MIAMI UNIVERSITY the Graduate School Certificate for Approving the Dissertation

Home , Bicosoecida, Blood Falls, Cryptomonadales, Glaciecola punicea, Isochrysidales, Lake Bonney (Antarctica)

MIAMI UNIVERSITY

The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation of Wei Li

Candidate for the Degree:

Doctor of Philosophy

Dr. Rachael Morgan-Kiss, Director

Dr. Annette Bollmann, Reader

Dr. Thomas Crist, Reader

Dr. Michael Vanni, Reader

Dr. Richard Edelmann, Graduate School Representative

Dr. Rebecca Gast, Distinguished Off Campus Scholar

ABSTRACT

INFLUENCE OF ENVIRONMENTAL DRIVERS AND INTERACTIONS ON THE MICROBIAL COMMUNITY STRUCTURE IN PERMANENTLY STRATIFIED MEROMICTIC ANTARCTIC LAKES

by Wei Li

The microbial loop plays important roles in the cycling of energy, carbon and elements in aquatic ecosystems. Viruses, bacteria, Archaea and microbial eukaryotes are key players in global carbon cycle and biogeochemical cycles. Investigating microbial diversity and community structure is crucial first step for understanding the ecological functioning in aquatic environment. Meromictic lakes are bodies of water and exhibit permanent stratification of major physical and chemical environmental factors. Microbial consortia residing in permanently stratified lakes exhibit relatively constant spatial stratification throughout the water column and are adapted to vastly different habitats within the same water. Pristine perennially-ice-covered lakes (Lake Bonney, Lake Fryxell and Lake Vanda) are meromictic lakes located in the McMurdo Dry Valleys (MDV) of Southern Victoria Land, Antarctica. The lakes have isolated water bodies and extremely stable strata that vary physically, chemically, and biologically within and between the water columns. The unique characteristics support microbially dominated food webs in these lakes.

In the research presented here, we gathered new understanding of how environmental drivers influence microbial community structure in these aquatic ecosystems. We explored the lake microbial ecology from three major approaches: 1). Assess trophic activities in the natural environment and identify potential environmental drivers impacting heterotrophic (β Glucosaminidase) and autotrophic (Ribulose 1,5 bisphosphate carboxylase) enzyme activities; 2). Resolve the protist community composition (i.e. autotrophic, heterotrophic and mixotrophic groups) based on high throughput sequencing and bioinformatics. Identify how the community structures correlate with specific environmental and biological factors; 3). Reveal the diversity of potential microbial interactions between the microorganisms in the MDV lakes at individual cell level, and investigate how the interactions vary between organisms with different nutritional strategies.

Studies of polar microbial communities on the cusp of environmental change will be important for predicting how microbial communities in low latitude aquatic systems will respond. This study expands the understanding of how environmental drivers interact with microbial communities in the Antarctica lakes, and provide new information to predict how the community structure will alter as response to climate changes. INFLUENCE OF ENVIRONMENTAL DRIVERS AND INTERACTIONS ON THE MICROBIAL COMMUNITY STRUCTURES IN PERMANENTLY STRATIFIED MEROMICTIC ANTARCTIC LAKES

A DISSERTATION

Submitted to the faculty of Miami University In partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Microbiology Ecology, Evolution and Environmental Biology Program

Wei Li Miami University Oxford, OH 2016 Advisor: Dr. Rachael Morgan-Kiss

TABLE OF CONTENTS

PAGE List of Tables v List of Figures vi Acknowledgments vii

Chapter I: Introduction 1.1. introduction 2

Chapter II: Spatial heterogeneity and the impact of biotic and abiotic drivers on microbial autotrophic and heterotrophic activities in three chemically stratified Antarctic lakes 2.1. Introduction 16 2.2. Materials and methods 20 2.2.1. Field sampling and limnological parameters 20 2.2.2. Rubisco carboxylase activity assay 21 2.2.3. Glucosaminidase activity assay 21 2.2.4. Protein concentration determination 22 2.2.5. Bacterial enumeration 22 2.2.6. Statistical analyses 23 2.3. Results 24 2.3.1. Lake Bonney 24 2.3.2. Lake Fryxell 25 2.3.3. Lake Vanda 25 2.3.4. Cluster analyses of lake physicochemical parameters 26 2.3.5. Correlation of autotrophic and heterotrophic activities with physicochemical parameters 26 2.4. Discussion 28 2.5. Acknowledgements. 32 2.6. References 41

Chapter III: Influence of environmental drivers and potential interactions on the distribution of microbial communities from three permanently stratified Antarctic lakes 3.1. Introduction 45 3.2. Materials and methods 48 3.2.1. Site description and sample collection 48 3.2.2. Environmental parameter measurements 48 3.2.3. DNA library preparation and Illumina MiSeq Sequencing 49 3.2.4. Sequence analysis 49 3.2.5. Diversity assessment 50 3.2.6. Network analysis 50 3.3. Results 52 3.3.1. Environmental parameters 52 3.3.2. Ilumina MiSeq sequencing summary 52

iii 3.3.3. Bacterial community composition 53 3.3.4. Eukaryotic community composition 54 3.3.5. Co-occurrence microbial network and associated environmental factors 55 3.4. Discussion 57 3.5. Acknowledgements. 61 3.6. References 76

Chapter IV: Ultrastructural and single-cell level characterization reveals metabolic versatility in a microbial eukaryote community from an ice-covered Antarctic lake 4.1. Introduction 87 4.2. Materials and methods 90 4.2.1. Site description, sample collection and enrichment cultures 90 4.2.2. Single cell sorting 91 4.2.3. DNA template preparation for PCR 91 4.2.4. Sanger sequencing 92 4.2.5. Illumina sequencing 92 4.2.6. Analysis of the sequences 93 4.2.7. Confocal Laser Scanning Microscopy 93 4.2.8. Scanning Electron Microscopy 94 4.3. Results and discussion 95 4.3.1. Identities and trophic modes of sorted eukaryotes 95 4.3.2. Isolation and description of two key photosynthetic protists 98 4.3.3. Community composition of organisms co-sorted with Lake Bonney eukaryotes 99 4.3.4. Potential interactions between Dry Valley protists and bacteria 100 4.4. Conclusions 105 4.5. Acknowledgements 105 4.6. References 112

Chapter V: Conclusions 5.1 Conclusions 127 5.2 References 133

LIST OF TABLES

PAGE 1. Pearson correlation coefficient values (R) for average RubisCO and βGAM activity with lake physical, chemical, and biological parameters for all samples 39 2. Linear regression models explaining the enzyme activities in different portions of water columns 40 3. Summary of major physicochemical parameters in the studied lakes 68 4. Pairwise comparison of species richness between lakes using Tukey’s HSD test 69 5. ANOSIM of communities from different lakes and layers 70 6. List of nodes in co-occurrence network 72 7. Basic growth physiology of Chlamydomonas sp. ICE-MDV and Isochrysis sp. MDV isolates 121 8. Diversity of major algal classes in Lake Bonney enrichment cultures 122

v LIST OF FIGURES

PAGE 1. Schematic of Microbial loop and energy flows in aquatic environment 7 2. Map showing locations of study sites within the McMurdo Dry Valleys, Antarctica 8 3. Depth profiles of physical and chemical characteristics for the MDV lakes in this study 9 4. Map showing locations of study sites within the McMurdo Dry Valleys, Antarctica 33 5. Salinity profiles for Lake Bonney (east lobe, ELB; west lobe, WLB), Lake Fryxell (FRX) and Lake Vanda (VAN) 34 6. Depth profiles of physical and chemical characteristics for the three lakes in this study 35 7. Depth profiles for phytoplankton biomass (Chl a) and bacterial abundance for study sites 36 8. Spatial variability in enzyme activity of RubisCO and β-Glucosaminidase (B-GAM) in study sites 37 9. Hierarchical clustering analysis (HCA) and principal component analysis (PCA) 38 10. Observed and estimated alpha diversity of 16S and 18S OTUs. 62 11. 16S OTUs relative abundance at phylum level 63 12. 18S OTUs relative abundance at phylum level 64 13. NMDS plot of bacterial communities 65 14. NMDS plot of eukaryal communities 66 15. Association network of concurrent of bacteria, eukaryotes and correlation with environmental parameters 67 16. Rarefaction curves of 16S and 18S OTUs 84 17. Representive SEM micrographs of protists found in studied lakes 85 18. Sample cytogram showing flow cytometric sort regions for microbial eukaryotes from an enrichment culture (MDV87) generated from the Antarctic Dry Valley Lake Bonney. 106 19. Maximum-likelihood tree (1000 bootstrap) showing identity of Lake Bonney microbial eukaryote single amplified genomes (EUK-SAGS) recovered from enrichment culture MDV87 107 20. Confocal microscopic images of Chlamydomonas sp. ICE-MDV and Isochrysis sp. MDV isolates 108 21. Principle Coordinates Analysis (PCoA) 109 22. SEM micrographs and the diversity of microbial partners associated with the heterotrophic nanoflagellate Pteridomonas 110 23. Evidence of parasite-host interactions between Lake Bonney microbial eukaryotes. a, Diversity of 16S rRNA gene OTUs recovered from Pirsonia EUK-SAGs 111 24. Confocal microscopic images of Isochrysis sp. MDV isolates without LysoTracker Green staining 123 25. Confocal microscopic images of Chlamydomonas sp. ICE-MDV isolates without LysoTracker Green staining 124 26. Heat map of communities associated with sorted eukaryotic organisms based on 16s rRNA sequence abundance 125 27. Working models of our current understanding of major carbon and energy cycles in Lake Bonney (west) vs. Lake Fryxell 132

vi Acknowledgements

I would like to express my sincere appreciation and thanks to my mentor and Team Protist leader, Dr. Rachael Morgan-Kiss, who has been extremely supportive to guide me through the journey in pursuit of my Ph. D. degree. You have been so generous and patient through the years, and taught me not just much of what I know about science with your knowledge and wisdom, but more importantly, how to think and solve scientific puzzles. I appreciate the guidance and freedom that you have given to me during my research in your lab. You have provide me with so many opportunities including attending conferences, building a publication record, and the amazing journeys to Antarctica Thank you for opening the door for me to be a scientist in polar environmental science. Your advice on both research and my career have been invaluable to me.

I would also like to thank my committee members, Drs. Annette Bollmann, Thomas Crist, Richard Edelmann, Rebecca Gast and Michael Vanni for the support, brilliant comments and suggestions and even hardship.

Thank you to my lab mates: Amber, Greg, and Isha and past lab mates: Jenna, Nick, and Sarah, as well as a special Team Protist member Chris for all of your support, help, friendship and “craziness”. A special thank you to Drs. Jenna Dolhi, Amber Teufel and Chris Sedlacek for your continued support and friendship, and for the tremendous effort in the fieldwork that we executed in Antarctica, and for being part of the amazingness of these unforgettable journeys. To the rest of my fellow graduate students who have become dear friends; you have made this more than just a “work place”. Your camaraderie and support have been more than I could have asked for and I think I owe my sanity to you all. You listened when I spoke, gave advice when I asked,

The Microbiology Department as a whole has been an amazing place to learn to be a scientist. I appreciate the opportunities to teach, attend meetings, receive achievement awards, and network with other scientists. I am also indebted to Barb Stahl, Darlene Davidson, Bev Scaggs and Amy Corrington who always have the answers.

Finally, I would like to give my greatest gratefulness to my wife, Jing. I understand how hard you have been through when I was away for field works. Thank you for giving me a place called home, and for your unconditional support, so I can continue chasing my dreams.

vii CHAPTER I

Introduction

Wei Li, Rachael M. Morgan-Kiss

CHAPTER I. 1.1.Introduction

More than 70% of the Earth’s surface is covered by water bodies, and planktonic microorganisms dominate these aquatic ecosystems in terms of both species and biomass (Auguet et al 2010, Brown et al 2009, Šlapeta et al 2005, Whitman et al 1998). The importance of microbes in such systems has been widely recognized ever since interactions between organic matter and microbial plankton were put into a hypothetical context (Pomeroy 1974) and the term “microbial loop” was coined to describe linkages between phytoplankton, heterotrophic bacteria and protozoa (Azam et al 1983) (Figure 1.1). As a significant portion of primary production is transformed into dissolved organic matter (DOM) through extracellular releasing mechanisms (intentional release or through lysis and death), this DOM is utilized by heterotrophic bacteria, and microbial eukaryotes eventually graze on smaller autotrophic cells and heterotrophic bacteria (Azam et al 1983). The microbial loop forms complex interactions between microbial Eukaryotes, Bacteria, Archaea and Viruses and linkages between primarily produced organic matters reaching larger organisms (Azam et al 1983, Azam et al 1994, Fenchel 2008, Wilhelm and Suttle 1999). In pelagic ecosystems, microbial loop contributes predominantly to carbon and nutrient cycling (Field et al 1998, Pernthaler 2005b, Pomero et al 2007).

Single-celled eukaryotic microorganisms (i.e., protists) are ubiquitous in every ecosystem on earth and play critical ecological roles in food web dynamics and global carbon and nutrient cycles (Montagnes et al 2012). Diverse protist lineages possess multiple nutritional modes; playing key roles as producers, decomposers, parasites, and predators. Phototrophic protists are important producers that contribute to global primary production and incorporate a significant portion of inorganic carbon into the food web (Caron et al 2009). Predatory protists are major consumers of planktonic phytoplankton and bacteria, providing control over the abundance of these organisms as well as linking primary producers/consumers with higher trophic levels (i.e., metazoans) (Sherr and Sherr 2002a). Mixotrophic protists (combined ability for photosynthesis and phagotrophic ingestion of food particles) are widespread in aquatic ecosystems (Moorthi et al 2009a, Sanders et al 2000b) and providing additional organic carbon sources under light- limiting conditions (Caron et al 1990). Complex relationships form between protists and bacteria,

2 including commensalism (e.g., dependence of bacteria on photosynthetic carbon from the photosynthetic organisms or endosymbionts ) and predation or pathogenesis (e.g., bacteria as prey for the heterotrophic flagellates or pathogens infect host cells ) (Amin et al 2012, Faust and Raes 2012a, Ferrantini et al 2009, Medina-Sánchez et al 2004).

- - + 3- nutrient levels (NO2 , NO3 , NH4 , and PO4 ) at and below the permanent chemoclines. Irradiance is the main energy source for light-driven production of organic carbon in the lakes; however, it does not saturate photosynthesis and is a major limiter of yearly production (Priscu 1995). Despite these commonalities in physical and chemical characteristics, each water column exhibits a unique physicochemical signature (Lyons et al 2000b) (Major physicochemical parameter profiles of studied lakes are shown in Figure 1.3).

The studied lakes support microbially dominated and truncated (lack metazoans) food webs, with the exception of low numbers of copepods and rotifers in some lakes (Laybourn-Parry and Pearce 2007). The majority of dissolved organic carbon is autochthonous (i.e., derived from new photosynthetic activity within the water column). Protists represent the major producers of

3 organic matter in the MDV lake food web (Morgan-Kiss et al 2006, Neale and Priscu 1995), while heterotrophic nanoflagellates and ciliates are the top predators of bacteria and smaller protists (Roberts and Laybourn-Parry 1999, Roberts et al 2004c). Despite the energetic cost of maintaining and regulating both photosynthetic and heterotrophic cellular apparatus, mixotrophy appears to be very prevalent in MDV aquatic food webs (Bell and Laybourn-Parry 2003, Roberts and Laybourn-Parry 1999). Mixotrophic metabolism is likely a survival strategy for alternate nutrient acquisition in extreme oligotrophic conditions as well as energy during the winter (Laybourn-Parry 2002).

Studies in other regions of Antarctica suggested that environmental drivers, in particular, salinity, oxygen and light may play critical roles in the biogeography of lake microorganisms (Cavicchioli 2015b). Heterogeneous distribution (i.e. vertical distribution) of microbes in the water columns has been observed in several studies in MDV lakes (refs). For example, recent applications of 18S rRNA sequencing in Lake Bonney provided the first insight into the phylogenetic diversity of the microbial eukaryote populations, 85% of which was related to photosynthetic species (Bielewicz et al 2011b). The water column of Lake Bonney is dominated by chlorophytes in the shallow layers and haptophytes in the chemocline, while Lake Fryxell is dominated by cryptophytes throughout the water column (Bielewicz et al 2011b, Dolhi et al 2015b, Kong et al 2012c). The water column of Lake Vanda harbors cyanobacteria and nanoplankton related to nonflagellated chlorophytes (Dolhi et al 2015b). Despite that these studies have begun to describe the biogeographic distribution of protists from these extreme and isolated aquatic systems, the question whether the diversity and distribution of microorganisms, especially protists are driven by specific environmental factor(s) has not been examined and a full understanding of the trophic abilities of MDV aquatic protists as well as potential interactions with other MDV microorganisms is currently lacking.

The Calvin-Benson-Bassham (CBB) cycle represents a major pathway for autotrophic carbon dioxide fixation in plants, most phytoplankton (eukaryotic algae and cyanobacteria) and many chemolithoautotrophic bacteria. The enzyme Ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) catalyzes the first and rate-limiting reaction, that is condensation of a 5-carbon sugar, ribulose-1,5-bisphosphate (RuBP) with either CO2 (carboxylation) or O2 (oxygenation or

4 photorespiration). Products of RubisCO carboxylation are reduced to sugars which are incorporated into cellular biomass (Tabita et al 2008). The planktonic autotrophic activity therefore can be measured using the proxy of in situ RubisCO activity (Dolhi et al 2012). In contrast, unicellular, phagotrophic protists rely on grazing of bacteria/small phytoplankton or particulate organic matter for energy, carbon and nutrient sources. Prey are sequestered into food vacuoles and large levels of digestive enzymes are released from lysosomes into the food vacuoles to rapidly hydrolyze the organic compounds (Vrba et al 1996). Planktonic heterotrophic activity in aquatic environments can be detected and quantified based on specific lysozyme activity associated with β- D- glucosaminidase enzymatic activity (i.e., βGAM) (Zubkov and Sleigh 1998). Therefore, hydrolytic activity of phagotrophic protists can be specifically measured based on the activity of hydrolysis enzymes extracted from lysosomes of protists in low condition, and this measurement can be used as a diagnostic tool for estimating heterotrophic protist activity (Zubkov and Sleigh 1998). These two enzymes can be used as proxies to access the autotrophic and heterotrophic activities of eukaryotic microbial communities and potentially yield rate measurement of these two critical processes in terms of carbon and nutrient cycling in the aquatic ecosystems.

In order to study the composition, organization and spatiotemporal patterns of microbial communities, small subunit rRNA gene amplicon sequencing (16S rRNA gene in Bacteria and Archaea or 18S rRNA gene in Eukarya) is a widely applied approach (Olsen et al 1986). During the last decade, high-throughput sequencing technology based on next generation sequencing platforms (NGS) and the application of barcode indexing are allowing the collection of thousands to ten thousands of sequences from a large number of samples simultaneously (Caporaso et al 2010, Caporaso et al 2011). Compared to the “old-fashion” methods (i.e., TRFLP or ARISA or clone library in combination with Sanger sequencing) to analyze 16S rRNA gene amplicons (Fisher and Triplett 1999, Liu et al 1997), the NGS approaches provide significant higher sequencing depth which generate deeper insights into the diversity of microbial communities(Eiler et al 2012b, Herlemann et al 2011, Sogin et al 2006). Employment of this technique to study the MDV lakes would potentially expand our understanding these unique ecosystems more thoroughly (Cavicchioli 2015b, Vick-Majors et al 2014a).

The overall goal of this project is to gather new understanding of how environmental drivers influence microbial community structure in ice-covered aquatic ecosystems. The following objectives will be addressed: (1) Assess trophic activities in the natural environment and identify potential environmental drivers impacting heterotrophic (β Glucosaminidase) and autotrophic (Ribulose 1,5 bisphosphate carboxylase) enzyme activities. (2) Resolve the protist community composition (i.e. autotrophic, heterotrophic and mixotrophic groups) and its correlation with specific environmental and biological factors. (3) Identify the nutritional mode and metabolic potential of key MDV protists. (4) Reveal the diversity of potential microbial interactions between the microorganisms in the MDV lakes, and investigate differential interactions among organisms with various nutritional strategies.

Figure 1.1. Schematic of Microbial loop and energy flows in aquatic environment. Adapted from (Okuda et al 2014)

Figure 1.2. Map showing locations of study sites within the McMurdo Dry Valleys, Antarctica: Lake Bonney (red arrow), Lake Fryxell (yellow arrow) and Lake Vanda (orange arrow). Map was adapted from Landsat Image Mosaic of Antarctica (LIMA) project from the Antarctic Geospatial Information Center (http://lima.usgs.gov) with courtesy of NASA, BAS, NSF and USGS.

Figure 1.3. Depth profiles of physical and chemical characteristics for the MDV lakes in this study. PAR: photosynthetically active radiation; DIN: dissolved inorganic nitrogen. Note that

PAR for Lake Vanda and PO4 for Lakes Vanda and Fryxell are plotted on secondary x-axes.

1.2. Reference

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76: 667-684.

Auguet J-C, Barberan A, Casamayor EO (2010). Global ecological patterns in uncultured Archaea. The ISME journal 4: 182-190.

Azam F, Fenchel T, Field J, Gray J, Meyer-Reil L, Thingstad F (1983). The ecological role of water-column microbes in the sea. Marine ecology progress series Oldendorf 10: 257-263.

Azam F, Smith D, Steward G, Hagström Å (1994). Bacteria-organic matter coupling and its significance for oceanic carbon cycling. Microbial Ecology 28: 167-179.

Bell EM, Laybourn-Parry J (2003). Mixotrophy in the Antarctic phytoflagellate Pyramimonas gelidicola. J Phycol 39: 644-649.

Bielewicz S, Bell EM, Kong W, Friedberg I, Priscu JC, Morgan-Kiss RM (2011). Protist diversity in a permanently ice-covered Antarctic lake during the polar night transition. ISME J 5: 1559-1564.

Brown MV, Philip GK, Bunge JA, Smith MC, Bissett A, Lauro FM et al (2009). Microbial community structure in the North Pacific ocean. The ISME journal 3: 1374-1386.

Campbell AG, Campbell JH, Schwientek P, Woyke T, Sczyrba A, Allman S et al (2013). Multiple single-cell genomes provide insight into functions of uncultured Deltaproteobacteria in the human oral cavity. PloS one 8: e59361.

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK et al (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods 7: 335- 336.

Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA, Turnbaugh PJ et al (2011). Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc Natl Acad Sci U S A 108: 4516-4522.

Caron DA, Porter KG, Sanders RW (1990). Carbon, nitrogen, and phosphorus budgets for the mixotrophic phytoflagellate Poterioochromonas malhamensis (Chrysophyceae) during bacterial ingestion. Limnology and Oceanography 35: 433-443.

Caron DA, Worden AZ, Countway PD, Demir E, Heidelberg KB (2009). Protists are microbes too: a perspective. The ISME journal 3: 4-12.

Cavicchioli R (2015). Microbial ecology of Antarctic aquatic systems. Nat Rev Microbiol 13: 691-706.

Chela-Flores J (2011). On the possibility of biological evolution on the moons of Jupiter. The Science of Astrobiology. Springer. pp 151-170.

Dolhi JM, Ketchum N, Morgan-Kiss RM (2012). Establishment of microbial eukaryotic enrichment cultures from a chemically stratified antarctic lake and assessment of carbon fixation potential. J Vis Exp 62: e3992.

Dolhi JM, Teufel AG, Kong W, Morgan-Kiss RM (2015). Diversity and spatial distribution of autotrophic communities within and between ice-covered Antarctic lakes (McMurdo Dry Valleys). Limnol Oceanogr 60: 977-991.

Eiler A, Heinrich F, Bertilsson S (2012). Coherent dynamics and association networks among lake bacterioplankton taxa. The ISME journal 6: 330-342.

Faust K, Raes J (2012). Microbial interactions: from networks to models. Nat Rev Microbiol 10: 538-550.

Fenchel T (2008). The microbial loop–25 years later. Journal of Experimental Marine Biology and Ecology 366: 99-103.

Ferrantini F, Fokin SI, Modeo L, Andreoli I, Dini F, Gortz HD et al (2009). "Candidatus Cryptoprodotis polytropus," a novel Rickettsia-like organism in the ciliated protist Pseudomicrothorax dubius (Ciliophora, Nassophorea). J Eukaryot Microbiol 56: 119-129.

Field CB, Behrenfeld MJ, Randerson JT, Falkowski P (1998). Primary production of the biosphere: integrating terrestrial and oceanic components. Science 281: 237-240.

Fisher MM, Triplett EW (1999). Automated approach for ribosomal intergenic spacer analysis of microbial diversity and its application to freshwater bacterial communities. Applied and environmental microbiology 65: 4630-4636.

Herlemann DP, Labrenz M, Jürgens K, Bertilsson S, Waniek JJ, Andersson AF (2011). Transitions in bacterial communities along the 2000 km salinity gradient of the Baltic Sea. The ISME journal 5: 1571-1579.

Kong W, Ream DC, Priscu JC, Morgan-Kiss RM (2012). Diversity and expression of RubisCO genes in a perennially ice-covered Antarctic lake during the polar night transition. App Env Microbiol 78: 4358-4366.

Laybourn-Parry J (2002). Survival mechanisms in Antarctic lakes. Philos Trans R Soc Lond B Biol Sci 357: 863-869.

Laybourn-Parry J, Pearce DA (2007). The biodiversity and ecology of Antarctic lakes: models for evolution. Philos Trans R Soc Lond B Biol Sci 362: 2273-2289.

Liu W-T, Marsh TL, Cheng H, Forney LJ (1997). Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and environmental microbiology 63: 4516-4522.

Lyons WB, Fountain AG, Doran PT, Priscu J, Neumann K (2000). The importance of landscape position and legacy: The evolution of the Taylor Valley Lake District. Freshwat Biol 43: 355- 367.

Marcy Y, Ouverney C, Bik EM, Lösekann T, Ivanova N, Martin HG et al (2007). Dissecting biological “dark matter” with single-cell genetic analysis of rare and uncultivated TM7 microbes from the human mouth. Proceedings of the National Academy of Sciences of the United States of America 104: 11889-11894.

Medina-Sánchez JM, Villar-Argaiz M, Carrillo P (2004). Neither with nor without you: a complex algal control on bacterioplankton in a high mountain lake. Limnology and Oceanography 49: 1722-1733.

Montagnes D, Roberts E, Lukeš J, Lowe C (2012). The rise of model protozoa. Tren Microbiol 20: 184-191.

Moorthi S, Caron DA, Gast RJ, Sanders RW (2009). Mixotrophy: a widespread and important ecological strategy for planktonic and sea-ice nanoflagellates in the Ross Sea, Antarctica. Aquat Microb Ecol 54: 269-277.

Morgan-Kiss RM, Priscu JP, Pocock T, Gudynaite-Savitch L, Hüner NPA (2006). Adaptation and acclimation of photosynthetic microorganisms to permanently cold environments. Microbiol Mol Biol Rev 70: 222-252.

Neale PJ, Priscu JC (1995). The photosynthetic apparatus of phytoplankton from a perennially ice-covered Antarctic lake: acclimation to an extreme shade environment. Plant Cell Physiol 36: 253-263.

Okuda N, Watanabe K, Fukumori K, Nakano S-i, Nakazawa T (2014). Biodiversity researches on microbial loop in aquatic systems. Biodiversity in Aquatic Systems and Environments. Springer. pp 51-67.

Olsen GJ, Lane DJ, Giovannoni SJ, Pace NR, Stahl DA (1986). Microbial ecology and evolution: a ribosomal RNA approach. Annual Reviews in Microbiology 40: 337-365.

Pernthaler J (2005). Predation on prokaryotes in the water column and its ecological implications. Nature Reviews Microbiology 3: 537-546.

Pomero LR, Williams PJ, Azam F, Hobbie J (2007). The microbial loop. Oceanography 20: 28.

Pomeroy LR (1974). The ocean's food web, a changing paradigm. Bioscience 24: 499-504.

Reynolds RT, Squyres SW, Colburn DS, McKay CP (1983). On the habitability of Europa. Icarus 56: 246-254.

Roberts EC, Laybourn-Parry J (1999). Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwat Biol 41: 737-746.

Roberts EC, Priscu JC, Wolf C, Lyons WB, Laybourn-Parry J (2004). The distribution of microplankton in the McMurdo Dry Valley Lakes, Antarctica: response to ecosystem legacy or present-day climatic controls? Polar Biol 27: 238-249.

Sanders RW, Berninger U-G, 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: 103-118.

Sherr EB, Sherr BF (2002). Significance of predation by protists in aquatic microbial food webs. A Van Leeuw J Microb 81: 293-308.

Šlapeta J, Moreira D, López-García P (2005). The extent of protist diversity: insights from molecular ecology of freshwater eukaryotes. Proceedings of the Royal Society of London B: Biological Sciences 272: 2073-2081.

Sogin ML, Morrison HG, Huber JA, Welch DM, Huse SM, Neal PR et al (2006). Microbial diversity in the deep sea and the underexplored “rare biosphere”. Proceedings of the National Academy of Sciences 103: 12115-12120.

Tabita FR, Satagopan S, Hanson TE, Kreel NE, Scott SS (2008). Distinct form I, II, III, and IV Rubisco proteins from the three kingdoms of life provide clues about Rubisco evolution and structure/function relationships. J Exp Bot 59: 1515-1524.

Vick-Majors TJ, Priscu JC, Amaral-Zettler LA (2014). Modular community structure suggests metabolic plasticity during the transition to polar night in ice-covered Antarctic lakes. The ISME journal 8: 778-789.

Vrba J, Šimek K, Pernthaler J, Psenner R (1996). Evaluation of extracellular, high-affinity β-N- acetylglucosaminidase measurements from freshwater lakes: an enzyme assay to estimate protistan grazing on bacteria and picocyanobacteria. Microbial ecology 32: 81-97.

Whitman WB, Coleman DC, Wiebe WJ (1998). Prokaryotes: the unseen majority. Proceedings of the National Academy of Sciences 95: 6578-6583.

Wilhelm SW, Suttle CA (1999). Viruses and nutrient cycles in the sea viruses play critical roles in the structure and function of aquatic food webs. Bioscience 49: 781-788.

Yoon HS, Price DC, Stepanauskas R, Rajah VD, Sieracki ME, Wilson WH et al (2011). Single- cell genomics reveals organismal interactions in uncultivated marine protists. Science 332: 714- 717.

Zubkov MV, Sleigh MA (1998). Heterotrophic nanoplankton biomass measured by a glucosaminidase assay. FEMS Microbiology Ecology 25: 97-106.

CHAPTER II

Spatial heterogeneity and the impact of biotic and abiotic drivers on microbial autotrophic and heterotrophic activities in three chemically stratified Antarctic lakes

Wei Li, Jenna M. Dolhi, Rachael M. Morgan-Kiss Author contributions: WL and JD (equal contribution) developed methods, performed data collection and analyses, and wrote the manuscript related to βGAM and RubisCO activities, respectively

Most of this chapter appeared as: Li W, Dolhi JM, Morgan-Kiss RM. Spatial heterogeneity and the impact of biotic and abiotic drivers on microbial autotrophic and heterotrophic activities in three chemically stratified Antarctic lakes. Near submission to Aquatic Microbial Ecology

CHAPTER II 2.1. Introduction Permanently ice-covered lakes located in the McMurdo Dry Valleys (MDV) of South Victoria Land, Antarctica have closed basins and extremely stable strata which vary physically, chemically, and biologically within and between the water columns. These unique systems are part of the McMurdo Long Term Ecological Research (mcmlter.org) site and have been monitored annually since 1993. The 3-4 m thick ice-covers restrict water column mixing and external inputs; however, minimal mixing occurs during the austral summer when ice melt forms liquid water moats around the lake perimeters. Glacier melt streams which flow for a few weeks in the austral summer are the primary mechanism of water and nutrient input to the lakes (Lyons et al 2000b). The duration and magnitude of stream flow is predicted to increase in future years as a consequence of climate-related change (i.e., warmer, wetter summers) (Doran et al 2008).

Permanently physicochemical stratification in the MDV lakes is due to the lack of wind-driven mixing combined with salinity gradients that form distinct haloclines. The lower depths of these lakes contain dense saline water while shallow depths are approximately freshwater. This study focused on three MDV lakes, Lake Bonney, Lake Fryxell and Lake Vanda. All lakes are

- - oligotrophic in the shallow, photic layers and increase sharply in nutrient levels (NO2 , NO3 , + 3- NH4 , and PO4 ) at and below the permanent chemoclines. Irradiance is the main energy source for light-driven production of organic carbon in the lakes; however, it does not saturate photosynthesis and is a major limit of yearly production (Priscu 1995). Despite these commonalities in physical and chemical characteristics, each water column exhibits a unique physicochemical signature (Lyons et al 2000b).

Due to limitations in nutrient availability and under-ice solar radiation, MDV lake food webs are microbially dominated and truncated (lack metazoans). Microbial eukaryotes (i.e., protists) play dual roles at the bottom and the top of the food webs as primary producers (i.e., light driven CO2 fixation into simple sugars) and top predators (i.e., consumers of bacteria and small phytoplankton) (Priscu et al 1999b). Phytoplankton and chemolithoautotrophic bacteria fix inorganic carbon using light-dependent and -independent pathways, respectively, and form the base of the food web in the MDV lakes. Recent applications of 18S rRNA sequencing in Lake

Bonney provided the first insight into the phylogenetic diversity of the microbial eukaryote populations, 85% of which was related to photosynthetic species (Bielewicz et al 2011b). The water column of Lake Bonney is dominated by chlorophytes in the shallow layers and haptophytes in the chemocline, while Lake Fryxell is dominated by cryptophytes (Bielewicz et al 2011b, Dolhi et al 2015b, Kong et al 2012c). The water column of Lake Vanda harbors cyanobacteria and nanoplankton related to nonflagellated chlorophytes (Dolhi et al 2015b).

In aquatic environments heterotrophic unicellular protists heavily graze on prokaryotic (i.e., cyanobacteria) and small eukaryotic communities (Pernthaler 2005b, Sˇimek et al 2000). They plays a major role in top-down control of phytoplankton and bacterioplankton population, trophic transfer of energy and nutrient, as well as microbial turnover in these ecosystems (Boenigk and Arndt 2002, Buitenhuis et al 2010, Pierce and Turner 1992). In the microbial- dominated MDV lakes, heterotrophic protists are the tertiary predators of the aquatic food chain. Consumers which engulf particulate organic carbon (i.e., heterotrophic activity via phagocytosis) in the lakes include heterotrophic flagellates, ciliates, and rotifers (Priscu et al 1999b). Heterotrophic protists play important roles in energy and nutrient cycling (Azam and Malfatti 2007, Strom et al 1997). In oligotrophic systems, a significant amount of biomass is grazed and eventually transferred to higher trophic levels by phagotrophic protists (Moorthi et al 2009b, Sanders et al 2000a, Sherr and Sherr 2002b). Heterotrophic grazing activity by nanoflagellates and phytoplankton capable of heterotrophy (i.e., mixotrophy) has been reported in several MDV lakes (Marshall and Laybourn-Parry 2002, Roberts and Laybourn-Parry 1999, Thurman et al 2012). Recent studies on the 18S rRNA taxonomic diversity in Lake Bonney and Lake Fryxell (Bielewicz et al 2011b, Vick-Majors et al 2014b) also reported protists related to important heterotrophic nanoflagellates in freshwater and marine environments (e.g., ciliates, choanoflagellates, chrysomonads, Spumella; (Boenigk and Arndt 2002, Pierce and Turner 1992, Sˇimek et al 2000). Mixotrophy appears to be widespread in ice-covered Antarctic lakes and is an adaptive advantage for acquiring limited nutrients under oligotrophic conditions (Nygaard and Tobiesen 1993) and providing additional organic carbon sources under light-limiting conditions (Caron et al 1990). When mixotrophic phytoplankton dominate the photosynthetic community, such as in high mountain lakes (Medina-Sánchez et al 2004), complex relationships form between protists and bacteria, including commensalism (i.e., dependence of bacteria on

17 photosynthetic carbon from the mixotrophs) and predation (i.e., bacterial as prey for the mixotrophs). In this study we measured enzyme activities representative of autotrophy (RubisCO) and heterotrophy (N-acetyl-beta-glucosaminidase; βGAM) and characterize the trophic modes of protistian communities throughout the water columns of ELB, WLB, Fryxell, and Vanda.

Unicellular, phagotrophic protists rely on grazing of bacteria/small phytoplankton or particulate organic matter for energy, carbon and nutrient sources. Prey are sequestered into food vacuoles and large levels of digestive enzymes are released from lysosomes into the food vacuoles to rapidly hydrolyze the organic compounds (Vrba et al 1996). Planktonic heterotrophic activity in aquatic environments can be detected and quantified based on specific lysozyme activity associated with β-D- glucosaminidase enzymatic activity (Zubkov and Sleigh 1998). Copiotrophic bacteria utilize similar enzymes to breakdown complex organic carbon particles (Arnosti 2011); however, the bacterial enzymes are excreted into their environment, while protist βGAM enzymes are restricted to the acidic environment of their food vacuoles. Thus, the optimal pH condition of protist βGAM tends to be lower (pH < 5) compared to that of heterotrophic bacteria (neutral or slightly basic) (Gonzalez et al 1993, Vrba et al 1993, Vrba et al 1996). Therefore, hydrolytic activity of phagotrophic protists can be specifically measured based on the activity of hydrolysis enzymes extracted from lysosomes of protists in low pH condition, and this measurement can be used as a diagnostic tool for estimating heterotrophic protist activity (Zubkov and Sleigh 1998).

In the microbially dominated MDV lakes, protistian carbon fixation and phagotrophic activity are critical to the carbon and nutrient cycling. Few studies have reported the autotrophic and heterotrophic eukaryotes distribution and activities in these lakes based on artificial prey feeding experiment or microscopic numeration (James et al 1998, Laybourn-Parry et al 1995, Roberts et al 2004a), but direct rate measurement of these two critical processes is still lacking. In addition, the microbial communities in these lakes are strongly controlled by “bottom-up” effect, and studies suggest that the microplankton distribution is strongly affected by physicochemical environment (Roberts et al 2004a, Roberts et al 2004b). However, it is unclear how the environmental factors will drive the autotrophic and heterotrophic activities in protistian communities in the MDV lake ecosystems. We hypothesize that 1) the trophic activities of protists are stratified in lake water columns due to highly stratified environmental conditions; 2) these activities are correlated with specific environmental driver(s). In this study, we investigated autotroph and protist-specific heterotroph distribution using RubisCO and acid-activities in protistian communities in the MDV lake ecosystems and correlated activity with specific biotic and abiotic factors.

2.2. Materials and methods

2.2.1. Field sampling and limnological parameters

Locations of study sites within the Taylor (Lakes Bonney and Fryxell) and Wright (Lake Vanda) Valleys of the McMurdo Dry Valleys, Antarctica, are shown in Figure 2.1. Water column samples were collected during the austral summer (Nov to Dec) of 2012 field season with the exception of samples for Lake Vanda RubisCO activity which were collected in Dec 2011. All sampling depths were measured from the piezometric water level in the ice hole using a depth- calibrated hand winch. Water samples were collected with a 5 L Niskin bottle and stored under low temperature and dark conditions until processing. Water samples for RubisCO carboxylase activity (2-5 L) and βGAM activity (0.75-2 L) were vacuum filtered (0.3 mBar) onto 47 mm GF/C filters (Whatman, UK) and hydrophilic nylon membrane filters (Millipore, MA) with a

0.45 µm pore size, respectively. Filters were immediately flash frozen in liquid N2 and shipped to the US laboratory on dry ice. Samples were stored at -80 °C for less than two months before determining enzyme activity. Prior to the field season, several both enzymatic assays were optimized. Fresh vs frozen mock filters were also compared to assess the effect of freezing.

Photosynthetically active radiation (PAR), temperature, chlorophyll a (Chl a), primary

+ - 3- productivity (PPR), and nutrient (NH4 , NO3 , and PO4 ) concentrations were determined through the water columns of ELB, WLB, FRX and VAN.. PAR was measured with a Li-Cor LI-193 spherical quantum sensor (Li-Cor Biosciences, NE). Temperature was measured with a Seabird model 25 profiler (Spigel and Priscu 1998). Chl a concentrations were determined using a profiling spectrofluorometer (BBE Moldaenke FluoroProbe). Light mediated PPR was

14 determined by measuring NaH CO3 incorporation in duplicate over a 24 h in situ incubation. Nutrients were measured as part of the NSF-funded McMurdo MCM-LTER program according to the methodology outlined in the MCM-LTER limnological manual (http://www.mcm.lter.org). Briefly, inorganic nitrogen species were determined with a Lachat autoanalyzer and soluble reactive phosphorus (SRP) was measured manually using the antimony-molybdate method (Strickland and Parsons 1972b).

2.2.2. Rubisco carboxylase activity assay

GF/C filters were thawed on ice and cut in half. All subsequent extraction steps were carried out on ice. Each cut half filter was transferred to a 1.5 mL screw cap tube containing 0.1 mm zirconia/silica beads and 650 µL ice cold RubisCO extraction buffer (50 mM bicine, 10 mM

-1 MgCl2, 1 mM EDTA, 5 mg mL BSA, and 0.1% triton X-100; add fresh: 20 mM NaHCO3, 10 mM DTT, 0.1 mg mL-1, pH 7.8). Filters were disrupted with a Minibead-Beater-16(Biospec Products, OK) for three cycles of 30 s at speed setting 48 with alternating 1 min on ice incubations. The lysate was transferred to a cold 1.5 mL microcentrifuge tube and cleared by centrifugation at 4 °C for 2 min at 15,000 x g. The soluble lysate was used in RubisCO carboxylase assays.

RubisCO carboxylase activity was measured in the soluble cell lysates (200 µL) within 30 min of extraction. Soluble cell lysates were incubated at 25 °C for 1 min before RubisCO activity was

14 measured using a standard NaH CO3 assay in duplicate reactions as described previously(Kong 14 et al 2012a) . Briefly, RubisCO was activated with MgCl2 (8 mM) and NaH CO3 (ViTrax, Placentia, CA, specific activity in final reaction: 0.03 µCi mmol-1) for 5 min at 25 °C. The carboxylase assay was initiated by addition of 20 µL of the substrate, Ribulose-1, 5-bisphosphate (RuBP; Sigma-Aldrich, USA; 15 mM) and the reaction was allowed to proceed for5 min at 20 °C. The reaction was stopped with 100 µL 100% propionic acid and unincorporated 14C was exhausted by centrifugation in a fume hood for 1.25 hr at 2,000 x g. Soluble cell lysate with no RuBP substrate added was used in the assay as a negative control. A multipurpose scintillation counter LS6500 (Beckman Coulter, FL) was used to determine cpm of acid stable end products. RubisCO activity was calculated per liter of lake water filtered. In order to optimize the RubisCO carboxylase activity method, preliminary tests were performed on a pure culture of a green algal isolate originated from Lake Bonney (Chlamydomonas sp. UWO241) and environmental samples from a local reservoir, Acton Lake in College Corner, OH. We also tested carboxylase assays under a range of assay temperatures and determined that 20oC was optimal across a range of samples from isolated organisms and environments.

2.2.3. Glucosaminidase activity assay

The βGAM activity was measured as particulate enzyme activity retained on filters. Nylon filters were thawed on ice and extraction was carried out on ice. Filters were disrupted by 3 cyles of bead beating in 2 mL of ice cold extraction buffer (0.1 M Acetate, 0.1% 23 lauryl ether [Brig 35], pH 4.6 with glacial acetic acid). βGAM enzyme activity was measured in soluble lysates (250 µL) using a β-N-Acetyl-glucosaminidase Assay Kit (Sigma-Aldrich, USA). Cell lysates were incubated with the fluorogenic substrate 4-methylumbelliferyl-n-acetyl- a -D- glucosaminide in the dark at 20 ˚C for 4 hrs. Nano-pure water and purified β-N-Acetyl- glucosaminidase from Jack beans (standard with assay kit) were used as negative and positive controls, respectively. Fluorescence intensity was quantified in a Lambda-35 spectrofluorometer (Perkin-Elmer, MA). A standard curve was made with known amounts of the product of the enzymatic reaction, 4-MUF (Vrba et al 1993, Zubkov and Sleigh 1998). βGAM activity was expressed as amount of molecules been hydrolyzed per unit time per unit volume of original samples. In order to test the assay and choose filters with optimal enzyme recovering capability, filters made of various materials (glass fiber, polyethersulfone, and nylon) were tested using environmental samples from a local lake (Acton Lake, OH). The test results indicated that samples extraction from nylon filters produced the highest enzyme yield (data not shown).

2.2.4. Protein concentration determination

Protein concentration was determined in the soluble cell lysate extracted for RubisCO and βGAM assays according to the Bradford method using bovine serum albumin (Sigma-Aldrich, MO) as a standard (Bradford 1976) .

2.2.5. Bacterial enumeration

Quantitation of free-living bacteria in the lake water columns was carried out using a method modified from Lebaron et al. (1994). Lake water (2 mL) from each depth was fixed with paraformaldehyde at a final concentration of 2 % v/v for 30 min. Fixed samples were filtered on 25 mm black polycarbonate membrane with a pore size of 0.2 µm. Bacteria cells were stained with a fluorochrome (4',6-diamidino-2-phenylindole, DAPI) followed by epifluorescent

22 microscopic enumeration. Filters were examined using an Olympus AX-70 Multi-mode System with a specific filter set (EX 360/40 nm, EM 460/50 nm) for DAPI staining, and digital images of at least 15 random views were taken. Images were then calibrated, and bacterial counts from each image were recorded using ImageJ software (V1.47, National Institutes of Health, USA). The bacterial concentration of each depth was determined by using the average bacterial counts of each image.

2.2.6. Statistical analyses

+ - 3- Abiotic parameter data (PAR, temperature, NH4 , NO3 , PO4 , N:P ratio, DIC, and conductivity) were normalized using z-scores. Hierarchical cluster analysis (HCA) and principal component analysis (PCA) was carried out in PAST (V3.07, Oslo, Norway) software (Hammer et al 2001b). All other statistical analyses were performed using JMP Pro 11 software (SAS Institute Inc., Cary, NC). Data of physical (except temperature), chemical, and biological parameters were log- transformed, and the correlation between these parameters and enzyme activities were determined using Pearson correlation (R) analysis. Linear models of enzyme activities and abiotic factors were generated using stepwise multiple linear regressions.

2.3. Results

2.3.1. Lake Bonney

Lake Bonney is separated into two lobes, east and west. The water columns of each lobe remain isolated from each other for most of the year, with the exception of a narrow passage which connects the upper photic zones for a few weeks during the austral summer. Thus, east and west lobe Lake Bonney share some common characteristics, including strong thermohaline stratification (Figure 2.2); however, the deeper layers of each lobe are separated from each other. PAR values were at or below 20 µmol photons m-2 s-1 (Figure 2.3B,D). Temperatures in the shallow depths and the chemoclines ranged between 2 to 4 oC in both lobes, but declined below 0 oC in the monimolimnion of WLB (Figure 2.3 B,D). Both lobes exhibit SRP levels near the limit of detection throughout the water columns and early studies reported that phytoplankton and bacterial

+ - communities are limited by phosphate (Priscu 1995, Ward et al 2003). Nitrogen (NH4 , NO3 ) was low in the surface waters of both lobes. Spatial trends in several chemical variables are distinct in the monimolimnia below the chemoclines. Most notably is the chemistry of the deep waters below the chemoclines, which is substantially different between the two lobes (Figure 2.3 A,C). The east

- lobe exhibits higher levels of NO3 at depths below the chemocline, indicating the absence of a normally functioning nitrogen cycle (Ward et al 2003). Chl a maxima were present in the shallow and deep layers of both lobes of Lake Bonney; the deep chlorophyll maximum (DCM) was well defined in WLB, reaching a maximum Chl a of 15 µg L-1 (Figure 2.4A,B). Spatial trends in bacterial biomass were uncoupled from phytoplankton in WLB. Maximum bacterial biomass occurred above the chemocline in both lobes, but was higher in WLB (4.2 x 104 cell L-1) and peaked at different depths (5 m for ELB and 10 m for WLB; Figure 2.4A,B).

Early studies by Laybourn-Parry and colleagues (Laybourn-Parry et al 2005b, Roberts and Laybourn-Parry 1999) reported that dry valley lake protist communities exhibit versatile metabolic ability to generate energy and acquire carbon and nutrients. More recent studies have applied molecular approaches to describe the diversity and distribution of phytoplankton (Dolhi et al 2015b, Kong et al 2012c, Kong et al 2014a) and microbial eukaryotes (Bielewicz et al 2011b, Vick-Majors et al 2014b). It is unknown how variability in key environmental factors

24 within and between lakes influences metabolic potential. We used RubisCO and βGAM as estimators of autotrophic and heterotrophic metabolism, respectively. Enzyme activity was expressed on the basis of total extracted protein (Figure 2.5 A-D) or volume of filtered lake water (Figure 2.5E-H). Activity for both enzymes was higher in ELB relative to WLB. On a protein basis, RubisCO activity peaked at 13 m in both lobes. Spatial trends in βGAM activity were distinct from RubisCO activity, indicating that the importance of the two metabolic processes is variable at different layers within the lakes. Both lobes exhibited a peak in βGAM activity in shallow depths (5 m); however WLB also exhibited a distinct peak in the chemocline (15 m depth). Spatial trends in βGAM were comparable on a protein or lake water basis; however, trends in RubisCO activity were dependent upon which variable was used for normalized the data.

2.3.2. Lake Fryxell

Lake Fryxell is the least saline (maximum conductivity ~7 mS cm-1) and shallowest (20 m maximum depth) of the three lakes in this study (Figure 2.2) and exhibited very low PAR levels (maximum PAR <10 µmol photons m-2 s-1) (Figure 2.3F). In common with Lake Bonney, nutrient concentrations were low or below detection limits in the upper oxic zone, but increased sharply at the chemocline. SRP levels were significantly higher in the deeper zones of FRX relative to Lake Bonney, and earlier studies showed that FRX phytoplankton communities are limited for nitrogen (Priscu 1995). Higher nutrient levels supported the highest levels of Chl a and bacterial biomass in the water column of FRX compared with other MDV lakes (Figure 2.4C). Peaks in Chl a and bacterial biomass occurred at 9 m and 7 m, respectively; indicating that primary and bacterial production are also uncoupled in this lake. On a protein basis, RubisCO activity peaked above the chemocline (7 m); however, on a lake water volume basis RubisCO activity peaked in the deep anoxic zone (14 m) (Figure 2.5C,G). βGAM activity levels in Lake Fryxell were comparable with Lake Bonney; however, no spatial trends were noted.

2.3.3. Lake Vanda

Lake Vanda is the deepest lake (maximum depth 80 m) in our study and exhibited stratification which was comparable to Lake Bonney (Figure 2.2). PAR levels under the ice are 5-10 fold higher relative to Lakes Bonney and Fryxell, and the temperature rises to above 20oC below the chemocline (Figure 2.3F). Nutrient concentrations for most of the water column (above 60 m depth) were low or at undetectable levels, and SRP levels in Lake Vanda were comparable with Lake Bonney (i.e., maximum SRP < 10 µM) (Figure 2.3G). Lake Vanda is considered one of the most oligotrophic bodies of water in the world; Chl a and bacterial biomass levels peaked within the chemocline (68 m depth) and were the lowest of the three study lakes (Figure 2.4D). Peaks in Chl a and bacterial biomass co-occurred. Similarly, both RubisCO and βGAM activities both peaks in the deep layer of Lake Vanda, albeit at a lower depth compared with Chl a and bacterial biomass (72 m) (Figure 2.5D,H).

2.3.4. Cluster analyses of lake physicochemical parameters

In agreement with past reports, lake physicochemistry and biology exhibited strong vertical stratification as well as lake-specific differences (Figures 2 to 4). This complexity was also reflected in spatial trends of both RubisCO and βGAM enzyme activities (Figure 2.5). In an effort to determine whether specific abiotic factors might drive autotrophic or heterotrophic activity, we first performed HCA and PCA analyses of major abiotic factors (i.e., temperature, PAR, conductivity and dissolved inorganic nutrients) across all sampling sites (Figure 2.6). Despite that the whole lake physical and chemical conditions are largely distinct from one lake to another, the HCA and PCA results indicated that physicochemical parameters collected from 93 total samples (31 sampling depths with triplicated samples at each depth) clustered strongly into two major groups representing sampling depths located in the shallow layers (above chemocline) versus the deep layers (chemocline and deeper). PCA analyses also showed that PAR was

- positively correlated with layers above the chemocline while all other parameters (SRP, NO3 , + NH4 , N:P ratio, temperature and DIC) were positively correlated with layers at and below the chemocline (Figure 2.6B,C). .

2.3.5. Correlation of autotrophic and heterotrophic activities with physicochemical parameters

Given that lake physicochemical parameters strongly clustered into two groups (i.e., the upper water column and at/below the chemocline), we grouped autotrophic and heterotrophic enzyme activity across the lakes into these two groups and performed Pearson correlation coefficient (PCC) analysis between enzyme activity and a suite of biological and environmental parameters. Several environmental and biological parameters exhibited significant correlations with RubisCO and βGAM (Table 2.1). Spatial trends in Log RubisCO activity in the shallow water column layers weakly correlated with abiotic parameters such as conductivity, temperature,

- 3- NO3 , PO4 and N:P ratio, as well as biotic parameters such as PPR and abundance of cryptophytes. However, in deep water columns Log RubisCO activity negatively corresponded

+ - with the environmental parameter conductivity as well as NH4 , NO3 and N:P ratio, and positively corresponded with the biological parameters Chl a, protein concentration, as well as abundance of cryptophytes and haptophytes.

The relationship between spatial distribution of heterotrophic activity (i.e., βGAM) and various lake parameters was distinct from that of RubisCO. Statistically significant correlations for βGAM activity in shallow water column layers include those with PAR, temperature, bacteria counts, protein concentration, Chl a as well as green algae and cyanobacteria abundance. In contrast, the enzyme activity in deep water column layers showed correlations with conductivity, temperature, NH4+ and PO43-, protein concentration (Table 2.1). Interestingly, in the deep layers βGAM activity was positively correlated with temperature while in the shallow layers, it had negative correlation with this parameter.

From linear models (Table 2.2) built with stepwise multiple linear regressions, RubisCO activity

- in the shallow layers was largely explained by abiotic parameters (i.e., conductivity, NO3 and 3- PO4 ), while in the deep layers, biotic factors (i.e., Chl a, cryptophytes, haptophytes) explained most of the RubisCO activity variance. On other hand, βGAM activity in the shallow layers was mostly explained by biotic parameters (i.e., bacterial biomass, Chl a and chlorophytes), while

+ abiotic parameters (i.e., conductivity, NH4 etc.) were the significant explanatory variables for βGAM activity in the deep layers.

2.4. Discussion

MDV lakes vary in their distribution, diversity and abundance of plankton communities, which is driven by physicochemical differences within and between the lakes (Laybourn-Parry et al. 1997, Roberts et al. 2004b, Kong et al. 2012b). Recent studies applied combined sequencing and qPCR approaches to describe the diversity and vertical distribution of the autotrophic communities residing in Lake Bonney (Kong et al. 2012a, b), as well as Lakes Fryxell and Vanda (Dolhi et al. 2015). These studies predicted that light availability was a strong environmental driver for both gene copy number and transcript level of form IA/B RubisCO (large subunit encoded by rbcL) in Lake Bonney and Lake Vanda (Kong et al. 2012a; Dolhi et al. 2015). Spatial trends in Cryptophyte rbcL abundance were positively correlated with PAR in Lake Bonney; however, in Lake Fryxell form I A/B RubisCO was negatively correlated with light, and a second major RubisCO isoform, form ID was not correlated with PAR (Kong et al. 2012a). These studies provided evidence that light availability is not necessarily a limiting growth factor for the MDV protists and that the interactions between MDV protists and environmental drivers is complex. More recently, single cell sorting of MDV protists identified specific groups of MDV protists from Lake Bonney which are capable of diverse metabolic ability (e.g. photosynthesis, mixotrophy, heterotrophy, parasitism). . Thus, while the recent focus on molecular applications has advanced our understanding of the activity and function of the microbial eukaryotes residing in the MDV lakes, metabolic potential and the impact of environmental factors on protist trophic mode is not fully understood. Here, we complemented earlier predictions from molecular studies by directly measuring potential rates for the autotrophic enzyme, RubisCO, and heterotrophic enzyme, βGAM.

Autotrophic (RubisCO) and heterotrophic (βGAM) enzyme activities were highest at the chemocline (ELB, WLB) or began increasing at the chemocline (Lakes Fryxell and Vanda). WLB exhibited the lowest overall RubisCO activity which was at first surprising as the water column exhibited a well-defined Chl a maxima compared with ELB (see Figure 2.4) which was consistent with earlier reports (Bielewicz et al. 2011; Kong et al. 2012). However, this finding fit well with an observation by Kong et al (2012b) that while WLB exhibited higher levels of Chl a and rbcL DNA levels, the transcriptional activity of both rbcL ID and rbcL IA/B (i.e. the ratio of

28 mRNA to DNA) was 25 and 500-fold lower in WLB compared with ELB (Kong et al. 2012b). In support of the molecular evidence, rates of phytoplankton primary productivity (PPR) were also 4 to 5-fold lower in WLB compared with ELB (Kong et al. 2012b). The transcriptionally active RubisCO harboring community within the chemocline of both lobes includes haptophytes dominated by an Isochrysis sp. and a stramenopile related to Nannochloropsis (Kong et al. 2012b). Presumably, these organisms contribute the majority of light-dependent carbon fixation observed in Lake Bonney. Last, there is recent molecular evidence in WLB for bacterial communities capable of chemolithoautotrophy, (Kong et al. 2012a; Dolhi et al. 2015). Thus, despite seasonally connectivity between the upper water columns of ELB and WLB, there are clearly distinct influences on the autotrophic activity between the lobes of Lake Bonney.

Ciliates and dinoflagellates were reported to make up the Lake Bonney heterotrophic community (Bielewicz et al 2011b, Vick-Majors et al 2014b). In addition, mixotrophic phytoflagellates were shown to have higher grazing rates than heterotrophic flagellates in Lake Bonney especially in the months leading up to the austral winter (Thurman et al. 2012). A recent study reported that the haptophyte Isochrysis, a nanoplankton which dominates the chemoclines of both lobes of Lake Bonney, is capable of mixotrophy (Li et al 2016b). In addition, the spatial trends in βGAM in both ELB and WLB match that of haptophyte rbcL abundance (Kong et al. 2012b). Thus, Thus, haptophytes may contribute to both RubisCO and βGAM activity within the chemoclines of ELB and WLB. Chemoclines represent steep redox transitions and are typically associated with high microbial activity and biomass in meromictic lakes (Van Gemerden and Mas 1995), representing a rich source of POC for heterotrophic and mixotrophic protists. Conversely, βGAM activity the shallow depths of both lobes may reflect an alternative mechanism for the acquisition of additional nutrients (nitrogen and phosphorus) in the highly oligotrophic layers of Lake Bonney (see Figure 2.5) (Laybourn-Parry 2002).

Communities of Lake Fryxell exhibited high RubisCO activity corresponding to its status as the most productive lake in this study. Cryptophytes, capable of mixotrophy (i.e., combined photosynthesis and heterotrophy) reside at and below the chemocline, (Dolhi et al 2015b, Laybourn-Parry 1997, Marshall and Laybourn-Parry 2002, Roberts and Laybourn-Parry 1999) and are therefore likely to be major contributors to the enzyme activities measured in Lake

Fryxell. The deep chemocline of Lake Vanda (72 m) had greatest βGAM activity among all of the lakes which matched spatial trends in relatively high RubisCO activity at depths below the chemocline. The peak in RubisCO activity is likely influenced by stramenopiles related to Heterococcus sp. at 68 m (Dolhi et al 2015b). Ciliates and heterotrophic nanoflagellates, which may include mixotrophic Ochromonas sp., were reported at 60-64 m in Lake Vanda (James et al 1998) and may contribute to the heterotrophic activity measured in this lake. The warm water temperature at this depth in Lake Vanda may contribute to optimal enzyme function and corresponds with the positive relationship between βGAM activity and temperature in deep water layers (Figures. 3H, 5D and 5H).

The PCC analysis of enzyme activity and biotic and abiotic environmental parameters provided insight on drivers of auto- and heterotrophic groups in shallow and deep lake depths. The meromictic nature of these lakes with little mixing between shallow and deep layers, allows for grouping of sampling depths (i.e., shallow and deep) according to biotic and abiotic parameters. The correlation between RubisCO activity and biotic parameters (protein concentration, Chl a, and haptophyte and cryptophyte abundance) in deep layers indicates that haptophytes and cryptophytes are likely major contributors to inorganic carbon fixation within the chemoclines of Lakes Bonney and Fryxell, respectively. Oppositely, βGAM activity correlated strongly with biotic parameters (bacterial counts, protein concentration, Chl a, and green algae abundance) in shallow lake layers, whereas βGAM activity correlated strongly with abiotic parameters

+ 3- (temperature, salinity, NH4 , and PO4 ) in deep lake layers. This correlation fits well with the assumption that heterotrophic protists heavily graze on bacterioplankton and small phytoplankton (Laybourn-Parry et al 1995, Laybourn-Parry et al 1997, Laybourn-Parry 2002). The positive correlation between βGAM activity and biotic parameters in these depths suggests that prey availability might be a very important limiting factor of heterotrophic activity in shallow layers. In addition, ultra-low nutrient and dissolved organic carbon concentration in these layers can only support low phytoplankton and bacteria biomass. In contrast, PCC analysis indicates that conductivity which is mostly affected by salinity in the water column is negatively correlated with heterotrophic activity, which can be explained by the fact that higher salinity depths (e.g., hyper-saline water in deep Lake Bonney) may impact the physiology of heterotrophic protists. Since many heterotrophic organisms are known for their delicate cell

30 structure, it is not surprising that MDV heterotrophic protists could be adversely affected by the salinity levels in the deep zone of Lake Bonney.

RubisCO activity was measureable at depths below the photic zones in all MDV lakes studied, indicating that there is the potential for carbon fixation using alternative energy sources, such as reduced organic or inorganic compounds (i.e., chemolithoautotrophic metabolism) (Figure 2.5). Evidence for chemolithoautotrophic bacteria in the MDV aquatic ecosystems has been reported for several MDV lakes. The anoxic, sulfidic bottom waters of Lake Fryxell have yielded culture- dependent (Sattley and Madigan 2006) and independent (Dolhi et al. 2015) evidence for sulfur- oxidizing bacteria (SOB) related to Thiobacillus thioparus. This region is suitable for carbon fixation by chemolithoautotrophs which may utilize sulfide and oxygen as an electron source and acceptor, respectively (Sattley and Madigan 2006). Evidence for the existence of a SOB community below the photic zone was also found for WLB(Dolhi et al 2015b, Kong et al 2012a) . It is predicted that these organisms utilize dimethyl sulfoxide (DMSO), a breakdown product of dimethylsulfoniopropionate (DMSP) which is produced by algae including haptophytes, chrysophytes, and dinoflagellates as a cryo- or osmo-protectant (Trevana et al. 2000), as an electron acceptor. A second breakdown product, dimethylsulfide (DMS) may be utilized as an electron donor (Raina et al 2010). The potential for dark carbon fixation also exists in ELB and Lake Vanda. PCR based detection of ammonia oxidizing bacteria (AOB) was found at depths above 25 m in ELB, and the lack of detection at deeper depths was likely due to inhibition of the PCR reaction (Voytek et al 1998, Voytek et al 1999). AOB were detected by in situ nitrification assays at 50-57 m in Lake Vanda, where reduced nitrogen compounds (electron donor) begin to increase and oxygen is available as an electron acceptor (Vincent et al. 1981, Voytek et al. 1999).

While the RubisCO enzyme assay measured the potential for a community to fix inorganic carbon, βGAM activity is a proxy for community respiration, specifically breakdown of complex organic carbon sources by acidic food vacuoles. Other sources of organic carbon include upward diffusion from a deep water pool or fall out of particulate organic matter that exists within the ice cover (Priscu et al 1999b). Bacterial prey is likely the major food source in the lakes; however

31 excretion from benthic algal mats may also serve as a source of carbon (Priscu et al 1999b, Wharton Jr et al 1983).

Using functional enzyme activity, this study showed that autotrophic and heterotrophic spatial trends within the MDV lakes are complex. While autotrophic metabolism was generally dominant at depths at and below the chemocline, heterotrophic metabolism was dominant at both shallow, nutrient deficient depths (Lake Bonney), within the chemocline (Lake Bonney and Vanda), or below the chemocline (Lake Fryxell). Previous studies of RubisCO gene diversity and distribution (Kong et al. 2012a, b) helped to attribute enzyme function to groups of organisms. Pairing diversity and function studies with environmental parameters will improve understanding of microbial community structure and how this may be impacted by climate change, an area of research with many unanswered questions (Caron and Hutchins 2012). Studies of polar microbial communities on the cusp of environmental change will be important for predicting how microbial communities in low latitude aquatic systems will respond.

2.5. Acknowledgements.

The authors thank the McMurdo LTER, Antarctic Support Contract and PHI helicopters for logistical assistance in the field. This work was supported by NSF Office of Polar Programs Grant OPP-1056396.

Figure 2.1. Map showing locations of study sites within the McMurdo Dry Valleys, Antarctica: Lake Bonney (BON), Lake Fryxell (FRX) and Lake Vanda (VAN). Image was edited from a LIMA based map from the Antarctic Geospatial Information Center (http://www.agic.umn.edu/maps/Antarctic).

Conductivity (mS cm-1) FRX 0 5 10 15 20 0

FRX

20 ELB WLB

Depth(m)

VAN

80 0 20 40 60 80 100 120 140 Salinity (mS cm-1) BON, VAN

Figure 2.2. Salinity profiles for Lake Bonney (east lobe, ELB; west lobe, WLB), Lake Fryxell (FRX) and Lake Vanda (VAN). Note that Lake Fryxell is plotted on a secondary x-axis.

SRP (µM) Temperature (oC) ELB, WLB, FRX 0 10 20 30 40 50 -2 0 2 4 6 8 10 0 0 A - ELB B - ELB 10 10

20 20

SRP (m) Depth Depth (m) Depth 30 NH + 30 4 Temperature NO - 3 PAR 40 40

0 0 C - WLB D - WLB

10 10

20 20

Depth (m) Depth Depth (m) Depth 30 30

40 40

0 0 E - FRX F - FRX

5 5

10 10 Depth (m) Depth Depth (m) Depth 15 15

0 10 20 30 40 50 -2 -1 PAR (µmol photons m s ) ELB, WLB, FRX

Temperature (oC) VAN 0 10 20 30 0 0 G - VAN H - VAN

20 20

40 40 Depth (m) Depth Depth (m) Depth 60 60

0 250 500 750 1000 1250 1500 0 100 200 300 -2 -1 + - PAR (µmol photons m s ) VAN NH4 , NO3 (µM)

Figure 2.3. Depth profiles of physical and chemical characteristics for the three lakes in this study: Lake Bonney east (A, B) and west (C,D) lobes, Lake Fryxell (E,F), and Lake Vanda (G,H). Note that PAR and Temperatures for Lake Vanda are plotted on secondary x-axes.

0 0 A - ELB B - WLB

10 10

20 20 Depth (m) Depth (m) Depth

30 Chl a 30 Bacteria

0 10 20 30 40 50 0 10 20 30 40 50 -1 -1 -1 -1 Chl a (µg L ) or Bacteria (x 100 cells L ) Chl a (µg L ) or Bacteria (x 100 cells L )

0 0 C - FRX D - VAN

5 20

10 40 Depth (m) Depth (m) Depth

15 60

20 80 0 10 20 30 40 50 0 5 10 15 20 -1 -1 -1 -1 Chl a (µg L ) or Bacteria (x 100 cells L ) Chl a (µg L ) or Bacteria (x 100 cells L )

Figure 2.4. Depth profiles for phytoplankton biomass (Chl a) and bacterial abundance for study sites.

Figure 2.5. Spatial variability in enzyme activity of RubisCO and β-Glucosaminidase (B-GAM) in study sites. Enzyme activity is represented as either nmol µg protein-1 hr-1 (A-D) or nmol hr- 1 L-1 lake water (E-H). Note that the scales are the x-axes are difference across all panels and the break in the x-axis in Panel A.

Figure 2.6. a. Hierarchical clustering analysis (HCA) and b, c. principal component analysis (PCA) based on major abiotic factors (i.e. PAR, temperature, dissolved carbon, nitrogen and phosphorus etc.) of all four studied lakes. Shallow and deep layers were coded as orange and blue respectively. HCA was performed using Ward’s method and bootstrapped 10,000 time. In PCA plots, shaded area and dashed circles indicated 70% and 85% similarity respectively within each group. Symbols indicated samples from ELB (plus), WLB (circle), FRX (cross) and VAN (diamond) respectively.

Table 2.1. Pearson correlation coefficient values (R) for average RubisCO and βGAM activity with lake physical, chemical, and biological parameters for all samples (n= 57-93). The significant of the correlation coefficient is indicated by *, 0.01

+ - 3- Cond PPR PAR Temp NH4 NO3 PO4 N:P DIC Bact Prot CHL G. AL. Cyan Hapt Crypt

Shallow (n= 20) *-0.53 *0.5 0.27 *-0.49 -0.35 *-0.48 *-0.52 *-0.49 -0.04 0.3 -0.08 0.46 -0.29 -0.31 0.26 *0.51

Log RubisCO

Deep (n=37) ***-0.66 -0.21 0.26 -0.19 **-0.52 **-0.42 -0.23 ***-0.69 -0.14 0.25 **0.42 ***0.57 *-0.36 0.04 **0.5 ***0.75

Shallow (n=42) 0.1 -0.26 **-0.47 *-0.39 0.03 0.04 0.06 0.08 -0.01 ***0.54 *0.39 ***0.67 ***0.54 0.15 0.17 -0.15

βGAM

Deep (n=51) **-0.47 -0.05 -0.22 **0.41 ***0.5 -0.18 ***0.45 -0.22 -0.22 -0.06 *0.32 0.01 0.03 0.25 -0.09 -0.09

Table 2.2. Linear regression models explaining the enzyme activities in different portions of water columns. The models result from a

mixed stepwise procedure and explain log-transformed autotrophic enzyme activity in shallow and deep layers: Log RubisCOshallow and

Log RubisCOdeep respectively, as well as heterotrophic enzyme activity in these two portions: βGAMshallow and βGAMdeep respectively. The significant of the regression coefficient is indicated by *, 0.01

No Dependent Linear model R2 p F n . variable adj

Log - 3- 0.4 1 *1.05 – 3.03 Log PPR – *0.85 Log Cond – *0.34 Log NO3 - 0.43 Log PO4 0.0292 3.81 18 RubisCOshallow 0 Log ***0.91 +*0.11 Log NH + – 0.14 Log Chl – ***0.14 Log G. AL. + 0.02 Log Hapt + 0.7 <0.000 19.0 2 4 37 RubisCOdeep ***0.36 Log Crypt 1 1 6 0.6 <0.000 18.5 3 βGAM 0.17 – *0.04 Temp –*0.42 Log Bact + **0.61 Log Chl + *0.17 Log G. AL. 32 shallow 9 1 0 0.4 <0.000 10.0 4 βGAM ***3.20 + ***0.27 Temp + *0.01 Log NH + + 0.04 Log Prot –***0.04 Log Cond 42 deep 4 7 1 3

2.6. References

Arnosti C (2011) Microbial extracellular enzymes and the marine carbon cycle. Ann Rev Mar Sci 3:401-425 Azam F, Malfatti F (2007) Microbial structuring of marine ecosystems. Nature Reviews Microbiology 5:782-791 Bielewicz S, Bell EM, Kong W, Friedberg I, Priscu JC, Morgan-Kiss RM (2011) Protist diversity in a permanently ice-covered Antarctic lake during the polar night transition. ISME J 5:1559-1564 Boenigk J, Arndt H (2002) Bacterivory by heterotrophic flagellates: community structure and feeding strategies. Antonie van Leeuwenhoek 81:465-480 Bradford MM (1976) A rapid and sensitive method for the quantification of microgram quantities of protein using th eprinciples of dye-binding. Anal Biochem 72:248-254 Buitenhuis ET, Rivkin RB, Sailley S, Le Quéré C (2010) Biogeochemical fluxes through microzooplankton. Global biogeochemical cycles 24 Caron DA, Hutchins DA (2012) The effects of changing climate on microzooplankton grazing and community structure: drivers, predictions and knowledge gaps. Journal of Plankton Research:fbs091 Caron DA, Porter KG, Sanders RW (1990) Carbon, nitrogen, and phosphorus budgets for the mixotrophic phytoflagellate Poterioochromonas malhamensis (Chrysophyceae) during bacterial ingestion. Limnology and Oceanography 35:433-443 Dolhi JM, Teufel AG, Kong W, Morgan-Kiss RM (2015) Diversity and spatial distribution of autotrophic communities within and between ice-covered Antarctic lakes (McMurdo Dry Valleys). Limnol Oceanogr 60:977-991 Doran PT, McKay CP, Fountain AG, Nylen T, McKnight DM, Jaros C, Barrett JE (2008) Hydrologic response to extreme warm and cold summers in the McMurdo Dry Valleys, East Antarctica. Ant Sci 20:499-509 Gonzalez JM, Sherr BF, Sherr E (1993) Digestive enzyme activity as a quantitative measure of protistan grazing: the acid lysozyme assay for bacterivory. Hammer Ø, Harper D, Ryan P (2001) PAST: Paleontological Statistics Software Package for education and data analysis. . Palaeontologia Electronica 4:9 James MR, Hall JA, Laybourn-Parry J (1998) Protozooplankton in the Dry Valley lakes of Southern Victoria Land. In: Priscu JC (ed) Ecosystem Dynamics in a Polar Desert, Book 72. Antarctic Research Series, Washington, D.C. Kong W, Dolhi JM, Chiuchiolo A, Priscu JC, Morgan-Kiss RM (2012a) Evidence of form II RubisCO (cbbM) in a perennially ice-covered Antarctic lake. FEMS Microbiol Ecol 82:491-500 Kong W, Li W, Prášil O, Romancova I, Morgan-Kiss RM (2014) An integrated study of photochemical function and expression of a key photochemical gene (psbA) in photosynthetic communities of Lake Bonney (McMurdo Dry Valleys, Antarctica). FEMS Microbiol Ecol 89:293-302 Kong W, Ream DC, Priscu JC, Morgan-Kiss RM (2012b) Diversity and expression of RubisCO genes in a perennially ice-covered Antarctic lake during the polar night transition. App Env Microbiol 78:4358-4366 Laybourn-Parry J (1997) The microbial loop in Antarctic lakes.

Laybourn-Parry J (2002) Survival mechanisms in Antarctic lakes. Philos Trans R Soc Lond B Biol Sci 357:863-869 Laybourn-Parry J, Bayliss P, Ellis-Evans JC (1995) The dynamics of heterotrophic nanoflagellates and bacterioplankton in a large ultra-oligotrophic Antarctic lake. Journal of Plankton Research 17:1835-1850 Laybourn-Parry J, James MR, McKnight DM, Priscu J, Spaulding SA, Shiel R (1997) The microbial plankton of Lake Fryxell, southern Victoria Land, Antarctica during the summers of 1992 and 1994. Polar Biology 17:54-61 Laybourn-Parry J, Marshall WA, Marchant HJ (2005) Flagellate nutritional versatility as a key to survival in two contrasting Antarctic saline lakes. Freshw Biol 50:830-838 Lebaron P, Troussellier M, Got P (1994) Accucary of epifluorescence microscopy counts for direct estimates of bacterial numbers. Journal of microbiological methods 19:89-94 Li W, Podar M, Morgan-Kiss RM (2016) Ultrastructural and single-cell level characterization reveals metabolic versatility in a microbial eukaryote community from an ice-covered Antarctic lake. Applied and Environmental Microbiology 82:3659-3670 Lyons WB, Fountain AG, Doran PT, Priscu J, Neumann K (2000) The importance of landscape position and legacy: The evolution of the Taylor Valley Lake District. Freshwat Biol 43:355-367 Marshall W, Laybourn-Parry J (2002) The balance between photosynthesis and grazing and Antarctic mixotrophic cryptophytes during summer. Freshwat Biol 47:2060-2070 Medina-Sánchez JM, Villar-Argaiz M, Carrillo P (2004) Neither with nor without you: a complex algal control on bacterioplankton in a high mountain lake. Limnology and Oceanography 49:1722-1733 Moorthi S, Caron DA, Gast RJ, Sanders RW (2009) Mixotrophy: a widespread and important ecological strategy for planktonic and sea-ice nanoflagellates in the Ross Sea, Antarctica. Aquat Microbial Ecol 54:269-277 Nygaard K, Tobiesen A (1993) Bacterivory in algae: a survival strategy during nutrient limitation. Limnology and Oceanography 38:273-279 Pernthaler J (2005) Predation on prokaryotes in the water column and its ecological implications. Nature Reviews Microbiology 3:537-546 Pierce R, Turner J (1992) Ecology of planktonic ciliates in marine food webs. Rev Aquat Sci 6:139-181 Priscu JC (1995) Phytoplankton nutrient deficiency in lakes of the McMurdo Dry Valleys, Antarctica. Freshwater Biology 34:215-227 Priscu JC, Wolf CF, Takacs CD, Fritsen CH, Laybourn-Parry J, Roberts JKM, Berry-Lyons W (1999) Carbon transformations in the water column of a perennially ice-covered Antarctic Lake. Biosci 49:997-1008 Raina J-B, Dinsdale EA, Willis BL, Bourne DG (2010) Do the organic sulfur compounds DMSP and DMS drive coral microbial associations? Tren Microbiol 18:101-108 Roberts EC, Laybourn-Parry J (1999) Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwat Biol 41:737-746 Roberts EC, Priscu JC, Laybourn-Parry J (2004a) Microplankton dynamics in a perennially ice- covered Antarctic lake–Lake Hoare. Freshwater Biology 49:853-869 Roberts EC, Priscu JC, Wolf C, Lyons WB, Laybourn-Parry J (2004b) The distribution of microplankton in the McMurdo Dry Valley Lakes, Antarctica: response to ecosystem legacy or present-day climatic controls? Polar Biology 27:238-249

Sˇimek K, Jürgens K, Nedoma J, Comerma M, Armengol J (2000) Ecological role and bacterial grazing of Halteria spp.: small freshwater oligotrichs as dominant pelagic ciliate bacterivores. Aquatic Microbial Ecology 22:43-56 Sanders RW, Berninger U-G, 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 192:103-118 Sherr EB, Sherr BF (2002) Significance of predation by protists in aquatic microbial food webs. Anton Van Leewen Int J Gen Molec Microbiol 81:293-308 Strickland JDH, Parsons TR (1972) A Practical Handbook of Seawater Analysis, Vol 167. Fisheries Research Board of Canada, Ottawa, Canada Strom SL, Benner R, Ziegler S, Dagg MJ (1997) Planktonic grazers are a potentially important source of marine dissolved organic carbon. Limnology and Oceanography 42:1364-1374 Tabita FR, Satagopan S, Hanson TE, Kreel NE, Scott SS (2008) Distinct form I, II, III, and IV Rubisco proteins from the three kingdoms of life provide clues about Rubisco evolution and structure/function relationships. J Exp Bot 59:1515-1524 Thurman J, Parry J, Hill PJ, Priscu JC, Vick TJ, Chiuchiolo A, Laybourn-Parry J (2012) Microbial dynamics and flagellate grazing during transition to winter in Lakes Hoare and Bonney, Antarctica. FEMS Microbiol Ecol 82:449-458 Van Gemerden H, Mas J (1995) Ecology of phototrophic sulfur bacteria. In: Anoxygenic photosynthetic bacteria. Springer Vick-Majors TJ, Priscu JC, Amaral-Zettler LA (2014) Modular community structure suggests metabolic plasticity during the transition to polar night in ice-covered Antarctic lakes. ISME J 8:778-789 Voytek MA, Priscu J, Ward BB (1999) The distribution and relative abundance of ammonia- oxidizing bacteria in lakes of the McMurdo Dry Valley, Antarctica. Hydrobiol 401:113- 130 Voytek MA, Ward BB, Priscu JC (1998) The Abundance of Ammonium-Oxidizing Bacteria in Lake Bonney, Antarctica Determined by Immunofluorescence, Pcr and In Situ Hybridization, Vol. Wiley Online Library Vrba J, Šimek K, Nedoma J, Hartman P (1993) 4-Methylumbelliferyl-β-N-Acetylglucosaminide Hydrolysis by a High-Affinity Enzyme, a Putative Marker of Protozoan Bacterivory. Applied and Environmental Microbiology 59:3091-3101 Vrba J, Šimek K, Pernthaler J, Psenner R (1996) Evaluation of extracellular, high-affinity β-N- acetylglucosaminidase measurements from freshwater lakes: an enzyme assay to estimate protistan grazing on bacteria and picocyanobacteria. Microbial ecology 32:81-97 Ward B, Granger J, Maldonado M, Wells M (2003) What limits bacterial production in the suboxic region of permanently ice-covered Lake Bonney, Antarctica? Aquatic microbial ecology 31:33 Wharton Jr RA, Parker BC, Simmons Jr GM (1983) Distribution, species composition and morphology of algal mats in Antarctic dry valley lakes. Phycologia 22:355-365 Zubkov MV, Sleigh MA (1998) Heterotrophic nanoplankton biomass measured by a glucosaminidase assay. Fems Microbiology Ecology 25:97-106

CHAPTER III

Influence of environmental drivers and potential interactions on the distribution of microbial communities from three permanently stratified Antarctic lakes

Wei Li, Rachael M. Morgan-Kiss

Most of this chapter appeared as: Li W, Morgan-Kiss RM. Influence of environmental drivers and potential interactions on the distribution of microbial communities from three permanently stratified Antarctic lakes. In preparation

CHAPTER III

3.1. Introduction

The microbial loop (i.e. microbial food web) plays important roles in the cycling of energy, carbon and elements in aquatic ecosystems. viruses, bacteria, Archaea and microbial eukaryotes are key players in global carbon cycle and biogeochemical cycles (Azam et al 1994, Azam 1998, Cotner and Biddanda 2002, Fenchel 2008, Fenchel et al 2012, Jiao et al 2010). Investigating microbial diversity and community structure is crucial first step for understanding the ecological functioning in aquatic environment. Meromictic lakes are bodies of water located mainly in the north and south polar regions as well as other high-latitude areas such as the Tibetan Plateau, which unlike the majority of lakes on earth, do not mix seasonally and exhibit permanent stratification of major physical (i.e., temperature, light availability) and chemical (i.e., oxygen, major nutrients, conductivity) environmental factors. Water masses exhibit year-round high physical stability due in large part to the presence of prominent haloclines. Microbial consortia residing in permanently stratified lakes exhibit relatively constant spatial stratification throughout the water column and are adapted to vastly different habitats within the same water column (e.g. stable transition zone between oxic mixolimnion and anoxic monolimnion). There is also often a dense and diverse microbial community residing in the redox transition zone (Van Gemerden and Mas 1995). Thus, meromictic lakes are interesting model systems for many questions in aquatic biology research, and more specifically ideal for studying the impact of environmental drivers on microbial community structure and interactions within the microbial loop (Eiler et al 2012a, Van der Gucht et al 2007, Yang et al 2015).

Antarctica harbors numerous marine-derived meromictic lakes, located in ice-free areas such as Vestfold Hills in Princess Elizabeth Land and the McMurdo Dry Valleys (MDV) in Southern Victoria Land. Lakes in the Vestfold Hills have recently been sites where next generation sequencing has been exploited to thoroughly described the identity and functional capacity of microbial populations (Cavicchioli 2015a, Lauro et al 2011, Laybourn-Parry and Bell 2014). In contrast, microbial community structure within the water columns of the MDV lakes is 45 significantly less well understood; although, biogeochemical processes and microbial production has been intensely studied since the establishment of the McMurdo Long Term Ecological Research (MCM-LTER) site in 1993. The east and west lobes of Lake Bonney (ELB and WLB) as well as Lake Fryxell (FRX) are three examples of three heavily studied MDV lakes located. The MDV is a polar desert with average air temperatures of -20°C and precipitation rates of <10 cm per year (Chela-Flores 2011, Priscu et al 1998, Reynolds et al 1983). The lakes represent only year-round liquid water body for life in this polar desert. Stratification occurs in these lakes due to the absent of wind mixing as well as extremely stable chemical gradients. Photosynthetically available radiation (PAR) is strongly attenuated by the presence of the permanent ice cover. In general, the shallow layers (above permanent chemoclines) have moderate to low solar irradiance but extremely low nutrient levels. Light level decreases quickly in the lakes along the depth, while nutrient concentration and salinity increases dramatically in the chemocline. In Lake Bonney, the deep water columns are in hypersaline condition (maximum 150 PSU) (Doran et al 2010, Lyons et al 2000a, Spigel and Priscu 2013).

The food web in these lakes are dominated by microorganisms and metazoans are largely absent with the exception of a few small invertebrates (copepods, rotifers, tardigrades). Biological communities residing in the MDV lakes can be generally categorized into four main zones: (i) a planktonic community dominated by photosynthetic protists in the oxygenated photic zone of mixolimnia; (ii) a zone of elevated microbial activity and a deep chlorophyll maximum (DCM) within the chemoclines; (iii) a microbial community residing in anoxic monimolimnia; and (iv) benthic cyanobacterial mat communities within the littoral zone above the chemoclines. Microbial eukaryotes occupied both positions of primary producers and predators in the trophic cascade. While a number of molecular studies have provided initial insights on the microbial community structures in MDV lakes (Bielewicz et al 2011a, Gordon et al 2000, Kong et al 2012b, 2014b, Vick-Majors et al 2014a), past work was either limited to clone library sequencing or next generation sequencing (NGS) of a few depths of the water columns. Thus, this study represents the first to utilize Illumina NGS (16S and 18S rRNA gene amplicon sequencing) across a larger number of replicated samples collected from the whole water column of three thoroughly characterized MDV lakes. We investigated the potential correlation patterns

46 between microbes, and identified the environmental drivers that impact on the community structure of the microbial eukaryote and bacterial communities. Due to inherent characteristics of meromictic lakes, permanently stratification creates distinct physico-chemical conditions in different layers in water column. Based on earlier evidence of environmental drivers on certain function groups (e.g., autotrophic eukaryotes and heterotrophic flagellates) (Kong et al 2012b, Laybourn-Parry et al 1995, Laybourn-Parry et al 2005a), we hypothesized that 1) due to the geographic isolation and lack of connectivity of the MDV lakes, lake microbial communities vary between lakes and exhibit niche-specific characteristics; 2) as the water column in the lakes is permanently stratified, microbial communities are stratified as well, and 3). interactions between specific organisms form ecological functional groups.

3.2. Materials and methods

3.2.1. Site description and sample collection

Water column samples were collected during the 2012 field season (November – December) according to Dolhi et al. (2015). All sampling depths were measured from the piezometric water level in the ice hole using a depth-calibrated hand winch. Water samples for DNA extraction were collected through boreholes in the ice cover of three lakes (GPS coordinates: ELB - 77o44'S 162 o10'E; WLB - 77 o 43S 162 o17'E; FRX - 77 o 37'S 163 o 11'E) using a 5 L Niskin bottle (General Oceanics). Sampling depths in the water column were selected to represent major chlorophyll a (chl a) maxima (Dolhi et al 2015c) as well as various locations within the epilimnion, thermocline and hypolimnion. Duplicated samples (0.75 – 2 L of water) from each depth were gently vacuum filtered (0.3 mBar) onto 47 mm Pall Supor® 450 polyethersulfone membranes (Pall Corporation, NY) according to Kong et al. (2012). Filters were immediately flash frozen in liquid N2 and shipped to the US laboratory on dry ice. Samples were stored at -80 ˚C before DNA extraction.

3.2.2. Environmental parameter measurements

Photosynthetically active radiation (PAR), temperature, chl a, primary productivity (PPR), and

+ - 3- nutrient (NH4 , NO3 , and PO4 ) concentrations were determined through the water columns of ELB, WLB, and FRX and matched sampling dates and locations of environmental DNA extractions. PAR was measured with a Li-Cor LI-193 spherical quantum sensor (Li-COR Biosciences, NE). Temperature was measured with a Seabird model 25 profiler (Spigel and Priscu 1998). Chl a concentration and relative abundance was determined was determined using a diving spectral fluorometer (FluoroProbe, BBE Moldaenke; Dolhi et al. 2015).. Light mediated

14 PPR was determined by measuring NaH CO3 incorporation in duplicate over a 24 h in situ

48 incubation. Nutrients were measured as part of the NSF-funded McMurdo LTER program according to the methodology outlined in the McMurdo LTER manual (http://www.mcm.lter.org). Briefly, inorganic nitrogen species were determined with a Lachat autoanalyzer and soluble reactive phosphorus (SRP) was measured manually using the antimony-molybdate method (Strickland and Parsons 1972a). Vertical profiles of chl a concentration, temperature, major nutrient and oxygen levels are summarized in Figures 2.2-2.4 in previous chapter.

3.2.3. DNA library preparation and Illumina MiSeq Sequencing

Environmental DNA was isolated from whole filters using the MP DNA kit (MP Biomedicals, CA) following the kit’s instruction and according to Bielewicz et al. (2011). DNA concentration was measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, DE). DNA was diluted if needed to the appropriate concentration for efficient amplification by PCR. The hypervariable regions V4 of the 16S rRNA gene (bacteria) and V9 of the 18S rRNA gene (eukaryotes) were amplified using the primer sets which encode F515/R806 for bacteria and F1391/R1501 for eukaryotes, respectively, and attached barcodes and linkers. Each sample was independently amplified in triplicate. PCR and MiSeq sequencing reactions were strictly followed the protocol provided by the Earth Microbiome Project (http://www.earthmicrobiome.org) (Amaral-Zettler et al 2009, Caporaso et al 2011, Caporaso et al 2012, López-García et al 2001). Samples were be sequenced according to the manufacturer’s recommendations using a 300-cycle MiSeq Reagent Kit v2 (Illumina®) on a MiSeq platform in house with a 2 X 150 bp paired-end run in the presence of 25 % PhiX sequencing control DNA.

3.2.4. Sequence analysis

Sequences generated on the MiSeq were analyzed with Quantitative Insights Into Microbial Ecology (MacQIIME v 1.9.0) pipeline (Caporaso et al 2010). Paired-end reads were first quality filtered (reads with minimum quality score of 25 retained) and stitched using fastq-join method (Aronesty 2011). Operational taxonomic unit (OTU) picking and taxonomic classification were performed following open-reference clustering procedure: sequences were clustered against Greengenes (v 13.8) database for prokaryotes and SILVA (v 119) database for eukaryotes respectively using UCLUST algorithm with cutoff of 97% and 95% similarity for prokaryotes and eukaryotes respectively. Sequences which failed to hit the reference databases were next clustered de novo to assigned new OTUs (Conlan et al 2012, Edgar 2010, Pruesse et al 2007, Quast et al 2013, Rideout et al 2014). OTUs with one sequence per sample or only appeared in a single sample were discarded to reduce the potential diversity inflation due to sequence errors.

3.2.5. Diversity assessment

Samples were rarefied to 5000 sequences/sample based the number of samples in the library with the lowest sequence number. Alpha diversity (number of OTUs, chao1 and Shannon index) was assessed in QIIME (Kuczynski et al 2012). Beta diversity of bacteria and eukaryotes were calculated using Bray-Curtis dissimilarity matrix in square-root-transformed OTU counts (Bray and Curtis 1957, Kuczynski et al 2012). Nonmetric multidimensional scaling (nMDS) was used to assess the relationship between microbial communities and environmental parameters. To test the significance of microbial communities and environmental factors, one-way analysis of similarity (ANOSIM) was applied with 9999 permutations (Anderson 2001, Anderson et al 2011). Statistics were carried out in PAST (v 3.07, Oslo, Norway) software (Hammer et al 2001a).

3.2.6. Network analysis

For network inference, the potential co-occurrence between OTUs as well as their correlation with environmental factors was assessed in CoNet (v 1.1.0 beta) (Faust et al 2012), and the results were visualized in Cytoscape (v 3.4.0) (Shannon et al 2003). Prior to importing dataset into CoNet, OTUs counts of duplicated samples were averaged, samples from all three lakes were pooled and z-score-transformed environment parameters were used for the metadata file. Pairwise scores between OTUs were generated using four measures (Bray-Curtis and Kullback- Leibler dissimilarities, as well as Pearson and Spearman correlations) with the threshold setting for retaining the initial network with 1000 positive and 1000 negative edges supported by all four measures. To alleviated compositional bias, 1000 renormalized permutation and bootstrap were generated with threshold of Benjamini-Hochberg p value < 0.05, following the ReBoot procedure (Faust et al 2012).

3.3. Results

3.3.1. Environmental parameters

As expected, the water columns of ELB, WLB and FRX exhibited high spatial stratification at the level of major physical and chemical parameters (Table 3.1). The chemoclines in these lakes are 18-20m, 15-18m and 10-12m for ELB, WLB and FRX respectively. Above the chemocline, the water columns were generally oligotrophic; and phosphorus concentration was extremely low throughout the water column of both lobes of Lake Bonney (Dolhi et al. 2015). Ammonia and phosphorus level increased dramatically in the chemoclines and deeper water layers. Rapidly increasing of nitrate concentration was only observed in ELB and has been implicated in incomplete N-cycle in this lake (Ward and Priscu 1998). Nutrient ratios in these lakes were generally unbalanced relative to that of the Redfield ratio (Martiny et al 2014, Redfield 1934). Irradiance within the water columns was < 10% of incident in all lakes; particularly in FRX, where Only 1% of the incident reaches the water column as a consequence of low transparency of the ice covers. Both ELB and WLB exhibited chl a peaks under the ice cover (5-6 m depth) as well as DCM at a depth of 15 m. at the DCM in FRX was located at 9-11 m depth.

3.3.2. Ilumina MiSeq sequencing summary

After demultiplexing and initial quality filtering, total numbers of 6,411,268 and 1,962,313 reads for 16S and 18S rRNA genes were obtained, respectively. The sequences were clustered into OTUs using open-OTU-picking procedure which generated 1155 and 320 bacterial and eukaryal OTUs, respectively. Known chloroplast sequences were filtered from the 16S rRNA OTUs to reduce the influence of chloroplast 16S rRNA genes when estimating the richness of the samples. Samples were randomly subsampled to smallest sample size (5000 sequences). Rarefaction analysis based on OTUs indicated that most of the libraries reached the plateau level

(Figure A3.1). The observed and estimated (Chao1) OTU numbers of each sample are shown in Figure 3.1. The number of prokaryotic species per sample ranged from 115 to 273 for observed and from 142 to 387 for Chao1 estimated, while the number of eukaryotic species per sample ranged from 16 to 47 for observed and from 22 to 84 for estimated. At the level of community richness among the three lake, FRX exhibited higher prokaryotic species richness than ELB and WLB in observed species numbers (Tukey’s HSD test, p < 0.05) while FRX only showed higher richness than ELB in estimated species numbers (Tukey’s HSD test, p < 0.05). No other significant difference was found within pairwise comparison between lakes (Table 3.2). Relative abundance of unclassified bacterial OTUs ranged from 0.9% - 12% of the total OTUs; in contrast unclassified eukaryotic sequences were in the range between 7% - 70% of the total OTUs in individual samples. Deep samples (ELB 30 m and FRX 14 m) had highest percentage of both bacterial and eukaryal unclassified OTUs. For community composition and further analyses, unclassified OTUs were removed.

3.3.3. Bacterial community composition

Bacterial community diversity was assessed by sequencing the V4 region of the 16S rRNA gene. Phylotyping of the bacterial communities from the water columns of ELB, WLB and FRX revealed that the most abundant OTUs among the MDV lake communities belonged to the phyla Bacteroidetes, Actinobacteria and Proteobacteria, which together contributed over 85% of the total OTU counts in each lake (Figure 3.2). In FRX samples representative sequences belonging to 28 bacterial phyla were identified in which 9 phyla (Armatimonadetes, Caldiserica, Lentisphaerae, GN04, Hyd24-12, NKB19, OP3, WS1 and WWE1) were only detected in FRX samples. All three lakes shared 13 phyla, with 3 phyla (Tenericutes, Spirochaetes and OP9 clade) only found in WLB and FRX, and 2 distinct phyla (SR1 and GN02 clades) only shared between ELB and FRX. In WLB, Actinobacteria were abundant in the upper layers and were replaced by Gammaproteobacteria below chemocline. In the oxic-anoxic transition zone (25m) and anoxic zone (30 and 35m), the abundance of Fimicutes increased rapidly. In the oxic zone of FRX (5-11m), Bacteroidetes (32 ± 8%) and Actinobacteria (42 ± 8%) dominated the oxic zone,

53 while the abundance declined to 10% and 6% respectively in the anoxic zone (12 and 14m). In the deep anoxic layers of FRX, the relative abundance of Deltaproteobacteria, Hyd24-12 and OP9 increased.

Patterns of bacterial community structure were examined using non-metric multidimensional scaling based on Bray-Curtis dissimilarity matrix (Figure 3.4). Bacterial communities from the upper oxic zones of ELB and WLB clustered together (ANOSIM R = 0.06, p = 1), while communities from oxic layers of FRX, as well as anoxic waters of FRX and WLB differed significantly from each other and clustered separately. The results of the full ANOSIM analysis are reported in Table 3.3.

3.3.4. Eukaryotic community composition

A total of 12 phyla were detected among the microbial eukaryote communities from the three MDV lakes (Figure 3.3). In ELB Cryptophyta and Chlorophyta were abundant throughout the water columns, representing n 33± 21% and 24 ± 7% of total 18S rRNA OTU counts, respectively. While sequences related to the genus Geminigera dominated in the Cryptophyta group (Figure A3.2F), chlorophytes were more diverse. OTUs related to Mychonastes, Chlamydomonas (Figure A3.2D) and Chlorella were the most abundant Chylorophyta genera. OTUs related to as Prasinophyceae (Figure A3.2E) dominated the chlorophyte population in the chemocline. Within the phylum Haptophyta, >90% of the OTUs were identified as in the Isochrysidales (Figure A3.2G) order. Haptophyte sequences were detected in all layers of ELB, but only 1 to 2.6% of total OTUs were haptophyta in the mixolimnion which is lower than the abundance with a range of 16 to 35% at or below chemocline. In contrast, in WLB, Cryptophyta (mostly Geminigera) was abundant (52 to 70% of the eukaryal community) in the shallow layers (5-13m), but declined in the chemocline (15m) and were rare in the deep water column (20-35m) (Figure 3.3). Chlorophytes replaced the cryptophytes in the deeper layers of WLB; however, the deep chlorophyte population was related to a nonmotile nanophytoplankton, Mychonastes, while

54 shallow populations were dominated by Chlamydomonas, a large biflagellate species. The Haptophyta distribution in WLB was comparable with that of ELB: haptophyte abundance increased from less than 8% above chemocline (5-13m) to the range of 10 to 24% in the deep layers (15-35m). In both ELB and WLB, Stramenopile sequences were detected throughout the water columns, with peaks in abundance near the bottom of chemoclines in both lakes. In addition, WLB had higher abundance of Stramenopiles in the deep layer compare with ELB, which were dominated by a non-pigmented Chrysophyte (Paraphysomonas) (Figure A3.2H). In contrast with Lake Bonney communities, the FRX eukaryotic community was dominated by Cryptophyta throughout the water column which represented from 76 to 97% of the total sequences. Stramenopiles were the second most abundant eukaryote group in FRX samples, and were dominated by sequences related to diatoms (Figure 3C). Discoba and Metamonads (in the Excavata pylum) were only found in FRX samples, and no other excavates were identified in the ELB or WLB samples.

The ordination based on Bray-Curtis dissimilarity matrix indicated that ELB and WLB eukaryote communities clustered closely and were separate from the FRX cluster (Figure 3.5). In addition, eukaryote diversity and abundance in FRX was largely distinct from ELB or WLB (ANOSIM p < 0.001). When comparing ELB and WLB communities, despite ANOSIM p < 0.05, the small R value (0.293) indicated only slightly differences between these two lakes. Our results also indicated that eukaryotic communities in freshwater mixolimnion were significantly different with chemocline and monimolimnion (Table 3.3).

3.3.5. Co-occurrence microbial network and associated environmental factors

We combined several methods to quantify the compositional dissimilarity and correlation (Bray- Curtis and Kullback-Leibler dissimilarities as well as Pearson and Spearman correlations) to investigate co-occurrence patterns within and between bacterial and eukaryotic communities. A total of 102 nodes (OTUs) (listed in Table 3.4) and 326 interactions (edges) were identified

(Figure 3.6). The network analysis detected 255 positive and 71 negative interactions between OTUs. Most of the negative interactions were between eukaryotic groups. Cryptophytes were negatively correlated with haptophytes (related to the genus Isochrysis), chlorophytes (related to the genus Mychonastes), and choanoflagellates(related to the Acanthocidae), while no direct interactions were identified between the latter three (Figure 3.6). Few interactions between eukaryotes and bacteria were predicted. The only recognized bacteria that directly interacted with eukaryotes were a few Actinobacteria in the order of Actinomycetales and Acidimicrobiales. Environmental factors (i.e., N to P ratio and conductivity) exhibited distinct correlations with eukaryotic organisms. For example, N to P ratio and conductivity had a positive correlation with haptophytes while both environmental factors were negatively correlated with cryptophytes.

Interactions between bacteria were generally positive, with exception of a few OTUs related to Proteobacteria. Despite these negative relationships, the bacterial networks divided into several sub-networks. All Gammaproteobacteria and Verrucomicrobia were found in Sub-Network 1 along with several heterotrophic Bacteroidetes and Actinobacteria. All Planctomycetes were found in Sub-Network 2 with few Rickettsiaceae and Microthrixaceae. In the third sub-network, the average edges (interactions) was 4.37 which is significant higher than the average 3.07 edges per node for the whole net-work (p < 0.05). Establishment of these sub-networks indicated that certain bacteria might form functional groups in the natural environment.

3.4. Discussion

To date spatial distribution patterns in microbial communities within the stratified aquatic environment of the MDV lakes has been limited to date due to a lack of depth of (sequence) sampling. In this study, small subunit rRNA gene amplicon sequencing for both the bacterial and protistian was performed using Illumina next generation sequencing platform from samples spanning the water columns of three highly studied lakes Thus, the data from the study provided the new insights into MDV lake microbial community structure and their potential environmental and biological drivers in these aquatic ecosystems.

Due to the strong vertical chemical gradients, redox transitions and solar irradiance availability, stratified microbial communities are the major feature of meromictic lakes (Andrei et al 2015, Comeau et al 2012, Laybourn-Parry et al 1995, Newton et al 2011). As expected, stratification of both bacterial and eukaryotic microbial communities was observed in all three lakes in this study (Figure 3.3-3.5, Table 3.3). In individual lakes, the community structure of microbial populations was more similar within same regions in the water column: this was particularly true for the upper water columns of ELB and WLB which are connected for several weeks in the austral summer. However, major factors controlling the distribution of bacteria and eukaryotes differed: bacterial communities were distinct in the oxic vs. the anoxic zones, while eukaryotic communities were different in the shallow mixolimnion vs. the deep chemocline and monimolimnion (Table 3.3). This differential clustering of bacterial vs. eukaryal communities likely reflects the strong influence of oxygen vs. light availability, respectively.

Results based on statistical analysis of bacterial OTUs generated from the sequencing were similar to outcomes from previous studies using different molecular and microscopic methods (Bielewicz et al 2011a, Kong et al 2012b, Kong et al 2014b, Laybourn-Parry et al 1997). The water column in ELB is mostly oxygenated while WLB is anoxic zone below 25m, and FRX has deep anoxic region (below 12m) as well. Our sampling depths in this study covered several

57 depths in the anoxic zone of WLB and FRX. Direct comparison of the major phyla in both lakes indicated that although Bacteroidetes, Actinobacteria and Betaproteobacteria were most abundant groups of bacteria in the oxic zones of all three lakes, the anoxic bacterial communities were very different (Figure3 and 5; Table 3.3). In the WLB anoxic zone, Gammaproteobacteria (majorly Alteromonadaceae and Marinobacter) and Firmicutes (majorly Acidaminobacteraceae) exhibited relatively high abundance, by contrast, Deltaproteobacteria (majorly Desulfobulbaceae, Geobacteraceae and Syntrophaceae) were abundant in the deep FRX layers. WLB is characterized by a hypersaline deep layer water and both Alteromonadaceae and Acidaminobacteraceae were reported to be abundant in other high salinity aquatic environments (Andrei et al 2015, Kharroub et al 2011, López-Pérez and Rodriguez-Valera 2014); The Alteromonadaceae family comprises a diverse group of gammaproteobacteria which are mostly marine in origin and require sodium for growth. Many display diverse potential for degrading a variety of substrates, and are often associated with particulate material and marine snow (Ivanova et al 2004). One Antarctic strain, Glaciecola punicea, was identified to be associated with sea-ice diatom assemblages (Bowman et al 1998). Acidominobacter is an obligate anaerobe: several strains exhibit the ability to degrade various amino acids and are often associated with hydrogen-consuming organisms (Meijer et al 1999) .The WLB water column is also in direct contact during the summer with a unique geochemical feature associated with the Taylor Glacier, the so called “Blood Falls”. Blood falls represents a unique ecosystem of an ancient brine pool which is enriched in reduced iron (Mikucki et al 2004, Mikucki et al 2009). In contrast, the deep layer of FRX water column is associated with high biogenic methane and reduce forms of sulfur compounds (e.g., hydrogen sulfide and sulfite) (Karr et al 2006, Sattley and Madigan 2006). We found that Desulfobulbaceae dominated Deltaproteobacteria sequences in samples from these layers. All members of the Desulfobulbaceae family have been isolated from anoxic environment and are involved in anaerobic oxidation of methane or other simple organic carbon molecules in environment of similar conditions (Kuever 2014, Lloyd et al 2006, Teske et al 2002). Members of this group are typically sulfate reducers, and it is likely that this organism may play a role in the active sulfur cycle in FRX Sulfate reducing Desulfovibrio spp. have recently been isolated from FRX (Sattley and Madigan 2010). Interestingly, Sattley and Madigan (2010) observed that carbon substrate preference of the SRB population residing in FRX differed throughout the water column. Communities residing below the chemocline were stimulated by

58 lactate, while the SRB community in the deep water column exhibit a preference for acetate. Thus, the dominant organisms detected in this study in the deep anoxic water might form niche specific communities and possibly play important roles in biogeochemical cycling. The dissimilarity of the organism distribution in anoxic waters between WLB and FRX is likely indicative of additional environmental conditions influencing bacterial community structure in these dark, anoxic environments (for e.g., the influence of Blood Falls).

Spatial differences in the eukaryotic community structure within and between the three study lakes were also noted in this study (Table 3.3). A diverse group of photosynthetic and mixotrophic protists were detected throughout the water columns, even in depths where PAR is not available. In lake Bonney, Chlamydomonas was highly abundant in shallow, relatively fresh water, but diminished in deep waters which are characterized by higher nutrients, but limited PAR. Similar trends were reported in previous studies based on abundance and diversity of functional genes: the larger biflagellate chlorophyte species appear to favor high PAR and low nutrient conditions. Since members of this genus are often obligately photosynthetic, PAR availability may be a major driver in structuring the distribution of this group of MDV phytoplankton (Bielewicz et al 2011a, Dolhi et al 2015c, Kong et al 2014b). Cryptophytes were abundant in all three lakes, but the distribution was different among lakes. They dominated the entire water column in FRX, and the similar phenomenon was observed by Marshall and Laybourn-Parry (2002). As the mixotrophic cryptophytes do not solely require solar irradiance to provide energy via photosynthesis, they can live in the aphotic zone of the water column by grazing on bacteria (LAYBOURN-PARRY et al 2005, Marshall and Laybourn-Parry 2002). In addition, tolerance to low oxygen environment and high sulfate in FRX may also allow these organisms to dominate the protist communities in FRX (Gasol et al 1993). In contrast to FRX, cryptophytes tended to be abundant in the shallow layers, but were generally restricted from deeper, hypersaline waters. Network analysis in this study indicates that cryptophyte populations are negatively correlated with salinity which suggests the influence of specific environmental factors on the distribution of these organisms (Figure 3.6). In addition, competition between mixotrophs and pure heterotrophs (e.g., choanoflagellates and other heterotrophic nanoflagellates etc.) could be another negative factor contraining spatial distribution of the of cryptophytes in

Lake Bonney (Roberts and Laybourn-Parry 1999). In support of this suggestion, Chrysophyta (Paraphysomonas) dominated in the Stramenopile supergroup in ELB and WLB, peaking within the chemocline, and agrees with other reports that Chrysophta are frequently detetected in suboxic to anoxic habitats where they play important roles in transferring carbon and nutrients from secondary primary production to higher trophic levels (Park and Simpson 2010, Stock et al 2009, Wylezich and Jürgens 2011). Last, the eukaryal data represents classified OTUs. In the shallower depths of the lakes, >95% of the 18S rRNA sequences were related to known eukaryal sequences in the databases. However, in some of the deeper samples, >70% of our 18S rRNA sequences were unclassified, suggesting that the monimolimnia may harbor novel microbial life. This is currently being pursued in a metagenomics project in collaboration with JGI.

While networks based on phylogenetic information cannot identify the underlying mechanism of metabolic or other interactions, they can aid in defining predicted interactions between microbial partners. This study applied co-occurrence network analysis to identify non-random patterns of co-occurrence between: (i) eukaryotes-eukaryotes, (ii) bacteria-eukaryotes, (iii) bacteria-bacteria, and (iv) either group with a variety of environmental factors. Interestingly, several groups of bacteria participated in multiple sub-networks. These functional groups might responsible for biogeochemical processes across different organisms in the MDV aquatic ecosystem. In the natural environment, microbial interaction are complicated (Carpenter 2012, Morris et al 2013), but the outcomes led by interactions between microorganisms are usually including win, loss or neutral. Despite that we identified many interconnection between organisms in the lakes samples, however it is hard to investigate the community dynamics based on the pairwise correlation of abundance of a single time point (Cram et al 2015). To identify, for example, parasitism or predation (Faust and Raes 2012b) or cross-feeding between photosynthetic organisms and heterotrophic bacteria (Amin et al 2012, Wingender et al 2012), establishing short term (days to months) and/or long term (year round) community dynamic models (Cram et al 2015) or examining organisms at higher resolution to gather more directly information of their interactions (Heywood et al 2011, Li et al 2016a, Martinez-Garcia et al 2012).

3.5. Acknowledgements:

Figure 3.1Observed (orange) and estimated (blue) alpha diversity of 16S and 18S OTUs.

Figure 3.2 16S OTUs relative abundance at phylum level.

Figure 3.3 18S OTUs relative abundance at phylum level.

Figure 3.4. NMDS plot of bacterial communities (OTU counts, Bray-Curtis dissimilarity, stress = 0.072). Oxic and anoxic: upper oxic zone and deep anoxic zone in the water columns respectively. Shaded areas indicate 75% similarity within each lake.

Figure 3.5. NMDS plot of eukaryotic communities (OTU counts, Bray-Curtis dissimilarity, stress = 0.077). Shaded areas indicate 90% similarity within each lake. Shallow: mixolimnion. Deep: chemocline +monolimnion.

Figure 3.6. Association network of concurrent of bacteria, eukaryotes and correlation with environmental parameters. Nodes: OTUs and environmental factors; Edges: positive (blue solid lines) and negative (red dash lines) interactions; Degrees: size of the nodes (edges of individual nodes).

Table 3.1. Summary of major physicochemical parameters in the studied lakes. Mixolimnion; Chemocline; Monolimnion; Oxic; Anoxic

+ - 3- Depth PAR Temperature Conductivity DO NH4 NO3 PO4 N:P DIC Chl

-2 -1 -1 -1 -1 (m) (µm photons m s ) (˚C) (mS cm ) mg O2 L uM uM uM uM ug L

5 16.0 -4.7 1.3 26.9 1.0 8.0 0.1 16.4 1.1 6.9

6 15.3 -3.5 1.4 29.1 1.1 8.8 0.1 17.2 1.3 5.7

10 12.0 -1.0 3.6 38.4 0.8 10.8 0.0 23.7 2.2 4.3

13 9.0 0.5 8.5 39.0 2.1 22.4 0.0 47.1 6.2 3.9

15 7.2 0.7 15.7 45.2 11.4 33.9 0.1 83.6 11.7 4.0

18 4.9 1.0 30.2 37.5 20.2 56.0 0.1 122.8 10.7 2.3 ELB

20 4.0 0.8 51.0 29.7 63.2 80.0 0.2 207.4 16.3 1.7

22 3.4 0.6 79.1 13.6 237.1 116.6 0.4 411.4 19.0 3.0

25 2.2 0.0 106.7 0.4 170.1 132.7 1.3 172.3 13.8 0.3

30 0.9 -1.4 114.0 0.3 196.0 169.3 0.9 268.8 5.4 0.2

5 12.6 0.0 1.2 25.4 1.2 6.3 0.5 14.4 0.7 11.1

10 6.6 0.1 3.5 33.3 1.0 8.4 0.5 17.4 2.5 11.9

13 4.1 -2.5 7.6 47.5 1.3 8.9 0.5 19.0 5.1 5.2

15 2.5 -3.4 17.9 48.5 44.1 18.1 0.6 97.1 41.9 6.6

WLB 20 1.0 -6.3 65.6 0.5 170.1 12.4 0.7 260.6 63.9 2.4

25 0.5 -9.6 73.1 0.0 210.3 0.5 0.8 273.8 78.6 0.0

30 0.0 -11.4 78.0 0.0 255.2 0.8 0.9 278.2 88.1 0.0

5 13.5 -8.8 0.6 25.5 0.1 0.1 0.0 2.1 2.2 0.0

7 8.3 -7.1 2.1 25.9 0.0 0.0 0.1 1.8 5.4 7.6

8 5.0 -6.2 2.8 29.0 0.0 0.0 0.1 1.7 10.9 9.0

9 3.0 -5.6 3.4 28.4 0.0 0.0 0.1 1.6 13.3 18.8 FRX 11 0.7 -4.9 5.0 6.6 20.3 0.2 2.3 7.8 19.7 22.0

12 0.2 -4.8 5.8 0.0 77.9 1.4 7.6 9.9 30.8 7.7

14 0.0 -4.8 7.1 0.0 260.5 2.0 22.6 11.4 42.4 1.0

Table 3.2. Pairwise comparison of species richness between lakes using Tukey’s HSD test.

* p-value less than 0.05

Pairs Bacteria (observed) Bacteria (estimated) Eukaryotes (observed) Eukaryotes (estimated)

Q score p-value Q score p-value Q score p-value Q score p-value

ELB-WLB 0.956 0.779 1.733 0.445 2.257 0.259 2.370 0.226

ELB-FRX 4.984 *0.003 3.571 *0.040 0.072 0.999 1.887 0.385

WLB-FRX 4.028 *0.018 1.837 0.404 2.329 0.238 0.483 0.938

Table 3.3 ANOSIM of communities from different lakes and layers (p value larger than 0.05 is indicated in red). (EO: ELB oxic; WO: WLB oxic, WA: WLB anoxic; FO; FRX oxic; FA; FRX anoxic; ES: ELB shallow; ED: ELB deep; WS: WLB shallow; WD: WLB deep; FS: FRX shallow; FD: FRX deep)

Pair Data Type R value p (corrected)

EO-WO Bacteria 0.057 1.0000

EO-WA Bacteria 0.954 0.0010

EO-FO Bacteria 0.663 0.0010

EO-FA Bacteria 0.960 0.0070

WO-WA Bacteria 0.688 0.0040

WO-FO Bacteria 0.553 0.0030

WO-FA Bacteria 0.758 0.0150

WA-FO Bacteria 1.000 0.0010

WA-FA Bacteria 1.000 0.0170

FO-FA Bacteria 1.000 0.0130

ELB-WLB Eukaryotes 0.283 0.0015

ELB-FRX Eukaryotes 0.971 0.0003

WLB-FRX Eukaryotes 0.818 0.0003

ES-ED Eukaryotes 0.570 0.0030

ES-WS Eukaryotes 0.293 0.4620

ES-WD Eukaryotes 0.857 0.0135

ES-FS Eukaryotes 1.000 0.0420

ES-FD Eukaryotes 1.000 0.0045

ED-WS Eukaryotes 0.844 0.0030

ED-WD Eukaryotes 0.554 0.0075

ED-FS Eukaryotes 1.000 0.0015

ED-FD Eukaryotes 1.000 0.0015

WS-WD Eukaryotes 0.771 0.0105

WS-FS Eukaryotes 0.946 0.0345

WS-FD Eukaryotes 0.960 0.0060

WD-FS Eukaryotes 1.000 0.0060

WD-FD Eukaryotes 1.000 0.0030

FS-FD Eukaryotes 0.870 0.0045

Table 3.4. List of nodes in co-occurrence network.

NodeLabel kingdom phylum class order family genus

Act_1 Bacteria Actinobacteria Actinobacteria Actinomycetales none none

Act_2 Bacteria Actinobacteria Thermoleophilia Solirubrobacterales none none

Act_3 Bacteria Actinobacteria Actinobacteria Actinomycetales ACK-M1 none

Act_4 Bacteria Actinobacteria Thermoleophilia Solirubrobacterales none none

Act_5 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales Microthrixaceae none

Act_6 Bacteria Actinobacteria Thermoleophilia Solirubrobacterales none none

Act_7 Bacteria Actinobacteria Actinobacteria1 Actinomycetales none none

Act_8 Bacteria Actinobacteria Actinobacteria Actinomycetales ACK-M1 none

Act_9 Bacteria Actinobacteria Thermoleophilia none none none

Act_10 Bacteria Actinobacteria Acidimicrobiia none none none

Act_11 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales Microthrixaceae none

Candidatus Act_12 Bacteria Actinobacteria Actinobacteria1 Actinomycetales Microbacteriaceae Rhodoluna

Act_13 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales C111 none

Candidatus Act_14 Bacteria Actinobacteria Actinobacteria Actinomycetales Microbacteriaceae Rhodoluna

Act_15 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales C111 none

Act_16 Bacteria Actinobacteria Actinobacteria1 Actinomycetales Microbacteriaceae none

Act_17 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales none none

Act_18 Bacteria Actinobacteria Actinobacteria Actinomycetales none none

Act_19 Bacteria Actinobacteria Actinobacteria1 none none none

Candidatus Act_20 Bacteria Actinobacteria Actinobacteria1 Actinomycetales Microbacteriaceae Aquiluna

Act_21 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales none none

Candidatus Act_22 Bacteria Actinobacteria Actinobacteria1 Actinomycetales Microbacteriaceae Aquiluna

Act_23 Bacteria Actinobacteria Actinobacteria Actinomycetales none none

Act_24 Bacteria Actinobacteria Actinobacteria Actinomycetales Microbacteriaceae none

Candidatus Act_25 Bacteria Actinobacteria Actinobacteria Actinomycetales Microbacteriaceae Aquiluna

Act_26 Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales C111 none

Act_27 Bacteria Actinobacteria Actinobacteria1 Actinomycetales ACK-M1 none

Bact_1 Bacteria Bacteroidetes Saprospirae Saprospirales Chitinophagaceae Sediminibacterium

Bact_2 Bacteria Bacteroidetes Saprospirae Saprospirales Chitinophagaceae none

Bact_3 Bacteria Bacteroidetes [Saprospirae] [Saprospirales] Chitinophagaceae Sediminibacterium

Bact_4 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Cryomorphaceae Fluviicola

Bact_5 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Flavobacteriaceae none

Bact_6 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Flavobacteriaceae Polaribacter

Bact_7 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Flavobacteriaceae Sediminicola

Bact_8 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Flavobacteriaceae Sediminicola

Bact_9 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales Flavobacteriaceae Polaribacter

Bact_10 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales NS11-12 none

Bact_11 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales NS11-12 none

Bact_12 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales Sphingobacteriaceae none

Bact_13 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales Sphingobacteriaceae none

Bact_14 Bacteria Bacteroidetes Saprospirae none none none

Bact_15 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales none none

Bact_16 Bacteria Bacteroidetes Sphingobacteriia none none none

Bact_17 Bacteria Bacteroidetes Flavobacteriia none none none

Bact_18 Bacteria Bacteroidetes Flavobacteriia Flavobacteriales none none

Bact_19 Bacteria Bacteroidetes Saprospirae Saprospirales none none

Bact_20 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales none none

Bact_21 Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales none none

Bact_22 Bacteria Bacteroidetes none none none none

Alpha_1 Bacteria Proteobacteria Alphaproteobacteria Rickettsiales Rickettsiaceae none

Alpha_2 Bacteria Proteobacteria Alphaproteobacteria Rickettsiales Rickettsiaceae none

Alpha_3 Bacteria Proteobacteria Alphaproteobacteria Rickettsiales none none

Alpha_4 Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales none none

Alpha_5 Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales none none

Alpha_6 Bacteria Proteobacteria Alphaproteobacteria none none none

Beta_1 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae RS62

Beta_2 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae RS62

Beta_3 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae none

Beta_4 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae none

Beta_5 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae none

Beta_6 Bacteria Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae none

Beta_7 Bacteria Proteobacteria Betaproteobacteria none none none

Gamma_1 Bacteria Proteobacteria Gammaproteobacteria Acidithiobacillales none none

Gamma_2 Bacteria Proteobacteria Gammaproteobacteria Acidithiobacillales none none

Gamma_3 Bacteria Proteobacteria Gammaproteobacteria Alteromonadales HTCC2188 HTCC

Gamma_4 Bacteria Proteobacteria Gammaproteobacteria Alteromonadales HTCC2188 none

Gamma_5 Bacteria Proteobacteria Gammaproteobacteria Alteromonadales HTCC2188 HTCC

Gamma_6 Bacteria Proteobacteria Gammaproteobacteria Alteromonadales none none

Gamma_7 Bacteria Proteobacteria Gammaproteobacteria none none none

NC unclassified0 none none none none none

Plan_1 Bacteria Planctomycetes Planctomycetia Pirellulales Pirellulaceae none

Plan_10 Bacteria Planctomycetes none none none none

Plan_2 Bacteria Planctomycetes Planctomycetia Pirellulales Pirellulaceae none

Plan_3 Bacteria Planctomycetes Planctomycetia Pirellulales none none

Plan_4 Bacteria Planctomycetes Planctomycetia Planctomycetales Planctomycetaceae Planctomyces

Plan_5 Bacteria Planctomycetes Planctomycetia Planctomycetales Planctomycetaceae Planctomyces

Plan_6 Bacteria Planctomycetes Planctomycetia Planctomycetales Planctomycetaceae none

Plan_7 Bacteria Planctomycetes Planctomycetia Planctomycetales Planctomycetaceae Planctomyces

Plan_8 Bacteria Planctomycetes Planctomycetia Planctomycetales none none

Plan_9 Bacteria Planctomycetes Planctomycetia none none none

Prot Bacteria Proteobacteria none none none none

Ver_1 Bacteria Verrucomicrobia Verrucomicrobiae Verrucomicrobiales Verrucomicrobiaceae Verrucomicrobium

Ver_2 Bacteria Verrucomicrobia Verrucomicrobiae Verrucomicrobiales Verrucomicrobiaceae Verrucomicrobium

Ver_3 Bacteria Verrucomicrobia none none none none

Mychonastes sp. Chloro_1 Eukaryota Archaeplastida Chloroplastida Chlorophyta none LBMa-1

Mychonastes sp. Chloro_2 Eukaryota Archaeplastida Chloroplastida Chlorophyta none LBMa-1

Nannochloris sp. Chloro_3 Eukaryota Archaeplastida Chloroplastida Chlorophyta Trebouxiophyceae ANR-9

Nannochloris sp. Chloro_4 Eukaryota Archaeplastida Chloroplastida Chlorophyta Trebouxiophyceae ANR-9

Geminigera Crypto_1 Eukaryota Cryptophyceae Cryptomonadales Geminigera none cryophila

Geminigera Crypto_2 Eukaryota Cryptophyceae Cryptomonadales Geminigera none cryophila

Geminigera Crypto_3 Eukaryota Cryptophyceae Cryptomonadales Geminigera none cryophila

Crypto_4 Eukaryota Cryptophyceae Cryptomonadales Geminigera none none

uncultured marine Crypto_5 Eukaryota Cryptophyceae Cryptomonadales none none eukaryote

uncultured marine Crypto_6 Eukaryota Cryptophyceae Cryptomonadales none none eukaryote

Geminigera Crypto_7 Eukaryota Cryptophyceae Cryptomonadales none none cryophila

uncultured Hapto_1 Eukaryota Haptophyta Prymnesiophyceae Isochrysidales ESS2202 haptophyte

uncultured Hapto_2 Eukaryota Haptophyta Prymnesiophyceae Isochrysidales ESS2202 haptophyte

Hapto_3 Eukaryota Haptophyta Prymnesiophyceae Isochrysidales ESS2202 none

Hapto_4 Eukaryota Haptophyta Prymnesiophyceae Isochrysidales none none

Hapto_5 Eukaryota Haptophyta Prymnesiophyceae none none none

Hapto_6 Eukaryota Haptophyta none none none none

3.6. References

Amaral-Zettler LA, McCliment EA, Ducklow HW, Huse SM (2009). A method for studying protistan diversity using massively parallel sequencing of V9 hypervariable regions of small- subunit ribosomal RNA genes. PLoS ONE 4: e6372.

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76: 667-684.

Anderson MJ (2001). A new method for non-parametric multivariate analysis of variance. Austral Ecol 26: 32-46.

Anderson MJ, Crist TO, Chase JM, Vellend M, Inouye BD, Freestone AL et al (2011). Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecol Lett 14: 19-28.

Andrei A-Ş, Robeson MS, Baricz A, Coman C, Muntean V, Ionescu A et al (2015). Contrasting taxonomic stratification of microbial communities in two hypersaline meromictic lakes. ISME J 9: 2642-2656.

Aronesty E (2011). ea-utils: Command-line tools for processing biological sequencing data. Durham, NC: Expression Analysis.

Azam F, Smith D, Steward G, Hagström Å (1994). Bacteria-organic matter coupling and its significance for oceanic carbon cycling. Microbial Ecology 28: 167-179.

Azam F (1998). Microbial control of oceanic carbon flux: the plot thickens. Science 280: 694.

Bielewicz S, Bell E, Kong W, Friedberg I, Priscu JC, Morgan-Kiss RM (2011). Protist diversity in a permanently ice-covered Antarctic lake during the polar night transition. The ISME journal 5: 1559-1564.

Bowman JP, Mccammon S, Brown J, McMeekin T (1998). Glaciecola punicea gen. nov., sp. nov. and Glaciecola pallidula gen. nov., sp. nov.: psychrophilic bacteria from Antarctic sea-ice habitats. Int J Syst Evol Microbiol 48: 1213-1222.

Bray JR, Curtis JT (1957). An Ordination of the Upland Forest Communities of Southern Wisconsin. Ecological Monographs 27: 326-349.

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK et al (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods 7: 335- 336.

Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N et al (2012). Ultra- high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. The ISME journal 6: 1621-1624.

Carpenter SR (2012). Complex interactions in lake communities. Springer Science & Business Media.

Cavicchioli R (2015). Microbial ecology of Antarctic aquatic systems. Nature Reviews Microbiology 13: 691-706.

Chela-Flores J (2011). On the possibility of biological evolution on the moons of Jupiter. The Science of Astrobiology. Springer. pp 151-170.

Comeau AM, Harding T, Galand PE, Vincent WF, Lovejoy C (2012). Vertical distribution of microbial communities in a perennially stratified Arctic lake with saline, anoxic bottom waters. Sci Rep 2: 604.

Conlan S, Kong HH, Segre JA (2012). Species-level analysis of DNA sequence data from the NIH Human Microbiome Project. PLoS One 7: e47075.

Cotner JB, Biddanda BA (2002). Small players, large role: microbial influence on biogeochemical processes in pelagic aquatic ecosystems. Ecosystems 5: 105-121.

Cram JA, Xia LC, Needham DM, Sachdeva R, Sun F, Fuhrman JA (2015). Cross-depth analysis of marine bacterial networks suggests downward propagation of temporal changes. The ISME journal 9: 2573-2586.

Doran PT, Lyons WB, McKnight DM (2010). Life in Antarctic deserts and other cold dry environments: astrobiological analogs, vol. 5. Cambridge University Press.

Edgar RC (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26: 2460-2461.

Eiler A, Heinrich F, Bertilsson S (2012). Coherent dynamics and association networks among lake bacterioplankton taxa. ISME J 6: 330-342.

Faust K, Raes J (2012). Microbial interactions: from networks to models. Nat Rev Micro 10: 538-550.

Faust K, Sathirapongsasuti JF, Izard J, Segata N, Gevers D, Raes J et al (2012). Microbial Co- occurrence Relationships in the Human Microbiome. PLoS Comput Biol 8: e1002606.

Fenchel T (2008). The microbial loop–25 years later. Journal of Experimental Marine Biology and Ecology 366: 99-103.

Fenchel T, Blackburn H, King GM (2012). Bacterial biogeochemistry: the ecophysiology of mineral cycling. Academic Press.

Gasol JM, García-Cantizano J, Massana R, Guerrero R, Pedrós-Alió C (1993). Physiological ecology of a metalimnetic Cryptomonas population: relationships to light, sulfide and nutrients. Journal of Plankton Research 15: 255-275.

Gordon DA, Priscu J, Giovannoni S (2000). Origin and Phylogeny of Microbes Living in Permanent Antarctic Lake Ice. Microb Ecol 39: 197-202.

Hammer Ø, Harper D, Ryan P (2001). PAST-PAlaeontological STatistics, ver. 1.89. Palaeontologia electronica 4: 1-9.

Heywood JL, Sieracki ME, Bellows W, Poulton NJ, Stepanauskas R (2011). Capturing diversity of marine heterotrophic protists: one cell at a time. The ISME journal 5: 674-684.

Ivanova EP, Flavier S, Christen R (2004). Phylogenetic relationships among marine Alteromonas-like proteobacteria: emended description of the family Alteromonadaceae and proposal of Pseudoalteromonadaceae fam. nov., Colwelliaceae fam. nov., Shewanellaceae fam.

78 nov., Moritellaceae fam. nov., Ferrimonadaceae fam. nov., Idiomarinaceae fam. nov. and Psychromonadaceae fam. nov. Int J Syst Evol Microbiol 54: 1773-1788.

Jiao N, Herndl GJ, Hansell DA, Benner R, Kattner G, Wilhelm SW et al (2010). Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nature Reviews Microbiology 8: 593-599.

Karr EA, Ng JM, Belchik SM, Sattley WM, Madigan MT, Achenbach LA (2006). Biodiversity of methanogenic and other archaea in the permanently frozen Lake Fryxell, Antarctica. Appl Environ Microbiol 72: 1663-1666.

Kharroub K, Aguilera M, Jiménez-Pranteda ML, González-Paredes A, Ramos-Cormenzana A, Monteoliva-Sánchez M (2011). Marinobacter oulmenensis sp. nov., a moderately halophilic bacterium isolated from brine of a salt concentrator. International journal of systematic and evolutionary microbiology 61: 2210-2214.

Kong W, Ream DC, Priscu JC, Morgan-Kiss RM (2012). Diversity and expression of RubisCO genes in a perennially ice-covered Antarctic lake during the polar night transition. Appl Environ Microbiol 78: 4358-4366.

Kong W, Li W, Romancova I, Prasil O, Morgan-Kiss RM (2014). An integrated study of photochemical function and expression of a key photochemical gene (psbA) in photosynthetic communities of Lake Bonney (McMurdo Dry Valleys, Antarctica). FEMS microbiology ecology 89: 293-302.

Kuczynski J, Stombaugh J, Walters WA, González A, Caporaso JG, Knight R (2012). Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc Microbiol: 1E. 5.1-1E. 5.20.

Kuever J (2014). The Family Desulfobulbaceae. The Prokaryotes. Springer. pp 75-86.

Lauro FM, DeMaere MZ, Yau S, Brown MV, Ng C, Wilkins D et al (2011). An integrative study of a meromictic lake ecosystem in Antarctica. The ISME journal 5: 879-895.

Laybourn-Parry J, Bayliss P, Ellis-Evans JC (1995). The dynamics of heterotrophic nanoflagellates and bacterioplankton in a large ultra-oligotrophic Antarctic lake. Journal of Plankton Research 17: 1835-1850.

Laybourn-Parry J, James MR, McKnight DM, Priscu J, Spaulding SA, Shiel R (1997). The microbial plankton of Lake Fryxell, southern Victoria Land, Antarctica during the summers of 1992 and 1994. Polar Biology 17: 54-61.

Laybourn-Parry J, Marshall WA, Marchant HJ (2005). Flagellate nutritional versatility as a key to survival in two contrasting Antarctic saline lakes. Freshwater Biology 50: 830-838.

Laybourn-Parry J, Bell EM (2014). Ace Lake: three decades of research on a meromictic, Antarctic lake. Polar Biology 37: 1685-1699.

LAYBOURN-PARRY J, Marshall WA, Marchant HJ (2005). Flagellate nutritional versatility as a key to survival in two contrasting Antarctic saline lakes. Freshwater Biology 50: 830-838.

Li W, Podar M, Morgan-Kiss RM (2016). Ultrastructural and single-cell level characterization reveals metabolic versatility in a microbial eukaryote community from an ice-covered Antarctic lake. Applied and environmental microbiology: AEM. 00478-00416.

Lloyd KG, Lapham L, Teske A (2006). An anaerobic methane-oxidizing community of ANME- 1b archaea in hypersaline Gulf of Mexico sediments. Appl Environ Microbiol 72: 7218-7230.

López-García P, Rodríguez-Valera F, Pedrós-Alió C, Moreira D (2001). Unexpected diversity of small eukaryotes in deep-sea Antarctic plankton. Nature 409: 603-607.

López-Pérez M, Rodriguez-Valera F (2014). The Family Alteromonadaceae: 69-92.

Lyons W, Fountain, R, Doran P, Priscu, J et al (2000). Importance of landscape position and legacy: the evolution of the lakes in Taylor Valley, Antarctica. Freshwater Biology 43: 355-367.

Marshall W, Laybourn-Parry J (2002). The balance between photosynthesis and grazing in Antarctic mixotrophic cryptophytes during summer. Freshwater Biology 47: 2060-2070.

Martinez-Garcia M, Brazel D, Poulton NJ, Swan BK, Gomez ML, Masland D et al (2012). Unveiling in situ interactions between marine protists and bacteria through single cell sequencing. The ISME journal 6: 703-707.

Martiny AC, Vrugt JA, Lomas MW (2014). Concentrations and ratios of particulate organic carbon, nitrogen, and phosphorus in the global ocean. Scientific Data 1: 140048.

Meijer WG, Nienhuis-Kuiper ME, Hansen TA (1999). Fermentative bacteria from estuarine mud: Phylogenetic position of Acidaminobacter hydrogenoformans and description of a new type of Gram-negative, propionigenic bacterium as Propionibacter pelophilus gen. nov., sp. nov. Int J Syst Evol Microbiol 49: 1039-1044.

Mikucki JA, Foreman CM, Priscu JC, Lyons WB, Sattler B, Welch KA (2004). Geomicrobiology of Blood Falls: A Saline, Iron-rich Subglacial Feature of Taylor Glacier, Antarctica. Aquat Geochem 10: 199-220.

Mikucki JA, Pearson A, Johnston DT, Turchyn AV, Farquhar J, Schrag DP et al (2009). A contemporary microbially maintained subglacial ferrous "ocean". Science 324: 397-400.

Morris BE, Henneberger R, Huber H, Moissl-Eichinger C (2013). Microbial syntrophy: interaction for the common good. FEMS microbiology reviews 37: 384-406.

Newton RJ, Jones SE, Eiler A, McMahon KD, Bertilsson S (2011). A guide to the natural history of freshwater lake bacteria. Microbiology and Molecular Biology Reviews 75: 14-49.

Park JS, Simpson AG (2010). Characterization of halotolerant Bicosoecida and Placididea (Stramenopila) that are distinct from marine forms, and the phylogenetic pattern of salinity preference in heterotrophic stramenopiles. Environmental microbiology 12: 1173-1184.

Priscu JC, Fritsen CH, Adams EE, Giovannoni SJ, Paerl HW, McKay CP et al (1998). Perennial Antarctic lake ice: an oasis for life in a polar desert. Science 280: 2095-2098.

Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J et al (2007). SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35: 7188-7196.

Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P et al (2013). The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41: D590-596.

Redfield AC (1934). On the proportions of organic derivatives in sea water and their relation to the composition of plankton. university press of liverpool Liverpool, UK.

Reynolds RT, Squyres SW, Colburn DS, McKay CP (1983). On the habitability of Europa. Icarus 56: 246-254.

Rideout JR, He Y, Navas-Molina JA, Walters WA, Ursell LK, Gibbons SM et al (2014). Subsampled open-reference clustering creates consistent, comprehensive OTU definitions and scales to billions of sequences. PeerJ 2: e545.

Roberts EC, Laybourn-Parry J (1999). Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwater Biology 41: 737-746.

Sattley WM, Madigan MT (2006). Isolation, characterization, and ecology of cold-active, chemolithotrophic, sulfur-oxidizing bacteria from perennially ice-covered Lake Fryxell, Antarctica. Appl Environ Microbiol 72: 5562-5568.

Sattley WM, Madigan MT (2010). Temperature and nutrient induced responses of Lake Fryxell sulfate-reducing prokaryotes and description of Desulfovibrio lacusfryxellense, sp. nov., a pervasive, cold-active, sulfate-reducing bacterium from Lake Fryxell, Antarctica. Extremophiles 14: 357-366.

Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D et al (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome research 13: 2498-2504.

Spigel RH, Priscu JC (2013). Physical Limnology of the Mcmurdo Dry Valleys Lakes: 153-187.

Stock A, Jurgens K, Bunge J, Stoeck T (2009). Protistan diversity in suboxic and anoxic waters of the Gotland Deep (Baltic Sea) as revealed by 18S rRNA clone libraries. Aquatic Microbial Ecology 55: 267.

Strickland JD, Parsons TR (1972). A practical handbook of seawater analysis.

Teske A, Hinrichs KU, Edgcomb V, de Vera Gomez A, Kysela D, Sylva SP et al (2002). Microbial Diversity of Hydrothermal Sediments in the Guaymas Basin: Evidence for Anaerobic Methanotrophic Communities. Applied and Environmental Microbiology 68: 1994-2007.

Van der Gucht K, Cottenie K, Muylaert K, Vloemans N, Cousin S, Declerck S et al (2007). The power of species sorting: local factors drive bacterial community composition over a wide range of spatial scales. Proceedings of the National Academy of Sciences 104: 20404-20409.

Van Gemerden H, Mas J (1995). Ecology of phototrophic sulfur bacteria. Anoxygenic photosynthetic bacteria. Springer. pp 49-85.

Ward BB, Priscu JC (1998). l THE ABUNDANCE OF AMMONIUM—OXIDIZING BACTERIA IN LAKE 3 BONNEY, ANTARCTICA DETERMINED BY IMMUNOFLUORESCENCE, PCR 5 ‘AND IN SITU HYBRIDIZATION.

Wingender J, Neu TR, Flemming H-C (2012). Microbial extracellular polymeric substances: characterization, structure and function. Springer Science & Business Media.

Wylezich C, Jürgens K (2011). Protist diversity in suboxic and sulfidic waters of the Black Sea. Environmental microbiology 13: 2939-2956.

Yang N, Welch KA, Mohajerin TJ, Telfeyan K, Chevis DA, Grimm DA et al (2015). Comparison of arsenic and molybdenum geochemistry in meromictic lakes of the McMurdo Dry Valleys, Antarctica: Implications for oxyanion-forming trace element behavior in permanently stratified lakes. Chemical Geology 404: 110-125.

3.7. Appendix

Figure A3.1 Rarefaction curves of 16S (upper) and 18S (bottom) OTUs

Figure A3.2 representive SEM micrographs of protists found in studied lakes.

CHAPTER IV

Ultrastructural and single-cell level characterization reveals metabolic versatility in a microbial eukaryote community from an ice-covered Antarctic lake

Wei Li, Mircea Podar, Rachael M. Morgan-Kiss Author contributions: WL developed methods, performed sequencing and data analyses, microscopy and imaging, and wrote the manuscript. MP performed single-cell isolation and whole genome amplification.

Most of this chapter appeared as: Li W, Podar M, Morgan-Kiss RM (2016). Ultrastructural and single-cell level characterization reveals metabolic versatility in a microbial eukaryote community from an ice-covered Antarctic lake. Applied and environmental microbiology: AEM. 00478-00416.

CHAPTER IV 4.1. Introduction

The microbial loop links the transfer of nutrients and carbon between the microorganisms of the food web, and includes complex ecological interactions between microbial Eukaryotes, Bacteria, Archaea and Viruses. Microbial activity is influenced by bottom-up (e.g., availability of nutrient and energy sources) and top-down (e.g., predation, parasitism, viral lysis) controls, as well as microbe-microbe interactions (e.g., bacterial consortia). Interaction partnerships between microorganisms lead to combinations of win, loss or neutral outcomes for each microbial partner. For example, parasitism or predation are examples of win-loss outcomes (Faust and Raes 2012b), while mutualistic interactions within biofilms and syntrophic interactions (cross- feeding of metabolic products between two species) between an alga and a heterotrophic bacterium are examples of win-win partnerships (Amin et al 2012). Understanding microbial interactions and the influence of environmental factors on microbial distribution patterns is a critical component of deciphering global nutrient fluxes and biogeochemical cycles.

Single-celled eukaryotic microorganisms (i.e., protists) are ubiquitous in every ecosystem on earth and play critical ecological roles in food web dynamics and global carbon and nutrient cycles (Montagnes et al 2012). Diverse protist lineages possess multiple nutritional modes; playing key roles as producers, decomposers, parasites, and predators. Phototrophic protists are important producers that contribute to global primary production and incorporate a significant portion of inorganic carbon into the food web (Caron et al 2009). Predatory protists are major consumers of planktonic phytoplankton and bacteria, providing control over the abundance of these organisms as well as linking primary producers/consumers with higher trophic levels (i.e., metazoans) (Sherr and Sherr 2002a). Mixotrophic protists (combined ability for photosynthesis and phagotrophic ingestion of food particles) are widespread in aquatic ecosystems (Moorthi et al 2009a, Sanders et al 2000b). Recent molecular surveys from marine and freshwater environments have revealed a high diversity of 18S rRNA sequences which are unrelated to existing cultured protists (Caron et al 2004, Epstein and Lopez-Garcia 2008). The application of high-throughput sequencing has begun to reveal protist biogeography (Lara et al 2011, Simon et al 2015a, Zinger et al 2012) as well as seasonal variability (Nolte et al 2010, Simon et al 2015b). 87

The McMurdo Dry Valleys (MDV) of Southern Victoria Land, Antarctica, is a polar desert: with average air temperatures of -20°C and precipitation rates of <10 cm per year (Chela-Flores 2011, Reynolds et al 1983). Numerous marine-derived, perennially-ice covered lakes are the only source of year-round liquid water for life on the Antarctic continent. Permanent ice caps prevent wind mixing and significant nutrient inputs: MDV lake chemistry is vertically stratified in the water column and lakes exhibit oxygen-rich/ultra-oligotrophic surface waters which are separated by permanent chemoclines from ancient, anoxic/saline waters (Green and Lyons 2009). Thus, the MDV lakes represent aquatic environments with strong selective pressures on microbial evolution including low temperatures, low annual levels of photosynthetically active radiation (PAR), and limited nutrient (N and P) availability (Laybourn-Parry 2009). Salinity, oxygen and light appear to play important roles in the biogeography of Antarctic lake microorganisms (Cavicchioli 2015b). Each lake supports simple food webs which harbor little to no metazoans, with the exception of low numbers of copepods and rotifers in a few lakes (Laybourn-Parry and Pearce 2007). Protists play dominating roles in carbon and nutrient cycling in the MDV microbial food web. The majority of dissolved organic carbon is autochthonous (i.e., derived from new photosynthetic activity within the water column). Protists represent the major producers of organic matter in the MDV lake food web (Morgan-Kiss et al 2006, Neale and Priscu 1995), while heterotrophic nanoflagellates and ciliates are the top predators of bacteria and smaller protists (Roberts and Laybourn-Parry 1999, Roberts et al 2004c). Despite the energetic cost of maintaining and regulating both photosynthetic and heterotrophic cellular apparatus, mixotrophy appears to be very prevalent in MDV aquatic food webs (Bell and Laybourn-Parry 2003, Roberts and Laybourn-Parry 1999). Mixotrophic metabolism is a survival strategy for MDV protists to exploit alternate sources of either nutrients (i.e., under oligotrophic conditions) or energy (i.e., in the absence of light) (Laybourn-Parry 2002).

As part of the NSF Long Term Ecological Research (LTER) Program, several lakes within the MDV have been the focus of more than two decades of intensive study. Lake Bonney is one of the most well-studied MDV lakes, and the lake physical and chemical characteristics have been thoroughly described by Spigel and Priscu (Spigel and Priscu 1998). Lake Bonney is divided into two basins (west and east lobes) which are separated by a narrow passage which allows mixing

88 of the upper photic zones for a few weeks in the summer. The shallow waters of both lobes are generally freshwater; however, salinity levels increase steeply at the permanent chemocline (18 – 20 m depth for the east lobe). Below the chemocline, the deep anoxic zone is hypersaline (maximum salinity 125-150 PSU). Photosynthetically active radiation (PAR) is relatively low under the ice (<50 µmol m-2 s-1) due to the attenuation of the ice cover and declines below detectable levels at depths >25 m (29). Recently, the diversity and spatial distribution of the planktonic microbial eukaryotic communities residing in Lake Bonney has been reported (Bielewicz et al 2011a, Dolhi et al 2015a, Kong et al 2012b, Vick-Majors et al 2014a). Protist populations residing in the MDV lakes are vertically stratified and exhibit lake-specific differences in their distribution patterns. The water column of Lake Bonney is dominated by large chlorophytes (Chlamydomonas) in the shallow, nutrient-poor depths, while the dominant photosynthetic protists are nanoplankton (haptophytes, stramenopiles) which exhibit peak abundances within the permanent chemocline where PAR is extremely limited (Bielewicz et al 2011a). While these studies have begun to describe the biogeographic distribution of the protist communities residing in these ice-covered polar lakes, a full understanding of the trophic abilities of MDV protists and potential interactions with other MDV microorganisms is currently lacking.

The vast diversity of microbial eukaryotes has been largely inaccessible by conventional cultivation methods. Over the past decade the technology of single-cell genomics has been exploited to recover genomic information from single uncultivated cells from their natural habitats, and has revealed cell-specific interactions such as symbioses, predation and parasitism (Campbell et al 2013, Marcy et al 2007, Yoon et al 2011). In this study, single cells of protists originating from an enrichment culture from the zone of maximum primary production in the permanent chemocline of Lake Bonney (east lobe) were isolated on the basis of fluorescent properties which related to their potential nutritional mode (photosynthetic, heterotrophic, mixotrophic). We chose to utilize an enrichment cultivation approach to allow us to work with live organisms and avoid the logistical issues associated with preserving and shipping natural protist communities from Antarctica to our US laboratory. Our hypothesis is MDV lake protists possess wide metabolic versatility which drives nutritional mode-dependent microbial interactions. Specifically, we addressed these open questions regarding the ecology of the MDV

89 lake microbial eukaryote communities: What is the range of trophic versatility in the MDV planktonic protist communities? Do MDV phototrophic and heterotrophic protists interact with specific microbial partners? In this study, whole genome amplification (WGA) was performed to produce single amplified genomes (SAGs) from a variety of the sorted eukaryote cells (with their microbial partners). Combining Illumina® sequencing with various microscopic methods, we were able to 1) identify nutritional mode and metabolic potential of several key MDV protists; 2) reveal a diversity of potential microbial interactions, including predation, parasitism and endosymbiosis; 3) provide new information to understand microbial metabolic strategies for survival under permanent light limiting and nutrient poor conditions.

4.2. Materials and methods

4.2.1. Site description, sample collection and enrichment cultures

Samples were collected from the water column of the east lobe of Lake Bonney, an ice-covered meromictic lake located in the Taylor Valley, McMrdo Dry Valleys. Samples were collected through the ice hole located in the middle of the lake (77o42.825S, 162 o 26.832E) during the austral summer on December 21, 2012. A sampling depth (13 m piezometric depth) was selected to reflect the depth of maximum phytoplankton biomass (Bielewicz et al 2011a), which was verified by in situ chlorophyll fluorescence with a submersible FluoroProbe (BBE Moldaenke GmbH) (Dolhi et al 2015a). Lake water samples were collected with a 5 L Niskin bottle (General Oceanics), transferred to 1-L amber bottles, and stored at 4 ˚C in the dark until processing.

An enrichment cultivation approach was employed to allow for the transport of live cultures with higher biomass to our US laboratory (Miami University, Ohio). To ensure maximum recovery of protist diversity, a series of enrichment cultures were started in the presence of a number of autotrophic media types (Bischoff and Bold 1963, Guillard and Ryther 1962, Stanier et al 1971) and growth media concentration (10 – 100%). Lake water was inoculated into 25-mL sterile culture flasks and incubated in a photoincubator at 5 ˚C /25 µmol photons m-2 s-1 in McMrdo Station, Antarctica, for 2 weeks prior to shipment. Cultures were regularly inspected under the microscope for growth and the presence of microbial eukaryotes. Cultures were then shipped to

90 our US laboratory at Miami University at 4 ˚C in the dark. Upon arrival, enrichment cultures were transferred to a photoincubator set at the same temperature/light regime until cell sorting. Last, while cryopreservation in the presence of glycerol buffer works well for preserving bacterial communities(Rinke et al 2014, Swan et al 2011), preliminary experiments showed that this preservation approach failed to preserve internal structures of eukaryotic cells and caused lysis of some wall-less protists.

4.2.2. Single cell sorting

Prior to cell sorting, all enrichment cultures were visually inspected by light microscopy for growth and morphological diversity of the protist communities. Based on these microscopic observations, one enrichment culture (grown in 10% F medium) which exhibited relatively high protist morphological diversity was chosen to be shipped to Oak Ridge National Laboratory for single-cell sorting. Protist food vacuoles were stained using a pH-sensitive probe, LysoTracker Green DND-26 (Invitrogen, Carlsbad, CA, USA) (Rose et al 2004). A 2mL aliquot of the culture was gently homogenized by several inversions and incubated in the dark on ice for 15 min in the presence of 75 nmol L-1 of LysoTracker. Microbial eukaryotic cells of various size, morphology and trophic ability (phototrophic, heterotrophic, mixotrophic) were identified and sorted with a Cytopeia INFLUX sorter (BD, Franklin Lakes, NJ, USA) in a Class 1000 clean room using a 488 nm argon laser for excitation. Green (528-538 nm) and red (670-730 nm) fluorescent emission indicated LysoTracker green fluorescence and chlorophyll autofluorescence, respectively. Targeted single eukaryotic cells were sorted into four 96-well plates which contained 3 µL UV- sterilized TE buffer in each well (Campbell et al 2013). Individual cells were sorted from discrete sectors using the three criteria. Plate 1 (P1) contained cells showing high LysoTracker fluorescence only (ie., heterotrophic/phagotrophic); Plate 2 (P2) contained cells exhibiting high red fluorescence (ie., photosynthetic) and the absence of green fluorescence; and Plates 3 and 4 (P3, P4) contained smaller cells that exhibited intermediate levels of red and green fluorescence (ie., mixotrophic; Fig. 4.1). Plates were stored at -80 ˚C until whole genome amplification.

4.2.3. DNA template preparation for PCR

For DNA preparation, single-sorted eukaryote cells were lysed using alkaline treatment by the addition of 3 µL of buffer that consisted of 0.13 M KOH, 3.3 mM EDTA pH 8.0 and 27.7 mM Dithiothreitol, heated to 95°C for 30 sec, and immediately placed on ice for 10 min, followed by addition of 3 µL neutralization buffer (0.13 M HCl, 0.42 M Tris pH 7.0, 0.18 M Tris 8.0). Genomic DNA from the cell lysates was amplified using multiple displacement amplification (MDA) by the addition of 11 µL of MDA master mix containing 90.9 µM random hexamers with two protective, phosporothioate bonds on the 3ʹ end (Integrated DNA Technologies, Coralville, IA, USA), 1.09 mM dNTPs (Roche Indianapolis, IN, USA), 1.8X phi29 DNA polymerase buffer (New England BioLabs, Ipswich, MA, USA), 4 mM DTT and ~100 U phi29 DNA polymerase enzyme (purified in house) (Blainey and Quake 2011). Whole genome amplification was performed for 10 hrs at 30°C followed by inactivation at 80°C for 20 min. MDA products were diluted 100-fold in sterile 1X TE buffer, then screened by PCR and sequencing.

4.2.4. Sanger sequencing

A fragment of the 18S rRNA gene was amplified from each of the SAGs using universal eukaryote primers [EK-82F (5’ -GAAACTGCGAATGGCTC) and EK-1520R (5’ - CYGCAGGTTCACCTAC)] (López-García et al 2001) to generate PCR products for sequencing. PCR was performed in triplicate using 25 cycles of 95°C for 1 min, 52°C for 1 min, and 72°C for 2 min. Sequencing reactions were performed using the BigDye Terminator v3.1 cycle sequencing kit (ABI, CA) with M13R primer and the fragments were sequenced on an Applied Biosystems 3730×l DNA Analyzer (ABI, CA) located in the Center for Bioinformatics and Functional Genomics (CBFG) at Miami University.

4.2.5. Illumina sequencing

Following successful identification of the individual eukaryote SAGs, 79 samples were selected for sequencing of the associated bacterial communities on an Illumina® MiSeq Platform in the CBFG. 16S rRNA genes from the SAGs were amplified using the primer set (F515/R806) which encoded sequence against the highly variable V4 region of 16s rRNA, barcodes and linkers. PCR

92 reactions and MiSeq sequencing reactions were strictly followed the protocol provided by the Earth Microbiome Project (http://www.earthmicrobiome.org) (Amaral-Zettler et al 2009, Caporaso et al 2011, Caporaso et al 2012, López-García et al 2001). Samples were sequenced according to the manufacturer’s recommendations using a 300-cycle MiSeq Reagent Kit v2 (Illumina®) in a 2 X 150 bp paired-end run in the presence of 25 % PhiX DNA.

4.2.6. Analysis of the sequences

For 18S rRNA gene sequences derived from Sanger sequencing, representative and closest sequences were selected from GenBank and aligned using MUSCLE. A maximum-likelihood tree was generated using MEGA6 software. Bootstrapping was used to estimate node support of 1,000 replicate trees.

16S rRNA gene sequences generated on the MiSeq instrument were analyzed with QIIME (v1.8.0). OTUs were identified at 97% cutoff using Greengenes database (v13.8) (Pruesse et al 2007). All OTUs with one sequence per sample were discarded. Samples were 120 times rarefied based on the number of sequences in the library with the lowest sequence number. Alpha diversity (number of OTUs, chao1 and Shannon index) was assessed (data not shown). Beta- diversity was integrated using weighted unifrac distance metric (Hamady and Knight 2009, Lozupone et al 2007). Principal coordinates analysis (PCoA) and ANOSIM similarity analysis was performed to identify co-existence of eukaryotic and prokaryotic organisms in the single-cell samples (Anderson 2001, Anderson et al 2011, Caporaso et al 2010, Kuczynski et al 2012).

4.2.7. Confocal Laser Scanning Microscopy

Isochrysis sp. MDV and Chlamydomonas sp. ICE-MDV cultures were treated with LysoTracker Green DND-26 probe using the same procedure previously described for FACS. A Zeiss LSM- 710 (Carl Zeiss Microscopy GmbH, Germany) confocal laser scanning microscope was used for images in Fig. 4.3. A 488 nm argon ion laser was used as elimination/excitation source. LysoTracker and chlorophyll fluorescence were detected within 520-540 nm and 650-750 nm

93 respectively. Differential interference contrast (DIC) images were also generated using transmission light mode. Images were pseudocolored in Zeiss ZEN 2011 (Carl Zeiss Microscopy GmbH, Germany) software.

4.2.8. Scanning Electron Microscopy

Samples from the original enrichment culture were fixed with 1% paraformaldehyde and 1.25% glutaraldehyde, and a secondary fixation was applied in 1% osmium tetroxide. Specimens were dehydrated through an ethanol series, critical-point dried with CO2, and sputter-coated with gold. Micrographs were generated with a Zeiss SUPRA-35 FEG SEM (Carl Zeiss Microscopy GmbH, Germany).

4.3. Results and discussion

4.3.1. Identities and trophic modes of sorted eukaryotes

To overcome the logistical constraints of transporting fresh samples from Antarctica to our US laboratory as well as low biomass issues associated with MDV planktonic communities, we used an enrichment cultivation-based approach to test FACS/ single cell sorting on McMrdo Dry Valley lake protist communities. A series of enrichment cultures were started in Antarctica using a range of growth media as described in the Experimental Procedures and incubated for 2 weeks prior to shipment to our US laboratory. Upon arrival in our US laboratory, we monitored algal diversity using a bbe FluoroProbe (see Table A2.2 in Appendix) and visually inspected each culture using bright-field and fluorescence microscopy to qualitatively assess protist diversity. Inspection of one of the cultures (Enrichment MDV87) revealed large (>15 µm) biflagellate chlorophytes, several nanophytoplankton identified as haptophytes and cryptophytes (~ 5 µm), as well as dinoflagellates cells exhibiting variable pigmentation (owing to the presence or absence of phytoplankton prey) and several non-pigmented heterotrophic nanoflagellates (2-5 µm). Based on our recent work on natural protist diversity in Lake Bonney (Bielewicz et al 2011a), we concluded that MDV87 harbored representatives of several key members of the Lake Bonney protist community, including chlorophytes, haptophytes, choanoflagellates, dinoflagellates, and nanoflagellates, all of which have been detected in environmental sequence libraries generated from the photic zone of the east lobe of Lake Bonney.

Based on LysoTracker and chlorophyll fluorescence signals combined with cell size information, we gated four eukaryote populations for single cell sorting: a group with high LysoTracker fluorescence (i.e., presence of protist food vacuole; Fig. 4.1, Plate 1 – green group), a group with high red fluorescence (i.e., presence of chlorophyll; Fig. 4.1, Plate 2 – red group) and two groups with smaller cells that exhibited intermediate levels of red and green fluorescence (Fig. 4.1, Plates 3 and 4 – blue groups). Targeted single cells were deposited into four 96-well plates according to distinct gates that separated various types of organisms based on the size and fluorescent characteristics mentioned above. Following cell lysis and whole genome amplification using phi29 DNA polymerase, 65% (250 out of 384 samples) of the SAGs were

95 successfully amplified using 18S rRNA primers based on electrophoretic separation of the PCR products on a 1% agarose gel. All SAGs which exhibited a band of the correct size (~1400 bp) were Sanger-sequenced. We recovered 79 high quality partial 18S rRNA sequences (~600bp) for phylogenetic analysis (GenBank accession numbers: KU196097- KU196166). Representative sequences were aligned with SILVA and GenBank databases using SINA and BLAST. Phylogenetic analysis revealed that the SAGs library harbored a total of 16 OTUs (based on a cut-off of 97% similarity) related to the Stramenopila supergroup and several other major phyla, including Chlorophyta, Choanoflagellida, Dinoflagellata, Cryptophyta and Haptophyta. The majority of SAG 18S rRNA sequences were identical to uncultivated eukaryote clones recovered in an earlier study on natural protist populations residing in Lake Bonney (Bielewicz et al 2011a).

All SAGs recovered from the photosynthetic group (i.e., high autofluorescence, low LysoTracker fluorescence; Fig. 4.1; SAG Plate 2; n=13) were related to a large biflagellate Chlorophyte, Chlamydomonas, which was closely related (99% identity) to an Antarctic marine species, Chlamydomonas sp. Antarctic 2E9, and two Antarctic ice algal strains, Chlamydomonas sp. ICE- W and ICE-L (Liu et al 2006). The Chlamydomonas SAGs were also closely related to sequences from clone libraries generated from the depth of maximum productivity (13 m depth) in the east lobe of Lake Bonney (Bielewicz et al 2011a) (Fig. 4.2). Recent studies on phylogenetic and functional genes indicated that this organism is the dominant chlorophyte in the east and west lobes of Lake Bonney (Bielewicz et al 2011a, Kong et al 2012b). Last, our FACS results which suggest that this organism is photoauotrophic were supported by earlier physiological studies on Chlamydomonas ICE sp. (Liu et al 2006) as well as a highly studied Lake Bonney chlorophyte, Chlamydomonas sp. UWO241 (Dolhi et al 2013, Morgan-Kiss et al 2006).

SAG sequences related to a nanoflagellate haptophyte (Isochrysis galbana; 97% identity) represented a significant proportion of samples (80 %) recovered from SAG Plate 4 (n=17; Fig. 4.2), which represented one of two plates with intermediate levels of green and red fluorescence (Fig. 4.1, SAG Plate 4). 18S rRNA sequences from these SAGs were very similar to sequences recovered from clone libraries generated from 13 m-deep waters from east and west lobe Lake

Bonney (Bielewicz et al 2011a), and recent reports suggest that haptophytes communities related to this organisms are key primary producers in this lake (Dolhi et al 2015a, Kong et al 2012b, Kong et al 2014b). Communities of Isochrysis dominate the chemoclines of the east and west lobes of Lake Bonney (Dolhi et al 2015a) Several recent studies using quantitative PCR to monitor abundance of photosynthetic functional genes (rbcL, psbA) indicates that in contrast with chlorophyte populations, abundance (DNA) and activity (RNA) of haptophyte populations is not correlated with light availability (Dolhi et al 2015a, Kong et al 2014b). These data suggest that the dry valley Isochrysis may rely on alternative metabolic strategies, such as mixotrophy.

Stramenopiles represent a poorly understood supergroup of microbial eukaryotes that exhibit broad trophic ability (e.g., phototrophy, mixotrophy, predation, parasitism). The marine stramenopiles (MASTs) lineage is a diverse group of microbial eukaryotes and represent a large proportion of heterotrophic nanoflagellates (HNFs) in marine environments (Massana et al 2004). HNFs are small microbial eukaryotes (2 to 20 µm) which graze on bacteria and picophytoplankton and are recognized as the dominant consumers of picoplanktonic biomass in marine environments (Fenchel 1982, Sanders et al 2000b). Our 18S rRNA gene sequence library contained several SAGs that were related to stramenopiles (Fig. 4.2). The most abundant stramenopile SAGs were closely related to Pteridomonas danica (SAG Plate 4; 99% identification, n=22) and Pirsonia verrucosa (SAG Plate 3; 94-100% identification, n=22). Sequences related to both stramenopiles were recovered in clone libraries from the west and east lobes of Lake Bonney (Bielewicz et al 2011a) (Fig. 4.2). Recent work has reported that stramenopiles make up a significant proportion of the protist population (up to 60%) in both lobes of Lake Bonney (Dolhi et al 2015a, Vick-Majors et al 2014a). Pirsonia is a phycovorous nanoflagellate which feeds on marine centric diatoms (Kuhn et al 2004). Pteridomonas is a ubiquitous marine heterotrophic nanoflagellate (Sherr and Sherr 2002a). In addition, two OTUs within the Stramenopile supergroup (SAG clones P3E2 and P4C7) exhibited very low similarity (<81%) to the closest related identified sequence in the database (Thraustochytrium sp.). Both OTUs were, however, closely related to the sequences generated in Lake Bonney clone libraries from a previous study (Bielewicz et al 2011a).

Several other protists groups were represented in the SAG libraries at low abundance. A single SAG was recovered for two choanoflagellates and one dinoflagellate. One SAG related to a psychrophilic marine cryptophyte Geminigera cryophila was also recovered. G. cryophila is abundant in the shallow layers of Lake Bonney (Bielewicz et al 2011a, Dolhi et al 2015a). Dinoflagellates have been detected in relatively high abundance in the west lobe of Lake Bonney, while sequences related to choanoflagellates were less abundant (Bielewicz et al 2011a, Vick-Majors et al 2014a). All low abundance SAGs were recovered from the two mixotrophic groups (SAG plates 3 and 4; Fig. 4.1) and were closely related to sequences recovered from clone libraries generated from east or west lobe Lake Bonney, with the exception of one choanoflagellate sequence (SAG P4B8; Fig. 4.2).

4.3.2. Isolation and description of two key photosynthetic protists

Recent work on the natural communities of Lake Bonney suggested that a chlorophyte related to Chlamydomonas sp. ICE-L and a haptophyte related to Isochrysis galbana occupy important yet distinct roles in the Antarctic lake food web (Bielewicz et al 2011a, Kong et al 2012b, Vick- Majors et al 2014a). Our FACS analyses indicated that the chlorophyte is a pure photoautotrophic species while the haptophyte may possess the ability to combine photosynthetic metabolism with heterotrophic activity, such as digestion of captured bacterial prey or particulate carbon. To confirm our hypotheses regarding the trophic capability of these key MDV protists, we purified isolates of both organisms from enrichment cultures. A culture of the Chlamydomonas sp. (hereafter named Chlamydomonas sp. ICE-MDV) was recovered by plating of an enrichment culture on Bold’s Basal Medium (Nichols and Bold 1965) with 1.5% agar. The Isochrysis strain (hereafter named Isochrysis sp. MDV) was isolated by dilution to extinction of an enrichment culture in 0.5X seawater supplemented with F/2 medium. The identity of both strains was confirmed by 18S rRNA gene sequencing (data not shown). Confocal microscopy showed that Chlamydomonas sp. ICE-MDV is a large (15 - 20 µm) biflagellate cell exhibiting high autofluorescence (Fig. 4.3a, d and e). As LysoTracker probe is fluorescent acidotrophic; therefore, we wondered if the acidic compartment (lumen) of chloroplasts (Alberts et al 2002) might cause false positive fluorescence signal under flow cytometry. We did not detect green fluorescence in the presence of LysoTracker in Chlamydomonas sp. ICE-MDV cells, despite the

98 presence of a large chloroplast (Fig. 4.3d and e). This also confirms that it is unlikely that any of the SAGs sorted in the mixed SAG plates (Plates 2 and 3; Fig. 4.1) were a product of spurious chloroplast staining. Isochrysis sp. MDV is a small (2 - 5 µm) brown-pigmented biflagellate alga. It possesses bi-lobed chloroplasts which exhibited high autofluorescence (Fig. 4.3i and j). In contrast with the chlorophyte strain, when cells of Isochrysis sp. MDV were treated with LysoTracker probe, green fluorescence was localized to a single vacuole (Fig. 4.3g). The presence of green fluorescence was associated only with LysoTracker-stained cells, as neither strain exhibited green fluorescence in the absence of LysoTracker (see Figs. A2.1 and A2.2 in Appendix). We have also confirmed in growth experiments that Isochrysis sp. MDV can grow either in the dark (heterotrophic) or the light (phototrophic), but grows optimally in the presence of a variety of organic carbon sources and low irradiance (mixotrophic). In contrast, Chlamydomonas ICE-MDV cannot grow in the dark in the presence of organic carbon, but exhibits ability to grow under a broader range of light intensities (see Table A2.1 in Appendix for description of basic growth physiology). Therefore, the chlorophyll and LysoTracker staining fluorescence information from confocal microscopy combined with growth physiology confirmed results from the FACS analyses that Chlamydomonas sp. ICE-MDV is a pure photosynthetic protist while Isochrysis sp. MDV is likely capable of mixotrophic metabolism. These results fit well with data from functional gene analyses that Lake Bonney chlorophyte populations that occupy the under-ice layers of the water column are strongly correlated with light availability on spatial and seasonal scales (Kong et al 2014b). By contrast, haptophyte populations dominating planktonic communities in the permanent chemocline are not influenced by light availability but can supplement light-dependent photosynthesis with ingestion of bacterial prey or particulate carbon.

4.3.3. Community composition of organisms co-sorted with Lake Bonney eukaryotes

To gain further insight into the trophic modes as well as potential microbial partners of the MDV microbial eukaryotes, we selected a number of eukaryote SAGs of varying trophic potential for 16S rRNA amplicon community sequencing. From the population of successfully amplified and sequenced SAGs, we sequenced a fragment of the 16S rRNA gene from 79 eukaryote SAGs using the Illumina® MiSeq platform (NCBI BioProject ID PRJNA304193). We recovered both

99 bacterial 16S and plastid SSU rRNA genes from these sequencing libraries. The plastid sequences were filtered from the 16S rRNA gene libraries of all SAG OTUs known to be photosynthetic, including Chlamydomonas sp. ICE-MDV, Isochrysis sp. MDV, and Geminigera. To assess both the diversity and co-occurrence patterns of organisms that co-sorted with the individual eukaryote SAGs, we generated a heat map of 16S rRNA genes associated with each individual SAG sample (see Fig. A4..3 in Appendix). The distribution of recovered 16S rRNA gene sequences was highly variable and dependent upon both the trophic mode and identity of the microbial eukaryote partner. Bacterial sequences associated with the photosynthetic Chlamydomonas sp. ICE-MDV and the mixotrophic Isochrysis sp. MDV were generally diverse, with a predominance of the phyla Actinobacteria and Bacteriodetes. In contrast, 16S rRNA gene sequences recovered from the two stramenopile SAGs related to Pteridomonas and Pirsonia, which dominated the heterotrophic (Plate 1) and one of the mixotrophic (Plate 4) populations, respectively, were enriched with plastid sequences (see Fig. A4..3 in Appendix).

Principal Coordinates Analysis (PCoA) based on UniFrac distance metrics of 16S rRNA OTUs from the four most abundant SAGs (Isochrysis sp. MDV, Chlamydomonas sp. ICE-MDV, Pteridomonas and Pirsonia) exhibited clustering by protist trophic mode (Fig. 4.4). Similarity analysis using ANOSIM method indicated that bacterial OTUs associated with individual eukaryote SAGs were significantly different to each other (p<0.001, R2=0.40). Bacterial OTUs associated with Isochrysis sp. MDV and Chlamydomonas sp. ICE-MDV were clustered separately from each other, but were more closely related compared with either of the straemophile SAGs. 16S rRNA gene libraries generated from SAGs of the heterotrophic nanoflagellates (Pteridomonas and Pirsonia) were more related to each other than to either of the photosynthetic protists or the environmental samples (Fig. 4.4).

4.3.4. Potential interactions between Dry Valley protists and bacteria

Complex ecological interactions exist between protists and other microorganisms, including predation of bacteria by heterotrophic protists and syntrophic interactions between photosynthetic algae and heterotrophic bacteria (Amin et al 2012, Buchan et al 2014).

100

Traditionally, heterotrophic protists were identified via culture-dependent and microscopic methods (Auinger et al 2008, Jezbera et al 2005, Pernthaler 2005a, Sherr et al 1993). Recent studies have shown that a single-cell genomic approach can provide direct evidence of specific predator-prey interactions (Heywood et al 2011, Martinez-Garcia et al 2012). We recovered 22 samples that had high LysoTracker probe-associated fluorescence and were closely related to the stramenopile Pteridomonas danica (Figs. 2.1 and 2.2). Pteridomonas is a ubiquitous heterotrophic nanoflagellate in marine systems, and is an important linkage between bacteria and higher trophic levels in the food web (Caron et al 1999, Pelegr et al 1999, Zubkov and Sleigh 2005). Interestingly, we observed that more than 90% of 16S rRNA gene sequences from Pteridomonas SAGs were closely related to a stramenopile chloroplast sequence that was neither reported in environmental sequence libraries (Bielewicz et al 2011a, Vick-Majors et al 2014a) nor detected in other SAG 16S rRNA libraries in this study. A previous study reported that Pteridomonas danica harbors non-pigmented non-photosynthetic plastids or vestigial chloroplasts (Sekiguchi et al 2002); therefore, it is likely that the chloroplast sequences we recovered in 16S rRNA gene sequence libraries generated from Pteridomonas SAGs represent plastid sequences from vestigial chloroplasts. This interpretation was supported by a lack of pigmentation under light microscopy in Pteridomonas cells (data not shown). We removed the stramenopile plastid sequences from the 16S rRNA gene libraries generated from Pteridomonas SAGs and investigated the remaining 16S rRNA gene sequences in the Pteridomonas libraries (Fig. 4.5a). The most abundant sequences were related to a haptophyte plastid rRNA gene (32 ± 23% in total OTUs) and a firmicute, Streptococcus (13±8% in total OTUs) (Fig. 4.6a). HNFs represent a diverse ecological group cable of various trophic preferences (e.g., bacteriovorous, phycovorous or omnivorous). P. danica feeds on both bacteria and picophytoplankton and is therefore an omnivore (Zwirglmaier et al 2009); however, we did not find any reports of this genus predating on haptophytes. Picocyanobacteria are largely absent from the water column of the MDV lakes (Dolhi et al 2015a, Kong et al 2012b); therefore, we suggest that in addition to heterotrophic bacteria, the haptophyte population which dominates the photic zone of Lake Bonney could be the natural prey for the MDV Pteridomonas.

A number of feeding strategies have been reported for heterotrophic flagellates such as filtration of food particles with an array of radial pseudopodia or tentacles or direct interception of prey by

101 flagella. In addition, flagellates are often attached to particles such as marine snow when feeding (Christensen-Dalsgaard and Fenchel 2003, Kiørboe et al 2004). Examination of the Lake Bonney Pteridomonas by scanning electron microscopy revealed the presence of a single long flagellum surrounded by a radial array of tentacles (Fig. 4.5 b-d) which are characteristic morphological features of Pteridomonas species (Caron et al 1999). In addition, we observed particles attached to the tentacles and cells were often attached to a surface of particle by a stalk (Fig. 4.5b). To our knowledge, our images are some of the first published SEMs of this genus. Thus, the Lake Bonney Pteridomonas appears to rely on a filter feeding mechanism and may be particle- attached within the water column. This would be an advantageous strategy within the permanent chemocline where potential prey would be abundant and particulate organic carbon from the upper photic zone may accumulate. In support of this suggestion, Roberts et al. (Roberts et al 2004c) reported that HNF dominated heterotrophic protozoa in Lake Bonney and peaked within the permanent chemocline of the west lobe of Lake Bonney.

A second highly abundant SAG (n=22) within the Stramenopile supergroup exhibited a relatively high proportion of chloroplast sequences within the 16S rRNA gene sequence libraries (Figs. 2.6 and A2.3 in Appendix). These SAGs were recovered from Plate 4, which exhibited relatively high green fluorescence and intermediate levels of autofluorescence, and were closely related to a parasitoid nanoflagellate, Pirsonia sp. (Figs. 2.1 and 2.2). Parasitoid protists comprise a diverse taxonomic group which are thought play major roles in controlling their prey abundance; however, their role in microbial food webs is poorly understood (Skovgaard 2014). Plastid sequences recovered from the Pirsonia SAGs were all related to Chlamydomonas chloroplast sequences (Fig. 4.6a); however, unlike Pteridomonas, Pirsonia is not known to harbor a vestigial plastid. Therefore, the presence of plastid sequences within the 16S rRNA gene sequencing libraries suggested that Pirsonia potentially interacts with Chlamydomonas in a predator-prey relationship. However, microscopic examination indicated that the average size of Pirsonia was ~5 µm which was noticeable smaller than Chlamydomonas (~15 µm; Fig. 4.6b, c). Past studies have observed Pirsonia species are host specific for marine centric diatoms. The feeding mechanism involved attachment to a host diatom cell, formation of a trophosome inside the host cell and transfer of digested material into the parasite cell (Kuhn et al 2004, Schweikert and Schnepf 1997). We observed an extracellular organelle on the Dry Valley Pirsonia which likely

102 functions as part of this feeding mechanism (Fig. 4.6b, arrow). While Diatoms are abundant in the ephemeral streams and microbial mats around the Dry Valleys (Doran et al 1994, Spaulding et al 1997, Sumner et al 2015), they are rare in MDV lake water columns (Bielewicz et al 2011a, Dolhi et al 2015a, Vick-Majors et al 2014a). To our knowledge there are no reports of Pirsonia interacting with chlorophytes; however, we observed numerous Pirsonia cells attached to Chlamydomonas cells (Fig. 4.6 d and e), which is a critical step for infection of host cells.

Phytoplankton-bacteria interactions are well known in aquatic environments and are largely driven by the dependence of heterotrophic bacteria on the production of alga-derived organic substrates in the form of extracellular phytoplankton products or decaying algal biomass. There is clear evidence that phytoplankton community distribution and dynamics influences specific bacterial assemblages (Murray et al 2007, Paver et al 2013, Simek et al 2008). Algal-bacteria interactions involve either free-living, non-particle attached communities of heterotrophic bacteria or cell-to-cell contact between specific bacteria and algal hosts: bacteria groups associated with either interaction appear to be significantly different from each other (Grossart et al 2005, Rösel and Grossart 2012). Bacterioplankton can exhibit differential abilities for incorporation of specific algal-derived substrates (Salcher et al 2013). The Dry Valley lakes are essential closed systems for most of the year, and thus heterotrophic bacteria are heavily reliant on newly fixed carbon from phytoplankton production. Recent studies based on environmental sequencing of phytoplankton and bacterial communities have shown that spatial distribution of planktonic communities strongly varies within and between lakes (Bielewicz et al 2011a, Dolhi et al 2015a, Vick-Majors et al 2014a); however, interactions between MDV phytoplankton and heterotrophic bacteria have not been resolved. In this present study, SAGs of either the chlorophyte Chlamydomonas sp. ICE-MDV or the haptophyte Isochrysis sp. MDV were associated with a number of bacterial OTUs (Figs. 2.5 and A2.3 in Appendix). In general, the evidence for strong algal-bacteria interactions was lacking. However, we did note two potential bacterial associations with the Lake Bonney phototrophic SAGs. First, a significant proportion of Chlamydomonas ICE-MDV single cells were associated with OTUs related to the Methylobacteriaceae. Members of this group are obligate methylotrophs that can oxidize a number of methylated compounds including methanol and methylamines. These organisms are relatively abundance in marine environments, particularly during phytoplankton blooms (Morris 103 et al 2006, Sekar et al 2004). Many marine microalgae produce a number of single-carbon compounds as osmolytes (Sieburth and Keller 1989); however, less is known about algal- methylotroph interactions in freshwater environments (Salcher et al 2015). While the presence of methylotrophy in the MDV lakes has not been reported, it seems probable that algal production of osmolytes would be a likely adaptation to survive harsh conditions such as low temperatures combined with high salinity.

In contrast with the Chlamydomonas sp. ICE-MDV SAGs, Isochrysis sp. MDV SAGs were not generally associated with methylotrophic bacterial sequences. Instead, the majority of haptophyte single cells were associated with a number of bacterial sequences from Flavobacteriaceae (see Fig. A4..3 in Appendix). In the marine environment, Flavobacteria represent a dominant Bacteriodetes which breakdown complex organic matter and biopolymers by either direct attachment to algal cells or algal-derived detritus (Kirchman 2002, Teeling et al 2012). Haptophyte blooms in the Southern ocean are associated with peaks in abundance in Flavobacteria (Williams et al 2013), and sequences related to Flavobacteria are highly abundant in the bacterioplankton communities residing in the chemocline of Lake Bonney (Vick-Majors et al 2014a). Our work supports a putative interaction between Lake Bonney haptophytes and Flavobacteria.

The relatively high number of bacterial OTUs associated with the phototrophic SAGs could also represent evidence of low algal-bacterial coupling in the MDV lakes. Phytoplankton exudate quantity and composition are influenced by environmental factors including nutrient limitation, light availability and temperature (Eckert et al 2012, Teeling et al 2012). The conditions of the MDV lakes (e.g., limited light and nutrients, low temperatures) may have selected for algal species which excrete a low fraction of algal-derived compounds, and thus bacterioplankton rely more heavily upon exudate release during algal lysis or death. In support of this latter hypothesis, we have observed heavy recruitment and colonization of bacteria on dead and dying algal cells in our enrichment cultures; an interaction which would have been excluded in the FACS sorting. There is also evidence that natural bacterioplankton communities obtain a significant fraction of organic carbon from ancient, relict pools of organic matter (Priscu et al 1999a).

104

4.4. Conclusions

Microbial eukaryotes represent the majority of diversity across the eukaryotic tree of life and are capable of complex metabolic modes and interactions. While important in marine ecosystems, their influence reaches profound levels in aquatic systems where the microbial loop dominates. In our current study, we discovered that the MDV protists possess a diverse array of metabolic capabilities (pure photosynthetic, to mixotrophy, heterotrophy and parasitism) which likely contributes to the strong vertical layering of key protists communities in the water column of Lake Bonney. Metabolic versatility of MDV protists underpins specific microbe-microbe interactions in some cases (e.g., Pteridomonas-chlorophyte), while for other protists (e.g., haptophytes), interactions with heterotrophic bacteria do not appear to be an important survival strategy.

4.5. Acknowledgements:

The authors thank the McMrdo LTER, Antarctic Support Contract and PHI helicopters for logistical assistance in the field. We thank Andor J. Kiss and the Center for Bioinformatics and Functional Genomics at Miami University for assistance with Illumina sequencing. We thank Richard E Edelmann, Matthew Duley and Center for Advanced Microscopy and Imaging at Miami University for assistance with microscopy and image analysis. This work was supported by NSF Office of Polar Programs Grant OPP-1056396. MP has been supported by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory (ORNL). ORNL is managed by UT-Battelle, LLC, for the U.S. Department of Energy under contract DE-AC05-00OR22725. We thank Steven Allman for technical assistance with the flow cytometry cell sorting.

105

Figure 4.1. Sample cytogram showing flow cytometric sort regions for microbial eukaryotes from an enrichment culture (MDV87) generated from the Antarctic Dry Valley Lake Bonney. Prior to sorting the sample was stained with LysoTracker and samples were gated on green fluorescence (LysoTracker-stained food vacuoles) vs. red fluorescence (chlorophyll autofluorescence). Ovals indicate groups that were selected for single cell sorting. Colors represent high chlorophyll autofluorescence (red), high LysoTracker fluorescence (green) or intermediate of either (blue). P1 – P4, sorted 96-well plates 1 to 4.

106

Figure 4.2 Maximum-likelihood tree (1000 bootstrap) showing identity of Lake Bonney microbial eukaryote single amplified genomes (EUK-SAGS) recovered from enrichment culture MDV87. SILVA and GenBank data bases were used to assign phylogenetic position of EUK- SAGS. Colors of the SAG sequences are correlated with Figure 1 sorting parameters. * indicates sequences from the natural eukaryote populations in Lake Bonney generated in a previous study (Bielewicz et al 2011a). Genbank accession numbers for EUK-SAG 18S rRNA gene sequences are: KU196097- KU196166.

107

Figure 4.3. Confocal microscopic images of Chlamydomonas sp. ICE-MDV (a - e) and Isochrysis sp. MDV (f - j) isolates stained with LysoTracker Green and DAPI. a and f, DIC images; b and g, LysoTracker fluorescence; c and h, DAPI fluorescence; d and i, chlorophyll autofluorescence; e and j, overlays of a – d and f – i respectively. Bars indicate 5 µm.

108

Figure 4.4. Principle Coordinates Analysis (PCoA) of weighted UniFrac distances of 16S rRNA genes associated with EUK-SAGs of Isochrysis sp. MDV (crosses), Chlamydomonas sp. ICE- MDV (diamonds), Pteridomonas (squares) and Pirsonia (circles). Each data point represents one SAG sample. Ovals represent the 50% similarities within each sample group.

109

Figure 4.5. SEM micrographs and the diversity of microbial partners associated with the heterotrophic nanoflagellate Pteridomonas (n=22). a. Diversity of 16S rRNA gene OTUs recovered from Pteridomonas EUK-SAGs. b-d, SEM images of Pteridomonas sp. cells from the enrichment culture. Bars: 1µm.

110

Figure 4.6. Evidence of parasite-host interactions between Lake Bonney microbial eukaryotes. a, Diversity of 16S rRNA gene OTUs recovered from Pirsonia EUK-SAGs (n=22). b, c, SEM images showing the size comparison of Pirsonia vs. Chlamydomonas sp. ICE-MDV, respectively. The arrow indicated the extrusive organelle of Pirsonia. d, e, SEM images showed Pirsonia cells attached to Chlamydomonas sp. ICE-MDV. Bars: 5 µm.

111

4.6. References

Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P (2002). Chloroplasts and Photosynthesis. Molecular Biology of the Cell, 4 edn. Garland Science: New York.

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76: 667-684.

Anderson MJ (2001). A new method for non-parametric multivariate analysis of variance. Austral Ecol 26: 32-46.

Anderson MJ, Crist TO, Chase JM, Vellend M, Inouye BD, Freestone AL et al (2011). Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecol Lett 14: 19-28.

Auinger BM, Pfandl K, Boenigk J (2008). Improved methodology for identification of protists and microalgae from plankton samples preserved in Lugol's iodine solution: combining microscopic analysis with single-cell PCR. Appl Environ Microbiol 74: 2505-2510.

Bell EM, Laybourn-Parry J (2003). Mixotrophy in the Antarctic phytoflagellate Pyramimonas gelidicola. J Phycol 39: 644-649.

Bischoff HW, Bold HC (1963). Some soil algae from Enchanted Rock and related algal species. Phycological studies IV. University of Texas Press: Austin.

Blainey PC, Quake SR (2011). Digital MDA for enumeration of total nucleic acid contamination. Nucleic Acids Res 39: e19.

Buchan A, LeCleir GR, Gulvik CA, Gonzalez JM (2014). Master recyclers: features and functions of bacteria associated with phytoplankton blooms. Nat Rev Microbiol 12: 686-698.

112

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK et al (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods 7: 335- 336.

Caron D, Gast R, Lim E, Dennett M (1999). Protistan community structure: molecular approaches for answering ecological questions. Molecular Ecology of Aquatic Communities. Springer. pp 215-227.

Caron DA, Countway PD, Brown MV (2004). The growing contributions of molecular biology and immunology to protistan ecology: molecular signatures as ecological tools. J Eukaryot Microbiol 51: 38-48.

Caron DA, Worden AZ, Countway PD, Demir E, Heidelberg KB (2009). Protists are microbes too: a perspective. The ISME journal 3: 4-12.

Cavicchioli R (2015). Microbial ecology of Antarctic aquatic systems. Nat Rev Microbiol 13: 691-706.

Chela-Flores J (2011). On the possibility of biological evolution on the moons of Jupiter. The Science of Astrobiology. Springer. pp 151-170.

Christensen-Dalsgaard KK, Fenchel T (2003). Increased filtration efficiency of attached compared to free-swimming flagellates. Aquat Microb Ecol 33: 77-86.

113

Dolhi JM, Maxwell DP, Morgan-Kiss RM (2013). Review: The Antarctic Chlamydomonas raudensis: An emerging model for cold adaptation of photosynthesis. Extremophiles 17: 711- 722.

Dolhi JM, Teufel AG, Kong W, Morgan-Kiss RM (2015). Diversity and spatial distribution of autotrophic communities within and between ice-covered Antarctic lakes (McMrdo Dry Valleys). Limnol Oceanogr 60: 977-991.

Doran PT, Wharton Jr RA, Lyons WB (1994). Paleolimnology of the McMrdo dry valleys, Antarctica. J PALEOLIMNOL 10: 85-114.

Eckert EM, Salcher MM, Posch T, Eugster B, Pernthaler J (2012). Rapid successions affect microbial N-acetyl-glucosamine uptake patterns during a lacustrine spring phytoplankton bloom. Environ Microbiol 14: 794-806.

Epstein S, Lopez-Garcia P (2008). "Missing protists: a molecular perspective. Biodiv Conserv 17: 213-218.

Faust K, Raes J (2012). Microbial interactions: from networks to models. Nature Reviews in Microbiology 10: 538-550.

Fenchel T (1982). Ecology of heterotrophic microflagellates. IV. Quantitative occurrence and importance as bacterial consumers. Mar Ecol Prog Ser 9: 35-42.

Green W, Lyons W (2009). The saline lakes of the McMrdo Dry Valleys, Antarctica. Aquat Geochem 15: 321-348.

Grossart HP, Levold F, Allgaier M, Simon M, Brinkhoff T (2005). Marine diatom species harbour distinct bacterial communities. Environ Microbiol 7: 860-873.

Guillard RR, Ryther JH (1962). Studies of marine planktonic diatoms. I. Cyclotella nana Hustedt and Detonula confervacea Cleve. Can J Microbiol 8: 229-239.

Hamady M, Knight R (2009). Microbial community profiling for human microbiome projects: Tools, techniques, and challenges. Genome Res 19: 1141-1152.

Heywood JL, Sieracki ME, Bellows W, Poulton NJ, Stepanauskas R (2011). Capturing diversity of marine heterotrophic protists: one cell at a time. The ISME journal 5: 674-684.

114

Jezbera J, Hornak K, Simek K (2005). Food selection by bacterivorous protists: insight from the analysis of the food vacuole content by means of fluorescence in situ hybridization. FEMS microbiology ecology 52: 351-363.

Kiørboe T, Grossart H-P, Ploug H, Tang K, Auer B (2004). Particle-associated flagellates: swimming patterns, colonization rates, and grazing on attached bacteria. Aquat Microb Ecol 35: 141-152.

Kirchman DL (2002). The ecology of Cytophaga–Flavobacteria in aquatic environments. FEMS microbiology ecology 39: 91-100.

Kong W, Li W, Romancova I, Prasil O, Morgan-Kiss RM (2014). An integrated study of photochemical function and expression of a key photochemical gene (psbA) in photosynthetic communities of Lake Bonney (McMrdo Dry Valleys, Antarctica). FEMS microbiology ecology 89: 293-302.

Kuczynski J, Stombaugh J, Walters WA, González A, Caporaso JG, Knight R (2012). Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc Microbiol: 1E. 5.1-1E. 5.20.

Kuhn S, Medlin L, Eller G (2004). Phylogenetic position of the parasitoid nanoflagellate Pirsonia inferred from nuclear-encoded small subunit ribosomal DNA and a description of Pseudopirsonia n. gen. and Pseudopirsonia mucosa (Drebes) comb. nov. Protist 155: 143-156.

Lara E, Mitchell EA, Moreira D, Lopez Garcia P (2011). Highly diverse and seasonally dynamic protist community in a pristine peat bog. Protist 162: 14-32.

Laybourn-Parry J (2002). Survival mechanisms in Antarctic lakes. Philos Trans R Soc Lond B Biol Sci 357: 863-869.

Laybourn-Parry J, Pearce DA (2007). The biodiversity and ecology of Antarctic lakes: models for evolution. Philos Trans R Soc Lond B Biol Sci 362: 2273-2289.

Laybourn-Parry J (2009). No place too cold. Science 324: 1521-1522.

115

Liu C, Huang X, Wang X, Zhang X, Li G (2006). Phylogenetic studies on two strains of Antarctic ice algae based on morphological and molecular characteristics. Phycologia 45: 190- 198.

López-García P, Rodríguez-Valera F, Pedrós-Alió C, Moreira D (2001). Unexpected diversity of small eukaryotes in deep-sea Antarctic plankton. Nature 409: 603-607.

Lozupone CA, Hamady M, Kelley ST, Knight R (2007). Quantitative and qualitative diversity measures lead to different insights into factors that structure microbial communities. Appl Environ Microbiol 73: 1576-1585.

Massana R, Castresana J, Balagué V, Guillou L, Romari K, Groisillier A et al (2004). Phylogenetic and ecological analysis of novel marine stramenopiles. Appl Environ Microbiol 70: 3528-3534.

Montagnes D, Roberts E, Lukeš J, Lowe C (2012). The rise of model protozoa. Tren Microbiol 20: 184-191.

Morris RM, Longnecker K, Giovannoni SJ (2006). Pirellula and OM43 are among the dominant lineages identified in an Oregon coast diatom bloom. Environ Microbiol 8: 1361-1370.

116

Murray A, Arnosti C, De La Rocha C, Grosart H-P, Passow U (2007). Microbial dynamics in autotrophic and heterotrophic seawater mesocosms. II. Bacterioplankton community structure and hydrolytic enzyme activities. Aquat Microb Ecol 49: 123-141.

Neale PJ, Priscu JC (1995). The photosynthetic apparatus of phytoplankton from a perennially ice-covered Antarctic lake: acclimation to an extreme shade environment. Plant Cell Physiol 36: 253-263.

Nichols HW, Bold HC (1965). Trichosarcina polymorpha gen. et sp. nov. J Phycol 1: 34-38.

Nolte V, Pandey RV, Jost S, Medinger R, Ottenwalder B, Boenigk J et al (2010). Contrasting seasonal niche separation between rare and abundant taxa conceals the extent of protist diversity. Mol Ecol 19: 2908-2915.

Paver SF, Hayek KR, Gano KA, Fagen JR, Brown CT, DavisRichardson AG et al (2013). Interactions between specific phytoplankton and bacteria affect lake bacterial community succession. Environ Microbiol 15: 2489-2504.

Pelegr S, Christaki U, Dolan J, Rassoulzadegan F (1999). Particulate and Dissolved Organic Carbon Production by the Heterotrophic Nanoflagellate Pteridomonas danica Patterson and Fenchel. Microb Ecol 37: 276-284.

Pernthaler J (2005). Predation on prokaryotes in the water column and its ecological implications. Nat Rev Microbiol 3: 537-546.

Priscu JC, Wolf CF, Takacs CD, Fritsen CH, Laybourn-Parry J, Roberts EC et al (1999). Carbon transformations in a perennially ice-covered Antarctic lake. Bioscience 49: 997-1008.

Reynolds RT, Squyres SW, Colburn DS, McKay CP (1983). On the habitability of Europa. Icarus 56: 246-254.

Rinke C, Lee J, Nath N, Goudeau D, Thompson B, Poulton N et al (2014). Obtaining genomes from uncultivated environmental microorganisms using FACS-based single-cell genomics. Nat Protoc 9: 1038-1048.

117

Roberts EC, Laybourn-Parry J (1999). Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwat Biol 41: 737-746.

Roberts EC, Priscu JC, Wolf C, Lyons WB, Laybourn-Parry J (2004). The distribution of microplankton in the McMrdo Dry Valley Lakes, Antarctica: response to ecosystem legacy or present-day climatic controls? Polar Biol 27: 238-249.

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

Rösel S, Grossart H-P (2012). Contrasting dynamics in activity and community composition of free-living and particle-associated bacteria in spring. Aquat Microb Ecol 66: 169-181.

Salcher MM, Posch T, Pernthaler J (2013). In situ substrate preferences of abundant bacterioplankton populations in a prealpine freshwater lake. The ISME journal 7: 896-907.

Salcher MM, Neuenschwander SM, Posch T, Pernthaler J (2015). The ecology of pelagic freshwater methylotrophs assessed by a high-resolution monitoring and isolation campaign. The ISME journal 9: 2442-2453.

Schweikert M, Schnepf E (1997). Light and electron microscopical observations on Pirsonia punctigerae spec, nov., a nanoflagellate feeding on the marine centric diatom Thalassiosira punctigera. Eur J Protistol 33: 168-177.

Sekar R, Fuchs BM, Amann R, Pernthaler J (2004). Flow sorting of marine bacterioplankton after fluorescence in situ hybridization. Appl Environ Microbiol 70: 6210-6219.

Sekiguchi H, Moriya M, Nakayama T, Inouye I (2002). Vestigial chloroplasts in heterotrophic stramenopiles Pteridomonas danica and Ciliophrys infusionum (Dictyochophyceae). Protist 153: 157-167.

Sherr EB, Caron DA, Sherr BF (1993). Staining of heterotrophic protists for visualization via epifluorescence microscopy. Handbook of methods in aquatic microbial ecology: 213-228.

118

Sherr EB, Sherr BF (2002). Significance of predation by protists in aquatic microbial food webs. A Van Leeuw J Microb 81: 293-308.

Sieburth JM, Keller MD (1989). Methylaminotrophic bacteria in xenic nanoalgal cultures: incidence, significance, and role of methylated algal osmoprotectants. Biol Oceanogr 6: 383-395.

Simek K, Hornak K, Jezbera J, NEDOMA J, Znachor P, Hejzlar J et al (2008). Spatio-temporal patterns of bacterioplankton production and community composition related to phytoplankton composition and protistan bacterivory in a dam reservoir. Aquat Microb Ecol 51: 249-262.

Simon M, Jardillier L, Deschamps P, Moreira D, Restoux G, Bertolino P et al (2015a). Complex communities of small protists and unexpected occurrence of typical marine lineages in shallow freshwater systems. Environ Microbiol 17: 3610-3627.

Simon M, Lopez-Garcia P, Deschamps P, Moreira D, Restoux G, Bertolino P et al (2015b). Marked seasonality and high spatial variability of protist communities in shallow freshwater systems. The ISME journal 9: 1941-1953.

Skovgaard A (2014). Dirty tricks in the plankton: Diversity and Role of Marine Parasitic Protists: diversity and role of marine parasitic protists. Acta Protozool 53: 51-62.

Spaulding S, McKnight D, Stoermer E, Doran P (1997). Diatoms in sediments of perennially ice- covered Lake Hoare, and implications for interpreting lake history in the McMrdo Dry Valleys of Antarctica. J PALEOLIMNOL 17: 403-420.

Spigel RH, Priscu JC (1998). Physical limnology of the McMrdo Dry Valleys lakes. Wiley Online Library.

Stanier R, Kunisawa R, Mandel M, Cohen-Bazire G (1971). Purification and properties of unicellular blue-green algae (order Chroococcales). Bacteriol Rev 35: 171.

Sumner DY, Hawes I, Mackey TJ, Jungblut AD, Doran PT (2015). Antarctic microbial mats: A modern analog for Archean lacustrine oxygen oases. Geology 43: 887-890.

Swan BK, Martinez-Garcia M, Preston CM, Sczyrba A, Woyke T, Lamy D et al (2011). Potential for Chemolithoautotrophy Among Ubiquitous Bacteria Lineages in the Dark Ocean. Science 333: 1296-1300.

119

Teeling H, Fuchs BM, Becher D, Klockow C, Gardebrecht A, Bennke CM et al (2012). Substrate-controlled succession of marine bacterioplankton populations induced by a phytoplankton bloom. Science 336: 608-611.

Williams TJ, Wilkins D, Long E, Evans F, DeMaere MZ, Raftery MJ et al (2013). The role of planktonic Flavobacteria in processing algal organic matter in coastal East Antarctica revealed using metagenomics and metaproteomics. Environ Microbiol 15: 1302-1317.

Yoon HS, Price DC, Stepanauskas R, Rajah VD, Sieracki ME, Wilson WH et al (2011). Single- cell genomics reveals organismal interactions in uncultivated marine protists. Science 332: 714- 717.

Zinger L, Gobet A, Pommier T (2012). Two decades of describing the unseen majority of aquatic microbial diversity. Mol Ecol 21: 1878-1896.

Zubkov MV, Sleigh MA (2005). Assimilation efficiency of Vibrio bacterial protein biomass by the flagellate Pteridomonas : assessment using flow cytometric sorting. FEMS microbiology ecology 54: 281-286.

Zwirglmaier K, Spence E, Zubkov MV, Scanlan DJ, Mann NH (2009). Differential grazing of two heterotrophic nanoflagellates on marine Synechococcus strains. Environ Microbiol 11: 1767-1776.

120

4.7. Appendix

Table A4.1. Basic growth physiology of Chlamydomonas sp. ICE-MDV and Isochrysis sp. MDV isolates.

Growth Growth Organism Growth Media Temperature Irradiance Range Organic Carbon Range (˚C) (μmol m-2 s-1)

Chlamydomonas Bolds 2 - 15 5 - 500 None sp. ICE-MDV

Isochrysis sp. cereal grass, rice, L1/2, f/2 2 - 10 0 - 50 MDV yeast extract

121

Table A2.2. Diversity of major algal classes in Lake Bonney enrichment cultures. Algal diversity was determined by spectral chlorophyll a fluorescence (bbe FluoroProbe). Enrichment used in this study is highlighted.

Enrichment Depth Algal Class (µg Chl a L-1) Lake Medium Name (m) Green Cyano Hapto Crypto

73.1 ELB 6 F/2 58.90 0.00 5.78 1.38 74.1 ELB 6 DYV 21.33 0.25 0.00 3.09 75.1 ELB 6 BBM 26.06 0.35 2.00 0.00 76.1 ELB 6 F/2 26.00 0.16 1.32 1.17 77.1 ELB 6 10% DYV 46.04 0.00 0.00 1.13 78.1 ELB 6 BBM 17.98 0.26 0.00 1.08 79.1 ELB 6 F/2 22.01 0.34 1.32 0.00 80.1 ELB 6 DYV 18.30 0.36 0.00 4.73 81.1 ELB 6 10% BBM 48.34 0.00 5.44 1.80 82.1 ELB 6 10% F/2 29.81 0.00 2.07 1.40 83.1 ELB 6 10% DYV 34.04 0.00 0.00 1.17 84.1 ELB 13 BBM 51.53 0.00 19.88 1.12 85.1 ELB 13 F/2 47.35 0.00 38.76 0.89 86.1 ELB 13 DYV 125.29 0.00 43.31 0.00 87.1 ELB 13 10% F/2 60.43 0.00 46.42 6.41 88.1 ELB 13 10% BBM 25.53 0.00 12.49 0.21 89.1 ELB 13 10% DYV 20.67 0.00 18.86 0.12 90.1 ELB 13 BBM 85.55 0.00 22.05 0.00 91.1 ELB 13 F/2 68.49 0.00 137.60 1.04 92.1 ELB 13 DYV 42.25 0.00 0.59 0.44 93.1 ELB 13 10% BBM 37.90 0.00 20.13 0.85 94.1 ELB 13 10% F/2 16.77 0.00 12.63 0.00 95.1 ELB 13 10% DYV 6.68 0.00 6.04 0.16

122

Figure A4.1. Confocal microscopic images of Isochrysis sp. MDV isolates without LysoTracker Green staining. a. green channel; b, DIC image; c, chlorophyll autofluorescence; d, overlays of a – c. Bars indicate 5 μm.

123

Figure A4.2. Confocal microscopic images of Chlamydomonas sp. ICE-MDV isolates without LysoTracker Green staining. a. green channel; b, DIC image; c, chlorophyll autofluorescence; d, overlays of a – c. Bars indicate 5 μm.

124

Figure A4.3. Heat map of communities associated with sorted eukaryotic organisms based on 16s rRNA sequence abundance. 16s OTUS were identified to phylum level. Color code on the left indicates the abundance of sequence of each phylum. Color code on the right side represent high chlorophyll autofluorescence (red), high LysoTracker fluorescence (green) or intermediate of either (blue) of the eukaryotic organisms. The communities associated with heterotrophic organisms were predominated by chloroplasts of phytoplankton (blue box), indicating that they were the potential prey of the heterotrophic eukaryotes.

125

CHAPTER V

Conclusion

Wei Li, Rachael M. Morgan-Kiss

126

CHAPTER V

5.1. Conclusions

Conclusions

In aquatic ecosystems, microorganisms form the foundation of food webs and play critical roles in global carbon and nutrient cycles as well as ecosystem services (Barber 2007, Schulze and Mooney 2012). The linkages between phytoplankton, heterotrophic bacteria, protozoa and viruses are described as the concept of the “microbial loop” (Azam et al 1983), which has been recognized as a major contributor to carbon and elementary biological cycling (Field et al 1998, Pernthaler 2005, Pomero et al 2007). Single-celled eukaryotic microorganisms (i.e., protists) are ubiquitous in every ecosystem on earth and play critical roles in the microbial loop in the cycling of carbon, energy and nutrients (Montagnes et al 2012). Diverse protist lineages possess multiple nutritional modes; playing key roles as producers, decomposers, parasites, and predators. Complex interactions between protists and bacteria as well as environmental conditions directly influence the function of microbial loop and further the stability of whole ecosystem (Amin et al 2012, Faust and Raes 2012, Ferrantini et al 2009, Medina-Sánchez et al 2004).

Meromictic lakes are bodies of water which are characterized by permanent stratification of major physical and chemical environmental factors. Water masses exhibit year-round high physical stability. Microbial consortia residing in permanently stratified lakes exhibit relatively constant spatial stratification throughout the water column and are adapted to vastly different habitats within the same water column (e.g. stable transition zone between oxic mixolimnion and anoxic monolimnion). There is also often a dense and diverse microbial community residing in the redox transition zone (Van Gemerden and Mas 1995). Thus, meromictic lakes are interesting model systems for many questions in aquatic biology research, and more specifically ideal for studying the impact of environmental drivers on microbial community structure and interactions within the microbial loop (Eiler et al 2012, Van der Gucht et al 2007, Yang et al 2015).

127

Pristine perennially-ice-covered lakes (Lake Bonney, Lake Fryxell and Lake Vanda) are meromictic lakes located in the McMurdo Dry Valleys (MDV) of Southern Victoria Land, Antarctica. The lakes have isolated water bodies and extremely stable strata that vary physically, chemically, and biologically within and between the water columns (Lyons et al 2000). The unique characteristics that support microbial dominated food webs. These lakes are considered as ideal model systems to study fundamental ecological questions. In the research presented here, the overall goal is to gather new understanding of how environmental drivers influence microbial community structure in these aquatic ecosystems. In order to achieve the goal, we explored the lake microbial ecology from three major approaches:

1 Assess trophic activities in the natural environment and identify potential environmental drivers impacting heterotrophic (β Glucosaminidase) and autotrophic (Ribulose 1,5 bisphosphate carboxylase) enzyme activities. (Chapter 2) 2 Resolve the protist community composition (i.e. autotrophic, heterotrophic and mixotrophic groups) based on high throughput sequencing and bioinformatics. Identify how the community structures correlate with specific environmental and biological factors.(Chapter 3) 3 Reveal the diversity of potential microbial interactions between the microorganisms in the MDV lakes at individual cell level, and investigate how the interactions vary between organisms with different nutritional strategies. (Chapter 4) The major conclusions of this research are summarized below:

Eukaryotic carbon fixation and sequestation are critical processes in the carbon cycle in the lake ecosystem. In this research the community trophic potential was directly measured via RubisCO and βGAM enzyme activities. Based on functional enzyme activity, autotrophic and heterotrophic spatial trends within the MDV lakes are complex. While autotrophic metabolism was generally dominant at depths at and below the chemocline, heterotrophic metabolism was dominant at both shallow, nutrient deficient depths (Lake Bonney), within the chemocline (Lake Bonney and Vanda), or below the chemocline (Lake Fryxell) (Figure 2.5). Both autotrophic and heterotrophic enzyme activities are impacted by environmental factors, however the level of activities are correlated with distinct environmental parameters depending on whether it is in mixolimnion or in chemocline and monolimnion (Table 2.2). Pairing this function studies with

128 environmental parameters will improve understanding of microbial community structure and how this may be impacted by climate driven fluctuation, an area of research with many unanswered questions (Caron and Hutchins 2012).

Each MDV lakes have distinct community structures, and in general, east and west lobes of Lake Bonney are more similar to each other than to Lake Fryxell. Both eukaryotic and prokaryotic distributions in the MDV lakes are stratified meaning the microbial communities adapt to the physicochemical conditions in different depth in the lakes. Despite the variation among lakes, the eukaryotic communities in the shallow mixolimnions are significant different compared with them at or below chemoclines, while bacterial communities are differentiated by the oxic and anoxic zones. Coherence of functional groups with diverse metabolic capabilities might responsible for biogeochemical processes and feather affect the whole ecosystems in the MDV lakes. (Figure 5.1). Strong correlation between photosynthetic protists to the nutrient (especially nutrient balance) which is anticipated to be impacted by climate related events provides new insights into how the microbial structure will be altered with the global changes.

129

According to our current knowledge we propose integrated food webs of west Lake Bonney and Lake Fryxell which are two of most intensively studied lakes in the this region. In the mixolimnions of both WLB and FRX, mixotrophic cryptophytes dominate the protist populations (Figure 3.3). Despite that light availability is relatively high in the shallow water column of WLB, in the nutrient deficient environment, due to the capability of utilizing both solar energy and grazing bacteria, mixotrophs are favorable over the large pure autotrophic chlorophytes (Chlamydomonas which have highest abundance in the chlorophyta in these layers). In addition, bacteria are abundant even in the shallow water layers (Figure 2.4), predation on bacteria might support the mixotrophic flagellate population ((Marshall and Laybourn-Parry 2002)). This suggestion supports our observation of high heterotrophic activities above chemocline at the bacteria concentration peak (Figure 2.5). The strong negative correlation between salinity and cryptophytes population (Figure 3.6) might explain the difference of cryptophytes distribution in the WLB and FRX. Cryptophytes diminish as reaching deep saline layers in WLB but dominate FRX throughout the water column which is fairly fresh water. In the deep aphotic zone of WLB, the populations of heterotrophs (i.e., chrysophytes; Figure A3.2-H) and small salt-tolerant haptophytes (Figure A3.2) increase, in contrast, heterotrophic nanoflagellates (i.e., choanoflagellates) are abundant in the deep layers of FRX. We were unable to classify more than 50% of the 18S OTUs from the deep FRX samples based on the available database, although is not uncommon when using similar approach(Oikonomou et al 2015, Triadó-Margarit and Casamayor 2015). Since the distribution of eukaryotes presented here was based on known taxa information in the data base, the abundance of some organisms might be overestimated. The bacteria communities are distinct between WLB and FRX. In FRX our data support the assumption that products of anaerobic bacterial metabolism originating from the sediment and the monimolimnion such as sulfide, methane and ammonia, support a rich and abundant bacterial community in the chemocline (Sattley and Madigan 2006), while in WLB phytoplankton originated dimethylsulfoniopropionate (DMSP) supports a sulfur oxidizing bacteria community including wide range of organisms who can utilize organic sulfur compounds (Cavicchioli 2015, Raina et al 2010). It has been predicted that climate changes would have complex effect on the lake physicochemical conditions including light availability, nutrient conditions as well as chemocline depth etc. (Doran et al 1994, Doran et al 2002, Roberts et al 2004). Our models as well as the findings in this research will help better understand how the microbial community

130 structure would be altered. Also studies of polar microbial communities on the cusp of environmental change will be important for predicting how microbial communities in low latitude aquatic systems will respond.

131

Figure 5.1 Working models of our current understanding of major carbon and energy cycles in Lake Bonney (west) vs. Lake Fryxell. Colors for phytoplankton represent dominant groups: green – chlorophytes; brown – haptophytes; orange – cryptophytes. Other abbreviations: SRB- sulfate reducing bacteria; SOB-sulfide oxidizing bacteria; GSB-green sulfur bacteria; PNSB- purple non-sulfur bacteria.

132

Reference:

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76: 667-684.

Azam F, Fenchel T, Field J, Gray J, Meyer-Reil L, Thingstad F (1983). The ecological role of water-column microbes in the sea. Marine ecology progress series Oldendorf 10: 257-263.

Barber RT (2007). Oceans: picoplankton do some heavy lifting. Science 315: 777-778.

Caron DA, Hutchins DA (2012). The effects of changing climate on microzooplankton grazing and community structure: drivers, predictions and knowledge gaps. Journal of Plankton Research: fbs091.

Cavicchioli R (2015). Microbial ecology of Antarctic aquatic systems. Nature Reviews Microbiology 13: 691-706.

Doran PT, Wharton Jr RA, Lyons WB (1994). Paleolimnology of the McMurdo dry valleys, Antarctica. Journal of Paleolimnology 10: 85-114.

Doran PT, Priscu JC, Lyons WB, Walsh JE, Fountain AG, McKnight DM et al (2002). Antarctic climate cooling and terrestrial ecosystem response. Nature 415: 517-520.

Faust K, Raes J (2012). Microbial interactions: from networks to models. Nat Rev Microbiol 10: 538-550.

Field CB, Behrenfeld MJ, Randerson JT, Falkowski P (1998). Primary production of the biosphere: integrating terrestrial and oceanic components. Science 281: 237-240.

Lyons WB, Fountain AG, Doran PT, Priscu J, Neumann K (2000). The importance of landscape position and legacy: The evolution of the Taylor Valley Lake District. Freshwat Biol 43: 355- 367.

133

Marshall W, Laybourn-Parry J (2002). The balance between photosynthesis and grazing in Antarctic mixotrophic cryptophytes during summer. Freshwater Biology 47: 2060-2070.

Medina-Sánchez JM, Villar-Argaiz M, Carrillo P (2004). Neither with nor without you: a complex algal control on bacterioplankton in a high mountain lake. Limnology and Oceanography 49: 1722-1733.

Montagnes D, Roberts E, Lukeš J, Lowe C (2012). The rise of model protozoa. Tren Microbiol 20: 184-191.

Oikonomou A, Filker S, Breiner HW, Stoeck T (2015). Protistan diversity in a permanently stratified meromictic lake (Lake Alatsee, SW Germany). Environmental microbiology 17: 2144- 2157.

Pernthaler J (2005). Predation on prokaryotes in the water column and its ecological implications. Nature Reviews Microbiology 3: 537-546.

Pomero LR, Williams PJ, Azam F, Hobbie J (2007). The microbial loop. Oceanography 20: 28.

Raina J-B, Dinsdale EA, Willis BL, Bourne DG (2010). Do the organic sulfur compounds DMSP and DMS drive coral microbial associations? Trends in microbiology 18: 101-108.

Schulze E-D, Mooney HA (2012). Biodiversity and ecosystem function. Springer Science & Business Media.

Triadó-Margarit X, Casamayor EO (2015). High protists diversity in the plankton of sulfurous lakes and lagoons examined by 18s rRNA gene sequence analyses. Environmental microbiology reports 7: 908-917.

134