Variabilité Temporelle, Diversité Et Biogéographie Des Ciliés Et Dinoflagellés Dans L’Océan Arctique

Home , Blastodiniales, Dinophysis ovum, Nassophorea, Picozoa

Variabilité temporelle, diversité et biogéographie des ciliés et dinoflagellés dans l’Océan Arctique

Thèse

Deo Florence Onda

Doctorat interuniversitaire en océanographie Philosophiae Doctor (Ph. D.)

Québec, Canada

Variabilité temporelle, diversité et biogéographie des ciliés et dinoflagellés dans l’Océan Arctique

Thèse

Deo Florence Onda

Sous la direction de :

Connie Lovejoy, directrice de recherche Marcel Babin, codirecteur de recherche

RÉSUMÉ

Des impacts du changement climatique sur les communautés de phytoplancton microbien ont déjà été reportés dans l'océan Arctique. Cependant, peu d’attention a été portée sur le microzooplancton aux rôles écologiques multiples tels que les ciliés et les dinoflagellés. Le but de cette thèse était d'améliorer les connaissances et la compréhension de l'écologie du microzooplancton, mais aussi de permettre la prédiction de leur réponse dans un Arctique en mutation. Nous avons utilisé la technique de séquençage d'amplicons à haut débit du gène 18S ARNr et de l'ARNr 18S (à partir d'ADNc) afin d’étudier le profil des communautés ainsi que leur diversité. Par la suite, nous avons testé différentes hypothèses reposant sur les variations temporelles et spatiales de ces assemblages microzooplanctoniques. Les résultats ont montré que le microzooplancton présente une forte saisonnalité dans le golfe d'Amundsen en réponse aux conditions changeantes. Des assemblages estivaux semblables ont été observés de 2003 à 2010, à l'exception de juillet 2008, après le record de minimum de glace en été 2007. Cet évènement particulier a permis de nous indiquer une sensibilité de ces espèces face aux conditions de glace. Les communautés de dinoflagellés du bassin du Canada sont régies par des processus déterministes et stochastiques qui dépendent de la variabilité de l'environnement, indiquant une sensibilité potentielle aux changements environnementaux. Nous en avons déduit que les dinoflagellés et autres taxons apparentés, ayant des rôles fonctionnels similaires, peuvent fournir une stabilité des flux alimentaires et énergétiques dans des conditions de limitation de lumière ou de nutriments, associées à l’approfondissement de nitracline. Étant donné que de nombreux ciliés et dinoflagellés sont mixotrophes, ils pourraient indirectement influencer les cycles biogéochimiques par le broutage sur les bactérivores et le petit plancton, et ainsi relier la boucle microbienne avec les niveaux trophiques supérieurs. La grande diversité et l'ubiquité des ciliés et des dinoflagellés suggèrent également une complexité dans les réseaux trophique microbiens et de nouvelles possibilités de recherche pour les océanographes. .

iii

ABSTRACT

Impacts of climate change on microbial communities in the Arctic Ocean have been mostly reported for major phytoplankton groups, with less attention to microzooplankton, such as ciliates and dinoflagellates, which have multiple ecological roles. For example, many ciliates and dinoflagellates are mixotrophic and could indirectly influence biogeochemical cycles by grazing on bacterivores and small plankton and linking the microbial loop with the higher trophic levels. The aim of this thesis was to address knowledge gaps in microzooplankton phylogeny, ecology and distribution with a goal of providing information needed to eventually predict of microzooplankton responses to the changing Arctic. We used high throughput amplicon sequencing of the 18S rRNA gene and 18S rRNA (as cDNA) to generate community and diversity profiles, which were used to test hypotheses on microzooplankton assembly across time and space. Results showed that microzooplankton exhibited strong seasonality in response to changing conditions in Amundsen Gulf. Similar summer assemblages were seen from 2003- 2010 with the exception in July 2008 following the summer ice minimum record in 2007. Canada Basin dinoflagellate communities were governed by both deterministic and stochastic processes that were dependent on the variability of the environment, indicating potential sensitivity to environmental change. We inferred that dinoflagellates and other taxa with similar functional roles could provide stability to food and energy flows under conditions of light- or nutrient-limitation associated with a deepening nitracline. The high diversity and ubiquity of ciliates and dinoflagellates also suggest a complexity within microbial food webs and new research opportunities for oceanographers.

TABLE DES MATIÈRES

RÉSUMÉ ...... ii ABSTRACT ...... iv TABLE DES MATIÈRES ...... v LISTE DES TABLEAUX ...... viii LISTE DES FIGURES ...... ix LISTE DES TABLEAUX SUPPLÉMENTAIRES ...... xi LISTE DES FIGURES SUPPLÉMENTAIRES ...... xii LISTE DES ABRÈVIATIONS ...... xiv REMERCIEMENTS ...... xvii AVANT-PROPOS ...... xix

CHAPITRE 1: INTRODUCTION GÉNÉRALE ...... 1 1.1 L’Océan Arctique et l’Arctique canadien ...... 1 1.1.1 Golfe d’Amundsen ...... 3 1.1.2 bassin Canada ...... 3 1.2 La succession d’eucaryotes microbiens dans un Arctique stratifié ...... 4 1.2.1 Banquise et eau de surface ...... 5 1.2.2 L’Halocline du Pacifique et le Maximum de Sub-surface en Chlorophylle (MSC) ...... 7 1.3 Vue d’ensemble des ciliés et des dinoflagellés ...... 8 1.3.1 Ciliés ...... 9 1.3.2 Dinoflagellés...... 9 1.3.3 Les fonctions trophiques des ciliés et des dinoflagellés ...... 10 1.4 Identification et classification des communautés microbiennes ...... 12 1.5 Méthodes statistiques...... 13 1.5.1 Mesures de Dissimilarité et regroupement ...... 13 1.5.2 L'analyse canonique de la correspondance (ACC) ...... 14 1.6 Les communautés microbiennes dans un Arctique en changement ...... 14 1.7 Objectifs et structure de la thèse ...... 16

Chapter 2: Seasonal and interannual changes in ciliate and dinoflagellate species assemblages in the Arctic Ocean (Amundsen Gulf, Beaufort Sea, Canada) ...... 19 Résumé ...... 19 Abstract ...... 20 2.1 Introduction ...... 21 2.2 Materials and Methods ...... 22 2.2.1 Sample collection, extraction and sequencing...... 22 2.2.2 Post-sequence data processing and taxonomic classification ...... 24 2.2.3 Statistical and diversity analyses ...... 25

2.3 Results ...... 26 2.3.1 Physico-chemical regimes of the Amundsen Gulf ...... 26 2.3.2 Community clustering based on 18S rRNA and 18S rRNA genes ...... 26 2.3.3 Seasonal succession and distribution ...... 27 2.3.4 Interannual variability ...... 28 2.4 Discussion ...... 28 2.4.1 Interpretation of DNA and RNA-derived abundances and diversity ...... 29 2.4.2 Environmental influences on microzooplankton ...... 30 2.4.3 Ecological functions ...... 31 2.4.4 Interannual microzooplankton turnover and the changing Arctic ...... 31 2.5 Acknowledgments ...... 33 2.6 Author Contributions ...... 34

Chapter 3: Dinoflagellates (Dinophyceae) in size fractionated samples from Canada Basin, Western Arctic Ocean ...... 50 Résumé ...... 50 Abstract ...... 51 3.1 Introduction ...... 52 3.2 Materials and Methods ...... 54 3.2.1 Sampling and sequencing ...... 54 3.2.2 Bioinformatics and small fraction-associated Dinophyceae OTU picking ...... 56 3.2.3 Phylogenetic placement of short reads ...... 57 3.2.4 Statistical and ecological network analyses ...... 58 3.3 Results ...... 58 3.3.1 Physico-chemical profiles of the sampling sites ...... 58 3.3.2 Small fraction-associated dinoflagellate diversity ...... 59 3.3.3 Community structuring and correlation with environmental variables ...... 59 3.4 Discussion ...... 60 3.4.1 Linking ratio of fractionated rDNA reads to potential cell characteristics ...... 61 3.4.2 Phylogenetic identity of small-fraction associated Dinophyceace ...... 62 3.4.3 Distribution and potential ecologies ...... 63 3.5 Conclusions ...... 65 3.6 Acknowledgments ...... 65

Chapter 4: Assembly rules governing dinoflagellates and their potential responses to changing Western Arctic Ocean ...... 81 Résumé ...... 81 Abstract ...... 82 4.1 Introduction ...... 83 4.2 Materials and Methods ...... 85 4.2.1 Study sites, sample collection and processing ...... 85 4.2.2 DNA-RNA extraction, sequencing and bioinformatics processing ...... 86 4.2.3 Community diversity and statistical analyses ...... 87 4.3 Results ...... 89 4.3.1 Environmental characteristics and general features ...... 89 4.3.2 Community Similarity and Phylogenetic structuring ...... 89 4.3.3 rRNA to rDNA ratios and taxonomic distribution ...... 91 4.4 Discussion ...... 92 4.4.1 Environmental variability and selection processes ...... 92 4.4.2 Community structuting and variability in life cycle strategies ...... 95 4.4.3 Dinoflagellate communities in the deepening nitracline ...... 95 4.5 Conclusion ...... 97 4.6 Acknowledgments ...... 97

CHAPTER V. GENERAL CONCLUSIONS ...... 111 5.1 Synthesis of the study ...... 111 5.2 Perspectives ...... 113 5.2.1 Databases and taxonomic classification ...... 114 5.2.2 Limitations of 18S rRNA gene for ciliate and dinoflagellate studies ...... 115 5.2.3 Microzooplankton in Arctic food web models ...... 116

BIBLIOGRAPHY ...... 117

vii

LISTE DES TABLEAUX

Table 2.1. Dates of collection 2007 -2008 (Date), Stations (Stn), collection Latitude and Longitude (Lat-Long), physico-chemical parameters and chlorophyll a (mg.m3 Chl a) concentrations of the samples used for amplicon tag pyrosequencing. Other column names refer to day length (DayL), depth (Z) of sampling, temperature (T), salinity (S), nitrate+nitrite (NO3¯mmol m3), 3- 3 3 phosphate (PO4 mmol m ),dissolved oxygen (DO mmolm ) and Photosynthetically Active Radiation (µE m-2 s-1 PAR). na: not available...... 34 Table 3.1. Stations (Stn), dates of collection (Date), collection Latitude and Longitude (Lat-Long), physico-chemical parameters and chlorophyll a (Chl a) concentrations of the samples used for amplicon high throughput sequencing. Other column names refer to depth (Z) of sampling, depth category (Layer), temperature (Temp), salinity (S), dissolved oxygen (DO), nitrate (NO3 3- ¯), silicate (Si(OH)4), phosphate (PO4 ), colored dissolved organic matter (CDOM) and dissolved inorganic carbon (DIC)...... 67 Table 4.1. Station name (Stn), Dates of collection 2012-2013 (Date), collection Latitude and Longitude (Lat-Long), and physico-chemical parameters of the samples used for amplicon tag Illumina sequencing. Other column names refer to depth (Z) of sampling, temperature (T), 3 salinity (S), dissolved oxygen (DO), Chlorophyll a fluorescence (Chl a mg m ), nitrate (NO3¯ 3 3 3 3 mmol m ), silicate (SiO4 mmol m ), phosphate (PO4 mmol m ),conductivity (Cond), colored dissolved organic mtter (CDOM mg m3), dissolved inorganic carbon (DIC x102 μmol kg-1), alkalinity (Alk x102 μmol kg-1), bacterial (Bact x104 cell ml-1), picoeukaryotic (Pico cell ml-1) and nanoeukaryotic (Nano cell ml-1) cell counts. na: not available...... 99

viii

LISTE DES FIGURES

Figure 1.1. Projection polaire de la region de l’Arctique (60-90⁰ N)...... 1 Figure 1.2 Emplacements du Bassin Canadien (image à gauche) et du Golfe d’Amundsen (boîte rouge), qui ont été les sites d’étude. Image generee par Ocean Data Viewer (ODV; Schlitzer, 2016)… ...... 2 Figure 1.3.Un schéma simplifié du réseau trophique arctique où la préférence de taille joue un rôle important. Notez la position des ciliés et des dinoflagellés en tant que prédateurs de petit phytoplancton que de proies pour le plus large zooplancton (redessiné de Glöckner et al.,2012, p. 27, original de Sandaa, 2009) ...... 5 Figure 2.1. Sites and dates of collection of water samples used for A) seasonal (November 2007-July 2008) and B) interannual (2003-2010) studies from the Amundsen Gulf region and adjacent Darnley and Franklin Bays...... 36 Figure 2.2. A dendrogram of the unweighted UniFrac dissimilarity matrix for both DNA- (filled shapes) and RNA-based (open shapes) communities collected from surface (triangles) and subsurface chlorophyll maxima/halocline (SCM; circles) layers representing different seasons including Autumn-Winter (AW) and Spring -Summer (SS), collected over the course of the IPY-CFL study...... 37 Figure 2.3. Distribution of potentially active (based on rRNA reads) major taxa of the lowest possible ranks in (A, C) ciliates and (B, D) dinoflagellates (98% level) with the changing seasons (sampling date) in surface and SCMhalocline layers (see text). Samples were collected from Amundsen Gulf, Darnley Bay and Franklin Bay...... 38 Figure 2.4. A network source-target plot showing the significant correlations of OTUs binned at the lowest possible ranks (relative abundances) and square root transformed environmental parameters. Only vertices (circles) and edges (arrows) that have Spearman’s rho >0.3 significant at p<0.001 were retained. Black circles are dinoflagellates and grey are ciliates. Environmental variables tested include dissolved oxygen (DO), total chlorophyll a (chl a), daylength (DayL), temperature (Temp), depth (Dep), salinity (Sal), nitrate+nitrite (NO3¯), 3- phosphate (PO4 ) and ice. Numbers correspond to the identities on the right...... 39 Figure 2.5. (A) The pairwise unweighted UniFrac community dissimilarity of combined ciliate and dinoflagellate communities. In the boxplots, one point (dot) represents the pairwise Unweighted UniFrac dissimilarity (Y-axis) of a particular sample against another sample. Thus, for each date there are 10 dots (overlapping dots are obscured) representing that date compared against 10 dates. The black line within the box, is the mean UniFrac distance of a particular sample against all the other samples. The broken line indicates the mean dissimilarity value among all samples. (B) Combined relative abundances of the major taxa contributing to the high July 8 dissimilarity, particularly of Strombidium, Laboea, Monodinium sp. and other unclassified ciliates...... 40 Figure 3.1. Map of the stations (red circles) where samples used in this study were collected from Western Canada Basin and Northwind Ridge in August 2012 onboard the CGGS Louis S St. Laurent...... 68 Figure 3.2. Salinity (a), temperature (b) and chlorophyll a fluorescence (c) plots of the six stations (color coded) used in this study...... 68 Figure 3.3. Dinoflagellate OTUs were grouped according to the ratio of their rDNA reads in large (>3 µm) over small (0.2-3 µm) fractions that were generated by high throughput sequencing from the surface, DCM and PWW m depths (see text) collected from Western Arctic Ocean

(Canada Basin). Negative Log10 values indicate small fraction-associated OTUs (small- associated) and positive are large fraction-associated OTUs (large-associated). Not shown are OTUs that had equal chances (0 Log10) of being found in both fractions...... 69 Figure 3.7. An ecological network showing strong and positive co-occurrences (Spearman’s rho >0.6, p<0.01) of the Syndiniales and their potential microplankton hosts (ciliates, dinoflagellates and radiolarians) generated from highly abundant OTUs (>100 reads). The different taxonomic groups are represented by the colored nodes, while varied thickness of the edges (black lines) refer to the level of Spearman’s correlation. The numbers refer to significant clusters or grouping of highly connected co-occurrences, while the shapes correspond to depths where the OTUs were most abundant based on rRNA reads. The 21 small-associated dinoflagellate OTUs were excluded from this particular analysis since they were assumed to be inactive cysts...... 73 Figure 4.1. Map of the stations in the northern and western regions of Canada Basin including the Northwind Ridge (Stn. TU-1) White circles indicate stations uniquely visited in 2012 and black in 2013, while gray circles were visited in both years...... 101 Figure 4.2. Relationship between selection processes and environmental variability. A.) Bar graphs showing the range of variation of the standard deviations of within depth Bray-Curtis Dissimilarities (SDBC) based on depth, temperature, salinity, DO, Chl a, transmissivity, nitrate, silicate, phosphate, conductivity and CDOM values. SDBC in the surface was significantly higher than the DCM and PWW also indicating highest overall variability. B) Phylogenetic structuring of communities (beta-diversity) based on the presence-absence of OTUs (Unweighted-UniFrac) also showed strong depth-category association clustered using PCoA. Shapes represent year of collection: 2012 (circles) 2013 (triangles) C.) Depth-dependent differences in the relative influences of stochastic and deterministic processes along the three water masses inferred from the within-depth NTI was also observed. All values were > -2 indicating lack of signals of overdispersion or homogenous selection. Colors represent depths in A-C (green=surface, blue=DCM, violet=PWW)...... 102 Figure 4.3. Log transformed rRNA:rDNA ratio based on rarefied read counts of each taxa binned on lowest possible rank represented by different colors on the right. The size of the circles corresponds to the number of OTUs of the same taxonomic ranking. A) surface, B) DCM and C) PWW waters ...... 103 Figure 4.4. Heatmap based on the level of activity (rRNA reads) of the most abundant OTUs binned to the lowest possible ranks. Distribution of these taxa mostly clustered by depth, where photosynthetic-mixotrophic groups were more abundant in the surface and DCM while heterotrophic taxa dominated in the DCM and mostly in PWW. Colored bars represent depths (green= surface, blue=DCM, violet=PWW) and clustering was determined using UPGMA approach...... 104

LISTE DES TABLEAUX SUPPLÉMENTAIRES

Supplementary Table S2.1. Proportion of large (>3 µm) and small (0.2-3 µm) fractions in the normalized tag pyrosequencing based on size-fractionated chlorophyll a concentrations estimated from the original samples...... 41 Supplementary Table S2.2. Summary of physico-chemical data associated with the interannual samples collected during summer to fall in the Amundsen Gulf and Beaufort Sea from 2003- 2010. na: not available...... 41 Supplementary Table S2.3. Summary of the OTUs in the surface shared by Jan 3 and Jun 14 samples but not present in the other summer samples from June 24 and July 21...... 42 Supplementary Table S2.4. Summary of dinoflagellate OTUs detected in the seasonal data per depth (surface and halocline/SCM) and their corresponding identities and abundances in RNA and DNA libraries...... 43 Supplementary Table S2.5. Summary of ciliate OTUs detected in the seasonal data per depth (surface and halocline/SCM) and their corresponding identities and abundance in RNA and DNA libraries...... 43 Supplementary Table S3.1. Summary of the stations and depths sampled (surface, deep chlorophyll maxima [DCM], PWW), and the fraction (small/large) and type of samples (RNA/DNA) used in this study...... 74

LISTE DES FIGURES SUPPLÉMENTAIRES

Supplementary Figure S2.1. The mean number of OTUs in the different seasons, summer (Jul-Sep) and autumn (Oct-Nov), of ciliates and dinoflagellates collected from 2003 to 2010. No significant difference was observed between these seasons within each taxa/group. Dinoflagellates however were generally more abundant than ciliates (p<0.001). OTU picking was carried out using USEARCH 7 and compared against the SILVA database at 98% similarity...... 45 Supplementary Figure S2.2. Bray-Curtis similarity clustering of 9 environmental variables of the seasonal surface and halocline/SCM samples in the Amundsen Gulf and Franklin Bay. Shapes indicate surface (triangle) and halocline/SCM (circle) depths. Samples from Darnley Bay (DB) are marked with red stars while those from Franklin Bay are indicated with blue stars. The samples grouped by depth and season rather than geographical location, except for the June 14 samples from DB...... 46 Supplementary Figure S2.3. Species richness (Chao1 index) based on DNA also changed with season (Julian Day). A significant decrease in richness in the total eukaryotic community was observed in the months after mixing and loss of ice (broken lines). Halocline/SCM-based richness (circles) was also significantly higher than the surface (triangle) in the samples collected from Amundsen Gulf (AG), Darnley Bay (DB) and Franklin Bay (FB)...... 47 Supplementary Figure S2.4. Annual mean relative abundances of both A) ciliates and B) dinoflagellates relative to the total eukaryotic microbial community based on 11 samples collected over an 8-year period from 2003 to 2010 in the SCM of the Amundsen Gulf and Beaufort Sea...... 48 Supplementary Figure S2.5. A maximum likelihood gene tree generated from nearly full length 18S rRNA gene sequences showing the placement of the most abundant “Other Ciliates” OTUs from the interannual data, here represented by their most similar reference sequences (OTU 1- 14). Arrows indicate significant increase or decrease in abundance after 2007, tested using Krustal Wallis H-Test implemented in STAMP (Parks et al., 2014). The tree was constructed using RaxML with bootstrap support repeated 1000 times. Only bootstrap values higher than 50 are shown. The short HTS OTU reads were mapped back to the tree using RaxML- Evolutionary Placement Algorithm (RaxML-EPA)...... 49 Supplementary Figures S3.1. An 18S rRNA gene reference trees generated by RaxML approach, which was used to identify HTS-derived reads of small-associated core dinoflagellates OTUs, also described in Figure 3.4A...... 75 Supplementary Figures S3.2. An 18S rRNA gene reference trees generated by RaxML approach, which was used to identify HTS-derived reads of small-associated Syndiniales OTUs in Syndiniales Group I (A) and Syndiniales Groups II-V (B), also described in Figure 3.4B...... 77 Supplementary Figure S3.3. Principal components analysis (PCA) ordination of the samples based 3- on the nutrients (Nitrate, NO3¯; Phosphate, PO4 ; Silicate, Si(OH)4), salinity (Sal), temperature (Temp), dissolved organic carbon (DIC), dissolved oxygen (DO) and chlorophyll a (Chl a) fluorescence values. Only variables that had significant contributions (marked by red arrows) are shown. Shapes indicate depth categories surface (square), deep chlotophyll maxima (triangle) and PWW (circle)...... 78 Supplementary Figure S3.4. Principal coordinates analysis (PCoA) plot based on Unweighted UniFrac (presence-absence) computed from small rDNA of the 143 selected putative picodinoflagellate OTUs showed that samples clustered by depth (surface=square, DCM=triangle, PWW=circle) with strong correlation (R2=0.95, p<0.001)...... 78

xii

Supplementary Figure S3.5. OTU-level likelihood occurrences of picoplanktonic Mamiellophyceae using ratios of rDNA reads in fractionated samples (large/small). The log10 values of the ratios were mostly negative, suggesting that most Mamiellophyceae were small fraction- associated and truly picoplanktonic. OTUs that were more positive or large fraction-associated may have been concentrated together with predators or associated with flocs of marine snow...... 79 Supplementary Figure S3.6. Order/genus-level likelihood occurrences using ratios of rDNA and rRNA relative abundances (from binned OTUs) in fractionated samples (large/small) showed that dinoflagellates could be grouped according to these ratios. Class 1 is composed of taxa whose rDNA and rRNA both occur more on the larges. Class 2 is made up of taxa whose rDNA are more on the large fraction, but rRNA with equal odds of being found in both fractions. Class 3 are taxa that have rDNA found more in the large, but rRNA in small fraction, while Class 4 are taxa that have higher odds of being found in the small fraction for both rRNA and rDNA reads...... 80 Supplementary Figure S4.1. Abundances of the different major taxonomic groups relative to the total microbial community in the different depths (surface, DCM and PWW – see text) and fractions (small and large) based on (A) rRNA and (B) rDNA. Dinoflagellates were more abundant in the surface and PWW than the DCM. Samples from 2012 and 2013 were arranged starting with the stations from the shallower Northwind Ridge (TU1) northward to CBN2. Included are OTUs present only in either rRNA or rDMA. Not included are the taxa lower than 0.01%...... 106 Supplementary Figure S4.2. A summarized log10 representation of the fraction association of dinoflagellate OTU reads in small (0.2-3 um) over large (>3 um) fractions. Gray bars represent 1og10 rDNA reads and black are for log10 rRNA reads. This graph shows that in some OTUs, rRNA were more often found in the small fraction while rDNA reads were more associated with the large fraction, which may indicate effects of size fractionation artitifacts...... 107 Supplementary Figure S4.3. Comparison of the distribution of rRNA (-log10) and rDNA (+log10) after aggregation of reads from small and large fractions of the same OTUs. It was apparent that there was a stronger correspondence in the rRNA and rDNA reads of some OTUs after aggregation while others were more abundant in the rRNA than their rDNA and vice-versa, which could indicate environmental filtering. Many OTUs however, were either only present in the rRNA or in the rDNA, which were all removed from the final dataset...... 107 Supplementary Figure S4.4. A depth-dependent change in the relative influences of stochastic and deterministic processes (box plots) along the stratified layers of the Arctic inferred from the within-depth NRI was observed based on Net Relatedness Index (NRI) similar to the results of the NTI. Colours represent depths (green=surface, blue=DCM, violet=PWW)...... 108 Supplementary Figure S4.5. A triplot based on Canonical Correspondence Analysis (CCA) with forward selection showing the association of the samples (sites) and selected dinoflagellate taxa with environmental variables or factors including silicate (S), nitrate (N), phosphate (P), depth (D), coloured dissolved organic matter (CDOM), salinity (Sal), chlorophyll a (Chl a), dissolved oxygen (DO), nanoplankton (Nano), picoplankton (Pico) and bacteria (Bac). Colors represent depths (green=surface, blue=DCM, violet=PWW)...... 109 Supplementary Figure S4.6. Heatmap based on the relative abundance of rRNA reads of the less abundant taxa based on the lowest possible ranks. Coloured circles represent depths (green= surface, blue=SCM, violet=PWW) and clustering was determined using UPGMA approach.110

xiii

LISTE DES ABRÈVIATIONS

18S: Large subunit of the ribosomal gene in eukaryotes ACC : Analyse Corréspondence Canonique ADN ou DNA: Acide désoxyribonucléiques ou deoxyribonucleic acid AG: Amundsen Gulf ANOVA: Analysis of variance AO: Arctic Ocean ARN ou RNA: Acide ribonucléiques ou Ribonucleic acid BC: Bassin canadien BG: Beaufort Gyre BLAST: Basic local alignment search tool bp: Base pairs Bray-Curtis: Bray Curtis Dissimilarity CB: Canada Basin CBDW: Canada Basin Deep Water (l’eau profonde du basin canadien) CCA: Canonical Correspondence Analysis CCGS: Canadian Coast Guard Ship cDNA: complementary DNA generated from RNA, here using reverse transcription CDOM: coloured dissolved organic matter CFL: circumpolar flaw lead study Chl a: Chlorophyll a CO2: Carbon dioxide, dioxide du carbon CTD: conductivity, temperature, depth profiler DB: Darnley Bay DCM: Deep Chlorophyll Maxima (favoured term for the Canada Basin) DIC: Dissolved Inorganic Carbon DMSP: Dimethylsulfopropionate DO: Dissolved Oxygen EOL: Encyclopedia of Life EPA: Evolutionary Placement Algorithm EPE: l’eau du Pacificque d’été EPH: l’eau du Pacificque d’hiver EPM: d’Eau polaire modifiée FB: Franklin Bay GB: Gyre de Beaufort HA: Halocline de l’Atlantique HP: Halocline du Pacifique HSD: Honest significant difference HTS: High throughput sequencing IBIS: Institut de Biologie Intégrative et des Systèmes IPY: International polar year JOIS-BGEP: Joint Ocean Ice Studies – Beaufort Gyre Exploration Project MALVs: Marine alveolates MSC : Maximum de chlorophyll sous-surface NCBI: National Center for Biotechnology Information NMDS: Non-metric multidimensional scaling NO3 ¯: Nitrate NPZD: Nutrient-Phytoplankton-Zooplankton-Detritus model NRI: Net Relatedness Index

xiv

NTI: Nearest Taxon Index NWR: Northwind Ridge OA: L’Océan Arctique OTU: Operational Taxonomic Unit PAR: Photosynthetically active region PAST: Paleontogical Statistics PC: Polycarbonate PCA: Principal Component Analysis PCoA: Principal Coordinate Analysis PCR: Polymerase Chain Reaction PML: Polar Mixed Layer 3- PO4 : Phosphate PSW: Pacific Summer Water PWW: Pacific Winter Water QIIME: Quantitative Insights into Microbiology RaxML: Random AXelerated Maximum Likelihood rDNA: ribosomal DNA genes, amplicons generated from 18S rRNA gene from DNA rRNA: ribosomal RNA amplicons generated from 18S rRNA from cDNA SCM: Subsurface Chlorophyll Maxima, more general term used for Amundsen Gulf and adjacent seas SG: Syndiniales Group Si (OH)4: Silicate SRA: Short Read Archive (NCBI) Stn: Station TS: Temperature–Salinity UPGMA: Unweighted Pair Group Method with Arithmetic Mean

To my family, especially my Tatay and Nanay, Rodantes C. Onda and Carmencita L. Onda. #ParaSaBayan

xvi

REMERCIEMENTS

“If I have seen further, it is by standing on the shoulders of giants” – Isaac Newton, 1676 Succeeding in this endeavour would not have been possible without the help of many people who believed in my capabilities and supported me throughout this journey.

My deepest and heartfelt thanks to my adviser, Connie Lovejoy, for taking the risk of sending a Pacific Islander to the wonderful Arctic. Your mentorship, guidance and expertise in all stages of my doctoral studies shaped me to become a better researcher and person. It was a realization of my aspirations, from just reading your work to training under your supervision. Thank you for opening doors of endless opportunities for which I will be always greateful. My gratitude also goes to my co- director, Marcel Babin for the support and guidance, whose enthusiasm for the Arctic and science inspires younger scientists like us to pursue our goals.

Many thanks to my thesis committee members, Yves Graton (INRS) and André Rochon (UQAR), who had always been there from my doctoral exam to my thesis completion. Your scientific advice guided and helped me direct my research to more productive objectives. I am also grateful to David Morse of Université de Montreal, and Alexander Culley and Juan Carlos Villareal Aguilar of U. Laval for accepting the responsibility of being in my evaluation committee and their time for reading and commenting on this thesis.

Many people also guided and continue to guide me and my scientific career. Many thanks to my former MSc advisers, Dr. Rhodora Azanza and Dr. Arturo Lluisma who continued to provide me with their support and guidance even after my stint in UP-MSI. To Warwick Vincent, your enthusiasm in science inspires us to do better. To the great scientists I admire: Chris Bowler, Leila Tirichine, David Montagnes, Micah Dunthorn, Helena Yap, Coke Montaňo, John Clamp: my utmost gratitude for the support, advice, and opportunities, which greatly helped my scientific career.

I was lucky and privileged to have worked with many people. To my partner in crime and PhD buddy, Nathalie Joli, the past 4 years would not have been bearable without you. Thank you for all the moral and emotional support, the existential crisis conversations, and for always pushing me to do more. To Vincent Carrier, my academic brother, thanks for helping me keep my sanity, for helping me become a Quebecois, and for being there through all ups and mostly downs. Same goes to your entire family. To the Lovejoy Labo past and current, especially Mary Thaler, Vani Mohit, André Comeau, Cindy Dasilva, Jerome Comté, Bérangère Péquén, Marianne Potvin, Sophie Creveceour, and Dimitri

xvii

Kalenitchenko for the patience and your readiness to always help me. To Adam Monier, whom I had the opportunity to share great experiences with in the ship, the friendship we built while freezing during deployments will never be forgotten! To Nastasia Freyria and Loic Jacquémot, thanks for the help especially with my French troubles. To my fellow Takuvik Peeps, and batchmates whom I started this journey with, especially Moritz Schmid and Nicholas Schiffrine, and also to Debbie Christiansen-Stowe and Marie-Heléne Forget, it had been a privilege knowing all of you.

I have met and made a lot of friends in Québec who made me felt at home and loved. To my first Québec gang - Sophie, Antoine, Joannie, Arnaud, Alex, Mike, Simon, I could not have hoped for a better circle of friends in Québec! You made my Québec experiences unforgettable, worthwhile, and fun! To my ASNUUL family and co-execs, thank you for the opportunity of sharing the wonderful UN experience and friendship with all of you. To the Filipino-Québecers Association (FQA), especially Ate Noxie and Kuya Nico, Chloe, Pia and Tita Tia, thanks for being a family to me when I needed it most, and to the entire FQA for always keeping me well fed and loved.

My PhD stint would not have been successful and memorable without experiencing one of the most fascinating regions of the earth – the Arctic! For this, I am greateful to all the Captains and Crew of CCGS Louis S St. Laurent, scientists, and the DFO-IOS people onboard the JOIS expeditions from 2013 to 2015 – the cruises had not only been scientifically productive but also fun. My special thanks to Bill Williams, Sarah Zimmerman, Jenny Hutchings and David Walsh, for your insights in my research, learning science onboard had been more fun with you.

I have survived for 4 years of being away from home because of many people who were always there for me even if they were half across the globe. My friends - Tin, Dang, Eric, Jords, Jojo and MSI peeps, thank you for proving that distance is never a hindrance to friendships. My highest gratitudes to Majih, thank you for the inspiration, support, understanding and love. You have always been my number one supporter even in my moments of self-doubt. This is for us!

Lastly, to my family - Nanay, Tatay, Ate and Kuyas, thank you for letting me explore the world, for supporting and believing in me. Everything I have, and everything I am will not be possible without you. This page will not be enough to say all my thanks!

-deo, Québec

xviii

AVANT-PROPOS

Cette thèse présente les résultats de mon doctorat réalisé sous la supervision du Dr. Connie Lovejoy, à titre de directeur, et du Dr. Marcel Babin, à titre de codirecteur, tous deux professeurs au Département de biologie de l’Université Laval à Québec. Cette thèse sera divisée en cinq chapitres ; on y retrouva une introduction générale (chapitre 1), suivie d’une présentation sous forme d’articles scientifiques (chapitres 2 à 4), en terminant par une conclusion générale des perspectives de recherche (chapitre 5). Ma thèse doctorale se concentre autour des les eucaryotes microbiens marins dans l’Océan Arctique, principalement sur les ciliés et dinoflagellés dans l’ouest de l’Océan Arctique. Les données biologiques, physico-chimiques et accessoires utilisés dans cette recherche ont été collectées par des collègues ou par moi-même lors de mes nombreuses expéditions. Les échantillons provenant du golfe d’Amundsen et de la mer de Beaufort ont été recueillis pendant les Canadian Arctic Shelf Exchange Study 2003-2004, ArcticNet 2005-2010, Canada’s Three Oceans (C3O) 2007, Circumpolar Flaw Lead Study 2007-2008 et le projet France-Canada Malina 2009. En ce qui concerne le bassin canadien, les études ont été réalisées en collaboration avec les Joint Ocean Ice Studies (JOIS) et Beaufort Gyre Exploration Project (BGEP) 2012-2013. J’ai principalement utilisé la technologie séquençages haut debit (SHD) et j’ai personnellement réalisé la majorité du travail en laboratoire et la collecte de données. La rédaction à titre de coauteurs sur les articles scientifiques reflète la contribution apportée aux banques de données, à l’analyse et à la direction des interprétations. De plus, j’ai dirigé toutes les analyses bio-informatiques, généré et testé statistiquement l’hypothèse, en plus d’écrire les manuscrits, le tout sous la supervision de mon directeur de recherche. Plus spécifiquement :

Chapitre 1 : Introduction Générale Chapitre 2 : Onda D.F.L., Medrinal E., Comeau A., Babin M., Lovejoy C. (2017). Seasonal and interannual changes in ciliate and dinoflagellate species assemblages in the Arctic Ocean (Amundsen Gulf, Beaufort Sea, Canada). Frontiers in Marine Science, Aquatic Microbiology. 4(16), doi:10.3389/fmars.2017.00016 Chapitre 3 : Onda D.F.L., Lovejoy C. Dinoflagellates (Dinophyceae) in the size-fractionated samples from Canada Basin, Western Arctic Ocean. Soumis. Chapitre 4 : Onda D.F.L., Monier, A., Babin, M., Lovejoy C. Assembly rules governing dinoflagellates and their potential responses to nitracline deepening in the Western Arctic Ocean. Ce chapitre sera soumis sous peu pour publication. Chapitre 5 : Conclusion Générale Dans le cadre de mon doctorat, j’ai eu l’opportunité de faire un stage, supporté par Québec Océan, à l’Institut de biologie – École Normale Supérieure de Paris en France en 2014 sous la supervision des Drs Chris Bowler et Leila Tirichine, où j’ai appris des techniques moléculaires de pointe et où j’ai été

xix introduit au programme Tara Oceans. Cette expérience a été grandement enrichissante pour ma recherche. De plus, mes études antérieures à la The Marine Science Institute de University of the Philippines ont contribuées à mon doctorat, mais en m’orientant davantage vers la dynamique au niveau communautaire plutôt que les modèles bases sur des espèces. Mon but était d’intégrer ces deux lignes de recherche pour découvrir les principes généraux qui sous-tendent la structure du réseau alimentaire microbien, et comment les espèces interagissent dans un plus large contexte. J’ai également été inspiré par mes expériences antérieures aux Phillipines réalisées dans le cadre de ma maîtrise et par mes recherches professionnelles pendant lesquelles je me concentrais sur l’identification et le suivi des espèces dinoflagellés toxiques et dangereuses, en utilisant la microscopie et l’approche moléculaire. J’ai poursuivi mes publications sur ce sujet tout au long de mon doctorat, ce qui fût contributoire à la réalisation des mes objectifs de recherche. Ces travaux et collaborations ont résulté sur les publications suivantes :

1. Onda DFL, Lluisma AO and Azanza RV. (2015). Potential DMSP-degrading bacterial assemblage dominate endosymbitoic microflora of Pyrodinium bahamense var. compressum (Dinophyceae) in vitro. Archives of Microbiology, 197:965-71.

2. Onda DFL, Azanza RV and Lluisma AO. (2014). Development, morphological characteristics and viability of temporary cysts of Pyrodinium bahamense var. compressum (Dinophyceae) in vitro. The European Journal of Phycology, 49:265-275.

3. Onda DFL, Benico G, Sulit AF, Gaite PL, Azanza RV and Lluisma AO. (2013). Morphological and molecular characterization of some HAB-forming dinoflagellates from Philippine waters. The Philippine Science Letters, 6(1):1-10.

4. Azanza RV, Vargas VMD, Fukami K, Keshavmurthy S, Onda DF and Azanza PV. (2013). Culturable algalytic bacteria isolated from seaweeds in the Philippines and Japan. Journal of Environmental Science and Management, 1:1-10.

5. Manset KJV, Azanza RV and Onda DFL. (2013). Algicidal bacteria from fish cultures in Bolinao, Pangasinan, Northern Philippines. Journal of Environmental Science and Management, 1:11-20.

6. Marquez GPB, Reichardt WT, Azanza RV, Onda DFL, Lluisma AO, Hasegawa T and Montaño NME. (2015). Dominance of hydrogenotrophic methanogens at the peak of biogas production in thalassic digesters. Waste and Valorization, DOI 10.1007/s12649-014-9325-4.

La partie phylogénétiquede cette thèse, laquelle inclut la curation et la classification de haute résolutionet également l’amélioration de l’annotation et la classification taxonomique dans la plus récente version du Northern Microbial 18S rRNA Gene Reference Database sur laquelle j’ai travaillé à titre de collaborateur.

1. Lovejoy, C., Comeau, A., Thaler, M. 2016. Curated reference database of SSU rRNA for northern marine and freshwater communities of Archaea, Bacteria and microbial eukaryotes, v. 1.1 (2002-2008). Nordicana D23, doi: 10.5885/45409XD-79A199B76BCC4110. Les résultats de ces travaux ont également été présentés dans plusieurs conférences scientifiques aux niveaux local, national et international :

1. Onda, DFL and Lovejoy C. (2016). Eukaryotic microbial communities in the changing Arctic. Presented at the Green Talents Alumni Conference 2016, Octobre 2016, Berlin, Allemagne.

2. Onda, DFL and Lovejoy C. (2016). Diversity and biogeography of ciliates in the Arctic Ocean. Roundtable Meeting of the International Research Coordination Network for Biodiversity of Ciliates, Washington D.C., Septembre 2016.

3. Onda, DFL and Lovejoy C. (2016). The Missing Link: Ciliates in the changing Arctic. Poster, International Society of Microbial Ecology 2016, Montreal, Québec, Juliet 2016.

4. Onda, DFL and Lovejoy, C. (2016). Ciliates and Dinoflagellates exhibit strong seasonality and sensitivity to climatological changes in the Amundsen Gulf (Arctic). EMBL Symposia: New Age for Marine Eukaryote Discovery, EMBL-EMBO, Janvier 2016, Heidelberg, Allemagne.

5. Onda DFL and Lovejoy C. (2016). Patterns, Processes and Responses: understanding microbial community assembly in the changing Arctic. IBIS-Interlab, Novembre 2016.

6. Onda DFL. (2016). The Missing Link: ciliate in the changing Arctic. Journee des Etudiants. IBIS, Université Laval. August 2016. Winner: Best Poster Presentation.

7. Onda DFL. (2016). Ciliate and dinoflagellates in the changing Arctic. IBIS-Interlab, Université Laval, April 2016.

8. Onda DFL, Joli N, Potvin M and Lovejoy C. (2014). Eukaryotic microbial diversity, biogeography and potential interactions in the Canadian Arctic and surrounding seas. Oral presentation, Quebec-Ocean Annual General Assembly. Rivier du Loup, Quebec, Novembre 2014.

9. Onda, DFL and Lovejoy C. (2013). Genetic diversity and distribution of planktonic Gymnodinoid species in the Arctic waters. Poster presentation, Quebec-Ocean Annual General Assembly. Rivier du Loup, Quebec, Novembre 2013.

Finalement, au cours de mon doctorat, j'ai été récipiendaire de plusieurs bourses provenant d'organisations locales et internationales pour ma participation à des ateliers scientifiques et stages internationaux :

1. Green Talents Competition 2016,sponsorisé parle German Ministry on Research and Education, Berlin, Allemagne.

xxi

2. Ecological Dissertations in Aquatic Sciences (ECO-DAS XIII), sponsorisé par leAmerican Society for Limonology and Oceanography, SOEST-C-MORE-University of Hawa’ii (déclinée, conflit d’horaire avec 1). 3. Marine Eukaryote Early Career Scientist Training and Pre-Summitfor Advanced PhDs and Post-Doc, sponsorisé par leGordon and Betty Moore Foundation – Marine Microbes Initiative (MMI), tenu en European Molecular Biology Laboratory (EMBL), Heidelberg, Allemgane, Janvier 2016. 4. Fonds de recherche du Québec - Nature et technologies (FRQNT) International Training FellowshipparQuébec-Océan, stagedansl’École Normale Superieur, Paris, France, Mai-Julliet 2014. 5. Bourse Leadership et developpment durable, Université Laval, 2012-2014. 6. Canadian Excellence Research Chair – Takuvik Joint International Laboratory, bourse Doctorat, 2012-2016. 7. Natural Science and Engineering Research Council, Discovery Granta Connie Lovejoy 8. Bourse Richard Bernard, Hiver 2017, Fondacion Richard-Bernard et La Sociéte Provancher par Département de Biologie.

xxii

CHAPITRE 1: INTRODUCTION GÉNÉRALE

1.1 L’Océan Arctique et l’Arctique canadien L’Océan Arctique (OA; Figure 1.1) est la plus petite des cinq régions océaniques, couvrant approximativement 3% (18,8 million km2) de la surface planétaire totale (Jakobsson, 2002). Avec une profondeur moyenne de 1050 m, c’est également l’océan lemoins profond, entouré de terres et ainsi considéré comme un double estuaire (Carmack & Wassmann, 2006). La majeure partie de l’OA est située au-dessus du Cercle Polaire et est couverte saisonnièrement ou de façon pérenne de glace (Morison et coll., 2012). Malgré sa petite taille, l’OA est fortement influencé par différents apports d’eaux plus chaudes de densités différentes résultant en des régimes océaniques grandement stratifiés. Cette stratification rend cet océanplus stable comparativement aux autres océans du monde (Rainville et coll., 2011).

Figure 1.1. Projection polaire de la region de l’Arctique (60-90⁰ N). Image generée par Ocean Data Viewer (ODV; Schlitzer, 2016)

Dans l’ouest de l’Océan Arctique au nord de l’Amérique (du 90⁰ au 180⁰, Figure 1.1), les couches stratifiées sont catégorisées en quatre principales masses. Les principales couches, en allant de la surface vers le fond, sont (i) la couche d’eau polaire mélangée (EPM) ou Polar Mixed Layer (PML) ayant une faible salinité dû à la fonte saisonnière de la glace et des apports riverains (S = 27-31), (ii) les eaux du Pacifique riches en nutriments (S = 31-33), (iii) la masse d’eau plus salée mais moins riche

1 en nutriments de l’Atlantique (S> 34) et (iv) l’eau profonde du bassin canadien (CBDW), une eau plus froide et plus dense associée avec la formation des canaux de saumure de la glace de mer (Shimada et coll., 2005; McLaughlin et coll., 2005; Timmermans, et coll., 2008; Bjork et coll., 2010). Les haloclines entre ces masses d’eau sont robustes et maintiennent la structure physique du basin avec l’Halocline du Pacifique (HP) et l’Halocline de l’Atlantique (HA) dans la partie supérieure des eaux du Pacifique et de l’Atlantique respectivement (Steele, 2004). Á cause de sa profondeur, l’Halocline du Pacifique joue un rôle important dans le réapprovisionnement de nutriments dans les eaux de surface (Steele, 2004; Pickart & Stossmeister, 2008). Plus précisément, le refroidissement périodique de la surface résulte en un mélange causant la diminution rapide de la profondeur des eaux d’origine du Pacifique et le transport des nutriments dans les couches supérieures. La stratification dans l’Océan Arctique est également fortement influencée par les systèmes de vents dominants (voir les sections suivantes). La forte réciprocité entre les processus météorologiques et physicochimiques a des impacts significatifs sur les communautés biologiques, particulièrement sur les organismes marins planctoniques. Cette thèse se concentre sur les communautés de deux régions de l’Arctique canadien où nous avons testé des hypothèses sur l’écologie des protistes microzooplanctoniques.Des études temporelles (Chapitre 2) ont été réalisées en utilisant des échantillons recueillis au sud de la région de la Mer de Beaufort, dans le golfe d’Amundsen (Figure 1.2, encadré rouge), alors que les échantillons pour les analyses de diversité (Chapitre 3) et spatiales (Chapitre 4) ont été amassés au nord et à l’ouest du Bassin Canadien (Figure 1.2).

Figure 1.2 Emplacements du Bassin Canadien (image à gauche) et du Golfe d’Amundsen (carré rouge), qui ont été les sites d’étude. Image générée par Ocean Data Viewer (ODV; Schlitzer, 2016).

1.1.1 Golfe d’Amundsen Le golfe d’Amundsen (GA) est un grande détroit (longueur de 400 km largeur de170 km) représentant une extension du sud-est de la mer de Beaufort bornée par l’île de Banks au nord, le plateau du Mackenzie à l’ouest, les côtes continentales du Canada au sud et ses archipels à l’est (Shadwick et coll., 2011; Figure 1.2 rectangle). Son courant de surface est fortement influencé par le Gyre de Beaufort (GB) qui est principalement située dans la partie septentrionale du golfe d’Amundsen (voir Section 1.2, encadré rouge). Alors que le courant de surface est principalement contrôlé par la circulation anticyclonique de la gyre, les courants en profondeur de la mer de Beaufort demeurent principalement contrôlés par une circulation direction cyclonique, entraînés en un transport vers l’est des eaux du Pacifique et de l’Atlantique (MacDonald et coll., 1987). Les directions opposées des courantsde surface et ses masses d’eau sous-jacentes causent des remontées qui fournissent des nutriments aux plateaux de l’Arctique (McLaughlin et coll., 2004). Malgré le fait que les côtes arctiques sont couvertes d’une banquise côtière durant l’hiver, son océan intérieur et ses marges continentales demeurent libres de glace pour une grande partie de l’année (Barber et coll., 2010). La présence de polynies, qui sont des régions d’eau libre de glace créées par les systèmesde vents, les remontées d’eau plus chaude ou une combinaison des deux (Smith et coll., 1990) est une caractéristique importantedu GA. La polynie du GA fait partie du Système du Cap Bathurst, qui est également connecté au chenal de séparation circumpolaire qui s’étend sur toute la côte de l’Océan Arctique (Barber & Massom, 2007). Cette polynieestune zone sensible pour la formation de saumure, les échanges de chaleur et de gaz et les activités biologiques intenses, faisant du GA une région caractérisée par une productivité biologique relativement forte (Barber et coll., 2010, 2012; Fransson et coll., 2013). Durant l’été, la plupart des détroits deviennent libres de glace et sont fortement influencés par les apports riverains et les écoulements continentaux, provenant en particulier du Delta du Mackenzie (Carmack & MacDonald, 2002). Ceci s’accompagne également de variations saisonnières des profils de salinité et de température de la colonne d’eau dues au mouvement vertical des masses d’eau (Garneau et coll., 2008). 1.1.2 bassin Canada Le bassin Canada (BC) représentant la zone la plus grande et la plus profonde (ca. 4 km) de l’Arctique canadien. Il est délimité par le continent américain au sud, l’archipel canadien à l’est, la crête de Northwind à l’ouest et la crête de Lomonosov au nord (Figure 1.2). Le BC est également le dernier sous bassin de l’Océan Arctique qui reçoit un apport d’eau Atlantique (Aagard et coll., 1981). Il est unique par la présence d’un courant appelé le gyre de Beaufort (GB), principalement influencé par les vents, les oscillations atmosphériques, le forçage géostrophique et la formation et la fonte saisonnière de la glace. Le GB est reconnu pour être lié aux systèmes du climat global (Proshutinsky et coll., 2009; Morison et coll., 2012; Krishfield et coll., 2014). Le GB couvre la mer de Beaufort près du plateau

3 continental canadien jusqu’au nord du BC (Proshutinsky, 2002; Steele, 2004; McPhee, 2012). Sa circulation anticyclonique cause la formation d’un réservoir d’eau douce et éventuellement d’une coulée d’Ekman au centre du, menant à un renforcement de la stratification particulièrement durant l’été (Proshutinsky et coll., 2009; Yamamoto-Kawai et coll. 2009; McLaughlin & Carmack, 2010). On estime que ce réservoir contiendrait 17,000 à 22,410 km3 d’eau douce, correspondant approximativement au tiers du budget en eau douce de l’OA (Zhan & Zhang, 2001; Krishfield et coll., 2014). De plus, cela crée un gradient de densité horizontal empêchant la dispersion d’eaux continentales riches en nutriments vers le basin, augmentant en étendue latitudinale et en robustesse au cours de la dernière décennie (Nishino et coll., 2011a). Il a été observé que l’oscillation du GB s’inversait périodiquement, menant au relâchement de l’eau douce vers l’Atlantique Nord et contrôlant ainsi des anomalies de salinité mesurables dans la région (Proshutinsky, 2002). Ces fortes interactions air-mer ont des impacts significatifs sur la disponibilité des nutriments et conséquemment sur la productivité biologique de la région (Nishino et coll., 2008). Cela résulte en une productivité primaire se produisant dans les eaux plus profondes ou sous la surface au cours d’une grande partie de l’année dans le BC (McLaughlin & Carmack, 2010).

1.2 La succession d’eucaryotes microbiens dans un Arctique stratifié À plus basses latitudes, les cyanobactéries sont connues pour dominer les communautés phytoplanctoniques. Étant absentes de l’Océan Arctique, les communautés d’eucaryotes microbiens sont les composantes élémentaires des réseaux trophiques (Vincent, 2000; Lovejoy et coll., 2007). Selon les données de télédétection de la couleur de l’océan, la productivité primaire nette en milieu pélagique des microorganismes eucaryotes est estimés à 400-600 Tg C année-1 (voir Babin et coll., 2015) mais les calculs peuvent varier selon des modèles utilisés et n’incluent généralement pas les strates sous la surface (Ardyna et coll., 2013). Le carbone assimilé est transféré vers les niveaux trophiques supérieurs ou perdu par respiration via les innombrables interactions biologiques. Ces interactions sont en grande partie façonnées par la taille des organismes (Figure 1.3), c.-à-d. le petit phytoplancton est ingéré par le microzooplancton, lui-même qui est par la suite chassé par le mésozooplancton et ainsi de suite (Forest et coll., 2011; Glöckner et coll., 2012; LeFouest et coll., 2015). La plupart des transferts trophiques se produit dans la partie supérieure au-dessus de 200 m de l’OA où les ressources sont disponibles autant pour les fonctions photosynthétiques qu’hétérotrophes. De plus, les premiers 200 m sont influencés par la variabilité du couplage physico-météorologique, qui influence la dynamique chimique et biologique du système (McLaughlin & Carmack, 2010; Nishino et coll., 2011b). Mon travail de doctorat se focalise sur les communautés microbiennes ainsi que leurs

4 interactions dans les eaux supérieures (<200 m) de l’OA, comprenant principalement les eaux de mélange de surface(EPM) et les couches d’eau Pacifique.

Figure 1.3. Un schéma simplifié du réseau trophique marin arctique où la préférence de taille joue un rôle important. Notez la position des ciliés et des dinoflagellés en tant que prédateurs de petit phytoplancton et proies du plus large zooplancton (redessiné de Glöckner et coll., 2012, p. 27, original de Sandaa, 2009).

1.2.1 Banquise et eau de surface Les communautés microbiennes arctiques sont caractétisées par une forte succession saisonnière et une biogéographie relative aux conditions et ressources environnementales (e.g. Ardyna et coll., 2011; Comeau et coll., 2011; Marquardt et coll., 2016). À la fin de l’hiver et au début du printemps, les algues de glace dominantes sont les diatomées, particulièrement adaptées aux basses températures et à la faible luminosité. Elle profite des nutriments à la base de la banquise et constituent une portion significative de la productivité primaire annuelle (Syvertsen, 1991; Melnikov et coll., 2002; Arrigo et coll., 2011; Boetius et coll., 2013). Elles sont aussi une source nutritive importante pour le zooplancton migrant vers la surface tôt au cours du printemps (Darnis et coll., 2012) et pour les hétérotrophes flagellés (Laurion et coll., 1995). Toutefois, les algues de glaces jouent un rôle important seulement lorsqu’elles sont relâchées par la fonte de la banquise pour couler hors de la zone photique durant la transition des efflorescences printanières de plancton (Hill & Cota, 2005; Boetius et coll., 2013). Par la suite, les efflorescences de phytoplancton peuvent se former dans les eaux nouvellement

5 libres de glace le long de la lisière de labanquise en fonte, survenant dans les zones couvertes de glace saisonnière et pouvant persister durant quelques mois sur de grandes étendues (20-100 km2; Perette et coll., 2011). Le réchauffement de la surface, l’augmentation de la stratification et de l’intensité lumineuse ainsi que le renouvellement en nitrate durant les évènements de remontée périodiques favorisent la croissance d’espèces photosynthétiques plus grosses telles que les diatomées et les dinoflagellés. Ces cellules peuvent éventuellement couler plus rapidement dû à leur plus grande taille et leur agrégation, faisant d’eux un des principaux moteurs de la pompe biologique (Perette et coll., 2011; Kluijver et coll., 2013). Une partie des grosses cellules peut également être recyclée par broutage et par lyse, maintenant le carbone organique la productivité primaire dans la zone euphotique. (Pesant et coll., 1998). Les premières efflorescences du printemp épuisent éventuellement les nutriments des eaux de surface ce qui interrompt la prolifération des grosses espèces (Pommier et coll., 2009). La variabilité intra-annuelle dans les compositions des communautés a également été observée dû aux variations saisonnières (Wang et coll., 2005; Comeau et coll., 2011). Les eaux océaniques du large, telles que dans le BC, sont généralement moins productives que les mers continentales, telles que le golfe d’Amundsen où le mélange turbulent a plus d’impact sur le renouvellement de nutriments. Cette différence est particulièrement visible lors des saisons libres de glace dans l’Arctique (Tremblay et coll., 2008; Lee et coll., 2012). Vers l’été, la diminution des nutriments favorise des communautés de surface dominées par les pico- et nanoeucaryotes. Leur rapport surface-volume l’assimilation de nutriments à faible concetration (Li et coll., 2009). Dans l’ouest de l’Arctique canadien, les Mamiellales et autres petites cellules persistent sous la banquise durant la nuit polaire (Lovejoy et coll., 2007; Terrado et coll., 2008) et constituent une composante importante du phytoplancton de l’été à l’automne (Vidussi et coll., 2004). Les efflorescences sous la glace suivies des efflorescences printanières de phytoplancton survenant en lisière de banquise et dans les zones de glace marginales, épuisent les nutriments dans la PML. L’augmentation de l’apport en eau douce faible en nutriments au système durant le printemps et l’été, couplée au forçage atmosphérique, accentuent d’avantage l’approfondissement de la PML. Bien que la diminution de la banquise permette une plus grande pénétration et une plus grande disponibilité de lumière pour la photosynthèse, la limitation en nutriments à la surface inhibe la production par les microalgues et affecte ainsi les transferts d’énergie et de carbone (Nishinoet coll., 2011 b). Pendant ce temps, la forte stratification due à la différence de salinité empêche le transport des eaux riches en nutriments du Pacifique vers la surface. Le résultatestune surface faible en nitrates, mais occasionnellement forte en ammonium (Lee &Whitledge, 2004). La plupart des nutriments

(principalement NO3) approvisionnant la surface par diffusion, par tourbillon, par remontéeconvective

6 ou encore par mélange turbulent (Nishino et coll. 2011a; Carpenter & Timmermans, 2012), proviennent des eaux sous-jacentes du Pacifique.

1.2.2 L’Halocline du Pacifique et le Maximum de Sub-surface en Chlorophylle (MSC) Les eaux Pacifique (EP) avec un écart de température variant de 0˚C a-2˚C se retrouvent sous la PML et sont modifiées de façon saisonnière avec une composante d’été (EPÉ), ci-après le Pacific Summer Water (PSW) ayant une salinité de 30-32, et une d’hiver (EPH) ou ci-après le Pacific Winter Winter (PWW) avec une salinité de 33-34 (Steele, 2004). Les eaux du Pacifique entrent dans le bassin Canadapar trois voies différentes, mais la plus large passe par des tourbillons dansle canyon de Barrow au nord de la mer des Tchouktches faisant de cette région unezone sensible où l’on observe de forte productivité (Watanabe, 2011; Nishino et coll., 2011b). Les EP contiennent plus de nutriments (c.-à-d. N, P et Si) que les eaux de surface avec le maximum de nitrates observé au centre du PWW (S = 33.1; McLaughlin et coll., 2005). Ceci est dû à la formationd’efflorescences de phytoplancton, consommant le NO3dans PSW lorsque la lumière estivale est disponible. En revanche, durant l’hiver, les fortes concentrations de NO3 sont conservées dû à la faible luminosité, ainsi, le NO3est transporté avec succès jusqu’au basin Canada (Nishino et coll., 2011a). En parallèle, la décomposition et le recyclage de la matière organique produisent de fortes concentrations d’ammonium (NH4), transportées dans le Bassin canadien et la mer de Beaufort (Yamamoto-Kawai et coll. 2006, 2008; Nishino et coll. 2008). Les EP combinées circulent sous le PML à des profondeurs comprises entre 40 et 150 m durant l’été (Steele, 2004). Les nutriments, principalement le nitrate, atteignent des concentrations faibles ou indétectables dans la PML durant l’été. Ces conditions ne sont pas optimales pour la production primaire par lesplus grosses cellules (Li et coll., 2009) mais peuvent favoriser certaines espèces (Lee &Whitledge, 2004; Lovejoy et coll., 2007). Également, l’apport d’eau douce en surface dû à la fonte de la glace et aux rivières résulte en un approfondissement de la nitracline, la strate d’eau où la concentration en nitrates augmente abruptement dans la partie inférieure de la zone photique (Nishino et coll. 2011a). Les organismes photosynthétiques d’adaptent en suivant la nitracline, plus profonde et caractérisée par une faible lumière. Ceci mène à l’établissement d’un maximum de chlorophylle sous la surface (MSC), ci- après le « subsurface chlorophyll maxima » (SCM; Tremblay et coll. 2008; Pommier et coll., 2009; Martin et coll., 2012) dans la partie supérieure des eaux Pacifique. Dans le bassin Canadaet le golfe d’Amundsen, le SCM survient normalement dans la masse d’eau du PSW (McLaughlin et coll., 2004). À cette profondeur, les fortes concentrations en nutriments permettent aux cellules de croître mais la limitation en lumière nécessite une plus grande production de pigments photosynthétiques (Lovejoy et coll., 2007; Martin et coll., 2010). Ainsi, malgré le fait que le SCM coïncide avec une accumulation de

7 de biomasse photosynthétique, ceci n’implique pas forcément une plus forte productivité dans la colonne d’eau puisqu’il pourrait s’agir seulement d’une stratégie adaptative pour répondre à la faible disponibilité en lumière (Huisman et coll., 2006; Martin et coll., 2010). Puisque la lumière n’est pas limitante à la SCM peu profonde et que le NO3 est aussi abondant, elle peut supporter le développement de plus grosses cellules phytoplanctonique, telles que les diatomées et les dinoflagellés photosynthétiques-mixotrophes. En général, le NH4est une source d’azote énergétiquement préféréesous des conditions de faible luminosité puisque l’assimilation du NO3 implique une étape de réduction avant l’utilisation par les cellules (Kristiansen&Farbrot, 1991; Martin et coll., 2010, 2012;

Coupel et coll., 2012). Le recyclage de l’azote organique, incluant le NH4, fait également du SCM un site important de productivités secondaire et tertiaire, particulièrement sous des conditions de forte stratification (Weston, 2005; Tremblay et coll., 2008). De plus, le SCM peut agir comme un piège à nutrimentsvenant du bas, affaiblissant d’avantage la production primaire dans la partie supérieure de la zone euphotiqueau-dessus de l’halocline (Weston, 2005; Mundy et coll., 2009; Martin et coll., 2010). Martin et coll. (2012) ont observé une forte contribution du SCM à l’assimilation de l’azote et du carbone du printemps jusqu’à la fin de l’automne dans l’Arctique canadien. Toutefois, le SCM demeure le site où l’on observe la plus grande productivité dans le Bassin canadien durant l’été (Hill & Cota, 2005; Steiner et coll., 2016). Également des modéles ont permisd’estimer qu’elleétait responsable d’approximativement 70% de la production primaire annuelle dans le Golfe d’Amundsen (Martin et coll., 2013). Dans d’autres régions, par exemple dans la Mer du Nord, le SCM contribue à 10% de la production primaire annuelle (Leeuwen et coll., 2012) et jusqu’à 58% dans la partie centrale durant l’été (Weston, 2005). Alors que la dynamiquede succession du phytoplancton a été intensément étudiée dans l’Arctique, l’évolution temporelle et les réponses des autres composantes des communautés, telles que le microzooplancton arctique (c.-à-d. les ciliés et les dinoflagellés) demeurent méconnues (e.g. Levinsen&Neilsen, 2002; Sherr&Sherr, 1998, 2009; Sherr et coll., 2013; Nelson et coll., 2014).

1.3 Vue d’ensemble des ciliés et des dinoflagellés D’un point de vue moléculaire comme morphologique, les ciliés et les dinoflagellés constituentdes groupes d’eucaryotes microbiensles plus divers. Ils appartiennent au super-groupe Alveolata, qui incluent également l’Apicomplexa (Leander& Keeling, 2003; Adl et coll., 2005, 2012). Leurs classifications taxonomiques et généralesont subi de nombreuses révisions dans le passé. Ils sont également parmi plus anciens groupes de protistes étudiés et les espèces de microzooplancton les plus abondantes à de différentes latitudes, et leurs habitats incluent les régions côtières, les océans, les estuaires et même les lacs (Levinsen&Neilsen, 2002; Monti &Minocci, 2013). Certaines espèces des

8 deux taxa produisent des efflorescences massives, pouvant être nocives (Anderson et coll., 2002; Liu, 2012).

1.3.1 Ciliés Les ciliés sont un grand groupe de microzooplancton à cils avec plus de 8000 espèces décrites. Toutefois, comparativement aux dinoflagellés, les ciliés sont considérés comme monophylétiques (Lynn, 2008). La taxonomie de ce groupe est confortéepar les classifications basées sur les caractères morphologiques tels que les motifs des kinétides somatiques, les structures orales ou encore les modèles de division cellulaire (Lynn, 2012). Présentement, les ciliés comprennent 12 lignées acceptées incluant les Oligohymenophorea, Plagiopylea, Colpodea, Prostomatea, Nassophorea, Phyllopharyngea, Litostomatea, Armophorea, Spirotrichea, Cariacotrichea, Heterotrichea et Karyorelictea (Orsi et coll., 2012; Lynn, 2012). La majorité des espèces vivent librement et sont distribuées globalement, incluant les régions polaires (Finlay et coll., 1996; Foissner et al., 2008). En dépit des nombreuses études sur les ciliés et la richesse des données morphologiques archivées, plusieurs nouvelles espèces et taxon non-décrits sont signalés. Comparées à la microscopie, les analyses génétiques (c.-à-d. 18S ADNr) ont révélé d’avantage d’informations, telles que l’existence et l’importance de la biosphère des ciliés rares (Dunthorn et coll., 2014b). Par exemple, les phylotypes marins, autrefois considérés comme associés strictement aux colpodes d’eaux douces, ont été détectés en utilisant le séquençage à haut débit et la haute résolution des placements au sein d’arbres phylogénétiques (Dunthorn et coll., 2014a). Dans l’Arctique, les études ont principalement porté sur des taxon spécifiques, tels que les Tintinnida (Dolan et coll., 2016), à cause des difficultés de préservation et d’identification. Les analyses moléculaires ont toutefois indiqué que la plupart des séquences environnementales sont associées à Strombidium, les Strom A et les Strom B (Lovejoy et coll., 2006; Lovejoy & Potvin, 2011) ainsi qu’à d’autres taxon non-classifiés (Terrado et coll., 2009; Bachy et coll., 2012). La réconciliation entre la diversité morphologique et les séquences non- identifiées et récupérées directement depuis l’environnement reste un défi important (Santoferarra et coll., 2014; 2016).

1.3.2 Dinoflagellés Les dinoflagellés sont des protistes flagellésreprésentés par une grandegamme de taille (5-200 µm) mais généralement considérés comme des espèces nano- et microzooplanctoniques (Gomez et coll., 2005; Not et coll., 2012). La caractérisation du groupe est principalement basée sur la morphologie cellulaire générale, comprenant les motifs des plaques thécales et les positions des sillons

9 dans les thèques et des flagelles. Toutefois, pour plusieurs espèces, les informations de base demeurent manquantes ou débattues (voir Hoppenrath, 2016). Les dinoflagellés ont des morphologies, des compositions de pigments et des stades de vie variées et près de la moitié (2000) des espèces décrites sont considérées photosynthétiques (Not et coll., 2012). Selon les analyses moléculaires et morphologiques, neuf ordres sont reconnus au sein de la classe des Dinophyceae sous la superclasse des Dinoflagellées (Guiry & Guiry, 2016 sur AlgaeBase, accédéle 30 Décembre, 2016), incluant les Gonyaulacales, Peridiniales, Gymnodiniales, Suessiales, Prorocentrales, Diniphysiales, Blastodiniales, Phytodiniales et les Pyrocystales (Saldarriaga et al, 2004; Not et coll., 2012; Orr et coll., 2012). D’autres ordres, les Noctilucales, Thoracosphaeriales et Lophodiniales, demeurent incertains (Not et coll., 2012). Bien que la phylogénie déduite des séquences génétiques soit compatible avec certaines classifications basées sur la morphologie, le classement de plusieurs espèces demeure incohérent et souffre de paraphylie (Hoppenrath & Leander, 2007; Logares et coll., 2007; Orr et coll., 2012). De récentes observations de phylotranscriptomique suggèrent que les dinoflagellés thécates (portant des plaques) sont monophylétiques alors que les espèces athécates sont paraphylétiques (Janouskovec et coll., 2017). À ce jour, il existe entre 4000 et 4500 espèces de dinoflagellés affectées à 550 genres (Taylor et coll., 2008; Hoppenrath, 2016; Gomez, 2005, 2007). L’utilisation récente de marqueurs génétiques a toutefois révélé une plus grande diversité parmi les dinoflagellés, mais plusieurs n’ont pas de paires morphologiques correspondantes (Lin et coll., 2009; Stern et coll., 2010; Kohli et coll., 2014). Par exemple, en Arctique, alors que plusieurs séquences sont associées à des espèces connues ou décrites appartenant principalement au groupe de Gymnodiniales-Prorocentrales-Péridiniales (GPP; Saldarriaga et coll., 2004), une partie importante des séquences se regroupent avec d’autres séquences environnementales (Terrado et coll., 2009; Lovejoy & Potvin, 2011; Comeau et coll., 2011; Bachy et coll., 2012), suggérant la possibilité de découvrir une plus grande diversité de dinoflagellés. Ainsi, le nombre d’espèces actuel pourrait être bien plus important que ce qui est actuallement admis.

1.3.3 Les fonctions trophiques des ciliés et des dinoflagellés Les ciliés et les dinoflagellés (collectivement de microzooplancton) jouent un rôle important dans les réseaux trophiques puisqu’ils peuvent agir comme herbivores et proies en même temps. Le large spectre de leurs activités trophiques a des impacts significatifs sur le réseau alimentaire marin. En tant que prédateurs, ils consomment en moyenne 60-70% de la production primaire des océans et constituent le principal facteur de mortalité du phytoplancton (Landry &Calbet, 2004). Ils rivalisent ainsi avec le méso- et le macrozooplancton pour le phytoplancton (Sherr & Sherr, 2002). Par exemple, dans le Golfe d’Amundsen, le zooplancton, tel que les copépodes, consomme seulement 17-28% des producteurs primaires à la surface et 36-59% dans la totalité de la colonne d’eau (Darnis & Fortier, 2012). En

10 revanche, par exemple, dans les Grands Bancs de Terre-Neueve, le microzooplancton consomme 50- 70% du phytoplancton (Paranjape, 1990), et cette valeur peut atteindre 100% qu’observé dans un système de fjord (Calbet et coll., 2011). Dans l’ouest de l’Océan Arctique, on estime que les ciliés dénudés et les dinoflagellés gymnodinoïdes dominants consomment en moyenne jusque 22% de la production quotidienne de phytoplancton (Sherr & Sherr, 2009). Le microzooplancton peut former des efflorescences épisodiques suivant celles de leurs proies présumées, telles que les petites diatomées durant le printemps. Ainsi l’abondance de certains taxons plus ou moins forte en fonction des saisons, suggère l’existence d’une saisonnalité chez ces espèces (Paranjape, 1990). Le rôle potentiel du macrozooplancton en tant que est très important, en particulier dans les eaux profondes où les phototrophes sont rares dû à la faible luminosité. Plusieurs études indiquent que la plupart des organismes planctoniques vivant dans les eaux profondes sont détritivores mais la faible composition en nutriments des débris ne suffit pas aux besoins nutritifs des organismes plus gros. Ainsi, le microzooplancton pourrait répondre à cette demande (Stoecker & Capuzzo, 1990). Ils transfèrent le carbone fixé et l’énergie au zooplancton, qui en retour supporte les niveaux trophiques supérieurs. Les nauplius et les adultes des espèces de Calanusse nourrissent préférentiellement sur les ciliés et les dinoflagellés (Seuthe et coll., 2007). Bien que la plupart des espèces soient des proies pour le méso- et le zooplankton, des études suggèrent qu’ils ont également des proies potentielles pour certaines larves de poisson préfèrantdes proies plus large que 5 µm (Sherr & Sherr, 1988; Levinsen et coll., 1999). Par exemple, des lorica de certains ciliés tintinnidés ont été retrouvés dans les intestins de larves de poissons (Conover, 1982) mais d’autres espèces pourrait également être consommées sans pouvoir être identifiées a cause du manque de débris reconnaissables. Ciliés et dinoflagellés peuvent également se nourrir l’un sur l’autre. Par exemple, des études sur Dinophysis et Mesodinium dans le Golfe de Mexico ont montré des corrélations positives entre leurs abondances, suggérant que la présence de certains ciliés pourrait être un indicateur d’une efflorescence de certains dinoflagellés (Harred et Cambell, 2014). Certaines espèces de microzooplancton pourrait être mixotrophes et ainsi agir comme producteur et consommateur, en fonction de la disponibilité en nutriments et des conditions environnmentales (Jeong et coll., 2010; Johnson, 2011). Les connaissances sur les organismes catégorisés comme mixotrophes demeurent toutefois limitées, ce qui pourrait mener à une identification inappropriée ou erronée (Flynn et coll. 2012). Il est donc primordial de mieux connaitre la fonctionnalité intermédiaire des ciliés et des dinoflagellés.

1.4 Identification et classification des communautés microbiennes Les approches conventionnelles pour l’identification taxonomique de microzooplanctonreposant sur la microscopie dépendent fortement de la morphologie et nécessitent parfois la mise en culture de l’organisme pour une caractérisation approfondie (ex. Taylor et coll., 2008; Montagnes et coll., 2010; Medinger et coll., 2010; Santoferrara et coll., 2014). De plus, certaines espèces sont petites ou fragiles et s’abiment durant la préservation alors que d’autres peuvent présenter différents morphotypes avec différents stades de vie rendant la classification plus ardue (Lovejoy et coll., 1993; McManus & Katz, 2009). La plupart des espèces hétérotrophes ou mixotrophes sont difficiles à cultiver (Lin et coll., 2009; Dunthorn et coll., 2014b), limitant encore plus notre capacité à caractériser ces espèces. Par conséquent, plusieurs eucaryotes microbiens demeurent non-identifiés et inconnus. Comparé aux méthodes traditionnelles, l’avènement des approches moléculaires, telles que la réaction en chaîne par polymérase(RCP) combiné avec le clonage et le séquençage Sanger a permis d’établir plus précisément et plus rapidement la taxonomique des organismes issus de différentes communautés (Handelsman et coll., 1998). Les avancées technologiques récentes permettent maintenant de générer plusieurs milliers de séquences à l’aide du séquençage à haut-débit (SHD) à partir d’une seul RCP. Les produits séquencés, allant d’un gène ciblé (souvent les 18 S du gène de l’ARNr) jusqu’au séquençage global des gènes, fournissent une gamme d’orthologues ainsi qu’un profil de tous les gènes présents dans une communauté ciblée (Gilbert & Dupont, 2011). De plus, grâce aux avancées en bio-informatique, combiner et analyser plusieurs échantillons pour le séquençage (séquençagemultiplexe) provenant de différents sites d’échantillonnage est désormais possible (Kunin et coll., 2008). Le séquençage multiplexe et ciblé a permis la découverte et la caractérisation d’espèces importantes de phytoplancton noncultivé appartenant par exemple au groupe des picoprasinophytes (Cuvelier et coll. 2010; Lovejoy et coll. 2010), aux straménopiles marins non-cultivés (Massana et coll., 2014) et aux espèces de micro algues rares (Medinger et coll., 2010). L’utilisation de l’ARN au lieu de l’ADN permet également de fournir des informations sur la diversité taxonomique ainsi quel’activité potentielle des microorganismes. L’utilisation du SHD est bénéfique à la modélisation à grande échelle des réponses et des changements des communautés microbiennes. De plus, la plupart des séquences générées sont déposées dans des bases de données publiques (ex. NCBI, RDP, JGI, GOS, etc.), qui permettent aux scientifiques de comparer les résultats entre différents écosystèmes, environnements et localisations géographiques (Maidak et coll.,2000; Meyer et coll., 2008), même à l'échelleglobale (e.g. Tara Ocean, Bork et coll., 2015). La profondeur du séquençage et l’information générée à l’aide de cette technologie a permis de comprendre les communautés microbiennes à un niveau inaccessible auparavant avec d’autres

12 méthodes. Depuis que le SHD a été introduit, plusieurs études sur différents systèmes, particulièrement en milieu marin, ont été réalisées dans le but de comprendre le rôle global des communautés microbiennes (procaryotes et eucaryotes) dans les différents cycles biogéochimiques et les différents écosystèmes (voir Gilbert & Dupont, 2011; de Vargas et coll., 2015; Lima-Mendez et coll., 2015).

1.5 Méthodes statistiques

Les études écologiques impliquent généralement l’examin des interactions entre plusieurs espèces et variables environnementales, et il est rare qu'une seule variable de réponse soit suffisante pour expliquer un système écologique (Buttigieg & Ramette, 2014; Guide to StatisticalAnalysis in MicrobialEcology - GUSTA ME, https://sites.google.com/site/mb3gustame/). Les études en écologie microbienne, en particulier celles utilisant le séquençage à haut débit (SHD), traitent généralement de grandes quantités de données, ce qui peut constituer un défi tant dans l'exploration que dans l'analyse. Il est donc nécessaire d'utiliser des approches statistiques robustes mais aussi pratiques pour interpréter de manière efficace et précise ces grands ensembles de données. Les analyses statistiques multivariées sont utiles pour de telles applications en raison de leur capacité à faire face à la complexité des variables à réponse multiple (Ramette, 2007). Dans cette étude, un certain nombre de méthodes statistiques multivariées ont été utilisées pour explorer les données et tester les hypothèses, dont certaines sont décrits dans cette section.

1.5.1 Mesures de Dissimilarité et regroupement

Une façon d’explorer le regroupement des communautés en utilisant l’abondance d’espèces ou les facteurs environnementauxest de calculer une matrice de distance en fonction de leurs similarités, de leurs dissemblances, ou de leurs caractéristiques partager ou non. Par exemple, dans cette étude, nous avons utilisé l’indice dedissimilarité Bray-Curtis pour quantifier la distance entre les compositions des différents sites ou échantillons (Legendre & Legendre, 1998) sur la base des variables environnementales mesurées (par exemple, température, salinité, nutriments, profondeur, chlorophylle). Pour les données biologiques générées par le SHD, nous avons utilisé la distance ‘unweighted UniFrac’ (Lozupone & Knight, 2005), une approche qui calcule aussi la dissimilarité de composition comme Bray-Curtis mais incorpore des informations phylogénétiques. L’unweighted UniFrac utilise des informations sur la présence et l'absence d'unités taxonomiques opérationnelles (OTU) dans chacun des échantillons et intègre des distances phylogénétiques par paires entre des ensembles de taxons ou OTUs dans un arbre phylogénétique. Ces distances peuvent ensuite être

13 utilisées pardes approches de regroupement, telles que le regroupement hiérarchique ou les statistiques dimensionnelles non métriques (NMDS) pour visualiser le regroupement en fonction de la similarité ou de la dissemblance des échantillons. De manière alternative, la mise au point dimensionnelle métrique ou l'Analyse des Coordonnées Principales (PCoA) pourrait être utilisée pour projeter des dissemblances de la communauté (par exemple, UniFrac) dans un espace cartésien avec pour but d'expliquer la plus grande partie de la variance dans les ensembles de données originales, laquelle est représentée par le pourcentage de contribution de chaque axe (Ramette, 2007). Les principales valeurs de coordonnées (ou axes) peuvent ensuite être extraites et utilisées pour rechercher les facteurs potentiels du regroupement.

1.5.2 L'analyse canonique de la correspondance (ACC) L'analyse canonique de la correspondance (ACC) est une technique utile pour modéliser la réponse d'une espèce à un certain nombre de variations ou gradients environnementaux dans une ordination bidimensionnelle simplifiée, permettant d'associer une variable explicative qualitative ou associée à une réponse unimodale (abondance d'espèces ou nombre des OTUs). Au lieu d'inclure toute la variable dans la représentation finale ou l'ordination, seules les correspondances expliquées par les variables sont incluses dans le résultat final (Buttigieg & Ramette, 2014). Par exemple, l’ACC a été principalement utilisée dans des études visant à identifier des paramètres environnementaux spécifiques qui pourraient expliquer les profils ou patron d'abondance et de diversité d'un groupe d'espèces ou de taxons, manifestés par des relations linéaires unimodales (voir Ramette, 2007).Toutes les approches statistiques décrites et utilisées dans cette étude ont leurs propres limites, mais elles restent utiles pour explorer et simplifier de vastes ensembles de données, afin de mieux comprendre l’écologie des communautés microbiennes.

1.6 Les communautés microbiennes dans un Arctique en changement

Les changements globaux de température affectent directement les modèles climatiques locaux et l’océanographie de la région arctique (Morison et coll., 2012) et plusieurs études ont démontré que les communautés microbiennes sont influencées par ces changements (ex. Li et coll., 2009; Comeau et coll., 2011; Monier et coll., 2015; Thaler et coll., 2017). Par exemple, les manifestations les plus forte des effets du changement climatique sont la diminution de la couverture estivale de la banquise et l’augmentation du volume d’eau douce qui s’accumule dans l’OA à cause de l’augmentation des interactions océan-atmosphère (McLaughlin &Carmack, 2010; Morison et coll., 2012; Krishfield et

14 coll., 2014). Ce dernier phénomène cause un renforcement de la stratification, inhibe le transport de nutriments provenant des eaux plus profondes et approfondit la nutricline, laissant le MSC et les eaux de surface limitées en nutriments (McLaughlin & Carmack, 2009; Comeau et coll., 2011; Bergeron & Tremblay, 2014). Les nutriments devenant de plus en plus rares dans la zone euphotique, et l’OA devenant de plus en plus libre de glace, des modifications des communautés microbiennes sont anticipées (Lee et coll., 2012; Coupel et coll., 2012, 2014; Riebesell et coll., 2013; Thoisen et coll., 2015). Par exemple, le déclin général de la biomasse du gros phytoplancton corrèle avec les plus faibles concentrations de NO3 et l’adoucissement de la couche superficielle du bassin Canada (Li et coll., 2009; Nishino et coll., 2011b). L’incursion normale d’eau Pacifique combinée avec des températures de surface plus élevées peuvent également causer l’établissement de communautés typiques du Pacifique. Lovejoy et Potvin (2010) ont rapporté pour la première fois l’occurrence de plusieurs taxons dans la mer de Beaufort normalement associés aux eaux du Pacifique à plus basses latitudes. Le volume d’eau du Pacifique entrant dans la gyre de Beaufort a augmenté au cours des dernières années et contribué à la fonte de la glace pérenne, particulièrement en 2007 (Zhang et coll., 2008). Ainsi, les perturbations climatiques causent des modifications des conditions océaniques et peuvent avoir un impact significatif sur les régions biogéographiques de l’Arctique (Ardyna et coll. 2011; Terrado et coll., 2014), particulièrement dans le transfert d’énergie et le flux de carbone dans les écosystèmes arctiques (Li et coll., 2009). La modélisation et les analyses environnementales des écosystèmes aquatiques tempérés et tropicaux ont révélé qu’une stratification renforcée et une limitation sévère de nutriments, favorisant le petit phytoplancton. En même temps, une proéminence d’espèces mixotrophes, telles que les ciliés et les dinoflagellés, peut également se manifester (Skjodlborg et coll., 2003; O’Connor et coll., 2009; Lewandowska et coll., 2014; Jassey et coll., 2015; D’Alelio et coll., 2016). La capacité des espèces mixotrophes à exploiter l’énergie en étant photosynthétique et phagotrophe leur permet de bien proliférer dans des eaux claires mais oligotrophes et même de dépasser les espèces spécialistes (Tittel et coll., 2003). De plus, ils contribuent significativement au niveau trophique supérieur en broutant les petites cellules et en devenant les proies préférentielles du zooplancton (Calbet & Landry, 2004; Seuthe et coll., 2007; 2011; Lewandowska et coll., 2014; D’Alelio et coll., 2016). Dans l’Arctique, les ciliés et les dinoflagellés sont importants dans le cycle de la production primaire et doivent être pris en compte dans les modélisations des cycles de carbone et de nutriments (Levinsen & Nielsen, 2002). Ils semblent également être moins sujets à l’influence négative des évènements associés aux changements climatiques. Par exemple, des études au sein des régions côtières arctiques du Svalbard ont démontré que le microzooplancton demeure inaffecté par les changements de pCO2/pH (Aberle et coll., 2013). Il a également été signalé que les ciliés mixotrophes ont augmenté en abondance et en diversité lors des

épisodes records de fonte de la banquise en 2007 et en 2012 dans l’ouest de l’Océan Arctique (Comeau et coll., 2011; Jiang et coll., 2013). Toutefois, en dépit de leur rôle clé dans les réseaux trophiques marins, plusieurs aspects concernant la distribution et l’écologie potentielle de la grande majorité des ciliés et dinoflagellés non-cultivés demeurent peu compris (Calbet & Landry, 2004; Dolan et coll., 2013), particulièrement dans l’Arctique (Nelson et coll., 2014).

1.7 Objectifs et structure de la thèse

Les 3 articles scientifiques qui forment cette thèse traitent des variations dans le temps et dans l’espace de la diversité des ciliés et des dinoflagellés dans l’Océan Arctique.Le chapitre 2 étudie la variation temporelle des ciliés et des dinoflagellés. Le chapitre 3 se focalise sur dinoflagellés, et en particulier explore la diversité des picodinoflagellés. Le chapitre 4 présente une analyse des effets de la variabilité spatiale sur les communautés de dinoflagellés.

Chapitre 2 (Article 1): Seasonal and interannual changes in ciliate and dinoflagellate species assemblages in the Arctic Ocean (Amundsen Gulf, Beaufort Sea, Canada)

Publié: 2017, Frontiers in Marine Science, section Aquatic Microbiology, 4(16),doi: 10.3389/fmars.2017.00016

Cet article se penche principalement sur la dynamique temporelle et la succession des ciliés et dinoflagellés (microzooplankton) dans l’Arctique. Les changements des conditions physicochimiques à chaque saison sont accompagnés par une variabilité dansla disponibilité des ressources, qui sont sélectionnés par les groupes fonctionnels dominants. Bien que plusieurs études aient déjà été consacrées à la compréhension de l’évolution des communautés de phytoplancton en ce qui concerne les nutriments et la lumière, on dispose encore de peu de donnéesconcernant l’adaptation du microzooplancton, y compris sur les ciliés et les dinoflagellés pendant la saison hivernale. Des études récentes ont mis l’emphase sur les conséquences du changement climatique sur les communautés de phytoplanctons, mais peu d’études ont été réalisées sur l’importance du microzooplancton. Par conséquent, cette partie de l’étude vise à :

1. Raffiner notre connaissance de la variabilité saisonnière du microzooplancton avec une meilleure résolution taxonomique, et identifier les tendances qui sont compatibles avec la sélection environnementale dans le microzooplancton arctique;

2. Comprendre les différences concernant la distribution et la composition des communautés de microzooplanctons relative à la stratification, principalement l’observation des communnautés de surface et du SCM dans le golfe d’Amundsen ; 3. Étudier les tendances des communautés de microzooplancton estivales interannuelles et détecter les réponses potentielles à des événements de fonte des glaces.

Chapitre 3 (Article 2): Dinoflagellates in size fractionated samples from Canada Basin, Western Arctic Ocean

Article en préparation pour une soumission a un journal scientifique

Les dinophycées (core + syndiniale parasitaire) font partie des espèces les plus diverses de protistes. Toutefois, l’étendue exacte de leur diversité reste méconnue et débattue. Plus précisément, les études basées sur la microscopie ont seulement signalé des cellules libres de dinoflagellés dans la gamme de taille du microplancton (> 5 µm). Toutefois, de récentes études utilisant des échantillons fractionnés par filtrations (grand : > 3.0 μm, petit : 0.2–3.0 μm) couplé au séquençage Sanger ou au séquençage haut-débit (SHD) ont démontré l'abondance de séquences de dinoflagellés picoplanctonique (0.2–3 µm), communément désignés par le terme « picodinoflagellé » et principalement non-cultivés et non-classifiés. L’absence de preuves miscropcopiques de l’existence des picodinoflagellés, spécialement pour les « core » picodinoflagélles, consititue une énigme et pourrait potentiellement limiter l’usage et l’interprétation des données SHD fractionnées. L’existence de dinoflagelles picoplanctonique est envisageable sous la forme de kystes temporaires, de zoospores et de dinospores, mais celles-ci n’ont jamais été investiguées en utilisant la méthode SHD. Par conséquent, cette partie de l’étude vise à :

1. Découvrir comment les données fractionnées de SHD peuvent être utilisées pour vérifier l'existence potentielle des petits Dinophyceae libres (3 um) dans l'Océan Arctique en utilisant les rapports fractionnés de la 18S ADNr (>3.0 μm / 0.2-3.0 μm) et les statistiques de probabilité.

2. Déterminer les effets des variables spatiales et environnementales dans la composition des communautés de dinoflagellés.

3. Mieux comprendre les réponses potentielles des dinoflagellés mixotrophes à l'approfondissement de la nitracline découlant du changeant climatique.

Chapitre 4 (Article 3): Assembly rules governing dinoflagellates and their potential responses to nitracline deepening in the Western Arctic Ocean

Article en préparation pour une soumission a un journal scientifique

Le régime physico-chimique de l’océan Arctique est rudement affecté par les changements climatiques, particulièrement par l’augmentation de l’entrée d’eau douce et l’approfondissement de la nitracline. Les taxons mixotrophiques comme les dinoflagellés semble moins sensibles à ces changements que les autotrophes grâce à leur capacité à assimiler le carbone inorganique par voie photosynthétique, et à consommer le carbone organique disponible dans l’eau de mer par voie hétérotrophe. Toutefois, en Arctique, on connait mal la diversité des dinoflagellés, la structure de leur communautés, et resilence face aux perturbations des changements climatiques qui affectent présentement les propriétés physiques océaniques.

1. Déterminer les influences relatives des processus stochastiques et déterministes dans l'assemblage des dinoflagellés dans les trois masses d'eau des 200 m supérieurs de l'Océan Arctique à l'aide de modèles phylogénétiques aléatoire ; 2. Déterminer les effets des variabilités spatiales et environnementales dans la composition et les assemblages des communautés de dinoflagellés ; 3. Obtenir des informations sur les réponses potentielles des taxons mixtotrophes de dinoflagellés à l'approfondissement de la nitracline associé au changement climatique.

Chapter 2: Seasonal and interannual changes in ciliate and dinoflagellate species assemblages in the Arctic Ocean (Amundsen Gulf, Beaufort Sea, Canada)

Résumé

Les récentes études se sont focalisées sur les effets des changements climatiques pouvant entraîner des changements au sein des communautés phytoplanctoniques en Arctique. Cependant, malgré leur contribution importante à la mortalité du phytoplancton, les ciliés et les dinoflagellés ont reçu une attention moindre. Certaines espèces de dinoflagellés et ciliés peuvent aussi contribuer à la photosynthèse nette, ce qui suggère que la composition en espèces pourrait refléter la complexité du réseau trophique. Afin d'identifier de potentiels modèle d'occurrence saisonnière et annuels et de relier les espèces aux conditions environnementales, nous avons pour la première fois examiné le patron de variation saisonnière du microzooplancton, puis réalisé une analyse en profondeur de la variabilité interannuelle des espèces. Nous avons utilisé une méthode de séquençage d'amplicons haut débit afin d'identifier les ciliées et les dinoflagellés au niveau taxonomique le plus bas avec une base de données organisée d'ARNr 18S en Arctique. Les fragments d'ADN et d'ARN ont été générés à partir d'échantillons collectés dans l'Arctique Canadien entre novembre 2007 et juillet 2008. La proportion de fragments affiliés aux ciliés augmente en surface pendant l'été lorsque la salinité est au plus bas et que les plus petites proies du phytoplancton sont abondantes. A l'inverse, la proportion de fragments affiliés aux espèces de dinoflagelléschloroplastidiques augmentent dans la chlorophylle maximum sous-superficielle où la concentration en nutriments inorganique est la plus élevée. En comparant les communautés collectées durant l'été de 2003 à 2010, nous avons trouvé que les changements de composition des communautés microzooplanctoniques étaient associés au record de minimum de glace de 2007. Spécifiquement, les fragments associés aux espèces de petits prédateurs comme Laboea, Monodinium et Strombodinium ainsi que plusieurs ciliés non classifiés augmentent après l'été 2007, alors que les autres dinoflagellés habituellement dominants durant l'été diminuent. La capacité des organismes d'exploiter de plus petites proies est un mode de nutrition prédit pour devenir dominant dans le futur en Arctique et pourrait être un avantage pour ces petits ciliés dans l'optique du changement climatique.

Abstract

Recent studies have focused on how climate change could drive changes in phytoplankton communities in the Arctic. In contrast, ciliates and dinoflagellates that can contribute substantially to the mortality of phytoplankton have received less attention. Some dinoflagellate and ciliate species can also contribute to net photosynthesis, which suggests that species composition could reflect food web complexity. To identify potential seasonal and annual species occurrence patterns and to link species with environmental conditions, we first examined the seasonal pattern of microzooplankton and then performed an in-depth analysis of interannual species variability. We used high-throughput amplicon sequencing to identify ciliates and dinoflagellates to the lowest taxonomic level using a curated Arctic 18S rRNA gene database. DNA- and RNA-derived reads were generated from samples collected from the Canadian Arctic from November 2007 to July 2008. The proportion of ciliate reads increased in the surface towards summer, when salinity was lower and smaller phytoplankton prey were abundant, while chloroplastidic dinoflagellate species increased at the subsurface chlorophyll maxima (SCM), where inorganic nutrient concentrations were higher. Comparing communities collected in summer and fall from 2003-2010, found that microzooplankton community composition change was associated with the record ice minimum in the summer of 2007. Specifically, reads from smaller predatory species like Laboea, Monodinium and Strombidium and several unclassified ciliates increased in the summer after 2007, while the other usually summer-dominant dinoflagellate taxa decreased. The ability to exploit smaller prey, which are predicted to dominate the future Arctic, could be an advantage for these smaller ciliates in the wake of the changing climate.

2.1 Introduction

The climate driven changes now occurring across the Arctic are expected to affect marine phytoplankton communities in many ways, for example, by limiting nutrient supply, lowering surface salinities and potentially by ocean acidification (Coupel et al., 2012, 2014; Riebesell et al., 2013; Thoisen et al., 2015). The loss of multi-year ice has implications for the sea-ice communities (Comeau et al., 2013), and consequently the Arctic food chain (Soreide et al., 2010). But less intuitively, the loss of multi-year ice can have an impact on the pelagic communities, for example by lengthening the open water season and influencing the depth of the upper mixed layer in this salinity stratified ocean. In the Canada Basin and elsewhere, changes in the timing and character of phytoplankton production have already been reported (Li et al., 2009; Ardyna et al., 2014). Less attention has been given to the microzooplankton (ciliates and dinoflagellates) that consume from 22-75% of the daily phytoplankton production in the Arctic and other cold water oceans (Paranjape, 1990; Landry and Calbet, 2004; Sherr and Sherr, 2009; Sherr et al., 2013). Ciliates and dinoflagellates are also influenced by top-down processes, where changes in macrozooplankton species are linked to the successional progression of the microzooplankton communities (Riisgaard et al., 2014). Microzooplankton species, especially dinoflagellates, use a variety of techniques to capture prey (e.g. Nielsen and Kiørboe, 2015), with some ciliates being fastidious in terms of nutritional requirements (e.g. Johnson, 2011), and others having a narrow size specificity range (e.g. Park and Kim, 2010). Because of such selectivity, microzooplankton could have a top-down effect on phytoplankton and protist community composition (Irigoien, 2005), with implications for microbial food webs, nutrient and carbon cycles, as well as zooplankton nutrition and higher food webs. In addition to their major role as phagotrophs, many microzooplankton are mixotrophic and able to carry out photosynthesis. In particular, about half of the described species of dinoflagellates are photosynthetic with integrated chloroplasts (Saldarriaga et al., 2001; Hansen, 2011). Other dinoflagellates, along with some ciliates, host photosynthetic symbionts, or are kleptoplasidic (Esteban et al., 2010). The range of trophic roles among microzooplankton suggests that the microzooplankton species composition could be used as indicators of differences in microbial food webs. Such an approach would require testing for associations among particular taxa and the conditions where they occur. However, there have been few studies focusing on occurrence patterns of ciliates and dinoflagellates in the Arctic (Nelson et al., 2014). The use of 18S rRNA gene surveys has highlighted the prevalence of dinoflagellates and ciliates in the Arctic, but most studies have been restricted to summer and early fall (Comeau et al., 2011). Dinoflagellates and ciliates persist over winter darkness in the Arctic(Terrado et al., 2011; Marquardt et al., 2016), but their seasonal progression is poorly understood.To resolve temporal patterns at finer

21 taxonomic resolution, and to test for trends consistent with environmental selection among Arctic microzooplankton, we investigated samples collected from November 2007 to July 2008 during the International Polar Year Circumpolar Flaw Lead Study (IPY-CFL; Barber et al., 2010, 2012). The IPY samples from both surface waters and the top of the Pacific halocline, where a subsurface chlorophyll maximum (SCM) usually develops (McLaughlin and Carmack, 2010; Monier et al., 2015). We hypothesized that the strong seasonality and stratification in Amundsen Gulf (Beaufort Sea) would be sufficient to lead to niche partitioning and select different species assemblages at the two depths and over time. Both DNA and RNA (converted to cDNA) were used as templates for targeted amplicon sequencing to compare the community with potential for active protein synthesis (from RNA) and communities from DNA, which could include free DNA and cells in state of dormancy or encystment (Jones and Lennon, 2010; Hunt et al., 2013). Specifically, we applied high throughput sequencing (HTS) targeting the V4 region of 18S rRNA (referred to as rRNA) and the 18S rRNA gene (referred to as rDNA) to determine species level composition in the communities. To test for interannual variability versus trends over time, we re-analyzed the 8 years of ciliate and dinoflagellate amplicon data from Amundsen Gulf reported in Comeau et al. (2011). For both the seasonal and interannual studies, we classified the microzooplankton reads using an improved reference database (Lovejoy et al., 2015) and constructed robust phylogenies. We applied multivariate statistics to link changes in community structure and composition with prevalent environmental drivers, to provide insights on how microzooplankton communities respond to seasonal changes and inter-annual variability.

2.2 Materials and Methods

2.2.1 Sample collection, extraction and sequencing

Seasonal samples were collected onboard the Research Icebreaker CCGS Amundsen every 2 to 4 weeks from November 2007 to July 2008 from Amundsen Gulf and adjacent bays (Figure 2.1A). We targeted the polar mixed layer (PML) near the surface (5-20 m), and the top of a halocline (22 to 80 m), which separates the PML from Pacific Summer Water (PSW). In late spring the Beaufort Sea SCM typically forms at this halocline (Bergeron and Tremblay, 2014). During winter and early spring, the halocline sample depth was identified by salinity and temperature profiles with salinities of 31-32, and from May to July confirmed by the deeper chlorophyll fluorescence peak.Samples used to investigate interannual variability of ciliate and dinoflagellate communities in the SCM were collected during multiple missions to the Amundsen Gulf and Beaufort Sea between 2003 and 2010, aboard the CCGS Amundsen (2003-2006, 2009-2010), CCGS Louis S. St. Laurent (2007) and CCGS Sir Wilfred Laurier (2008). Details of these missions have been reported elsewhere (Barber et al., 2010, 2012; Comeau et al., 2011; Terrado et al., 2011).

Except for 2010, all samples were preserved using the same methods, and following extraction, all were amplified and sequenced using the same protocols. Briefly, water samples for DNA (ca. 6 L) and RNA (ca. 4 L) were collected into cleaned acid rinsed carboys from 12-L (CCGS Amundsen) or 10-L (CCGS Louis St Laurent, CCGS Sir Wilfred Laurier) Niskin-type bottles mounted on a Rosette system equipped with a conductivity, temperature and depth (CTD; Sea-Bird Electronics Inc.,), chlorophyll fluorescence (Seapoint Sensors Inc.) and relative nitrate from the ultraviolet spectrophotometer (ISUS, Satlantic) profilers. Target depths were identified by fluorescence, salinity, temperature and relative nitrate from the downcast and water collected on the upcast. All samples were filtered and preserved within 2 hours of collection. Following pre-filtration through a 50-µm mesh to remove macrozooplankton, samples were sequentially filtered using a peristaltic pump (Cole-Parmer) onto a 3-µm pore-size 47-mm polycarbonate filter (PC, AMD Manufacturing), and 0.2 µm pore-size Sterivex™ unit (Millipore) for the DNA and a 47-mm 0.2 µm pore-size PC filter for the RNA. All PC filters were placed in 2-mL cryovials. Prior to 2010, DNA was preserved by adding 1.8 mL of lysis buffer to the Sterivex units (0.2-3 µm fraction) and cryovials (3-50 µm fraction), and RNA was preserved in RLT buffer (Qiagen) as in Terrado et al. (2011) and Thaler and Lovejoy (2015). Samples were either frozen immediately at -80 °C or placed in liquid nitrogen after adding buffer. Samples collected in 2010 were preserved using RNALater (Ambio). After 30-60 min in buffer, the filters were flash frozen in liquid nitrogen and then kept at -80 °C until processing in the laboratory. We note that preliminary tests on paired samples from both marine and freshwater environments showed no significant differences in DNA and RNA recovery or community compositions between the protocols (Lovejoy pers. comm.). For samples collected before 2010, DNA was extracted from the filters as in Terrado et al. (2011) using a salt based method (Aljanabi and Martinez, 1997) while RNA was extracted using the RNAEasy® Micro Kit (Qiagen). The sample collected in 2010 was extracted using the AllPrep DNA/RNA Mini Kit (Qiagen) with DNA and RNA from the same filters. Conversion of RNA to cDNA was carried out using the High Capacity Reverse Transcription Kit (Applied Biosystems). The V4 region of 18S rRNA in both DNA and cDNA samples was amplified using eukaryotic specific forward primers E572F with the Roche A adapter and reverse primer E1009R as described by Comeau et al. (2011). For the seasonal samples, large and small fractions were pooled after normalization based on relative chlorophyll concentrations (Supplementary Table S2.1). Only the 0.2-3 µm fraction was used in the interannual study. Separate sequencing runs were carried out for the seasonal and interannual studies. For each run, equimolar concentrations of amplicons from each sample were pooled for multiplex sequencing using the GS FLX Titanium Roche 454 platform (Roche/454 Life Sciences,) at

IBIS/Université Laval Plateforme d’Analyses Génomiques (Quebec, QC, Canada). All reads have been deposited in the Short Read Archive of NCBI under accession codes: PRJNA283142 (IPY-CFL) and SRA029114 (Amundsen Gulf Time Series, Comeau et al., 2011).

2.2.2 Post-sequence data processing and taxonomic classification

The seasonal data reads were quality filtered using the Quantitative Insights into Microbial Ecology program (QIIME; Caporaso et al., 2010b). Short sequences and primers were identified in mothur and removed (Schloss et al., 2009). Data were then de-noised following Reeder and Knight (2010). Chimeric sequences were identified in UCHIME (Edgar et al., 2011) and removed. Interannual data was processed using mothur (Schloss et al., 2009) as described in Comeau et al. (2011). The two data sets were then processed for OTU picking at 98% similarity utilizing USEARCH (Edgar, 2010) against the Silva Reference Database v.102 (Pruesse et al., 2007). All OTU representative reads were aligned in PyNast (Caporaso et al., 2010a), manually curated in BioEdit v.7.2.5 (Hall, 1999) and used to construct a phylogenetic tree (FastTree; Price et al., 2010). Assignment of taxonomic identity was performed in mothur with a 0.8 confidence threshold against the Northern Reference Database v.1.0 (Lovejoy et al., 2015), which follows the Silva taxonomy but includes high-quality longer environmental sequences from the Arctic and North Atlantic (Charvet et al., 2012; Dasilva et al., 2014; Terrado et al., 2009, 2011;Monier et al., 2013). However, using this database, around 30-45% of the reads were still not classified beyond “Other Dinoflagellates” or “Other Ciliates”. To further increase taxonomic resolution, dinoflagellate sequences from Saldarriaga et al. (2001, 2004), Logares et al. (2007) and Potvin et al. (2013), and ciliate sequences from Dunthorn et. al. (2014), along with Evolutionary Placement Algorithm –Randomized Axelerated Maximum Lieklihood (EPA-RAxML) classified abundant Arctic-derived HTS reads were added to the v.1.0 of Northern Reference Database (v.1.1, Lovejoy et al. 2016). Taxonomic classifications of the dinoflagellate sequences used as references were verified in the AlgaeBase.org (Guiry and Guiry, 2015) and by literature searches, while ciliate taxonomic classification followed Lynn (2012). Using the v.1.1 of the database, taxonomy assignment was re-performed to generate the final OTU matrix (OTU reads per sample) with improved results (i.e. “Other Dinoflagellates” and “Other Ciliates” went down to 5-10%). OTUs matching metazoa, fungi, higher plants and reads that could not be classified to any taxonomic level below Eukaryota were removed and not analyzed further. This cleaned data set was considered as the total microbial eukaryotic community. To facilitate diversity comparisons, the OTU matrix was then rarefied, based on the sample with the fewest reads in the separate data sets, resulting in 4953 reads per sample for the seasonal study, and 7041 reads per sample for the interannual comparison.

To investigate the OTUs that contributed to the changes in July 2008 for the interannual data, (see Results), the remaining abundant “Other Ciliate” OTUs were searched for in NCBI and their most similar nearly full length 18S rRNA gene sequences were used to construct reference trees as in Thaler and Lovejoy (2015). Then, the abundant but unclassified ciliate OTUs were then mapped back onto the reference trees using EPA-RAxML v.8 (Stamatakis, 2014). We then inferred their probable taxonomic identity based on the nearest or most similar annotated reference sequences.

2.2.3 Statistical and diversity analyses

Statistical tests were carried out in the R environment v 3.0 (R Development Core Team, 2008). Spearman’s Rank Correlation (rho) in the Vegan package (Dixon and Dixon, 2003) was used to test correlations between taxa and environmental parameters, and results wereplotted in igraph(Csardi and Nepusz, 2006). Only correlations with Spearman’s rho >0.3 and significant at p < 0.001 were retained (Barberán et al., 2012). Species-variable relationships were then visualized by network analysis as target-source plots in Cytoscape 3.0 (Shannon et al., 2003). Analysis of variance (ANOVA) was applied to determine significant differences between samples and linear regression tests were used to infer relationships, both analysis were carried out using PAST v3.0 (Hammer et al., 2001). Alpha diversity (Chao1 index) of the ciliate and dinoflagellate communities in each seasonal sample was estimated as implemented in QIIME (Chao and Shen, 2003; Caporaso et al., 2010b). Since the samples used for the interannual part of the study were collected in different months (Jul to Nov) over the 8 years, we tested for variability associated with the month of collection and found no significant differences in species richness of OTUs in ciliate or dinoflagellate abundances between summer and autumn (Supplementary Figure S2.1). Phylogenetic unweighted UniFrac dissimilarities (beta diversity) among the taxa were computed separately for seasonal and interannual datasets by the jackknife method and generalized UniFrac procedure (Lozupone and Knight, 2005; Chen et al., 2012). All dinoflagellate and ciliate OTUs in the seasonal study were included in the beta diversity measure. To avoid potential rarefication artifacts (Ramette, 2007) in the interannual data, rare OTUs that were defined as < 1% of the total ciliate or dinoflagellate reads, were not included in the beta diversity and remaining analysis. The pairwise dissimilarity matrices were used to generate dendrograms using UPGMA (Price et al., 2010). The weighted contributions of seasons, depths (surface or SCM–halocline) and templates (RNA or DNA) to community structuring were computed using ADONIS (Fierer et al., 2012). The checkerboard score (C-score) was used to test for non-random co-occurrence patterns under the null hypothesis parameter using the oecosimu function in the R library bipartite with 1000 simulations (Dormann et al., 2008).

2.3 Results

2.3.1 Physico-chemical regimes of the Amundsen Gulf

For the seasonal study (Figure 2.1A), surface PML waters remained cold (-1.7 to 2 ˚C) from mid-January to mid-May and increased towards summer reaching 8 ˚C in July, when the region was ice-free. Phosphate concentration changed little over time with concentrations from 0.6 to 1.6 mg m-3. Over the 9 months, nitrate concentrations were significantly greater in the halocline compared to the surface (t-test, p<0.001), with greatest concentrations from February to April (Table 2.1). Nitrate concentrations started to decrease beginning in early April in the surface and in May at the halocline. The chlorophyll a (Chl a) concentrations, as estimated from in situ fluorescence, were negligible until 9 April and were greater in the surface waters compared to halocline through 19 May. Concentrations were below 1 mg m-3 in both the surface and halocline until July when a strong SCM was apparent at the halocline (Table 2.1). The physical characteristics of the time series stations (Figure 2.1B) have been reported elsewhere (Comeau et al., 2011). Briefly, over the 8 years, nutrients in the SCM ranged from 0.09 to

-3 -3 8.22 mmol m for NO3¯, 0.46 to 1.29 mmol m for PO4 and the highest Chl a (total) was detected on September 2005 reaching 2.14 mg m-3 (Supplementary Table S2.2). After binning the data from before and after 2007, total summer-fall Chl a values went from 0.43 mg m-3 to 0.31 mg m-3 in the SCM.

2.3.2 Community clustering based on 18S rRNA and 18S rRNA genes

The seasonal rDNA and rRNA OTU communities separated into distinct surface and halocline clusters, with clear separation between the rDNA and rRNA communities within the depth defined clusters (Figure 2.2). Within the template clusters samples separated by season as Autumn-Winter (Nov-April) and Spring-Summer (May-July) categories, except for one rRNA surface sample collected on June 14, which clustered with the winter rRNA samples (Supplementary Figure S2.2). The similarity was mainly driven by the presence of several OTUs that occurred in the rRNA surface community on Jan 3 and June 14 but were absent in the other summer samples (Supplementary Table S2.3). We note that, because of ship time constraints, the June 14 samples were collected from Darnley Bay, which is an extension of Amundsen Gulf. For all samples, the ADONIS test also showed individual contributions to clustering of around 14% (adj. R2=0.148 and 0.146, p<0.001) grouping by depth category and sample type, while season contributed the least to the variance observed (adj. R2=0.073, p<0.001). The C-score test further supported the non-random distribution of the microzooplankton communities (p<0.001).

2.3.3 Seasonal succession and distribution

Since DNA can be retrieved from both living and senescent cells, and free DNA can persist in the environment (Torti et al., 2015), the RNA-sourced samples were used to investigate potential influence by local conditions. Based on rRNA reads, ciliates and dinoflagellates were found in all samples (Figure 2.3), with greater proportions during summer (ca. 40%) and lowest in spring (ca. 5%) when larger photosynthetic taxa (mostly diatoms) dominated the planktonic community. The combined mean seasonal alpha diversity (Chao 1 index) of both ciliates and dinoflagellates based on rDNA, decreased from 125 in winter to 83 after spring (Supplementary Figure S2.3). In terms of total relative abundance, ciliate proportions were similar in both depths during winter (Figure 2.3A, 2.3C) while dinoflagellates were better represented in the surface (Figure 2.3B, 2.3D). Ciliate read relative abundance increased in the surface starting early June (Figure 2.3A), while dinoflagellates increased in the halocline (Figure 2.3D). Significant associations (Spearman’s rho > 0.3, p < 0.001) among taxa and environmental variables were detected in the network analysis (Figure 2.4). The associations were consistent with the seasonal succession and depth (water mass) seen in the UniFrac clustering (Figure 2.2). For example, surface-associated ciliates: Strombidium, Pseudotontonia, Askenasia and dinoflagellates Gymnodinium-like, Adenoides-like, Gyrodinium c.f. gutrula and Scripsiella (Figure 2.4, taxa 1–7) were associated with higher DO, which was a characteristic of the surface PML. In addition, abundances of Amphidoma, “Arctic Clade 1”, Blastodinium, Monodinium and Novostrombidium (Figure 2.4, taxa 8– 13) were higher when Chl a concentrations were greater, which was associated with longer day lengths and warmer water, suggesting characteristic surface summer taxa. Tintinnidium, Parauronema, Oligohymenophorea, Gyrodinium helveticum and Amphidinium-like taxa (Figure 2.4, taxa 15–18) were more associated with the deeper, more saline and nutrient-richer SCM or halocline layer. Although Laboea (Figure 2.4, taxon 14) was also strongly correlated with the SCM/halocline conditions, it was most abundant when the days were longer.In contrast, Woloszynskia, Gymnodinium sp. and “Other Gyrodinium” (Figure 2.4, taxa 19, 21, 22)were more abundant in the halocline during winter when ice cover was more extensive. These species, correlated with specific environmental variables, provide evidence for succession and seasonality within the two water masses. Binned OTUs of poorly defined groups that were not classified beyond their respective ranks, including “Other” Dinoflagellates, Ciliophora, Spirotrichea, Litostomatea and Strombidium (Figure 2.4 taxa 23–30), did not correlate with any of the environmental variables tested.

2.3.4 Interannual variability

Ciliate communities were consistently dominated by “Other Spirotrichea” with annual mean abundance of 9.4% + 2% relative to the total microbial eukaryotic reads. “Other ciliates” (1.7% + 1%), Strombidium sp., Pelagostrombidium sp., “Other Oligohymenophorea”, “Other Litostomatea”, Askenasia sp. and Monodinium sp. abundances ranged from 0.01% to 1.5% of total microbial eukaryote relative abundance (Supplementary Figure S2.4A). The most abundant dinoflagellate OTUs were from Gymnodiniales, Blastodinium-like, “Arctic clade 1”, Gyrodinium, Adenoides-like and unclassified Dinoflagellates, each accounting for 1% to 6% of the total microbial eukaryotic reads (Supplementary Figure S2.4B). Here, using our refined taxonomic database we found that microzooplankton community composition also changed, especially in summer 2008. Specifically, based on abundant OTUs (>1%), the July 2008 sample showed the greatest dissimilarity from other communities (Figure 2.5A). This dissimilarity was driven by the July 2008 increase in both abundance and OTU richness of Strombidium, Laboea, Monodinium and “Other Ciliates” (Figure 2.5B), with a significant correlation with decreasing salinity (r2 = 0.82, p < 0.01). After 2008, the relative abundances of these groups returned to previous levels by 2010. Their relative abundances were also inversely correlated (multiple 2 2 2 regression) with SiO3 (r =0.29, p<0.001), salinity (r =0.22, p=0.1), and nitrate (r =0.18, p=0.2). Phylogenetic placement of the “Other Ciliate” reads further revealed that these were in the Class Spirotrichea, and included sub-classes Oligotrichia and Choreotrichia. Most of these Spirotrichea clustered with environmental sequences, which we refer to as Environmental Clades 1, 2, 3, 4, 5 and 6 (Supplementary Figure S2.5). Aggregated read counts of Oligotrichia, and Environmental Clades 1, 4, 5 and 6 showed significant increase after 2007 (Krustal Wallis, p< 0.01), while Choreotrichia, and Environmental Clades 2 and 3 decreased.

2.4 Discussion

Temporal and distributional studies of ciliates and dinoflagellates are usually restricted to morphologically recognizable species (Levinsen and Nielsen, 2002; Montagnes, 1996). The use of HTS coupled with the improved reference database provided higher resolution of potential species that suggests high diversity compared to previous reports from the Arctic. Overall, we discriminated around 30 taxa per 100 reads, with the total number of reads always higher for dinoflagellates than for ciliates (Supplementary Tables S2.4 and S2.5). This translated into 251 dinoflagellates and 141 ciliate species based on OTUs defined as 98% similar as estimated using Chao 1. These species estimates are considerably higher than the 55 ciliate species reported in the Western Canada Basin (Jiang et al., 2013) and the < 20 dinoflagellates species reported in the Beaufort Sea (Okolodkov and Dodge, 1996) based

28 on microscopy. The high proportion of environmental clades here suggests that there are likely ciliate and dinoflagellate species confined to the Arctic, as is the case for diatoms (Luddington et al., 2016) and prasinophytes (Lovejoy et al., 2007).

2.4.1 Interpretation of DNA and RNA-derived abundances and diversity

Eukaryotic microbial communities from rDNA have been used to infer water mass history (Hamilton et al., 2008; Monier et al., 2013; Zhang et al., 2014), whereas taxa from rRNA are thought to more closely indicate active representatives of the community (Campbell and Kirchman, 2013;Hunt et al., 2013). Here, we found that for both DNA and RNA sourced samples, the surface communities always clustered together separate from the SCM communities. Earlier microscopy (Lovejoy et al., 1993; Okolodkov and Dodge, 1996; Jiang et al., 2013) and clone library studies using longer 18S rRNA sequences (Bachy et al., 2011; Lovejoy and Potvin, 2011) have also highlighted the differences between SCM and surface communities over the Arctic. Within the two water masses, communities separated by template suggests a pool of historic DNA, which could include dormant or less active stages, e.g. cysts (Verni and Rosati, 2011; Bravo and Figueroa, 2014), advected non-active species (Lovejoy and Potvin, 2011), preserved free DNA, or non-living material in marine snow (Nielsen et al., 2007; Boere et al., 2011). All of these sources can be transported by currents and persist over long distances (Heiskanen, 1993; Brocks and Banfield, 2009). Although RNA-derived reads cannot be directly used to estimate biomass (Medinger et al., 2010; Blazewicz et al., 2013), the patterns of relative read abundance were consistent with specific environmental drivers operating on particular species or clades, with communities separating by season. One exception was the June 14 rRNA surface sample from Darnley Bay, which based on OTU composition clustered with the January 3 rRNA sample. The placement of the June sample in an otherwise winter cluster was consistent with deeper unmodified Pacific Winter Water appearing in Darnley Bay, which was evident in the physico-chemical sample clustering (Supplementary Figure S2.2). Upwelling and on-shelf transport of deeper Pacific Water from Amundsen Gulf has been reported previously (Garneau et al., 2006; Paulic et al., 2012). Unfortunately, CTD transects were not taken along the shelf break in June and the source of this water remains speculative. The biological clustering was driven by 15 OTUs belonging to “Arctic Clade 1”, Gyrodinium, Gymnodinium, “Other Dinoflagellates, Askenasia, Tintinndinium, “Other Spirotrichea” and “Other Ciliates”.

2.4.2 Environmental influences on microzooplankton

While the dependence on phytoplankton prey can generate annual bimodal or unimodal microzooplankton bloom patterns in some regions, with temperature having an influence (Levinsen and Nielsen, 2002; Godhantaraman and Uye, 2003), there is little understanding of species composition over annual cycles. Here we found that ciliate and dinoflagellate taxa showed distinct seasonal patterns that could be partially explained by salinity, temperature and nutrient changes from winter to summer (Figure 2.4). Specifically, the proportion of ciliate reads were generally higher under ice free conditions, when the surface waters were less saline and at the end of the spring bloom when nitrate concentrations were drawn down byphotosynthetic activity (Forest et al., 2010; Tremblay et al., 2011). Although many ciliates have a wide range of salinity tolerance (Montagnes, 1996), this has not been investigated for Arctic species. In contrast dinoflagellate proportions were higher in the surface in winter, but more abundant in the SCM towards summer. Dinoflagellates can produce the osmolyte dimethylsulfoniopropionate (DMSP), which allow cells to regulate internal homeostasis (see review by Stefels, 2000), which may provide an advantage relative to ciliates under conditions of slightly higher salinity. During late spring and towards summer with increasing light availability, the putative plastid- containing ciliates and dinoflagellates increased proportionally in the rRNA OTUs, suggesting that photosynthesis may have provided additional energy for photosynthetic or mixotrophic taxa. In Amundsen Gulf, in summer, the higher sun angle and longer day-length mean that more light reaches the nutrient-rich halocline leading to the establishment of the SCM. We found that ciliate and dinoflagellate reads made up on average ca. 60% of the total microbial eukaryotic reads in early summer in the SCM. In contrast, a microscopy study in 2005-2006 reported that photosynthetic and non- photosynthetic dinoflagellates constitute only around 3-6% of the total mean phytoplankton cell counts in the SCM (Martin et al., 2010). The discrepancy is likely related to cell breakage during collection, misidentification with smaller dinoflagellates grouped with other flagellates, and loss during storage. Ciliates, can also be fragile and tend to be either ignored or under sampled using standard Lugol’s techniques (Lovejoy et al., 1993). Ciliate and dinoflagellate dominance in 18S rRNA gene libraries has also been attributed to their comparatively larger genomes and multiple copies of the rRNA gene (Godhe et al., 2008). Irrespective of the actual cell abundance, our analysis clearly revealed otherwise undetected changes in species over seasons (Supplementary Tables 2.4 and 2.5). Our seasonal data ended in July, but the interannual data in the SCM showed that both alveolate groups remain through summer to late autumn.

2.4.3 Ecological functions

Season and depth were also the major factors that grouped ciliates and dinoflagellates using the network analyses. The seasonal changes in microbial composition also suggested shifts in dominant function within the two groups (Figure 2.4). For example, winter communities were mainly dominated by heterotrophic taxa. The surface winter communities would have access to small planktonic species such as Micromonas (Lovejoy et al., 2007), haptophytes and smaller heterotrophic flagellates (Terrado et al., 2011). The relative scarcity of food resources during winter likely selected for particular groups as reported for Kongsfjorden, Svalbard (Seuthe et al., 2011) and Disko Bay (Levinsen et al., 2000; Levinsen and Nielsen, 2002). Dinoflagellates can also graze on smaller ciliates using pseudopodia or feeding veils (Jacobson and Anderson, 1986, 1996). Such trophic interactions, with dinoflagellates most likely preying on ciliates, could contribute to differences in relative abundances in different depths and times (Hansen et al., 1999; Møller et al., 2006; Seuthe et al., 2011). Additional trophic interactions were indicated in the summer when primary production was higher and zooplankton were more abundant in surface waters (Seuthe et al., 2007; Forest et al., 2011), with the appearance of the putative copepod parasite Blastodinium. After spring and towards summer, thecorrelation of some ciliates and dinoflagellates with Chl a might also be associated with predation on phytoplankton. Ciliates and dinoflagellates consume from 30% (15.6 g C m-2) of gross primary production in Amundsen Gulf to as much as 56% in some Arctic fjords (Seuthe et al., 2007; Forest et al., 2011). The availability of light in spring and summer could favor mixotrophy as well. For example, the obligate mixotrophic Laboea and most litostomatids,and the potentially plastidic Scripsiella-like, “Other Gyrodinium”, “Other Gymnodiniales” and Adenoides further increased in abundance in the SCM over this period when both light and prey would be available.

2.4.4 Interannual microzooplankton turnover and the changing Arctic

The summer-autumn ciliate and dinoflagellate assemblages tended to be similar in the different years, with a consistent predictable re-establishment of communities most years, despite samples being collected from across a large geographical distance. The SCM in the Beaufort Sea forms near the upper surface of the Pacific Winter Water, which is transported as a distinct entity over long distances (Carmack and MacDonald, 2002). Previous studies consistently show that protist communities in the SCM of Beaufort Sea and Amundsen Gulf change little over geographical space and for a given year are strongly associated with their water mass of origin (e.g. Lovejoy and Potvin, 2011; Comeau et al., 2011; Monier et al., 2015). The strong association with water mass was evident in the two samples

31 collected in July 2008 that were analysed separately because of differences in sample preparation protocols between the seasonal study and the interannual study. The internannual study was only from the small fraction, while the seasonal study included amplicons from large and small fractions. Despite this, the dinoflagellate and ciliate OTUs from Amundsen Gulf collected 8 July, were similar to those collected further offshore in the Beaufort Sea on 21 July, consistent with both 2008 samples from the same water mass and similar conditions. Lovejoy et al. (2002) showed in microcosm experiments in the Arctic that ciliates and dinoflagellates dominated when nutrients were depleted following blooms of large phytoplankton. Ciliates are able to exploit oligotrophic conditions, by grazing on smaller phytoplankton and bacteria either as strict heterotrophs or by way of mixotrophy. By the end of July 2008, the Beaufort Sea SCM was nitrate limited and had the lowest mean phosphate concentrations measured from 2003 to 2010. Overall primary productivity was low (Martin et al., 2013) with small cells (< 3 µm) dominating the Chl a biomass (Comeau et al., 2011). The mean salinity at the SCM was lower in 2008 compared to the preceding years (2003–2007), which was due to the input of additional freshwater from multi-year ice melt in 2007. This event was part of the ongoing trend in decreasing summer ice extent that has been recorded since the onset of the satellite era. The continuing low ice volume over winter was associated with the 2008 ice break-up two weeks before the historic average. This precocious ice breakup stimulated an earlier pelagic spring bloom and earlier depletion of nutrients in the euphotic zone (Forest et al., 2011). These July 2008 conditions were reflected in the microzooplankton assemblage, characterized by the greater prevalence of Strombidinium, Laboea, Monodinium sp., other unclassified ciliates and relative decrease in most dinoflagellates, which was also noted in the seasonal data set. The relative abundance of ciliate taxa in the above community gradually returned to earlier levels along with the partial recovery of summer ice extent. At the OTU level, Spirotrichea showed different trends, before and after 2007. Spirotrichea ciliates are obligate mixotrophs, feeding on smaller prey and presumably able to use retained chlorophyte or prymnesiophyte-derived plastids for photosynthesis (McManus and Katz, 2009; Stoecker et al., 1988, 2009). Similar observations were reported by Jiang et al. (2013) in Western Arctic Ocean where a few ciliate groups became dominant during the 2012 record ice melt, supporting the notion that microzooplankton are highly sensitive to changing physico-chemical regimes, including those associated with low-ice or freshening events. The mean vertical positions of the SCM and the nitracline in the Beaufort Sea and Amundsen Gulf deepened from 2003 to 2011 (Bergeron and Tremblay, 2014). The increasing distance between the depth at which light is still sufficient for photosynthesis and the nitracline will continue to accelerate under a changing climate regime (Yamamoto-Kawai et al., 2008; McLaughlin and Carmack, 2010; Krishfield et al., 2014), whichcould drive both the surface and SCM layers to become more nutrient-

32 limited. Such an effect has already been reported for the Canada Basin with an increase in smaller phytoplankton and lower nitrate levels in the upper 200 m (Li et al., 2009). Under these new conditions ciliates and dinoflagellates that are better able exploit smaller prey would be selected for. Our results support recent mesocosm and modelling experiments that suggest that under future conditions, the microzooplankton and microbial loop will become more prominent as efficient intermediates between bacteria-picoplankton and the classical food webs (Skjoldborg et al., 2003;O’Connor et al., 2009; Montagnes et al., 2010; Lewandowska et al., 2014; D’Alelio et al., 2016). In summary, this study shows that in Amundsen Gulf, ciliates and dinoflagellates exhibit complex temporal dynamics and are influenced by biological and physico-chemical regulators. Ciliate reads were more abundant in the surface when salinity was lower towards summer while dinoflagellates dominated in the SCM where nutrients and light were available for potential mixotrophic or photosynthetic activity. Despite this strong seasonality, the microzooplankton assemblages were also highly similar every summer from 2003-2010 except in 2008, following the 2007 summer ice minimum. Mixotrophic taxa increased drastically when nutrient concentrations were low and small prey were presumably available. Low nutrient-small cell conditions are predicted to occur more frequently in the Arctic, and this may lead to a change in dominant microzooplankton species that link to the classical food webs.

2.5 Acknowledgments

The seasonal data was collected during Circumpolar Flaw Lead – International Polar Year (CFL-IPY) study supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), and the Network of Centers of Excellence ArcticNet. NSERC Discovery grants and ArcticNet funding to CL facilitated completion of the study. DFLO received scholarships from Université Laval and the Canadian Excellence Research Chair – Remote Sensing of Canada’s New Arctic Frontier (CERC) grant, and additional support from the Fonds de recherche du Québec Nature et Technologies (FRQNT) to Quebéc-Océan aided in this research.The authors acknowledge the help and assistance of staff at the Institute of Ocean Sciences, Department of Fisheries and Oceans Canada, the Captains and Crews of CCGS Amundsen, Louis S St. Laurent and Wilfred Laurier. The authors acknowledge the help and assistance of staff at the Institute of Ocean Sciences, Department of Fisheries and Oceans Canada, the Captains and Crews of CCGS Amundsen, Louis S St. Laurent, and Wilfred Laurier. We are also grateful to Jonathan Gagnon and Jean-Éric Tremblay for the nutrient data, and to Ramon Terrado colleagues from ICM, CSIC Barcelona, Claire Evans from NIOZ and Marianne Potvin in carrying out some of the sampling and laboratory work.

2.6 Author Contributions

DFLO and CL conceived the project. CL and EM collected samples. EM and AC carried out the laboratory work, and pre-processed the sequence data. DFLO, AC, and CL analyzed the data, DFLO and CL wrote the manuscript and prepared the figures. MB was responsible oceanographic analysis supervision. All authors commented on the text and agreed to this submission.

3- Date Lat Long DayL Z T S Chl a NO3¯ PO4 DO PAR Stn (N) (E) hrs (m) (˚C) (PSU) 19-Nov 70.62 -123.001 1.47 10 -1.7 30.1 0.08 0.71 0.66 5380 0.048 405 80 -1.4 32.3 0.05 5.41 1.21 5300 0 7-Dec 71.31 -124.788 0 10 -1.7 30.6 0.04 1.06 0.77 5280 0.022 5D 50 -0.9 31.4 0.03 5.23 1.06 5160 0 3-Jan 71.23 -124.451 0 10 -1.7 31.1 0.02 1.41 0.77 5200 0.001 14D 75 -1.6 32 0.02 5.04 0.9 5160 0 18-Feb 71.31 -124.497 7.54 12 -1.7 31.7 0.02 4.11 0.88 5200 0.104 22D 50 -1.7 32 0.02 7.8 1.25 5100 0 10-Mar 71.04 -123.899 10.9 12 -1.7 31.8 0.05 4.32 0.74 5340 0.008 29D 65 -1.5 32.5 0.01 11 0.9 5060 0 17-Mar 70.91 -123.899 11.98 10 -1.7 31.9 0.07 5.13 1.45 5060 0

D29 50 -1.6 32.2 0 10.58 1.97 4070 0.052 9-Apr 71.31 -124.578 15.57 10 -1.7 31.9 0.6 3.18 1.41 5380 0.288

D36 50 -1.7 32 0.36 5.02 1.54 5300 0.18 30-Apr 70.76 -123.727 19.47 11 -1.7 32 0.76 3.08 1.37 5280 1.999

D43 55 -1.7 32 0.78 11.06 2.12 5160 0.079 19-May 70.66 -122.88 24 12 -1.4 31.9 10.62 0.95 1.22 5200 2.138

450b 30 -1.3 32.3 3.12 4.38 1.58 5160 1.581 14-Jun 69.99 -125.552 24 0 -1.3 31.8 0.58 1.5 na 5200 0.574

FB01 22 -1.4 32.1 0.49 2.5 na 5100 0.909 24-Jun 69.82 -123.648 24 5 2.1 31.3 0.7 1.25 na 5340 33.819

F7b 33 -1.1 32.1 0.25 2.3 na 5060 2.248 21-Jul 70.7 -122.932 24 4 8.4 30.5 0.29 0.79 0.68 5060 2.038 405- 42 -1 32.2 4.64 14.58 1.53 4070 10a 0.039

Figure 2.1. Sites and dates of collection of water samples used for A) seasonal (November 2007-July 2008) and B) interannual (2003-2010) studies from the Amundsen Gulf region and adjacent Darnley and Franklin Bays.

Figure 2.2. A dendrogram of the unweighted UniFrac dissimilarity matrix for both DNA- (filled shapes) and RNA-based (open shapes) communities collected from surface (triangles) and subsurface chlorophyll maxima/halocline (SCM; circles) layers representing different seasons including Autumn- Winter (AW) and Spring -Summer (SS), collected over the course of the IPY-CFL study.

Figure 2.3. Distribution of potentially active (based on rRNA reads) major taxa of the lowest possible ranks in (A, C) ciliates and (B, D) dinoflagellates (98% level) with the changing seasons (sampling date) in surface and SCMhalocline layers (see text). Samples were collected from Amundsen Gulf, Darnley Bay and Franklin Bay.

Figure 2.4. A network source-target plot showing the significant correlations of OTUs binned at the lowest possible ranks (relative abundances) and square root transformed environmental parameters. Only vertices (circles) and edges (arrows) that have Spearman’s rho >0.3 significant at p<0.001 were retained. Black circles are dinoflagellates and grey are ciliates. Environmental variables tested include dissolved oxygen (DO), total chlorophyll a (chl a), daylength (DayL), temperature (Temp), depth 3- (Dep), salinity (Sal), nitrate+nitrite (NO3¯), phosphate (PO4 ) and ice. Numbers correspond to the identities on the right.

Figure 2.5. (A) The pairwise unweighted UniFrac community dissimilarity of combined ciliate and dinoflagellate communities. In the boxplots, one point (dot) represents the pairwise Unweighted UniFrac dissimilarity (Y-axis) of a particular sample against another sample. Thus, for each date there are 10 dots (overlapping dots are obscured) representing that date compared against 10 dates. The black line within the box, is the mean UniFrac distance of a particular sample against all the other samples. The broken line indicates the mean dissimilarity value among all samples. (B) Combined relative abundances of the major taxa contributing to the high July 8 dissimilarity, particularly of Strombidium, Laboea, Monodinium sp. and other unclassified ciliates.

Polar Mixed Layer Pacific Halocline Water Proportion of Proportion of Proportion of Proportion of Small fraction Large fraction Small fraction Large fraction 19 Nov 2007 41 % 59 % 65 % 35 % 07 Dec 2007 41 % 59 % 59 % 41 % 03 Jan 2008 50 % 50 % 50 % 50 % 18 Feb 2008 44 % 56 % 67 % 33 % 10 Mar 2008 50 % 50 % 82 % 18 % 17 Mar 2008 72 % 28 % 50 % 50 % 09 Apr 2008 62 % 38 % 65 % 35 % 30 Apr 2008 15 % 85 % 54 % 46 % 19 May 2008 10 % 90 % 50 % 50 % 14 Jun 2008 51 % 49 % 32 % 68 % 24 Jun 2008 65 % 35 % 20 % 80 % 21 Jul 2008 85 % 15 % 21 % 79 %

Supplementary Table S2.2. Summary of physico-chemical data associated with the interannual samples collected during summer to fall in the Amundsen Gulf and Beaufort Sea from 2003-2010, na: not available. Mean monthly 3- Z T NO3¯ Si(OH)4 PO4 Ice extent Chl a total Chl a small Sample (m) (˚C) S mmol m3 mmol m3 mmol m3 (x10 km2) mg m-3 mg m-3 Aug 2003 29 -0.78 30.621 0.22 2.07 0.71 7.58 0.07 0.03 Oct 2003 35 -1.44 31.485 6.20 16.03 0.70 9.99 0.08 0.04 Jul 2004 50 -1.10 31.880 6.56 10.35 1.14 9.11 0.22 0.11 Aug 2004 54 -1.13 31.934 8.22 16.48 1.29 6.65 0.22 0.08 Sep 2005 32 -1.04 32.038 4.82 9.23 1.14 5.39 2.14 0.16 Oct 2006 46 -1.22 32.089 6.00 11.42 1.01 7.35 0.25 0.03 Nov 2007 10 -1.65 30.118 0.71 3.96 0.66 9.42 0.08 0.03 Jul 2008 43 -0.92 31.813 0.40 2.50 0.83 8.41 0.16 0.13 Aug 2009 50 -1.21 31.194 0.96 5.48 0.91 6.00 0.30 0.27 Oct 2009 25 -0.13 31.165 0.09 3.76 0.46 6.78 0.45 0.37 Oct 2010 40 -1.20 31.947 na na na 6.84 0.57 0.03

Supplementary Table S2.3. Summary of the OTUs in the surface shared by Jan 3 and Jun 14 samples but not present in the other summer samples from June 24 and July 21. Surface Taxonomy #OTU ID 3-Jan 14-Jun 17 12 1 Arctic Clade1 44 37 5 Spirotrichea 82 1 2 Spirotrichea 154 4 2 Other Dinoflagellates 184 3 1 Spirotrichea 197 1 1 Spirotrichea 281 2 6 Askenasia 396 1 1 Tintinnidium mucicola 493 2 3 Other Dinoflagellates 529 1 2 Spirotrichea 689 4 1 Other Dinoflagellates 738 19 1 Gyrodinium clone AN0630L10 778 3 2 Other Dinoflagellates 895 1 2 Spirotrichea 1704 3 2 Gymnodinium like 1803 1 1 Oligohymenophorea

Supplementary Table S2.4. Summary of dinoflagellate OTUs detected in the seasonal data per depth (surface and halocline/SCM) and their corresponding identities and abundances in RNA and DNA libraries. Surface SCM/Halocline RNA DNA RNA DNA No. of No. of No. of No. of No. of No. of No. of No. of Taxa OTUs reads OTUs reads OTUs reads OTUs reads Adenoides-like 2 269 2 15177 1 3 2 623 Amphidinium-like 1 280 1 56 1 226 1 60 Amphidoma like 2 67 1 28 0 0 2 623 Arctic Clade 1 8 1071 8 2349 10 2687 12 1624 Blastodinium like 2 211 1 46 1 44 0 0 Cochlodinium fulvescens 1 23 1 6 1 6 0 0 Dinoclone North Pole Sw0 70 1 13 2 9 2 7 0 0 Dinoclone SCM15C21 0 0 1 2 0 0 0 0 Dinophysis 1 103 1 21 0 0 1 5 Gymnodiniales 4 1306 6 733 8 2801 8 1088 Gymnodinium AF60-like clade3 0 0 0 0 1 7 0 0 Gymnodinium sp. 2 22 2 411 1 7 4 295 Gymnodinium-like 2 209 1 12 1 11 2 74 Gyrodinium 10 1756 10 402 10 520 11 381 Gyrodinium AN0630L10 1 21 0 0 0 1 17 Gyrodinium cf. gutrula 2 581 1 13 1 21 0 0 Gyrodinium helveticum 1 1 1 1 1 11 1 2 Nematodinium like 1 10 1 6 0 0 1 8 Other dinoflagellates 49 4694 56 5159 57 3921 69 2403 Perdiniopsis like 0 0 0 0 1 12 0 0 Peridinium-like 0 0 0 0 1 6 1 8 Prorocentrum like 0 0 0 0 0 1 4 Protoperidiniales-like 0 0 0 0 0 0 1 49 Protoperidinium 0 0 0 0 1 5 0 0 Protoperidinium tricingulatum 0 0 0 0 1 6 0 0 0705311 Scripsiella like 1 855 1 764 1 875 0 0 Woloszynkia halophila WHTV 1 82 1 25 1 145 1 35 S1 Total 92 11574 98 25220 102 11321 119 7299

Supplementary Table S2.5. Summary of ciliate OTUs detected in the seasonal data per depth (surface and halocline/SCM) and their corresponding identities and abundance in RNA and DNA libraries.

Surface SCM/Halocline RNA DNA RNA DNA No. of No. of No. of No. of No. of No. of No. of No. of Taxa OTUs reads OTUs reads OTUs reads OTUs reads Askenasia 3 149 2 54 2 20 3 40 Balanion masanensis 1 5 0 0 0 0 0 0 Ciliophora 10 948 13 941 11 1375 11 1373 Laboea strobila 1 83 2 222 3 165 2 19 Litostomatea 3 649 2 613 5 678 1 7 Monodinium sp. 3 57 2 61 2 244 3 58 Novistrombidium 2 15 1 99 1 11 1 3 Oligohymenophorea 3 48 2 157 6 187 1 21 Parauronema 0 0 0 0 1 122 1 136 Pelagostrombilidium 2 175 1 84 3 142 2 88 neptuni Pseudotontonia 1 76 1 4 1 2 1 2 Spirotrichea 49 7155 58 9317 58 7031 69 14148 Strombidium 7 1261 9 599 8 1264 9 313 Strombidium SNB99 2 0 0 0 0 0 0 2 15 Tintinnidium mucicola 1 7 0 0 1 52 1 8 Urotricha sp. 0 0 0 0 1 15 0 0

Total 86 10628 93 12151 103 11308 107 16231

Supplementary Figure S2.2. Bray-Curtis similarity clustering of 9 environmental variables of the seasonal surface and halocline/SCM samples in the Amundsen Gulf and Franklin Bay. Shapes indicate surface (triangle) and halocline/SCM (circle) depths. Samples from Darnley Bay (DB) are marked with red stars while those from Franklin Bay are indicated with blue stars. The samples grouped by depth and season rather than geographical location, except for the June 14 samples from DB.

Supplementary Figure S2.3. Species richness (Chao1 index) based on DNA also changed with season (Julian Day). A significant decrease in richness in the total eukaryotic community was observed in the months after mixing and loss of ice (broken lines). Halocline/SCM-based richness (circles) was also significantly higher than the surface (triangle) in the samples collected from Amundsen Gulf (AG), Darnley Bay (DB) and Franklin Bay (FB).

Supplementary Figure S2.4. Annual mean relative abundances of both A) ciliates and B) dinoflagellates relative to the total eukaryotic microbial community based on 11 samples collected over an 8-year period from 2003 to 2010 in the SCM of the Amundsen Gulf and Beaufort Sea.

Supplementary Figure S2.5. A maximum likelihood gene tree generated from nearly full length 18S rRNA gene sequences showing the placement of the most abundant “Other Ciliates” OTUs from the interannual data, here represented by their most similar reference sequences (OTU 1-14). Arrows indicate significant increase or decrease in abundance after 2007, tested using Krustal Wallis H-Test implemented in STAMP (Parks et al., 2014). The tree was constructed using RaxML with bootstrap support repeated 1000 times. Only bootstrap values higher than 50 are shown. The short HTS OTU reads were mapped back to the tree using RaxML-Evolutionary Placement Algorithm (RaxML-EPA).

Chapter 3: Dinoflagellates (Dinophyceae) in size fractionated samples from Canada Basin, Western Arctic Ocean

Résumé

Le terme dinoflagellé fait habituellement référence aux organismes libres et symbiotiques de référence au sein des Dinophycées. Les études basées sur les séquences codantes pour l'ARN 18S dans l’environnement marin reportent fréquemment des Dinophycées non identifiée parmi la classe de taille du picoplancton (0.2-3 µm). Ces organismes sont désignés sous le terme de picodinoflagellés. Les dinoflagellés occupent une grande variété de niveaux trophiques, et leur potentielle abondance nécessite de prendre en compte leurs interactions avec la communauté microbienne et l'ensemble du réseau trophique. Dans cette étude, nous utilisons un filtrage bioinformatique et une approche probabiliste pour identifier des unités taxonomiques opérationnelles (OTUs avec 98% de similarité) de Dinophycées picoplanctoniques au sein d'une communauté environnementale en assignant un rapport de vraisemblance sur la région V4 des gènes d’ARNr 18S et sur des fragments d’ARNr 18S provenant d’échantillons fractionnés en différentes tailles (0.2-3 µm et >3 µm) générés par séquençage à haut débit. L'identité taxonomique des OTUs associés aux petites fractions a été établie par alignement et par cartographie pour obtenir une phylogénie robuste. Environ 7% des OTUs ont été provisoirement assignés au picoplancton. La plupart de ces OTUs (85%) étaient étroitement liés à des Alvéolés marins non cultivables parmi des parasites de l'ordre des Syndiniales alors que les OTUs restants ont été assignés au taxon de référence des core dinoflagellés. Nous avons alors examiné la distribution verticale et horizontale de ces organismes dans le bassin Canada de l'océan Arctique en comparant les échantillons prélevés dans les couches de surface, de maximum de chlorophylle et dans les eaux du Pacifique d’été. La phylogénie et la distribution ont été cohérentes avec les OTUs picoplanctonique appartenant à des stades de vie tels que des kystes temporaires ou des dinospores libres.

Abstract

Dinophyceae, core dinoflagellates and putative parasitic Syndiniales, are ubiquitous in marine environments and environmental 18S rRNA gene surveys frequently report unidentified Dinophyceae in the picoplanktonic (0.2-3 µm) size range. These sequences are often referred to as picodinoflagellates but are rarely classified further. Core dinoflagellates are trophically diverse and the ecologies of free living species <3 μm could impact our understanding of marine microbial food webs. Here, we collected samples from 3 water masses in the Canada Basin (Arctic Ocean) to identify and ecologically classify Dinophyceae from the environment. We applied a bioinformatics filtering and probabilistic approach to identify candidate environmental picoplanktonic dinophyceaen operational taxonomic units (OTUs with 98% similarity) by assigning likelihood ratios of V4 region 18S rRNA gene and 18S rRNA reads from filter-fractionated samples (0.2-3 µm and >3 µm) generated by high throughput sequencing (HTS). The taxonomic identities of small fraction-associated OTUs were established by alignment, and mapping short reads back to robust phylogenies using an evolutionary placement algorithm. We found that the majority of the Dinophyceae OTUs in the 0.2-3 µm fraction were likely filtration artifacts, but ca. 16% were provisionally retained as picoplanktonic. Most (85%) of these were related to uncultivated marine alveolates within in the Syndiniales and only 15% were placed within the core dinoflagellate taxa. We then examined their horizontal and depth distributions in the Canada Basin by comparing surface, deep chlorophyll maxima and Pacific Summer Water samples. Both phylogeny and distribution were consistent with all of the small fraction dinophyceae OTUs being life stages such as temporary cysts or dinospores. In conclusion, our more accurate interpretation of fractionated HTS data showed that, at least in the Arctic, small Dinophyceae-related reads should be considered predominantly as life stages of larger taxa.

3.1 Introduction

Dinoflagellates (Phylum Dinoflagellata) in the Class Dinophyceae are flagellated protists that are diverse (Janouskovec et al., 2017), have multiple trophic roles, and can account for high proportion of diversity and biomass in marine food webs (Taylor et al., 2008). This class is divided into two main clades made up of mostly free-living core dinoflagellates, sometimes referred to as Dinokaryota, with 9 to 12 Orders (Saldarriaga et al., 2004; Not et al, 2012; Janouskovec et al., 2016) and the parasitic Order Syndiniales (following the World Register of Marine Species accessed 15 April 2017, http://www.marinespecies.org/aphia.php?p=taxdetails&id=146213).

For the core dinoflagellates, with characteristic fibrillar chromosomes that are always condensed, referred to as dinokaryon (thus Dinokaryota), the taxonomy is based on morphological features such as types of surface ornamentation (e.g. pores, areolae, spines, ridges), arrangement of vesicles in the cell cortex, and for thecate taxa the arrangement of thecal plates (Taylor et al., 1985, 2008; Daugbjerg et al., 2000; Hoppenrath, 2016). The majority of described marine core dinoflagellate species are >10 µm (Tomas, 1997; Okolodkov, 2005; Not et al., 2012; Hoppenrath, 2016). Although, some Syndiniales are morphologically distinct with specific host associations, many are uncultivated and known only from amplified environmental gene sequences (Groisillier et al., 2006; Guillou et al., 2008; Siano et al. 2011). Known Syndiniales exhibit alternating life forms of host-associated trophonts, and the smaller (usually <3 µm) motile free-living dinospores (Siano et al., 2011).

The advent of PCR based 18S rRNA gene (rDNA) environmental surveys, including high throughput sequencing (HTS), has been marked by reports of Dinophyceae in samples that had been pre-filtered through 2 or 3 µm pore size filters (Diez et al., 2001a; Lopez-Garcia et al., 2001). Since then, sample fractionation from many seas and oceans, including the Arctic, consistently report high relative abundance of sequences classified as Dinophyceae in the < 3 µm size fraction (Romari et al., 2004; Lin, 2006; Lovejoy et al., 2006; Terrado et al., 2009; Comeau et al. 2011; Terrado et al. 2011). This raises the possibility that picoplanktonic Dinophyceae could be widespread. Given the large literature suggesting size effects on growth rates, temperature tolerance, stoichiometry and food web properties (Huerte-Ortega et al., 2012; Sal et al., 2015), cell size is implicit when attributing potential ecological niche space and function of a species in the environment. As size-fractionated amplicon data becomes routine, it is becoming imperative to agree on taxonomic assignations within size fractions. For example, based on the TARA global ocean survey, Bescot et al. (2016) estimated that dinoflagellate rDNA recovered from the 0.2-5 µm fraction accounted for ca. 41% of the total dinoflagellate (mainly core dinoflagellates based on Not et al., 2012) operational taxonomic unit (OTU) richness in the Ocean. But in the absence of a robust categorization of the uncultivated reads, the ecological implications of

52 these findings are uncertain (Not et al., 2009; Sørensen et al., 2013). While Syndiniales reads in the small fraction may be consitent with the presence of the dinospores, the association of core dinoflagellate reads in small fractions is still enigmatic. Specifically, to our knowledge, Prorocentrum nux, with measured lengths of 3-5 µm (Puigserver and Zingone, 2002) is the smallest described core dinoflagellate and no core dinoflagellate species with a diameter of < 3 µm has been reported. While the possibility of ‘core picodinoflagellate’ species cannot be discounted, the taxonomic diversity in 18S rRNA gene surveys would also be consistent with the reads originating from small life stages, such as temporary cysts and zoospores, and would include the dinospores in Syndiniales. Alternatively, the frequent occurrence of core dinoflagellate reads in the smaller size fractions may be due to cell flexibility, or debris from sloppy feeding by zooplankton. Other sources of core dinoflagellate DNA in the smaller fraction could be due to more direct methodological artifacts, such as breakage during sample collection and handling (Diez et al., 2001b; Cheung et al., 2008; Massana et al., 2011). The usage of DNA as a template could also result in reads coming from broken dead cells or extracellular DNA amplified by PCR (Not et al., 2009;Cordinaldesi et al., 2011; Sørensen et al., 2013; Torti et al., 2015).

While breakage, filtration and PCR biases certainly account for some of the Dinophyceae sequences detected in filters < 3 µm, novel clades of many heterotrophic (Lukeš et al., 2015) and plastidic taxa (Choi et al., 2017)continue to be discovered especially using environmental sequencing approach, and more discovery is predicted as phylogenies improve with increased sampling and more regions of the ocean are explored (e.g. Lin, 2006; Lin et al., 2006; Bachy et al. 2012; Bescot et al. 2016; de Vargas et al., 2015). It is now practical to classify uncultivated taxa by placement of the HTS- generated short sequences onto phylogenetic trees based on longer reference sequences (e.g. Dunthorn et al., 2014; Thaler and Lovejoy, 2015). Following phylogenetic placement and comparison with closely related species, the ecology of uncultivated taxa or ecotypes inferred from the environment or environmental conditions where the OTUs predominantly occur, provide additional information of their likely niche space (Thaler and Lovejoy, 2015; Luddington et al., 2016). The Western Canadian Arctic Ocean is attractive for such studies because of the well characterized water masses that circulate within relatively narrow depth ranges that are easily identified from temperature-salinity (TS) characteristics (McLaughlin and Carmack, 2010). The water masses that enter and circulate in the Canada Basin of the Arctic Ocean with origins from different oceans and seasons provide distinct environments for microbial communities (Monier et al. 2013; Thaler and Lovejoy 2015). For example, the upper boundary of the Pacific water entering the Canada Basin and Amundsen Gulf is marked by a deep chlorophyll maxima (DCM), which tends to have similar communities, across large horizontal distances (Lovejoy and Potvin, 2011; Monier et al., 2013, 2015). Above this, the surface is well lit but nutrient

53 poor, whereas below the DCM there is insufficient light for photosynthesis. Although, Dinophyceae are routinely recovered in both small (0.2–3 µm) and large fractions (3–50 µm) of size fractionated samples in the Arctic (Bachy et al., 2011; Comeau et al., 2011; Terrado et al., 2011), their potential niches and normal cell size remain unknown. In addition, questions remain as to whether the Dinophyceae in the two size fraction categories are physiologically active. Provided that reads from RNA are also found in the DNA, the use of RNA as a template for 18S rRNA reads comes closer to identifying taxa that are active in protein synthesis (e.g. Blazewicz et al., 2013;Egge et al., 2013; Hu et al., 2016).

In this study, we used high throughput amplicon sequencing targeting the V4 region of 18S rRNA gene (rDNA) and 18S rRNA (rRNA) to identify nominally “pico” environmental Dinophyceae (core dinoflagellates and Syndiniales) in size fractionated samples with the goal to constrain their probable cell size and determine their taxonomy based on phylogeny. To achieve this, we used a probability of association method based on read counts of individual OTUs in the small and large fractions. Specifically, we calculated the log10 ratio of rDNA reads in the 0.2–3 µm fraction over the > 3 µm fraction. The OTUs with a “greater probability” of being small, referred to as small-associated OTUs, were then checked against the rRNA read libraries to restrict the analysis to more likely living cells. To this end, we separated core dinoflagellates and Syndiniales, which are functionally distinct and then identified OTUs to the lowest possible taxonomic rank using a curated reference database and phylogenetic mapping. We hypothesized that biotic and abiotic environmental conditions in the three Arctic water masses would harbor different communities ofDinophyceae, which was tested by way of multivariate statistical analysis.To infer the ecology and environmental distribution patterns of the small fraction-associated Dinophyceae in the different environments, we then investigated occurrence patterns of the small-associated OTUs and related these patterns to potential niches that would be consistent with putative phylogenetic identities.

3.2 Materials and Methods 3.2.1 Sampling and sequencing Samples for nucleic acids, nutrients, and ancillary environmental data were collected during the 2012 expedition of the Joint Ocean Ice Studies – Beaufort Gyre Exploration Program (JOIS–BGEP) on board the Canadian Coast Guard Icebreaker CCGS Louis S St. Laurent. Rationale of the cruise, and details of hydrography and physical oceanography are reported in Krishfield et al. (2014). The samples were collected August 25-30 (76°– 82° N) from different major hydrographic regions along the Northwind Ridge (NWR), and in the western and northern Canada Basin (CB). Specifically, we targeted samples from NWR (Stn CB11), the northern edge of the NWR (Stns CB9 and CB11), the

54 northern reaches of Canada Basin (Stn CB16) and under ice and ice-edge waters (Stns CBN and CBN2; Fig. 3.1). Target depths were selected based on water column characteristics from conductivity, temperature and depth (CTD), in situ chlorophyll fluorescence, dissolved oxygen and relative nitrate sensors mounted on the ship’s Rosette system (McLaughlin et al. 2012). Based on the downcast profiles, samples were collected with 10-L Niskin bottles that were closed during the up-cast, except in Stn CB16. For this station, a separate Rosette on the ship’s foredeck was used, which was programmed to close the bottles at the target depths based on a main rosette cast taken the previous hour. Samples were collected from three water masses identified as the surface layer (5-6 m), the deep chlorophyll maxima (DCM), which is near the top of the Pacific Summer Water halocline, and from Pacific Winter Water (PWW), which was sampled at around 100 m. Depths and salinity were verified by bottle salinity, and CTD data are available at the BGEP site (http://www.whoi.edu/beaufortgyre/expeditions).

Dissolved inorganic carbon (DIC) was measured on board as in Yamamoto-Kawai et al. (2013) and dissolved oxygen (DO) was verified using Winkler-based Titration and Kit B (Scripps Institute of Oceanography, San Diego, CA, USA). Samples for nutrient analyses (nitrate, phosphate, silicate) were collected directly from the Niskin bottles and analyzed using a 3-channel Technicon Auto Analyzer (SEAL Analytical, Mequon WI, USA) following the methods of Barwell-Clarke and Whitney (1997). Nutrient samples were not taken at some of the PWW depths, and missing data was extrapolated from the same station water mass and nearest depth. For rDNA and rRNA, samples were collected from the Niskin bottles into acid-cleaned and sample-rinsed carboys, and filtered immediately on board. At least ca. 6 L of water from each depth were filtered sequentially using a peristaltic pump and in-line filtration system, consisting of a 50-µm mesh to remove macrozooplankton, a 3 µm 47-mm polycarbonate filter (PC, AMD Manufacturing; large fraction), and a 0.2 µm Sterivex™ filter unit (Millipore; small fraction). Large fraction filters were placed in a 2-mL microfuge tubes, and 1.8 ml of RNALater™ (Invitrogen™, ThermoFisher Scientific) was added to both Sterivex filters and the microfuge tubes and left at room temperature for 30 mins before storing at -80 °C until processing in the laboratory.

DNA and RNA were extracted from the same filter using the All-Prep DNA/RNA Minikit (Qiagen) as described by Dasilva et al. (2014). The RNA was converted to cDNA using the High Capacity Reverse Transcription Kit (Applied Biosystems, ThermoFisher Scientific) following the provided protocol. The V4 region of 18S rRNA was amplified using the forward primer E572F and reverse primer E1009R (Comeau et al., 2011) each conjugated with a MiSeq© specific linking primer (Illumina) under the following PCR conditions: initial denaturation at 98 °C for 5 minutes, then 33 cycles of denaturation at 98 °C for 10 secs, annealing at 55 °C for 30 secs, another 30 secs for extension at 72 °C, and a final extension at 72 °C for 5 mins. Samples were tagged with a unique pair of indices

(barcodes) at the 5’ end of both amplicon strands using the TruSeq® and Nextera® (Illumina) barcode sets in a nested PCR with the following conditions: initial denaturation at 98 °C for 30s, then 13 cycles of 98 °C for 10s, 55 °C for 30s, 72 °C for 30s, and final elongation at at 72 °C for 4 mins and then left at 4° C. All amplicons were purified using the Agencourt AMPure XP (Beckman Coulter) magnetic bead method and checked by gel electrophoresis. Concentrations of purified barcoded samples were measured spectrophotometrically with a Nanodrop 2000c UV-Vis (ThermoScientific) and then pooled at a final concentration of 121 ng µl-1 per sample before sequencing. Multiplex sequencing was carried out using the MiSeq-Illumina© platform at IBIS/Université Laval Plateforme d’Analyses Génomiques (Québec, QC, Canada). Short read libraries are deposited in NCBI GenBank Sequence Read Archive (SRA) under the BioProject Canada Basin Microbial Communities PRJNA362945, BioSamples SAMN6281933-2000.

3.2.2 Bioinformatics and small fraction-associated Dinophyceae OTU picking Separate forward and reverse reads were demultiplexed by CASAVA v.1.8.2 (Illumina) and merged using USEARCH v.7 with at least 25 bp overlap (Edgar, 2010). Merged paired-end sequences were filtered for short sequences and collated in Quantitative Insights into Microbial Ecology (QIIME; Caporaso et al., 2010b). Singletons and chimeras were determined de novo and against SILVA database v.102 (Quast et al., 2013) and removed. Clustering to determine operational taxonomic units (OTUs) was carried out at 98% similarity (Comeau et al., 2011;Wolf et al., 2015) in USEARCH v.7 (Edgar, 2010). Representative reads were aligned in PyNast (Caporaso et al., 2010a) and a phylogenetic tree was generated using FastTree (Price et al., 2010). OTU maps and tables were generated in UPARSE (Edgar, 2013), and taxonomy assignments were carried out in mothur (Schloss et al., 2009) against our 18S rRNA gene reference database v.1.1 (Lovejoy et al., 2016), which includes curated Arctic sequences. Fungi, metazoan and sequences unclassified only to the major group level were filtered out. For abundance and diversity comparisons the OTU table was resampled (rarefied) to 19,767 reads per sample, which was based on the sample with the fewest number of reads. The Dinophyceae-related reads (core dinoflagellate taxa and Syndiniales) were selected for the remaining analysis.We further restricted the OTUs to those also found in rRNA to select against OTUsthat were consistently from non-active cells, debris or free-nucleic acids and OTUs that did not appear in corresponding rRNA libraries at least once were not considered for further analysis. OTUs were considered small fraction-associated (small-associated) when they were detected exclusively in the 0.2–3 µm fractionfrom rDNA. These included core dinoflagellates and syndinean alveolates in Groups I-IV in Guillou et al. (2008), which are referred to as MALVS I and II in Groisillier et al. (2006). For each retained OTU that was present in both size fractions, the mean rDNA read ratio for each depth category from all stations, in large compared to small fractions, was log10 transformed.

The log10 ratio is a method to find the odds or probability of a specific OTU occurrence in either fraction, where the log10 of 1 is 0 and indicates equal proportions in both fractions, negative values indicate higher probability of being small and positive values with being large (Ganesh et al., 2014). Because of the initial aggregation by sample depth (mean reads per depth), small-associated OTUs may not have been present or abundant in both the rDNA and rRNA in a single sample. OTUs that had negative log10 values in one depth category but 0 in others were kept, since clogging could result in retention of small cells in larger fractions, while OTUs determined to be small-associated in one depth but large-associated in another were removed in the interest of identifying taxa which were consistently small. To compare, we also calculated log10 ratios of both rRNA and rDNA reads of binned OTUs based on their lowest taxonomic ranks.

To test our assumptions inferringsize from read counts, we applied the same methods to known picoplanktonic groups in the Mamiellophyceae, which in these samples would be mostly Micromonas (Lovejoy et al., 2006; Lovejoy et al., 2007). We expected that OTUs related to these groups would besmall-fraction associated, and thus wouldhave negative log10 values.

3.2.3 Phylogenetic placement of short reads To determine the phylogenetic identity of the small-associated OTUs, phylogenetic reference trees were generated using Randomized Axelerated Maximum Likelihood (RAxML v.8; Stamatakis, 2014). For this, the small-associated OTUs were searched against the NCBI GenBank using BLASTn to identify the best hits (Altschul et al., 1990). Most of the high similarity sequences (> 98%, e-value 1e-10) were selected and aligned in MUSCLE (Edgar, 2004), with the alignment manually edited and trimmed in MEGA v.7.0 (Kumar et al., 2016), and used to build robust clade-specific reference trees with RAxML v.8 using the GTRGAMMA models, run 1000 times (Stamatakis, 2014). Reference trees were generated separately for the core dinoflagellates and the Syndiniales (Groisillier et al., 2006; Lin et al., 2006; Guillou et al., 2008). The reference trees, including outgroups and related taxa were constructed with a total of 165 curated sequences for core dinoflagellates (Fig. S3.1), and 247 for Syndiniales (Fig. S3.2), which were mainly retrieved from the NCBI, andboth with > 800 parsimony informative positions. Assignment down to the clade or species levels in Syndiniales was based on Guillou et al. (2008) but in the case of conflicts they were only assigned as Syndiniales Groups I to V (Guillou et al.. 2008). Clades that were not previously described in the literature were labeled ‘unassigned clade’. The small-associated Dinophyceae OTUs were then mapped back onto the reference trees (core dinoflagellates and Syndiniales) using the -n option in the Evolutionary Placement

Algorithm in-RAxML v.8.2 (EPA-RAxML, Berger et al., 2011). Trees were visualized in Dendroscope v.3.0 (Huson and Scornavacca, 2012).

3.2.4 Statistical and ecological network analyses We further investigated the patterns of occurrences of the small-associated OTUs and compared their depth distribution with the distribution based on their phylogenetic identification. In addition, phylogenetic community clustering based on the presence-absence of small-associated OTUs were estimated by Unweighted UniFrac distance, which is a measure of phylogenetic beta diversity, and Principal Coordinates Analyses (PCoA; Gower, 1996) in QIIME (Caporaso et al. 2010b). The PC1 values derived from the PCoA were tested for correlations with the environmental parameters to select the strongest drivers of community phylogenetic clustering. The UniFrac distances of the small- associated OTUs were also clustered using Non-Metric Dimensional Scaling (NMDS) implemented in ‘Vegan’ overlain to salinity data (Oksanen et al., 2015).

Co-occurrence of the small-associated potential parasitic or symbiotic taxa with potential hosts based on literature searches was explored using ecological network analysis. For this analysis, we extracted OTUs matching putative hosts of syndinean taxa including ciliates, core dinoflagellates and the radiolarians in the Acantharea and Polycystinea (Coats, 1999; Chambouvet et al., 2008; Ikenoue et al., 2016). To increase the resolution of the network, only OTUs with > 100 rRNA reads from the combined large and small fractions were used. The interaction matrix was generated in ‘vegan’ and plotted as an undirected network in ‘igraph’ implemented in R (Csardi and Nepusz, 2006). Networks were visualized in Cytoscape (Shannon et al., 2003) and only edges connected by Spearman’s rho > 0.6 and significant at p < 0.01 were included (Barberán et al., 2012). The analysis of variance (ANOVA), multiple regressions, Canonical Correspondence Analysis (CCA) and Principal Components Analysis (PCA) were carried out in PAST v3.0 (Hammer et al., 2001).

3.3 Results 3.3.1 Physico-chemical profiles of the sampling sites During the 2012 expedition, most of the Canada Basin was sea ice-free, with sea ice limited to above 80° N (Stn CBN). Strong salinity stratification and layering of different water masses was typical of the Western Canadian Arctic (McLaughlin et al., 2005) at all stations (Table 3.1). PCA ordination based on nutrient, salinity, temperature, CDOM, DIC, DO and Chl a values clustered the samples into the three target water masses (Fig. S3.3), the near surface Polar Mixed Layer (PML) with salinities ranging from 24.3 to 28.8, the top of the Pacific Summer Water Halocline where the DCM forms during

58 summer with salinities from 31.2 to 32 was always slightly warmer than the deeper Pacific Winter Water (PWW) with salinities from 32.2 to 33.2 (Figs. 3.2A &3.2B). Stn TU-1 on the NWR differed the most with the freshest and warmest upper waters (Fig. 3.2B), and lowest nitrate concentrations at the three depths (Table 3.1). In addition, near surface temperature maxima of -0.2 °C to 0.02 °C (Jackson et al., 2010) were evident at depths around 40 to 60 m in stations inside the Canada Basin, while Stn. TU-1 had a temperature minimum of -1.0 °C at ca. 49 m, suggesting a remnant winter mixed layer (rML; Jackson et al., 2011, Fig. 3.2B). Maximum Chl a concentration estimated from fluorescence in the DCM was greatest in the shallower ridge Stn TU-1 (1.10 µg L-1), and minimal at Stns CBN (0.13 µg L-1) and CBN2 (0.19 µg L-1) in the north of Canada Basin (Fig. 3.2C).

3.3.2 Small fraction-associated dinoflagellate diversity In total, we analyzed 6 stations, with 3 depths each except for Stn CB9 with 2 depths, 2 size fractions each from the 2 nucleic acid templates (RNA and DNA; Table S3.1). After quality filtering, 4,666,650 reads were generated from the tagged paired-end reads with a range of390-490 bp and average of 450 bp. Total microbial eukaryote reads (excluding fungi and “Other Eukaryotes”) ranged from 19,767 to 495,781 per sample. Dinophyceae OTUs accounted for 38.70% (1,806,029) of the total reads and clustered into 883 OTUs. Out of the 883 Dinophyceae OTUs, based on their DNA log10 odd ratios across all depths (Fig. 3.3), 143 had higher likelihoods of being found in small fraction or small- associated, accounting for around 16% of the total Dinophyceaen OTUs but only ca. 3.5% of the total microbial OTUs (all fractions).

In-depth phylogenetic placement of all the small-associated OTUs using EPA-RAxML further classified the Dinophyceae sequences within the dinoflagellate core taxa (Fig. 3.4A) and the major groups of syndinean taxa (Fig. 3.4B). The core dinoflagellate OTUs (21 OTUs; Fig. 3.4A) clustered with established clades based on the most similar reference sequences from cultures, with representatives from Gyrodinium, Gymnodinium, Pfiesteria, Prorocentrum, Heterocapsa, Karenia, Amphidinium and Akashiwo and a Noctilucales. The remaining OTUs were Syndiniales, with representatives from Groups I (SG-I, 18 OTUs), SG-II (84 OTUs), SG-III (7 OTUs) and SG-IV (5 OTUs). SG-V small-associated OTUs, which are fish parasites, were not found in this study.

3.3.3 Community structuring and correlation with environmental variables Phylogenetic community clustering based on the unweighted UniFrac, grouped samples primarily by depth category in both NMDS (Fig. 3.5) and PCoA (Fig. S3.4). Depth category, rather

59 than station, explained most of the variation (49.76%; PC1 vs. depth R2 = 0.93, p<0.001). Stn TU-1 in NWR emerged as the most dissimilar, especially in the surface and DCM, but the community still clustered with other samples in the same depth category.

Most (112 OTUs) of the rRNA reads of the 143 small-associated Dinophyceae were in the PWW samples and with a lower number (80 OTUs) in the surface waters. The DCM and PWW shared the greatest number of OTUs with 70% of OTUs shared. CCA ordination placed most core dinoflagellate OTUs with high DO, Chl a and temperature, indicating that they were most active in the upper surface and DCM waters (Fig. 3.6A). In contrast, most OTUs (>100) belonging to Syndiniales were associated with higher salinity, nutrients, DIC and CDOM, indicating taxa active in deeper waters, consistent with the PCA results (Fig. S3.3). Syndiniales Groups I and II (SG-I and SG-II) were present in all depths, while Groups III and IV were mostly limited to PWW. The rRNArelative abundance of some of the OTUs varied with depth. For example, the most widespread and abundant OTU belonging to SG-I (OTU 29; Fig. 3.6B) was mostly found in the surface waters. The other abundant rRNA (OTUs 90, 99 and 126) were only in the DCM. In contrast, more reads from OTUs 47, 80, 95, 101 and 1024 were in the PWW, with fewest read in the surface waters.

There was a strong and positive co-occurrence of several syndinean taxa with potential hosts (Fig. 3.7). A total of eight significant clusters, connected by Spearman’s rho > 0.6 (p < 0.001) were identified. Syndiniales Groups SG-I and SG-II seemed to co-occur more often, with both appearing in Clusters 4, 6 and 7. Specific associations were also apparent with SG-IV associated with Litostomatea, Gymnodiniales and Polycistenea OTUs (Clusters 3 and 5), while SG-III was associated with ciliates; Choreotrichida, Oligohymenophorea, and several unclassified dinoflagellates. SG-I and SG-II were less specific with significant connections to the majority of ciliate, dinoflagellate and radiolarian taxa. One cluster consisted of only potential hosts with OTUs from ciliates, dinoflagellates and one Radiolarian. The distribution of the hosts and syndinean taxa was also associated with depth categories. Specifically, both host and syndinean OTUs in Clusters 2, 3 and 6 consisted of active OTUs (based on rRNA reads) in the PWW, while those in Clusters 4 and 8 were more abundant in the DCM. Cluster 7 OTUs were associated with both surface and PWW, while Cluster 1 was made up of highly abundant OTUs but unconnected nodes.

3.4 Discussion Although Dinophyceaen dinoflagellates are among the most diverse and highly studied protistan groups, many aspects of their ecology and diversity are still unknown, especially for the uncultivated taxa from environmental surveys. A primary example of this is the high proportion of Dinophycean reads associated with the small fraction (0.2–3 μm) of HTS data. Here, we found that many of these

60 small-associated reads were within clades of Syndiniales, consistent with these parasitic taxa having life forms smaller than 3 μm (i.e. motile dinospores), as reported elsewhere (Guillou et al., 2008, Siano et al., 2011). However, the core dinoflagellates reported in small fractions in environmental surveys has baffled many taxonomists and microbial ecologists, since pico-size motile representatives of core dinoflagellates have not been seen using microscopy or even flow cytometry (Marie et al., 2010). This is despite taxonomic revelations from scanning and transmission microscopies, which has revealed the diversity of small species for over half a century (Manton and Parke, 1960). The disparity between molecular and microscopic evidence is further confounded by the confusion in the use of terminology by non-specialists, which is a problem not only for Dinophyceae but for dinoflagellates (Dinoflagellata) in general as discussed by Hoppenrath (2016). This confusion arises from the long history of dinoflagellate classification. In essence, dinoflagellates have been considered as both microalgae and as microzooplankton over time and fall under two codes of nomenclature. For Dinophyceae specifically, in the phycological literature, Dinoflagellata (Dinophyta) have separate classes for Dinophyceae F.E Fritsch (core dinoflagellates) and the Syndinea Chatton (Guirry and Guiry, 2016; AlgaeBase, searched 17 April 2017,http://www.algaebase.org/browse/taxonomy). In the Encyclopedia of life (EOL) non-curated entry (accessed 17 April 2016, http://www.eol.org/pages/4758/overview) and World Registry of Marine Species (accessed 11 April 2017, http://www.marinespecies.org/), Dinophyceae Fritsch, 1927 includes both core dinoflagellates and Syndinea. In the non-ranking classification in Adl et al. (2005, 2012) and the Open Tree of Life (Hinchliff et al., 2015; https://tree.opentreeoflife.org) however, Syndiniales are separately classifed under the Protalveolata. The SILVA 18S rRNA gene reference database (Quast et al., 2013) frequently used in many HTS-surveys tends to follow the non-ranking systems. Since automated pipelines based on different classifications often fail to distinguish the two major lineages, when small fraction is reported as picodinoflagellates at face value, ecological insights can be lost.

3.4.1 Linking ratio of fractionated rDNA reads to potential cell characteristics Here we analyzed the V4 region of the 18S rRNA gene and the 18S rRNA from size fractionated samples collected from a mostly oligotrophic region of the Arctic Ocean with the goal of verifying the occurrence of the dinophycean “picodinoflagellates”. We found that with few exceptions, core dinoflagellate and syndinean related OTUs at 98% similarity level, and with representatives in both rRNA and rDNA, occurred in both size fractions. On the surface, this would suggest that majority, if not all, of the Dinophyceae taxa detected have picoplanktonic representatives or are vulnerable to cell breakage. However, using the log10 ratio approach, we only identified 143 out of 883 OTUs that had higher probability of being found in the smaller fraction, while the rest, especially most of the core

61 dinoflagellates, were more associated with the large fractions. First, this was consistent with taxon specific morphological descriptions of core dinoflagellates being larger (Tomas, 1997; Hoppenrath, 2016). Second, the greater proportion of likely ‘pico’ cells belonging to Syndinales was consistent with the production of dinospores by these mostly parasitic groups (Coats, 1999; Siano et al., 2011). This approach of associating the OTU reads with size class was validated on the Mamilliophyceae that are unequivocally smaller than 3 µm, with the OTUs placed overwhelmingly into the small size fraction (Fig. S3.5). However, the separation was not perfect. Reasons why the same OTUs can appear in both size fractions are well known; although filter fractionation physically separates the small and large- celled organisms, there are caveats (Padilla et al., 2015), with cell breakage being a major factor. For fragile larger cells, this presumably would have led to inconsistent log10 ratio patterns, and when this occurred, we removed suspect OTUs.

The presence of truly small pico-sized cells in large fractions could also be due to attachment to larger particles, filter clogging (Terrado et al., 2009) or association with their hosts in the case of the parasites (Bescot et al., 2016). Although reads from both major taxonomic divisions of the Dinophyceaeoccurred in both fractions, these were found in different relative abundances, with the core dinoflagellates forming different categories of fraction ratios (Fig. S3.6) suggesting a variety of life cycles or degree of cell fragility (Not et al. 2009; Bescot et al., 2016) among the different orders and genera. Specifically, consistent with the taxonomic mapping, OTU representatives closely related to larger dinoflagellate species had higher occurrences in > 3 µm fraction, while those related to smaller, flexible or fragile taxa were more abundant in the smaller 0.2–3.0 µm fraction. The 2% total small- associated core dinoflagellate OTUs we identified was considerably less than the 19% to 47% of the total dinoflagellate reads reported elsewhere (Vaulot et al., 2008; Cheung et al., 2010; Bescot et al., 2016), and while this may have been specific to our study region, other reports warrant critical examinantion.

3.4.2 Phylogenetic identity of small-fraction associated Dinophyceace Out of 143 small-associated Dinophyceae OTUs, only 21 aligned with the core dinoflagellate genera or clades (Saldarriaga et al., 2001; Orr et al., 2012; Rene et al., 2015). Many were related to heterotrophic Gyrodinium spp. and potential toxin-producers Akashiwo and Karenia. Other small- associated OTUs clustered with environmental clones, which we designated “unidentified core dinoflagellates” 1 to 4 (Fig. 3.4A). Others were related to thecate Prorocentralesand other Peridiniales such as Heterocapsa and Pfiesteria. Interestingly, one OTU clustered with a Noctiluca sequence (Fig. 3.4A), but its placement might have been influenced by the limited phylogenetic resolution provided

62 by the 18S rRNA gene (Saldarriaga et al., 2004; Gómez, 2014). The remaining 122 small-associated Dinophyceae OTUs belonged to syndinean Groups I-IV as defined by Guillou et al. (2008). Other gene surveys consistently reported that these groups constitute majority of the alveolate small-fraction reads (< 5 µm) in marine systems including the Arctic (Bachy et al., 2011; Sørensen et al., 2012; Monier et al., 2015; Wolf et al., 2015; Marquardt et al., 2016) but their taxonomic classification has proven difficult because the phylogenetic positioning and arrangement of the clades within Syndiniales are unclear with many clades not supported by bootstrap analyses (Groisillier et al., 2006; Guillou et al., 2008), and the syndinean complex should be revisited. Despite this, in our study, individual OTUs consistently clustered together forming distinct clades. Our updated reference tree also revealed potential clades that have not been described before. For example, we noted several sequences that did not cluster with other reference sequences but were all well within the syndinean complex (i.e. ‘unassigned clades’; Fig. S3.2), consistent with endemicity in the Arctic (Balzano et al., 2012; Luddington et al., 2016). In contrast, in the Southern Ocean there were few region-specific Syndinales OTUs (Wolf et al., 2015), suggesting geographic isolation, ecological selection or high host specificity in the Arctic.

Since most of the small-associated core dinoflagellate OTUs were highly similar to known taxa that do not have any reported picoplanktonic cell representatives, it is possible then they could be the smaller life stages. Most GPP species for example, form single-walled temporary cysts or zoospores that could be smaller than their vegetative counterparts (Oestmann and Lewis, 1996; Zhou and Yan, 2002). Small symbiotic dinoflagellate species are also plausible (Gast and Caron, 1996; Pochon et al., 2014), although they would be expected to be more abundant in the larger fraction if associated with their hosts. Several studies have reported the abundance of temporary cysts of core dinoflagellates (Garcés et al., 2002; Olli, 2004; Bravo and Figueroa, 2014) and dinospores of syndinean Dinophyceae (Siano et al., 2011; Piwosz et al., 2013) in the field, indicating their widespread occurrence. Our results are the first HTS-inferred occurrence of these under reported life stages, which cannot be easily recognized by conventional microscopic methods but would need multi-staining techniques (e.g. Onda et al., 2014) or in situ hybridization approaches for higher resolution for identification (e.g. Chambouvet et al., 2008; Siano et al., 2011).

3.4.3 Distribution and potential ecologies Occurrence of uncultivated microorganisms under specific environmental conditions could provide insight into their potential ecology. As with other Arctic microbial groups, the assemblages of small- associated Dinophyceae OTUs were found to correspond to water mass (e.g. Monier et al., 2013; Thaler

63 and Lovejoy, 2015; Onda et al., 2017) and could suggest strong environmental filtering. Even for the Northwind Ridge Stn TU-1, where the target Pacific water masses were closer to the surface compared to the Canada Basin Beaufort Gyre, communities were true to their Pacific history in keeping with Atlantic- or Pacific water-associated dinoflagellate communities in the Arctic described by Okolodkov and Dodge (1996).

Arctic core dinoflagellates are thought to be either mixotrophic or heterotrophic (e.g. Levinsen and Nielsen, 2002; Sherr and Sherr, 2009; Sherr et al., 2013) with few reports of true autotrophs (Richerol et al., 2012). Here, the small-associated core dinoflagellate OTUs with matches to Gymnodinium, Akashiwo and Karenia, which have chloroplasts (Stoecker, 1999; Janouskoveca et al., 2017), were only found in the surface and DCM and could contribute to carbon fixation and productivity in these layers. However, field and in situ experiments also suggest that low nutrients in the surface or light limitation in the DCM and PWW could induce encystment by dinoflagellate populations presumably to avoid competition (Anderson et al., 1985; Garcés et al., 2002; Olli and Anderson, 2002; Olli, 2004; Rintala et al., 2007). In contrast, lack of light at the PWW potentially favored heterotrophic dinoflagellates such as Pfiesteria, Amphidinium, Gyrodinium (Hansen, 1991) and some potentially mixotrophic Heterocapsa (Millette et al., 2016). The dinoflagellates with mixotrophic potential could survive by grazing on bacteria or other protists (Jeong et al., 2010, 2014).

Syndiniales species are mainly regarded as parasites with a trophont stage alternating with a free- living dinospore stage (Coats, 1999; Guillou et al., 2008), and their distribution usually follows their hosts (Siano et al., 2011). The network analysis was consistent with the potential for close host-parasite associations. For example, the highly abundant SG-I OTU (OTU 29) was mostly found in the surface waters of the northernmost stations nearer the ice edges, where potential litostomatid ciliate hosts were also abundant (Fig. 3.7, Cluster 5). SG-I Duboscquella-related sequences along with litostomatid ciliates were also reported near the North Pole (Bachy et al., 2011) and Amundsen Gulf upper waters (Terrado et al., 2011). The SG-II Amoebophrya (Clade 2) OTUs, which are reported as specialist parasites of core dinoflagellates (Coats and Park, 2002; Jephcott et al., 2016) and some radiolarians (Bråte et al., 2012), significantly co-occurred with Gymnodiniales, Polycystinea and Acantharea in the DCM and PWW. Amoebophrya dinospores have been microscopically detected using fluorescence in situ hybridization (FISH) in first year sea ice in the Arctic albeit in low number (Bachy et al., 2011; Piwosz et al., 2013). Given the reported strong association with dinoflagellates, the co-occurrence of Amoebophrya with ciliates may have been a consequence of ciliates grazing of dinospores rather than parasitism (Johansson and Coats, 2002). OTUs of copepod parasites from the environmental clade SG- III and Hematodinium-related sequences from SG-IV (Skovgaard et al., 2005; Stentiford and Shields,

2005) occurred more consistently in the PWW, where downward migrating zooplankton are usually abundant during summer (Seuthe et al., 2007; Darnis et al., 2012). Potentially, the syndinean OTUs at this depth represent free-living dinospores released from infected copepods. Terrado et al. (2011) also detected these groups during spring in Amundsen Gulf before the break-up of ice coinciding with the upward migration of zooplankton. The association of many described Syndiniales howeverwith their hosts may be non-specific to extremely host-specific (Kim 2006; Kim et al., 2008; Brate et al. 2012; Ikenoue et al. 2016), and specific associations require further study.

3.5 Conclusions Using a probability approach to differentiate likely small and large fraction-associated Dinophyceae OTUs, we found that only 143 out of 883 OTUs were small-associated, indicating that majority of the OTU reads detected in the small fractions (< 3 µm) were more likely artefactual. The small-associated OTUs were closely related to known environmental clades or described species. Overall, available evidence suggest that the picodinoflagellates (Dinophyceae) in the Arctic Ocean are dominated by Sydiniales, with very few representatives from the core dinoflagellates, and would mostly pertain to the picoplankonic life stages such as temporary cysts or dinospores. We also did not find compelling evidence based on fractionation, phylogeny, distribution, and inferred ecology that would suggest existence of potential novel core picodinoflagellates. However, the existence of ‘true core picodinoflagellates’ cannot be completely dismissed since our study was limited over time and confined to a single ecozone. Results of this study however should serve as a caution in interpreting HTS-derived data especially in the case of filter-fractionated Dinophyceae reads. Overall, more studies are needed to accurately link the ‘core picodinoflagellate’ reads with morphological evidence.

3.6 Acknowledgments

Samples were collected during the Joint Ocean Ice Studies – Beaufort Gyre Exploration Project (JOIS-BGEP), which was a collaboration between the Woods Hole Oceanographic Institution (WHOI) and the Institute of Ocean Sciences – Department of Fisheries and Oceans Canada (DFO- Canada). Major funding was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant, and the Network of Centers of Excellence ArcticNet funds to CL.

DFLO received scholarships from Université Laval, the Canadian Excellence Research Chair – Remote Sensing of Canada’s New Arctic Frontier (CERC) grant, Takuvik Joint International Laboratory, and additional support from the Fonds de recherche du Québec Nature et Technologies (FRQNT) to Quebéc-Océan.

The authors acknowledge the Captain and Crew of CCGS Louis S St. Laurent and the staff of IOS-DFO for their assistance and help during the cruise. We are also grateful to Jane Eert, Mike Dempsey, Linda White, Sarah-Ann Quesnel, Mary Thaler, Emmanuelle Medrinal, Cindy Dasilva and Marianne Potvin for logistics, sampling and assistance with some of the laboratory work. We also thank the IBIS bioinformatics team, especially Jerome Laroche, for their computational support.

Table 3.1. Stations (Stn), dates of collection (Date), collection Latitude and Longitude (Lat-Long), physico-chemical parameters and chlorophyll a (Chl a) concentrations of the samples used for amplicon high throughput sequencing. Other column names refer to depth (Z) of sampling, depth 3 category (Layer), temperature (Temp), salinity (S), dissolved oxygen (DO), nitrate (NO3¯), silicate (Si(OH)4), phosphate (PO4 -), colored dissolved organic matter (CDOM) and dissolved inorganic carbon (DIC).

Lat Long Z DO (mmol Chl a NO ¯ Si(OH) PO 3- CDOM DIC Stn Date Layer T (C⁰) Sal 3 4 4 (N) (W ) (m) m3) (mg m3) (mmol m3) (mmolm3) (mmolm3) (mg m3) (μmol kg-1)

TU1 30-Aug-12 76.02 -160.26 6.853 Surface -0.26 25.26 385.73 0.04 0 2.50 0.50 3.07 1769.84

69.342 DCM -0.05 31.20 363.18 0.45 1.84 6.29 0.93 4.60 2068.16 92.753 PWW -0.76 31.86 335.84 0.15 6.55 13.81 1.32 5.15 2113.22 CB9 29-Aug-12 78.01 -149.97 7.5 Surface -0.66 26.62 387.34 0.05 0.01 2.65 0.57 3.16 1826.63 68.776 DCM -0.47 31.66 349.15 0.30 4.85 10.47 1.15 4.77 2071.99 CB11 25-Aug-12 79.01 -149.99 6.809 Surface -0.77 26.66 386.84 0.03 0 2.6 0.55 3.20 1840.21

79.412 DCM -0.86 31.89 340.00 0.16 6.7 13.13 1.3 4.87 2104.47 104.678 PWW -1.29 32.32 314.09 0.04 11.1 23.18 1.63 5.19 2148.97 CB16 26-Aug-12 78.00 -140.01 6.613 Surface -0.74 27.35 384.25 0.07 0 2.36 0.57 3.38 1872.57

56.358 DCM -0.37 31.30 367.28 0.62 2.08 6.6 0.96 4.71 2054.12 116.987 PWW -1.28 32.36 309.99 0.05 11.42 24.09 1.66 5.18 2142.69 CBN 26-Aug-12 80.88 -137.43 7.015 Surface -1.34 28.32 401.36 0.05 0 2.74 0.64 3.20 1918.19

87.745 DCM -1.42 32.35 324.90 0.08 12.19 26.34 1.74 5.21 2160.12 153.447 PWW -1.49 33.15 276.09 0.05 16.19 39.50 2.01 5.16 2202.52 CBN2 28-Aug-12 80.22 -130.10 6.664 Surface -1.40 28.78 401.36 0.06 0.19 2.83 0.66 3.11 1995.01

64.1 DCM -1.13 31.95 324.90 0.28 7.52 14.83 1.35 4.89 2132.44 153.594 PWW -1.49 33.18 276.09 0.05 16.02 37.35 1.97 5.25 2235.06

Figure 3.1. Map of the stations (red circles) where samples used in this study were collected from Western Canada Basin and Northwind Ridge in August 2012 onboard the CGGS Louis S St. Laurent.

Figure 3.2. Depth profiles from the six stations (colour coded) used in this study.A) Salinity, B) temperature, and C) in situ chlorophyll a fluorescence

Figure 3.3. Dinoflagellate OTUs were grouped according to the ratio of their rDNA reads in large (>3 µm) over small (0.2-3 µm) fractions that were generated by high throughput sequencing from the surface, DCM and PWW m depths (see text) collected from Western Arctic Ocean (Canada Basin). Negative Log10 values indicate small fraction-associated OTUs (small-associated) and positive are large fraction-associated OTUs (large-associated). Not shown are OTUs that had equal chances (0 Log10) of being found in both fractions.

(see caption next page…)

Figure 3.4. Collapsed dinophyceae phylogenetic trees reconstructed from nearly full length 18s rRNA gene sequences showing the placement of the HTS reads represented by the most similar sequences belonging to (A) the “core dinoflagellate” (GPP+Noctilucales) and (B) the Syndiniales Groups I-V. The number of OTUs similar to the clades or species are indicated by the red numbers in parenthesis. Also indicated in the Syndiniales tree are the old MALVs I and II groupings. Only bootstrap supports higher than 50 are shown.

Figure 3.5.A Non-Metric Dimensional Scaling (NMDS) plot showing the community structuring of 143 small associated dinophyceae OTUs based on the rDNA abundance in small fraction and overlaid with the salinity values where the samples were collected. Samples clustered by depth (surface=square, DCM=triangle, PWW/>100m=circle) with strong correlation (R2=0.95, p<0.001).

A B

Figure 3.6.(A)Canonical correspondence analysis (CCA) ordination revealed the differences in the patterns of distribution and potential activity of the 143 OTUs based on rRNA abundance in small fraction, belonging to 5 main groups: GPP+Noctilucales (green), Syndiniales Group I (SG-1; blue), SG-II (orange), SG-III (pink) and SG-IV (red). These differences in level of activity (rRNA abundance) were very apparent in the most abundant OTUs (B).

Figure 3.7. An ecological network showing strong and positive co-occurrences (Spearman’s rho >0.6, p<0.01) of the Syndiniales and their potential microplankton hosts (ciliates, dinoflagellates and radiolarians) generated from highly abundant OTUs (>100 reads). The different taxonomic groups are represented by the colored nodes, while varied thickness of the edges (black lines) refer to the level of Spearman’s correlation. The numbers refer to significant clusters or grouping of highly connected co- occurrences, while the shapes correspond to depths where the OTUs were most abundant based on rRNA reads. The 21 small-associated dinoflagellate OTUs were excluded from this particular analysis since they were assumed to be cysts.

Supplementary Table S3.1. Summary of the stations and depths sampled (surface [surf], deep chlorophyll maxima [DCM], Pacific Winter Water [PWW]), and the fraction (small/large) and type of samples (RNA/DNA) used in this study.

Small (0.2-3.0 um) Large (>3.0 um) Station Depths sampled DNA RNA DNA RNA TU-1 surf, DCM, PWW x x x x

CB9 surf, DCM x x x x CB11 surf, DCM, PWW x x x x CB16 surf, DCM, PWW x x x x CBN surf, DCM, PWW x x x x CBN2 surf, DCM, PWW x x x x

Supplementary Figures S3.1. An 18S rRNA gene reference tree generated by RaxML approach, which was used to identify HTS-derived reads of small-associated core dinoflagellates OTUs, also described in Figure 3.4A. Taxonomy based on Guiry and Guriy (2017).

A (Legend next page…)

Supplementary Figures S3.2. An 18S rRNA gene reference trees generated by RaxML approach, which was used to identify HTS-derived reads of small-associated Syndiniales OTUs in Syndiniales Group I (A) and Syndiniales Groups II-V (B), also described in Figure 3.4B. Taxonomy based on Guillou et al. (2008). Sequences belonging to clades that have not been described previously are labeled “unassigned Syndiniales Group”.

Supplementary Figure S3.3. Principal components analysis (PCA) ordination of the samples based 3- on the nutrients (Nitrate, NO3¯; Phosphate, PO4 ; Silicate, Si(OH)4), salinity (Sal), temperature (Temp), dissolved organic carbon (DIC), dissolved oxygen (DO) and chlorophyll a (Chl a) fluorescence values. Only variables that had significant contributions (marked by red arrows) are shown. Shapes indicate depth categories surface (square), deep chlotophyll maxima (triangle) and PWW (circle).

Supplementary Figure S3.4. Principal coordinates analysis (PCoA) plot based on Unweighted UniFrac (presence-absence) computed from small rDNA of the 143 selected putative picodinoflagellate OTUs showed that samples clustered by depth (surface=square, DCM=triangle, PWW=circle) with strong correlation (R2=0.95, p<0.001).

Supplementary Figure S3.6. Order/genus-level likelihood occurrences using ratios of rDNA and rRNA relative abundances (from binned OTUs) in fractionated samples (large/small) showed that dinoflagellates could be grouped according to these ratios. Class 1 is composed of taxa whose rDNA and rRNA both occur more on the larges. Class 2 is made up of taxa whose rDNA are more on the large fraction, but rRNA with equal odds of being found in both fractions. Class 3 are taxa that have rDNA found more in the large, but rRNA in small fraction, while Class 4 are taxa that have higher odds of being found in the small fraction for both rRNA and rDNA reads. Taxonomic assignments were based on the Northern 18S rRNA Reference Database v.1.1 (Lovejoy et al., 2016).

Chapter 4: Assembly rules governing dinoflagellates and their potential responses to changing Western Arctic Ocean

Résumé La structure physique de l'Océan Arctique est sélective pour différentes populations microbiennes, avec des implications majeures pour la qualité nutritionnelle et sa disponibilité pour les niveaux trophiques supérieurs, la séquestration du carbone et les cycles de nutriments inorganiques. En particulier, la région du gyre de Beaufort, située dans le bassin ouest de l'Arctique Canadien est fortement stratifiée avec des séparations des masses d'eau dans les premiers 200 mètres qui contiennent des communautés microbiennes eucaryotes distinctes pouvant révéler différentes fonctions écologiques. Cependant, on en sait peu sur les espèces, les lois d'assemblages et, par conséquent, sur la résilience de ces assemblages aux perturbations climatiques qui affectent actuellement les propriétés physiques de l'océan. Afin de répondre à cette problématique, nous avons ciblé les régions situées au Nord du bassin Canadaet de la crête de Northwind, dans une zone éloignée des influences anthropogénique non climatiques, afin d'identifier les processus influençant la structure des communautés dans la partie supérieure de la colonne d'eau. Nous avons collecté des échantillons dans la couche des eaux polaires modifiée de surface (PML), la couche de maximum de chlorophylle sous- surface (DCM) située au sein des eaux du Pacifique d’été (PSW) ainsi que dans la couche plus froide des eaux du Pacifique d’hiver (PWW). Nous avons porté une attention particulière aux dinoflagellés, considérés comme ubiquistes dans les 200 premiers mètres de la colonne d'eau en Arctique, afin d'estimer l'influence relative des processus déterministes et stochastiques sur l'assemblage des communautés. Nous avons d'abord identifié les dinoflagellés en utilisant une approche de séquençage haut débit de la région V4 de l'ARNr 18S avec une assignation taxonomique en utilisant à la fois les ADN et ARN comme modèles. Afin de déterminer le processus sélectif dominant au sein des masses d'eau, nous avons analysé la structure des communautés et les patrons de distribution des unités taxonomiques opérationnelles (OTUs) en utilisant des modèles phylogénétiques aléatoires. Nous avons trouvé que les processus déterministes sont dominants dans les eaux de surface ce qui contraste avec les eaux PMLou les processus stochastiques sont plus importants, suggérant que la communauté est le produit d'événements historiques. Les communautés de la PSW et de la DCM ont été influencées à la fois par des processus déterministes et stochastiques ce qui suggèrent une vulnérabilité aux changements environnementaux. Sous certaines conditions, des phylotypes différents, potentiellement des dinoflagellés mixotrophes lutant contre la limitation par la lumière et les nutriments, pourraient persister plus longtemps dans un scénario ou la PSW s'approfondirait.

Abstract

The physical oceanographic structure of the Arctic Ocean is associated with distinct microbial populations with implications for food quality and availability for higher trophic levels, carbon sequestration and inorganic nutrient cycles. In particular, the Western Arctic Canada Basin - Beaufort Gyre region is strongly stratified with separate water masses in the upper 200 m that contain distinct microbial eukaryotic communities and potential ecological functions. However, little is known about the species, rules of assembly, and by implication, the resilience of assemblages, to climate perturbations currently affecting physical oceanic properties. To address this, we targeted the northern reaches of the Canada Basin and Northwind Ridge, in an area that is distant from non-climate driven anthropogenic influences, with an aim to identify processes driving community structure in the upper water column. We collected samples from the surface polar mixed layer (PML), the deep chlorophyll maxima (DCM) within Pacific Summer Water (PSW) and the colder Pacific Winter Water (PWW). We specifically examined dinoflagellates, which are ubiquitous in the upper 200 m of the Arctic, to infer the relative influences of deterministic and stochastic processes of community assembly. We first identified the dinoflagellates using high throughput amplicon V4 18S rRNA sequencing and taxonomic binning using both DNA and RNA as templates. To determine the dominant selective processes within the water masses, we analyzed community structure and distribution patterns of operational taxonomic units (OTUs) using randomized phylogenetic models. We found that deterministic processes dominated in the surface PML which was in contrast to the deeper PWW water mass, where stochastic processes were stronger, suggesting a community that was the product of upstream events. The PSW– DCM communities were influenced by both deterministic and stochastic processes suggesting vulnerability to change. Under current conditions, different phylotypes of putative mixotrophic dinoflagellates persisted under light- and nutrient limitations, and could persist longer in a scenario where the PSWand associated nitracline deepens.

4.1 Introduction

Pelagic protist communities support higher trophic levels in the Arctic, with the majority of protist biomass including microzooplankton in the upper 200 m, making this an active zone for trophic transfer (Nishino et al., 2008). Although Atlantic Water occupies the the deep basin, the upper 200 m of the Canada Basin, one of the two major deep basins of the Arctic Ocean, is stratified becausefresher and colder surface Polar Mixed Layer (PML; Shimada et al., 2005) water is underlain by the warmer, more saline Pacific Summer Water (PSW) and colder Pacific Winter Water (PWW; Steele, 2004; Shimada et al., 2005; Carmack et al., 2016). Sufficient light for photosynthesis penetrates only into the upper 40 to 60 m, and in the absence of upwelling events, nutrients are trapped in the deeper the Pacific- origin waters. Over much of the summer,maximum phytoplankton biomass and production occurs at the top of this Pacific halocline system (Lee and Whitledge, 2004; Steiner et al., 2016), forming therecurring summer subsurface or deep chlorophyll maximum (DCM) layer associated with PSW(McLaughlin et al., 2005; Martin et al., 2010, 2013).The environment of the upper mixed layer is the most physically variable, with rapid changes in temperature, periodic freshening with melting ice, and drifting ice floes, exposure to wind-induced turbulence and high levels of ultraviolet radiation in summer. The deeper Pacific waters are relatively more stable and temperature–salinity (TS) and nutrient characteristics change little over long distances in the Western Canada Basin (Nishino et al., 2008; Shimada et al., 2005). Below the DCM where PWW dominates light is limiting (Li et al., 2009). Thus, each of the three layers (surface, DCM in PSW, PWW) could be considered distinct habitats that likely selectfor dissimilar microbial assemblages as reported in the adjacent Beaufort Sea and Amundsen Gulf, which share the same water mass structure (Comeau et al., 2011; Monier et al., 2015).

Arctic food webs are often reported as short and highly dependent on specialist photosynthetic and heterotrophic interactions (e.g. Forest et al., 2011). Although phytoplankton such as diatoms provide much of the fixed carbon and nutrition during spring and ice edge blooms, they tend to be absent or rare for much of the year, and small flagellates and dinoflagellates tend to dominate biomass. Dinoflagellates also dominate most gene-based surveys (Terrado et al., 2009; 2011; Comeau et al., 2011; Monier et al., 2015; Onda et al., 2017) and make up to 37% of microbial eukaryotic cells in microscopy-based studies (e.g. Martin et al., 2010; Ardyna et al., 2011; Coupel et al.,2012). Dinoflagellates are taxonomically and trophically diverse, and only around half of described species are reported to have integrated chloroplasts (Saldarriaga et al. 2004; Taylor et al., 2008). Most chloroplastidic species investigated to date however retain the capacity for phagotrophy and are considered mixotrophic (Hansen, 2011; Stoecker, 1999). The number of mixotrophic taxa is likely underestimated since in addition to species with integrated chloroplasts, are those that host a

83 photosynthetic symbiont or temporarily retain prey chloroplasts through kleptoplastidy (Esteban et al., 2010; Kim et al., 2015, Flynn et al., 2013). Their heterotrophic-mixotrophic lifestyles have been implicated in the top-down control of some phytoplankton communities and indirectly control populations of bacterial decomposers by high consumption of small bacterivores (Irigoein, 2015; Jassey et al., 2016).

Recent changes in the Arctic, associated with climate, have been mostly reported for size fractioned phytoplankton and the seasonal distribution patterns of chlorophyll a (Li et al. 2009, Ardyna et al., 2014). These studies have been facilitated by the use of flow cytometry for counting chlorophyll containing cells and satellite imagery of the surface ocean. Such synoptic tools are not available for monitoring changes in heterotrophic cells or specific enough to detect potentially mixotrophic species such as dinoflagellates. However, given that recent studies in tropical and temperate systems indicate that increasing proportions of mixotrophic taxa may be directly linked to climate change (Jassey et al., 2015; Lewandowska et al., 2014), it is imperative to identify heterotrophs and mixotrophs in Arctic microbial food webs. In the Arctic, mixotrophs could be favoured by light or nutrient limitation associated with the deepening of the nitracline due to increased freshwater input and stratification (Comeau et al., 2011; Bergeron and Tremblay 2014; Brown et al., 2015). Modelling and field-based studies of oligotrophic open oceans indicate that mixotrophic nutrition contributes flexibility and stability to overall ecosystem functions (Jost et al., 2004; Troost et al., 2005; Mitra et al., 2014; 2016). Further understanding of the ecology of dinoflatgellate communities, in particular the assembly rules governing their spatio-temporal distribution/biogeography will provide important baseline data useful in predicting the ecosystem dynamics under climate change scenarios (Stegen et al., 2012; Mitra et al., 2014).

Here we investigated dinoflagellate communities to extract patterns of dinoflagellate distribution and assemblage in the upper 200 m of the Arctic under well-defined environmental conditions. We first used high throughput amplicon sequencing targeting the V4 region of 18S rRNA gene (here as rDNA) and 18S rRNA (here as rRNA), providing a high level of detail of the microbial community at the Northern limits of the Canada Basin, where non-climate related anthropogenic influences are minimal. Samples were collected in 2012 and 2013 from three distinct water masses. We applied randomized phylogenetic models to generate null dinoflagellate communities that were compared with phylogenetic clustering of Arctic communities from the three depth categories. Resulting patterns were then used to evaluate the dominant selection processes that governed the dinoflagellate assembly. Phylogenetic and alignment approaches established a high-level phylogeny, which we related to potential functional roles based on literature searches. We then applied multivariate statistics to associate patterns with

84 environmental selectors, and used ratios of rRNA to rDNA to infer potential life stage differences of taxa within water masses. We then examined potential trajectories of the SCM community subjected to a deepening of the nitracline.

4.2 Materials and Methods 4.2.1 Study sites, sample collection and processing Nucleic acid samples and ancillary data were collected from the Northwind Ridge, and western and northern regions of the Canada Basin (Figure 4.1) in August 2012 and 2013.Sampling was carried out on board the Canadian Icebreaker CCGS Louis S St. Laurentusing a Rosette Systems equipped with conductivity, temperature and depth (CTD) profilers (Sea-Bird Electronics, Inc., California, USA) coupled with dissolved oxygen (DO), coloured dissolved organic matter (CDOM), chlorophyll a (Chl a) fluorescence and relative nitrate (ISUS probe, Satlantic) sensors. Sample collection in 2012 (6 stations: TU-1, CB9, CB11, CB16, CBN and CBN2) was described previously (Chapter 3) and the same methods were used in 2013 (5 stations: TU-1, CB9, CB11, CB12, CB16) with slight differences. In 2012, profiling and water collection (10-L Niskin bottles) were carried out from the same cast using the Main Rosette system. However, due to logistical limitationsin 2013, profiling was done during downcast on the Main Rosette while sample collection from a second Rosette system deployed from the foredeck usingtime-dependent auto-firing mechanism (AFM, Sea-Bird). The rosette and Niskin bottles were lowered down to the target depths calculated fromtheocean surface. Consistency of the targeted and actual sampling depths were verified by the pressure data in the AFM trigger file and by comparing bottle salinities of the 2 rosettes. Salinity samples were analyzed onboard using a Guideline Portasal and IAPSO standard seawater (McLaughlin et al., 2005).On board, dissolved inorganic carbon (DIC) was also measured using a colourimetric method (Coulometrics Model 5011, VINDTA, Germany), while dissolved oxygen (DO) wasdetermined by Winkler-based titration (Kit B, Scripps Institute of Oceanography, USA).Nutrient analyses (nitrate, phosphate, silicate) were also measured on board using a 3-channel Technicon Auto Analyzer following the methods of Barwell-Clarke and Whitney (1997). Nutrient samples were not taken at some of the 100 m PWW depths, and values were extrapolated from available samples collected from the same water mass and nearest depth. All physico- chemical data and CTD logs are available at the BGEP site (http://www.whoi.edu/beaufortgyre/expeditions) and other details of the cruise are reported in Krishfield et al. (2014).

Around 6L of seawater for nucleic acid extraction were collected from the Niskin bottles into acid- rinsed carboys. Samples were collected from near surface at 5-6 m, considered as the polar mixed layer (PML), the subsurface chlorophyll maxima (SCM) that coincides with PSW from 52-87 m, which

85 wasidentified from Chl a fluorescence, and from92 m to 158 min PWW. The water samples were serially filtered through a 50-µm mesh, and material collected onto 3 µm pore size 47–mm polycarbonate (PC) filters (AMD Manufacturing) and then a 0.2 µm Sterivex™ filter unit (Millipore;) using a peristaltic pump (Cole-Parmer, USA). The PC filters were placed into 2-mL sterile microfuge tubes containing 1.8 mL RNALater (Ambion™). The RNALater was added to the Sterivex filters using pipette. Following 20-60 min at room temperature, both filter typeswere stored in -80° C.

Samples for flow cytometry were collected following Li and Dickie (2001) and Li et al. (2011). Briefly, a 100 mL PC bottle was rinsed with sample water and then filled directly from the Niskin bottles. Duplicate 1.8 mL aliquots were immediately transferred to 2.0 mL cryovialsand preserved with 200 µl 10%EM grade paraformaldehyde (Sigma-Aldrich). The cryovials were briefly vortexed and left at room temperature in the dark for at least 10 min and then kept at -80° C. The 2012 samples were analyzed using a FACSort (Becton Dickinson, San Jose, CA) equipped with a 477-nm argon-ion laser. Following cross calibrations, the 2013 samples were processed using an EPICS ALTRA flow cytometer (Beckman Coulter; Tremblay et al., 2009) and a BD AccuriTM C6 flow cytometer (BD Biosciences) both equipped with blue (488 nm) laser. Phytoplankton were distinguished by Chl a autofluorescence while bacteria were first stained with nucleic acid dye Sybr-Green (Life Technologies, USA).

4.2.2 DNA-RNA extraction, sequencing and bioinformatics processing DNA and RNA were extracted from the same filters as described in Dasilva et al. (2014) using the RNA/DNA Mini-Kit (Qiagen). RNA was first converted to cDNA using the High Capacity Reverse Transcription Kit (Applied Biosystems) before amplification following the kit instructions. The V4 region of the rRNA gene and cDNA were then amplified using primer pairs E572F (forward) and E1009R (reverse) (details in Comeau et al., 2011) conjugated with MiSeq© linking primers to generate one library (collection of amplicons) per sample. Each library was then tagged with TruSeq© and Nextera© barcodes in the 5’-ends of both amplicon strands as described in Chapter 2. Amplicons were pooled with a final concentration of 121 ng µl-1 before paired-end multiplex sequencing using a MiSeq- Illumina© at IBIS/Université Laval Plateforme d’Analyses Génomiques (Quebec, QC, Canada). Short read libraries are available from the NCBI GenBank Short Reads Archives (SRA) under BioProject Canada Basin Marine Microbes PRJNA362945, BioSamples SAMN06285023-5082 (2013).

HTS reads were demultiplexed, and barcodes and primers removed in CASAVA v.1.8.2 (Illumina). Quality filtering for short sequences, de novo chimera removal and paired-end reads with at least 25 bp overlap were merged using options in Quantitative Insights into Microbial Ecology (QIIME; Caporaso et al., 2010). Operational taxonomic units (OTUs) were clustered at 98% similarity and representative

86 readspicked in UPARSE (Edgar, 2010). Representative sequences were then aligned in PyNast (Caporasso et al., 2010) and a phylogenetic tree generated using FastTree (Price et al., 2010). UPARSE (Edgar, 2013) was then used to generate an OTU matrix, and taxonomy assignments in mothur with a minimum confidence threshold of 0.8 (Schloss et al., 2009). These assignments followedthe expanded 18S rRNA gene reference database v.1.1 (Lovejoy et al., 2016), which includes curated Arctic and environmental sequences (Chapter 2, Onda et al., 2017).

All metazoans and reads that were unclassified below “Eukaryota” were removed, and the OTU matrix was rarefied to a uniform depth of 16,044 reads per sample for abundance and diversity comparisons. The rarefied OTU matrix was considered as the total microbial community and inspected for differences and consistency. Subsequently, ‘dinoflagellate’-related reads (see Chapter 3) were retained for the detailed comparative ecological analysis. We further limited our analysis to OTUs that were present in both DNA and RNA at least once, irrespective of which size fraction the OTUs originated (Chapter 3). The new OTU matrix was used for downstream statistical analyses. The combined OTUs from the two fractions provided acomprehensiveview of the dinoflagellate community composition since OTUs that appear only in one fraction were represented. OTUs that clustered with environmental sequences were further identified using the Random Axelerated Maximum Likelihood - Evolutionary Placement Algorithm v.8.1 (RAxML-EPA) for in-depth phylogenetic placement as described in Chapter 3 (this thesis). Sequences were then classified based on the nearest annotated reference sequences and functional groups were assigned based on literature searches

4.2.3 Community diversity and statistical analyses Dinoflagellate community beta-diversity was assessed on 18SrRNA and rDNA and computed from the unweighted UniFrac distances (Lozupone and Knight, 2005) and represented with Principal Coordinate Analysis (PCoA; Gower, 1966) in QIIME (Caporasso et al. 2010). Linear regressions were then applied to the PC1 coordinates and environmental correlates were identified to determine structuring drivers. Non-random community distribution was tested using a Checkerboard score (C- score) under the null hypothesis parameter in package ‘bipartite’ in R (Dormann et al., 2008).

The Nearest Taxon Index (NTI), which combines phylogenetic distances and species occurrences, was used to assess phylogenetic clustering and infer dominant selection processes acting on the communities. The NTI based on the mean nearest taxon distances (MNTD) of the tips of the phylogenetic tree (‘tree label’) was used since the study was focused on a defined taxonomic group (dinoflagellates). A phylogenetic tree was first generated from dinoflagellate OTUs that were present in both sample types (rRNA and rDNA) and used to compute for the MNTD using the ‘ses.mntd’

87 function in the R-package ‘Picante’ v.1.6under the null modelparameter (null.model = ‘taxa.labels’) resampling 999 times,from which NTI was computed (mntd.obs.z × -1). The model was run using the combined small and large rarefied OTU table. An alternative null model to generate NTI indexes was also compared based on the independent swap algorithm (null.model = ‘independentswap’; Gotelli and Enstminger, 2003; Kembel, 2009). For comparison, Net Relatedness Index (NRI) based on distances from the inner branches was also computed. To investigate trends between selection pressures and α- diversity, we computed Faith’s Phylogenetic Diversity (PD) based on the total branch length of the OTU phylogenetic tree using the package ‘Picante’ (Kembel et al., 2010). NTI values were regressed with the environmental parameters or abiotic factors to determine the likely drivers of phylogenetic structuring. To investigate the effects of environmental heterogeneity (here as ‘variability’) on community assembly, we first computed the pairwise Bray-Curtis dissimilarity matrix using the ‘vegdist’ function in Vegan package in R (Dixon and Dixon, 2003) of the more variable environmental parameters depth, temperature, salinity, DO, Chl a, transmissivity, nitrate, silicate, phosphate, conductivity and CDOM. Then, the mean Bray-Curtis (subscript BC) values were determined and the standard deviations (SD) were calculated for each sample. We then used the SDBC as the index of within-depth variability; higher

SDBC within water masswas interpreted as high station to station (sample to sample) variability, which were then regressed against the NTI per sample and their means per depth. To further detect global spatial autocorrelation of parameters within and between depth categories, Global Moran’s I values were calculated using water mass parameters as indicators (salinity, temperature, nutrients, CDOM, Chl a) by means of the ‘Moran’s.I’ function in ‘ape’ package in R (Paradis et al., 2004). Samples were also clustered based on the environmental variables using Principal Component Analysis (PCA) in Past 3.0 (Hammer et al., 2001). Spearman’s rank correlation test was carried out in the Vegan package in R (Dixon and Dixon, 2003). Phylogenetically related taxa are more likely to share traits that are adaptive to a particular environment (Webb, 2002), and by proxy, closely related species would be expected to co-occur in a given environment. To determine any depth-dependent patterns, we identified the relative level of activity of taxonomically binned dinoflagellate OTUs in the 3 depth categories by plotting rRNA abundances as a heatmap, clustered by sample (column) using UPGMA in the package ‘ggplot2’ in R (Wickham, 2009). We then examined the contribution of different environmental processes to the activity of the communities by comparing the rRNA to rDNA ratios. OTUs were first pooled based on their lowest taxonomic rank (i.e. genus or clade), and rRNA to rDNA ratios were transformed to log ratiosand plotted. Correlations of specific taxa with environmental parameters were tested and plotted as biplots using Canonical Correspondence Analysis (CCA) with forward selection implemented in

CANOCO 4.5 (ter Braak and Smillauer, 2002). Analysis of variance (ANOVA), significant differences (t-tests and Mann Whitney U-test) and linear regressions were implemented in PAST 3.0 (Hammer et al., 2001).

4.3 Results 4.3.1 Environmental characteristics and general features The two years were comparable and general physical features, nutrients and Chl a concentrations did not significantly differ between years within the target layers. Samples collected across Canada Basin (Figure 4.1) grouped by physical attributes into the three distinct depth categories; the surface Polar Mixed Layer (PML) with salinity from 25.26 to 28.78; the PSW where the deep chlorophyll maxima developed (DCM) with salinities from 31.08 to 32.35; and the PWW with salinities from 31.86 to 33.15. TS diagrams of CTD profiles confirmed water mass assignments. Biologically, based on fluorescence, Chl a concentrations in the SCM were 0.2-1.0 mg Chl a m3 and higher than in the surface or deeper layers, where concentrations were < 0.1 mg Chl a m3(Table 4.1). The depth of the DCM also varied within the sampling area, and occurred deeper and had weaker fluorescence signals near the center of the Canada Basin compared to in and near the Northwind Ridge. Allnutrient concentrations were greater in the DCM and PWW compared to the surfacewhere NO3¯ was< 0.01

3 3- 3 3 mmol m , PO4 was < 0.6 mmol m , and SiO3 was< 2.8 mmol m . The stations showed some spatial autocorrelation (insignificant positive Moran’s I) for all depths using the physical and chemical parameters indicating samples geographically closer to each other also tended to be strongly similar within depth categories. The biologically labile parameters Chl a and CDOM had negative values (p>

0.01) indicating a lack of autocorrelation. The variability (SDBC) from station to station was greatest in the surface (Figure 4.2A), which had the highest SDBC values (0.079 + 0.1), significantly higher than the DCM (0.07 + 0.08; p< 0.01) and PWW (0.06 + 0.08, p< 0.001Mann-Whitney U test). Eukaryotic pico- and nano-phytoplanktonic cells had maxima in the DCM and lowest in the PWW and bacterial concentrations generally decreased with depth (Table 4.1).

4.3.2 Community Similarity and Phylogenetic structuring From the total of 128 libraries (size-fractionated rRNA and rDNA), over 7.5 million (7,587,456) high ca. 450 bp length quality reads remained, ranging from 16,044 to 494,758 reads per sample. Using 98% sequence similarity threshold for OTU clustering, there was a total of 2,402 OTUs from the dataset. Of these, 420 OTUs were identified as core dinoflagellates (see Chapter 3) with 497,700 reads, accounting for 24.23% of the total microbial eukaryotic reads in the rarefied resampling. Relative to the total microbial community, dinoflagellates tended to be more represented in the large

89 fractions than the small, and were more abundant in both surface and PWW than the DCM (Supplementary Figure S4.1). However, often many of the dinoflagellate rRNA reads of the same OTUs were more frequent in the small fraction while the rDNA reads were mostly in the large fraction, or the ther way around, indicating potential effects of fractionation (Supplementary Figure S4.2). To facilitate the analysis and lessen effects of fractionation, rarefied small and large fraction reads (16,044 per library) of the total community (all microbial eukaryotes) were combined, totaling to 32,000 reads per rRNA or rDNA of each sample. After combining the fractions, around half of the dinoflagellate OTUs (212) were recovered from DNA and RNA (Supplementary Figure S4.3), and the rest were only either detected in DNA (24.29%) or RNA (25.23%). OTUs that were not found in both templates were removed and not analyzed further. For the remaining OTUs, we calculated the log ratios of rRNA to rDNA, which we then used for other ecological analysis.

Phylogenetic beta diversity of dinoflagellate OTUs reflected strong depth category association (Figure 4.2B). Both the surface and PWW samples formed tight watermass defined clusters, regardless of sampling year. The DCM samples were more dispersed but still distinct. The one exception was the Stn. TU-1 PWW sample that clustered with other Canada Basin DCM samples. The C-test (p < 0.001) further showed the non-random distribution of the dinoflagellate OTUs and suggested community structuring. The axis 1 coordinates (PC1) of the PCoA contributed to around 40.15% of the variation with strong correlation with depth (R2 = 0.93, p< 0.01) and salinity (R2 = 0.91, p< 0.01), both reflecting the stratification in the Arctic Ocean.

Phylogenetic clustering based on NTI is used to infer potential processes shaping the community, where values < -2 are generally associated with homogenous selection, values from -2 to +2 are attributed to stochastic processes, and those > +2 suggest environmental selection or habitat filtering (Hardy, 2008; Stegen et al., 2012, 2013; Dini-Andreote et al., 2015). Here, depth-dependent differences inferred from the NTI within-depth category (mean p= 0.003) in the relative influences of stochastic and deterministic processes on dinoflagellate communities were found within the layers of the Arctic Ocean (Figure 4.3C). Surface samples were characterized by strong phylogenetic clustering of component OTUs, which generated a mean NTI greater than +2 for both RNA (4.3 +1.39) and DNA (5.24 +1.18), suggesting the dominance of deterministic processes in the surface while stochastic processes dominated in the PWW with mean NTI less than 2 for both RNA (1.76 +1.39) and DNA (1.38 +0.86). Both processes were detected in the DCM, with a mean NTI of 2.43+ 1.39 and 2.03 +0.81 for RNA and DNA-derived libraries, respectively. No overdispersion (NTI < -2) was observed in any of the samples. The same pattern was found using the NRI and the alternative ‘interswap’ method (Supplementary Figure S4.4). The NTI distributions corresponded to differences in several

90 environmental parameters. In particular, NTI values were all significantly negatively correlated (p< 0.001 based on regression) with markers of water masses or depth categories such as salinity (R2 = 0.52), nitrate (R2 = 0.29), silicate (R2 = 0.26) and phosphate (R2 = 0.38), which all increased with depth.

Environmental variability represented by the mean within-depth SDBC was also positively correlated with the mean within-depth NTI, indicating linear relationship between environmental variability and selection pressure, which was observed in both DNA and RNA–derived libraries (both R2 = 0.99). NTI was also significantly negatively correlated with PD (Spearman’s rho, r = -0.346, p < 0.01). PD was highest in the PWW (4.02 + 0.68) based on combined rRNA and rDNA diversities, followed by the surface (2.89 + 0.45) and DCM (2.83 + 0.89).

4.3.3 rRNA to rDNA ratios and taxonomic distribution The dominant selection processes and potential drivers with depth were accompanied by the changes in the rRNA to rDNA read ratios (Figure 4.3), which suggest variability in the dinoflagellate life stages related to environmental conditions. Since all selected OTUs were both present (rDNA) and presumably contained ribosomes (rRNA) in at least one depth, and the reads had been corrected for potential biases of size fractionation and amplification-sequencing, we considered rRNA to rDNA ratios > 1 as indications of high potential for metabolic activity while the ratios < 1 indicated less metabolically less active taxa. In the surface, most taxa had ratios near 2:1 (slope m = 0.48) with a mean of 1.77 + 2.5 (Figure 4.3A). Most of these were also putatively active in the DCM where the ratios were ca. 15:1 (m = 6.59; Figure 4.3B), which was significantly greater compared to the surface and PWW. In the deeper PWW, most taxa were also closer to 2:1 (1.77 + 2.5, m = 1.55) rRNA to rDNA. The ratios indicate less activity in the surface and PWW compared to the SCM (Figure 4.3C). One exception was the ratios for a Gymnodinium sp. Art135d30, an environmental clade that were consistently lower, suggesting low activity in all depths. Other taxa had distinct depth distribution such as Gymnodiniumsp.SW072 and “Other Gymnodiniales” where the ratio was lower in the deeper samples, whereas the ratiosfor Pseudopfiesteria-like OTUs increased with depth.

At theOTU level, unclassified dinoflagellates consistently had the highest number of OTUs over all depths (Figure 4.3). Based on rRNA reads, dinoflagellates were more abundant relative to the entire communityin the surface and PWW compared to the DCM where more phototrophs including diatoms, Phaeocystales, pelagophytes and heterotrophic taxa such as Acantharea and ciliates dominated (Supplementary Figure S4.1). CCA ordination (Supplementary Figure S4.5) of the most abundant rRNA OTUs (Figure 4.4) and less abundant taxa (Supplementary Figure S4.6) showed significant association of potential activity with specific environmental variables. For example, mixotrophic genera such as those from Gymnodiniales, Protoperidinium andNematodinium were more common

91 under high light or high nutrient conditions in the surface and DCM but were very low in the PWW samples. Other mixotrophic taxa such as Pseudopfiesteria and Adenoides were associated with lower light in the DCM and higher nutrients in PWW. The putatively strictly heterotrophic Ankistrodinium- likegroup appeared active in all depths with maxima in the deeper waters, indicating adaptation to multiple conditions. Identities of the OTUs were derived from phylogenetic placement (see Chapter 3).

4.4 Discussion The physical oceanographic structure in the Canadian Arctic limits exchange between water masses (Winton, 1988) where distinct communities are potentially selected by the environmental factors within each layer (e.g. Hamilton et al., 2008; Galand et al., 2010; Terrado et al., 2011; Comeau et al., 2011; Monier et al., 2013; 2015), and likely contributes to the biodiversity in the Arctic (Lovejoy et al., 2017). Here, we investigated the taxonomic and phylogenetic structuring of dinoflagellates that are a key resource for zooplankton over much of the year (Falk-Petersen et al., 2009; Seuthe et al., 2011) within the upper 3 main water masses of the Canada Basin in the Arctic Ocean. There are caveats associated with the use of 18S rDNA and rRNA to estimaterelative abundance including variability in gene copy number among species and groups, cell size, and the physiological state of the cells (Zhu et al., 2005; Godhe et al., 2008; Gong et al., 2013). Dinoflagellates in particulartend to be larger cells (Hoppenrath, 2016) and have large genomes (LaJeunesse et al., 2005; Lin, 2006; Godhe et al., 2008; Hou and Lin, 2009). Nevertheless, the use of rRNA and rDNA, and the rRNA to rDNA ratios, remains a tool for investigating the diversity and variability of microbial taxa relative to environmental gradients (Hu et al., 2016). To overcome these potential issues, we removed potential artefactual rDNA and doubtful rRNA OTUs, and limited our analysis to OTUs that were present in both rDNA and also active in protein synthesis (rRNA). Further, by limiting our analyses to dinoflagellates, we minimized potential inter-taxon variability of 18S gene copy number and cell size.

4.4.1 Environmental variability and selection processes Overall, we found evidence that both deterministic and stochastic processes influenced the assembly of dinoflagellate communities, and the relative weight of the two processes differed among water mass categories (Figure 4.2). Stochastic events include unpredictable disturbances that affect mortality and growth. Suchevents are probabilistic and resultin ecological drift when immigration is limited. In contrast, deterministic processes include niche-based selection or habitat filtering imposed by the environment, resource availability, and antagonistic or synergistic species interactions (see Chase and Myers, 2011). This framework is used to predict community responses to changing

92 conditions from the perspective of community ecology and evolutionary dynamics (see Mouquet et al., 2012). The dominant process is thought to depend on the heterogeneity or variability of the landscape and the distribution of resources (e.g. Vellend et al., 2014; Bar-Massada et al., 2014), and reflects the stage of community succession over time (Stegen et al., 2012; Dini-Andreote et al., 2015).

In the surface samples of the the Canada Basin, deterministic processes dominated and could be attributed to strong environmental selection due to nutrient limitation and variability in surface salinities, limiting the range of species able to grow. This environmental variability, estimated from the

SDBC, which was a measure of variability in specific parameters and physical factors among stations within depth category favoured deterministic processes in the surface and may have driven selection for dinoflagellates with a wider range of trophic responses. In contrast, in the less variable deeper PWW, stochastic processes were dominant.

The PWW in the Canada Basin is more isolated from surface processes because of greater density stratification that does not break down even under ice-free conditions and increased sea surface- wind interactions (Lincoln et al., 2016). Conservation in PWW TS properties was reflected in the low variability of the samples (lowest SDBC; Figure 3A) despite having large range of sampling depths (92- 182 m). The similarity in conditions and distribution of resources provided multiple taxonomic groups suitable environments for their establishment over large distances (Bar-Massada et al., 2015), and increased probability of stochastic events. However, Pacific waters from the shallower Bering Strait to Chukchi Sea then to the deeper Canada Basin (Steele, 2004; Shimada, 2005; McLaughlin et al., 2005) would have been the source of taxa in PWW and there may have been habitat filtering during the transport, with stochastic processes dominating and operating on the surviving community under the relatively stable conditions inside the Canada Basin (Ferrenberg et al., 2013; Powell et al., 2015). The absence of strong within depth distance-decay patterns supports this notion (Martiny et al., 2006; Monier et al., 2015). Unlike the surface waters that are modified seasonally, the Pacific waters inside the Beaufort Gyre have a residence time of around 10 years (MacDonald, 2000), sufficient for stochastic processes to dominate after the initial selection for heterotrophic species dominating in these aphotic waters (Ferrenberg et al., 2013; Powell et al., 2015).

Between these two extremes the DCM community, with NTIs ranging from +1.5 to +2.5, would have been governed by both deterministic and stochastic processes, suggesting adaptation to current conditions and a contribution from more random selection. Both deterministic and stochastic processes have been previously inferred to operate within a single ecological community and the relative contributions of eachhas been used to attribute environmental perturbation patterns (Hubbell, 2001; Chase and Myers, 2011; Stegen et al., 2013; Ferrenberg et al., 2013;Gravel et al., 2006; Kraft et al.,

2015). Due to the depth of the DCM, factors that influence light penetration (ice cover, CDOM, Chl a) could have also contributed environmental variability and consequential deterministic processes. Narrow distribution or low availability of key resources can also limit the number of species able to establish or proliferate, thus increasing deterministic effects (Horner-Devine and Bohannan, 2006; Kembel et al., 2010; Amaral-Zettler et al., 2011; Bar-Massada et al., 2015). However, the stability of nutrient input from below and a homogenous light field from above would result in a stochastic influence on species composition. The water masses represent environmentally dissimilar ecosystems because of the different histories and environmental influences and the strong relationship between variability and selection processes (SDBC vs. NTI) in the three watermasses was consistent with the continuum hypothesis, where both stochastic and deterministic processes influence the assembly of microbial communities (Gravel et al., 2006; Kraft et al., 2014) but is dependent on environmental heterogeneity (Dini-Andreote et al., 2015; Bar-Massada et al., 2015). Similar observations were reported for prokaryotic communities in other stratified systems such as wells and subsurface soil environments (Stegen et al., 2012, 2013; Dini-Andreote et al., 2015), indicating that rules governing spatial patterns in the relative significance of selection processes may be similar across taxa and ecosystems. The PML and PWW represented ends of the spectrum where variability decreased with depth. Higher variability in the PML (SDBC) was found within the narrow range in sampling depths (5- 8 m; Figure 4A). The physical variability would have been due to local weather patterns, wind strength and the presence of ice at the different stations. On a basin wide scale, PML waters are also modified by the extent of precipitation, ice melt, conditions during freeze up with brine rejection, basin wide transport patterns of riverine inputs and upwelled waters from diapycnal mixing (Woodgate et al., 2005; Yang, 2006; Yamamoto-Kawai et al., 2008). Consistent with environmental selection were lower phylogenetic diversity (PD) of communities that persisted under these highly variable conditions. Generally, the negative correlation of NTI with PDindicated that deterministic events selected formore phylogenetically related taxa that were sharing the same traits adapted to a particular or set of conditions in the surface waters. For example, the deterministic processes operating in the PML, characterized by high variability and resource limited conditions, may become more prevalent during longer ice free conditionsin a future Arctic Ocean. Similar selective patterns within depth categories in the photic zone microbial community were previously reported by Monier et al. (2015) in the southern Beaufort Sea, water mass vertical structureconsists of the same watermasses as in the Canada Basin (Wassman et al., 2015).

4.4.2 Community structuting and variability in life cycle strategies For both surface and PWW, rRNA to rDNA ratios were near 2:1 (Figure 4.3), suggesting similarly low activity (defined as capacity for protein synthesis) in the two water masses. In the first case, surface waters were most likely limited by nitrogen availability, as reported previously for the upper waters of the summer Arctic (Tremblay et al., 2011; Nishino et al., 2008; Bergeron and Tremblay, 2014). Under these conditions, smaller phototrophs, including Mamiellophyceae (Supplementarty Figure S4.1B), with higher surface to volume ratios arefavoured (Menden-Deuer and Lessard, 2000; Barton et al., 2013). Gymnodinoids, which are potentially mixotrophic (Sherr and Sherr,2002) and dominated the dinoflagellatesin the the surface (Figure 4.4), but the low rRNA to rDNA ratio suggest that they were not rapidly growing, suggesting low prey availability as well as inorganic nutrient limitation. In the second case, the absence of light in the PWWmeans that predation would be aprimary means of energy acquisition. The low rRNA to rDNA ratios suggest that dinoflagellates were not doing well and that allothounous, either vertical flux or via advection, carbon did not support high bacterial abundance.

In the DCM, both nutrients and light were available, and overall showed higher beta diversity compared to the PML and PWW. In particular, there were a number of additional heterotrophic taxa (e.g. Rhizaria and Picozoa; Supplementary Figure S4.1) presumably enabled multiple trophic strategies within the dinoflagellates. The dinoflagellates that were present were apparentlyhealthy, as inferred from the high average ribosomal rRNA to coding gene ratio of of 15:1. The overall microbial eukaryotic community was more diverse compared to the surface and PWW, with dinoflagellateread abundancerelatively low compared to the total. Dinoflagellates have high respiratory losses due to their large cell sizes and genomes (Crawford, 1992; Geider and Osborne, 1989; Rizzo, 2003; Lopez- Sandoval et al., 2014), contributing to their lower growth rates (Tang, 1996) and may have been disadvantaged compared to other protist heterotrophs.

4.4.3 Dinoflagellate communities in the deepening nitracline The decreasing summer ice extent and increasing freshwater budget in the Beaufort Gyre are manifestations of climate-related changes in the Arctic (Krishfield et al., 2014). These changes will not only have a direct impact on the surface community, but could disrupt the light-nutrient balance at the DCM and alteringthe structure and composition microbial communities (Li et al., 2009; Comeau et al., 2011; Lovejoy, 2014;Monier et al., 2015). For example, lightlimitation will be more prevelant and heterotrophic processes are likely to increase as the DCM community follows a deepening nitracline. However, model simulations of the continued deepening of the nitracline (Steiner et al., 2016) suggest

95 that mixotrophic communities could be favored, and could maintain a microbial food web, providing nutrition for macrozoplankton, temporing the negative consequences for energy and carbon flow (Mitra et al., 2014).

The DCM offshore of Beaufort Sea and southern Canada Basinhas followed the deepening nitracline (Bergeron and Tremblay, 2014; Coupel et al., 2015). Although early seasonal loss of ice in these regions could increase overall light availability, the high CDOM and non-algal particles from riverine discharges and ice meltwater could reduce surface PAR by as much as8% (Bélanger et al., 2013; Guéguen et al., 2015). Mixotrophs could be expected to proliferate under low light or even in the absence of light if they switched to an entirely heterotrophic mode. However, we found that many of the putative mixotrophic taxa (e.g. Dinophysis, Other Gymnodiniales, Gymnodinium sp.; Figure 5) were less active or absent in the light-limited PWW, consistent with beingpoor competitorsagainst specialist heterotrophsand likely having lower grazing rates compared to the specialists (Rothhaupt, 1996; Calbet et al., 2011). In general, many mixotrophs require light for positive growth (Tittel et al., 2003; Ptacnik et al., 2016) and Hansen (2011) reported that only 3 out of 34 of mixotrophic dinoflagellates tested were able to grow in complete darkness. In part, this could be because some mixotrophs may use light for proton pumping and ATP generation (Wilken et al., 2014), while others are capable of photoregulation but this is coupled with reduced growth rate (Hansen et al., 2016). Such studies suggest thatas the DCM deepens the dinoflagellates would come to resemble a PWW-like community, with a shift towards heterotrophic dinoflagellates and by extention a shift towardsstochastic selection.

The distance between the depths where light is still available for net positive growth and the nitracline is expected to increase in the future as the nitracline deepens (Bergeron and Tremblay, 2014; Brown et al., 2015), resulting to nutrient drawdown in both DCM and surface layers. Towards this oligotrophication, the mixotrophic (plastidic) taxa including dinoflagellates could become the primary pico-nanoplanktonconsumer andalso fix CO2photsynthetically (Hartmann et al., 2012). Mixotrophic taxa are reported to have acompetitive advantage against specialists in low nutrient-low prey environments (Mitra et al., 2016). Extreme nutrient limitation in combination with some light could force the system to become more deterministic and extend the depth where communities are more phylogenetically similar. In support of this notion, Monier et al. (2015) reported the dominance of heterotrophic and mixotrophic dinoflagellates in the weak SCM in the coastal Beaufort Sea where nutients were lacking. In Baffin Bay, Lovejoy et al. (2002) also reported the dominance of mixotrophic- heterotrophic dinoflagellates after blooms of diatoms, when nutrients were depleted.

Aside from the DCM being governed by a light-nutrient intersection (Martin et al., 2010), we found that the DCM dinoflagellate communitieswere also governed by assembly rules with dynamic

96 interactions. These selection processes were strongly dependent onthe variability of the environment in each of the water masses. Thus, any significant phenomenon or event that could disturb or change this environment would also alter or affect assembly dynamics, making the DCM vulnerable to slight physical oceanographic changes. In both cases of light and nutrient limitations in the DCM, dinoflagellates could provide resiliency and stability for food and energy flows as potential producers, prey and grazers, which may also be true for other mixotrophic taxa. These groups could indirectly influence biogeochemical cycles by grazing bacteria, bacterivores and small plankton, and by linking the microbial loop with the higher trophic levels. Mesocosm, environmental and modelling studies in the tropics and temperate systems revealed that climate related changes are coupled with the shifts in ecosystem functions and structure that emphasize the increasing central role of mixotrophy (Jassey et al., 2003; Skjoldborg et al., 2003; O’Connor et al., 2009; Montagnes et al., 2010; Caron and Hutchins, 2013;Lewandowska et al., 2014; D’Alelio et al., 2016). These contributions have also been modeled to be significant in the global scale (Ward and Follows, 2016), which warrants careful re-examination on the role of these functional groups in the Arctic.

4.5 Conclusion Evaluating the responses of microbial communities to climate related changes is challenging due to the high diversity of unknown taxa and the complexity of their interactions. Here, we showed that the assemblies of the functionally and taxonomically diverse dinoflagellate communities across Arctic water masses are governed by both deterministic and stochastic processes. The DCM specifically was characterized by both stochastic and deterministic processes sensitive to physico-chemical changes associated with the changing climate, such as the deepening of the nitracline. The physical oceanographic changes could result to community replacement in upper waters that would determine overall ecosystem functions, with the selection of heterotrophs under light limitation and the mixotrophs following nutrient drawdown.

4.6 Acknowledgments

Samples were collected during the Joint Ocean Ice Studies – Beaufort Gyre Exploration Project (JOIS-BGEP) which was a collaboration between the Woods Hole Oceanographic Institution (WHOI) and the Institute of Ocean Sciences – Department of Fisheries and Oceans Canada (DFO-Canada). Major funding was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant, and the Network of Centers of Excellence ArcticNet funds to CL. DFLO received scholarships from Université Laval, the Canadian Excellence Research Chair – Remote

Sensing of Canada’s New Arctic Frontier (CERC) grant, Takuvik Joint International Laboratory, and additional support from the Fonds de recherche du Québec Nature et Technologies (FRQNT) to Quebéc-Océan.

The authors acknowledge the Captain and Crew of CCGS Louis S St. Laurent and the staff of IOS-DFO for their assistance and help during the cruise. We are also grateful toBill Williams, Sarah Zimmermann, Jane Eert, Mike Dempsey, Sarah-Ann Quesnel and Adam Monier for logistics, sampling and laboratory work. We also would also like to thank the IBIS bioinformatics teams especially Jerome Laroche for computational support.

Table 4.1. Station name (Stn), Dates of collection 2012-2013 (Date), collection Latitude and Longitude (Lat, Long), and physico-chemical parameters of the samples used for amplicon tag Illumina sequencing. Other column names refer to depth (Z) of sampling, temperature (T), salinity 3 3 3 3 (S), dissolved oxygen (DO), Chlorophyll a fluorescence (Chl a mg m ), nitrate (NO3¯ mmol m ), silicate (SiO4 mmol m ), phosphate (PO4 mmol m3), conductivity (Cond), colored dissolved organic mtter (CDOM mg m3), dissolved inorganic carbon (DIC x102 μmol kg-1), alkalinity (Alk x102 μmol kg-1), bacterial (Bact x104 cell ml-1), picoeukaryotic (Pico cell ml-1) and nanoeukaryotic (Nano cell ml-1) cell counts. na: not available.

Lat Long Z Temp Stn Date Sal DO Chl a NO ¯ SiO PO 3- Cond CDOM DIC Alk Bac Pico Nano (N) (W ) (m) (C⁰) 3 4 4 6.9 -0.26 25.26 358.73 0.04 0 2.50 0.50 21.16 3.07 17.7 18.3 26.82 1660 193 69.3 -0.05 31.20 363.18 0.45 1.84 6.29 0.93 26.16 4.60 20.68 21.8 12.67 1092 521 92.8 -0.76 31.86 335.84 0.15 6.55 13.81 1.32 26.11 5.15 21.13 22.0 15.32 203 138 7.5 -0.66 26.62 387.34 0.05 0.01 2.65 0.57 22.19 3.16 18.27 19.3 21.28 1136 67 68.8 -0.47 31.66 349.15 0.30 4.85 10.47 1.15 26.10 4.77 20.72 22.0 14.69 921 345 6.8 -0.77 26.66 386.84 0.03 0 2.6 0.55 22.14 3.20 18.4 19.3 20.09 1367 109 79.4 -0.86 31.89 340.00 0.16 6.7 13.13 1.3 26.04 4.87 21.04 22.1 7.84 304 122 104.7 -1.29 32.32 314.09 0.04 11.1 23.18 1.63 26.03 5.19 21.49 22.3 9.85 44 29 6.6 -0.74 27.35 384.25 0.07 0 2.36 0.57 22.69 3.38 18.73 19.8 22.13 1217 120 56.4 -0.37 31.30 367.28 0.62 2.08 6.6 0.96 25.99 4.71 20.54 21.9 14.04 1962 870 117.0 -1.28 32.36 309.99 0.05 11.42 24.09 1.66 26.07 5.18 21.43 22.3 9.19 34 16 7.0 -1.34 28.32 401.36 0.05 0 2.74 0.64 22.91 3.20 19.18 20.2 34.46 1401 104 87.7 -1.42 32.35 324.90 0.08 12.19 26.34 1.74 25.94 5.21 21.6 22.3 13.23 65 18 153.4 -1.49 33.15 276.09 0.05 16.19 39.50 2.01 26.51 5.16 22.03 22.7 12.28 n/a n/a 6.7 -1.40 28.78 401.36 0.06 0.19 2.83 0.66 22.54 3.11 19.95 21.0 16.12 944 101 64.1 -1.13 31.95 324.90 0.28 7.52 14.83 1.35 25.80 4.89 21.32 22.1 15.36 745 272 153.6 -1.49 33.18 276.09 0.05 16.02 37.35 1.97 26.54 5.25 22.35 22.7 8.08 na Na 6.2 -1.13 26.28 394.75 0.04 0.07 2.79 0.54 21.66 2.89 17.9 19.2 4.24 0 0 54.8 0.16 31.16 365.90 0.96 1.68 6.5 0.91 26.19 4.22 20.24 21.9 2.84 420 250 116.4 -1.18 32.35 317.40 0.04 11.19 23.03 1.62 26.13 4.87 21.29 22.4 4.31 130 110

CB9 12-08-13 78.02 -150.11 5.6 -1.41 27.27 397.70 0.05 0.01 2.56 0.58 22.17 2.92 18.8 19.7 6.34 50 10 65.9 0.03 31.08 377.42 0.43 1.17 5.57 0.87 26.07 4.17 20.4 21.8 3.61 290 290 116.7 -1.21 32.29 317.80 0.05 10.76 22.18 1.6 26.07 4.81 21.58 22.3 3.95 210 50 5.4 -1.41 27.78 403.15 0.06 0.04 2.52 0.59 22.28 3.01 18.15 20.0 10.47 210 10 84.0 -0.90 31.92 332.18 0.13 7.88 14.8 1.36 26.00 4.56 20.49 22.1 9.10 800 600 113.6 -1.27 32.33 317.00 0.04 11.55 24.15 1.64 26.05 4.80 21.01 22.4 7.27 100 80 6.1 -1.40 27.20 398.50 0.05 -0.04 2.46 0.56 22.12 2.93 18.78 19.6 4.36 40 0 62.7 -0.06 31.13 369.87 0.61 1.34 5.59 0.89 26.07 4.27 20.58 21.9 3.24 970 450 116.5 -1.23 32.35 316.55 0.04 11.2 23.08 1.62 26.10 4.93 21.64 22.4 3.63 450 250 6.2 -1.49 28.32 399.66 0.05 0.02 2.78 0.6 22.89 3.06 na na 11.13 130 70 52.6 -0.18 31.18 373.71 0.42 1.66 6.46 0.9 25.87 4.06 na na 5.69 2140 780 108.3 -1.27 32.33 317.31 0.06 11.47 24.08 1.63 26.05 4.81 na na 5.32 160 40

100

Figure 4.1. Map of the stations in the northern and western regions of Canada Basin including the Northwind Ridge (Stn. TU-1) White circles indicate stations uniquely visited in 2012 and black in 2013, while gray circles indicate stations that were visited in both years.

101

Figure 4.2. Relationship between selection processes and environmental variability. A.) Bar graphs showing the range of variation of the standard deviations of within depth Bray-Curtis Dissimilarities (SDBC) based on depth, temperature, salinity, DO, Chl a, transmissivity, nitrate, silicate, phosphate, conductivity and CDOM values. SDBC in the surface was significantly higher than the DCM and PWW (Mann-Whitney U test) also indicating highest overall variability. B) Phylogenetic structuring of communities (beta-diversity) based on the presence-absence of OTUs (Unweighted UniFrac) also showed strong depth-category association clustered using PCoA. Shapes represent year of collection: 2012 (circles) 2013 (triangles) C.) Depth-dependent differences in the relative influences of stochastic and deterministic processes along the three water masses inferred from the within-depth NTI was also observed. All values were > -2 indicating lack of signals of overdispersion or homogenous selection. Colors represent depths in A-C (green=surface, blue=DCM, violet=PWW).

102

Figure 4.3. Log transformed rRNA:rDNA ratio based on rarefied read counts of each taxa binned on lowest possible rank represented by different colors on the right. The size of the circles corresponds to the number of OTUs of the same taxonomic ranking. A) surface, B) DCM and C) PWW.

103

Figure 4.4. Heatmap based on the level of activity (rRNA reads) of the most abundant OTUs binned to the lowest possible ranks. Distribution of these taxa mostly clustered by depth, where photosynthetic- mixotrophic groups were more abundant in the surface and DCM while heterotrophic taxa dominated in the DCM and mostly in PWW. Coloured bars represent depths (green= surface, blue=DCM, violet=PWW) and clustering was determined using UPGMA approach.

104

105

Supplementary Figure S4.1. Abundances of the different major taxonomic groups relative to the total microbial community in the different depths (surface, DCM and PWW – see text) and fractions (small = left; large = right) based on (A) rRNA and (B) rDNA. Dinoflagellates were more abundant in the surface and PWW than the DCM. Samples from 2012 and 2013 were arranged starting with the stations from the shallower Northwind Ridge (TU1) northward to CBN2. Included are OTUs present only in either rRNA or rDMA. Not included are the taxa lower than 0.01%.

106

Supplementary Figure S4.2. A summarized log10 representation of the fraction association of dinoflagellate OTU reads in small (0.2-3 um) over large (>3 um) fractions. Gray bars represent 1og10 rDNA reads and black are for log10 rRNA reads. This graph shows that in some OTUs, rRNA were more often found in the small fraction while rDNA reads were more associated with the large fraction, which may indicate effects of size fractionation artitifacts.

Supplementary Figure S4.3. Comparison of the distribution of rRNA (-log10) and rDNA (+log10) after aggregation of reads from small and large fractions of the same OTUs. It was apparent that there was a stronger correspondence in the rRNA and rDNA reads of some OTUs after aggregation while others were more abundant in the rRNA than their rDNA and vice-versa, which could indicate environmental filtering. Many OTUs however, were either only present in the rRNA or in the rDNA, which were all removed from the final dataset.

107

Supplementary Figure S4.4. A depth-dependent change in the relative influences of stochastic and deterministic processes (box plots) along the stratified layers of the Arctic inferred from the within-depth NRI was observed based on Net Relatedness Index (NRI) similar to the results of the NTI. Colours represent depths (green=surface, blue=DCM, violet=PWW).

108

Supplementary Figure S4.5. A triplot based on Canonical Correspondence Analysis (CCA) with forward selection showing the association of the samples (sites) and selected dinoflagellate taxa with environmental variables or factors including silicate (S), nitrate (N), phosphate (P), depth (D), coloured dissolved organic matter (CDOM), salinity (Sal), chlorophyll a (Chl a), dissolved oxygen (DO), nanoplankton (Nano), picoplankton (Pico) and bacteria (Bac). Colours represent depths (green=surface, blue=DCM, violet=PWW).

109

Supplementary Figure S4.6. Heatmap based on the relative abundance of rRNA reads of the less abundant taxa based on the lowest possible ranks. Coloured squares represent depths (green, surface; blue, SCM; violet, PWW) and clustering was determined using UPGMA (see text).

110

CHAPTER V. GENERAL CONCLUSIONS

Arctic marine ecosystems are highly dependent on the productivity of the microbial food webs, which means that knowledge of all of the components is a prerequisite in understanding overall energy and nutrient flows in the region. Compared to other protist groups such as phytoplankton and a few heterotrophic bacterivores, our understanding of the ecology of Arctic microzooplankton that are largely dinoflagellates and ciliates is limited. Although HTS-based surveys of microbial communities have been carried out in the Arctic (e.g. Terrado et al., 2009, 2011; Comeau et al., 2011; Monier et al., 2015, Wolf et al., 2015, Marquadt et al., 2016), ciliates and dinoflagellates were never examined in detail. The work presented in this thesis is the first, to our knowledge touse HTS data to focus on the diversity and biogeographyof ciliates and dinoflagellates. The general goals of this thesis were to fill specific knowledge gaps and answer outstanding questions on microzooplankton ecology, specifically to 1) investigate temporal dynamics of microzooplankton looking at both seasonal and interannual patterns and trends, 2) explore their diversity in the Arctic Ocean especially of the uncultivated taxa, 3) and understand the spatial variability and underlying rules governing their community assembly. Generating new knowledge on these issues provide us with new tools to understand and potentially predict responses of microbial communities towards the changing Arctic.

5.1 Synthesis of the study

In Chapter 2 (Article 1), we investigated the variability of ciliates and dinoflagellates across seasons in the surface and halocline, where the SCM forms, in Amundsen Gulf. We showed that as with other protist groups, ciliates and dinoflagellates exhibited strong seasonality and depth partitioning reflected in both DNA- and RNA-derived reads. These community changes could be partially explained by changes in salinity, temperature, light and nutrients over annual cycles. Relative to the total microbial community, relative microzooplankton abundances were highest in mid-winter and towards the end of summer, with both periods dominated by reads belonging to taxa matching mixotrophic- heterotrophic related microzooplankton. Their lower relative abundances during spring however, were consistent with limited sequencing space due to the dominance of other microbial eukaryotes (e.g. diatoms). We also observed depth partitioning between ciliates and dinoflagellates towards summer, where dinoflagellate reads were greater in the SCM while ciliates tended to dominate in the surface waters, where nutrientswere depleted. The ciliates in the surface may have also avoided potential grazingloss by dinoflagellate (Löder et al., 2011). We found recurring annual summer composition, suggesting consistency in microbial assemblages under similar environmental conditions. This supported our hypothesis that ciliate and dinoflagellate assemblages were likely strongly influenced by abiotic factors, and anomalies in those conditions would be reflected by the changes in microbial

111 community composition. For example, following the minimum ice record in 2007, the summer of 2008 was characterized by a significant increase in some mixotrophic ciliate taxa and decrease in other usually abundant groups. The summer of 2008 in Amundsen Gulf was dominated by smaller phytoplankton. The ability of some mixotrophic ciliates, such as Laboea, Strombidium and Monodinium, to exploitsmall prey dominating the low nutrient conditions would potentially providea competitive advantage over other taxa. The association of these taxa sithlow-ice events make them good candidate indicator species of such changes. Lastly, this study highlights the importance of long- term periodic sampling to detect trends and potential responses to large scale environmental changes, such as those related to climate change.

We identified potential novel unclassified environmental clades (Chapter 2) using alignment and phylogenetic approaches, suggesting high unexplored diversity within ciliates and dinoflagellates. However, compared to unclassified ciliates or “Other Ciliates”, we observed higher proportions of unclassified reads in core dinoflagellates or “Other Dinoflagellates” even in the pico-size fractions (0.2- 3 μm; Supplementary Figure S2.4), which were potential undescribed species or clades (Lin et al., 2006; Bescot et al., 2015). This was intriguing since apart from rare reports of small life stages, studies based on microscopy have not reported core dinoflagellate cells in the pico-size range. If true core picodinoflagellates exist, reports of high abundance could suggest an underappreciated but significant ecological role. This leads to the idea that interactions and potential functional roles must be investigated to deepen understanding of the microbial food webs. To address this, Chapter 3 (Article 2) focused on the diversity of dinoflagellates (Dinophyceae) in the fractionated HTS data from the Canada Basin and extended our investigations to the deeper Pacific Winter Waters. Since cell breakage during fractionation could contribute to the detection of taxa in both small and large fractions, and we tested the hypothesis that the abundance of the 18S rRNA gene readsreflected size-range associations. Specifically, we predicted that true large-celled Dinophyceae would havethe majority of reads in the larger fractions, while known picoplanktonic groups in the genus Micromonas and parasitic dinophycean Synidniales with small dinospores were more often found in the small fractions. Using this probabilistic approach, we investigated the phylogenetic identities and distribution of small fraction-associated core dinoflagellate and syndinean OTUs. Evidence inferred from phylogenies and distribution suggest that majority of the picodinoflagellates we observed belonged to the Order Syndiniales, and were consistent with them having picoplanktonic life stages such as temporary cysts in core dinoflagellates and dinospores in syndinean dinophyceae. Further, we showed that most of the core dinoflagellate reads detected in small fractions were potential artifacts, with implications for the interpretation of HTS data. The effect of fractionation may have also partially obscured some of the

112 interannual patterns for dinoflagellates compared to ciliates in Chapter 2, since ciliates were strongly associated with the small fractions, presumably because the tended to break up during filtation.

Unlike the specialist phototrophs, which are largely confined to the photic layers, microzooplanktonare ubiquitous in the water column. We initially observed their strong association with water masses (Chapters 2 and 3), and potential co-occurrences and interactions (Chapter 3) that could influence their assembly in the different depths. However, the effects of this spatial variability to the underlying assembly rules and their potential responses to changing conditions remained unclear. Thus, Chapter 4 (Article 3) examined ecological theories of deterministic and stochastic processes to gain insights into how mixotrophic-heterotrophic communities assemble under spatially variable and contrasting conditions. For this, we chose dinoflagellates as the model group. Due to the potential effects of size fractionation and the presence of artefactual reads (Chapter 3), we limited the study to OTUs that were present in both rRNA and rDNA from the combined rarefied small and large reads. This allowed us to lessen effects free DNA or nucleic acids associated with dead cells. Based on random phylogenetic models using Nearest Taxon Index (NTI), our results showed that surface communities were more phylogenetically clustered than expected by chance (compared to the null community), suggesting strong environmental selection or habitat filtering. In contrast, the communities in the deeper samples from Pacific Winter Water, where conditions were more stable and less variable, were more governed by stochastic processes. The DCM communities, between the two other water masses, were governed by both selection processes. Using dinoflagellates, we found that the DCM community was governed by a dynamic interaction of assembly rules that would be influenced by the availability of resources. Based on our observations in contrasting conditions in the surface and PWW, we inferred that the mixotrophic-heterotrophic dinoflagellates would contribute to stability and resiliency of food and energy flows in conditions associated with the deepening nitracline. This conclusion is relevantin the wake of the changing climate and the upper surface freshening, which result in a deepening nitracline. Future work could extend these conclustion to other taxonomic groups such as the ciliates that have similar functional roles.

5.2 Perspectives

This study has generated new information on the ecology of ciliates and dinoflagellates, and provided theoretical frameworks useful for field-based and focused experimental investigations. In the following sections, we note several methodological and theoretical limitations that remain and need to be addressed for future studies on these highly diverse and functional groups.

113

5.2.1 Databases and taxonomic classification Comprehension and accurate interpretation of HTS data are dependent on the capacity to identify and classify the generated reads (Santamaria et al., 2012; Guillou et al., 2013). Although ciliates and dinoflagellates tend to be one of the most abundant groups in these types of surveys, they have the highest percentage of unknown or unclassified reads. In Chapter 2 for example, even after using the most recently updated databases (e.g. Ribosomal Database Project – RDP and SILVA) and modified in-house databases (e.g. Comeau et al., 2011; Monier et al., 2013), as many as 40% of ciliate and dinoflagellate reads remained unknown. This highlights potential novel diversity in the Arctic, and the lack of comprehensive and up-to-date reference sequence database for both ciliates and dinoflagellates. These limit researchers in gaining deeper insights into the ecology of microbial communities.

Another problem seen is the lack of consensus in ranking systems used for classification, particularly in dinoflagellates (see Chapter 3). This system creates confusion and results in the accumulation of “problematic” reference sequences. For example, although the NCBI nr database is rich in environmental sequences, most of these sequences were not classified beyond “Alveolata” or “Dinophyceae” or labelled as “uncultured/environmental clone” (Chapters 2, 3 and 4), needing greater efforts to classify them to become useful. Also, many sequences deposited in the NCBI, which tends to be the most used reference database, were either misidentified or never updated. This problem is not confined to NCBI, and for the RDP, around 20% of deposited sequences in 2012 was reported to be misidentified, chimeric or unidentified (Guillou et al., 2013). Using these sequences without careful culling or curation results in further misidentification of sequences, which are later on uploaded resulting to an accumulation of “problematic” references for future studies. Thus, the creation of a universally usable and highly curated database is needed. The community of microbial eukaryote researchers has begun to address this, and is the motivation behind large collaborations for 18S rRNA gene curation such as the UniEuk (http://unieuk.org/) and the EukRef (http://eukref.org/). Such collaborative community efforts to curate ribosomal databases for environmental sequencing will eventually enable microbial ecologists to exploit data much more efficiently from across studies.

In this study, we expanded existing reference databases by adding annotatedsequences from the literature, careful re-examination and curation of environmental sequences, and phylogenetic placement of abundant but unclassified HTS reads from the Arctic. However, despite these efforts, around 5-10% of sequences remained unclassified using similarity search approaches. Binning of these unclassified ranks could limit analysis due to lack of coherent patterns relevant or correlating with the environment (Chapter 2, Figure 2.4). These environmental sequences coupled with robust phylogenies can be used as reference trees to classify the HTS data that would allow possible identification of

114 regional or global populations of uncultivated taxa (e.g. Dunthorn et al, 2014a; Thaler and Lovejoy, 2014). For this, we used the EPA-RAxML approach, which allowed more plausible identification. In Chapter 3 for example, aside from establishing reference sequences for dinoflagellates, we also generated an updated phylogenetic tree for the dinophycean Syndiniales, which identified new clades using newly published sequences not includedin reference papers of Guillou et al. (2008) and Groissier et al. (2006).

Compared to ciliates, however, for which taxonomic consensus are largely communicated through the International Research Coordination Network for Biodiversity of Ciliates (M. Dunthorn, pers. comm.; J. Clamp, pers. comm.; IRCNBC Meeting 2016), efforts to consolidate and unify morphological and molecular diversities are less coordinated in dinoflagellates (M. Hoppenrath, pers. comm.; Taylor, 2004; Gomez et al., 2014; Hoppenrath, 2016). This was also one of the motivations for the efforts to focus on dinoflagellates in Chapters 3 and 4 in this study, which were contributions to the Northern Microbial 18S rRNA Gene Reference Database (Lovejoy et al., 2015, 2016). These databases are following the EukRef and UniEuk guidelines and aim to enable Arctic investigators in the accurate classification of HTS data. We also highlight the advantage and necessity of moving towards phylogenetic placement-based classification especially of abundant but unclassified reads.

5.2.2 Limitations of 18S rRNA gene for ciliate and dinoflagellate studies

The advantage of using universal markers (e.g. 18S rRNA gene) for identifying protists is that it puts the variability of certain taxa in the context of the entire or total microbial community. Although this approach has been useful for most microbial ecology studies, taxa-focused investigations have proven to be more challenging. The use of a fragment of the ribosomal gene (e.g. V4 or V9 regions) raises concerns that the marker does not provide enough taxonomic resolution compared to the long reads generated by cloning-Sanger sequencing approach. Bescot et al. (2015) for example, who did a planetary survey of dinoflagellates in photic waters limited their classification to the clear-cut ranks at the level of order since the V9 region of the 18S rRNA gene has limited taxonomic resolution. Studies, however, showed that V4 region can generate similar results in terms of diversity comparable to the whole gene sequences (Hu et al., 2015), as long as careful curation, quality filtering, sequence clustering and denoising have been carried out (Bachy et al., 2013).

The usage of more hypervariable genes or gene fragments has also been suggested for the microzooplankton (Bachvaroff et al., 2014). For example, the mitochondrial genes cytochrome oxidase 1 (cox1) and cytochrome oxidase b (cob) have been shown to be useful especially in barcoding some species and conducting diversity surveys (Lin et al., 2009; Strüder-Kypke and Lynn, 2010; Stern et al.,

115

2010; Kohli et al., 2014). The gene coding for actin has also been tested, but results showed significant deviation between the abundance based on reads and percentage of cell samples, andwas more efficient in reflecting actual community structure, beta diversity and high resolution at the generic level than the rDNA (Guo et al., 2016). However, the richness of the databases for such genes is still limited, and while useful for taxa-focused studies would become challenging for total community investigations.

We also showed the applicability of randomized phylogenetic models (Webb et al., 2002) to understand assembly rules in protist communities. To our knowledge, this is the first study that applied the NTI-based theoretical models first developed for prokaryotic communities to dinoflagellates (Stegen et al., 2012, 2013; Dini-Andreote et al., 2015). Our study showed promising results especially in inferring possible responses of communities to changing conditions and suggested that the relative contributions of deterministic and stochastic processes across stratified environments could be similar. However, more robust and simulation studies should be done to verify our results.

5.2.3 Microzooplankton in Arctic food web models

The Arctic has been recently experiencing drastic changes associated with the global climate change. Reports on the responses of microbial communities however have primarily focused on phytoplankton communities (Li et al., 2009; Arrigo et al., 2011; Ardyna et al., 2014; Irwin et al., 2015; Park et al., 2015) with less attention on other heterotrophic and mixotrophic groups. Results of this study showed that similar patterns observed in tropical and temperate systems relative to the microzooplankton could also be occurring in the Arctic, where mixotrophic taxa like ciliates and dinoflagellates are expected to play a central role especially with strengthened stratification. In Chapter 2, we showed that some mixotrophic ciliates significantly increased in abundance and diversity following low ice-related events. In Chapter 4, we showed the vulnerability of the SCM layer to such changes and how dinoflagellates could be favoured in both nutrient- or light-limited conditions with the deepening nitracline. There is an immediate need then to include these important taxonomic groups in Arctic food web models to capture a more accurate picture of food and energy flows in the polar region. Although we identified environmental factors that likely drive the responses of microzooplankton communities, more laboratory- or in situ-based studies are needed to confirm these correlations before they can be used to parameterize ecological numerical models. Nevertheless, we have also shown how molecular data could be exploited to test hypotheses beyond the usual distribution and diversity patterns by integrating phylogenetic information with species occurrences and environmental gradients.

116

BIBLIOGRAPHY

Aagaard, K., Coachman, L. K., and Carmack, E. C. (1981). On the halocline of the Arctic Ocean. Deep Sea Res., 28, 529-545. Aberle, N., Schulz, K. G., Stuhr, A., Malzahn, A. M., Ludwig, A., and Riebesell, U. (2013). High tolerance of microzooplankton to ocean acidification in an Arctic coastal plankton community. Biogeosciences 10, 1471–1481. doi:10.5194/bg-10-1471-2013. Adl, S., Simpson, A. G. B., Farmer, M., Andersen, R., Anderson, O. R., Barta, J. R., et al. (2005). The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. J. Eukaryot. Microbiol. 52, 399–451. doi:10.1111/j.1550-7408.2005.00053.x. Adl, S., Simpson, A. G. B., Lane, C., Lukes, J., Bass, D., Bowser, S. S., et al. (2012). The revised classification of eukaryotes. J. Eukaryot. Microbiol. 59, 429–493. doi:10.1111/j.1550- 7408.2012.00644.x. Aljanabi, S. M., and Martinez, I. (1997). Universal and rapid salt-extraction of high quality genomic DNA for PCR-based techniques. Nucleic Acid Res. 25, 4692–4693. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990). Basic Local Alignment Search Tool. J. Mol. Biol. 215, 403–410. Amaral-Zettler, L. A., Zettler, E. R., Theroux, S. M., Palacios, C., Aguilera, A., and Amils, R. (2011). Microbial community structure across the tree of life in the extreme Río Tinto. ISME J. 5, 42– 50. doi:10.1038/ismej.2010.101. Anderson, D. M., Coats, D. W., and Tyler, M. A. (1985). Encystment of the dinoflagellate Gyrodinium uncatenatum: temperature and nutrient effects. J. Phycol. 21, 200–206. Anderson, D.M., Glibert, P.M. and Burkholder, J.M. (2002). Harmful algal blooms and eutrophication: nutrient sources, composition, and consequences. Estuaries, 25, 704-726. Ardyna, M., Babin, M., Gosselin, M., Devred, E., Bélanger, S., Matsuoka, A., and Tremblay, J.-É. (2013). Parameterization of vertical chlorophyll a in the Arctic Ocean: impact of the subsurface chlorophyll maximum on regional, seasonal and annual primary production estimates. Biogeosciences, 10, 4383-4404. Ardyna, M., Babin, M., Gosselin, M., Devred, E., Rainville, L., and Tremblay, J. Èric (2014). Recent Arctic Ocean sea ice loss triggers novel fall phytoplankton blooms. Geophys. Res. Lett. 41, 1–6. doi:10.1002/2014GL061047. Ardyna, M., Gosselin, M., Michel, C., Poulin, M., and Tremblay, J. (2011). Environmental forcing of phytoplankton community structure and function in the Canadian High Arctic: contrasting oligotrophic and eutrophic regions. Mar. Ecol. Prog. Ser. 442, 37–57. doi:10.3354/meps09378. Arrigo, K. R., Matrai, P. A., and Van Dijken, G. L. (2011). Primary productivity in the Arctic Ocean: Impacts of complex optical properties and subsurface chlorophyll maxima on large-scale estimates. J. Geophys. Res., 116 (C11). Babin, M., Bélanger, S., Ellingsen, I., Forest, A., Fouest, V. Le, Lacour, T., et al. (2015). Estimation of primary production in the Arctic Ocean using ocean colour remote sensing and coupled physical – biological models: strengths, limitations and how they compare. Prog. Oceanogr. 139, 197-220. doi:10.1016/j.pocean.2015.08.008. Bachvaroff, T. R., Gornik, S. G., Concepcion, G. T., Waller, R. F., Mendez, G. S., Lippmeier, J. Casey, et al. (2014). Dinoflagellate phylogeny revisited: using ribosomal proteins to resolve deep branching dinoflagellate clades. Mol. Phylogenet. Evol. 70, 314–322. doi:101016/j.ympev.2013.10.007.

117

Bachy, C., Dolan, J. R., López-García, P., Deschamps, P., and Moreira, D. (2012). Accuracy of protist diversity assessments: morphology compared with cloning and direct pyrosequencing of 18S rRNA genes and ITS regions using the conspicuous tintinnid ciliates as a case study. ISME J. 7, 244–255. doi:10.1038/ismej.2012.106. Bachy, C., López-García, P., Vereshchaka, A., and Moreira, D. (2011). Diversity and vertical distribution of microbial eukaryotes in the snow, sea ice and seawater near the North Pole at the end of the polar night. Front. Microbiol. 2, 1-2. doi:10.3389/fmicb.2011.00106. Balzano, S., Gourvil, P., Siano, R., Chanoine, M., Marie, D., Lessard, S., et al. (2012). Diversity of cultured photosynthetic flagellates in the northeast Pacific and Arctic Oceans in summer. Biogeosciences 9, 4553–4571. doi:10.5194/bg-9-4553-2012. Barber, D. G., Asplin, M. G., Raddatz, R. L., Candlish, L. M., Nickels, S., and Prinsenberg, S. J. (2012). Change and variability in sea ice during the 2007 – 2008 Canadian International Polar Year program. Climate Change, 115, 115–133. doi:10.1007/s10584-012-0477-6. Barber, D., Leitch, D., Asplin, M. G., Gratton, Y., Lukovich, J. V, and Galley, R. J. (2010). The International Polar Year (IPY) Circumpolar Flaw Lead (CFL) system study: Overview and the physical system. Atmosphere-Ocean 48, 225–243. doi:10.3137/OC317.2010. Barber, D.G., and Massom, R.A. (2007). The role of sea ice in Arctic and Antarctic polynyas. In: Smith W.O. Jr, Barber, D.G. (eds.) Polynyas: windows to the world. Elsevier Oceanography Series 74, pp 1–54. Barberán, A., Bates, S. T., Casamayor, E. O., and Fierer, N. (2012). Using network analysis to explore co-occurrence patterns in soil microbial communities. ISME J. 6, 343–351. doi:10.1038/ismej.2011.119. Bar-Massada, A., Kent, R., and Carmel, Y. (2014). Environmental heterogeneity affects the location of modelled communities along the niche-neutrality continuum. Proc. Biol. Sci. 281, 20133249. doi:10.1098/rspb.2013.3249. Barton, A. D., Finkel, Z. V, Ward, B. a, Johns, D. G., and Follows, M. J. (2013). On the roles of cell size and trophic strategy in North Atlantic diatom and dinoflagellate communities. Limnol. Oceanogr. 58, 254–266. doi:10.4319/lo.2013.58.1.0254. Barwell-Clarke, J.E., and Whitney, F.A. (1996). Institute of Ocean Sciences nutrient methods and analysis. Canadian Technical Report of Hydrography and Ocean Sciences, 182, Sidney, 43 pp. Bélanger, S., Cizmeli, S.A., Ehn, J., Matsuoka, A., Doxaran, D., Hooker, S., and Babin, M. (2013). Light absorption and partitioning in Arctic Ocean surface waters: impact of multiyear ice melting. Biogeosciences, 10, 6433–6452 http://dx.doi.org/10.5194/bg-10-6433-2013. Berger, S. a., Krompass, D., and Stamatakis, A. (2011). Performance, accuracy, and web server for evolutionary placement of short sequence reads under maximum likelihood. Syst. Biol. 60, 291– 302. doi:10.1093/sysbio/syr010. Bergeron, M., and Tremblay, J.E. (2014). Shifts in biological productivity inferredfromnutrient drawdown in the southern Beaufort Sea (2003–2011) and northern BaffinBay (1997–2011), Canadian Arctic. Geophys. Res. Lett. 41, 3979–3987. doi:10.1002/2014GL061266.Received. Bescot, N. Le, Mahé, F., Audic, S., Dimier, C., Garet, M., Poulain, J., et al. (2016). Global patterns of pelagic dinoflagellate diversity across protist size classes unveiled by metabarcoding. Environ. Microbiol. 18, 609–26. doi:10.1111/1462-2920.13039. Björk, G., Anderson, L. G., Jakobsson, M., Antony, D., Eriksson, B., Eriksson, P. B., et al. (2010). Flow of Canadian basin deep water in the Western Eurasian Basin of the Arctic Ocean. Deep Sea Res. Part I Oceanogr. Res. Pap. 57, 577–586. doi:10.1016/j.dsr.2010.01.006.

118

Blazewicz, S. J., Barnard, R. L., Daly, R. A., and Firestone, M. K. (2013). Evaluating rRNA as an indicator of microbial activity in environmental communities: limitations and uses. ISME J. 7, 2061–8. doi:10.1038/ismej.2013.102. Boere, A. C., Rijpstra, W. I. C., Lange, G. J. D. E., Damste, J. S. S., Coolen, M. J. L., Burg, D., et al. (2011). Preservation potential of ancient plankton DNA in pleistocene marine sediments. Geobiology 9, 377–393. doi:10.1111/j.1472–4669.2011.00290.x. Boetius, A., Albrecht, S., Bakker, K., Bienhold, C., Felden, J., Fernández-Méndez, M., Hendricks, S., et al. (2013). Export of algal biomass from the melting Arctic sea ice. Science, 339, 1430-1432. DOI: 10.1126/science.1231346. Bork, P. Bowler, C., de Vargas, C., Gorsky, G., Karsenti, E. and Wincker, P. (2015). Tara Oceans studies plankton at planetary scale. Science, 348, 873. Bråte, J., Krabberød, A. K., Dolven, J. K., Ose, R. F., Kristensen, T., Bjørklund, K. R., et al. (2012). Radiolaria associated with large diversity of Marine Alveolates. Protist 163, 767–777. doi:10.1016/j.protis.2012.04.004. Bravo, I., and Figueroa, R. (2014). Towards an ecological understanding of dinoflagellate cyst functions. Microorganisms 2, 11–32. doi:10.3390/microorganisms2010011. Brocks, J. J., and Banfield, J. (2009). Unravelling ancient microbial history with community proteogenomics and lipid geochemistry. Nat. Rev. Microbiol. 7, 601–609. doi:10.1038/nrmicro2167. Brown, Z. W., Lowry, K. E., Palmer, M. a., van Dijken, G. L., Mills, M. M., Pickart, R. S., et al. (2015). Characterizing the subsurface chlorophyll a maximum in the Chukchi Sea and Canada Basin. Deep Sea Res. Part II Top. Stud. Oceanogr. 118, 1–17. doi:10.1016/j.dsr2.2015.02.010. Buttigieg, P. L. and Ramette, A. (2014). A guide to statistical analysis in microbial ecology: a community-focused, living review of multivariate data analyses. FEMS Microbiol Ecol, 90, 543– 550. doi:10.1111/1574-6941.12437. Calbet, A., Riisgaard, K., Saiz, E., Zamora, S., Stedmon, S., and Nielsen, T. G. (2011). Phytoplankton growth and microzooplankton grazing along a sub-Arctic fjord (Godthabsfjord, west Greenland). Mar. Ecol. Prog. Ser. 442, 11–22. doi:10.3354/meps09343. Campbell, B. J., and Kirchman, D. L. (2013). Bacterial diversity, community structure and potential growth rates along an estuarine salinity gradient. ISME J. 7, 210–220. doi:10.1038/ismej.2012.93. Caporaso, J. G., Bittinger, K., Bushman, F. D., Desantis, T. Z., Andersen, G. L., and Knight, R. (2010a). PyNAST: A flexible tool for aligning sequences to a template alignment. Bioinformatics 26, 266–267. doi:10.1093/bioinformatics/btp636. Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F. D., Costello, E. K., et al. (2010b). QIIME allows analysis of high- throughput community sequencing data intensity normalization improves color calling in SOLiD sequencing. Nat. Methods 7, 335–336. doi:10.1038/nmeth0510-335. Carmack, E. C., and Macdonald, R. W. (2002). Oceanography of the Canadian Shelf of the Beaufort Sea: a setting for marine life. Arctic 55, 29–45. Carmack, E. C., Yamamoto-Kawai, M., Haine, T. W. N., Bacon, S., Bluhm, B. A., Lique, C., et al. (2016). Freshwater and its role in the Arctic Marine System: Sources, disposition, storage, export, and physical and biogeochemical consequences in the Arctic and global oceans. J. Geophys. Res. Biogeosciences 121, 675–717. doi:10.1002/2015JG003140.

119

Carmack, E., and Wassman, P. (2006). Food webs and physical–biological coupling on pan-Arctic shelves: unifying concepts and comprehensive perspectives. Prog. Oceanogr. 71, 446–477. doi:10.1016/j.pocean.2006.10.004. Caron, D. A., and Hutchins, D. a. (2013). The effects of changing climate on microzooplankton grazing and community structure: drivers, predictions and knowledge gaps. J. Plankton Res. 35, 235–252. doi:10.1093/plankt/fbs091. Carpenter, J. R., and Timmermans, M.-L. (2012). Deep mesoscale eddies in the Canada Basin, Arctic Ocean. Geophys. Res. Let. 39, L20602. doi:10.1029/2012GL053025 Chambouvet, A., Morin, P., Marie, D., and Guillou, L. (2008). Control of toxic marine dinoflagellate blooms by serial parasitic killers. Science 322, 1254-7. Chao, A., and Shen, T. (2003). Non-parametric estimation of Shannon's index of diversity when there are unseen species in sample. Environ. Ecol. Stat. 10, 429–443. Charvet, S., Vincent, W. F., Comeau, A., and Lovejoy, C. (2012). Pyrosequencing analysis of the protist communities in a High Arctic meromictic lake: DNA preservation and change. Front. Microbiol. 3, 422. doi:10.3389/fmicb.2012.00422. Chase, J. M., and Myers, J. A. (2011). Disentangling the importance of ecological niches from stochastic processes across scales. Philos. Trans. R. Soc. B Biol. Sci. 366, 2351–2363. doi:10.1098/rstb.2011.0063. Chen, J., Bittinger, K., Charlson, E. S., Hoffmann, C., Lewis, J., Wu, G. D., et al. (2012). Associating microbiome composition with environmental covariates using generalized UniFrac distances. Bioinformatics 28, 2106–2113. doi:10.1093/bioinformatics/bts342. Cheung, M. K., Au, C. H., Chu, K. H., Kwan, H. S., and Wong, C. K. (2010). Composition and genetic diversity of picoeukaryotes in subtropical coastal waters as revealed by 454 pyrosequencing. ISME J. 4, 1053–1059. doi:10.1038/ismej.2010.26. Cheung, M. K., Chu, K. H., Li, C. P., Kwan, H. S., and Wong, C. K. (2008). Genetic diversity of picoeukaryotes in a semi- enclosed harbour in the subtropical western Pacific Ocean. 53, 295– 305. doi:10.3354/ame01247. Choi, C.J., Bachy, C., Jaeger, G.S., Proirier, C., Sudek, L., Sarma, V.V.S.S., Mahadevan, A., Giovanni, S. J. and Worden, A. (2017). Newly discovered deep-branching marine plastid lineages are numerically rare but globally distributed. Curr. Biol. 27, pR15-R16. DOI: http://dx.doi.org/10.1016/j.cub.2016.11.032. Coats, D. W. (1999). Parasitic life styles of marine dinoflagellates. J. Eukaryot. Microbiol. 46, 402– 409. Coats, D. W., and Park, M. G. (2002). Parasitism of photosynthetic dinoflagellates by three strains of Amoebophrya (Dinophyta): parasite survival, infectivity, generation. J. Phycol. 528, 520–528. Comeau A. M., Philippe B., Thaler M., Gosselin M., Poulin M., and Lovejoy C. (2013). Protists in Arctic drift and land-fast sea ice.J. Phycol. 49, 229–240. DOI: 10.1111/jpy.12026. Comeau, A. M., Li, W. K. W., Tremblay, J.-É., Carmack, E. C., and Lovejoy, C. (2011). Arctic Ocean microbial community structure before and after the 2007 record sea ice minimum. PLoS One 6, e27492. doi:10.1371/journal.pone.0027492. Conover, R.J., Jr. (1982). Interrelations between microzooplankton and other plankton organisms. Ann. Inst. Oceanogr. 59, 31-46. Cordinaldesi, C., Barucca, M., Luna, G. M., Dell-Anno, A. (2011). Preservation, origin and genetic imprint of extracellular DNA in permanently anoxic deep-sea sediments. Mol. Ecol. 20, 642– 654. doi:10.1111/j.1365-294X.2010.04958.x.

120

Coupel, P., Jin, H. Y., Joo, M., Horner, R., Bouvet, H. A., Garçon, V., et al. (2012). Phytoplankton distribution in unusually low sea ice cover over the Pacific Arctic. Biogeosciences, 9, 4863- 4850. doi:10.5194/bgd-9-2055-2012. Coupel, P., Ruiz-pino, D., Sicre, M. A., Chen, J. F., Lee, S. H., Schiffrine, N., et al. (2015). The impact of freshening on phytoplankton production in the Pacific Arctic Ocean. Prog. Oceanog, 131, 113-125. doi:10.1016/j.pocean.2014.12.003. Crawford, D. W. (1992). Metabolic cost of motility in planktonic protists: Theoretical considerations on size scaling and swimming speed. Microb. Ecol. 24, 1–10. doi:10.1007/BF00171966. Csardi, G., and Nepusz, T. (2006). The igraph software package for complex network research. InterJournal Complex Syst. 1695, 1-9. Cuvelier, M. L., Allen, A. E., Monier, A., McCrow, J. P., Messié, M., Tringe, S. G., Woyke, T., et al. (2010). Targeted metagenomics and ecology of globally important uncultured eukaryotic phytoplankton. PNAS, 107, 14679–84. D’Alelio, D. D., Libralato, S., Wyatt, T., and Ribera, M. (2016). Ecological-network models link diversity, structure and function in the plankton food-web. Sci. Rep. 6, 21806. doi:10.1038/srep21806. Darnis, G., and Fortier, L. (2012). Zooplankton respiration and the export of carbon at depth in the Amundsen Gulf (Arctic Ocean). J. Geophys. Res. Ocean. 117, 1–12. doi:10.1029/2011JC007374. Darnis, G., Robert, D., Pomerleau, C., Link, H., Archambault, P., Nelson, R. J., et al. (2012). Current state and trends in Canadian Arctic marine ecosystems: II. Heterotrophic food web, pelagic- benthic coupling, and biodiversity. Clim. Change 115, 179-205. doi:10.1007/s10584-012-0483- 8. Dasilva, C. R., Li, W. K. W., and Lovejoy, C. (2014). Phylogenetic diversity of eukaryotic marine microbial plankton on the Scotian Shelf, Northwestern Atlantic Ocean. J. Plankton Res. 36, 344– 363. doi:10.1093/plankt/fbt123. Daugbjerg, N., Hansen, G., and Moestrup, Ø. (2000). Phylogeny of some of the major genera of dinoflagellates based on ultrastructure and partial LSU rDNA sequence data, including the erection of three new genera of unarmoured dinoflagellates. Phycologia 39, 302–317. doi:10.2216/i0031-8884-39-4-302.1. de Vargas, C., Audic, S., Henry, N., Decelle, J., Mahé, F., Logares, R., et al. (2015). Eukaryotic plankton diversity in the sunlit ocean. Science 384, 1–11. Diez, B., Pedrós-Alió, C., and Massana, R. (2001a). Study of genetic diversity of eukaryotic picoplankton in different oceanic regions by small-subunit rRNA gene cloning and sequencing. Appl. Environ. Microbiol. 67, 2932–2941. doi:10.1128/AEM.67.7.2932. Diez, B., Pedrós-Alió, C., Marsh, T. L., &Massana, R. (2001b). Application of denaturing gradient gel electrophoresis (DGGE) to study the diversity of marine picoeukaryotic assemblages and comparison of DGGE with other molecular techniques. Appl Environ Microbiol, 67:2942-2951. Dini-Andreote, F., Stegen, J. C., Dirk, J., Elsas, V., and Falcão, J. (2015). Disentangling mechanisms that mediate the balance between stochastic and deterministic processes in microbial succession. PNAS, E1326–E1332. doi:10.1073/pnas.1414261112. Dixon, P., and Dixon, P. (2003). VEGAN, a package of R functions for community ecology. J. Veg. Sci. 14, 927–930. doi:10.1111/j.1654-1103.2003.tb02228.x.

121

Dolan, J. R., Pierce, R. W., and Jin, E. (2016). Tintinnid ciliates of the marine microzooplankton in Arctic Seas: a compilation and analysis of species records. Polar Biol. n/a, n/a. doi:10.1007/s00300-016-2049-0. Dolan, J. R., Yang, E. J., Kim, T. W., and Kang, S. (2013). Microzooplankton in a warming Arctic: A comparison of Tintinnids and Radiolarians from summer 2011 and 2012 in the Chukchi Sea. Acta Protozoologica. 53, 101-124. Dormann, C. F., Gruber, B., and Fründ, J. (2008). Introducing the bipartite package: Analysing ecological networks. R News 8, 8–11. ISSN 1609-3631. Dunthorn, M., Otto, J., Berger, S. A., Stamatakis, A., Consortium, B., Stock, A., et al. (2014a). Placing environmental next-generation sequencing amplicons from microbial eukaryotes into a phylogenetic context. Mol. Biol. Evol. 31, 993–1009. doi:10.1093/molbev/msu055. Dunthorn, M., Stoeck, T., Clamp, J., and Warren, A. (2014b). Ciliates and the rare biosphere: a review. J. Eukaryot. Microbial. 61, 404–409. doi:10.1111/jeu.12121. Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acid Res. 32, 1792–1797. doi:10.1093/nar/gkh340. Edgar, R. C. (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. doi:10.1093/bioinformatics/btq461. Edgar, R. C. (2013). UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–1001. doi:10.1038/nmeth.2604. Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C., and Knight, R. (2011). UCHIME improves sensitivity and speed of chimera detection. 27, 2194–2200. doi:10.1093/bioinformatics/btr381. Egge, E. S., Bittner, L., Andersen, T., Audic, S., and de Vargas, C. (2013). 454 Pyrosequencing to describe microbial eukaryotic community composition, diversity and relative abundance?: a test for marine Haptophytes. PLoS One 8, e74371. doi:10.1371/journal.pone.007437 Esteban, G. F., Fenchel, T., and Finlay, B. J. (2010). Mixotrophy in Ciliates. Protist 161, 621–641. doi:10.1016/j.protis.2010.08.002. Falk-Petersen, S., Mayzaud, P., Kattner, G., and Sargent, J. R. (2009). Lipids and life strategy of Arctic Calanus. Mar. Biol. Res. 5, 18–39. doi:10.1080/17451000802512267. Fensome, R. A., Saldarriaga, J. F., and Taylor, M. F. J. R. (1999). Dinoflagellate phylogeny revisited: reconciling morphological and molecular based phylogenies. Grana 38, 66–80. doi:10.1080/00173139908559216. Ferrenberg, S., O’Neill, S. P., Knelman, J. E., Todd, B., Duggan, S., Bradley, D., et al. (2013). Changes in assembly processes in soil bacterial communities following a wildfire disturbance. ISME J. 7, 1102–1111. doi:10.1038/Ismej.2013.11. Fierer, N., Leff, J. W., Adams, B. J., Nielsen, U. N., Bates, S. T., Lauber, C. L., et al. (2012). Cross- biome metagenomic analyses of soil microbial communities and their functional attributes. Proc. Natl. Acad. Sci. U. S. A. 109, 21390–5. doi:10.1073/pnas.1215210110. Finlay, B.J., Corliss, J.O., Esteban, G. et al. (1996). Biodiversity at the microbial level: the number of free-living ciliates in the biosphere. Q. Rev. Biol., 71, 221–237 Flynn, K. J., Stoecker, D. K., Mitra, A., Ra, J. A., Glibert, P. M., Hansen, P. E. R. J., et al. (2012). Misuse of the phytoplankton – zooplankton dichotomy: the need to assign organisms as mixotrophs within plankton functional types. J. Plank. Res. 0, 1–9. doi:10.1093/plankt/fbs062. Foissner, W., Chao, A., and Katz, L.A. (2008). Diversity and geographic distribution of ciliates (Protista: Ciliophora).17, 345-363. doi:10.1007/s10531-007-9254-7.

122

Forest, A., Bélanger, S., Sampei, M., Sasaki, H., Lalande, C., and Fortier, L. (2010). Three-year assessment of particulate organic carbon fluxes in Amundsen Gulf (Beaufort Sea): Satellite observations and sediment trap measurements. Deep. Res. Part I Oceanogr. Res. Pap. 57, 125– 142. doi:10.1016/j.dsr.2009.10.002. Forest, A., Tremblay, J.-É., Gratton, Y., Martin, J., Gagnon, J., Darnis, G., et al. (2011). Biogenic carbon flows through the planktonic food web of the Amundsen Gulf (Arctic Ocean): A synthesis of field measurements and inverse modeling analyses. Prog. Oceanogr. 91, 410–436. doi:10.1016/j.pocean.2011.05.002. Fransson, A., Chierici, M., Miller, L. A., Carnat, G., Shadwick, E., Thomas, H., et al. (2013). Impact of sea-ice processes on the carbonate system and ocean acidification at the ice-water interface of the Amundsen Gulf, Arctic Ocean. 118, 1–23. doi:10.1002/2013JC009164 Galand, P. E., Lovejoy, C., Hamilton, A. K., Ingram, R. G., Pedneault, E., and Carmack, E. C. (2009). Archaeal diversity and a gene for ammonia oxidation are coupled to oceanic circulation. Environ. Microbiol. 11, 971–80. doi:10.1111/j.1462-2920.2008.01822.x. Ganesh, S., Parris, D. J., DeLong, E. F., and Stewart, F. J. (2014). Metagenomic analysis of size- fractionated picoplankton in a marine oxygen minimum zone. ISME J. 8, 187–211. doi:10.1038/ismej.2013.144. Garcés, E., Maso, M., and Camp, J. (2002). Role of temporary cysts in the population dynamics of Alexandrium taylori (Dinophyceae). J. Plankton Res. 24, 681–686. Garneau M-V., Roy, S., Lovejoy, C., Gratton, Y., and Vincent, W.F. (2008). Seasonal dynamics of bacterial biomass and production in a coastal arctic ecosystem: Franklin Bay, western Canadian Arctic. Can. Arct. J. Geophys. Res., 113: 7–91. Garneau, M.-È., Vincent, W. F., Alonso-Sáez, L., Gratton, Y., and Lovejoy, C. (2006). Prokaryotic community structure and heterotrophic production in a river-influenced coastal Arctic ecosystem Aquat. Microb. Ecol. 42, 27–40. doi:10.3354/ame042027. Gast, R. J., and Caron, D. A. (1996). Molecular Phylogeny of symbiotic dinoflagellates from planktonic Foraminifera and Radiolaria. Mol. Biol. Evol. 13, 1192–1197. Geider, R. J., and Osborne, B. A. (1989). Respiration and microalgal growth?: a review of the quantitative relationship. 112, 327–341. Gilbert, J. A., and Dupont, C. L. (2011). Microbial metagenomics: beyond the genome. Ann. Rev. Mar. Sci. 3, 347–371. Glöckner F.O., Stal, L.J., Sandaa, R.-A., Gasol, J.M., O’Gara, F., Hernandez, F., Labrenz, M., Stoica, E., Varela, M.M., Bordalo, A., and Pitta, P. (2012). Marine microbial diversity and its role in ecosystem functioning and environmental change. Marine Board Position Paper 17. Calewaert, J.B. and McDonough N. (Eds.). Marine Board-ESF, Ostend, Belgium. Godhantaraman, N., and Uye, S. (2003). Geographical and seasonal variations in taxonomic composition, abundance and biomass of microzooplankton across a brackish-water lagoonal system of Japan. J. Plankton Res. 25, 465–482. Godhe, A., Asplund, M. E., Härnström, K., Saravanan, V., Tyagi, A., and Karunasagar, I. (2008). Quantification of diatom and dinoflagellate biomasses in coastal marine seawater samples by real-time PCR. Appl. Environ. Microbiol. 74, 7174–7182. doi:10.1128/AEM.01298-08. Gómez, F. (2005). A list of free-living dinoflagellate species in the world's oceans. Acta Bot. Croat. 64, 129–212. Gómez, F. (2007). Gymnodinioid dinoflagellates (Gymnodiniales, Dinophyceae) in the open Pacific Ocean. Algae, 22, 273–286.

123

Gómez, F. (2014). Problematic biases in the availability of molecular markers in protists: the example of the Dinoflagellates. Acta Protozool. 53, 63–75. doi:10.4467/16890027AP.13.0021.1118. Gong, J., Dong, J., Liu, X., and Massana, R. (2013). Extremely high copy numbers and polymorphisms of the rDNA operon estimated from single cell analysis of Oligotrich and Peritrich ciliates. Protist 164: 369-379. Gotelli, N.J., and Entsminger, G.L. (2003). Swap algorithms in null model analysis. Ecology, 84, 532–535. Gower, J.C. (1966). Some distance properties of latent root and vector methods used in multivariate analysis. Biometrika 5, 325-338. Gravel, D., Canham, C. D., Beaudet, M., and Messier, C. (2006). Reconciling niche and neutrality: The continuum hypothesis. Ecol. Lett. 9, 399–409. doi:10.1111/j.1461-0248.2006.00884.x. Groisillier, A., Massana, R., Valentin, K., Vaulot, D., and Guillou, L. (2006). Genetic diversity and habitats of two enigmatic marine alveolate lineages Genetic diversity and habitats of two enigmatic marine alveolate lineages. Aquat. Microb. Ecol. 42, 277–291. doi:10.3354/ame042277. Guéguen, C., Itoh, M., Kikuchi, T., Eert, J., and Williams, W. J. (2015). Variability in dissolved organic matter optical properties in surface waters in the Amerasian Basin. Front. Mar. Sci., 2, 1–9. doi:10.3389/fmars.2015.00078. Guillou, L., Viprey, M., Chambouvet, A., Welsh, R. M., Kirkham, A. R., Massana, R., et al. (2008). Widespread occurrence and genetic diversity of marine parasitoids belonging to Syndiniales (Alveolata). 10, 3349–3365. doi:10.1111/j.1462-2920.2008.01731.x. Guillou, L., Bachar, D., Audic, S., Bass, D., Berney, C., Bittner, L. et al. (2013). The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukaryote small subunit rRNA sequences with curated taxonomy. Nucleic Acids Res, 41, D597–D604. Guiry, M.D. and Guiry, G.M. (2015). AlgaeBase. World-wide electronic publication, National University of Ireland, Galway. http://www.algaebase.org; searched on 11 July 2015. Guiry, M.D. and Guiry, G.M. (2016). AlgaeBase. World-wide electronic publication, National University of Ireland, Galway. http://www.algaebase.org; searched on 11 December 2016. Guo, L., Sui, Z., and Liu, Y. (2016). Quantitative analysis of dinoflagellates and diatoms community via MiSeq sequencing of actin gene iand V9 region of 18S rDNA. Scientific Reports, 6, 37049. doi:10.103srep34709. Hall, T. (1999). BioEdit: A user-friendly biological sequence alignment program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Hamilton, A. K., Lovejoy, C., Galand, P. E., Gk, C., and Ingram, R. G. (2008). Water masses and biogeography of picoeukaryote assemblages in a cold hydrographically complex system. Limnol. Oceanogr. 53, 922–935. Hammer, Ø., Harper, D. A. T., and Ryan, P. D. (2001). PAST: Paleontological Statistics Software Package for Education and Data Analysis. Paleontol. Electron. 4, 1–9.http://palaeo- electronica.org/2001_1/past/issue1_01.htm Handelsman, J., Rondon, M.R., Brady, S.F., Clardy, J., Goodman, R.M. (1998). Molecular biological access to the chemistry of unknown soil microbes: a new frontier for natural products. Chem. Biol. 5:245–49. Hansen, B. W., Nielsen, T. G., and Levinsen, H. (1999). Plankton community structure and carbon cycling on the western coast of Greenland during the stratified summer situation. III. Mesozooplankton. Aquat. Microb. Ecol. 16, 233–249.

124

Hansen, P. J. (2011). The role of photosynthesis and food uptake for the growth of marine mixotrophic dinoflagellates. J. Eukaryot. Microbiol. 58, 203–214. doi:10.1111/j.1550- 7408.2011.00537.x. Hansen, P. J., Ojamäe, K., Berge, T., Trampe, E. C. L., Nielsen, L. T., Lips, I., et al. (2016). Photoregulation in a kleptochloroplastidic dinoflagellate, Dinophysis acuta. Front. Microbiol. 7, 1–11. doi:10.3389/fmicb.2016.00785. Hardy, O. J. (2008). Testing the spatial phylogenetic structure of local communities: Statistical performances of different null models and test statistics on a locally neutral community. J. Ecol. 96, 914–926. doi:10.1111/j.1365-2745.2008.01421.x. Harred, L. B., and Campbell, L. (2014). Predicting harmful algal blooms: a case study with Dinophysis ovum in the Gulf of Mexico. J. Plankton Res. 36, 1434–1445. doi:10.1093/plankt/fbu070. Hartmann, M., Grob, C., Tarran, G. A., Martin, A. P., Burkill, P. H., Scanlan, D. J., et al. (2012). Mixotrophic basis of Atlantic oligotrophic ecosystems. Proc. Natl. Acad. Sci. U. S. A. 109, 5756–5760. doi:10.1073/pnas.1118179109. Heiskanen, A. (1993). Mass encystment and sinking of dinoflagellates during a spring bloom. Mar. Biol. 116, 161–167. Hill, V., and Cota, G. (2005). Spatial patterns of primary production on the shelf, slope and basin of the Western Arctic in 2002. Deep Sea Res. Part II: Topical Studies in Oceanography, 52(24-26), 3344–3354. Hinchcliff, C.E., Smith, S.A., Allman, J.F., Burleigh, J.G., Chaudhary, R., Coghill, L. M., Crandall, K.A., Drew, B.T., Gazis, R., Gude, K., Hibbett, D.S., Katz, L.A., et al. (2015). Synthesis of phylogeny and taxonomy into a comprehensive tree of life. PNAS, 112, 12764-12769. Hoppenrath, M. (2016). Dinoflagellate taxonomy — a review and proposal of a revised classification. Mar. Biodivers. doi:10.1007/s12526-016-0471-8. Hoppenrath, M., and Leander, B. S. (2009). Molecular phylogeny of Parvilucifera prorocentri (Alveolata, Myzozoa): insights into Perkinsid character evolution. J. Eukaryot. Microbiol. 56, 251–256. doi:10.1111/j.1550-7408.2009.00395.x. Horner-Devine, M.C., and Bohannan, B.J.M. (2006). Phylogenetic clustering and overdispersion in bacterial communities. Ecology, 87, 100–108. Hou, Y., and Lin, S. (2009). Distinct gene number-genome size relationships for eukaryotes and non- eukaryotes: Gene content estimation for dinoflagellate genomes. PLoS One, 4, e6978. doi:10.1371/journal.pone.0006978. Hu, S. K., Campbell, V., Connell, P., Gellene, A. G., Liu, Z., Terrado, R., et al. (2016). Protistan diversity and activity inferred from RNA and DNA at a coastal ocean site in the eastern North Pacific. FEMS Microbiol. Ecol. 92, 1–13. doi:10.1093/femsec/fiw050. Hubbell, S. P. (2001). The unified neutral theory of biodiversity and biogeography. Princeton, NJ: Princeton University Press. Huerte-Ortega, M., Cermeno, P., Calvo-Diaz, A., and Maranon, E. (2012). Isometric size-scaling of metabolic rate and the size abundance distribution of phytoplankton. Proc. R. Soc. B 279, 1815– 1823. doi:10.1098/rspb.2011.2257. Huisman, J., Pham Thi, N.N., Karl, D.M.., and Sommeijer, B. (2006). Reduced mixing generates oscillations and chaos in the oceanic deep chlorophyll maximum. Nature 439, 322–325

125

Hunt, D. E., Lin, Y., Church, M. J., Karl, D. M., Tringe, S. G., Izzo, L. K., et al. (2013). Relationship between abundance and specific activity of bacterioplankton in open ocean surface waters. Appl. Environ. Microbiol. 79, 177–184. doi:10.1128/AEM.02155-12. Huson, D. H., and Scornavacca, C. (2012). Dendroscope 3: An interactive tool for rooted phylogenetic trees and networks. Syst. Biol. 61, 1061–1067. doi:10.1093/sysbio/sys062. Ikenoue, T., Bjørklund, K. R., Dumitrica, P., Krabberød, A. K., Kimoto, K., Matsuno, K., et al. (2016). Marine Micropaleontology Two new living Entactinaria (Radiolaria) species from the Arctic province: Joergensenium arcticum n. sp. and Joergensenium clevei n. sp. Mar. Micropaleontol. 124, 75–94. doi:10.1016/j.marmicro.2016.02.003. Irigoien, X. (2005). Phytoplankton blooms: a loophole in microzooplankton grazing impact? J. Plankton Res. 27, 313–321. doi:10.1093/plankt/fbi011. Irwin, A.J., Finkel, Z.V., Muller-Karger, F.E. and Ghinaglia, L.T. (2015). Phytoplankton adapt to changing ocean environments. PNAS. 112, 5762-5766. Jackson, J. M. (2011). Changes to the near surface waters of the Canada Basin, Arctic Ocean from. J. Geog. Res. 116, C10008. 10.1029/2011JC007069 Jackson, J. M., Carmack, E. C., Mclaughlin, F. A., Allen, S. E., and Ingram, R. G. (2010). Identification , characterization , and change of the near surface temperature maximum in the Canada Basin , 1993 – 2008. J. Geophys. Res. Oceans 115, C05021. doi:10.1029/2009JC005265.

Jacobson, D. M., and Anderson, D. M. (1986). Thecate heterotrophic dinoflagellates: feeding behavior and mechanisms. J. Phycol. 22, 249–258. doi:10.1111/j.1529-8817.1986.tb00021.x. Jacobson, D. M., and Anderson, D. M. (1996). Widespread phagocytosis of ciliates and other protists by marine mixotrophic and heterotrophic thecate dinoflagellates. J. Phycol. 32, 279–285. Jakobsson, M. (2002). Hypsometry and volume of the Arctic Ocean and its constituent seas. Geochemistry, Geophys. Geosystems 3, 1–18. doi:10.1029/2001GC000302. Janouškovec, J., Gavelis, G. S., Burki, F., Dinh, D., Bachvaroff, T. R., Gornik, S. G., et al. (2017). Major transitions in dinoflagellate evolution unveiled by phylotranscriptomics. Proc. Natl. Acad. Sci. U. S. A. 114, E171–E180. doi:10.1073/pnas.1614842114. Jassey, V. E. J., Signarbieux, C., Hättenschwiler, S., Bragazza, L., Buttler, A., Delarue, F., et al. (2015). An unexpected role for mixotrophs in the response of peatland carbon cycling to climate warming. Sci. Rep. 5, 16931. doi:10.1038/srep16931. Jeong, H. J., du Yoo, Y., Kim, J. S., Seong, K. A., Kang, N. S., and Kim, T. H. (2010). Growth, feeding and ecological roles of the mixotrophic and heterotrophic dinoflagellates in marine planktonic food webs. Ocean Sci. J. 45, 65–91. doi:10.1007/s12601-010-0007-2. Jeong, H. J., Lim, A. S., Yoo, Y. Du, Lee, M. J., Lee, K. H., Jang, T. Y., et al. (2014). Feeding by heterotrophic dinoflagellates and ciliates on the free-living dinoflagellate Symbiodinium sp. (clade E). J. Euk. Microbiol. 61, 27–41. doi:10.1111/jeu.12083. Jephcott, T. G., Alves-de-souza, C., Ogtrop, F. F. Van, Sime-Ngando, T., Karpov, S. A., and Guillou, L. (2016). Ecological impacts of parasitic chytrids, syndiniales and perkinsids on populations of marine photosynthetic dinoflagellates. Fungal Ecol. 19, 47–58. doi:10.1016/j.funeco.2015.03.007. Jiang, Y., Yang, E. J., Min, J. O., Kang, S. H., and Lee, S. H. (2013). Using pelagic ciliated microzooplankton communities as an indicator for monitoring environmental condition under impact of summer sea-ice reduction in Western Arctic Ocean. Ecol. Indic. 34, 380–390. doi:10.1016/j.ecolind.2013.05.026.

126

Johansson, M., and Coats, D. W. (2002). Ciliate grazing on the parasite Amoebophrya sp. decreases infection of the red-tide dinoflagellate Akashiwo sanguinea. Aquat. Microb. Ecol. 28, 69–78. Johnson, M. D. (2011). Acquired phototrophy in ciliates: a review of cellular interactions and structural adaptations. J. Eukaryot. Microbiol. 58, 185–195. doi:10.1111/j.1550- 7408.2011.00545.x. Jones, S. E., and Lennon, J. T. (2010). Dormancy contributes to the maintenance of microbial diversity. Proc. Natl. Acad. Sci. U. S. A. 107, 5881–5886. doi:10.1073/pnas.0912765107. Jost, C., Lawrence, C. A., Campolongo, F., Van De Bund, W., Hill, S., and De Angelis, D. L. (2004). The effects of mixotrophy on the stability and dynamics of a simple planktonic food web model. Theor. Pop. Biol. 66, 37–51. doi:10.1016/j.tpb.2004.02.001. Kembel, S.W. (2009). Disentangling niche and neutral influences on community assembly: assessing the performance of community phylogenetic structure tests. Ecol. Lett., 12, 949–960. Kembel, S.W., Cowan, P.D., Helmus, M.R., Cornwell, W.K., Morlon, H., Ackerly, D.D. et al. (2010). Picante: R tools for integrating phylogenies and ecology. Bioinformatics, 26, 1463–1464. Kim, S. (2006). Patterns of host range for two strains of Amoebophrya (Dinophyta) infecting thecate dinoflagellate: Amoebophrya spp. ex Alexandrium affine and ex Gonyaulax polygramma. J. Phycol. 1173, 1170–1173. doi:10.1111/j.1529-8817.2006.00277.x. Kim, S., Park, M. G., Kim, K.-Y., Kim, C.-H., Yih, W., Park, J. S., et al. (2008). Genetic Diversity of Parasitic dinoflagellates in the genus Amoebophrya and its relationship to parasite biology and biogeography. J. Euk. Microbiol. 55, 1–8. doi:10.1111/j.1550-74080.2007.00295.x. Kim, S., Yoon, J., and Park, M. G. (2015). Obligate mixotrophy of the pigmented dinoflagellate Polykrikos lebourae (Dinophyceae, Dinoflagellata). Algae 30, 35–47. Kluijver, a., Soetaert, K., Czerny, J., Schulz, K. G., Boxhammer, T., Riebesell, U., & Middelburg, J. J. (2013). A 13C labelling study on carbon fluxes in Arctic plankton communities under elevated CO2 levels. Biogeosciences, 10(3), 1425–1440. Kohli, G.S., Neilan, B.A., Brown, M.V., Hoppenrath, M., and Murray, S. (2014). Cob gene pyrosequencing enables characterization of benthic dinoflagellate diversity and biogeography. Environ. Microbiol. 16, 467-485. Kraft, N. J. B., Adler, P. B., Godoy, O., James, E. C., Fuller, S., and Levine, J. M. (2015). Community assembly, coexistence and the environmental filtering metaphor. Funct. Ecol. 29, 592–599. doi:10.1111/1365-2435.12345. Krishfield, R. A., Proshutinsky, A., Tateyama, K., Williiams, W., Carmack, E. C., McLaughlin, F., et al. (2014). Deterioration of perennial sea ice in the Beaufort Gyre from 2003 to 2012 and its impact on the oceanic freshwater cycle. J. Geophys. Res. Ocean. 119, 1271–1305. doi:10.1002/2013JC008999. Kristiansen, S., and Farbrot, T. (1991). Nitrogen uptake rates in phytoplankton and ice algae in the Barents Sea, Polar Res. 10, 187-192. Kumar, S., Stecher, G., and Tamura, K. (2016). MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Mol. Biol. Evol., 1–5. doi:10.1093/molbev/msw054. Kunin, V., Copeland, A., Lapidus, A., Mavromatis, K., and Hugenholtz, P. (2008). A bioinformatician’s guide to metagenomics. Microbiol. Mol. Biol. Rev. 72(4), 557–78. Lajeunesse, T. C., Andersen, R. A., and Galbraith, D. W. (2005). Symbiodinium (Pyrrophyta) genome sizes (DNA content) are smallest among dinoflagellates. J. Phycol., 41, 880–886. doi:10.1111/j.1529-8817.2005.00111.x.

127

Landry, M. R., and Calbet, A. (2004). Microzooplankton production in the oceans. ICES J. Mar. Sci. 61, 501–507. doi:10.1016/j.icesjms.2004.03.011. Laurion, I., Demers, S., and Vezina, A. (1995). The microbial food web associated with the ice algal assemblage: biomass and bacterivory of nanoflagellate protozoans in Resolute Passage. Mar. Ecol. Prog. Ser. 120, 77–87. Le Fouest, V., Manizza, M., Tremblay, B., and Babin, M. (2015). Modelling the impact of riverine DON removal by marine bacterioplankton on primary production. Biogeosciences 12, 3385– 3402. doi:10.5194/bg-12-3385-2015. Leander, B. S., and Keeling, P. J. (2003). Morphostasis in alveolate evolution. Trends Ecol. Evol. 18, 395–402. doi:10.1016/S0169-5347(03)00152-6. Lee, S. H., and Whitledge, T. E. (2004). Primary and new production in the deep Canada Basin during summer 2002. Polar Biol. 28, 190–197. doi:10.1007/s00300-004-0676-3. Lee, S. H., Joo, H. M., Liu, Z., Chen, J., and He, J. (2012). Phytoplankton productivity in newly opened waters of the Western Arctic Ocean. Deep Sea Res. Part II: Topical Studies in Oceanography, 81-84, 18–27. Leeuwen, S. M., Molen, J., Ruardij, P., Fernand, L., and Jickells, T. (2012). Modelling the contribution of deep chlorophyll maxima to annual primary production in the North Sea. Biogeochem. 113, 137–152. doi:10.1007/s10533-012-9704-5. Legendre, P., and Legendre, L. (1998). Numerical ecology, second English ed. Elsevier Science BV, Amsterdam, pp. 853. Levinsen, H., and Nielsen, T. G. (2002). The trophic role of marine pelagic ciliates and heterotrophic dinoflagellates in arctic and temperate coastal ecosystems: A cross-latitude comparison. Limnol. Oceanogr. 47, 427–439. doi:10.4319/lo.2002.47.2.0427. Levinsen, H., Nielsen, T. G., and Hansen, B. W. (2000). Annual succession of marine pelagic protozoans in Disko Bay, West Greenland, with emphasis on winter dynamics. Mar. Ecol. Prog. Ser. 206, 119–134. doi:10.3354/meps206119. Levinsen, H., Nielsen, T.G., and Hansen, B.W. (1999). Plankton community structure and carbon cycling on the western coast of greenlandduring the stratified summer situation. II. Heterotrophic dinoflagellates and ciliates. Aquat. Microb. Ecol. 16, 217–232. Lewandowska, A. M., Boyce, D. G., Hofmann, M., Matthiessen, B., Sommer, U., and Worm, B. (2014). Effects of sea surface warming on marine plankton. Ecol. Lett. 17, 614–623. doi:10.1111/ele.12265. Li, W. K. W., Andersen, R. A., Gifford, D. J. et al. (2011). Planktonic microbes in the Gulf of Maine Area. PLoS ONE, 6, e20981. Li, W. K. W., Mclaughlin, F. A., Lovejoy, C., and Carmack, E. C. (2009). Smallest algae thrive as the Arctic Ocean freshens. Science, 326, 539. Li, W. K. W., Dickie, P.M. (2001). Monitoring phytoplankton, bacterioplankton, and virioplankton in a coastal inlet (Bedford Basin) by flow cytometry. Cytometry 44, 236–246 Lima-Mendez, G., Faust, K., Henry, N., Decelle, J., Colin, S., Carcillo, F., et al. (2015). Determinants of community structure in the global plankton interactome. Science, 348, doi: 10.1126/science.1262073. Lin, S. (2006). The smallest dinoflagellate genome is yet to be found: A comment on LaJeunesse et al. Symbiodinium (Pyrrhophyta) genome sizes (DNA content) are smallest among dinoflagellates. J. Phycol., 748, 746–748. doi:10.1111/j.1529-8817.2006.00213.x.

128

Lin, S., Zhang, H., Hou, Y., Miranda, L., and Bhattacharya, D. (2006). Development of a dinoflagellate-oriented PCR primer set leads to detection of picoplanktonic dinoflagellates from Long Island Sound. Appl. Environ. Microbiol. 72, 5626–5630. doi:10.1128/AEM.00586-06. Lin, S., Zhang, H., Hou, Y., Zhuang, Y., and Miranda, L. (2009). High-level diversity of dinoflagellates in the natural environment, revealed by assessment of mitochondrial cox1 and cob genes for dinoflagellate DNA barcoding. Appl. Environ. Microbiol. 75, 1279–1290. doi:10.1128/AEM.01578-08 Lincoln, B., Rippeth, T., Lenn, Y.-D., Timmermans, M., Williams, W. J., and Bacon, S. (2016). Wind-driven mixing at intermediate depths in an ice-free Arctic Ocean. Geophys. Res. Lett. 43, 9749–9756. doi:10.1002/2016GL070454.Received. Liu, H. (2012). Potential risk of Mesodinium rubrum bloom in the aquaculture area of Dapeng’ao cove, China: diurnal changes in the ciliate community structure in the surface water. Oceanologia 54, 109-117. doi:10.5697/oc.54-1.109. Loder, G.J., Meunier, C., Wiltshire, K.H., Boersna, M., and Aberle, N. (2011). The role of ciliates, heterotrophic dinoflagellates and copepods in structuring spring plankton communities at Helgoland Roads, North Sea. Mar. Biol., 158, 1551-1580. Logares, R., Shalchian-Tabrizi, K., Boltovskoy, A., and Rengefors, K. (2007). Extensive dinoflagellate phylogenies indicate infrequent marine-freshwater transitions. Mol. Phylogenet. Evol. 45, 887–903. doi:10.1016/j.ympev.2007.08.005. Lopez-Garcia, P., Rodriguez-Valera, F., Pedrós-Alió, C., and Moreira, D. (2001). Unexpected diversity of small eukaryotes in deep-sea Antarctic plankton. Nature 409, 603–607. López-Sandoval, D. C., Rodríguez-Ramos, T., Cermeño, P., Sobrino, C., and Marañón, E. (2014). Photosynthesis and respiration in marine phytoplankton: Relationship with cell size, taxonomic affiliation, and growth phase. J. Exp. Mar. Bio. Ecol. 457, 151–159. doi:10.1016/j.jembe.2014.04.013. Lovejoy and Plankton Network Experts, CBMP. (2016). State of the Arctic Marine Biodiversity Report (SAMBR). Circumpolarbiodiversity Monitoring Program CAFF. Iceland. Lovejoy, C., Comeau, A. and Thaler, M. (2015). Curated reference database of SSU rRNA for northern marine and freshwater communities of archaea, bacteria and microbial eukaryotes.Nordicana, D23. DOI: 10.5885/45409XD-79A199B76BCC4110. Lovejoy, C. (2014). Changing views of arctic protists (marine microbial eukaryotes) in a changing Arctic. Acta Protozoologica, 53, 91-100. DOI: 10.4467/16890027AP.14.009.1446. Lovejoy, C., and Potvin, M. (2011). Microbial eukaryotic distribution in a dynamic Beaufort Sea and the Arctic Ocean. J. Plankton Res. 33, 431–444. doi:10.1093/plankt/fbq124. Lovejoy, C., Comeau, A., and Thaler, M. (2016). Curated reference database of SSU rRNA for northern marine and freshwater communities of archaea, bacteria and microbial eukaryotes, v. 1.1 (2002-2008). Nordicana D23, doi: 10.5885/45409XD-79A199B76BCC4110. Lovejoy, C., Galand, P. E., and Kirchman, D. L. (2010). Picoplankton diversity in the Arctic Ocean and surrounding seas. Marine Biodiversity, 41, 5–12. Lovejoy, C., Legendre, L., and Price, N. (2002). Prolonged diatom blooms and microbial food web dynamics: experimental results from an Arctic polynya. Aquat. Microb. Ecol. 29, 267–278. doi:10.3354/ame029267. Lovejoy, C., Massana, R., and Pedrós-Alió, C. (2006). Diversity and distribution of marine microbial eukaryotes in the Arctic Ocean and adjacent seas. Appl. Environ. Microbiol. 72, 3085-3095. doi:10.1128/AEM.72.5.3085

129

Lovejoy, C., Vincent, W. F., Bonilla, S., Roy, S., Martineau, M.-J., Terrado, R., et al. (2007). Distribution, phylogeny, and growth of cold-adapted picoprasinophytes in Arctic Seas. J. Phycol. 43, 78–89. doi:10.1111/j.1529-8817.2006.00310.x. Lovejoy, C., Vincent, W. F., Frenette, J., and Dodson, J. (1993). Microbial gradients in a turbid estuary: application of a new method for protozoan community analysis. Limnol. Oceanogr. 38, 1295–1303. Lovejoy, C., Vincent, W.F., Bonilla, S., Roy, S., Martineau, M.-J., Terrado, R., Potvin, M., Massana, R., Pedrós-Alió, C. (2007). Distribution, phylogeny, and growth of cold-adapted picoprasinophytes in Arctic seas. J. Phycol., 43:78–89. Lozupone, C., and Knight, R. (2005). UniFrac: a new phylogenetic method for comparing microbial communities. Appl. Environ. Microbiol. 71, 8228–8235. doi:10.1128/AEM.71.12.8228. Luddington, I. A., Lovejoy, C., and Kaczmarska, I. (2016). Species-rich meta-communities of the diatom order Thalassiosirales in the Arctic and northern Atlantic Ocean. J. Plank. Res. 38, 781- 797. doi: 10.1093/plankt/fbw030. Lukeš, J., Flegontova, O., and Horák, A. (2015). Diplomenids. Curr. Biol. 25, R702–R704. doi:10.1016/j.cub.2015.04.052. Lynn, D. H. (2012). Ciliophora. eLS, 1–12. doi:10.1002/9780470015902.a0001966.pub3. Lynn, D.H. (2008). The Ciliated Protozoa: characterization, classification, and guide to the literature. 3rd ed. Springer, Dordrecht. Macdonald, R. W. (2000), Arctic estuaries and ice: A positive-negative estuarine couple, in The Freshwater Budget of the Arctic Ocean, edited by E. L. Lewis, pp. 383 – 407, Kluwer Acad., Dordrecht, Netherlands. Macdonald, R. W., Wong, C. S., and Erickson, P. E. (1987). The distribution of nutrients in the southeastern Beaufort Sea: Implications for water circulation and primary production. J. Geophys. Res. 92, 2939. doi:10.1029/JC092iC03p02939. Maidak, B. L., Cole, J. R., Lilburn, T. G., Parker, C. T., Saxman, P. R., Stredwick, J. M., Garrity, G. M., et al. (2000). The RDP (Ribosomal Database Project) continues. Nuc. Acids Res., 28, 173–4. Manton, I., and Parke, M. (1960). Further observations on small green flagellates with special reference to possible relatives of Chromulina pusilla Butcher. J. Mar. Biollogical Assoc. UK 39, 275–298. Marie, D., Shi, X. L., Rigaut-Jalabert, F., and Vaulot, D. (2010). Use of flow cytometric sorting to better assess the diversity of small photosynthetic eukaryotes in the English Channel. FEMS Microbiol. Ecol. 72, 165–178. doi:10.1111/j.1574-6941.2010.00842.x. Marquardt, M., Vader, A., Stubner, E., Reigstad, M., and Gabrielsen, T. (2016). Strong seasonality of marine microbial eukaryotes in a high-Arctic fjord (Isfjorden, West Spitsbergen). Appl. Environ. Microbiol. 82, 1866-1880. doi:doi: 10.1128/AEM.03208-15. Martin, J., Dumont, D., and Tremblay, J.-É. (2013). Contribution of subsurface chlorophyll maxima to primary production in the coastal Beaufort Sea (Canadian Arctic): A model assessment. J. Geophys. Res. Ocean. 118, 5873–5886. doi:10.1002/2013JC008843. Martin, J., Tremblay, J. É., and Price, N. M. (2012). Nutritive and photosynthetic ecology of subsurface chlorophyll maxima in Canadian Arctic waters. Biogeosciences 9, 5353–5371. doi:10.5194/bg-9-5353-2012. Martin, J., Tremblay, J., Gagnon, J., Tremblay, G., Lapoussière, A, Jose, C., Poulin, M., et al. (2010). Prevalence, structure and properties of subsurface chlorophyll maxima in Canadian Arctic waters. Mar. Ecol. Prog. Ser., 412, 69–84.doi:10.3354/meps08666.

130

Martiny, J. B. H., Bohannan, B. J. M., Brown, J. H., Colwell, R. K., Fuhrman, J. A., Green, J. L., et al. (2006). Microbial biogeography: putting microorganisms on the map. Nat. Rev. Microbiol. 4, 102–112. DOI:10.1038/nrmicro1341. Massana, R., Campo, J., and Sieracki, M. E. (2014). Exploring the uncultured microeukaryote majority in the oceans: reevaluation of ribogroups within stramenopiles. ISME J. 8, 854–866. doi:10.1038/ismej.2013.204. Massana, R., Pernice, M., Bunge, J. A., Campo, J., and Csic, M. (2011). Sequence diversity and novelty of natural assemblages of picoeukaryotes from the Indian Ocean. ISME J. 5, 184–195. doi:10.1038/ismej.2010.104. McLaughlin, F. A., and Carmack, E. C. (2010). Deepening of the nutricline and chlorophyll maximum in the Canada Basin interior, 2003-2009. Geophys. Res. Lett. 37, L24602. doi:10.1029/2010GL045459. McLaughlin, F., Carmack, E., Macdonald, R.., Melling, H., Swift, J., Wheeler, P., et al. (2004). The joint roles of Pacific and Atlantic-origin waters in the Canada Basin, 1997–1998. Deep Sea Res. Part I Oceanogr. Res. Pap. 51, 107–128. doi:10.1016/j.dsr.2003.0910. Mclaughlin, F., Shimada, E. K., and Carmack, E. E. (2005). The hydrography of the southern Canada Basin, 2002. Polar Biol. 28, 182–189. doi:10.1007/s00300-004-0701-6. McManus, G. B., and Katz, L. A. (2009). Molecular and morphological methods for identifying plankton: what makes a successful marriage? J. Plank. Res. 31, 1119–1129. doi:10.1093/plankt/fbp061. McPhee, M. (2012). Intensification of geostrophic currents in the Canada Basin, Arctic Ocean. J. Clim., 26, 1-11. Medinger, R., Nolte, V., Pandey, R. A. M. V., Jost, S. (2010). Diversity in a hidden world: potential and limitation of next- generation sequencing for surveys of molecular diversity of eukaryotic microorganisms. Mol. Ecol. 19, 32–40. doi:10.1111/j.1365-294X.2009.04478.x.Diversity. Melnikov, I. a, Kolosova, E. G., Welch, H. E., and Zhitina, L. S. (2002). Sea ice biological communities and nutrient dynamics in the Canada Basin of the Arctic Ocean. Deep Sea Res. Part I: Oceanographic Research Papers, 49(9), 1623–1649. Menden-Deuer, S., and Lessard, E. J. (2000). Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 45, 569–579. doi:10.4319/lo.2000.45.3.0569. Meyer, F., Paarmann, D., D’Souza, M., Olson, R., Glass, E. M., Kubal, M., Paczian, T., et al. (2008). The metagenomics RAST server - a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC bioinformatics, 9, 386. Mitra, A., Flynn, K. J., Burkholder, J. M., Berge, T., Calbet, A., Raven, J. A., et al. (2014). The role of mixotrophic protists in the biological carbon pump. Biogeosciences 11, 995–1005. doi:10.5194/bg-11-995-2014. Mitra, A., Flynn, K. J., Tillmann, U., Raven, J. A., Caron, D., Stoecker, D. K., et al. (2016). Defining planktonic protist functional groups on mechanisms for energy and nutrient acquisition: incorporation of diverse mixotrophic strategies. Protist 167. doi:10.1016/j.protis.2016.01.003.

Møller, E. F., Nielsen, T. G., and Richardson, K. (2006). The zooplankton community in the Greenland Sea: Composition and role in carbon turnover. Deep. Sea Res. I 53, 76–93. doi:10.1016/j.dsr.2005.09.007.

131

Monier, A., Terrado, R., Thaler, M., Comeau, A. M., Medrinal, E., and Lovejoy, C. (2013). Upper Arctic Ocean water masses harbor distinct communities of heterotrophic flagellates. Biogeosciences 10, 4273–4286. doi:10.5194/bg-10-4273-2013. Monier, A., Comte, J., Babin, M., Forest, A., Matsuoka, A., and Lovejoy, C. (2015). Oceanographic structure drives the assembly processes of microbial eukaryotic communities. ISME J. 9, 990– 1002. doi:10.1038/ismej.2014.197. Monier, A., Terrado, R., Thaler, M., Comeau, A., Medrinal, E., and Lovejoy, C. (2013). Upper Arctic Ocean water masses harbor distinct communities of heterotrophic flagellates. Biogeosciences 10, 4273–4286. doi:10.5194/bg-10-4273-2013. Montagnes, D. J. S. (1996). Growth responses of planktonic ciliates in the genera Strobilidium and Strombidium. Mar. Ecol. Prog. Ser. 130, 241–254. Montagnes, D. J. S., Allen, J., Brown, L., Bulit, C., Davidson, R., Fielding, S., et al. (2010). Role of ciliates and other microzooplankton in the Irminger Sea (NW Atlantic Ocean). Mar. Ecol. Prog. Ser. 411, 101–115. doi:10.3354/meps08646. Monti, M., and Minocci, M. (2013). Microzooplankton along a transect from northern continental Norway to Svalbard. Polar Res., 32, 1–9. doi:10.3402/polar.v32i0.19306. Morison, J., Kwok, R., Peralta-Ferriz, C., Alkire, M., Rigor, I., Andersen, R., et al. (2012). Changing Arctic Ocean freshwater pathways. Nature 481, 66–70. doi:10.1038/nature10705. Mouquet, N., Devictor, V., Meynard, C. N., Munoz, F., Bersier, L. F., Chave, J., et al. (2012). Ecophylogenetics: Advances and perspectives. Biol. Rev. 87, 769–785. doi:10.1111/j.1469- 185X.2012.00224.x. Mundy, C. J., Gosselin, M., Ehn, J., Gratton, Y., Rossnagel, A., Palmer, M., et al. (2009). Contribution of under-ice primary production to an ice-edge upwelling phytoplankton bloom in the Canadian Beaufort Sea. Geophys. Res. Lett. 36, 1–5. doi:10.1029/2009GL038837. Nelson, R. J., Ashjian, C. J., Bluhm, B. A., Conlan, K. E., Gradinger, R. R., Grebmeier, J. M., et al. (2014). Biodiversity and biogeography of the lower trophic taxa of the Pacifi c Arctic Region?: Sensitivities to climate change, in J.M. Grebmeier and W. Maslowski (eds.), The Pacific Arctic Region: Ecosystem status and trends in a rapidly changing environment. Spinger, Dorrdrecht.. 269–336. doi:10.1007/978-94-017-8863-2. Nielsen, K. M., Johnsen, P. J., Bensasson, D. and Daffonchio, D. (2007). Thematic issue on horizontal gene transfer release and persistence of extracellular DNA. Environ. Biosafety Res. 6, 37–53. Nielsen, L. T. and Kiørboe, T. (2015). Feeding currents facilitate a mixotrophic way of life. ISME J. 9, 2117–2127. doi:10.1038/ismej.2015.27. Nishino, S., Itoh, M., Kawaguchi, Y., Kikuchi, T., and Aoyama, M. (2011b). Impact of an unusually large warm-core eddy on distributions of nutrients and phytoplankton in the southwestern Canada Basin during late summer/early fall 2010. Geophys. Res. Let., 38, L16692. DOI: 10.1029/2011GL047885. Nishino, S., Kikuchi, T., Yamamoto-Kawai, M., Kawaguchi, Y., Hirawake, T., and Itoh, M. (2011a). Enhancement/reduction of biological pump depends on ocean circulation in the sea-ice reduction regions of the Arctic Ocean. J. Oceanog., 67, 305–314. Nishino, S., Shimada, K., Itoh, M., Yamamoto-Kawai, M., and Chiba, S. (2008). East–west differences in water mass, nutrient, and chlorophyll a distributions in the sea ice reduction region of the western Arctic Ocean. J. Geophys. Res. 113, C00A01. doi:10.1029/2007JC004666.

132

Not, F., del Campo, J., Balague, V., de Vargas, C., and Massana, R. (2009). New insights into the diversity of marine picoeukaryotes. PLoS One 4, e7143. doi:10.1371/journal.pone.0007143. Not, F., Siano, R., Kooistra, W. H. C. F., Simon, N., Vaulot, D., and Probert, I. (2012). Diversity and ecology of eukaryotic marine phytoplankton. In Advances in Botanical Research (Elsevier Ltd.), 1–53. O’Connor, M. I., Piehler, M. F., Leech, D. M., Anton, A., and Bruno, J. F. (2009). Warming and resource availability shift food web structure and metabolism. PLoS Biol. 7, 3–8. doi:10.1371/journal.pbio.1000178. Oestmann, D. J., and Lewis, D. H. (1996). Improved cell culture propagation of Amyloodinium ocellatum. Dis. Aquat. Organ. 24, 173–178. Okolodkov, Y. B. (2005). The global distributional patterns of toxic, bloom dinoflagellates recorded from the Eurasian Arctic. Harmful Algae, 4, 351–369. doi:10.1016/j.hal.2004.06.016. Okolodkov, Y. B., and Dodge, J. D. (1996). Biodiversity and biogeography of planktonic dinoflagellates in the Arctic Ocean. J. Exp. Mar. Bio. Ecol. 202, 19–27. Oksanen, J., F. G. Blanchet, R. Kindt, P. Legendre, P. R. Minchin, R. B. O’Hara, G. L. Simpson, P. Solymos, M. H. M. Stevens, and H. Wagner. (2015). Vegan: community ecology package. https://CRAN.R-project.org/package=vegan. Olli, K. (2004). Temporary cyst formation of Heterocapsa triquetra (Dinophyceae) in natural populations. Mar. Biol. 145, 1–8. doi:10.1007/s00227-004-1295-9. Olli, K., and Anderson, D. M. (2002). High encystment success of the dinoflagellate Scripsiella cf. lachrymosa in culture experiments. J. Phycol. 156, 145–156. Onda, D. F. L., Lluisma, A. O., & Azanza, R. V (2014). Development, morphological characteristics and viability of temporary cysts of Pyrodinium bahamense var. compressum (Dinophyceae) in vitro Eur. J. Phycol. 49, 265–275. doi:10.1080/09670262.2014.915062. . Onda, D. F. L., Medrinal, E., Comeau, A. M., Babin, M., and Lovejoy, C. (2017). Seasonal and interannual changes in ciliate and dinoflagellate species assemblages in the Arctic Ocean (Amundsen Gulf, Beaufort Sea, Canada). Front. Mar. Sci. 4, article 16. doi:10.3389/FMARS.2017.00016. Orr, R. J. S., Murray, S. a, Stüken, A., Rhodes, L., and Jakobsen, K. S. (2012). When naked became armored: an eight-gene phylogeny reveals monophyletic origin of theca in dinoflagellates. PLoS One 7, e50004. doi:10.1371/journal.pone.0050004. Orsi, W., Edgcomb, V., Faria, J., Foissner, W., Fowle, W. H., Hohmann, T., et al. (2012). Class Cariacotrichea, a novel ciliate taxon from the anoxic Cariaco Basin, Venezuela. Int. J. Syst. Evol. Microbiol. 62, 1425–1433. doi:10.1099/ijs.0.034710-0. Padilla, C. C., Ganesh, S., Gantt, S., Huhman, A., Parris, D. J., and Gonzalez, J. M. (2015). Standard filtration practices may significantly distort planktonic microbial diversity estimates. Front. Microbiol. 6, 1–10. doi:10.3389/fmicb.2015.00547. Paradis, E., Claude, J., and Strimmer, L. (2004). APE: analyses of phylogenetics and evolution in R language. Bioinformatics 20: 289290. Paranjape, M. A. (1990). Microzooplankton herbivory on the Grand Bank (Newfoundland, Canada): a seasonal study. Mar. Biol. 328, 321–328. Park, J-Y., Kug, J-S., Bader, J., Rolph, B. and Kwon, M. (2015). Amplified Arctic warming by phytoplankton under greenhouse warming. Proc. Natl. Acad. Sci. U. S. A. 112, 5921-5926.

133

Park, M. G., and Kim, M. (2010). Prey specificity and feeding of the thecate mixotrophic dinoflagellate Fragilidium duplocampaforme. J. Phycol. 46, 424–432. doi:10.1111/j.1529- 8817.2010.00824.x. Paulic, J., Bartzen, B., Bennett, R., Conlan, K., Hardwood, L., Howland, K., et al. (2012). Ecosystem overview report for the Darnley Bay area of interest (AOI). DFO Can. Sci. Advis. Sec. Res. Doc. 2011/062, 1–63. Perrette, M., Yool, A., Quartly, G. D., and Popova, E. E. (2010). Near-ubiquity of ice-edge blooms in the Arctic. Biogeosciences, 8, 515-524. Pesant, S., Legendre, L., Gosselin, M., Ashjian, C., Booth, B., and Fortier, L. (1998). Pathways of carbon cycling in the euphotic zone: the fate of large- sized phytoplankton in the Northeast Water Polynya. J Plank. Res., 20, 1267–1291. Pickart, R. S. and G. Stossmeister (2009). Outflow of Pacific water from the Chukchi Sea to the Arctic Ocean. Chinese Journal of Polar Science 19, 135-148, ark:/85065/d7m61kx9 Piwosz, K., Wiktor, M., Niemi, A., Tatarek, A., and Michel, C. (2013). Mesoscale distribution and functional diversity of picoeukaryotes in the first-year sea ice of the Canadian Arctic. ISME J. 7, 1461–1471. doi:10.1038/ismej.2013.39. Pochon, X., Putnam, H. M., and Gates, R. D. (2014). Multi-gene analysis of Symbiodinium dinoflagellates: a perspective on rarity, symbiosis, and evolution. PeerJ 2, e394. doi:10.7717/peerj.394. Pommier, J., Gosselin, M., and Michel, C. (2009). Size-fractionated phytoplankton production and biomass during the decline of the northwest Atlantic spring bloom. J. Plank. Res., 31(4), 429– 446. Potvin, É., Rochon, A., and Lovejoy, C. (2013). Cyst-theca relationship of the arctic dinoflagellate cyst Islandinium minutum (Dinophyceae) and phylogenetic position based on SSU rDNA and LSU rDNA. J. Phycol., 49, 848–866. doi:10.1111/jpy.12089. Powell, J. R., Karunaratne, S., Campbell, C. D., Yao, H., Robinson, L., and Singh, B. K. (2015). Deterministic processes vary during community assembly for ecologically dissimilar taxa. Nat. Commun. 6, 8444. doi:10.1038/ncomms9444. Price, M. N., Dehal, P. S., and Arkin, A. P. (2010). FastTree 2 – Approximately Maximum- Likelihood Trees for Large Alignments. PLoS One 5, e9490. doi:10.1371/journal.pone.0009490.

Proshutinsky, A. (2002). The role of the Beaufort Gyre in Arctic climate variability: Seasonal to decadal climate scales. Geophys. Res. Let., 29(23), 2100. Proshutinsky, A., Krishfield, R., and Barber, D. (2009). Preface to special section on Beaufort Gyre Climate System Exploration Studies: Documenting key parameters to understand environmental variability. J. Geophys. Res., 114, C00A08. Pruesse, E., Quast, C., Knittel, K., Fuchs, B. M., 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. doi:10.1093/nar/gkm864. Ptacnik, R., Gomes, A., Royer, S. J., Tsagaraki, T. M., Sazhin, A. F., Berger, S. A., et al. (2016). A light-induced shortcut in the planktonic microbial loop. Sci. Rep. 6, 29286. doi:10.1038/srep29286. Puigserver, M., and Zingone, A. (2002). Prorocentrum nux sp. nov. (Dinophyceae), a small planktonic dinoflagellate from the Mediterranean Sea, and discussion of P. nanum and P. pusillum. Phycologia 41, 29–38. doi:10.2216/i0031-8884-41-1-29.1.

134

Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Glo, F. O., et al. (2013). The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acid Res. 41, 590–596. doi:10.1093/nar/gks1219. R Development Core Team (2008). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3–900051–07–0, URL http://www.R–project.org. Rainville, L., Lee, C. M., and Woodgate, R. A. (2011). Impact of wind-driven mixing in the Arctic Ocean. Oceanography 24: 136–145. Ramette, A. (2007). Multivariate analyses in microbial ecology. FEMS Microbiol. Ecol. 62, 142–160. doi:10.1111/j.1574-6941.2007.00375.x. Reeder, J., and Knight, R. (2010). Rapidly denoising pyrosequencing amplicon reads by exploiting rank-abundance distributions. Nat. Methods 7, 668–669. doi:10.1038/nmeth0910-668b. Rene, A., Camp, J., and Garcés, E. (2015). Diversity and phylogeny of Gymnodiniales (Dinophyceae) from the NW Mediterranean Sea revealed by a morphological and molecular approach. Protist 166, 234–263. doi:10.1016/j.protis.2015.03.001. Richerol, T., Pienitz, R., and Rochon, A. (2012). Marine micropaleontology modern dinoflagellate cyst assemblages in surface sediments of Nunatsiavut fjords. Mar. Micropaleontol. 88–89, 54– 64. doi:10.1016/j.marmicro.2012.03.002. Riebesell, U., Gattuso, J. P., Thingstad, T. F., and Middelburg, J. J. (2013). Arctic Ocean acidification: pelagic ecosystem and biogeochemical responses during a mesocosm study. Biogeosciences 10, 5619–5626. doi:10.1594/PANGAEA.769833. Riisgaard, K., Swalethorp, R., Kjellerup, S., Juul-Pedersen, T., and Nielsen, T. G. (2014). Trophic role and top-down control of a subarctic protozooplankton community. Mar. Ecol. Prog. Ser. 500, 67–82. doi:10.3354/meps10706. Rintala, J.-M., Spilling, K., and Blomster, J. (2007). Temporary cyst enables long-term dark survival of Scrippsiella hangoei (Dinophyceae). Mar. Biol. 152, 57–62. doi:10.1007/s00227-007-0652-x. Rizzo, P. J. (2003). Those amazing dinoflagellate chromosomes. Cell Res. 13, 215–217. doi:10.1038/sj.cr.7290166. Romari, K., Vaulot, D., Curie, M., and Cedex, R. (2004). Composition and temporal variability of picoeukaryote communities at a coastal site of the English Channel from 18S rDNA sequences. Limnol. Oceanogr. 49, 784–798. Rothhaupt, K.O. (1996) Laboratory experiments with a mixotrophic chrysophyte and obligately phagotrophic and phototrophic competitors. Ecology 77: 716–724. Sal, S., Alonso-Saez, L., Bueno, J., Garcia, F. C., and Lopez-Urrutia, A. (2015). Thermal adaptation, phylogeny, and the unimodal size scaling of marine phytoplankton growth. Limnol. Oceanogr. 60, 1212–1221. doi:10.1002/lno.10094. Saldarriaga, J. F., Taylor, F. J. R., Cavalier-Smith, T., Menden-Deuer, S., and Keeling, P. J. (2004). Molecular data and the evolutionary history of dinoflagellates. Eur. J. Protistol. 40, 85–111. doi:10.1016/j.ejop.2003.11.003. Saldarriaga, J. F., Taylor, F. J. R., Keeling, P. J., and Cavalier-smith, T. (2001). Dinoflagellate nuclear SSU rRNA phylogeny suggests multiple plastid losses and replacements. J. Mol. Evol. 53, 204–213. doi:10.1007/s002390010210. Sandaa, R.A. (2009). Viral pool, environmental. In: Encyclopedia in Microbiology, third edition, Ed: Schaechter M. Section: Environmental Microbiology and Ecology. pp 553-567. Elsevier Ltd.

135

Santamaria, M., Fosso, B., Consiglio, A., de Caro, G., Grillo, G., Licciuli, F., Liuni, S., Marzano, M., Alonso-Alemany, D., Valiente, G., and Pesole, G. (2012). Reference databases for taxonomic assignment in metagenomics. Brief. Bioinform., 13, 682-695. Santoferrara, L. F., Grattepanche, J., Katz, L. A., and McManus, G. B. (2014). Pyrosequencing for assessing diversity of eukaryotic microbes: analysis o.f data on marine planktonic ciliates and comparison with traditional methods. Environ Microbiol. 16, 2752–2763. doi:10.1111/1462- 2920.12380 Santoferrara, L. F., Grattepanche, J., Katz, L. A., and Mcmanus, G. B. (2016). Patterns and processes in microbial biogeography: do molecules and morphologies give the same answers? ISME J. 10, 1779–1790. doi:10.1038/ismej.2015.224. Schlitzer, R. (2016). Ocean Data View, http://odv.awi.de Schloss, P. D., Westcott, S. L., Ryabin, T., Hall, J. R., Hartmann, M., Hollister, E. B., et al. (2009). Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537–7541. doi:10.1128/AEM.01541-09. Seuthe, L., Darnis, G., Wexels, C., Wassmann, P., and Fortier, L. (2007). Winter – spring feeding and metabolism of Arctic copepods: insights from faecal pellet production and respiration measurements in the southeastern Beaufort Sea. Polar Biol. 30, 427–436. doi:10.1007/s00300- 006-0199-1. Seuthe, L., Iversen, K. R., and Narcy, F. (2011). Microbial processes in a high-latitude fjord (Kongsfjorden, Svalbard): II. Ciliates and dinoflagellates. Polar Biol. 34, 751–766. doi:10.1007/s00300-010-0930-9. Shadwick, E. H., Thomas, H., Chierici, M., Else, B., Fransson, A., Michel, C., et al. (2011). Seasonal variability of the inorganic carbon system in the Amundsen Gulf region of the southeastern Beaufort Sea. Limnol. Oceanogr. 56, 303–322. doi:10.4319/lo.2011.56.1.0303. Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., et al. (2003). Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res., 13, 2498–2504. doi:10.1101/gr.1239303.metabolite. Sherr, E. B., and Sherr, B. F. (2002). Significance of predation by protists in aquatic microbial food webs. Antonie van Leeuwenhoeak, 81, 293–308. Sherr, E. B., and Sherr, B. F. (2009). Capacity of herbivorous protists to control initiation and development of mass phytoplankton blooms. Aquat. Microb. Ecol. 57, 253–262. doi:10.3354/ame01358. Sherr, E. B., Sherr, B. F., and Ross, C. (2013). Microzooplankton grazing impact in the Bering Sea during spring sea ice conditions. Deep. Res. II 94, 57–67. Sherr, E., and Sherr, B. (1988). Role of microbes in pelagic food webs: A revised concept. Limnol. Oceanogr. 33, 1225–1227. doi:10.4319/lo.1988.33.5.1225. Shimada, K., Itoh, M., Nishino, S., Mclaughlin, F., Carmack, E., and Proshutinsky, A. (2005). Halocline structure in the Canada Basin of the Arctic Ocean. Geophys. Res. Lett., 32, 1–5. doi:10.1029/2004GL021358. Siano, R., Alves-de-souza, C., Foulon, E., Bendif, E. M., Simon, N., Guillou, L., et al. (2011). Distribution and host diversity of Amoebophryidae parasites across oligotrophic waters of the Mediterranean Sea. Biogeosciences 8, 267–278. doi:10.5194/bg-8-267-2011.

136

Skjöldborg, A., Gissel, T., Levinsen, H., Madsen, S. D., Thingstad, T. F., and Winding, B. (2003). Impact of changing ice cover on pelagic productivity and food web structure in Disko Bay, West Greenland: a dynamic model approach. Deep. Res. Part I 50, 171–187. Skovgaard, A., Massana, R., Balague, V., Mar, P., and Saiz, E. (2005). Phylogenetic position of the copepod-infesting parasite Syndinium turbo (Dinoflagellata, Syndinea). Protist 156, 413–423. doi:10.1016/j.protis.2005.08.002. Smith, S.D., Muench, R.D., Pease, C. (1990). Polynyas and leads: an overview of physical processes and environment. J. Geophys. Res., 95, 9461–9480. Søreide, J. E., Leu, E., Berge, J., Graeve, M., and Falk-Petersen, S. (2010). Timing of blooms, algal food quality and Calanus glacialis reproduction and growth in a changing Arctic. Glob. Chang. Biol. 16, 3154–3163. doi:10.1111/j.1365-2486.2010.02175.x. Sørensen, N., Daugbjerg, N., and Richardson, K. (2013). Choice of pore size can introduce artefacts when filtering picoeukaryotes for molecular biodiversity studies. Microbiol. Aquat. Syst. 65, 964–968. doi:10.1007/s00248-012-0174-z. Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics, 30, 1312-3. doi:10.1093/bioinformatics/btu033. Steele, M. (2004). Circulation of summer Pacific halocline water in the Arctic Ocean. J. Geophys. Res., 109(C2), C02027. doi:10.1029/2003JC002009 Stefels, J. (2000). Physiological aspects of the production and conversion of DMSP in marine algae and higher plants. J. Sea Res. 43, 183–197. doi:10.1016/S1385-1101(00)00030-7. Stegen, J. C., Lin, X., Fredrickson, J. K., Chen, X., Kennedy, D. W., Murray, C. J., et al. (2013). Quantifying community assembly processes and identifying features that impose them. ISME J. 7, 2069–79. doi:10.1038/ismej.2013.93. Stegen, J. C., Lin, X., Konopka, A. E., and Fredrickson, J. K. (2012). Stochastic and deterministic assembly processes in subsurface microbial communities. ISME J. 6, 1653–1664. doi:10.1038/ismej.2012.22. Steiner, N. S., Sou, T., Deal, C., Jackson, J. M., Jin, M., Popova, E., et al. (2016). The future of the subsurface chlorophyll-a maximum in the Canada Basin-A model intercomparison. J. Geophys. Res. Ocean. 121, 387–409. doi:10.1002/2015JC011232. Stentiford, G. D., and Shields, J. D. (2005). A review of the parasitic dinoflagellates Hematodiniumspecies and Hematodinium-like infections in marine crustaceans. Dis. Aquat. Organ. 66, 47–70. Stern, R.F., Horak, A., Andrew, R.L., Coffroth, M.A., Andersen, R.A., Kupper, F.C., Jameson, I. et al. (2010). Environmental barcoding reveals massive dinoflagellate diversity in marine environments. PLoS ONE 5, e13991. doi:10.1371/journal.pone.0013991. Stoecker, D. K. (1999). Mixotrophy among dinoflagellates. J. Eukaryot. Microbiol. 46, 397–401. Stoecker, D. K., and Capuzzo, J. M. (1990). Predation on protozoa: its importance to zooplankton. J. Plank. Res. 12, 891–908. doi:10.1093/plankt/12.5.891. Stoecker, D. K., Johnson, M. D., De Vargas, C., and Not, F. (2009). Acquired phototrophy in aquatic protists. Aquat. Microb. Ecol. 57, 279–310. doi:10.3354/ame01340. Stoecker, D. K., Taniguchi, A., and Michaels, A. E. (1988). Abundance of autotrophic, mixotrophic and heterotrophic planktonic ciliates in shelf and slope waters. Mar. Ecol. Prog. Ser., 50, 241– 254.

137

Strüder-Kypke, M. and Lynn, D.H. (2010). Comparative analysis of the mitochondrial cytochrome c oxidase subunit I (COI) gene in ciliates (Alveolata, Ciliophora) and evaluation of its suitability as a biodiversity marker. System. Biodiv., 8, 131-148, DOI: 10.1080/14772000903507744. Syvertsen, E. E. (1991). Ice algae in the Barents Sea: types of assemblages, origin, fate and role in the ice-edge phytoplankton bloom. Polar Res. 10, 277–288. doi: 10.1111/j.1751- 8369.1991.tb00653.x Tang, E.P.Y. (1996). Why do dinoflagellates have lower growth rates? J. Phycol. 32, 80–84. Taylor, F.J.R.M. (2004). Illumination or confusion? Dinoflagellate molecular phylogenetic data viewed from a primarily morphological standpoint. Phycol. Res., 52, 308-334. DOI: 10.1111/j.1440-183.2004.00360.x. Taylor, F. J. R., Hoppenrath, M., and Saldarriaga, J. F. (2008). Dinoflagellate diversity and distribution. Biodivers. Conserv. 17, 407–418. doi:10.1007/s10531-007-9258-3. Taylor, M. F. J. R. (1985). The taxonomy and relationships of red tide flagellates, in Toxic Dinoflagellates (Eds), eds. D. M. Anderson, A. W. Whie, and D. G. Baden (New York: Elsevier Science Publishing Co.), 11–26. Ter Braak, C.J.F. and Smilauer, P. (2002). CANOCO Reference manual and CanoDraw for Windows User’s guide: Software for Canonical Community Ordination (version 4.5). Microcomputer Power, Ithaca, New York. 500 pp. http://www.canoco.com. Terrado, R., Lovejoy, C., Massana, R., and Vincent, W. F. (2008). Microbial food web responses to light and nutrients beneath the coastal Arctic Ocean sea ice during the winter–spring transition. J. Mar. Syst., 74(3-4), 964–977. Terrado, R., Medrinal, E., Dasilva, C., Thaler, M., Vincent, W. F., and Lovejoy, C. (2011). Protist community composition during spring in an Arctic flaw lead polynya. Polar Biol. 34, 1901– 1914. doi:10.1007/s00300-011-1039-5. Terrado, R., Scarcella, K., Thaler, M., Vincent, W. F., and Lovejoy, C. (2013). Small phytoplankton in Arctic seas: vulnerability to climate change. Biodiversity, 14(1), 2–18. Terrado, R., Vincent, W., and Lovejoy, C. (2009). Mesopelagic protists: diversity and succession in a coastal Arctic ecosystem. Aquat. Microb. Ecol. 56, 25–39. doi:10.3354/ame01327. Thaler, M., and Lovejoy, C. (2015). Biogeography of heterotrophic flagellate populations indicates the presence of generalist and specialist taxa in the Arctic Ocean. Appl. Environ. Microbiol. 81, 2137–2148. doi:10.1128/AEM.02737-14. Thaler, M., Vincent, W. F., Lionard, M., Hamilton, A. K., and Lovejoy, C. (2017). Microbial community structure and interannual change in the last Epishelf Lake ecosystem in the North Polar Region. Front. Mar. Sci. 3, 275. doi:10.3389/fmars.2016.00275. Thoisen, C., Riisgaard, K., Lundholm, N., and Nielsen, T. G. (2015). Effect of acidification on an Arctic phytoplankton community from Disko Bay, West Greenland. Mar. Ecol. Prog. Ser. 520, 21–34. Timmermans, M.-L., Toole, J., Proshutinsky, A, Krishfield, R., and Plueddemann, A. (2008). Eddies in the Canada Basin, Arctic Ocean, observed from ice-tethered profilers. J. Phys. Oceanog., 38(1), 133–145. Tittel, J., Bissinger, V., Zippel, B., Gaedke, U., Bell, E., Lorke, A., et al. (2011). Mixotrophs combine resource use to outcompete specialists: Implications for aquatic food webs. Proc. Natl. Acad. Sci. U.S.A. 100, 12776–12781. doi:10.1073/pnas.2130696100. Tomas, C. R. (1997). Identifying marine phytoplankton. San Diego, CA: Academic Press.

138

Tong, M., Smith, J. L., Kulis, D. M., and Anderson, D. M. (2015). Role of dissolved nitrate and phosphate in isolates of Mesodinium rubrum and toxin- producing Dinophysis acuminata. Aquat. Microb. Ecol. 75, 169-185. doi:10.3354/ame01757. Torti, A., Alexander, M., and Barker, B. (2015). Marine genomics origin, dynamics, and implications of extracellular DNA pools in marine sediments. Mar. Genomics 24, 185–196. doi:10.1016/j.margen.2015.08.007. Tremblay, G., Belzile, C., Gosselin, M., Poulin, M., Roy, S., and Tremblay, J. (2009). Late summer phytoplankton distribution along a 3500 km transect in Canadian Arctic waters: strong numerical dominance by picoeukaryotes. Aquat. Microb. Ecol. 54, 55–70. doi:10.3354/ame01257. Tremblay, J.-É., Bélanger, S., Barber, D. G., Asplin, M., Martin, J., Darnis, G., et al. (2011). Climate forcing multiplies biological productivity in the coastal Arctic Ocean. Geophys. Res. Lett. 38, L18604, doi:10.1029/2011GL048825. Tremblay, J.-É., Simpson, K., Martin, J., Miller, L., Gratton, Y., Barber, D., and Price, N. M. (2008). Vertical stability and the annual dynamics of nutrients and chlorophyll fluorescence in the coastal, southeast Beaufort Sea. J. Geophys. Res., 113, C7S90. doi:10.1029/2007JC004547 Troost, T. A., Kooi, B. W., and Kooijman, S. A. L. M. (2005). Ecological specialization of mixotrophic plankton. Am. Nat. 166, E45--E61. doi:10.1086/432038. Vaulot, D., Eikrem, W., and Viprey, M. (2008). The diversity of small eukaryotic phytoplankton (<3 µm) in marine ecosystems. FEMS Microbiol., 32, 795-820. doi:10.1111/j.1574- 6976.2008.00121.x. Vellend, M., Srivastava, D. S., Anderson, K. M., Brown, C. D., Jankowski, J. E., Kleynhans, E. J., et al. (2014). Assessing the relative importance of neutral stochasticity in ecological communities. Oikos, 123, 1420–1430. doi:10.1111/oik.01493. Verni, F., and Rosati, G. (2011). Resting cysts: A survival strategy in protozoa Ciliophora. Ital. J. Zool. 78, 134–145. doi:10.1080/11250003.2011.560579. Vidussi, F., Roy, S., Lovejoy, C., Gammelgaard, M., Thomsen, H., Booth, B., Tremblay, J. E. and Mostajir, B. 2004. Spatial and temporal variability of the phytoplankton community structure in the North Water Polynya, investigated using pigment biomarkers. Can. J. Fish. Aquat. Sci. 61:2038–52. Vincent, W. F. (2000). Cyanobacterial dominance in the Polar Regions. In: Whitton BA, Potts M (eds) Ecology of the cyanobacteria: their diversity in time and space. Kluwer, Dordrecht, 321– 340, 669 pp. Wang, J., Cota, G. F., and Comiso, J. C. (2005). Phytoplankton in the Beaufort and Chukchi Seas: Distribution, dynamics, and environmental forcing. Deep Sea Res. Part II Top. Stud. Oceanogr. 52, 3355–3368. doi: 10.1016/j.dsr2.2005.10.014. Ward, B. A., and Follows, M. J. (2016). Marine mixotrophy increases trophic transfer efficiency, mean organism size, and vertical carbon flux. Proc. Natl. Acad. Sci. U. S. A. 113, 2958–2963. doi:10.1073/pnas.1517118113. Wassmann, P., Kosobokova, K.N., Slagstad, D., et al. (2015). The contiguous domains of Arctic Ocean advection: Trails of life and death. Prog. Oceanog. 139, 42-65. Watanabe, E. (2011). Beaufort shelf break eddies and shelf-basin exchange of Pacific summer water in the western Arctic Ocean detected by satellite and modeling analyses. J. Geophys. Res., 116, C08034. doi:10.1029/2010JC006259 Webb, C.O., Ackerly, D.D., McPeek, M.A., and Donoghue, M.J. (2002). Phylogenies and community ecology. Annu Rev Ecol Syst 33: 475–505.

139

Weston, K. (2005). Primary production in the deep chlorophyll maximum of the central North Sea. J. Plank. Res., 27(9), 909–922. Wickham, H. (2009). ggplot2: Elegant graphics for data analysis. Springer-Verlag New York. Wilken, S., Schuurmans, J. M., and Matthijs, H. C. P. (2014). Do mixotrophs grow as photoheterotrophs? Photophysiological acclimation of the chrysophyte Ochromonas danica after feeding. New Phytol. 204, 882–889. doi:10.1111/nph.12975. Winton, M. (1999). Polar water column stability. J. Phys. Oceanogr. 29, 1368–1371. doi:10.1175/1520-0485(1999)029<1368: PWCS>2.0.CO;2. Wolf, C., Kilias, E., and Metfies, K. (2015). Protists in the Polar Regions: comparing occurrence in the Arctic and Southern oceans using pyrosequencing. Pol. Res., 34, 23225. Woodgate, R. A., Aagaard, K., Swift, J. H., Falkner, K. K., and Smethie, W. M. (2005). Pacific ventilation of the Arctic Ocean’s lower halocline by upwelling and diapycnal mixing over the continental margin. Geophys. Res. Lett. 32, 1–5. doi:10.1029/2005GL023999. Yamamoto-Kawai, M., McLaughlin, F. A. and Carmack, E. C. (2013), Ocean acidification in the three oceans surrounding northern North America, J. Geophys. Res. Oceans, 118, 6274–6284, doi:10.1002/2013JC009157. Yamamoto-Kawai, M., Carmack, E. C. and McLaughlin F. A. (2006). Nitrogen balance and Arctic throughflow, Nature, 443, 43-43, doi:10.1038/443043a. Yamamoto-Kawai, M., McLaughlin, F.A., Carmack, E. C., Nishino, S., Shimada, K., and Kurita, N. (2009). Surface freshening of the Canada Basin, 2003–2007: River runoff versus sea ice meltwater. J. Geophys. Res., 114, C00A05. doi:10.1029/2008JC005000 Yamamoto-Kawai, M., McLaughlin, F.A., Carmack, E. C., Nishino, S., and Shimada, K. (2008). Freshwater budget of the Canada Basin, Arctic Ocean, from salinity, 18O, and nutrients. J. Geophys. Res. 113, C01007. doi:10.1029/2006JC003858. Yang, J. (2006). The seasonal variability of the Arctic Ocean Ekman transport and its role in the mixed layer heat and salt fluxes. J. Clim. 19, 5366–5387. doi:10.1175/JCLI3892.1. Zhan, X., and J. Zhang (2001), Heat and freshwater budgets and pathways in the Arctic Mediterranean in a coupled ocean/sea-ice model, J. Oceanogr., 57, 207–234. Zhang, J., Lindsay, R., Steele, M., and Schweiger, A. (2008). What drove the dramatic retreat of arctic sea ice during summer 2007? Geophys. Res. Let., 35(11), L11505. doi:10.1029/2008GL034005. Zhang, Y., Zhao, Z., Dai, M., Jiao, N., and Herndl, G. J. (2014). Drivers shaping the diversity and biogeography of total and active bacterial communities in the South China Sea. Mol. Ecol. 23, 2260–2274. doi:10.1111/mec.12739. Zhou, C., and Yan, X.-J. (2002). New observations on the meiotic process in the marine dinoflagellate Noctiluca scintillans (Noctilucales, Dinophyceae). Chinese J. Oceanol. Limnol. 20, 67–73. Zhu, F., Massana, R., Not, F., Marie, D., and Vaulot, D. (2005). Mapping of picoeucaryotes in marine ecosystems with quantitative PCR of the 18S rRNA gene. FEMS Microbiol. Ecol. 52, 79–92. doi: 10.1016/j.femsec.2004.10.006.

140