sign in sign up
<<
Home , Ciliate, Cyanobiont, Dinoflagellate, Dinophysis acuminata, Eukaryote, Histioneis

FACULTY OF SCIENCE UNIVERSITY OF COPENHAGEN

PhD thesis

Woraporn Tarangkoon

Mixotrophic among Marine and : Distribution, and

Academic advisor: Associate Professor Per Juel Hansen

Submitted: 29/04/10 Contents

List of publications 3 Preface 4 Summary 6 Sammenfating (Danish summary) 8 สรุป (Thai summary) 10

The sections and objectives of the thesis 12 Introduction 14 1) Mixotrophy among marine planktonic protists 14 1.1) The role of light, concentration and nutrients for 17 the growth of marine mixotrophic planktonic protists 1.2) Importance of marine mixotrophic protists in the 20 planktonic food web 2) Marine symbiont-bearing dinoflagellates 24 2.1) Occurrence of symbionts in the order Dinophysiales 24 2.2) The spatial distribution of symbiont-bearing dinoflagellates in 27 marine waters 2.3) The role of symbionts and phagotrophy in dinoflagellates with symbionts 28 3) and mixotrophy in the marine genus 30 3.1) Occurrence of symbiosis in Mesodinium spp. 30 3.2) The distribution of marine Mesodinium spp. 30 3.3) The role of symbionts and phagotrophy in marine 33 and Mesodinium pulex Conclusion and future perspectives 36 References 38 Paper I Paper II Paper III Appendix-Paper IV Appendix-I Lists of publications

The thesis consists of the following papers, referred to in the synthesis by their roman numerals. Co-author statements are attached to the thesis (Appendix-I).

Paper I Tarangkoon W, Hansen G Hansen PJ (2010) Spatial distribution of symbiont-bearing dinoflagellates in the Indian in relation to oceanographic regimes. Aquat Microb Ecol 58:197-213.

Paper II Tarangkoon W, Hansen PJ (Submitted) Prey selection, ingestion and growth responses of the common marine ciliate Mesodinium pulex in the light and in the dark. Aquat Microb Ecol

Paper III Hansen PJ, Moldrup M, Tarangkoon W, Garcia-Cuetos L, Moestrup Ø (draft manuscript) Does the marine red ciliate Mesodinium rubrum have replaceable symbionts?

Appendix-Paper IV Farnelid H, Riemann L, Tarangkoon W, Hansen G, Hansen PJ (Submitted)

Putative N2-fixing heterotrophic associated with - consortia in the low- Indian Ocean Aquat Microb Ecol

Paper I is reprinted with kind permission from Inter-Research

3 Preface

This thesis is written as part of the fulfillment of PhD degree from the Faculty of Science, University of Copenhagen. My PhD grant was supported from Rajamangala University Srivijaya, Thailand.During the PhD period, I was based at the Marine Biological Laboratory (MBL), Helsingør. However, Paper I & Appendix-paper IV were based on results from materials sampled during the “Galathea 3” expedition (Leg 7).

Heartfelt thanks to my supervisor, Per Juel Hansen, for all his help and kindness throughout my study. I could not have done it without him. Apart from him being a superb supervisor in an academic way (to open my eyes on culture experiments, contribute ecological thinking, guide the writing etc.), he has also solved my many abroad obstacles during my study (language, finding a place to stay, visa problem, etc). He is always patient, always stimulate me during my difficult moments, and a companion along the way. “Thank you for being you, Per”. I am also grateful for the stimulating contribution of the other co-authors in my 4 papers. It has been a great pleasure to collaborate with you. I am particularly grateful to Torkel G. Nielsen for providing me with the opportunity to join Leg 7 in the “Galathea 3” expedition and sharing environmental data. I also thank other scientists on board, especially Thomas Kiørboe, Andre Visser, Karen Marie Hillingsø, Maria Hastrup Jensen, Carsten Smidt (Captain of HMDS ‘Vædderen’) and his crew for their help during Leg7. Sincere gratitude to Michael Olesen, regardless of what questions I brought up to him e.g. Danes life, religion, relationship, He was always open for an enthusiastic discussion.

I am grateful to all members of the group at the MBL, particularly Lasse T. Nielsen, Karen Riisgaard, Louise K. Poulsen, and Morten Moldrup, who always tried to understand my Thai-English accent and were helpful. Special thanks to Morten, who was frequently bothered (questioning, asking for help, etc) by me during all these . All my officemates, Jane W. Berens, Herik Staahl, Jon Svendsen, Michael van Deurs, Maria F. Steinhausen, Bjørn Tirsgård are thanked for their help and friendly atmosphere. Another special thank is given to Marianne Ernsted, for her help with finding books and big warm hugs when I was frustrated and homesick. Birgit Thorell Lyck, Marriane Saietz, and other staffs at the MBL and the Aquarium also deserve my thanks for their support and help during my study.

4

Many thanks to Marc Staal and Carola Simon for your generous and all thoughtfulness, you guys are my Dutch big brother and sister who have taken good care of me, in my academic as well as during free through these years. Cátia Carreirra, Christian Lønborg, Alejandro M. Garcia, are thanked for kindly help, making me laugh and having good even though we have only met at the last phase of my stay in Denmark but your friendship and kindness are memorized. I am also lucky that I met my new Thai friends and their in laws, Worasiri&Urlik Pederson, Akkaraya&Allan Nielsen, Chanchira&Dennis Mølbæk, Atchaneey Chamnansinp. Thanks for your warm hospitality and help.

I am deeply grateful to my colleagues at the Marine Science Department, Faculty of Science and Fisheries Technology, Rajamangala University Srivijaya for their support and hard work while I was away. Thanks are specially given to Suwat Tanyaros, for the achieved grant and encouragement, and to Pornthep Wiruchawong for listening to all my problems and being helpful in a general sense. Suree Satapoomin and Ajcharaporn Piumsomboon are appreciated for their help, advice and support. Without their help I might not end up having my PhD in Denmark. I am indebted to Suriyan Saramul for his advice in using Surfer program, Patama Singhruck, Itchika Sivaipram for always being there, listening and understanding all my issues.

Last but not least, million thanks are not enough for my beloved family, mom, dad, my sister, my brother, and my grandparents. Without their love, support and beliefs in me, I could never have been strong enough to endure the cold and dark winters in Denmark and finish my PhD.

28th of April 2010

Woraporn (Mam)

5 Summary

The mixotrophic protists (= protists which combine heterotrophy and ) are common in marine waters around the world. They vary widely in their photosynthetic and ingestion capabilities and they add a further complication to the marine planktonic food web. This thesis focused on 2 groups of mixotrophic protists: 1) symbiont-bearing dinoflagellates and 2) ciliates belonging to the genus Mesodinium spp.

The spatial distribution (horizontal and vertical) of symbiont-bearing dinoflagellates (order Dinophysiales, genera , Parahistioneis, Cithoristes, Amphisolenia, Triposolenia) was investigated along a transect from the deep ocean (Indian Ocean) to shallow coastal waters (North West off ), as well as on a transect outside Broome (Australia). The symbionts of these dinoflagellates are either prokaryotic (e.g. heterotrophic bacteria, cyanobacteria) or eukaryotic . concentrations of these dinoflagellates were very low in these waters (< 4 cells L-1). The ectosymbionts-bearing dinoflagellates were most common and had the highest diversity in waters characterized by high temperatures (> 28 ºC) and very low nitrogen concentrations (< 0.4µM). Using light and transmission electron , we could demonstrate that Ornithocercus spp. ingested not only their ectosymbionts but also other prey items (i.e. ciliates). For future research on their physiology, the successful establishment of these in laboratory culture is required.

The ciliate genus Mesodinium contains heterotrophic and mixotrophic species (so far only one symbiont containing species M. rubrum has been described). This study investigated the prey selection, photo and feeding physiology of a non-symbiont containing Mesodinium species, M. pulex. The results showed that Mesodinium pulex ingests a variety of prey cells, but that ingestion rates and especially growth rates varied depending upon the diet. The effects of light and prey concentration on photosynthesis, ingestion and growth rate of M. pulex was studied in detail when fed the dinoflagellate rotundata, The photosynthetic performance of Mesodinium pulex was quite small, amounting for less then 4 % of its carbon uptake, indicating that M. pulex is primarily a heterotrophic species. Despite this, light affected ingestion rates. Ingestion rates increased by a factor of 2 in the light compared to in the dark. Consequently, growth rates also increased in the light.

6 The ciliate Mesodinium rubrum contains symbionts of cryptophyte origin. In the laboratory, our strain of Mesodinium rubrum is normally cultured on cryptophytes within the “Teleaulax ”. Prey selection of M. rubrum was investigated by offering different prey types (i.e. cryptophytes, dinoflagellate). Mesodinium rubrum ingested all the offered prey, but it could only maintain sustained growth when fed on Teleaulax amphioxeia. To whether the symbionts of M. rubrum are permanent or temporary (replaceable), M. rubrum cultures were offered preys from 4 different cryptophyte . TEM pictures of M. rubrum revealed no evidence of sequestered from the prey. Also, the molecular data could not confirm that chloroplasts of M. rubrum can be replaced when offered other cryptophyte prey outer the “Teleaulax clade”.

7 Sammenfating

Mixotrofe protister (dvs. protister, der kombinerer en heterotrof og autotrof levevis) er yderst almindelige i verdenshavene. De varierer utroligt meget i deres evne til at fotosyntetisere og optage føde. Dette komplicerer unægtelig det planktoniske fødenet yderligere. Denne afhandling har fokuseret på 2 grupper af mixotrofe organismer: 1) symbiontbærende dinoflagellater og 2) ciliater tilhørende slægten Mesodinium spp.

Den rummelige fordeling (horisontalt og vertikalt i vandsøjlen) af de symbiontbærende dinoflagellater (i ordenen Dinophysiales mere specifikt slægterne Ornithocercus, Histioneis Parahistioneis, Cithoristes, Amphisolenia, Triposolenia) blev undersøgt langs et transekt gående fra det dybe Indiske Ocean til grunde kystnære vande ved Nordvest Australien. Desuden undersøgtes et transekt udenfor Broome (Australien). Symbionterne hos disse dinoflagellater er enten prokaryoter (dvs. heterotrofe bakterier eller cyanobakterier) eller alger. Cellekoncentrationerne af disse dinoflagellater var meget lave i disse vande (< 4 celler L-1). Blandt disse var de ectosymbiontbærende dinoflagellater de mest almindelige, og vi fandt den største artsdiversitet i vande med høj temperatur (> 28 ºC) og lave nitrogen koncentrationer (< 0.4µM). Ved hjælp af lys og transmissions elektronmikroskopi viste vi at Ornithocercus spp. udover at æde deres deres ectosymbionter, også ernærede sig af andre fødeemner (f. eks. ciliater). Fremtidige studier af disse dinoflagellaters fysiologi vil være dybt afhængige af at kunne etablere laboratoriekulturer.

Ciliatslægten Mesodinium indeholder heterotrofe samt mixotrofe arter (endnu er der kun beskrevet en enkelt symbiontbærende art). Dette studie undersøgte bytteselektion samt foto- og fødeoptagelses-fysiologi hos en Mesodinium-art uden symbionter, Mesodinium pulex. Her kunne vi vise, at M. pulex optager et bredt udvalg af arter, men at fødeoptagelsesrater, og især vækstrater varierer alt efter byttet. Effekten af lys og fødekoncentration blev undersøgt i detaljer, når M. pulex blev tilbudt dinoflagellaten Heterocapsa rotundata som føde. Den målte fotosyntese var ret lille, og udgjorde mindre end 4 % af det totale kulstofoptag. Heraf følger at M. pulex formodentligt primært er en heterotrof art. Til trods for dette, så vi at lys påvirkede fødeoptagelsesraten, og denne fordobledes i lys i forhold til mørke. Dette påvirker så igen vækstraten, således at denne er højere i lys end i mørke.

8 Ciliaten Mesodinium rubrum indeholder en symbiont af rekylalge (crytophycé) oprindelse. For at undersøge om symbionten er permanent eller midlertidig, undersøgtes fødeselektion også her. Mesodinium rubrum blev tilbudt et bredt udvalg af føde i passende størrelse (crytophycéer og dinoflagellater). Alle byttetyper blev optaget af Mesodinium rubrum, men vedvarende vækst sås kun når den blev tilbudt Teleaulax amphioxeia, som tilhører ”Teleaulax kladen”. Så for at teste hvorvidt symbionten var permanent eller midlertidig (dvs. udskiftelig), blev den tilbudt cryptophycéer fra 4 forskellige klader. TEM billeder kunne ikke vise at Mesodinium rubrum havde tilegnet sig kloroplastre fra sit bytte. Heller ikke molekylære metoder kunne vise udskiftning af symbionter, hvis ciliaten fik tilbudt byttealger uden for ”Teleaulax kladen”.

9 สรุป (Thai Summary)

โปรโตซัวที่ดํารงชีพแบบ (=โปรโตซัวที่ดํารงชีพแบบผสมโดยกินอินทรียสารและสังเคราะหแสง แบบพืช) เปนกลุมที่พบไดทั่วไปในทะเลทั่วโลก ความสามารถที่หลากหลายในการสังเคราะหแสงและการ กินอาหารของพวกมันทําใหสายใยอาหารในทะเลซับซอนยิ่งขึ้น วิทยานิพนธฉบับนี้มุงเนนโปรโตซัว Mixotroph 2 กลุม คือ ไดโนแฟลกเจลเลตที่มี symbionts อาศัยอยูรวม และซิลิเอตในสกุล Mesodinium

การศึกษาครั้งนี้ศึกษาการแพรกระจายตามพื้นที่ในแนวนอนและแนวดิ่งของไดโนแฟลกเจลเลตที่มี symbionts อาศัยอยูรวม ไดแก ไดโนแฟลกเจลเลตในวงศ Dinophysuales เฉพาะสกุล Ornithocercus, Histioneis Parahistioneis, Cithoristes, Amphisolenia, Triposoleniaจากบริเวณน้ําลกมหาสมึ ุทรอินเดียไปยัง แนวชายฝงที่ตนฝื้ งตะวันตกเฉียงเหนือของประเทศออสเตรเลีย และรวมถึงบริเวณนอกฝงเมือง Broome ประเทศออสเตรเลีย พบวาประเภทของ symbionts ในไดโนแฟลกเจลเลต คือ prokaryotic (ตัวอยางเชน แบคทีเรีย ไซยาโนแคทีเรีย) หรือ สาหรายเซลลเดียว ซึ่งพบไดโนแฟลกเจลเลตที่มี symbiont อาศัยอยูรวมใน ความหนาแนนต่ํา (นอยกว า 4 เซลลตอลิตร) ตลอดพื้นที่ศึกษา นอกจากนั้นไดโนแฟลกเจลเลตที่ม ี ectosymbionts (symbiontsที่อาศัยอยูดานนอกเซลล) สามารถพบไดทั่วไปและมีความหลากชนิดสูงในน้ําที่มี อุณหภูมิสูง (สูงกวา 28 ºC ) และสารอาหารไนโตรเจนต่ํา (ต่ํากวา 0.4μM) การศกษาองคึ ประกอบภายใน เซลลของไดโนแฟลกเจลเลตสกุล Ornithocercus แสดงวาไดโนแฟลกเจลเลตสก ุลนี้สามารถกินอาหารที่เปน ทั้ง symbionts ของพวกมันเองและสิ่งมีชีวตอิ ื่น เชน ซิลิเอต ทั้งนี้การศึกษาวจิ ัยเกี่ยวกับสรีรวิทยาของไดโน- แฟลกเจลเลตที่มี symbionts อาศัยอยูรวมในอนาคตขึ้นอยูกับความสําเร็จในการเลี้ยงไดโนแฟลกเจลเลตกลุม นี้ในหองปฏิบัติการ

ซิลิเอตในสกุล Mesodinium ประกอบดวยชนิดที่ดํารงชีพแบบกินอินทรียสารซึ่งพบไดหลากกลาย ชนิดและชนิดที่ดํารงชีพแบบกินอินทรียและสังเคราะหแสงแบบพ ืชผสมกัน ซึ่งในปจจุบันพบเพียงชนิดเดียว คือ Mesodinium rubrum ที่มี symbionts อยูดวย การศึกษาครั้งนี้มุงศึกษาการเลือกสรรเหยื่อ การสังเคราะห แสงและสรีรวทยาการกิ ินอาหารของ Mesodinium อีกชนิดหนึ่ง คือ Mesodinium pulex ซึ่งเปนชนิดที่ไมมี symbiont อาศยอยั ูรวม วาสามารถดํารงชีพแบบ mixotroph ไดหรือไม ผลการศึกษาพบวา M. pulexสามารถ กินเหยื่อไดหลากชนิดแตอัตราการกินและอัตราการเติบโตของ Mesodinium pulex ผันแปรตามชนิดของเหยื่อ การศึกษาอิทธพลของความเขิ มแสงและปริมาณเหยื่อวามีผลตออัตราการสังเคราะหแสง อัตราการกินและ การเติบโตของ M. pulex เมื่อใหไดโนแฟลกเจลเลตชนิด Heterocapsa rotundata เปนเหยื่อ พบวาอ ัตราการ สังเคราะหแสงต่ํากวา 4% ของคารบอนจากการกิน ซึ่งบงชี้วา M. pulex มีการดํารงชีพแบบกินอินทรียสาร

10 เปนหลัก นอกจากนี้แสงมีอิทธิพลตออัตราการกินโดยเพิ่มอัตราการกินขึ้นเปน 2 เทาในที่มีแสงพอเพียงเมื่อ เทียบกับที่มืดสนิท ซึ่งสงผลใหการเติบโตในที่มีแสงสูงขึ้นดวยเชนกัน

ซิลิเอตอีกชนิดในสกุลเดียวกันนี้ Mesodinium rubrum ซึ่งเปนซิลิเอตที่มี symbionts ชนิดที่ตนกําเนิด เปนสาหราย cryptophyte การศึกษาเรื่องการเลือกสรรเหยื่อของ M. rubrum สายพันธุเดนมารก ที่เลี้ยงดวย สาหราย cryptophyte ในกลุม “Teleaulax clade” ในหองปฏิบัติการ โดยใหเหยื่อหลายชนิด (ตัวอยางเชน ไดโนแฟลกเจลเลต สาหราย cryptophyte) พบวาซิลิเอตชนิดนี้กินเหยื่อทุกชนิด แตมีการเติบโตเมื่อไดรับ อาหารชนิด Teleaulax amphioxeia เพียงชนิดเดียว ศึกษาวา symbionts นั้นเปนแบบอาศัยรวมแบบถาวรหรอื เปลี่ยนแทนที่ไดในซิลิเอต M. rubrum โดยการใหเหยื่อเปนสาหราย cryptophyte ตางชนิดจาก 4 กลุม (clade) ที่ตางกัน สัณฐานวิทยาภายในของเซลลซิลิเอตเมื่อกินสาหราย cryptophyte ไมพบหลักฐานวา chloroplasts ของเหยื่อถูกเก็บรักษาไวโดยซิลิเอต และผลการศึกษาทางชวโมเลกี ุลยืนยันวา chloroplasts ของซิลิเอต M. rubrum นั้นไมสามารถเปลี่ยนแทนที่โดยสาหราย cryptophyte นอกกลุม Teleaulax clade ได

11 The sections and objectives of the thesis

Mixotrophy among protists is widespread in marine waters, especially among ciliates and dinoflagellates. Some types of mixotrophic protists have not yet been established in laboratory culture, mainly because our knowledge of their feeding preferences and physiology is very restricted. In those cases, important information on their biology can be gathered from studying their distribution in the in relation to environmental factors like light, inorganic nutrients, temperature etc. Other types of mixotrophic protists have recently been brought into culture, but our knowledge of their physiology and prey preferences is in many cases still very limited.

The first part of this thesis focuses on how the distribution of the symbiont-bearing dinoflagellates (Order Dinophysiales) is related to environmental factors in the Indian Ocean. None of these species have ever been established in laboratory culture and very little is known about their distribution and biology. This study was part of larger project carried out during the “leg 7” of the Galathea 3 expedition, autumn 2006. The objective of this work was to 1) study the spatial distribution (vertical and horizontal distribution) of the symbiont-bearing dinoflagellates from oceanic stations to coastal stations during the cruise from Cape Town, South of Africa to North West Australia, 2) relate the cell distributions to physical and chemical variables, 3) study the relationship between the symbionts and the host cells. This includes the use of light and microscopy to reveal the contents of food (Paper I). Finally, I collected a large number of species during the cruise in order to study the potential of these symbionts to fix N2 (i.e. having NifH ). Since the molecular work, which is the bulk part of the work, was carried by colleagues at the Linnaeus University and I therefore just attached this part of my work as an appendix (Appendix-Paper IV).

The second part of my thesis deals with the ciliates Mesodinium spp, which are common and important species in marine waters. This part aimed at understanding the prey selection and feeding physiology of a non-symbiont containing Mesodinium pulex and the symbiont containing species M. rubrum (Paper II, Paper III). I specifically investigated the photosynthetic rate of M. pulex cells to test to what extent this species rely on photosynthesis for growth and survival (Paper II). I also studied the effects of light on ingestion and growth rates of M. pulex. There is an ongoing discussion on whether the symbionts of M. rubrum are

12 replaceable or not. Thus, I studied this by offering M. rubrum different types of cryptophytes and monitored the growth and ingestion rates of M. rubrum in each case. The fate of ingested prey cells were investigated using a combination of ultrastructural and molecular techniques (Paper III).

13 Introduction

Introduction

This introduction comprises of 3 parts of which the first part presents the current knowledge of marine mixotrophic protists in general. The second and third parts of the introduction present what we do know on the specific groups of marine mixotrophic protists (i.e. symbiont- bearing dinoflagellates and Mesodinium spp.), which I have studied during my PhD study.

Part 1. Mixotrophy among marine planktonic protists

Mixotrophy is a term often used to describe the use of mixed nutrition by an . The most popular use of this term refers to the combined use of photosynthetic and (e.g. phagotrophy, , saprotrophy) in a single organism (Caron 2000, Stoecker 1999, Jones 1994, Burkholder et al. 2008).

Mixotrophic planktonic protists are common in aquatic habitats around the world. They can be found in polar to equatorial regions, in coastal to oceanic waters, in freshwater as well as in ; they can even be encountered in super euryhaline waters. Also, they live in nutrient poor as well as in eutrophic waters (reviewed in Stoecker 1999, Burkholder et al. 2008). Mixotrophic protists vary widely in their photosynthetic and ingestion capabilities and may be considered to occupy a wide spectrum of nutritional strategies from absolute autotrophy to absolute heterotrophy (Jones 1994). Therefore, mixotrophic planktonic protists complicate the flow of and nutrients in food webs by functioning both as producers and consumers, rendering classical models of aquatic incomplete (Fig. 1). For example, mixotrophic planktonic protists compete with and bacteria for dissolved inorganinc and organic nutrients and prey on them at the same time (Ptacnik et al. 2004). Mixotrophic protists may not only compete with phototrophs for soluble nutrients, but may in some cases even actually facilitate them (Rothhaupt 1996). Mixotrophic protists prey on a variety species and sizes of prey depending upon their feeding habits. Thus, different feeding strategies may put them on different trophic levels in planktonic food webs (Hansen & Calado 1999).

14 Introduction

Zooplankton

Mixotrophs protists Herbivorous Bacterivorous protists protists

Viral lysis DOM Heterotrophic & bacteria POM

Fig. 1 Marine mixotrophic protists in the marine . Solid arrows represent pathways of consumption of organic ; dashed arrows represent pathways by which organic matter, both dissolved (DOM) and particulate (POM) is released from living organisms (adapt from Caron 2000).

Roger I. Jones (Jones 1994) was the first to describe mixotrophy nutrition as a spectrum of nutritional strategies from absolute heterotrophy over mixotrophy to absolute autotrophy in aquatic protists (Fig. 2). He separated the organisms into three groups. The first group comprises of organisms which have chloroplasts of their own (left side on Fig. 2). In this group the entire mixotrophic continuum can be found from largely phototrophic to large heterotrophic organisms. The second group is made up of organisms which can sequester (steal) functional chloroplasts from algal prey (right side bottom). These organisms are considered mainly heterotrophic organisms which mainly use photosynthesis to survive periods of starvation. The third group comprises of organisms which form symbiosis with alga. These are considered mainly to belong to the phototrophic end of the spectrum.

15 Introduction

Fig. 2 Schematic representation of the continuum of nutritional strategies amongst planktonic protists. Examples are shown of genera which include mixotrophic forms amongst their species. (Jones 1994)

Harriet JL Jones (Jones 1997) further discussed the role of light and prey concentration for aquatic (but mainly freshwater) with their own chloroplasts (= mixotrophy in phytoflagellates). She categorized these phytoflagellates into 4 groups of mixotrophics (A, B, C and D) depending upon the respective role of photosynthesis and phagotrophy: Group A comprises of organisms that are primarily heterotrophic and phototrophy is employed only when prey concentrations limit heterotrophic growth. In all the other 3 groups (B, C and D) phototrophy is the dominant mode of nutrition. In group B, phagotrophy supplements growth when light is limiting, therefore growth is inversely proportional to light intensity. In group C, phagotrophy provides essential substances for growth and ingestion is proportional to light intensity. Group D includes protists which have very low ingestion rates, ingesting prey only, for example, for cell maintenance during prolong dark periods. Later, Stoecker (1998) made a review in which the goal was to include all types of mixotrophy, thus including the protists with symbionts and those which sequester chloroplasts from their prey. In this review she treated, besides light and prey concentration, also the role of nutrients and growth factors. She

16 Introduction came up with 3 main types for which she made predictions for their dependency on light, prey concentration and nutrients. In short, Type I are the ‘ideal mixotrophs’ which are able to utilize phototrophy and phagotrophy equally well. The type II, covers all what she called the primarily phototrophic “algae”, which for different reasons supplements photosynthesis with food uptake. Finally, her type III mixotroph, covers predominantly heterotrophic which use photosynthesis for survival in the light when food is scarce or use photosynthesis as a supplement in carbon . The three reviews mentioned above (Jones 1994, Jones 1997, Stoecker 1999) are all more than 10 years old and were all build upon a very limited number species of which many species were freshwater species. Since then many more have been studied.

In the following part (part 1), I will try in a short way to summarize our current knowledge on role of light, prey concentration and nutrients for the growth of marine mixotrophic planktonic protists and their role in the planktonic food web

1.1) The role of light, food concentration and nutrients for the growth of marine mixotrophic planktonic protists

Mixotrophic protists with their own chloroplasts (=phytoflagellates)

Mixotrophic protists with chloroplasts of their own display a growth response towards irradiance like an ordinary autotrophic alga. However their growth response to addition of preys varies quite bit among species. In the case of the dinoflagellate Fragilidium subglobosum, addition of prey will increase the maximum growth rate at all irradiances. However, the difference in growth rates is by far largest at low irradiances and this species can even grow quite fast in the dark if supplied with fresh food (Skovgaard 1996). Overall this species grows equally well as a phototrophic and a heterotrophic. However, it grows fastest as a mixotrophic. Some down regulation of the photosynthetic apparatus occurs when F. subglobosum is well fed and the contribution of phagotrophy to the total carbon uptake is more than 70%, even at high irradiances.

The prymnesiophytes spp. function completely different. In these species the addition of even large amounts of prey will only affect the growth rate at low

17 Introduction irradiances and thus food serves only as carbon supplement when light becomes limiting for growth (Jones 1997, Hansen & Hjorth 2002). Feeding in these species does not seem to boost the photosynthetic apparatus (Jones 1997). However, light is required for growth in Chrysochromulina spp (Jones 1997). In other cases, like in the dinoflagellates armiger and K. veneficum/micrum (Li et al. 1999, Berge et al. 2008a), the maximum growth rate in a is quite low (between 0.01-0.25 d-1 depending upon species and the use of soil extract) even at high irradiances. Addition of prey will lead to considerably higher growth rates (0.7 – 0.94 d-1) at relative high irradiance. So in these species, feeding is very much dependent upon light and they will not feed in the dark (Li et al. 1999). Feeding also seems to boost the photosynthetic apparatus Karlodinium spp, as chloroplasts tend to increase in size when the cells are fed (Berge et al. 2008a,b, Li et al. 1999).

The dinoflagellate Gymnodinim resplendens behaves very much like the Karlodinium species. This species has chloroplasts of its own, but it can not grow in the light in an inorganic nutrient medium, if it does not feed on another dinoflagellate, Prorocentrum minimum. The ingested prey supplies the dinoflagellate with organic carbon as well as essential growth factors for phototrophy growth. Like in Karlodinium, feeding seems to boost the photosynthetic apparatus, as rates of photosynthesis go up when the cells are fed. In contrast to Karlodinium spp, resplendens can grow in the dark if fed, but with much lower growth rates (Skovgaard 2000). An even more extreme example is represented by mixotrophic dinoflagellates belonging to the genus . These species have chloroplasts of their own (of either cryptophyte or origin), but they cannot grow in an inorganic medium without food (Riisgaard & Hansen 2009). All Dinophysis species in culture so far have been grown with the ciliate Mesodinium rubrum as prey (Park et al. 2006, Kim et al. 2008, Nishitani et al. 2008a, b, Cuetos-Gracia 2010). Food uptake may provide the dinoflagellate with more than 70 % of its carbon needs (Riisgaard & Hansen 2009) However, at natural prey concentrations food uptake will only make up for less than 10 % of its carbon requirements (Riisgaard & Hansen 2009). Mixotrophic Dinophysis are dependent upon light for growth (Kim et al. 2008, Riisgaard & Hansen 2009).

Some mixotrophic dinoflagellates only feed when they are nutrient limited and stop feeding when nutrients become available e.g. furca (Smalley et al. 2003). Likewise, Heterocapsa triquetra and Procentrum minimum only ingest prey when inorganic nutrients are limiting (Stoecker et al. 1997, Legrand et al. 1998) and ingestion of prey by the

18 Introduction mixotrophic dinoflagellate Prorocentrum minimum is in fact inhibited by addition of nitrate and (Stoecker et al. 1997).

Mixotrophic protist with symbionts and kleptochloroplasts

In the case that do not originate from the mixotrophic itself. The photosynthetic component in mixotrophic can be in different forms. “Symbiosis” and “kleptochloroplastidy” have been found commonly among marine protists (e.g. Norris 1967, Taylor 1982, Laval- Peuto & Rassoulzadegan 1988, reviewed by Stoecker 2009). “Symbiosis” refers to the relationship between two or more organisms. Typically, a symbiont living inside (as ) or is attached to the outside/surface (as ectosymbiont) of another organism (the host). Kleptochloroplastidy (or kleptoplastidy) is the process by which the chloroplasts of a prey are sequestered by a predator and photosynthetically still active for a while in predator. Thus, stolen chloroplasts are called kleptochloroplasts or just kleptoplastids (Schnepf 1993).

Mixotrophic protists with (hosts) tend to largely depend on the photosynthetic activity of their endosymbiont. In many cases, however, the hosts need to ingest prey to get essential substances for sustaining their symbionts. Therefore, when starved under optimal light conditions, the photosynthetic activity supports and prolongs survival of the hosts. An example of this type of symbiosis is found in the green , which is a common bloom former in South East Asia. This species contains thousands of cells of the pedinophyte Pedinomonas noctilucae. Some strains can grow photoautotrophically in an inorganic medium with some vitamins added, while others strains of the green Noctiluca scintillans are dependent upon ingestion of food (Hansen et al. 2004, Saito et al. 2006). The phagotrophy can contribute as much as 30% to the growth rate of the green N. scintillans under saturated prey concentrations and optimum light intensity. However, at natural prey concentrations typically found where it blooms, the contribution of ingested prey is much smaller (a few %; Hansen et al. 2004).

Another example of a marine protist with endosymbionts is the ciliate Mesodinium rubrum. This ciliate contains a cryptophyte symbiont. However, it can only grow if supplied with cryptophyte prey (Gustafson et al. 2000). It gets most of its carbon from photosynthesis, very much like in the case of the green Noctiluca (Johnson & Stoecker 2005, Smith & Hansen 2007).

19 Introduction

Some dinoflagellates, within the order Dinophysiales (i.e. Ornithocercus, Histioneis), bear cyanobacterial ectosymbionts (Norris 1967, Lucas 1991). However, the exact relationship between ectosymbionts and their hosts is not known.

Some marine protists only sequester the chloroplasts from their prey. In all known cases in marine waters, the marine kleptochloroplastidic protists depends on largely phagotrophy for growth, while photosynthesis of the chloroplasts only seem to be enough to cover for respiration and excretion. (Putt 1990a). The sequestered chloroplasts need to replaced, because they only will be functional for a short time; from few to several days depends on species (Stoecker & Silver 1990, Skovgaard 1998, Lewitus et al. 1999). Thus the main role of kleptochloroplasts seems to be that they help the hosts to survive periods of low prey abundances (Stoecker et al. 1988, Stoecker & Silver 1990, Skovgaard 1998,). An example of such a protist is the ciliate Laboea strobila (Putt 1990a), but quite a few “oligotrichs” have the same ability (e.g many Strombidium spp. ; Stoecker & Silver 1990, Jonsson 1987). These ciliates are all able to grow in the dark, although their growth rates are higher in the light. Among the dinoflagellates, kleptoplastidity has been reported for a number dinoflagellates, i.e. poechilochroum (Larsen 1988; Gymnodinium graciletum (Skovgaard 1998, Jakobsen et al. 2000), Cryptoperidiniopsis sp, and piscicida (e.g. Lewitus et al. 1999, Eriksen et al. 2002). Like for the ciliates, the growth rate of the kleptoplastidic dinoflagellate Cryptoperidiniopsis sp. is strongly influenced by light intensity when fed prey cells in excess ( major, cryptophyte).

1.2) Importance of marine mixotrophic protists in the planktonic food web

Scientists dealing the marine plankton food web often split protists into either phototrophic protists (=phytoplankton) and heterotrophic protists (= protozoa) and relatively few papers have incorporated mixotrophs into the web. One of the reasons for this is that our knowledge on which species are mixotrophs is still somewhat limited. Another reason is that it is difficult to “identify” mixotrophs in field samples. In many cases you will not be able to apply a species name to fixed cells in water samples. In those cases, mixotrophy is “confirmed” by addition of stained prey or plastic beads. In a few cases, where the grazing of large protists has been considered, beads have been added to some smaller protists which then again have been fed to the mixotrophs in question (Smalley & Coats 2002). Pitta &

20 Introduction

Giannakourou (2000) reviewed the obstacles in distinguishing mixotrophic from phototrophic or heterotrophic protists with recently ingested algae. They also discussed the use of different methodology in sampling and preserved which led either underestimated or overestimated number of mixotrophic planktonic protists. Overall, however, mixotrophs are every where; it is just a matter of identifying them and their impact on the food web.

Mixotrophic nanoflagellates can contribute up to 39 - 50 % of the phototrophic nanoflagellates abundance in surface waters of Sargasso Sea (Arenovski et al. 1995, Sanders et al. 2000). The prey of nanoflagellates is usually consisting of bacteria and in some cases picoplanktonic algae. Zublov & Tarran (2008) revealed that the small mixotrophic (<5μm) carried out 40-95% and 37-70% of the bacterivory in the euphotic layer of temperate North Atlantic Ocean in summer and in the surface waters of the tropical North –East Atlantic Ocean, respectively. Marine mixotrophic nanoplankton are also important grazers on (i.e. ) or (Sanders et al. 2000, Tsai et al. 2007, Chan et al. 2009). Safi & Hall (1999) revealed that the mixotrophic nanoflagellates contributed 57 % of measured grazing impact on picophytoplankton-sized , 40% of grazing on bacteria-sized particles and 55% of grazing on stained bacteria per day. Also, in the northwestern during the summer 1995, the mixotrophic planktonic protists (nano- and microplankton) contributed to 14% and 24% of the ingestion of bacteria and nanoplankton, respectively even though their were significantly lower than heterotrophic planktonic protists (Bouvier et al. 1998). Recently, Unrein et al (2010) estimated that only one single mixotrophic species, Dinobyon faculiferum represent 4.5% of the total bacterial grazing losses in coastal Mediterranean.

Very little is known about the role of mixotrophic ciliates and dinoflagellates as grazers in natural communities. Smalley & Coats (2002) investigated the feeding rate of dinoflagellate in Chespeake Bay. They obtained feeding rates of 0-0.11 prey C. furca-1h-1 and estimated that Ceratium furca alone removed on average 16% of Choreotrichid (<20µm size) d-1and 67% of Strobilidium spp per day.

Li et al. (2000) revealed the mean number of ingested cryptophyte per Gyrodinium galatheanum was high as 0.46 for G. galatheanum populations in the surface waters of the mid- and upper of Chesapeake Bay. However, G. galatheanum has only minor grazing impact on cryptophyte prey populations, removing about 0-4% of the cryptophyte standing stock on a

21 Introduction daily basis. Mixotrophic ciliates can form an important part of the photosynthetic biomass. Estimates of their contribution to total range up to 24% for surface samples from Nordic Seas (Putt 1990b). Mixotrophs can likewise make up for a substantial biomass of grazers. Stoecker et al. (1996) estimated the biomass of mixotrophs (i.e. ciliates, , acantharia, ) in upper 90 m in Equatorial Pacific (at 1400W) made up for as much as 27%, 47% and 56% of the protozooplankton biomass in size ranges 20-64 µm ,>64-200 µm and > 200µm, respectively and this does even not include the mixotrophic dinoflagellates. Pitta & Giannakourou (2000) found that mixotrophic ciliates (i.e mixotrophic oligotrichs) contributed 17-54% of ciliate abundance and 13-62% of ciliate biomass in oligotrophic Eastern Mediterranean. Especially one mixotrophic ciliate tends to dominate in many waters, Mesodinium rubrum. This species is primarily a phototrophic mixotroph, but when it blooms it will have a huge impact on the occurrence of other protists (Smith & Hansen 2007). Even in non bloom occurrences, Stoecker et al. (1991) reported Mesodinium rubrum made up for from <1 to ≥ 70% of the community primary production in the surface water.

One of the disadvantages of the methods used in the papers mentioned above is that these techniques do not give the full picture of what the protists eat. Recently, a new technique has been invented, which in principle will allow for the determination of more realistic ingestion rates (i.e. a measure of the total amount of food ingested) (Rose et al 2004). The principle of the method is that it uses a pH stain (a fluorescent acidotropic probe) which will penetrate the predator cell and stain food vacuoles (with low pH) (Carvalho & Granéli 2006). Carvalho et al. (2008) combined this technique with flow cytometry to verify feeding frequency (i.e. the percentage of cells containing food vacuoles in a given ) in natural populations in the Baltic Sea. Using this technique a greater number of D. norvegica cells can be examined in the short time and it allows the analysis of live cells from cultures or natural populations in real time. Also, a much higher frequency of food vacuoles were obtained with this technique than previously reported. The limitation of the acidotropic probe technique is that it will stain other acidic vesicles (i.e. autophagic, parasitophorus) and not only food vacuoles (Carvalho & Granéli 2006). Recently, Bowers et al. (2010) combined flow cytometry and real-time PCR to demonstrate consumption of a cryptophyte species ( sp.) by a toxic mixotrophic haptophyte ( parvum). Using flow cytometry, the feeding frequency of a population of P. parvum cells was calculated using the phycoerythrin (PE) autofluorescence signal from Rhodomonas sp. and the

22 Introduction of an acidotropic probe labeling the food vacuoles. Thus, flow cytometry allowed for a rapid enumeration of food vacuoles and enumeration of consumed prey cells based on their pigments, while real-time PCR confirmed the identity of the species within the food vacuoles. These recent advanced techniques (i.e. Carvalho et al. 2008, Bowers et al. 2010) now allow us to study the role of phagotrophy by mixotrophs in natural populations much better.

23 Introduction

Part 2) Marine symbiont-bearing dinoflagellates

Part 2.1 Occurrence of symbionts in the order Dinophysiales

The dinoflagellate order Dinophysiales includes the families Amphisoleniaceae, Dinophysiaceae, and Oxyphysaceae (Steidinger & Tangen 1997). So far, symbionts have only been found associated with species within the families Amphisoleniaceae and Dinophysiaceae (Table1). The family Amphisoleniaceae consists of the genera Amphisolenia and Triposolenia, which both of which have species that contain endosymbionts of prokaryotic as well as of eukaryotic origin (Kofoid & Skogsberg 1928, Lucas 1991). Recently, Daugbjerg et al. (unpublished) revealed that symbionts of A. bidentata were in fact closely related to some free-living pelagophytes.

Fig. 3 Some marine symbionts-bearing dinoflagellates in order Dinophysiales: a) Amphisolenia b) Triposolenia c) Citharistes d) Histioneis e) Parahistioneis f) Ornithocercus (Kofoid & Skogsberg 1928, Taylor 1976)

24 Introduction

In the family Dinophysiaceae, the genera Citharistes, Histioneis, Parahistioneis, Ornithocercus , all bear ectosymbionts (Fig. 3). Originally, the ectosymbionts were named “Phaeosomes” by Schütt (1895), who first gave this name to the brown coloured bodies he discovered in the girdle list of these dinoflagellates. Norris (1967) was the first to identify that these structures contained symbionts and after studying the living specimens, he finally concluded that these symbionts were cyanobacteria. The species he found were named Synechococcus carcerarius and Synechocystis consortia. Apart from these species he also found some symbionts, which he thought might be eukaryotic algae. In 1991, Lucas identified 4 types of cyanobacterial symbionts by examination of the cytoplasmic content of cells (Lucas 1991). Foster et al. (2006b) later provided the first phylogenetic description of the diversity of symbionts of dinophysoid genera (Amphisolenia, Citharistes, Histioneis, Ornithocercus). They revealed that majority of the cyanobacterial symbionts sequences were most similar to Synechococcus and , while some sequences had a low identity (<92%) to eukaryotic chloroplasts. Some sequences were, however, more similar to a variety of heterotrophic bacteria. In the same , Foster et al (2006a) identified 8 different morphotypes of cyanobacterial ectosymbionts and 2 bacteria morphotypes from Ornithocercus spp. and Histioneis spp. based on TEM.

Other members of the family Dinophysiaceae include the genus Dinophysis, which contains phototrophic species which have either cryptophyte (e.g. D. acuminta, D. acuta, D. norvegica) or haptophyte (only D. mitra) chloroplasts (Table 1) . Molecular data have suggested that Dinophysis spp. relies on sequestered chloroplasts from ingested prey (kleptochloroplasts; e.g. Janson 2004, Minnhagen & Janson 2006, Nagai et al. 2008, Park et al. 2008). Recently, Garcia-Cuetos et al. (2010), however, using a combination of molecular, TEM and experimental techniques revealed that the chloroplasts of D. acuminata in fact are chloroplasts, which are now fully incorporated in the dinoflagellate, and not kleptochloroplasts, which the cells get from ingested prey. However, more research is needed to clarify if this is true for all phototrophic Dinophysis spp. Nevertheless, in this thesis, the Dinophysis spp. are not considered symbiont bearing.

25 Introduction

Table 1. Mixotrophy in marine Dinophysiales

Host species Status of Symbionts species Distribution References symbiont/chloroplast (or taxa)/chloroplast origin Family Amphisoleniaceae Amphisolenia spp. endosymbiont Eukaryotic algae (chrysophyte? or Oceanic, sometimes associated Paper I & references therein dinoflagellate?), cyanobacteria e.g. with ; cosmopolitan in Synechococcus carcerarius, bacteria warm temperate to tropical waters Pelagophyte species Daugbjerg et al. unpublished Triposolenia spp. endosymbiont? Small irregular, very pale yellowish- Probably exclusive eupelagic; Paper I & references therein green chromatophores cosmopolitan in warm temperate to tropical waters Family Dinophysiaceae Citharistes spp. ectosymbiont Cyanobacteria Oceanic, tropical, subtropical and Paper I & references therein warm temperate seas Histioneis/ ectosymbiont Eukaryotic algae, cyanobacteria e.g. Oceanic; cosmopolitan in warm Paper I & references therein Parahistioneis spp. Synechococcus carcerarius, temperate to tropical waters Synechocystis consortia

26 Ornithocercus spp. ectosymbiont Cyanobacteria e.g. Synechococcus Neritic,Oceanic; cosmopolitan in Paper I & references therein carcerarius, bacteria warm temperate to tropical waters chloroplast yellow-orange autofluorescence, Neritic, typically cold and warm Lessard & Swift 1986, Cryptophyte origin temperate waters, worldwide Takishita et al. 2002, Garcia-Cuetos et al. 2010 Dinophysis caudata chloroplast Cryptophyte origin Neritic and estuarine in warm Park et al. 2008 temperate to tropical waters, worldwide; rarely found in cold water, possibly an intruder in warm water masses Dinophysis fortii chloroplast Cryptophyte origin Oceanic and neritic; cold temperate Nagai et al. 2008, Takishita to tropical waters, worldwide et al. 2002 Dinophysis infundibulus chloroplast Cryptophyte origin Temperate waters Nishitani et al. 2008b Dinophysis mitra chloroplast Haptophyte origin Temperate to tropical waters Koike et al. 2005 Dinophysis norvegica chloroplast yellow-orange autofluorescence, Neritic,Oceanic; temperate waters Lessard & Swift 1986 Cryptophyte origin Takishita et al. 2002 ?: Status not confirmed

26 Introduction

Part 2.2 The spatial distribution of symbiont-bearing dinoflagellates in marine waters

The geographical distribution of symbiont-bearing dinoflagellates (i.e. Amphisolenia, Citharistes Histioneis Ornithocercus, Parahistioneis, and Triposolenia) belonging to the order Dinophysiales, is restricted to tropical, subtropical and warm-temperate seas from about 460N to 400S (Table 6&Fig 10 in Paper I) and they are exclusively found in oligotrophic waters (Gordon et al. 1994, Jyothibabu et al. 2006, Paper I). Very little is known about these dinoflagellates as most of the papers published so far have been qualitative studies. The reason for this is undoubtedly due to their scarce numbers even in their most favored offshore environments (0.1-5 cells l-1) and to fact that none of them have been successfully established in the laboratory cultures yet. Few researches have investigated the seasonal distribution of symbiont-bearing dinoflagellates. Gordon et al. (1994) investigated the seasonal distribution of ectosymbiont-bearing dinoflagellates (genera Ornithocercus, Hisitoneis/Parahistioneis and Citharistes) during almost 5 years in the Gulf of Aqaba, Red Sea. They found that the ectosymbiont-bearing dinoflagellates showed peak in numbers during periods characterized by thermal stratification and nitrogen limitation. Similar results were obtained by Jyothibabu et al. (2006) in the Bay of Bangal. Here the highest cell concentrations of ectosymbiont- bearing dinoflagellates were found in the spring intermonsoon, which is characterized by strong thermal stratification and nitrogen depletion in the upper water layers. To the best of my knowledge, there are no reported on the temporal distribution of endosymbiont-bearing dinoflagellates.

In Paper I, I investigated the horizontal and vertical distribution of symbiont-bearing dinoflagellates along the Indian Ocean and in Broome Sea, Australia. My study revealed that the highest cell concentrations and the highest species diversity of symbiont-bearing dinoflagellates were found at surface temperature between 20-300C and associated with low nitrogen concentrations (N-limitation). However, the symbiont dinoflagellates were not common in coastal waters, with high turnover rates of , even though nitrogen concentrations were low. This indicates that turnover rates of nutrients also play a role. Symbiont-bearing dinoflagellates have typically been found in surface waters down to 100-400 m’s depth (Kofoid & Skogsberg 1928, Gómez 2005, Paper I). Paper I suggested that their vertical distributions were related to the depth of optimal irradiance for the growth of the symbionts (in the upper 100 m). However, other factors (i.e prey cells) may play a role to in constraining the vertical distribution of the symbiont-bearing dinoflagellates.

27 Introduction

Part 2.3. The role of symbionts and phagotrophy in dinoflagellates with symbionts

What do the otherwise heterotrophic dinoflagellates get out of having ectosymbionts? Light microscopical photographs (Fig 7 in Paper I) showed food vacuoles inside the dinoflagellates, which had resemblance in size and color to their symbionts. TEM pictures confirmed food vacuoles inside the dinoflagellates, but the remains could not in most cases be identified. However, one food revealed remnants of a chloroplast close to . The trichocysts seemed to come from a ciliate and thus it was speculated that this was remains from a ciliate, which had ingested a eukaryotic prey (Fig 9 in Paper I). Likewise, Lucas (1991) found that some cells of Ornithocercus magnificus, Histioneis dolon and Parahistineis para contained food vacuoles with unidentified prey inside. Unfortunately, no direct observations of the ingestion process in symbiont-bearing dinoflagellates have been observed so far.

The ectosymbionts in Dinophysiales are cyanobacteria and has been shown among marine strains of small unicellelular cyanobacteria: Synechococcus, Synechocystis, and Gleocapsa sp. (Spiller & Shanmugam 1987, Reddy et al 1993). Since the cyanobionts of symbiont-bearing dinoflagellates have been shown to be closely related to the N2 fixing genera Synechococcus and Cyanotheca, it seems likely that they indeed fix nitrogen (Foster et al 2006b). Also, it is well known that high temperatures and N-limitation seem to promote nitrogen fixation in non diazotrophic cyanobacteria (Gordon et al.1994, Capone 2000, Karl et al. 2002, Staal et al. 2003, Breitbarth et al. 2006). It may in fact be one of reasons why the -bearing dinoflagellates are restricted to warm nutrient depleted waters.

So far, cyanobacteria symbionts have been found as ectosymbionts in the genera Ornithocercus spp., Histioneis spp., Parahistioneis, Citharistes spp. and confirmed as endosymbionts in Amphisolenia spp. (Norris 1967, Lucas 1991, Janson et al. 1995, Foster et al. 2006a, b, Paper I, Appendix-Paper IV). However, Janson et al. (1995) did not detect the (nitrogen fixing ) from cyanobionts of Ornithocercus spp. by labeling antisera against the enzyme nitrogenase, which is involved in nitrogen fixation. Interestingly, the cyanobiont type 4 of Histioneis depressa is so far the only one, which has been shown to be capable of N2 fixation (Foster et al 2006a). Recently, Farnelid et al. (Appendix-Paper IV)

28 Introduction revealed that some heterotrophic bacteria, which were found in association with the symbiont- bearing dinoflagellates, were the ones containing the nifH genes. Much more research is however required on this topic

Some Amphisolenia species (i.e A. bidentata, A. thrinax), possess eukaryotic endosymbionts (Lucas 1991, Appendix-Paper IV, Daugberg et al. umpublished). The symbionts appear to be completely intact eukaryotic cells (pelagophytes). No food vacuoles have been found in Amphisolenia species, indicating the species do not digest their symbionts and thus rely on photosynthetate leaking from the symbionts (Lucas 1991).

29 Introduction

Part 3) Symbiosis and mixotrophy in the marine ciliate genus Mesodinium

Part 3.1. Occurrence of symbiosis in Mesodinium spp.

The ciliate genus Mesodinium (Stein 1863) is a small genus belonging to the order Cyclotrichiida ( ; Lynn 2008). This genus is globally distribution in aquatic habitats and includes species which are mixotrophic as well as completely heterotrophic. The mixotrophic species (the red M. rubrum (Lohmann 1908) Hamburger & Buddenbrock 1911 (=Myrionecta rubra), possesses a symbiont which originates from the Teleaulax / clade (Hansen & Fenchel 2006, Garcia-Cuetos et al. 2010). This species is found in saline lakes, brackish and marine waters (Lindholm 1985, Perriss et al. 1995). Isolates of Mesodinium rubrum are now in culture from Antarctic waters (Gustafson et al 2000) and temperate waters (Korea: Yih et al. 2004, Denmark: Hansen & Fenchel 2006). To what extent all these isolates are the same species is unknown at present.

Heterotrophic non-symbiont containing species i.e. M. pulex Claparède & Lachmann 1858,1859, M. acarus Stein 1867, M. fimbriatum Stokes 1887, M. cinctum Calkins 1902, M. pupula Kahl 1933, and M. velox Tamar 1986 have been found in both freshwater and marine waters, except for M. velox, which has only been reported from (Tamar 1986, Foissner et al. 1999). Interestingly, some species (i.e. M. pulex, M. acarus) have been found in both planktonic and benthic habitats. However, M. pulex can easily be confused with M. acarus and M. fimbriatum (Noland 1937, Foissner et al. 1999). Recently, Bass et al. (2009) discovered that clones of so-called M. pulex were located in two different phylogenetic clades, indicating the existence of an additional undescribed species. Clearly, more research on the species diversity in this genus is needed.

Part 3.2. The distribution of marine Mesodinium spp.

Marine Mesodinium spp. includes species without symbionts i.e. M. pulex, M. acarus , M. pupula and M. velox and the symbiont containing M. rubrum (Lindholm 1985, Tamar 1986, Tamar 1992, Madoni 2006). Best known among the species without symbionts is M. pulex, which is common in brackish and marine waters and the only one of this type which is established successfully in laboratory cultures (Dolan & 1991, Jakobsen et al. 2006, Table 2).

30

Introduction

Table 2 Occurrence of Marine Mesodinium spp.

Length Width Abundance Species Living form Location references (μm) (μm) (x103μm3) (cells l-1) (mean) Heterotrophic species M. .acarus - - - - benthic Lesina , Adriatic Sea Madoni 2006 - - - - benthic Chernaya River ,Kandalaksha Bay, Mazei & Burkovsky 2005 White Sea M. pulex 18 - 1.3 - planktonic in L'Houmeau, near La Rochelle, Dupuy et al.2007 French Atlantic coast - - - 15x103-≥ 3.00x105 benthic Molenplaat, Schelde estuary, Netherlands Hamels et al. 2005 20 15 345-6250 planktonic solar saltern of the Yellow Sea, China Lei et al. 2009 - - - - benthic Chernaya River estuary,Kandalaksha Bay, Mazei & Burkovsky 2005 White Sea - - - - planktonic Baltic Sea Mironova et al. 2009 18-25 15-18 - - planktonic Yellow Sea, China Zhang et al. 2002 - - - - planktonic Nervión River estuary, Bay of bisca, Spain Urrutxurtu et al. 2003 12 - - - planktonic Oyster farming area, Thau Lagoon, France Dupuy et al. 2000 31 M. pulex var pupula - - - - benthic Chernaya River estuary,Kandalaksha Bay, Mazei & Burkovsky 2005 Kahl 1933 White Sea M. pupula 25-30 18-20 planktonic Yellow Sea, China Zhang et al. 2002 - - - - planktonic Baltic Sea Mironova et al. 2009 M.velox 40 18 - 12 planktonic solar saltern of the Yellow Sea, China Lei et al. 2009 Phototrophic species M. rubrum 20 - - 1134 planktonic solar saltern of the Yellow Sea, China Lei et al. 2009 25-30 20-25 - - planktonic Yellow Sea, China Zhang et al. 2002 38 - 9.3 - planktonic pond in L'Houmeau, near La Rochelle, Dupuy et al.2007 French Atlantic coast 30 - - 180 (a) planktonic Wester basin of the Mediterranean Sea Dolan & Marrasé 1995 - - - 1.0x106 (a) planktonic Southampton Water and Test and Itchen Crawford et al. 1997 , England M. rubrum large 33 28 - 402-8794 planktonic open northen Baltic Sea proper Johansson et al. 2004 M. .rubrum small 17 14 - 1257-9487 planktonic open northen Baltic Sea proper Johansson et al. 2004 M. rubrum - - - 50-3.7x104 planktonic estuary of the Gulf of Main, USA Sanders 1995 - - - - planktonic Nervión River estuary, Bay of bisca, Spain Urrutxurtu et al. 2003 (a) maximum concentration

31 Introduction

The symbiont containing M. rubrum forms regularly blooms in many areas i.e. coastal, upwelling zones around the world (e.g. Stoecker et al.1989, Lindholm 1985, Wilkerson & Grunseich 1990, Williams 1996, Table 2), and thus has received much more attention than other marine Mesodinium spp (Crawford et al. 1997).

Mesodinium rubrum is generally present throughout the water column down to 200 m (i.e. Lindholm 1985, Dolan & Marrasé 1995), however it can form dense patches in certain parts of the water column, which is probably due to that fact that M. rubrum exhibits a strong phototactic response (e.g. Lindholm 1985, Fenchel & Hansen 2006). The swimming behaviour of M. rubrum is quite different from most other ciliates. Quick jumps are often succeeding periods where the ciliate does not move. During jumps M. rubrum can travel at a speed exceeding 5 mm s-1 (Lindholm 1985) and during jumps M. rubrum can reach a speed up to 1.2 cm s-1(Fenchel & Hansen 2006). Mesodinium rubrum usually blooms in surface waters, however blooms have been found subsurface (e.g. Lindholm 1985, Owen et al. 1992). The factors related to subsurface bloom formation may be availability of nutrients or a negative phototaxic response of M. rubrum to very high light intensities which can be found in surface waters (Fenchel & Hansen 2006, Stoecker et al. 2009). Surprisingly, Montagnes et al. (2008) found no relationship between M. rubrum abundance and suspected prey (cryptophyte) in the open waters of the North Atlantic, but instead found a correlation between abundance and temperature in the spring.

Most quantitative data on M. pulex come from freshwaters, i.e. lakes, reservoirs, where it can be found in sediments and in ground waters as well (Barbieri&Godinho-Orlandi 1989, Wiackowski et al. 2001, Zingel et al. 2007,Andrushchyshyn et al. 2007). Very few investigations have related M. pulex distributions to biotic or abiotic factors, but for example, Gomes & Godinho (2003) found high abundances of M. pulex in a eutrophic lake near the sediment, characterized by low dissolved levels. Also, it was found that M. pulex can be positively correlated to concentrations in highly eutrophic shallow lakes, like in the Lake Köyliönjärvi (SW Finland) (Wiackowski et al. 2001). Although Mesodinium pulex is a common ciliate in marine waters, quantitative data are rare and mostly derive from investigation of ciliate associated with the sediment. Reported cell densities range from 345-6250 cells l-1 in the solar saltern (10-30 cm deep) of the Yellow Sea (Lei et al. 2009) to I 5x103 to ≥3.00x105 cells l-1 in sandy sediments of an estuarine intertidal flat (Table 2, Hamels et al. 2005).

32 Introduction

Part 3.3. The role of symbionts and phagotrophy in marine Mesodinium rubrum and Mesodinium pulex

During the 1970’s it was shown that Mesodinium rubrum contains cryptophyte chloroplasts, , cryptophyte mitochondria, cryptophyte cytoplasma and a so-called symbiont nucleus (Taylor et al. 1971, Hibberd 1977). Later, molecular data revealed that this symbiont is very similar to (or identical) free-living cryptophytes belonging to the Teleaulax/Geminigera clade and it was suggested that the ciliate got its chloroplasts and symbiont nuclei from ingested prey (e.g. Johnson & Stoecker 2005, Johnson et al. 2006). However, about the same time it was shown that M. rubrum can divide its chloroplasts and at least in some isolates (Danish) also its symbiont nucleus (Hansen & Fenchel 2006, Garcia- Cuetos et al. 2010). At present we do not know if the different isolates of M. rubrum in fact are the same species, but the different results obtained with the Danish and Antarctic strains may suggest that they are not the same species. Nevertheless, there is an ongoing debate whether or not the symbionts in M. rubrum are permanent or indeed replaceable (i.e. Gustafson et al. 2000, Hansen & Fenchel 2006, Johnson et al. 2007, paper III).

Photosynthesis is the main source of the nutrition of all studied M. rubrum strains (e.g. Stoecker et al. 1991, Smith & Hansen 2007) and the ingestion of just 1 cryptophyte cell per day is enough for M. rubrum in order to maintain its maximum growth rate. Besides this, the symbiont allows M. rubrum to survive for extended periods (months) of time in the light when subjected to sudden starvation (Johnson & Stoecker 2005, Smith & Hansen 2007). The role of light for the non-symbiont containing Mesodinium species has never been investigated. It is possible that the often so-called heterotrophic Mesodinium species in fact do carry out some photosynthesis, which might increase their growth rate in the light or allow them to survive longer in the light when starved. In Paper II, this was investigated for the first time using M. pulex. It turned out that photosynthetic rates of M. pulex were quite low and never exceeded 4% of the total carbon requirements of M. pulex.

Phagotrophy is important for both M. rubrum and M. pulex although it plays quite different roles in the two species (i.e. Johnson & Stoecker 2005, Smith & Hansen 2007, Paper II, Paper III). Mesodinium rubrum is able to feed on a variety of prey items, i.e. cryptophytes, dinoflagellates and bacteria (Gustafson et al. 2000, Yih et al. 2004, Myung et al. 2006, Smith & Hansen 2007, Park et al. 2007, Paper III). However so far, M. rubrum has

33 Introduction only been maintained successfully when fed cryptophytes, belonging to the Teleaulax/Geminigera clade (i.e. Danish isolate: Teleaulax amphioxeia, Teleaulax sp., Korean isolate:Teleaulax spp. and Antarctic isolate: Geminigera cryophila, Table 4 in Paper III). Maximum growth rates of M. rubrum are obtained at very low prey concentrations (1000 prey cells ml-1) and at very low food uptake rates (1-2% of total carbon uptake; Smith & Hansen 2007). Why M rubrum has to feed it still not fully resolved. Either, Mesodinium rubrum needs to ingest a “growth factor”, which it only can get from the ingestion of specific species of cryptophytes, or it needs to replace its symbionts by eating specific cryptophyte species from time to time (Paper III).

Marine Mesodinium pulex gets most of its nutrition from phagotrophy (Paper II), and thus maximum ingestion rates of M. pulex are far higher than that of M. rubrum (10-20 times; Table 3, Smith & Hansen 2007, Paper II, Paper III). Like in the case of M. rubrum, M. pulex ingests many different types of preys, including cryptophytes, dinoflagellates and ciliates (Dolan &Coats 1991, Jakobsen et al. 2006, Paper II). Maximum growth rates of Mesodinium pulex are higher than those of the symbiont containing M. rubrum, when incubated in the sufficient light and prey concentrations (Smith & Hansen 2007, Paper II). However, M. rubrum survives much better periods of starvation than M. pulex does. All Mesodininum species catch their prey by the use of tentacles which are equipped with mucocysts. Despite this, M.rubrum and M.pulex displayed quite different preferences for prey species (Paper II, Paper III).

Table 3 Comparison of ingestion rates and growth rates between M.rubrum and M.pulex when fed on different prey M.rubrum M.pulex Prey species Ingestion rate ± SE Growth rate ± SE Ingestion rate ± SE Growth rate ± SE (cells predator-1 d-1) (d-1) (cells predator-1 d-1) (d-1) Heterocapsa rotundata 0.84±0.11 0.08±0.04a 14.93±0.85 1.13±0.03 theta 0.43±0.01 0.37±0.01 6.51±2.80 -0.10±0.29 tepida 0.47±0.02 0.32±0.01 4.21±2.63 -0.37±0.03 Monoculture of Predator - 0.11±0.04a - -0.65±0.09 - 0.31±0.02 - - a : the same experiment

34 Introduction

Interestingly, light seems to affect ingestion and growth rates of the marine non- symbiont containing M. pulex, which is unexpected for so-called heterotrophic protists (Paper II). Most studies of functional and numerical responses of heterotrophic protists have been carried out in dim light or in the dark (e.g. Verity 1991, Montagnes et al. 1996, Strom & Morello 1998, Jeong et al.1999, Gismervik 2005). When supplied with sufficient prey, the maximum ingestion rate of M. pulex was 49 prey cells predator-1 d-1 in the light, while it in the dark the maximum ingestion rate was considerably lower (27 prey cells predator-1 d-1). As a consequence the maximum growth rate of M. pulex in the light was 1.41±0.03 d-1, as compared to 1.19±0.07 d-1 in the dark (Paper II). Effects of light on ingestion and growth may be due to both direct effects of light on digestion rates and to indirect effects due to light effects on food quality (Skovgaard 1998, Strom 2001). The experiments carried out on M. pulex (Paper II) were not designed to differentiate between direct and indirect effects of light. However, the starvation experiments gave some indication that it may well be direct light effects we observed. To our surprise we found higher mortality rates of M. pulex in the light than in the dark (Paper II). This is in contrast to previous studies on heterotrophic protists which showed no effect of light on survival rates when they were starved (Skovgaard 1998, Strom 2001). It is well known that M. pulex can ingest prey of its own size (Dolan & Coats 1991) and recently, Moestrup et al. (in prep) found cannibalism in marine M. pulex. Therefore if M. pulex has higher ingestion rates in the light than in the dark, this may explain the higher mortality rates in the light compared to in the dark.

35 Introduction

Conclusion and future perspectives

Protists which mix phototrophy and phagotrophy (mixotrophs) are common and important members of the marine plankton around the world. They occur in small numbers in oligotrophic tropical offshore waters as well as in coastal eutrophic waters where they sometimes form blooms. They potentially play great role both as primary producers and as grazers on bacteria and other protists. This thesis focuses on distribution and biology of protists which form symbiosis with phototrophic protists and bacteria. I have mainly worked with two groups little studied groups: 1) Dinoflagellates which bear ecto and endo-symbionts and 2) Ciliates with cryptophyte endosymbionts

Quite a few genera belonging to the class Dinophysiales contain species which have either ectosymbionts (cyanobacterial) or intact endosymbionts (cyanobacterial or ). This study revealed that the species diversity and cell concentrations were highest in the in offshore warm waters characterized by low nitrogen concentrations and low turnover rates. They seem to live where most algae cannot make a living. We found evidence for that the dinoflagellates bearing ectosymbionts ate their symbionts (ectosymbionts) as well as other prey (ciliates). Besides this we found nifH genes associated N2 fixation in the consortia. The big surprise was that these genes were not associated with the cyanobacteria, but rather with heterotrophic bacteria, which were attached to the dinoflagellates. This group of dinoflagellates has still not been cultured and we do not know much their biology. Questions like: 1) how fast do these consortia grow, 2) how much of the carbon need of the dinoflagellate come from the ectosymbionts, 3) what are the rates N-fixation and how important is this for the success of the consortia, remains to be answered. We also found some dinoflagellates (Amphisolenia spp) which had endosymbionts. The functional importance of these endosymbionts to the dinoflagellates is completely unknown.

The ciliate genus Mesodinium contains species with cryptophyte endosymbionts as well as species without symbionts. The study on the symbiont containing species Mesodinium rubrum dealt with to what extent it can replace its symbionts or not. Despite feeding it with different cryptophytes belonging to all major marine cryptophyte clades, we found no proof of the ability of this ciliate to be able to exchange its symbionts. It would feed and digest all the offered preys, but it could not sustain growth of any of them, except for when fed a Telaulax

36 Introduction clade species. So either it can only exchange symbionts within this clade or it needs a growth factor which it can only get by preying on Teleaulax clade species. Future studies should test if it can exchange symbionts via ingestion of prey cells by offering it different species (which can be separated genetically) within the Teleaulax clade.

I also studied the non symbiont containing Mesodinium species, M. pulex, in order to test if photosynthesis is of any importance to such species. It turned out the photosynthesis plays a little role for its carbon metabolism. Instead, to my surprise I found that ingestion rates and growth rates were indeed increased in the light as compared to in the dark. This finding has potential huge implications for our understanding on how light influences digestion rates in large heterotrophic protists. If this is a general phenomenon, most published rates of heterotrophic protists may have been underestimated a great deal.

Over the past decades, the studies on the functional biology of mixotophic protists have revealed that mixotrophy is common among marine protists. However, due to methodological problems in quantifying food uptake in natural communities, the role of mixotrophs in marine food webs is not well understood. New promising tools like the acidotropic stains, which in principle allow for detection of food vacuoles in protists in general, have recently been developed. Combined with modern flow cytometry and molecular tools, it may for the first time allow us to proper quantify food uptake in mixotrophs in natural communities.

37 Introduction

References

Andrushchyshyn OP, Wilson KP, Williams DD (2007) Ciliate communities in shallow groundwater: seasonal and spatial characteristics. 52: 1745-1761

Arenovski AL, Lim EL, Caron DA (1995) Mixotrophic nanoplankton in oligotrophic surface waters of the Sargasso Sea may employ phagotrophy to obtain major nutrients. J Plankton Res 17:801-820

Barbieri SM, Godinho-Orlandi MJL (1989) Planktonic protozoa in the tropical reservoir: Temporal variations in abundance and composition. Rev Hydrobiol trop 22:275-285

Bass D, Brown N, Mackenzie-Dodds J, Dyal P, Nierzwicki-Bauer SA, Vepritskiy AA, Richards TA (2009) A molecular perspective on ecological differentiation and biogeography of cyclotrichiid ciliates. J Eukaryot Microbiol 56:559-567

Berge T, Hansen PJ, Moestrup Ø (2008a) Feeding mechanism, prey specificity and growth in light and dark of the plastidic dinoflagellate Karlodinium armiger. Aquat Microb Ecol 50:279-288

Berge T, Hansen PJ, Moestrup Ø (2008b) Prey size spectrum and bioenergetics of the mixotrophic dinoflagellate Karlodinium armiger. Aquat Microb Ecol 50:289-299

Bouvier T, Becquevort S, Lancelot C (1998) Biomass and feeding activity of phagotrophic mixotrophs in the northwestern Black Sea during the summer 1995. Hydobiologia 363:289- 301

Bowers HA, Brutemark A, Carvalho WF, Granéli E (2010) Combining flow cytometry and real-time PCR methodology to demonstrate consumption by Prymnesium parvum1. J Am Wat Res Assoc 46:133-143

38 Introduction

Breitbarth E, Oschlies A,LaRoche J (2006) Physiological constraints on the global distribution of —effect of temperature on diazotrophy. Biogeosciences- Discussions 3:779-801

Burkholder JM, Glibert PM, Skelton HM (2008) Mixotrophy, a major mode of nutrition for harmful algal species in eutrophic waters. Harmful Algae 8:77-93.

Capone DG (2000) The marine nitrogen . In: Kirchman DL (ed) of the Ocean. Wiley-Liss, New York pp, 455-493

Caron DA (2000) Symbiosis and mixotrophy among pelagic organisms. In: Kirchman DL (ed) Microbial ecology of the . Wiley-Liss, New York, p 495-523

Carvalho WF, Granéli E (2006) Acidotropic probes and flow cytometry: a powerful combination for detecting phagotrophy in mixotrophic and heterotrophic protists. Aquat Microb Ecol 44:85-96

Carvalho WF, Minnhagen S, Granéli E (2008) Dinophysis novergica (), more a predator than a producer? Harmful Algae. 7:174-183

Chan YF, Tsai AN, Chiang KP, Hsieh C (2009) Pigmented nanoflagellates grazing on Synechococcus: Seasonal variations and effect of size in the coastal of subtropical western Pacific. Microb Ecol 58:548-557

Crawford DW (1989) Mesodinium rubrum: the phytoplankter that wasn't. Mar Ecol Prog Ser 58: 161-174

Crawford DW, Purdie DA, Lockwood APM, Weissman P (1997) Recurrent red- in the Southampton Water estuary caused by the phototrophic ciliate Mesodinium rubrum. Estuarine 45:799-812

Dolan JR, Coats DW (1991) A study of feeding in predacious ciliates using prey ciliates labeled with fluorescent microspheres. J Plankton Res 13:609-627

39 Introduction

Dolan JR, Pérez MT (2000) Costs, benefits and characteristics of mixotrophy in marine oligotrichs. Freshwater Biol 45:227-238

Dolan JR, Marrasé C (1995) Planktonic ciliate distribution relative to a : Catalan Sea, N.W. Mediterranean, June 1993. Deep-Sea Res Part I 42:1965-1987

Dupuy C, Pastoureaud A, Ryckaert M, Sauriau PG, Montanié H (2000) Impact of the oyster Crassostrea gigas on the microbial community in Atlantic coastal near La Rochelle. Aquat Microb Ecol 22:227-242

Dupuy C, Ryckaert M, LeGall S, Hartmann HJ (2007) Seasonal variations in planktonic community structure and production in an Atlantic Coastal pond: The importance of nanoflagellates. Mirob Ecol 53:537-548

Eriksen NT, Hayes KC, Lewitus AJ (2002) Growth responses of the mixotrophic dinoflagellates, Cryptoperidiniopsis sp. and , to light under prey-saturated conditions. Harmful Algae 1:191-203

Fenchel T, Hansen PJ (2006) Swimming behaviour in Mesodinium rubrum. Mar Biol Res 2: 33-40

Foissner W, Berger H, Schaumburg J (1999) Identification and ecology of limnetic plankton ciliates. - Informationsberichte des Bayerischen Landesamtes für Wasserwirtschaft, 3/99: 1- 793

Foster RA, Carpenter EJ, Bergman B (2006a) Unicellular cyanobacterial ectosymbionts in open ocean dinoflagellates, radiolarians, and : ultrastructural characterization and immuno-localization of phycoerythrin and nitrogenase. J Phycol 42:453-463

Foster RA, Collier JL, Carpenter EJ (2006b) Reverse PCR amplification of cyanobacterial symbiont 16S rRNA sequences from single non-photosynthetic eukaryotic marine planktonic host cells. J Phycol 42:243-250

40 Introduction

Garcia-Cuetos L, Moestrup Ø, Hansen PJ, Daugbjerg N (2010) The toxic dinoflagellate Dinophysis acuminata harbors permanent chloroplasts of origin, not kleptochloroplasts. Harmful Algae 9: 25-38

Gismervik I (2005) Numerical and functional response of choreo- and oligotrich planktonic ciliates. Aquat Microb Ecol 40:163-173

Gomes EAT, Godinho MJL (2003) Structure of the protozooplankton community in a tropical shallow and eutrophic lake in Brazil. Acta Oecol 24:S153-S161

Gordon N, Angel DL, Neori A, Kress N, Kimor B (1994) Heterotrophic dinoflagellates with symbiotic cyanobacteria and nitrogen limitation in the Gulf of Aqaba. Mar Ecol Prog Ser 107:83-88

Gustafson JrDE, Stoecker DK, Johnson MD, Van Heukelem WF, Sneider K (2000) Cryptophyte algae are robbed of their by the marine ciliate Mesodinium rubrum. 405: 1049-1052

Gómez F (2005) Histioneis (Dinophysiales, Dinophyceae) from the western Pacific Ocean. Bot Mar 48: 421-425

Hamels I, Muylaert K, Sabbe K, Vyverman W (2005) Contrasting dynamics of ciliate communities in sandy and silty sediments of an estuarine intertidal flat. Eur J Protistol 41:241-250

Hansen PJ (1991) Dinophysis – a planktonic dinoflagellate genus which can act both as a prey and a predator of a ciliate. Mar Ecol Prog Ser 69:201-204

Hansen PJ, Calado A (1999) Phagotrophic Mechanisms and Prey Selection in Free-living Dinoflagellates. J Eukaryot Microb 46: 382-389

Hansen PJ, Fenchel T (2006) The bloom-forming ciliate Mesodinium rubrum harbors a single permanent endosymbiont. Mar Biol Res 2:169-177

41 Introduction

Hansen PJ, Hjorth M (2002) Growth and grazing responses of the prymnesiophyte Chrysochromulina ericina: the role of irradiance, prey concentration and pH. Mar Biol 141: 975-983

Hansen PJ, Miranda L, Azanza R (2004) Green Noctiluca Scintillans: a dinoflagellate with its own greenhouse. Mar Ecol Prog Ser 275:79-87

Hibberd DJ (1977) Observation on ultrastructure of crypotmonad endosymbiont of red-water ciliate Mesodinium rubrum. J Mar Biol Assoc UK 57:45-61

Jakobsen HH, Hansen PJ, Larsen J (2000) Growth and grazing responses of two chloroplast- retaining dinoflagellates: effect of irradiance and prey species. Mar Ecol Prog Ser 201:121- 128

Jakobsen HH, Everett LM, Strom SL (2006) Hydromechanical signaling between the ciliate Mesodinium pulex and motile protist prey. Aquat Microb Ecol 44:197-206

Janson S (2004) Molecular evidence that in the -producing dinoflagellate genus Dinophysis originate from the free-living cryptophyte Teleaulax amphioxeia. Environmental 6:1102-1106

Janson S, Carpenter EJ, Bergman B (1995) Immunolabelling of phycoerythrin, ribulose 1,5- bisphosphate carboxylase/oxygenase and nitrogenase in the unicellular cyanobiont of Ornithocercus spp. (Dinophyceae). Phycologia 34:171-176

Jeong HJ (1999) The ecological roles of heterotrophic dinoflagellates in marine planktonic community. J Eukaryot Microbiol 46:390-396

Jeong HJ, Shim JH, Lee CW, Kim JS, Koh SM (1999) Growth and grazing rates of the marine planktonic ciliate Strombidinopsis sp. on red-tide and toxic dinoflagellate. J Eukaryot Microbiol 46: 69-76

Johansson M,Gorokhova E, Larsson U (2004) Annual variability in ciliate community structure, potential prey and predators in the open Baltic Sea proper. J Plankton Res 26: 67-80

42 Introduction

Johnson MD, Stoecker DK (2005) The role of feeding in growth and the photophysiology of Myrionecta rubra. Aquat Microb Ecol 39:303-312

Johnson MD, Tengs T, Oldach D, Stoecker DK (2006) Sequestration, performance, and functional control of cryptophyte plastids in the ciliate Myrionecta rubra (Ciliophora). J Phycol 42:1235-1246

Johnson MD, Oldach D, Delwiche CF, Stoecker DK (2007) Retention of transcriptionally active cryptophyte nuclei by the ciliate Myrionecta rubra. Nature 445:426-428

Jones HLJ (1997) A classification of mixotrophic protists based on their behaviour. Freshw Biol 37:35-43

Jones RI (1994) Mixotrophy in planktonic protists as a spectrum of nutritional strategies. Mar Microb Food Webs 8:87-96

Jonsson PR (1987) Photosynthetic assimilation of inorganic carbon in marine oligotrich ciliates (Ciliophora, Oligotrichina). Mar Microb Food Webs 2:55-68

Jyothibabu R, Madhu NV, Maheswara PA, Asha-Devi CR, Balasbramanian T, Nair KKC, Achuthankutty CT (2006) Environmentally-related seasonal variation in symbiotic associations of heterotrophic dinoflagellates with cyanobacteria in the western Bay of Bengal. Symbiosis 42:51-58

Karl D, Michaels A, Bergman B, Capone D, Carpenter E, Letelier R, Lipschultz F, Paerl H, Sigman D, Stal L (2002) Dinitrogen fixation in the world's oceans. Biogeochem 57/58: 47-98

Kim S, Kang YG, Kim HS, Yih W, Coats DW, Park M G (2008) Growth and grazing responses of the mixotrophic dinoflagellate Dinophysis acuminata as functions of light intensity and prey concentration. Aquat Microb Ecol 51:301–10

Kofoid CA, Skogsberg T (1928) The dinoflagellateseae.- Mem. Mus. Comp. Zool. Harv. 51: 1-766, 31pl

43 Introduction

Koike K, Sekiguchi H, Kobiyama A, Takishita K, Kawachi M, Koike K, Ogata T (2005) A novel type of kleptoplastidy in Dinophysis (Dinophyceae): presence of a haptophyte-type in Dinophysis mitra. Protist 156:225-237

Larsen J (1988) An ultrastructural study of Amphidinium poecilochroum (Dinophyceae), a phagotrophic dinoflagellate feeding on small species of cryptophytes. Phycologia 27:366-377

Laval-Peuto M, Rassoulzadegan F (1988) Autofluorescence of marine planktonic Oligotrichina and other ciliates. Hydrobiologia 159:99-110

Legrand C, Granéli E, Carlsson P (1998) Induced phagotrophy in the photosynthetic dinoflagellate Heterocapsa triquetra. Aquat Microb Ecol 15:65-75

Lei Y, Xu K, Choi JK, Hong HP, Wickham SA (2009) Community structure and seasonal dynamics of planktonics ciliates along salinity gradients. Europ J Protistol 45:305-319

Lessard E, Swift E (1986) Dinoflagellates from the North Atlantic classified as phototrophic or heterotrophic by epifluorescence microscopy. J Plankton Res 8:1209-1215

Lewitus AJ, Glasgow HB, Burkholder JM (1999) Kleptoplastidy in the toxic dinoflagellate Pfiesteria piscicida (Dinophyceae)1. J phycol 35:303-312

Li A, Stoecker DK, Adolf JE (1999) Feeding, pigmentation, photosynthesis and growth of the mixotrophic dinoflagellate Gyrodinium galatheanum. Aquat Microb Ecol 19:163-76

Li A, Stoecker DK, Coats DW (2000) Mixotrophy in Gyrodinium galatheanum (Dinophyceae): grazing responses to light intensity and inorganic nutrients. J Phycol 36:33-45

Lindholm T (1985) Mesodinium rubrum - a unique photosynthetic ciliate. Adv Aqvat Microbiol 3:1-48

Lucas IAN (1991) Symbionts of the tropical Dinophysiales (Dinophyceae). Ophelia 33:213- 224

44 Introduction

Lynn DH (2008) The Ciliated Protozoa - Characterization, Classification, and Guide to the Literature, Springer Science, 605 p

Madoni P (2006) Benthic ciliates in Adriatic Sea . Europ J Protistol 42:165-173

Mazei YA, Burkovsky IV (2005) Species composition of benthic ciliate community in the Chernaya River estuary (Kandalaksha Bay, White Sea) with a total checklist of the White Sea benthic ciliate fauna. 4:107-120

Minnhagen S, Janson S (2006) Genetic analyses of Dinophysis spp. support kleptoplastidy. FEMS Microbiol Ecol 57:47-54

Mironova EI, Telesh IV, Skarlato SO (2009) Planktonic Ciliates of the Baltic Sea (a Review). Inland Water Biology 2:13-24

Montagnes D, Allen J, Brown L, Bulit C, Davidson R, Díaz-Ávalos C, Fielding S, Heath M, Holliday NP, Rasmussen J, Sanders R, Waniek JJ, Wilson D (2008) Factors controlling the abundance and size distribution of the phototrophic ciliate Myrionecta rubra in open waters of the North Atlantic. J Eukaryot Microbiol 55:457-465

Montagnes DJS, Berger JD, Taylor FJR (1996) Growth rate of the marine planktonic ciliate Strombidinopsis cheshiri Snyder and Ohman as a function of food concentration and interclonal variability. J Exp Mar Biol Ecol 206:121-132

Myung G,Yih W, Kim HS, Park JS, Cho BC (2006) Ingestion of bacterial cells by the marine photosynthetic ciliate Myrionecta rubra. Aquat Microb Ecol 44:175-180

Nagai S, Nishitani G, Tomaru Y, Sakiyama S, Kamiyana T(2008) by the toxic dinoflagellate Dinophysis fortii on the ciliate Myrionecta rubra and observations on the sequestration of ciliate chloroplasts. J Phycol 44:909-922

Nishitani G, Nagai S, Sakiyama S, Kamiyama T (2008a) Successful cultivation of the toxic dinoflagellate Dinophysis caudata (Dinophyceae). Plankton Res 3:78-85

45 Introduction

Nishitani G, Nagai S, Takano Y, Sakiyama S, Baba K, Kamiyama T (2008b) Growth characteristics and phylogenetic analysis of the marine dinoflagellate Dinophysis infundibulus (Dinophyceae). Aquat Microb Ecol 52:209-221

Noland LE (1937) Observations on marine ciliates of the gulf coast of Florida. Trans Am Microsc Soc 56:160-171

Norris RE (1967) Algal consortisms in marine plankton. In: Kirhnamurthy V(ed) Proceedings of a seminar on Sea, Salt and . Central Salt and Marine Chemicals Research Institute, Bharnagar, p178-189

Owen RW, Gianesella-Galvão SF, Kutner MBB (1992) Discrete, subsurface layers of the autotrophic ciliate Mesodinium rubrumoff Brazil. J Plankton Res 14: 97-105

Park MG, Kim S, Kim HS, Myung G, Kang YG, Yih W (2006) First successful culture of the marine dinoflagellate Dinophysis acuminata. Aquat Microb Ecol 45:101-106

Park JS, Myung G, Kim HS, Cho BC, Yih W (2007) Growth responses of the marine photosynthetic ciliate Myrionecta rubra to different cryptomonad strains. Aquat Microb Ecol 48:83-90

Park MY, Park JS, Kim M, Yih W (2008) Plastid dynamics during survival of Dinophysis caudata without its ciliate prey. J Phycol 44:1154-1163

Perriss SJ, Laybourn-Parry J, Marchant HJ (1995) Widespread occurrence of populations of the unique autotrophic ciliate Mesodinium rubrum (Ciliophora: Haptorida) in brackish and saline lakes of the Vestfold Hills (Eastern Antarctica). Polar Biology 15:423-428

Pitta P,Giannakourou A (2000) Planktonic ciliates in the oligotrophic Eastern Mediterranean: vertical , spatial distribution and mixotrophy. Mar Ecol Prog Ser 194:269-282

Ptacnik R, Sommer U, Hansen T, Martens V (2004) Effects of microzooplankton and mixotrophy in an experimental planktonic food web. Limnol Oceanogr 49: 1435-1445

46 Introduction

Putt M (1990a) Metabolism of photosynthate in the chloroplast-retaining ciliate Laboea strobila. Mar Ecol Prog Ser 60: 271-282

Putt M (1990b) Abundance, chlorophyll content and photosynthesis rates of ciliates in the Nordic Sea during summer. Deep-Sea Res Part I 37:1713-1731

Reddy KJ, Haskell JB, Sherman DM, Sherman LA (1993) Unicellular, aerobic nitrogen- fixing cyanobacteria of the genus Cyanothece. J Bacteriol 175:1284-1292

Riisgaard K, Hansen PJ (2009). Role of food uptake for photosynthesis, growth and survival of the mixotrophic dinoflagellate Dinophysis acuminata. Mar Ecol Prog Ser 381:51-62

Rose JM, Caron DA, Sieracki ME, Poulton N (2004) Counting heterotrophic nanoplankton protists in cultures and aquatic communities by flow cytometry. Aquat Micro Ecol 34:263- 277

Rothhaupt KO (1996) Utilization of substitutable carbon and phosphorus sources by the mixotrophic chrysophyte Ochromonas sp. Ecology 77:706-715

Safi KA, Hall JA (1999) Mixotrophic and heterotrophic nanoflagellate grazing in the convergence zone east of New Zealand. Aquat Microb Ecol 20:33-93

Saito H, Furuya K, Lirdwitayaprasit T (2006) Photoautotrophic growth of Noctiluca scintillans with an endosymbiont Pedinomonas noctilucae. Plankton Benthos Res. 1:97-101

Sanders RW (1995) Seasonal distributions of the photosynthesizing ciliates Laboea strobila and Myrionecta rubra (=Mesodinium rubrum) in an estuary of the Gulf of Maine. Aquat Microb Ecol 9:237-242

Sanders RW, Berninger UG, Lim EL, Kemp PF, Caron DA (2000) Heterotrophic and mixotrophic nanoplankton predation on picoplankton in the Sargasso Sea and on Georges Bank. Mar Ecol Prog Ser 192:103-18

47 Introduction

Schnepf E (1993) From prey via endosymbiont to plastid: comparative studies in dinoflagellates. In: Lewin RA (ed) Origins of Plastids: , Prochlorophytes, and the origins of chloroplasts. Chapman and Hall, New York, P 53-72

Schütt, F. 1895. Die Peridineen der Plankton-Expedition-Ergebn. Plankton-Expedition der Humboldt Stiftung 4: 1-170

Skovgaard A (1996) Mixotrophy in Fragilidium subglobosum (Dinophyceae): growth and grazing responses as functions of light intensity. Mar Ecol Prog Ser 143: 247-253

Skovgaard A (1998) Role of chloroplast retention in a marine dinoflagellate. Aquat Microb Ecol 15: 293-301

Skovgaard A (2000) A phagotrophically derivable growth factor in the plastidic dinoflagellate Gyrodinium resplendens (Dinophyceae). J Phycol 36: 1069-1078

Smalley GW, Coats DW (2002) Ecology of the dinoflagellate Ceratium furca: distribution, mixotrophy, and grazing impact on ciliate populations of Chesapeake Bay. J Eukaryot Microbiol 49:64-74

Smalley GW, Coats DW, Stoecker DK (2003) Feeding in the mixotrophic dinoflagellate Ceratium furca is influenced by intracellular nutrient concentrations. Mar Ecol Prog Ser 262:137-151

Smith M, Hansen PJ (2007) Interaction between Mesodinium rubrum and its prey: importance of irradiance, prey concentration, and pH. Mar Ecol Prog Ser 338:61-70

Spiller H, Shanmugam KT (1987) Physiological conditions for nitrogen fixation in a unicellular marine cyanobacterium, Synechococcus sp. strain SF1. J Bacteriol 169:5379-5384

Staal M, Meysman FJR, Stal LJ (2003) Temperature excludes N2-fixing heterocystous cyanobacteria in the tropical oceans. Nature 425:504-507

Steidinger KA, Tangen K (1997) Dinoflagellates. In: Tomas CR (ed) Identifying marine phytoplankton. Academic Press, California, p 387-584

48 Introduction

Stoecker DK (1998) Conceptual models of mixotrophy in planktonic protists and some ecological and evolutionary implications. Europ J Protistol 34:281-290.

Stoecker DE (1999) Mixotrophy among dinoflagellates. J Eukaryot Microbiol 24:397-401

Stoecker DK, Gustafson DE, Verity PG (1996) Micro- and mesoprotozooplankton at 140°W in the equatorial Pacific: and mixotrophs. Aquat Microb Ecol 10:273-282

Stoecker DK, Johnson MD, de Vargas C, Not F (2009) Acquired phototrophy in aquatic protists. Aquat Microb Ecol 57:279-310

Stoecker DK, Li A, Coats DW, Gustafson DE, Nannen MK (1997) Mixotrophy in the dinoflagellate Prorocentrum minimum. Mar Ecol Prog Ser 152:1-12

Stoecker DK, Putt M, Davis LH, Michaels AE (1991) Photosynthesis in Mesodinium rubrum: species-specific measurements and comparison to community rates. Mar Ecol Prog Ser 73: 245-252

Stoecker DK, Silver MW (1990) Replacement and aging chloroplasts in Strombidium capitatum (Ciliophora: Oligotrochia). Mar Biol 107:491-502

Stoecker DK, Silver MW, Michaels AE, Davis LH (1988) Obligate mixotrophy in Laboea strobila, a ciliate which retains chloroplasts. 99:415-423

Stoecker DK, Taniguchi A, Michaels AE (1989) Abundance of autotrophic, mixotrophic and heterotrophic planktonic ciliates in shelf and slope waters. Mar Ecol Prog Ser 50:241-254

Strom SL (2001) Light-aided digestion, grazing and growth in herbivorous protists. Aquat Microb Ecol 23:253-261

Strom SL, Morello TA (1998) Comparative growth rates and yields of ciliates and heterotrophic ciliates. J Plankton Res 20:571-584

49 Introduction

Tamar H (1986) Four marine species of Mesodinium (Ciliophora : Mesodiniidae) I. Mesodinium velox. Trans Am microsc Soc 105:130-140

Tamar H (1992) Four marine species of Mesodinium (Ciliophora : Mesodiniidae) II. Mesodinium pulex CLAP & LACHM, 1858 Arch Protistenkd 141:284-303

Taylor FJR (1976) Dinoflagellates from the International Indian Ocean Expedition. Bibliotheca Botanica 132:1-234

Taylor FJR (1982) Symbiosis in marine microplankton. Ann. Inst. Oceanogr. Paris 58: 61-90

Taylor FJR, Blackbourn DJ, Blackbourn J (1971) The red-water ciliate Mesodinium rubrum and its “incomplete symbionts”: a review including new ultrastructural observations. J Fish Res Brd Canada 28:391-407

Tsai AY, Chiang KP, Chan YF, Lin YC, Chang J (2007) Pigmented nanoflagellates in the western subtropical Pacific are important grazers on Synechococcus populations. 29:71-77

Unrein F, Gasol JM, Massana R (2010) Dinobryon faculiferum (Chrysophyta) in coastal Mediterranean seawater: presence and grazing impact on bacteria. J Plankton Res 32:559-564

Urrutxurtu I, Orive E, de la Sota A (2003) Seasonal dynamics of ciliated protozoa and their potential food in an eutrophic estuary (Bay of Biscay). Est Coast Shelf Sci 57:1169-1182

Verity PG (1991) Measurement and simulation of prey uptake by marine planktonic ciliates fed plastidic and aplastidic nanoplankton. Limnol Oceanogr 36: 729-750

Wiackowski K, Ventela AM, Moilanen M, Saarikari V, Vuorio K, Sarvala J (2001) What factors control planktonic ciliates during summer in a highly eutrophic lake? Hydrobiologia 443:43-57

Wilkerson FP, Grunseich G (1990) Formation of bloom by the symbiotic ciliate Mesodinium rubrum: the significance of nitrogen uptake. J Plankton Res 12:973-989

50 Introduction

Williams J (1996) Blooms of Mesodinium rubrum in Southampton Water. J Plankton Res 18:1685-1697

Yih W, Kim HS, Jeong H J, Myung G & Kim YG (2004) Ingestion of cryptophyte cells by the marine photosynthetic ciliate Mesodinium rubrum. Aquat Microb Ecol 36:165-170

Zhang W, Xu K, Wan R, Zhang G, Meng T, Xiao T, Wang R, Sun S, Choi JK(2002) Spatial distribution of ciliates, nauplii and eggs, Engraulis japonicus post-larvae and microzooplankton herbivorous activity in the Yellow Sea, China. Aquat Microb Ecol 27:249- 259

Zingel P, Agasild H, Noges T, Kisand V (2007) Ciliates are the dominant grazers on pico- and nanoplankton in a shallow, naturally highly eutrophic lake. Microb Ecol 53: 134-142

51

PAPER I

© Reprinted with kind permission from Inter-Research

PAPER II

Prey selection, ingestion and growth responses of the common marine ciliate Mesodinium pulex in the light and in the dark

Woraporn Tarangkoon1,2, Per Juel Hansen1*

1Marine Biological Laboratory, Strandpromenaden 5, DK-3000 Helsingør, Denmark 2Faculty of Science and Fisheries Technology, Rajamangala University of Technology Srivijaya, 92150 Trang, Thailand

*Corresponding author. E-mail [email protected]

ABSTRACT

The ciliate Mesodinium pulex (Class Litostomatea) ingested all 5 species of cryptophytes and the autotrophic dinoflagellate Heterocapsa rotundata offered as prey. Despite this, it could only grow on the cryptophytes Teleaulax sp. and Guinardia theta and the dinoflagellate H. rotundata, because ingestion rates of the other prey cells, even at very high prey concentrations, were too low to support growth. The numerical and functional responses of Mesodinium pulex fed Heterocapsa rotundata were investigated in the laboratory in the light (100 μmol photons m-2 s-1) and in the dark. In the light the growth rate was significantly higher than in the dark at all prey concentrations. Active photosynthesis was measured in M. pulex and the measured rates could not explain the increased growth rates in the light. Instead, the increased growth rates could mainly be explained by elevated ingestion rates in the light. We also studied the starvation response at different irradiances and in the dark (100, 50 μmol photons m-2 s-1 and dark). M. pulex survived for up to 2 weeks without food, but mortality rates of in the light were larger than in the dark.

Key words: Mesodinium pulex, growth, ingestion, ciliate Tarangkoon & Hansen

INTRODUCTION

The ciliate genus Mesodinium (Stein 1863) is a small genus belonging to the Order Cyclotrichiida (Class Litostomatea; Lynn 2008). Presently, the genus includes the red M. rubrum (Lohmann 1908) Hamburger & Buddenbrock 1911 (=Myrionecta rubra), which forms symbiosis with cryptophytes belonging to the Teleaulax /Geminigera clade (i.e. Garcia-Cuetos et al. 2010) and 6 species without symbionts: M. pulex Claparède & Lachmann 1858,1859, M. acarus Stein 1867, M. fimbriatum Stokes 1887, M. cinctum Calkins 1902, M. pupula Kahl 1933, and M. velox Tamar 1986.

Mesodinium acarus, M. fimbriatum and M. pulex have all been reported from freshwater, but at least M. acarus and M. pulex also occurs in brackish and marine waters (i.e. Foissner et al 1999). The remaining species are commonly found in brackish and marine waters (Lindholm 1985, Foissner et al. 1999). Recently, Bass et al. (2009) reported that clones identified as M. pulex fall into two different phylogenetic clades, indicating the existence of additional undescribed species.

Our knowledge of which diet supports the growth of Mesodinium spp derives mainly from studies of M. rubrum and M. pulex. Mesodinium rubrum has been shown to ingest many different species of cryptophytes (Gustafson et al. 2000, Yih et al. 2004, Park et al. 2007, Hansen et al. in prep). However, it can also feed on other types of prey (i.e a dinoflagellate; Hansen et al. in prep). Despite this, successful cultures of Mesodinium rubrum have so far only been established on cryptophytes within the Teleaulax clade as prey (i.e Teleaulax amphioxeia, T. acuta and Geminigera cryophila; Gustafson et al. 2000, Yih et al. 2004, Hansen & Fenchel 2006, Park et al. 2007, Hansen et al in prep). Mesodinium pulex has so far been reported to feed on the dinoflagellate Heterocapsa rotundata and the ciliate Metanophrys sp and successful cultures have been established on both types of prey (Dolan & Coats 1991, Jakobsen & Strom 2004, Jakobsen et al. 2006). It has been shown to be a very inefficient grazer on the cryptophyte Rhodomonas salina or the dinoflagellate Gymnodinium simplex, mainly due a very low successful capture rate. Also it will not ingest artificial beads (Dolan & Coats 1991, Carrias et al. 1996). A recent study has suggested that M. rubrum also ingests bacteria (Myung et al. 2006);

1 Tarangkoon & Hansen however the capture mechanism of the small sized bacteria is unknown at this stage and it is currently unknown if M. rubrum can sustain growth on bacteria.

Detailed physiological studies have only been carried out on temperate strains and an antarctic strain of Mesodinium rubrum (Yih et al. 2004, Johnson & Stoecker 2005, Hansen & Fenchel 2006, Park et al. 2007, Smith & Hansen 2007). This species, which forms symbiosis with cryptophytes, is predominantly phototrophic and can only grow in the light. Maximum growth rates are obtained at very low prey concentrations (1000 prey cells ml-1) and at very low food uptake rates (1-2% of total carbon uptake; Smith & Hansen 2007). When subjected to sudden starvation, M. rubrum will perform 3-4 cell divisions and to a large extent replicate its cryptophyte chloroplasts and at least in some isolates also its symbiont nucleus (Hansen & Fenchel 2006, Johnson et al. 2007, Smith & Hansen 2007). Thereafter it is able to starve for up to a month depending on the incubation temperature. Very little is known about the physiology of the non-symbiont containing species, and their photosynthetic potential has never been studied (Jakobsen 2006). However, prior to this study we observed that M. pulex cells often are filled with autotrophic food even after the prey has been depleted, indicating possible active photosynthesis.

The aim of the present work was to study the prey selection and feeding physiology of a non-symbiont containing Mesodinium species. We chose M. pulex because it is a very common species around the world and because it is the only species which at the moment can be held in laboratory culture. We hypothesized that: 1) Mesodinium pulex is an omnivorous feeder that relies on its ability to sense and successfully catch prey cells. 2) Mesodinium pulex can carry out photosynthesis in the light and that this carbon uptake in light will increase its growth rate or make it survive better when suddenly starved.

MATERIALS AND METHODS

Culture of organisms. Mesodinium pulex was provided by the culture collection of the Marine Biological Laboratory (Helsingør, Denmark). It had originally been isolated from water samples collected from a boat launch near Shannon Point Marine Center, Washington, USA by H.H.

2 Tarangkoon & Hansen

Jakobsen (Jakobsen et al. 2006). Cultures of M. pulex were maintained on the dinoflagellate Heterocapsa rotundata unless otherwise stated. The organisms were grown at a temperature of 20+1 0C in enriched filtered (0.22µm) f/20 seawater medium (Guillard 1983) at salinity 30+1 psu without silicate. Light was provided by cool-white fluorescent light at a light:dark cycle of 14:10h with an illumination of ca. 100 µmol photons m-2 s-1. Other prey items (cryptophytes, Table1) used in some experiments were grown in enriched filtered h/20 seawater media (modified from medium ‘h/2’ Guillard 1975), otherwise the growth conditions were as stated above. All cultures were non-.

Table 1. Algae used in the experiments, their cell volumes, estimated spherical diameters (ESD; n=20), and culture identification.

Cell volume±SE ESD Species (taxa) Culture collection –ID number (µm3) (µm) Heterocapsa rotundata 142.2±12.8 6.3 SCCAP-K0441 (Dinophyceae) Teleaulax amphioxeia 127.4±9.5 6.1 SCCAP () Teleaulax sp. 105.3±2.9 5.8 MBL - 1 (Cryptophyceae) Guillardia theta 100.7±4.5 5.6 CCMP-2712 (Cryptophyceae) Hemiselmis rufescens 72.4±4.3 5.0 CCMP-440 (Cryptophyceae) Hemiselmis tepida 58.2±4.8 4.7 CCMP-442 (Cryptophyceae)

Cell volume. Cell length and width of preserved organisms were measured in a Sedgewick- Rafter chamber under an inverted at 400x magnification. Suitable geometrical forms of organisms were used and the cell volumes were calculated according to Hillebrand et al. (1999). The following geometrical forms were used: A cone with hemisphere was used for M.

3 Tarangkoon & Hansen pulex, H. rotundata, Teleaulax sp. and T. amphioxeia, while the shape of polate spheroid was used for G. theta, H. rufescens, H. tepida. The ESD (Equivalent Spherical Diameter) was estimated by the equation: ESD = (biovolume/0.523)0.33. Carbon contents of M. pulex and the prey organisms were estimated with the carbon conversion factor: pg C cell-1 = 0.216 x volume0.939 and pg C cell-1 = 0.760 x volume0.819, respectively (Menden-Deuer&Lessard 2000).

Feeding and growth of M. pulex fed different prey items. This experiment was designed to test whether M. pulex was able to feed and grow on a variety of cryptophytes and a dinoflagellate (Table 1). For comparison we ran control experiments with M. pulex in monocultures (thus exposed to starvation). Preliminary experiments had shown that some of the cryptophytes cannot grow on the f/20 medium, because they cannot use nitrate as an inorganic N-source. We therefore switched to the medium “h/20”, which is basically the same as f/20, except that it also includes (Guillard 1975). Prior to the initiation of the experiments, M. pulex had been grown on H. rotundata and allowed to deplete H. rotundata as prey (= residual prey concentrations less than ca < 5 cells ml-1). This experiment was carried out at an irradiance of 100 µmol photons m-2 s-1.

Preliminarily, the initial predator-prey concentration ratio was set at ca. 25:1250 cells ml- 1, to ensure sufficient prey in mixed culture during the experiment period (4 days). However, it turned out that M. pulex could not control the cryptophyte prey populations. We therefore used an initial predator-prey concentration ratio of ca. 200:1000 cells ml-1 (1:5 ratio) on Day 0. Monocultures of cryptophytes were initiated at a concentration of ca. 1000 cells ml-1. Subsamples (5-10ml) were withdrawn at Day 0, 2 and 4 from triplicates of 65-ml culture bottles and cells counted. After subsampling on Day 2, all experimental bottles were refilled to capacity with fresh filtered h/20 medium. In cases where prey populations had been almost depleted during the first 2 days of the experiment, additional prey was added to the experimental bottles. Only the experiments in which M. pulex displayed growth between Day 2 and 4 were allowed to continue to Day 6. pH was measured in all experimental bottles during experimental period with a Sentron® pH meter (model 2001) equipped with Red line probe, with a detection limit of 0.01 pH units. The pH meter probe was calibrated with Sentron buffers of 7 and 10.

4 Tarangkoon & Hansen

The growth/mortality rate of M. pulex ( μ y ) and prey cells ( μ x ) was calculated assuming exponential growth/mortality:

(ln N1 − ln N 0) μ (d −1) = y,x t where N0 and N1 are cell number at start (t0) and at the end (t1) of each incubation experiment, respectively, and t is the interval time experiment (d).

The ingestion rate U is the per capita ingestion rate, which is dependent on prey (x). This was estimated using the following 2 equations. dx = μ −Uy dt x dy = μ y dt y

Ingestion rate of M. pulex was calculated assuming that predator (y) and prey (x) grow exponentially with rate constant of μ y and μx , respectively. The decrease in prey due to grazing isUy . This ingestion rate was calculated by using software as described in Jakobsen & Hansen (1997).

Rates of growth and ingestion of M. pulex were fitted to the Michelis-Menten equation, respectively, using the software SigmaPlot 10 (Systat Software, Inc) : μ (x − x ) μ(d −1) = max 0 K m + (x − x0 )

Where μmax is the maximal growth rate of M. pulex, x is the actual prey concentration, x0 is the threshold prey concentration for growth (where μ y = 0) and Km is the prey concentration sustaining ½ μmax ,and: U (x) U (d −1) = max K m + (x)

5 Tarangkoon & Hansen

-1 -1 Where U max ís the maximal ingestion rate per M. pulex (prey cells predator d ), x is the prey

-1 concentration (cells ml ) and Km is the prey concentration sustaining ½ U max For statistic analysis, t-test analyses were used to test for differences found in rates between dark and light treatments. T-test and ANOVA were used to compare means (ingestion rate and growth rate) between each prey to zero value and between each other, respectively.

Photosynthetic performance of Mesodinium pulex and the prey Heterocapsa rotundata. The photosynthetic performance of Mesodinium pulex cultures when fed the dinoflagellate Heterocapsa rotundata was measured at an irradiance of 100 µmol photons m-2s-1. The photosynthetic performance was measured in both well fed M. pulex cultures and in cultures which had almost depleted their prey. In the first case, M. pulex cells was separated from its prey by individually picking the ciliates with a drawn Pasteur pipette and transferring them to 0.2 µm filtered growth medium. This was repeated 3 times to exclude all H. rotundata. In the second case, M. pulex cultures were allowed to almost deplete their prey, before bulk measurements of the mixed cultures were carried out (Table 2). Subsamples were taken for enumeration of prey and M. pulex cells. This allowed for subtraction of the contribution of H. rotundata to the total photosynthesis and thus an estimation of the photosynthetic performance due to M. pulex cells.

Photosynthetic rates were measured by a modification of the “single cell method” 14 - (Stoecker et al. 1988, Skovgaard et al. 2000). A NaH CO3 stock solution (specific activity 100 µCi ml-1) was then added to each vial, containing 2 ml cell suspension of resulting in a specific activity of ~ 1.0 µCi ml-1. The 2 ml cell suspension contained either 20 or 40 ciliates in the case where the cells were picked individually, while it was in the range of 400-1300 cells when the photosynthetic performance was carried out on the mixed cultures (Table 2). The vials were incubated on a glass shelf with light coming from beneath for 2 h. All measurements were carried out in triplicates. The vials were always accompanied by parallel dark vials, which were treated similarly, except that they were wrapped in aluminum foil during incubation.

After incubation, the specific activity of the medium was measured by transferring 100 µl from each vial to new vials containing 200 µl phenylethylamine. The remaining suspension received 2.0 ml of 10 % acetic acid in methanol to remove all inorganic C. The vials were dried

6 Tarangkoon & Hansen overnight at 60 °C and the residue was then re-dissolved in 1.5 ml distilled water. Finally, 10 ml of Packard Insta-Gel Plus scintillation flour were added to all vials (including those for specific activity) and activity was measured using a Packard 1500 Tri-Carb liquid scintillation counter. Calculations of photosynthetic rates were based on Parsons et al. (1984). Total dissolved inorganic carbon content was measured with a 225-Mk3 infrared gas analyzer (Analytic Development Co. Ltd. Hoddesdon, England).

Functional and numerical response in the light and in the dark. These experiments were designed to test the effect of light on growth and ingestion rates of M. pulex. Heterocapsa rotundata was selected as prey based on the prey selection experiments. Experiments were initiated by mixing exponentially growing cultures of M. pulex and H. rotundata. Cultures of H. rotundata were also run as monocultures, thereby allowing the calculation of ingestion rates. All experiments were carried out in triplicate in 65 ml tissue culture bottles filled to capacity. The experiments were carried out in darkness (culture bottles were wrapped in aluminum foil), and at an irradiance of 100 µmol photons m-2 s-1. Irradiance was measured using a LI-COR, LI-1000 sensor equipped with a spherical probe. The initial prey concentrations during these experiments were approximately 450-18,200 cells ml-1 and a prey:predator cell concentration ratio of > 10 (based on preliminary data). The initial two days of the incubation served as acclimation to light and prey concentration. Subsamples, 5-10 ml, were taken for cell counts every 1-2 days, depending on growth rates, during experimental period (3-6 days) at a fixed time of day to eliminate potential diurnal variations in ingestion and growth rate of M. pulex. Subsamples were fixed in Lugol’s (final concentration 1%) and counted on inverted microscope at 100 x magnification. A Sedgewick-Rafter cell was used for counting when cell densities were above 200 cells ml-1. Otherwise 2 or sometimes 25 ml sedimentation chambers were used in cases where cell numbers were lower. To avoid pH effects on the growth of H. rotundata and M. pulex, pH was monitored during all the experiments with high prey concentrations (above 8000 cells ml-1) directly in bottles. Subsequently, the experimental bottles were refilled to capacity with fresh filtered f/20 media. In order to keep prey concentrations at a certain concentration and to avoid pH effects dilution of the experimental bottles and/or addition of prey cells were often required.

7 Tarangkoon & Hansen

Responses of M. pulex to starvation in the light and in the dark. This experiment was designed to examine whether or not light affects growth/survival responses of M. pulex when starved. This experiment was conducted in darkness, 50 and 100 µmol photons m-2 s-1 for 18-21 days. Initial predator-prey concentration ratio was ca 200:2800 cells ml-1. Monocultures of H. rotundata were also run parallel with all mixed cultures. The experiments were adapted to each irradiance level for 2 days (Day 0-2). Subsamples (5-10ml) were collected at Day 2, 3, 4 and thereafter every 2 days until the termination of each experiment and the experimental bottles were filled to capacity with fresh filtered f/20 media. The mortality rate was calculated in the same way as growth (see previous section). For comparison between light treatments, t-test analyses were used to test for differences between means. Mean values were averaged using only data after Day 4 (= no residual prey in all treatments).

RESULTS

Feeding and growth of M. pulex fed different prey items

The culture of M. pulex subjected to starvation decreased in cell concentration throughout the duration of the experiments and estimated mortality rates were in the range to 0.85-1.07 d-1 after Day 2 (Fig. 1A, 2A). In the experiment where M. pulex was fed H. rotundata at an initial prey concentration of 1000 cells ml-1, the ciliate almost depleted the prey during the first 2 days of the incubation and the negative growth of the ciliate was observed (-0.02 d-1) (Fig. 1B). Additional prey was therefore added on Day 2, thereby increasing the H. rotundata cell concentration to 10,000 cells ml-1. On Day 4, the prey populations had been considerably reduced, and M. pulex had grown with an average rate of 0.78 d-1 during this period (Fig 2A). Additional prey was not added on Day 4, which lead to a total depletion of prey on Day 6, and a decrease in M. pulex concentration.

The response of M. pulex to the cryptophytes was quite different than to that of H. rotundata. Mesodinium pulex ingested all of the offered species, however, it was unable to control the growth of any of them (Fig. 1, 2). Highest ingestion rates were obtained with

8 Tarangkoon & Hansen

Teleaulax sp and Guillardia theta as prey (22-25 and 12-15 cells predator-1 d-1, respectively), while lower ingestion rates were obtained with the other prey species (2-10 cells predator-1 d-1) (Fig 2B). Taking the different cell volumes of the individual preys (Table 1), allow for a calculation of ingested biovolume in each case (Fig. 2C). From this it is evident that the difference in growth rates on the different preys are reflected in different levels of ingestion rates (in terms of biovolume).

The highest growth rate of M. pulex in this experiment was obtained on Teleaulax sp. (0.8-1.0 d-1). However, this growth rate was obtained at a much higher cell concentration than when M. pulex was fed H. rotundata (Fig.1, 2). Mesodinium pulex did also grow well when fed G. theta, although growth rates were lower (0.5 d-1). In cases where M. pulex were fed Teleaulax amphioxeia, H. rufescens or H. tepida, M. pulex either maintained its population (on T. amphoxeia) or died out during the incubation even though cell concentrations became high (H. rufescens and H. tepida). Overall, it seems like M. pulex needs to ingest a biovolume of >800- 1000 µm3 d-1 to support growth.

9 Tarangkoon & Hansen

Fig. 1. Cumulative growth (corrected for dilutions) of M. pulex when fed on different prey: (A) no prey, (B) H. rotundata, (C) H. rufescens, (D)H. tepida, (E) G. theta, (F) Teleaulax sp., (G) T. amphioxeia. Dash line indicates addition of prey and arrow indicates half volume dilution. Dot line represents the speculated of died out H.rotundata. Symbols represent treatment means± 1SE

10 Tarangkoon & Hansen

Fig. 2. Mesodinium pulex. (A) Growth rate of M. pulex fed on different prey items (B) Ingestion rate of M. pulex (C) Ingestion rate in terms of biovolume of M. pulex. Black bars represent the measurement data between Day 2 to Day 4. Bars represent treatment means± 1SD.Grey bars represent the measurement data between Day 4 to Day 6. Asterisk indicates that the experiment was ended on Day 4.

11 Tarangkoon & Hansen

The pH of the medium was monitored in both monocultures and in the mixed cultures throughout the duration of the experiments (Fig. 3). In the mixed cultures it remained below pH 8.1, except for in the mixed culture with Teleaulax sp. as prey. Here it increased to pH 8.6 at the end of the experiment. In the algal monoculture, pH increased to reach high pH values in the cases of Teleaulax sp. and Heterocapsa rotundata (pH 9). This coincides with the entry into stationary growth observed in the monocultures of these species between Day 4 and 6 (Fig.1 B, F).

Fig. 3. pH dynamics during the experiment of M. pulex fed different prey types (A) monocultures of prey items ,(B) monocultures of M. pulex and mixed cultures of prey and predator. Symbols represent treatment means ± 1SE.

Photosynthetic performance of Mesodinium pulex and the prey Heterocapsa rotundata

The photosynthetic performance of Mesodinium pulex was 2.3 ± 0.86pg C cell-1 h-1 (average ± SEM) when fed the dinoflagellate Heterocapsa rotundata in excess at an irradiance of 100 µmol photons m-2s-1 (Table 2). This is equivalent to a daily rate of ca. 32 pg C cell-1 (14:10 light dark period). In M. pulex cultures that have just about depleted the prey in mixed cultures that

12 Tarangkoon & Hansen photosynthetic performance was much lower, 0.37 ± 0.08 pg C cell-1 h-1 (average ± SEM) (Table 2). This is equivalent to ca. 5.2 pg C cell-1 d-1. For comparison, the photosynthetic performance of H. rotundata cells in monocultures were 3.76 ± 0.18 pg C cell-1 h-1 or 52 ± 2.7 pg C cell-1 d-1 (average ± SEM, n=21).

Table 2. Photosynthetic performance of well fed and starved M. pulex cells when fed H.rotundata as prey. n = number of replicates.

Range in average ciliate Range in average prey Photosynthetic rate Physiological state concentrations concentrations average±SEM n (cells ml-1) (cells ml-1) (pgC cell-1h-1) Starving cells 210-645 4-83 0.37±0.08 21 Well fed cells 131-441 3090-18266 2.32±0.86 12

Functional and numerical responses in the dark and in the light

The functional response of M. pulex fed H. rotundata was investigated in the light (100 μmol photons m-2 s-1) and in the dark (Fig. 4A). Maximum ingestion rate was 49 H. rotundata cells predator-1 d-1 in the light, while it in the dark the maximum ingestion rate was considerably lower, 27 H. rotundata cells predator-1 d-1 (t-test, p<0.01). Half saturation constants were 1210 and 1580 cells ml-1 for the light and dark treatments. Ingestion rates of ~35 and 15 H. rotundata cells predator-1 d-1 were sufficient to maintain maximum growth rate of M. pulex in the light and in the dark, respectively (Fig. 4A).

The growth rate of M. pulex fed H. rotundata followed Michaelis-Menten kinetics both in light and in the dark (Fig. 4B). In the light the growth rate of M. pulex reached the maximum growth rate of 1.41±0.03 d-1 at average prey concentration of ~6,700 cells ml-1. A prey concentration of ~40 cells ml-1 was required for maintenance (μ=0). In the dark M. pulex reached a maximum growth rate of 1.19±0.07 d-1, which was significantly (t-test, p<0.01) lower than the

13 Tarangkoon & Hansen maximum growth rate obtained in the light (Fig 4B). Maintenance (μ=0) in the dark was obtained at a prey concentration of ca.10 cells ml-1, which was also lower than that found in the light.

Response of M. pulex to starvation in the dark and at two light levels

The effect of starvation on survival and cell volume of M. pulex when exposed to irradiances of 100 and 50 μmol photons m-2 s-1 and to darkness was studied (Fig. 5). During acclimation period (Day 0-2), prey concentrations were high in all cases leading to initial high growth rates of M. pulex (Fig.5D). Prey was depleted at Day 3 and 4 in the light and in the dark, respectively, and M. pulex subjected to sudden starvation. Initial growth rates were 0.99, 0.95 and 0.71 d-1 (Day 0-2) for the light treatments (100 and 50 μmol photons m-2 s-1) and the dark, respectively. These initial growth rates changed to mortality rates of 0.55, 0.62 and 0.09 d-1 for the light treatments (100 and 50 μmol photons m-2 s-1) and the dark, respectively during Day 4-6 (Fig. 5D). After Day 6, the mortality rates stayed fairly constant over the next 10-14 days, maybe with the exception of a slightly increased mortality rate from 0.46 to 0.73 d-1 during Day 12-16 in the light treatments. From Day 6, the mortality rates of the light treatments were significantly higher than the dark (t-test, p<0.05). No different of the mortality rates between high light and medium light showed in the same period (t-test, p>0.05).

14 Tarangkoon & Hansen

Fig. 4. Mesodinium pulex. (A) Ingestion rate of M . pulex fed on H.rotundata at 2 irradiances (100 and 0 μmol photons m-2 s-1) as function of prey concentration (Cp), (B) Growth rate (μ) of M. pulex when fed on H. rotundata at 2 irradiances. Symbols represent treatment means± 1SE. The curves were fitted to Michaelis-Menten kinetics.μ = 1.44 x (Cp- 40)/(707.13+( Cp-40) for the high light and μ = 1.10 x (Cp- 10)/(1745.44+( Cp-10) for the dark incubated

The cell volumes of M. pulex decreased slightly from Day 2 to Day 6 of the starvation experiment in all treatments (Fig 5E). After Day 6 the cell volumes were constant. The observed changes were not significantly different among treatments (t-test, p>0.05).

Growth/survival experiments were also carried out on monocultures of H. rotundata in the 2 light levels and dark to follow the fate of the prey organisms in three treatments (Fig 5A-C). In the high and medium light treatments, H. rotundata grew at a rate of 1.11 and 0.90 d-1, respectively for the first 6 days of the incubation. In the dark treatment, H. rotundata maintained its cell concentration for the first 4 days of the incubation. However, at Day 6 no cells were left any more.

15 Tarangkoon & Hansen

Fig. 5. Mesodinium pulex. Cumulative growth (corrected for dilutions) of M. pulex at (A)100 μmol photon m-2 s-1, (B) 50 μmol photon m-2 s-1, (C) Dark, (D) Growth rate of M. pulex at 3 different irradiances, (E) Cell volume of M. pulex at different irradiances. Symbols represent treatment means± 1SE. Dot line represents the speculated of died out H. rotundata from the monoculture.

16 Tarangkoon & Hansen

DISCUSSION

Prey selection in Mesodinium pulex

All Mesodinium species are phagotrophic and they seem to rely on motile prey for food (i.e. Dolan & Coats 1991, Gustafson et al. 2000, Jakobsen et al. 2006). The motile prey is detected by a band of equatorial cirri encircling the cell by sensing hydrodynamic disturbances of the medium made by the swimming prey. Prey capture in M. pulex has been studied in some detail (Jakobsen et al. 2006). Upon detection Mesodinium pulex immobilizes the prey cell using the oral tentacles, which have attached to each tentacle tip (Lindholm et al. 1988, Dolan&Coats 1991, Tamar 1992, Jakobsen et al. 2006). The prey is then transported to the mouth by retraction of the tentacles (Jakobsen et al. 2006). The entire process takes typically less than 1/30 of a second from detection to ingestion of H. rotundata (Jakobsen et al. 2006).

In our study, M. pulex successfully ingested all the offered prey organisms, which were in the size range of 4.7-6.3 µm (estimated spherical diameter, ESD). However, the ingestion rates varied in terms of both cells-1 h-1 and as biovolume h-1, allowing it only to grow when fed 2 out of the 5 different preys offered. Previously, it has been shown that M. pulex cannot ingest algae like the haptophyte globosa (4 µm, single cells) or the cryptophyte Rhodomonas salina (ESD 7 µm) (Tang et al. 2001, Jakobsen et al. 2006). Also, quite low ingestion rates were obtained with the dinoflagellate Gymnodinium simplex (ESD 9 µM) as prey. This raises the question of what causes this differentiated ingestion rate of quite similarly-sized prey cells. Jakobsen et al. (2006) showed that fast swimming prey like the dinoflagellate Gymnodinium simplex will make deformation rates in the water of a magnitude that disguises its as a predator and the ciliate performs jumps upon encountering it in the same way as it does when escaping an approaching predator. None of the cryptophytes studied in the present investigation were particularly fast swimming cells, so that explanation can be ruled out. However, since M. pulex feeds rheotactically on motile prey, swimming prey cells need to generate a sufficient deformation rate to trigger an attack from M. pulex. That is; slow swimming among prey cells may be a strategy aimed at reducing detection. This us with two possible mechanisms that can account for the low capture success of M. pulex on the cryptophytes. Either the ciliate is

17 Tarangkoon & Hansen unable to physically capture the prey cells or cells are able to a large extent to evade being caught by jumping away from the ciliate or by reducing its deformation rate due to slow swimming.

Physical capture of prey cells in M. pulex involves 3 overall steps: Prey immobilizing using the extrusomes, followed by adhesion of the prey to the tentacle tips, and finally ingestion of the prey (Jakobsen et al. 2006). At each of the capture steps there is chance that M. pulex will fail to ingest the prey. Jakobsen et al. (2006) demonstrated that M. pulex paralyzed the cryptophyte R. salina less efficiently than it did the dinoflagellate H. rotundata (Jakobsen et al. 2006), indicating that the initial attack is an important step in the capture success of M. pulex. However, attaching prey to the tentacles may involve a complex and often highly specific recognition process between prey surface lectines and glycoprotein binding sites on the predator that acts as prey discriminating component in itself as has been shown in other protists (Sakaguchi et al. 2002, Wotton et al. 2007). Our visual investigation in the inverted microscope did not suggest that the cryptophytes we offered to M. pulex could escape by jumping away from the ciliate upon attack. Hence, the low capture success is most likely a product of both the inability of the extrusomes to properly attack to the prey and an inefficient predator prey match of binding glycoproteins and prey lectines.

If this explanation is true, why can not the low capture success be compensated for by high concentrations of prey cells? Even though cryptophyte concentrations were very high in some cases, ingestion rates were still low. The reasons for this lie most probably in the Mesodinium spp capture mechanism. Mesodinium spp use tentacles equipped with extrusomes (Lindholm et al.1988). Extrusomes can most likely only be fired once, where after they will have to be replaced. In Mesodinium spp. this probably means that the tentacles will have to be withdrawn into the cell and reabsorbed when the extrusomes have been fired, and new ones synthesized. This will put a maximum limit to the number of capture attempts by the ciliate and may thus explain our results.

18 Tarangkoon & Hansen

Effects of light on growth and ingestion by Mesodinium pulex

Mesodinium pulex displayed considerably higher growth rates in the light (using a 14:10 light:dark cycle) compared to the dark, irrespective of prey concentration (Fig 4). This may suggest that it is mixotrophic and that it gains a considerable input in terms of carbon from photosynthesis. Rates of photosynthesis of M. pulex were ~ 32 pg C cell-1 d-1 in well-fed cultures, which is about ten times lower than the rates measured in the predominantly photosynthetic mixotroph Mesodinium rubrum using the same techniques (Hansen & Fenchel 2006). Using a growth efficiency of the carbon uptake due to photosynthesis of 50 % in phototrophs (Hansen et al. 2000), suggests that 16 pg C will be available for M. pulex growth per day. For comparison, a M. pulex cell contains ~210 pg C (Table 3). The maximum growth rate of M. pulex in the light was ~ 1.4 d-1 or 2 doublings per day (Fig. 4B). This means that the expected growth increase due to photosynthesis of a well-fed M. pulex culture is less than 4 %. Thus, the increased growth rates in M. pulex in the light, were by and large due to increased ingestion rates in the light (Fig. 4).

Table 3. Cell volumes, estimated spherical diameter and estimated cell carbon contents of M. pulex (well fed) and H. rotundata cultured in f/20 medium at an irradiance of 100 μmol photons m-2 s-1 and in the dark (n=20). Carbon content was estimated using the equations given by Menden-Deuer & Lessard (2000).

Irradiance Cell volume±SD Estimated Carbon content Species ESD(µm) (μmol photons m-2 s-1) (µm3) ±SD (pg C cell-1)

100 Mesodinium pulex 1510.93±445.22 13.7 208.30±57.93 Heterocapsa rotundata 116.94±41.55 5.9 37.21±2.43 Dark Mesodinium pulex 1431.93±413.46 13.5 198.12±53.59 Heterocapsa rotundata 89.44±21.35 5.4 29.96±7.37

The finding of increased ingestion and growth rates in M. pulex in the light is not unique among heterotrophic protists, although very little information is presently available. Elevated ingestion and growth rates in the light compared to in the dark (or very low irradiance levels; i.e. ≤ 20 µmol photons m-2s-1) have previously been shown for a heterotrophic dinoflagellate,

19 Tarangkoon & Hansen

Gymnodinium sp (Skovgaard 1998) and 2 ciliates: Coxliella sp. and Strombidinopsis acumicatum (Strom 2001).

Effects of light on ingestion and growth can be due to both direct effects of light on digestion rates and to indirect effects due to light effects on food quality. Our growth and grazing experiments were not designed to differentiate between direct and indirect light effects, such as prey quality, or prey carbon content etc. However, Strom (2001) found direct positive effect of light on ingestion rates of 2 ciliate species, which increased by a factor of 2-7 at all irradiances ≥20 µmol photons m-2s-1 using short term incubations (mins) and FLAs (fluorescently labeled algae) as prey. Thus, although very few data are available on light effects on ingestion and growth of heterotrophic ciliates, these data tend to suggest that our observations of increased ingestion and growth rates in the light at least partly can be explained by direct light effects on ingestion rates. What causes the increased ingestion rates in the light? Previous studies on the topic have suggested “light aided” digestion of prey (Skovgaard 1998, Strom 2001). This is not unlikely because it is well established that dissolved organic matter is photochemically degradable (reviewed by Moran & Zepp 1997). No data are however available on the effect of light on digestion rates in heterotrophic protists, apart from Strom (2001), who found a 40-fold increase in loss of food vacuoles (a proxy for digestion rate) of the heterotrophic dinoflagellate Noctiluca scintillans, in the light compared to in the dark, when fed algal prey. More research on this topic is required to see if this might be a general phenomenon among heterotrophic protists.

Survival response in the light and in the dark

Mesodinium pulex cultures, which were allowed to deplete their prey to very low cell concentrations, displayed very low rates of photosynthesis (Table 2). Also, when M. pulex was subjected to starvation, it survived better in the dark than in the two light treatments in our experiments. Thus, we can reject our hypothesis that photosynthesis in the light slows down mortalities compared to in the dark. Previous studies of survival responses of algivorous heterotrophic protists (2 ciliates and 1 dinoflagellate) have not been able to demonstrate any differences between light and dark treatments (Skovgaard 1998, Strom 2001). So why does M. pulex differ so much from the other species tested?

20 Tarangkoon & Hansen

Mesodinium pulex is known to ingest preys of its own size (i.e. the ciliate Metanophrys sp., Dolan & Coats 1991) and has very recently been shown to be cannibalistic (Moestrup et al. in prep). None of the species tested by Skovgaard (1998) and Strom (2001) have been reported to be cannibalistic. Thus, if M. pulex indeed eats its own kind when starved and if ingestion and digestion rates in the light are higher compared to those in the dark this may explain the higher mortality rates observed in the light. This question would be worth studying in much more detail.

CONCLUDING REMARKS

Many different groups of protists capture single prey items using a variety of capture mechanisms. Dinoflagellates often use capture filaments, while often use the haptonema for prey capture (Kawachi & Inouye 1995, Hansen & Calado 1999). The cyclotrichs, including Mesodinium spp, use extrusomes for prey capture. Our knowledge on the mechanisms of prey selection in many protist predators, which rely in single prey capture, is very limited. While some studies indicate chemical as well as mechanical cues for detection of prey items (Jacobson & Anderson 1986, Hansen & Calado 1999, Naustvoll 2000a,b, Jakobsen et al. 2006, Riisgaard & Hansen 2009), the role of capture success of such predator types is largely unexplored (e.g. Jakobsen et al. 2006). In Mesodinium pulex, we show that prey species with a similar size are ingested at very different rates, suggesting that the cell surface of the prey is important in avoiding being consumed by predators like Mesodinium. Much more work is needed on this topic and its significance in species succession, prey/predator and carbon flow in marine waters.

An unexpected result of our work was the apparent role of light for ingestion and growth of M. pulex. Our hypothesis of a significant role of photosynthesis for the growth and survival of M. pulex was rejected. Thus, our data instead support the suggestions by Skovgaard (1998) and Strom (2001) that light plays a direct role in the digestion of prey by heterotrophic protists. Most studies dealing with the growth and grazing rates of heterotrophic ciliates have been carried out in complete darkness or at very low levels of irradiance (< 25 µmol photons m-2s-1; Verity 1991, Montagnes et al. 1996, Jeong et al.1999, Gismervik 2005). The reasons for the incubation of such experiments at low or no light among most researchers have primarily been to reduce prey

21 Tarangkoon & Hansen growth rate during the experiments and the fact that light has not been considered an important factor for grazing of entirely heterotrophic protists. If light turns out to play an important role for digestion of prey by protists in general it may have considerable implications for the application of ingestion data from laboratory experiments to field abundances of protists aiming to estimate the role of protists in microbial food webs (e.g. Hansen et al. 1997). Thus, work on this topic is strongly urged.

Acknowledgements

We are indebted to Hans Henrik Jakobsen, Øjvind Moestrup and Lydia Garcia-Cuetos for comments and suggestions, which improved this paper significantly. We thank Hans Henrik Jakobsen for the use of this isolate of Mesodinium pulex. This study was supported by the Danish Research Council to Per Juel Hansen, grant no 272-06-0485, and a PhD grant from Rajamangala University of Technology Srivijaya, Thailand to Woraporn Tarangkoon.

LITERATURE CITED

Bass D, Brown N, Mackenzie-Dodds J, Dyal P, Nierzwicki-Bauer SA, Vepritskiy AA, Richards TA (2009) A molecular perspective on ecological differentiation and biogeography of cyclotrichiid ciliates. J Eukaryot Microbiol 56: 559-567

Calkins GN (1902) Marine protozoa from Woods Hole. Bull US Fish Commn 21 (year 1901): 415-468

Carrias JF, Amblard C, Bourdier G (1996) Protistan bacterivory in an oligomesotrophic lake: importance of attached ciliates and flagellates. Microbial Ecology 31:249-268

Claparède E, Lachmann J (1858) Études sur les infusoires et les rhizopodes. Mémoirs de l’Institut National Genevois, vol. 5, années 1857. Kessmann, Genève. 1-260 plus planches 1-13

22 Tarangkoon & Hansen

Claparède E, Lachmann J (1859) Études sur les infusoires et les rhizopodes. Mémoirs de l’Institut National Genevois, vol. 6, années 1858. pp. 261-482 plus planches 1-24. Kessmann, Genève.

Dolan JR, Coats DW (1991) A study of feeding in predacious ciliates using prey ciliates labeled with fluorescent microspheres. J Plankton Res 13:609–627

Foissner W, Berger H, Schaumburg J (1999) Identification and ecology of limnetic plankton ciliates. Informationsberichte des Bayer. Landesamtes für Wasserwirtschaft, Heft 3/99, 793 pp

Garcia-Cuetos, L, Moestrup,Ø., Hansen PJ, Daugbjerg, N (2010). The toxic dinoflagellate Dinophysis acuminata harbors permanent chloroplasts of cryptomonad origin, not kleptochloroplasts. Harmful Algae 9: 25-38

Gismervik I (2005) Numerical and functional response of choreo- and oligotrich planktonic ciliates. Aquat Microb Ecol 40:163–173

Guillard RRL (1975) Culture of phytoplankton for feeding marine . In: WL Smith WL and Chanley MH (eds) Culture of marine invertebrate . Plenum Pub., New York, p. 29-60

Guillard RRL (1983) Culture of phytoplankton for feeding invertebrate animals. In: Berg CJ (ed) Culture of marine . Hutchinson Ross, Stroudsberg, PA, p 123-128

Gustafson JrDE, Stoecker DK, Johnson MD, Van Heukelem WF, Sneider K (2000) Cryptophyte algae are robbed of their organelles by the marine ciliate Mesodinium rubrum. Nature. 405: 1049- 1052.

Hamburger C, Buddenbrock W von (1911) XIII. Nordische mit Anschluss der Tintinnoidea. Nordisches Plankton, Zool. Teil. Lipsius und Tischer, Kiel und Leipzig. 211 pp.

23 Tarangkoon & Hansen

Hansen PJ, Calado A (1999) Phagotrophic mechanisms and prey selection in free-living dinoflagellates. J Eukaryot Microb 46: 382-389

Hansen PJ, Fenchel T (2006) Mesodinium rubrum harbours a permanent endosymbiont. Marine Biology Research 2: 169-177

Hansen PJ, Moldrup M, Tarangkoon W, Garcia-Cuetos L, Moestrup Ø (in prep). Does the marine red tide ciliate Mesodinium rubrum have replaceable symbionts?

Hansen PJ, Hansen B, Bjørnsen PK (1997) grazing and growth: scaling within the size range 2mm to 2000 mm. Limnol & Oceanogr 42:687-704

Hansen PJ, Skovgaard A, Glud RN, Stoecker DK (2000) Physiology of the mixotrophic dinoflagellate Fragilidium subglobosum. II. Effects of time scale and prey concentration on photosynthetic performance. Mar Ecol Prog Ser 201: 137-146

Hillebrand H, Durselen CD, Kirschtel D, Pollingher U, Zohary T (1999) Biovolume calculation for pelagic and benthic . J Phycol 35:403–424

Jacobson DM, Anderson DM (1986) Thecate heterotrophic dinoflagellates: feeding behavior and mechanisms. J Phycol 22: 249–258

Jakobsen HH, Hansen PJ (1997). Prey size selection, growth and grazing responses of a small heterotrophic dinoflagellate Gymnodinium sp. and a ciliate Balanion comatum: a comparative study. Mar Ecol Prog Ser 158: 75-86

Jakobsen HH, Strom SL (2004) Circadian cycles in growth and feeding rates of heterotrophic protist plankton. Limnol Oceanogr 49: 1915-1922

Jakobsen HH, Everett LM, Strom SL (2006) Hydromechanical signaling between the ciliate Mesodinium pulex and motile protist prey. Aquat Microb Ecol 44: 197-206.

24 Tarangkoon & Hansen

Jeong HJ, Shim JH, Lee CW, Kim JS, Koh SM (1999) Growth and grazing rates of the marine planktonic ciliate Strombidinopsis sp. on red-tide and toxic dinoflagellate. J Eukaryot Microbiol 46: 69–76

Johnson MD, Stoecker DK (2005) Role of feeding in growth and photophysiology of Myrionecta rubra. Aquat Microb Ecol 39: 303–312

Johnson MD, Oldach D, Delwiche CF, Stoecker DK (2007) Retention of transcriptionally active cryptophyte nuclei by the ciliate Myrionecta rubra. Nature 445: 426-428

Kahl A (1935) Die Tierwelt Deutschlands und der angrenzenden Meeresteile. 30. Urtiere oder Protozoa. I. Wimpertiere oder Ciliata (). 4. Peritricha und Chonotricha, pp. 651-886. Jena, Gustav Fischer Verlag.

Kawachi M, Inouye I, Madea O, Chihara M (1991) The haptonema as a food-capturing device – observations on Chrysochromulina hirta (). Phycologia 30: 563-573

Lindholm T (1985) Mesodinium rubrum – a unique photosynthetic ciliate. Adv Aquat Microbiol 3:1-48.

Lindholm T, Lindroos P, Mörk AC (1988) Ultrastructure of the photosynthetic ciliate Mesodinium rubrum. BioSystems 21:141–149

Lohmann H (1908) Untersuchungen zur feststellung des vollständigen gehaltes des meeres an plankton. Wiss Meeresunters. Abt Kiel (N.F.) 10:129-370

Lynn DH (2008) The Ciliated Protozoa - Characterization, Classification, and Guide to the Literature, Springer Science, 605 p.

Menden-Deuer SM, Lessard EJ (2000) Carbon to volume relationships for dinoflagellates, and other protist plankton. Limnol Oceanogr 45 (3):569–579

25 Tarangkoon & Hansen

Moestrup Ø, Gracia-Cuetos L. Daugbjerg N, Hansen PJ (in prep) Studies on the ciliate genus Mesodinium II. Heterotrophic and red autotrophic marine species, with a discussion of the of the genus and description of Mesodinium minor comb. nov.

Montagnes DJS, Berger JD, Taylor FJR (1996) Growth rate of the marine planktonic ciliate Strombidinopsis cheshiri Snyder and Ohman as a function of food concentration and interclonal variability. J Exp Mar Biol Ecol 206:121-132

Moran MA, Zepp RG (1997) Role of photoreactions in the formation of biologically labile compounds from dissolved organic matter. Limnol Oceanogr. 42:1307–1316

Myung G,Yih W, Kim HS, Park JS, Cho BC (2006) Ingestion of bacterial cells by the marine photosynthetic ciliate Myrionecta rubra. Aquat Microb Ecol 44:175-180

Naustvoll LJ (2000a) Prey size spectra and food preferences in thecate heterotrophic dinoflagellates. Phycologia 39: 187-198

Naustvoll LJ (2000b) Prey size spectra and food preferences in naked heterotrophic dinoflagellates. Phycologia 39: 448-455

Park JS, Myung G, Kim HS, Cho BC, Yih W (2007) Growth responses of the marine photosynthetic ciliate Myrionecta rubra to different cryptomonad strains. Aquat Microb Ecol 48:83-90

Parsons TR, Maita Y, Lalli CM (1984) A Manual of Chemical and Biological Methods for Seawater Analysis. Pergamon Press, Oxford, 173 pp.

Riisgaard K, Hansen PJ (2009) Role of food uptake for photosynthesis, growth and survival of the mixotrophic dinoflagellate Dinophysis acuminata. Mar Ecol Prog Ser 381: 51-62

26 Tarangkoon & Hansen

Sakaguchi M, Murakami H, Suzaki T (2002) Involment of a 40-kDa glycoprotein in food recognition, prey capture, and induction of in the protozoon Actinophrys sol. Protist 152:33-41

Skovgaard A (1998) Role of chloroplast retention in a marine dinoflagellate. Aquat Microb Ecol 15:293-301

Skovgaard A, Hansen PJ, Stoecker DK (2000) Physiology of the mixotrophic dinoflagellate Fragilidium subglobosum. I. Effects of phagotrophy and irradiance on photosynthesis and carbon content. Mar Ecol Prog Ser 201:129-136

Smith M, Hansen PJ (2007) Interaction between Mesodinium rubrum and its prey: importance of irradiance, prey concentration, and pH. Mar Ecol Prog Ser 338: 61-70

Stein F (1863) Neue infusorienformen in der Ostsee. Amtliche berichte deutscher naturforscher und Ærzste in Karlsbad, 37. Versammlung in september 1862: 161-162, 165-166

Stein F (1867) Der Organismus der Infusionsthiere nach eigenen Forschungen. II. Abtheilung. 1. Darstellung der neuesten Forschungsergebnisse über Bau, Fortpflanzung und Entwickelung der Infusionsthiere. 2. Naturgeschichte der heterotriche Infusorien. Leipzig, Engelmann. III-VI and 1- 355 pp plus 16 Tafeln.

Stoecker DK, Silver MW, Michaels AE, Davis LH (1988) Obligate mixotrophy in Laboea strobila, a ciliate which retains chloroplasts. Mar Biol 99:415 423.

Stokes AC (1887) Notices of new American freshwater infusoria. Jl R Microsc Soc, year 1887:35-40

Strom SL (2001) Light-aided digestion, grazing and growth in herbivorous protists. Aquat Microb Ecol 23:253-261

Tamar H (1986) Four marine species of Mesodinium (Ciliophora : Mesodiniidae) I. Mesodinium velox. Transactions of the American microscopical Society 105:130-140

27 Tarangkoon & Hansen

Tamar H (1992) Four marine species of Mesodinium (Ciliophora : Mesodiniidae) II. Mesodinium pulex Clap & Lachm, 1858. Arch Protistenkd 141:284-303

Tang KW, Jakobsen HH, Visser AW (2001) Phaeocystis globosa (Prymnesiophyceae) and the planktonic food web: feeding, growth, and trophic interactions among grazers. Limnol Oceanogr 46:1860–1870

Verity PG (1991) Measurement and simulation of prey uptake by marine planktonic ciliates fed plastidic and aplastidic nanoplankton. Limnol Oceanogr 36: 729–750

Wootton EC, Zubkov MV, Jones DH, Jones RH, Martel CM, Thornton CA, Roberts EC (2007) Biochemical prey recognition by planktonic protozoa. Environm Microb 9:216-222

Yih W, Kim HS, Jeong H J, Myung G & Kim YG (2004) Ingestion of cryptophyte cells by the marine photosynthetic ciliate Mesodinium rubrum. Aquat Microb Ecol 36:165–170

28

PAPER III

Does the marine red tide ciliate Mesodinium rubrum have replaceable symbionts?

Hansen PJ1*, Moldrup M1, Tarangkoon W1,3, Garcia-Cuetos L2, Moestrup Ø2

1Marine Biological Laboratory, Department of Biology, University of Copenhagen, Strandpromenaden 5, DK-3000 Helsingør 2Phycology Laboratory, Department of Biology, University of Copenhagen, Øster Farimagsgade 2D, DK-1353 Copenhagen 3Faculty of Science and Fisheries Technology, Rajamangala University of Technology Srivijaya, 92150 Trang, Thailand

*Corresponding author: [email protected] KEY WORDS: Mesodinium rubrum, symbionts, ingestion, cryptophytes

ABSTRACT:

The red tide ciliate Mesodinium rubrum (=Myrionecta rubra) is known to contain a symbiont of cryptophyte origin. Molecular data have shown that the symbiont is very closely related or similar to free-living species belonging to the “Teleaulax clade”. This suggests that the symbiont of M. rubrum is either a temporary symbiont or a quite recently established symbiont. Here we present data from a number of experiments in which we tried to replace the symbionts in M. rubrum by supplying a number of different cryptophyte species belonging to different cryptophyte clades. Growth and ingestion rates of M. rubrum fed these cryptophytes were measured. In addition, cells of M. rubrum were analyzed for type of chloroplast using transmission electron microscopy and DNA sequences of the LSU. We found that M. rubrum ingested all the offered cryptophyte species, but it was unable to incorporate any of the offered prey species as symbionts. Also, M. rubrum can only sustain growth on cryptophyte species belonging to the “Teleaulax clade”. Hansen et al.

INTRODUCTION

Mesodinium rubrum is a ciliate which hosts a cryptophyte symbiont including cryptophyte chloroplasts, mitochondria, nucleomorph, and a so-called “symbiont nucleus” (Taylor et al. 1971, Hibberd 1977, Hansen & Fenchel 2006). Several isolates are now in culture from Antarctic (Gustafson et al. 2000) and temperate waters (e.g. Yih et al. 2004, Hansen & Fenchel 2006). To what extent they belong to the same species is presently unknown as some morphological and physiological differences exist between them. For example the Danish and the Antarctic strains differ in terms of chloroplast number and the occurrence and size of the symbiont nucleus. The Danish isolate has on average about 17-20 chloroplasts per cell (Hansen & Fenchel 2006), while only about 8 chloroplasts have been found in the Antarctic isolate (Gustafson et al. 2000, Johnson & Stoecker 2005, Johnson et al. 2006). Also, while the Danish isolate apparently has a permanent large symbiont nucleus (Hansen & Fenchel 2006), the Antarctic isolate may lose its symbiont nucleus but acquire it again (Johnson et al. 2007). Also, the nucleus of the Antarctic isolate seems to be able to grow in size. This suggests that the strains belong to different species, but it is currently unresolved. It has previously been shown that the Danish isolate of M. rubrum only needs to ingest 1 cryptophyte cell per day to grow at its maximum growth rate (Smith & Hansen 2007). In term of ingested carbon, the food uptake only accounts 1-2 % of the carbon requirements per day for M. rubrum. This raises the question why M. rubrum has to eat to sustain growth? One explanation could be that M. rubrum harbors chloroplasts of its own but needs to ingest prey to obtain some essential micronutrients. Another alternative could be a renewal of the chloroplast after it underwent several divisions within the host. This implies that the chloroplasts are replaceable and have to be acquired from prey cells for sustained growth. So far, the Korean and the Danish isolates have been cultured on Teleaulax spp, while the Antarctic isolate has been cultured on Geminigera cryophila. All these cryptophytes are closely related and belong to the same cryptophyte clade ( Hoef-Emden, 2008) Recently however, Park et al. (2007), found some growth of the Korean M. rubrum when fed a cryptophyte belonging to the “Rhodomonas” clade suggesting that M. rubrum might be able to utilize different types of cryptophytes. However, to what extent, M. rubrum can ingest and grow on a large variety of different kinds of cryptophytes is presently unknown.

2 Hansen et al.

The aim of present investigation was to study ingestion and growth responses of the Danish isolate of M. rubrum when offered 7 different species of cryptophytes belonging to different marine cryptophyte clades and 1 dinoflagellate species. Moreover, we tried to determine to what extent M. rubrum can replace its symbionts using ultrastructural and molecular techniques

MATERIALS AND METHODS

Cultures. A culture of the photosynthetic ciliate Mesodinium rubrum was established from single cells isolated from surface sea water samples collected in Frederikssund, Denmark, during a bloom event on April 17th 2007. Single cells were isolated from the samples using a drawn-out Pasteur micropipette. The isolates were washed by several transfers through sterile-filtered seawater (0.20 μm Satorius Minisart filter) before placing in multi-dish wells (3 cells per well), each containing 2 ml sterile filtered seawater-based h/20 (Guillard 1975) medium with a salinity of 30. The cryptophyte Teleaulax amphioxeia (K-0434) was used as prey species for M. rubrum and was established from seawater samples collected from the Øresund in March 1990, Denmark, and provided by the Scandinavian Culture Collection of Algae and Protozoa of the University of Copenhagen. Every 2-3 d, 1 drop of dilute T. amphioxeia suspension in sterile filtered h/20 medium was added to the wells containing M. rubrum. When M. rubrum was not able to consume all added T. amphioxeia, it was transferred to a new vial containing new sterile- filtered h/20 medium. After one month the wells contained dense suspensions of M. rubrum. All cultures were kept in h/20 medium at a salinity of 30 and kept on a glass table. Light (cool white, 100 µmol photons m-2 s-1) was provided from beneath in a light:dark cycle of 14:10. All experiments were performed at a temperature of 20±1°C.

3 Hansen et al.

Growth and survival responses of Mesodinium rubrum when fed 7 different cryptophytes and a dinoflagellate. Seven cryptophyte species and 1 dinoflagellate were tested as food for Mesodinium rubrum (Table 1). The selection covers the 4 marine clades found within the cryptophytes (Fig. 1). Information on the strains is listed in Table 1. All experiments were carried out in 65-ml tissue culture bottles filled to capacity. Samples (2ml) were withdrawn at Day 4, 8, 12 (or 10) and bottles were refilled to capacity with fresh h/20 medium. It has previously been shown that growth of M. rubrum is affected when pH exceeds 8.5 and that it dies when pH exceeds 8.8 (Smith & Hansen 2007). Thus, in order to interpret data correctly, pH was measured directly in the bottles on each sampling occasion, using a Sentron Argus pH meter equipped with a HOT-Fet line pH probe, which was calibrated using standard buffers of pH 7 and 10. Prior to all experiments, M. rubrum was grown on the cryptophyte T. amphioxeia. Only M. rubrum cultures that had just depleted the prey were used for experiments.

Table 1. Protist strains used as prey in the experiments. Species Culture collection - ID number GenBank accession number Heterocapsa rotundata K-0483 (SCCAP) NA vectensis CCMP-432 HM126534 Guillardia theta CCMP-2712 X57162 Hanusia phi CCMP-325 U53126 Hemiselmis tepida CCMP-442 HM126533 Proteomonas sulcata CCMP-321 HM126536 Rhodomonas salina K-1487 (SCCAP) HM126532 Teleaulax amphioxeia K-0434 (SCCAP) AJ421146

In the first set of experiments, the two cryptophytes T. amphioxeia and Proteomonas sulcata and the dinoflagellate Heterocapsa rotundata were used as prey for M. rubrum. Initial cell concentrations were 1000 cells ml-1 of both prey and predator. This prey concentration was chosen based on the previous observation that maximum growth of M. rubrum occurs during these conditions when fed T. amphioxeia. Moreover, it has been shown that M. rubrum ingests ~1 prey cell (T. amphioxeia) d-1 at this prey concentration (Smith & Hansen 2007), which means that M. rubrum should be able to control its prey if it ingests the offered prey. In cases where prey was depleted totally in the experimental bottles, M. rubrum from that particular experiment was subcultured (i.e. diluted) and prey concentrations were increased and/or sampling frequency

4 Hansen et al. increased. Control experiments were carried out with both a monoculture of unfed M. rubrum and monocultures of the given prey species. All experiments were carried out in triplicate.

In the second set of experiments, the growth responses of M. rubrum were studied when offered five species of cryptophytes: Chroomonas vectensis, Rhodomonas salina, Hanusia phi, Guillardia theta, Hemiselmis tepida. The setup of the experiment was as in the first set of experiments, except that all the mixed cultures were diluted on Day 4 to diminish effects of elevated pH on the outcome of the experiments.

Measurements of ingestion rates Ingestion rates of M. rubrum were determined from the decrease in prey concentrations over 2-4 day periods when comparing with the growth of the control cultures as described by Jakobsen & Hansen (1997). The ingestion rate U was estimated using the following 2 equations:

dx = µ −Uy (2) dt x dy = µ y (3) dt y where (x) is ingested by grazer (y). It is assumed that the grazer (y) grows exponentially with the rate constant of μy and that the prey (x) grows with the rate constant of μx. The mortality of the cryptophytes due to grazing is Uy, where U (cells predator-1d-1) is the per capita ingestion rate, which is independent of x. The ingestion rate (U) was iteratively calculated using “Prey” (by B. Vismann) software (Jakobsen & Hansen 1997).

5 Hansen et al.

Fig.1. Phylogeny based on nuclear SSU rDNA sequences (1572 bp) inferred from Bayesian analysis. Glaucocystis nostochinearum, Gloeochaete wittrockiana, Cyanophora paradoxa and Cyanoptyche gloeocystis constituted the outgroup. Branch support was obtained from Bayesian posterior probabilities and bootstrap (100 replicates) in maximum likelihood analyses. At internodes, posterior probabilities (1) are written first followed by bootstrap values (in percentage) from ML. (*) Highest possible posterior probability (1.0) and bootstrap value (100%). Species in bold face were used for the experiments in this study. 6 Hansen et al.

Transmission electron microscopy (TEM) A culture of Mesodium rubrum fed Hemiselmis tepida was mixed 1:1 with 4% glutaraldehyde in 0.2 M cacodylate buffer at pH 7.4 and containing 0.4 M sucrose. After 1 hour at 4 ºC, the cells were concentrated by centrifugation. Subsequently, they were rinsed 3 times in cold cacodylate buffer of decreasing sucrose content. Once rinsed, the material was post-fixed overnight in 2% osmium tetroxide in 0.2 M cacodylate buffer at pH 7.4 at 4 ºC. Before dehydration, the material was rinsed briefly in buffer. Each step of the dehydration lasted 20 min at 4 ºC in the following ethanol concentrations: 30%, 50%, 70%, 90% and 96%. The material was transferred to room temperature while in 96% ethanol and dehydration completed in two changes of absolute ethanol, 20 min in each change. Following two brief rinses in propylene oxide, the material was transferred to a 1:1 mixture of Spurr’s embedding mixture (Spurr) and propylene oxide and left uncovered overnight, followed by 5 hours in a fresh mixture of Spurr. The material was then moved to a new recipient and Spurr was added. Finally, it was polymerized at 70 ºC overnight. Sectioning was carried out on a Reichert Ultracut E ultramicrotome using a diamond knife. The sections were collected on slot grids (Rowley & Moran 1975) and stained for 15 min with 2% uranyl acetate in methanol, followed by Reynold’s lead citrate. The grids were examined in a JEM-1010 electron microscope (JEOL, Tokyo, Japan), fitted with a digital camera.

DNA extraction, cell isolation, PCR amplification, cloning and The DNA extractions of the cryptophytes were performed as previously described in Hansen et al. (2003) on the cryptophyte species used as prey during the present experiment (Table 1). Samples from the experimental flasks were fixed in acid Lugol at Day 4 for the experiments carried out with C. vectensis, G. theta, H. phi and H. tepida as well as Day12 for the two latter species. Lugol-fixed cells of M. rubrum were isolated using drawn Pasteur glass pipette under an Olympus inverted microscope CKX31 (Olympus, Tokyo, Japan) and washed at least three times in ddH2O under the inverted microscope to avoid any cryptophyte present in the fixed sample to be carried with the cell of interest. Finally, the five washed M. rubrum cells were transferred into a 0.2 ml PCR tube (StarLab, Ahrensburg, Germany) and kept frozen at -20 ºC until further processing.

7 Hansen et al.

PCR reactions were carried out in 50 µl volume. PCR amplifications of the nuclear SSU rDNA (nSSU rDNA) of the cryptophytes were performed as outlined in Hoef-Emden et al. 2002, with a combination of primers CrN1F and SSUBR . While, PCR amplifications of the nucleomorph LSU rDNA (nmLSU rDNA) were carried out as described in Garcia-Cuetos et al. (2010) with the primer combination nmLSUCr3F and D3B (Nunn et al. 1996). Prior to amplification, physical disruption was conducted using glass beads (Sigma-Aldrich, Gillingham, UK) to ensure cell disruption of the M. rubrum cells (Frommlet & Iglesias-Rodríguez 2008). A semi-nested PCR amplification was subsequently performed using nmLSUCr3F and D2C (Scholin et al. 1994) applying the same PCR profile. All PCR reactions were carried out on a MJ Research PTC-200 Peltier Thermal Cycler (MJ Research Inc, Waltham, MA, USA).

To discriminate between possible copies of the nucleomorph LSU present in M. rubrum, all amplifications were cloned with the TOPO TA Cloning Kit (Catalogue nr. K4500-01) from Invitrogen (Carlsbad, CA). Following plating, transformed clones were selected and the nucleomorph LSU was amplified, as described above.

All DNA fragments were purified using Nucleofast, following the manufacturer’s recommendations (Macherry-Nagel Inc., Bethlehem, Pennsylvania, USA). 500 ng PCR product was air-dried over night and sent to the sequencing service at Macrogen (Seoul, Korea) for determination in both directions using the same primers employed for amplification.

Alignments and phylogenetic analyses To determine the phylogenetic position of our cryptophyte isolates, a data set with numerous cryptophyte taxa, consisting of nuclear SSU rDNA sequences was analyzed. The sequences were first aligned using MAFFT 6.624 (Katoh & Toh 2008) and then improved manually using BioEdit 7.0.5 sequence alignment software (Hall 1999). The data set was composed of 103 sequences including four glaucocystophyte sequences as outgroup taxa based on previous phylogenetic studies based on non-coding genes (Bhattacharya et al. 1995, Hoef-Emden 2008).

8 Hansen et al.

A Bayesian method was used to infer phylogeny, using the program MrBayes v.3.2 (Huelsenbeck & Ronquist 2001). Two simultaneous Monte Carlo Markov chains (MCMC; Yang & Rannala 1997) were run from random trees for a total of 2,000,000 generations (metropolis- coupled MCMC). One of every 50 trees was sampled. AWTY (Wilgenbusch et al. 2004) was used to graphically evaluate the extent of the MCMC analysis. After excluding the first sampled trees categorized as the ‘‘burn-in period’’, a consensus tree was constructed using PAUP* 4.0.b10 software (Swofford 2002) based on 39.000 trees. Then, Modeltest (Posada & Crandall 1998), implemented in the PAUP* 4.0.b10 software (Swofford 2002), identified GTR model as the best. Using these settings, a tree was reconstructed with the online version of the PhyML software (Guidon & Gascuel 2003) available on the Montpellier bioinformatics platform at http://www.atgc-montpellier.fr/phyml using the maximum likelihood (ML) method (Felsenstein 1981). The reliability of internal branches was assessed using the bootstrap method with 100 replicates (Felsenstein 1985).

RESULTS

Growth response and food uptake of Mesodinium rubrum when fed Teleaulax amphioxeia as prey

Mesodinium rubrum grew initially in the unfed control experiments during the first 4-8 days of the incubation in both experiments (Figs 2A, 3A, Table 2). When M. rubrum was fed Teleaulax amphioxeia at a prey:predator ratio of 1:1 the growth of M. rubrum was not significantly different from the unfed control during the first 4 days of incubation. During these first 4 days the prey was completely depleted (Fig. 2B). When the M. rubrum culture was re-fed T. amphioxeia at a prey predator ratio of 5:1 on Day 4, growth increased considerably compared to the control during the subsequent 4 days of incubation (Fig. 2A), and the prey was again completely depleted (Fig. 2B). A second re-feeding of the M. rubrumon Day 8, in combination with a shortening of the incubation period to 2 days, resulted in a further increase in M. rubrum growth rate (Fig. 2A). However, T. amphioxeia was not completely depleted on Day 10 (Fig. 2B). The growth of T. amphioxeia in monoculture was also studied (Fig. 2D), allowing the calculation of crude

9 Hansen et al.

estimates of average ingestion rates for the periods 0-4 days, 4-8 days and 8-10 days. Calculated ingestion rates varied from 0.4 to 2.5 prey cells M. rubrum d-1, at average prey concentrations of 200-1000 cells ml-1 (see Table 2 or Figure 4 for details).

106 M. rubrum mono A T.amphioxeia-mix B ) M. rubrum fed T.amphioxeia A ) P.sulcatum-mix B -1 ) ) -1 -1 M. rubrum fed P.sulcatum -1 105 H.rotundata-mix M. rubrum fed H.rotundata

104

103 103

102 Cell concentration (cells ml Cell concentration (cells ml Cell concentration (cells ml Cell concentration (cells ml 101 024681012024681012

10.0 6 10 M. rubrum mono C D

) D M.rubrum/T.amp-mix

9.5 -1 M.rubrum/P.sul-mix 105 M.rubrum/H. rotun-mix 9.0 4 10

pH 8.5 pH 103 8.0 T. amphioxeia-mono 2 7.5 10 Psulcata-mono Cell concentration(cells ml H.rotundata-mono

7.0 101 024681012 024681012 Time (days) Time (days) TimeTime (days) (days)

Fig. 2. Experiment 1. A. Changes in cell concentrations of M. rubrum as a function of incubation time (day) in monoculture and when grown in mixed cultures with the cryptophytes Teleaulax amphioxeia, Proteomonas sulcata, and the dinoflagellate Heterocapsa rotundata. B. Changes in prey cell concentrations in the mixed cultures. Development of pH in the cultures (C) and changes in prey cell concentrations cells in monocultures (D).

10 Hansen et al.

Table 2. Growth rates (average +/- SE) of all prey species (except T. amphioxeia) and M. rubrum in monocultures and mixed cultures during the first 4 days of the incubations (based on data shown on Fig. 2&3). The growth rate of M. rubrum in monoculture was 0.11±0.04 d-1 in the first set of experiments and 0.31±0.02 d-1 in the second set. The number of replicates was 3 in all experiments (n=3). For all species, prey growth rate in the mixed culture was significantly lower than growth rates in monoculture (t-test, p<0.05). The growth rate of M. rubrum in monoculture was not significantly lower than in the growth rate in mixed cultures.

Growth rate (mean ± std) (µ), d-1 Species Prey Prey M. rubrum in monoculture in mixed culture in mixed culture Experiment 1 Heterocapsa rotundata 1.19±0.03 0.77±0.03 0.08±0.04 Proteomonas sulcata 0.91±0.02 0.84±0.01 0.11±0.04

Experiment 2 Chroomonas vectensis 0.82±0.03 0.60±0.03 0.29±0.05 Guillardia theta 1.00±0.01 0.67±0.01 0.37±0.01 Hanusia phi 0.86±0.03 0.49±0.04 0.38±0.02 Hemiselmis tepida 0.78±0.01 0.33±0.01 0.32±0.01 Rhodomonas salina 1.16±0.05 1.02±0.06 0.28±0.04

Growth responses and food uptake of Mesodinium rubrum when fed other prey items

Mesodinium rubrum could not sustain growth when fed the dinoflagellate Heterocapsa rotundata and the other cryptophytes species tested (Figs 2A&3A). Initial growth of M. rubrum in the mixed cultures containing these prey types was observed for the first 4-8 days in all cases, but the growth rate never exceeded the growth of M. rubrum in monoculture. In some cases, M. rubrum died in the mixed cultures between Day 4 and Day 12, depending upon prey species (Fig.3A). The of M. rubrum always coincided with an increase in pH above 9 (Figs 2C&3C).

In many cases the prey items grew fast in both algal monocultures and in the mixed cultures (Figs 2D&3 D). pH increased above 9 after 8-12 days of incubation, and growth of the algae decreased and in some cases stopped. A comparison of the growth response of the prey in monocultures and in the mixed cultures with M. rubrum during the first 4 days of the incubation, revealed that prey concentrations in the mixed cultures were always lower than in the monocultures (Table 2). The lower growth rate found in the mixed culture (Table 2) indicated that M. rubrum ingested all types of the offered prey items. Calculation of ingestion rates during

11 Hansen et al.

the first 4 days of the incubation gave ingestion rates around 0.25-0.8 prey cells M. rubrum d-1 at average prey concentrations of 3200-15300 cells ml-1 (see Table 3 and Fig. 4 for details). These are considerable lower ingestion rates than when T. amphioxeia was used as prey at lower concentrations (Table 3). For all species the estimated ingestion rates were significantly different from zero (p<0.05). It should be stressed that the prey concentration changed considerably in the experimental flasks, thus the estimated rates are crude average rates for the 2-4 day incubations, and should not to be regarded as precise measurements.

106 M.rubrum, mono C.vectensis, mix A B M.rubrum fed C.vectensis R.salina-mix M.rubrum fed R.salina H.phi-mix ) ) -1 -1 M.rubrum fed H.phi G.theta-mix 104 M.rubrum fed G.theta 5 H.tepida-mix M.rubrum fed H.tepida 10

104 103 Cell concentration (cells ml (cells concentration Cell Cell concentration (cells (cells ml concentration Cell

3 10 102 0246810121402468101214 Time (days)

10.5 106 C M.rubrum/C.vect-mix D 10.0 M.rubrum/R.sal-mix )

M.rubrum/H.phi-mix 1 - M.rubrum/G.theta-mix 5 9.5 M.rubrum/H.tepida-mix 10

9.0 pH 104 8.5 C.vectensis, mono R. salina, mono H. phi, mono 8.0 cell concentration (cells ml G.theta, mono H. tepida, mono 103

7.5 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Time (days) Time (days)

Fig. 3. Experiment 2. A. Changes in cell concentrations of M. rubrum as a function of incubation time (day) in monoculture and when grown in mixed cultures with the cryptophytes Chroomonas vectensis, Rhodomonas salina, Hanusia phi, Guillardia theta, Hemiselmis tepida. B. Changes in prey cell concentrations in the mixed cultures. Development of pH in the cultures (C) and changes in prey cell concentrations cells in monocultures (D).

12 Hansen et al.

Table 3. Average prey concentrations and end prey concentrations are also listed. In cases where T. amphioxeia was used as prey item, data are also given for Days 4-8 and 8-10. Initial prey concentrations were around 1000 cells ml-1, exception the experiments involving T. amphioxeia, Day 4 and 8, where initial prey concentrations were 5000 cells ml-1. For all species the estimated ingestion rates were significantly different from zero (p<0.05)

Average prey concentration in mixed End prey concentration culture with M. rubrum during the (cells ml-1) Species experiment (cells ml-1) Heterocapsa rotundata 6817 22000 Chroomonas vectensis 3590 9500 Guillardia theta 4279 12350 Hanusia phi 3172 7300 Hemiselmis tepida 2536 4600 Proteomonas sulcata 12204 42600 Rhodomonas salina 15264 63700 Teleaulax amphioxeia (Day 0-4) 203 7 Teleaulax amphioxeia (Day 4-8) 287 0 Teleaulax amphioxeia (Day 8-10) 1018 33

3.0

2.5 ) -1

2.0

1.5

1.0

Ingestion rate (cells d 0.5

0.0 i ta h a a 4 8 id t - - he p lina 4 ndata t H.p lca a u te su G. R. s , day 0 , day H. P. p H. rot C.vectensis mp mp day 8-10 .a T.am T.a T

Prey type Fig. 4. Mesodinium rubrum. Estimated average ingestion rates calculated based on data shown in Fig. 2 and 3. Only data for the first 4 days (Day 0-4) of the incubations were used, except for the prey T. amphioxeia where ingestion rates also were estimated for Day 4-8 and Day 8-10.

13 Hansen et al.

Transmission electron microscopy

Figures 5A-F are a series of micrographs obtained from a culture of Mesodinium rubrum fed the cryptomonad Hemiselmis tepida. The longitudinal section of Mesodinium (Fig. 5A) shows several chloroplasts in the cell, located in both the oral and the aboral end of the cell. Collectively we designate these as chromatophores as their origin is different (see below). Hemiselmis is ingested at the oral end of Mesodinium, which here possesses, in addition to a small mouth, bands of and a number of tentacles (Fig. 5B). Hemiselmis cells are ingested whole (Figs 5C&D), and Fig 5C illustrates a cell in which both the cryptomonad nucleus, the cryptomonad chloroplasts and the two types of trichocysts () of Hemiselmis are visible. One of the two macronuclei of Mesodinium is also visible Fig. 5A, while the small nucleus in the upper part of the cell is from a prey cell. Details of the two types of chromatophore in Mesodinium are shown in Figs 5E&F. In one type, the permanent chromatophore or chloroplast of Mesodinium, the thylakoids are grouped in lamella of three thylakoids (Fig. 5E). For comparison, Fig. 5F shows a chloroplast from a Hemiselmis cell located within Mesodinium, demonstrating the 2-thylakoid lamella characteristic of Hemiselmis. In contrast to the 3-thylakoid lamellae, the 2-thylakoid lamellae of Hemiselmis were often not well preserved in the sections, indicating that they were being subjected to digestion .

Molecular result

The nuclear SSU rDNA alignment consisted of 1572 bp. The molecular phylogeny based on this alignment and inferred from Bayesian analysis yielded the tree topology shown in Fig. 1. Glaucocystophytes rooted the tree and the cryptophytes were divided in five clades. The first clade included four genera: Hemiselmis, Chroomonas, Komma and Falcomonas. The second clade comprised four genera: Rhodomonas, , and Storeatula. The third clade was formed by two genera: Hanusia and Guillardia. The fourth clade was composed of genus only. Finally, the fifth clade included three genera: Teleaulax, and Geminigera. Proteomonas sulcata and Falcomonas daucoides were isolated and did not belong to any clade. Yet, the relationship between the clades was unresolved.

14 Hansen et al.

Fig. 5. Mesodinium rubrum grown in culture and fed Hemiselmis tepida. A. longitudinal section of M. rubrum from the oral end (Oe) to the aboral end (Ae) illustrating the cilia, chloroplasts (Chl), starch grains (St), the (ma-N) and the bands of microtubules around the mouth. B. section of the mouth (M) of M. rubrum through the tentacles (T) about to engulf a cell of H. tepida (Hr) showing part of the bands of microtubules (Bm). C. Engulfed cell of H. tepida (arrows) within M. rubrum. Visible organelles are the chloroplast, the small (sTri) and large trichocysts (lTri) and the cryptomonad nucleus (N). D. Chloroplast of M. rubrum (arrowheads) with triplets of thylakoids (t-Thy) next to an engulfed cell of H. tepida (arrows) with a chloroplast showing thylakoids in pairs (p-Thy). E. Detail of the chloroplast of M. rubrum harbouring thylakoids in triplets. F. Detail of the chloroplast within H. tepida, which holds thylakoids in pairs.

15 Hansen et al.

DISCSSION

Ingestion and growth responses of Mesodinium rubrum when fed different types of prey

Cultures of Mesodinium rubrum from various parts of the world have only been established by offering them cryptophytes from the Teleaulax/Geminigera clade as food (Table 4). Recently, Park et al. (2007) found some growth of a strain of M. rubrum fed the cryptophyte strain (CR- MAL03) belonging to the “Rhodomonas” clade. However, to what extent M. rubrum can ingest and grow on a variety of cryptophyte prey is presently unknown. In this study, our M. rubrum strain ingested all the offered prey items. Estimated ingestion rates were in the range of 0.5 to 1 cell d-1 for all prey types exception T. amphioxeia, where ingestion rates were higher at comparable cell concentrations (Fig. 2, Table 2). From a previous study we know that this level of ingestion rate suffices to support good growth of M. rubrum (Smith & Hansen 2007). Yet, M. rubrum did only grow when T. amphioxeia was offered as food.

Table 4. List of successful M. rubrum cultures and preys that they can be maintained on. Place of origin of culture Prey reference Antarctica Geminigera cryophila Gustafson et al. 2000, Johnson et al. 2005, 2006, 2007, Hacket et al. 2009 Denmark Teleaulax amphioxeia, Hansen & Fenchel 2006, Smith & Teleaulax sp. Hansen 2007, Riisgaard & Hansen 2009 Korea Teleaulax spp, Unidentified Yih et al. 2004, Park et al. 2006, strains belong to the 2007, 2010 Teleaulax clade Japan Teleaulax amphioxeia Nagai et al. 2008, Nishitani et al. 2008a,b

Can Mesodinium rubrum sequester chloroplasts from cryptophyte prey?

This raises the question of why M. rubrum cannot grow when fed other prey types than Teleaulax spp. although it ingests all kinds of prey items. One explanation could be that M. rubrum relies on the replacement (from time to time) of chloroplasts for sustained growth. T. amphioxeia is characterized by having chloroplasts with thylakoids that are arranged in triplets (like in M.

16 Hansen et al. rubrum). With the exception of P. sulcata that shares this thylakoid structure, the other cryptophytes used in the present study have thylakoids arranged in pairs (Cuetos-Garcia 2010). Thus, if functional chloroplasts had been taken up from these other cryptophytes by M. rubrum, this would be revealed in TEM sections of M. rubrum cells. However, in the case of M. rubrum fed H. tepida that we studied in detail, prey cells taken up were found in food vacuoles and often partly digested. The same holds true when we used molecular methods to test whether functional chloroplasts from other prey species were taken up by M. rubrum. In all cases investigated, we only found sequences matching T. amphioxeia within M. rubrum cells and never sequences that would match that of the other prey fed to M. rubrum. Thus, we found no proof of chloroplast sequestration by M. rubrum when fed these other types of cryptophytes.

Why is Mesodinium rubrum not selective in its choice of prey?

One of the most striking observations of the present study was that M. rubrum ingested all the offered prey items, including a dinoflagellate of the same size as the cryptophyte prey. This appears to be a bad strategy for an organism that seems to rely on the ingestion of specific cryptophyte prey. Like other Mesodinium species, M. rubrum is a raptorial feeder that senses individual prey items before it attacks and captures them, using the tentacles at the oral end of the cell (see Jakobsen et al. 2006). Apparently, the ciliate cannot recognize its prey. As long as the prey has the right size and it is actively swimming it will be ingested. This in turn explains why M. rubrum, when the prey concentration is high, is able to ingest much more food that it actually needs (see Smith & Hansen 2007). Therefore, in waters where Teleaulax spp are present all year round such as in the Kattegat (Denmark), M. rubrum will always eat enough of the right kind of prey to sustain its growth despite the fact that it might also eat preys that cannot. This feeding mechanism might seem inadequate, yet it works.

Conclusions and future perspectives

We found no evidence of chloroplast replacement in our strain of Mesodinium rubrum when offered cryptophytes from a variety of cryptophyte clades. This might suggest that the cryptophyte symbiont in M. rubrum is a permanent symbiont that cannot be replaced by prey

17 Hansen et al. ingestion. It implies that M. rubrum feeds to obtain some important, as yet unknown, growth factor(s) which it can get only through ingestion of cryptophytes from the Teleaulax/Geminigera clade. Yet, we can not discard the possibility that M. rubrum can replace its symbionts with other species from the Teleaulax/Geminigera clade. Unfortunately, the only temperate Teleaulax species available in culture is T. amphioxeia. Thus, to solve the problem of symbiont replacement in M. rubrum, a cryptophyte species belonging to the Teleaulax/ Geminigera clade that posseses a distinctive chloroplast or nucleomorph molecular signature from T. amphioxeia needs to be established in culture.

Acknowledgements This study was supported by the Danish Research Council to Per Juel Hansen, grant no 272-06- 0485, and a PhD grant from Rajamangala University of Technology Srivijaya, Thailand to Woraporn Tarangkoon.

Literature cited

Bhattacharya D, Helmchen T, Bibeau C (1995) Comparisons of nuclear-encoded small subunit ribosomal reveal the evolutionary position of the Glaucocystophyta. Mol Biol Evol12:415- 420

Felsenstein J (1981) Evolutionary trees from DNA sequences: a maximum likelihood approach. J Mol Evol 17:368-376

Felsenstein J (1985) Confidence limits on phylogenies: an approach using the bootstrap. 39:783-791

Frommlet J, Iglesias-Rodríguez D (2008) Microsatelite genotyping of single cells of the dinoflagellate species polyedrum (Dinophyceae): a novel approach for marine microbial population studies. J Phycol 44:1116-1125

18 Hansen et al.

Garcia-Cuetos L, Moestrup Ø, Hansen PJ, Daugbjerg N (2010) The toxic dinoflagellate Dinophysis acuminata harbors permanent chloroplasts of cryptomonad origin, not kleptochloroplasts. Harmful Algae 9:25-38

Guidon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52:696-704

Guillard RRL (1975) Culture of phytoplankton for feeding marine invertebrate. In: WL Smith WL and Chanley MH (eds) Culture of marine invertebrate animals. Plenum Pub., New York, p. 29-60

Gustafson JrDE, Stoecker DK, Johnson MD, Van Heukelem WF, Sneider K (2000) Cryptophyte algae are robbed of their organelles by the marine ciliate Mesodinium rubrum. Nature 405:1049- 1052

Hackett JD, Mengmong T, Kulis DM, Fux E, Hess P, Bire R, Anderson DM (2009) DSP toxin production de novo in cultures of Dinophysis acuminata (Dinophyceae) from North America. Harmful Algae 8:873-879

Hall TA (1999) Bioedit: a user friendly biological sequence alignment editor and analysis program from windows 95/97/NT. Nucleic Acids Symp Ser 41:95–98.

Hansen G, Daugbjerg N, Franco JM (2003) Morphology, toxin composition and LSU rDNA phylogeny of Alexandrium minutum (Dinophyceae) from Denmark, with some morphological observations on other European strains. Harmful Algae 2: 317-335

Hansen PJ, Fenchel T (2006) The bloom-forming ciliate Mesodinium rubrum harbors a single permanent endosymbiont. Mar Biol Res 2:169–177

Hibberd DJ (1977) Observation on ultrastructure of crypotmonad endosymbiont of red-water ciliate Mesodinium rubrum. J Mar Biol Assoc UK 57:45-61

19 Hansen et al.

Hoef-Emden K (2008) Molecular phylogeny of phycocyanin containing cryptophytes : Evolution of biliproteins and geographic distribution. J Phycol 44:985-993

Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:754-755

Jakobsen HH, Everett LM, Strom SL (2006) Hydromechanical signaling between the ciliate Mesodinium pulex and motile protist prey. Aquat Microb Ecol 44:197-206

Jakobsen HH, Hansen PJ (1997) Prey size selection, growth and grazing responses of a small heterotrophic dinoflagellate Gymnodinium sp. and a ciliate Balanion comatum: a comparative study. Mar Ecol Prog Ser 158:75-86

Johnson MD, Oldach D, Delwiche CF, Stoecker DK (2007) Retention of transcriptionally active cryptophyte nuclei by the ciliate Myrionecta rubra. Nature 445: 426-428

Johnson MD,Stoecker, DK (2005) Role of feeding in growth and photophysiology of Myrionecta rubra. Aquat Microb Ecol 39:303-312

Johnson MD, Tengs T, Oldach D, Stoecker DK (2006) Sequestration, performance, and functional control of cryptophyte plastids in the ciliate Myrionecta rubra (Ciliophora). J Phycol 42:1235-1246

Katoh K,Toh H(2008) Recent developments in the MAFFT multiple sequence alignment program. Brief Bioinform 9:286-298

Nagai S, Nishitani G, Tomaru Y, Sakiyama S, Kamiyama T (2008) Predation by the toxic dinoflagellate Dinophysis fortii on the ciliate Myrionecta rubra and observation of sequestration of ciliate chloroplasts. J Phycol 44:909–22

20 Hansen et al.

Nishitani G, Nagai S, Sakiyama S, Kamiyama T (2008) Successful cultivation of the toxic dinoflagellate Dinophysis caudata (Dinophyceae). Plankton Benthos Res 3:78–85

Nishitani G, Nagai S, Takano Y, Sakiyama S, Baba K, Kamiyama T (2008) Growth characteristics and phylogenetic analysis of the marine dinoflagellate Dinophysis infundibulus (Dinophyceae). Aquat Microb Ecol 52:209–21

Nunn GB, Theisen BF, Christensen B, Arctander P (1996) Simplicity-correlated size growth of the nuclear 28S ribosomal RNA D3 expansion segment in the crustacean order Isopoda. J Mol Evol 42:211–223.

Park MG, Kim M, Kim S, Yih W (2010) Does Dinophysis caudata (Dinophyceae) have permanent plastids? J Phycol 46: 236-242

Park MG, Kim S, Kim H S, Myung G, Kang YG, Yih W (2006). First successful culture of the marine dinoflagellate Dinophysis acuminata. Aquat Microb Ecol 45:101–6

Park JS, Myung G, Kim HS, Cho BC, Yih W (2007) Growth responses of the marine photosynthetic ciliate Myrionecta rubra to different cryptomonad strains. Aquat Microb Ecol 48:83-90

Posada D, Crandall KA (1998) MODELTEST: Testing the model of DNA substitution. Bioinformatics 14:817-818

Riisgaard K, Hansen PJ (2009) Role of food uptake for photosynthesis, growth and survival of the mixotrophic dinoflagellate Dinophysis acuminata. Mar Ecol Prog Ser 381:51–62

Rowley JC, Moran DT (1975) Simple procedure for mounting wrinkle-free sections on formvar- coated slot grids. Ultramicroscopy 1: 151–155.

21 Hansen et al.

Scholin CA, Herzog M, Sogin M, Anderson DM (1994) Identification of group and strain- specific genetic markers for globally distributed Alexandrium (Dinophyceae). II. Sequence analysis of a fragment of the LSU rRNA gene. J Phycol 30:999-1011

Smith M, Hansen PJ (2007) Interaction between Mesodinium rubrum and its prey: importance of irradiance, prey concentration, and pH. Mar Ecol Prog Ser 338:61-70

Swofford DL (2002). PAUP* 4.0: Phylogenetic analysis using parsimony (*and other methods). Sinauer Associates, Sunderland, MA

Taylor FJR, Blackbourn DJ, Blackbourn J (1971) The red-water ciliate Mesodinium rubrum and its “incomplete symbionts”: a review including new ultrastructural observations. J Fish Res Brd Canada 28:391-407

Wilgenbusch JC, Warren DL,Swofford DL (2004). AWTY: A system for graphical exploration of MCMC convergence in Bayesian phylogenetic inference. http://cebcsitfsuedu/awty

Yang Z, Rannala B (1997) Bayesian phylogenetic inference using DNA sequences: A Markov Chain Monte Carlo method. Mol Biol Evol 14:717-724

Yih W, Kim HS, Jeong H J, Myung G, Kim YG (2004) Ingestion of cryptophyte cells by the marine photosynthetic ciliate Mesodinium rubrum. Aquat Microb Ecol 36:165–170

22

APPENDIX-PAPER IV

Putative N2-fixing heterotrophic bacteria associated with dinoflagellate-cyanobacteria consortia in the low-nitrogen Indian Ocean

Hanna Farnelid1, Woraporn Tarangkoon2,3, Gert Hansen4, Per Juel Hansen2 and Lasse Riemann1*

1Department of Natural Sciences, Linnaeus University, SE-39182 Kalmar, Sweden 2Marine Biological Laboratory, Strandpromenaden 5, 3000 Helsingør, Denmark 3Faculty of Science and Fisheries Technology, Rajamangala University of Technology Srivijaya, 92150 Trang, Thailand 4Department of , Ø. Farimagsgade 2D, 1353, Copenhagen, Denmark

*Corresponding author. E-mail [email protected]; Tel. (+46)480447334; Fax: (+46)480447340.

Farnelid et al.

ABSTRACT

Heterotrophic dinoflagellates bearing unicellular cyanobacterial symbionts are common within the order Dinophysiales. However, the ecological role of these symbionts is unclear. Due to the occurrence of such consortia in oceanic waters characterized by low nitrogen concentrations, we hypothesized that the symbionts fix gaseous nitrogen (N2). Individual heterotrophic dinoflagellates containing cyanobacterial symbionts were isolated from the open Indian Ocean and off Western Australia, and characterized using light microscopy, transmission electron microscopy (TEM), and nitrogenase (nifH) gene amplification, cloning, and sequencing. Cyanobacteria, heterotrophic bacteria and eukaryotic algae were recognized as symbionts of the heterotrophic dinoflagellates. NifH gene sequences were obtained from 23 of 37 (62%) specimens of dinoflagellates (Ornithocercus spp. and Amphisolenia spp.). Interestingly, only two specimens contained cyanobacterial nifH sequences, while 21 specimens contained nifH genes related to heterotrophic bacteria. Of the 137 nifH sequences obtained 68% were most similar to Alpha-, Beta- and Gamma-, 8% clustered with anaerobic bacteria, and 5% were related to second alternative (anfH). Twelve sequences from five host cells formed a discrete cluster which may represent a not yet classified nifH Cluster. Eight dinoflagellates contained only one type of nifH sequence (>99% sequence identity) but overall the putative N2-fixing symbionts did not appear host specific and mixed assemblages were often found in single host cells. This study provides the first insights into the nifH diversity of dinoflagellate symbionts and suggests a symbiotic co- existence of non-diazotrophic cyanobacteria and N2-fixing heterotrophic bacteria in heterotrophic dinoflagellates.

Key words: symbionts, nitrogen fixation, nifH, heterotrophic bacteria, Ornithocercus, Amphisolenia, Histioneis, dinoflagellates, Indian Ocean, Galathea 3

1 Farnelid et al.

INTRODUCTION

Cyanobacterial symbionts (cyanobionts) of non-photosynthetic dinophysoids (Dinophyceae) were first observed more than 100 years ago (Schütt 1895). They are thought to function as photosynthetic partners in their relationship with the host (Taylor 1982) but although many different types of cyanobacterial and bacterial symbionts have been described for several dinophysoid genera (such as Amphisolenia, Histioneis, Ornithocercus and Parahistioneis) little is known about the identity and diversity of these symbionts. Further, their ecological significance is essentially unknown.

Recently, we found that symbiont-bearing dinoflagellates were most common in the photic zone of the Indian Ocean characterized by low nutrients and low phytoplankton biomass (Tarangkoon et al. 2010). This is consistent with previous observations from the Indian Ocean (Jyothibabu et al. 2006) and the Red Sea (Gordon et al. 1994). Due to this distribution it has been suggested that the photosynthetic symbionts are N2-fixers (diazotrophs), providing their hosts with reduced N (Gordon et al. 1994; Jyothibabu et al. 2006). To date only a single study has demonstrated nitrogenase in a cyanobiont of a heterotrophic dinoflagellate (Foster et al. 2006a); however, several studies indicate that cyanobionts may be diazotrophic. For instance, even though most of the 65 cyanobacterial 16S rRNA gene sequences retrieved from individual heterotrophic eukaryotic host cells were related to Synechococcus, three sequences from two Histioneis sp. hosts were related to the unicellular N2-fixing cyanobacterium Cyanothece sp. (Foster et al. 2006b). Further, using fluorescent in situ hybridization putative unicellular diazotrophic cyanobionts associated with dinoflagellates have been observed in the Mediterranean Sea (Le Moal & Biegala 2009) and the Southwest Pacific Ocean (Biegala & Raimbault 2008).

In tropical and subtropical waters the ubiquitous filamentous cyanobacterium Trichodesmium sp. (Capone et al. 1997) and intracellular cyanobacterial symbionts of diatoms (Rhizolenia-; Carpenter et al. 1999) were long believed to be solely responsible for pelagic N2 fixation. However, recent molecular studies targeting the nifH gene, encoding the iron component of the nitrogenase enzyme, have shown that free-living unicellular cyanobacteria are also abundant and can account for a significant fraction of the N2 fixation (Montoya et al. 2004; Zehr et al. 2001). Similarly, it has recently been recognized that non-

2 Farnelid et al. cyanobacterial diazotrophs, mostly heterotrophic bacteria, are widespread in marine waters, and their ecological function and importance is currently unknown (Farnelid & Riemann 2008).

In the present study we sought to identify potential N2-fixing symbionts of heterotrophic dinoflagellates in the Indian Ocean. The symbionts were characterized using light microscopy and transmission electron microscopy (TEM) and nifH genes were amplified from individual symbiont-bearing dinoflagellate hosts using a nested PCR approach. Our study points to a hitherto unrecognized importance of heterotrophic bacteria for N acquisition in cyanobacteria-dinoflagellate consortia in tropical waters.

MATERIALS AND METHODS

Sample collection. Sampling was carried out onboard the Danish Navy surveillance frigate “F359 Vædderen” during Leg 7 of the 3rd Danish Galathea expedition (October–November 2006). Samples were obtained from 21 stations located across the Indian Ocean and along a transect perpendicular to Broome in North Western Australia (Fig. 1 in Tarangkoon et al. 2010). For nifH gene analysis samples were obtained at stations BR5 to BR9 (17° 03' S, 120° 49' E; 16° 50' S, 120° 34' E; 16° 26' S, 119° 56' E; 16° 15' S, 119° 38' E; 16° 01' S 119° 19' E) in the Broome transect. Live plankton samples were collected at each station with a 20 µm mesh size plankton net as vertical hauls, from about 70 m depth to the surface or from water samples (30 L) from 10 m and 30 m depths collected by Niskin bottles attached to a conductivity, temperature and depth profiler rosette. Subsequently, plankton was concentrated using a 20 µm Nitex mesh size filter. The filters were kept immersed during the filtration to facilitate the retention of live cells. The concentrated samples were transferred to 100 ml of filtered seawater from which single cells were isolated using a drawn Pasteur pipette. Cells for nifH gene analysis were then successively washed in three baths of 2 ml 0.2 µm filtered seawater, placed individually in a 0.2 ml PCR tube, and immediately frozen at -20oC. The cells included a range of dinoflagellates species (Table 3), though, no Histioneis species were obtained.

3 Farnelid et al.

Microscopy. Dinoflagellates were examined shortly after sampling using an Olympus BX51 light microscope fitted with a Soft-Imaging ColorView III digital camera and identified as described in (Tarangkoon et al. 2010). In total ~100 cells were examined in the study. Seven cells of O. magnificus and two of O. quadratus were collected at station 5 in the Indian Ocean transect (29° 35' S, 95° 15' E) and preserved for TEM (Tarangkoon et al. 2010). In the laboratory, sectioning was done on an Ultracut E ultramicrotome using a diamond knife, and the sections were collected on slot grids and placed on Formvar film. After staining in uranyl acetate and lead citrate, sections were examined in a JEOL JEM-1010 electron microscope operated at 80 kV. Micrographs were taken using a GATAN 792 digital camera.

DNA extraction and nifH amplification. DNA was extracted from individual heterotrophic dinoflagellates with symbionts using an enzyme/phenol-chloroform protocol (Riemann et al. 2008) and 200 μl SET lysis buffer (20% sucrose, 50 mM EDTA, 50 mM Tris-HCl, pH 8.0). An extraction without added sample served as a control on the purity of the extraction chemicals. Seven of the samples were instead of the extraction procedure subjected to three cycles of freeze/thawing (-80°C for 1 min and 75°C for 1 min constituted one cycle), which lyses the cells (Sebastian & O'Ryan 2001). To amplify nifH, degenerate primers purified by high-performance liquid chromatography and polyacrylamide gel electrophoresis (Sigma- Aldrich; Zani et al. 2000; Zehr & McReynolds 1989) were used according to a nested PCR protocol (Zehr & Turner 2001) using Pure Taq Ready-To-Go PCR Beads (GE Healthcare). A negative control reaction with UV-treated water was included in each PCR batch. To minimize the risk of contamination, mixing of reagents was done in a UV-treated sterile flow bench in a UV-treated room, template was added in a PCR/UV workstation in a separate room, and single tubes (not strips) were used. For the initial PCR reaction 3-6 μl of the extracted DNA or freeze/thawed solution was added as template and 1 μl PCR product was transferred to the subsequent PCR reaction. Five μl from the second PCR reaction was examined on a 1% agarose gel and for samples that produced a ~359 bp product, the remaining 20 μl was gel purified (E.Z.N.A Gel extraction kit, VWR). The negative PCR control and the negative extraction control were always blank. For the negative control PCR reaction, although there was no visible product the gel region corresponding to the correct product size was excised, gel purified and cloned. All purified products were cloned using the TOPO TA Cloning Kit (Invitrogen). DNA was obtained using the R.E.A.L Prep96

4 Farnelid et al.

Plasmid Kit (Qiagen) according to manufacturer’s protocol and sequencing was done commercially (Macrogen, Korea).

Sequence and phylogenetic analysis. Vector sequences and primers were removed manually and the sequences were translated and aligned using the Lasergene 7 package (DNASTAR). The most similar uncultured and cultured relatives as identified from BLASTN comparisons from the NCBI GenBank database were added to the dataset and a neighbor-joining was constructed in MEGA4 (Tamura et al. 2007). The partial nifH sequences have been deposited in GenBank under accession numbers GU196835-GU196971.

RESULTS

Microscopy analyses of symbionts

The morphologies (e.g. color, shape and size) of symbionts of heterotrophic dinoflagellates (~100 cells) were compared to published data on ectosymbionts (Table 1) and endosymbionts (Table 2). All Ornithocercus spp. cells had orange and elongated cyanobacterial ectosymbionts located within the cingulum, while some also had large rod-shaped non- cyanobacterial on their sulcal lists (Fig. 1A, large arrow and arrowhead, respectively; Table 1). These putative ectosymbiotic heterotrophic bacteria were not observed on Histioneis or Amphisolenia. Histioneis spp. contained two other types of cyanobacterial ectosymbionts (Fig. 1B, Table 1). In Amphisolenia spp. only endosymbiotic spheres of 3 - 7 µm were found (Fig. 1C, large arrow; Table 2). The endosymbionts in A. bidentata contained a single yellow chloroplast and a nucleus demonstrating its eukaryotic origin. The symbionts of A. thrinax had a more brownish color, but whether these symbionts were of a eukaryotic origin is unclear (not shown).

5 Farnelid et al.

Table 1. Types and characteristics of ectosymbionts of heterotrophic dinoflagellates.

Group Cell shape Length Width Characteristics / Internal structure of symbionts Heterotrophic References (μm) (μm) dinoflagellates (name type) Indian Ocean Synechococcus carcerarius Ellipsoid, 8-10 3-5 Light pink to purplish-red in color Ornithocercus formosus, Norris 1967 Short rod, O. heteroporus, Cylindrical O. magnificus, O. quadratus, O. splendidus, O. thumii, H. carinata, H. dolon, H. pacifica, 6 H. striata Parahistioneis sp. H. carinata, Synechocystis consortia Spherical 6-8 Grey-Blue in color Norris 1967 Parahistioneis sp. Ornithocercus sp. Lucas 1991 Cyanobacteria Rod/ 1.5-2.8 1.2-1.5 3-4 concentric thylakoids,carboxysomes central Histioneis sp. (Type I) Parahistioneis sp. Ellipsoid Ornithocercus sp. Cyanobacteria Rod/ 1.0-1.7 2 Peripheral/central thylakoids, occasional carboxysomes Lucas 1991 Citharistes apsteinii Ellipsoid (Type II) Short, irregular thylakoids, few large carboxysomes, many Histioneis sp. Lucas 1991 Cyanobacteria Spherical 3.5-4.8 cyanophycin granules Parahistioneis sp. (Type III) 2-3 peripheral thylacoids, several transverse thylakoids, cluster O. magnificus, Cyanobacteria Elongate 8-10 1.7-3.3 This study of carboxysomes, occasional putative cyanophycin granule O. quadratus Cyanobacteria Ellipsoid 1.25 Orange color Histioneis spp. This study Cyanobacteria Spherical 2.5-5 - Pale light greenish color Histioneis spp. This study

a : Mean length

6 Farnelid et al.

Table 1. (continue) Group Cell shape Length Width Characteristics / Internal structure of symbionts Heterotrophic References (μm) (μm) dinoflagellates (name type) Pacific Ocean Cyanobacteria Spherical/ 1.6±0.6a 1.3±0.4a Prominent glycogen clusters throughout Ornithocercus sp. Foster et al. 2006a Oblong the cytoplasm, carboxysomes scattered often in clusters (Type 1) Cyanobacteria Ellipsoid 2.4±0.6a 1.9±0.5a Sheath, 4-5 concentric peripheral/central Ornithocercus sp Foster et al. 2006a thylakoids, carboxysomes, and glycogen stores central (Type 2) Cyanobacteria Spherical 3.7±0.7a 2.3±0.8a Sheath, large glycogen inclusions throughout cytoplasm, H. depressa Foster et al. 2006a no distinct thylakoids (Type 4) Cyanobacteria Rod/ 1.4±0.5a 1.0±0.3a Glycogen in smaller packets scattered, H. depressa Foster et al. 2006a Spherical diffuse thylakoids and no carboxysomes apparent (Type 5) Cyanobacteria Rod 2.8±0.2a 1.3±0.2a Sheath, 3-4 peripheral thylakoids, small packets Histioneis sp. Foster et al. 2006a of glycogen or occasional as larger inclusion, (Type 6) no visible carboxysomes Cyanobacteria Ellipsoid 1.7±0.6a 0.9±0.1a 4-6 peripheral thylakoids with small packets of glycogen Histioneis sp. Foster et al. 2006a 7 scattered in between, no visible carboxysomes (Type 7) Prochlorococcus? Spherical/ 0.6±0.2a 0.3±0.1a 2-3 peripheral thylakoids, small scattered glycogen packets, Histioneis sp. Foster et al. 2006a Oblong no visible carboxysomes (Type 8) Atlantic Ocean Foster et al. 2006a Cyanobacteria Spherical/ 3.5±0.7a 2.8±0.3a Thylakoids throughout cytoplasm, carboxysomes scatted in O. magnificus (Type 3), Rod clusters, glycogen in larger bands between thylakoids Janson et al. 1995 Cyanobacteria Rod/ 2.5 3-4 concentric thylakoids O. magnificus Janson et al. 1995 (type I of Lucas) Ellipsoid Cyanobacteria Rod/ 1-2 Peripheral/central thylakoids O. magnificus Janson et al. 1995 (type II of Lucas) Ellipsoid Cynabacteria Rod 10 1.5 Peripheral thylakoids O. magnificus Janson et al. 1995 (type IV of Lucas) 0.4±0.1a 0.4±0.1a - Foster et al. 2006a Heterotrophic bacteria Coccoid - Glycogen scattered throughout the cytoplasm O. magnificus 0.3±0.1a (Type b1) 0.5±0.2a 1.0±0.4 0.4±0.1a Foster et al. 2006a Heterotrophic bacteria Coccoid a - Glycogen scattered throughout the cytoplasm O. magnificus 0.3±0.3a (Type b2) 0.3±0.1a a : Mean length

7 Farnelid et al.

Table 2. Types and characteristics of endosymbionts of heterotrophic dinoflagellates in the Indian Ocean.

Group Cell shape Length Width Characteristics / Internal structure of symbionts Heterotrophic References (μm) (μm) dinoflagellates (name type) S. carcerarius Ellipsoid, 8-10 3-5 Light pink to purplish-red in color A. globifera Norris 1967 Short rod, Cylindrical Cyanobacteria Rod 10 1.5 Peripheral thylakoids, carboxysomes in rosettes. Amphisolenia sp. Lucas 1991 Oblique division (Type IV) Eukaryotic Spherical - - Golden cells, possibly Chrysophyceae or A. thrinax, Norris 1967 Dinophyceae A. palmata Eukaryotic Spherical 2-3 - 1-2 plastids, a nucleus, A. bidentata, Lucas 1991 A. thrinax Eukaryotic Spherical 3-5 - Single yellow chloroplast and a nucleus A. bidentata This study - Spherical 4-7 - Brownish color A. thrinax This study Heterotrophic bacteria Coccoid, - 0.5 A central core of DNA fibrils, numerous A. bidentata, Lucas 1991 8 Short rod A. thrinax

8 Farnelid et al.

Fig. 1. Light microscopy of live cells. (A) Ornithocercus thumii; cyanobacterial ectosymbionts (large arrow). Notice large bacteria on the sulcal list (arrowhead). The small arrows indicate LCL (lower cingular list), UCL (upper cingular list), LSL (left sulcal list). (B) Histioneis biremis; two different types of cyanobacterial ectosymbionts are present (large and small arrows, respectively). (C) Amphisolenia bidentata with numerous eukaryotic endosymbionts (large arrow). Inset: Details of the endosymbionts. Chloroplast (small arrow); nucleus (arrowhead).

9 Farnelid et al.

TEM revealed that the ectosymbionts of O. magnificus and O. quadratus were cyanobacteria and heterotrophic bacteria (Fig. 2A, large and small arrows respectively). In both species, cyanobacterial ectosymbionts were present in the cingulum (Fig. 2A); though a substantial number was lost during the TEM fixation process. These all appeared to be of the same type, i.e. containing 2 - 3 peripheral thylakoid bands in addition to several bands traversing the cell (Fig. 2E, large arrow). Clusters of polyhedral granules, carboxysomes (Lucas 1991), were present in all the cyanobacterial ectosymbionts examined (Fig. 2B, D). In some, electron translucent granules were present (Fig. 2D), similar to putative cyanophycin granules (Lucas 1991). A typical eubacterial Gram-negative wall, consisting of a thin wall in-between two membranes, surrounded the cyanobiont cells (Fig. 2C, arrows). In some Ornithocercus cells, the cingulum also contained small rod-shaped, 1.5 x 0.2 µm, heterotrophic bacteria (Fig. 2A, F, G). Unfortunately, the large rod- shaped heterotrophic bacteria seen on the sulcal list by light microscopy (Fig. 1A, arrowhead) were lost during the TEM preparation procedure.

10 Farnelid et al.

Fig. 2. TEM of Ornithocercus magnificus. (A). Longitudinal section of whole cell of O. magnificus. Four cyanobacterial ectosymbionts and small bacteria are present in the cingulum (large and small arrows, respectively). Numerous rhabdosomes are present within the cell (arrowhead). (B) Longitudinal section of a cyanobiont. Peripheral- (small arrow) and central thylakoid membranes (large arrow); putative carboxysomes (c). (C) The triple-layered cyanobiont wall (small arrows). (D) The putative carboxysomes. (E) Traversing thylakoids (large arrow) and putative cyanophycin granule (arrowhead) of a cyanobiont. (F,G) Details of the bacterial ectosymbionts, in longitudinal (large arrow) and transverse view (arrowhead).

NifH sequence composition and phylogeny

NifH amplicons were obtained from 23 of the 37 analyzed symbiont-bearing heterotrophic dinoflagellates. The 137 nifH sequences obtained were related to nifH Cluster I (Cyanobacteria and Alpha-, Beta- and Gamma-proteobacteria), Cluster II (alternative nitrogenases; anfH) and Cluster III (anaerobic bacteria) as defined by (Chien & Zinder 1996). Sixty-eight% of the sequences

11 Farnelid et al.

(originating from 17 samples) were most similar to proteobacterial nifH sequences (Fig. 3). All ten sequences from the negative control formed a cluster within (>98% within cluster sequence identity; Fig. 3) related to a previously reported PCR reagent contaminant sequence. Four sample sequences (A. bidentata, samples P60 and P62; Table 3) were affiliated with this cluster, but were not identical to the negative control sequences. Other sequences clustering with Betaproteobacteria were most similar to Ideonella dechloratans (18 sequences from five samples; 91-92% sequence identity) and Burkholderia vietnamiensis (12 sequences from five samples; 97-98% sequence identity) but were clearly distinguished from the negative control sequences (Fig. 3).

Two of the dinoflagellates contained nifH sequences clustering with cyanobacteria (Fig. 3). Sample P7 contained sequences of 97 % nucleotide similarity to punctiforme while the sequences from P1 were only distantly related to known (<78 % sequence identity). Three samples contained nifH sequences within nifH Cluster III and two samples contained sequences related to anfH genes, encoding the iron-only nitrogenase, within nifH Cluster II (Table 3). Twelve sequences, originating from five samples, formed a well supported cluster (bootstrap 99%, Fig. 3), which may represent a novel nifH cluster. These sequences clustered with Caldicellulosiruptor saccharolyticus (87-98% sequence identity) and with environmental nifH sequences (EU978414 and EU693383). Five sequences formed a separate cluster only distantly related to known nifH phylotypes (<69% sequence identity; Fig. 3).

To link sequence composition to dinoflagellate hosts we examined whether a sample contained single or several nifH sequence types and whether specific nifH sequences were associated with specific host species. Eight host cells, among which all examined host species were represented, contained only one nifH sequence type (>99% sequence identity) while nine host cells contained two or three nifH sequence types each (Table 3). In addition, similar nifH sequences were found in several hosts and different host species (Fig. 3). For example, 13 nifH gammaproteobacterial sequences from four samples of different species formed a distinct cluster with an uncultured nifH from the Pacific Ocean (DQ481270, 98-99 % sequence identity) and 13 sequences originating from three samples of different species clustered with Klebsiella pneumoniae (98-99% sequence identity).

12 Farnelid et al.

Fig. 3. Neighbor-joining phylogenetic tree of nifH-deduced sequences from symbiont bearing heterotrophic dinoflagellates. Bootstrap values >50% (1000 replications) are shown. Scale indicates the number of amino acid substitutions per site. Multiple sequences clustering together are collapsed into triangles. Sample number and the individual host with the symbionts are indicated in bold with the number of sequences in brackets

13 Farnelid et al.

Table 3. Phylogenetic affiliations of the nitrogenase gene (nifH) sequences obtained from various dinoflagellate species. The % sequence identity among the sequences clustering together is indicated in brackets after the number of sequences per cluster. Below the nifH Cluster (I – III) is indicated with the phylogenetic affiliation in brackets as designated in Fig. 3.

Species Sample Accession numbersNumber of sequences per phylogenetic group Total number of ID sequences per sample Amphisolenia sp. P26 GU196890 - GU196894 5 (99.1) 5 I (β) P29 GU196895 - GU196899 4 (99.1) 1 5 I (α) ? A. bidentata P45 GU196923 - GU196930 2 (99.7) 2 (98.2) 1111 8 ? I (γ)I (γ)I (γ)I (β) III P50 GU196931 - GU196938 5 (99.7) 2 (96.3) 1 8 ?? I (α) ? P60 GU196947 - GU196949 3 (99.4) 3 I (β)* P62 GU196950 - GU196953 2 (92.7) 1 1 4 I (γ)I (β)I (β)* P63 GU196954 - GU196960 7 (99.1) 7 I (β) O. heteroporus P39 GU196918 - GU196922 3 (98.1) 2 (100) 5 II I (β) P85 GU196966 - GU196971 6 (99.4) 6 I (γ) O. magnificus P01 GU196845 - GU196850 6 (97.8) 6 I (**) P02 GU196851 - GU196855 5 (98.8) 5 I (γ) P53 GU196939 - GU196946 5 (99.4) 2 (99.4) 1 8 I (γ)I (β) ? O. quadratus P06 GU196856 - GU196862 5 (99.4) 2 (98.5) 7 I (γ)I (β) P13 GU196874 - GU196879 6 (96.9) 6 I (β) P14 GU196880 - GU196885 6 (98.8) 6 I (γ) O. steinii P07 GU196863 - GU196869 3 (99.4) 3 (99.1) 1 7 I (**) I (γ)I (γ) P08 GU196870 - GU196873 4 (99.1) 4 I (γ) P33 GU196916 - GU196917 2 (99.7) 2 I (β) O. thumii P21 GU196886 - GU196889 4 (98.1) 4 II P31 GU196900 - GU196906 7 (99.4) 7 ? P32 GU196907 - GU196915 4 (99.1) 2 (98.8) 1 1 1 9 III III III I (β)I (γ) P66 GU196961 - GU196963 3 (95.9) 3 III P67 GU196964 - GU196965 2 (99.4) 2 I (β) Negative control NTC GU196835 - GU196844 10 (98.5) 10 I (β)*

Total 137

α - β -Betaproteobacteria γ - * putative PCR reagent contaminant ** Cyanobacteria I -nifH Cluster I II -nifH Cluster II III -nifH Cluster III ? Unknown cluster related to Caldicellulosiruptor saccharolyticus ?? Unknown cluster distantly related to known nifH phylotypes

14 Farnelid et al.

DISCUSSION

The role of heterotrophic dinoflagellate symbionts has been a mystery for many years. Due to the apparent restriction of these consortia to marine waters deplete of inorganic reduced N it has been suggested that the cyanobacterial symbionts provide their hosts with N through N2 fixation (Gordon et al. 1994; Jyothibabu et al. 2006; Tarangkoon et al. 2010). In this first report of nifH genes from dinoflagellate-cyanobacteria consortia, we show that 23 of the 37 investigated dinoflagellate cells carried putative diazotrophs, and that 21 of these carried nifH genes exclusively related to heterotrophic bacteria. Hence, our analysis suggests that heterotrophic diazotrophs rather than cyanobacteria supply the dinoflagellates with reduced N.

Identification of symbionts through microscopy

The identification of symbionts of heterotrophic dinoflagellates has so far primarily been based on size, shape, pigmentation, and in some cases, ultrastructure. Heterotrophic bacterial ectosymbionts and/or cyanobacterial ectosymbionts of heterotrophic dinoflagellates have been described from the Indian, Pacific and Atlantic Oceans (Table 1). The types of ectosymbionts that we observed are similar to those previously described from the Indian Ocean (Table 1). For instance, the fairly large ectosymbionts (8 - 10 x 3 - 5 µm) in Ornithocercus magnificus and O. quadratus were similar in pigmentation, size and shape to Synechococcus carcerarius (Norris 1967, Table 1). This was supported by a molecular study targeting cyanobacterial 16S rRNA gene sequences from symbionts of eukaryotic hosts where the majority of the cyanobacterial sequences were closely related to Synechococcus (>96% similarity; Foster et al. 2006b). Further, the ectosymbionts of Histioneis carinata and H. biremis both contained at least two types of reddish cyanobacterial ectosymbionts, a large one (2.5 - 5 µm) and a smaller one (1.25 µm; Fig. 1B), similar to types III and I, respectively (Table 2 in Lucas 1991).

In accordance with previous observations from the Indian Ocean we found cyanobacterial and eukaryotic endosymbionts in Amphisolenia spp. (Table 2). Photosynthetic endosymbionts were observed in both A. bidentata (Fig. 1C) and A. thrinax. So far only one type of prokaryotic endosymbiont, S. carcerarius, has been reported; originating from A. globifera (Hallegraeff & Jeffrey 1984; Lucas 1991; Norris 1967), while eukaryotic endosymbionts have been reported from different species of Amphisolenia (Lucas 1991; Norris 1967). We observed an additional type of eukaryotic symbiont in A. bidentata. Interestingly, Foster et al. (2006b) also recovered 16S rRNA

15 Farnelid et al. genes <92% identical to eukaryotic plastids from an A. bidentata host, which could represent the eukaryotic symbionts as reported by Lucas (1991) and/or in this study (Table 2).

Heterotrophic bacteria have previously been reported as both ecto- and endosymbionts of heterotrophic dinoflagellates (Tables 1 and 2). For instance, Foster et al. (2006a) detected two morphotypes of heterotrophic bacteria associated with Ornithocercus magnificus sp. Likewise, we observed heterotrophic bacterial ectosymbionts of Ornithocercus spp. (Fig. 1A) and both cyanobacterial and heterotrophic bacterial ectosymbionts for O. magnificus and O. quadratus (Fig. 2A). In O. magnificus and O. steinii, Janson et al. (1995) observed heterotrophic bacteria between the upper and the lower girdle list of the cingular groove in all samples examined by TEM. Also, groups of heterotrophic bacterial endosymbionts have been observed in the cytoplasm of A. thrinrax and A. bidentata (Lucas 1991). Interestingly, although using primers targeting cyanobacteria, 26% of the 16S rRNA sequences recovered from eukaryotic marine hosts by Foster et al. (2006b) originated from heterotrophic bacteria; however, none of these could be directly linked to diazotrophic species (based on BLASTN search results on sequences provided by R. A. Foster). Thus, the occurrence of heterotrophic ecto- and endosymbionts of heterotrophic dinoflagellates is not unusual but to our knowledge there is no previous documentation of a N2-fixing potential in these symbionts.

Identities of nifH genes obtained from symbiont-bearing dinoflagellates

Although cyanobacterial symbionts were visible in all examined dinoflagellates only two of the 23 cells, which yielded nifH sequences, had sequences related to cyanobacteria. Intriguingly, sample P7 (O. steinii) contained nifH sequences 97% similar to the filamentous heterocystous cyanobacterium , which is known from freshwater and for its endosymbiotic associations with plants (Meeks et al. 2002). Similarly, 16S rRNA gene sequences 92% identical to Nostoc spp. were found in an A. bidentata host (Foster et al. 2006b). Taken together, these results may suggest that symbiosis facilitates the survival of Nostoc species in the saline marine environment. NifH sequences related to cyanobacteria were also obtained from sample P1 (O. magnificus) but these were only distantly related to known phylotypes (Fig. 3). Since very few nifH sequences related to cyanobacteria were found we find it unlikely that the role of the cyanobacteria in the symbiosis should be to supply the host with reduced N. Importantly, in a parallel study using the same primer sets, we detected representatives from the major groups of unicellular cyanobacteria (e.g., Crocosphaera watsonii, Cyanothece and Group A; Bergman et al. 1997; Stal &

16 Farnelid et al.

Zehr 2008). Thus, the lack of these known cyanobacteria in the present data set is not due to a primer mis-match.

Twenty-one samples contained sequences clustering in nifH Cluster I, with Alpha-, Beta- and Gamma-proteobacteria (Fig. 3, Table 3). NifH gene contamination of PCR reagents, particularly with Alpha- and Beta-proteobacterial sequences, may occur in the nested PCR (Goto et al. 2005; Zehr et al. 2003). However, the ten nifH sequences we obtained from non-visible negative control samples clustered with only four sample sequences (Fig. 3, Table 3) and were not identical to any sample sequences. Hence, reagent contamination appeared negligible in our study. However, hypothetical sources of error such as amplification of nifH genes derived from bacteria ingested by the host or from free-living bacteria which may have been passed through the three washing steps with 0.2 µm filtered seawater cannot be ruled out. In addition, as the detection limit of the nested nifH assay is unknown, samples which did not yield a nifH product could theoretically have contained putative diazotrophs.

Diverse Proteobacteria within nifH Cluster I are commonly detected in marine waters (e.g., Church et al. 2005; Hewson et al. 2007; Langlois et al. 2005; Moisander et al. 2008; Zehr et al. 1998). Associated with dinoflagellates, we found eight diverse clusters of nifH sequences related to Gammaproteobacteria while only two sequences were related to Alphaproteobacteria (Fig. 3). Interestingly, 30 sequences (22% of all sequences) were affiliated with two betaproteobacterial clusters, distinct from the negative control sequences (Fig. 3). Similarly, bacteria associated with the photosynthetic dinoflagellate Gyrodinium instriatum were dominated by Betaproteobacteria (Alverca et al. 2002). Hence, although rare in marine ecosystems (Barberán & Casamayor 2010), Betaproteobacteria appear common as symbionts of dinoflagellates.

Sequences clustering in nifH Cluster II were obtained from two samples (O. heteroporus P39 and O. thumii P21, Table 3). Mo- independent nitrogenases are present in a diverse group of diazotrophs and second alternative nitrogenases are expressed under Mo- and V- deficient conditions (Betancourt et al. 2008). Bacteria containing alternative nitrogenase genes have been isolated from diverse marine environments (Loveless et al. 1999) but interestingly anfH related genes seem to be absent in sub-tropical and tropical open waters (e.g., Church et al. 2005; Hewson et al. 2007; Langlois et al. 2005; Moisander et al. 2008; Zehr et al. 1998). Thus, the recovery of anfH related genes suggests that symbionts of dinoflagellates may be an environmental niche in open water where second alternative nifH genes can be used.

17 Farnelid et al.

Sequences from Cluster III, which contains nifH genes from anaerobic bacteria, have been detected (Church et al. 2005) but appear uncommon in the open ocean (Langlois et al. 2005; Langlois et al. 2008). The presence of Cluster III sequences in three dinoflagellates therefore suggests that the cyanobacteria-dinoflagellate consortia provide low oxygen (O2) habitats required for N2 fixation (Paerl & Prufert 1987). Similarly, Cluster III sequences from strict anaerobes and nitrogenase activity have been detected in association with zooplankton (Braun et al. 1999). However, since our Cluster III sequences were only distantly related to cultivated anaerobic bacteria (76-87% sequence identity) the they represent are rather uncertain. Survival of strict anaerobes associated with dinoflagellates would presumably require vertical inheritance of these symbionts as the dinoflagellate host divides. However, given the observed non-host specificity for the symbionts (see below), it may be more likely that the obtained Cluster III sequences derive from facultatively anaerobic bacteria. Twelve sequences originating from five samples formed a discrete nifH cluster separate from the known nifH clusters I-IV (Chien & Zinder 1996; Fig. 3). These sequences were 87-98% similar to the nifH gene of C. saccharolyticus (van de Werken et al. 2008), which can grow in the absence of reduced N (van Niel et al. 2002). It is surprising to find sequences closely related to an anaerobic extreme in the ; however, the cluster also contains nifH sequences from a marine bloom and from symbionts of . Hence, surface-associated growth in the marine environment, like in association with dinoflagellates, may be characteristic for these bacteria.

Eleven clone libraries yielded two or three different nifH sequence types per dinoflagellate. This suggests the presence of mixed assemblages of diazotrophic symbionts in host cells (Table 3), consistent with previous microscopy observations of mixed populations of cyanobionts and/or bacterial cell types in one host cell (Foster et al. 2006a). In addition, observations of several specimens of the same dinoflagellate species with diverging nifH sequences and different species of dinoflagellates hosting identical nifH sequences suggested that the nifH phylotypes were not host specific. Similar patterns of non-host specific 16S rRNA gene phylotypes were also observed for cyanobacterial symbionts in ciliates, dinoflagellates, and radiolarians (Foster et al. 2006b). In contrast, in the Richelia intracellularis- symbiosis a divergence of hetR and nifH sequences of symbionts from different host species was interpreted as an indication of host specificity (Janson et al. 1999, Foster & Zehr 2006). Thus, it appears that at any one time dinoflagellate hosts may contain multiple symbionts but the low degree of specificity also indicates that their dependence on specialized symbionts is not fundamental.

18 Farnelid et al.

Putative ecological roles of the consortia

In dinoflagellate-cyanobacteria consortia the host’s requirement for fixed carbon as well as N is presumably the driving force for the relationship. Our results show that heterotrophic bacterial symbionts rather than cyanobionts have the genetic potential for fixing N2. Consequently, we speculate that the widespread, and somewhat counter intuitive distribution of these large (50 - 1000 µm) species of heterotrophic dinoflagellates in the oligotrophic subtropical and tropical oceans is partly made possible by symbiont-mediated photosynthesis (cyanobacteria) and N2 fixation (heterotrophic bacteria).

Acknowledgements

Danish Galathea expedition and the Captain of HMDS ‘Vædderen’, Carsten Smidt, and his crew are thanked for excellent assistance in connection with sampling. We thank R.A. Foster for generously providing unpublished sequence data. The project was supported by grants from Knud Højgaards Fond, Danish Natural Sciences Research Council (272-05-0333 and 272-06-0485 to P.J.H. and 277- 05-0421 to G.H.) and Dansk Expeditions fond. The work of H.F. was supported by the Swedish Research Council FORMAS (217-2006-342 to L.R.). The present work was carried out as part of the Galathea 3 expedition under the auspices of the Danish Expedition Foundation. This is Galathea 3 contribution no. xx.

LITERATURE CITED

1. Alverca E, Biegala IC, Kennaway GM, Lewis J, Franca S (2002) In situ identification and localization of bacteria associated with Gyrodinium instriatum (, Dinophyceae) by electron and confocal microscopy. Eur J Phycol 37:523-530

2. Barberán A, Casamayor EO (2010) Global phylogenetic community structure and β−diversity patterns in surface metacommunities. Aquat Microb Ecol 59:1-10

3. Bergman B, Gallon JR, Rai AN, Stal LJ (1997) N2-fixation by non-heterocystous cyanobacteria. FEMS Microbiol Rev 19:139-185

19 Farnelid et al.

4. Betancourt DA, Loveless TM, Brown JW, Bishop PE (2008) Characterization of Diazotrophs Containing Mo-Independent Nitrogenases, Isolated from Diverse Natural Environments. Appl Environ Microbiol 74:3471-3480

5. Biegala IC, Raimbault P (2008) High abundance of diazotrophic picocyanobacteria (<3 μm) in a Southwest Pacific coral lagoon. Aquat Microb Ecol 51:45-53

6. Braun ST, Proctor LM, Zani S, Mellon MT, Zehr JP (1999) Molecular evidence for zooplankton-associated nitrogen-fixing anaerobes based on amplification of the nifH gene. FEMS Microbiol Ecol 28:273-279

7. Capone DG, Zehr JP, Paerl HW, Bergman B, Carpenter EJ (1997) Trichodesmium, a Globally Significant Marine Cyanobacterium. Science 276:1221-1229

8. Carpenter E, Montoya JP, Burns JA, Mulholland MR, Subramaniam A, Capone DG (1999)

Extensive bloom of a N2-fixing diatom/cyanobacterial association in the tropical Atlantic Ocean. Mar Ecol Prog Ser 185:273-283

9. Chien Y-T, Zinder SH (1996) Cloning, functional organization, transcript studies, and phylogenetic analysis of the complete nitrogenase stuctural genes (nifHDK2) and associated genes in the Archaeon Methanosarcina barkeri 227. J Bacteriol 178:143-148

10. Church MJ, Jenkins BD, Karl DM, Zehr JP (2005) Vertical distributions of nitrogen-fixing phylotypes at Stn ALOHA in the oligotrophic North Pacific Ocean. Aquat Microb Ecol 38:3- 14

11. Farnelid H, Riemann L (2008) Heterotrophic N2-fixing Bacteria: Overlooked in the Marine Nitrogen Cycle? In: Couto GN (ed) Nitrogen Fixation Research Progress. Nova Science Publishers Inc., New York

12. Foster RA, Carpenter EJ, Bergman B (2006a) Unicellular cyanobionts in open ocean dinoflagellates, radiolarians, and tintinnids: Ultrastructure characterization and immuno- localization of phycoerythrin and nitrogenase. J Phycol 42:453-463

13. Foster RA, Collier JL, Carpenter EJ (2006b) Reverse transcription PCR amplification of cyanobacterial symbiont 16S rRNA sequences from single non-photosynthetic eukaryotic marine planktonic host cells. J Phycol 42:243-250

20 Farnelid et al.

14. Foster RA, Zehr JP (2006) Characterization of diatom-cyanobacteria symbioses on the basis of nifH, hetR and 16S rRNA sequences. Env Microbiol 8:1913-1925

15. Gordon N, Angel DL, Neorl A, Kress N, Kimor B (1994) Heterotrophic dinoflagellates with symbiotic cyanobacteria and nitrogen limitation in the Gulf of Aqaba. Mar Ecol Prog Ser 107:83-88

16. Goto M, Ando S, Hachisuka Y, Yoneyama T (2005) Contamination of diverse nifH and nifH- like DNA into commercial PCR primers. FEMS Microbiol Lett 246:33-38

17. Hallegraeff GM, Jeffrey SW (1984) Tropical Phytoplankton Species and Pigments of Continental Shelf Waters of North and North-West Australia. Mar Ecol Prog Ser 20:59-74

18. Hewson I, Moisander P, Achilles K, Carlson C, Jenkins BD, Mondragon E, Morrison AE, Zehr JP (2007) Characteristics of diazotrophs in surface to abyssopelagic waters of the Sargasso Sea. Aquat Microb Ecol 46:15-30

19. Janson S, Carpenter EJ, Bergman B (1995) Immunolabelling of phycoerythrin, ribulose 1,5- biphosphate carboxylase/oxygenase and nitrogenase in the unicellular cyanobionts of Ornithocercus spp. (Dinophyceae). Phycologia 34:171-176

20. Janson S, Wouters J, Bergman B, Carpenter EJ (1999) Host specificity in the Richelia-diatom symbiosis revealed by hetR gene sequence analysis. Env Microbiol 1:431-438

21. Jyothibabu R, Madhu NV, Maheswaran PA, Asha Devi CR, Balasubramanian T, Nair KKC, Achuthankutty CT (2006) Environmentally-related seasonal variation in symbiotic associations of heterotrophic dinoflagellates with cyanobacteria in the western Bay of Bengal. Symbiosis 42:51-58

22. Langlois RJ, Hümmer D, La Roche J (2008) Abundances and Distributions of the Dominant nifH Phylotypes in the Northern Atlantic Ocean. Appl Environ Microbiol 74:1922-1931

23. Langlois RJ, LaRoche J, Raab PA (2005) Diazotrophic diversity and distribution in the tropical and subtropical Atlantic Ocean. Appl Environ Microbiol 71:7910-7919

24. Le Moal M, Biegala IC (2009) Diazotrophic unicellular cyanobacteria in the northwestern Mediterranean Sea: A seasonal cycle. Limnol Oceanogr 54:845-855

21 Farnelid et al.

25. Loveless TM, Saah JR, Bishop PE (1999) Isolation of Nitrogen-Fixing Bacteria Containing -Independent Nitrogenases from Natural Environments. Appl Environ Microbiol 65:4223-4226

26. Lucas IAN (1991) Symbionts of the tropical dinophysiales. Ophelia 33:213-224

27. Meeks J, Campbell E, Summers M, Wong F (2002) Cellular differentiation in the cyanobacterium Nostoc punctiforme. Arch Microbiol 178:395-403

28. Moisander PH, Beinart RA, Voss M, Zehr JP (2008) Diversity and abundance of diazotrophic in the South China Sea during intermonsoon. ISME J 2:954-967

29. Montoya JP, Holl CM, Zehr JP, Hansen A, Villareal TA, Capone DG (2004) High rates of N2 fixation by unicellular diazotrophs in the oligotrophic Pacific Ocean. Nature 430:1027-1031

30. Norris RE (1967) Algal consortisms in marine plankton. In: V.Kirhnamurthy (ed) Proceedings of a seminar on Sea, Salt and Plants. Central Salt and Marine Chemicals Research Institute, Bharnagar

31. Paerl HW, Prufert LE (1987) Oxygen-poor microzones as potential sites of microbial N2 fixation in nitrogen-depleted aerobic marine waters. Appl Environ Microbiol 53:1078-1087

32. Riemann L, Leitet C, Pommier T, Simu K, Holmfeldt K, Larsson U, Hagström Å (2008) The Native Bacterioplankton Community in the Central Baltic Sea Is Influenced by Freshwater Bacterial Species. Appl Environ Microbiol 74:503-515

33. Schütt F (1895) Die Peridineen der Plankton-Expedition.-Ergebn. Plankton-Expedition der Humboldt Stiftung 4:1-170

34. Sebastian CR, O'Ryan C (2001) Single-cell sequencing of dinoflagellate (Dinophyceae) nuclear ribosomal genes. Mol Ecol Notes 1:329-331

35. Stal LJ, Zehr JP (2008) Cyanobacterial Nitrogen Fixation in the Ocean: Diversity, Regulation, and Ecology. In: Herrero A, Flores E (eds) The Cyanobacteria , and Evolution. Caister Academic Press, Norfolk, UK

36. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Analysis (MEGA) software version 4.0. Mol Biol Evol 24:1596-1599

22 Farnelid et al.

37. Tarangkoon W, Hansen G, Hansen P (2010) Spatial distribution of symbiont-bearing dinoflagellates in the Indian Ocean in relation to oceanographic regimes. Aquat Microb Ecol 58:197-213

38. Taylor FJR (1982) Symbioses in marine microplankton. Ann Inst Oceanogr 58:61-90

39. van de Werken HJG, Verhaart MRA, VanFossen AL, Willquist K, Lewis DL, Nichols JD, Goorissen HP, Mongodin EF, Nelson KE, van Niel EWJ, Stams AJM, Ward DE, de Vos WM, van der Oost J, Kelly RM, Kengen SWM (2008) Hydrogenomics of the Extremely Thermophilic Bacterium Caldicellulosiruptor saccharolyticus. Appl Environ Microbiol 74:6720-6729

40. van Niel EWJ, Budde MAW, de Haas GG, van der Wal FJ, Claassen PAM, Stams AJM (2002) Distinctive properties of high producing extreme , Caldicellusiruptor saccharolyticus and Thermotoga elfii. Int J Hydrogen Energ 27:1391-1398

41. Zani S, Mellon MT, Collier JL, Zehr JP (2000) Expression of nifH genes in natural microbial assemblages in Lake George, New York, detected by PCR. Appl Environ Microbiol 66:3119-3124

42. Zehr JP, Crumbliss LL, Church MJ, Omoregie EO, Jenkins BD (2003) Nitrogenase genes in PCR and RT-PCR reagents: implications for studies of diversity of functional genes. BioTechniques 35:996-1005

43. Zehr JP, McReynolds LA (1989) Use of degenerate oligonucleotides for amplification of the nifH gene from the marine cyanobacterium Trichodesmium thiebautii. Appl Environ Microbiol 55:2522-2526

44. Zehr JP, Mellon MT, Zani S (1998) New nitrogen-fixing microorganisms detected in oligotrophic oceans by amplification of nitrogenase (nifH) genes. Appl Environ Microbiol 64:3444-3450

45. Zehr JP, Turner PJ (2001) Nitrogen Fixation: Nitrogenase Genes and . In: Paul JH (ed) Methods in Microbiology. Academic Press, New York

46. Zehr JP, Waterbury JB, Turner PJ, Montoya JP, Omoregie E, Steward GF, Hansen A, Karl

DM (2001) Unicellular cyanobacteria fix N2 in the subtropical North Pacific Ocean. Nature 412:635-638

23 Appendix-I

Contributions in paper

Paper I II III Appendix-IV Original idea PJH, WT PJH, WT PJH,WT, MM PJ, WT Study design PJH, WT PJH, WT PJH,WT, MM PJ, WT, HF, LR and method TEM work GH, WT - ØM GH,WT Molecular - - LG HF, LR Work Data gathering WT WT PJ, WT, MM, HF ØM, LG Writing WT, PJ, GH WT,PJ WT, PJ, MM, HF, LR, PJ, ØM, LG WT, GH

Woraporn Tarangkoon (WT), Per Juel Hansen (PJH), Gert Hansen (GH), Morten Moldrup (MM), Øjvind Moestrup (ØM), Lydia Garcia-Cuetos (LG), Hanna Farnelid (HF), Lasse Riemann (LR)