: Growth rate (1), doubling time (2), cell density at steady state (3) total cells grown until day 29 (4). Data points represent mean values (Mono-cultures Chlorella: N+ & N-: n = 1; Dolichospermum: N+ & N-: n = 2; Co-cultures: N+: n = 3; N-: n = 2). Endpoints “cell density at steady state” and “total cells” represent fitted values estimated using the Gompertz function.

MANUSCRIPT 2

3.4.3. Western Blot

Western blots of mono-cultures showed a much higher nitrogenase expression for all analysed days in the N- treatments compared to the N+ treatments (Figure 3-3A). A similar trend was observed for the co-culture chemostats, all samples of the N- treatments showed a higher expression of nitrogenase compared to the samples from the same day of the N+ treatment (Figure 3-3B). Furthermore, nitrogenase expression showed a clear temporal increase from day 7 to day 19 of the co-culture experiment (Figure 3-3B).

Figure 3-3 Western-blot analysis of nitrogenase (NifH, ~35 kDa). Expression of nitrogen-limiting (N-) and nitrogen-saturating (N+) chemostat experiments. For (A) Dolichospermum mono-culture 6.2 µg total protein were applied per lane. For (B) Dolichospermum and Chlorella co-culture, 4.42 µg total protein were applied per lane. NifH was detected using a chicken anti-nitrogenase antibody as primary antibody followed by a HRP-conjugated rabbit anti-chicken IgY as second antibody. Analysis was performed for day (d) 7, 10, 13, 16, 19 of the experiments.

3.4.4. Nutrient concentrations

Dissolved N:P ratios of all N- chemostats stayed ≤2.5 throughout the experiments (Figure S 3-1), indicating a continuous strong N limitation. N:P ratios of N+ treatments showed a high variability between replicates and throughout the mono- and co-culture experiments. The N:P of the N+ mono-culture varied between 29 and 71 indicating a temporary N limitation (Figure S 3-1A). N:P of the N+ co-cultures varied in one replicate from 24 to 55 in the other replicate from 5 to 917 (Figure S 3-1A). Variations in N:P were mainly due to strong fluctuations in DIN concentrations (Figure S 3-1B). Both, Dolichospermum mono- and co-cultures showed a decreasing trend of DIN and SRP concentration throughout the experiments (Figure 3-4). In all chemostats, SRP concentrations were continuously <15 µM (Figure 3-4C, D).

MANUSCRIPT 2

Figure 3-4 Nutrients measured in chemostat experiments. Dissolved inorganic nitrogen (DIN) and soluble reactive phosphorus (SRP) measured in (A & C) nitrogen-saturating and (B & D) nitrogen-limiting Dolichospermum mono-culture (n = 1) and Dolichospermum and Chlorella co-cultures (n = 2).

3.5. Discussion

In this study we investigated the ability of diazotrophic Dolichospermum to compensate for DIN depletion by switching to N2 fixation, both in mono-cultures and in co-cultures with the green alga Chlorella.

Contradictory to the initial hypothesis, Dolichospermum growth in mono-cultures was not affected by N-limitation in this study. Previous studies of other N2-fixing cyanobacteria showed inconsistent results. In agreement to our study, Brauer et al. (2015, 2013) found that steady state population densities and specific growth rates of Cyanothece did not differ between high- N (100 µM and 1.5 mM, respectively) and N-free conditions at light saturation. In contrast, Agawin et al. (2007) described a > 2-fold higher steady state population density of Cyanothece in N-rich (8 mM) compared to intermediate (0.5 mM) and N-free conditions (Agawin et al., 2007). The observed effect was the same under low light (20 µmol m-2 s-1) and high light (40 µmol m-2 s-1) conditions, though steady state cell densities of all N treatments were > 2- fold higher in the high light compared to the low light conditions. These inconsistent results may be partially explained by differences in experimental setups, e.g. varying temperatures, light intensities and N concentrations. Temperatures below 21°C were shown to hamper the efficiency of the nitrogenase and thus N2 fixation (Brauer et al., 2013), but since culture conditions in all studies exceeded 21°C, temperature differences can be ruled out as

MANUSCRIPT 2 explanatory factor. While the definitions of low N varied on a relatively small scale (0-25 µM) between the studies, differences of high N concentrations were tremendous (32-8000 µM), and might account for a major share of the varying outcome. However, interspecies differences in N assimilation and critical light intensities might also play an important role (Agawin et al., 2007; Huisman et al., 1999). Cyanobacteria require high light intensities to maintain growth during N2 fixation (De Nobel et al., 1998). Optimal light intensity for maximum growth and N2 fixation rates seemingly varies among cyanobacterial species (De Nobel et al., 1998, 1997; Moisander et al., 2008; Willis et al., 2016). Dolichospermum is known to require high light intensities, severely light-limited conditions occur at <17 µmol photons m−2 s−1 (De Nobel et al., 1998). The here applied light intensity of 55 μmol photons m-2 s-1 can be considered as intermediate light intensity for Dolichospermum (moderately light-limited conditions at 60 µmol photons m−2 s−1; De Nobel et al., 1998). Therefore, we cannot exclude, that light intensity in this was too low to enable maximal growth rates of Dolichospermum.

The fact that nitrogenase expression was higher under N-limiting conditions suggests that nitrogenase was upregulated to compensate for low DIN concentrations. Since Dolichospermum growth was not affected by N-limitation, it can be assumed that increased expression of nitrogenase enabled higher N2 fixation rates and could fully compensate for low

DIN concentrations. However, N2 fixation rates were not directly measured in this study. Nitrogenase activity under different N scenarios was previously measured via nifH gene expression (Moisander et al., 2008), N2 fixation rates via acetylene reduction assay (Brauer et

15 al., 2013; Moisander et al., 2008) or N-labeled N2 method (Higgins et al., 2018; Mulholland et al., 2006). To our knowledge, upregulation of nitrogenase under N-limitation is shown here for the first time on the protein level.

3.5.1. Growth competition between Dolichospermum and Chlorella

According to competition theory, the species with the lowest critical nitrogen requirements (nitrogen concentration below which species will have negative net growth rate) will be the superior competitor under low N conditions (Agawin et al., 2007; Tilman, 1982). However, at high nutrient concentrations, the species with lowest critical light intensity competitively excluded the other species (Agawin et al., 2007). The non-N2-fixing phytoplankton species are in general considered to have lower light requirements than N2- fixing cyanobacteria and hence are expected to be better competitors at low light conditions (Agawin et al., 2007; Huisman et al., 1999). In agreement to this, we show here that Chlorella displaced Dolichospermum at N-saturating conditions, whereas Dolichospermum displaced

MANUSCRIPT 2

Chlorella at N-limiting conditions. Similar to mono-culture experiments, nitrogenase expression was continuously upregulated under N-limiting conditions, indicating that Dolichospermum compensated limited DIN sources by increased N2 fixation.

To our knowledge, the competition between Dolichospermum and green algae at N-saturating and N-limiting conditions has not been explored yet experimentally. However, several publications performed co-culture experiments with green algae and other N2-fixing cyanobacteria. Competitive exclusion of non-N2-fixing species by N2-fixing cyanobacteria was shown for Cyanothece, being the superior competitor over Chlorella under low (0.1 mmol L-1) but also high N (8 mmol L-1) if provided with high light (40 µmol photons m−2 s−1) conditions (Agawin et al., 2007). Under low light conditions (20 µmol photons m−2 s−1) Cyanothece in turn was replaced by the non-N2-fixing cyanobacterium Synechococcus bacillaris (S. bacillaris) under low and high N treatment (Agawin et al., 2007). Similar to this, Huisman et al. (1999) reported Aphanizomenon to be displaced by Chlorella under high N and high light conditions, while Aphanizomenon replaced the green alga Scendesmus under the same conditions. The latter example shows, that light requirements of N2-fixing species do not necessarily exceed the ones from non-N2-fixing species. Agawin et al. (2007) also found that Chlorella has a quite high critical light requirement, exceeding the one from Cyanothece.

Apart from competitive exclusion, N2-fixing cyanobacteria can also facilitate growth of non- fixing phytoplankton species (Agawin et al., 2007), as well as heterotrophic bacteria (Brauer et al., 2015) by releasing substantial amounts of fixed nitrogen (Mulholland et al., 2006). Agawin et al. (2007) found that at low light and low N concentration, S. bacillaris did not displace Cyanothece, but both coexisted. Release of fixed nitrogen by Cyanothece enabled S. bacillaris to become four times more abundant in the low nitrate co-culture than it would have been in mono-culture. However, this effect was not observed for Chlorella in the present study (compare Figure 3-1A and C). Taken together, this suggests that the capability of diazotrophic taxa to outcompete non-N2-fixing species highly depends on the individual light requirements of both competitors. This concept was previously described and tested by others (Agawin et al., 2007). Nevertheless, this study expands this finding with experimental data for Dolichospermum.

A reduction in nutrient loads (N and P), either by reducing external inputs or by reducing internal nutrient concentrations, is the prioritized bloom mitigation strategy. While there is consensus regarding the efficiency of P reduction (Dolman et al., 2012; Jeppesen et al., 2005), the effects of N reduction on cyanobacterial bloom formation are less clear. It has been proposed that N reduction would merely lead to a shift from non-N2-fixing species, such as

MANUSCRIPT 2

Microcystis, to N2-fixing species, such as Dolichospermum or Aphanizomenon (Downing et al., 2001; Schindler, 2012; Schindler et al., 2008), rather than decreasing the total biomass of toxic cyanobacteria. However, a general decrease in cyanobacteria biomass with decreasing N loads has been also reported, without promoting N2-fixing taxa (Dolman et al., 2012). The data we present here suggest that Dolichospermum dominance in natural systems is promoted by reducing N loads, as Dolichospermum could completely compensate for DIN depletion by switching to N2 fixation, without detrimental effect on growth parameters.

Furthermore, dominance of N2 fixing cyanobacteria in natural phytoplankton communities might be promoted in future by several effects linked to global warming. These include higher light intensity, increased water surface temperatures and elevated CO2 and are expected to decrease physiological costs of N2 fixation (Paerl and Paul, 2012; Visser et al., 2016). Further testing of Dolichospermum competitive growth under different light, temperature and CO2 regimes may elucidate this issue. Complementing existing phytoplankton competition models with further experimental data from competition experiments between N2-fixing cyanobacteria and other phytoplankton species is crucial to predict how changes in N load affect species composition in natural systems.

3.6. Acknowledgements

The work was financially supported by the Ministry of Science, Research and the Arts of the Federal State Baden-Württemberg, Germany (grant: Water Research Network project: "Challenges of Reservoir Management - Meeting Environmental and Social Requirements").

MANUSCRIPT 2

3.7. Supplemental Information

Figure S 3-1 Nutrient ratios in chemostat experiments. Dissolved inorganic nitrogen (DIN) to soluble reactive phosphorus (SRP) ratio measured in (A) nitrogen-saturating and (B) nitrogen-limiting Dolichospermum mono-culture (n = 1) and Dolichospermum and Chlorella co-cultures (n = 2).

Table S 3-1 Endpoints of Dolichospermum and Chlorella mono-cultures. SEM = Standard error of the mean.

Endpoints Student's t-test Mean ± SEM p-value Growth rate (d-1) Chlorella (N+) vs. Chlorella (N-) 0.11 vs. 0.43 NA NA Dolichospermum (N+) vs. Dolichospermum (N-) 0.37 ± 0.13 vs. 0.49 ± 0.15 0.61 ns Doubling time (d) Chlorella (N+) vs. Chlorella (N-) 6.35 vs. 1.62 NA NA Dolichospermum (N+) vs. Dolichospermum (N-) 2.18 ± 0.78 vs. 1.57 ± 0.49 0.58 ns Cell density at steady state (cells mL-1) Chlorella (N+) vs. Chlorella (N-) 2.62×107 vs. 732,400 NA NA Dolichospermum (N+) vs. Dolichospermum (N-) 3.05×106 ± 189,350 vs. 2.39×106 ± 431,846 0.3 ns AUC (total cells) Chlorella (N+) vs. Chlorella (N-) 2.70×108 vs. 3.42×107 NA NA + - 7 6 7 6 Dolichospermum (N ) vs. Dolichospermum (N ) 7.54×10 ± 8.03×10 vs. 6.10×10 ± 7.71×10 0.33 ns

MANUSCRIPT 2

Table S 3-2 Endpoints of Dolichospermum and Chlorella co-cultures. SEM = Standard error of the mean.

Endpoints Co-culture Treatment Mean ± SEM Growth rate (d-1) Dolichospermum N+ 0.58 ± 0.26 Dolichospermum N- 0.10 ± 0.002 Chlorella N+ 0.12 ± 0.02 Chlorella N- 0.86 ± 0.32 Doubling time (d) Dolichospermum N+ 1.70 ± 0.55 Dolichospermum N- 6.70 ± 0.12 Chlorella N+ 6.68 ± 1.63 Chlorella N- 0.94 ± 0.35 Cell density at steady state (cells mL-1) Dolichospermum N+ 1.43×106 ± 489,909 Dolichospermum N- 3.92×106 ± 521,797 Chlorella N+ 9.83×106 ± 139,090 Chlorella N- 514,960 ± 139,090 AUC (total cells) Dolichospermum N+ 3.67×107 ± 1.31×107 Dolichospermum N- 7.29×107 ± 1.86×106 Chlorella N+ 1.75×108 ± 6.02×107 - 7 6 Chlorella N 2.67×10 ± 1.33×10

MANUSCRIPT 3

4. MANUSCRIPT 3

Can toxin warfare against fungal parasitism influence seasonal Dolichospermum bloom dynamics? – A field observation

B. Weisbrod1, E. Riehle1, M. Helmer2, D. Martin-Creuzburg2, D.R. Dietrich1

1Human and Environmental Toxicology, University of Konstanz, Universitätsstrasse 10, 78464 Konstanz, Germany

2Limnological Institute, University of Konstanz, Mainaustrasse 252, 78464 Konstanz, Germany

4.1. Abstract

Cyanobacterial blooms often consist of numerous co-existing cyanobacterial species, with predominant taxa dynamically varying intra-annually. Parasitism by fungi (chytrids) has come into focus as an important factor driving short-term bloom dynamics. Using microscopic analysis, Illumina sequencing and cyanobacterial toxin analyses, we monitored the seasonal succession of Dolichospermum blooms in a reservoir along with environmental parameters. We identified two consecutive Dolichospermum blooms that were characterized by a straight and a coiled morphotype, separated by a complete bloom collapse. Phylotyping provided evidence for three putative Dolichospermum amplicon sequence variants (ASVs); i.e. Dolichospermum1 & 2 in the first bloom (straight filaments) and Dolichospermum3 in the second bloom (coiled filaments). Morphotype succession as well as total filament concentration did not correlate with any of the measured environmental parameters. Fungal parasitism by the chytrid Rhizosiphon crassum occurred in straight Dolichospermum filaments only. Coiled filaments showed no infection despite ambient presence of chytrids, deduced from fungal ASVs, throughout the entire observation period. Toxin concentrations (microcystins (MCs) and anabaenopeptins) correlated significantly with the abundance of the straight Dolichospermum morphotype. Enhanced cyanotoxin biosynthesis in the straight Dolichospermum morphotype, interpreted as a defensive reaction to fungal parasitism, appeared to come at the expense of lowered competitiveness with the co-occurring coiled morphotype. Our findings support the hypothesis that selective parasitism by chytrids is an important factor driving short-term morphotype and toxin dynamics within cyanobacterial blooms.

MANUSCRIPT 3

4.2. Introduction

Cyanobacterial blooms (hereinafter blooms) are no static phenomenon but highly diverse and dynamic events (Wood et al., 2017). Long-term (years, months) bloom dynamics can be observed as succession of the predominant bloom genus, primarily correlated to changes in environmental parameters, such as temperature, CO2, nutrients and pH (Fernández et al., 2015; Visser et al., 2016; Wood et al., 2017). However, molecular environmental surveys have demonstrated that the bloom community and the predominant species (identified via 16S rRNA metabarcoding) can vary in space and time also on a short- term scale, i.e. within weeks or even days (Berry et al., 2017; Steiner et al., 2017; Woodhouse et al., 2016). The same holds true for spatio-temporal variation of bloom toxicity, which also can change within days or even hours (Orr et al., 2018; Wood et al., 2011). Short-term changes in bloom toxicity can result from varying intracellular toxin production (toxin cell quota) or from changes in the abundance of toxic and non-toxic genotypes within the bloom population (Orr et al., 2018). Changes in intra- and extracellular toxin (microcystin (MC)) concentrations were demonstrated to result from highly variable transcription of toxin gene clusters (Wood et al., 2012). The co-occurrence of toxic and non-toxic genotypes was shown by the presence or absence of gene clusters coding for cyanobacterial toxins (Briand et al., 2009; Kurmayer et al., 2004; Vaitomaa et al., 2003). These findings suggested that classification of bloom subpopulations based on their intracellular oligopeptide composition, i.e. chemotypes, would allow for an even higher resolution of bloom dynamics (Agha and Quesada, 2014). Indeed, the co-occurrence of different chemotypes within natural blooms and their spatio-temporal variation was reported for Microcystis and Planktothrix blooms (Agha et al., 2014; Johansson et al., 2019; Lezcano et al., 2018), demonstrating the necessity of chemotyping for better understanding of short-term bloom dynamics.

Short-term dynamics could only be partially explained by environmental parameters, such as nutrient concentrations or temperature, known to drive long-term bloom dynamics. Thus, biotic factors, such as grazing, allelopathic and symbiotic interactions with heterotrophic bacteria and parasites (viruses (Gons et al., 2002), pathogenic bacteria, protists and fungi (Gerphagnon et al., 2015)), have come into focus as bloom influencing forces (Berry et al., 2017; Sommer, 2012; Steiner, 2017; Woodhouse et al., 2016). Fungi (Chytridiomycota; chytrids) are known to parasitize a wide variety of phytoplankton taxa, including filamentous cyanobacteria, e.g. Dolichospermum and Planktothrix (Gerphagnon et al., 2015; Rasconi et al., 2012; Sønstebø and Rohrlack, 2011). However, the extent to which chytrid parasitism can affect bloom dynamics, or even result in the disintegration of a bloom, is currently unresolved (Frenken et

MANUSCRIPT 3 al., 2017; Gleason et al., 2015). Reported chytrid infection intensities, leading to the disintegration of seemingly identical Dolichospermum macrosporum blooms, varied greatly: a 17 % prevalence of infection (PrF) with the chytrid Rhizosiphon crassum (R. crassum) was reported in Lake Aydat (France) by Gerphagnon et al. (2013); while a PrF of 98 % was reported by Rasconi et al. (2012) for a different Dolichospermum macrosporum bloom in the same lake. Differences in sensitivities to chytrid parasitism among Dolichospermum macrosporum phenotypes suggest differences at the level of the hosts defence reaction that are detectable only via genotyping or chemotyping approaches. Cyanobacterial toxins have been proposed to be part of the host’s defence reaction to fight back parasites (Gerphagnon et al., 2015; Gleason et al., 2015; Holfeld, 1998). To date, however, there is only one experimental study investigating toxin production in response to chytrid parasitism in cyanobacteria (Rohrlack et al., 2013). The authors of this infection study showed that several chytrid strains were less infective in a Planktothrix wild-type strain than in knock-out mutants incapable of producing known bioactive oligopeptides (MCs, microviridins, anabaenopeptins).

In this field study, aimed to address several of the outlined research gaps with regard to short- term bloom dynamics, a fungal infection of the Dolichospermum sp. bloom was observed right at the start of the field investigation. Accordingly, the influence of the observed chytrid parasitism was included in the holistic monitoring approach, to understand short-term bloom dynamics. Our monitoring approach included the recording of abiotic (water temperature, rainfall, water level fluctuation, nitrogen and phosphorus) and biotic parameters (chytrid infection intensity). In addition, cyanobacterial taxa descriptors (phenotyping, phylotyping based on 16S rRNA sequencing, toxin gene cluster detection) were recorded and toxin identification (MCs, anatoxin-a, anabaenopeptins) was carried out to allow distinction of different Dolichospermum chemotypes.

4.3. Material & Methods

4.3.1. Sampling location

Samples were taken from an oligo- to mesotrophic reservoir (Schwarzenbachtalsperre) located in the northern Black Forest (48° 39′ 17″ N, 8° 19′ 46″ E), Southwest Germany. The hydropower dam has a surface area of 0.66 km² and a storage volume of 14 Million m3. The Schwarzenbach reservoir is fed by natural tributaries as well as via water pumped from the water compensation reservoirs located in Forbach (48° 40′ 17″ N, 8° 21′ 16″ E) and supplied by the river Murg. In the early 2000, the Schwarzenbach reservoir experienced strong

MANUSCRIPT 3

Microcystis sp. blooms, which were replaced by intense, recurring Dolichospermum sp. blooms as of 2016 (Figure S 4-1).

4.3.2. Field sampling

Water samples were taken every two weeks from April to October 2018 at the centre of the Schwarzenbach reservoir approximately 200 m distance from the dam. This sampling position was chosen based on observations from previous monitoring campaigns, indicating bloom accumulation at this position within the lake as a result of wind and currents. Samples were taken at 1 m water depth, in respect to the water surface, using a water sampler. Samples were prefiltered through a 140 µm mesh to exclude larger zooplankton. 100 mL aliquots (n = 1 per sampling day) of the filtered samples were immediately fixed with Lugol’s solution for microscopic analyses, i.e. Dolichospermum quantification and assessment of infection intensities. Subsamples of 140-400 mL were taken for Illumina sequencing, toxin and nutrient analyses and filtered onto GF/F glass fibre filters (Whatman™, GE Healthcare Life Science, Chicago, USA). The filtrate was used for analysis of dissolved nutrients. All samples were stored at -20 °C until further processing.

4.3.3. Dolichospermum quantification

Dolichospermum taxa were identified based on morphological criteria (Komarek and Zapomelova, 2008, 2007). Technical triplicates of 10 mL were taken from the fixed water samples (see chapter 4.3.2.) and were left to settle overnight in an Utermöhl counting chamber. Subsequent counting was performed at 200x magnification with an inverted microscope (Zeiss, Axiovert 135). Mean number of filaments per mL and mean cell number per filament was determined for every sampling day. To enumerate filament number per mL, five diameter transects were counted within the counting chamber. Mean cell number per filament was determined for each sample by counting the cells of 20 randomly chosen filaments. Mean cell number per mL was calculated by multiplying the mean filament number per mL with the mean cell number per filament.

4.3.4. Genomic DNA extraction

GF/F filters were lysed with a FastPrep-24 Tissue and Cell Homogenizer (MP Biomedicals™, Eschwege, Germany) and DNA was extracted using a DNA

MANUSCRIPT 3

Clean&Concentrator™ Kit (Zymo Research, Freiburg, Germany). Extracted DNA was stored at -20 °C until further processing.

4.3.5. High Throughput Sequencing

Illumina™ sequencing was used to amplify the bacterial-specific V3-4 region of the 16S rRNA gene as well as fungi-specific internal transcribed spacer 2 (ITS2) region (Table S 4-1 Primer pairs used in the study.Table S 4-1). Locus-specific sequences of primers were modified to include Illumina™ adapters (Herlemann et al., 2011; Klindworth et al., 2013). First PCR reactions were performed in 35 µL volumes containing 1 U of Novagen® KOD polymerase (Merck, Darmstadt Germany), 3.5 µl buffer, 2.1 µg MgSO4, 3.5 µl dNTPs, 20.7 µl

H2O ultrapure, 2.1 µl BSA, 0.7 µl of each primer (0.2 µM) and 1 µl template DNA (15-45 ng DNA). The PCR mixture was cycled with: 95 °C for 2 min, followed by 29-31 cycles of 95 °C for 20 s, 61/65 °C for 30 s, 72 °C for 2 min, and a final extension of 72 °C for 7 min (Herlemann et al., 2011). Sufficient concentration of the first PCR product was confirmed using gel- electrophoreses. 20 µl of the first PCR products were submitted to Eurofins Genomics (Ebersberg, Germany), where the second PCR was performed, adding sequencing adapters and sample-specific indices to each amplicon. Amplicons were pooled into a single library and paired-end sequences (2 × 300) were generated on a MiSeq (v3 chemistry) instrument generating an average of 80,000-90,000 read pairs per amplicon. Sequence data were automatically demultiplexed and forward and reverse reads were assigned to samples.

Bioinformatic analysis of Illumina sequences Bioinformatic analysis of metabarcoding data was performed using QIIME 2 version 2019.4 (Bolyen et al., 2018). Raw sequence reads were demultiplexed and quality filtered using q2- demux plugin. Denoising of sequences was followed by chimera identification and removal using DADA2 (Callahan et al., 2016).

Taxonomic assignment of Illumina sequences For taxonomic assignment of quality-filtered amplicon sequences variants (ASVs), two different reference databases were used: For assignment of bacterial 16S rRNA sequences, the SILVA (Quast et al., 2013) QIIME 2 release (version 132) with the corresponding 99 % threshold level reference database was used. For assignment of fungal ITS2 sequences, the UNITE QIIME release for fungi (Nilsson et al., 2019; UNITE Community, 2019) was used, with

MANUSCRIPT 3 a reference database resulting from dynamic threshold clustering of species. Taxonomic classification was performed using the QIIME 2 implemented q2-feature-classifier-plugin (a scikit-learn naive Bayes machine-learning classifier) (Bokulich et al., 2018; Pedregosa et al., 2012). The classifier was pre-trained with the respective primer sequences prior to taxonomic assignment. For ITS2 sequences, the VSEARCH consensus taxonomic classifier (Rognes et al., 2016) was additionally used. Features that were assigned as unclassified or classified only to kingdom level were excluded from further downstream analysis. Rare taxa, defined as taxa accounting for less than 0.01 % of the community, were also excluded from further analysis. Analysis of sequences resulting from Illumina sequencing was performed using the phyloseq package (McMurdie and Holmes, 2013) of the R software (R Core Team 2020).

4.3.6. Detection of toxin genes

To detect genes involved in the biosynthesis of MCs and anatoxin-a, all samples were screened for the respective toxin gene clusters using PCR. As target regions, the anatoxin synthetase gene C (anaC) and the MC synthetase gene E (mcyE) were amplified. The primer pairs anaCF/anaCR (Shams et al., 2015) (Table S 4-1) were used for the amplification of the anaC gene and the strain Osc-193 (UHCC) served as positive control. The primer pair mcyE- F2/AnamcyE-12R (Vaitomaa et al., 2003) was used for amplification of the mcyE gene and the strain Anabaena-90 (Ana-90; UHCC) was used as a positive control. Amplicons were submitted for Sanger sequencing and resulting sequences were compared for sequence similarity using BLAST (Altschul et al., 1990) in the NCBI non-redundant nucleotide database (O’Leary et al., 2016). The Geneious software version 8.1.8 was used for all alignments.

4.3.7. Toxin quantification

Commercially-available enzyme-linked immunosorbent assay (ELISA) kits were used for the quantification of cell-bound MCs, anatoxin-a, and anabaenopeptins from water samples. MCs were additionally analysed by ultra-performance liquid chromatography - tandem mass spectrometry (UPLC-MS/MS) to determine congener profiles.

Toxin extraction Cell bound toxins were extracted from GF/F filters in three sequential extraction rounds. Filters were first incubated 30 min with 3 mL aqueous methanol (50 % v/v) and vortexed for 10 min followed by 15 min incubation in an ultra-sonic bath. Extracts were centrifuged and the supernatants were collected. The extraction procedure was repeated twice with the remaining

MANUSCRIPT 3 pellet yet without the initial incubation. The supernatants of all three extractions rounds were combined and the resulting 9 mL extract was dried overnight, using a speed-vacuum system (Univapo 100H). Samples were resuspended with 500 µL aqueous methanol (50 % v/v) and stored in glass vials at -20 °C until analysis.

Toxin quantification via ELISA Cell-bound MCs, anatoxin-a, and anabaenopeptins were quantified using commercial ELISA kits (520011, 520060, 520070, ABRAXIS Inc, Warminster, PA, USA), with a limit of detection (LOD) ≥ 0.1 ng mL-1 for MC and anatoxin-a, and ≥0.08 µg mL-1 for anabaenopeptin. Prior to analysis crude extracts were diluted 1:20 with manufacturer diluent solution to a final methanol concentration of 2.5 %. Samples were analysed in technical triplicates for every ELISA following the manufacturer’s standard protocol.

Microcystin quantification via UPLC-MS/MS MC quantification via UPLC-MS/MS was performed for 16 MC congeners (MC-RR, MC-YR, MC-LR, MC-FR, MC-WR, MC-RA, MC-Raba, MC-LA, MC-FA, MC-WA, MC-LAba, MC-FAba,

MC-WAba, MC-LY, MC-LF, MC-LW). Synthetic, deuterated MC-LR (D7-MC-LR) and MC-LF

(D5-MC-LF) were used as internal standards. The synthesis of D5-MC-LF (containing phenylalanine with a deuterated phenyl group (five D-atoms) in position 4) was published previously (Zemskov et al., 2017). D7-MC-LR was synthesized in analogy to the published synthesis of D5-MC-LF with leucine deuterated in the side chain (seven D-atoms) at position 2 (Altaner et al., 2019). Samples were analyzed using an UPLC-MS/MS system, consisting of an Acquity H-class liquid chromatograph equipped with an Acquity BEH C18 column (1.6 µm, 2.1 x 50 mm) and a guard column (kept at 40 °C) coupled to a Waters XEVO TQ-S mass spectrometer operating in the electrospray ionization mode. Solvents were 10 % ACN (solvent

A) and 90 % ACN (solvent B), both containing 100 mM CH2O2 and 6 mM NH3 (total flow of 0.4 mL min-1). The gradient started with 25 % B (held for 30 s) and the proportion of B then increased first to 45 % within 30 s, then to 60 % within 180 s, and eventually to 99 % within 12 s (held for additional 30 s). Re-equilibration back to 25 % B was achieved over 78 s, where it was held for another 60 s before the next sample was injected. The injection volume was 5 µL. Compounds were ionized using a capillary voltage of 3 kV and a nebulizer pressure of 7.0 bar. Dissolution was achieved using a nitrogen flow of 1000 L h-1 at 500 °C. Analysis of the congeners was divided into five analysis windows to reduce the number of parallel analyzed congeners in order to maximize the time spent scanning for each individual compound. MC congener concentrations in the analyte were converted into ng mL-1 water sample and into MC

MANUSCRIPT 3 cell quota using mean cell number mL- 1 (see chapter 4.3.3.). The LOD of the UPLC-MS/MS was ≥ 0.5 ng mL-1.

4.3.8. Fungal infection intensity

Fungal infection intensity of the Dolichospermum bloom was assessed microscopically for every sampling day from the Lugol’s fixed water samples (see chapter 4.3.2.). Infection intensity was calculated according to Bush et al. (1997) and expressed as prevalence of infection (PrF), which is the proportion of infected host individuals (Ni) in a given total host population (N): PrF (%) = Ni *N-1 * 100.

In analogy to Gerphagnon et al. (2013), the PrF was calculated based on the percentage of infected filaments from the total number of filaments. Filaments were defined as infected if any life stage of the chytrid life cycle could be observed attached to or inside of the filament. Life stages were defined as zoospores, buds, sporangia or papilla. Chytrid taxa were identified using recent morphological criteria (Rasconi et al., 2012).

4.3.9. Physical and chemical parameters

Water surface temperature and chlorophyll a concentrations were measured continuously from May to October with a multiparameter probe (CTD probe, RBR Ltd., Ottawa, Canada) placed at the sampling point (see chapter 4.3.2.). To derive water level fluctuations as a result of Schwarzenbach reservoir management (pumping, turbination), the water level was continuously measured with a pressure sensor located at 0.3 m above sediment (TDR- 2050, RBR Ltd., Ottawa, Canada). Daily mean rainfall data at the Freudenstadt weather station (situated approximately 22 km from the Schwarzenbach reservoir) were extracted from the Climate Data Center of the German Weather Service. Dissolved inorganic and organic particulate bound nitrogen (DIN and NPart) and soluble reactive and particulate bound phosphorus (SRP and PPart) were analyzed on a 2-week interval (see chapter 4.3.2.) from GF/F filters and filtrate (Basen et al., 2017).

4.3.10. Data analysis

Single and multiple linear regression models were used to test for association between bloom parameters and physicochemical parameters as well as association between toxin concentration and Dolichospermum morphotypes. Daily chlorophyll a concentration was correlated to daily water level fluctuation, water temperature and rainfall data. Total

MANUSCRIPT 3

Dolichospermum filament concentration and filament concentration of the different Dolichospermum morphotypes were correlated to DIN and SRP, DIN:SRP ratio, water level fluctuation, water temperature and rainfall data. The latter three parameters were averaged over a time span of seven days prior to the respective filament sampling date. To test for significant differences in toxin production and abiotic parameters between the first and second bloom phase, unpaired Student’s t-tests were performed, comparing total MC and anabaenopeptin concentration as well as DIN, SRP and water temperature. Furthermore, MC congeners and anabaenopeptin were correlated to filament concentration of the different Dolichospermum morphotypes. Regression models were performed using the lm() function of R version 3.6.1 (R Core Team, 2020).

GraphPad prism version 5.0 for Windows (GraphPad Software, San Diego California USA) was used for Student’s t-tests, all figure preparations and to calculate the standard error of the mean (SEM) for technical triplicates of PrF, microscopic filament counts and MC and anabaenopeptin concentrations determined by ELISA.

4.4. Results

4.4.1. Dolichospermum quantification and correlation to abiotic parameters

The Schwarzenbach reservoir experienced two cyanobacterial blooms in 2018 that were dominated by two different Dolichospermum morphotypes (Figure 4-1A). The first bloom, consisting of straight Dolichospermum filaments (hereafter straight bloom) (Figure 4-2A), persisted from May to June, with peak filament concentrations in June (541,571 ± 117,066 filaments L-1), and subsequently disappeared completely. Based on literature criteria, the straight morphotype was assumed to represent either D. heterosporum (Anabaena heterospora Nygaard), D. planctonicum (A. solitaria f. planctonica (Brunnthaler)) or D. smithii (A. solitaria f. smithii (Komarek)). The second bloom, consisting of coiled Dolichospermum filaments (hereafter coiled bloom) (Figure 4-2B), became prominent in July with peak filament concentrations in mid-September (>106 filaments L-1) and then declined end of October when filaments were undetectable. The coiled morphotype was assumed to represent either D. crassa (A. spiroides var. crassa (Lemmermann)) or D. circinale (A. circinalis (Rabenhorst ex Bornet & Flahault)). While low filament numbers of the coiled morphotype (Figure 4-2B) were also observed during the straight bloom, no filaments of the straight morphotype (Figure 4-2A) were detectable during the subsequent coiled bloom (Figure 4-1A). The two bloom phases were separated by a sudden bloom collapse at the beginning of July, when total filament numbers decreased to <16,000 filaments L-1.

MANUSCRIPT 3

Figure 4-1 Dolichospermum morphotypes and nutrient concentrations in the Schwarzenbach reservoir during the bloom season 2018. (A) Abundance of Dolichospermum morphotypes: Straight and coiled filaments L-1, prevalence of infection (PrF) of the straight Dolichospermum morphotype. Error bars indicate the standard error of the mean of technical replicates (±SEM, n = 3); (B) Nutrient concentrations: Dissolved inorganic nitrogen (DIN) and soluble reactive phosphorus (SRP). Samples taken from 1 m water depth.

Figure 4-2 The two different Dolichospermum morphotypes found in the Schwarzenbach reservoir during the bloom season 2018. Magnified images of heterocysts in insets. (A) Straight morphotype possibly representing D. heterosporum (A. heterospora Nygaard), D. plantonicum (A. solitaria f. planctonica (Brunnthaler)) or D. smithii (A. solitaria f. smithii (Komarek)); (B) Coiled morphotype possibly representing D. crassa (A. spiroides var. crassa (Lemmermann)) or D. circinale (A. circinalis (Rabenhorst ex Bornet & Flahault)). Samples taken from 1 m water depth.

SRP concentrations remained relatively constant (1.3-4.9 µg L- 1) throughout the growth period, except for the last three sampling dates when concentrations strongly increased to 69.0-86.0 µg L- 1 (Figure 4-1B). DIN concentrations showed great temporal variations throughout the whole season (2.5-99.0 µg L- 1). Most prominent were two peaks at the beginning of June (99.0 µg L- 1) and July (80.0 µg L- 1). DIN concentrations strongly decreased concomitantly with the strong increase of SRP (Figure 4-1B) at the end of the sampling period (last three sampling dates), which was accompanied by a decrease in total Dolichospermum

MANUSCRIPT 3 filament concentration (Figure 4-1A). DIN, SRP and water temperature did not significantly differ (p > 0.05) between the two different blooms phases (29/05 – 09/07 vs. 23/07 – 28/09).

Mean daily rainfall data showed only one strong rain event on July 7th with 58 mm  58 L m-2 (Figure S 4-2). Single and multiple regression models showed no significant correlation of total Dolichospermum filament concentration to any of the measured abiotic parameter (Table S 4-2, Figure S 4-5). Neither rainfall nor water level fluctuations were significantly correlated with total Dolichospermum filament concentration or chlorophyll a concentration (Table S 4-2, Figure S 4-5), suggesting that the observed Dolichospermum bloom dynamics were not affected by weather associated events or abiotic parameters. Only chlorophyll a concentration provided for a mathematically moderately positive correlation to water temperature (r2 = 0.522, p < 0.0001).

4.4.2. Bloom phylotype dynamics

In total, 14 different bacteria phyla were identified via Illumina sequencing throughout the growth period (Figure 4-3A). The four most abundant phyla were Cyanobacteria, Actinobacteria, Proteobacteria and Bacteroidetes. Relative abundance of the phylum cyanobacteria varied during the year (Figure 4-3A), with peaks in mid-June, August and mid- to end-September, corresponding to filament counts (Figure 4-1A). 18 different ASVs were assigned to seven cyanobacterial genera: Aphanizomenon, Cyanobium, Dolichospermum, Microcystis, Planktothrix, Pseudanabaena and Snowella (Figure 4-3B). Dolichospermum was the most abundant genus, accounting for >70 % of the cyanobacterial community in 10 out of the 12 samples. Only two sampling dates, i.e. at the bloom collapse beginning of July and at the final sampling date end of October, were characterized by a low relative abundance of cyanobacteria (Figure 4-3A), i.e. specifically of Dolichospermum (Figure 4-3B).

MANUSCRIPT 3

Figure 4-3 Bacterial community composition in the Schwarzenbach reservoir during the bloom season 2018. Relative abundances of the individual (A) bacteria phyla in the overall bacterial community and (B) cyanobacteria genera in the cyanobacterial community based on 16S rRNA reads. Samples taken from 1 m water depth.

Three individual ASVs were assigned to the genus Dolichospermum (Dolichospermum.NIES41), occurring in varying relative abundances throughout the growth period (Figure 4-3B). Classification further down to species level was not possible due to the fact that the straight and coiled Dolichospermum were not successfully grown as axenic cultures, thereby not allowing for full-genome sequencing. The available 407 bp 16S rRNA sequences of the ASVs available from Illumina sequencing were too short for distinctions at the species level. Therefore, these ASVs were designated as Dolichospermum1, Dolichospermum2 and Dolichospermum3 for further analysis. Global alignment of the Dolichospermum1/2/3 sequences showed very high sequence similarity with >99.5 % identical nucleotides (Table S 4-3). Dolichospermum3 showed two nucleotide mismatch positions when aligned to Dolichospermum1 and Dolichospermum2. Dolichospermum1 and Dolichospermum2 showed one mismatch position when aligned to each other. Sequences of Dolichospermum1/2/3 were additionally aligned against the NCBI nucleotide database for further identification at the species level. Due to high sequence similarity, the database search resulted in similar hits for all three queries. 100 % identity match and query cover were found for several Dolichospermum species, including D. planctonicum and D. circinale.

The ASVs of Dolichospermum1 and Dolichospermum2 were detected only during the straight bloom in June (Figure 4-1A, Figure 4-3B). In contrast, Dolichospermum3 ASVs were found throughout the year, but were low in relative abundance during the straight bloom (May - June). As of July, Dolichospermum3 ASV abundance increased, coinciding with the emerging coiled bloom. Dolichospermum3 ASVs represented <90 % of the total cyanobacteria community within the bloom peak (August - October), thus most likely reflecting the ASV of the coiled morphotype (Figure 4-2B).

MANUSCRIPT 3

4.4.3. Toxin and toxins genes

Cell-bound MC, whether determined via ELISA or UPLC-MS/MS, were detected primarily during the straight bloom and at the start of the coiled bloom (Figure 4-4). Total MC concentration did not differ significantly between the first and the second bloom phase (p > 0.05, Figure S 4-6). MCs were represented mostly by congeners MC-YR, MC-LR and MC-FR, as revealed by UPLC-MS/MS were predominant (Figure 4-4B). Other MC congeners appeared to be present, as suggested by the higher total MC values obtained with the ELISA, but could not be identified via UPLC-MS/MS due to the LOD for the identification of specific MC congeners (Figure 4-4A, Table S 4-4). MC-YR and MC-FR were primarily found during the first, straight bloom and their concentrations were positively correlated with filament concentrations of the straight morphotype (MC-YR: r2 = 0.841, p = 0.002, MC-FR: r2 = 0.795, p = 0.006). Although MC-LR predominated during the coiled bloom, no positive correlation was found with the filament concentration of the coiled morphotype (r2 = 0.086, p = 0.814).

Figure 4-4 Quantification of cell-bound cyanobacterial toxins from the Schwarzenbach reservoir during the bloom season 2018. (A) Microcystin and anabaenopeptin concentrations analyzed via ELISA; error bars indicate the standard error of the mean (SEM) of three technical replicates, asterisks refer to PCR detection of the mcyE gene. (B) Concentrations of microcystin (MC) congeners detected via UPLC-MS/MS and MC cell quota based on total MC (∑ of all MC congeners). Color shading indicates first bloom of straight Dolichospermum morphotype (blue), first (light red) and second phase (dark red) of the second bloom of the coiled Dolichospermum morphotype. Samples taken from 1 m water depth.

Total MC concentrations correlated well with the presence of Dolichospermum-specific mcyE (Figure 4-4A). Alignment of mcyE sequences from all samples (210-262 bp) showed a high nucleotide overlap of 205 bp. This consensus sequence searched against the NCBI nucleotide database detected the highest sequence similarity (100 % query cover, identity >99 %) with

MANUSCRIPT 3

Dolichospermum lemmermannii 66A mcyE gene (EU151888.1) and that of other Dolichospermum species (CP043056.1, CP003284.1, AJ536156.1). Anabaenopeptin concentrations were significantly higher in the first, straight bloom (p = 0.0016, Figure S 4-6) than in the second, coiled bloom, and showed a strong positive correlation with filament concentrations of the straight Dolichospermum morphotype (r2 = 0.729, p = 0.015; Table S 4-2). Anabaenopeptin concentrations remained just above the LOD (Figure 4-4A) in both phases of the second, coiled bloom. Neither the anaC gene nor anatoxin-a was detected in any of the samples, suggesting that anatoxin-a producing species were either absent or below the LOD throughout the 2018 field study. Cylindrospermopsin and saxitoxins were not analyzed in this study, however, both were analyzed during previous blooms (2011 & 2015) by a commercial water analysis company and found to be below the LOD.

4.4.4. Chytrid parasitism

Microscopical inspection revealed that the straight Dolichospermum morphotype was heavily parasitized with chytrids, whereas the coiled morphotype did not show any signs of chytrid infections. Illumina sequencing identified three major fungi divisions: Ascomycota, Basidiomycota and Chytridiomycota (Figure 4-5), with a total of 48 ASVs assigned to Chytridiomycota. No ASV could be assigned further to class level.

Figure 4-5 Fungal community composition from the Schwarzenbach reservoir during the bloom season 2018. Relative abundances of the individual fungal phyla in the overall fungal community based on ITS 2 (internal transcribed spacer 2) reads. Samples taken from 1 m water depth.

MANUSCRIPT 3

Based on morphological features, e.g. outgrowth of tubular rhizoids through cells, endobiotic thallus, epiphytic bud and pyriform sporangium (Figure 4-6), and the fact that only vegetative cells of Dolichospermum were infected, the chytrid parasite was assumed to represent R. crassum (Rasconi et al., 2012). The life cycle of the chytrid parasite in the Schwarzenbach reservoir samples was comparable to the life cycle of R. crassum described previously (Canter, 1954; Gerphagnon et al., 2013). The infection process comprised six stages: encystment, prosporangium, expansion, budding, mature and empty stages (Figure 4-6).

Figure 4-6 Chytrid (R. crassum) life cycle in the Schwarzenbach reservoir. Samples of the straight Dolichospermum morphotype: Free zoospores (A), encystment (B), prosporangium and expansion (C), budding and mature stage (D), empty stage (E), fragmentation of infected filament (F), complete lysis of infected filaments (G).

PrF values (46 ±4.6 % and 38 ±10.6 %) at the onset of the straight bloom (Figure 4-1A) showed that almost half of the straight Dolichospermum morphotype population was infected. Most infected filaments showed multiple infection sites (Figure S 4-4). In the declining phase of the straight bloom, the relative number of infected filaments decreased to 0 % (Figure 4-1A). Expansion of the fungal rhizoid within the infected filament were observed to stop once the infection reached a heterocyst (Figure 4-7). Zoospores were not observed to attach directly to heterocysts nor were any infected heterocysts observed. Infected filaments with heterocysts showed an asymmetric infection pattern, with healthy vegetative cells on one side and infected degraded cells on the other (Figure 4-7), suggesting that heterocysts present a natural barrier for within filament chytrid expansion.

MANUSCRIPT 3

Figure 4-7 Asymmetric infection of straight Dolichospermum morphotype by the chytrid R. crassum.

4.5. Discussion

4.5.1. Morphotype, phylotype and toxin dynamics

The sequential appearance of Dolichospermum morphotypes in the Schwarzenbach reservoir suggested the presence of three partially overlapping individual blooms (Figure 4-1A, Figure 4-3B, Figure 4-4). The first, straight bloom (Dolichospermum1 and 2) was genetically and morphologically different from the second, coiled bloom (Dolichospermum3). Only the straight bloom was massively parasitized by R. crassum (Figure 4-6, Figure 4-7) and characterized by the presence of cyanobacterial toxins (MCs and anabaenopeptins, Figure 4-4). During the first phase of the coiled bloom (Dolichospermum3), the mcyE gene and MCs (Figure 4-4A) were also present, while anabaenopeptins were not detectable. MC congener abundance also varied between bloom phases. While the straight morphotype primarily produce MC-YR and MC-FR, the coiled morphotype produced MC-LR (Figure 4-4B). Moreover, the MC cell quota appeared to be much higher in the coiled than in the straight Dolichospermum morphotype. No toxin was found throughout the second phase of the coiled bloom, suggesting the presence of two different coiled Dolichospermum strains in the first and second phase of the coiled bloom, albeit indistinguishable with the phylotyping approach employed here (Figure 4-3B). This interpretation is in line with previous reports, showing a sequence of different chemotypes, despite >99 % 16S rRNA gene sequence similarity, within a Planktothrix (Rohrlack et al., 2009) and Microcystis bloom (Lezcano et al., 2018).

Seasonal successive dominance of different morphotypes in natural lakes was previously described for Microcystis (Xiao et al. 2018) as a result of changing environmental conditions. In contrast to the findings by Xiao et al. (2018), environmental parameters of the

MANUSCRIPT 3

Schwarzenbach reservoir, e.g. SRP, DIN, and weather events, usually associated with bloom dynamics, failed to explain the Dolichospermum bloom dynamics observed, i.e. the morphotype succession and the sudden collapse of the bloom in July (Table S 4-2). It is to be noted however that we cannot exclude an influence of CO2, as our CO2 measurements lacked robustness to allow any correlation with the two different blooms. The increase of SRP and decrease of DIN for the last three sampling dates was most likely a result of the fall turnover, i.e. the mixing of the surface water with deeper layers of the Schwarzenbach reservoir. The latter, however, does not explain the strong variability of DIN throughout the 2018 bloom season (Figure 4-1). Previous monitoring campaigns had shown that nutrient inputs from natural tributaries of the Schwarzenbach reservoir were negligible (LUBW, 2010). Thus, the variability in nutrient concentrations in the Schwarzenbach reservoir is most likely the direct effect of water pumped up into the Schwarzenbach reservoir from compensation reservoirs fed by the river Murg in the valley (LUBW, 2010).

Heavy parasitism by R. crassum appeared to be an important factor driving the straight Dolichospermum bloom dynamics. This was also suggested earlier by observations by Gerphagnon et al. (2015) and Rasconi et al. (2012). The concomitant increase in MCs and anabaenopeptins concentrations with decreasing chytrid infection during the straight bloom (Figure 4-4) may reflect a defence reaction of the straight Dolichospermum morphotype towards fungal parasitism. This interpretation is supported by several reports indicating that cyanobacterial toxins produced may have properties that could harm chytrids (see chapter 4.5.2.).

4.5.2. Cyanobacterial toxins as potential defence compounds against fungal infections

Chytrids express ser/thr protein phosphatase (PP) 1 and 2A (NCBI: XP_006675182.1 and ORY49437.1), most likely as critical regulators of cell growth, differentiation, apoptosis, motility, DNA damage response and cell cycle progression, as was demonstrated for other eukaryotes (Shi, 2009). More importantly, PP2A, as a component of the STRIPAK-like complex (Ariño et al., 2019; Hwang and Pallas, 2014), appears to play a role in the virulence of some plant pathogenic filamentous fungi. The potential role of PPs in modulating the virulence of chytrids has not been investigated yet.

MCs are highly specific and effective PP inhibitors (Runnegar et al., 1995). MC-LR, for instance, was demonstrated to be a much more potent PP inhibitor of human PPs than MC-

MANUSCRIPT 3

YR (Altaner et al., 2019). Assuming similar sensitivity of chytrid PPs to MC -mediated inhibition as observed for human PPs (Altaner et al., 2019), MC-LR primarily produced by the coiled morphotype could provide for a more efficient protection against chytrid infection than MC-YR of the straight morphotype. However, the potential inhibition of chytrid PP1 and 2A by MCs has not been studied yet. Sequence comparison of the catalytic subunits of chytrid PP1 and PP2A with mouse and human PP1 and PP2A revealed a >85 % amino acid sequence homology (Table S 4-5). The latter finding and the fact that PPs are highly conserved (Shi, 2009) and inhibited by MCs across eukaryotes (Fontanillo and Köhn, 2018) strongly suggests that chytrid PP2A activity could also be affected by MCs. This line of thought supports the hypothesis that MCs could serve as defence compounds against chytrid infection, as inferred already from earlier studies on Planktothrix (Rohrlack et al., 2013; Sønstebø and Rohrlack, 2011). Moreover, anabaenopeptins, beyond inhibiting PPs, have also been shown to effectively inhibit proteases (Spoof et al., 2015), thus potentially also inhibiting metalloproteases of chytrids expressed during host colonization and expansion (Farrer et al., 2017). Considering that PPs and proteases are critical players within the fungal mechanism of infection and host cell colonization, it appears reasonable to assume that their inhibition by MCs and anabaenopeptins could diminish the susceptibility of a host to chytrid infection (Farrer et al., 2017). However, polypeptide toxin synthesis may come at the cost of lower competitiveness, as is discussed below (see chapter 4.5.4.).

4.5.3. Morphotype dynamics in the context of chytrid infection

The fact that coiled Dolichospermum filaments were not affected by chytrid parasitism, despite lower toxin production, suggests that the coiled morphotype offers some protection compared to the straight morphotype. To our knowledge this is a unique observation as there are no previous publications, whether from field observations or laboratory experiments, demonstrating increased protection of coiled morphotypes against chytrid parasitism when compared to straight morphotypes. Although we could speculate that coiling results in physical hindrance of chytrid penetration of individual cells of the filament, resolution to this question would require culturing of the coiled and straight form in the laboratory for chytrid infection experimentation. It is also possible that the difference in filament morphology is associated with differences in cell wall composition. Cyanobacteria cell walls are known for their thick peptidoglycan and mucilage layers which may differ in thickness and composition between different species (Hoiczyk and Hansel 2000). Lower chytrid infection of the coiled morphotype could therefore also be attributed to thicker cell walls, protecting from penetration of zoospores into host cells.

MANUSCRIPT 3

Besides Dolichospermum, several other phytoplankton taxa, including the cyanobacteria Planktolyngbya, Raphidiopsis (former Cylindrospermopsis) and Pseudanabaena, the green alga Gloeotila, and the diatom Aulacoseira granulate (Padisák et al., 2003), can occur as coiled and straight morphotypes. Coiling may also protect against rotifer grazing, as observed in a Planktothrix bloom consisting of straight and coiled morphotypes (Fabbro and Duivenvoorden, 1996).

4.5.4. Selective parasitism as a potential driver of seasonal bloom dynamics

Selective parasitism on a predominant species has the potential to affect the seasonal succession of phytoplankton (Donk and Ringelberg, 1983) and heterotrophic bacteria (Thingstad and Lignell, 1997). Thingstad and Lignell (1997) named this concept ‘kill the winner’ hypothesis, which describes pathogens to infect the most abundant species, thereby steering interspecific competition, ultimately leading to a higher plankton diversity (Sommer, 2012). Thus, selective parasitism of the straight Dolichospermum morphotype by chytrids is suggested to favour the growth of the non-parasitized coiled morphotype. Why the coiled morphotype was not parasitized by chytrids, despite the omni-presence of chytrid ASVs in the water column remains unclear. Dolichospermum spp. are capable of producing a variety of other cyanotoxins, including cylindrospermopsins and saxitoxins (Li et al., 2016; Welker and Döhren, 2006). The toxin analyses presented here were focused on MCs, anatoxin-a and anabaenopeptins. Beside these toxins, cyanobacteria produce an incredible diversity of hundreds of less investigated cyanopeptides, including aeruginosins, cyanopeptolins, microginins, microviridins and aerucyclamides, (Janssen, 2019; Welker and Döhren, 2006). Some of the latter toxins were shown to occur in surface waters just as frequently as MCs and were demonstrated to inhibit proteases in the nanomolar range (Janssen, 2019). It however remains to be experimentally demonstrated, whether or not any of these cyanopeptides are capable of inhibiting chytrid proteases and thus could be critical for the protection of cyanobacteria from chytrid infection.

The maximum infection rate of 46 ± 4.6 % (Figure 4-1A) observed in this study is much higher than most of the infection rates reported in the literature resulting in the demise of whole populations (Gerphagnon et al., 2017). Therefore, it is highly likely that the high intensity of chytrid parasitism found in our study presumably was causal for the complete bloom collapse observed in July in the Schwarzenbach reservoir. The latter finding and the fact that environmental parameters were not able to explain the observed Dolichospermum bloom

MANUSCRIPT 3 collapse suggest that chytrid parasitism and competition with other morphotypes are major factors driving short-term bloom dynamics.

4.6. Acknowledgements

We thank the Institute for lake research (ISF Langenargen) of the State Office for the Environment, Measurements and Nature Conservation of the Federal State of Baden- Württemberg (LUBW) for providing field data from 1990-2015. Furthermore, we thank Desiree Helmer and Jorge Encinas (Limnological Institute, University of Konstanz) for support in field sampling, Aswin Mangerich and Sarah Krassnig (Molecular Toxicology, University of Konstanz) for UPLC-MS/MS support. The work was financially supported by the Ministry of Science, Research and the Arts of the Federal State Baden-Württemberg, Germany (grant: Water Research Network project: Challenges of Reservoir Management - Meeting Environmental and Social Requirements). While the German Research Foundation (DFG, grant INST 38-537-1) provided funding for the UPLC-MS/MS.

4.7. Supplemental Information

Table S 4-1 Primer pairs used in the study. Primers used for Illumina sequencing are indicated in italics.

Amplicon size Target region/gene Primer Primer sequence (3'-5') Original publication (bp) V3-4 region of 16S 341F_ill ACT CCT ACG GGA GGC AGC AG Modified after Herlemann et al. 500 rRNA 802R_ill GGAC TAC HVG GGT WTC TAA T 2011 ITS2f GCATCGATGAAGAACGCAGC ITS2 500 White et al. 1990 ITS2r TCCTCCGCTTATTGATATGC mcyE_F2 GAAATTTGTGTAGAAGGTGC mcyE 247 Vaitomaa et al. 2003 AnamcyE_12R CAATCTCGGTATAGCGGC anaC_F TCTGGTATTCAGTMCCCTCYAT AnaC 366 Shams et al. 2014 anaC_R CCCAATARCCTGTCATCAA

MANUSCRIPT 3

Table S 4-2 Spearman’s rank correlation coefficients between abiotic and biotic factors.

Variable1 Variable 2+ r2 P-value Chlorophyll a water temp. 0.522 < 0.0001 Chlorophyll a daily rainfall 0.001 0.991 Chlorophyll a water level fluc. 0.040 0.592 Chlorophyll a water temp + daily rainfall + water level fluc. 0.274 < 0.0001

Total filaments L-1 water temp. 0.073 0.831 Total filaments L-1 DIN 0.217 0.499 Total filaments L-1 DRP -0.172 0.593 Total filaments L-1 mean rainfall -0.210 0.514 Total filaments L-1 water level fluc. -0.175 0.588

MC Total Straight filament L-1 0.656 0.039 anabaenopeptin Straight filament L-1 0.729 0.015 MC Total Coiled filament L-1 -0.358 0.313 anabaenopeptin Coiled filament L-1 -0.485 0.156 MC Total Straight filament L-1 + Coiled filament L-1 -0.234 0.865 anabaenopeptin Straight filament L-1 + Coiled filament L-1 0.668 0.009

MC-YR Straight filament L-1 0.841 0.002 MC-LR Straight filament L-1 -0.378 0.282 MC-FR Straight filament L-1 0.795 0.006 MC-YR Coiled filament L-1 -0.551 0.104 MC-LR Coiled filament L-1 0.086 0.814 MC-FR Coiled filament L-1 -0.765 0.010

MC-YR Straight filament L-1 + Coiled filament L-1 0.440 0.055 MC-LR Straight filament L-1 + Coiled filament L-1 -0.206 0.799 MC-FR Straight filament L-1 + Coiled filament L-1 0.473 0.044

Table S 4-3 Nucleotide sequences (407 bp) of the three Dolichospermum ASVs. Based on Illumina sequencing of the 16S rRNA reads taken from water surface samples during the Schwarzenbach reservoir bloom season 2018.

MANUSCRIPT 3

Table S 4-4 Microcystin (MC) congener concentrations in analytes determined using via UPLC-MS/MS. Concentrations are given in ng mL-1, concentrations below limit of detection (0.5 ng mL-1) are indicated as

Date (dd/mm/yy) MC-RR MC-YR MC-LR MC-FR MC-WR MC-RA MC-RAba MC-LA MC-FA MC-WA MC-LAba MC-FAba MC-WAba MC-LY MC-LF MC-LW D7-LR D5-LF

20/05/18

04/06/18

14/06/18

09/07/18

23/07/18

06/08/18

20/08/18

07/09/18

17/09/18

MANUSCRIPT 3

Table S 4-5 Alignments of chytrid protein phosphatase (PP) 1 and 2A with human and mouse PP1 and 2A.

MANUSCRIPT 3

Figure S 4-1 Total phosphorous (TP), total nitrogen (TN) and TN to TP ratio in the Schwarzenbach reservoir from 1990 to 2018 and the observed periods with cyanobacterial blooms.

Figure S 4-2 Mean daily precipitation (blue) (mm ≡ L m-2), water temperature (red) and Chlorophyll a concentrations (green) at the Schwarzenbach reservoir 2018.

MANUSCRIPT 3

Figure S 4-3 Particulate bound P (PPart) and N (NPart) concentrations in water surface samples of the Schwarzenbach reservoir in 2018.

Figure S 4-4 Infection of Dolichospermum filaments: Single (A) and multiple (B) attached zoospores.

Figure S 4-5 Water depth (solid line) daily water level fluctuation (dashed line) in the Schwarzenbach reservoir 2018.

MANUSCRIPT 3

Figure S 4-6 Results of the unpaired Student’s t-test of mean total microcystin (MC) and anbaenopeptin (AP) concentrations between the first and the second bloom phase; * p < 0.05.

MANUSCRIPT 4

5. MANUSCRIPT 4

Interdisciplinary Water Reservoir Management

M. Daus1, K. Koberger1, K. Koca2, F. Beckers2, J. Encinas-Fernandes3, B. Weisbrod4, D. Dietrich4, S. Gerbersdorf2, R. Glaser1, S. Haun2, H. Hofmann3, D. Martin-Creuzburg3, F. Peeters3, S. Wieprecht2

1Physical Geography, Albert-Ludwigs-University Freiburg, Schreiberstrasse 20, 79098, Freiburg, Germany

2Institute for Modelling Hydraulic and Environmental Systems, Pfaffenwaldring 61, 70569, Stuttgart, Germany

3Limnological Institute, University of Konstanz, Mainaustrasse 252, 78464 Konstanz, Germany

4Human and Environmental Toxicology, University of Konstanz, Universitätsstrasse 10, 78464, Konstanz, Germany

5.1. Abstract

Water resources, especially freshwater, are amongst of the most important resources for humankind. The available resources have been managed for centuries to ensure accessibility and continuous use of water over the entire year. Reservoirs are a common way to store and retain water enabling a multitude of purposes like storage of drinking and irrigation water, recreation, flood protection, navigation and hydro energy production and are being built since centuries. Today, few reservoirs serve only one purpose, thereby requiring a management of demands and interests present. Since the projects will cause negative impacts alongside desired advantages both on a local and global scale, it is even more urgent to develop a common management framework that is able to mitigate negative impacts, incorporate different demands and make them visible within the discourse in order to avoid conflicts from early on. The scientific publications on reservoirs are manifold, yet a comprehensive and integrative information tool is not available. Therefore, the tool presented herein is based on the results from the CHARM (CHAllenges of Reservoir Management) project as well as the condensed outcome of relevant literature to aid and enhance knowledge in reservoir management. The project focused on five different aspects related to existing reservoirs in southern Germany, namely: sedimentation of reservoirs, biostabilization of fine sediments, toxic cyanobacteria(l) (blooms), greenhouse gas emissions from reservoirs and social contestation, respectively consent. These five research foci contributed to the topics and

MANUSCRIPT 4 setup of an information tool which can be found alongside this publication to provide insights for experts and laymen.

5.2. Introduction

Water and electrical energy are two of the most important resources on our planet. Dams and reservoirs provide and store both resources based on the implementation of infrastructure to retain water. Due to population growth and changes in lifestyle the demands on water bodies are constantly rising and diversifying. This incorporates interests like drinking and irrigation water supply, flood protection, energy production or navigation (Glaser, 2014, DPA [05.07.2019]). It is estimated that around 58,000 large dams are built, planned or under construction worldwide, a trend which will likely accelerate in the future (Mulligan et al., 2020). In Germany, dam construction with an emphasis on hydropower began around 1900 and reached its peak during the second half of the last century (Blackbourn, 2007). At present, there are 371 reservoirs that meet the ICOLD (International Commission on Large Dams) definition of large dams (more than 15 m dam height or 3 million cubic meter storage capacity) (DTK, 2013). The construction of dams and reservoirs does not only radically change the area directly affected by the artificial lake itself, but the entire runoff-system as well as the catchment area (Döll et al., 2009). It will cause changes in the flow regime, sediment deposition in the reservoirs, accumulation of nutrients in the (impounded) water body that may foster cyanobacterial blooms as well as methane and carbon dioxide emissions from the water column and possibly lead to conflict in the local communities to name but a few (Beckers et al., 2018). These issues in reservoir management have motivated the project team of CHARM (CHallenges of Reservoir Management) with a consortium comprised of the Universities Stuttgart, Konstanz and Freiburg in Baden-Württemberg (Germany) to explore these challenges and positively contribute to future demands regarding managing (large) water reservoirs.

5.3. Sediments

The global construction of approx. 58,000 large dams (Mulligan et al. 2020; ICOLD, 2019), has considerably influenced delivery of sediment from land to oceans by fragmenting the river network (Nilsson et al., 2005) resulting in an estimated over 100 Gt sediment trapped in reservoirs (Syvitski et al., 2005). When a river enters a reservoir, sediments carried by the river settle due to decreasing flow velocities and turbulence levels. Consequently,

MANUSCRIPT 4 sedimentation reduces reservoir storage that was designed for water and thus reduces the services of the reservoirs. Due to the binding properties of fine sediments (large surfaces compared to their volume) coupled with biostabilization (see Section 4), pollutants, nutrients and organic matter also accumulate in the reservoirs (Pohlert et al., 2011; Majerovà et al.,

2018) impairing water quality (see Section 5) and leading to emissions of CO2 and CH4 (see Section 6). Additionally, since incoming and outgoing sediment loads are different, hydrology, morphology and ecology of downstream channels are also negatively impacted (Graf, 2006; Magilligan and Nislow, 2005; Wu et al., 2019; Collier, 2000; Vörösmarty et al., 2003, 1997). It is estimated that approximately 40 % of the global river discharge have been blocked by large reservoirs with a storage capacity greater than 0.5 km3, yielding an estimation of 4-5 Gt sediments trapped in reservoirs annually. Mahmood (1987), Yoon (1992) and Bruk (1996) estimate a reservoir storage reduction by 1 % per year. Coupling this information with current storage of major dams (6863.5 km3) and assuming a dry bulk density of 1200 kg/m3, an annual storage loss is calculated to be ~82 Gt yr-1, which is approx. 20-fold larger than the estimation by Vörösmarty et al. (2003). It must, however, be noted that storage loss is not equally distributed across the world since the quantity of sediments deposited in the reservoirs depends strongly on physical (e.g. shape) and operational characteristics of the reservoir (e.g. pumping) as well as the characteristics of the catchment (e.g. slope, hydrology) (Haun et al., 2013). In addition, hydrological variability as a result of climate change (Haddeland et al., 2014) and human interactions may also increase sediment loads, leading to increased reservoir sedimentation. Therefore, development and implementation of sustainable sediment management techniques to maintain sufficient reservoir storage over the long term has become increasingly important in meeting rising energy and water demands. Traditionally, reservoirs are designed for a lifetime of 50 to 100 years (Annandale 1987; Morris and Fan, 1998), with abandoning or decommissioning the reservoir in mind, leading to societal, environmental and economic problems beyond the designed lifetime. Often the dead storage volume accounts for sedimentation. However, depending on the given boundaries (e.g. sediment yield, shape and volume of the reservoir), reservoir sedimentation may claim more volume than the designed dead storage volume and an active reservoir management becomes necessary. Indeed, based on the estimated approx. 69 km3 annual storage loss, 30 % and 80 % of the global reservoir capacity will be lost by 2050 and 2100, respectively, if the reservoirs are managed ineffectively. Therefore, particularly in views of competing land uses as well as pushing social (see Section 7) and environmental issues, it is critical that already during the design and construction phase of new reservoirs sediment management strategies and facilities are planned considering long-term reservoir sedimentation to ensure that the implemented measures are most effective and sustainable over a long-term period (>100

MANUSCRIPT 4 years). Furthermore, existing reservoirs must be adapted to sustainable management practices as much as it is possible (Annandale 1987). Since the 1990s several guidelines have been released related to sediment management in reservoirs (e.g. Morris and Fan, 1998; Batuca and Jordaan, 2000). These documents provide specific standards to evaluate the feasibility of different sediment management strategies (Atkinson and Shen 1999; White, 2001). As each reservoir is unique, the final decision on an appropriate management strategy must be taken with care. Sediment management strategies to reduce reservoir sedimentation and maintain reservoir volume can be summarized under three main approaches (Kondolf et al., 2014): - reducing the sediment yield originating from the catchment by controlling erosion upstream - minimizing the sediment depositions in the reservoir by managing flows during periods of high flows - recovering already lost reservoir volume by removing deposited sediment applying various techniques.

5.4. Biostabilization of fine sediments

The predominant microbial lifestyle in aquatic ecosystems is characterized by multicellular and multispecies communities (Fischer and Pusch, 2001; Flemming and Wingender, 2010; Geesey et al., 1978; Lozupone and Knight, 2007), flourishing between their self-synthesized three-dimensional matrix of extracellular polymeric substances (EPS) (Flemming and Wingender, 2010; Stoodley et al., 2002). These diverse microbial communities are capable of colonising various solid-water interfaces, with sediment being excellent substrata (Battin et al., 2016; Gerbersdorf and Wieprecht, 2015) and possess common features that distinguish them as biofilm (Flemming, 2020; Flemming and Wingender, 2010; Gerbersdorf and Wieprecht, 2015). The transition of microorganisms from planktonic to biofilm lifestyle is controlled by a range of environmental conditions among which the local hydrodynamics are of paramount importance (Berke et al., 2008; Gerbersdorf and Wieprecht, 2015). Surface-attached microorganisms deactivate the expression of the genes involved in motility and activate the genes involved in adhesion and biofilm development (Tuson and Weibel, 2013). Surface association and subsequent biofilm formation provides these microorganisms with critical advantages. As opposed to a single planktonic lifestyle, enhanced availability of essential nutritional resources, increased interactions of microorganisms within the surface associated biofilm-matrix and distinct, mostly diverse compositions lead to higher metabolic activity. This results in high survival, wide tolerance to environmental conditions and higher reproduction potential to the embedded microorganisms (Dang and Lovell, 2015;

MANUSCRIPT 4

Flemming and Wingender, 2010). The community composition and productivity of the microorganisms are continuously controlled by various reciprocal interactions between chemical, biological and physical (e.g. hydrological) factors (Gerbersdorf and Wieprecht, 2015). The operation of reservoirs can lead to significant accumulations of fine sediment and nutrients, which can result in formation and development of biofilms on the sediment surface. Microorganisms and their metabolic activities support fundamental ecological and biochemical processes, e.g., biodegradation of organic matter and toxins, water self-purification and the cycling of nutrients (Battin et al., 2016; Madsen, 2011; Nicolella et al., 2005; Schultz and Urban, 2008; Shannon et al., 2008), and the community composition eventually determines their ecological and environmental functions (Danovaro and Pusceddu, 2007; Foshtomi et al., 2015; Gilbertson et al., 2012). The microorganisms settled on fine sediments are also known to glue sediment grains together through EPS matrix (Jones, 2017; Paterson et al., 2018) and permeate their void space, which can, in turn, alter sediment properties, e.g., density, morphology, size gradation, architecture (Fang et al., 2012; Gibbs, 1983; Huiming et al., 2011; Shang et al., 2014) erosion and transport behavior of sediments (Banasiak et al., 2005; Chen et al., 2017; Droppo et al., 2015; Fang et al., 2016, 2015, 2017; Gerbersdorf et al., 2008; Malarkey et al., 2015; Righetti and Lucarelli, 2010; Vignaga et al., 2013) and associated contaminants (Burns and Ryder, 2001; Förstner et al., 2004). The ability of biofilms to increase erosion thresholds by biological actions is named “biostabilization" (de Brouwer et al., 2005; Passarelli et al., 2014; Paterson et al., 2018) and has been reported to mediate the cycle of sediment erosion, transportation, deposition and consolidation (ETDC) in aquatic ecosystems, including reservoirs (Fang et al., 2015). Although the importance of biostabilization at the sediment surface and deeper layers has been increasingly recognized through laboratory and field studies (Chen et al., 2017; Paterson et al., 2018 (and references therein) over the past two decades, yet little is known about the transport processes of biofilm-bound sediment and its effect on aquatic environment and bed morphology (Gerbersdorf et al., 2020). Thus, sediment transport models as well as river and reservoir management strategies disregard and therefore underestimate the effect of biofilm growth on sediment.

5.5. Cyanobacteria

Reservoirs and dams used for hydropower or drinking water supply are subject to major anthropogenic and environmental influences, potentially favoring cyanobacterial harmful algal blooms (cHABs). Pelagic and benthic cyanobacteria are known to produce a large variety of toxins (e.g., microcystins, saxitoxins, anatoxins), frequently resulting in human and animal poisoning events (Dietrich et al., 2008). There is consensus that increased nutrient input

MANUSCRIPT 4

(eutrophication) in conjunction with increasing surface water temperatures promoting water column stratification, are closely linked to more frequent and pronounced cHABs (O'Neil et al., 2012; Paerl and Huisman, 2008; Quiblier et al., 2013). Nutrients that have been mainly linked to the occurrence of eutrophication associated cHABs are phosphorus (P) and nitrogen (N). However, past decades of research revealed that blooms are not restricted to eutrophic or hypertrophic systems, but also occur in nutrient poorer (oligo-mesotrophic) lakes and reservoirs and can even be promoted by oligotrophication (Callieri et al., 2014; Ernst et al., 2009; Posch et al., 2012; Salmaso, 2010; Salmaso et al., 2012).

5.5.1. Which species form blooms?

It is essential to realize that cyanobacteria are not a homogenous group when trying to understand cHAB dynamics and bloom response to mitigating or promoting factors and management strategies. Therefore, the first step of any bloom management should be a careful analysis of the in situ situation, including the identification of the predominant bloom species. cHABs forming cyanobacteria can be clustered into two groups based on their nutrient preferences and their likelihood to form blooms: Cyanobacteria genera typically forming surface blooms in N-limited systems are N2(dinitrogen)-fixing taxa (Dolichospermum, Cylindrospermopsis, Aphanizomenon, Lyngbya and Nodularia) as they can compensate for N limitation by using atmospheric N2. Planktothrix blooms are typically found in mesotrophic lakes (Ernst et al., 2009), where they produce deep chlorophyll maxima. Genera typically occurring in eutrophic, P- and N-rich systems are Microcystis and Cylindrospermopsis (Dolman et al., 2012; Paerl and Otten, 2013).

5.5.2. Management tools

Several studies have summarized cHAB monitoring and management strategies (Chorus and Bartram, 1999; Ibelings et al., 2014; Paerl et al., 2016). The suitability and success of the respective strategy might vary and should be considered carefully on a case- by-case basis (O'Neil et al., 2012; Paerl and Otten, 2013). Nutrient control and reduction is still considered to be the most efficient management tool. Here again, a detailed analysis of the in situ situation should be the first step. This includes a characterization of the reservoir by its trophic level index and the identification of nutrient sources and their pathways. Nutrients can enter the system via point sources like natural and artificial tributaries or water pumping storages (especially in the case of hydropower reservoirs) or can originate from di use sources

MANUSCRIPT 4 from the catchment area. The German environment agency (Umweltbundesamt (UBA)) developed a decision support system for reservoir managements (https://toxic- cyanobacteria.com/), which summarizes background information and questionnaires linked with cyanobacterial blooms and toxins. This tool might deliver an overview of the situation and help to develop a water safety plan. Besides the described direct control of nutrient input, a reduction might also be achieved via sediment dredging/flushing and binding or precipitation of nutrients via occulating agents, these invasive strategies should however be considered carefully, as they often come with financial and ecological costs (Bullerjahn et al., 2016; Paerl and Otten, 2013). Finally, physical factors are often manipulated to mitigate cHABs. Physical measures include disintegration of vertical stratification and increased flushing rates in reservoirs (Paerl et al., 2016). Vertical mixing devices to break down stratification have been successfully applied in small lakes and reservoirs (Paerl and Otten, 2013). The efficacy and applicability of these physical manipulation primarily depends on the size and purpose of the reservoir, the predominant bloom species and should be applied in parallel with nutrient control (Paerl et al., 2016).

5.6. Greenhouse gas emission

Dammed systems release substantial amounts of greenhouse gases (GHG) that contribute considerably to the global budget of GHG (Barros et al., 2011; Beaulieu et al., 2014),

Hertwich, 2013; Luyssaert et al., 2012; Raymond et al., 2013). Especially CH4 is of major

-1 concern with reservoirs emitting 8.9-22.2 t yr CH4 globally (Deemer et al., 2016). GHG emissions depend on the location and construction of the reservoir and on its management. The conditions in the catchment determine influx of nutrients and the amount of organic carbon that is eventually stored in the sediments. The meteorological conditions affect the strength of stratification, the duration of the season, the intensity of vertical mixing in the water column and of the gas exchange at the water surface. The morphometry of the reservoir together with meteorological conditions and amount of organic material in the sediments determines, whether anoxic conditions develop in the deep water which support accumulation of CH4 and

CO2 and eventually lead to GHG emission during fall overturn, or to dam-downstream emissions if the outlet of the reservoir is located in the anoxic layer. Reservoir management may alter the efficiency of different GHG emission pathways. The choice of the depth and the temporal sequence of water influx and withdrawal can have substantial effects on GHG emissions. Continuous deep-water withdrawal may remove GHG and nutrients released from the sediments to dam-downstream. Nutrient removal may result in reduced within-system production, thus in a smaller accumulation of organic material in the sediments and

MANUSCRIPT 4 consequently a reduced GHG production. The removal of cold deep-water reduces stratification, thus supports vertical mixing of e.g., oxygen. In contrast, withdrawal near the water surface may directly remove algae and cyanobacteria from the system and thus reduce organic material, but favors development of strong stratification with reduced mixing. The latter increases the risk of anoxia and GHG accumulation in the deep water. In case water removal is not continuous but operated as a single major drawdown of the reservoir, the pressure release on the CH4 saturated pore-water causes formation of CH4 bubbles and thus a substantial increase in CH4 emissions via ebullition (Engle and Melack, 2000; Deborde et al., 2010; Varadharajan and Hemond, 2012; Maeck et al., 2014; Harrison et al., 2017). If this drawdown is conducted after the stratification period, it may additionally cause substantial downstream GHG emissions if CO2 and CH4 had accumulated in anoxic waters. In case small drawdowns are conducted at regular time intervals overall CH4 emissions via ebullition may be smaller than for an operation with no withdrawal followed by a single large drawdown per year. Inflow of water at largest depth may be advantageous for reducing GHG emissions. Because the inflowing water is typically warmer and enriched in oxygen compared to the deep water, stratification is reduced which supports vertical transport by mixing, and the deep water is oxygenated which enhances oxidation of CH4 at the sediment water interface and prevents accumulation of CH4 that could be released during fall overturn. Pump-storage operation with turbination and inflow of water in the deep water may be particularly advantageous with respect to GHG emissions, because the ebullition fluxes during regular small drawdowns lead to comparatively small CH4 emissions and the inflow of water in the deep water supports oxygenation of the deep water and oxidation of CH4 at the sediment-water interface, thus preventing accumulation of CH4 (Encinas Fernández et al., 2014). Another effect of regular water fluctuations in a pump-storage system is the periodic drying and wetting of near shore sediments which may reduce CH4 fluxes from these sediments.

5.7. Societal implications

The social surrounding is linked in a myriad of ways to most processes within and around the reservoir (Zar et al., 2015; Kirchherr et al., 2016; Kirchherr and Charles, 2016; World Commission on Dams, Kornijów, 2009; Siegmund-Schultze et al., 2018; Votruba and Broža, 1989; Cernea, 1997). The land use within the upstream catchment area influences sediment erosion (Annandale, 1987), nutrient input, e.g., from agricultural uses or sewage treatment plants but also hydrological aspects (Zubala, 2009). Downstream riparian may also have demands regarding water quantity and quality (sediments, nutrients, general suspended load). There are many ecological concerns as some reservoirs, due to their history, do not

MANUSCRIPT 4 meet minimal discharge requirements or incorporate pass-through installations for aquatic organisms (BMUB/UBA, Aguiar et al., 2016; Dittmann et al., 2009; EU 2000). People living nearby may benefit from the reservoirs in various ways: water storage ensures drinking and irrigation water supply, there are recreational opportunities as well as affiliated economic advantages like employment in hydropower enterprises or tourism. But differing interest of stakeholders can lead to conflict, e.g., in connection to agricultural uses and nature protection versus recreation and other economic uses, especially energy production (Nguyen et al., 2017). Invasive species, like aquatic birds, might also be an issue (Kleinhenz and Koenig, 2018). Nature protected areas will also limit possible interference in the hydrological system therefore will be influencing water in- and output.

5.8. Occupancy of water reservoirs and their direct and indirect implications

Renewable energy production is naturally fluctuating. If the reservoir is connected to a pump-storage system, it is possible to follow that variable demand structure particularly if hydropower production is the only or the dominant function (Giesecke and Mosonyi, 2009). As hydroelectricity production will lead to water level changes within the reservoir but also downstream, recreational facilities on the shore, nature protection areas or downstream riparian floodplain will be subject to drastic water level changes thereby impede full flexibility (Hanson et al., 2002). Depending on the overall system, pump storage might introduce water from a different catchment into the reservoir leading to new challenges e.g., in respect to nutrient content (Godde et al., 2015) If the reservoir is part of a pro table touristic system, these interests may play a major role in the management of the reservoir. Frequent water level fluctuations, water quality issues and health risks (e.g., due to cHABs) will be a topic of considerable concern (Daus et al., 2019; Schrenk-Bergt et al., 2004). There may be conflicts regarding accessibility of the reservoir if it is restricted due to hydro-energy installation or nature protection areas. In case of drinking water reservoirs, regulations are in place that ensure water quality and govern water use, restrict access and regulate possible emitters of pollutants within the catchment (Bundesministerium der Justiz und für Verbraucherschutz, 15.07.2020, 21.05.2001). Drinking water supply will usually be the predominant purpose and other uses will only be authorized if they do not impair this vital function. If the implementation of drinking water protection area leads to economic losses for other usages (e.g., agriculture, forestry), compensation is required (§ 52 Abs. 5 WHG, § 96 Abs. 2 WHG of the German water law). Furthermore, water quantity might be an issue considering dry years. Many reservoirs are part of a decentralized flood protection system. In case of an imminent flood event, the

MANUSCRIPT 4 water level has to be lowered to provide additional protection volume (EU, 2007). Downstream riparians will benefit from a more constant water flow that will usually not exceed safe water levels. However, hydropower companies will not be in favor of discharging large amounts of water at a time when there might not be much revenue. Overall there are lots of interests present at water reservoirs, which the management, or Integrated Water Resources Management (Borchardt et al., 2016) ideally all incorporates.

5.9. Review tool for management

During the CHARM project, extensive discussions took place in order to consolidate the knowledge on five aspects of reservoir management discussed in the previous sections and to identify the challenges of reservoir management through the analysis of project-related results (laboratory, field, stakeholder analysis) and pivotal papers (https://milan- daus.github.io/Literatureon-Water-Reservoir-Management). Accordingly, as an attempt to support the achievement of the Sustainable Development Goals (SDG) objectives in the water sector (Bhaduri et al., 2016), more precisely in the storage of this fundamental resource, a conceptual tool was developed that provides an overview of the interdependencies and interactions between different factors of reservoir management in a case study perspective from Germany (https://github.com/milan-daus/Literature-on-Water-Reservoir-Management). The application of this tool for experts and non-experts and the evaluation of the presented interdependencies for other multi-purpose reservoirs can be seen as encouraging examples to create transparency and to improve the management of storage systems, which are of central importance to water-food-energy nexus in pursuit of meeting the SDG.

5.10. Summary and conclusions

The management of water reservoirs affects a multitude of interests and aspects in and around the actual water body depending on the main function of the system. Most reservoirs serve multiple purposes and are therefore prone to interest conflicts between those uses. The CHARM research project developed a tool to showcase some important key aspects of management-related issues connected to the five topics under consideration, namely: sedimentation, biostabilization (of fine sediments), cyanobacteria and possible cHABs, greenhouse gas emissions (mainly CH4 and CO2) from reservoirs and possible social conflicts and trade-offs. The tool provided, shows the possible interrelations of these topics. Furthermore, a matrix based on the idea of cross-impact-matrices (Weimer-Jehle, 2006) can

100

MANUSCRIPT 4 identify certain interrelations of typical reservoir management issues and is open to be modified by the individual users of the tool to get a comprehensive overview and support decision making.

5.11. Availability of data

The excel tool that presents key aspects of management-related issues and their interrelations can be accessed at https://github.com/milan-daus/Literature-on-Water- Reservoir-Management, while the relevant literature is available at https://milan- daus.github.io/Literature-on-Water-Reservoir-Management/.

5.12. Acknowledgements

The authors would like to thank all project members of the CHARM project to make this publication possible with great effort. Especially Prof. Markus Noack, Michelle Helmer, Dr. Thomas Haas and the Ministry of Science, Research and the Arts of the State of Baden- Württemberg which funded the project.

101

GENERAL DISCUSSION

6. GENERAL DISCUSSION

Occurrences of harmful cyanobacterial blooms in lakes and reservoirs are reported worldwide, with a significant increase over the last decades. Simultaneously, weather extremes related to climate change cause an increased demand for reliable and sustainable water sources. Cyanobacterial blooms are dynamic and complex phenomena, showing spatiotemporal variations of bloom toxicity and species composition in long and short-term. Consequently, it is important to improve our knowledge on parameters causing cyanobacterial blooms as well as understanding factors that influence intra-annual bloom dynamics. In this context, three main topics regarding bloom dynamics were addressed in this work: Using lake sediment to reconstruct cyanobacterial bloom history of lakes (Manuscript 1); N2-fixing Dolichospermum blooms and nitrogen load reduction as bloom mitigation strategy (Manuscript 2 and 4); and association between short-term bloom dynamics and chytrid parasitism (Manuscript 3). Detailed information and discussions of the results can be found in the respective manuscripts. In the following chapters a short summary is given and the individual key findings are discussed in the common context of global warming, bloom mitigation strategies and reservoir management.

6.1. Using lake sediment to reconstruct cyanobacterial bloom history (Manuscript 1)

The availability of long-term data on the abundance and diversity of potentially toxic cyanobacterial populations in lakes is still very limited and relies often on sparse historic phytoplankton records. In New Zealand environmental data exist for only <5 % of the 3,800 lakes, and these data sets typically date back only 20 to 30 years. This lack of data results in uncertainty as to whether bloom-forming and toxin-producing species have always been present in a lake, or whether increased proliferating in response to eutrophication or climate change occurred through recent history, or if there has been a succession of new (invasive) species. Consequently, there is a need for robust evidence documenting if, when and why changes occurred. While the analysis of sedimentary DNA is well developed and has been applied to various topics in paleolimnological research, it is a fairly new approach in cyanobacterial bloom research. The work presented in this manuscript shows how cyanobacterial bloom-related proxies in lake sediment can be used to trace back occurrences of cyanobacterial blooms in the past. Our results indicate that potentially toxin-producing cyanobacterial blooms are a rather recent phenomenon in Lake Rotorua. Furthermore, the current sampling method employed in paleolimnological studies was validated. We showed

102

GENERAL DISCUSSION that one sample from the center of the lake can reflect the whole lake bacterial community. However, Lake Rotorua is a rather small lake and our findings might not be transferable to larger lakes, which might exhibit a higher spatial heterogeneity. The finding that Microcystis spp. have been present in Lake Rotorua since around the 1950’s agrees with other lake sediment studies, which state an increase of cyanobacteria populations during the 1900s in North America (Pal et al., 2015; Taranu et al., 2015) and Europe (Legrand et al., 2017; Zastepa et al. 2017; Monchamp et al. 2018, 2019). These studies identified nutrient enrichment as a result of intensified agriculture, and global warming as the strongest environmental drivers of the observed increase of cyanobacteria. New Zealand’s freshwater ecosystems show a decreasing trend in lake health and a strong increase of cyanobacterial blooms over the past decades, which is primarily linked to intensified dairy farming. An ongoing effort to understand the lakes bloom histories and bloom drivers in New Zealand, is the Lakes380 project (https://lakes380.com/). In this project sediment cores from 10 % of New Zealand lakes are collected to analyze past ecosystem health and anthropogenic induced changes in the aquatic environment, but first publications are pending. So far, there is to our knowledge just one other sediment core study which reconstructed the historical composition of cyanobacterial communities for a New Zealand lake (Wood et al., 2009). However, the species identification method applied by Wood et al. (Automated ribosomal intergenic spacer analysis) could not provide evidence if cyanobacteria increased disproportional to other phytoplankton species. Whereas in our study we could show that there was no Microcystis spp. and a generally low abundance of cyanobacteria before the 1950’s, based on relative abundances in the bacterioplankton. We assessed the potential toxin production of previous bloom populations by the occurrence of mcyE genes, but actual MC concentrations were only analyzed for sediment surface samples. Including quantification of MCs in sediment core layers as proxy for past MC concentrations, similar to the study of Zastepa et al. (2017) would improve our study and may help in further studies to identify possible drivers of toxigenic cyanobacteria. Why there was an increase in Microcystis and cyanobacteria in Lake Rotorua could not be readily explained. Due to its shallow depth, large catchment area and small outflow of the lake it can be assumed that Lake Rotorua has always been eutrophic. To which extend the general nutrient enrichments in the region’s waterways also changed nutrient loads of Lake Rotorua would need further analysis. One way to do this is the use of diatom assemblages preserved in lake sediments (Stoermer and Smol, 2001). With their specific ecological preferences and rapid response to changes in nutrient concentrations, diatoms can be used as bioindicator to track past water quality dynamics and shifts in nutrient regimes (Adams et al., 2014). This method was already successfully applied in cyanobacterial bloom research by Zastepa et al.

103

GENERAL DISCUSSION

(2017) who correlated sediment MC concentrations with historical N and P concentrations, inferred from diatom assemblages, covering a time span of 200 years.

Despite being a promising tool, it needs to be noted that lake sediment core studies may not be feasible for artificial water reservoirs. Hydraulic flushing of deposited sediments has been employed in the management of many reservoirs to sustain the useful storage capacity (Lai and Shen, 1996; Schleiss et al., 2016). However, it can be supposed that availability of long- term monitoring data of nutrients and phytoplankton is better for most reservoirs compared to natural lakes, at least if they are used for drinking-water supply and recreation.

To sum up, cyanobacterial bloom proxies stored in lake sediments provide measures of current and historical toxic cyanobacterial populations. This information helps to assess the natural condition of a lake and, for instance, if current cyanobacterial blooms are a consequence of shifts in water quality and nutrient loads. This might be used as the first step to decide upon possible bloom mitigation strategies, including nutrient load reduction and lake restoration (see chapter 6.3).

6.2. Fungal parasitism and its influence on bloom toxicity (Manuscript 3)

Changes in bloom toxicity are assumed to primarily result from changes in the cell concentrations of toxic and non-toxic cyanobacterial genotypes within bloom populations (Orr et al., 2018). While the succession of genotypes during the course of a bloom is well documented (see chapter 1.5.1), little is known about the factors influencing these genotype dynamics. Parasitism is increasingly recognized as an important factor steering phytoplankton succession and has therefore been implemented in phytoplankton models (Frenken et al., 2017; Sommer et al., 2012, Figure 6-1).

104

GENERAL DISCUSSION

Figure 6-1 Seasonal (winter through autumn) biomass patterns of phytoplankton in a eutrophic (left) and an oligotrophic (right) water body. Seasonal (winter through autumn) biomass patterns of phytoplankton in a eutrophic (left) and an oligotrophic (right) water body. Gray line, classic scenario (moderate fish predation, overwintering of metazoan plankton unimportant); blue line (OZ), overwintering zooplankton important; red line (F−−), high metazoan density in fishless water bodies; orange line (F++), metazoan plankton suppressed by high fish predation. Shading indicates the mean vulnerability of phytoplankton against metazoan. The thickness of the horizontal bars indicates the seasonal change in regard to the relative importance of physical factors such as grazing by protists and metazoa, nutrient limitation and parasitism. Adapted from Sommer et al. 2012

The regulating effects of parasites on phytoplankton dynamics include (a) delayed timing or decreased peak population densities of algal blooms, (b) steering the outcome of interspecific competition among (sub)dominant species, and (c) directing intraspecific succession and maintaining genetic diversity (Sommer et al., 2012). In Manuscript 3, we documented for the first time how chytrid parasitism can control cyanobacterial genotype shifts in a natural bloom and can thereby alter bloom toxicity. Contrary to expectations, we observed that the Dolichospermum genotype that showed higher toxin production was more susceptible to the chytrid parasite, resulting in a shift towards lower bloom toxicity. In contrast, Rohrlack et al. (2013) observed a lower virulence of chytrids to a Planktothrix wildtype strain producing several cyanotoxins, compared to its knock-out mutants lacking the production of at least one of the toxins. However, besides MCs and anabaenopeptins, the wildtype strain used by Rohrlack et al. also produced aeroginosins, microviridin, cyanopeptolins which were not tested in our study. Therefore, direct comparison of these two studies is difficult. The question whether and if so which cyanotoxins actually function as antiparasite defense system, for example by inhibiting chytrid proteases, can only be resolved via mechanistic studies with the individual cyanotoxins.

Chytrid parasitism is not only determined by interactions between host and parasite, but also depends on environmental conditions, in particular on temperature (Gsell et al., 2013; Ibelings

105

GENERAL DISCUSSION et al., 2011). Earlier experimental studies have shown that chytrids are almost non-infective at water temperatures below 3ºC (Sommer et al., 2012). Ibelings et al. (2011) described the dynamics of host-parasite interactions for >30 years between the freshwater diatom Asterionella formosa and two highly virulent chytrid parasites. They observed that milder winters, as a result of global warming, promoted ongoing chytrid infections of its diatom host but at low prevalence. This in turn allowed other diatoms to take over as dominant species. Similar to this, Gsell et al. (2013) described a genotype-specific susceptibility of diatoms towards chytrid virulence which also varied with temperature level. The duration and intensity of chytrid parasite pressure on cyanobacterial host populations is therefore likely to be affected by changes in lake temperature patterns due to global warming. This, in turn may affect the selective strength of the parasite on the genetic structure of the host population (Gsell et al., 2013). However, whether increased chytrid parasitism selects for more toxic or less toxic cyanobacterial genotypes is closely linked to the putative antiparasital function of cyanotoxins and remains to be further investigated.

6.3. Nitrogen load control as bloom mitigation strategy (Manuscript 2 & 4)

There is a number of physical, chemical, and biological approaches that may decrease bloom intensities: (1) reducing nutrient input from external sources, (2) reducing internal release of nutrients that are already in the system, (3) increase flushing rates, (4) destratifying waters by artificial mixing, (5) application of algaecides and (6) biological manipulation (Paerl et al., 2016a; Paerl, 2018). However, most scientists emphasize that nutrient load reduction (targeting N and P) is the primary bloom mitigation strategy to ensure long term and sustainable success. There are multiple approaches to target this nutrient reduction and their suitability depends on the size, usage and nature of the water body. Sediment dredging for example is commonly applied for artificial reservoirs. This strategy is primarily used to ensure the storage capacity of reservoirs, but also prevents nutrient regeneration from sediment to the water body. Sediments act as the storage bank of nutrients that can be exchanged rapidly between the bottom and water column. Furthermore, sediments store cyanobacterial resting cells and thereby act as a “seed bank” for blooms. While dredging sounds attractive from a bloom mitigation perspective, it is expensive, alters biogeochemical cycling and is a highly invasive strategy which disturbs the benthic floral and faunal habitat (Paerl, 2018). Therefore, where possible, the source water should be managed to prevent the development of cyanobacterial blooms.

106

GENERAL DISCUSSION

Despite the general agreement on the linkage between eutrophication and cyanobacterial blooms, there is less consensus concerning the exact strategy on how to reverse eutrophication in lakes: Should the reduction be focused on N loads, P loads, both nutrients or should a certain N:P ratio be aimed? This is also a question of financial costs, as reduction of both elements is more expensive than reduction of just one. It is therefore important to deduce when eutrophication and cyanobacterial blooms can be reduced by controlling P alone, and when more expensive control of N and P may be required (Schindler, 2012).

Another important issue regarding nutrient load reduction is the response time of the lake, meaning the time period it takes until a new nutrient equilibrium is reached. Lake response to P reduction is known to be rather slow. At least three retention times are needed to wash out 95 % of the excess P pool in the water column of fully mixed lakes (Jeppesen et al., 2005; Sas, 1990). The long response time is due to internal loadings continuously replenishing the P pool in the water column (Sas, 1990). In contrast, N loss by denitrification results in negligible internal N loading, resulting in a quick response of the N concentration (Jensen et al., 1992).

The efficiency of P load reduction is documented for many lakes (Schindler et al., 2008). However, P-only reduction is not always successful. For example, Lac du Bourget, located in the French Alps, has undergone major restoration programs. Nutrient-reducing efforts focusing on P loads (N: 750 to 600 µg L-1, P: 100 µg L-1 to 30 µg L-1), lead to an increase in the N:P ratio from 7 to 20 and caused ultimately the occurrence of Planktothrix blooms (Jacquet et al., 2005;

Vrede et al., 2009). This genus, as well as other non-N2 fixing taxa (e.g., non-N2 fixing Lyngbya and Oscillatoria spp.), points to N-over-enrichment in the system and indicate that N input reduction is additionally needed to reverse this trend (Paerl, 2018; Paerl et al., 2016b).

Reduction of N-only loads in turn is assumed to cause a shift towards N2-fixing cyanobacteria, potentially without decreasing overall cyanobacterial biomass (see chapter 1.3.1, Schindler et al., 2008). Experimental data for individual diazetrophic species are especially important, since efficiency of nitrogenase and N2-fixation seems to differ tremendously among diazotrophic taxa. The results of Manuscript 2 indicate that Dolichospermum can fully compensate for N- limitation by N2-fixation. Therefore, N-only reduction might indeed cause a shift towards Dolichospermum. However, assumptions resulting from short-term laboratory or mesocosm studies should be interpreted with caution and cannot directly be transferred to natural scenarios (Schindler, 2012). Nevertheless, this potential scenario should be carefully considered for waterbodies in which N-only input reductions are being considered to reverse eutrophication.

107

GENERAL DISCUSSION

The key priority for every management strategy is to establish the extent to which N and P reduction is needed to effectively mitigate the bloom. Unfortunately, there is still no precise threshold that simply needs to be complied and the task can only be carried out on a case- level. The N:P ratio approach might help to determine whether a system is N- or P-over- enriched. However, the ratio alone is neither effective nor meaningful without considering the total N and P and concentrations (Paerl, 2018). Overall, the most efficient strategy to mitigate cyanobacterial blooms is dual N and P reduction (Dolman et al., 2012; Lewis and Wurtsbaugh, 2008; Paerl et al., 2016b). In case of blooms in oligo- to mesotrophic lakes and reservoirs, e.g. Dolichospermum blooms in the Schwarzenbach reservoir, Planktothrix blooms in alpine lakes, reduction of P- or N-only might be carefully considered.

Finally, it needs to be pointed out that eutrophic and mesotrophic lakes do naturally occur and not every nutrient-rich lake has been an oligotrophic system before. In these cases, nutrient load reduction cannot be considered as a renaturation strategy but an invasive intervention into a productive ecosystem. Paleolimnological studies might help to decide upon the lakes previous trophic state and bloom history (see chapter 6.1).

6.4. Risk assessment and short-term bloom dynamics

This thesis did not aim to revise current risk assessment strategies for cyanobacterial blooms. However, some of our findings might help to improve current and future risk assessment and will therefore be discussed in this context in the following paragraph.

Water bodies used for drinking water supply or recreational purposes are typically managed by local water managers, who are also responsible for undertaking the risk assessment of cyanobacterial blooms. This includes decision making whether responses, such as bathing restrictions, should be undertaken to ensure public health protection. Short-term bloom dynamics are of particular concern for water managers, as rapid changes might lead to an over- or underestimation of the actual risk posed by a bloom. Therefore, the methodical approach of risk assessment but also the timing and frequency of monitoring are important parameters to ensure save recreational and drinking waters. Approaches of bloom risk assessment and monitoring differ between countries (see Table 1 & 2 in Chorus, 2012b) but are primarily based on the World Health Organization (WHO) guidelines for drinking water (WHO, 2017) and recreational water (WHO, 2003), which are implemented in the European law in the respective guidelines of the European Union (EU) (Directive 98/83/EC and 2006/7/EC, EU, 1998, 2006). In the case of drinking water, national monitoring methods and

108

GENERAL DISCUSSION frequencies reflect the severity of toxin-producing blooms in the respective country. Germany for example has no specific regulations for cyanotoxin monitoring or guidance values for drinking water, as only about 20 % of drinking water supply is from surface water, and that mostly from well protected reservoirs (Chorus, 2012b; UBA, 2015). Nevertheless, where cyanobacteria do occur, the common WHO guideline value for MC-LR in drinking water (1 μg L−1) is applied. New Zealand in turn faces more problems with toxin-producing blooms, and obtains 75 % of drinking water from surface water supplies. Therefore, the government has implemented a comprehensive risk management plan, including monitoring of cyanotoxins in up to weekly frequency and very specific maximum acceptable values for the most common cyanotoxins (Chorus, 2012b).

In the case of recreational waters, the EU guideline demands the establishment of a 'bathing water profile', i.e. the assessment of the potential for contamination (EU, 2006). If this profile, or historic data indicate a potential for cyanobacterial proliferation in the respective water body, appropriate monitoring shall be carried out to enable timely identification of health risks. When cyanobacterial proliferation occurs and a health risk has been identified or presumed, adequate management measures shall be taken immediately to prevent exposure. However, further definitions of 'appropriate monitoring', 'timely identification of health risk' and 'adequate management measures' are left to the individual country. Most countries implemented the three-tiered concept recommended by the WHO, in which health risk of a bloom is not directly measured via cyanotoxin quantification, but deduced from the cyanobacterial biomass, cell concentration or chlorophyll-a concentration (Table 6-1). Cell densities >20,000 cells mL-1 (~MC concentration >4 µg L-1) result in bathing restrictions (EU, 2006; UBA, 2015).

Table 6-1 World Health Organization guideline values for the health risk assessment of cyanobacteria in recreational water and recommended actions. Chlorophyll-a µg L-1 Microcystin Cells mL-1 with dominance of Recommended action concentration µg L-1 cyanobacteria

Low risk for adverse health Post on-site risk advisory signs; 20,000 10 2–4 (maximal 10) outcomes Inform relevant authorities

Watch for scums; Restrict bathing Intermediate risk for 20 (maximal 100,000 50 and further investigate hazard; Post adverse health outcomes 50–100) on-site risk advisory signs Immediate action to prevent contact with scums; Possible prohibition of High risk for adverse health Scum (Bloom) swimming and other water contact outcomes formation activities; Post on-site risk advisory signs; Inform relevant authorities.

The usage of biomass parameters as basis for risk assessment has the advantage of encompassing all cyanotoxins, including those yet unidentified or for which the toxicological

109

GENERAL DISCUSSION data are insufficient for the derivation of a guidance value (Chorus, 2012b). The downside is that in blooms where the presence of toxin-producing genotypes is low, responses such as bathing restrictions may be more restrictive than necessary for health protection (Chorus, 2012b; Turner et al., 2018). As was described in previous publications (see chapter 1.5.1) and in Manuscript 3 of this thesis, total cyanotoxin concentrations but also MC congener composition can differ tremendously between genotypes. Even in the course of a toxic bloom, toxin concentrations may vary significantly. Toxin-producing and non-producing genotypes as well as levels of toxin production cannot be distinguished via visual inspection. Routine analysis of bloom samples by LC-MS/MS would therefore provide a beneficial confirmatory approach to the biomass parameters, aiding both public health and the needs of water users and managers (Turner et al., 2018).

Another critical point of the current risk assessment is the timing and frequency of measurements. While some states implemented a fortnightly (e.g. Netherlands) or even weekly (e.g. New Zealand) monitoring frequency of cell densities during bathing season, German implementation only asks for intensification of monitoring during bathing season, once cyanobacteria cell density exceeds 10,000 cells mL-1(UBA, 2015). The formation and breakdown of surface blooms is highly dynamic, and varies at a time scale of hours (Ibelings et al., 2003; Wood et al., 2011). Monitoring these dynamics with routine means is impossible (Chorus, 2012b). However, monitoring in higher frequency will always result in a more accurate and realistic risk assessment. Therefore, weekly monitoring should be taken into account by water managers, once a surface bloom occurs.

Lastly the bathing water profile is discussed, which assess the potential of a water body to experience cyanobacterial blooms in the first place (in the EU), and therefore also decides upon the monitoring frequency. Whether the profile indicates increased potential for cyanobacterial proliferation depends on the nutrient load. Lakes with a eutrophic nutrient status or with TP loads of 25-50 µg L-1 are expected to have a high potential to experience blooms (Umweltbundesamt (UBA), 2015). Nevertheless, they sometimes also occur in oligotrophic environments (see chapter 1.3.1, Carey et al., 2012; Sterner et al., 2020). A common species occurring at lower nutrient loads is P. rubescens. Planktothrix blooms are considered to pose a low health risk, as they occur deeper in the water column without interfering with water sports (UBA, 2015). In the recent study it was shown that also mesotrophic lakes and reservoirs can experience blooms, e.g. Dolichospermum blooms in the Schwarzenbach reservoir. In contrast to P. rubescens, Dolichospermum builds surface blooms and therefore has a greater potential to expose swimmers with cyanotoxins. Measurements of TP in the Schwarzenbach reservoir were frequently <20 mg L-1. There is no doubt that eutrophic lakes have an increased risk for

110

GENERAL DISCUSSION cyanobacterial blooms. Nevertheless, also oligotrophic and mesotrophic bathing waters should be monitored closely once any sign of cyanobacterial proliferation occurs. This might be especially relevant assuming a future increase in occurrences of N2-fixing species in oligo- mesotrophic waters. Several publication state that global warming will promote the growth of diazotroph taxa (Brauer et al., 2013; Visser et al., 2016). Increasing temperature will result in a substantially higher N2-fixation activity and greater competitive advantage for diazotrophic cyanobacteria in N-limited waters. Rising CO2 concentrations may also enhance N2-fixation rates, as has been reported for the marine cyanobacterium Trichodesmium (Levitan et al., 2007). Furthermore, warmer water in spring is supposed to increase nutrient consumption by the phytoplankton community which causes N limitation later in the growing season to the advantage of N2-fixing cyanobacteria (Elliott, 2012).

111

CONCLUSIONS AND OUTLOOK

7. CONCLUSIONS AND OUTLOOK

Global increase of harmful cyanobacterial blooms demands an improved understanding of ecosystem changes promoting bloom growth, as well as environmental factors controlling bloom dynamics. This work deals with different aspects of the dynamics of toxin producing cyanobacterial blooms in natural lakes and artificial water reservoirs.

We showed that a single sample taken in the center of a lake is sufficient to capture microbial diversity across a lake. This study will help to investigate lakes cyanobacterial bloom history and to identify past shifts in lake ecology that may have caused an increase of cyanobacterial blooms appearance. This knowledge is needed to protect and restore lake health. Apart from bloom research, results of this study are relevant to other paleolimnological studies which are now incorporating molecular microbial analysis into their suite of techniques.

Understanding short-term changes of bloom intensity and toxicity are crucial for safe bloom risk assessment. We demonstrated for the first time, in a natural bloom how fungal parasitism can influence genotype shifts over the course of a bloom, thereby influencing bloom toxicity. Albeit more studies are needed on the putative antiparasital function of cyanotoxins, our study provides valuable information on the biological function of cyanotoxins. Global warming might alter parasitism intensity and therefore also bloom toxicity in the future. Further studies are needed to investigate the influence of fungal but also viral parasitism on cyanobacterial bloom toxicity.

On a global scale toxin producing cyanobacterial blooms are expected to increase as a consequence of climate change. Therefore, sustainable management and efficient bloom mitigation strategies are needed to ensure safe and clean recreational waters and drinking water supplies in the future. In this study, we showed that reversing eutrophication of surface water by N-only reduction might promote growth of N2-fixing cyanobacteria, e.g. Dolichospermum and can therefore not be recommended as bloom mitigation strategy. We presented an information tool which aids and enhances multidisciplinary knowledge in reservoir management. Despite the current limitations of the tool it but will be extended and improved in future.

112

RECORD OF CONTRIBUTION

8. RECORD OF CONTRIBUTION

MANUSCRIPT 1

The study was conceptualized by S.A. Wood and B. Weisbrod in discussion with and suggestions by J. Puddick and D.R. Dietrich. Field sampling of sediment samples was done by B. Weisbrod, S. Waters and R. Whyte-Wilding. Toxin extraction and sediment spiking was done by B. Weisbrod under the supervision of J. Puddick. HPLC measurements and data analysis was done by J. Puddick. Illumina PCR and qPCR of surface sediments was done by B. Weisbrod and K. Steiner. ddPCR and Illumina-PCR of sediment cores was done by R. Whyte-Wilding under the supervision of K. Steiner. Bioinformatic analysis of Illumina sequencing data was done by O. Laroche and B. Weisbrod. Data analysis and graphical representation of all other data was done by B. Weisbrod. B. Weisbrod wrote the original draft of the manuscript. All authors edited and reviewed the manuscript.

MANUSCRIPT 2

The study was conceptualized by B. Weisbrod, D.R. Dietrich and D. Martin-Creuzburg. All experimental investigations were performed by J. Breiholz under the supervision of B. Weisbrod. Formal data analysis, validation and graphical representation of data was done by B. Weisbrod and J. Breiholz with suggestions of D. Martin-Creuzburg and D.R. Dietrich. B. Weisbrod wrote the original draft of the manuscript. All authors edited and reviewed the manuscript.

MANUSCRIPT 3

The study was designed by B. Weisbrod, D.R. Dietrich and D. Martin-Creuzburg. Field sampling of water samples was done by M. Helmer, D. Helmer and B. Weisbrod. Microscopic phytoplankton analysis and ELISAs were performed by B. Weisbrod. E. Riehle performed HPLC analysis. Formal data analysis, validation and graphical representation of data was done by B. Weisbrod. B. Weisbrod wrote the original draft of the manuscript. All authors edited and reviewed the manuscript.

113

RECORD OF CONTRIBUTION

MANUSCRIPT 4

Conceptualization, M. Daus., K. Koberger; Methodology, M. Daus, K. Koberger, K. Koca. and F. Beckers; Data curation, M. Daus., K. Koberger, K. Koca. and F. Beckers; Writing - original draft preparation, M. Daus, K. Koberger, K. Koca, F. Beckers, J. Encinas-Fernandes, B. Weisbrod, D. Dietrich, S. Gerbersdorf, R. Glaser, S. Haun, H. Hofmann, D. Martin- Creuzburg, F. Peeters; Writing - review and editing, M. Daus, K. Koberger, K. Koca, F. Beckers, and B. Weisbrod; Visualization, M. Daus, K. Koberger, K. Koca and F. Beckers; Supervision, M. Daus, K. Koberger.

114

REFERENCES

9. REFERENCES

Adams, K.E., Taranu, Z.E., Zurawell, R., Cumming, B.F., Gregory-Eaves, I., 2014. Insights for lake management gained when paleolimnological and water column monitoring studies are combined: A case study from Baptiste Lake. Lake and Reservoir Management 30, 11–22. Agawin, N.S., Rabouille, S., Veldhuis, M., Servatius, L., Hol, S., van Overzee, H.M.J., Huisman, J., 2007. Competition and facilitation between unicellular nitrogen-fixing cyanobacteria and non- nitrogen-fixing phytoplankton species. Limnology and Oceanography 52. Agha, R., Lezcano, M.À., Labrador, M. del M., Cirés, S., Quesada, A., 2014. Seasonal dynamics and sedimentation patterns of Microcystis oligopeptide-based chemotypes reveal subpopulations with different ecological traits. Limnol. Oceanogr. 59, 861–871. Agha, R., Quesada, A., 2014. Oligopeptides as biomarkers of cyanobacterial subpopulations. Toward an understanding of their biological role. Toxins 6, 1929–1950. Aguiar, F.C., Martins, M.J., Silva, P.C., Fernandes, M.R., 2016. Riverscapes downstream of hydropower dams. Effects of altered flows and historical land-use change. Landscape and Urban Planning 153, 83-98. Albay, M., Akcaalan, R., Tufekci, H., Metcalf, J.S., Beattie, K.A., Codd, G.A., 2003. Depth profiles of cyanobacterial hepatotoxins (microcystins) in three Turkish freshwater lakes. Hydrobiologia 505, 89–95. Alexova, R., Fujii, M., Birch, D., Cheng, J., Waite, T.D., Ferrari, B.C., Neilan, B.A., 2011a. Iron uptake and toxin synthesis in the bloom-forming Microcystis aeruginosa under iron limitation. Environmental Microbiology 13, 1064–1077. Alexova, R., Haynes, P.A., Ferrari, B.C., Neilan, B.A., 2011b. Comparative protein expression in different strains of the bloom-forming cyanobacterium Microcystis aeruginosa. Molecular & Cellular Proteomics 10. Altaner, S., Puddick, J., Fessard, V., Feurstein, D., Zemskov, I., Wittmann, V., Dietrich, D.R., 2019. Simultaneous detection of 14 microcystin congeners from tissue samples using UPLC- ESI- MS/MS and two different deuterated synthetic microcystins as internal standards. Toxins 11, 388. Al-Tebrineh, J., Merrick, C., Ryan, D., Humpage, A., Bowling, L., Neilan, B.A., 2012. Community composition, toxigenicity, and environmental conditions during a cyanobacterial bloom occurring along 1,100 Kilometers of the Murray River. Appl Environ Microbiol 78, 263–272. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403–410. Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 26, 32–46. Annandale, G.W., 1987. Reservoir sedimentation, 1st ed, Developments in water science. Elsevier Science. Appleby, P.G., Oldfield, F., 1978. The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. CATENA 5, 1–8. Ariño, J., Velázquez, D., Casamayor, A., 2019. Ser/Thr protein phosphatases in fungi: structure, regulation and function. Microb. Cell 6, 217–256. Arnaud, F., Lignier, V., Revel, M., Desmet, M., Beck, C., Pourchet, M., Charlet, F., Trentesaux, A., Tribovillard, N., 2002. Flood and earthquake disturbance of 210Pb geochronology (Lake Anterne, NW Alps). Terra Nova 14, 225–232. Arnaud, F., Magand, O., Chapron, E., Bertrand, S., Boës, X., Charlet, F., Mélières, M.-A., 2006. Radionuclide dating (210Pb, 137Cs, 241Am) of recent lake sediments in a highly active geodynamic setting (Lakes Puyehue and Icalma—Chilean Lake District). Sci. Total Environ. 366, 837–850. Atkinson, E. 1996. The feasibility of flushing sediment from reservoirs, TDR Project R5839, Report OD 137, HR Wallingford. Avşar, U., Jónsson, S., Avşar, Ö., Schmidt, S., 2016. Earthquake-induced soft-sediment deformations and seismically amplified erosion rates recorded in varved sediments of Köyceğiz Lake (SW Turkey). J. Geophys. Res. Solid Earth 121, 4767–4779.

115

REFERENCES

Azevedo, S.M.F.O., Carmichael, W.W., Jochimsen, E.M., Rinehart, K.L., Lau, S., Shaw, G.R., Eaglesham, G.K., 2002. Human intoxication by microcystins during renal dialysis treatment in Caruaru-Brazil. Toxicology 181–182, 441–446. Bagu, J.R., Sykes, B.D., Craig, M.M., Holmes, C.F.B., 1997. A molecular basis for different interactions of marine toxins with protein phosphatase-1. J. Biol. Chem. 272, 5087–5097. Banasiak, R., Verhoeven, R., De Sutter, R., Tait, S., 2005. The erosion behaviour of biologically active sewer sediment deposits: Observations from a laboratory study. Water Res. 39, 5221– 5231. Barros, N., Cole, J.J., Tranvik, L.J., Prairie, Y.T., Bastviken, D., Huszar, V.L.M., del Giorgio, P., Roland, F., 2011. Carbon emission from hydroelectric reservoirs linked to reservoir age and latitude. Nature Geoscience, 4, 593-596. Bar-Yosef, Y., Sukenik, A., Hadas, O., Viner-Mozzini, Y., Kaplan, A., 2010. Enslavement in the water body by toxic Aphanizomenon ovalisporum, inducing alkaline phosphatase in phytoplanktons. Current Biology 20, 1557–1561. Basen, T., Fleckenstein, K.M., Rinke, K., Rothhaupt, K.-O., Martin‐Creuzburg, D., 2017. Impact of temperature and nutrient dynamics on growth and survival of Corbicula fluminea: A field study in oligotrophic Lake Constance. Int. Rev. Hydrobiol. 102, 15–28. Battin, T.J., Besemer, K., Bengtsson, M.M., Romani, A.M., Packmann, A.I., 2016. The ecology and biogeochemistry of stream biofilms. Nat. Rev. Microbiol. 14, 251–263. Batuca, D., Jordaan, J., 2000. Silting and desilting of reservoirs. Journal of hydraulic engineering 126, 874–875. Beaulieu, J.J., Smolenski, R.L., Nietch, C.T., Townsend-Small, A., Elovitz, M.S., 2014. High methane emissions from a midlatitude reservoir draining an agricultural watershed. Environmental Science & Technology, 48 (19), 11100-11108. Beckers, F., Haun, S., Gerbersdorf, S., Noack, M., Dietrich, D., Martin-Creuzburg, D., Peeters, F., Hofmann, H., Glaser, R., Wieprecht, S., 2018. CHARM - Challenges of Reservoir Management - Meeting environmental and social requirements. Hydrolink 2018, 16–18. Benza, B., 2020. Botswana says toxins in water killed hundreds of elephants. Reuters. Berke, A.P., Turner, L., Berg, H.C., Lauga, E., 2008. Hydrodynamic attraction of swimming microorganisms by surfaces. Phys. Rev. Lett. 101, 038102. Berry, J.P., Gantar, M., Perez, M.H., Berry, G., Noriega, F.G., 2008. Cyanobacterial toxins as allelochemicals with potential applications as algaecides, herbicides and insecticides. Mar Drugs 6, 117–146. Berry, M.A., Davis, T.W., Cory, R.M., Duhaime, M.B., Johengen, T.H., Kling, G.W., Marino, J.A., Uyl, P.A.D., Gossiaux, D., Dick, G.J., Denef, V.J., 2017. Cyanobacterial harmful algal blooms are a biological disturbance to Western Lake Erie bacterial communities. Environ. Microbiol. 19, 1149–1162. Blackbourn, D.; Rennert, U. 2008. Die Eroberung der Natur - Eine Geschichte der deutschen Landschaft, 3rd ed.; Pantheon: München, Germany. Bláha, L., Babica, P., Maršálek, B., 2009. Toxins produced in cyanobacterial water blooms – toxicity and risks. Interdiscip Toxicol 2, 36–41. Block, J.C., Haudidier, K., Paquin, J.L., Miazga, J., Levi, Y., 1993. Biofilm accumulation in drinking water distribution systems. Biofouling 6, 333–343. Bokulich, N.A., Kaehler, B.D., Rideout, J.R., Dillon, M., Bolyen, E., Knight, R., Huttley, G.A., Gregory Caporaso, J., 2018. Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2’s q2-feature-classifier plugin. Microbiome 6, 90. Bolyen, E., Rideout, J.R., Dillon, M.R., Bokulich, N.A., Abnet, C., Al-Ghalith, G.A., Alexander, H., Alm, E.J., Arumugam, M., Asnicar, F., Bai, Y., Bisanz, J.E., Bittinger, K., Brejnrod, A., Brislawn, C.J., Brown, C.T., Callahan, B.J., Caraballo-Rodríguez, A.M., Chase, J., Cope, E., Silva, R.D., Dorrestein, P.C., Douglas, G.M., Durall, D.M., Duvallet, C., Edwardson, C.F., Ernst, M., Estaki, M., Fouquier, J., Gauglitz, J.M., Gibson, D.L., Gonzalez, A., Gorlick, K., Guo, J., Hillmann, B., Holmes, S., Holste, H., Huttenhower, C., Huttley, G., Janssen, S., Jarmusch, A.K., Jiang, L., Kaehler, B., Kang, K.B., Keefe, C.R., Keim, P., Kelley, S.T., Knights, D., Koester, I., Kosciolek, T., Kreps, J., Langille, M.G., Lee, J., Ley, R., Liu, Y.-X., Loftfield, E., Lozupone, C., Maher, M.,

116

REFERENCES

Marotz, C., Martin, B.D., McDonald, D., McIver, L.J., Melnik, A.V., Metcalf, J.L., Morgan, S.C., Morton, J., Naimey, A.T., Navas-Molina, J.A., Nothias, L.F., Orchanian, S.B., Pearson, T., Peoples, S.L., Petras, D., Preuss, M.L., Pruesse, E., Rasmussen, L.B., Rivers, A., Michael S Robeson, I.I., Rosenthal, P., Segata, N., Shaffer, M., Shiffer, A., Sinha, R., Song, S.J., Spear, J.R., Swafford, A.D., Thompson, L.R., Torres, P.J., Trinh, P., Tripathi, A., Turnbaugh, P.J., Ul- Hasan, S., Hooft, J.J. van der, Vargas, F., Vázquez-Baeza, Y., Vogtmann, E., Hippel, M. von, Walters, W., Wan, Y., Wang, M., Warren, J., Weber, K.C., Williamson, C.H., Willis, A.D., Xu, Z.Z., Zaneveld, J.R., Zhang, Y., Zhu, Q., Knight, R., Caporaso, J.G., 2018. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 37 (8):852–857. Borges, H., Wood, S.A., Puddick, J., Blaney, E., Hawes, I., Dietrich, D.R., Hamilton, D.P., 2016. Intracellular, environmental and biotic interactions influence recruitment of benthic Microcystis (Cyanophyceae) in a shallow eutrophic lake. J. Plankton Res. 38, 1289–1301. Brauer, V.S., Stomp, M., Bouvier, T., Fouilland, E., Leboulanger, C., Confurius-Guns, V., Weissing, F.J., Stal, L., Huisman, J., 2015. Competition and facilitation between the marine nitrogen- fixing cyanobacterium Cyanothece and its associated bacterial community. Front. Microbiol. 5. Brauer, V.S., Stomp, M., Rosso, C., Beusekom, S.A. van, Emmerich, B., Stal, L.J., Huisman, J., 2013. Low temperature delays timing and enhances the cost of nitrogen fixation in the unicellular cyanobacterium Cyanothece. ISME J 7, 2105–2115. Braun, A., Pfeiffer, T., 2002. Cyanobacterial blooms as the cause of a Pleistocene large mammal assemblage. Paleobiology 28, 139–154. Briand, E., Escoffier, N., Straub, C., Sabart, M., Quiblier, C., Humbert, J.-F., 2009. Spatiotemporal changes in the genetic diversity of a bloom-forming Microcystis aeruginosa (cyanobacteria) population. ISME J. 3, 419–429. Bruce, D., Westerhoff, P., Brawley-Chesworth, A., 2002. Removal of 2-methylisoborneol and geosmin in surface water treatment plants in Arizona. AQUA (SUPPL.) 51, 183–197. Brunberg, A.-K., Blomqvist, P., 2003. Recruitment of Microcystis (Cyanophyceae) from lake sediments: The importance of littoral inocula. J. Phycol. 39, 58–63. Brussaard, C.P.D., 2004. Viral control of phytoplankton populations - A review. Journal of Eukaryotic Microbiology 51, 125–138. Bullerjahn, G.S., McKay, R.M., Davis, T.W., Baker, D.B., Boyer, G.L., Anglada, L.V., Doucette, G.J., Ho, J.C., Irwin, E.G., Kling, C.L., Kudela, R.M., Kurmayer, R., Michalak, A.M., Ortiz, J.D., Otten, T.G., Paerl, H.W., Qin, B., Sohngen, B.L., Stumpf, R.P., Visser, P.M., Wilhelm, S.W., 2016. Global solutions to regional problems: Collecting global expertise to address the problem of harmful cyanobacterial blooms. A Lake Erie case study. Harmful Algae 54, 223–238. Buratti, F.M., Manganelli, M., Vichi, S., Stefanelli, M., Scardala, S., Testai, E., Funari, E., 2017. Cyanotoxins: producing organisms, occurrence, toxicity, mechanism of action and human health toxicological risk evaluation. Arch Toxicol 91, 1049–1130. Bürgi, H., Stadelmann, P., 2002. Change of phytoplankton composition and biodiversity in Lake Sempach before and during restoration. Hydrobiologia 469, 33–48. Burns, A., Ryder, D.S., 2001. Potential for biofilms as biological indicators in Australian riverine systems. Ecol. Manag. Restor. 2, 53–64. Bush, A.O., Lafferty, K.D., Lotz, J.M., Shostak, A.W., 1997. Parasitology meets ecology on its own terms: Margolis et al. revisited. J. Parasitol. 83, 575. Butterwick, C., Heaney, S.I., Talling, J.F., 2005. Diversity in the influence of temperature on the growth rates of freshwater algae, and its ecological relevance. Freshwater Biology 50, 291– 300. Callahan, B.J., McMurdie, P.J., Rosen, M.J., Han, A.W., Johnson, A.J.A., Holmes, S.P., 2016. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581– 583. Callieri, C., 2010. Single cells and microcolonies of freshwater picocyanobacteria: A common ecology. J Limnol 69, 257–277. Callieri, C., Bertoni, R., Contesini, M., Bertoni, F., 2014. Lake level fluctuations boost toxic cyanobacterial "Oligotrophic Blooms”. PLoS ONE 9.

117

REFERENCES

Callieri, C., Coci, M., Corno, G., Macek, M., Modenutti, B., Balseiro, E., Bertoni, R., 2013. Phylogenetic diversity of nonmarine picocyanobacteria. FEMS Microbiol. Ecol. 85, 293–301. Callieri, C., Cronberg, G., Stockner, J.G., 2012. Freshwater picocyanobacteria: Single cells, microcolonies and colonial forms. In: Whitton, B.A. (Ed.), Ecology of cyanobacteria II: Their diversity in space and time. Springer Netherlands, Dordrecht, pp. 229–269. Canter, H.M., 1954. Fungal parasites of the phytoplankton. III. Trans. Br. Mycol. Soc. 37, 111-133. Capo, E., Debroas, D., Arnaud, F., Perga, M.-E., Chardon, C., Domaizon, I., 2017. Tracking a century of changes in microbial eukaryotic diversity in lakes driven by nutrient enrichment and climate warming. Environ. Microbiol. 19, 2873–2892. Carey, C.C., Ewing, H.A., Cottingham, K.L., Weathers, K.C., Thomas, R.Q., Haney, J.F., 2012. Occurrence and toxicity of the cyanobacterium Gloeotrichia echinulata in low-nutrient lakes in the northeastern United States. Aquat Ecol 46, 395–409. Carey, C.C., Weathers, K.C., Cottingham, K.L., 2008. Gloeotrichia echinulata blooms in an oligotrophic lake: helpful insights from eutrophic lakes. J Plankton Res 30, 893–904. Carlson, R.E., 1977. A trophic state index for lakes1. Limnology and Oceanography 22, 361–369. Carmichael, W., Drapeau, C., Anderson, D., 2000. Harvesting of Aphanizomenon flos-aquae Ralfs ex Born. & Flah. var. flos-aquae (Cyanobacteria) from Klamath Lake for human dietary use. Journal of Applied Phycology 12, 585–595. Carmichael, W.W., 1995. Toxic Microcystis and the environment, in: Toxic Microcystis. CRC Press, pp. 1–12. Carmichael, W.W., Boyer, G.L., 2016. Health impacts from cyanobacteria harmful algae blooms: Implications for the North American Great Lakes. Harmful Algae 54, 194–212. Carnea, Michael, M., 1997. Hydropower dams and social impacts. A sociological perspective. Carpenter, S.R., Stanley, E.H., Vander Zanden, M.J., 2011. State of the world’s freshwater ecosystems: Physical, chemical, and biological changes. Annu. Rev. Environ. Resour. 36, 75– 99. Chapra, S.C., Boehlert, B., Fant, C., Bierman, V.J., Henderson, J., Mills, D., Mas, D.M.L., Rennels, L., Jantarasami, L., Martinich, J., Strzepek, K.M., Paerl, H.W., 2017. Climate change impacts on harmful algal blooms in U.S. freshwaters: A screening-level assessment. Environ. Sci. Technol. 51, 8933–8943. Chen, X.D., Zhang, C.K., Zhou, Z., Gong, Z., Zhou, J.J., Tao, J.F., Paterson, D.M., Feng, Q., 2017. Stabilizing Effects of bacterial biofilms: EPS Penetration and redistribution of bed stability down the sediment profile. J. Geophys. Res. Biogeosciences 122, 3113–3125. Chia, M.A., Jankowiak, J.G., Kramer, B.J., Goleski, J.A., Huang, I.-S., Zimba, P.V., do Carmo Bittencourt-Oliveira, M., Gobler, C.J., 2018. Succession and toxicity of Microcystis and Anabaena (Dolichospermum) blooms are controlled by nutrient-dependent allelopathic interactions. Harmful Algae 74, 67–77. Chorus, I., 2012a. Cyanotoxins: Occurrence, causes, consequences. Springer Science & Business Media. Chorus, I., 2012b. Current approaches to cyanotoxin risk assessment, risk management and regulations in different countries 151. Chorus, I., Bartram, J., 1999. Toxic cyanobacteria in water: a guide to their public health consequences, monitoring and management. xv + 416 pp. Chorus, I., Falconer, I.R., Salas, H.J., Bartram, J., 2000. Health risks caused by freshwater cyanobacteria in recreational waters. Journal of Toxicology and Environmental Health, Part B 3, 323–347. Christiansen, G., Fastner, J., Erhard, M., Börner, T., Dittmann, E., 2003. Microcystin biosynthesis in Planktothrix: Genes, Evolution, and Manipulation. J Bacteriol 185, 564–572. Christoffersen, K., Lyck, S., Winding, A., 2002. Microbial activity and bacterial community structure during degradation of microcystins. Aquat. Microb. Ecol. 27, 125–136. Clarke, K.R., Gorley, R.N., 2015. PRIMER Version 7.0.12: User Manual/Tutorial. (PRIMER-E, Plymouth).

118

REFERENCES

Codd, G.A., Metcalf, J.S., Beattie, K.A., 1999. Retention of Microcystis aeruginosa and microcystin by salad lettuce (Lactuca sativa) after spray irrigation with water containing cyanobacteria. Toxicon 37, 1181–1185. Cox, P.A., Banack, S.A., Murch, S.J., 2003. Biomagnification of cyanobacterial neurotoxins and neurodegenerative disease among the Chamorro people of Guam. Proc. Natl. Acad. Sci. U.S.A. 100, 13380–13383. Crush, J.R., Briggs, L.R., Sprosen, J.M., Nichols, S.N., 2008. Effect of irrigation with lake water containing microcystins on microcystin content and growth of ryegrass, clover, rape, and lettuce. Environmental Toxicology 23, 246–252. Dang, H., Lovell, C.R., 2015. Microbial surface colonization and biofilm development in marine environments. Microbiol. Mol. Biol. Rev. MMBR 80, 91–138. Danovaro, R., Pusceddu, A., 2007. Biodiversity and ecosystem functioning in coastal lagoons: Does microbial diversity play any role? Estuar. Coast. Shelf Sci., Biodiversity and Ecosystem Functioning in Coastal and Transitional Waters 75, 4–12. Daus, M., Koberger, K., Gnutzmann, N., Hertrich, T., Glaser, R., 2019. Transferring water while transforming landscape: New societal implications, perceptions and challenges of management in the reservoir system franconian lake district. Water 11(12), 2469 Davis, T.W., Orr, P.T., Boyer, G.L., Burford, M.A., 2014. Investigating the production and release of cylindrospermopsin and deoxy-cylindrospermopsin by Cylindrospermopsis raciborskii over a natural growth cycle. Harmful Algae 31, 18–25. de Bernardi, R., Giussani, G., 1990. Are blue-green algae a suitable food for zooplankton? An overview. Hydrobiologia 200, 29–41. de Brouwer, J.F.C., Wolfstein, K., Ruddy, G.K., Jones, T.E.R., Stal, L.J., 2005. Biogenic stabilization of intertidal sediments: The importance of extracellular polymeric substances produced by benthic diatoms. Microb. Ecol. 49, 501–512. De Nobel, W.T. (Pim), Huisman, J., Snoep, J.L., Mur, L.R., 1997. Competition for phosphorus between the nitrogen-fixing cyanobacteria Anabaena and Aphanizomenon. FEMS Microbiol Ecol 24, 259–267. De Nobel, W.T. (Pim), Matthijs, H.C.P., von Elert, E., Mur, L.R., 1998. Comparison of the light-limited growth of the nitrogen-fixing cyanobacteria Anabaena and Aphanizomenon. The New Phytologist 138, 579–587. Deborde, J., Anschutz, P., Guérin, F., Poirier, D., Marty, D., Boucher, G., Thouzeau, G., Canton, M., Abril, G., 2010. Methane sources, sinks and fluxes in a temperate tidal Lagoon: The Arcachon lagoon (SW France). Estuarine, Coastal and Shelf Science, Elsevier, 89, 256-266. Deemer, B.R., Harrison, J.A., Li, S., Beaulieu, J.J., DelSontro, T., Barros, N., Bezerra-Neto, J.F., Powers, S.M., dos Santos, M.A., Vonk, J.A., 2016. Greenhouse gas emissions from reservoir water surfaces: A new global synthesis. BioScience 66, 949–964. Dellow, S., Massey, C.I., McColl, S.T., Townsend, D.B., Villeneuve, M., 2017. Landslides caused by the 14 November 2016 Kaikoura earthquake, South Island, New Zealand. 20th NZGS Symposium DelSontro, T., McGinnis, D.F., Sobek, S., Ostrovsky, I., Wehrli, B., 2010. Extreme methane emissions from a Swiss hydropower reservoir: Contribution from bubbling sediments. Environmental Science & Technology, 44 (7), 2419-2425. DeMott, W.R., Gulati, R.D., Donk, E.V., 2001. Daphnia food limitation in three hypereutrophic Dutch lakes: Evidence for exclusion of large-bodied species by interfering filaments of cyanobacteria. Limnology and Oceanography 46, 2054–2060. DeMott, W.R., Zhang, Q.-X., Carmichael, W.W., 1991. Effects of toxic cyanobacteria and purified toxins on the survival and feeding of a copepod and three species of Daphnia. Limnology and Oceanography 36, 1346–1357. Deutscher Bundestag, 2009. Gesetz zur Ordnung des Wasserhaushalts (Wasserhaushaltsgesetz). Dietrich, D., Hoeger, S., 2005. Guidance values for microcystins in water and cyanobacterial supplement products (blue-green algal supplements): a reasonable or misguided approach? Toxicology and applied pharmacology, risk cyanobacterial toxins: Occurance, ecology, detection, toxicology and health effects assessment 203, 273–289.

119

REFERENCES

Dietrich, D.R., Fischer, A., Michel, C., Hoeger, S.J., 2008. Toxin mixture in cyanobacterial blooms – a critical comparison of reality with current procedures employed in human health risk assessment, in: Hudnell, H.K. (Ed.), Cyanobacterial harmful algal llooms: State of the Sscience and research needs, advances in experimental medicine and biology. Springer New York, pp. 885–912. Dittmann, E., Fewer, D.P., Neilan, B.A., 2013. Cyanobacterial toxins: biosynthetic routes and evolutionary roots. FEMS Microbiology Reviews 37, 23–43. Dittmann, E., Wiegand, C., 2006. Cyanobacterial toxins – occurrence, biosynthesis and impact on human affairs. Molecular Nutrition & Food Research 50, 7–17. Dittmann, R., Froehlich, F., Pohl, R., Ostrowski, M. 2009. Optimum multi-objective reservoir operation with emphasis on flood control and ecology. Nat. Hazards Earth Syst. Sci., 9. Dokulil, M.T., Teubner, K., 2000. Cyanobacterial dominance in lakes. Hydrobiologia 438, 1–12. Döll, P., Fiedler, K., Zhang, J., 2009. Global-scale analysis of river flow alterations due to water withdrawals and reservoirs. Hydrol. Earth Syst. Sci. 13, 2413–2432. Dolman, A.M., Rücker, J., Pick, F.R., Fastner, J., Rohrlack, T., Mischke, U., Wiedner, C., 2012. Cyanobacteria and cyanotoxins: The influence of nitrogen versus phosphorus. PLOS ONE 7, e38757. Donk, E.V., Ringelberg, J., 1983. The effect of fungal parasitism on the succession of diatoms in Lake Maarsseveen I (The Netherlands). Freshw. Biol. 13, 241–251. Downing, J.A., Watson, S.B., McCauley, E., 2001. Predicting cyanobacteria dominance in lakes. Can. J. Fish. Aquat. Sci. 58, 1905–1908. Drobac, D., Tokodi, N., Lujić, J., Marinović, Z., Subakov-Simić, G., Dulić, T., Važić, T., Nybom, S., Meriluoto, J., Codd, G.A., Svirčev, Z., 2016. Cyanobacteria and cyanotoxins in fishponds and their effects on fish tissue. Harmful Algae 55, 66–76. Droppo, I.G., D'Andrea, L., Krishnappan, B.G., Jaskot, C., Trapp, B., Basuvaraj, M., Liss, S.N., 2015. Fine-sediment dynamics: towards an improved understanding of sediment erosion and transport J. Soils Sediments 15, 467–479. Dunnington, D.W., Spooner, I.S., 2018. Using a linked table-based structure to encode self- describing multiparameter spatiotemporal data. FACETS 3, 326–337. Dvorak, P., Casamatta, D., Hasler, P., Jahodářová, E., Norwich, A., Poulíčková, A., 2017. Diversity of the cyanobacteria, in: Modern topics in the phototrophic prokaryotes: Environmental and applied aspects. pp. 3–46. Edwards, K.F., Klausmeier, C.A., Litchman, E., 2011. Evidence for a three-way trade-off between nitrogen and phosphorus competitive abilities and cell size in phytoplankton. Ecology 92, 2085–2095. Ehsani, N., Vörösmarty, C.J., Fekete, B.M., Stakhiv, E.Z., 2017. Reservoir operations under climate change: Storage capacity options to mitigate risk. Journal of Hydrology 555, 435–446. Elliott, J.A., 2010. The seasonal sensitivity of cyanobacteria and other phytoplankton to changes in flushing rate and water temperature. Global Change Biology 16, 864–876. Elliott, J.A., 2012. Is the future blue-green? A review of the current model predictions of how climate change could affect pelagic freshwater cyanobacteria. Water Research 46, 1364–1371. Encinas Fernandez, J., Peeters, F., Hofmann, H., 2014. Importance of the autumn overturn and anoxic conditions in the hypolimnion for the annual methane emissions from a temperate lake. Environmental Science & Technology, 48, 7297-7304. Engle, D., Melack, J.M., 2000. Methane emissions from an Amazon floodplain lake: Enhanced release during episodic mixing and during falling water. Biogeochemistry, 51 (1), 71-90. EPA, 2015. Recommendations for public water systems to manage cyanotoxins in drinking water (No. EPA 815-R-15-010). U.S. Environmental Protection Agency (U.S. EPA). Ernst, B., Hoeger, S.J., O’Brien, E., Dietrich, D.R., 2006. Oral toxicity of the microcystin-containing cyanobacterium Planktothrix rubescens in European whitefish (Coregonus lavaretus). Aquatic Toxicology 79, 31–40. Ernst, B., Höger, S.J., O´Brien, E., Dietrich, D.R., 2009. Abundance and toxicity of Planktothrix rubescens in the pre-alpine Lake Ammersee, Germany. Harmful Algae 8, 329–342.

120

REFERENCES

Esteves-Ferreira, A.A., Cavalcanti, J.H.F., Vaz, M.G.M.V., Alvarenga, L.V., Nunes-Nesi, A., Araújo, W.L., 2017. Cyanobacterial nitrogenases: phylogenetic diversity, regulation and functional predictions. Genet Mol Biol 40, 261–275. EU, 1998. Council Directive 98/83/EC on the quality of water intended for human consumption. EU, 2000. Directive 2000/60/EC of the European Parliament and of the Council establishing a framework for Community action in the field of water policy EU, 2006. Directive 2006/7/EC of the European Parliament and the Council concerning the management of bathing water quality. EU, 2007. Directive 2007/60/EC of the European Parliament and the Council on the assessment and management of flood risks. Fabbro, L.D., Duivenvoorden, L.J., 1996. Profile of a bloom of the cyanobacterium Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju in the Fitzroy River in tropical central Queensland. Mar. Freshw. Res. 47, 685–694. Fallon, R.D., Brock, T.D., 1981. Overwintering of Microcystis in Lake Mendota. Freshw. Biol. 11, 217–226. Fan, J. and Morris, G.L. 1992. Reservoir Sedimentation. II: Reservoir desiltation and long-term storage capacity. Journal of Hydraulic Engineering, ASCE, 118(3). Fang, H., Fazeli, M., Cheng, W., Dey, S., 2016. Transport of biofilm-coated sediment particles J. Hydraul. Res. 54, 631–645. Fang, H., Fazeli, M., Cheng, W., Huang, L., Hu, H., 2015. Biostabilization and transport of cohesive sediment deposits in the three gorges reservoir. PLoS ONE 10. Fang, H., Zhao, H., Shang, Q., Chen, M., 2012. Effect of biofilm on the rheological properties of cohesive sediment. Hydrobiologia 694, 171–181. Fang, H.W., Lai, H.J., Cheng, W., Huang, L., He, G.J., 2017. Modeling sediment transport with an integrated view of the biofilm effects. Water Resour. Res. 53, 7536–7557. FAO/IAEA. 2017. Use of 137Cs for soil erosion assessment. Fulajtar, E., Mabit, L., Renschler, C.S., Lee Zhi Yi, A., Food and Agriculture Organization of the United Nations, Rome, Italy. 64p. Farrer, R.A., Martel, A., Verbrugghe, E., Abouelleil, A., Ducatelle, R., Longcore, J.E., James, T.Y., Pasmans, F., Fisher, M.C., Cuomo, C.A., 2017. Genomic innovations linked to infection strategies across emerging pathogenic chytrid fungi. Nat. Commun. 8. Fernández, C., Estrada, V., Parodi, E.R., 2015. Factors triggering cyanobacteria dominance and succession during blooms in a hypereutrophic drinking water supply reservoir. Water. Air. Soil Pollut. 226, 73. Ferrão-Filho, A. da S., Kozlowsky-Suzuki, B., 2011. Cyanotoxins: Bioaccumulation and effects on aquatic animals. Mar Drugs 9, 2729–2772. Fischer, A., Hoeger, S.J., Stemmer, K., Feurstein, D.J., Knobeloch, D., Nussler, A., Dietrich, D.R., 2010. The role of organic anion transporting polypeptides (OATPs/SLCOs) in the toxicity of different microcystin congeners in vitro: A comparison of primary human hepatocytes and OATP-transfected HEK293 cells. Toxicology and Applied Pharmacology 245, 9–20. Fischer, H., Pusch, M., 2001. Comparison of bacterial production in sediments, epiphyton and the pelagic zone of a lowland river Freshw. Biol. 46, 1335–1348. Fischer, W.J., Altheimer, S., Cattori, V., Meier, P.J., Dietrich, D.R., Hagenbuch, B., 2005. Organic anion transporting polypeptides expressed in liver and brain mediate uptake of microcystin. Toxicology and applied pharmacology, risk cyanobacterial toxins: Occurance, ecology, detection, toxicology and health effects assessment 203, 257–263. Flemming, H.-C., 2020. Biofouling and me: My Stockholm syndrome with biofilms Water Res. 173, 115576. Flemming, H.-C., Wingender, J., 2010. The biofilm matrix Nat. Rev. Microbiol 8, 623–633. Flemming, H.-C., Wuertz, S., 2019. Bacteria and archaea on Earth and their abundance in biofilms Nat. Rev. Microbiol. 17, 247–260. Flint, E., 1975. Phytoplankton in some New Zealand lakes. In: Jolly VH, Brown JMA, (Ed.), New Zealand Lakes. Auckland University Press., Auckland (New Zealand), pp. 63–192. Flombaum, P., Gallegos, J.L., Gordillo, R.A., Rincón, J., Zabala, L.L., Jiao, N., Karl, D.M., Li, W.K.W., Lomas, M.W., Veneziano, D., Vera, C.S., Vrugt, J.A., Martiny, A.C., 2013. Present and future

121

REFERENCES

global distributions of the marine cyanobacteria Prochlorococcus and Synechococcus. Proc Natl Acad Sci U S A 110, 9824–9829. Fontanillo, M., Köhn, M., 2018. Microcystins: Synthesis and structure–activity relationship studies toward PP1 and PP2A. Bioorg. Med. Chem., Constrained Peptides and Biological Targets 26, 1118–1126. Förstner, U., Heise, S., Schwartz, R., Westrich, B., Ahlf, W., 2004. Historical contaminated sediments and soils at the river basin scale. J. Soils Sediments 4, 247. Foshtomi, M.Y., Braeckman, U., Derycke, S., Sapp, M., Gansbeke, D.V., Sabbe, K., Willems, A., Vincx, M., Vanaverbeke, J., 2015. The link between microbial diversity and nitrogen cycling in marine sediments is modulated by macrofaunal bioturbation. PLOS ONE 10, e0130116. Foy, R.H., Gibson, C.E., Smith, R.V., 1976. The influence of daylength, light intensity and temperature on the growth rates of planktonic blue-green algae. British Phycological Journal 11, 151–163. Francis, G., 1878. Poisonous Australian Lake. Nature 18, 11–12. Frenken, T., Alacid, E., Berger, S.A., Bourne, E.C., Gerphagnon, M., Grossart, H.-P., Gsell, A.S., Ibelings, B.W., Kagami, M., Küpper, F.C., Letcher, P.M., Loyau, A., Miki, T., Nejstgaard, J.C., Rasconi, S., Reñé, A., Rohrlack, T., Rojas‐Jimenez, K., Schmeller, D.S., Scholz, B., Seto, K., Sime‐Ngando, T., Sukenik, A., Waal, D.B.V. de, Wyngaert, S.V. den, Donk, E.V., Wolinska, J., Wurzbacher, C., Agha, R., 2017. Integrating chytrid fungal parasites into plankton ecology: research gaps and needs. Environ. Microbiol. 19, 3802–3822. Fristachi, A., Sinclair, J.L., Hall, S., Berkman, J.A.H., Boyer, G., Burkholder, J., Burns, J., Carmichael, W., DuFour, A., Frazier, W., Morton, S.L., O’Brien, E., Walker, S., 2008. Occurrence of cyanobacterial harmful algal blooms workgroup report, in: Hudnell, H.K. (Ed.), Cyanobacterial harmful algal blooms: State of the science and research needs, Advances in experimental medicine and biology. Springer, New York, NY, pp. 45–103. Gallon, J.R., 1992. Reconciling the incompatible: N2 fixation and O2. New Phytologist. Gan, N., Xiao, Y., Zhu, L., Wu, Z., Liu, J., Hu, C., Song, L., 2012. The role of microcystins in maintaining colonies of bloom-forming Microcystis spp. Environmental Microbiology 14, 730– 742. Garcia-Pichel, F., Belnap, J., Neuer, S., Schanz, F., 2003. Estimates of global cyanobacterial biomass and its distribution. Algological Studies 109, 213–227. Geesey, G.G., Mutch, R., Costerton, J.W., Green, R.B., 1978. Sessile bacteria: An important component of the microbial population in small mountain streams 1. Limnol. Oceanogr. 23, 1214–1223. Gerbersdorf, S.U., Jancke, T., Westrich, B., Paterson, D.M., 2008. Microbial stabilization of riverine sediments by extracellular polymeric substances. Geobiology 6, 57–69. Gerbersdorf, S.U., Wieprecht, S., 2015. Biostabilization of cohesive sediments: revisiting the role of abiotic conditions, physiology and diversity of microbes, polymeric secretion, and biofilm architecture. Geobiology 13, 68–97. German Ministry of Justice, 2020. Verordnung über die Qualität von Wasser für den menschlichen Gebrauch (Trinkwasserverordnung) Gerphagnon, M., Colombet, J., Latour, D., Sime-Ngando, T., 2017. Spatial and temporal changes of parasitic chytrids of cyanobacteria. Sci. Rep. 7, 6056. Gerphagnon, M., Latour, D., Colombet, J., Sime-Ngando, T., 2013. Fungal Parasitism: Life cycle, dynamics and impact on cyanobacterial blooms. PLOS ONE 8, e60894. Gerphagnon, M., Macarthur, D.J., Latour, D., Gachon, C.M.M., Ogtrop, F.V., Gleason, F.H., Sime‐ Ngando, T., 2015. Microbial players involved in the decline of filamentous and colonial cyanobacterial blooms with a focus on fungal parasitism. Environ. Microbiol. 17, 2573–2587. Gevers, D., Cohan, F.M., Lawrence, J.G., Spratt, B.G., Coenye, T., Feil, E.J., Stackebrandt, E., Peer, Y.V. de, Vandamme, P., Thompson, F.L., Swings, J., 2005. Re-evaluating prokaryotic species. Nat Rev Microbiol 3, 733–739. Gibbs, R.J., 1983. Effect of natural organic coatings on the coagulation of particles Environ. Sci. Technol. 17, 237–240.

122

REFERENCES

Giesecke, J., Mosonyi, E., 2009. Wasserkraftanlagen: Planung, Bau und Betrieb, 5th ed. Springer- Verlag, Berlin Heidelberg. Giguet-Covex, C., Ficetola, G.F., Walsh, K., Poulenard, J., Bajard, M., Fouinat, L., Sabatier, P., Gielly, L., Messager, E., Develle, A.L., David, F., Taberlet, P., Brisset, E., Guiter, F., Sinet, R., Arnaud, F., 2019. New insights on lake sediment DNA from the catchment: importance of taphonomic and analytical issues on the record quality. Sci. Rep. 9, 14676. Gilbert, J.J., 1996. Effect of food availability on the response of planktonic rotifers to a toxic strain of the cyanobacterium Anabaena flos-aquae. Limnology and Oceanography 41, 1565–1572. Gilbertson, W.W., Solan, M., Prosser, J.I., 2012. Differential effects of microorganism–invertebrate interactions on benthic nitrogen cycling FEMS Microbiol. Ecol. 82, 11–22. Gilroy, D.J., Kauffman, K.W., Hall, R.A., Huang, X., Chu, F.S., 2000. Assessing potential health risks from microcystin toxins in blue-green algae dietary supplements. Environ Health Perspect 108, 435–439. Gkelis, S., Lanaras, T., Sivonen, K., 2006. The presence of microcystins and other cyanobacterial bioactive peptides in aquatic fauna collected from Greek freshwaters. Aquatic Toxicology 78, 32–41. Glaser, R. 2014. Global change. Das neue Gesicht der Erde. Primus Verlag, Darmstadt, Germany. Gleason, F.H., Jephcott, T.G., Küpper, F.C., Gerphagnon, M., Sime-Ngando, T., Karpov, S.A., Guillou, L., van Ogtrop, F.F., 2015. Potential roles for recently discovered chytrid parasites in the dynamics of harmful algal blooms. Fungal Biol. Rev. 29, 20–33. Godde, D., Engels, K., Schmid, S., Achatz, R., Haupt, O., Beer, C., Höller, S., Jaberg, H., Miller, B., Heigerth, G., Niemann, A., Perau, E., Schreiber, U., Koch, M.K., Schüttrumpf, H., Pummer, E., Günther, M., Plenker, D., 2015. In: Pumpspeicherkraftwerke, Heimerl, S. (Ed.), Wasserkraftprojekte, Band 2. Ausgewählte Beiträge aus der Fachzeitschrift WasserWirtschaft. Springer Vieweg, Wiesbaden, Germany, pp. 277-354. Gons, H.J., Ebert, J., Hoogveld, H.L., van den Hove, L., Pel, R., Takkenberg, W., Woldringh, C.J., 2002. Observations on cyanobacterial population collapse in eutrophic lake water. Antonie Van Leeuwenhoek 81, 319–326. Griffiths, D.J., Saker, M.L., 2003. The Palm Island mystery disease 20 years on: A review of research on the cyanotoxin cylindrospermopsin. Environmental Toxicology 18, 78–93. Grossman, A.R., Schaefer, M.R., Chiang, G.G., Collier, J.L., 1993. The phycobilisome, a light- harvesting complex responsive to environmental conditions. Microbiology and Molecular Biology Reviews 57, 725–749. Gsell, A.S., de Senerpont Domis, L.N., Verhoeven, K.J., van Donk, E., Ibelings, B.W., 2013. Chytrid epidemics may increase genetic diversity of a diatom spring-bloom. ISME J 7, 2057–2059. Gucht, K.V. der, Cottenie, K., Muylaert, K., Vloemans, N., Cousin, S., Declerck, S., Jeppesen, E., Conde-Porcuna, J.-M., Schwenk, K., Zwart, G., Degans, H., Vyverman, W., Meester, L.D., 2007. The power of species sorting: Local factors drive bacterial community composition over a wide range of spatial scales. Proc. Natl. Acad. Sci. 104, 20404–20409. https://doi.org/10.1073/pnas.0707200104 Guillard, R.R.L., Lorenzen, C.J., 1972. Yellow-green algae with Chlorophyllide C1,2. Journal of Phycology 8, 10–14. Guiry, M.D., 2012. How many species of algae are there? Journal of Phycology 48, 1057–1063. Guiry, M.D., Guiry, G.D., 2020. AlgaeBase. World-wide electronic publication, National University of Ireland, Galway. https://www.algaebase.org; searched on 6 July 2020. Tuson, H.H., Weibel, D.B., 2013. Bacteria–surface interactions Soft Matter 9, 4368–4380. Haddeland, I., Heinke, J., Biemans, H., Eisner, S., Flörke, M., Hanasaki, N., Konzmann, M., Ludwig, F., Masaki, Y., Schewe, J., Stacke, T., Tessler, Z.D., Wada, Y., Wisser, D., 2014. Global water resources affected by human interventions and climate change Proc. Natl. Acad. Sci. 111, 3251–3256. Halland, N., Brönstrup, M., Czech, J., Czechtizky, W., Evers, A., Follmann, M., Kohlmann, M., Schiell, M., Kurz, M., Schreuder, H.A., Kallus, C., 2015. Novel small molecule inhibitors of activated thrombin activatable fibrinolysis inhibitor (TAFIa) from natural product anabaenopeptin. J. Med. Chem. 58, 4839–4844.

123

REFERENCES

Halstvedt, C.B., Rohrlack, T., Andersen, T., Skulberg, O., Edvardsen, B., 2007. Seasonal dynamics and depth distribution of Planktothrix spp. in Lake Steinsfjorden (Norway) related to environmental factors. J Plankton Res 29, 471–482. Hamilton, D., Wood, S., Dietrich, D., Puddick, J., 2013. Costs of harmful blooms of freshwater cyanobacteria. In: Cyanobacteria: An Economic Perspective. pp. 245–256. Hanson, T.R., Upton Hatch, L., Clonts, H.C. 2007. Reservoir water level impacts on recreation, property and non user values. In: Journal of the American water resource association. 38/4, 1007-1018. Harada, K., Fujii, K., Shimada, T., Suzuki, M., Sano, H., Adachi, K., Carmichael, W.W., 1995. Two cyclic peptides, anabaenopeptins, a third group of bioactive compounds from the cyanobacterium Anabaena flos-aquae NRC 525-17. Tetrahedron Letters 36, 1511–1514. Harb, G. 2013. Numerical modeling of sediment transport processes in alpine reservoirs dissertation, Schriftenreihe zur Wasserwirtschaft der Technischen Universität Graz, Band 73. Harms, H., Kurita, K.L., Pan, L., Wahome, P.G., He, H., Kinghorn, A.D., Carter, G.T., Linington, R.G., 2016. Discovery of anabaenopeptin 679 from freshwater algal bloom material: Insights into the structure–activity relationship of anabaenopeptin protease inhibitors. Bioorganic & Medicinal Chemistry Letters 26, 4960–4965. Harpole, W.S., Ngai, J.T., Cleland, E.E., Seabloom, E.W., Borer, E.T., Bracken, M.E.S., Elser, J.J., Gruner, D.S., Hillebrand, H., Shurin, J.B., Smith, J.E., 2011. Nutrient co-limitation of primary producer communities. Ecology Letters 14, 852–862. Harrison, J.A., Deemer, B.R., Birchfield, M.K., O’Malley, M.T., 2017. Reservoir water-level drawdowns accelerate and amplify methane emission Environmental Science & Technology, 51(3), 1267-1277. Haugland, R.A., Siefring, S.C., Wymer, L.J., Brenner, K.P., Dufour, A.P., 2005. Comparison of Enterococcus measurements in freshwater at two recreational beaches by quantitative polymerase chain reaction and membrane filter culture analysis. Water Res. 39, 559–568. Haun, S., Kjærås, H., Løvfall, S. and Olsen, N.R.B., 2013. Three-dimensional measurements and numerical modelling of suspended sediments in a hydropower reservoir. Journal of Hydrology 479, 180-188. Havens, K.E., James, R.T., East, T.L., Smith, V.H., 2003. N:P ratios, light limitation, and cyanobacterial dominance in a subtropical lake impacted by non-point source nutrient pollution. Environmental Pollution 122, 379–390. Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison, W.C., Dortch, Q., Gobler, C.J., Heil, C.A., Humphries, E., Lewitus, A., Magnien, R., Marshall, H.G., Sellner, K., Stockwell, D.A., Stoecker, D.K., Suddleson, M., 2008. Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae, HABs and Eutrophication 8, 3–13. Hereman, T.C., Bittencourt‐Oliveira, M. do C., 2012. Bioaccumulation of Microcystins in Lettuce. Journal of Phycology 48, 1535–1537. Herlemann, D.P., Labrenz, M., Jürgens, K., Bertilsson, S., Waniek, J.J., Andersson, A.F., 2011. Transitions in bacterial communities along the 2000 km salinity gradient of the Baltic Sea. ISME J. 5, 1571–1579. Herrero, A., Muro-Pastor, A.M., Flores, E., 2001. Nitrogen control in cyanobacteria. Journal of Bacteriology 183, 411–425. Hertwich, E.G., 2013. Addressing biogenic greenhouse gas emissions from hydropower in LCA Environ. Sci. Technol., 47(17), 9604-11. Higgins, S.N., Paterson, M.J., Hecky, R.E., Schindler, D.W., Venkiteswaran, J.J., Findlay, D.L., 2018. Biological nitrogen fixation prevents the response of a eutrophic lake to reduced loading of nitrogen: Evidence from a 46-Year whole-lake experiment. Ecosystems 21, 1088–1100. Hilborn, E.D., Roberts, V.A., Backer, L., DeConno, E., Egan, J.S., Hyde, J.B., Nicholas, D.C., Wiegert, E.J., Billing, L.M., DiOrio, M., Mohr, M.C., Hardy, F.J., Wade, T.J., Yoder, J.S., Hlavsa, M.C., 2014. Algal bloom–associated disease outbreaks among users of freshwater lakes - United States, 2009–2010. MMWR Morb Mortal Wkly Rep 63, 11–15. Hoiczyk, E., Hansel, A., 2000. Cyanobacterial cell walls: News from an unusual prokaryotic envelope. Journal of Bacteriology 182, 1191–1199.

124

REFERENCES

Holfeld, H., 1998. Fungal infections of the phytoplankton: seasonality, minimal host density, and specificity in a mesotrophic lake. New Phytol. 138, 507–517. Holl, C.M., Montoya, J.P., 2005. Interactions between nitrate uptake and nitrogen fixation in continuous cultures of the marine diazotroph Trichodesmium (cyanobacteria)1. Journal of Phycology 41, 1178–1183. Holland, A., Kinnear, S., 2013. Interpreting the Possible Ecological Role(s) of Cyanotoxins: Compounds for competitive advantage and/or physiological aide? Mar Drugs 11, 2239–2258. Holst, T., Jørgensen, N.O.G., Jørgensen, C., Johansen, A., 2003. Degradation of microcystin in sediments at oxic and anoxic, denitrifying conditions. Water Res. 37, 4748–4760. Holzner, C.P., Aeschbach-Hertig, W., Simona, M., Veronesi, M., Imboden, D.M., Kipfer, R., 2009. Exceptional mixing events in meromictic Lake Lugano (Switzerland/Italy), studied using environmental tracers. Limnology and Oceanography 54, 1113–1124. Howarth, J.D., Fitzsimons, S.J., Norris, R.J., Jacobsen, G.E., 2014. Lake sediments record high intensity shaking that provides insight into the location and rupture length of large earthquakes on the Alpine Fault, New Zealand. Earth Planet. Sci. Lett. 403, 340–351. Huiming, Z., Hongwei, F., Minghong, C., 2011. Floc architecture of bioflocculation sediment by ESEM and CLSM. Scanning 33, 437–445. Huisman, J., Codd, G.A., Paerl, H.W., Ibelings, B.W., Verspagen, J.M.H., Visser, P.M., 2018. Cyanobacterial blooms. Nat. Rev. Microbiol. 16, 471. Huisman, J., Jonker, R.R., Zonneveld, C., Weissing, F.J., 1999. Competition for light between phytoplankton species: Experimental tests of mechanistic theory. Ecology 80, 211–222. Huisman, J., Sharples, J., Stroom, J.M., Visser, P.M., Kardinaal, W.E.A., Verspagen, J.M.H., Sommeijer, B., 2004. Changes in turbulent mixing shift competition for light between phytoplankton species. Ecology 85, 2960–2970. Humpage, A.R., Falconer, I.R., 1999. Microcystin-LR and liver tumor promotion: effects on cytokinesis, ploidy, and apoptosis in cultured hepatocytes. Environmental Toxicology 14, 61– 75. Humpage, A.R., Fontaine, F., Froscio, S., Burcham, P., Falconer, I.R., 2005. Cylindrospermopsin genotoxicity and cytotoxicity: Role of cytochrome P-450 and oxidative stress. Journal of Toxicology and Environmental Health, Part A 68, 739–753. Humpage, A.R., Rositano, J., Bretag, A.H., Brown, R., Baker, P.D., Nicholson, B.C., Steffensen, D.A., 1994. Paralytic shellfish poisons from Australian cyanobacterial blooms. Mar. Freshwater Res. 45, 761–771. Hunter, P.D., Tyler, A.N., Willby, N.J., Gilvear, D.J., 2008. The spatial dynamics of vertical migration by Microcystis aeruginosa in a eutrophic shallow lake: A case study using high spatial resolution time-series airborne remote sensing. Limnol. Oceanogr. 53, 2391–2406. Hwang, J., Pallas, D.C., 2014. STRIPAK complexes: Structure, biological function, and involvement in human diseases. Int. J. Biochem. Cell Biol. 47, 118–148. Hyenstrand, P., Burkert, U., Pettersson, A., Blomqvist, P., 2000. Competition between the green alga Scenedesmus and the cyanobacterium Synechococcus under different modes of inorganic nitrogen supply. Hydrobiologia 435, 91–98. Ibelings, B.W., Backer, L.C., Kardinaal, W.E.A., Chorus, I., 2015. Current approaches to cyanotoxin risk assessment and risk management around the globe. Harmful Algae 49, 63–74. Ibelings, B.W., Bormans, M., Fastner, J., Visser, P.M., 2016. CYANOCOST special issue on cyanobacterial blooms: synopsis - a critical review of the management options for their prevention, control and mitigation. Aquat. Ecol. 50, 595–605. Ibelings, B.W., Bruin, A.D., Kagami, M., Rijkeboer, M., Brehm, M., Donk, E.V., 2004. Host parasite interactions between freshwater phytoplankton and chytrid fungi (Chytridiomycota). Journal of Phycology 40, 437–453. Ibelings, B.W., Gsell, A.S., Mooij, W.M., Donk, E.V., Wyngaert, S.V.D., Domis, L.N.D.S., 2011. Chytrid infections and diatom spring blooms: paradoxical effects of climate warming on fungal epidemics in lakes. Freshwater Biology 56, 754–766. Ibelings, B.W., Havens, K.E., 2008. Cyanobacterial toxins: a qualitative meta–analysis of concentrations, dosage and effects in freshwater, estuarine and marine biota, in: Hudnell, H.K.

125

REFERENCES

(Ed.), Cyanobacterial harmful algal blooms: State of the science and research needs, advances in experimental medicine and biology. Springer, New York, NY, pp. 675–732. Ibelings, B.W., Vonk, M., Los, H.F.J., Molen, D.T. van der, Mooij, W.M., 2003. Fuzzy modeling of cyanobacterial surface waterblooms: Validation with NOAA-AVHRR satellite images. Ecological Applications 13, 1456–1472. International Commission on Large Dams (ICOLD), 2013. Talsperren in Deutschland. Springer Vieweg: Wiesbaden, Germany, 2013. IPCC, 2018. Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty. [Masson-Delmotte, V., P. Zhai, H.-O. Pörtner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani, W. Moufouma-Okia, C. Péan, R. Pidcock, S. Connors, J.B.R. Matthews, Y. Chen, X. Zhou, M.I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, and T. Waterfield (eds.)]. In Press. Ishitsuka, M.O., Kusumi, T., Kakisawa, H., Kaya, K., Watanabe, M.M., 1990. Microviridin. A novel tricyclic depsipeptide from the toxic cyanobacterium Microcystis viridis. J. Am. Chem. Soc. 112, 8180–8182. Izaguirre, G., Hwang, C.J., Krasner, S.W., McGuire, M.J., 1982. Geosmin and 2-Methylisoborneol from cyanobacteria in three water supply systems. Appl Environ Microbiol 43, 708–714. Jacquet, S., Briand, J.-F., Leboulanger, C., Avois-Jacquet, C., Oberhaus, L., Tassin, B., Vinçon- Leite, B., Paolini, G., Druart, J.-C., Anneville, O., Humbert, J.-F., 2005. Proliferation of the toxic cyanobacterium Planktothrix rubescens following restoration of the largest natural french lake (Lac du Bourget). Harmful Algae 4, 651–672. Jähnichen, S., Ihle, T., Petzoldt, T., Benndorf, J., 2007. Impact of inorganic carbon availability on microcystin production by Microcystis aeruginosa PCC 7806. Appl. Environ. Microbiol. 73, 6994–7002. Jang, M.-H., Ha, K., Joo, G.-J., Takamura, N., 2003. Toxin production of cyanobacteria is increased by exposure to zooplankton. Freshwater Biology 48, 1540–1550. Janssen, E. M. L., 2019. Cyanobacterial peptides beyond microcystins-a review on co-occurrence, toxicity, and challenges for risk assessment. Water Research, 151, 488-499. Jensen, J.P., Jeppesen, E., Kristensen, P., Christensen, P.B., Søndergaard, M., 1992. Nitrogen loss and denitrification as studied in relation to reductions in nitrogen loading in a Shallow, hypertrophic lake (Lake Søbygård, Denmark). Internationale Revue der gesamten Hydrobiologie und Hydrographie 77, 29–42. Jeppesen, E., Søndergaard, M., Jensen, J.P., Havens, K.E., Anneville, O., Carvalho, L., Coveney, M.F., Deneke, R., Dokulil, M.T., Foy, B., Gerdeaux, D., Hampton, S.E., Hilt, S., Kangur, K., Köhler, J., Lammens, E.H.H.R., Lauridsen, T.L., Manca, M., Miracle, M.R., Moss, B., Nõges, P., Persson, G., Phillips, G., Portielje, R., Romo, S., Schelske, C.L., Straile, D., Tatrai, I., Willén, E., Winder, M., 2005. Lake responses to reduced nutrient loading – an analysis of contemporary long-term data from 35 case studies. Freshwater Biology 50, 1747–1771. Johansson, E., Legrand, C., Björnerås, C., Godhe, A., Mazur-Marzec, H., Säll, T., Rengefors, K., 2019. High diversity of microcystin chemotypes within a summer bloom of the cyanobacterium Microcystis botrys. Toxins 11, 698. Jöhnk, K.D., Huisman, J., Sharples, J., Sommeijer, B., Visser, P.M., Stroom, J.M., 2008. Summer heatwaves promote blooms of harmful cyanobacteria. Global Change Biology 14, 495–512. Jones, S., 2017. Goo, glue, and grain binding: Importance of biofilms for diagenesis in sandstones. Geology 45, 959–960. Jungblut, A.D., Lovejoy, C., Vincent, W.F., 2010. Global distribution of cyanobacterial ecotypes in the cold biosphere. ISME J 4, 191–202. Kaebernick, M., Neilan, B.A., 2001. Ecological and molecular investigations of cyanotoxin production. FEMS Microbiol Ecol 35, 1–9. Kaplan, A., Harel, M., Kaplan-Levy, R.N., Hadas, O., Sukenik, A., Dittmann, E., 2012. The languages spoken in the water body (or the biological role of cyanobacterial toxins). Front Microbiol 3, 138.

126

REFERENCES

Kaplan, A., Schwarz, R., Lieman-Hurwitz, J., Reinhold, L., 1991. Physiological and molecular aspects of the inorganic carbon-concentrating mechanism in cyanobacteria 1. Plant Physiol 97, 851–855. Kardinaal, W.E.A., Janse, I., Agterveld, M.K., Meima, M., Snoek, J., Mur, L.R., Huisman, J., Zwart, G., Visser, P.M., 2007. Microcystis genotype succession in relation to microcystin concentrations in freshwater lakes. Aquatic Microbial Ecology 48, 1–12. Kearns, K.D., Hunter, M.D., 2001. Toxin-producing Anabaena flos-aquae induces settling of Chlamydomonas reinhardtii, a competing motile alga. Microb Ecol 42, 80–86. Kellmann, R., Mihali, T.K., Jeon, Y.J., Pickford, R., Pomati, F., Neilan, B.A., 2008. Biosynthetic intermediate analysis and functional homology reveal a saxitoxin gene cluster in cyanobacteria. Appl Environ Microbiol 74, 4044–4053. Kelly, D., Shearer, K., Schallenberg, M., 2013. Nutrient loading to shallow coastal lakes in Southland for sustaining ecological integrity values. Cawthron Report 2375. Prepared for Environment Southland. Kirchherr, J., Charles, K.J., 2016. The social impacts of dams. A new framework for scholarly analysis. Environmental Impact Assessment Review 60, 99-114. Kirchherr, J., Pohlner, H., Charles, K.J., 2016. Cleaning up the big muddy. A meta-synthesis of the research on the social impact of dams. Environmental Impact Assessment Review 60, 115- 125. Kitchens, C.M., Johengen, T.H., Davis, T.W., 2018. Establishing spatial and temporal patterns in Microcystis sediment seed stock viability and their relationship to subsequent bloom development in Western Lake Erie. PLOS ONE 13, e0206821. Kleinhenz, A., Koenig, A., 2018. Home ranges and movements of resident graylag geese (Anser anser) in breeding and winter habitats in Bavaria, South Germany. PloS one 13, e0202443. Kleinteich, J., Hildebrand, F., Wood, S.A., Ciŕs, S., Agha, R., Quesada, A., Pearce, D.A., Convey, P., Küpper, F.C., Dietrich, D.R., 2014. Diversity of toxin and non-toxin containing cyanobacterial mats of meltwater ponds on the Antarctic Peninsula: a pyrosequencing approach. Antarctic Science 26, 521–532. Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., Glöckner, F.O., 2013. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 41. Kling, H., Watson, S., Mccullough, G., Stainton, M., 2011. Bloom development and phytoplankton succession in Lake Winnipeg: A comparison of historical records with recent data. Aquat. Ecosyst. Health Manag. 14, 219–224. Komárek, J., 2010. Recent changes (2008) in cyanobacteria taxonomy based on a combination of molecular background with phenotype and ecological consequences (genus and species concept). Hydrobiologia 639, 245–259. Komárek, J., 2016. A polyphasic approach for the taxonomy of cyanobacteria: principles and applications. European Journal of Phycology 51, 346–353. Komarek, J., Zapomelova, E., 2007. Planktic morphospecies of the cyanobacterial genus Anabaena = subg. Dolichospermum - 1. part: coiled types. Fottea 7, 1–31. Komarek, J., Zapomelova, E., 2008. Planktic morphospecies of the cyanobacterial genus Anabaena = subg. Dolichospermum - 2. part: straight types. Fottea 8, 1–14. Kondolf, G.M., Gao, Y., Annandale, G.W., Morris, G.L., Jiang, E., Zhang, J., Cao, Y., Carling, P., Fu, K., Guo, Q., Hotchkiss, R., Peteuil, Ch., Sumi, T., Wang, H.-W., Wang, Z., Wei, Z., Wu, B., Wu, C. and Ted Yang, Ch.T., 2014. Sustainable sediment management in reservoirs and regulated rivers: Experiences from five continents. Earth’s Future 2(5), 256-280. Kornijów, R., 2009. Controversies around dam reservoirs. Benefits, costs and future. Ecohydrology & Hydrobiology 9, 141-148. Kosol, S., Schmidt, J., Kurmayer, R., 2009. Variation in peptide net production and growth among strains of the toxic cyanobacterium Planktothrix spp. European Journal of Phycology 44, 49– 62.

127

REFERENCES

Kosten, S., Huszar, V.M., Bécares, E., Costa, S., Donk, E. van, Hansson, L., Lurling, M.F.L.L.W., 2012. Warmer climates boost cyanobacterial dominance in shallow lakes. Global Change Biology 18, 118–126. Kumar, K., Mella-Herrera, R.A., Golden, J.W., 2010. Cyanobacterial heterocysts. Cold Spring Harb Perspect Biol 2. Kurmayer, R., 2011. The toxic cyanobacterium Nostoc sp. strain 152 produces highest amounts of microcystin and nostophycin under stress conditions. J Phycol 47, 200–207. Kurmayer, R., Christiansen, G., Fastner, J., Börner, T., 2004. Abundance of active and inactive microcystin genotypes in populations of the toxic cyanobacterium Planktothrix spp. Environ. Microbiol. 6, 831–841. Kurmayer, R., Kutzenberger, T., 2003. Application of Real-Time PCR for Quantification of microcystin genotypes in a population of the toxic cyanobacterium Microcystis sp. Appl. Environ. Microbiol. 69, 6723–6730. Lacap-Bugler, D.C., Lee, K.K., Archer, S., Gillman, L.N., Lau, M.C.Y., Leuzinger, S., Lee, C.K., Maki, T., McKay, C.P., Perrott, J.K., de los Rios-Murillo, A., Warren-Rhodes, K.A., Hopkins, D.W., Pointing, S.B., 2017. Global diversity of desert hypolithic cyanobacteria. Front. Microbiol. 8. Lai, J.-S., Shen, H.W., 1996. Flushing sediment through reservoirs. Journal of Hydraulic Research 34, 237–255. Layzell, D.B., Turpin, D.H., Elrifi, I.R., 1985. Effect of N source on the steady state growth and N assimilation of P-limited Anabaena flos-aquae. Plant Physiol. 78, 739–745. Leflaive, J., Ten‐Hage, L., 2007. Algal and cyanobacterial secondary metabolites in freshwaters: a comparison of allelopathic compounds and toxins. Freshwater Biology 52, 199–214. Legrand, B., Lamarque, A., Sabart, M., Latour, D., 2017. Benthic archives reveal recurrence and dominance of toxigenic cyanobacteria in a eutrophic lake over the last 220 years. Toxins 9. Lévesque, B., Gervais, M.-C., Chevalier, P., Gauvin, D., Anassour-Laouan-Sidi, E., Gingras, S., Fortin, N., Brisson, G., Greer, C., Bird, D., 2014. Prospective study of acute health effects in relation to exposure to cyanobacteria. Science of The Total Environment 466–467, 397–403. Levitan, O., Rosenberg, G., Setlik, I., Setlikova, E., Grigel, J., Klepetar, J., Prasil, O., Berman‐Frank, I., 2007. Elevated CO2 enhances nitrogen fixation and growth in the marine cyanobacterium Trichodesmium. Global Change Biology 13, 531–538. Lewis, W.M., Wurtsbaugh, W.A., 2008. Control of lacustrine phytoplankton by nutrients: Erosion of the phosphorus paradigm. International Review of Hydrobiology 93, 446–465. Lezcano, M., Agha, R., Cirés, S., Quesada, A., 2018. Spatial-temporal survey of Microcystis oligopeptide chemotypes in reservoirs with dissimilar waterbody features and their relation to genetic variation. Harmful Algae 81. Li, S. Y., Zhang, Q. F., Bush, R. T. and Sullivan, L. A. 2015. Methane and CO2 emissions from China’s hydroelectric reservoirs: a new quantitative synthesis. Environmental Science and Pollution Research 22(7), 5325–5339. Li, X., Dreher, T.W., Li, R., 2016. An overview of diversity, occurrence, genetics and toxin production of bloom-forming Dolichospermum (Anabaena) species. Harmful Algae 54, 54–68. Li, Z., Hobson, P., An, W., Burch, M.D., House, J., Yang, M., 2012. Earthy odor compounds production and loss in three cyanobacterial cultures. Water Research 46, 5165–5173. Lindström, E.S., Langenheder, S., 2012. Local and regional factors influencing bacterial community assembly. Environ. Microbiol. Rep. 4, 1–9. Lozupone, C.A., Knight, R., 2007. Global patterns in bacterial diversity. Proc. Natl. Acad. Sci. 104, 11436. LUBW, 2010. ISF Jahresbericht 2009/2010 (in german). Landesanstalt für Umwelt, Messungen und Naturschutz Baden-Württemberg, Karlsruhe. Lukač, M., Aegerter, R., 1993. Influence of trace metals on growth and toxin production of Microcystis aeruginosa. Toxicon 31, 293–305. Luyssaert, S., Abri,l G., Andres, R., Bastviken, D., Bellassen, V., Bergamaschi, P., Bousquet, P., Chevallier, F., Ciais, P., Corazza, M., Dechow, R., Erb, K.H., Etiope, G., Fortems-Cheiney, A., Grassi, G., Hartmann, J., Jung, M., Lathière, J., Lohila, A., Mayorga, E., Moosdorf, N., Njakou, D.S., Otto, J., Papale, D., Peters, W., Peylin, P., Raymond, P., Rödenbeck, C., Saarnio, S.,

128

REFERENCES

Schulze, E.D., Szopa, S., Thompson, R., Verkerk, P.J., Vuichard, N., Wang, R., Wattenbach, M., Zaehle, S., 2012. The European land and inland water CO2, CO, CH4 and N2O balance between 2001 and 2005. Biogeosciences, 9, 3357-3380. Madsen, E.L., 2011. Microorganisms and their roles in fundamental biogeochemical cycles. Curr. Opin. Biotechnol., Energy biotechnology – Environmental biotechnology 22, 456–464. Maeck, A., DelSontro, T., McGinnis, D. F., Fischer, H., Flury, S., Schmidt, M., Fietzek, P. and Lorke, A., 2013. Sediment trapping by dams creates methane emission hot spots. Environmental Science and Technology 47(15), 8130–8137. Maeck, A., Hofmann, H., Lorke, A., 2014. Pumping methane out of aquatic sediments - ebullition forcing mechanisms in an impounded river. Biogeosciences, 11 (11), 2925-2938. Mahmood, K., 1987. Reservoir sedimentation: impact, extent, and mitigation. Technical paper (No. PB-88-113964/XAB; WORLD-BANK-TP-71). International Bank for Reconstruction and Development, Washington, DC (USA). Malarkey, J., Baas, J.H., Hope, J.A., Aspden, R.J., Parsons, D.R., Peakall, J., Paterson, D.M., Schindler, R.J., Ye, L., Lichtman, I.D., Bass, S.J., Davies, A.G., Manning, A.J., Thorne, P.D., 2015. The pervasive role of biological cohesion in bedform development. Nat. Commun. 6. Malbrouck, C., Kestemont, P., 2006. Effects of microcystins on fish. Environmental Toxicology and Chemistry 25, 72–86. Manganelli, M., Stefanelli, M., Vichi, S., Andreani, P., Nascetti, G., Scialanca, F., Scardala, S., Testai, E., Funari, E., 2016. Cyanobacteria biennal dynamic in a volcanic mesotrophic lake in central Italy: Strategies to prevent dangerous human exposures to cyanotoxins. Toxicon 115, 28–40. Martin, C., Oberer, L., Ino, T., König, W.A., Busch, M., Weckesser, J., 1993. Cyanopeptolins, new depsipeptides from the cyanobacterium Microcystis sp. PCC 7806. J. Antibiot. 46, 1550–1556. Martin-Creuzburg, D., Elert, E. von, Hoffmann, K.H., 2008. Nutritional constraints at the cyanobacteria - Daphnia magna interface: The role of sterols. Limnology and Oceanography 53, 456–468. Massey, C., Townsend, D., Rathje, E., Allstadt, K.E., Lukovic, B., Kaneko, Y., Bradley, B., Wartman, J., Jibson, R.W., Petley, D.N., Horspool, N., Hamling, I., Carey, J., Cox, S., Davidson, J., Dellow, S., Godt, J.W., Holden, C., Jones, K., Kaiser, A., Little, M., Lyndsell, B., McColl, S., Morgenstern, R., Rengers, F.K., Rhoades, D., Rosser, B., Strong, D., Singeisen, C., Villeneuve, M., 2018. Landslides triggered by the 14 November 2016 Mw 7.8 Kaikōura Earthquake, New Zealand. Bull. Seismol. Soc. Am. 108, 1630–1648. https://doi.org/10.1785/0120170305 McMurdie, P.J., Holmes, S., 2013. phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8 (4): e61217. Medlin, L.K., Orozco, J., 2017. Molecular techniques for the detection of organisms in aquatic environments, with emphasis on harmful algal bloom species. Sensors 17, 1184. Mehnert, G., Leunert, F., Cirés, S., Jöhnk, K.D., Rücker, J., Nixdorf, B., Wiedner, C., 2010. Competitiveness of invasive and native cyanobacteria from temperate freshwaters under various light and temperature conditions. J Plankton Res 32, 1009–1021. Merel, S., Walker, D., Chicana, R., Snyder, S., Baurès, E., Thomas, O., 2013. State of knowledge and concerns on cyanobacterial blooms and cyanotoxins. Environment International 59, 303– 327. Meriluoto, J., Spoof, L., Codd, G.A., 2017. Handbook of cyanobacterial monitoring and cyanotoxin analysis. John Wiley & Sons. Milliman, J.D., Farnsworth, K.L., 2011. River discharge to the coastal ocean: A global synthesis. Cambridge University Press, Cambridge. Milliman, J.D., Syvitski, J.P.M., 1992. Geomorphic/tectonic control of sediment discharge to the ocean: the importance of small mountainous rivers. Journal of Geology 100, 325– 344. Ministry for the Environment & Stats NZ, 2019. New Zealand’s environmental reporting series: Environment Aotearoa 2019. Available from www.mfe.govt.nz and www.stats.govt.nz.

129

REFERENCES

Mitrovic, S.M., Bowling, L.C., Buckney, R.T., 2001. Vertical disentrainment of Anabaena circinalis in the turbid, freshwater Darling River, Australia: Quantifying potential benefits from buoyancy. J Plankton Res 23, 47–55. Moernaut, J., Daele, M.V., Heirman, K., Fontijn, K., Strasser, M., Pino, M., Urrutia, R., Batist, M.D., 2014. Lacustrine turbidites as a tool for quantitative earthquake reconstruction: New evidence for a variable rupture mode in south central Chile. J. Geophys. Res. Solid Earth 119, 1607– 1633. Moisander, P.H., Paerl, H.W., Zehr, J.P., 2008. Effects of inorganic nitrogen on taxa-specific cyanobacterial growth and nifH expression in a subtropical estuary. Limnology and Oceanography 53, 2519–2532. Monchamp, M.-E., Spaak, P., Domaizon, I., Dubois, N., Bouffard, D., Pomati, F., 2018. Homogenization of lake cyanobacterial communities over a century of climate change and eutrophication. Nat. Ecol. Evol. 2, 317–324. Monchamp, M.-E., Spaak, P., Pomati, F., 2019. High dispersal levels and lake warming are emergent drivers of cyanobacterial community assembly in peri-Alpine lakes. Sci. Rep. 9. Monchamp, M.-E., Walser, J.-C., Pomati, F., Spaak, P., 2016. Sedimentary DNA reveals cyanobacterial community diversity over 200 years in two perialpine lakes. Appl. Environ. Microbiol. 82, 6472–6482. Morris, G.L. and Fan, J., 1998. Reservoir Sediment Handbook McGraw-Hill Book Co., New York. Mulholland, M.R., Bernhardt, P.W., Heil, C.A., Bronk, D.A., O’Neil, J.M., 2006. Nitrogen fixation and release of fixed nitrogen by Trichodesmium spp. in the Gulf of Mexico. Limnology and Oceanography 51, 1762–1776. Müller-Navarra, D.C., Brett, M.T., Park, S., Chandra, S., Ballantyne, A.P., Zorita, E., Goldman, C.R., 2004. Unsaturated fatty acid content in seston and tropho-dynamic coupling in lakes. Nature 427, 69–72. Mulligan, M., van Soesbergen, A., Sáenz, L., 2020. GOODD, a global dataset of more than 38,000 georeferenced dams. Sci. Data 7, 31. Mullineaux, C.W., 2001. How do cyanobacteria sense and respond to light? Molecular Microbiology 41, 965–971. Murakami, M., Okita, Y., Matsuda, H., Okino, T., Yamaguchi, K., 1994. Aeruginosin 298-A, a thrombin and trypsin inhibitor from the blue-green alga Microcystis aeruginosa (NIES-298). Tetrahedron Letters 35, 3129–3132. Murakami, M., Suzuki, S., Itou, Y., Kodani, S., Ishida, K., 2000. New anabaenopeptins, potent carboxypeptidase-A inhibitors from the cyanobacterium Aphanizomenon flos-aquae. J. Nat. Prod. 63, 1280–1282. Murray, S.A., Mihali, T.K., Neilan, B.A., 2011. Extraordinary conservation, gene loss, and positive selection in the evolution of an ancient neurotoxin. Mol Biol Evol 28, 1173–1182. Neilan, B.A., Jacobs, D., Therese, D.D., Blackall, L.L., Hawkins, P.R., Cox, P.T., Goodman, A.E., 1997. rRNA sequences and evolutionary relationships among toxic and nontoxic cyanobacteria of the genus Microcystis. Int. J. Syst. Evol. Microbiol. 47, 693–697. Neilan, B.A., Pearson, L.A., Moffitt, M.C., Mihali, K.T., Kaebernick, M., Kellmann, R., Pomati, F., 2008. The genetics and genomics of cyanobacterial toxicity, in: Hudnell, H.K. (Ed.), cyanobacterial harmful algal blooms: State of the science and research needs, advances in experimental medicine and biology. Springer New York, pp. 417–452. Neilan, B.A., Pearson, L.A., Muenchhoff, J., Moffitt, M.C., Dittmann, E., 2013. Environmental conditions that influence toxin biosynthesis in cyanobacteria. Environmental Microbiology 15, 1239–1253. Nguyen, H., Pham, T., Lobry de Bruyn, L., 2017. Impact of hydroelectric dam development and resettlement on the natural and social capital of rural livelihoods in Bo Hon village in central Vietnam. Sustainability 9, 1422. Nicolella, C., Zolezzi, M., Rabino, M., Furfaro, M., Rovatti, M., 2005. Development of particle-based biofilms for degradation of xenobiotic organic compounds. Water Res. 39, 2495–2504. Nilsson, C., Reidy, C.A., Dynesius, M., Revenga, C., 2005. Fragmentation and flow regulation of the world’s large river systems. Science 308, 405–408.

130

REFERENCES

Nilsson, R.H., Larsson, K.-H., Taylor, A.F.S., Bengtsson-Palme, J., Jeppesen, T.S., Schigel, D., Kennedy, P., Picard, K., Glöckner, F.O., Tedersoo, L., Saar, I., Kõljalg, U., Abarenkov, K., 2019. The UNITE database for molecular identification of fungi: handling dark taxa and parallel taxonomic classifications. Nucleic Acids Res. 47, D259–D264. Nishiwaki-Matsushima, R., Ohta, T., Nishiwaki, S., Suganuma, M., Kohyama, K., Ishikawa, T., Carmichael, W.W., Fujiki, H., 1992. Liver tumor promotion by the cyanobacterial cyclic peptide toxin microcystin-LR. J Cancer Res Clin Oncol 118, 420–424. Nutman, A.P., Bennett, V.C., Friend, C.R.L., Van Kranendonk, M.J., Chivas, A.R., 2016. Rapid emergence of life shown by discovery of 3,700-million-year-old microbial structures. Nature 537, 535–538. O’Donnell, J.L., Kelly, R.P., Shelton, A.O., Samhouri, J.F., Lowell, N.C., Williams, G.D., 2017. Spatial distribution of environmental DNA in a nearshore marine habitat. PeerJ 5, e3044. O’Leary, N.A., Wright, M.W., Brister, J.R., Ciufo, S., Haddad, D., McVeigh, R., Rajput, B., Robbertse, B., Smith-White, B., Ako-Adjei, D., Astashyn, A., Badretdin, A., Bao, Y., Blinkova, O., Brover, V., Chetvernin, V., Choi, J., Cox, E., Ermolaeva, O., Farrell, C.M., Goldfarb, T., Gupta, T., Haft, D., Hatcher, E., Hlavina, W., Joardar, V.S., Kodali, V.K., Li, W., Maglott, D., Masterson, P., McGarvey, K.M., Murphy, M.R., O’Neill, K., Pujar, S., Rangwala, S.H., Rausch, D., Riddick, L.D., Schoch, C., Shkeda, A., Storz, S.S., Sun, H., Thibaud-Nissen, F., Tolstoy, I., Tully, R.E., Vatsan, A.R., Wallin, C., Webb, D., Wu, W., Landrum, M.J., Kimchi, A., Tatusova, T., DiCuccio, M., Kitts, P., Murphy, T.D., Pruitt, K.D., 2016. Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res. 44, D733– D745. O’Neil, J.M., Davis, T.W., Burford, M.A., Gobler, C.J., 2012. The rise of harmful cyanobacteria blooms: The potential roles of eutrophication and climate change. Harmful Algae, 14, 313– 334. Oh, H.-M., Lee, S.J., Jang, M.-H., Yoon, B.-D., 2000. Microcystin production by Microcystis aeruginosa in a phosphorus-limited chemostat. Appl Environ Microbiol 66, 176–179. Okino, T., Matsuda, H., Murakami, M., Yamaguchi, K., 1993. Microginin, an angiotensin-converting enzyme inhibitor from the blue-green alga Microcystis aeruginosa. Tetrahedron Letters, The International Journal for the Rapid Publication of Preliminary Communications in Organic Chemistry 34, 501–504. Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P.R., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H., Szoecs, E., Wagner, H., 2019. vegan: Community Ecology Package. Oliveros, J.C., 2015. Venny. An interactive tool for comparing lists with Venn’s diagrams. Olsson-Francis, K., de la Torre, R., Towner, M.C., Cockell, C.S., 2009. Survival of akinetes (resting- state cells of cyanobacteria) in low Earth orbit and simulated extraterrestrial conditions. Orig Life Evol Biosph 39, 565. Orr, P.T., Jones, G.J., 1998. Relationship between microcystin production and cell division rates in nitrogen-limited Microcystis aeruginosa cultures. Limnology and Oceanography 43, 1604– 1614. Orr, P.T., Willis, A., Burford, M.A., 2018. Application of first order rate kinetics to explain changes in bloom toxicity—the importance of understanding cell toxin quotas. J. Ocean. Limnol. 36, 1063– 1074. Ouellette, A.J., Wilhelm, S.W., 2003. Toxic cyanobacteria: the evolving molecular toolbox. Frontiers in Ecology and the Environment 1, 359–366. Padisák, J., Soróczki-Pintér, É., Rezner, Z., 2003. Sinking properties of some phytoplankton shapes and the relation of form resistance to morphological diversity of plankton – an experimental study. Hydrobiologia 500, 243–257. Paerl, H., 2017. The cyanobacterial nitrogen fixation paradox in natural waters. F1000Res 6. Paerl, H.W., 2014. Mitigating harmful cyanobacterial blooms in a human- and climatically-impacted world. Life 4, 988–1012. Paerl, H.W., 2018. Mitigating toxic planktonic cyanobacterial blooms in aquatic ecosystems facing increasing anthropogenic and climatic pressures. Toxins (Basel) 10.

131

REFERENCES

Paerl, H.W., Fulton, R.S., Moisander, P.H., Dyble, J., 2001. Harmful freshwater algal blooms, with an emphasis on cyanobacteria. ScientificWorldJournal 1, 76–113. Paerl, H.W., Gardner, W.S., Havens, K.E., Joyner, A.R., McCarthy, M.J., Newell, S.E., Qin, B., Scott, J.T., 2016. Mitigating cyanobacterial harmful algal blooms in aquatic ecosystems impacted by climate change and anthropogenic nutrients. Harmful Algae, 54, 213–222. Paerl, H.W., Hall, N.S., Calandrino, E.S., 2011. Controlling harmful cyanobacterial blooms in a world experiencing anthropogenic and climatic-induced change. Science of The Total Environment 409, 1739–1745. Paerl, H.W., Huisman, J., 2008. Blooms like It hot. Science 320, 57–58. Paerl, H.W., Huisman, J., 2009a. Climate change: a catalyst for global expansion of harmful cyanobacterial blooms. Environ Microbiol Rep 1, 27–37. Paerl, H.W., Otten, T.G., 2013. Harmful cyanobacterial blooms: Causes, consequences, and controls. Microb Ecol 65, 995–1010. Paerl, H.W., Paul, V.J., 2012. Climate change: Links to global expansion of harmful cyanobacteria. Water Research, Cyanobacteria: Impacts of climate change on occurrence, toxicity and water quality management 46, 1349–1363. Paerl, H.W., Scott, J.T., McCarthy, M.J., Newell, S.E., Gardner, W.S., Havens, K.E., Hoffman, D.K., Wilhelm, S.W., Wurtsbaugh, W.A., 2016. It takes two to tango: When and where dual nutrient (N & P) reductions are needed to protect lakes and downstream ecosystems. Environ. Sci. Technol. 50, 10805–10813. Pal, S., Gregory-Eaves, I., Pick, F.R., 2015. Temporal trends in cyanobacteria revealed through DNA and pigment analyses of temperate lake sediment cores. J Paleolimnol 54, 87–101. Palinska, K.A., Surosz, W., 2014. Taxonomy of cyanobacteria: a contribution to consensus approach. Hydrobiologia 740, 1–11. Passarelli, C., Olivier, F., Paterson, D.M., Meziane, T., Hubas, C., 2014. Organisms as cooperative ecosystem engineers in intertidal flats. J. Sea Res., 92, 92–101. Paterson, D.M., Hope, J.A., Kenworthy, J., Biles, C.L., Gerbersdorf, S.U., 2018. Form, function and physics: the ecology of biogenic stabilisation. J. Soils Sediments 18, 3044–3054. Pearson, L.A., Hisbergues, M., Börner, T., Dittmann, E., Neilan, B.A., 2004. Inactivation of an ABC transporter gene, mcyH, results in loss of microcystin production in the cyanobacterium Microcystis aeruginosa PCC 7806. Appl Environ Microbiol 70, 6370–6378. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., 2012. Scikit-learn: Machine learning in Python. Mach. Learn. PYTHON 6. Pettersson, K., Herlitz, E., Istvánovics, V., 1993. The role of Gloeotrichia echinulata in the transfer of phosphorus from sediments to water in Lake Erken. Hydrobiologia 253, 123–129. Pflugmacher, S., 2004. Promotion of oxidative stress in the aquatic macrophyte Ceratophyllum demersum during biotransformation of the cyanobacterial toxin microcystin-LR. Aquatic Toxicology 70, 169–178. Pilon, S., Zastepa, A., Taranu, Z.E., Gregory-Eaves, I., Racine, M., Blais, J.M., Poulain, A.J., Pick, F.R., 2019. Contrasting histories of microcystin-producing cyanobacteria in two temperate lakes as inferred from quantitative sediment DNA analyses. Lake Reserv. Manag. 35, 102– 117. Pinheiro, C., Azevedo, J., Campos, A., Loureiro, S., Vasconcelos, V., 2013. Absence of negative allelopathic effects of cylindrospermopsin and microcystin-LR on selected marine and freshwater phytoplankton species. Hydrobiologia 705, 27–42. Planavsky, N.J., Asael, D., Hofmann, A., Reinhard, C.T., Lalonde, S.V., Knudsen, A., Wang, X., Ossa Ossa, F., Pecoits, E., Smith, A.J.B., Beukes, N.J., Bekker, A., Johnson, T.M., Konhauser, K.O., Lyons, T.W., Rouxel, O.J., 2014. Evidence for oxygenic photosynthesis half a billion years before the Great Oxidation Event. Nature Geoscience 7, 283–286. Pohlert, T., Hillebrand, G., and Breitung, V., 2011. Trends of persistent organic pollutants in the suspended matter of the river rhine. Hydrological Processes, 25, 3803–3817. Posch, T., Köster, O., Salcher, M.M., Pernthaler, J., 2012. Harmful filamentous cyanobacteria favoured by reduced water turnover with lake warming Nat. Clim. Change 2, 809–813.

132

REFERENCES

Pouria, S., de Andrade, A., Barbosa, J., Cavalcanti, R., Barreto, V., Ward, C., Preiser, W., Poon, G.K., Neild, G., Codd, G., 1998. Fatal microcystin intoxication in haemodialysis unit in Caruaru, Brazil. The Lancet 352, 21–26. Preston, T., Stewart, W.D.P., Reynolds, C.S., 1980. Bloom-forming cyanobacterium Microcystis aeruginosa overwinters on sediment surface. Nature 288, 365–367. Puddick, J., Prinsep, M.R., Wood, S.A., Kaufononga, S.A.F., Cary, S.C., Hamilton, D.P., 2014. High levels of structural diversity observed in microcystins from Microcystis CAWBG11 and characterization of six new microcystin congeners. Mar. Drugs 12, 5372–5395. Puddick, J., Wood, S.A., Hawes, I., Hamilton, D.P., 2016. Fine-scale cryogenic sampling of planktonic microbial communities: Application to toxic cyanobacterial blooms. Limnol. Oceanogr. Methods 14, 600–609. Punčochářová, M., 1990. Culture collection of algae at the department of botany, Faculty of Natural Science, Charles University in Prague (CAUP). Archiv für Protistenkunde 138, 143–158. Puschner, B., Galey, F.D., Johnson, B., Dickie, C.W., Vondy, M., Francis, T., Holstege, D.M., 1998. Blue-green algae toxicosis in cattle. J. Am. Vet. Med. Assoc. 213, 1605–1607, 1571. Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J., Glöckner, F.O., 2013. The SILVA ribosomal RNA gene database project: improved data processing and web- based tools. Nucleic Acids Res. 41, D590–D596. Quiblier, C., Susanna, W., Isidora, E.-S., Mark, H., Aurélie, V., Jean-François, H., 2013. A review of current knowledge on toxic benthic freshwater cyanobacteria – Ecology, toxin production and risk management. Water Res. 47, 5464–5479. R Core Team, 2020. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Rai, A.N., Söderbäck, E., Bergman, B., 2000. Cyanobacterium-plant symbioses. New Phytologist 147, 449–481. Rajendhran, J., Gunasekaran, P., 2011. Microbial phylogeny and diversity: small subunit ribosomal RNA sequence analysis and beyond. Microbiol. Res. 166, 99–110. Rantala, A., Fewer, D.P., Hisbergues, M., Rouhiainen, L., Vaitomaa, J., Börner, T., Sivonen, K., 2004. Phylogenetic evidence for the early evolution of microcystin synthesis. PNAS 101, 568– 573. Rantala, A., Rajaniemi-Wacklin, P., Lyra, C., Lepistö, L., Rintala, J., Mankiewicz-Boczek, J., Sivonen, K., 2006. Detection of microcystin-producing cyanobacteria in finnish lakes with genus-specific microcystin synthetase gene E (mcyE) PCR and associations with environmental factors. Appl. Environ. Microbiol. 72, 6101–6110. Rapala, J., Sivonen, K., Lyra, C., Niemelä, S.I., 1997. Variation of microcystins, cyanobacterial hepatotoxins, in Anabaena spp. as a function of growth stimuli. Appl Environ Microbiol 63, 2206–2212. Rasconi, S., Niquil, N., Sime‐Ngando, T., 2012. Phytoplankton chytridiomycosis: community structure and infectivity of fungal parasites in aquatic ecosystems. Environ. Microbiol. 14, 2151–2170. Rastogi, R.P., Madamwar, D., Incharoensakdi, A., 2015. Bloom dynamics of cyanobacteria and their toxins: Environmental health impacts and mitigation strategies. Front. Microbiol. 6. Raymond, P.A., Hartmann, J., Lauerwald, R., Sobek, S., McDonald, C., Hoover, M., Butman, D., Striegl, R., Mayorga, E., Humborg, C., Kortelainen, P., Dürr, H., Meybeck, M., Ciais, P., Guth, P., 2013. Global carbon dioxide emissions from inland waters. Nature, 503 (7476), 355-9. Revelle, W. R., 2017. psych: Procedures for Personality and Psychological Research. Software Reynolds, C.S., Jaworski, G.H.M., Cmiech, H.A., Leedale, G.F., Lund, J.W.G., 1981. On the annual cycle of the blue-green alga Microcystis aeruginosa Kütz. Emend. Elenkin. Philos. Trans. R. Soc. Lond. B Biol. Sci. 293, 419–477. Richards, J.G., Farrell, A.P., Brauner, C.J., 2009. Fish Physiology: Hypoxia. Academic Press. Righetti, M., Lucarelli, C., 2010. Resuspension phenomena of benthic sediments: The role of cohesion and biological adhesion. River Res. Appl. 26, 404–413.

133

REFERENCES

Rigosi, A., Carey, C.C., Ibelings, B.W., Brookes, J.D., 2014. The interaction between climate warming and eutrophication to promote cyanobacteria is dependent on trophic state and varies among taxa. Limnology and Oceanography 59, 99–114. Rinta-Kanto, J.M., Ouellette, A.J.A., Boyer, G.L., Twiss, M.R., Bridgeman, T.B., Wilhelm, S.W., 2005. Quantification of toxic Microcystis spp. during the 2003 and 2004 blooms in western Lake Erie using quantitative real-time PCR. Environ. Sci. Technol. 39, 4198–4205. Rognes, T., Flouri, T., Nichols, B., Quince, C., Mahé, F., 2016. VSEARCH: a versatile open source tool for metagenomics. PeerJ 4. Rohrlack, T., Christiansen, G., Kurmayer, R., 2013. Putative antiparasite defensive system involving ribosomal and nonribosomal oligopeptides in cyanobacteria of the genus Planktothrix. Appl. Environ. Microbiol. 79, 2642–2647. Rohrlack, T., Christoffersen, K., Dittmann, E., Nogueira, I., Vasconcelos, V., Börner, T., 2005. Ingestion of microcystins by Daphnia: Intestinal uptake and toxic effects. Limnology and Oceanography 50, 440–448. Rohrlack, T., Hyenstrand, P., 2007. Fate of intracellular microcystins in the cyanobacterium Microcystis aeruginosa (Chroococcales, Cyanophyceae). Phycologia 46, 277–283. Rohrlack, T., Skulberg, R., Skulberg, O.M., 2009. Distribution of oligopeptide chemotypes of the cyanobacterium Planktothrix and their persistence in selected lakes in Fennoscandia. J. Phycol. 45, 1259–1265. Romanowska‐Duda, Z., Tarczyńska, M., 2002. The influence of microcystin-LR and hepatotoxic cyanobacterial extract on the water plant Spirodela oligorrhiza. Environmental Toxicology 17, 434–440. Rouhiainen, L., Jokela, J., Fewer, D.P., Urmann, M., Sivonen, K., 2010. Two alternative starter modules for the non-ribosomal biosynthesis of specific anabaenopeptin variants in Anabaena (cyanobacteria). Chemistry & Biology 17, 265–273. Rouhiainen, L., Vakkilainen, T., Siemer, B.L., Buikema, W., Haselkorn, R., Sivonen, K., 2004. Genes Coding for hepatotoxic heptapeptides (microcystins) in the cyanobacterium Anabaena Strain 90. Appl Environ Microbiol 70, 686–692. Rueckert, A., Cary, S.C., 2009. Use of an armored RNA standard to measure microcystin synthetase E gene expression in toxic Microcystis sp. by reverse-transcription QPCR. Limnol. Oceanogr. Methods 7, 509–520. Rühland, K., Priesnitz, A., Smol, J.P., 2003. Paleolimnological evidence from diatoms for recent environmental changes in 50 lakes across Canadian arctic treeline. Arct. Antarct. Alp. Res. 35, 110–123. Runnegar, M., Berndt, N., Kaplowitz, N., 1995. Microcystin uptake and inhibition of protein phosphatases: Effects of chemoprotectants and self-inhibition in relation to known hepatic transporters. Toxicol. Appl. Pharmacol. 134, 264–272. Sabart, M., Pobel, D., Briand, E., Combourieu, B., Salençon, M.J., Humbert, J.F., Latour, D., 2010. Spatiotemporal variations in microcystin concentrations and in the proportions of microcystin- producing cells in several Microcystis aeruginosa populations. Appl. Environ. Microbiol. 76, 4750–4759. Salmaso, N., 2000. Factors affecting the seasonality and distribution of cyanobacteria and chlorophytes: a case study from the large lakes south of the Alps, with special reference to Lake Garda. Hydrobiologia 438, 43–63. Salmaso, N., 2010. Long-term phytoplankton community changes in a deep subalpine lake: responses to nutrient availability and climatic fluctuations. Freshw. Biol. 55, 825–846. Salmaso, N., Buzzi, F., Garibaldi, L., Morabito, G., Simona, M., 2012. Effects of nutrient availability and temperature on phytoplankton development: A case study from large lakes south of the Alps. Aquat. Sci. 74, 555–570. Salmaso, N., Mosello, R., 2010. Limnological research in the deep southern subalpine lakes: Synthesis, directions and perspectives. Advances in Oceanography and Limnology 1, 29–66. Sano, T., Usui, T., Ueda, K., Osada, H., Kaya, K., 2001. Isolation of new protein phosphatase inhibitors from two cyanobacteria species, Planktothrix spp. J. Nat. Prod. 64, 1052–1055.

134

REFERENCES

Sas, H., 1990. Lake restoration by reduction of nutrient loading: Expectations, experiences, extrapolations. SIL Proceedings, 1922-2010 24, 247–251. Savichtcheva, O., Debroas, D., Kurmayer, R., Villar, C., Jenny, J.P., Arnaud, F., Perga, M.E., Domaizon, I., 2011. Quantitative PCR enumeration of total/toxic Planktothrix rubescens and total cyanobacteria in preserved DNA isolated from lake sediments. Appl. Environ. Microbiol. 77, 8744–8753. Scavia, D., David Allan, J., Arend, K.K., Bartell, S., Beletsky, D., Bosch, N.S., Brandt, S.B., Briland, R.D., Daloğlu, I., DePinto, J.V., Dolan, D.M., Evans, M.A., Farmer, T.M., Goto, D., Han, H., Höök, T.O., Knight, R., Ludsin, S.A., Mason, D., Michalak, A.M., Peter Richards, R., Roberts, J.J., Rucinski, D.K., Rutherford, E., Schwab, D.J., Sesterhenn, T.M., Zhang, H., Zhou, Y., 2014. Assessing and addressing the re-eutrophication of Lake Erie: Central basin hypoxia. Journal of Great Lakes Research 40, 226–246. Schatz, D., Keren, Y., Vardi, A., Sukenik, A., Carmeli, S., Börner, T., Dittmann, E., Kaplan, A., 2007. Towards clarification of the biological role of microcystins, a family of cyanobacterial toxins. Environmental Microbiology 9, 965–970. Scheffer, M., Rinaldi, S., Gragnani, A., Mur, L.R., van Nes, E.H., 1997. On the dominance of filamentous cyanobacteria in shallow, turbid lakes. Ecology 78, 272–282. Schillereff, D.N., Chiverrell, R.C., Macdonald, N., Hooke, J.M., 2014. Flood stratigraphies in lake sediments: A review. Earth-Sci. Rev. 135, 17–37. Schindler, D.W., 1977. Evolution of phosphorus limitation in lakes. Science 195, 260–262. Schindler, D.W., 2012. The dilemma of controlling cultural eutrophication of lakes. Proc Biol Sci 279, 4322–4333. Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton, M.P., Parker, B.R., Paterson, M.J., Beaty, K.G., Lyng, M., Kasian, S.E.M., 2008. Eutrophication of lakes cannot be controlled by reducing nitrogen input: Results of a 37-year whole-ecosystem experiment. PNAS 105, 11254–11258. Schirrmeister, B.E., Gugger, M., Donoghue, P.C.J., 2015. Cyanobacteria and the Great Oxidation Event: evidence from genes and fossils. Palaeontology 58, 769–785. Schleiss, A.J., Franca, M.J., Juez, C., Cesare, G.D., 2016. Reservoir sedimentation. Journal of Hydraulic Research 54, 595–614. Schmidt, J.R., Wilhelm, S.W., Boyer, G.L., 2014. The fate of microcystins in the environment and challenges for monitoring. Toxins 6, 3354–3387. Schopf, J.W., 2002. The fossil record: Tracing the roots of the cyanobacterial lineage, in: Whitton, B.A., Potts, M. (Eds.), The ecology of cyanobacteria: Their diversity in time and space. Springer Netherlands, Dordrecht, pp. 13–35. Schrenk-Bergt, C., Krause, D., Lewandowski, J., Steinberg, C.E.W., 2004. Eutrophication problems and their potential solutions in the artificial shallow lake Altmühlsee (Germany). Studia Quarternaria 21, 73-86. Schultz, P., Urban, N.R., 2008. Effects of bacterial dynamics on organic matter decomposition and nutrient release from sediments: A modeling study. Ecol. Model. 210, 1–14. Seitzinger, S.P., Mayorga, E., Bouwman, A.F., Kroeze, C., Beusen, A.H.W., Billen, G., Drecht, G.V., Dumont, E., Fekete, B.M., Garnier, J., Harrison, J.A., 2010. Global river nutrient export: A scenario analysis of past and future trends. Global Biogeochemical Cycles 24. Sevilla, E., Martin‐Luna, B., Vela, L., Bes, M.T., Fillat, M.F., Peleato, M.L., 2008. Iron availability affects mcyD expression and microcystin-LR synthesis in Microcystis aeruginosa PCC7806. Environmental Microbiology 10, 2476–2483. Shams, S., Capelli, C., Cerasino, L., Ballot, A., Dietrich, D.R., Sivonen, K., Salmaso, N., 2015. Anatoxin-a producing Tychonema (Cyanobacteria) in european waterbodies. Water Res. 69, 68–79. Shang, Q., Fang, H., Zhao, H., He, G., Cui, Z., 2014. Biofilm effects on size gradation, drag coefficient and settling velocity of sediment particles. Int. J. Sediment Res. 29, 471–480. Shannon, M.A., Bohn, P.W., Elimelech, M., Georgiadis, J.G., Mariñas, B.J., Mayes, A.M., 2008. Science and technology for water purification in the coming decades. Nature 452, 301–310.

135

REFERENCES

Shatwell, T., Köhler, J., 2019. Decreased nitrogen loading controls summer cyanobacterial blooms without promoting nitrogen-fixing taxa: Long-term response of a shallow lake. Limnology and Oceanography 64, S166–S178. Shen, H.W., 1999. Flushing sediment through reservoirs. Journal of Hydraulic Research 37, 743– 757. Shi, Y., 2009. Serine/Threonine phosphatases: Mechanism through structure. Cell 139, 468–484. Siegmund-Schultze, M., do Carmo Sobral, M., Alcoforado de Moraes, M.M.G., Almeida-Cortez, J.S., Azevedo, J.R.G., Candeias, A.L., Cierjacks, A., Gomes, E.T.A., Gunkel, G., Hartje, V., Hattermann, F.F., Kaupenjohann, M., Koch, H., Köppel, J., 2018. The legacy of large dams and their effects on the water-land nexus. Reg Environ Change 18, 1883-1888. Sikorski, J., 2019. A new method for constructing Pb-210 chronology of young peat profiles sampled with low frequency. Geochronometria 46, 1–14. Singh, S., Kate, B.N., Banerjee, U.C., 2005. Bioactive compounds from cyanobacteria and microalgae: An Overview. Critical Reviews in Biotechnology 25, 73–95. Sinha, R., Pearson, L.A., Davis, T.W., Burford, M.A., Orr, P.T., Neilan, B.A., 2012. Increased incidence of Cylindrospermopsis raciborskii in temperate zones – Is climate change responsible? Water Research, Cyanobacteria: Impacts of climate change on occurrence, toxicity and water quality management 46, 1408–1419. Sivonen, K., 2009. Cyanobacterial Toxins. In: Schaechter, M. (Ed.), Encyclopedia of Microbiology. pp. 290–307. Sivonen, K., Jones, G.J. (Eds.), 1999. Cyanobacterial toxins, in: Toxic cyanobacteria in water: A guide to their public health consequences, monitoring, and management. E & FN Spon, London ; New York. Sivonen, K., Namikoshi, M., Evans, W.R., Carmichael, W.W., Sun, F., Rouhiainen, L., Luukkainen, R., Rinehart, K.L., 1992. Isolation and characterization of a variety of microcystins from seven strains of the cyanobacterial genus Anabaena. Appl Environ Microbiol 58, 2495–2500. Smith, R.B., Bass, B., Sawyer, D., Depew, D., Watson, S.B., 2019. Estimating the economic costs of algal blooms in the Canadian Lake Erie Basin. Harmful Algae 87, 101624. Smith, V.H., 1983. Low nitrogen to phosphorus ratios favor dominance by blue-green algae in lake phytoplankton. Science 221, 669–671. Soares, R.M., Yuan, M., Servaites, J.C., Delgado, A., Magalhães, V.F., Hilborn, E.D., Carmichael, W.W., Azevedo, S.M.F.O., 2006. Sublethal exposure from microcystins to renal insufficiency patients in Rio de Janeiro, Brazil. Environmental Toxicology 21, 95–103. Sobek, S., DelSontro, T., Wongfun, N., Wehrli, B., 2012. Extreme organic carbon burial fuels intense methane bubbling in a temperate reservoir. Geophysical Research Letters, 39. Sommer, 2012. Beyond the Plankton Ecology Group (PEG) model: Mechanisms driving plankton succession. Sommer, U., Adrian, R., De Senerpont Domis, L., Elser, J.J., Gaedke, U., Ibelings, B., Jeppesen, E., Lürling, M., Molinero, J.C., Mooij, W.M., van Donk, E., Winder, M., 2012. Beyond the Plankton Ecology Group (PEG) Model: Mechanisms driving plankton succession. Annual Review of Ecology, Evolution, and Systematics 43, 429–448. Sønstebø, J.H., Rohrlack, T., 2011. Possible Implications of chytrid parasitism for population subdivision in freshwater cyanobacteria of the genus Planktothrix. Appl. Environ. Microbiol. 77, 1344–1351. Sotton, B., Domaizon, I., Anneville, O., Cattanéo, F., Guillard, J., 2014. Nodularin and cylindrospermopsin: a review of their effects on fish. Reviews in Fish Biology and Fisheries. Spoof, L., Błaszczyk, A., Meriluoto, J., Cegłowska, M., Mazur-Marzec, H., 2015. Structures and activity of new anabaenopeptins Produced by Baltic Sea cyanobacteria. Mar. Drugs 14. Srinivasan, R., Sorial, G.A., 2011. Treatment of taste and odor causing compounds 2-methyl isoborneol and geosmin in drinking water: A critical review. Journal of Environmental Sciences 23, 1–13. Staal, M., te LintelHekkert, S., Herman, P., J Stal, L., 2002. Comparison of models describing light dependence of N2 fixation in heterocystous cyanobacteria. Applied and environmental microbiology 68, 4679–83.

136

REFERENCES

Ståhl-Delbanco, A., Hansson, L.-A., Gyllström, M., 2003. Recruitment of resting stages may induce blooms of Microcystis at low N:P ratios. J Plankton Res 25, 1099–1106. Stal, L.J., 2015. Cyanobacteria, diversity and evolution of, in: Gargaud, M., Irvine, W.M., Amils, R., Cleaves, H.J. (Jim), Pinti, D.L., Quintanilla, J.C., Rouan, D., Spohn, T., Tirard, S., Viso, M. (Eds.), Encyclopedia of Astrobiology. Springer, Berlin, Heidelberg, pp. 595–599. Steffen, M.M., Belisle, B.S., Watson, S.B., Boyer, G.L., Wilhelm, S.W., 2014. Status, causes and controls of cyanobacterial blooms in Lake Erie. Journal of Great Lakes Research 40, 215–225. Steffen, W., Crutzen, P.J., McNeill, J.R., 2007. The Anthropocene: Are humans now overwhelming the great forces of nature. ambio 36, 614–621. Steffensen, D.A., 2008. Economic cost of cyanobacterial blooms, in: Hudnell, H.K. (Ed.), Cyanobacterial harmful algal blooms: State of the science and research needs, advances in experimental medicine and biology. Springer, New York, NY, pp. 855–856. Steiner, K., Wood, S.A., Puddick, J., Hawes, I., Dietrich, D.R., Hamilton, D.P., 2017. A comparison of bacterial community structure, activity and microcystins associated with formation and breakdown of a cyanobacterial scum. Aquat. Microb. Ecol. 80, 243–256. Sterner, R.W., Reinl, K.L., Lafrancois, B.M., Brovold, S., Miller, T.R., 2020. A first assessment of cyanobacterial blooms in oligotrophic Lake Superior. Limnology and Oceanography n/a. Stewart, I., Seawright, A.A., Shaw, G.R., 2008. Cyanobacterial poisoning in livestock, wild mammals and birds – an overview, in: Hudnell, H.K. (Ed.), Cyanobacterial harmful algal blooms: State of the science and research needs, advances in experimental medicine and biology. Springer, New York, NY, pp. 613–637. Stoermer, E.F., Smol, J.P., 2001. The Diatoms: Applications for the environmental and earth sciences. Cambridge University Press. Stoodley, P., Sauer, K., Davies, D.G., Costerton, J.W., 2002. Biofilms as complex differentiated communities. Annu. Rev. Microbiol. 56, 187–209. Stüken, A., Rücker, J., Endrulat, T., Preussel, K., Hemm, M., Nixdorf, B., Karsten, U., Wiedner, C., 2006. Distribution of three alien cyanobacterial species (Nostocales) in northeast Germany: Cylindrospermopsis raciborskii, Anabaena bergii and Aphanizomenon aphanizomenoides. Phycologia 45, 696–703. Sukenik, A., Hadas, O., Kaplan, A., Quesada, A., 2012. Invasion of Nostocales (cyanobacteria) to subtropical and temperate freshwater lakes – Physiological, regional, and global driving forces. Front Microbiol 3. Sukenik, A., Quesada, A., Salmaso, N., 2015. Global expansion of toxic and non-toxic cyanobacteria: effect on ecosystem functioning. Biodivers Conserv 24, 889–908. Sumi, T. and Kantoush, S.A., 2011. Sediment management strategies for sustainable reservoir. Eds. Schleiss, A.J. and Boes R.M. CRC Press. ISBN 9780415682671. Syvitski, J.P.M., Vörösmarty, C.J., Kettner, A.J., Green, P., 2005. Impact of humans on the flux of terrestrial sediment to the global coastal ocean. Science 308, 376–380. Takeya, K., Kuwata, A., Yoshida, M., Miyazaki, T., 2004. Effect of dilution rate on competitive interactions between the cyanobacterium Microcystis novacekii and the green alga Scenedesmus quadricauda in mixed chemostat cultures. J Plankton Res 26, 29–35. Taranu, Z.E., Gregory‐Eaves, I., Leavitt, P.R., Bunting, L., Buchaca, T., Catalan, J., Domaizon, I., Guilizzoni, P., Lami, A., McGowan, S., Moorhouse, H., Morabito, G., Pick, F.R., Stevenson, M.A., Thompson, P.L., Vinebrooke, R.D., 2015. Acceleration of cyanobacterial dominance in north temperate-subarctic lakes during the Anthropocene. Ecology Letters 18, 375–384. Thiel, T., 2005. Nitrogen fixation in heterocyst-forming cyanobacteria. In: Klipp, W., Masepohl, B., Gallon, J.R., Newton, W.E. (Eds.), Genetics and regulation of nitrogen fixation in free-living bacteria, nitrogen fixation: Origins, applications, and research progress. Springer Netherlands, Dordrecht, pp. 73–110. Thingstad, T., Lignell, R., 1997. Theoretical models for the control of bacterial growth rate, abundance, diversity and carbon demand. Aquat. Microb. Ecol. 13, 19–27. Tillett, D., Dittmann, E., Erhard, M., von Döhren, H., Börner, T., Neilan, B.A., 2000. Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide–polyketide synthetase system. Chem. Biol. 7, 753–764.

137

REFERENCES

Tilman, D., 1982. Resource competition and community structure. Princeton University Press. Tonk, L., Welker, M., Huisman, J., Visser, P.M., 2009. Production of cyanopeptolins, anabaenopeptins, and microcystins by the harmful cyanobacteria Anabaena 90 and Microcystis PCC 7806. Harmful Algae 8, 219–224. Tsoularis, A., Wallace, J., 2002. Analysis of logistic growth models. Math Biosci 179, 21–55. Turner, A.D., Dhanji-Rapkova, M., O’Neill, A., Coates, L., Lewis, A., Lewis, K., 2018. Analysis of microcystins in cyanobacterial blooms from freshwater bodies in England. Toxins (Basel) 10. UBA, 2015. Empfehlung zum Schutz von Badenden vor Cyanobakterien-Toxinen. Bundesgesundheitsbl. 58, 908–920. UNITE Community, 2019. UNITE QIIME release for Fungi. Version 18.11.2018. UNITE Community. Utkilen, H., Gjølme, N., 1995. Iron-stimulated toxin production in Microcystis aeruginosa. Appl. Environ. Microbiol. 61, 797–800. Vaitomaa, J., Rantala, A., Halinen, K., Rouhiainen, L., Tallberg, P., Mokelke, L., Sivonen, K., 2003. Quantitative real-time PCR for determination of microcystin synthetase E copy numbers for Microcystis and Anabaena in lakes. Appl. Environ. Microbiol. 69, 7289–7297. Vandamme, P., Pot, B., Gillis, M., de Vos, P., Kersters, K., Swings, J., 1996. Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol. Rev. 60, 407–438. Vander Woude, A., Ruberg, S., Johengen, T., Miller, R., Stuart, D., 2019. Spatial and temporal scales of variability of cyanobacteria harmful algal blooms from NOAA GLERL airborne hyperspectral imagery. J. Gt. Lakes Res. 45, 536–546. Varadharajan, C., Hemond, H.F., 2012. Time-series analysis of high-resolution ebullition fluxes from a stratified, freshwater lake. Journal of Geophysical Research: Biogeosciences, 117. Vargas, C. de, Audic, S., Henry, N., Decelle, J., Mahé, F., Logares, R., Lara, E., Berney, C., Bescot, N.L., Probert, I., Carmichael, M., Poulain, J., Romac, S., Colin, S., Aury, J.-M., Bittner, L., Chaffron, S., Dunthorn, M., Engelen, S., Flegontova, O., Guidi, L., Horák, A., Jaillon, O., Lima- Mendez, G., Lukeš, J., Malviya, S., Morard, R., Mulot, M., Scalco, E., Siano, R., Vincent, F., Zingone, A., Dimier, C., Picheral, M., Searson, S., Kandels-Lewis, S., Coordinators, T.O., Acinas, S.G., Bork, P., Bowler, C., Gorsky, G., Grimsley, N., Hingamp, P., Iudicone, D., Not, F., Ogata, H., Pesant, S., Raes, J., Sieracki, M.E., Speich, S., Stemmann, L., Sunagawa, S., Weissenbach, J., Wincker, P., Karsenti, E., 2015. Eukaryotic plankton diversity in the sunlit ocean. Science 348. https://doi.org/10.1126/science.1261605 Verspagen, J.M.H., Waal, D.B.V. de, Finke, J.F., Visser, P.M., Donk, E.V., Huisman, J., 2014. Rising CO2 levels will intensify phytoplankton blooms in eutrophic and hypertrophic lakes. PLOS ONE 9, e104325. Vignaga, E., Sloan, D.M., Luo, X., Haynes, H., Phoenix, V.R., Sloan, W.T., 2013. Erosion of biofilm- bound fluvial sediments Nat. Geosci. 6, 770–774. Visser, P.M., Verspagen, J.M.H., Sandrini, G., Stal, L.J., Matthijs, H.C.P., Paerl, H.W., Huisman, J., 2016. How rising CO2 and global warming may stimulate harmful cyanobacterial blooms. Harmful Algae, 54, 145-159. Visser, P.M., Ibelings, B.W., Mur, L.R., 1995. Autumnal sedimentation of Microcystis spp. as result of an increase in carbohydrate ballast at reduced temperature. J Plankton Res 17, 919–933. Visser, P.M., Ibelings, B.W., Mur, L.R., Walsby, A.E., 2005. The ecophysiology of the harmful cyanobacterium Microcystis, in: Huisman, J., Matthijs, H.C.P., Visser, P.M. (Eds.), Harmful cyanobacteria, aquatic ecology series. Springer Netherlands, Dordrecht, pp. 109–142. Visser, P.M., Passarge, J., Mur, L.R., 1997. Modelling vertical migration of the cyanobacterium Microcystis. Hydrobiologia 349, 99–109. Vitousek, P., Mooney, H., Lubchenco, J., Melillo, J., 1997. Human domination of earth’s ecosystems. Science 277. Vörösmarty, C.J., Meybeck, M., Fekete, B. and Sharma, K., 1997. The potential impact of neo- Castorization on sediment transport by the global network of rivers. Walling, D.E., and Probst, J.L. (Eds.). Human Impact on Erosion and Sedimentation. IAHS Publication 245, Wallingford, UK, 261–273

138

REFERENCES

Vörösmarty, C.J., Meybeck, M., Fekete, B., Sharma, K., Green, P. and Syvitski, J.P.M., 2003. Anthropogenic sediment retention: major global impact from registered river impoundments. Global and Planetary Change, Vol. 39. Amsterdam, Elsevier Science, 169–90. Votruba, L., Broža, V., 1989. Water management in reservoirs. Elsevier, Amsterdam, Netherlands. Vrede, T., Ballantyne, A., Mille‐Lindblom, C., Algesten, G., Gudasz, C., Lindahl, S., Brunberg, A.K., 2009. Effects of N : P loading ratios on phytoplankton community composition, primary production and N fixation in a eutrophic lake. Freshwater Biology 54, 331–344. Waal, D.B.V. de, Verspagen, J.M.H., Lürling, M., Donk, E.V., Visser, P.M., Huisman, J., 2009. The ecological stoichiometry of toxins produced by harmful cyanobacteria: an experimental test of the carbon-nutrient balance hypothesis. Ecology Letters 12, 1326–1335. Walker, S., Lund, J., Schumacher, D., Brakhage, P., McManus, M., Miller, J., Augustine, M., Carney, J., Holland, R., Hoagland, K., Holz, J., Barrow, T., Rundquist, D., Gitelson, A., 2008. Chapter 6: Nebraska Experience, in: Cyanobacterial harmful algal blooms: State of the science and research needs. Springer, New York, NY, p. 14. Walsby, A.E., 1994. Gas vesicles. Microbiol Rev 58, 94–144. Walsby, A.E., 2007. Cyanobacterial heterocysts: terminal pores proposed as sites of gas exchange. Trends in Microbiology 15, 340–349. Wang, H., Sivonen, K., Rouhiainen, L., Fewer, D.P., Lyra, C., Rantala-Ylinen, A., Vestola, J., Jokela, J., Rantasärkkä, K., Li, Z., Liu, B., 2012. Genome-derived insights into the biology of the hepatotoxic bloom-forming cyanobacterium Anabaena sp. strain 90. BMC Genomics 13, 613. Wang, H., Zhang, Z., Liang, D., du, H., Pang, Y., Hu, K., Wang, J., 2016. Separation of wind’s influence on harmful cyanobacterial blooms. Water Res. 98, 280–292. Ward, D.M., Castenholz, R.W., 2002. Cyanobacteria in Geothermal Habitats, in: Whitton, B.A., Potts, M. (Eds.), The ecology of cyanobacteria: Their diversity in time and space. Springer Netherlands, Dordrecht, pp. 37–59. Watanabe, M.F., Oishi, S., Harada, K.-I., Matsuura, K., Kawai, H., Suzuki, M., 1988. Toxins contained in Microcystis species of cyanobacteria (blue-green algae). Toxicon 26, 1017–1025. Weimer-Jehle, W., 2006. Cross-impact balances: A system-theoretical approach to cross-impact analysis. Technological Forecasting and Social Change 73, 334–361. Welker, M., Von Döhren, H., 2006. Cyanobacterial peptides — Nature’s own combinatorial biosynthesis. FEMS Microbiol Rev 30, 530–563. White, R., 2001. Evacuation of sediments from reservoirs. Thomas Telford Publishing. WHO, 2003. World Health Organization. Guidelines for safe recreational water environments. World Health Organization, Geneva. WHO, 2017. World Health Organization. Guidelines for drinking-water quality, 4 edn. World Health Organization, Geneva. Wiedner, C., Rücker, J., Brüggemann, R., Nixdorf, B., 2007. Climate change affects timing and size of populations of an invasive cyanobacterium in temperate regions. Oecologia 152, 473–484. Wiese, M., D’Agostino, P.M., Mihali, T.K., Moffitt, M.C., Neilan, B.A., 2010. Neurotoxic Alkaloids: Saxitoxin and Its Analogs. Mar Drugs 8, 2185–2211. Wilhelm, B., Nomade, J., Crouzet, C., Litty, C., Sabatier, P., Belle, S., Rolland, Y., Revel, M., Courboulex, F., Arnaud, F., Anselmetti, F.S., 2016. Quantified sensitivity of small lake sediments to record historic earthquakes: Implications for paleoseismology. J. Geophys. Res. Earth Surf. 121, 2–16. Willis, A., Chuang, A.W., Burford, M.A., 2016. Nitrogen fixation by the diazotroph Cylindrospermopsis raciborskii (Cyanophyceae). Journal of Phycology 52, 854–862. Willis, A., Chuang, A.W., Woodhouse, J.N., Neilan, B.A., Burford, M.A., 2016b. Intraspecific variation in growth, morphology and toxin quotas for the cyanobacterium, Cylindrospermopsis raciborskii. Toxicon 119, 307–310. Wilson, A.E., Wilson, W.A., Hay, M.E., 2006. Intraspecific variation in growth and morphology of the bloom-forming cyanobacterium Microcystis aeruginosa. Appl Environ Microbiol 72, 7386– 7389. Wood, S.A., Borges, H., Puddick, J., Biessy, L., Atalah, J., Hawes, I., Dietrich, D.R., Hamilton, D.P., 2017. Contrasting cyanobacterial communities and microcystin concentrations in summers

139

REFERENCES

with extreme weather events: insights into potential effects of climate change. Hydrobiologia 785, 71–89. Wood, S.A., Dietrich, D.R., 2011. Quantitative assessment of aerosolized cyanobacterial toxins at two New Zealand lakes. J. Environ. Monit. 13, 1617–1624. Wood, S.A., Dietrich, D.R., Cary, S.C., Hamilton, D.P., 2012. Increasing Microcystis cell density enhances microcystin synthesis: a mesocosm study. Inland Waters 2, 17–22. Wood, S.A., Rueckert, A., Hamilton, D.P., Cary, S.C., Dietrich, D.R., 2011. Switching toxin production on and off: intermittent microcystin synthesis in a Microcystis bloom. Environ. Microbiol. Rep. 3, 118–124. Woodhouse, J.N., Kinsela, A.S., Collins, R.N., Bowling, L.C., Honeyman, G.L., Holliday, J.K., Neilan, B.A., 2016. Microbial communities reflect temporal changes in cyanobacterial composition in a shallow ephemeral freshwater lake. ISME J 10, 1337–1351. World Commission on Dams, 2000. Dams and development. A new framework for decision-making Earthscan. London, UK. Wörmer, L., Cirés, S., Quesada, A., 2011. Importance of natural sedimentation in the fate of microcystins. Chemosphere 82, 1141–1146. Wurts, W.A., Durborow, R.M., 1992. Interactions of pH, carbon dioxide, alkalinity and hardness in Fish Ponds 4. Xiao, M., Li, M., Reynolds, C. S., 2018. Colony formation in the cyanobacterium Microcystis. Biological Reviews, 93(3), 1399-1420. Xie, L., Xie, P., Guo, L., Li, L., Miyabara, Y., Park, H.-D., 2005. Organ distribution and bioaccumulation of microcystins in freshwater fish at different trophic levels from the eutrophic Lake Chaohu, China. Environmental Toxicology 20, 293–300. Yin, X., Goudriaan, J., Lantinga, E.A., VOS, J., Spiertz, H.J., 2003. A flexible sigmoid function of determinate growth. Ann Bot 91, 361–371. Yoon, Y.N., 1992. The state and the perspective of the direct sediment removal methods from reservoirs International Journal of Sediment Research, 7(2), 99–116. Young, F.M., Thomson, C., Metcalf, J.S., Lucocq, J.M., Codd, G.A., 2005. Immunogold localisation of microcystins in cryosectioned cells of Microcystis. Journal of Structural Biology 151, 208– 214. Zamyadi, A., MacLeod, S.L., Fan, Y., McQuaid, N., Dorner, S., Sauvé, S., Prévost, M., 2012. Toxic cyanobacterial breakthrough and accumulation in a drinking water plant: A monitoring and treatment challenge. Water Research, Cyanobacteria: Impacts of climate change on occurrence, toxicity and water quality management 46, 1511–1523. Zarfl, C. Lumsdon, A.E. Berlekamp, J. Tydecks, L. Tockner, K. 2015. A global boom in hydropower dam construction. Aquat Sci, 77, 161–170. Zastepa, A., Pick, F., Blais, J., Saleem, A., 2015. Analysis of intracellular and extracellular microcystin variants in sediments and pore waters by accelerated solvent extraction and high performance liquid chromatography-tandem mass spectrometry. Anal. Chim. Acta 872. Zastepa, A., Taranu, Z.E., Kimpe, L.E., Blais, J.M., Gregory-Eaves, I., Zurawell, R.W., Pick, F.R., 2017. Reconstructing a long-term record of microcystins from the analysis of lake sediments. Sci. Total Environ. 579, 893–901. Zemskov, I., Altaner, S., Dietrich, D.R., Wittmann, V., 2017. Total synthesis of microcystin-LF and derivatives thereof. J. Org. Chem. 82, 3680–3691. Zilliges, Y., Kehr, J.-C., Meissner, S., Ishida, K., Mikkat, S., Hagemann, M., Kaplan, A., Börner, T., Dittmann, E., 2011. The cyanobacterial hepatotoxin microcystin binds to proteins and increases the fitness of Microcystis under oxidative stress conditions. PLoS One 6. Zubala, T., 2009. Influence of dam reservoir on the water quality in a small upland river. Ecohydrology & Hydrobiology 9, 165-173.

140

REFERENCES

141